Anintal-Sedintent

Relations
The Biogenic
Alteration of Sediments
 TOPICS IN GEOBIOLOGY
Series Editor: F. G. Stehli, University of Florida

Volume 1 SKELETAL GROWTH OF AQUATIC ORGANISMS

Biological Records of Environmental Change
Edited by Donald C. Rhoads and Richard A. Lutz

Volume 2 ANIMAL-SEDIMENT RELATIONS

The Biogenic Alteration of Sediments
Edited by Peter L. McCall and Michael J. S. Tevesz
Animal-Sediment
Relations
The Biogenic
Alteration of Sediments

Edited by
Peter L. McCall
Case Western Reserve University
Cleveland, Ohio

and
Michael J. S. Tevesz
Cleveland State University
Cleveland, Ohio

Springer Science+Business Media, LLC

 Library of Congress Cataloging in Publication Data

Main entry under title:

Animal-sediment relations.
(Topics in geobiology; v. 2)
Includes bibliographical references and index.
I. Benthos. 2. Sediments (Geology). 3. Marine sediments. 1. McCall, Peter L.,
1948- . II. Tevesz, Michael J. S. III. Series.
QH90.8.B46A54 1982 574.5'2636 82-16523
ISBN 978-1-4757-1319-0 ISBN 978-1-4757-1317-6 (eBook)
DOI 10.1007/978-1-4757-1317-6

© 1982 Springer Science+Business Media New York

Originally published by Plenum Press, New York in 1982
Softcover reprint of the hardcover I st edition 1982
A Division of Springer Science+Business Media, LLC

All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the Publisher
Contributors

Robert C. Aller Department of the Geophysical Sciences, University of

Chicago, Chicago, Illinois 60637
Larry F. Boyer Department of Geology and Geophysics, Yale University,
New Haven, Connecticut 06511
Charles W. Byers Department of Geology and Geophysics, University
of Wisconsin, Madison, Wisconsin 53706
J. Berton Fisher Department of Geological Sciences, Case Western Re-
serve University, Cleveland, Ohio 44106. Present address: Research
Center, Amoco Production Company, Tulsa, Oklahoma 74102
Gerald Matisoff Department of Geological Sciences, Case Western Re-
serve University, Cleveland, Ohio 44106
Peter L. McCall Department of Geological Sciences, Case Western Re-
serve University, Cleveland, Ohio 44106
Donald C. Rhoads Department of Geology and Geophysics, Yale Uni-
versity, New Haven, Connecticut 06511
Michael J. S. Tevesz Department of Geological Sciences, Cleveland State
University, Cleveland, Ohio 44115

v
Preface

In 1881, Charles Darwin published The Formation of Vegetable Mould

through the Action of Worms. In his book he described the feeding ac-
tivities of terrestrial oligochaetes and their effect on the physical and
chemical properties of the soil and soil fertility. In his autobiography he
confided:
I have now (May 1, 1881) sent to the printers the MS. of a little book on The
Formation of Vegetable Mould through the Action of Worms. This is a subject
of but small importance; and I know not whether it will interest any readers,
but it has interested me. It is the completion of a short paper read before the
Geological Society more than forty years ago, and has revived old geological
thoughts.

Darwin was wrong. In the two years following the publication of the book,
8500 copies were sold. The book was more immediately popular than
The Origin of Species. (Of course, it also followed the publication of that
by now famous book on evolution.) Later workers were to confirm Dar-
win's observations on the effects of oligochaetes on soil structure and
fertility and further document soil alterations produced by other animals
(Taylor, 1935; Satchell, 1958; Edwards and Lofty, 1972).
Still other studies have shown that the biogenic alteration of sedi-
ments is a pervasive process in aquatic environments as well. Dapples
(1942) authored one of the first large-scale reviews of aquatic
animal-sediment interactions and emphasized that benthic macroinver-
tebrates profoundly affect sediment properties and processes in the ma-
rine realm:
In many, but not all, environments in which marine deposits accumulate
benthonic life is abundant. Certain of these organisms burrow for shelter or
food, or they may ingest the sediments in search of any contained organic
material. Either of these two behaviors results in some alteration of the sedi-
ments already deposited. A list of such changes includes obliteration of strat-
ification, destruction of gradation of grain size resulting from normal settling

vii
viii Preface

through an aqueous medium, trituration and solution of rock masses and frag-
ments, formation of tubes and burrows, addition of faecal matter, initiation of
cementation, bleaching of the sediments, and reduction of the amount of con-
tained organic material.
On the basis of present meagre quantitative information, there is reason
to believe that the larger organisms which contribute to diagenetic changes are
holothurians and worms, whereas, rock boring organisms and echinoderms are
secondary in importance. Obviously, the degree to which any of the alterations
already listed may take place is dependent upon the quantity of these organisms
dwelling in any particular area and, hence, is in turn dependent upon the
localities which possess the optimum living conditions.

Rhoads (1974) recently summarized the effects of marine bottom-

dwelling organisms on the properties of mud bottoms in shallow marine
environments, and Rowe (1974) produced a sketchier outline for the less
well-known deep-sea environments. Their conclusion, based on quanti-
tative results, was that benthic flora and fauna, especially macroinver-
tebrates lprimarily molluscs, annelids, and arthropods) are able to pro-
duce large changes in the physical and chemical properties of marine
sediments. Petr (1977) produced a short review of the early literature on
the effects of invertebrates on freshwater sediments.
As a perusal of the reference lists of any of the following chapters
demonstrates, there has been a great deal of work on animal-sediment
relations in the last several years. And, as the text of these chapters dem-
onstrates, there have appeared a number of new ideas concerning the
ways in which organisms affect sediments. We have reviewed
animal-sediment relations in comparable aquatic environments where
organisms are liable to have large and observable effects on bottom prop-
erties-Recent marine (Part I) and freshwater (Part II) soft bottoms (un-
consolidated fine clastics )-and have tried to add a time dimension by
examining what kinds of information have been extracted from the rock
record of these two environments (Part III). The volume and sophistication
of our present knowledge of animal-sediment relations descend in the
same order. There has been a greater number of workers working for a
longer period of time on marine than on nonmarine animal-sediment
relations. There is a smaller amount of less conclusive work on ancient
marine animal-sediment relations. We know the least about ancient fresh-
water sediments.
In Chapter 1, Rhoads and Boyer eschew a review of the kinds and
rates of sediment mixing in marine sediments, since Aller (1978) and Lee
and Schwarz (1980) have already reviewed the area. They have instead
concentrated on biogenic changes in physical and chemical properties
that occur in the course of the ecological succession of organisms follow-
ing some disturbance of the bottom. They pay detailed attention to the
Preface ix

many activities of organisms that affect sediment transport and particu-

larly emphasize the interrelated influence of tubes, pelletal layers, and
microbial mucopolysaccharide binders on entrainment properties. In
Chapter 2, Aller examines the numerous ways in which macrobenthos
alter the horizontal and vertical distribution of particles and solutes in
marine sediments and alter the rate of diagenesis of organic matter. Ma-
crobenthos not only alter the flux of materials between sediments and the
overlying water column, but also influence the temporal and spatial dis-
tributions of other benthic organisms.
In Chapter 3, McCall and Tevesz review the present state of knowl-
edge concerning the effects of the major groups of freshwater lake ma-
crobenthos on the physical properties and transport of clastic lake sedi-
ments. They conclude with speculation that the animal-sediment
interactions they describe may be of great import to the study of the
evolution of benthic communities as well as to the study of sedimentologic
processes. In Chapter 4 Fisher reviews the present state of knowledge
concerning the effects of the same major groups of freshwater benthos on
chemical properties of the sediments and chemical mass transfer between
sediments and overlying water.
In Chapter 5, Byers reviews recent work on the relation of marine
trace fossils to sedimentology that has appeared since the volume of Frey
(1975). He ends with some ideas about what these biogenic sedimentary
structures can tell us about the history of life. In Chapter 6, Tevesz and
McCall review the meager knowledge concerning freshwater trace fossils.
The few excellent studies available serve as examples of the potential for
reconstructing the physical-chemical properties of ancient aquatic substrata.
In Chapter 7, Matisoff reviews the various kinds of quantitative
models that have been constructed to describe the effects of organisms
on particle mixing and chemical mass transfer between sediment pore
fluids and overlying waters. He compares results obtained from different
kinds of models and suggests their utility in deciphering the sedimentary
record of biological mixing.
Some readers may be unfamiliar with several terms that appear
throughout this volume. The book concerns the effects of benthos (bottom-
dwelling organisms) on sediment properties. Most authors have written
about macrobenthos (the adult stages of which remain in a 1-mm mesh
sieve during the process of separating organisms from sediment). Some
attention is also paid to meiobenthos (organisms from 1 mm to about 100
/-Lm in diameter)-including nematodes, harpacticoid copepods, ostra-
cods, and various aschelminthes-and microbenthos (organisms smaller
than 100 /-Lm in diameter)-which include protists, fungi, and bacteria.
Macrobenthos are further classified according to life habits (Fig. 1). Epi-
x Preface

Filter Feeder Deposit Feeder Herbivore Carnivore

Mobile Crustaceans Gastropods' Gastropods

EchlnOids Crustaceans
Asteroids

EPIFAUNAL Sedentary Coelenterates

Attached Sponges Coelenterates

Bivalves ( Corals)
Polychaetes
Bryozoans

Mobile Polychaetes Bivalves Gastropods

Polychaetes Crustaceans
Oligochaetes Polychaetes

INFAUNAL Sedentary Bivalves Polychaetes Coelenterates

Chlronomlds Crustaceans

Attached Bivalves

Figure 1. Life habits of benthic macroinvertebrates, with examples. There is no one-to-one

correspondence between taxa and life habits, and this list of taxa is not exhaustive. Asterisk
(*) indicates the predominant life habitat of the taxon. After Eicher and McAlester (1980).

fauna spend the majority of their time on the substratum surface; infauna
live primarily beneath the substratum surface. Vagile benthos move ac-
tively on or in the substratum, sedentary benthos do not, and attached
benthos are incapable of movement along the bottom. Filter feeders gather
their food from the water overlying the bottom or, in rare cases, from
sediment pore water. Deposit feeders gather their food by feeding directly
on unconsolidated sediment deposits. Herbivores feed selectively on live
plant material and carnivores prey on other animals.
We have not made many freshwater-marine comparisons of ani-
mal-sediment relations to identify or explain differences and common-
alities. This attempt is premature given our present state of knowledge.
We can only conclude now that functionally similar organisms have a
similar effect on similar bottoms in both marine and freshwater environ-
ments. We think that future comparisons will be more cogent. However,
we hope at least that the co-occurrence in one book of work on
animal-sediment relations in freshwater and marine environments and
in ancient rocks will promote a cross-fertilization of methods and ideas
among workers that will result in geologically and ecologically interesting
and useful conclusions.

Peter 1. McCall
Michael J. S. Tevesz
Preface xi

References
Aller, R. C., 1978, Experimental studies of changes produced by deposit feeders on pore
water, sediment, and overlying water chemistry, Am . J. Sci. 217:1185-1234 .
Dapples, E. C., 1942, The effect of macro-organisms upon near shore marine sediments, J.
Sediment. Petro1. 12:118-126.
Edwards, C. A., and Lofty, J. R., 1972, Biology of Earthworms, Chapman and Hall, London.
Eicher, D. 1., and McAlester, A. 1., 1980, History of the Earth, Prentice-Hall, Englewood
Cliffs, New Jersey.
Frey, R. W. (ed.), 1975, The Study of Trace Fossils, Springer-Verlag, New York.
Lee, H., and Schwartz, R. C., 1980, Biological processes affecting the distribution of pollutants
in marine sediments. Part II. Biodeposition and bioturbation, in: Contaminants and
Sediments (R. A. Baker, ed.), Volume 2, pp. 555-605, Science Publishers, Ann Arbor,
Michigan.
Petr, T., 1977, Bioturbation and exchange of chemicals in the mud-water interface, in:
Interactions Between Sediments and Fresh Water (H. 1. Golterman, ed.), pp. 216-226,
Junk, The Hague.
Rhoads, D. C., 1974, Organism-sediment relations on the muddy sea floor, Oceanogr. Mar.
Bioi. Annu. Rev. 12:263-300.
Rowe, G. T., 1974, The effects of the benthic fauna on the physical properties of deep-sea
sediments, in: Deep Sea Sediments (A. L. Inderbitzen, ed.), pp. 381-400, Plenum Press,
New York.
Satchell, J. E., 1958, Earthworm biology and soil fertility, Soil Fert. 21:209-219.
Taylor, W. P., 1935, Some animal relations to soils, Ecology 16:127-136.
Contents

I. Recent Marine Environments

Chapter 1 . The Effects of Marine Benthos on Physical

Properties of Sediments: A Successional
Perspective
Donald C. Rhoads and Larry F. Boyer
1. Introduction ................. .......................................... .................... 3
2. Ecological Succession and Organism-Sediment Relations .... 5
3. Ecological Succession and Geotechnical Properties: A
Laboratory Experiment ........... .............. .................. ........ ........ ... 10
4. Sediment Transport and Biogenic Activity................ ........ ..... 19
5. A Qualitative Predictive Model................................................ 36
6. Recommendations for Future Work ......................................... 40
References ............................................. ................ ..................... 43

Chapter 2 . The Effects of Macrobenthos on Chemical

Properties of Marine Sediment and Overlying
Water
Robert C. Aller
1. Introduction .............. ..... ............................................................ 53
2. Diagenetic Reactions ................................................................. 54
3. Reactive Particle Redistribution ............................................... 56
4. Solute Transport ........................................................................ 71
5. Macrofaunal Influence on Sediment-Water Exchange Rates. 83
6. Reaction Rates ....... .................................................. .... .............. 87
7. Chemistry of the Burrow Habitat ....... ....................................... 89
xiii
xiv Contents

8. Spatial and Temporal Patterns in Sediment Chemistry.......... 92

9. Summary.............. ...................................... ................................ 93
Appendix: Solutions to Model Equations ............ .................... 94
References .................................................................................. 96

II. Recent Freshwater Environments

Chapter 3 . The Effects of Benthos on Physical Properties

of Freshwater Sediments
Peter 1. McCall and Michael J. S. Tevesz
1. Introduction ............................................................................... 105
2. Freshwater Sediments and Macrobenthos ............................... 106
3. Macrobenthos Life-Styles .......................................................... 113
4. Mixing of Sediments ................................................................. 124
5. Biogenic Modification of Sediment Transport ......................... 150
6. Ecological Interactions and Sediment Properties .................... 161
7. Future Work ............................................................................... 166
References .................................................................................. 168

Chapter 4 . Effects of Macrobenthos on the Chemical

Diagenesis of Freshwater Sediments
J. Berton Fisher
1. Introduction ........ ...................................... ................................. 177
2. Freshwater Sediment System ................................................... 178
3. Mechanisms by Which Benthos Affect Chemical Diagenesis 186
4. Observed Effects of Freshwater Macrobenthos on Chemical
Diagenesis .................................................................................. 200
5. Conclusions ............................................................................... 209
References .................................................................................. 211

III. Ancient Environments

Chapter 5 . Geological Significance of Marine Biogenic
Sedimentary Structures
Charles W. Byers
1. Introduction ............................................................................... 221
2. Traces and Sedimentology........................................................ 222
Contents xv

3. Paleobathymetry ........................................................................ 233

4. Precambrian Traces ....... ......... ............................ ....................... 243
References .................................................................................. 252

Chapter 6 Geological Significance of Aquatic Nonmarine

Trace Fossils
Michael J. S. Tevesz and Peter L. McCall
1. Introduction ............................................................................... 257
2. Identification of Traces ........ ..................................................... 259
3. Trophic Level Reconstruction .................................................. 270
4. Paleoenvironmental Reconstruction ......................................... 271
5. Discussion ........ .......................................................................... 277
6. Concluding Remarks ..... .................................. .................. ........ 280
References .................................................................................. 281

IV. Models

Chapter 7 Mathematical Models of Bioturbation

Gerald Matisoff
1. Introduction ............. .................................................................. 289
2. Particle Transport Models ......................................................... 293
3. Fluid Transport Models ............................................................ 313
4. Conclusions .......... ................ ..................................................... 325
References ............. .................................................................... . 327

Index .............................. ................................................................... 331

 I
Recent Marine Environments
Chapter 1
The Effects of Marine Benthos on
Physical Properties of Sediments
A Successional Perspective
DONALD C. RHOADS and LARRY F. BOYER

1. Introduction ... ... ..... .. .... ....... ... ............ .. ... .. ......... .. .. .. ........ .. ... .. ...... ............. .. .. ....... .. ..... 3
2. Ecological Succession and Organism-Sediment Relations ........ ... ............. .......... .. .. 5
2.1. Pioneering Stages ...... ... .............................................................. ...................... .. .. 7
2.2 . Equilibrium Stages ... .. ............ .. ... .. ........ .. .... .. ............. ........ .. .. .. ............ .......... .. .... 9
3. Ecological Succession and Geotechnical Properties: A Laboratory Experiment ..... 10
3.1. Methods ........... .. ........................ .... .................................. .... ......... ... ........... .... ...... 13
3.2. Results .......... ....... .... .. ....... ... ... .... ....... .. ..... .. ... .... ... .. .. ....... .. .. .. .......... .. .. .. ... .. .. ... .. ... 15
3. 3. Conclusions .... .... .. .. .. ..... ... ............................... .. ....... ......... ..... .. .. .. ... ... .... ........... ... 17
4. Sediment Transport and Biogenic Activity............ .. ..... .. ................................... .. ..... 19
4.1. Bed Roughness .... .. .. ..... ................. .. ............. .... .. .. ... .. .. .... .. ... .. .... .. .. .. .......... .. .. ... ... 19
4.2. Pelletal Textures ....... ... ... .. ................ .. .... ... ... .... ............ .. .. ... ... .... ... ............ ........ .. 24
4.3. Geotechnical Mass Properties .................................................... .. .. .. ................... 29
5. A Qualitative Predictive Model.... ... .. ............. .. ............. .. .. .. ............................... .. .. .. .. 36
6. Recommendations for FuturE! Work .. .............. .. ........................................ .. ........ ....... 40
References .......... .. .. .......... ...... ............................ .......... ................................................ 43

1. Introduction
The effects of benthic organisms on the physical properties of granular
substrata are well documented . The range of effects has been presented
in H. B. Moore (1931, 1939), Schwartz (1932), Dapples (1942), D. G. Moore
and Scruton (1957), McMaster (1967), Rhoads (1974), Rowe (1974), Powell
(1974), Richards and Park (1976) , Myers (1977a,b), Self and Jumars (1978),
Lee and Swartz (1980), and Carney (1981). These papers relate the effects
of benthic species to changes in grain size, sorting, fabric, water content,

DONALD C. RHOADS and LARRY F. BOYER • Department of Geology and Geophysics,

Yale University, New Haven, Connecticut 06511.

3
4 Chapter 1

compaction, shear strength, and bottom stability. Those autecologic pa-

rameters that appear to be most highly correlated with physical modifi-
cations of sediments include: method of feeding, feeding selectivity, feed-
ing level relative to the sediment-water interface, degree of mobility,
organism size and population density, burrowing depth, and, if the or-
ganism is a tube dweller, the density, spacing, and length of tubes.
The influence of most invertebrate benthos on the biological benthic
boundary layer* is limited to a relatively small vertical zone, i.e., from
a few centimeters above the sediment surface to a few decimeters below
the bottom. Important exceptions exist to this generalization, most notably
shrimp that burrow to depths of 3-4 m (Pemberton et al., 1976; Myers,
1979) and certain fish that can burrow to depths of 60-70 cm (Cool, 1971;
Atkinson and Pullin, 1976).
In view of the generally surficial effects of benthos on the physical
properties of sediments, why study this phenomenon? Our interest in the
upper few centimeters of sediment is based on the fact that the interface
is both biologically active and chemically reactive. The physical prop-
erties of sediments, as controlled by biological processes, may have a
major effect on sedimentation, sediment transport, nutrient regeneration ,
and the fates and historical records of radiochemical species, pollutants,
and fossils. Biology, chemistry, and sediments are coupled through the
process of bioturbation, involving the transport of particles, as well as
pumping water into, and out of, the bottom. Aller (1977, 1978) has re-
viewed the literature on bioturbation, and makes the observation that
water pumping is about 100 times greater (on a weight-to-weight basis)
than particle bioturbation. We will see that both fluid and particle bio-
turbation can affect the physical properties of sediments.
The dispersal or concentration of potentially harmful environmental
contaminants may be controlled by the interactions of both particulate
and dissolved pollutants with the seafloor. The way a particular benthic
organism or suites of benthic organisms bioturbate, or otherwise process
the seafloor, can determine if a pollutant is entrapped or mixed into the
bottom or, alternatively, transported and dispersed (e.g., Phelps, 1966;
Haven and Morales-Alamo, 1972; Swartz and Lee, 1980; Lee and Swartz,
1980; Berger and Heath, 1968; Hanor and Marshall, 1971; Aller and Coch-
ran, 1976; Jumars et al., 1981).
A small but increasing body of data indicates that the response of
sediments to fluid shear stress is significantly affected by their biological

* Here defined as the height above and below the sediment-water interface that is occupied ,
or otherwise influenced, by benthic organisms. The physical effects of organisms extend
spatially beyond their immediate "biospace." For example, the current velocity profile,
as measured several centimeters away from the bed, may be related to topographic bound-
ary conditions controlled by biological features .
Effects of Marine Benthos on Sediment Physical Properties 5

components. Plants and invertebrates are known to trap and bind sedi-
ments or promote sedimentation (Ginsburg and Lowenstam, 1958; Fager,
1964; Neuman et 01., 1970; Marshall and Lucas, 1970; Scoffin, 1970; M.
Lynch and Harrison, 1970; Holland et 01., 1974; Rhoads et 01., 1978b;
Yingst and Rhoads, 1978; Boyer, 1980; Nowell et 01., 1981; Eckman et 01.,
1981). Sediments can also become destabilized by organisms relative to
azoic or bound conditions (Rhoads and Young, 1970; Southard et 01.,
1971; D. K. Young, 1971; R. A. Young and Southard, 1978; Rhoads et 01.,
1978b; Yingst and Rhoads, 1978; Boyer, 1980; Nowell et 01.,1981; Eckman
et 01., 1981; Grant et 01., 1982).
It is the purpose of this chapter to focus on those physical
organism-sediment relations that are of potential value in predicting the
physical properties of sediments and the transport fates of fine-grained
sediments and their associated contaminants. We will limit our discussion
primarily to macrofaunal effects on subtidal muds. The geochemical as-
pects of this problem are covered in Chapter 2 of this volume. We also
direct the reader to an excellent recent review of biodepositional and
bioturbational effects on the distribution of pollutants in sediments (Lee
and Swartz, 1980). We will not cover biogenic rate processes here (see
Aller, 1978, and Lee and Swartz, 1980). Rather, our contribution will
focus on how organism-sediment relations develop during ecological
succession of the seafloor. * Generalizations will be made about how early
and late successional stages influence the physical properties of sedi-
ments-some of the cause-and-effect relationships are well known, while
others remain speculative. The successional paradigm for interpreting
organism-sediment relations is new, and, therefore, poorly tested. New
kinds of measurements and observational techniques will be required to
explore and test this model. We will offer a few promising techniques and
approaches which we feel are necessary for conducting future work.

2. Ecological Succession and Organism-Sediment

Relations
Until recently, methods of describing bottom communities and their
relationships to sediments followed, more or less, the procedures outlined

* The term succession has many connotations in the literature (see McIntosh, 1980, for a
discussion of the history of the successional concept). We define primary succession as
the predictable appearance of macrobenthic invertebrates belonging to specific functional
types following a benthic disturbance. These invertebrates interact with sediment in spe-
cific ways. Because functional types are the biological units of interest for this study, our
definition does not demand a sequential appearance of particular invertebrate species or
genera.
6 Chapter 1

by C. G. J. Petersen (1913) and popularized by G. Thorson (1957). The

classical Danish method of defining benthic communities consists of
identifying characterizing species. Characterizing species are those that
are temporally and spatially persistent, dominate in biomass, and have
relatively long life-spans. These criteria select for "equilibrium" species,
which live on parts of the seafloor where physical disturbance is uncom-
mon. Early organism-sediment work was also done from this "steady-
state" point of view. In addition, much of this early work was limited to
the sedimentary effects of individuals of a species or single-species pop-
ulations (e.g., Crozier, 1918; Dapples, 1942; Rhoads, 1963; Gordon, 1966).
A notable exception to this generalization is the early German work in

Figure 1. Development of infaunal assemblages over time following a major disturbance (A)
and along chronic pollution gradients (B). Pioneering species (left side) tend to be tubicolous
or otherwise sedentary organisms that live near the sediment surface and feed near the
surface or from the water column. Particle bioturbation rates are low. High-order successional
stages (right side) tend to be dominated by bioturbating infauna that feed at depth within
the sediment. Particle mixing by bioturbation is intensive. Figure from Rhoads et oJ. (1978a).
The pollution gradient diagram (B) is modified from Pearson and Rosenberg (1976) .
Effects of Marine Benthos on Sediment Physical Properties 7

the North Sea, where the effects of whole faunas on sediments were
considered (see Schafer, 1972).
A more ecologically realistic concept of benthic communities was
proposed by Johnson (1971, 1972), who suggested that a benthic com-
munity is a temporal and spatial mosaic, "parts of which are at different
levels of succession . . . in this view the community is a collection of
relics of former disasters." This successional perspective has its origin in
plant ecology as popularized by Clements (1916).
Subsequent experimental and field work by McCall (1977), Myers
(1977a,b), Pearson and Rosenberg (1978), Rhoads et a1. (1977), Wolff et
a1. (1977), Santos and Bloom (1980), and Santos and Simon (1980) helped
to define the taxonomic and functional structure of these "temporal mos-
aics." This dynamic approach to community structure suggests that earlier
work on subtidal organism-sediment relations be reevaluated in terms
of a successional paradigm, and that future work test the validity of these
successional dynamics.
Figure 1 depicts organism-sediment relations as they develop during
succession of a subtidal granular bottom. Most of the information about
this successional pattern comes from mud bottoms (e.g., McCall, 1977;
Pearson and Rosenberg, 1978; Rhoads et a1., 1977, 1978a), but major qual-
itative features of this succession are apparently also shared with sand-
dwelling faunas (e.g., Myers, 1977a,b; Dauer and Simon, 1975).

2.1. Pioneering Stages

Pioneering stages usually occur near shore, above the mean shoaling
depth of storm waves. Disturbance assemblages may, however, be found
at any water depth-even in abyssal trenches (Jumars and Hessler,
1976)-and in any sediment type. The controlling factors are related to
the intensity and frequency of disturbance* (Osman and Whitlatch, 1978).
Small, opportunistic, tube-dwelling polychaetes are the first faunal
components to colonize a new or newly disturbed bottom. Their aggre-
gations may reach densities of 10 5 m - 2 within a few days to weeks after
a disturbance (McCall, 1977; Rhoads et a1., 1978a). Tubicolous amphipods
appear shortly after the polychaete acme.
Most pioneering species feed near the sediment surface or from the
water column. Tube walls serve to isolate the colonizing organisms from

* The example of disturbance used in our review is the physical reworking of the seafloor
by waves and currents. This is only one category of disturbance that can restructure benthic
communities. For a comprehensive review of the physical, biological, and chemical per-
turbations that can affect macrofauna I succession we refer the reader to Pearson and
Rosenberg (1978) and Dayton and Oliver (1980).
8 Chapter 1

the sediment by controlling the rate of diffusion of ambient pore water

solutes into the tube environment (Aller, 1980).
An organism colonizing a newly disturbed bottom or new sediment
is faced with formidable physiological problems. The pore water chem-
istry will be unpredictable, especially regarding the vertical concentration
gradients of pore water metabolites, metal sulfides, dissolved oxygen, *
and organic decomposition products. Reduced compounds present at high
concentrations in the subsurface sediment diffuse across the tube walls
and enter the tube interior. Aller (1980) has shown that organisms oc-
cupying closely spaced, small-diameter tubes efficiently share the work
(measured as tube irrigation) required to maintain a constant and low
concentration of solutes (e.g., NH 4 ) in their tubes. The unpredictable
chemical and trophic environment below the sediment-water interface
may also explain why most pioneering species feed at, or above, the
sediment-water interface.
Most pioneering species have adaptive strategies that have been de-
scribed as r-selected, while many infaunal deposit-feeding species share
K-selected attributes (McCall, 1977; Rhoads et a1., 1978a). Vermeij (1978)
has pointed out that the often-cited opportunistic-to-equilibrium (or r-
to-K) end-member classification of benthic marine communities does not
represent the full range of possible adaptive types. He recognizes a third
end-member called stress-tolerant, which is representative of species that
predominantly inhabit physiologically stressful areas such as the inter-
tidal zone. At this time, the organism-sediment successional model (Fig.
1) does not consider this stress-tolerant group of species. However, it is
interesting to note that many intertidal species such as Ensis directus
(bivalvia) make brief appearances in subtidal pioneering stages (McCall,
1977).
In summary, the sedimentary effects of pioneering species are limited
to the near-surface region of the bottom «2 cm) and they include the
following:
1. Construction of dense tube aggregations, which may affect micro-
topography and bottom roughness on a scale dictated by tube di-
ameter, tube height, and tube spacing.

* Mapping the depth of the redox potential discontinuity (RPO) below the surface of marine
sediments has proven useful for assessing the affect of organic pollution on the benthic
ecosystem (Pearson and Stanley, 1979; Rhoads and Germano. 1982). The cause-and-effect
relationships are complex. In areas of the seafloor experiencing organic enrichment, bi-
ological oxygen demand and chemical oxygen demand are highly correlated with organic
decomposition. This oxygen demand causes the Eh = 0 horizon to rebound toward the
sediment surface. The enrichment is also associated with the appearance of pioneering
species, which are relatively ineffective in exchanging ambient pore water with overlying
water; hence the RPO remains near the sediment surface.
Effects of Marine Benthos on Sediment Physical Properties 9

2. Fluid bioturbation, which pumps water into and out of the bottom
through vertically oriented tubes. Particle bioturbation, although
present, is of subordinate importance.
3. Surface deposit feeding and suspension feeding, which cover the
surface of the bottom with fecal pellets, especially the fusiform
pellets of opportunistic polychaetes.

2.2. Equilibrium Stages

High-order successional stages are found where major physical dis-

turbances such as intense annual storms rarely affect the bottom. Exam-
ples have been described for the Maldanid-Nucula-Syndosmya (poly-
chaete-bivalve) community in the Clyde Sea (H. B. Moore, 1931), The
Nucula-Nephtys (bivalve-polychaete) assemblage in Buzzards Bay
(Sanders, 1958; Rhoads and Young, 1970), the Molpadia-Euchone (hol-
othurian-polychaete) community of Cape Cod Bay (Rhoads and Young,
1971; D. K. Young and Rhoads, 1971), the Maldane-Amphiura (poly-
chaete-ophiuroid) community of Ria de Muros, Spain (Tenore et al.,
1982), and the Lumbrineris-Alpheus-Diolodonta (polychaete-shrimp-
bivalve) community in Kingston Harbor, Jamaica (Wade, 1972). A late
successional stage is dominated by infaunal deposit feeders, many
of which are represented by "head-down" conveyor belt feeders
(sensu Rhoads, 1974). Some of these species are tubicolous (e.g., the
polychaete family Maldanidae) but many others are free-living. Equi-
librium assemblages are associated with a deeply oxygenated sediment
surface where the RPD commonly reaches depths of over 10 cm. Feeding
is concentrated at, but not limited to, the RPD. The RPD is in fact related
to the feeding depth. * The redox zone appears to be a region of high
microorganism productivity (see Yingst and Rhoads, 1980). Both vertical
particle mixing and pore water exchange by respiratory pumping are
important bioturbational processes.
Organism-sediment relations are complex and well developed in
equilibrium systems. In summary, the sedimentary effects of equilibrium
species in shallow water environments appear to be as follows:
1. Transfer of both water and particles over vertical distances of up
to 10-20 cm takes place.

* The depth of the redox boundary can be seen as a brown oxidized layer at the surface of
the sediment, or locally around tube and burrow walls. In sediments high in iron content,
this colored layer may not, in many cases, reflect the presence of free molecular oxygen
in sediment pore water. Once oxidized, ferric iron may persist long after the associated
pore water has been depleted of dissolved oxygen (Revsbech et oj., 1979).
10 Chapter 1

2. Intensive particle mixing produces homogeneously mixed fabrics

and many of the particles at and below the sediment surface may
be in the form of fecal pellets.
3. Head-down feeding produces void spaces (feeding pockets) at
depth within the bottom.
4. The RPD is located over 2 cm into the bottom and commonly to
depths of 10-20 cm.
5. Surface microtopography may be featureless and planar if tidal
resuspension "smoothes" over biologically produced features at
the sediment surface, or, in the absence of the smoothing effect,
the surface may be covered with numerous feeding pits and fecal
or excavation mounds.
In some cases pioneering and equilibrium species may coexist in the
same sediment, if physical disturbance involves only near-surface sedi-
ment. Large and deeply burrowed infaunal species may not be affected
by such small-scale disturbances (e.g., Woodin, 1978). The pioneers may
arrive as settling larvae, stimulated by chemical cues associated with the
disturbed sediment (e.g., Gray, 1974). One of us (DCR) has seen mixing
of low- and high-order successional species in Long Island Sound by the
passive wash-in of opportunists from adjacent habitats, but this mixing
represents a transient state. Pioneering species may be excluded from
equilibrium assemblages by competitive exclusion mechanisms such as
trophic group amensalism (Rhoads and Young, 1970). Thayer (1979) has
described this phenomenon as biological "bulldozing."

3. Ecological Succession and Geotechnical Properties: A

Laboratory Experiment
If sedimentary properties are influenced by the stage of succession,
one should be able to observe changes in geotechnical parameters by
experimentally introducing the different successional stages into the same
sediments and observing the effects of dominant species on geotechnical
properties. We conducted such an experiment using sediments and or-
ganisms from Long Island Sound (Table I).
A study site called FOAM* is located in 14 m of water approximately
1 mile from shore on a mud bottom. This station is maintained in a low-
order successional stage by seasonal resuspension of the seafloor. Several

* Station FOAM is located near McCall's (1975, 1977) site A. The FOAM site has been the
location of intensive geochemical investigations. The initials are, appropriately, an ac-
ronym for Friends of Anoxic Muds (Goldhaber et 01., 1977).
Effects of Marine Benthos on Sediment Physical Properties 11

Table I. Experimental Design of the Successional Geotechnical Experiment

Sediment source (tank no.)

Station FOAM a Station NWC b

Introduced macrofauna 1 2 3 4 5 6 7 8 9 10

None (control) X X X
FOAM ambient fauna Xc Xd
NWC ambient fauna X· X!
250 Nucula annulata only X
49 Yoldia limatula only X
25 Nephtys incisa only X
a 85% Silt-clay. mean grain size 0.0125 ± 0.0074 mm (N = 10) (top 10 em).
b 96% Silt-clay. mean grain size 0.0046 ± 0.0005 mm (N = 20) (top 10 em).
c 6 Nephtys incisa and 4 Clymenella torquata.
d 3 Nephtys incisa . 184 Nucula annulata. 20 Cylichnella canaliculata . 1 Cylichnella oryza. plus nu-
merous spionids not retained on the 1.0-mm sieve .
• 15 Nephtys incisa. 39 Nucula annulata. 1 Cylichnella canaliculata. 1 Nassarius. plus numerous spion-
ids not retained on the 1.0-mm sieve.
f 3 Nephtys incisa. 107 Nucula annulata. 20 Cylichnella canaliculata. 1 Nassarius trivittatus. 1 Pinnixia
sayana. 1 Sipunculida goldfingia(?).

centimeters of sediment are resuspended during storms (McCall, 1978).

The distribution of species at FOAM is patchy, and their abundance and
diversity are unpredictable. This high degree of variance arises from the
short life spans and high population turnover rates of pioneering species.
At best, one may predict that, from the total list of opportunistic species
known for Long Island Sound, a small subset of these species will dom-
inate the faunal assemblage at anyone time.
Figure 2A shows structural evidence of storm erosion and resuspen-
sion of FOAM sediments. A sediment profile photograph (Rhoads and
Cande, 1971; Rhoads and Germano, 1982) also shows the density of small
tubicolous polychaetes typically present at FOAM (Fig. 2B). In some years,
the bivalve Mulinia lateralis may arrive as an early pioneer in great abun-
dance. In this case, tubicolous polychaetes are excluded from dense M.
lateral is patches (Fig. 2C). McCall (1977) has described the faunal dy-
namics near this sampling station.
Station NWC, located in 20 m of water near the center of the sound,
is below the mean shoaling depth of storm waves and supports an equi-
librium infaunal community (McCall, 1977; Rhoads et al., 1977). Three
species persistently dominate this station in both biomass and number:
the errant polychaete Nephtys incisa and two protobranch bivalves, Yol-
dia limatula and Nucula annulata. Figure 3A shows the relatively ho-
mogenous sedimentary fabric produced by deep and intensive bioturba-
tion at this station. The sediment surface may be covered with a layer of
organic-mineral aggregates (Fig. 3B).
12 Chapter 1

Figure 2. Sedimentary fabrics and biogenic structures ill muddy sediments representing a
physically disturbed bottom. (A) X-radiograph of station FOAM, showing current laminated
sediment lying below a bioturbated interval. Note the unconformity separating these two
fabrics. (B) In situ sediment profile photograph showing dense populations of pioneering
tubicolous polychaetes typically found at station FOAM. (C) X-radiograph of a dense popu-
lation of the pioneering mactrid bivalve Mulinia lateralis also sometimes found at station
FOAM.
Effects of Marine Benthos on Sediment Physical Properties 13

Figure 3. Sedimentary tabrics and biogenic structures in muddy sediments representing a

bottom infrequently disturbed by storm. (Al X-radiograph showing the bioturbated fabric
typical of station NWC. Note the burrow-mottled structure and random orientation of shells
within the sediment. (Bl In situ sediment profile photograph of organic-mineral aggregates
at the sediment surface at station NWC. Note the absence of near-surface-dwelling benthos.
Some of the aggregates are fecal pellets; others represent pieces of cohesive surface sediment
that have been broken and moved by particle bioturbation.

3.1. Methods

The successional/geotechnical experiment was set up by bringing

defaunated sediment from stations FOAM and NWC into the laboratory.
Large macrofauna were removed by passing sediment through a 1-mm
mesh sieve. Fourteen liters of the sieved mud were placed into each of
10 Plexiglas tanks, each measuring 36 cm long, 27 cm wide, and 30 cm
deep. After settling and compaction, the depth of sediment in each tank
14 Chapter 1

"
Cl.
:.:
~
J:
25
DIAMETER e 16cm
RANGE 8;!: 2-2!S KPa
f-
<!>
Z
W 20
a:
f-
(J')

a: 15
~
W
J:
(J')
10 DIAMETER a; 22 em
Figure 4. The relationship between
0
W RANGE a; 01- 9)(Po shear vane breakaway torque and
z the undrained shear strengths of
<f
a: DIAMETER a; 29cm
our experimental muds. Three vane
0 RANGE a; 05- 5 KPo
z~
05
sizes were used to measure shear
Q 50 JQ 50 ~ 50 ~ strength over the range < 0 .5-2.5
BREAKAWAY TORQUE DIAL READING kPa (1 kPa = 10 g·cm - 2 ) .

was approximately 15 cm. Ten liters of seawater covered the sediment

in each tank and was circulated at 30 liters' hr -1 between the test tanks
and a temperature-controlled, 148-liter reservoir. Water temperatures
were regulated during the experiment to match the changing ambient
water temperature in Long Island Sound (15-22°C during the experi-
mental period). Eighteen days after sediment had been placed in the test
tanks, benthic macrofauna were introduced into the tanks as outlined in
Table 1.
Undrained sediment shear strength and sediment water content were
measured in each tank over the course of the experiment (57 days du-
ration). Vertical profiles of these two variables were measured at intervals
of 2.0-2.5 cm to depths of about 12 cm. Water contents were determined
by sectioning small cores (2.5-cm diameter) and weighing each section
(wet weight). The samples were then dried at 105°C for 3 days. The sec-
tions were weighed, and percent water and water content determined.
Shear vane measurements were made with a modified torque screwdriver
with specially constructed vanes. The three vane sizes are shown in Fig.
4, which relates their breakaway torque to undrained shear strength in
muds.
The successional/geotechnical experiment began on July 7,1980, and
terminated on September 2,1980. Geotechnical measurements were made
13 times over the 57-day period. We report here only data for the beginning
and the end of the experiment (day 57). Because of the sediment treat-
ments involved in setting up this experiment (sieving and remolding) and
the relatively short term of the experiment, we cannot directly compare
field and laboratory values of water content and undrained shear strength
from FOAM and NWC sediments. The standards of comparison in this ex-
periment are the controls which lack macrofauna :::::1 mm in size. This
experiment addresses the problem of detecting a change in geotechnical
Effects of Marine Benthos on Sediment Physical Properties 15

properties relative to the controls. The magnitude of the change can be

used to estimate qualitatively the relative effects of pioneering and equi-
librium species on geotechnical properties.

3.2. Results

3.2.1. FOAM Sediments

Vertical profiles of undrained shear strength for tank 3 (containing
the ambient FOAM fauna) and tank 2 (containing the ambient NWC fauna)
are shown in Figs. 5A and B, respectively. The data are plotted as the
difference between values measured in the two experimental tanks and
values (from the same depth intervals) in the control tank (1). Horizontal
bars graphed to the right of the 0 percentage difference datum represent
an increase in shear strength relative to the control; values plotted to the
left represent decreases.
FOAM sediments containing near-surface-living pioneer species ex-
perience a 12% increase in shear strength in the 0- to 2.5-cm depth interval
relative to the control. In comparison, a 50% decrease in shear strength
was measured for this same depth interval with the presence of subsurface
deposit feeders. The 2.5- to 5.0-cm depth interval in both tanks shows a
decrease in shear strength. This decrease is most dramatic in tank 3 (90%).
Radiographs of this sediment show the presence of methane gas voids at
depth, which might explain this reduction in shear strength. Methane
forms after pore water sulfate has been depleted. The supply of interstitial
sulfate would be expected to be lower in the less intensively bioturbated

A (tonk 3)'FO A M AMBIENT FAUNA B (tonk 2)'NWC AMBIENT FAUNA

Figure 5. The effects of pioneering and equilibrium species on undrained sediment (mud)
shear strengths. (A) Changes in undrained shear strengths of experimental FOAM muds pop-
ulated with a pioneering macrofaunal benthic assemblage (FOAM). (B) Changes in undrained
shear strengths of experimental FOAM muds populated with station NWC equilibrium species.
In A and B, plotted values of percent change in shear strengths are relative to a control tank
containing FOAM sediment without macrofauna. Shear strength changes shown are for 57
days after introduction of macrofauna. See Table 1.
16 Chapter 1

tank (Fig. 5A) compared with the more intensively bioturbated tank (Fig.
5B). Below 5 cm, both tanks show an increase in shear strength relative
to the control. This depth interval is well below the living depths of the
introduced faunas. Possible "far-field" effects of benthos on sediments
will be discussed later.
No difference in sediment water content was observed between the
experimental tanks. Surface values ranged from 60% to 61 % and declined
to about 51% at 15 cm. The shapes of the curves were also similar.

3.2.2. NWC Sediments

Vertical profiles of undrained sediment shear strengths in NWC sed-
iments, containing both Pioneering and equilibrium species, are given in
Figs. 6A-D. In this experiment two controls were run. The average profile

A (tank 4)' Nucula annulata only B (tank 5), Yoldia limatula only
% DECREASE % INCREASE % DECREASE
&>20100

o
2.5

5.0

75

10.0

c (tank 7)'M!l1hty! incisa only o (tank 8)'FO A M AMBIENT FAUNA

Figure 6. The effects of pioneering and equilibrium species on undrained sediment (mud)
shear strengths. (A) Changes in undrained shear strengths of experimental NWC muds pop-
ulated with 250 bivalves (Nucula annulata). (B) Changes in undrained shear strengths of
experimental NWC muds populated with 49 bivalves (Yoldia limatula). (C) Changes in un-
drained shear strengths of experimental NWC muds populated with 25 errant polychaete
worms (Nephtys incisal. (D) Changes in undrained shear strengths of experimenal NWC muds
populated with the pioneering FOAM fauna. In A-D. plotted values of percent change in
shear strengths are relative to the mean of two control tanks containing NWC sediment without
macrofauna. Shear strength changes shown are for 57 days after introduction of macrofauna.
See Table I.
Effects of Marine Benthos on Sediment Physical Properties 17

change in shear strength (at day 57) is used as a datum to compare meas-
ured changes in the experimental tanks. Tank 4 (Fig. 6A) contains the late
successional or equilibrium-stage bivalve Nucula annulata. This proto-
branch bivalve has a small but measurable effect of decreasing shear
strengths in the interval 0-2.5 cm. Yoldia limatula, another equilibrium-
stage protobranch, has a relatively greater effect in decreasing near-surface
shear strengths (tank 5, Fig. 6B). Yoldia lives below N. annulata and can
burrow to 3-4 cm depending on its size. Below the living depths of these
two clams, shear strengths increase relative to the controls.
The burrowing depth of the errant polychaete Nephtys incisa depends
on its length. In tank 7 (Fig. 6C), most of the burrow systems are limited
to the upper 3-4 cm, but a few large specimens penetrated to 8.5 cm. The
effect of N. incisa on sediment shear strength above 5 cm is comparable
to that of N. annulata (Fig. 6A). The increase in sediment shear strength
below the burrow-penetrated area may represent another "far-field" effect.
The decrease in shear strength in the interval 5.0-7.5 cm may be related
to the presence of methane gas voids or local influence of a large N. incisa
burrow.
Tank 8 (Fig. 6D) shows that the ambient FOAM fauna increases sedi-
ment shear strength from the surface to 7.5 cm, but that below this depth
there is no difference between the test and control tanks. Although FOAM
and NWC sediments are somewhat different in organic content and grain
size, the addition of pioneering species to these two types of muds results
in near-surface increases in shear strength (tank 3, Fig. SA, and tank 8,
Fig.6B).
Water content profiles of tanks shown in Figs. 6A-D were not sig-
nificantly different from one another, nor did they differ from those re-
ported from the FOAM sediment experiments.

3.3. Conclusions

The geotechnical experiment has shown that the successional stage

can have an important effect on sediment shear strengths. Within the 0-
to 2.5-cm surface interval, particle advection and burrow excavation as-
sociated with equilibrium species "dilates," "fluffs," or otherwise de-
creases the shear strength of the surficial sediment. Although not meas-
ured in this experiment, equilibrium species are known to pelletize
sediment intensively (Rhoads and Young, 1970; D. K. Young, 1971). This
pelletization effectively increases the modal grain size of the sediment,
reducing the number of particle-to-particle contracts, and increases sed-
iment porosity.
Pioneering species, as described earlier, live and feed near the sed-
18 Chapter 1

iment surface and are not effective in mixing sediment. When produced,
pellets are deposited on the sediment surface. The numerous spionid
polychaetes present in the FOAM assemblage pump water into and out of
the sediment through their tubes. This irrigation may stimulate near-sur-
face populations of microorganisms and meiofauna. * The apparent in-
crease in sediment shear strengths associated with this assemblage may
be related to mucus binding of particles (e.g., Martin and Waksman, 1940;
Frankel and Meade, 1973). The presence of the tubes also increases sed-
iment resistance to horizontal shearing forces.
The most interesting results of these experiments are the most dif-
ficult to interpret. Every experimental tank showed an effect below the
living depths of the introduced macrofauna which we have called a "far-
field" effect. We are cautious in our interpretation of these apparent
effects . One would need to repeat this experiment with more control tanks
to better evaluate sources of variance in the data. However, we believe
that many of the observed "far-field" phenomena may represent real bi-
ological effects. Rowe (1974) has shown that burrowing anemones can
affect sediment shear strengths as far as 20 cm away from the burrow and
tube. Aller (1978) has shown that microbial ATP and production and
consumption of decomposition products in the sediment are affected well
below the sediment actually occupied by Y. limatula. In this case, the
far-field effect is related to how the clam influences the rate of exchange
of pore water constituents such as ammonium and sulfate. Bioturbation,
especially respiratory pumping, affects the rates of chemical reactions as
well as the concentration gradients of dissolved compounds in pore water.
Another far-field effect is related to the depth of the RPD below the
sediment surface. Respiratory pumping can maintain the RPD several
millimeters or centimeters below the surface. The RPD usually lies below
the zone of living macrofauna, and is a region of intensive microbial
production (Yingst and Rhoads, 1980). It is also possible that the chemical
and physical gradients set up by macrofaunal reworking affect both the
distribution and activities of interstitial meiofauna. This size class of
organisms can also influence the physical properties of sediment (e.g.,
Cullen, 1973; Boyer, 1980) .
A possible far-field effect is related to the density or size stratification
of detrital particles at the base of the reworked zone. Silt- or sand-sized
particles with specific gravities 22.0 are reworked through the zone of
bioturbation and become concentrated at the base of this zone as a graded

* Total meiofaunal and microbial adenosine triphosphate (ATP) at the sediment surface at
FOAM is more than twice that at the sediment surface at station NWC (Aller and Yingst.
1980).
Effects of Marine Benthos on Sediment Physical Properties 19

bed (van Straaten, 1952; Rhoads and Stanley, 1965; Rhoads and Young,
1970; Cadee, 1979).
An unexpected result of this experiment was that the water content
profiles did not track measured changes in sediment shear strength. We
offer two possible explanations. First, sediment water content may be
insensitive to the biological activity present in our tanks, at least on the
time scale of 57 days. FOAM sediments in the field have summer water
contents (percent water) of between 50% and 60%'within the upper 10
cm of the sediment surface (Aller et al., 1980), while NWC sediments
normally have water contents 2:70% all year long (Rhoads et aI., 1977).
Alternatively, our procedures for storing water content cores (7°C for
approximately 1 week) may have resulted in sediment compaction, which
"erased" differences in water content.
More geotechnical experiments like the one described here are
needed to evaluate fully the relationship between succession and sedi-
ment geotechnical properties . Separate experiments must be done for
muddy and sandy substrata. Sandy sediments may behave very differently
from muds under the same bioturbational regimes, as sands tend to com-
pact rather than "dilate" when mixed (Webb, 1969; Myers, 1977a; Powell,
1977).

4. Sediment Transport and Biogenic Activity

Several biogenic structures and biologically mediated changes in sed-
imentary properties are of potential importance for determining substra-
tum stability: (1) production of bed roughness; (2) increase in modal grain
size through the formation of fecal pellets and the secretion of mucopo-
lysaccharides, which coat pellets and particles and bind them to one
another; and (3) change in sediment packing, shear strength, and water
content. Each of the above factors will be discussed and then related to
the successional paradigm outlined earlier.

4.1. Bed Roughness

Organisms produce pits and depressions on the seafloor related to

burrowing and foraging activity. Benthos also produce elevations such as
mounds from burrow excavation, tracking, and feeding (Frey, 1975). The
length scales of these features range from meters (ray pits) to decimeters
(shrimp mud mounds) in diameter, and on down to sub millimeter scales,
in the case of fecal pellets deposited on the sediment surface.
Not all bioturbational activity results in an increase in bed roughness
 Chapter 1
20
on scales of centimeters. If most of the species are small and mobile, the
surface may be homogeneously reworked, resulting in a very "smooth"
and flat bottom. An example of this is described for the mud facies in
central Buzzards Bay (Rhoads and Young, 1970). However, if the re-
working species are large and relatively sedentary, each individual may
generate considerable topographic relief by its localized feeding activity.
Examples of this local topographic effect are described for central Cape
Cod Bay, Massachusetts, related to the presence of mounds made by the
holothurian Molpadia oolitica (Rhoads and Young, 1971), and for bur-
rowing shrimp mounds in Discovery Bay, Jamaica (Aller and Dodge,
1974).
Another commonly encountered and qualitatively important biolog-
ical roughness element consists of tubes that project a few millimeters
above the bottom. These tubes may be constructed of parchmentlike or-
ganic material or mucus-cemented sediment grains. Polychaetes and am-
phipod crustaceans commonly form dense aggregations of tubes in phys-
ically disturbed habitats. The spatial scales of the aggregations range from
several thousand square meters (e.g., Fager, 1964) to approximately 10 5
m -2 (e.g., Eckman, 1979). Tube populations may develop very quickly
following a disturbance through larval settlement, reaching densities of
10 5 m- 2 within a few days (McCall, 1977). These tube fields are usually
short-lived phenomena lasting 2 years or less, as the tube-forming species
are subject to mortality from intensive predation (Fager, 1964) or com-
petitive amensalistic interactions with burrowing species (e.g., Myers,
1977a,b). As we will see later, individual tubes themselves may enhance
bottom erosion, resulting in eventual washout of the population (Eckman
et al., 1981). On the other hand, some tube fields and their associated
benthic communities may be relatively permanent, with sediment-
stabilizing species coexisting with sediment-destabilizing species
(Wilson, 1979).
Inferences about the influence of tube fields on substratum stability
are largely based on the association of dense tube aggregations with in-
creased topographic elevation of the bottom, a decrease in mean grain
size, an increase in the quantity of organic-rich detritus settling between
tubes (Mills, 1967, 1969; Myers, 1977a; M. Lynch and Harrison, 1970;
Woodin, 1976; Featherstone and Risk, 1977; Bailey-Brock, 1979), or ab-
sence of traction bed forms within tube fields (Fager, 1964). Indirect in-
ferences about habitat stability have also been based on diversity patterns.
Tube patches have a higher abundance and diversity of non-tube-
constructing species living within them than are found on the adjacent
seafloor lacking tubes (Woodin, 1978; Wilson, 1979; Bailey-Brock, 1979;
Eckman, 1979).
An extensive fluids literature (see review by Wooding, 1973) exists
Effects of Marine Benthos on Sediment Physical Properties 21

on roughness elements; however, few flume experiments have been con-

ducted on the effects of animal tubes or other biologically produced rough-
ness elements on substratum stability. Rhoads et al. (1978b) did flume
work with the capitellid polychaete Heteromastus filiformis and showed
that, 3 days after a population was introduced into a mud, the critical
erosion velocity was increased by 80% relative to the critical velocity
required to entrain tube-free sediment. The mechanism of sediment sta-
bilization in this experiment is difficult to interpret, because the authors
also demonstrated, in a separate set of experiments, that bacterial mucus
films can increase entrainment velocities by up to 60%. The H. filiformis
study did not separate tube effects from microorganism effects. This study
also reported an apparent decrease in critical erosion velocity for muds
in central Long Island Sound associated with the appearance of a dense
set of the polychaete Owenia filiformis. Again, the correlation between
the presence of tubes and sediment stabilization does not allow one to
identify a simple cause-and-effect relationship because the effects of mi-
crobial binding were not measured.
While there is only one experimental study of the natural tube prob-
lem per S8 (Eckman et al., 1981), there are many experimental studies of
the effects of roughness spacing on flow resistance. The original study
was carried out by Schlichting (1936) (ef. Morris, 1955; Sayre and Al-
bertson, 1963; Lettau, 1969; Raupach, 1981). Of particular interest is a
study by Nowell and Church (1979), for they presented not only resistance
diagrams but also detailed velocity field and turbulent flux profiles. They
investigated the effect that different densities of boundary roughness ele-
ments have on the shape of the velocity, kinetic energy, and turbulent
energy dissipation profiles measured over the total boundary layer. The
roughness elements used in their experiments were Lego® construction
blocks. The density of the blocks was expressed as a ratio of the plan area
of N roughness elements (blocks) to the total bed area of the flume channel.
At low densities of blocks (less than 1 unit area of blocks per 22 unit
areas of flume bottom), each block acts as an isolated element. Turbulent
vortices are shed from individual roughness elements and the turbulent
energy is transferred to the bed. The kinetic energy and energy dissipation
functions show the characteristic monotonic increases all the way to the
wall, which is hydraulically rough. At intermediate densities (1/16-1/22),
the profile of kinetic energy is approximately constant over a distance of
1-2 roughness heights. At the highest densities (1/8-1112), maximum
turbulent intensity and the rate of turbulent energy dissipation were el-
evated to near the tops of the roughness elements, which "protected" the
bed within the tube field from higher-energy turbulence ("skimming
flow" sensu Morris, 1955).
The Nowell and Church (1979) study has important implications for
22 Chapter 1

STABILIZ ING Figure 7. The relationship between tube diameter and

tube spacing in affecting noncohesive bed stability.

"}
Tube values in the stabilizing field are proposed to
initiate "skimming flow," which protects the bed from
turbulence. Tube values in the destabilizing field cause
bed scour by reattachment of wake turbulence . Mod-

1
ified from Eckman et 01. (1981). (A) McCall and Fisher
03 - A 6 +1 (1980) (oligochaetes); (B) Fager (1964) (Owenia fusi-
formis); (C) Eckman et a1. (1981) (0. fusiform is); (0)
DESTABILIZING McCall (1977) (Streblospio benedicti) ; (E) Bailey-Brock
(1979) (Chaetopterid polychaetes); (F) Myers (1977a,b)
(Corophium insidiosum); (G) Mills (1969) (Ampelisca
~O~~2~~4~~6~~B~-J.O abdita); (H) Fager (1964) (Oweniafusiformis); (I) Myers
TUBE DIAMETER (mm) (1977a,b) (Microdentopus gryllotolpa).

predicting how the substratum might respond to the spacing and size of
biological features projecting above the bed. Using the 1/12 ratio as a
threshold roughness spacing for initiating skimming flow, Eckman et a1.
(1981) have shown that natural densities of tubicolous species span the
inferred range of stabilizing densities (Fig. 7). Of particular interest is the
prediction that densities of Owenia, reported by Fager (1964) to stabilize
a sand bottom from wave surge, should destabilize the bed (i.e., give values
for critical shear velocity lower than those predicted by the Shields
curve*). Average densities reported by Fager (1964) actually destabilized
sediment in the Eckman et a1. (1981) flume study. The cause for disparity
between Fager's field observations and the Owenia flume experiment is
unknown, but Eckman et al. (1981) suggest that the difference between
the field observations and their flume experiments is related to the pres-
ence of sediment-binding exudates from surface-living bacteria, diatoms,
or a commensal anemone that is attached to the Owenia tube.
Changes in substratum elevation, sediment quality, and grain size,
which are frequently cited as evidence for tube stabilization of the seaf-
loor, must be interpreted with care. Even though the tube density falls
within the destabilizing region of Fig. 7, the bed may be stabilized by
exudates of chemautotrophic microorganisms or diatoms. The presence
* The Shields curve takes into account important fluid and sedimentary parameters to
provide the ratio of dimensionless shear to a particle Reynolds number (i.e., ratio of
entrainment to stabilizing forces) . This curve allows one to compare data on measured
grain sizes and fluid shear stresses regardless of flume design. Moreover, laboratory results
can be compared with field measurements or estimates (see Madsen and Grant, 1976;
Miller et aJ., 1977).
Effects of Marine Benthos on Sediment Physical Properties 23

of tubes may be directly coupled with these exudates. Microorganism

productivity is enhanced through the increased supply of dissolved nu-
trients pumped across the sediment-water interface through tubes (Webb,
1969, Aller and Yingst, 1978; Aller, 1980).
Tube fields that fall well within the stabilizing region of Fig. 7 may
still be susceptible to scour and erosion, especially at the leading edge
and sides of a patch where flow accelerations occur. In a rotational tidal
current, all sides of a patch may be subject to scour. Most flume and field
studies have not addressed the problem of form drag associated with tube
patches. The question of whether "skimming flow" is actually induced
by tube-dwelling organisms is still unanswered, as flume work has not
yet been done with tube densities predicted to be stabilizing (Fig. 7).
Another biogenic factor that can armor or protect sediments from
erosion is the presence of dense concentrations of shell material at the
sediment-water interface. The quantity of shell at the sediment surface
is related to both the sedimentation rate and the shell production rate.

200

100

60
U
UJ
<10
(/)
..... 20
2:
~
10
} 6
Figure 8. Competency diagrams show- 4
ing the effects of biological activity on
2
critical threshold velocities. (A) The
maximum observed change in mean
rolling velocity of five glass bead beds
following exposure to microbial growth
(replotted from Ullman, 1975). Flow
data coverted to z = 100 cm and plot- GRAIN DIAMETER IN MICRONS
ted on a competency diagram. Trian-
gular symbols are initial (aseptic) bead
values; circles are values after micro-
bial binding. Microbial binding dis-
places threshold values above the
quartz critical erosion velocity bound-
ary. From Rhoads et al. (1978b). (B)
Effects of fecal pellets, tracks, and ex-
udate binding on natural sediments. 8 r.r AMBIENT SED.
Data on Calves Pasture Pt., Redfield > 6 • CALVES PIlSTURE I'T.
(iI PELLETS
.. REDFIELD 15L ANl
T AMBIENT SED.
Island, and Jules Island, Barnstable • ,JULES IS~D
(!') TRACKED SED.
Harbor, Massachusetts, from Boyer 2
(1980). Other data from Nowell et a1.
(1981). 10 <10 100 200 400 1000 4000
24 Chapter 1

5
>-
u
z
~ ~~~ ________r.~r._
o
W
IX:
"-

DESTABILIZATION STABILIZATION

Figure 9. Change in critical threshold velocity for initiating bed movement in fine sandy
intertidal sediments reworked by meiofauna. Horizontal scale is change in flume velocity
(cm ' sec-'); vertical axis is measurement frequency. Of the 41 flume runs, 78% (N = 32)
showed that particle bioturbation lowered the critical velocity. The change in critical velocity
is relative to a recently eroded bed surface. From Boyer (1980).

Pioneering assemblages are highly productive systems, and opportunistic

bivalve molluscs (e.g., species of the families Mactridae, Tellinidae, and
Pandoridae) may appear in abundances of 10 3 -10 4 m - 2 (McCall, 1977;
Rhoads et aI., 1978a). These clam populations usually persist for less than
1 year. As the death assemblage accumulates, the disarticulated valves
may be packed so densely that they armor or pave the interface (Fig. 2C).
Sediments may be destabilized by bed roughness produced by sur-
ficial tracks and trails (Eckman et aI., 1979). Nowell et al. (1981) have
conducted flume experiments to determine the threshold velocity for very
fine sand tracked by a small (approximately 4 mm) clam, Transenella
tantilla. The critical shear velocity was reduced by 20% from that of the
untracked sediment (Fig. 8B). Boyer (1980) has also observed decreases
in critical flume entrainment velocities related to the presence of tracks
and trails of small macrofauna as well as meiofauna that fluff-up the
interface (Fig. 9). Although macrofauna I and meiofaunal bioturbation can
erase or otherwise reduce the topographic relief of tracks and trails (Cul-
len, 1973; MacIlvaine and Ross, 1979; Boyer, 1980), meiofaunal reworking
produces a smaller-scale boundary roughness on the order of 1 mm or
less (Grant et al., 1982; see comment in Nowell et aI., 1981). In our flume
experiments, we have found that ostracods, copepods, and foraminifera
moving horizontally just below the sediment surface are effective in pro-
ducing a characteristic "hummocky" relief on the surface. Grant et al.
(1982) show that meiofaunal reworking of the bed may lower the critical
shear velocity to near the predicted Shields values.

4.2. Pelletal Textures

Much of the potential food for detritus-feeding invertebrates is as-

sociated with sediments with high proportions of particles within the
silt-clay size range (Sanders, 1958; Sanders et aI., 1962; Purdy, 1964;
Effects of Marine Benthos on Sediment Physical Properties 25

Whitlatch, 1976). The high surface-area-to-volume ratios of small particles

provide a large expanse for the attachment and growth of microbial pop-
ulations that produce mucopolysaccharide exudates (Zobell, 1943; D. L.
Lynch and Cotnoir, 1956; Hobbie and Lee, 1980). The size fraction of
particles selected in the feeding process varies among species. Many de-
posit feeders preferentially select particles larger than 10 /-Lm (Whitlatch,
1976) and below 1 mm (Rhoads and Stanley, 1965). In addition to or-
ganic-mineral aggregates, relatively "clean" mineral particles may be
ingested and stripped of adsorbed or absorbed organic films (Sulanowski,
1978; Pinck et a1., 1954). Particle selection may also be based on grain
density and surface texture (Self and Jumars, 1978; Taghon et a1., 1978).
In addition to the size specificity of food particles, much of the po-
tential food for detritus-feeding invertebrates is located within the upper
few centimeters of the seafloor. In the intertidal zone, usable organic
material may be present to a depth of 20 cm (Johnson, 1977; Sulanowski,
1978) and within the upper 2 cm of subtidal coastal sediments (Johnson,
1974, 1977). The above generalizations are based on detailed analyses of
vertical profiles of particle morphology, organic content, and total protein
and carbohydrate in marine muds and sands (Johnson, 1974, 1977; Whit-
latch, 1977, 1980). Detritus feeding is therefore concentrated within, but
not limited to, these depths.
With few exceptions, marine benthic invertebrates produce fecal pel-
lets* and these pellets or castings are deposited at, or near, the sediment
surface. This generalization also holds true for head-down conveyor belt
feeders, which may ingest particles at depths of 20 cm below the surface.
Ingested material is usually transported upward through the digestive
tract and egested at the sediment surface. In equilibrium communities,
these pellets may be subsequently advected downward by bioturbation
and distributed throughout the bioturbation zone (Rhoads, 1967). Pi-
oneering species are less efficient in vertical mixing; therefore, pellets
remain near the sediment surface. Pelletization may result in a change in
the grain size of surface sediments. In muds, this process commonly
involves the addition of a very fine pelletal sand fraction to an unpelle-
tized silt-clay matrix (Rhoads and Young, 1970). Because many detriti-
vores preferentially select for particles within the silt-clay fraction, sand-
sized fecal pellets are avoided until they break down and the constituent
particles are repopulated by algae and/or microbes (Frankenberg and
Smith, 1967; Fenchel, 1970; Hargraves, 1976; Levinton and Lopez, 1977;
Levinton, 1980). Most fecal pellets are produced by the resident benthos,

* Pseudofeces (particles manipulated but not ingested by an organism) may also be ejected
or deposited at the sediment surface. Pseudofeces are usually less consolidated than fecal
pellets and castings.
26 Chapter 1

but H. B. Moore (1931) describes pelletal muds from the Clyde Sea that
are composed predominantly of zooplankton feces.
Jumars et a1. (1981) have modeled the fates of sediments that contain
deposit feeders. Their Markov model involves the probabilities of particle
selection, pellet disintegration, and burial, and the probability of transport
of pellets versus "free" unpelletized sediment. The authors show how the
steady-state abundance of a given particle type comprising a pellet is a
function of pellet durability and particle (feeding) selectivity. High den-
sities of a given particle type in surface sediments result from packaging
of particles into durable fecal pellets.
The distribution of pellets in the sediment column depends on their
chance of burial and preservation. Burial may be brought about by a net
input of "new" sediment or by downward advection by particle bioturbation.
Particle residence times are also affected by the feeding types present
(Jumars et a1., 1981): surface deposit feeders only, subsurface deposit
feeders only, or both of these feeding types occurring together. Particle
residence times are also sensitive to the relative erodibilities of pellets
compared to unpelletized sediment.
The production of very fine sand-size pellets from silt-clay-size par-
ticles would, a priori, appear to favor selective transport of pellets relative
to unpelletized sediment. The threshold shear velocity for entrainment
of fine sand is lower than that for cohesive silt and clay. Flume obser-
vations in natural sediment containing organisms, however, show that
pellets do not always have lower critical shear velocities than silt or clay,
and may have higher critical shear velocities (Fig. 8) (Nowell et aI., 1981).
In nature, it is probably a rare phenomenon to have single, discrete par-
ticles of fine sand or smaller grain sizes move independently of other
particles. Most particles below fine sand size are associated with or-
ganic-mineral aggregates (Rhoads, 1973; Johnson, 1974) held together
with microbial mucus (Rhoads, 1973).
Existing competency curves are of limited use in predicting the sta-
bility of natural fine-grained sediments because their use is restricted to
the particular apparatus or experimental location chosen. Moreover, hy-
draulic engineers avoid biological growth in their flume channels by add-
ing growth inhibitors to the water. It is not surprising then, to expect
major departures in threshold velocities of natural sediments from those
predicted by "azoic" competency diagrams (Boyer, 1980; Nowell et 01.,
1981). However, biologically reworked intertidal sediments fall well
within the values predicted from the Shields curve (Shields, 1936; Madsen
and Grant, 1976; Grant et a1., 1982).
Nowell et a1. (1981) have conducted flume studies on the threshold
velocities of polychaete and bivalve pellets and castings in natural and
artificial sediment. Individual pellets, as well as mounds of fecal castings,
Effects of Marine Benthos on Sediment Physical Properties 27

were held together and to the bed by "elastic" mucus. These pellets and
castings were abraded in situ by greater than critical shear velocities.
Mucus cemented individual pellets and fecal mounds to the bottom (Fig.
BB). Boyer's (19BO) fl ume observations of intertidal sediments su pport these
observations. The most transportable feces are those produced by species
that eject their pellets several centimeters into the water above the bed.
Pseudofeces may also be forcefully pumped into the water column
(Rhoads, 1963).
The transport of pellets appears to be largely controlled by the pres-
ence of binding microbial, meiofaunal, or diatom mucus (excluding algae).
Below the plant "photosynthetic compensation depth," microbial and
meiofaunal exudates are probably the major binding agents. Microbial
binding of medium, sand-sized, to coarse, silt-sized, spherical glass beads
has been shown to increase critical entrainment velocities by as much as
60%* in the Rhoads flume (Fig. BA). The onset of a mucus binding effect,
starting from initially sterile glass bead surfaces, was observed to be as
short as 3 days in laboratory flume experiments (Rhoads et al., 197Bb).
In nature, bacterial colonization of surfaces is enhanced by the ad-
sorption of glycoproteinaceous films onto surfaces (Baier, 1973; Neihof
and Loeb, 1973). Organic films can be adsorbed onto sterile and clean
surfaces immersed in seawater within a period of 10 min (Goupil et al.,
1973). Clay minerals bear net negative surface charges, promoting the
rapid adsorption of mucopolysaccharide films (Khailov and Finenko,
1970; Neihof and Loeb, 1972). Microbial recolonization of newly pro-
duced pellets may take place within a few hours in the summer (J. Y.
Yingst, personal communication). In the absence of intensive particle
bioturbation, iron hydroxide encrustations may also serve to bind surface
sediments (Parthenaides and Paswell, 1970). Subtidal carbonates, not sub-
jected to bioturbation, may lithify in situ by precipitation of cements. This
condition has been proposed to explain the abundant flat pebble (car-
bonate) conglomerates observed in Cambrian strata (Sepkoski, 19B2).
In our flume work with Long Island Sound muds, we have rarely
seen individual pellets roll along the bed. As velocity is increased, or-
ganic-mineral aggregates t (including pellets) that project above the bed
start to vibrate and rock in place.

* Alcian blue staining of the glass beads showed the presence of carbohydrate-rich mucus
at very small point contacts between grains. These contact films were barely visible under
50 x magnification.
t Many of these aggregates may have their origins from the water column by the formation
of particulates from dissolved organic matter, adsorbed onto bubble surfaces (micelle
formation), or onto other surfaces in the water column (Riley, 1963; Krank, 1973, 1975).
When the aggregates reach the seafloor they are mixed into the bottom and become part
of the detrital food chain.
28 Chapter 1

We have called this the critical excitation velocity (Rhoads et aI.,

1978b). This same phenomenon has been described by Nowell et al. (1981)
for fecal mounds of polychaetes and by Boyer (1980) for polychaete and
Hydrobia sp. (gastropod) pellets. The excitation velocity can be well above
the predicted critical flume entrainment velocity for detached aggregates.
The elastic mucus that binds most particles to one another and to the bed
only fails when the flow rate is increased well above the excitation ve-
locity. * As pieces of organic-mineral aggregates break away from the bed,
they move downstream as traction or suspended load (depending on the
density of the detached aggregate and the flow velocity relative to aggre-
gate diameter) . These aggregates can be quite large (1-2 mm) and include
pellets and pellet aggregates, as well as irregularly shaped detrital par-
ticles. With sustained flow, strings of mucus and mucus-bound particles
are observed on the surface of the bed; one end of the strand is attached
to the bed while the free end vibrates and "flails" at a high frequency in
the turbulent vortices. These filaments represent coiled sheets of mucus
and are probably the "fibrous binding material" observed in broken sed-
iment surfaces in flume experiments described by MacIlvaine and Ross
(1979), Nowell et a1. (1981), and Boyer (1980).
It is important for future experimental flume work to measure quan-
titatively the binding mucus at the sediment-water interface. This is a
formidable problem. Mucopolysaccharides are a complex and poorly de-
fined group of macromolecules. Quantitative separation from sediments
is difficult, as mucopolysaccharides are closely associated with proteins
and inorganic salts (Hobbie and Lee, 1980).
Another problem involves characterizing seasonal changes in mucus
production and degradation. Yingst and Rhoads (1980) summarized ear-
lier work on soil microbes: Soil microorganisms are known to be tem-
perature-controlled in their production of polysaccharides (Martins and
Craggs, 1946; Harris et al., 1966). Microorganism assemblages in soils
produce a range of chemically different binding agents (Aspiras et aI.,
1971), which may differ in their biodegradability as well as in the physical
properties that they confer on soil aggregates. The effect of environmental
variables such as temperature on the binding of sediment particles by
microorganisms may involve the production of agents of different bio-
degradability (Martins and Craggs, 1946). Aspiras et al. (1971) found that
maximum structural stability of microbially bound solid aggregates was
reached sooner at high temperatures, but was maintained longer at lower
temperatures, once maximum stability was reached.
* Using microbially bound glass beads. Ullman (1975) found that the mucus bonds failed
after the flume velocity was increased approximately 77% above the excitation velocity.
Boyer (1980) also found that mucus-coated. intertidal sands had a "tear-away" velocity
approximately 70% higher than the excitation velocity using the same flume.
Effects of Marine Benthos on Sediment Physical Properties 29

Mucopolysaccharides also appear to be chiefly produced in mi-

croaerophilic environments (Mitchell and Nevo, 1964). Therefore, the
production of mucus may vary with redox potential (Eh) and the position
of the RPD in sediments. Even if one could make quantitative extractions
of mucopolysaccharides from sediments, we question the value of such
a parameter for predicting the bed response to fluid shear stress. The
critical shear velocity may not be simply a linear function of the absolute
quantity of mucus present. Once grain-to-grain contacts are bridged by
an elastic film, additional mucus (increasing film thickness) may have
only a relatively minor effect on binding strength. To develop a complete
understanding of fine-grained sediment transport we must begin to un-
derstand the binding of sediments by mucus (Richards and Parks, 1976).
However, we foresee serious analytical and methodological problems in
characterizing biological adhesion.
Some pellets may move as discrete particles. Risk and Moffat (1977)
describe the transport of both pellets and pseudofeces of the intertidal
tellinid bivalve Macoma baltica from muddy sediments of the Minas
Basin, Nova Scotia. Feces are picked up on the flood tide and transported
shoreward. These findings are especially interesting, as Boyer (1980) com-
monly found that intertidal pellets composed of fine sand and exposed
during low tide are tightly bound to the bottom by microbial and algal
mucus.
When pellets or pellet aggregates are entrained, they may move as
bed load. This can result from high settling velocities relative to their
critical shear velocities (Nowell et al., 1981). If the whole bed is suspended
into the water column by wave surge, the presence of fecal pellets can
increase the settling velocity of pellet-bound mud by two orders of mag-
nitude (Haven and Morales-Alamo, 1966; McCall, 1979).

4.3. Geotechnical Mass Properties

The mass or bulk geotechnical properties of sediments, such as den-

sity, water content, Atterberg limits (Lambe and Whitman, 1969), and
macroscopic shear strength, are all influenced to a substantial degree by
benthic organisms (Chapman, 1949; McMaster, 1962, 1967; Webb, 1969;
Rhoads, 1970; Silva and Hollister, 1973; Richards and Parks, 1976; Bok-
uniewicz et al., 1975; Myers, 1977a,b; Boyer, 1980).
The ease with which free-living infaunal burrowers move through,
and feed upon, sediments is known to be a function of sediment water
content and state of compaction (Chapman, 1949). The limiting force with
which burrowing infaunal metazoans can displace sediment is deter-
mined either by the internal coelomic pressures that can be generated by
30 Chapter 1

muscular contraction of their flexible and extensible body wall (Chapman

and Newell, 1947; Clark, 1964) or by hydrostatic pressures exerted by
blood or water vascular systems (Nicol, 1967). To facilitate burrowing and
feeding, some metazoa, especially bivalves, also liquify the sediment by
injecting water anteriorly into the bottom. This causes an instantaneous
local increase in pore water pressure, and the liquid limit of the sediment
is temporarily exceeded. At this instant, the organism moves forward into
the liquified zone. In sands, this fluidized sediment represents a transient
state, as overburden pressure soon causes the sediment to collapse on
itself. In cohesive silts and clays, the sediment may remain dilated long
after a burrowing organism has passed through or otherwise processed
the sediment. This activity is evident from the presence of abandoned
burrows and feeding voids. The mass properties of muds, therefore, reflect
the cumulative burrowing history to a greater degree than do rapidly
consolidating sands. If burrowing organisms are relatively active, a few
individuals per unit area of the bottom can have a major impact on re-
molding and dilating muds. * The effect of burrowers on a specific sedi-
ment is related to the rate of burrowing versus the rate of consolidation.
In the sense that muds compact less rapidly than cohesionless sands, the
relative effect of burrowing organisms on the bulk density of muds would
be expected to be greater than on sands.
Burrowing in muds has the effect of increasing sediment water con-
tent (Harrison et al., 1964; Harrison and Wass, 1965; Rhoads, 1970, 1973;
Rhoads and Young, 1970). Intensive particle bioturbation, characteristic
of equilibrium communities, is associated with fine-grained sediment
water contents that are greater than 60%, and commonly over 70% (Table
II). Burrowing also decreases near-surface sediment compaction (Boku-
niewicz et al., 1975) and decreases undrained shear strength, as shown
in the geotechnical experiments described earlier (Section 3.2). Because
many pioneering species are not efficient in mixing particles vertically,
these bottom types may compact and have water contents of less than
60%. However, generalizations are difficult to make because the water
content will also depend on the frequency and intensity of physical re-
suspension by currents. Such a "fluid" pioneering bottom has been de-
scribed for the Ria Arosa, Spain, by Tenore et al. (1982). These mud
bottoms are apparently resuspended quite frequently (more than once a
month) and have a water content of over 60% in the upper 12 cm.
Rowe (1974) made in situ sediment shear strength measurements in
a highly burrowed mud in Buzzards Bay, Massachusetts, and found av-
erage shear strengths to be 0.98 kPa (10 g' cm -2). Located on this burrowed

* Burrowing may operate together with pelletization to cause sediment dilation. The pres-
ence of pellets can decrease packing through the production of interpellet void space.
Effects of Marine Benthos on Sediment Physical Properties 31

Table II. Near-Surface Water Contents of Muds Bioturbated by Errant Infaunal

Deposit Feeders
Water content Depth interval
Location (percent)" (cm) Reference

Buzzards Bay, Massachusetts, >60 0-1 Rhoads and Young (1970)

USA
Cape Cod Bay. Massachusetts, >60 0-3 Rhoads and Young (1971)
USA
Long Island Sound, >70 0-1 Rhoads et a1. (1977)
Connecticut, USA
Ria Muros, Galicia, Spain -70 0-2 Tenore et a1. (1982)
Clyde Sea, Scotland >80 0-6 H. B. Moore (1931)

Q Weight of water expressed as a percentage of the initial weight of solids plus water.

bottom are sedentary anemones (Ceriantheopsis americanus). Shear

strengths of the sediment within 20 cm of the anemones increase to 1.83
kPa (18.7 g' cm -2). The mechanisms by which C. americanus and other
macrofauna increase sediment shear strengths, or otherwise promote or
permit compaction, are poorly understood. In some cases, the cause and
effect may be obvious, i.e., walled tube structures within the bottom ob-
viously increase the breakaway torque of a shear vane apparatus or impede
the vertical penetration of the bottom by a penetrometer or fall cone.
Moreover, large tube dwellers may laterally compact sediments several
millimeters away from the immediate tube (e.g., Aller and Yingst, 1978).
In the case of C. americanus, the mucus shed from the oral crown may
have spread over the sediment surface and caused particle-to-particle
adhesion.
Other stabilization mechanisms may be less obvious, such as the
interactions between macrofauna, meiofauna, and microorganisms. For
example, pore water irrigation is associated with enhanced production
of microbial populations and their mucopolysaccharide exudates (Webb,
1969). These viscous and elastic-binding mucus secretions, generated by
bacteria as well as by macro- and meiofauna, may fill intergranular pore
spaces (Frankel and Meade, 1973). Because many geotechnical properties
are related to water content, the inferred range of influence of benthic
organisms on the rheological properties of sediments is shown in Fig. 10.
In order to evaluate how biological changes in mass properties might
influence sediment transport, we have examined much of the literature
on the erosion of cohesive sediments. This literature does not present a
consensus on the relationships between measured geotechnical properties
and critical threshold velocities (e.g., Partheniades, 1965; Einstein and
Krone, 1962; Gularte et al., 1980). Richards and Parks (1976) state that
32 Chapter 1

WATER Figure 10. Mass geotechnical proper-

ties of sediments related to grain size
and water content. In clayey or silt-
rich sediments, the properties of the
sediment can be significantly affected
by the successional stage populating
the sediment. The inferred range of
effects is shown by the large doubl e
arrow. Pioneering species may fa ci li-
tate the compaction of muds (plas-
PLASTIC- DILATANT
ti c-dilatant field), but this depends
on the frequency and intensity of bot-
tom resusp ension by currents. High-
order successional species tend to en-
hance the water content of the bottom
CLODDY
through intensive particle bioturba-
tion (thixotropic-fluid fi eld). Dia-
SILT 8 SAND 50 CLAY gram modified from Boswell (1961).

"at present there is not a satisfactory way to predict the critical (fluid)
shear stress of a cohesive bed surface or erosion rates as a function of
excess strain." In part, the problem is procedural. Intercomparison of data
is difficult owing to the diversity of methods used to measure geotechnical
properties, the range of properties measured, the variety of open and
closed channel flumes used, the different criteria applied to identify ini-
tiation of motion and mass erosion, and the different manipulation his-
tories and mineralogical compositions of test sediments.
Parthenia des (1965), in his review of the literature on cohesive sed-
iment erosion, concludes that shear strength and Atterberg limits are poor
predictors of cohesive sediment erodibility. Because incipient bed erosion
is related to how fluid shear stress is transferred to the sediment-water
interface, one might expect that subsurface geotechnical properties might
prove to be uncoupled from processes at the immediate sediment-water
interface. *
It is clear that most geotechnical measurements are made on vertical
and horizontal scales that are too coarse to address adequately the problem
of a natural sediment-water interface. As we have seen in Figs. 2 and 3,
and described in earlier sections, the biological and physical structure of
the sediment surface must be measured and described on the scale of a
millimeter or less.
Partheniades (1965) emphasizes the importance of physiochemical
properties of the interface, particularly the configuration and strength of

* Subsurface mass properties might be more appropri ately described for determining the
extent of massive bed erosion.
Effects of Marine Benthos on Sediment Physical Properties 33

90
®
_ 1\0 1
1 I I

t~
•~ 80 -
I-
~
I- 70
z
- -
UJ
I-
Z
8 60 l-
• <> -
0::
UJ •
0.<>
ti
~ 50 I-
<><>~ -

I I I
"- I I
40
0102030405060
CRITICAL VELOCITY (CM/SEC)

@
Figure 11. The relationship among sediment water 5Or----,,--- , ----r----,.-,
content (weight of water expressed as a percentage o o
initial weight of solids and water), salinity, and
IO ·IM /
critical entrainment velocity. (A) Data from fine- SAUNITY
grained shallow and deep-water muds. Velocity
given for z = 100 cm. 0, From Postma (1967) ; 0,
from Migniot (1968); ., from Lonsdale and Sou- o ·IM

.
thard (1974). Redrawn from Southard (1974). (B) SALIMTY

Fine-grained sediment (grundite) erosion as a func- ...J

tion of sediment water content and ambient water -~ 20 ~. ---- .~

salinity. Velocity given for closed flume channel. !::: I~ ""'--

Much of the flocculation between clay particles 5 O~=--~~~-~~~-~41r--~~'~

" .
takes place at salinities ::;10%0. From Gularte (1978).
WATER CONTENT (WT %)

ow37
4Orr"rrTT'-rr"rrTT"rrTO'-rr~rrrT~~

".
~&i
...
~
o
;:30 • ... ... •
....
~
1325
9
~20 #
W
~ 15
.... •
...J
u. 10 • REOfIELD I~D
Figure 12. The critical velocity ...J
<I
... JULES ISLAND
for bed motion in sandy interti- ~ 5
dal sediments containing differ-
ent concentrations of particulate
5 o~~~wu~~~~~~~~~~~
J 2 3 A ~ & 1 ~ 9 ~ U ~ ~ ~ ~
organic carbon at the sediment % PARTICULATE ORGANIC CARBON AT THE
surface. From Boyer (1980) . IMMEDIATE INTERFACE (-2mm)
 34 Chapter 1

12 -=::-- - -- , - - -- . -- _ _ __ --,--,
U
10
.9
U .8
\jj .7
"-
:::;; .6 Figure 13. Change in critical ve-
8 ~
~:::> ~
locity (UzrilJ of muddy sedi-
.3 ments from Buzzards Bay, Mas-
2 sachusetts, subjected to bioturba-
J tion over a period of 61 days.
°o~-----~~~-~~-----~~~~ Data plotted from Young and
DAYS AFTER REWORKNG Southard (1978).

aggregates located at the sediment surface. Although his experiments were

done with azoic* clay-rich sediments, his perspective is very relevant to
the conditions present on a natural interface composed of mucus-bound
organic-mineral aggregates and fecal pellets. Gularte et a1. (1980) also
suggest that physiochemical interparticle forces at the sediment surface
are extremely important for determining the erosional response of azoic
sediments composed of clay minerals.
Some workers have found reasonably good inverse correlations be-
tween erosion resistance and sediment water content (Postma, 1967; Sou-
thard et a1., 1971; Lonsdale and Southard, 1974). Void space, salinity,
and sediment shear strength have also been found to be important (Einsele
et a1., 1974; Gularte et a1., 1980) (Fig. 11). Critical flume erosion velocity
has been found to be correlated with the organic carbon content of in-
tertidal sediments (Boyer, 1980) (Fig. 12). R. A. Young and Southard
(1978) found that the threshold shear velocity for particle movement in-
creased as the Leco carbon content of experimental sediments was in-
creased over the range of 0.8-2.0 weight percent. [Organic matter content
is correlated with water content (r = 0.62) in muddy Long Island Sound
sediments (Yingst and Rhoads, 1978).]
We previously noted that intensive burrowing and particle biotur-
bation can produce surface water contents in muds of greater than 60%
(Table II), and that bioturbation decreases sediment shear strengths. Par-
ticle bioturbation would than appear to lower critical shear velocities of
muds. R. A. Young and Southard (1978) have experimentally shown that
Utril. (critical bottom shear velocity) can be substantially decreased in
sediments reworked by errant deposit feeders (Fig. 13). Grant et a1. (1982)

* Insufficient data are given in most engineering studies to determine whether or not bacterial
or plant activity was excluded from the sediments under examination. By azoic we mean
that no micro- or macroorganisms were deliberately introduced into, or maintained in,
the sediment during the experiment.
Effects of Marine Benthos on Sediment Physical Properties 35

indicate that small «1 mm) macrofaunal and meiofaunal invertebrates

reworking intertidal fine sands may lower the critical shear velocity of
previously stabilized sediments.
The flume measurements of Rhoads et a1. (1978b) suggest a seasonal
change in subtidal mud erodibility in central Long Island Sound over a
2~-year period (Fig. 14). The critical rolling velocity for organic-mineral
aggregates ranged from 16 cm' sec- 1 to 28 cm' sec- 1 (velocities calculated
for a depth off the bottom of 1 m). The seafloor was most easily eroded
in spring and early summer, and most resistant to erosion in winter. These

'"
ci
U 16

t::
Q:

U 4 J A S - O NO J F M A MJ J A SON 0 J F M A M J J A SON 0
1974 1975 1976
DATE

25rB----------________________________-,

J ASON DIJ F M A M J JSON

A DIJ F MAM J J ASON D
1974 1975 1976
DATE
Figure 14. Seasonal changes in seafloor erodibility in central Long Island Sound over a 2!-
year period . Measurements made from diver-obtained box cores that were inserted into a
laboratory flume. (A) Seasonal change in mean critical rolling velocity and mean suspension
velocity at station N WC, central Long Island Sound . Vertical bars represent one standard
deviation. Curves fitted to the average of two replicate samples except for April and October,
1975, and October, 1976, when data points were determined from single box cores. Data
points are significantly different at the 0.95 level if separated by ~0.8 velocity units on the
y axis. (B) Seasonal change in bottom sediment temperature at station NWC. Circled data
points are times when bottom temperatures were taken without sampling with flume cores.
From Rhoads et 01 . (1978b) .
36 Chapter 1

flume observations are supported by seasonal field measurements of near-

bottom turbidity in Long Island Sound (Rhoads and Boyer, 1983).
R. A. Young and Southard (1978) made in situ Seaflume (a flume
deployed on the seafloor; R. A. Young, 1977) measurements in Buzzards
Bay, Massachusetts, over the period August through February, 1977. The
station selected for study appears to be similar to the Long Island Sound
bottom studied by Rhoads et 01. (1978b). Young and Southard found
utrit. to vary by a factor of two at measurement sites located only a few
meters apart. They attribute this variance to spatial patchiness in un-
measured physical or biological parameters rather than bulk geotechnical
properties. No systematic change in U(~rit. was detected over the time span
of the study. The area of Buzzards Bay where Young and Southard made
their measurements was severely affected in 1969 by an oil spill, which
killed off many bioturbating species. As of this writing, the equilibrium
fauna has still not completely recovered. This fact, plus the relatively
short period of measurement, may explain the spatial patchiness in the
results of the Buzzards Bay study.

5. A Qualitative Predictive Model

There have been two earlier attempts to place organism-sediment
relationships into a time-dependent successional framework (Webb et 01.,
1976; Rhoads et 01.,1977). We have updated these earlier models in Table
III.
The underlying assumptions in Table III are that we are relating
sedimentary properties to end-member series of succession on a subtidal
muddy seafloor. A late pioneering stage may be associated with physical
properties of the seafloor that are between the extreme end-member cases
outlined in the table. * These transition states are poorly studied. A second
assumption is that the described biological processes are operating at high
rates. The effects of seasonality are not considered. In environments with
a large temperature range, rates of biological activities may change by a
factor of two or more with each lOoC change in temperature. This means
that the rate of particle bioturbation by equilibrium species will be dra-
matically reduced in winter. The physical properties of sediments in late
winter may therefore approach "azoic" values. Fluid bioturbation or ir-

* The time involved in a succession such as the one outlined in Table III and in Fig. 1
depends on many factors, and is poorly known. Our best estimate is that a primary succes-
sion, starting at a macrofaunal free state, will take tens of years to develop into a persistent
equilibrium community. This estimate assumes that the system remains undisturbed in
the course of its recovery.
Effects of Marine Benthos on Sediment Physical Properties 37

rigation will similarly be decreased in winter. The temperature effect on

benthic metabolism is important for explaining seasonal changes in pore
water chemistry, and rates at which nutrients are pumped into, and out
of, the bottom (Rhoads et a1., 1977; Aller, 1978, 1980; Nixon et a1., 1980;
Propp et a1., 1980; Zeitschel, 1980). If nutrient fluxes are coupled to
benthic microbial and mucopolysaccharide production, one might expect
a strong seasonal control on sediment cohesion. These interacting factors
may be the main reasons for the observed seasonal changes in critical
flume entrainment velocities described for Long Island Sound sediments
(Rhoads et a1., 1978b).
Seasonal changes in pelagic primary production and sedimentation
of these products to the seafloor affect metabolic rates of benthos (Har-
grave, 1980). This pulse of organic matter could temporarily increase
utril. (R. A. Young and Southard, 1978). However, utrit. may decrease if
the organic matter is in the form of large aggregates (Riley, 1963), which
are not firmly bound to the bottom. Seasonal changes in benthic primary
productivity may be important in determining temporal patterns in the
critical shear velocity of intertidal sediments (Grant et a1., 1982).
Recruitment patterns of the benthos can also affect rate-dependent
processes, as seen in Table III. Recruitment is often correlated with
changes in water temperature and primary production cycles. During
recruitment, the density of organisms on the bottom increases by orders
of magnitude over long-term, mean abundances. This is especially true
of pioneering stages (Rhoads et a1., 1978a). The recruited organisms are
all of meiofaunal size «0.5 mm) and occur in a thin zone near the sed-
iment-water interface (Mare, 1942; Yingst, 1978). During the recruitment
period there is intensive meiofaunal processing of the sediment surface
(Cullen, 1973). Competition between permanent and temporary mei-
ofauna results in high mortality, and abundances of total meiofauna may
drop dramatically over a short period of time (Bell and Coull, 1980). In
this regard, the muddy seafloor in Long Island Sound is most easily eroded
in the spring, a period of intensive benthic recruitment (Rhoads et a1.,
1978b).
We now return to the original intention of this paper: to propose
hypotheses about the transport fates of fine-grained sediments and their
contaminants. The organism-sediment-fluid interactions outlined in
Table III predict that particulate and dissolved pollutants will experience
different transport fates depending on the ecological conditions of the
seafloor with which they came into contact. For example, a suspended
particulate contaminant passing over a pioneering assemblage has a rel-
atively high probability of being intercepted by the numerous suspension
and surface deposit feeders present. Once the pollutant has arrived at the
sediment surface, it also has a higher probability of being ingested and
 w
=

Table III. Geotechnical and Transport Properties of Nearshore Marine Muds as Influenced by Faunal Successional StageO

Pioneering stages Equilibrium stages

Physical sedimentary
properties Observations Interpretations Observations Interpretations

1. Shear strength Increased in upper 0- to 2- May be related to Shear strengths in the Intensive particle
cm interval relative to pioneering species upper 0- to 2-cm interval bioturbation breaks up
macrofauna-free muds pumping water across are decreased relative to mucus-bound surface,
(3.2). the sediment surface, macrofauna-free muds producing a "granular"
stimulating surface (3.2). texture at the sediment
primary production and surface (4.2). Sediment
microbial growth at the is maintained in a
sediment surface dilated state (3.2).
(mucopolysaccharide
binding) (4 .2-4.3). Slow
rate of particle advection
allows sediment to
compact.
2. Water content (and Depends on frequency and Low water contents Commonly > 60% in the Related to intensive
density/void ratio intensity of bottom «60%) related to low upper 0- to 2-cm interval particle reworking and
correlates) resuspension by water rate of particle of muds (4.3). presence of intergranular
turbulence. advection. However, or interpelletal void
frequent bottom space as well as n
res us pension may keep presence of open ::r
po
'd
sediment near the liquid burrows (4.3). co...,
limit (4 .3).
3. Biogenic texture Surface pelletized, Most particle feeding is Pellets found distributed Most pellets are deposited ~
especiall y by done from the water throughout the actively near the sediment ct
;::;.
polychaetes (4.2). column or sediment bioturbated depth surface but many are 0
'"......,
surface. Pellets are interval (4.2). advected downward by
deposited at the surface biogenic mixing (4.2). s::
OJ
...,
and are not advected S·
co
downward (4.2) . tIl
4. Surface relief Patches of densely packed Sediment accretion within Relief depends on species High water content and co
::s
tubes. Patches may be dense patches (4.1). present and their modes low shear strength of S-
o
associated with of reworking. If species sediment do not permit '"0
topographic elevations are laterally mobile, topographic relief to ::s
(/)
composed of organic- relief may be small. If develop on a large scale. co
0-
rich fine-grained the species are Relief limited to
sediment (2.1-4.1). sedentary, the surface activities and structures
S'
co
may be covered with of individual sedentary 8-
"0
feeding pits and organisms (2.2-4.1). ::r
'<
excavation or fecal and '"
pseudofecal mounds
n'
e:..
(4.1). ...,"0
0
5. Sediment transport Based on observations 1, 2, Related to Based on observations 1 Related to the presence of "d
co
and 4 above, we predict mucopolysaccharide and 2 above U;rit. low-density organic ::l
binding (4.2-4.3) and/or (ii'
an increase in U;rit. at decreased relative to mineral aggregates at the
'"
high population "skimming" flow (4.1). macrofauna-free mud sediment surface.
densities. At low (4.3). Seasonal changes in
densities, bottom may be erodibility may be
destabilized relative to a related to efficiency of
macrofauna-free mud mucopolysaccharide
(4.1). binding and rates of
bioturbation (4.3) .

a Numbers in parentheses refer to sections of the text in which observations and interpretations are discussed in more detail.

~
t.C
40 Chapter 1

incorporated into fecal pellets. The pellets will be deposited at the surface
and may be bound there by mucopolysaccharides or trapped in stable
eddies between densely packed tubes. Dissolved pollutants can come into
contact with anaerobic subsurface sediment by being pumped downward
1 or 2 cm through tubes and diffusing through tube walls into the ambient
sediment. The residence time of pollutants in a pioneering stage depends
on the frequency and intensity of major physical disturbances, which are
associated with massive mortality and washout of both sediments and
fauna. Solid-phase pollutants transported in suspension over an equilib-
rium community are less likely to be actively filtered or otherwise inter-
cepted by feeding benthos because most of the infauna feed below the
sediment surface. However, once a particle settles to the bottom, there is
a higher probability that the contaminant will be advected downward,
ingested, and incorporated into the deposit in the form of an aggregate.
It is also likely that some of the particulate pollutants will reappear at the
interface and be resuspended and dispersed by bottom currents with low
mean critical shear velocities (e.g., U;ril. = 2.0 for 42-fLm particles in Long
Island Sound). The resuspended particles may scavenge other dissolved
pollutants from the water column (e.g., Aller, 1978; Aller et al., 1980).
Dissolved pollutants can be actively drawn into the bottom to depths of
10-20 cm by the respiratory pumping of burrowing or tube-dwelling
species. Muds occupied by equilibrium species are chemically reactive
owing to intensive particle and fluid bioturbation, and the potential for
particle resuspension by weak bottom currents is increased (Aller, 1980).
Moreover, the mean residence times of pollutants in sediments populated
by equilibrium communities are expected to be longer than in areas of
the seafloor dominated by pioneering assemblages.

6. Recommendations for Future Work

In order to test further how geotechnical properties of sediments are
related to benthic successional history, it is necessary to make in situ
observations of organism-sediment relations. Conventional means of bot-
tom sampling disturb or destroy important structures near the sedi-
ment-water interface. * Free-fall samplers are preceeded by a pressure
wave, which washes away small tubes, pelletal layers, and fecal mounds.
Existing methods of faunal analysis, such as grab sampling and sieving,
are similarly inadequate because important information about the vertical

* The only ship-deployed sampler that we have seen that recovers a nearly undisturbed
sediment-water interface sample is a spade box core that has been fitted with internal
vertical partitions. These partitions dampen out water motion over the bottom as the
sample is recovered (e.g., Hessler and Jumars. 1974).
Effects of Marine Benthos on Sediment Physical Properties 41

stratification of infauna is lost (Rhoads et aI., 1981). Similarly, infor-

mation about the dispersion patterns of tubes and feeding structures,
surface pellet layers, the depth of the RPD, and the depth of subsurface
feeding voids is also lost.
The Rhoads-Cande Profile Camera was developed specifically for
the purpose of observing organism-sediment relations in situ (Rhoads
and Cande, 1971). The ability of this photographic technique to identify
successional stages accurately has been documented by Bosworth et al.
(1980) and Rhoads and Germano (1982). The profile camera technique
has been used successfully for over 10 years. We have also found sediment
radiography (e.g., Figs. 2A and 3A) to provide useful information about
sedimentary processes related to the successional paradigm (Tenore et
al., 1982).
A newly developed and promising remote sensing technique is ul-
trasound imaging (Orr and Rhoads, 1982). Our initial laboratory work has
been done with a 1.6-MHz backscattering system. Small-scale (approxi-
mately l-cm) structures can be imaged acoustically at the surface of sed-
iments; more importantly, subsurface structures can be seen to a depth
of 10 cm. The ultrasound technique can be adapted to a towed sled vehicle
pulled over the seafloor. Such a system might allow one to map rapidly
the distribution of successional stages by providing acoustic information
on sedimentary and biological structures at and below the sediment-water
interface. At the Woods Hole Oceanographic Institution, A. J. Williams
and W. D. Grant (personal communication) have constructed a I-MHz
acoustic profiler capable of measuring bed form topography to 2:1 mm.
High-frequency sound may have potential for remotely measuring
and detecting changes in mass geotechnical properties (Richardson and
Young, 1981). This technique requires inserting an acoustic probe verti-
cally into the surface of the sediment. Sound is propagated horizontally
and the difference in travel times along two different path lengths is
measured. Vertical profiles in acoustic properties of the sediments can
be obtained by sequentially moving the probe downward. This study has
shown that bioturbated muds have lower impedance, a lower Rayleigh
reflection coefficient, lower bulk and shear moduli, and a lower shear
wave velocity propagation than nonbiotubated muds (see Richardson and
Young, 1981).
Future work must involve measurement of geotechnical properties
on smaller scales than has been done in the past. For problems relating
to sediment transport, these measurements may have to be made within
1 or 2 mm of the sediment-water interface. We are not aware of any
existing geotechnical instruments that can make measurements on this
scale.
In situ measurement of critical shear stresses and shear velocity at
42 Chapter 1

the ocean floor is also desirable. In this regard, the Seaflume concept of
R. A. Young (1977) is probably the most appropriate means for deter-
mining the in situ critical shear velocity of natural sediments populated
by organisms. In order to better understand complex organism-
sediment-fluid interactions, more laboratory flume work is required.
Recent advances in forward and backscattering laser-doppler velo-
cimetry technology open up a new approach for accurately measuring
water flow and suspended solids within a flume channel. Vertical velocity
profiles may be made to within a fraction of a millimeter of the bed
without interfering with flow or organisms (Grant et aI., 1982). Once again,
high-resolution profiling is necessary in organism-sediment work be-
cause many biological structures and their effects are of small scale.
The influence of tubes, pelletal layers, and mucopolysaccharide bind-
ers on sediment stability must be investigated under highly controlled
laboratory conditions. Of particular importance is the characterization of
mucus binders. This is both a biochemical and a mechanical problem. No
one has yet measured the elasticity or breaking strength of these com-
pounds, nor are their kinetics or the conditions of their breakdown com-
pletely known.
It is clear from our earlier discussion of successional processes and
metabolic rate factors that the geotechnical and transport properties of
sediments are temporally and spatially dynamic. A sampling program
designed to address the problem of seafloor stability should take into
account important biological events. In environments that experience a
large annual range in temperature and solar radiation, sampling should
be extensive in the spring (when rising temperatures and increasing day
length stimulate recruitment and primary production). Sampling in late
summer and early autumn will characterize biological processes working
at their maximum rates. Early winter is associated with declining meta-
bolic rates and sediment processing. Late winter sampling may closely
approximate "azoic" sediment conditions, as benthic organisms, although
present, are metabolically dormant.
Thermally uniform deep-water environments may also experience
subdued seasonality. Cycles in surface water production result in the
sedimentation of detritus to the seafloor in the form of zooplankton fecal
pellets (Honjo, 1976). This pulse of detritus may be associated with an
increase in benthic activity. In shallow-water tropical environments, pe-
riodic changes in nearshore salinity and water turbulence during monsoon
periods can similarly cause major restructuring of the benthic community.
An areal sampling plan for benthic environs can be more equitably
constructed if the physical disturbance history of the seafloor is known
or can be estimated. One need not know this history precisely. Estimates
may be made on the basis of a reconnaissance survey of sedimentary
Effects of Marine Benthos on Sediment Physical Properties 43

fabrics obtained in X-radiographs of box cores or with an interface camera

survey (Rhoads et al., 1981; Rhoads and Germano, 1982). The geologist
or sedimentologist need not be highly proficient in the identification of
benthic species to make judgments about successional history. If the ge-
ologist can recognize the functional aspects of benthic communities as
outlined in Fig. 1, this information is sufficient to recognize end-member
disturbance regimes. Once this organism-sediment information is re-
corded on a hydrographic chart, the selection of stations for measuring
sedimentary or biological properties is facilitated.

ACKNOWLEDGMENTS. This work was supported by NSF grant OCE 7826211

and EPA Cooperative Agreement 807331-01 to Donald C. Rhoads. J. Ger-
mano and D. Muschenheim procured cores of the seafloor while scuba
diving. J. Germano identified the macrofauna in the microcosm experi-
ments. M. Pimer of Professional Sea Services and the R/V Gale provided
support at sea. D. Muschenheim and D. Kramer aided in the construction
and fabrication of the experimental set-up and in the measurement pro-
gram. A. Goodhue constructed the experimental tanks. C. Phelps of Draft-
A-Line helped with the figures; W. Sacco provided the finished prints.
B. Gorecki typed the original manuscript. Grain size information was
provided by D. K. Young and M. Richardson of NORDA, Bay St. Louis,
Mississippi. R. Gularte of Levingston Marine Corporation, Annapolis,
Maryland, built and calibrated the shear vane device for our use. We also
thank the many people who have reviewed the manuscript, especially J.
Germano, P. Jumars, A. Nowell, J. Eckman, G. Taghon, W. Grant, and the
editors, P. L. McCall and M. J. S. Tevesz.

References
Aller, R. c., 1977, The influence of macrobenthos on chemical diagenesis of marine sedi-
ments, Ph.D. dissertation, Department of Geology and Geophysics, Yale University, New
Haven, Connecticut.
Aller, R. C., 1978, Experimental studies of changes produced by deposit feeders on pore
water, sediment, and overlying water chemistry, Am. J. Sci. 217:1185-1234.
Aller, R. C., 1980, Relationships of tube-dwelling benthos with sediment and overlying water
chemistry, in: Marine Benthic Dynamics (K. R. Tenore and B. C. Coull, eds.). pp.
285-308, University of South Carolina Press, Columbia.
Aller, R. C., and Cochran, J. K., 1976, 234Th/238U disequilibrium in nearshore sediment:
Particle reworking and diagenetic time scales, Earth Planet. Sci. Lett. 20:37-50.
Aller, R. C" and Dodge, R. E., 1974, Animal-sediment relations in a tropical lagoon, Dis-
covery Bay, Jamaica, J. Mar. Res. 32:209-232.
Aller, R. C., and Yingst, J. Y., 1978, Biogeochemistry of tube-dwellings: A study of the
sedentary polychaete Amphitrite ornata (Leidy). J. Mar. Res. 36:201-254.
Aller, R. C" and Yingst, J. Y., 1980, Relationships between microbial distributions and the
44 Chapter 1

anaerobic decomposition of organic matter in surface sediments of Long Island Sound,

U.S.A., Mar. Biol. 56:29-42.
Aller, R. C., Benninger, 1. K., and Cochran, J. K., 1980, Tracking particle associated processes
in nearshore environments by use of 234Th/238U disequilibrium, Earth. Planet. Sci. Lett.
47:161-175.
Aspiras, R. B., Allen, O. N., Harris, R. F., and Chesters, G., 1971, The role of micro-organisms
in the stabilization of soil aggregates, Soil Biol. Biochem. 3:347-353.
Atkinson, R. J. A., and Pullin, R. S. V., 1976, The red band-fish, Cepola rubescens L. at
Lundy, Rep. Lundy Fld. Soc. 27:1-6.
Baier, R. E., 1973, Influence of the initial surface condition of materials in bioadhesion, in:
Proceedings of the Third International Congress on Corrosion and Fouling (R. F. Acker,
B. F. Brown, J. R. DePalma, and W. D. Iverson, eds.). pp. 633-639, National Bureau of
Standards, Washington, D.C.
Bailey-Brock, J. H., 1979, Sediment trapping by chaetopterid polychaetes on a Hawaiian
fringing reef, J. Mar. Res. 37:643-656.
Bell, S. S., and Coull, B. c., 1980, Experimental evidence for a model of juvenile macro-
fauna-meiofauna interactions, in: Marine Benthic Dynamics (K. R. Tenore and B. C.
Coull, eds.). pp. 179-192, University of South Carolina Press, Columbia.
Berger, W. H., and Heath, G. R., 1968, Vertical mixing in pelagic sediments, J. Mar. Res.
26:134-143.
Bokuniewicz, H. J., Gordon, R. B., and Rhoads, D. C., 1975, Mechanical properties of the
sediment-water interface, Mar. Geol. 18:263-278.
Boswell, P. G. H., 1961, Muddy Sediments: Some Geotechnical Studies for Geologists, En-
gineers, and Soil Scientists, Heffer, Cambridge.
Bosworth, W. S., Germano, J., Hartzband, D. J., McCusker, A. J., and Rhoads, D. C., 1980,
Use of benthic sediment-profile photography in dredging impact analysis and moni-
toring, Ninth World Dredging Conference, 29-31 October, 1980, Vancouver, British
Columbia, Canada.
Boyer, 1. F., 1980, Production and preservation of surface traces in the intertidal zone, Ph.D.
dissertation, Department of the Geophysical Sciences, The University of Chicago, Chi-
cago, Illinois.
Cadee, G. c., 1979, Sediment reworking by the polychaete Heteromastus fiIiformis on a tidal
flat in the Dutch Wadden Sea, Neth. J. Sea Res. 13:441-456.
Carney, R. S., 1981, Bioturbation and biodeposition, in: Principles of Benthic Marine Pa-
leoecology (A. J. Boucot, ed.). pp. 357-400, Academic Press, New York.
Chapman, G. A., 1949, The thixotropy and dilatancy of a marine soil, J. Mar. BioI. Assoc.
U.K. 28:123-140.
Chapman, G. A., and Newell, G. E., 1947, The role of the body-fluid in relation to movement
in soft bodied invertebrates. i. The burrowing of Arenicola, Proc. R. Soc. London Ser.
B 134:431-455.
Clark, R. B., 1964, Dynamics of Metazoan Evolution, Clarendon Press, Oxford, 313 pp.
Clements, F. E., 1916, Plant succession: An analysis of the development of vegetation,
Carnegie Institute, Washington, Publication 242, 512 pp.
Cool. D.O., 1971, Depressions in shallow marine sediments made by benthic fishes, J.
Sediment. Petrol. 41:577-578.
Crozier, W., 1918, The amount of bottom material ingested by holothurians (Stichopus). J.
Exp. Zool. 26:379-389.
Cullen, D. J., 1973, Bioturbation of superficial marine sediments by interstitial meiobenthos,
Nature 242:323-324.
Dapples, E. c., 1942, The effect of macro-organisms upon near shore marine sediments, J.
Sediment. Petrol. 12:118-126.
Effects of Marine Benthos on Sediment Physical Properties 45

Dauer, D. M., and Simon, J. L., 1975, Repopulation of the polychaete fauna of an intertidal
habitat following natural defaunation: Species equilibrium, Oecologia (Berlin) 22:99-117.
Dayton, P. K., and Oliver, J. S., 1980, An evaluation of experimental analyses of population
and community patterns in benthic marine environments, in: Marine Benthic Dynamics
(K. R Tenore and B. C. Coull, eds.), pp. 93-120, University of South Carolina Press,
Columbia.
Eckman, J. E., 1979, Small-scale patterns and processes in a soft substratum, intertidal
community, J. Mar. Res. 37:437-457.
Eckman, J. E., Nowell, A. R M., and Jumars, P. A., 1979, The influence of animal motility
on sediment entrainment, Eos 60:847 (abs.).
Eckman, J. E., Nowell, A. R M., and Jumars, P. A., 1981, Sediment destabilization by animal
tubes, J. Mar. Res. 39:361-374.
Einsele, G., Overbeck, R, Schwarz, H. U., and Unsold, G., 1974, Mass physical properties,
sliding and erodibility of experimentally deposited and differently consolidated clayey
muds, Sedimentology 21:339-372.
Einstein, H. R, and Krone, R B., 1962, Experiments to determine modes of cohesive sediment
transport in salt water, J. Geophys. Res. 67:1451-1461.
Fager, E. W., 1964, Marine sediments; effects of a tube-building polychaete, Science
143:356-359.
Featherstone, R P., and Risk, M. J., 1977, Effects of tube-building polychaetes on intertidal
sediments of the Minas Basin, Bay of Fundy, J. Sediment. Petrol. 47:446-450.
Fenchel, T., 1970, Studies on the decomposition of organic detritus derived from the turtle
grass Thalassia testudinium, Limnol. Oceanogr. 15:14-20.
Frankel, L., and Meade, D. J., 1973, Mucilaginous matrix of some estuarine sands of Con-
necticut, J. Sediment. Petrol. 43:1090-1095.
Frankenberg, D., and Smith, K. L., Jr., 1967, Coprophagy in marine animals, Limnol. Ocean-
ogr. 12:443-450.
Frey, R. W. (ed.), 1975, The Study of Trace Fossils, Springer-Verlag, New York.
Ginsburg, R N., and Lowenstam, H. A., 1958, The influence of marine bottom communities
on the depositional environment of sediments, ]. Geol. 66:310-318.
Goldhaber, M. B., Aller, R c., Cochran, J. K., Rosenfeld, J. K., Martens, C. S., and Berner,
R A., 1977, Sulfate reduction, diffusion, and bioturbation in Long Island Sound sedi-
ments: Report of the FOAM group, Am. J. Sci. 277:193-237.
Gordon, D. C., Jr., 1966, The effects of the deposit feeding polychaete Pectinaria gouldii on
the intertidal sediments of Barnstable Harbor, Limnol. Oceanogr. 11:327-332.
Goupil, D. W., DePalma, V. A., and Baier, R. E., 1973, Prospects for non-toxic fouling-resistant
paints, in: Proceedings of the 9th Marine Technology Society Conference, pp. 445-458,
Marine Technology Society, Washington, D.C.
Grant, W. D., Boyer, L. F., and Sanford, L. P., 1982, The effect of biological processes on the
initiation of sediment motion in non-cohesive sediment, J. Mar. Res. 40: (in press).
Gray, J. 5., 1974, Animal-sediment relationships, Oceanogr. Mar. Biol. Annu. Rev.
12:223-261.
Gularte, R C., 1978, The erosion of cohesive marine sediments as a rate process, Ph.D.
dissertation, University of Rhode Island, Kingston, Rhode Island.
Gularte, R C., Kelley, W. E., and Nacci, V. A., 1980, Erosion of cohesive sediments as a rate
process, Ocean Eng. 7:539-551.
Hanor, J. 5., and Marshall, N. F., 1971, Mixing of sediment by organisms, in: Trace Fossils:
A Field Guide to Selected Localities in Pennsylvanian, Permian, Cretaceous, and Ter-
tiary Rocks of Texas and Related Papers (B. F. Perkins, ed.), pp. 127-135, Louisiana
State University Press, Miscellaneous Publication 71-1, Baton Rouge.
Hargrave, B. T., 1976, The central role of invertebrate faeces in sediment decomposition, in:
46 Chapter 1

The role of Terrestrial and Aquatic Organisms in Decomposition Processes 0. M. An-

dersen and A. Macfadyen, eds.), pp. 301-321, Blackwell, Oxford.
Hargrave, B. T., 1980, Factors affecting the flux of organic matter to sediments in a marine
bay, in: Marine Benthic Dynamics K. R. Tenore and B. C. Coull, eds.), pp. 243-264,
University of South Carolina Press, Columbia.
Harris, R. F., Chesters, G., and Allen, O. N., 1966, Dynamics of soil aggregation, Adv. Agron.
18:107-169.
Harrison, W., and Wass, M. L., 1965, Frequencies of infaunal invertebrates related to water
content of Chesapeake Bay sediments, Southeast. Geol. 6:177-187.
Harrison, W., Lynch, M. P., and Altschaefel, A. G., 1964, Sediments of lower Chesapeake
Bay with emphasis on mass properties, J. Sediment. Petrol. 34:727-755.
Haven, D. S., and Morales-Alamo, R., 1966, Aspects of biodeposition by oysters and other
invertebrate filter feeders, Limnol. Oceanogr. 11:487-498.
Haven, D. S., and Morales-Alamo, R., 1972, Biodeposition as a factor in sedimentation of
fine suspended solids in estuaries, Geol. Soc. Am. Mem. 133:121-130.
Hessler, R. R., and Jumars, P. A., 1974, Abyssal community analysis from replicate box cores
in the central No. Pacific, Deep Sea Res. 21:185-209.
Hobbie, J. E., and Lee, C., 1980, Microbial production of extracellular material: Importance
in benthic ecology, in: Marine Benthic Dynamics (K. R. Tenore and B. C. Coull, eds.),
pp. 341-346, University of South Carolina Press, Columbia.
Holland, A. F., Zingmark, R. G., and Dean, J. M., 1974, Quantitative evidence concerning
the stabilization of sediments by marine benthic diatoms, Mar. Bioi. 27:191-196.
Honjo, S., 1976, Coccoliths: Production, transportation and sedimentation, Mar. Micropa-
leontol. 1:65-79.
Johnson, R. G., 1971, Animal-sediment relations in shallow water benthic communities,
Mar. Geol. 11:93-104.
Johnson, R. G., 1972, Conceptual models of benthic marine communities, in: Models in
Paleobiology (T. J. M. Schopf, ed.), pp. 149-159, Freeman and Cooper, San Francisco.
Johnson, R. G., 1974, Particulate matter at the sediment-water interface in coastal environ-
ments, J. Mar. Res. 33:313-330.
Johnson, R. G., 1977, Vertical variation in particulate matter in the upper twenty centimenters
of marine sediments, J. Mar. Res. 35:273-282.
Jumars, P. A., and Hessler, R. R., 1976, Hadal community structure: Implications from the
Aleutian Trench , J. Mar. Res. 34:547-560.
Jumars, P. A., Nowell, A. R. M., and Self, R. L. F., 1981, A simple model of
flow-sediment-organism interactions, Mar. Geol. 42:155-172.
Khailov, K. M., and Finenko, Z. Z., 1970, Organic macromolecular compounds dissolved
in sea-water and their inclusion into food chains, in: Marine Food Chains 0. H. Steele,
ed.), pp. 6-18, University of California Press, Berkeley.
Krank, K., 1973, Flocculation of suspended sediment in the sea, Nature 246:348-350.
Krank, K., 1975, Sediment deposition from flocculated suspensions, Sedimentology 22:111-
123.
Lambe, T. W., and Whitman, R. V., 1969, Soil Mechanics, John Wiley and Sons, New York,
553 pp.
Lee, H., II, and Swartz, R. c., 1980, Biological processes affecting the distribution of pollutants
in marine bioturbation, in: Contaminants and Sediments (R. A. Baker, ed.), Volume 2,
pp. 555-605, Science Publishers, Ann Arbor, Michigan.
Lettau, P., 1969, Note on aerodynamic roughness parameter estimation, J. Appl. Meteorol.
8:828-832.
Levinton, J. S., 1980, Particle feeding by deposit-feeders: Models, data, and a prospectus,
in: Marine Benthic Dynamics (K. R. Tenore and B. C. Coull, eds.), pp. 423-441, Uni-
versity of South Carolina Press, Columbia.
Effects of Marine Benthos on Sediment Physical Properties 47

Levinton, J. S., and Lopez, G. R., 1977, A model of renewable resources and limitation of
deposit-feeding benthic populations, Oecologia (Berlin) 31:117-190.
Lonsdale, P., and Southard, J. B., 1974, Experimental erosion of North Pacific red clay, Mar.
Geol. 17:M51-M60.
Lynch, D. L., and Cotnoir, 1. J. , Jr., 1956, The influence of clay minerals on the breakdown
of certain organic substrates, Soil Sci. Soc. Am. Proc. 20:367-370.
Lynch, M., and Harrison, W., 1970, Sedimentation caused by a tube-building amphipod, J.
Sediment. Petrol. 40:434-435.
McCall, P. 1., 1975 , The influence of disturbance on community patterns and adaptive
strategies of the infaunal benthos of central Long Island Sound, Ph.D. dissertation, Yale
University , New Haven, Connecticut.
McCall, P. 1., 1977, Community patterns and adaptive strategies of the infaunal benthos of
Long Island Sound, J. Mar. Res. 35:221-266.
McCall, P. 1., 1978, Spatial-temporal distributions of Long Island Sound infauna: The role
of bottom disturbance in a nearshore marine habitat, in: Estuarine Interactions (M. L.
Wiley, ed.), pp. 191-219, Academic Press, New York.
McCall, P. 1., 1979 , The effects of deposit-feeding oligochaetes on particle size and settling
velocity of Lake Eerie sediments, J. Sediment. Petrol. 49:813-818.
McCall, P. 1., and Fisher, J. B., 1980, Effects of tubificid oligochaetes on physical and
chemical properties of Lake Erie sediments, in: Aquatic Oligochaete Biology (R. O.
Brinkhurst and D. G. Cook, eds .), pp. 253-317 , Plenum Press, New York.
Maclivaine, J. C., and Ross, D. A., 1979, Sedimentary processes on the continental slope of
New England, J. Sediment. Petrol. 49:565-574 .
McIntosh, R. P. , 1980, The relationship between succession and the recovery process in
marine sediments, in: The Recovery Process in Damaged Ecosystems 0. Cairns , Jr. , ed.),
pp. 11-62, Ann Arbor Scientific Publishers, Ann Arbor, Michigan.
McMaster, R. L., 1962, Seasonal variability in compactness in marine sediments: A laboratory
study, Geol. Soc. Am. Bull. 73:643-646.
McMaster, R. 1. , 1967, Compactness variability of estuarine sediments : An in situ study, in:
Estuaries (G. Lauff, ed.), pp . 261-267, American Association for the Advancement of
Science, Washington, D.C.
Madsen, O. S., and Grant, W. D., 1976, Sediment transport in the coastal environment, Ralph
M. Parsons Laboratory for Water Resources and Hydrodynamics, Report 209, Massa-
chusetts Institute of Technology, Cambridge, Massachusetts, 105 pp.
Mare, M. F., 1942, A study of the marine benthic community with special reference to the
microorganisms, J. Mar. BioI. Assoc. U. K. 25:517-554 .
Marshall, N., and Lucas, K ., 1970, Preliminary observations on the properties of bottom
sediments with an without eelgrass, Zostera marina, Proc. Natl. Shellfish Assoc.
60:107-111.
Martin, J. P., and Waksman, S. A., 1940, Influence of microorganisms on soil aggregation
and erosion, Soil Sci. 50:29-47.
Martins, J. P. , and Craggs, B. A., 1946, Influence of temperature and moisture on the soil-
aggregating effect of organic residues , J. Am. Soc. Agron. 38:322-339.
Migniot, C., 1968, Etude des proprif~tes physiques de differents sediments tres fins et de
leur comportement sous des actions hydrodynamiques, La Houille Blanche 23:591-620.
Miller, M. C., McCave, I. N., and Komar, P. D., 1977, Threshold of sediment motion under
unidirectional currents , Sedimentology 24:507-527.
Mills, E. 1., 1967, The biology of an ampeliscid amphipod crustacean sibling species pair,
J. Fish. Res. Board Can . 24:305-355.
Mills, E. L., 1969, The community concept in marine zoology, with comments on continua
and instability in some marine communities: A review, J. Fish. Res. Board Can.
26:1415-1428.
48 Chapter 1

Mitchell, R., and Nevo, Z., 1964, Effects of bacterial polysaccharide accumulation on infil-
tration of water through sand, Appl. Microbiol. 12:219-223.
Moore, D. G., and Scruton, P. c., 1957, Minor internal structure of some recent unconsoli-
dated sediments, Bull. Am. Assoc. Petrol. Geol. 41:2723-2751.
Moore, H. B., 1931, The muds of the Clyde Sea area. III. Chemical and physical conditions;
rate and nature of sedimentation; and fauna, J. Mar. Biol. Assoc. U. K. 17:325-358.
Moore, H. B., 1939, Faecal pellets in relation to marine deposits, in: Recent Marine Sediments
(P. Trask, ed.). pp. 516-523, American Association of Petroleum GeologistslDover Press,
New York.
Morris, H. M., 1955, A new concept of flow in rough conduits, Trans. Am. Soc. Civ. Eng.
120:373-398.
Myers, A. C., 1977a, Sediment processing in a marine subtidal sandy bottom community.
I. Physical aspects, J. Mar. Res. 35:609-632.
Myers, A. c., 1977b, Sediment processing in a marine subtidal sandy bottom community.
II. Biological consequences, J. Mar. Res. 35:633-647.
Myers, A. C., 1979, Summer and winter burrows of a mantis shrimp, Squilla empusa, in
Narragansett Bay, Rhode Island (U.S.A.). Estuarine Coastal Mar. Sci. 8:87-98.
Neihof, R. A., and Loeb, G. I., 1972, The surface charge of particulate matter in seawater,
Limnol. Oceanorgr. 17:7-16.
Neihof, R. A., and Loeb, G. I., 1973, Molecular fouling surfaces in seawater, in: Proceedings
of the Third International Congress on Marine Corrosion and Fouling (R. F. Aker, B.
F. Brown, J. R. dePalma, and W. P. Iverson, eds.). pp. 710-718, National Bureau of
Standards, Gaithersburg, Virginia.
Neuman, A. C., Gebelein, C. P., and Scoffin, T. P., 1970, The composition, structure, and
erodibility of subtidal mats, Abaco, Bahamas, J. Sediment. Petrol. 40:274-297.
Nicol. J. A. C., 1967, The Biology of Marine Animals, John Wiley and Sons, New York.
Nixon, S. W., Kelly, J. R., Furnas, B. N., Oviatt, C. A., and Hale, S. S., 1980, Phosphorus
regeneration and the metabolism of coastal marine bottom communities, in: Marine
Benthic Dynamics (K. R. Tenore and B. C. Coull eds.). pp. 219-242, University of South
Carolina Press, Columbia.
Nowell, A. R. M., and Church, M., 1979, Turbulent flow in a depth-limited boundary layer,
J. Geophys. Res. 84:4816-4824.
Nowell, A. R. M., Jumars, P. A., and Eckman, J. E., 1981, Effects of biological activity on the
entrainment of marine sediments, Mar. Geol. 42:133-153.
Orr, M., and Rhoads, D. C., 1982, Acoustic imaging of structures and macrofauna in the
upper 10 cm of sediments using a megahertz backscattering system, Mar. Geol.
(in press).
Osman, R. W., and Whitlatch, R. B., 1978, Patterns of species diversity: Fact or artifact?
Paleobiology 4:41-54.
Partheniades, E., 1965, Erosion and deposition of cohesive soils, Proc. Am. Soc. Civ. Eng.,
I. Hydraul. Div. 91:105-139.
Parthenaides, E., and Paswell, R. E., 1970, Erodibility of channels with cohesive boundary,
Proc. Am. Soc. Civ. Eng., I. Hydraul. Div. 96:755-771.
Pearson, T. H., and Rosenberg, R., 1978, Macrobenthic succession in relation to organic
enrichment and pollution of the marine environment, Oceanogr. Mar. Biol. Annu. Rev.
16:229-311.
Pearson, T. H., and Stanley, S. 0., 1979, Comparative measurement of the redox potential
of marine sediments as a rapid means of assessing the effect of organic pollution, Mar.
Biol. 53:371-379.
Pemberton, G. S., Risk, M. J., and Buckley, D. E., 1976, Supershrimp: Deep bioturbation in
the strait of Canso, Nova Scotia, Science 192:790-791.
Effects of Marine Benthos on Sediment Physical Properties 49

Petersen, C. G. J., 1913, Valuation of the sea. II. The animal communities of the sea bottom
and their importance for marine zoogeography, Rep. Danish BioI. Stat. 21:1-44.
Phelps, D. K., 1966, Partitioning of the stable elements, Fe, Zn, Se, Sm, within a benthic
community, Anasco Bay, Puerto Rico, in: Radioecological Concentration Processes (B.
Aberg and F. P. Hungate, eds.), pp. 721-734, Pergamon Press, New York.
Pinck, L. A., Dyal, R. S., and Allison, F. E., 1954, Protein-montmorillonite complexes, their
preparation and the effects of soil micro-organisms on their decomposition, Soil Sci.
78:109-118.
Postma, H., 1967, Sediment transport environment, in: Estuaries (G. H. Lauff, ed.), pp.
158-179, American Association for the Advancement of Science, Washington, D.C.
Powell, E. N., 1977, Particle size selection and sediment reworking in a funnel feeder,
Leptosynapta tenuis (Holothuroidea, Synoptidae), lnt. Rev. Ges. Hydrobiol. 62:385-403.
Propp, M. V., Tarasoff, V. G., Cherbadgi, I. I., and Lootnik, N. V., 1980, Benthic-pelagic
oxygen and nutrient exchange in a coastal region of the Sea of Japan, in: Marine Benthic
Dynamics (K. R. Tenore and B. C. Coull, eds.), pp. 265-284, University of South Carolina
Press, Columbia.
Purdy, E., 1964, Sediments as substrates, in: Approaches to Paleoecology (J. Imbrie and N.
Newell, eds.), pp. 238-271, John Wiley and Sons, New York.
Raupach, T., 1981, A wind tunnel study of turbulent flow close to regularly arrayed rough
surfaces, Boundary Layer Meteorol. 18:373-397.
Revsbech, N. P., Jorgensen, B. B., and Blackburn, T. H., 1979, Oxygen in the sea bottom
measured with a microelectrode, Science 207:1355-1356.
Rhoads, D. C., 1963, Rates of sediment reworking by Yoldia limatula in Buzzards Bay,
Massachusetts and Long Island Sound, J. Sediment. Petrol. 33:723-727.
Rhoads, D. C., 1967, Biogenic reworking of intertidal and subtidal sediments in Barnstable
Harbor and Buzzards Bay, Massachusetts, J. Geol. 75:461-476.
Rhoads, D. C., 1970, Mass properties, stability and ecology of marine muds related to bur-
rowing activity, in: Trace Fossils (T. P. Crimes and J. C. Harper, eds.), pp. 391-406,
Seel House Press, Liverpool.
Rhoads, D. C., 1973, The influence of deposit-feeding benthos on water turbidity and nutrient
recycling, Am. J. Sci. 273:1-22.
Rhoads, D. c., 1974, Organism-sediment relations on the muddy sea floor, Oceanogr. Mar.
BioI. Annu. Rev. 12:263-300.
Rhoads, D. C., and Boyer, L. F., 1983, Seasonal patterns in sediment resuspension in central
Long Island Sound, U.S.A., (in preparation).
Rhoads, D. C., and Cande, S., 1971, Sediment profile camera for in situ study of orga-
nism-sediment relations, Limnol. Oceanogr. 16:110-114.
Rhoads, D. C., and Germano, J., 1982, Characterization of benthic processes using sediment-
profile imaging: An efficient method of Remote Ecological Monitoring of the seafloor
(REMOTS system), Mar. Ecol. Prog. Ser. (in press).
Rhoads, D. C., and Stanley, D. J., 1964, Biogenic graded bedding, J. Sediment. Petrol.
35:956-963.
Rhoads, D. C., and Young, D. K., 1970, The influence of deposit-feeding organisms on
sediment stability and community trophic structure, J. Mar. Res. 28:150-178.
Rhoads, D. C., and Young, D. K., 1971, Animal-sediment relations in Cape Cod Bay, Mas-
sachusetts. II. Reworking by Molpadia oolitica (Holothuroidea), Mar. Bioi. 11:255-
261.
Rhoads, D. C., Aller, R. C., and Goldhaber, M., 1977, The influence of colonizing macro-
benthos on physical properties and chemical diagenesis of the estuarine seafloor, in:
Ecology of Marine Benthos (B. C. Coull, ed.), pp. 113-138, University of South Carolina
Press, Columbia.
50 Chapter 1

Rhoads, D. c., McCall, P. L., and Yingst, J. Y., 1978a, Disturbance and production on the
estuarine seafloor, Am. Sci. 66:577-586.
Rhoads, D. C., Yingst, J. Y., and Ullman, W., 1978b, Seafloor stability in central Long Island
Sound. Part I. Temporal changes in erodibility of fine-grained sediment, in: Estuarine
Interactions (M. L. Wiley, ed.), pp. 221-224, Academic Press, New York.
Rhoads, D. C., Germano, J., and Boyer, L. F., 1981, Sediment-profile imaging: An efficient
method of Remote Ecological Monitoring of the Seafloor (REMOTS system), Oceans
1:561-566, Publication No. 81CH1685-7, IEEE, Piscataway, New Jersey.
Richards, A. F., and Parks, J. M., 1976, Marine geotechnology: Average sediment properties,
selected literature and review of consolidation, stability and bioturbation-geotechnical
interactions in the benthic boundary layer, in: The Benthic Boundary Layer (I. N.
McCave, ed.), pp. 157-181, Plenum Press, New York.
Richardson, M. D., and Young D. K., 1981, Geoacoustic models and bioturbation, Mar. Ceol.
38:205-218.
Riley, G. A., 1963, Organic aggregates in seawater and the dynamics of their formation and
utilization, Limnol. Oceanogr. 8:372-381.
Risk, M. J., and Moffat, J. S., 1977, Sedimentological significance of fecal pellets of Macoma
baltica in the Minas Basin, Bay of Fundy, J. Sediment. Petrol. 47:1425-1436.
Rowe, G. T., 1974, The effects of the benthic fauna on the physical properties of deep-sea
sediments, in: Deep-Sea Sediments: Physical and Mechanical Properties (A. L. Inder-
bitzen, ed.), pp. 381-400, Plenum Press, New York.
Sanders, H. L., 1958, Benthic studies in Buzzards, Bay. I. Animal-sediment relationships,
Limnol. Oceanogr. 3:245-258.
Sanders, H. L., Goudsmit, E. L., Mills, E. L., and Hampson, G. E., 1962, A study of the
intertidal fauna of Barnstable Harbor, Massachusetts, Limnol. Oceanogr. 7:63-70.
Santos, S. L., and Bloom, S. A., 1980, Stability in an annually defaunated estuarine soft-
bottom community, Oecologia (Berlin) 46:290-294.
Santos, S. L., and Simon, J. L., 1980, Marine soft-bottom community establishment following
annual defaunation: Larval or adult recruitment? Mar. Ecol. Prog. Ser. 2:235-241.
Sayre, W. W., and Albertson, M. L., 1963, Roughness spacing in rigid open channels, Trans.
Am. Soc. Civ. Eng. 128:343-372.
Schafer, W., 1972, Ecology and Paleoecology of Marine Environments (I. Oertel and G. Y.
Craig, translators), University of Chicago Press, Chicago, Illinois.
Schlichting, H., 1936, Experimentelle Untersuchungen zum Rauhigkeitsproblem, Ing. Arch.
7:1-34.
Schwartz, A., 1932, Der tierische Einfluss auf die Meeressedimente, Senckenbergiana
14:118-172.
Scoffin, T. P., 1970, The trapping and binding of subtidal carbonate sediments by marine
vegetation in Bimini Lagoon, Bahamas, J. Sediment. Petrol. 40:249-273.
Self, R. F. L., and Jumars, P. A., 1978, New resource axes for deposit feeders, J. Mar. Res.
36:627-641.
Sepkoski, J. J., 1982, Flat-pebble conglomerates, storm deposits, and the Cambrian bottom
fauna, in: Cyclic and Event Stratification (G. Einsele and A. Seilacher, eds.), Springer-
Verlag, New York (in press).
Shields, A., 1936, Application of similarity principles and turbulent research to bed-load
movement (Mitteilungen der Preussicher Versuchsanstalt fur Wasserbau und Schiffen,
Berlin), in: W. M. Keck Laboratory of Hydraulics and Water Resources, Report 167 (W.
P. Ott and J. C. van Uchelen, translators), California Institute of Technology, Pasadena,
California.
Silva, A. J., and Hollister, C. D., 1973, Geotechnical properties of ocean sediments recovered
with giant piston corer. I. Gulf of Maine, J. Geophys. Res. 78:3597-3616.
Effects of Marine Benthos on Sediment Physical Properties 51
Southard, J, B., 1974, Erodibility of fine abyssal sediment, in: Deep-Sea Sediments: Physical
and Mechanical Properties (A. L. Inderbitzen, ed.), pp. 367-379, Plenum Press, New
York.
Southard, J. B., Young R. A., and Hollister, C. D., 1971, Experimental erosion of calcareous
ooze, J. Geophys. Res. 76:5903-5909.
Sulanowski, J. S. K., 1978, Field study of relationship between organic matter and sedi-
mentary particles, Ph.D. dissertation, The University of Chicago, Chicago, Illinois.
Swartz, R. C., and Lee, H., III, 1980, Biological processes affecting the distribution of pol-
lutants in marine sediments. Part 1. Accumulation, trophic transfer, biodegradation and
migration, in: Contaminants and Sediments, Volume 2 (R. A. Baker, ed.), pp. 533-553,
Ann Arbor Science Publishers, Ann Arbor, Michigan.
Taghon, G. L., Self, R. F. L., and Jumars, P. A., 1978, Predicting particle selection by deposit
feeders: A model and its implications, Limnol. Oceanogr. 23:752-759.
Tenore, K. R., Boyer, L. F., Corral, J., Garcia-Fernandez, c., Gonzalez, N., Gurrian, E. G.,
Hanson, R. B., Iglesias, J., Krom, M., Lopez-Jamar, E., McClain, J., Pamatmat, M., Perez,
A., Rhoads, D. c., Rodriguez, R. M., Santiago, G., Tietjen, J., Westrich, J., and Windom,
H. L. , 1982, Coastal upwelling in the Rias Bajas, N.W. Spain: Contrasting the benthic
regimes of the Ria de Arosa and de Muros, J. Mar. Res. 40 (in press).
Thayer, C. W., 1979, Biological bulldozing and the evolution of marine benthic communities,
Science 203:458-461.
Thorson, G., 1957, Bottom communities, in: Treatise on Marine Ecology and Paleoecology,
Volume I: Ecology (J. W. Hedgpeth, ed.), pp. 461-534, Geological Society of America
Memoir 67, Geological Society of America, New York.
Ullman, W., 1975, Stabilization of the sediment-water interface by the presence of the
extracellular products of microorganisms, Senior thesis, Department of Geology and
Geophysics, Yale University, New Haven, Connecticut.
van Straaten, L. M. J. U., 1952, Biogene textures and the formation of shell beds in the Dutch
Wadden Sea. I-II, Koninkl. Nederl. Akad. Wet. Proc. Ser. B 55:500-516.
Vermeij, G. J., 1978, Biogeography and Adaptation Patterns of Marine Life, Harvard Uni-
versity Press, Cambridge, Massachusetts.
Wade, B., 1972, A description of a highly diverse soft-bottom community in Kingston Har-
bour, Jamaica, Mar. Biol. 13:57-69 .
• Webb, J. E., 1969, Biologically significant properties of submerged marine sands, Proc. R.
Soc. London Ser. B 174:355-402.
Webb, J. E., Djorges, D. J. , Gray, J. S. , Hessler, R. R., van Andel, Tj. H., Werner, F., Wolff, T.,
Zijlstra, J. J., and Rhoads, D. C., 1976, Organism-sediment relationships (Working Group
Reports-Group E), in: The Benthic Boundary Layer (1. N. McCave, ed.), pp. 273-295,
Plenum Press, New York.
Whitlatch, R. B., 1976, Seasonal changes in the community structure of the macrobenthos
inhabiting the intertidal sand and mud flats of Barnstable Harbor, Massachusetts, Ph.D.
dissertation, The University of Chicago, Chicago, Illinois.
Whitlatch, R. B., 1977, Seasonal changes in the community structure of the macrobenthos
inhabiting the intertidal sand and mud flats of Barnstable Harbor, Massachusetts, Biol.
Bull. 152:275-294.
Whitlatch, R. B., 1980, Patterns of resource utilization and coexistence in marine intertidal
deposit-feeding communities, J. Mar. Res. 38:743-765.
Wilson, W. H., 1979, Community structure and species diversity of the sedimentary reefs
constructed by Petaloproctus socialis (polychaeta: Maldanidae), J. Mar. Res . 37:623-
641.
Wolff, W. J., Sandee, A. J. J., and DeWolf, L., 1977, The development of a benthic ecosystem,
Hydrobiologia 52:107-115.
 Chapter 1
52
Woodin, S. A., 1976, Adult-larval interactions in dense in faunal assemblages: Patterns of
abundance, J. Mar. Res. 34:24-41.
Woodin, S. A., 1978, Refuges, disturbance, and community structure: A marine soft-bottom
example, Ecology 59:274-284.
Wooding, R. A., 1973, Drag due to regular arrays of roughness element geometry, Boundary
Layer Meteorol. 5:285-308.
Yingst, J. Y., 1978, Patterns of micro- and meiofaunal abundance in marine sediments,
measured with the adenosine triphosphate assay, Mar. Bioi. 47:41-54.
Yingst, J. Y., and Rhoads, D. C., 1978, Seafloor stability in central Long Island Sound. Part
II. Biological interactions and their potential importance for seafloor erodibility, in:
Estuarine Interactions (M. N. Wiley, ed.). pp. 245-260, Academic Press, New York.
Yingst, J. Y., and Rhoads, D. C., 1980, The role of bioturbation in the enhancement of
microbial turnover rates in marine sediments, in: Marine Benthic Dynamics (K. R.
Tenore and B. C. Coull, eds.). pp. 407-421, University of South Carolina Press, Columbia.
Young, D. K., 1971, Effects of infauna on the sediment and seston of a subtidal environment,
Vie Milieu (Supplement) 22:557-571.
Young, D. K., and Rhoads, D. C., 1971, Animal-sediment relations in Cape Cod Bay, Mas-
sachusetts. I. A transect study, Mar. Bioi. 11:242-254.
Young, R. A., 1977, Seaflume: A device for in-situ studies of threshold erosion velocity and
erosional behavior of undisturbed marine muds, Mar. Ceo1. 23:Mll-M18.
Young, R. A., and Southard, J. B., 1978, Erosion of fine-grained marine sediments: Sea-floor
and laboratory experiments, Ceo1. Soc. Am. Bul1. 89:663-672.
Zeitzschel, B., 1980, Sediment-water interactions in nutrient dynamics, in: Marine Benthic
Dynamics (K. R. Tenore and B. C. Coull, eds.). pp. 195-218, University of South Carolina
Press, Columbia.
Zobell, C. E., 1943, The effect of solid surfaces on bacterial activity, J. Bacterio1. 46:38-59.
Chapter 2
The Effects of Macrobenthos on
Chemical Properties of Marine
Sediment and Overlying Water
ROBERT C. ALLER

1. Introduction ................................................................................................................. 53
2. Diagenetic Reactions ................................................................................................... 54
3. Reactive Particle Redistribution .................................. .. .......... .. ................................. 56
3.1. Homogeneous Reworking ......................................... ........................................... 56
3.2. Selective Reworking ................................................ .. .......................................... 67
4. Solute Transport ............................................................................................ .............. 71
4.1. Apparent Diffusion .............................................................................................. 71
4.2. Biogenic Advection ............................................................................................. 75
4.3. Average Diffusion Geometry............................................................................... 77
5. Macrofaunal Influence on Sediment-Water Exchange Rates ................................... 83
6. Reaction Rates .............................................................................................................. 87
7. Chemistry of the Burrow Habitat ............................................................................... 89
8. Spatial and Temporal Patterns in Sediment Chemistry........................................... 92
9. Summary ...................................................................................................................... 93
Appendix: Solutions to Model Equations .................................................................. 94
References .................................................................................................................... 96

1. Introduction
The composition of any environment or object is determined by a partic-
ular balance between material transport processes and chemical reactions
within and around it. In the case of marine sedimentary deposits, the
dominant agents of mass transport are often large bottom-dwelling ani-
mals that move particles and fluids during feeding, burrowing, tube con-
struction, and irrigation. Such biogenic material transport has major direct

ROBERT C. ALLER • Department of the Geophysical Sciences, University of Chicago,

Chicago, Illinois 60637.

53
54 Chapter 2

and indirect effects on the composition of sediments and their overlying

waters. In this chapter I review some of what is presently known about
these effects, their implications for both chemical and biological prop-
erties of a deposit, and how they can be conceptualized in quantitative
models.

2. Diagenetic Reactions

Many of the most important reactions taking place in sediments are

associated with the decomposition of organic matter and other biogenic
components such as CaC0 3 and Si0 2 -nH 2 0 (Berner, 1976a). These re-
actions influence pH and oxidation-reduction potential and cause the
depletion or buildup of characteristic reactants or products in both the
fluid and solid phases of a deposit (Baas-Beckling et a1., 1960; Thorsten-
son, 1970; Ben-Yaakov, 1973; Goldhaber and Kaplan, 1974). Because of

Table I. Idealized Decomposition Reactions

1. Aerobic respiration

2. Nitrate reduction
5(CHzO)x(NH 3)y(H 3 P0 4 )z + 4xN0 "3 -> xCOz + 3xHzO
+ xHCO"3 + 2xN z + 5yNH3 + 5zH3P0 4

3. Manganese reduction
(CHzO)ANH3)y(H 3 P0 4 )z + 2xMnOz + 3xCOz + xHzO -> 4xHC0"3
+ 2xMnz + + yNH3 + zH 3P0 4

4. Iron reduction
(CHzO)ANH3)y(H 3P0 4)Z + 4xFe(OHh + 7xCOz -> 8xHCO"
+ 3xHzO + 4xFe z + + yNH3 + zH 3P0 4

5. Sulfate reduction
2(CHzO)x(NH 3)y(H3P0 4 )z + xSO~ - -> 2xHC0 "3 + xHzS + yNH3 + 2zH3P0 4

6. Methane production
2(CHzO)ANH3)y(H 3 PO.)z -> xCO z + xCH 4 + 2yNH3 + 2zH3PO.

7. Fermentation (generalized)
12(CHzOlx(NH 3)y(H 3PO.)z -> xCH 3 CHzCOOH + xCH 3COOH
+ 2xCH3CHzOH + 3xCOz + xHz + 12yNH3 + 12zH3P0 4
Effects of Macrobenthos on Marine Sediment Chemical Properties 55

A B c

E
c
P->·_'-"-"--'-

"0
Q)
(j)

.... ............. .......

................. ......
...
.................. ......
.. .................

Figure 1. (AJ Classically assumed vertical zonation of electron acceptor use in sediments.
(BJ Reaction zonation around irrigated burrow microenvironment. (C) Reaction geometries
associated with fecal pellet microenvironments.

the central role of decomposition reactions in controlling the diagenesis

of sediments as well as their importance in biogeochemical cycles, a
subset of such reactions will be used here to illustrate the effects of ma-
crofaunal activity on sediment chemistry.
Idealized representations of major organic matter decomposition
pathways are listed in Table I (after Richards, 1965; Stumm and Morgan,
1970; Claypool and Kaplan, 1974; Froelich et a1., 1979). Stoichiometries
are variable but tend overall to x ~ 106, Y ~ 16, and z ~ 1 (Redfield, 1934;
Grill and Richards, 1964; Sholkovitz, 1973). The metabolic energy yields
available to an organism for each mole of carbon utilized during decom-
position depend in part on the electron acceptor involved and decrease
in the order O 2 > N0 3 2: Mn0 2 > FeOOH > SO~ - > CO 2 , This fact, to-
gether with the competitive exclusion principle of ecology, implies that
reactions (1)-(6) should occur successively in association with changes
in the types of bacteria that mediate them (Stumm and Morgan, 1970;
Claypool and Kaplan, 1974). Observations have generally supported these
inferences (Froelich et a1., 1979). In contrast, fermentation reactions like
reaction (7) are apparently temporally and spatially ubiquitous.
The concept of biogeochemical successions, together with the view
that sedimentary deposits are laterally homogeneous and essentially one-
dimensional bodies, has led to the dogma that decomposition reactions
are vertically stratified below the sediment-water interface (Fig. 1A). This
56 Chapter 2

dogma has guided both sampling strategies and interpretations of sedi-

ment properties.
Macrobenthos influence this hypothetical reaction distribution and
its maintenance in four major ways:

1. Material is translocated continuously between reaction zones dur-

ing feeding, burrowing, and tube construction.
2. Burrow and fecal pellet formation alters reaction and solute dif-
fusion geometries, creating a mosaic of biogeochemical microen-
vironments rather than a vertically stratified distribution (Figs.
1B,C).
3. New reactive organic substrates in the form of mucus secretions
may be introduced into the deposit independent of sedimentation
processes.
4. Feeding and mechanical disturbance may influence microbial
populations that mediate reactions.

Macrobenthos can also directly alter sedimentary minerals during passage

of material through their guts, but this class of effects is not considered
here (Dapples, 1942; Pryor, 1975; 1. S. Hammond, 1981).
The extent to which these effects are realized in any given case de-
pends on the functional groups of animals present, their abundance, cer-
tain taxonomic peculiarities, and the size of individuals. Functional
groups are defined on the basis of particular combinations of feeding type,
life habit, and mobility, i.e., the specific ways in which animals interact
with the seafloor (Brenchley, 1978). In the subsequent discussion em-
phasis will be placed on deposit-feeding benthos of varied life habits and
mobilities.

3. Reactive Particle Redistribution

3.1. Homogeneous Reworking

Organic and other particles are constantly supplied to the sea floor,
where they accumulate and undergo reactions such as those listed in
Table I. Within a single type of sedimentary deposit and in the absence
of disturbance by macrobenthos, spatial variation in the distribution of
any physical or chemical property is predominantly in the vertical di-
mension and is determined by a balance between the rate of upward
accretion of material and chemical reactions. For example, the concen-
tration, C, of a reactive organic matter substrate subject to microbial deg-
Effects of Macrobenthos on Marine Sediment Chemical Properties 57

radation is formally given by (after Berner, 1980):

ac a,
- = -- (we) + R (1)
at ax
where C is the mass per volume of sediment; t is time; x is the vertical
dimension, with its origin at the sediment-water interface, positive into
the deposit; w is the sedimentation or accretion rate; and R is the reaction
rate. In practice, compaction is usually ignored and a steady-state distri-
bution assumed, so that:

(2)

This equation can be solved by defining an appropriate reaction rate. As

a first approximation, organic matter decomposition is often assumed to
follow first-order kinetics with respect to reactant concentration, so that:

R = -kG (3)

where k is the first-order reaction rate constant. Models employing this

approximation have been successful in many cases, particularly for de-
scribing SO~ - reduction. Complications are that k is an approximation
to more complex Michaelis-Menton kinetics for multiple substrates, the
value of k is temperature-dependent, and over large depth intervals k can
vary significantly (J0rgenson, 1978; Berner, 1980).
The simplest influence macrobenthic activity has on this otherwise
stratightforward set of processes is the nonselective or homogeneous mix-
ing of sedimentary particles during feeding, burrowing, and construction
activities. Particles are biologically transported within the deposit at a
different rate and style than net accretion, w. There is no strict physical
analogue for such mixing even in the simple nonselective case.
On short time scales particles can move in streams around deposit-
feeding individuals (Fig. 2A). If only one kind of animal of a given size
is present, particle transport often occurs as advective loops between the
sediment-water interface and the depth of feeding (Rhoads, 1974; Aller
and Dodge, 1974; Amiard-Triquet, 1974; Fisher et aI., 1980). Particles are
also subject to local agitation and dispersion during burrowing or ma-
nipulation by appendages and mouth parts (Cullen, 1973; Fisher et aI.,
1980). Even with only one type of animal, natural variation in individual
size either within a given population or with time following larval set-
tlement results in various loop sizes (Whitlatch, 1974) (Fig. 2B). The
58 Chapter 2

A
_ n,n M B
Of' nn M
) ~.
· .. (.HJ · : , ~..

. .
. . . ...y J.· UU·;·
. .
UU· \JL) ...
... . ·.UU

]....f:
c o E

~I- .
. . . . . + . :
. ,

. . , ,

.. - .
. .· u~r ..

• •
Figure 2. (A) Hypothetical transport paths of particles around individual subsurface deposit
feeders of uniform size. (B) Multiple transport loop scales around organisms of varied size.
(C) Particle transport loops are subject to lateral motion depending on mobility of deposit
feeders. (D) Infilling of biogenic cavities results in particle transport independent of feeding
activity. (E) Vertical dimension of feeding loops goes to zero in case of surface deposit
feeders and in some cases may result in injection below the interface of surface-derived fecal
material.

lateral mobility and patchiness of infauna in addition to instantaneous

advective events such as burrow infilling further complicate particle mo-
tions (Hanor and Marshall, 1971; Benninger et a1., 1979) (Figs. 2C,D).
In the face of such complexity, it is often assumed that when a suf-
ficient number and variety of small-scale mixing events occur within a
sediment interval they can be treated as random particle eddies and quan-
tified in a mixing coefficient analogous to the physical eddy diffusion
coefficient in a fluid (Goldberg and Koide, 1962; Guinasso and Schink,
1975). A major inconsistency in this analogy is that the path length of
particle motion is often both specific in orientation and large in size scale
with respect to chemical or physical property gradients in the sediment.
For instance, reduced sulfide-rich material from depth may be placed
directly at the oxidized sediment-water interface without mixing with
material in between. Despite such conceptual inaccuracies, some insight
into the consequences of reworking as compared to simple sedimentation
can be gained from the analogy.
Effects of Macrobenthos on Marine Sediment Chemical Properties 59

Ignoring lateral variation in sediment properties and vertical com-

paction, the distribution of a sediment component in the bioturbated zone
is now described by:

ax ] _ (at)
atat = ~ax [DB (at) ax w +R (4)

where DB is the particle mixing coefficient. All other variables are the
same as for equation (1).
Because animal abundance and activity vary with depth, so does DB
(Mare, 1942; Myers, 1977; Guinasso and Schink, 1975). This variation can
be accommodated either by assigning a continuous functional form to DB,
such as an exponential decrease with depth (Schink and Guinasso, 1977;
Santschi et al., 1980; Olsen et al., 1981), or by assuming that DB is constant
over one or more empirically defined sediment intervals (Guinasso and
Schink, 1975; Benninger et a1., 1979; Nittrouer et a1., 1979) (Fig. 3). Al-
though this latter approach may appear completely unrealistic, reworking
is sometimes dominated by one or two species feeding at well-defined
horizons or burrowing within restricted intervals (Rhoads, 1974). The
kinds of data available to evaluate values of DB are often limited enough
that in any case only an average coefficient acting over an interval can
be determined.
Because both large and small "particle eddies" occur owing to var-
iation in size and functional groups of animals, it could also be reasoned
that DB may actually increase with depth in much the way that eddy
diffusion coefficients have length scale dependence in oceanic dispersion
processes (Okubo, 1971). This possibility is not being considered here in
part because of the imperfections in the diffusion analogy to bioturbation

A B c
DB DB

)(
LI
S- - LI ------
~l
o
L2
_S-
L3

Figure 3. Commonly assumed depth dependences of the particle transport coefficient DB

(AJ Exponentially decreasing magnitude with depth. (Bl Multiple step-function or composite
layer model. (Cl Single mixed layer overlying uninhabited zone.
60 Chapter 2

Table II. Isotopes Commonly Used in

Evaluation of Particle Reworking Rates
Isotope Half-life Input type
234Th 24.1 days Continuous
7Be 44 days Continuous
228Th 1.9 yr Continuous
210Pb 22 yr Continuous
137CS 30.1 yr Pulse
32Si 275 yr Continuous
He 5730 yr Continuous
239.24OpU 2.4 X 10 4 yr, 6540 yr Pulse

and in part because experimental data to test such dependence are not
available.
Values of DB are estimated from the distribution within a deposit of
particle-associated components having both well-defined inputs to the
deposit and known reaction rates. These may be stable tracers such as
colored particles, tektites, or foraminifera, all of which enter a deposit in
pulses, or naturally occurring radioactive tracers that are supplied con-
tinuously during deposition and decay away. Man-made tracers such as
the several isotopes of Pu and Cs are also used. The advantage of the
naturally occurring radioactive tracers is that continual supply often al-
lows assumption of a steady-state distribution (ac/at = 0) and easy eval-
uation of equations like (4).
The most commonly used isotopes in the evaluation of particle trans-
port, biogenic or otherwise, are listed in Table II. Because of their varied
half-lives, these isotopes are useful over a range of different depth and
time scales depending on biological activity and net sedimentation rate.
For example, consider the simple case of an inhabited and uniformly
mixed zone of sediment overlying an uninhabited interval, as depicted
in Fig. 3C. The steady-state vertical distribution of a continuously sup-
plied isotope, i, irreversibly adsorbed to particles is, for zone 1 (0 :S X
:S L):

-aA i -_ 0 -_ DB
at
(a-axAi)
Z
- -
z
W (aAi)
-
ax
- AiAi (5a)

and, for zone 2 (x 2: L):

o -w (-aA)
ax' - A·A
1 1
(5b)
Effects of Macrobenthos on Marine Sediment Chemical Properties 61

with the following boundary conditions:

1. x = 0, Jo = DB (aAdax) - WAi

2. x = L, AiZonel = Aizon e 2

3. x = L, aA)ax = 0

4.x~oo,Ai=0

where Ai is the activity of isotope i; hi is the decay constant of isotope i;

and Jo is the constant flux of constituent Ai supplied to the sediment at
x = O. These boundary conditions require continuity of material fluxes
between vertical zones and eventual decay of the tracer at depth. Neither
compaction nor loss of volume during reaction is considered. The solu-
tions to these equations are given in the Appendix for easy reference.
The kinds of vertical property distributions predicted by such as-
sumptions are illustrated in Fig. 4 for 234Th and 21OPb. Cases in which DB

234 Th (dpmlcc)

r.
0 2 4 0 20 40 0 200 400
,
4 0.4
, 0 .02 ,,- - -
E 8 De = 15.8 0 .8
~ De = 0 .158 0 .04 " De = 0
.t: (cm2{yr)
a.
Q)
12 1. 2 0 .06
0 16 1.6
20 2.0

210 Pb
(dpm/cc)
0 2 4 0 4

, .- -
8 2 6
-- 10

4 ,
E , I
I
I
/

~ 8
-: I
I

..c: ,- J I

a.
Q)
12
,,
.- .- I
I

0
16
,/ De = 15.8 De = 0 .158 De = 0
,
20
Figure 4. Predicted vertical profiles for 234Th (upper row) and 21uPb (lower row) assuming
single layer mixed at rates DB = 15.8 cm 2/yr (0.5 x 10- 6 cm2/sec); 0.158 cm 2/yr (0.5 x
10 - 8 cm2/sec); and 0 cm 2/yr with a sedimentation rate w = 0.1 cm/yr. The flux of 234Th and
21°Pb to the surface is fixed at 40 dpm/cm 2 per yr and 1 dpm/cm 2 per yr respectively. Note
changes in activity and depth scales in the three cases. These nuclides have decay time
scales in the same range as reactive organic matter.
 Table III. Representative Particle Reworking Coefficients from Marine Environments
Water Mixed layer Sedimentation
depth Sediment interval rate
Location Environment (m) type (cm) (cm/yr) DB (cm2/yr) Method Reference

Long Island Estuarine 8-35 Clay-silt-sand 0-5 <0.1-0.3 19 ± 16 (0.3-50) 234Th (1)
Sound 13-15 Clay-silt 0-3 <0.1-0.3 3-9 ?Be (2)
15 Silt-clay 3-15 <0.1-0.3 0.6-0.9 210Pb, 239, 24O pU (3)
Narragansett Bay Estuarine 10 Silt-clay 0-9 0.01 5-32 234Th, 210Pb, (4)
239.24OpU (4)
10-30 0.01 0.3-2.5 234Th, 21OPb,
239.24OpU
Hudson River Estuarine Silt-clay 0-8 1-3 0.25-1 137CS, 239. 24O pU (5)
New York Bight Inner shelf 23 Silt-sand 0-5 16 234Th (6)
Midshelf 69-87 Silt-clay 0-8 0.01-0.14 3.8-11 234Th, 210Pb, 239. (4)
240pU
8-40 0.01-0.14 1.7-2 234Th, 210Pb, 239. (4)
240pU
Slope 850-1800 Silt-clay 0-15 0.01-0.14 0.01-1.3 234Th, 21OPb, (4)
239.24OpU
Northwestern Midshelf 20-200 Silt 0-10 -0.3 10 ± 12 (0.75-44) 210Pb (7)
Pacific,
Washington
 Outer shelf 100-200 Silt-sand 0-10 - 0.2 5.5 ± 5.4 (0.28-19) 210Pb (7)
Slope 200-2000 Clay-silt-sand 0-10 -0.1 2.7 ± 3.6 (0.25-9.8) 21°Pb (7)
Southeastern Shelf 42 Silt-clay 0.5 30 234Th (8)
Atlantic.
Amazon
Mediterranean Pelagic 1000 0-12 0.38 239. 24OpU (9)
Sea
North Atlantic Pelagic 2705 Calcareous ooze 0-10 0.003 0.19 210Pb (10)
4810 0-8 0.22 239. 24OpU (9)
1410 0-9 0.25 239. 24OpU (9)
Southwestern Pelagic 4910 Calcareous ooze 0.063 21°Pb (10)
Atlantic 4920 0-4 0.14 Pu (9)
1345 0-6 0 .10 Pu (9)
Antarctic Pelagic 4080-4730 Siliceous ooze 0-8 0 .0004-0 .001 0.03-0.25 21°Pb (10)
Northern Pacific. Pelagic 4640-5050 Red clay. 0-7 0.00014-0.00031 0.25-0.44 21°Pb (10)
equatorial siliceous ooze
Western Pacific. Pelagic 1598 Calcareous ooze 0-8 0.001 0.12 21°Pb (11)
equatorial

a References: (1) Aller et a1. (1980). (2) Krishnaswami et a1. (1980). (3) Benninger et a1. (1979). (4) Santschi et a1. (1980). (5) Olsen et al. (1981). (6) Cochran and Aller
(1979) . (7) Carpenter et al. (1982). (8) DeMaster et a1. (1980). (9) Guinasso and Schink (1975). (10) Turekian et a1. (1978). (11) Peng et a1. (1979).
64 Chapter 2

is fixed at either 0.5 x 10- 5 cm 2/sec (15.8 cm2/yr), 0.5 x 10- 8 cm 2/sec
(0.158 cm2/yr), or 0 are shown for comparison. The nonzero values of DB,
and the other values chosen (w = 0.1 cm/yr, L = 10 cm, ITh = 40 dpm/
cm 2 per yr, IPb = 1 dpm/cm 2 per yr), are of realistic magnitudes for estuarine
and shelf sediments (Table III) (Aller et a1., 1980; Benninger et a1.,
1979; Nittrouer et a1., 1979; Carpenter et a1., 1982). When L is greater than
the depth at which A j becomes small, as in the case of 234Th, its exact
value does not affect the tracer distribution.
These examples illustrate that, in the range of sedimentation rates
usually encountered, biogenic reworking and the decay rate of an isotope
largely determine both the penetration patterns of a reactive component
in a deposit and the absolute quantity of a constituent decomposing at
any given depth. Sedimentation rate may be ignored entirely in any given
case if w2 «:; 4D B Aj. In the present example reactant supply or flux at the
sediment-water interface is held constant, so that only the vertical dis-
tribution of reaction rate and not the integrated quantity of decomposing
material is affected. The concentration at the sediment-water interface
is thus lowered by reworking while that at depth undergoes a relative
increase.
In most sedimentary basins, physical processes such as resuspension
and current transport act to buffer the interface concentration of a sedi-
ment component by lateral exchange and homogenization of surface ma-
terial. As a result, spatial variability in biogenic reworking together with
resuspension within a basin can cause a transfer of reactive particles from
regions of relatively low biogenic reworking into regions of high reworking
through a process of particle exchange during lateral and vertical mixing
(Aller and Cochran, 1976). Exchange increases the total amount, or in-
ventory, of a reactant in a deposit in areas of high biological activity. This
effect is shown quantitatively in Fig. 5, in which the interface boundary
condition (1) in model equations (5) has been changed to a fixed concen-
tration, x, = 0, A = Ao, to simulate surface buffering by resuspension.
The 234Th inventory increases with increased values of DB. In the real
world, the actual effect of reworking on surface concentrations (Fig. 4),
relative vertical distributions, and inventory (Fig. 5) will vary between
the two extremes illustrated by the constant flux and constant surface
concentration models (Figs. 4 and 5). For example, data from Long Island
Sound, an estuarine basin along northeast North America, indicate a gen-
eral increase in 234Th inventories with DB, although there is considerable
variation owing to localized peculiarities in supply, physical transport,
and biological reworking styles (Fig. 6A). 239, 240 pU inventories have also
been shown to increase in areas of increased biological reworking, as
reflected by depth of Pu penetration (Santschi et a1., 1980) (Fig. 6B).
Because the decomposition of organic matter can be approximately
Effects of Macrobenthos on Marine Sediment Chemical Properties 65

234 Th (dpm/cc)

0 2 3 4 0 2 3 4 0 2 3 4
---- - --
E ~

2 I
~
~

2 /
~ I

D : 15.8 1.58 0 .158

-
I
.£:. 4 , I B DB : DB
a. I = 4.9
Q) I 1.6 I 0.4
0 6
Figure 5. Predicted vertical distributions of 234Th assuming a single mixed layer (Fig. 3C)
and varied rates of mixing. In this case the surface activity is assumed to be buffered by
resuspension and supply from surrounding areas at a value of 4 dpm/cc. As a result the
inventory (1) varies with mixing rate from 4.9 dpm/cm 2 to 0.4 dpm/cm 2 when DB varies from
15.8 to 0.158 cm 2 /yr.

B
A

1.0 /0
10
'"E 0
01
I
...... 1 0
E
'"E 0 8 ~
a.
I
c. l
/
...... ,,- " /
E
•
,,- >. I
a. ,,-
::s
6 ,,- "
,,-
,,-
E
c: I
/
10
C.
,,-
...>. ,,-
Q)
> od°
0
,,-
• c: 0.5 /

."-
,,- " I
C
Q)
• q. c.

• ••
> I
c: 4
• "
,.-
,,-
c..
::J
°/ 6

• .- .,.
0 I
.s;:;
I-
,,-
,.-
N. /
cI ° ••
of
'"
2
•• -- " ,,- en
'" I
/
tIJt A --

.- -- --
N
,/
N /
I __ ~
(-
0
0 10 20 30
0 .5 1.0 1. 5 2.0
L (cm )
DB ( 166 cm 2 /sec)
Penetrat ion Depth
Vertica l Mobili ty
Figure 6. (A) 234Th inventories in sediments from Long Island Sound as a function of DB.
Inventory increases with increases in mixing rate, illustrating capture of resuspended par-
ticles during reworking. After Aller ef 01. (1980). (B) Variation in 239. 240 pU inventories as a
function of Pu penetration depth into sediment. Penetration depth varies with intensity of
biogenic or physical reworking. Nearshore stations (depth <100 m): 0, Buzzards Bay, Gay-
head (from Livingston and Bowen, 1979); 6, Narragansett Bay; 0, New York Bight. Offshore
stations (depth >100 m): e, Wilkinson Basin, Georges Bank (from Livingston and Bowen,
1979); . , New York Bight; \1, Atlantic deep sea (from Noshkin and Bowen, 1973). After
Santschi ef 01. (1980).
66 Chapter 2

described by first-order decay kinetics [equation (3)), the model distri-

butions derived for particle-associated nuclides are directly applicable to
the distribution of metabolizable organic matter or other reactive material
in a deposit. The effective first-order decomposition rate constants, k,
typical of organic components in surface sediments are in fact comparable
to Ai values for 234Th and 210Pb (e.g., Grill and Richards, 1964; Berner,
1981, J0rgenson, 1977a, 1978; Nixon et al., 1980). Disregarding the effects
of micro environments and solute transport, which will be discussed later,
these considerations and previous qualitative observations by many in-
vestigators (Rhoads, 1974) indicate that a larger proportion of organic
matter is decomposed at depth in a biogenic ally reworked deposit than
in unreworked deposits of comparable material supply. It is likely that,
as a result, a correspondingly larger proportion is remineralized by an-
aerobic rather than aerobic pathways. This can be seen qualitatively by
simply comparing the downward flux, h, of reactive material out of the
bioturbated zone at x = L in the mixed and unmixed cases: h = wA(L),
where A(L) mixed> A(L) unmixed. Reduced products of anaerobic de-
composition may not permanently build up in a deposit, however, because
particles can be subject to reoxidation before final burial.
Organic matter subduction during reworking together with resuspen-
sion must result in an increased inventory of reactive organic matter in
areas inhabited by deposit feeders (Fig. 7) through the same combination
of physical and biological reworking processes outlined previously for
234Th and Pu. Because deposit feeders utilize detritus and associated
bacteria for food, the resulting increase in food supply may itself en-
courage increases in animal size and abundance, further stimulating lo-
calized capture of resuspended reactive material. Biogenically mixed
patches should therefore act as biogeochemical hotspots in sedimentary
basins for this reason alone, other factors being equal. This also raises the
interesting possibility of competition for food between laterally separated
patches of deposit feeders that might not otherwise interact. The intensity
of such competition would be determined by both the nature of resus-
pension in the water column and the reworking styles of the benthic
infauna.
Average values of DB have been determined in a variety of nearshore,
shelf, and deep-sea environments. Some of these values, the techniques
used, and the depth interval over which they were determined are listed
in Table III. Although there can be substantial variability within any
region, or with depth in a deposit, in general there is a decrease in bio-
logical reworking from shallow (DB ~ 10 - 6 cm2/sec) to deep-sea (DB ~
10 - Bcm 2/sec) environments. This is consistent with the decreased met-
abolic activity generally observed in deeper water (Smith, 1978; Zeitschel,
1980). High DBs calculated for slope canyon deposits probably reflect
Effects of Macrobenthos on Marine Sediment Chemical Properties 67

Resuspension Lateral Exchange

Biogenic Subduc ti on
- Vert ica l Exchan e

Figure 7. Sketch of how vertical exchange of particles within the sediment during biogenic
reworking and lateral exchange of particles during resuspension can result in capture of a
reactive component into biogenically reworked patches at the expense of surrounding areas.
No net sedimentation takes place during exchange. After Aller and Cochran (1976).

physical rather than biological transport (Carpenter et al., 1982). In almost

all cases, biogenic particle reworking is on the average slower than solute
transport by molecular diffusion. Particle reworking per se is therefore
unlikely to control solute transport directly in most deposits. Indirect
effects are discussed later in this chapter.

3.2. Selective Reworking

Although the concept of homogeneous, random mixing as reflected

in the mixing coefficient, DB, is of great practical application and is useful
for quantifying and illustrating several fundamental effects of reworking,
it can be a misleading description of the processes taking place. During
feeding and burrow construction, animals are capable of selecting parti-
cles on the basis of particle position in the sediment, size, shape, surface
texture, and density (Dapples, 1942; Fager, 1964; Rhoads, 1974; Fenchel
et a1., 1975; Self and Jumars, 1978; Jumars et a1., 1981; Khripounoff and
Sibuet, 1980; Levinton, 1980). This selection may result in lateral and
vertical segregation of different classes of particles depending on the par-
ticle sizes available, the functional kinds of animals present, and their
age-size distributions.
Because many chemical properties of sediments correlate with grain
size, any spatial segregation of particle types will result in a corresponding
variability in chemical composition and reaction rates (Fig. 8). For ex-
68 Chapter 2

20

~
E
a.
:3 15
~
.s: ..
-0
<t
10
.c
a..
0
N

'"u0
-
5
~

:> Figure 8. Many reactive metals and organic com-

(f)
ponents associate preferentially with fine-grained
~ Sa nd ~ ~ Silt ---+
sediments. In this case, 210Pb activity in surface
sediments is shown to increase as sediment par-
2 3 4 5 6 tide size decreases in Washington Shelf sedi-
Median Grain S ize ments off northwest North America. After Nit-
(4) Units) trouer et a1. (1979).

ample, in the case of aerobic reaction rates, oxygen consumption by bac-

teria on particles is directly proportional to particle surface area (Hargrave,
1972; Hargrave and Phillips, 1977). Thus, the oxygen flux to any surface
is approximately constant (Fig. 9A). Because of differences in surface area,
fine-grained sediments should be exponentially more reactive on a mass
or volume basis than coarser-grained material when surface-dependent
processes such as remineralization reactions are involved (Figure 9B).
Reaction rate distribution can therefore be altered by selective feeding in
ways not accounted for by a general nonselective mixing coefficient DB.
In some cases where relatively fine material is preferentially ingested
at depth and defecated at the sediment-water interface, a biogenic stra-
tigraphy or graded bedding is formed (van Straaten, 1952; Rhoads and
Stanley, 1965; Cadee, 1976; Baumfalk, 1979) (Fig. 10). As a result, reaction
rates and properties associated with particle surface area are increased
near the sediment-water interface. Because of proximity of solute release
to overlying water, this should lower the buildup of metabolites in the
sediment, as discussed later. It will also accentuate interaction with over-
lying water by resuspension of small partcles known to be important in
adsorption and scavenging reactions in the water column (Aston and
Chester, 1973; Turekian et a1., 1980).
Other groups of deposit feeders may egest relatively coarse particles
at the interface or defecate fine material preferentially at depth (Whitlatch,
1974; Powell, 1977). Small particles placed at the sediment surface by
 A
0 .1 0

0 .08

1:
......
C>
E 0 .06

""0
Q)

E 0 .04
::0
VI
C
0
0

0 .02
0 '"

0 50 100 150 200

Surface Area (cm 2 )
B

600
,
,
I
I

\
I
I
C. 500
;;-
E ,,
S
,
I

400 I

VI
VI
C
~ ,,
..... 300 ,
,
C ,,\

«
~ ,
I

200

-
Q)
0 I
C I
.... \
::0 ,,

-
I
(/) 100 ,,
--+,-
,
--
Cloy _
Sill Sand

8 7 6 5 4 3 2 0 -I -2
Groin Size (~ Units)
Figure 9. Reaction rates are often a function of grain size or sediment surface area. (A) Oz
consumption by aerobic bacteria on particles and other oxidation reactions as a function of
surface area. The constant slope of the relation demonstrates that the oxygen flux to a surface
is constant. After Hargrave and Phillips (1977). (B) The theoretical relation between surface
area per mass of sediment and particle diameter or grain size. Particles are assumed to be
simple spheres of density 2.5 g/cc. Actual clay particles have considerably greater surface
areas per mass, ranging from -10-30 m 2/g (Kaolinite) to 40-100 mZ/g (Smectite). Fine-
grained material is therefore much more reactive than a comparable mass of coarse material
with respect to surface-dependent processes such as organic matter decomposition.
70 Chapter 2

A
B

0/0 Grain Size

T ""
20 40 60 80
0
'/ ' / / / // /
....ll.J
. . ;1 .. ' . 5 7.1
', ' fl . .
, . ' II ', ' • ..LL.L.LJl
E 10 ":':'1

ii0~~il~;t!~;r~~;'TIiZ ~

.s::
a.
15 '".l
'/

0
Q)
20 ':":';L

25

30
"• <80 urn

> 860
.
80 - 250
250 - 860

Figure 10, Segregation of particle sizes often occurs during feeding. (A) Subsurface feeding
by the lugworm Arenicola and surface defecation of ingested material can result in pockets
of coarse material at depth. After Rijken (1979). (B) Average verti cal grain size distribution
resulting from feeding by Arenicola in tidal flats of Dutch Wadden Sea. After Baumfalk
(1979).

one species inhabiting an area can also be preferentially subducted by

another or nonspecifically infill vacated burrows, For instance, selective
subduction occurs in some areas of Barnstable Harbor, Massachusetts,
inhabited by both the maldanid polychaete Clymenella torquata and the
terebellid Amphitrite ornata (Aller and Yingst, 1978) (Fig, 11), Amphitrite

"" ,-,'

'-, .... , "', ...

f
.,'

"':

. ,-..
',' ,"

,.,
- ..
.. .....:. ..•.':..:&~. ~ ---:'-1
'~. '
~ ' . ~ .:
:--:-.. ..
•
-/ -:--- - .,--------:-'

'

Figure 11. Small particles brought to the interface by th e feeding activiti es of some organisms
may be subducted by others during both burrow construction and fe eding. In this case the
interaction between the subsurface feeder Clymen ella torquata and th e surface deposit feeder
Amphitrite ornata results in biogeni c graded bedding interrupted by fine-grained shafts at
depth in areas of Barnstable Harbor, Massachusetts .
Effects of Macrobenthos on Marine Sediment Chemical Properties 71

builds its tubewall from small particles defecated at the interface by

Clymenella and subducts them as a distinct shaft. The specific micro en-
vironmental geometry of particle redistribution, and therefore reaction
rates, is important in determining solute buildup patterns and sediment
solute exchange with overlying water.

4. Solute Transport

The solutes consumed or generated in sediments by diagenetic re-

actions are subject to transport by molecular diffusion, advection, and
mixing processes caused by physical or biological activity. In the absence
of biological or physical disturbances, solute distributions can be quan-
titatively described by transport-reaction equations conceptually equiv-
alent to those used previously for the solid phase:

aC = Ds
at
(aaxC) _ (ac)
2

2
w
ax
+R (6)

where C is the solute concentration in pore water; Ds is the bulk sediment

diffusion coefficient; and R is the solute reaction rate per liter of pore
water. Other variables are defined as for equation (1). Compaction and
adsorption reactions are again ignored for ease of illustration. As before
it is assumed that variations in physical and chemical properties are
predominantly vertical and can be modeled by one-dimensional equa-
tions. Solutes diffuse along concentration gradients maintained by the
composition of an upper well-stirred water column reservoir and reactions
within the sediment pile. No independent pore water flow is allowed, so
that advective transport is equivalent to net sedimentation, w, and steady-
state conditions are usually assumed. Models of this type have been suc-
cessfully applied in a wide range of environments , particularly anoxic
basins and sediments below the bioturbated zone (Lerman, 1979; Berner,
1980).

4.1. Apparent Diffusion

The construction and irrigation of burrows by macroinfauna com-

plicate this otherwise straightforward one-dimensional conceptualization
of transport-reaction. In the bioturbated zone solutes can diffuse along
lateral concentration gradients into biogenic cavities as well as vertically
toward the sediment-water interface. Solute diffusion geometry is there-
fore three-dimensional and time-dependent.
Three classes of models have been proposed to quantify these pro-
72 Chapter 2

cesses. One class retains the one-dimensional formalism and assumes that
macrofaunal activities can be accounted for in an increased apparent
diffusion coefficient of solutres (D. E. Hammond et aI., 1975; Vanderborght
et aI., 1977; Goldhaber et aI., 1977; Aller, 1978). Like the particle mixing
coefficient, DB, this effective solute diffusion coefficient, D~, is analogous
to eddy transport coefficients used to describe physical mixing in the
overlying water column. In the strictest analogy, a single D~ should there-
fore apply to all solutes over a defined sediment depth interval, transport
in the underlying sediment being determined by molecular diffusion.
Because apparent coefficients are determined empirically, they include
the effects of coupling between particle reworking and solute transport
that result from reversible adsorption (Schink and Guinasso, 1978). Re-
ported values of D~ obtained from modeling pore water profiles range
from ~1O-5 to ~1O-4 cmz/sec (Vanderborght et al., 1977; Goldhaber et
al., 1977; Aller, 1978; Korosec, 1979; Filipek and Owen, 1980). These are
10-100 times higher than typical bulk sediment molecular diffusion coef-
ficients for the same deposits; Dss vary from ~1O-6 to ~1O-5 cmz/sec
depending on sediment structure and temperature (Li and Gregory, 1974).
The effect of a mixed layer on solute depletion or buildup is illus-
trated for NHt and SO~- (produced or consumed during sulfate reduc-
tion) (Table I) as follows. In this example, the composite layer model
equations used are, for Zone I (mixed) (0 s; x s; Ll):

(7a)

and, for Zone II (unmixed) (Ll s; x s; Lz ):

(7b)

where Ro , R1 , and ex are constants and with the following boundary

conditions:

1. x = 0, C = Gr
2. D 1 (aC 1 Iax) = Dz(aCzlax), x = Ll
3. C1 = Cz , X = Ll
4. aCzlax = 0, x = Lz

Compaction is ignored, advection is taken as unimportant relative to dif-

fusion over the depth interval considered, and a steady state is assumed
Effects of Macrobenthos on Marine Sediment Chemical Properties 73

for simplicity. Adsorption terms cancel at steady state (Berner, 1976b).

The solution is given in the Appendix for easy reference. Boundary con-
dition (4) requires an impermeable layer or saturation condition at depth
L2, as would occur in either an experimental aquarium setup or deep in
a deposit.
Reaction rates typical of terrigenous sediments from Mud Bay, South
Carolina, at T = 29°C are used in the calculations (Aller, 1980a). These
are RS04 = - 0.383exp( - 0.36x) - 0.061 and RNH 4' = 0.267exp( - 0.61x)
+ 0.0081 mM/day. Estimates of molecular diffusion coefficients at T =
29°C are made from the relation Ds - <l>2D where <I> is porosity and D is
the free solution diffusion coefficient (Li and Gregory, 1974; Lerman,
1978; Ullman and Aller, 1982). No viscosity correction for seawater den-
sity is made. Model predicted profiles for Gj' = 0.2 f-LM, L1 = 5 and 10
cm, D2(SO~-) = 0.717, and D2 (NHt) = 1.33 cm 2 /day (<I> = 0.851) are
plotted in Fig. 12. These are calculated to show the effect on a constituent
for variations in apparent diffusion coefficients relative to the molecular
diffusion coefficient of that ion. The values of D1 in the examples are
therefore not the same for both NHt and SO~ - . Identical values would

A B c
504 (mM )
5°4 (mM) RS (mM/day)

.
4 8. 12 16 4 8 12 16 01 0 .3 0 .5
o
0." • ,,
.
0, - 0, - Dz
4 ,,I' ..:

0, = 30z .:

0, - 10Dz
20

NH'; (rtlM) NH~ (mM) RN (mM/day)

1.0 2 .0 1.0 2 .0
0
\\:~"'''''''' '~- . 0 .2 0.4

4 :
,,
,,, 0, = Oz 0, • O2
, , I

E ,,\ , ,
.3 8 , ,
. ., I

. .,.
, ,,
12 ,
'. ,
a. '. .
~

C1) 16
0 ':0,>30.,, \ ~
•
~ DI=302 ~
I ,
20 0, = 10D z 0, = 10Dz

Figure 12. Vertical pore water profiles predicted for SO~- and NHt for mixed layers of 5
cm (A) and 10 cm (8) with varied ratios of apparent diffusion coefficient (D,) to molecular
diffusion coefficients (D 2 ) in underlying sediment. The assumed consumption rate, Rs , for
SO~- and production rate. RN • for NHt in each case are given in C.
74 Chapter 2

Alka li nity (meQ/I)

2 4 6 8 10

Figure 13. Vertical profiles of pore water alkalinity

(- HCO;) in experimental tanks with and without
Yoldia limatula, a mobile, subsurface deposit-feed-
ing bivalve (illustrated in Fig. 7) after 45 days. The
E 3
!:!. starting profile was vertically homogeneous. In this
case, a composite layer model in which a surface
g.
.J:;

5 I mixed layer overlies an unmixed layer, such as that

I
o I shown in Fig. 12, works well in explaining the ef-

I
: Control fects of macrofauna on the profile. Part of the de-
7 crease in alkalinity near the interface is caused by

I
sulfide oxidation in both tanks. After Aller (1978).
: Start
Yoldia
9
be required for a conceptually consistent application of the mixing model
in an actual deposit. This does not mean that solutes cannot be modeled
individually in this way, but such modeling becomes progressively more
arbitrary.
Several important qualitative conclusions can be drawn from these
examples. Even a relatively thin mixed layer can radically change the
buildup of reaction products or depletion of reactants both within and
below the mixed zone compared with the molecular diffusion case. This
effect is accentuated by the fact that mixing occurs near the sediment-water
interface where reactions, at least in these cases, are most rapid. Any

A B

Figure 14. Inverted oxidation-reduction stratigraphies can result from selective feeding and
irrigation at distinct horizons in a deposit. For example, when individual Clymenella tor-
quata (A) occur in patches (E) the zone of feeding can become a distinct oxidized plane
overlain by reduced material.
Effects of Macrobenthos on Marine Sediment Chemical Properties 75

constituent subject to diffusion should be influenced in this way. Con-

centration changes for solutes having relatively small molecular diffusion
coefficients, for instance, dissolved organic matter, will be particularly
great below the zone of elevated transport.
The use of apparent transport coefficients works well when animals
are highly mobile and restrict their activities to very near the sediment-water
interface (Aller, 1978) (Fig. 13). As the zone of animal influence increases
in thickness, a discrepancy between model and measured solute profiles
usually arises. The deeper the zone of bioturbation, for example, the larger
D~ must be to explain solute distributions at depth. The result is that
solute profiles near the sediment-water interface become less accurately
simulated at the expense of modeling deeper regions. This kind of model
is also not capable of simulating the maxima or minima that are sometimes
observed in concentration profiles.

4.2. Biogenic Advection

A portion of the overlying water drawn into burrows during irrigation

may be directly exchanged with surrounding pore waters, particularly in
permeable sands. Direct exchange also occurs during active burrowing or
tube construction. Some infauna, such as the lugworm Arenicola marina,
may use permeable sediments as a filter for respiratory currents to aug-
ment food supply either by capture of suspended material or by stimu-
lating local populations of bacteria (Kruger, 1959; Hylleberg, 1975). When
animals are closely spaced and their burrows or tubes are relatively uni-
formly oriented, distinct horizons within the sediment column can be-
come preferentially irrigated by such transport mechanisms. For instance,
beds of the maldanid worm C. torquata , a conveyor belt deposit feeder
that inhabits poorly sorted sands, often produce a second oxidized layer
at their feeding depth within the sediment column owing to both the
depletion of organic matter by selective feeding on fine-grained material
and irrigation of the resulting highly permeable, coarse lag layer (Fig. 14)
(Rhoads, 1974).
Such observations imply that solutes can be biogenically advected
between relatively well-mixed and distinct reservoirs within the sediment
and the overlying water column (Grundman is and Murray, 1977; D. E.
Hammond and Fuller, 1979; McCaffrey et al., 1980). Models constructed
on this basis assign biogenic exchange of pore water to an apparent ad-
vection velocity rather than a random diffusive mixing process. These
types of models have not been used to quantify solute profiles directly
but have been employed to qualitatively explain inverted oxi-
dation-reduction stratigraphies and pore water maxima or minima
76 Chapter 2

t
B
A
-----t--t , r------
Zone I

Zone 2

Figure 15. (AJ A portion of overlying water advected into burrows during irrigation may in
turn be advected into and through surrounding permeable sediment. (BJ Collapsing burrows
also result in rapid advective exchange of local pore water with ove rlying water.

(Grundmanis and Murray, 1977) (Fig. 15), or used as a mass balance

construct to account for observed fluxes across the sediment-water in-
terface (D. E. Hammond and Fuller, 1979; Korosec, 1979; McCaffrey et al.,
1980). Apparent advection velocities of 0.3 - 4 cm/day have been calcu-
lated for estuarine and shelf muds on the basis of in situ 222Rn profiles
or laboratory tracer experiments with 22Na (D. E. Hammond and Fuller,
1979; Luedtke and Bender, 1979; McCaffrey et al., 1980; Smethie et al.,
1981). These apparent rates compare well with actual irrigation velocities
of individual burrows and in that sense are of reasonable magnitude
(Mangum, 1964; Mangum et aI., 1968; Hoffman and Mangum, 1970; Coyer
and Mangum, 1973; Aller and Yingst, 1978).
The concept of advective transport implies that bulk flow of pore
water, usually in some specific direction, takes place in a deposit. Al-
though such flow represents a component of biogenic irrigation activity,
the model formalization of pore water solute transport in this way can
lead to serious misconceptions . Many tube dwellings are lined by poly-
saccharide membranes that are permeable to diffusive transport of solutes
but not advective flow (Foster-Smith, 1978; R. C. Aller, unpublished).
This can be readily demonstrated by removing a section of burrow lining
or tube and noting that in general it is extremely hard to push water
through the wall; however, surrounding sediment is often oxidized, im-
plying diffusive permeability. For this reason the major length of a lined
burrow or tube is unlikely to be involved in advective exchange.
Muddy sediments are themselves impermeable to advective flow. It
seems likely that much of the advective component of pore water ex-
change in muds is due not to the activities of oriented sedentary organisms
but to either mobile benthos or periodic localized collapse of burrow
walls. Such transport is not generally directional or specific to a particular
interval of sediment as implied by the previously described advection
Effects of Macrobenthos on Marine Sediment Chemical Properties 77

models. In contrast to muds, permeable sands are probably subject to

considerable pore water flow, biogenic or otherwise (Riedl et al., 1972).
Selection between application of the two end-member models of apparent
diffusion or apparent advection, neither of which is very realistic, should
therefore depend not only on the functional types of animals present but
also on sediment mass properties.

4.3. Average Diffusion Geometry

An alternative to incorporating the effects of macrobenthos in em-

pirically elevated transport coefficients can be derived by noting that the
bioturbated zone is not a homogeneous body but rather one permeated
by cylinders of water. These cylinders are the burrows or tubes of infauna
and are usually irrigated by their inhabitants. As a result, solute concen-
trations within the burrows are often close to those of the overlying water
(Aller and Yingst, 1978; R. C. Aller and J. Y. Yingst, in preparation).
Interstitial solutes generated within the sediment can diffuse along gra-
dients either toward the sediment-water interface or into nearby burrows.
The converse is true for solutes whose source is overlying water.
The geometry of solute diffusion is therefore altered from the one-
dimensional case by the presence of burrows. The exact geometry is three-
dimensionally complex and time-dependent. It is possible to simplify this
complexity for purposes of modeling by postulating that at any instant
the sediment is a mosaic of micro environments represented on the average
by a central irrigated tube or burrow and its immediately surrounding
sediment (Aller, 1980a). Knowing the composition of this average mi-
croenvironment is the same as knowing the composition of the bulk
deposit.
In the simplest imaginary case all animals inhabiting a deposit are
vertically oriented tube dwellers of equal size and uniform spacing. The
average microenvironment in this instance can be approximated by a
hollow cylinder or annulus of sediment, the inner boundary of which is
an inhabited tube and the outer boundary of which is half the distance
between adjacent tubes (Fig. 16). The bioturbated zone as a whole consists
of many such hollow cylinders packed together as a unit. Only the volume
of sediment at the center of any three adjacent cylinders is not assigned
to a given tube. This represents ~9% of the sediment volume and can be
ignored for reasons that will become obvious. A major advantage of as-
suming cylindrical symmetry is that it simplifies a three-dimensional
problem mathematically into two dimensions while retaining a relatively
realistic solute diffusion geometry. It is also the natural symmetry of the
system as most macrobenthos are themselves cylindrical in shape.
78 Chapter 2

A
I "

II "I .
R
I II
II
- '\1--
o r---~

x
Water
B I
I

,.
,_ -,--1-- .......
~:.j
I

Figure 16. (A) Sketch of upper region of a deposit idealized as packed hollow cylinders of
sediment. (B) Vertical cross-section of a deposit having the idealized diffusion geometry of
A. (C) The simple hollow cylinder of sediment imagined as the average microenvironment
within the bioturbated zone. The dimensions shown are those used in the transport-reaction
model described in the text (Aller, 1980a-c),

When populations of infauna of variable sizes and spacing are present

it is still possible to utilize the cylinder microenvironment concept, but
the dimensions represent a weighted average of those actually present in
a deposit. In this sense, the final cylinder size and orientation used in
modeling are fictitious. Steady-state solute distributions within any cyl-
inder are given by:

ac
at 0= Ds e:~) + Ds {:r [r(ac/ar))} + R (8)

where r is the radial distance from the tube axis. Advection owing to
sedimentation is ignored because of the small diffusion scales involved.
Boundary conditions on the cylinder are taken as:

1. x = 0, r = r1, C = CT
2. r = r2, ac/ar = 0
3. x = L, aC/ax = B

These conditions prescribe that (1) concentrations at the sediment-water

interface and in tubes are the same and constant, (2) concentrations go
through a maximum or minimum half way between any two tubes, and
(3) the flux of solutes at the base of a cylinder matches that from the
underlying, unburrowed zone. Like the cylinder microenvironment itself,
these boundary conditions are approximations. Burrow waters may not
be identical in solute concentration with overlying waters, the lining of
Effects of Macrobenthos on Marine Sediment Chemical Properties 79

some burrows hinders diffusion, and the outer boundary of a cylinder is

not tangent at all points to adjacent cylinders. The solution is given in
the Appendix for R = Roexp( -ax) + RI + k(C eq - C).
Despite these approximations and idealizations, the cylinder mi-
croenvironment produces good agreement with actual data. This is shown
in Fig. 17, which illustrates the solute profiles predicted for a site in Mud
Bay, South Carolina, along with measured profiles (Aller, 1980a). Pre-
dicted vertical profiles are obtained by integrating the solution to equation
(8) over finite vertical depth intervals comparable to sampling intervals.
The reaction rates for NH: and SO~ - are the same as used previously in
Fig. 12. Si is assumed to be produced by a reaction rate of the form R
= k(Ceq - C), where k and Ceq are constants. Measured values of B in
these profiles are -0.1,0.011, and 0.060 mM/cm for SO~-, NH:, and
SI(OH)4, respectively. The model profiles are generated in practice by
fixing r l and L and varying r2 until the best fit is obtained. An average
microenvironment in this example was initially estimated from the av-
erage burrow size (rl ~ 0.05 cm), burrowing depth (L ~ 15 cm), and
abundance (r2 ~ 2.5-5 cm; N = 50-500/m2) of Heteromastus filiformis,
a polychaete that numerically dominates the macroinfauna of Mud Bay.
The final value of r 2 ~ 2.1 cm (N = 722/m 2), necessary to obtain a good
fit, is not unrealistic, and its higher than estimated value reflects animal
mobility, error in estimating abundances, and the presence of other ben-
thos besides Heteromastus. Once the size of the average micro environ-

SO!- (mM) NH: (mM) Si (mM)

5 10 20 0 .1 0.2 0 .6 0.8
2 2 I' : 2 On.
4 I i
4 I' 4
! .
6 6 .~ 6

(T~.""
8 8 . : :11 Measured 8
E 10 10 10
2 12 12 12
.c 14 14 14
g. 16 16 16
o 18 18 18
20 20 20
Figure 17. SO~ - , NHt, and Si concentration profiles in pore water from Mud Bay, South
Carolina, used to illustrate behavior of the cylinder microenvironment model. Solid vertical
bars , measured concentrations; dashed vertical bars, cylinder model profiles; solid contin-
uous curves, one-dimensional model profiles having the same diffusion, reaction, and bound-
ary constants used in the cylinder model. The NHt profile predicted by the one-dimensional
model is off the scale and is not plotted. After Aller (1980a). Compare with the composite
layer models in Fig. 12, which use the same reaction rate distributions for SO~ - and
NHt.
 80 Chapter 2

ment was estimated from the NHt profile, which is very sensitive to
choice of geometry, the Si profile was determined by adjustment of the
first-order reaction rate k (= O.z/day); Ceq (= 577 f.1M) was estimated from
asymptotic concentrations in longer gravity cores. Molecular diffusion
coefficients are used and assumed to be isotropic.
The buildup of pore water solutes depends strongly on the size and
abundance of infauna. This is shown in Fig. 18, in which the mean pore
water concentration in the bioturbated zone is plotted as a function of
either r 1 or r 2 for both NHt and Si(OHk Reaction rates are those used
previously. It is assumed that solid-phase Si0 2 is not limiting. These plots
illustrate that concentration depends strongly on animal abundance and
to a lesser but still large extent on organism size.
The model demonstrates quantitatively that pore water concentra-
tions can be expected to vary in space to the same degree as do animal
populations. These graphs also show that the buildup of pore water con-
stituents subject to zeroth order reactions, such as NHt , is more affected

B
A 1.0 ~ ~u~o~. ~!.-~m!...""~ ::!,e' !!,Ca!... Co...!"_
rl :: O.l cm
' ,' O.Scm
'1= I.Oem

0 .1

c:
.9

~c:
., 0 .01
<>
§
u
c:
"
~ 0 .001 0 .1

+.
-------- ------ c;--
:x:
z 2 4 6 a 10 12
Ha lf Disl ance ae l ween Burrows (cm)

• 31S'3 r95 354 199 121 ~s 65 5'0 39 32 26 U

Holf Dis la nc e Betwe en Burr ows (em) Popula ti on Abundance (lm~)

31~3 795 ~4 199 '27 de ~5 ~o 39 3'2 26 2'2

Populal ion Abuodance (1m')

Figure 18. (A) Expected average concentration of NHt in 0- to 15-cm interval as a fun ction
of burrow spaci ng or abundance (rz) and fixed burrow size (r,). EqUivalent population
abundances (N) per m Z are also indicated . Concentrations are bounded above by the average
concentration predicted by the one-dimensional mod el and below by the overlying water
value. Reaction rate as in Fig. 12C. (B) Expected average concentration of Si in pore water
from 0-15 cm as function of burrow size and abund ance. The apparent solubility. Ceq. is
also plotted.
Effects of Macrobenthos on Marine Sediment Chemical Properties 81

A B c
NH+ (m M)
4
0.1 0 .2 0 .1 0 .2 0 ,1 0 .2
0
II I
I
E 4
/1
I I
~ I II
I I
I
8
\
I

I
£ II I
a.
Q) 12 Profil e ( I
I
0 I
II I I

NH4 (mM/doy)
0 .2 0 .4 0.2 0.4 0.2
0 ,
,,
,,
,
E 4 ,, ,
~ , ,,,
, Reoct ion
8 ,,
£ Rote
~ 12
0

Figure 19. The vertical pore water NH: profiles (upper row) expected for different reaction
rate distributions (lower row) when burrow size and abundance are fixed at r, = 0.05 cm
and r2 = 2,1 cm, respectively. The depth-integrated reaction rate is the same in each case,

by infauna than is the buildup of solutes controlled by first-order kinetics,

that is, the kinetics of reaction determine the response of an element to
animal activities. This comes about because, the greater the pore water
depletion in the latter case, the faster the solid phase dissolves to replenish
the dissolved phase, whereas in the former case production is relatively
independent of pore water buildup. These assumed kinetics are of course
idealized and representative of end-member behavior. For instance, pro-
duction by bacteria of metabolites like NHt is unlikely to be entirely
independent of pore water NHt concentrations.
The shape of vertical profiles depends strongly on the depth depen-
dence of reaction rates as well as the abundance and size of infauna. Fig.
19 illustrates the pore water NH: profile shape expected for three NH:
production rate distributions, each of which has a different depth depen-
dence of production but each with the same integrated or total rate of
NH: formation over the burrowed interval (f~Rdx = constant). The num-
ber and size of burrows present in the sediment are the same in each case.
Because the integrated production rate is constant, the total flux of
NH: from the bioturbated zone at steady state is also the same. Maxima
and minima in concentration distributions are produced by lateral dif-
fusion when reaction rates are strongly attenuated with depth (Fig. 19A).
82 Chapter 2

8 C
A
NH+ (mM)
4
0.1 0 .2 0 .1 0 .2
0 .1 0 .2
,
E I
~ 3 II
J::. ,I I
Q. 5 I I
Q)
0 I I
7
Figure 20. Vertical pore water profiles expected for the respective reaction rates A, B, and
C of Fig. 19 but assuming that the burrowed zone is restricted to 7 cm.

Burrows tap the low-production-rate sediment horizons and prevent sol-

ute buildup at depth much as a water well on land would deplete a
particular stratum of an aquifer having a low rate of groundwater replen-
ishment. When reaction rates are less attenuated, as would occur in areas
of rapid particle reworking, profiles have a parabolic shape that in itself
might not suggest extensive irrigation unless reaction rates were inde-
pendently known (Fig. 19C). The effect of varying burrowing depth for
the same reaction rates is also plotted in Fig. 20 to show that the resulting
solute profile shapes are comparable to the case of a lower attenuation
of reaction rate over a larger burrowed interval, 1. These examples illus-

NO; ~()()
NO; (uM)

2 4 6 8 10
- -...-:-..::.... .....~....
- .....
2 J
4
E
~ 6 "l ····j··l •• j .

-
J::.
a.
Q)
8
10
"t.."\.
"1

\
0
12

14
I
Figure 21. Multidimensional diffusion can produce subsurface maxima in pore water con-
stituents that otherwise would be depleted in the simple one-dimensional case. Expected
NO; distributions for RN = - O.Olexp( - 0.5x) and either no burrows (dashed line) or bur-
rows of radius [, = 0.05 cm, abundance [2 = 2.5 cm, and length L = 15 cm are plotted for
illustration. Depletion of reactive organic substrate at depth implied by the attenuation of
RN allows resupply of pore water NO; by lateral diffusion from burrows.
Effects of Macrobenthos on Marine Sediment Chemical Properties 83

trate that profile shape alone is not indicative of any unique reaction rate
or burrow size distribution. Independent determinations of both must be
made in order to interpret solute distributions.
Inverted stratigraphies of pore water constituents whose sources are
in overlying water can also be generated by multidimensional diffusion
when reactive substrate depletion with depth occurs. To illustrate this,
a hypothetical NO; profile is shown in Fig. 21 for an NO; reduction rate
of R = - O.Olexp( - 0.5x) and microenvironment dimensions of r 1 = 0.05
cm, r2 = 2.5 cm (N ~ 510/m 2 ), and L = 15 cm. The concentration gradient,
B, at the base of the irrigated zone is arbitrarily set here at B = 0 but
presumably would be <0 in the natural situation where NO; would be
depleted again beneath the burrowed zone. The rate function is approx-
imated by a zeroth order equation consistent with control of denitrifi-
cation by the quality of decomposed organic matter present rather than
simple first-order kinetics, as is appropriate when NO; itself is also lim-
iting (Vanderborght and Billen, 1975; Billen, 1978; S0rensen, 1978). Ro
and a were chosen to be in the range of reported NO; fluxes into sedi-
ments (Billen, 1978). In environments where substrate depletion takes
place very close to the sediment-water interface, such as in most areas
of the deep sea or eroded areas of seafloor, the depth scale of solute
minima such as shown in Fig. 21 would be greatly shortened and might
not be detected because of coarseness in sampling intervals.
Of the three models considered, the cylinder microenvironment
model is capable of most closely simulating solute distributions within
the bioturbated zone. Although idealized, because it incorporates the fun-
damental changes in solute diffusion geometry caused by macrobenthos,
it is relatively realistic and allows development of intuition in relating
one-dimensional pore water data to a multidimensional transport-reaction
regime. It is also capable of predicting the magnitude of effects on bottom
chemistry by benthic populations having different sizes and abundances
if reaction rates are known. Coupling between solid phase reworking and
solute transport owing to reversible adsorption is not taken into account
by the cylinder model, but this should not be significant unless the linear
adsorption coefficient is greater than ~10 (Schink and Guinasso, 1978).

5. Macrofaunal Influence on Sediment-Water Exchange

Rates
The exchange of solutes between sediments and overlying waters is
a major component of material cycles in aquatic environments. Because
macrobenthos control pore water solute concentration profiles, they in-
84 Chapter 2

fluence methods of predicting fluxes and in some cases the actual mag-
nitude of a flux. If sediment solute distributions were controlled by one-
dimensional vertical transport and reaction processes then in principle
the diffusive flux of a constituent across the sediment-water interface
could be simply calculated from Fick's first law:

J = - <pDs (ac)
ax x~o
(9)

where J is the solute diffusive flux at the interface, all other symbols are
defined as before, and advective flow is ignored.
In this context two end-member examples will be considered. The
first is pore water solutes controlled by zeroth-order or pseudo-zeroth-
order reaction kinetics so that R = Roexp( - ax) + Rl and the second is
solutes subject to first-order reactions, or R = k(C eq - C), where Ro, a,
R1 , k, and Ceq are constants. Dissolved NHt and Si(OH)4 approximate
these two cases, respectively (Berner, 1974; Schink et al., 1975; Hurd,
1973; Kamatani and Riley, 1979).
At steady state the diffusive flux of a solute produced or consumed
by zeroth-order reactions is independent of transport. If R = Roexp( - ax)
+ Rl over the interval 0 :::; x :::; L, then:

<pRo
J= - [1 - exp( - aLl] - <pR1L (10)
a

assuming no significant flux from the region below 1. Usually the term
exp( - aLl is insignificant for organic matter decomposition reactions. The
overall magnitude of diffusive exchange in this case is fixed by production
or consumption rates and in principle is not affected by macrofaunal
activities. The geometry of exchange can be greatly altered, however, and
prediction of fluxes from vertical pore water gradients by use of equation
(9) made invalid. The one-dimensional transport model is incorrect in
this case because a portion of solute exchange takes place both at the
sediment-water interface (x = 0) and across burrow walls at varying depths
within a deposit.
The vertical solute concentration gradient at the interface is a function
of the sizes and abundances of infauna, which determine the geometry
of diffusion (Fig. 18), and the depth dependence of the reaction rate (Figs.
19 and 20). In the case of the three production rate distributions and
corresponding solute profiles in Fig. 19, for example, the total flux of
NHt from the upper 15 cm is identical at 0.56 flmollcm 2 per day. In
Effects of Macrobenthos on Marine Sediment Chemical Properties 85
1.0
0 .9 rl = O.l cm
0.8 'I =0.5cm
'I = 1.0 em
0.7
I-
--, 0.6
"'-
0.5
--,"
+ - 0,4
•
:I: 0.3
z
0.2
0.1

4 6 8 10 12
Ha I f Distance Between Burrows (em)
, , , , , , I , ! , I !

3183 795 354 199 127 88 65 50 39 32 26 22

Popu lat ion Abundance (1m2 )

Figure 22. The ratio of the flux of NH: through the sediment-water interface at x = 0, lx,
to the total flux of NH: from both burrow walls and the sediment-water interface, h, as
a function of burrow size and abundance. The reaction rate assumed is that of Fig. lZC. The
ratio fxlh is a measure of the ratio of the molecular diffusion coefficient and the apparent
transport coefficient required to match the total flux with the vertical pore water concen-
tration gradient. After Aller (1980a).

contrast, the fluxes predicted from the vertical gradient alone in each case
are 0.26, 0.16, and 0.092 /-Lmollcm 2 per day from left to right, respectively
(Fig. 19) or 46%,29%, and 16% of the total expected value. The lower the
attenuation of reaction rate with depth in a burrowed sediment the less
accurate application of equation (9) is in predicting the flux from the
vertical concentration gradient alone. The cylinder microenvironment
model can be used to predict the variation in the proportion of NHt
released by vertical diffusion across the sediment-water interface relative
to total released. This is shown in Fig. 22 as a function of animal size and
abundance for the reaction rate distribution of Fig. 19A.
The ratio of the solute flux predicted from equation (9) to that actually
measured is sometimes used to determine a value of the effective transport
coefficient D~ necessary to match the vertical concentration gradient with
the measured flux (Martens and Klump, 1980). This coefficient is identical
to that described previously for modeling pore water distributions in the
bioturbated zone. The flux ratio plotted in Fig. 22 is therefore a measure
of the ratio of the bulk sediment molecular diffusion coefficient to the
effective diffusion coefficient, DslD~.
86 Chapter 2

16
14

% 12
~
~ 10
o
! 8
":> 6
LL 4

2 4 6 8 10 12
Hal f Distance Belween Burrows (ern)

3183 795 354 199 127 88 65 50 39 32 26 22

Populat ion Abundance (1m2 )

Figure 23. The total flux of dissolved silica from sediment as a function of burrow size and
abundance. Because of concentration-dependent dissolution. the flux increases as flushing
of the sediment by burrows increases. The assumed reaction rate is k = O.2/day with Ceq
= 577 f.LM. Solid-phase Si is taken as unlimited; in practice elevated fluxes could not be
maintained indefinitely. After Aller (1980a).

For solutes subject only to a first-order reaction rate R k(C eq - C)

the sediment-water flux at steady state is:

(11)

assuming that the concentration gradient across L is 0 and the value (kl
D)L is large. The presence of Ds in the expression demonstrates that the
flux in this case is not independent of transport. The greater the effective
transport rate the larger the flux. This is illustrated in Fig. 23 utilizing
the cylinder model for a rate constant k = 0.2/day, a diffusion coefficient
Ds = 0.687, an overlying water concentration CT = 0.074 mM, a basal
gradient B = 0.060 mM/cm, and a solubility Ceq = 0.577 mM. For sim-
plicity, solid-phase silica is assumed to be in excess and not limiting to
the reaction. In contrast to ammonium, the greater the flushing of sediment
by burrows in this instance, the greater the total flux of silica. The vari-
ation, as a function of animal size and abundance, in DsfD~ necessary to
explain a given silica flux from a vertical pore water gradient is similar to
but less pronounced than for the zeroth-order case (Aller, 1980a). Al-
though the total dissolved silica flux from the bottom may be greatly
increased, a corresponding increase in solute supply by additional dis-
solution of the solid phase tends to balance the diffusive loss into burrows
and lessens the effect of irrigation on the pore water concentration profile
as compared with constituents like NHt. .
Effects of Macrobenthos on Marine Sediment Chemical Properties 87

These examples show that irrigated tube structures and active bur-
rowing influence pore water constituents differently depending on the
kinetics of reactions controlling solute loss or supply and reaction rate
distribution. The ratio DslD~ determined from the discrepancy between
measured fluxes and that predicted from a vertical concentration gradient
for one constituent is therefore unlikely to be applicable to all solutes in
a deposit.

6. Reaction Rates
Aside from changing the geometrical distribution of reaction rates by
redistributing and repackaging particles, macrobenthos stimulate the rates
of microbially mediated decomposition reactions (Fenchel, 1970; Har-
grave, 1970, 1976; Harrison and Mann, 1975; Fenchel and Harrison, 1976;
Aller, 1978; Yingst and Rhoads, 1980). This occurs in several ways:
1. Surface area of organic detritus is mechanically increased or ex-
isting surfaces reexposed during feeding.
2. Grazing maintains microbial populations in a high-productivity
phase of rapid growth.
3. Metabolite buildup is decreased and electron acceptor supply in-
creased by irrigation, particle reworking, and multidimensional
diffusion.
4. Mucus secretions provide new reactive substrate.
5. Subduction or capture of resuspended reactive organic matter is
increased during reworking (discussed in Section 3).
An example of such stimulation is shown by measurement of the
oxygen uptake flux to various naturally occurring particles before and
after ingestion by different invertebrates (Fig. 24) (Hargrave, 1976). In-
creased uptake is presumably caused by one or more of the factors listed
above. Uptake decreases to preingestion rates after a few days.
Because at least one component of stimulation, reexposure of sur-
faces, is related directly to particle reworking activities, in the strictest
sense the reaction rate, R, in previous models is a function of the particle
mixing rate, DB. The extent of dependence remains to be determined.
Limits on the effect of mechanical disturbance can be set by measuring
metabolic rates in sediments that have been vigorously stirred (DB ~ 00)
compared with physically undisturbed material (DB ~ 0). Westrich (1983)
reports that SO~ - reduction rates in nearshore sediments are two
times higher in mixed than unmixed samples. Similar but less dramatic
results have been obtained in our own laboratory for a variety of anaerobic
decomposition reactants or products indicating elevated rates for periods
88 Chapter 2

Oxygen Uptake
(jJg 02icm2 i hr)
o 2 345 6
intert idal sand , (polychael.)

loke '5edimeo' (amphlpod)

lak e sedimen t .'" (Oli gochoeteJ

lake sedimenl .:,;;

fine sand .. · ·· '!21lIiJI (prolobronch bivalve)

T hola .. ;g .•.•... Parohyolell0 (amphl pod) Figure 24. Oxygen consumption rates of
(goslropod) different material before (cross hatched)
and after (stippled) ingestion by benthic
algae Wl.!.2r.iruI (gaslropod)
invertebrates. Elevated consumption rates
~ E~::::~~~~
-' llItl (echinoid) are sustained for several days following
pOrggmitt s Limnaea ( osl,o od) defecation. After Hargrave (1976).

of 1-2 weeks after mixing. This implies that decomposition rates in sed-
iments may vary between these two extremes of mixed and unmixed on
a spatial scale comparable to the size of individual mobile macrofauna
regardless of other stimulatory effects.
Greenwood (1968) suggests a simple model as to how such mechan-
ical stimulation may occur. Physical segregation of bacteria from organic
substrate and adsorption or physical hinderance of exoenzyme movement
by clays are envisioned as major factors limiting decomposition at the
microscale of sediment aggregates (Fig. 25). The mechanical disaggrega-
tion of such clusters by macro- and meiofauna results in a new arrange-
ment in which relatively unutilized substrate is exposed and pore water
exchanged. A short-term increase of microbial activity results.
In surface-oxidized material the interior of such aggregates is often

Figure 25. Model sedimentary aggregate

illustrating role of physical and chemical
~ Quar tz C] Organic Motter
(reactive clay surfaces) barriers to utiliza-
tion by microorganisms of organic matter
r;::;) Cl oy Domain • B acte ri a in a deposit. After Greenwood (1968) .
Effects of Macrobenthos on Marine Sediment Chemical Properties 89

anoxic despite abundant oxygen in surrounding pore waters. These ag-

gregates, which are usually fecal pellets, can be modeled as distinct sed-
imentary microenvironments themselves just as can be done for burrows
(Greenwood, 1968; J0rgenson, 1977b; Anderson and Meadows, 1978; Aller
and Yingst, 1978). In order to maintain a chemical microenvironment
distinct from that imposed by gradients from the sediment-water interface
or adjacent burrows, it is necessary that Ds (the bulk diffusion coefficient)
within the aggregate be less than Ds for the surrounding sediment or that
reaction rates be locally stimulated. The compacted, low-porosity nature
of many fecal pellets should indeed restrict diffusion within pellets com-
pared with ambient sediment, and data like those shown in Fig. 24 imply
that metabolic rates may be stimulated as well. Because fecal pellet size,
shape, mucus content, and compactness vary between individuals and
animal species, considerable microscale heterogeneity in biogeochemical
properties and reaction rates is expected at the sediment-water interface
where newly formed pellets are deposited.
Creation of chemical micro environments by macrofauna may also
determine local distribution of meiofauna around biogenic structures (Bell
et aI., 1978; Aller and Yingst, 1978; Thistle, 1979). Because meiofaunal
grazing of bacteria is known to influence decomposition (Fenchel and
Harrison, 1976) this effect will further accentuate spatial variation in
reaction rate.

7. Chemistry of the Burrow Habitat

The fact that burrow walls are permeable to solute diffusion means
that sedimentary reactions can modify the burrow habitat. A continual
flux of solutes from surrounding sediment into burrows takes place. The
magnitude of this flux is determined in any given case by concentration
gradients around a burrow and sediment permeability. In order to hold
the composition of burrow water constant or within a restricted range of
variation different from pore water, continual irrigation of the burrow
must balance the flux of solutes from the surrounding sediment. The
steady-state balance for a constituent not subject to precipitation in the
burrow is given by:

(12)

where F is the rate of change of burrow water solute concentration, J is

the solute flux (constant) from the tube walls, A is the inner surface area
of the burrow, Va is the burrow volume, v is the irrigation rate (volume/
90 Chapter 2

time), CB is the steady-state solute concentration in burrow water, and CT

is the solute concentration in overlying seawater. If the solute of interest
is also excreted by the burrow inhabitant, an additional source term must
be added to F.
This simple mass balance model has been shown to work well for an
intertidal population of Amphitrite ornata, a terebellid polychaete (Aller
and Yingst, 1978). The rate of change, F, in that case was estimated from
changes in the burrow water composition during tidal recession when
irrigation could not take place (Fig. 26). Mass balance models of this type
have also been used to calculate irrigation rates necessary to supply ox-
ygen to burrow inhabitants (Mangum and Burnett, 1975).
The flux of solutes from surrounding sediment determines in part the
excursions in burrow water composition during intertidal periods. Met-
abolic activity of burrow inhabitants (including bacteria) also alters bur-
row waters during the intertidal interval; oxygen, for example, is known
to be depleted in exposed burrows (Jones, 1955; Thompson and Pritchard,

B NH; (mM)
A
0.2 0.4 0.6 0.8 1.0 1.2 1.4
I
'!"
2
100 o /
1 3 .. -'. f. -
4
80 1

1
1
5
o
I
6
1
I 7
r.~ 40 /
1
E 8
z 1
3 9
09 1 ~ 10
20 /
1
0 g.1I : ;.
I
I
o 12 I
I

40 120 200 280 13 T .I. Burrow

I
Minutes Exposed 14 I
woll
15 I
I
16 I
17 J.

Figure 26. (A) NH: concentration in Amphitrite ornata burrows exposed on a tidal flat
versus time of exposure. Except for the first sample (t = 10 min), which represents a single
burrow, each point represents the average concentration in two or more burrows. (B)
NH: concentration in pore water from concentric sample intervals around one branch of
an A. ornata burrow and from surrounding sediment. ...... , 0.5- to 1.5-cm radial interval
from burrow axis; ~~~~, 1.5- to 3.75-cm radial interval; - - , sediment 30 cm from axis.
Each bar represents the vertical interval over which the sample was taken. After Aller and
Yingst (1978).
Effects of Macrobenthos on Marine Sediment Chemical Properties 91

1969). Assuming that CB must on the average be within a range consistent

with the physiological tolerances of a species, then the solute flux deter-
mines the minimum irrigation effort an animal must expend to control
the burrow environment when connection with overlying water is pos-
sible. Because the flux to an individual burrow is controlled by reaction
rates and solute buildup in surrounding sediment, it is necessarily influ-
enced by the presence or absence of other animals. This can be illustrated
using the cylinder model developed previously and equation (8) (Aller,
1980b). As an example, F for NHt has been calculated over a range of
burrow sizes and interburrow distances from the relation

(13)

using the reaction rate distribution of Fig. 12 and the remaining model
values as in Fig. 17 (see page 79), and integrating over burrow length
(15 cm). If CB and CT are assumed constant then the factor F· Vo is pro-
portional to the irrigation rate necessary to balance the solute influx. This
value is also plotted in Fig. 27 as a function of burrow size and density.
Excursions in burrow water compositions during intertidal periods,
the irrigation effort required to keep burrow waters at steady state, or the
standing concentration of sediment-derived solutes in the burrow at a
particular irrigation rate can all be lowered in two ways: (1) by decreasing
interburrow distances or (2) by increasing burrow size. Of the two stra-
tegies, crowding together in patches of high burrow abundance is the
most effective at reducing irrigation requirements. The construction of
impermeable burrow walls would accomplish the same result.
These simple consequences of diffusion geometry are apparently
utilized by animals inhabiting chemically undesirable sediments. It is
known that early colonizers of defaunated substrates such as waste dumps
or eroded areas of the bottom are often small tube dwellers in high abun-
dance (Grassle and Grassle, 1974; McCall, 1977; Rhoads et aI., 1977; Pear-
son and Rosenberg, 1978). These sediments commonly contain elevated
levels of metabolites or toxins. The small size of such animals is consistent
with the often relatively low oxygen contents of these environments (Pear-
son and Rosenberg, 1978), while crowding or high population density
can be interpreted as an extremely effective adaptation to a chemically
hostile substratum (Fig. 27) (Aller, 1980b). Crowding may also have the
beneficial effect of physically stabilizing the seafloor against erosion
(Schafer, 1972; Rhoads et aI., 1978; Rhoads and Boyer, Chapter 1).
Consideration of animal adaptations to sediment chemistry therefore
leads to the inference that high population density and rapid reproduction
 92 Chap ter Z

A B
10.0 0.8

07
rl = 0.05 em
0.6 VOfiable
Infer burrow
Distance

.....
L: 0 .1
:::;
5 r2=2.l em 03
u.. (N = 7221m2)
Variable
0.01 Burrow Size 0.2

Typical '2 r2 = 2.lem

0.1 Variable
Burraw Siz e

0.001 01 03 05 0.7 0 9. 0.1 0 .3 05 0,7 0.9

Burrow Radius, rl (em) Burrow Radius, r l (em)

o 2 4 6 8 10 024 6 8 ~
Half Distance Between Burrows, r2 (em) Half Distance Bet ween Burrows, r2 (em)

Figure 27. (A) The expected initial rate of change, F, of burrow water NH; concentrations
in exposed burrows as a function of burrow size and abundance of surrounding burrows.
The NH: production rate in surrounding sediment is that of Fig. lZC. tB) The irrigation rate
expressed as the factor F·Vo (see text) required to " stat " a burrow at a constant NH: con-
centration as a fun ction of burrow size and interburrow distances, Crowding togeth er dra-
matically decreas es the irrigation work required of individuals to detoxify the burrow habitat
or, correspondingly, decreases the concentration of sediment-derived solutes in the burrow
at any particular irrigation rate. After Aller (1980b).

in this context may represent specific adaptations to the chemical envi-

ronment rather than to the usual assumption that high fecundity is char-
acteristic of poor competitors. Overall, very little is known about the
chemical characteristics of the infaunal habit. This example, which is
speculative, illustrates that future research in this area will probably pro-
vide much insight into the ecology of soft-bottom benthos.

8. Spatial and Temporal Patterns in Sediment Chemistry

Because the effects of macrofauna on sediment chemistry are deter-
mined by the number, kinds, and sizes of animals, sediment composition
varies in space and time in accordance with the particular benthic com-
munity present (Rhoads et aI. , 1977; Aller, 1980c). This variation should
Effects of Macrobenthos on Marine Sediment Chemical Properties 93

occur regardless of otherwise uniform substrate characteristics such as

grain size or quantity of organic matter.
Temporal patterns in the kinds of animals inhabiting a deposit are
known to occur on ecological time scales (Pearson and Rosenberg, 1978;
Rhoads and Boyer, Chapter 1). Successional patterns in sediment chem-
istry occur in conjunction with these changes in community structure
(Rhoads et al., 1977). With respect to sediment chemistry, many of the
most important changes in animal type involve a simple increase in bur-
rowing depth and size of individuals with time. Progressively deeper
irrigation and particle reworking cause increased pore water exchange,
the subduction of fresh organic matter, reoxidation of reduced com-
pounds, and increased biogeochemical heterogeneity. The role of sedi-
ment chemistry in actually causing or allowing successive replacement
of fauna is at present speculative, as described previously.
On longer time scales, bioturbation as a process has changed pro-
gressively through geologic time with both the depth and intensity of
reworking increasing since the Precambrian (Thayer, 1979; Bambach and
Sepkoski, 1979). The corresponding pattern of change in chemical diage-
netic processes over geologic time must have been similar to spatial or
temporal patterns in sediment chemistry associated with differences in
faunal distributions in recent sediments. In the absence of significant
reworking, pore water solute profiles were presumably similar to those
predicted by one-dimensional transport reaction models. Saturation of
pore water with respect to a variety of reduced phases must have occurred
on the average closer to the sediment-water interface. Nutrients like
PO~ - , which may become incorporated into reduced phases, or like Si,
which are subject to concentration-dependent dissolution, were almost
certainly less efficiently recycled to the water column than in present-day
bioturbated sediment. Sulfide oxidation was probably less important be-
cause of less extensive particle reworking, and, partly as a result, the
dissolution of carbonate was probably less favored (Aller, 1982). Microen-
vironmental heterogeneity in diagenetic reactions must have also been
minimal in the absence of macrofaunal construction activity. Some sys-
tematic differences in the authigenic mineral components of sediments
through geologic time from otherwise equivalent depositional environ-
ments can therefore be expected to occur along with the differences in
sedimentary structures already documented (Bambach and Sepkoski,
1979; Sepkoski and Bambach, 1979).

9. Summary
In the preceding discussion I have emphasized some of the general
principles involved in understanding and quantifying the effects of ma-
94 Chapter 2

crofauna on sediment chemistry. Specific effects depend on the functional

groups of animals present in the benthic community, characteristics of
the depositional environment such as resuspension and net deposition
rates, and the kinetics of reaction of the constituents of interest. A lack
of detailed consideration of these factors in construction of quantitative
models can lead to unwarranted or at least misleading conclusions, such
as the often-repeated statement that the net flux of solutes from or to
bottom sediments must be increased by bioturbation. The interactions
between different groups of sediment-dwelling organisms as determined
by, or reflected in, sediment chemistry remain one of the most intriguing
and understudied aspects of marine biogeochemistry.

Appendix: Solutions to Model Equations

Equation (5) with conditions 5.1-4 has the following solutions:

Zone 1

Zone 2

A(x) f33 exp( - 'Axlw)

where:

a2 =
2DB
f31 = 'oa2 exp( a 2L1)
al(DBa2 - w) exp(alLd - a2(DBal - w) exp(a2Ll)
f32 = - 'oal exp(alLd
al(DBa2 - w) exp(a 1L1) - a2(DBal - w) exp(a 1L1)
f33 = Bl exp(a1L1) + f32 exp(a2L1) exp('ALlw)

Equation (7) with conditions 7.1-4 has the following solutions:

Effects of Macrobenthos on Marine Sediment Chemical Properties 95

Zone 1

Zone 2

+ Rl (L2 - Ld 2 R1U
+- RoLl
-- - - [exp ( - a L)]
2
2D2 2Dl Dla
Ro Ro
+ --2 [1 - exp( - aLd] + - (Ll - x)[exp( - aL 2 )]
Dla D2 a

+ -Ro
- 2 [exp( -aLd - exp( -ax)]
D2 a

Equation (8) with conditions 8.1-3 and R = Ro exp( -ax) + Rl + k(C eq

- C) has the following solution:

2
C(x,r) = CT + Bx + _D ~
£.J
[Gn] [ UUo(/-1nr)
-2 ( )- 1 ] sm(~nx)
.
L s n= 0 /-1
n 0 /-1n r 1

where:

n = 0,1,2 ...
~n = (n + i)(7T/L)
/-1n = (kIDs + ~~)1/2

Gn-- k(C T - Ceq) _ Rl (-1tkB R rae-aLl - I t - ~n]

+ 2 + 0 2 2
~n ~n ~n a + ~n

The functions Iv(z) and Kv(z) are the modified Bessel functions of the first
and second kind, respectively, of order v.
96 Chapter 2

ACKNOWLEDGMENTS. This work has been supported by NSF Grant OCE-

7900971 and successor OCE-811415. Fellowship support from the Sloan
Research Foundation is also acknowledged. J. Pasdeloup deciphered and
typed the manuscript. I thank J. Y. Yingst, J. E. Mackin, and P. W. McCall
for critical comments or editorial suggestions.

References
Aller. R C.. 1978, Experimental studies of changes produced by deposit feeders on pore
water, sediment, and overlying water chemistry, Am. f. Sci. 278:1185-1234.
Aller, R C., 1980a, Quantifying solute distributions in the bioturbated zone of marine sed-
iments by defining an average microenvironment, Geochim. Cosmochim. Acta
44:1955-1965.
Aller, R. C., 1980b, Relationships of tube-dwelling benthos with sediment and overlying
water chemistry, in: Marine Benthic Dynamics (K. R Tenore and B. C. Coull, eds.), pp.
285-308, University of South Carolina Press, Columbia.
Aller, R c., 1980c, Diagenetic processes near the sediment-water interface of Long Island
Sound. 1. Decomposition and nutrient element geochemistry (S, N, P), Adv. Geophys.
22:237-350.
Aller, R. c., 1982, Carbonate dissolution in shallow water marine sediments: Role of physical
and biological reworking, f. Geoi. 90:79-95.
Aller, R C., and Cochran, J. K., 1976, 234Th/238U disequilibrium in near-shore sediment:
Particle reworking and diagenetic time scales, Earth Planet. Sci. Lett. 29:37-50.
Aller, R c., and Dodge, R E., 1974, Animal-sediment relations in a tropical lagoon, Dis-
covery Bay, Jamaica, f. Mar. Res. 32:209-232.
Aller, R C., and Yingst, J. Y., 1978, Biogeochemistry of tube-dwellings: A study of the
sedentrary polychaete Amphitrite ornata (Leidy), J. Mar. Res. 36:201-254.
Aller, R. c., Benninger, 1. K., and Cochran, J. K., 1980, Tracking particle associated processes
in nearshore environments by use of 234Th/238U disequilibrium, Earth Planet. Sci. Lett.
47:161-175.
Anderson, J. G., and Meadows, P. S., 1978, Microenvironments in marine sediments, Proc.
R. Soc. Edinburgh Sect. B 76:1-16.
Amiard-Triquet, C., 1974, Etude experimentale de la contamination par Ie cerium 144 et Ie
fer 59 d'un sediment a Arenicola marina L. (Annelida Polychete), Cah. Bioi. Mar.
15:483-494.
Aston, S. R., and Chester, R, 1973, The influence of suspended particles on the precipitation
of iron in natural waters, Estuarine Coastal Mar. Sci. 1:225-231.
Baas-Beckling,1. C. M., Kaplan, 1. R, and Moore, D., 1960, Limits of the natural environment
in terms of pH and oxidation-reduction potentials, J. Ceoi. 68:243-284.
Bambach, R K., and Sepkoski. J. J., Jr., 1979, The increasing influence of biologic activity
on sedimentary stratification through the Phanerozoic, Ceoi. Soc. Am. 11:383.
Baumfalk, Y. A., 1979, Heterogeneous grain size distribution in tidal flat sediment caused
by bioturbation activity of Arenicola marina (Polychaeta), Neth. f. Sea Res. 13:428-440.
Bell, S. S., Watzin, M. C., and Coull, B. C., 1978, Biogenic structure and its effect on the
spatial heterogeneity of meiofauna in a salt marsh, J. Exp. Mar. Bioi. Ecoi. 35:99-107.
Benninger, L. K., Aller, R c., Cochran, J. K., and Turekian, K. K., 1979, Effects of biological
sediment mixing on the 210Pb chronology and trace metal distribution in a Long Island
Sound sediment core, Earth Planet. Sci. Lett. 43:241-259.
Effects of Macrobenthos on ·Marine Sediment Chemical Properties 97

Ben-Yaakov, S., 1973, pH buffering of pore water of recent anoxic marine sediments, Limnol.
Oceanogr. 18:86-94.
Berner, R. A., 1974, Kinetic models for the early diagenesis of nitrogen, sulfur, phosphorus
and silicon in anoxic marine sediments, in: The Sea (E. D. Goldberg, ed.). Volume 5,
pp. 427-450, John Wiley and Sons, New York.
Berner, R. A., 1976a, The benthic boundary layer from the viewpoint of a geochemist, in:
The Benthic Boundary Layer (1. N. McCave, ed.). pp. 33-55, Plenum Press, New York.
Berner, R. A., 1976b, Inclusion of adsorption in the modelling of early diagenesis, Earth
Planet. Sci. Lett. 29:333-340.
Berner, R. A., 1980, Early Diagenesis-A Theoretical Approach, Princeton University Press,
Princeton, New Jersey.
Berner, R. A., 1981, A rate model for organic matter decomposition during bacterial sulfate
reduction in marine sediments, Colloq. Int. C.N.R.S. 293:35-44.
Billen, G., 1978, A budget of nitrogen recycling in North Sea sediments off the Belgian coast,
Estuarine Coastal Mar. Sci. 7:127-146.
Brenchley, G. A., 1978, On the regulation of marine infaunal assemblages at the morpho-
logical level: A study of the interactions between sediment stabilizers, destabilizers,
and their sedimentary environment, Ph.D. dissertation, Johns Hopkins University, Bal-
timore, Maryland.
Cadee, G. C., 1976, Sediment reworking by Arenicola marina on tidal flats in the Dutch
Wadden Sea, Neth. J. Sea Res. 10:440-460.
Carpenter, R., Peterson, M. 1., and Bennett, J. T., 1982, 21OPb-derived sediment accumulation
and mixing rates for the Washington continental slope, Mar. Geol. 48: (in press).
Claypool, G. E., and Kaplan, 1. R., 1974, The origin and distribution of methane in marine
sediments, in: Natural Gases in Marine Sediments (1. R. Kaplan, ed.). pp. 99-139,
Plenum Press, New York.
Cochran, J. K., and Aller, R. C., 1979, Particle reworking in sediments from the New York
Right Apex: Evidence from 234Th/238U disequilibrium, Estuarine Coastal Mar. Sci.
9:739-747.
Coyer, P. E., and Mangum, C. P., 1973, Effect of temperature on active and resting metabolism
in polychaetes, in: Effects of Temperature on Ectothermic Organisms (W. Wieser, ed.).
pp. 173-180, Springer-Verlag, New York.
Cullen, D. J., 1973, Bioturbation of superficial marine sediments by interstitial meiobenthos,
Nature 242:323-324.
Dapples, E. C., 1942, The effect of macro-organisms upon near-shore marine sediments, J.
Sediment. Petrol. 12:118-126.
DeMaster, D. J., Nittrouer, C. A., Cutshall, N. H., Larsen, 1. 1., and Dion, E. P., 1980, Short
lived radionuclide profiles and inventories from Amazon continental shelf sediments,
Eos 61:1004.
Fager, E. W., 1964, Marine sediments: Effects of a tube-building polychaete, Science
143:356-359.
Fenchel, T., 1970, Studies on the decomposition of organic detritus derived from the turtle
grass Thalassia testudinum, Limnol. Oceanogr. 15:14-20.
Fenchel, T., and Harrison, P., 1976, The significance of bacterial grazing and mineral cycling
for the decomposition of particulate detritus, in: The Role of Terrestrial and Aquatic
Organisms in Decomposition Processes (J. M. Anderson and A. Macfadyen, eds.). Proc.
Symp. Br. Ecol. Soc. 17:285-299.
Fenchel, T., Kofoed, 1. H., and Lappalainen, A., 1975, Particle size-selection of two deposit
feeders: The amphipod Corophium volutator and the prosobranch Hydrobia ulval. Mar.
Biol. 30:119-128.
98 Chapter 2

Filipek, 1. H., and Owen, R M., 1980, Early diagenesis of organic carbon and sulfur in outer
shelf sediments from the Gulf of Mexico, Am. f. Sci. 280:1097-1112.
Fisher, J. B., Lick, W. J., McCall, P. 1., and Robbins, J. A., 1980, Vertical mixing of lake
sediments by tubificid oligochaetes, f. Geophys. Res. 85:3997-4006.
Foster-Smith, R 1., 1978, An analysis of water flow in tube-living animals, f. Exp. Mar. Bio1.
Eco1. 34:73-95.
Froelich, P. N., Klinkhammer, G. P., Bender, M. 1., Luedtke, N. A., Heath, G. R, Cullen, D.,
Dauphin, P., Hammond, D., Hartman, B., and Maynard, V., 1979, Early oxidation of
organic matter in pelagic sediments of the eastern equatorial Atlantic: Soboxic diage-
nesis, Geochim. Cosmochim. Acta 43:1075-1091.
Goldberg, E. D., and Koide, M., 1962, Geochronological studies of deep sea sediments by
the ionium/thorium method, Geochim. Cosmochim. Acta 26:417-450.
Goldhaber, M. B., and Kaplan, I. R, 1974, The sulfur cycle, in: The Sea (E. D. Goldberg,
ed.), Volume 5, pp. 569-655, John Wiley and Sons, New York.
Goldhaber, M. B., Aller, R C., Cochran, J. K., Rosenfeld, J. K., Martens, C. S., and Berner,
R A., 1977, Sulfate reduction diffusion and bioturbation in Long Island Sound sedi-
ments: Report of the FOAM group, Am. f. Sci. 277:193-237.
Grassle, J. F., and Grassle, J. P., 1974, Opportunistic life histories and genetic systems in
marine benthic polychaetes, f. Mar. Res. 32:253-284.
Greenwood, D. J., 1968, Measurement of microbial metabolism in soil, in: The Ecology of
Soil Bacteria (T. R. G. Gray and D. Parkinson, eds.), pp. 138-157, University of Toronto
Press, Toronto.
Grill, E. V., and Richards, R A., 1964, Nutrient regeneration from phytoplankton decom-
posing in sea water, f. Mar. Res. 22:51-69.
Grundmanis, V., and Murray, J. W., 1977, Nitrification and denitrification in marine sedi-
ments from Puget Sound, Limnol. Oceanogr. 22:804-813.
Guinasso, N. 1., Jr., and Schink, D. R, 1975, Quantitative estimates of biological mixing
rates in abyssal sediments, f. Geophys. Res. 80:3032-3043.
Hammond, D. E., and Fuller, c., 1979, The use of radon-222 as a tracer in San Francisco
Bay, in: San Francisco Bay: The Urbanized Estuary (T. J. Conomos, ed.), pp. 213-230,
American Association for the Advancement of Science, San Francisco, California.
Hammond, D. E., Simpson, H. J., and Mathieu, G., 1975, Methane and radon-222 as tracers
for mechanisms of exchange across the sediment-water interface in the Hudson River
Estuary, in: Marine Chemistry in the Coastal Environment (T. M. Church, ed.), ACS
Symposium Series 18, pp. 119-132, American Chemical Society, Washington, D.C.
Hammond, L. S., 1981, An analysis of grain size modification in biogenic carbonate sedi-
ments by deposit-feeding holothurians and echinoids (Echinodermata), Limno1. Ocean-
ogr. 26:898-906.
Hanor, J. S., and Marshall, N. T., 1971, Mixing of sediment by organisms, in: Trace Fossils
(B. F. Perkins, ed.), Louisiana State University Miscellaneous Publication 71-1, pp.
127-136, Louisiana State University Press, Baton Rouge.
Hargrave, B. T., 1970, The effect of a deposit-feeding amphipod on the metabolism of benthic
microfiora, Limno1. Oceanogr. 15:21-30.
Hargrave, B. T., 1972, Aerobic decomposition of sediment and detritus as a function of
particle surface area and organic content, Limnol. Oceanogr. 17:583-596.
Hargrave, B. T., 1976, The central role of invertebrate faeces in sediment decomposition, in:
The Role of Terrestrial and Aquatic Organisms in Decomposition Processes (J. M. An-
derson and A. Macfadyen, eds.), Proc. Symp. Br. Eco1. Soc. 17:301-321.
Hargrave, B. T., and Phillips, G. A., 1977, Oxygen uptake of microbial communities on solid
surfaces, in: Aquatic Microbial Communities (J. Cairns, Jr., ed.), pp. 545-587, Garland,
New York.
Effects of Macrobenthos on Marine Sediment Chemical Properties 99

Harrison, P. G., and Mann, K H., 1975, Detritus formation from eelgrass (Zostera marina
L.): The relative effects of fragmentation, leaching, and decay, Limnol. Oceanogr.
20:924-934.
Hoffman, R. J., and Mangum, C. P., 1970, The function of coelomic cell hemoglobin in the
polychaete Glycera dibranchiata, Compo Biochem. Physiol. 36:211-228.
Hurd, D. c., 1973, Interactions of biogenic opal, sediment and seawater in the central equa-
torial Pacific, Geochim. Cosmochim. Acta 37:2257-2282.
Hylleberg, J., 1975, Selective feeding by Abarenicola pacifica with notes on Abarenicola
vagabunda and a concept of gardening in lugworms, Ophelia 14:113-137.
Jones, J. D., 1955, Observations on the respiratory physiology and on the haemoglobin of
the polychaete genus Nephthys, with special reference to N. hombergii (Aud. et M.
Edw.), J. Exp. BioI. 32:110-125.
Jorgensen, B. B., 1977a, The sulfur cycle of a coastal marine sediment (Linfjorden, Denmark),
Limnol. Oceanogr. 22:814-831.
Jorgensen, B. B., 1977b, Bacterial sulfate reduction within reduced microniches of oxidized
marine sediments, Mar. Biol. 41:7-17.
Jorgensen, B. B., 1978, A comparison of methods for the quantification of bacterial sulfate
reduction' in coastal marine sediments. II. Calculation from mathematical models, Geo-
micro bioI. J. 1:29-48.
Jumars, P. A., Nowell, A. R. M., and Self, R. F. 1., 1981, A Markov model of
flow-sediment-organism interactions: Model formulation and sensitivity analysis, in:
Sedimentary Dynamics of Continental Shelves (c. A. Nittrouer, ed.), pp. 155-172,
Elsevier, Amsterdam.
Kamatani, A., and Riley, J. P., 1979, Rate of dissolution of diatom silica walls in seawater,
Mar. BioI. 55:29-35.
Khripounoff, A., and Sibuet, M., 1980, La nutrition d'echinodermes abyssaux.l. Alimentation
des holothuries, Mar. BioI. 60:17-26.
Korosec, M. A., 1979, The effects of biological activity on transport of dissolved species
across the sediment-water interface of San Francisco Bay, M.S. thesis, University of
Southern California, Los Angeles.
Krishnaswami, S., Benninger, 1. K, Aller, R. C., and Van Damm, K. L., 1980, Atmospheri-
cally-derived radionuclides as tracers of sediment mixing and accumulation in near-
shore marine and lake sediments: Evidence from 7Be, 21°Pb, and 239. 240pU, Earth Planet.
Sci. Lett. 47:307-318.
Kriiger, F., 1959, Zur Ernahrungsphysiologie von Arenicola marina 1., Zool. Anz. Suppl.
22:115-120.
Lerman, A., 1978, Chemical exchange across sediment-water interface, Annu. Rev. Earth
Planet. Sci. 6:281-303.
Lerman, A., 1979, Geochemical Processes: Water and Sediment Environments, John Wiley
and Sons, New York.
Levinton, J. S., 1980, Particle feeding by deposit feeders: Models, data and a prospectus, in:
Marine Benthic Dynamics (K R. Tenore and B. C. Coull, eds.), pp. 423-439, University
of South Carolina Press, Columbia.
Li, Y.-H., and Gregory, S., 1974, Diffusion of ions in sea water and in deep sea sediments,
Geochim. Cosmochim. Acta 38:703-714.
Livingston, H. D., and Bowen, V. T., 1979, Pu and 137CS in coastal sediments, Earth Planet.
Sci. Lett. 43:29-45.
Luedtke, N. A., and Bender, M. 1., 1979, Tracer study of sediment-water interactions in
estuaries, Estuarine Coastal Mar. Sci. 9:643-651.
McCaffrey, R. J., Myers, A. C., Davey, E., Morrison, G., Bender, M., Luedtke, N., Cullen, D.,
Froelich, P., and Klinkhammer, G., 1980, The relation between pore water chemistry
100 Chapter 2

and benthic fluxes of nutrients and manganese in Narragansett Bay, Rhode Island,
Limnol. Oceanogr. 25:31-44.
McCall, P. L., 1977, Community patterns and adaptive strategies of the infaunal benthos of
Long Island Sound, J. Mar. Res. 35:221-266.
Mangum, C. P., 1964, Activity patterns in metabolism and ecology of polychaetes, Compo
Biochem. PhysioI. 11:239-256.
Mangum, C. P., and Burnett, L. E., 1975, The extraction of oxygen by estuarine invertebrates,
in: Physiological Ecology of Estuarine Organisms (F. J. Vernberg, ed.), pp. 147-163,
University of South Carolina Press, Columbia.
Mangum, C. P., Santos, S. L., and Rhodes, W. R., Jr., 1968, Distribution and feeding in the
onuphid polychaete, Diopatra cuprea (Bosc), Marine BioI. 2:33-40.
Mare, M. F., 1942, A study of a marine benthic community with special reference to the
microorganisms, J. Mar. Bioi. Assoc. U. K. 25:517-574.
Martens, C. S., and Klump, ]. V., 1980, Biogeochemical cycling in an organic-rich coastal
marine basin. I. Methane sediment-water exchange processes, Geochim. Cosmochim.
Acta 44:471-490.
Myers, A. C., 1977, Sediment processing in a marine subtidal sandy bottom community. I.
Physical aspects, J. Mar. Res. 35:609-632.
Noshkin, V. E., and Bowen, V. T., 1973, Concentrations and distributions of long-lived
fallout radionuclides in open ocean sediments, in: Radioactive Contamination of the
Marine Environment, pp. 671-686, International Atomic Energy Agency, Vienna,
Austria.
Nittrouer, C. A., Sternberg, R. W., Carpenter, R., and Bennett,]. T., 1979, The use of Pb-210
geochronology as a sedimentological tool: Application to the Washington continental
shelf, Mar. Geol. 31:297-316.
Nixon, S. W., Kelly, J. R., Furnas, B. N., Oviatt, C. A., and Hale, S. S., 1980, Phosphorus
regeneration and the metabolism of coastal marine bottom communities, in: Marine
Benthic Dynamics (K. R. Tenore and B. C. Coull, eds.), pp. 219-242, University of South
Carolina Press, Columbia.
Okubo, A., 1971, Oceanic diffusion diagrams, Deep Sea Res. 18:789-802.
Olsen, C. R., Simpson, H. J., Ping, T.-H., Bopp, R. F., and Trier, R. M., 1981, Sediment mixing
and accumulation rate effects on radionuclide depth profiles in Hudson estuary sedi-
ments, J. Geophys. Res. 86:11020-11028.
Pearson, T. H., and Rosenberg, R., 1978, Macrobenthic succession in relation to organic
enrichment and pollution of the marine environment, Oceanogr. Mar. BioI. Annu. Rev.
16:229-311.
Peng, T. H., Broecker, W. S., and Berger, W. H., 1979, Rates of benthic mixing in deep-sea
sediment as determined by radioactive tracers, Quat. Res. (N.Y.) 11:141-149.
Powell, E. N., 1977, Particle size selection and sediment reworking in a funnel feeder,
Leptosynapta tenuis (Holothuroedea, Synaptidae), Int. Rev. Ges. Hydrobiol. 62:385-
408.
Pryor, W. A., 1975, Biogenic sedimentation and alteration of argillaceous sediments in
shallow marine environments, Geol. Soc. Am. Bull. 86:1244-1254.
Redfield, A. c., 1934, On the proportion of organic derivatives in sea water and their relation
to the composition of plankton, in: James Johnstone Memorial Volume, pp. 176-192,
University Press, Liverpool.
Rhoads, D. C., 1974, Organism-sediment relations on the muddy sea floor, Oceanogr. Mar.
BioI. Annu. Rev. 12:263-300.
Rhoads, D. C., and Stanley, D. J., 1965, Biogenic graded bedding, J. Sediment. Petrol.
35:956-963.
Rhoads, D. C., Aller, R. C., and Goldhaber, M. B., 1977, The influence of colonizing benthos
on physical properties and chemical diagenesis of the estuarine seafloor, in: Ecology
Effects of Macrobenthos on Marine Sediment Chemical Properties 101

of Marine Benthos (B. C. Coull. ed.). pp. 113-138, University of South Carolina Press,
Columbia.
Rhoads, D. c., Yingst. J. Y., and Ullman, W. J., 1978, Seafloor stability in central Long Island
Sound. Part 1. Temporal changes in erodibility of fine-grained sediment, in: Estuarine
Interactions (M. L. Wiley, ed .). pp. 221-244, Academic Press, New York.
Richards, F. A., 1965 , Anoxic basins and fjords, in : Chemical Oceanography 0. P. Riley and
G. Skirrow, eds.). Volume 1, pp. 611-645, Academic Press, New York.
Riedl, R J., Huang, N., and Machan, R, 1972,.The subtidal pump: A mechanism of interstitial
water exchange by wave action , Mar. Bio1. 13:210-221.
Rijken, M., 1979, Food and food uptake in Arenicola marina, Neth . J. Sea Res. 13:406-421.
Santschi, P. H., Li. Y. -H., Bell, J. J., Trier, R M. , and Kawtaluk, K., 1980, Pu in coastal marine
environments, Earth Planet. Sci. Lett. 51:248-265 .
Schafer, W., 1972, Ecology and Paleoecology of Marine Environments (G. Y. Craig, ed .; 1.
Oertel, translator). University of Chicago Press, Chicago.
Schink, D. R, and Guinasso, N. 1., Jr., 1977, Effects of bioturbation on sediment-seawater
interaction, Mar. Geo1. 23:133-154.
Schink, D. R, and Guinasso, N. 1., Jr. , 1978, Redistribution of dissolved and adsorbed
materials in abyssal marine sediments undergoing biological stirring, Am . J. Sci.
278:687-702 .
Schink, D. R, Guinasso, N. 1., Jr., and Fanning, K. A., 1975, Processes affecting the con-
centration of silica at the sediment-water interface of the Atlantic Ocean, J. Geophys.
Res. 80:3013-3031.
Self, R F. 1. , and Jumars, P. A. , 1978, New resource axes for deposit feeders? J. Mar. Res .
36:627-641.
Sepkoski, J. J., Jr. , and Bambach, R K., 1979, The temporal restriction of flat-pebble con-
glomerates: An example of co-evolution of organisms and sediments, Geo1. Soc. Am.
Abst. 11:256.
Sholkovitz, E. , 1973, Interstitial water chemistry of the Santa Barbara Basin sediments ,
Geochim. Cosmochim. Acta 37:2043-2073.
Smethie, W. M., Jr., Nittrouer, C. A., and Self, R F. 1., 1981 , The use ofradon-222 as a tracer
of sediment irrigation and mixing on the Washington continental shelf, in: Sedimentary
Dynamics of Continental Shelves (C. A. Nittrouer, ed.). pp. 173-200, Elsevier,
Amsterdam.
Smith, K. 1., Jr., 1978, Benthic community respiration in the N.W. Atlantic Ocean: In situ
measurements from 40 to 5200 m. , Mar. Biol. 47:337-347.
Sl:lrensen, J., 1978, Capacity for denitrification and reduction of nitrate to ammonia in a
coastal marine sediment, App1. Environ. Microbio1. 35:301-305.
Stumm, W., and Morgan, J. J., 1970, Aquatic Chemistry, John Wiley and Sons, New York.
Thayer, C. W., 1979, Biological bulldozers and the evolution of marine benthic communities,
Science 203:458-461.
Thistle , D. , 1979, Harpacticoid copepods and biogenic structures : Implications for deep-sea
diversity maintenance, in: Ecological Processes in Coastal and Marine Systems (R J.
Livingston, ed.), pp. 217-231, Plenum Press, New York.
Thompson, R K., and Pritchard, A. W., 1969, Respiratory adaptations of two burrowing
crustaceans, Callianassa californiensis and Upogebia pugettensis (Decapoda, Thalas-
sinidea). Biol. Bull. 136:274-287.
Thorstenson, D. c., 1970, Equilibrium distribution of small organic molecules in natural
waters, Geochim. Cosmochim . Acta 34:745-770.
Turekian, K. K. , Cochran, J. K. , and DeMaster, D. J. , 1978, Bioturbation in deep-sea deposits:
Rates and consequences, Oceanus 21:34-41.
Turekian, K. K. , Cochran, J. K., Benninger, 1. K., and Aller, R. C., 1980, The sources and
sinks of nuclides in Long Island Sound, Adv. Geophys. 22:129-164.
102 Chapter 2

Ullman, W. J., and Aller, R. C., 1982, Diffusion coefficients in nearshore marine sediments,
LimnoI. Oceanogr. 27:552-556.
Vanderborght, J. P., and Billen, G., 1975, Vertical distribution of nitrate in interstitial water
of marine sediments with nitrification and denitrification, Limnol. Oceanogr. 20:953-961.
Vanderborght, J. P., Wollast, R., and Billen, G., 1977, Kinetic models of diagenesis in dis-
turbed sediments. I. Mass transfer properties and silica diagenesis, LimnoI. Oceanogr.
22:787-793.
van Straaten, 1. M. J. U., 1952, Biogene textures and the formation of shell beds in the Dutch
Wadden Sea. I-II, Koninkl. NederI. Akad. Wet. Proc. Ser. B 55:500-516.
Westrich, J., 1983, Ph.D. dissertation, Yale University, New Haven, Connecticut (in
preparation) .
Whitlatch, R. B., 1974, Food-resource partitioning in the deposit-feeding polychaete Pecti-
naria gouldii, BioI. Bull. 147:227-235.
Yingst, J. Y., and Rhoads, D. C., 1980, The role of bioturbation in the enhancement of
microbial turnover rates in marine sediments, in: Marine Benthic Dynamics (K. R.
Tenore and B. C. Coull, eds.), pp. 407-422, University of South Carolina Press, Columbia.
Zeitzschel, B., 1980, Sediment-water interactions in nutrient dynamics, in: Marine Benthic
Dynamics (K. R. Tenore and B. C. Coull, eds.), pp. 195-218, University of South Carolina
Press, Columbia.
 II
Recent Freshwater Environments
Chapter 3
The Effects of Benthos on Physical
Properties of Freshwater Sediments
PETER L. McCALL and MICHAEL J. S. TEVESZ

1. Introduction ..... ... .... .......................................... ... .......... .. ........ .... .. ... ... ........................ 105
2. Freshwater Sediments and Macrobenthos ...... .................... .............................. ......... 106
3. Macrobenthos Life-Styles ........... .................... ....... ......... ..... .. .... .. .... ... ............... .. ... ... .. 113
3.1. Chironomids ...................... .. .. ................... ........ ..... .. ............ ... ....... ................. ..... . 113
3.2. Amphipods .................. ... .................... .... ......... ..... .. .......... ...... ........................... ... 116
3.3. Oligochaetes ..... ........... ... ... ................................................................................... 118
3.4. Bivalves ........... ... ..... ........ ............. ... .............. ....... .. .... .......... ..... ....... .. ... ... .. .......... 120
4. Mixing of Sediments .... ............ ....... ... ......... .. .... .. .. ..... .. ... .... ...... ... ......... .... .................. 124
4.1. Effects on Sediment Stratigraphy.......... .. ........ .......... ...... .. .. ... ......... ................ ... 124
4.2. Effects on Particle Size Distribution ................... .... ............. ... .... ..... ................... 139
4.3. Effects on Mass Properties of Sediments ........................................................... 146
5. Biogenic Modification of Sediment Transport ......... .... ...... ... .. ............................... ... 150
5.1. Field Observations .......... ...... ................. .. ............ .... ........ .. .................. .. .. ......... ... 150
5.2. Laboratory Experiments ... ........... ......... ....... ......... ............. .......... ........ .... ........... . 157
5.3. Some Comparisons ......................................... .. ................................................... 160
6. Ecological Interactions and Sediment Properties ...................................................... 161
6.1. Spatial-Temporal Variations of Fauna .............................................................. 161
6.2. Interactions among Macrofauna .......................................................................... 163
7. Future Work ............................................. .. .. ......................................... ...... ................. 166
References ................ ....... ........ .......... .... ........... ... ................... ...... .... ....... ....... ... .... ....... 168

1. Introduction
In this chapter we review the effects of freshwater benthos on the physical
(as opposed to chemical) properties of the bottom. Specifically, we will
focus our discussion on the effects of macrobenthos (adult length >1 mm)
on fine-grained bottoms (sediments that contain approximately 50% by

PETER L. McCALL • Department of Geological Sciences, Case Western Reserve Univer-

sity, Cleveland, Ohio 44106. MICHAEL]. S. TEVESZ • Department of Geological Sci-
ences, Cleveland State University, Cleveland, Ohio 44115.

105
106 Chapter 3

weight silt-clay-sized particles) of lakes and slow-flowing rivers. There

are three reasons for this approach. The first is that there is simply too
little known about freshwater meio- and microbenthos to merit a review
of their effects on sediment properties. More importantly, the macroben-
thos are probably the most potent modifiers of sediment properties by
virtue of their size relative to sediment grains, their population density,
their ability to move through a relatively large volume of sediment, and
their feeding and respiratory habits. Finally, we have restricted ourselves
to fine-grained bottoms because these comprise the bulk of freshwater
lake sediments and because the structure of this sediment is more easily
altered than is that of sediments with larger grain size. Some information
from slow-flowing river bottoms is included with lakes because their
sediments and faunas are very much alike (Hynes, 1970).
We will first describe some typical macrobenthic communities of
lakes and isolate some important organisms for study. Next we will look
at the biogenic modification of sediment grain size, mass properties, or-
ganic matter, stratigraphy, and erodibility by these organisms. Finally, we
will look briefly at the temporal effects on sediment properties produced
by succession and seasonality and suggest some directions for future
research.

2. Freshwater Sediments and Macrobenthos

In a recent review of sedimentary processes in lakes, Sly (1978) de-
lineated three major differences between lacustrine and marine environ-
ments. These are:
1. The smaller size of most lakes compared to oceanic environments
means that long-period waves are less frequent and that bottom
energy levels are below those of oceanic systems; for this reason,
well-sorted coarse sands and gravels are confined to the shallowest
portions of larger lakes and the average grain size of lake bottoms
is smaller than that in shelf environments (Fig. 1).
2. Because lakes are closed systems with respect to sediment trans-
port, and because of the usually larger ratio of drainage basin to
lake area, sedimentation rates in lakes are higher (by about a factor
of ten) than in many marine environments.
3. Lakes are almost tideless, so that intertidal areas are nearly absent.
In many respects, except for major differences in salinity fluctuations and
tidal motions, the shallower regions of large lakes resemble nearshore
marine estuarine environments, particularly in their often mutually high
Effects of Benthos on Freshwater Sediment Physical Properties 107

•
BAFFIN BAY
350

300

A TLANTIC SCOTIAN

250 SHELF ~

STRAITS

Vl \t";CKO<AC ,,••
,.,,.
II:
~ 200

••.* .,,'
w
~
~

1: t .1
~ 150
w , GREAT LAKES
a
t:,¥" MAJOR SUBBASINS
100
CLAIR J. 'I

I~,: • .j.,
50 __ .,'
EKALUGAD
F~RD
"I.I·I'
.,'
I,'~ BALTIC SEA

/
LAKE ERIE EASTERN BASIN

I.
.*: - ,
•
..J. '
"''It·
.J
~
SHALLOW LAKE ERIE
SEDIMENTS
• ." , _J.
o +-....,-..,..--,,..-...,....,.-;.;-....-----..,..--,,..-....- ___- - -
o 2345678910

MEAN SIZE ~ UNITS

Figure 1. Particle size-depth relationships in aquatic environments. Redrawn from Sly (1978).

sedimentation rates and dominance of silt-day-sized sediment. The num-

ber of important mineral phases in lakes is not large. Lacustrine sediments
are mostly quartz, feldspar, calcite, dolomite, the day minerals illite and
smectite, apatite, and iron sulfides and oxides (Jones and Bouser, 1978).
There is a wider range of organic matter in muddy lakes (gyttja, with
mostly autocthonously produced organic matter, as opposed to dy, or
peaty allocthonous sediment)-from dose to zero up to about 50% organic
matter by weight (Wetzel, 1975)-than in most marine environments,
where organic carbon content is typically less than 10% by weight.
108 Chapter 3

Soft-bottom marine communities are dominated in terms of numbers

and biomass by polychaete annelids, bivalve molluscs, and amphipods
(Sanders, 1956, 1968; Sanders and Hessler, 1969). The soft bottoms of
freshwater lakes and slow-flowing rivers are dominated by chironomid
insect larvae and oligochaete annelids that together appear to fill a niche
similar to that of the marine polychaetes, bivalve molluscs of two types,
and amphipods, not necessarily all together or in that order of importance.
Although lakes are variable bodies of water, examples will show some
typical patterns of community composition and distribution.
Lake Esrom is a well-studied moderately eutrophic lake located in
Denmark. It has an area of 17.3 km 2 , is about 8 km long and 2-3 km wide,
and has a maximum depth of 22 m. Rooted vegetation occurs in a relatively
flat narrow belt along the shore to a depth of about 6 m (28% of total lake
area). Steep slopes are present from 6-10 m with less steep slopes below
10 m. The bottom of the lake below 10 m (52% of total lake area) is covered
with silt-clay sediment, and at the greatest depth little oxygen is present
in the water from July to November (Jonasson, 1972). As in many other
lakes, sedimentation rate and organic matter content of the sediment in-
crease with depth below the epilimnion (Fig. 2).
In most lakes at least two macrofaunal communities can be distin-
guished: a shallower littoral community whose boundaries are deter-

SEDIMENTATION -2 -1
E 0 0 .5 1.0 1.5 2.0 g m d
~
J:
t
~ 5

10

ORGANIC MATTER
15 IN SEDIMENT

I
20

0.1 0.2 0.3

ORGANIC MATTER

Figure 2. Sedimentation rates and organic matter concentrations in Lake Esrom. Redrawn
from Jonasson (1978).
Effects of Benthos on Freshwater Sediment Physical Properties 109

2m DEPTH 20 m D EPTH
'" ..,.
...... ......
~......-.,.......,. ~c:s;dIIrr~~~~

~ 000
~.,.... ........ .....,. .,-:;:p~~~~

.,..... ~ ~~,,-:;,~~'
....., ................ ~ ..... ~.-..r~~
iii @I@@ oe", --=r~ ..-... .-. ..-..-. ~ ......... ~~

II! <a~<4~~cdc;~ca<a~~~<Q~~<:iQ~ <4 ........ ~ ....-........., ......... ~.....-..;o'~.-.-,

--.,. ...... ..-.,...-..:r ~ ~~~~

~~QQ~Q~Q~~~~~Q~~~~@~a ....... ...............

......., .,-:,. . . . . . . ~ ......... ~~~4lD'Q'
..",.... .--..~~~~
~O O & D©Q O OC')OO ........ .........
....
--=--~--.,.~~~~

... ............... ~c:=r:-v.....,.~~~~

>w >* >*t >* >It ...'"

~~~..-JJI~~C:-W-~~

)1M Q ~,.-w~~~~~~~
___ .............
)C ~ ~ >lit >'* )C:
~ ~~~-ca-:r ~

~ ~c--.~~~~~~~

'"~ ~~~;C~~ I/)

:I
~
~ """""~~~~~~pPIIf:f
~~~.-,,~~~~~
.........
;:c ~~...-r ~~~"""",~

!E>W~>C?lV#~~
....,,,.....-..,...,...,~~...-;:r~~
~ ~...-. ................. .-..::r~~....-.:::::r

~ >« ~ ~ >"* >1*

W
Q QO"QC CII IQ()'OOOOOOOf)OCCO(:lO
oo.OooC~.QCCO~~OQ~o oo e

~~*'II:k"~""_"'''''~ o o o a oo c aCOOQoooo.ce o ~

~ ~~_""'+'JIIH;Q.+"
UJ~~~~~.......-rr
O oo o o o o o a 006000QOO
....... "-""
f(!IIIfW I((!(({(\\/Ill)( 1111(( 1{()(j(111Wi IllfI(i II
'" ~lmlmllm)))l}l))))ll!!})@))) ~~ (~! Ill{!1 (II!llllll) jJllll)ll({I{(II (1IIIIl!! l((\ II!{II
~ ~t:.~*J:.t:* ~~~ ~ I--- 0'.00.0 0 0 < ) 0 0 0 0 0 0 0 0

:;C 45&- ~= o ~~)ooC;I-oI~lOood)0c-4

_ _
:t
~~~cw::a.~ ~._...~.__. _

e~ p=
:: ....-..,.ansn.c<@ii'iI>~....a2iIl>~~ ...
Q
.-...,...~~ ~ _.c;IIOIoa c_ooo_... ..,...,~~

: L - i -________________________
-~~-=- -- ~--.~~..-.~ --J

A B
Figure 3. (A) Littoral (2 m) community of Lake Esrom. Numbers per 0.02 m 2 • Herbivores:
Centroptilum luteolum , Micronecta minutissima , Lymnaea pereger, Gyraulus albus, Valvata
piscinalis, Bithynia tentaculata , Oxyethira flavicornis, Planorbis planorbis, Armiger cristo ,
and Eurycereue lamellatus. Detritus feeders: Dreissena polymorpha, Canthoeamptus sta-
phylinus, Asellus aquaticus, Caenis moesta, Chironomidae, Pisidium sp., and Oligochaeta
and Ceratopogonidae. Predators: Hydraearina, Helobdella stagnalis, Glossophonia com-
planata, Herpobdella octoeulata and H. testacea, Tanypodinae, Candona candida and C.
negleeta, Polyeelis tenuis , and Leptoceridae. (B) Profundal community (20 m) of Lake Esrom.
Numbers per 0.02 m 2 . Detritus feed ers : Chironomus anthracinus, Pisidium casertanum and
P. subtruneatum, and Potamothrix hammoniensis and coccoons . Predators: Proc1adius pee-
tinatus and Chaorborus flavieans. Redrawn from Jonasson (1978) .

mined by the shore of the lake and the depth to which rooted vegetation
extends, and a deeper profundal community that encompasses the lake
bottom below the zone of rooted vegetation. About 300 benthic metazoan
species are found in the littoral macrophyte community of Lake Esrom
(Jonasson, 1978). The relative abundance of macrofauna I dominants is
shown in Fig. 3A. The abundance of macrophytes in this shallow envi-
ronment gives rise to a multitude of micro environments and thus high
species diversity and a wide range of trophic types, notably grazing her-
bivores and carnivores in addition to deposit and suspension feeders. In
the muddy profundal of Esrom (>10 m), diversity drops to 20 species,
110 Chapter 3

Table la. Relative Abundances and Biomass of Different

Animal Groups in Zoo benthos Samples Taken from Lake
V6rtsjarv in July-August, 1965°
Relative abundance
(percent)

Sublittoral
Littoral and profundal

Chironomidae 35.8 64.5

Oligo chaeta 9.4 25.1
Mollusca
Small
Gastropoda 8.6 2.8
Pisidiidae 8.0 1.2
Big clams
Unionidae 0.3 0.1
Dreissena 2.3
Varia
Turbellaria <0.1 0.1
Nematoda 1.6 1.0
Hirudinea 4.6 0.3
Asellus 16.9
Gammarus 0.5
Araneida 0.1
Hydracarina 2.0 4.4
Ceratopogonidae 4.1 0.3
Diptera (others) 0.3
Ephemeroptera 0.8 0.1
Trichoptera 2.9
Lepidoptera <0.1
Coleoptera 1.4
Heteroptera <0.1
Sialis <0.1

Biomass

Total abundance m - 2 2270 907

Total weight (g . m 2 ) with large Mollusca 20.08 9.42
Total weight (g . m 2 ) 41.69 14.50

a From Timm (1975).

and the dominant species are all deposit and/or suspension feeders (Fig.
3B). Chironomid insect larvae, tubificid oligochaetes, and sphaerid bi-
valves are the most abundant organisms here. Jonasson (1972, 1978) de-
fines two other communities in Lake Esrom that are not found in all (or
even most) lakes: a muddy sublittoral community (6-10 m) distinguished
primarily by the overwhelming dominance in terms of numbers and bio-
Effects of Benthos on Freshwater Sediment Physical Properties 111

Table lb. Relative Abundance of the Bottom Fauna at 30

and 50 m Depth in the Central Part of Lake Miilaren,
1969-1976a

Relative abundance
(percent)

30 m 50 m

Crustacea 4.0 52.2

Pontoporeia affinis
Diptera, Chironomidae 27.5 6.5
Procladius spp.
Micropsectra spp.
Tanytarsus spp.
Phaenopsectra coracina
Monodiamesa sp.
Mollusca 6.3 4.8
Pisidium personatum
P. casertanum
P. henslowanum
P. subtruncatum
Oligochaeta 57.4 35.4
Potamothrix hammoniensis
Limnodrilus hoffmeisteri
Tubifex tubifex
T. ignotus
Aulodrilus pluriseta
Peloscolex ferox
Nemertini 2.5 0.7
Prostoma sp.
Turbellaria 2.1 0.4
Other taxa 0.1 0.1

a From Wiederholm (1978), O.6-mm sieve mesh.

mass of the large freshwater bivalve Dreissena polymorpha, and a gravel

surf zone community. Dreissena is restricted to Eurasian lakes; in North
America we find corbiculid and the larger unionid clams taking its place,
although they are not necessarily present as a distinct depth-defined com-
munity dominant.
Timm (1975) described the benthic macrofauna of a larger (270 km 2 )
but shallower (average depth 2.7 m) eutrophic lake in Estonia, Lake
V6rtsjarv. The littoral region, 0-1.5 m, covers 15% of the lake area, while
the profundal, 1.5-6 m, makes up 62% of the lake area. Once again we
see the dominance of chironomids, oligochaetes, and molluscs (Table Ia)
and the increasing importance of chironomids and oligochaetes in the
profundal zone (Fig. 4). Species lists from the profundal of the larger
oligotrophic Great Slave Lake in Canada and Lake Malaren in Sweden
112 Chapter 3

Table Ie. Comparisons of Relative Abundances and

Biomass of Organisms at Depths of 0-100 m in the Delta
Region and in the Main Part of Great Slave Lakeo

Relative abundance
(percent)

Organism Main lake Delta

Amphipoda 73 .6 60.8
Sphaeriidae 4.7 2.0
Oligo chaeta 10.5 31.2
Chironomidae 4.8 4.0
Ostracoda 4.2 0.9
Gastropoda 0.7 0
Nematoda 0.3 0.8
Miscellaneous 1.1 0.3

Biomass

Total organisms m - 2 1539 2328

Total weight (g . m 2 ) 2.5 3.29

a From Rawson (1953).

(Table Ib,c) illustrate the importance of these groups and also the occa-
sional dominance of crustacean amphipods.
The lessons we should take away from these descriptions are that
most of the bottom of most lakes is covered with fine-grained sediments
(silts and clays) with varying amounts of organic matter and that the

91m 2
D CH'"OHQM' OA.
32
spJm . OUGOCHAETA 30
2500 o MOlLUSCA 25
• OTHER
2000 20
1965
1500 15
1000 10
500 5
0
sp.lm 91m2
1500 15
1000 10
500 5
0 0
A B C 0 F
Figure 4. Vertical distribution of benthos in Lake V6rtsjarv. Redrawn trom Timm (1975) .
Effects of Benthos on Freshwater Sediment Physical Properties 113

dominant organisms on these bottoms in terms of numbers and biomass

are typically chironomids, oligochaetes, amphipods, and bivalves-a
small variety (pisidiids) and a large variety (unionids).

3. Macrobenthos Life-Styles
We will examine the activities of macrobenthos that affect sediment
properties in greater detail later, but it will be useful here to describe
briefly the life positions, feeding, mobility, and life histories of important
macrobenthos (see also Chapter 4).

3.1. Chironomids

The Chironomidae (common name "midges") are a family of dipteran

insects whose closest familiar relatives are the mosquitoes. The family is
distributed worldwide, mostly in aquatic environments. Members of the
subfamily Chironominae and genus Chironomus are particularly abun-
dant in the profundal zone of lakes. The life history of the group has been
reviewed by Oliver (1971). Most of the life cycle is spent in the larval
stage (Fig. 5), which lasts from several weeks in some tropical forms to
2 years in a few more northerly species. The adult flying stage lasts from
a few days to a few weeks at most and is used for dispersal, mating, and
oviposition. All the energy required to complete the life cycle is obtained
in the larval stage, because the adults do not feed. Most temperate species

INSTAR 1
d 2 3 4

Figure 5. Relative size of instars of Chironomus anthracinus. Redrawn from Jonasson (1972).
114 Chapter 3

b
/-.

Figure 6. (A) Resin case of U-shaped tubes of Chironomus plumosus. Scale in mm. Repro-
duced from McLachlan and Cantrell (1976) with permission of the publisher. (B) Chironomid
burrows from Lake Erie sediments. Depth of sediment column: 10 cm. (C) Conical burrow
of C. anthracinus from Lake Esrom. Reproduced from Jonasson (1972) with permission of
Gikos.
Effects of Benthos on Freshwater Sediment Physical Properties 115

Figure 6. (cont'd)

have one or two generations per year, and a few have one generation every
2 years. Eggs are deposited in water and larvae develop in the lake bottom,
where they undergo four molts (instars) before pupating and emerging
from the lake as flying adults. The size and weight of larval chironomids
depends on a variety of factors such as temperature and food supply, but
fourth-instar chironomids are likely to be 10-22 mm in length and weigh
1-2 mg dry (Jonasson, 1972; Mackey, 1977; Hilsenoff, 1967; M. S. Johnson
and Munger, 1930).
Larvae burrow into bottom sediments; most larvae live in the upper
8-10 cm of the substratum, with occasional occurrences at 40- to 50-cm
depths (Cole, 1953; Ford, 1967; Hilsenoff, 1966). The burrowing and feed-
ing behavior of a few chironomid species has been studied. Most chiron-
omids live in tubes lined with a transparent, fibrous salivary secretion.
It is thought that the very abundant and widespread species C. plumosus
typically constructs a U-shaped tube in the bottom (Fig. 6A) and filter
feeds by drawing water by means of anterioposterior undulations of its
body through a net spun in the lumen of the tube (Walshe, 1947, 1951).
But C. plumosus feeds in another very different way as well. Larvae may
perform as surface deposit feeders, scraping the top 1-2 mm of sediment
around their burrows for detritus and bacteria attached to fine-grained
sediment particles; they may also occasionally feed on pieces of higher
vegetation (Konstantinov, 1969). And not all burrows are U-shaped. Some
are straight tubes; some are of more complex designs (Fig. 6B) (McLachlan
and Cantrell, 1976).
116 Chapter 3

Table II. Faunal Groups Captured in Bottom Emergence Trap,

Western Basin, Lake Erie, 1979 Q

No. m - 2 captured No. m - 2 captured

28 August-26 September 26 September-17 October
Faunal group (29 days) (21 days)

Oligo chaeta (Naididae) 6 2

Chironomidae 50 (+ 112 pupae) 13
Nematoda o o
Bivalvia (Sphaeridae) 9 4
Amphipoda 246 110

Q Area = 0.46 m'.

The amount of time and effort allocated to one feeding method or the
other is not well known, but it is fairly well established that the bulk of
the food of mud-dwelling C. plumosus comes from bottom sediment
(Konstantinov, 1969; McLachlan, 1977). Burrowing behavior and feeding
method may be under environmental control. For instance, we have ob-
served simpler tube structure and evidence of more surface deposit feed-
ing (greater abundance of sticky, salivary threads extending radially from
tube openings) by C. plumosus in Lake Erie in sediments with a'very high
water content (>85% water by weight) and during low-oxygen conditions
under the summer hypolimnion. Jonasson (1972) described the burrowing
and feeding behavior of a related profundal species, C. anthracinus. This
species normally inhabits a conically shaped tube (Fig. 6C) that extends
above the interface like a chimney. During the summer stagnation period
in Lake Esrom, C. anthracinus feeds on surface deposits in a circular area
around its tube openings; when oxygen disappears from the water, feeding
stops and the "chimney" decays; when overlying water is rich in oxygen
and phytoplankton, the larvae apparently filter feed and construct small
fanlike extensions of their tubes that may close the tube under storm
conditions and prevent the entry of sediment into the burrow.
Mud-dwelling chironomids are very mobile, leaving the bottom
(usually at night) and undergoing migration into the water column (Mun-
die, 1959). C. plumosus is one of the most mobile benthos in Lake Erie,
as measured by its ability to leave the bottom and swim to a height of 1
m above the bottom (Table II). The individual frequency of reburrowing,
however, is unknown.

3.2. Amphipods
There are several hundred species of these freshwater crustaceans
(see Fig. 7 for morphology), but only a few-Hyalella azteca, several
Effects of Benthos on Freshwater Sediment Physical Properties 117

species of Gammarus (e.g., fasciatus, pseudolimnaeus, lacustris), and

Pontoporeia hoyi (= affinis)-are widespread and abundant constituents
of soft-bottom lake communities. Hyalella and Gammarus are usually
found in the warmer and shallower regions of lakes (Bousfield, 1958;
Hargrave, 1970a), while P. hoyi is more abundant in the deeper and colder
parts of lakes in northern Europe, Asia, and North America, where the
average temperature is less than 20°C (Bousfield, 1958; Smith, 1972). P.
hoyi is the most abundant and productive component of the macrobenthos
in the deeper regions of some northern lakes, including the upper Great
Lakes (Juday and Birge, 1927; Eggleton, 1937, 1939; Adams and Kregar,
1969; Henson, 1970; Mozley and Alley, 1973; Freitag et aI., 1976) . Pon-
toporeia is also found in shallower regions, especially in arctic and sub-
arctic lakes and rivers (J. W. Moore, 1979).
Adult amphipods are typically 5-10 mm long and weigh 1-2 mg dry
(Juday and Birge, 1927; Hargrave, 1970a; Alley and Chin, 1978). Their
population density in lakes ranges from a few hundred to over 14,000 per
square meter (Marzolf, 1965; Kraft, 1979). The life cycle is variable, de-
pending on food, temperature, and oxygen conditions, as well as on the
particular species under consideration. Gammarus species typically live
for less than 1 year, mature rapidly, and produce up to six groups of eggs
during their reproductive phase; gammarids are in reproductive condition
in all months except October-January in north temperate regions (Macan
and Mackereth, 1957; Hynes, 1955). P. hoyi is unusual, because its life-
span is about 2.5 years and reproduction does not begin until about 2 years.

Figure 7. The amphipod Pontoporeia (x 9). Modified from Pennack (1978).

118 Chapter 3

The female produces one brood, usually but not always in winter, and
then dies (Juday and Birge, 1927; Segerstrale, 1967).
While most freshwater amphipods are labeled as scavengers (Pennak,
1978), most profundal amphipods probably derive most of their nutrition
from bacteria and algae obtained from surface deposit feeding (Marzolf,
1965; Hargrave, 1970b,c). In soft muds, amphipods live in simple burrows
in the top 2 cm of sediment with the majority in the top 1 cm (Hargrave,
1970a; Krezoski et al., 1978; Heuschele, 1980). Like chironomids, they
are quite mobile and undergo extensive and frequent migrations into the
overlying water column (Table II) (Mundie, 1959).

3.3. Oligochaetes

Oligochaetes are typically bilaterally symmetrical, segmented, her-

maphroditic annelids. Eleven families of oligochaetes are found in fresh-
water environments (Brinkhurst and Jamieson, 1971), but only three-
the Tubificidae, Naididae, and Lumbriculidae-are widespread and abun-
dant in lakes. Of these, the Tubificidae are easily the most important, if
importance is measured by widespread distribution, population density
and biomass, and ability to alter sediment properties. Along with chiron-
omids, tubificid oligochaete worms are usually the dominant macrofauna
in lake profundal regions. Population densities are variable and may range
from a few hunmdred individuals per square meter to several million
individuals per square meter (Palmer, 1968). Higher population densities
are associated with fine-grained sediment (Fig. 8), higher inputs of organic
matter to sediments, and high temperatures (Brinkhurst and Jamieson,
1971; Ladle, 1971; M. G. Johnson and Brinkhurst, 1971; Birtwell and

a: •
w
~
c(
600
w 500
m
......
en 400
:Ii
a: 300
0 37 .8&1

~ 200 " D
1.1. 8174 9388 .. '
,"
0 100 1783 JlJ4g ~~~~~ .. .. ~ O- ------- Q
'
"88
0 0 "
z M A M J J A S 0 N 0 J
1967
Figure 8. Numbers of Tubifex tubifex in sand (0) and mud (e) cultures. Numbers represent
dry weight of worms in micrograms. Redrawn from Ladle (1971).
Effects of Benthos on Freshwater Sediment Physical Properties 119

Figure 9. Oligochaete burrows in Lake Erie sediments. Depth of sediment column: 15 cm.

Arthur, 1980). Individual dry weights range from less than 0.2 mg per
mature individual (Jonasson and Thorhauge, 1976b) to over 3.4 mg per
individual (Ladle, 1971; McCall et a1., 1979). Organism length is hard
to measure, since oligochaetes are capable of wide variation of extension
and contraction. Contracted lengths of preserved worms are typically 2-5
cm (e.g., Birtwell and Arthur, 1980), although lengths up to 15 cm have
been reported (Caspers, 1980).
Tubificid life histories are poorly known because it is often impos-
sible to identify field collections of immature tubificids to the species
level (Brinkhurst and Jamieson, 1971; Bonomi and DiCola, 1980). The life
history of even a single species is variable from locality to locality de-
pending on temperature and productivity, among other things. Higher
temperatures and high levels of food decrease the time required to reach
maturity and lengthen the breeding period (Kennedy, 1966; Bonomi and
DiCola, 1980).
Some tubificids take 2 years to reach maturity and breed for a short
period during the summer months (Brinkhurst, 1964). Other species like
the cosmopolitan species Limnodrilus hoffmeisteri and perhaps Tubifex
120 Chapter 3

tubifex have variable life histories, but usually reach maturity in a year
or less and often breed all year around (Brinkhurst and Kennedy, 1965;
Kennedy, 1966). There is one report of 1. hoffmeisteri having four or five
generations a year in a eutrophic reservoir in South Wales (Potter and
Learner, 1974). The majority of breeding worms appear to die after re-
production, but some worms are capable of resorbing sexual organs and
repeated breeding (Brinkhurst and Jamieson, 1971; Poddubnaya, 1980).
A few tubificids and most Naididae reproduce asexually.
Tubificids are infaunal burrowers (Fig. 9). They are found mostly in
the upper 20 cm of sediment, but they feed primarily in the top 2-8 cm
(Sorokin, 1966; Davis, 1974; McCall and Fisher, 1980). Tubificids typically
live and feed head down in the sediment. Some portion of the posterior
of the worms may project above the sediment-water interface. The pos-
terior is sometimes waved about under low-oxygen conditions for the
purpose of respiration. The worms selectively ingest silt and clay particles
at depth and feed on attached microflora, primarily bacteria (Brinkhurst
and Chua, 1969; Brinkhurst et a1., 1972; McCall et al., 1979). Fecal pellets
are deposited at the sediment-water interface. Tubificids move about in
sediment and are able to move laterally for short «1 m) distances in
laboratory microcosms, but they do not swim actively above the interface
in the manner of chironomids and amphipods (Table II) (Brinkhurst,
1974). Most lateral movement must be due to passive current transport
of adults and coccoons, which are millimeter-sized, ovoid egg containers
dropped into the sediment by breeding tubificids.

3.4. Bivalves

3.4.1. Unionacea
The Unionacea (particularly Unionidae and Margaritiferidae) are dis-
tributed worldwide, but they are most abundant and diverse (about 220
species) in eastern North America in the Mississippi River drainage basin
(Pennak, 1978). Unionids are found on all substratum types from gravel
to mud and in both rivers and lakes (R. 1. Johnson, 1970; Tevesz and
McCall, 1979). While it is true that unionids are typically found in quite
shallow water only a few meters deep, large populations have also been
recorded from deeper profundal muds. Wood (1953), for example, found
abundant unionids in the deepest part of the western basin of Lake Erie
at depths of 12 m.
Unionaceans are the largest members of the permanent freshwater
infauna. Adult lengths range from 4 to more than 14.6 cm, with average
lengths of about 7 cm (e.g., Matteson, 1948; Okland, 1963; Tudorancea,
Effects of Benthos on Freshwater Sediment Physical Properties 121

1969; Crowley, 1957; Hendelberg, 1960). Wood (1953) collected 14 species

of unionids from western Lake Erie and found that average individual
wet weights of species ranged from 1 to 25 g; the average for all species was
about 10 g. J. B. Fisher and Tevesz (1976) found the average dry weight
of Elliptio comp1anata in Lake Pocotopaug, Connecticut, to be about 1
g per individual (average length 7.4 cm). Population densities range from
a few tens of individuals (Wood, 1953; Okland, 1963; J. B. Fisher and
Tevesz, 1976) to several hundred per square meter (Tudorancea, 1969).
As a consequence, unionids typically comprise 90% or more of the stand-
ing crop biomass of the infaunal macrobenthic community in which they
occur (Tudorancea and Gruia, 1969; Wood, 1953; Okland, 1963; Negus,
1963; J. B. Fisher and Tevesz, 1976).
Among unionaceans sexes are separate, and the eggs of the female
are fertilized by sperm suspended in the water column and drawn into
the mantle cavity by the suspension-feeding female. Fertilized eggs are
carried in the gills, which function as a brood pouch (LeFevre and Curtis,
1912). Valved glochidia larvae are expelled from the brood pouch either
4-6 weeks after fertilization in short (summer) breeders or several months
later in long (winter) breeders, in which eggs are fertilized in the latter
part of the summer and carried through the winter to be discharged in
the spring and summer (LeFevre and Curtis, 1912; Matteson, 1948; Yokley,
1972). A single female may release a few thousand to over two million
glochidia larvae (Coker et a1., 1921). Glochidia larvae settle to the bottom,
where they die unless taken into the mouth of (hookless glochidia) or
brushed by (hooked glochidia) the proper species of fish. Glochidia attach
themselves to the gill filaments or sometimes the fins of fish (several
hundred to several thousand glochidia per fish), where they develop for
2-3 and, exceptionally, 4 weeks, after which they drop from the fish as
juveniles and take up their life on the bottom (LeFevre and Curtis, 1912;
Matteson, 1948; Coker et a1., 1921). Unionaceans grow rapidly in the first
2-5 years of life, after which growth slows markedly. Juveniles mature
in 2-5 years, may reproduce once per year for many years, and live to be
from 5 to 116 years old; most species live 15-40 years (Stansbery, 1967;
Haranghy, 1971; Yokley, 1972; LeFevre and Curtis, 1912; Stober, 1972;
Hendelberg, 1960; Crowley, 1957). Growth rates and longevity are of
course dependent on many factors, but where temperatures are held
approximately constant, growth rates of some species decline in deeper
water and finer sediments, while variability of growth rate decreases and
longevity increases (Fig. 10).
Unionaceans are filter feeders, so their feeding activity does not mix
sediments in the way that the feeding activity of oligochaetes does. The
typical life position of a unionid on soft sediments is shown in Fig. 11.
Life position, however, is variable. Some individuals live with their an-
122 Chapter 3

HABITAT
(SUBSTRATE)

IN

Figure 10. Unionid growth rates as a function of sediment type. Redrawn from Stansbery
(1970).

terior margin and apertures projecting a centimeter above the bottom,

while others have been reported living up to 20 cm below the sedi-
ment-water interface (McCall et al., 1979; Van Cleave, 1940; Baker, 1928).
While the food of unionaceans has been reported to consist of zooplank-
ton, phytoplankton, and organic detritus (Pennak, 1978; Coker et al.,
1921), the proximity of the inhalent aperture to the sediment-water in-
terface on soft, easily resuspended bottoms probably means that nutrition
is also obtained from bacteria growing on ingested fine-grained sediments
in a manner similar to that of the brackish-water filter-feeding bivalve

---:-_"'--..~.~. ". " ':. ~

.~~~"""-

Figure 11. Lampsilis radiata in infaunallife position. A. sediment unaffected by burrowing

activities; B,C, zone of high water content produced by burrowing activity; D, depression
in sediment-water interface produced by burrowing activity. Scale bar: 2 em. Reproduced
from McCall et a1. (1979) with permission of the Journal of Great Lakes Research.
Effects of Benthos on Freshwater Sediment Physical Properties 123

Rangia cuneata, described by Tenore (1968). At any rate, copious amounts

of mucus-laden sediment are deposited on the sediment surface as pseu-
dofeces by the unionid Lampsilis radiata in Lake Erie (McCall et al.,
1979). The clams are quite active at temperatures above lOOC (Stansbery,
1970) and frequently move about on top of and in the top 10-15 cm of
surface sediment (McCall et aI., 1979).

3.4.2. Pisidiidae
The Pisidiidae (formerly Sphaeriidae) are considerably different from
the Unionacea. Individual lengths (2-20 mm) and dry weights (0.5-2 mg)
of pisidiids are much smaller than those of unionaceans (Foster, 1932;
Thomas, 1965; Mackie et aI., 1976; Heard, 1977; Jonasson, 1972). Popu-
lation densities of pisidiids in the profundal zone of lakes are higher than
those of unionaceans, typically ranging from a few hundred to a few
thousand (2600) individuals per square meter. Pisidiids occur over a wider
range of depths and they reach their maximum abundance at greater
depths than unionaceans (Jonasson, 1972; Thut, 1969; Nalepa and
Thomas, 1976; Hamilton, 1971). In Lake Michigan, for example, unionids
occur no deeper than 30 m and are typically found at depths of less than
10 m (Reigle, 1967), while pisidiids reach their maximum abundance at
40-50 m and occur to depths of 225 m (Eggleton, 1937).
The life history of pisidiids has been recently summarized by Heard
(1977). Individuals are hermaphroditic. Fertilization is internal; young
are incubated in gill structures and are released as miniature adults. There
is no parasitic larval stage as there is in unionids. Relatively few young
are brooded at one time-usually 2-4 young in the genus Sphaerium,
more in Sphaerium species from temporary habitats, and up to 20 in the
genus Pisidium-and young at anyone time are typically of different
sizes and at different stages of development. Most individuals live 1.5-2
years and produce only one brood; some species live 3-4 years and some
produce two broods. Young are produced throughout the year, although
a particular species may reach a peak in reproduction in the summer or
winter months.
Many pisidiids are filter feeders on detritus, bacteria, algae, and pro-
tists (Monakov, 1972; Meier-Brook, 1969), but deposit feeding by common
and abundant species using the ciliated part of the bivalve foot to collect
fine-grained particles has also been reported (Mitropolsky, 1966; Mona-
kov, 1972). Most pisidiids are infaunal. Some species live in "normal"
bivalve position, with siphons just below the interface with the ventral
portion of the shell and foot pointing down into the sediment (Meier-
Brook, 1969; McCall, 1979). Other pisidiids of the genus Pisidium have
the unique life position shown in Fig. 12. The clam burrows into the
124 Chapter 3

Figure 12. Life position of Pisidium (right) and Sphaerium (left). Arrows indicate feeding
and respiratory currents. Scale bar: 5 mm. Adapted in part from Meier-Brook (1969).

bottom at a steep angle and then turns parallel to the interface and forms
a canal several millimeters below the interface. This becomes the exhalent
canal, water being drawn into the clam from surrounding interstitial
spaces (Meier-Brook, 1969). Jonasson (1972) records a similar life position
in a horizontal burrow for the common species P. casertanum and further
adds that burrows appear to be concentrated at the boundary between
oxidized and reduced sediments. Most pisidiids are indeed found in the
upper 2 cm of sediment, but Gale (1976) found that young and newborn
S. transversum are able to burrow to at least 16 cm below the sedi-
ment-water interface. Not all young burrow deeply, and those that do
become metabolically inactive and can live for more than a month in
anaerobic sediments. This deep-burrowing behavior is not unique to S.
transversum and may be an adaptation to escape temporarily unfavorable
circumstances at the interface (Gale, 1976).

4. Mixing of Sediments

4.1. Effects on Sediment Stratigraphy

4.1.1. Kinds of Mixing and Factors Affecting Mixing

Infaunal macrobenthos can mix the sediments in which they live by
either their feeding activities or their burrowing activities or both. Feeding
by freshwater deposit feeders can mix sediments by highly ordered or
random movement of sediment particles. The deposit feeding tubificid
Effects of Benthos on Freshwater Sediment Physical Properties 125

oligochaetes provide the best example of ordered mixing of freshwater

sediments.
Tubificids usually feed head and mouth down in the sediment with
their posterior end at or above the sediment-water interface. Sediment
particles and associated bacteria are ingested at depth. Bacteria are di-
gested on their passage through the gut (Wavre and Brinkhurst, 1971;
Brinkhurst et aI., 1972), whereupon sediment particles are packed into
mucus-bound fecal pellets and deposited at the sediment-water interface.
Thus the worm receives a food supply constantly renewed by gravity.
Moreover, the food supply is not diluted with recently digested sediment,
and the worm can remain sedentary yet be fairly well protected by sed-
iment from predators. Mixing by such feeding has been labeled "conveyor
belt" mixing by Rhoads (1974) and convective mixing by J. B. Fisher et
a1. (1980). Both terms refer to the circulation of sediment particles ver-
tically through the surface layers of the deposit under the feeding action
of benthos. A laboratory experiment by Robbins et a1. (1979) illustrates
this type of mixing. They placed a 1-mm layer of illite coated with 137CS
on top of a sediment column that contained 50,000 worms m - 2. 137CS is
a gamma emitter that sorbs strongly to illite clay particles. It is thus an
excellent material for tracing nondestructively the movement of sediment
particles by deposit feeders. 137CS activity throughout the sediment col-
umn was monitored with an external gamma detector. 137CS dropped
below the sediment-water interface as a fairly homogeneous layer as
worms fed below the layer and deposited fecal material on top of it. This
process continued until day 49 when a secondary peak was formed at the
sediment-water interface after the 137CS reached the top of the worm
feeding zone at ~3 cm depth (Fig. 13). This mixing process can be ana-
lyzed mathematically as a combination of convection (primarily) and
diffusion (to account for gradual peak broadening) of sediment particles.
Such a model has been constructed by J. B. Fisher et a1. (1980); it is
discussed in detail there and in Chapter 7.
The other type of biogenic sediment mixing is the random movement
of particles. It is not the peculiar result of a deposit feeding strategy, but
rather is created by suspension feeders, deposit feeders, as well as om-
nivores and is a result of burrowing as much as feeding. Thus it is probably
the more common type of biogenic mixing. Another experiment by Rob-
bins et a1. (1979) illustrates this kind of mixing. In this experiment, 137CS_
labeled clay was added to a sediment column containing the amphipod
Pontoporeia hoyi (10,000 m - 2). The activity profile evolved from a nearly
Gaussian shape to a uniform distribution in the top 1.5 cm in 17-20 days
(Fig. 14). The net effect of amphipod feeding and burrowing activity is
to mix sediments randomly. Particles can be treated as if they were dif-
fusing down from the interface. Appropriate models of this kind of mixing
126 Chapter 3

-'-
0..,.
17

37

4.

11
49

~
c
:::>
Q) ·1 0 I
0 2 •
>
:;::

~
0
Q)
0:::

>- o·

f-
> - ..
~
<->
«
0 • 0 2 4 6
• 0

SEDI MENT DEPTH (em)-

Figure 13. Effect of tubificids on the distribution of 137CS in laboratory sediments. From
Robbins et 01. (1979) with permission of Elsevier Scientific Publishing Co.

Odoy.

>-
.....
o 0

>
f=
~ A \
·1 0 1

Figure 14. Effect of an amphipod (Pontoporeio hoyi) on the distribution of 137Cs in laboratory
sediments. Redrawn from Robbins et 01. (1979) with permission of Elsevier Publishing Co.
Effects of Benthos on Freshwater Sediment Physical Properties 127

are discussed in Robbins et a1. (1979) and in Chapter 7. Unionids,

pisidiids, and chironomids also probably mix sediment in this fashion.
The configuration of the volume of biogenic sediment mixing de-
pends not only on the type of organism involved but also on population
density and distribution (horizontal and vertical) and individual organism
size. A single large active organism (e.g., a unionid) may have more im-
portant effects than many small individuals (e.g., pisidiids or tubificids).
Because large infaunal species tend to be more rare and patchily distrib-
uted than small infauna, the amount and the depth of sediment mixed
by large species will also be more patchily distributed than those of sed-
iment mixed by small, more evenly distributed organisms. The patchy
distribution of deep (>15 cm) mixing of sediment-associated trace metals
(Kovacik and Walters, 1973; Walters et a1., 1974) and pollen and 210Pb
(Robbins et a1., 1978) in the western basin of Lake Erie coincident with
the distribution of large unionid clams in the lake may be an example of
such patchy mixing. Nevertheless, most infaunal benthos, large and small,
are patchily distributed (e.g., Alley and Anderson, 1968; Brinkhurst, 1965;
Mozley and Garcia, 1972; Henson, 1962; Freitag et a1., 1976), and it may
appear unlikely that geochemists and geochronologists can tell anything
of value from the few cores that they are restricted to analyzing from any
one lake habitat. The reason they can is probably that over time the various
animal patches will have visited most spots in the habitat. If sediment-
associated particles are inert or slowly reactive, they will experience time-
averaged biogenic mixing. More rapidly reacting substances should more
closely track the shifts of infaunal populations.

4.1.2. Depth and Rate of Sediment Mixing

4.1.2a. Oligochaetes. Oligochaetes are among the most potent movers
of sediment in fresh water. Because of this and because they feed at depth
in the sediment, the vertical distribution of tubificids has received more
attention than that of any other freshwater invertebrates. Figure 15 shows
depth distributions of tubificids in cores collected from seven different
lakes. The depth distribution of these animals is extremely variable. While
in most lakes more than 90% of the tubificids will be found in the top
10-12 cm of sediment, there are exceptions. In Lake George, for instance
(Fig. 15E), 85% of the worm population is found below this depth. A
better generalization is that depth distributions are not fixed to a certain
small range worldwide, but are finely tuned to different lake environ-
ments. We will look at two factors influencing depth distribution, namely
worm population density and oxygen content of overlying water.
When sediment cores collected by the same investigator within one
lake (Figs. 15A,B) or from two nearby lakes (Figs. 15C,D) are compared,
worms are found to live deeper in cores with higher worm population
128 Chapter 3

40 60 80 100

2
I
,,
I

,,--,,---
,,"
.- ;
4
;/;"

,;; ' 4 .8 X 10 '

,
6
I
I
I
8 I
I

,
I
I

10

A
12 B

I
2 I
I

,,
I
I

,
,,
4 I
1497/m 2
,'2.27 X 10 '
6

"",
" PO M A T O T H RI X
8 HAMM O IENS I S
, ' ESTH WA ITE T OTA LS

10
///
12
I
,,
,
I
I

14 ,
C D
16

Figure 15. Depth distribution of tubificids in various lakes. The x axis represents percent
of fauna found deeper than depth indicated on y axis (in cm). (A) Dinton Brook (total
tubificids) . Data from Brinkhurst and Kennedy (1965). IB) Dinton Brook (Tubifex tubifex) .
Data from Brinkhurst and Kennedy (1965). (C] Esthwaite Water and Lake Wind emere. Data
from Stockner and Lund (1970). (D) Lakes Erken and Hovgardsfjiirden . Data from Milbrink
(197 3). (E) Lake George and Green Bay. Data from Burgis et 01. (1973) and J. A. Fisher and
Beeton (1975) . In A-D, solid lines denote low worm abundances and dashed lines denote
high worm abundances. In E, the solid lines indicate data from Green Bay, and the dashed
line indicates data from Lake George.
Effects of Benthos on Freshwater Sediment Physical Properties 129
o 20 40 60 80 100

OXYGENATED
I
I CONTROL
10 I
I
I
I
I
I
I
I
I
I
I
I
I
20 I
/1000-5000/m
I
I
I
I
I
I
I
I
I
I
30 I
I
I
I
I
I
I
I
I
I

40
E
Figure 15. (eont'd)

densities. In comparable cores with similar population densities, depth

distributions are also similar. For example, 50% of the worms in Dinton
Brook (Fig. 15A) are found between 2.5 and 4.5 cm when the population
density is ~200 , 000 m - 2 and only 10% of the population is found below
5 cm. When abundances increase to ~1,480,000 m - 2 , 50% of the popu-
lation is found between 4 and 7 cm, and over 40% of the population is
below 5 cm (Fig. 15A). The median depth of worms in cores is 1-2 cm
deeper in cores with high worm abundances (Figs. 15A-C). In comparable
cores with similar abundances, depth distributions are also similar, es-
pecially at low abundances (Fig. 15D). The correlation of abundance and
depth of occurrence in comparable cores probably has little to do with
the need for a greater volume of living space. There is no absolute relation
between abundance and depth of occurrence if all cores are compared
together. We think it is more likely that the worms adjust their feeding
depth lower when abundances are higher. McCall and Fisher (1980) report
that with densities of ~65,000 worms m -2 the zone of maximum feeding
was at a depth of 7.7 cm; with ~48,000, at about 5.5 cm. In an experiment
130 Chapter 3

by Robbins et al. (1979) in different sediment (Fig. 13), the feeding depth
of 3 cm was displaced downward to about 5-6 cm when more worms
were added to the experimental microcosm.
There is some evidence that oligochaetes are selective feeders on
sediment bacteria (Brinkhurst et al., 1972; Wavre and Brinkhurst, 1971)
and that they feed in the reduced zone of the sediment. A greater abun-
dance of worms will circulate the "sediment conveyor belt" faster (see
Section 4.1.1), and so sediment will likely need to fall to a greater depth
before the microbial food supply is replenished. In addition, the oxidized
surface layer is somewhat thickened by their feeding activity (see Chapter
4), so there is a smaller reducing environment for the growth of anaerobes.
If these hypotheses are correct, then depth distribution would be a func-
tion not only of worm abundance but also of sedimentological factors
such as grain size (surface area for bacterial attachment), organic matter
input (microbial food supply), and pore water geochemistry (microbial
environment), and it should not be surprising that the depth distribution
of oligochaetes varies widely among different lakes.
J. A. Fisher and Beeton (1975) described the effect of low dissolved
oxygen in water overlying sediment on the depth distribution of the oli-
gochaete Limnodrilus hoffmeisteri from Green Bay, Lake Michigan. Ex-
posure to anoxic water drove the worms down into the sediment for a
period of 3 days; the median depth of occurrence fell from 3 to 11 cm
(Fig. 15E). There was a subsequent upward migration for the next 4 days,
but the worms still lived deeper than they did in an oxygenated control
aquarium (Fig. 15E). Perhaps this is a partial explanation for the anom-
alously great depths of occurrence that Burgis et a1. (1973) report for Lake
George oligochaetes. Tropical Lake George is subject both to frequent
storms that disturb the surface layers of the sediment and to intervening
periods of stratification and anoxic hypolimnetic waters (Ganf and Viner,
1973). Additional factors such as sediment water content (Berg, 1938;
Stockner and Lund, 1970) and worm length and biomass (see McCall and
Fisher, 1980) are probably also important determinants of depth distri-
bution. Feeding depths may be different for different species. Krezoski
and Robbins (1980), for example, used the same radiotracer methods as
J. B. Fisher et a1. (1980) but utilized different sediments (from Lake Huron
instead of Lake Erie) and different oligochaetes (the lumbriculid Stylo-
drilus heringianus instead of the tubificids Tubifex tubifex and L. hoff-
meisteri) and observed different faunal feeding depths (3-5 cm instead
of 6-8 cm).
It is the depth of feeding that is primarily responsible for controlling
the depth of sediment mixing by oligochaetes. There is typically a fairly
narrow 2-cm zone of maximum feeding that is shallower than the deepest
penetration of the worms (Fig. 16). While worms burrowed to depths of
Effects of Benthos on Freshwater Sediment Physical Properties 131

FREQUENCY

O .2 .4 .6
~--~-----r----~----~~~~--~

..- 6
E
~ Feeding
I-
z 10
w
~
2i
w
I/)

~
:z: 16
I-
Q.
W
0
20

24

28
Figure 16. Feeding depth versus burrowing depth for Tubifex tubifex in a laboratory mi-
crocosm. Population density: 10 5 worms m - 2.

25 cm in this case, the depth of maximum feeding was 7-8 cm. Oligo-
chaetes burrow to greater depths than those at which they commonly live,
and they feed at even shallower depths. Burrowing itself contributes only
slightly to sediment mixing. While correlations of mixing depth and worm
distribution have been reported (Krezoski et al., 1978), this will not always
or even often be the case. It is known, for instance, that most tubificids
cease feeding at low oxygen levels (see McCall and Fisher, 1980), so that
even if worms migrate deeper under low-oxygen conditions-as in the
study of J. A. Fisher and Beeton (1975)-they do not necessarily mix the
sediments to greater depths.
A number of workers have performed experiments on factors affecting
the rate of mixing (feeding rate) by oligochaetes (see Table III). The re-
ported mixing rates are quite variable, and no single value can be used
to estimate mixing rate. At anyone temperature, mixing rates vary by
factors of from 3 to 160, indicating that mixing rate is strongly dependent
 ~
W
N

Table III. Comparison of Tubificid-Induced Subduction Velocities Reported by Various Authors

Subduction velocity (cm . day-l . m- 2 per 100,000 individuals)

Source/organism(s) 2-4°e 5-7°e 8-10 oe 11-13°e 14-16°e 17-19°e 20-22°e 23-25°e

Aisterberg (1925)
Limnodrilus and Tubifex 0.55
Irlev (1939)
Tubifex tubifex 0.02
Ravera (1955) 0.05
Peloscolex ferox 0.01 0.01
Bythonomus lemani 0.01 0.01
Psamnorytes barbatus 0 .01 0.01
Poddubnaya (1961)
Temperature experiments
Limnodrilus hoffmeisteri 0.04 0.09 0.20 0.30 0.60 0.68
L. udekemianus 0.06 0.12 0.22 0.36 0.66 0.77 n
Other experiments ::r
'0
L. hoffmeisteri 0.47 0.16 0.23, 1.67 '"Cii
L. udekemianus 0.41 0.19 0.27,1.62
..,
w
Wachs (1967) t'l
:::::
ctl
T. tubifex 0.04 0.07 0.14 0.17 0.26 0.52 0.64 0.73
~
Appleby and Brinkhurst (1970) en
0
.....,
T. tubifex 0.03 0.12 0.27 0.27 0.34
IJ:j
1. hoffmeisteri 0.02 0.08 0.13 0.38 ctl
::l
Peloscolex multisetosus 0.01 0.05 0.07 0.01 0.01 ;.
0
Davis (1974) en
0
Limnodrilus spp. 0.47 ::l
McCall and Fisher (1980) ...,'"'l
ctl
T. tubifex 0.02 0.11 0.20 en
~
M. J. S. Tevesz and D. M. Lipinski 0.03 ~
OJ
Mixed tubificids 0.10 co...,
Robbins et 01. (1979) Vl
ctl
T. tubifex 0 .10 P..
J. B. Fisher et 01. (1980) s·
ctl
T. tubifex 0.12
Krezoski and Robbins (1980)
a
'1:l
~
Stylodrilus heringionus 0.18 '<
en
ri·
e.
...,'1:l
0
'0
ctl
::1-
roO
en

~
W
W
134 Chapter 3

01
E
CiQ)s
--.
- ..
Q)
s:.
u ......

~ 0
'tI
E
~
•
...
L
P
01 >- • T
S- o.
'tI 2

~
IX
Z
0
1
~
U
w
u..
w
C

0
0 4 8 12 16 20
TEMPERATURE (1:)
Figure 17. Feeding rate as function of temperature for Limnodrilus hoffmeisteri (L), Tubifex
tubifex (T), and Peloscolex multisetosus (P). Redrawn from Appleby and Brinkhurst (1970).

on the particular species of worm studied. For anyone species, water

temperature appears to be the most important determinant of mixing rate.
Maximum rates are typically 5-10 times higher than minimum rates over
the seasonal range of temperatures experienced by one species; for any
one species-temperature combination, mixing rate ranges only by about
a factor of 3 (Table III). Mixing rates are not linearly dependent on tem-
perature, and different species of worms have different feeding optima
at different temperatures (Fig. 17) (Wachs, 1967; Appleby and Brinkhurst,
1970; McCall and Fisher, 1980). Smaller but statistically significant effects
on mixing rates have also been found for worm abundance, presence of
other worm species, and sediment type (Wachs, 1967; Zahner, 1967; Ap-
pleby and Brinkhurst, 1970; McCall and Fisher, 1980). McCall and Fisher
(1980) also found significant interactions between the effects of temper-
ature and substratum and temperature and abundance. This means that
the relation of feeding rate to temperature is different on different substrata
and at different worm abundances.
Given these variations in mixing depths and rates, few generalizations
concerning mixing by worms are safe. Given the positive correlation of
mixing depth and worm size and population density, the negative cor-
relation of mixing depth and oxygen level, and the positive correlation
of mixing rate and temperature, we expect that warm eutrophic lakes with
Effects of Benthos on Freshwater Sediment Physical Properties 135

a high input of organic matter that will support a greater number of larger
worms will have the most thoroughly and deeply mixed sediments. Of
course, if oxygen levels are so low that anoxia is frequent and of long
duration then mixing will cease. Burgis et 01. (1973) report mixing to 20
cm in shallow, eutrophic, tropical Lake George, but an undetermined
portion of this mixing is due to storm disturbances of the bottom. J. B.
Fisher et 01. (1980) estimate that mixing rates by tubificids can exceed
sedimentation rates by factors of ten or more in some parts of Lake Erie.
Robbins et 01. (1977, 1978) have found mixed profiles of radioisotopes
2-6 cm or more in thickness in other portions of the Great Lakes, which
they attribute largely to biological processes.
Stockner and Lund (1970) report mixing of sediments primarily by
tubificids but also by chironomids in Esthwaite Water and Lake Win de-
mere (see Fig. 15C for tubificid depth distribution). They used the pres-
ence of live phytoplankton in sediments as a tracer of mixing. The dif-
ficulty with this tracer was that the longevity of live algal species in muds
was not well known. Lund (1954) did laboratory experiments in which
live algae were incubated in dark anoxic muds. None survived longer
than 3 years. Three years would correspond to a sediment depth of only
0.45 cm (1.5 mm . year- 1 ) in these lakes and the presence of any live algae
deeper than this would be due to mixing. Stockner and Lund (1970) found
live algae deep in sediment cores but believed that biological mixing
could not account for their presence and so concluded that algae could
remain viable for 183-290 years. We think that their results are in fact
compatible with biological mixing. Many viable species were found to
8-10 cm in Wind em ere (lower tubificid abundance) and about 15 cm in
Esthwaite (higher tubificid abundance). The number of viable species
changed by 100% seasonally in the upper 5 cm of sediment, which is
compatible with rapid mixing or ventilation of sediments (probably by
chironomids). The greatest depth to which viable species were found was
25 cm in Windemere and 35 cm in Esthwaite. (No measure of relative
abundance of algae was available.) Both tubificids and chironomids were
present at these depths, although at greatly reduced abundances. Still, it
does not appear unreasonable to us to expect that a few algae could be
brought down to this depth, even though wholesale mixing is quite
unlikely.

4.1.2b. Other Macrobenthos. Mixing of sediment by benthos other

than oligochaetes has received little attention. Ford (1967) studied the
vertical distribution of Chironomidae in muddy stream bottoms. He found
that over 95% of the chironomids in both the laboratory and the field
occurred in the top 5 cm of sediment. Stockner and Lund (1970) found
chironomids almost entirely in the top 6 cm of the sediment of the two
136 Chapter 3

t day

0.8

Q)
.2
o
Q)
a:

•
1.5

Figure 18. Time evolution of 137CS activity profile for particle mixing by Pontoporeio hoyi.
Calculated activity is represented as a solid line; the points represent laboratory data. Re-
produced from Robbins et 01. (1978) with permission of Elsevier Scientific Publishing Co.

English lakes, Lake Windemere and Esthwaite Water, although they did
find two specimens at 55 and 65 em below the surface in Esthwaite
sediment. Milbrink (1973) found that most chironomids in four Swedish
lakes were restricted to the upper 3-4 cm of sediment. One larger species,
Chironomus plumosus, was found to depths of 15 cm. In Lake Erie, we
find that burrows of fourth instars of this same species reach 8-11 cm
below the sediment surface. Timm (1975) found a similar vertical distri-
bution of chironomids: Small chironomids were found primarily in the
top 5 cm of the mud of Lake V6rtsjiirv, while large C. plumosus were most
abundant 10-15 cm below the sediment-water interface. The actual mix-
ing of sediments by chironomids has hardly been examined. Stockner and
Lund (1970) report that Kleckner (1967) performed laboratory experiments
in which chironomids and tubificids mixed glass microbe ads into sedi-
Effects of Benthos on Freshwater Sediment Physical Properties 137

ment: "Generally only the top 3 cm were disturbed to a significant degree."

McLachlan (1977) found that C. plumosus thoroughly worked laboratory
sediments by feeding and active burrowing while employing grazing as
a feeding method. The same species left sediment "virtually undisturbed"
while suspension feeding. The depth and rate of mixing were not reported.
Robbins et al. (1979) studied the mixing of sediment by the amphipod
Pontoporeia hoyi from Lake Michigan in a laboratory experiment using
137Cs-labeled clay as a tracer of mixing. Pontoporeia mixes by burrowing
and surface deposit feeding rather than by conveyor belt deposit feeding.
Sediments were only mixed to a depth of 1.5-2 cm. Particle mixing by
Pontoporeia can be accurately described by a simple eddy diffusion model
of particle movement (Fig. 18). The upper 1.5 cm were uniformly mixed
at 7°C by a population of 16.0 x 10 3 amphipods m -2 in 17-30 days.
Pisidiid and unionid bivalves also mix sediment by their burrowing
activities. Although pisidiids are occasionally found several centimeters
below the sediment-water interface (Berg, 1938), most burrow in the top
2.5 cm, and it is unlikely that they account for any significant mixing
below this depth. McCall et a1. (1979) report that the unionid Lampsilis
radiata from Lake Erie (average length 6.3 cm) plows muddy sediments
to a depth of 2-3 cm during semiinfaunal crawling and mixes sediments
to a depth of 10 cm during burrowing. Larger unionids could mix sedi-
ments incompletely to a maximum depth of about 20 cm.

4.1.3. Complicated Stratigraphic Effects

Mixing alters sediment stratigraphy. The magnitude of this mixing
effect on sedimentary materials (and thus age determination) is dependent
on the mixing rate, mixing depth, and sedimentation rate (Chapter 7),
making any generalizations applicable to all lake environments suspect.
An observation that bears closer examination than it has received to date
is that not all sedimentary materials show the same response to biological
mixing. That is, examination of some sedimentary parameters may in-
dicate thorough biogenic mixing in the upper few centimeters of sediment,
while others show no such evidence. For example, the laboratory work
and field observations of Robbins et a1. (1978) on 210Pb and 137CS tell us
that the top 5-6 cm of sediment in Lake Erie are biogenically mixed, but
vertical profiles of some metals (Walters et aI., 1974) and organic carbon
and nitrogen (G. Matisoff, unpublished data) show no evidence of mixing.
This may be a result of very high and time-dependent inputs of these
materials. Pore water materials that react and establish equilibria faster
than the sediment is mixed may also be transparent to biogenic mixing.
Hakanson and Kallstrom (1978) show a radiograph of sediment from Lake
Ekoln in Sweden that they know to be mixed by tubificids, chironomids,
138 Chapter 3

0
3

2
5

4
7
E
2
~

a.
Cll 6
""0 9
~

c::
Cll
E
""0
Cll 8
Cf) 11

10
13

12
15

14
17

A B
Figure 19. Radiographs of sediment cores. (A) From 29 m in central Lake Ekoln . (B) From
32 m in southern Lake Ekoln. (C) From 8 m in western Lake Erie; depth of sediment column:
16 cm. A and B from Hakanson and Kallstrom (1978) with permission.

and pisidiid clams. The core shows laminae of light and dense materials
in an otherwise apparently homogeneous core of silt-clay material. The
laminae are most strongly developed in the top 4 cm of sediment. We
have observed a similar phenomenon in Lake Erie, where laminae are
most abundant in the top 7 cm of sediment (Fig. 19) . Laminae are not
normally present when biological mixing occurs and are usually good
evidence of dominance by physical processes (D. G. Moore and Scruton ,
1957; Rhoads, 1975). However, where mixing is of the ordered type that
results from tubificid feeding, this need not be the case. Laminae produced
near the sediment-water interface by waves and currents may remain in
existence until they descend to the base of the feeding zone (3-8 cm).
The base of the feeding zone itself may produce a lamina of large dense
(quartz instead of clay) material that is left behind as a lag deposit of
Effects of Benthos on Freshwater Sediment Physical Properties 139

c
Figure 19. (cont'd)

rejected food material by these selective deposit feeders. Kallstrom and

Axelson (1978) also suggest that some laminae are biologically produced,
with environmental factors such as oxygen in the overlying water deter-
mining the location of an ever-changing base of the biologically disturbed
zone. Finally, Howard and Elders (1970) suggest that some laminae can
survive intense mixing (at least in a marine sand with burrowing am-
phipods). Thus mixing is not always a simple process with simple effects,
and where sizable benthic populations are present, some caution needs
to be exercised before it is concluded that mixing does not occur in a
particular environment.

4.2. Effects on Particle Size Distribution

Alterations in the size of particles comprising soft bottoms can result

in alterations of the rate and mode of sediment transport from the bottom.
140 Chapter 3

In addition, the volume and arrangement of interparticle spaces can be

changed with a resulting change in the chemical flux of dissolved ma-
terials through the sediment. Macrobenthos alter particle size by making
or destroying fecal pellets and pseudofeces. *
The net effect of sizable populations of deposit feeders on particle
size distributions in mud bottoms is the addition of fine sand-size fecal
pellets to a silt-clay matrix. There has been little work on the pelletization
of freshwater muds. Iovino and Bradley (1969) studied the pelletization
of sediment by chironomid larvae of Mud Lake, which they believed to
be a modern analogue of Eocene oil shale environments. Mud Lake is
unusual because it is very shallow (only about 45 cm deep) and alkaline
(pH 7.7-10.2) and has no phytoplankton and no clastic sediments. The
sediment is comprised almost entirely of pellets of blue-green algae eg-
ested by Chironomus inhabiting the lake bottom at densities of a few
hundred per square meter. Laboratory populations of second instars fed
blue-green algae produced pellets averaging 168 fLm x 174 fLm. Fourth
instars produced pellets 342 fLm x 168 fLm. Pellets from the lake averaged
364 fLm x 142 fLm. Laboratory populations fed green algae produced no
coherent pellets at all. The method of feeding on the two food sources
was also different-grazing on the blue greens and filter feeding on the
greens. McLachlan and McLachlan (1976) studied the formation of bottom
sediment in a newly created temperate lake where the parent material for
sediment formation was both inorganic clastic sediment and organic mat-
ter (7-10% by weight). They found that Chironomus plumosus increased
the amount of both fine «105 fLm) and coarse (>250 fLm) material in the
sediment. Fines increased by about a factor of 4.5 and were the result of
mechanical disturbance of the mud by the chironomids. The amount of
coarse particles was over 200 times higher in sediment with chironomids
than in sediment with none, although most of the coarse particles (93%)
were associated with the tube silk rather than free fecal pellets (7%). This
squares with our casual observations of chironomids in laboratory aquaria.
When C. plumosus are introduced to sediments pelletized by tubificid
oligochaetes, pellets disappear in a circular area around chironomid bur-
rows, while massive aggregations of sediment, mucus, and silk become
associated with the chironomid tubes.
Similar laboratory observations on the mayfly nymph Hexagenia,
along with published accounts of its burrowing activities (Lyman, 1943;

* By particle size we mean the size of naturally produced aggregates of bottom material,
which may be amalgams of organic and inorganic material, and not the size of mineral
grains measured by standard sedimentological techniques. Pseudofeces are aggregates of
mineral grains and organic material that are manipulated by an organism but not ingested.
They are usually produced by suspension feeders and are less cohesive and more amor-
phous than fecal pellets.
Effects of Benthos on Freshwater Sediment Physical Properties 141

B. P. Hunt, 1953), suggest that Hexagenia may be another important pellet

disaggregator, but no experiments have been done to determine the mag-
nitude of its effects. Other fecal pellet producers have been identified
among the Gastropoda (Calow, 1975), Amphipoda (Hargrave, 1976; Lau-
tenschlager et a1., 1978), and Copepoda (Ferrante and Parker, 1977), but
their quantitative significance in altering physical properties of sediment
remains to be measured.
Tubificid oligochaetes are the pelletizers that have received the most
attention. McCall (1979) found that the top 0.5-1.0 cm of silt-clay mud
in western Lake Erie consisted of fecal pellets in various stages of decay.
Constituent mineral particles that made up the pellets had a median grain
diameter of 1.5 flm. Fresh pellets were cylindrical mineral-organic ag-
gregates averaging 280 flm x 70 flm. Even when pellets were subjected
to vigorous mechanical disturbance, mineral grains were still bound in
an organic matrix and some pellet pieces (80 flm long) remained (Fig. 20).
Laboratory experiments on pellet production and decay rates indicated
that in situ tubificid abundances of 10 4 m - 2 should be able to pelletize
the top 1-2 cm of sediment in this part of the lake and field observations
showed the presence of a pellet layer for most of the year. Settling velocity
measurements (Table IV) confirmed the existence of mineral grain aggre-
gates in lake deposits and showed a two-orders-of-magnitude increase in
settling velocity owing to pelletization. The net effect of pelletization of
these fine muds is to create a surface with larger grains that are more
difficult to erode and larger interparticle voids that are more permeable
to the flow of water, as experiments described later will show.
But pelletization can have an opposite effect, depending on the initial
particle size distribution of the sediment. Tevesz et a1. (1980) examined
the effects of tubificid feeding on a sandy river sediment (mean diameter
100 flm). They added worms (4 x 10 4 m- 2 ) to homogenized river sedi-
ment in the laboratory. A pronounced vertical sorting of sediment took
place owing to the size-selective feeding of the worms at depth (Figs. 21
and 22). The worms fed in layer C (Fig. 21) on the <62.5-flm size fraction
of particles, leaving behind a lag deposit with a higher sand concentration
than deeper unreworked layer D. The top layer, A, consisted of sand-size
fecal pellets composed primarily of <62.5-flm particles, and beneath this
was a decayed pellet layer of similar composition (Fig. 22). Coulter counter
analyses of the <62.5-flm fraction in the four layers further revealed that
the worms fed selectively but not exclusively on clay-size «4-flm) par-
ticles. The net effect of feeding on sandy sediments is to place lower-
density sand-size pellets at the sediment surface; both layers A and B had
higher water contents (by a factor of two) and more organic matter than
unreworked layer D and were more easily transported by river currents
than unreworked sediment. Such vertical sorting in sandy deposits has
142 Chapter 3

t 200,,", I
A

B
I 200E I
Figure 20. (A) Transmission light micrograph of the top 1 mm of sediment showing oligo-
chaete fecal pellets (x 12). (E) Transmission light micrograph of pellets disaggregated by
stirring. Most sediment is bound in an organic matrix; few discrete pellets are visible ( x 12).
(C) Transmission light micrograph of the top 1 mm of sediment from site 1 after treatment
with blender and dispersant (x 400). Note scale change.
Effects of Benthos on Freshwater Sediment Physical Properties 143

I 10,,", I
c
Figure 20. (cont'd]

also been reported by Kikuchi and Kurihara (1977) in submerged ricefield

soils where the 0- to 5-cm layer of soil was enriched in 100-f,Lm sand-size
particles, while the deeper soil at 6-8 cm was depleted and soil deeper
than 8 cm was unaffected.
Filter-feeding clams also aggregate sediment particles by the produc-

Table IV. Settling Velocities of Pelletized and Nonpelletized Sediment from

Western Lake Erie a
Median settling
velocity
Treatment (cm' sec - 1 ) Settling medium

Indi vidual tubificid pellets 1.03 Lake water

Top 1 mm of deposit 0.13 Lake water
Top 1 cm of deposit 0.06 Lake water
Highly degraded pellets 0.03 Lake water
Top 1 cm of deposit 0.02 Distilled water and dispersant
Top 1 cm of deposit by standard sizing 0.0002 Distilled water and dispersant
methods

a From McCall (1979).

144 Chapter 3

Figure 21. Sediment layering produced by tubificids in Vermilion (Ohio) River sediments.
Scale bar: 1 cm. Reproduced from Tevesz et oJ. (1980) with permission of the Society of
Economic Paleontologists and Mineralogists.

hon of mucally bound pseudofeces from silt- and clay-size material sus-
pended in the water column. This biodeposition phenomenon has been
quantified for only one freshwater species, Corbicula fluminea. Proko-
povich (1969) found that the silt- and clay-size sediment (10-20% of
sediment by weight) contained in a 113-mile-Iong concrete-lined canal
consisted almost entirely of the excreta of this clam. The 30-m-wide, 5-
m-deep canal contained 1-2 x 10 4 Corbicula in its upper portions. The
average clam (1 cm long; 1 g wet weight) was able to filter about 0.5 liters
of water per day and abstract completely silt-clay particles from the water
column (average concentration 30 ppm). Prokopovich calculated that 620
tons of clams in the upper portions of the canal produced 3350 tons of
excreta per year. Moreover, the amount of fine sediment was greater in
the upper, faster-flowing (1.2 m . sec -1) portions of the canal than in the
slower-moving lower reaches, just the opposite of what would be expected
from hydrodynamic considerations. The greater abundances of clams in
the upper portion of the canal and the binding properties of mucus prob-
Effects of Benthos on Freshwater Sediment Physical Properties 145

ably account for the discrepancy. Thus biodeposition by clams can dra-
matically alter both the size distribution and location of sediment de-
posits. Stanczykowska (1978) noted that the very common European
bivalve, Dreissena polymorpha, produced large amounts of excreta that
were saturated with mucus and enriched in bacteria that enabled large

100 10 0
La yer A La ye r 8

80 80

62.62
~5
't!- 60 60
.J::
0>

~ 40 40

r
23 .50
20 20
13.00

0.0 08
. 39 1 16 00 12 27

-I a 2 3 4- >4 -I 0 2 3 4->4
Ph i Ph i

100 100
Layer C Layer D

80 80

't!- 60 60
48.97
~ 44 .58
0>

3 40 40

20.84
20 20 17.93

2.85 1.14 3.0 1.08 1.10 32 1

-I a 2 3 4->4 -I 0 I 2 3 4 >4
-
Phi Phi
Figure 22. Mean grain size distribution (N = 3) of wet sieved sediment from laboratory aquaria
containing worms. Letters denote layers pictured in Fig. 21. Reproduced from Tevesz
et 01. (1980) with permission of the Society of Economic Paleontologists and Mineralogists.
146 Chapter 3

numbers of chironomids to live among the clams. McCall et al. (1979)

also noted that the large unionids accounting for over half the standing
crop biomass of macrobenthos in western Lake Erie produced copious
amounts of pseudofeces, although they did not quantify their observation.
The biodeposition phenomena are probably widespread and of great local
importance.

4.3. Effects on Mass Properties of Sediments

Mass properties refers not to the properties of individual sediment

particles such as size and density, but to those properties that arise from
the arrangement or fabric of the particles in a deposit. We will discuss
the effect of macrobenthos on water content and permeability.

4.3.1. Water Content

Water content is a measure of the compactness of a sediment: The
fewer interparticle connections, the more open the fabric and the higher
the water content. Water content is reported by benthic ecologists as a
percent by weight of water of the total weight of water plus sediment, and
by civil engineers and geotechnologists as the ratio of water weight to
sediment weight. We use the former measure here. Water content is usu-
ally taken as one measure of fine-grained bottom stability. Other things
being equal, bottoms with higher water contents are more easily eroded
than bottoms with lower water contents (Fig. 23). This property is further
discussed in Section 5. Water content is also related to sediment porosity,
which in turn is one factor controlling the diffusivity of chemical species
(see Chapter 4, Section 3.3).
The effect of tubificid oligochaetes on water content arises not so
much from their burrowing activity as from their selective feeding and
pelletizing activities. The size of the effect is dependent on the sand
content of the deposit they inhabit. McCall and Fisher (1980) report that
tubificids at densities of 50-100 x 10 3 m- 2 have no effect on the water
content of fine-grained Lake Erie muds (top 1 cm water content 70-80%,
<3% sand), despite the creation of many vertical burrows and a pellet
layer in the top 1 cm of deposit. Apparently a fabric that is opened in one
part of the deposit (burrows, interpellet voids) is compacted in an adjacent
portion of the sediment (pellets, deposit adjacent to burrows). This takes
place over such a short lateral distance that no net change in water content
is measured. However, where there is a significant admixture of sand,
selective feeding can greatly alter water content. Tevesz et al. (1980) found
that selective feeding of tubificids on a sandy river deposit (mean grain
Effects of Benthos on Freshwater Sediment Physical Properties 147

1000

'0L~~~:;:::""---

Deposition

10 10 ' 10'
SIZE DIAMETER IN MICRONS
Figure 23. The influence of water content on the critical erosion velocity of fine-grained
«100-fLm) sediments. Redrawn from Postma (1967).

size 110 f-Lm, 83% sand) resulted in a vertical sorting of particles and
increased water content in the top 1.5 cm of deposit (48% with worms,
27% with no worms).
Burrowing by mobile infauna can also increase water content of the
top few centimeters of fine-grained sediment. This phenomenon can be
seen in both fine muds and sands, but it is likely that the effect of bur-
rowing is greater where a vertical sorting by grain density can occur in
addition to the injection of water into the deposit during burrowing.
McCall et a1. (1979) have produced quantitative information on the effect
of mobile burrowers in freshwater sediments on water content. They
found that the unionid clam Lampsilis radiata, a mobile infaunal filter
feeder (mean length 6.3 cm), increased water contents from 62% to 72%
in the top centimeter of western Lake Erie sediment and from 42% to 63%
in the 3- to 4-cm layer. As an alternative to measuring water content we
have also used a rotational viscometer to measure the torque required to
turn a T-bar spindle at 0.5 rpm as another measure of sediment com-
pactness (see Myers, 1977). Vertical profiles of sediment compactness
close to (::;2 cm) and far away from (2:2 cm) four individual 1. radiata
have been measured (Fig. 24). Close to the clams, compactness was re-
duced by 16-70% to a depth of 7 cm. At greater distances, however,
148 Chapter 3

Figure 24. Mean compactness measured as torque (1 unit = 575 ergs) as a function of depth
in sediment. (A) Measurements made :s 2 em from bivalve; (E) measurements made 2: 2 em
from bivalve. Microcosms 1 and 2 contain Lampsijis mdiata.
Effects of Benthos on Freshwater Sediment Physical Properties 149

compactness was increased by from 0% to 58%. This is another of what

Rhoads and Boyer (Chapter 1, Section 3.3) call "far-field effects," and is
probably the result of foot probing during burrowing and valves adducting
during cleansing activities.

4.3.2. Permeability
Burrows shorten the path length through which water flows within
sediments. Fecal pellet layers also produce this effect by creating a smaller
number of larger interparticle voids. McCall and Fisher (1980) found no
evidence for biological advection of pore water through oligochaete bur-
rows, but found that enhanced diffusion in the pelletal layer of sediment
was sufficient to explain the flux of a chloride tracer through the sediment
(see Chapter 4, Section 3.3). These experiments were done in laboratory
aquaria without vigorous water circulation. Nevertheless, waves and cur-
rents may create pressure gradients sufficient to ventilate passively
muddy sediments made permeable by biological activity. McCall and
Fisher (1980) measured the permeability (ability of a column of water to
pass through a thickness of sediment) of western Lake Erie sediment with
and without tubificids. Permeability increased upon addition of 10 5
worms m -2 by a factor of two to four to a steady state value in a period
of 3 days. This increase was due to the burrows made by the worms, since
a thick pelletal layer could not form over this time interval. We have
performed one experiment to test the effect of a current flow over pel-

40

30
I
U
E 20 o FLUME - NO WORMS
Q.
Q.
• TUBE - WORMS
10 • FLUME - WORMS
o TUBE - NO WORMS

o 5 10 15 20 25 30

At (HOURS)
Figure 25. Effects of tubificid worms and flow on chloride diffusion across the sediment-water
interface.
150 Chapter 3

letized sediment. Chloride was added to western Lake Erie sediment to

achieve a pore water concentration of 280 ppm. Deionized water was then
placed over the sediment and the chloride concentration in the overlying
water measured over time. Chloride flux of sediment with and without
worms was measured in an annular flume with a current flow beneath
that required to move sediment particles and in a tube with aeration but
no currents (Fig. 25). Sediments with worms had a pellet layer of 0.5-cm
thickness when the experiment began. Pelletized sediment with a current
flow had higher initial flux rates of chloride than sediment without a
pelletized layer or without a flow. Although more extensive experimen-
tation is required, this experiment provides some support for the idea
that macrobenthos can increase passive ventilation of the top layer of
muddy sediments.

5. Biogenic Modification of Sediment Transport

The effects of grain size distribution, water content or porosity, and
sediment shear strength on erosion depend on the particular sediment
examined and on differing techniques of analysis (Sundborg, 1956;
Postma, 1967; Southard, 1974). For some cohesive sediments, there is no
discernible relation between measured physical properties and erosion
of surface sediment (Parthenaides and Paaswell, 1968). Examination of
the literature concerning the effects of marine biota on sediment transport
(see Chapter 1) reveals at least four biological activities that might also
be important in modifying sediment transport in freshwater environments:
1. Burrowing macrofauna and meiofauna can increase sediment
water content and increase erodability (Rhoads and Young, 1970;
Cullen, 1973; Rhoads et 01., 1978).
2. Deposit feeders in high-water-content muds can produce sand-size
pellets that are more difficult to move than fluid muds.
3. Dense patches of tube dwellers can increase sediment stability and
decrease erosion (Mills, 1969; Yingst and Rhoads, 1978).
4. Bacterial and algal films can stabilize sediments and make them
difficult to erode; surface deposit feeders can break up the film
and make sediment less stable (Scoffin, 1970; Holland et 01.,1974;
Rhoads et a1., 1978; Yingst and Rhoads, 1978; Boyer, 1980).

5.1. Field Observations

McCall and Fisher (1980) appear to be the only workers to have

examined the effects of biota on the transport of freshwater muds in any
Effects of Benthos on Freshwater Sediment Physical Properties 151

detail. We will review their results and add some previously unpublished
data from that study.
McCall and Fisher (1980) collected 38 box cores from nine locations
in Lake Erie (Fig. 26, Table V). The box cores were inserted into an annular
flume where the bottom stress needed to initiate rolling, T e , and observable
entrainment of particles into the flow, T e , were measured. The nine lo-
cation averages of sediment median grain size (measured by standard
geological techniques), percent clay, percent organic matter, and percent
water content were simultaneously regressed on location averages of crit-
ical entrainment stress obtained under aerobic bottom water conditions.
Average physical properties of sediment such as grain size, clay con-
tent, and water content are not good predictors of critical entrainment
stress of the fine silt and clay sediments in Lake Erie. More detailed
measurements of these properties at each sampling site might clarify the
relation to Te by unmasking small-scale variability hidden by average
measurements. The observations are worth collecting, but the small
amount of data currently available suggests that little would be gained.
Eight measurements of median grain size were made over 2 years at both
location 2 and location 6; the standard deviations of these measurements
are 11% and 14% of the mean, respectively. Since the median grain size
of sediment from locations 1-9 changed by a factor of 20, we might expect

{
N

I
'I
L __~!.. __
PENN ,

Figure 26. Lake Erie sampling localities for sediment entrainment studies, Redrawn from
McCall and Fisher (1980),
152 Chapter 3

Table V. Description of Box Cores Collected from Lake Erie

H2O
Date 'Ie Te content
Box number Location collected (dynes· cm - 2) (dynes· cm - 2) (percent) Comment a

1 1 6/16/76 0.21 0.82 71

2 1 10/1/76 0.75 1.67 75
3 1 10/1/76 0.56 1.52 78
4 2 7/5/76 0.18 0.61 70
5 2 10/1/76 0.61 1.44 71

6 2 10/1/76 0.58 1.26 74

7 3 7/6/76 0.35 0.74 72
8 3 7/6/76 0.43 1.21 82
9 3 5/23/77 0.35 1.21 78
10 3 5/23/77 0.32 1.61 73
11 3 9/1/77 0.32 1.51 86 A
12 3 9/1/77 0.38 1.14 84
13 4 7/5/76 0.21 0.74 88 T
14 4 7/5/76 0.33 0.66 84
15 4 9/29/76 0.46 1.07 87
16 4 9/29/76 0.38 0.99 88
17 4 9/29/76 0.38 1.05 84
18 5 10/3/76 0.52 1.08 66
19 5 10/3/76 0.48 1.05 60
20 6 10/3/76 0.62 1.17 61
21 6 10/3/76 0.86 1.39 59 T
22 7 5/19/77 0.33 0.84 68
23 7 5/19/77 0.53 1.46 66 T
24 7 9/4/77 0.20 2.08 65
25 7 9/4/77 0.68 2.15 71 A
26 7 9/4/77 1.05 2.68 71 A
27 8 5/6/77 0.45 1.20 64
27 8 5/6/77 1.20 4.33 64 Second run
28 8 5/6/77 2.47 4.27 74 A,T
29 8 5/5/77 0.56 1.93 71 T
30 8 5/19/77 0.45 1.23 59
31 8 5/19/77 0.73 2.20 61
32 8 5/19/77 0.59 2.15 55
33 8 9/4/77 0.67 2.08 A
34 8 9/4/77 0.45 2.06 64 A
35 8 9/4/77 0.41 1.55 68
36 8 9/4/77 0.65 1.99 57
37 9 5/19/77 1.44 3.27 70
38 9 5/19/77 0.51 3.02 67

a A, anoxic overlying water in lab; T, high tube density (>3000 tubes ill - ').
Effects of Benthos on Freshwater Sediment Physical Properties 153

that any difference in Te owing to grain size differences would not be

entirely masked by within-site variability. In fact, sediment water content
is probably the most variable of the physical properties studied, and water
content measurements were collected with each critical stress measure-
ment. Power law and linear regressions of individual box core water
content and T e explain only 17% and 11% of the data. The slope of the
regression is not significantly different from zero (0.023 ± 0.025, P =
0.05). Neither is there any correlation of Te and water content within the
nine collection sites.
These results appear to be contrary to those of other workers who
showed grain size and water content to be major controls of sediment
erosion (Postma, 1976; Southard, 1974; Fukuda, 1978; Fukuda and Lick,
1980). The results of Fukuda (1978) are particularly appropriate for com-
parison, because he studied the erosion of fine-grained sediments in Lake
Erie that were "azoic" with respect to macrofauna. Fukuda studied two
types of sediment from Lake Erie. One was an illitic clay derived from
heavily weathered Chagrin Shale, which crops out along the lake and is
a parent material for lake sediment. The other was lake bottom sediment
collected with grab samplers and subjected to sieving and storage in the
laboratory for several weeks. Some of his results are shown in Fig. 27.
Data on illitic clay were obtained by adding fluid clay-water mixtures
to box cores and then measuring Te as the sediment dewatered through
self-compaction. Data obtained in the same way with the single difference
that the whole flume bottom was covered with sediment gave the same
results as those obtained with the box cores, indicating that the box core
method is an acceptable way to determine Te'
The results show also that, without the action of other processes
acting to modify lake sediment properties, entrainment is indeed related
to water content: Higher water content results in lower Te (r = - 0.92, N
= 26). In the natural situation, however, other processes, many of them
biogenic, act to change sediment erodibility in such a way that the effect
of water content is masked. Furthermore, for a given water content all
lake sediments are more difficult to erode than parent illitic clay sedi-
ments. In addition, average TeS for sediments collected from sites 2 and
9 were 100% and 75% greater, respectively, in box cores collected with
minimal disturbance of biogenic ally modified sediment surfaces by
McCall and Fisher (1980) than in the grab-sampled and sieved sediment
used by Fukuda (1978). Some factor or factors act to aggregate or bind
particles together, making lake cores more difficult to erode. Finally, some
other processes must be acting in more or less random fashion, some
increasing and some decreasing erodibility, in such a way that the result
is a collection of data points through which it is not possible to draw one
smooth curve containing all the observations.
 154 Chapter 3

4.0

09

3.0 09

'"E
u
"-
en 0
<I> 0
c 2.0 0
'"
"0

~ 0
0

0 03 D3
1.0 0 DO

0
0

OL-__L-______L-______ ~ ______ ~ ______ ~ ______ ~ ______ ~

40 50 60 70 80 90 100
% WATER CON T ENT

Figure 27. Critical entrainment stress. T e , of oxidized box cores as a function of sediment
water content. e , Box cores of shale-based sediment run by Fukuda (1978) ; "" , runs made
with the entire flume covered with shale-based sediment (also by Fukuda, 1978); D, box
cores collected from locations 1-9. Area shaded with horizontal bars is to be compared with
box cores from location 9; area shaded with vertical bars is to be compared with box cores
from locations 2 and 3. Redrawn from McCall and Fisher (1980).

Some of the observed variations of Te may be related to biological

activity. Critical entrainment stress appears to be higher (1) in box cores
containing high densities of oligochaete worm tubes projecting out of the
bottom to a height of 3 mm or more; (2) in box cores collected in the fall
(mean temperature 1 m off bottom 11°C) compared to box cores collected
in summer (mean temperature 22°C); and (3) in box cores accidentally
exposed to water with low dissolved oxygen concentrations. The contrasts
of otherwise similar box cores are shown in Table VI. Given the small
sample sizes and the unknown underlying distributions of the variables,
a nonparametric test of the differences in T e , either the Wilcoxon signed
rank or sign test, is most desirable. Both substitute the sign of the differ-
ence between comparable box cores for the absolute difference, and both
Effects of Benthos on Freshwater Sediment Physical Properties 155

give the same result here. In all three contrasts all the differences between
comparable cores are negative. The null hypothesis is that there is no
difference between, say, summer and fall box cores. If that is true, then
there is a 50% chance of a positive difference and a 50% chance of a
negative difference for any pair of box cores. Since there are four pairs
of box cores in each contrast, the probability of observing four negative
differences in four comparisons is given by the binomial distribution as
(1/4)4 = 0.063. For a two-tail test of significance the probability of equally
extreme fluctuations in the opposite direction (all positive differences)
must be added and the probability rises to 0.125. This means that if the
null hypothesis were true deviations as large as those observed in these
contrasts would occur by chance 12.5% of the time.
We might expect dense fields of worm tubes to increase the critical
entrainment stress, since they will act to promote sediment binding and
prevent the transfer of stress to the sediment-water interface (but see
Chapter 1, Section 4.1). Chironomid tubes are probably even more effec-
tive at binding sediment than tubificid oligochaete tubes because they
have a larger diameter, greater length, and thicker walls constructed of
a sticky tube silk. Edwards (1958) provides testimony to the binding power
of chironomid tubes. He found masses of chironomid tubes floating in the
effluent channel of a sewage works; the tube mats were associated with
the spring emergence of Chironomus riparius (10 5 m- 2 ). The tube mats

Table VI. Contrasts of Box Cores Collected from Lake Erie

Stress
Comparable box difference Percent
Contrast cores (dynes· cm -2) increase

Low worm tube density-high worm 14-13 -0.08 12

tube density 27-29 -0.73 61
22-23 -0.62 74
20-21 -0.22 19
Mean = 41.5
High dissolved 02-low dissolved O 2 24-25,26 -0.34 16
36-33,34 -0.08 4
28-27 -3.07 256
12-11 -3.27 32
Mean = 77
Summer-fall 1-2,3 -0.78 95
4-5,6 -0.74 55
13,14-15,16,17 -0.34 33
22,23-24 -0.93 45
Mean = 57
156 Chapter 3

were also produced in laboratory experiments in which it was found that

methane bubbles generated at depth could lift a tube mass a few centi-
meters in thickness off the bottom of aquaria after emergence of the chi-
ronomids. Edwards hypothesized that irrigation currents of the insect
larvae removed gas bubbles from the sediment under normal conditions
so that the "vernal sloughing" of the tube layer, as he called it, occurred
only following spring emergence and elevated temperatures (higher gas
production). The effect of chironomids on sediment entrainment has not
yet been the subject of any flume studies.
The oxygen content of the water overlying the box cores occasionally
dropped to low levels (2 ppm or less), either because water circulation
in the aquarium holding the box core ceased for a period of time or because
a box core remained sealed for several days prior to leveling and insertion
into the flume. When this happened, increased mucus binding of the
surface was usually evident. In most cases, strings of mucus-bound sed-
iment were visible on the sediment surface; in a few cases, the entire
sediment surface was bound into a 1- to 2-mm-thick mat that could be
peeled off the box core surface intact with forceps. In only one case (box
core 8, location 3) was a similar mat found under fully oxidized condi-
tions; here also critical entrainment stress was increased over the com-
parable box core 7. It is hypothesized that mucus binding of surface
sediment increased Te in this case.
The summer-fall contrast is more difficult to account for. The only
obvious difference in comparable cores collected at these times is bottom
water temperature. It is possible, of course, that the seasonal change in
sediment erodibility may not be directly related to temperature. For in-
stance, microbial production of exopolymer decreases when nitrogen
availability is low (Unz and Farrah, 1976) and the availability of nitrogen
may change seasonally. However, many biological processes that affect
sediment erodibility are temperature-related, so the increase in critical
entrainment stress from summer to fall may also be related to biological
activity in the sediments. Yingst and Rhoads (1978) documented the ex-
istence of a seasonal pattern of erodibility of marine (Long Island Sound)
sediments in more detail. They proposed that differing temperature re-
sponses of sediment-stabilizing microbes, principally mucus-secreting
bacteria, and sediment destabilizing fauna, principally burrowing macro-
fauna, together with phase lags in the response activity to temperature
could produce seasonal changes in sediment properties. Such a seasonal
cycle can have large effects not only on bottom processes but also on the
overlying water column; McCall (1977), for instance, found that in warmer
months about twice as much bottom material was resuspended into the
water column in central Long Island Sound as in winter months, despite
a trend in wind and wave disturbances in the opposite direction.
Effects of Benthos on Freshwater Sediment Physical Properties 157

5.2. Laboratory Experiments

McCall and Fisher (1980) reported the results of an experiment in

which Lake Erie sediment was added to box cores with worms added,
with no worms added, and with formalin added. After incubating for 20
days at 7°C and 24°C, Te for each box was determined in the annular flume.
At 7°C, Te of pelletized sediment was 39% higher than Te of nonpelletized
sediment. There was no significant difference in Te between boxes with
and without formalin treatment, although the formalin treatment did make
sediment more difficult to erode. At this temperature the most important
biogenic modification is pelletization, which decreased erodibility by
increasing natural particle size (no tubes were built above the sediment
surface); microbial binding was evidently not effective in decreasing er-
odibility. The surface of the formalin-treated sediment may have been
relatively more compact than that of the untreated sediment and thus
more difficult to erode. Untreated sediment was subjected to surface mix-
ing by meiofaunal ostracods and copepods that may have made sediment
easier to erode. Indeed, the sediment treated at 7°C with no worms had
the lowest Te of any sediment, perhaps because there was no pelletization,
little microbial binding, and some loosening of surface sediment by
meiofauna.
Sediment in boxes incubated at 24°C with no worms was 28% more
difficult to erode than sediment incubated at 7°C with no worms. This
was likely due to greater microbial activity at the higher temperature.
There was no difference between sediments with and without worms,
indicating that microbial mucus production is evidently as important as
pelletization at this temperature. McCall and Fisher (1980) used para-
metric methods to discern these significant differences in average T e be-
tween treatments. Since the sediment in these laboratory box cores was
reasonably homogeneous within treatments, we would really not be sur-
prised if the errors were normally distributed, even though we would
need more samples to actually test this idea. If we can assume a Gaussian
distribution, then their method for making multiple comparisons (Tukey's
T test), which is more efficient at detecting differences that actually exist
than analogous non parametric tests, is appropriate. Our confidence in the
comparisons made should be tempered by the finding that only two con-
trasts are significant using nonparametric tests (Wilcoxon critical range
test; Colquhoun, 1971). These are: (1) that the Te of box cores at 7°C with
worms is greater than that of comparable cores without worms, and (2)
that sediment incubated at 24°C with no worms had a larger Te than
sediment incubated at 7°C with no worms.
Another experiment was performed to see if the increased mucus
binding effect that was seen in field box cores subjected to low dissolved
158 Chapter 3

oxygen concentrations could be reproduced. Box cores receiving the same

treatments as in the previous experiment were further incubated for 7
days in water low in dissolved oxygen (=1 ppm), after which their 'Te was
measured in the flume. As a further check on the dissolved oxygen effect,
'Te was measured a final time after incubation for 7 more days in fully

oxidized water (=10 ppm).

Critical entrainment stresses of box cores incubated under different
dissolved oxygen concentrations are shown in Table VII. The increase in
'Te in box cores moved from high to low dissolved oxygen concentrations

is also shown. In the comparisons that follow, both Gaussian and non-
parametric methods give similar results. The increase in 'Te with decreased
dissolved oxygen levels in box cores with active biota is significant; the
average increase is about 30% and may be as much as 55-65%. There is
no significant difference in 'Te between high and low oxygen levels in
formalin-treated box cores. Finally, the effect of anoxia on 'Te in sediments
with biological activity is temperature-dependent. There is a greater than
twofold increase in the anoxic effect between 7°C and 24°C. It is not known
with certainty which organisms produce all the mucal material observed
in box cores subjected to low oxygen levels. Most macro invertebrates,
molluscs and tubificids included, secrete mucopolysaccharides (S. Hunt,
1970), but smaller organisms-in particular protistans, algae, fungi, and
bacteria-are probably quantitatively more important in particle aggre-
gation [see Harris and Mitchell (1973) for a review]. Bacterial polysac-
charides are particularly effective in enhancing aggregation in soils (Mar-
tin, 1946; Geoghegan and Brian, 1948; Fehrmann and Weaver, 1978),
sewage wastes (Pavoni et al., 1972), and marine sediments (Rhoads et a1.,
1978; Yingst and Rhoads, 1978). Mucus-secreting algae are able to stabilize
shallow marine sediments (Frankel and Mead, 1973; Holland et a1., 1974).
Protists may also cause particle aggregation (Pillai, 1941; Curds, 1963).
But since an amount sufficient to coat the sediment surface was produced
in a short amount of time (7 days) under dark as well as light conditions,
it is thought that this binding material is bacterial in origin. Why this
material is produced when dissolved oxygen concentrations are low is
also unknown. It is known that exopolymer production is highest in the
death phases of bacterial cultures, perhaps because of cellular lysis (Pa-
voni et a1., 1972; Harris and Mitchell, 1973). Low-oxygen conditions may
induce this phenomenon in aerobic bacteria. Alternatively, low oxygen
may favor the growth of some bacteria that secrete greater amounts of
binding substance [J. Y. Yingst (personal communication) suggests Thio-
vulum as an example]. The decrease in critical entrainment stress upon
reaeration of box cores might be due to a decline in exopolymer produc-
tion and/or an increase in near-surface burrowing and deposit feeding
activity, which break up the mucus layer. Surface burrowing and deposit
Effects of Benthos on Freshwater Sediment Physical Properties 159

Table VII. The Effect of Anoxia on Critical Entrainment Stress

I. Critical entrainment stress (dynes . cm - 2) of sediment incubated at different levels of

dissolved oxygen

Dissolved oxygen

Temperature High oxygen Low oxygen Reaerated

(0C) Treatment (1) (2) (3)

7° No worms (nw) 0.72 0.94 1.08

0.62 0.77 0.83
Worms (w) 0.81 0.85 0.75
1.08 1.13 1.00
Formalin (f) 0.80 0.81
0.85 0.83
24° No worms 0.93 1.55 0.76
0.81 1.09 0.68
Worms 0.87 1.09 0.86
0.80 1.33 0.85
Formalin 0.81 0.83
0.84 0.85

II. Column (2) - column (1) differences

7°C 24°C

nw w nw w

0.05 0.22 0.01 0.23 0.62 0.02

0.05 0.16 -0.02 0.53 0.27 0.01
x= 0.045 0.19 -0.005 0.375 0.445 0.015

III. Statistics

A. Hypothesis: Mean column difference, a, for treatments with biological activity is zero.
1.
a -
Student's t(n- l) = ----:-F' = 3.568, P < 0.005 (one-tailed). N = 8, d = 0.264 ± 0.175
(S/ Vn)
(95% confidence).
2. The sum of ranks for paired samples, Wilcoxon's T, equals zero, p = 0.01 , N = 8.
a
The median difference, = 0.225. At the 0.95 level, 0.05 < < 0.53. a
B. Hypothesis: Formalin-treated samples show no increase in To from high to low oxygen
levels.
1. t = 0.588 , P < 0.75, N = 4.
2. Wilcoxon T = 3.5, P = 0.375, N = 4.
C. Hypothesis: The increase in To from high to low dissolved oxygen levels is greater at
24°C than at 7°C in those cores with active biota.
a
1. t = 3.074 , 0.025 < P < 0.01 , N = 4. At 0.95 level = 0.292 ± 0.232
2. Unpaired sum of ranks, T', equals 10, p = 0.05, n, = fl2 = 4.
a
At the 0.95 level = 0.29 ± 0.19.
160 Chapter 3

feeding are perhaps the most active processes in the destruction of the
mucus layer because exopolymers are resistant to microbial decay (Harris
and Mitchell, 1973; Martin and Richards, 1963), but the existence of easily
degradable "glues" cannot be ruled out.

5.3. Some Comparisons

These laboratory experiments are useful to demonstrate the existence

and potential importance of various biogenic modifications of sediment
erodibility (i.e., pelletization, microbial binding, locomotory movements).
However, they are not especially good predictors of field results . For
example, the laboratory experiment might lead us to believe that the
natural bottom should be most difficult to erode during periods of high
temperature, but this is not apparently the case. The laboratory experiment
lasted only a few weeks and employed fixed incubation temperatures
whereas the lake bottom "incubates" for a longer period of time at high
temperatures, and bottom temperature changes more slowly over a longer
period of time. In addition, having been sieved and fluidized, the initial
condition of laboratory sediment must have been different than that of
natural bottom sediment at the beginning of summer. Natural bottom
sediments record a history of past physical-chemical and biological mod-
ifications, and it may sometimes require a period of weeks or months for
new processes to erase this history.
Comparing the results of different workers is difficult at this time.
We all use flumes of different design with unknown effects on our results
and we measure erosion in different ways. Few data are available that are
reliable in the sense that the natural sediment surface is not severely
altered. But results of McCall and Fisher (1980), Rhoads et a1. (1978), and
Young and Southard (1978) are comparable, if only because these workers
preserved the biogenically created surface properties of the fine-grained
sediment that they examined. However, because each investigator used
a different criterion for measuring a "critical" erosion point, only internal
comparisons can be made. Young and Southard (1978) compared the
critical boundary shear velocity between (1) marine muds collected,
mixed, and settled into core boxes from suspension in the laboratory and
(2) in situ bottoms and bioturbated laboratory cores. They found a twofold
reduction of the critical erosion point in the latter case. McCall and Fisher
(1 g80) found a similar difference between disturbed and undisturbed
bottoms in freshwater Lake Erie muds, but the sign of the contrast was
reversed. Undisturbed cores collected in situ were 75-100% more difficult
to erode than biota-poor laboratory sediment at the same water content.
While it is clear that disturbing biogenically produced surfaces will dras-
Effects of Benthos on Freshwater Sediment Physical Properties 161

tically alter experimental results, the direction in which they will be al-
tered cannot be predicted in the general case. Further specification of
sediment physical properties and infauna is needed to make fair com-
parisons. Rhoads et a1. (1978) found a 60% increase in critical mean
suspension velocity from summer to fall (July-November) in marine muds
from Long Island Sound. We found a 57% increase. Rhoads et a1. (1978)
found a 56% increase in critical rolling velocity of fine silt beads upon
incubation in seawater containing bacteria. We found a 40-80% increase
in T e where evidence of increased binding was present. Rhoads et a1.
(1978) found an 80-100% increase in critical erosion velocity when ~104
tube dwellers m -2 were added to Long Island Sound muds. We found a
41 % inc ease in T e in Lake Erie box cores with > 3000 tubes m - 2. Signif-
icant biogenic modifications of fine-grained bottoms are a widespread
phenomenon. Despite great sedimentologic and faunal differences in
marine and freshwater bottoms, functionally similar organisms will likely
produce about the same degree of change in erodibility of fine-grained
surface sediments.

6. Ecological Interactions and Sediment Properties

Organisms influence substratum properties. Sediment properties in-
fluence the distribution and abundance of benthos. Although most of the
literature concerns marine environments [see reviews by Gray (1974) and
Rhoads (1974), for example], Cummins and Lauff (1969), Oliver (1971).
and Brinkhurst (1974). among others, have demonstrated the correlation
between organisms and physical and chemical properties of freshwater
bottoms. It is important to realize that benthic fauna and sediments con-
stitute a complex feedback system in which numerous influences on ben-
thos distribution and abundance can indirectly affect sediment properties.

6.1. Spatial-Temporal Variations of Fauna

In Lake Erie there appears to be a seasonal cycle of bottom erodibility.

We have hypothesized that this cycle is due in part to the different re-
sponse of groups of benthos to changing temperature (Section 5). There
may be other explanations of changing faunal abundances that can enable
us to predict changes in sediment properties. Rhoads and Boyer (Chapter
1) have described how sediment properties develop during ecological
succession on marine fine-grained bottoms. Following some disturbance
of the bottom that causes local mortality of the macrobenthos, a repeatable
pattern of faunal change occurs. Dense aggregations of small tube dwellers
and other near-surface dwelling infauna are followed in time by larger,
162 Chapter 3

deeper-dwelling infauna, which may be either deposit feeders or filter

feeders (McCall, 1977, 1978). Sediment shear strength, water content,
biogenic texture, microtopography, and erodibility are differentially al-
tered in early and late successional stages (Chapter 1, Table III).
The field experimental work required to establish the existence of an
orderly pattern of recovery from disturbance in freshwater lakes has yet
to be completed. However, for the past few years one of us (PLM) and his
students have been conducting such experiments in Lake Erie. The picture
that has been emerging is that, following a disturbance of the bottom,
large numbers of near-surface-dwelling ostracods and naid oligochaetes
are early colonizers. They are soon followed by abundant tube-dwelling
chironomids; later come deeper-dwelling deposit-feeding tubificids.
Large burrowing infauna (unionids) are absent from the early stages of
recovery. Since these organisms affect the bottom in different ways, it is
likely that the "successional paradigm" can be of use in predicting chang-
ing sediment properties on disturbed bottoms in freshwater as well as
marine environments. Whether or not such a model will be widely ap-
plicable is unknown at present. There is some suggestion from the lit-
erature that it might be. For example, Herdendorf and Reutter (1976)
investigated the effect of dredging on benthic populations near a nuclear
power station on Lake Erie. The benthic fauna was eradicated during
construction of water intake and discharge pipelines. Less than a year
was required for recolonization with populations in the dredged area
exceeding the original numbers. Sweeney et a1. (1975) sampled a dredge
spoil disposal site in Lake Erie 5 years after dumping had ceased. We find
no statistical difference in mean abundance of major macroinvertebrate
groups at dump and reference stations. Paterson and Fernando (1969),
Lellak (1969), and McLachlan (1974) documented macroinvertebrate col-
onization of newly created reservoirs, fishponds, and small lakes. Faunas
of chironomids followed by oligochaetes were established in less than 2
years, and in some cases less than 3 months.
Patchy distributions of benthos unrelated to measured substratum
properties could also be due to temporally varying local disturbances of
the bottom. In a correlation analysis of Lake Superior macrobenthos, Frei-
tag et a1. (1976) found no strong or consistent correlation of physical
parameters with benthos distributions. Partial correlation and multiple
regression analyses of Lake Ontario macrobenthos and physical-chemical
sediment properties were made by Nalepa and Thomas (1976), who found
that the distribution of the oligochaetes Limnodrilus hoffmeisteri and
Tubifex tubifex were related to chemical sediment properties only at deep-
water stations. The abundance of the amphipod Pontoporeia hoyi was
related to sediment grain size only at deep-water stations. At shallower
(more disturbed?) stations there was no significant relationship.
Effects of Benthos on Freshwater Sediment Physical Properties 163

6.2. Interactions among Macrofauna

Whether or not "successions" occur on lake bottoms, it is known that

freshwater benthos interact to alter their own abundances. A change in
community composition will also result in altered sediment properties.
Amensalistic, competitive, and mutualistic interactions have been de-
scribed. Jonasson and Thorhauge (1972) report that when Chironomus
anthracinus is abundant in Lake Esrom, the abundance of the tubificid
Pomatothrix hammoniensis decreases. In years when C. anthracinus is
absent, the number of tubificids increases. Brinkhurst and Kennedy (1965)
found a similar negative correlation between Chironomus thummi and
tubificid oligochaetes in different parts of a polluted stream. We have also
observed a negative interaction between two common species: Chirono-
mus plumosus, a surface deposit feeder, and the tubificid Limnodrilus
hoffmeisteri, a subsurface deposit feeder. In field colonization experi-
ments, the presence of large numbers of C. plumosus (>3000 m -2) was
associated with reduced abundances of oligochaetes. In laboratory ex-
periments, 1. hoffmeisteri avoid sediments inhabited by C. plumosus, and
migrate from sediment into which chironomids are introduced within 1
day. The reason for this avoidance behavior is presently unknown. Chi-
ronomids burrow using their pro legs and mouthparts, and this motion
can sever tubificids if they are encountered. Other possibilities are that
chironomids alter tubificid food resources by altering sediment chemistry
or that they excrete some toxic metabolite. Tubificids feed in the reducing
zone and chironomids oxidize sediment by pumping water through their
burrows. Lastly, surface deposit-feeding activities of chironomids may
inhibit tubificid respiration. Within the deposit feeding Tubificidae, both
intraspecific and interspecific competition have been found to influence
population abundance and distribution (Jonas son and Thorhauge, 1976a;
Bonomi and DiCola, 1980; Milbrink, 1980; Lang and Lang-Dobler, 1980).
Brinkhurst et al. (1972) suggest a mutualistic interaction among three
species of tubificids from Toronto Harbor, Tubifex tubifex, 1. hoffmeisteri,
and Peloscolex multisetous. Feeding (pelletization) and growth of these
species is greater in mixed culture than in pure culture, each species
stimulating growth of the others in an undetermined manner. Sephton
and Paterson (1980) found that the bivalve molluscs of Movice Lake in-
creased the numerical abundance of nonbivalve benthos (primarily chi-
ronomids) living close to them. Stanczykowsa (1978) found a similar
result near patches of the clam Dreissena. We have observed the presence
of a thickened and more well-defined black iron sulfide zone and greater
abundance and deeper penetration into the substratum by tubificid oli-
gochaetes in laboratory microcosms with unionids (Lampsilis radiata, 20
m -2) than in microcosms without clams (Fig. 28).
164 Chapter 3

Many of these mutualistic interactions are probably mediated by

changes in microbial food supply brought about by the activities of ma-
crobenthos. Hargrave (1970b,c) for example. found that the burrowing
and feeding activities of the freshwater amphipod Hyalella azteca stim-
ulated bacterial respiration in bottom sediments from Marion Lake. Fu-
kuhara et al. (1980) found that ammonia production in the 0- to 5-cm
layer of submerged ricefield soils was doubled under the influence of
feeding by L. hoffmeisteri and Branchyura sowerbyi; the number of aerobic
bacteria decreased, while the number of sulfate reducers increased. We
have found that the sediment-mixing activities of tubificids routinely
double the total number of bacteria (measured by epifluorescent micros-
copy) in the top 5 cm of Lake Erie sediment (Fig. 29). Pringle (1980)
reports an increase in diversity and biomass of diatoms in stream substrata
containing chironomid tubes. Aller (Chapter 2, Section 6) discusses the
ways in which stimulation of microbial production occurs. The effect of
this activity is to change sediment properties either directly by changing
the binding properties of the bottom or indirectly by altering the abun-
dance of other benthos. Many deposit-feeding benthos feed selectively on
bacteria (Marzolf, 1965; Hargrave, 1970b; Wavre and Brinkhurst, 1971).

Figure 28. Tubificid burrows and black sulfide layer in laboratory aquaria (A) with no union-
ids and (B) with unionids (20 individuals m -2). Scale bar: 1 em.
Effects of Benthos on Freshwater Sediment Physical Properties 165

NO. BACTERIAlg
DRY SEDIMENT x 10 9

o 4 8 12

~~
0 ......
4
..JW
W()
!XI-<

Figure 29. Effects of tubificid oligochaetes ~ffi

w!z
8

on bacterial abundances in Lake Erie sedi- Q-

ments . • , Microcosm with no worms; .,
microcosm with 5 x 104 worms m -2 12

Deposit-feeding tubificids are quite sensitive to changes in bacterial pop-

ulations, as the following experiment shows. 1. hoffmeisteri were placed
in a 37-cm 2 microcosm tube containing two sediments. One half contained
unaltered central basin sediment irradiated with a 60Co source to stop
bacterial activity. The tubificids avoided the irradiated sediment until

100

108 !zW
75 ~

-< is
a:W W
U)

~ ~
()
-<
!XI 50 a:
~
U)

u..
0
a: 10 7 ~
W
!XI !zW
~
::I 25 ()
a:
z W
Q.

o 5 10 15
TIME (IN DAYS)
Figure 30. Oligochaete abundance in presterilized sediment as a function of time . • , Abun-
dance of Limnodrilus hoffmeisteri; ., bacterial abundance in experimental cell; ., bacterial
abundance in sterilized central basin mud to which no worms were added.
166 Chapter 3

bacteria entered the log growth phase. Then the worms moved en masse
to the irradiated sediment (Fig. 30). Thus changes in bacterial numbers
can cause localization of macrofauna with resultant heterogeneities in
sediment properties.

7. Future Work
On the basis of either organism size, abundance, or casual observa-
tion, we suggest that mayfly nymphs, amphipods, unionid clams, and
meiofaunal ostracods, copepods, and nematodes are candidates for further
study. Also, although it is clear that the cosmopolitan chironomids are
very important in altering sediment chemistry (Chapter 4), probably too
little has been made of their effects on physical sedimentary properties.
Workers in the field have been concerned primarily with the organ-
ismal effects on sediment stratigraphy, physical properties that bear on
sediment transport, and physical-chemical properties that bear on di-
agenesis and flux of dissolved materials across the interface. (This last
topic is the subject of Chapters 2 and 4). Models of biogenic sediment
mixing are fairly well developed (Chapter 7). What is not well understood
is which sedimentary materials are transparent to mixing and which are
not. In addition, the mode of formation and preservation of laminae in
lake sediments is poorly understood. Such sedimentary structures are a
potentially useful tool in reconstructing the history of sediment transport
(Nittrouer and Sternberg, 1981; Chapter 6) and deserve further study.
There are no standard methods for examining biogenic modification
of sediment transport. Each worker uses a flume and bottom sampling
device of different construction, and each measures critical erosion points
in a different manner. Consequently, precise intercom paris on of results
is at present nearly impossible. We need some standard methods. Espe-
cially desirable would be an inexpensive quantitative way to measure the
critical erosion point. It is, however, quite likely that we will eventually
realize that the measurement of the point at which erosion is initiated is
not enough. Then the problem will become how best to measure the flux
rate of sediment from the bottom as a function of applied stress. One
interesting substantive problem for future work is the determination of
the rate of biogenic change of surface sediment properties that control
sediment transport-how long, for instance, does it take for biota to return
the bottom to an equilibrium condition following a large storm, or is there
no equilibrium? Another is the measurement of the vertical gradient of
erodibility of natural sediments. The few observations we have made
(Table VI) suggest that the gradient is quite steep over the top few mil-
limeters of sediment. Accurate calculation of the amount of sediment
Effects of Benthos on Freshwater Sediment Physical Properties 167

eroded during a storm, for instance, requires that we know this gradient.
Finally, we echo the sentiment of Rhoads and Boyer (Chapter 1) that the
quantitative assessment of the significance of microbial films is probably
the single most important problem in understanding biologically con-
trolled sediment transport.
We ought to begin also to examine the broader significance of our
results. Over short time scales (perhaps a year or less) biogenic modifi-
cations of the bottom can have important effects on bottom stability and
water column turbidity that can result in changes in sediment transport
and benthic and pelagic community composition. But what about longer
time scales? It is not clear, for example, how important biogenic control
of erosion is compared to sediment transport in the 5-year, 50-year, or
100-year storm.
It may turn out that biogenic modification of soft bottoms has greater
evolutionary significance than geological import, for it may be a powerful
selective force in benthic communities (Wilson, 1980). Wilson's idea is
that effects of organisms that are beneficial to their community can be
selected for; they are not mere epiphenomena. The idea that marine com-
munities function as "superorganisms" in which constituent species con-
tribute to the maintenance of the whole community is not new (Dunbar,
1960; Margalef, 1968), but Wilson (1980) is the first to put the idea on
what appears to be a firm evolutionary basis. Van Val en (1980) regards
this idea of selection for benefactor organisms as one of major significance
for evolutionary ecology. According to Wilson, species populations are
subdivided into groups that are homogeneous with respect to genetic
mixing (demes) and smaller groups that are homogeneous with respect
to the manifestation of individual traits (trait groups). Trait groups can
vary in their genetic composition. Variation among local groups of or-
ganisms that causes differential productivity of the groups can result in
an increase in the number of the organisms that increase the fitness of
their group. For example, earthworms improve the soil for plant growth
and plants provide food for worms. Some parts of a field will have worms
that aid plant growth better than other worms. More plants means more
earthworms and the patches with more beneficient worms will expand.
The temptation to explain the evolution of benefactors among marine and
freshwater soft bottom benthos by analogous reasoning is great, since soft-
bottom infauna so easily and greatly modify their surrounding environ-
ment and yet are sensitive to substratum properties. In addition, benthos
are relatively sedentary organisms for much of their life cycle, concen-
trating movement on a brief dispersal stage. Thus, subdivided populations
are probably common. It is probably not an accident that 20 of 42 examples
of nontrophic interactions (not predation and not competition) that affect
fitness (Wilson, 1980) are taken from benthic and soil communities.
168 Chapter 3

ACKNOWLEDGMENTS. The original research published in this chapter was

funded by grants from the National Science Foundation (OCFC-8005103)
and the National Oceanic and Atmospheric Administration
(NA80RAD0036), Office of Marine Pollution Assessment. Karen Toil
typed and edited several drafts of the manuscript. Fred Soster worked up
most of the data on colonization in Lake Erie. Walter Borowski did the
oligochaete sediment preference experiment. Sharon Matis provided the
chloride diffusion data. In developing ideas for this paper, we benefited
greatly from discussions with D. C. Rhoads, J. B. Fisher, and W. J. Lick.

References
Adams, C. E., and Kregar, R. D., 1969, Sedimentary and faunal environment of eastern Lake
Superior, in: Proceedings of the 12th Conference on Great Lakes Research, pp. 1-20,
International Association for Great Lakes Research, Ann Arbor, Michigan.
Alley, W. P., and Anderson, R. F., 1968, Small scale patterns of spatial distribution of the
Lake Michigan Macrobenthos, in: Proceedings of the 11th Conference on Great Lakes
Research, pp. 1-10, International Association for Great Lakes Research, Ann Arbor,
Michigan.
Alley, W. P., and Chin, J. J., 1978, Length-biomass relationships of the amphipod, Ponto-
poreia affinis of Lake Michigan, in: Abstracts of the 20th Conference on Great Lakes
Research, p. 2, International Association for Great Lakes Research, Ann Arbor, Michigan.
Alsterberg, G., 1925, Die Nahrungszirkulation einiger Binnensetypen, Arch. Hydrobiol.
15:291-338.
Appleby, A. G., and Brinkhurst, R. 0., 1970, Defecation rate of three tubificid oligochaetes
found in the sediment of Toronto Harbour, Ontario, J. Fish. Res. Board Can. 27:1971-1982.
Baker, F. c., 1928, The freshwater mollusca of Wisconsin. Part II. Pelecypods, Wisc. Geol.
Nat. Hist. Surv. Bull. 70.
Berg, K., 1938, Studies on the bottom animals of Esrom Lake, K. Dan. Vidensk. Selsk. Skr.
8:1-255.
Birtwell, I. R., and Arthur, D. R., 1980, The ecology of tubificids in the Thames estuary with
particular reference to Tubifex costatus (Clapanedi), in: Aquatic Oligochaete Biology
(R. O. Brinkhurst and D. G. Cook, eds.), pp. 331-382, Plenum Press, New York.
Bonomi, G., and DiCola, G., 1980, Population dynamics of Tubifex tubifex, studied by means
of a new model, in: Aquatic Oligochaete Biology (R. O. Brinkhurst and D. G. Cook, eds.),
pp. 185-204, Plenum Press, New York.
Bousfield, E. L. , 1958, Freshwater amp hi pod Crustacea of glaciated North America, Can .
Field Nat. 72:55-113.
Boyer, L. F., 1980, Production and preservation of surface traces in the intertidal zone, Ph.D.
dissertation, Department of the Geophysical Sciences, University of Chicago, Chicago,
Illinois.
Brinkhurst, R. 0., 1964, Observations on the biology of lake-dwelling Tubificidae, Arch.
Hydrobiol. 60:385-418.
Brinkhurst, R. 0., 1965, The use of sludge worms (Tubificidae) in the detection and as-
sessment of pollution, in: Proceedings of the 8th Conference on Great Lakes Research,
pp . R09-10, International Association for Great Lakes Research, Ann Arbor, Michigan.
Brinkhurst, R. 0., 1974, The Benthos of Lakes, St. Martin's Press, New York.
Brinkhurst, R. 0., and Chua, K. E., 1969, A preliminary investigation of some potential
Effects of Benthos on Freshwater Sediment Physical Properties 169

nutritional resources by three sympatric tubificid oligochaetes, ]. Fish. Res. Board Can.
26:2659-2668.
Brinkhurst, R 0., and Jamieson, B. G. M., 1971, Aquatic Oligochaeta of the World, University
of Toronto Press, Toronto.
Brinkhurst, R 0 ., and Kennedy, C. R., 1965, Studies on the biology of the Tubificidae
(Annelida, Oligochaeta) in a polluted stream, ]. Anim. Ecol. 34:429-443.
Brinkhurst, R 0., Chua, K. E., and Kaushik, N. K., 1972, Interspecific interactions and
selective feeding of tubificid oligochaetes, Limnol. Oceanogr. 17:122-133.
Burgis, M. J., Darlington, J. P. E. C., Dunn, I. G., Ganf, G. G., Gwahba, J. J., and McGowan,
L. M., 1973, The biomass and distribution of organisms in Lake George, Uganda, Proc.
R. Soc. London Ser. B 184:271-298.
Calow, P., 1975, Defecation strategies of two freshwater gastropods, Ancylus fluviatilis Mull
and Planorbis contortus Linn. (Pulmonata) with a comparison of field and laboratory
estimates of food absorption rates, Oecologia (Berlin) 20:51-63.
Caspers, H., 1980, The relationship of saprobial conditions to massive populations of tub-
ificids, in: Aquatic Oligochaete Biology (R O. Brinkhurst and D. G. Cook, eds.), pp.
503-506 , Plenum Press, New York.
Coker, R E., Clark, H. W., Shira, A. F., and Howard, A. D., 1921, Natural history and
propagation of freshwater mussels, Bull. U. S. Bur. Fish. 37:77-181.
Cole, G. A., 1953, Notes on the vertical distribution of organisms in the profundal sediments
of Douglas Lake, Michigan, Am. Midl. Nat. 49:252-56.
Colquhoun, 0., 1971 , Lectures on Biostatistics, Clarendon Press, Oxford.
Crowley, T. E., 1957, Age determination in Anodonta, ]. Conchol. 24:201-207.
Cullen, D. J., 1973, Bioturbation of superficial marine sediments by interstitial meiobenthos,
Nature 242:323-324.
Cummins, K. W., and Lauff, G. H., 1969, The influence of substrate particle size on the
microdistribution of stream macrobenthos, Hydrobiologia 34:145-181.
Curds, C. R, 1963, The flocculation of suspended matter by Paramecium caudatum,]. Gen.
Microbiol. 33:363-375.
Davis, R B., 1974, Stratigraphic effects of tubificids in profundallake sediments, Limnol.
Oceanogr. 19:466-488.
Dunbar, M. J., 1960, The evolution of stability in marine environments; natural selection at
the level of the ecosystem, Am . Nat. 94:129-136.
Edwards, R W., 1958, Vernal sloughing of sludge deposits in a sewage effluent channel,
Nature 180:100.
Eggleton, F. E., 1937, Productivity of the profundal benthic zone in Lake Michigan, Pap.
Mich. Acad. Sci. 22:593-611.
Eggleton, F. E., 1939, The deep water bottom fauna of Lake Michigan, Pap. Mich. Acad. Sci.
21:599-612.
Fehrmann, R C., and Weaver, R. W., 1978, Scanning electron microscopyof Rhizobium spp.
adhering to fine silt particles, Soil. Sci . Am. J. 42:279-281.
Ferrante, J. G., and Parker, J. I., 1977, Transport of diatom frustules by copepod fecal pellets
to the sediments of Lake Michigan, Limnol. Oceanogr. 22:92-98.
Fisher, J. A., and Beeton, A. M., 1975, The effect of dissolved oxygen on the burrowing
behavior of Limnodrilus hoffmeisteri (Oligo chaeta), Hydrobiologia 47:273-290.
Fisher, J. B., and Tevesz, M. J. S., 1976, Distribution and population density of Elliptio
complanata (Mollusca) in Lake Pocotopaug, Connecticut, Veliger 18:332-338.
Fisher, J. B., Lick, W. J., McCall, P. L., and Robbins, J. A., 1980, Vertical mixing of lake
sediments by tubificid oligochaetes, ]. Geophys. Res. 85:3997-4006.
Ford, J. B., 1967, The vertical distribution of larval Chironomidae in the mud of a stream,
Hydrobiologia 19:262-272.
170 Chapter 3

Foster, T. D., 1932, Observations on the life history of a fingernail shell of the genus Sphaer-
ium, J. Morpho!. 53:473-497.
Frankel, 1., and Mead, D. J., 1973, Mucilaginous matrix of some estuarine sands in Con-
necticut, J. Sediment. Petro!. 43:1090-1095 .
Freitag, R., Fung, P., Mothersill, J. S., and Prouty, G. K., 1976, Distribution of benthic
macroinvertebrates in Canadian waters of Northern Lake Superior, J. Great Lakes Res.
2:177-192.
Fukuda, M. K., 1978, The entrainment of cohesive sediments in freshwater, Ph.D. disser-
tation, Department of Mechanical and Aerospace Engineering, Case Western Reserve
University, 210 pp.
Fukuda, M. K. , and Lick, W., 1980, The entrainment of cohesive sediments in fresh water,
J. Geophys. Res . 85:2813-2824.
Fukuhara, H., Kikuchi, E., and Kurichara, Y., 1980, The effect of Branchiura sowerbyi on
bacterial populations in submerged ricefield soil, Oikos 34:88-93.
Gale, W. F., 1976, Vertical distribution and burrowing behavior of the fingernail clam,
Sphaerium transversum , Malacologia 15:401-409.
Ganf, G. G., and Viner, S. B., 1973 , Ecological stability in a shallow equatorial lake (Lake
George, Uganda). Proc. R. Soc. London Ser. B 184:321-346.
Geoghegan, M. J., and Brian, R. c. , 1948, Aggregate formation in soil. 1. Influence of some
bacterial polysaccharides on the binding of soil particles, Biochem . J. 43:5-13.
Gray, J. S., 1974, Animal-sediment relationships, Oceanogr. Mar. Bio!. Annu. Rev.
12:223-261.
Hakanson, L. , and Kallstrom , A., 1978, An equation of state for biologically active lake
sediments and its implications for interpretations of sediment data, Sedimentology
25:205-226.
Hamilton, A. 1., 1971, Zoobenthos of fifteen lakes in the experimental lakes area, North-
western Ontario, J. Fish. Res. Board Canada 28:257-263.
Haranghy, 1., 1971, Investigations on the life-span limitation of mussels as example of
telometric growing animals, Acta. Bio!. Acad. Sci. Hung. 22:3-7.
Hargrave, B. T. , 1970a, Distribution, growth, and seasonal abundance of HyalelJa azteca
(Amphipoda) in relation to sediment microflora, J. Fish. Res. Board Can. 27:685-699.
Hargrave, B. T., 1970b, The utilization of benthic microflora by Hyalella azteca (Amphipoda).
J. Anim. Eco!. 34:424-437.
Hargrave, B. T., 1970c, The effect of a deposit-feeding amphipod on the metabolism of
benthic microflora, Limno!. Oceanogr. 15:21-30.
Hargrave, B. T., 1976, The central role of invertebrate feces in sediment decomposition, in:
The Role of Terrestrial and Aquatic Organisms in Decomposition Processes 0. M. An-
derson and A. MacFayden, eds.). pp. 301-321, Blackwell Scientific Publications,
Oxford.
Harris , R. H., and Mitchell , R. , 1973, The role of polymers in microbial aggregation, Annu.
Rev. Microbiol. 27:27-50.
Heard, W. H., 1977, Reproduction of fingernail clams (Sphaeriidae: Sphaerium and Mus-
culium). Malacologia 16:421-455.
Hendelberg, J., 1960, The fresh-water pearl mussel. Margaritifera margaritifera (L.). Rep.
Inst. Freshwater Res. Drottningholm 41:149-171.
Henson, E. B., 1962, Notes on the distribution of the benthos in the Straits of Mackinac
region, in: Proceedings of the 5th Conference on Great Lakes Research, p. 174, Inter-
national Association for Great Lakes Research, Ann Arbor, Michigan.
Henson, E. B., 1970, Pontoporeia affinis (Crustacea, Amphipoda) in the Straits of Mackinac
region, in: Proceedings of the 13th Conference on Great Lakes Research, pp. 601-610,
International Association for Great Lakes Research, Ann Arbor, Michigan.
Effects of Benthos on Freshwater Sediment Physical Properties 171

Herdendorf, C. E., and Reutter, J. M., 1976, Pre-operational aquatic ecology monitoring
program for the Davis-Besse nuclear power station, Unit 1. Progress Report July
1-December 31, 1975, CLEAR, Ohio State University, Columbus, Ohio.
Heuschele, S., 1980, Vertical distribution of benthos in profundal Lake Superior sediments,
in: Titles and Abstracts, North American Benthological Society, 28th Annual Meeting,
Savannah, Georgia, p. 22.
Hilsenoff, W. 1., 1966, The biology of Chironomus plumosus in Lake Winnebago, Wisconsin,
Ann. Entomol. Soc. Am. 59:465-473.
Hilsenoff, W. 1., 1967, Ecology and population dynamics of Chironomus plumosus (Diptera,
Chironomidae) in Lake Winnebago, Wisconsin, Ecology 60:1183-1194.
Holland, E. F. , Zingmark, R G. , and Dean, J. M. , 1974, Quantitative evidence concerning the
stabilization of sediments by marine benthic diatoms, Mar. BioI. 27:191-196.
Howard, J. D., and Elders, C. A., 1970, Burrowing patterns of haustoriid amphipods from
Sapelo Island, Georgia, in: Trace Fossils (T. P. Crimes and J. C. Harper, eds.), pp.
243-262, Seel House Press, Liverpool.
Hung, B. P., 1953, The life history and economic importance of a burrowing mayfly, Hex-
agenia limbata, in southern Michigan lakes, Bulletin of the Institute of Fisheries Re-
search, Michigan Conservation Department, No.4, p. 151, Ann Arbor.
Hunt, S., 1970, Polysaccharide-Protein Complexes in Invertebrates, Academic Press,
London.
Hynes, H. B. N., 1955, The reproduction cycle of some British freshwater Gammaridae, J.
Anim. Ecol. 24:352-360.
Hynes, H. B. N.,1970, The Ecology of Running Waters, University of Toronto Press, Toronto.
Iovino, S. J., and Bradley, W. N., 1969, The role of larval chironomidae in the prodUction
of lacustrine copropel in Mud Lake, Marion County, Florida, Limnol. Oceanogr.
14:898-905.
Ivlev, V. S., 1939, Transformation of energy by aquatic animals, coefficient of energy con-
sumption by Tubifex tubifex (Oligochaeta), Int. Rev. Ges. Hydrobiol. 38:449-458.
Johnson, M. G., and Brinkhurst, R 0 ., 1971, Benthic community metabolism in Bay of
Quinte and Lake Ontario, J. Fish. Res. Board Can . 28:1715-1726.
Johnson, M. S., and Munger, F., 1930, Observations on excessive abundance of the midge
Chironomus plumosus at Lake Pepin, Ecology 11:110-126.
Johnson, R 1., 1970, The systematics and zoogeography of the Unionidae (Mollusca: Bivalvia)
of the southern Atlantic slope region, Harv. Univ. Mus. Compo Zool. Bull. 140:263-450.
Jonasson, P. M., 1972, Ecology and production of the profundal benthos in relation to
phytoplankton in Lake Esrom, Oikos Suppl. 14:1-148.
Jonasson, P. M., 1978, Zoobenthos of lakes, Verh. Int. Verein. Limnol. 20:13-37.
Jonasson, P. M., and Thorhauge, F., 1972, Life cycle of Pomatothrix hammoniensis (Tubi-
ficidae) in the profundal of a eutrophic lake, Oikos 23:151-158.
Jonasson, P. M., and Thorhauge, F., 1976a, Population dynamics of Pomatothrix hammon-
iensis in the profundal of Lake Esrom with special reference to environmental and
competitive factors, Oikos 27:193-203 .
Jonasson, P. M., and Thorhauge, F., 1976b, Production of Pomatothrix hammoniensis in the
profundal of eutrophic Lake Esrom, Oikos 27:204-209.
Jones, B. F., and Bouser, C. J., 1978, The mineralogy and related chemistry of lake sediments,
in: Lakes: Chemistry, Geology, Physics (A. Lerman, ed.), pp. 179-236, Springer-Verlag,
New York.
Juday, C., and Birge, E. A. , 1927, Pontoporeia and Mysis in Wisconsin lakes, Ecology
7:445-452.
Kennedy, C. R, 1966, The life history of Limnodrilus hoffmeisteri Clap. (Oligochaeta: Tub-
ificidae) and its adaptive significance, Oikos 17:158-168.
172 Chapter 3

Kikuchi, E., and Kurihara, Y., 1977, In vitro studies on the effects of tubificids on the
biological. chemical and physical characteristics of submerged ricefield soil and over-
lying water, Oikos 29:248-356.
Kleckner, J. F., 1967, The role of the bottom fauna in mixing lake sediments, M. S. thesis,
University of Washington, Seattle.
Konstantinov, A. S., 1969, Feeding of chironomid larvae and means of improving the food
base of water bodies, in: Transactions of the 6th Conference on Biology of Inland Waters
(N. S. Akatora and B. K. Shtegman, eds.). pp. 274-284, Israel Program for Scientific
Translation, IPSP No. 5136, Jerusalem.
Kovacik, T. 1., and Walters, 1. J., Jr., 1973, Mercury distribution in sediment cores from
western Lake Erie, in: Proceedings of the 15th Conference on Great Lakes Research, pp.
252-259, International Association for Great Lakes Research, Ann Arbor, Michigan.
Kraft, K. J., 1979, Pontoporeia distribution along the Keweenaw shore of Lake Superior
affected by copper tailings, J. Great Lakes Res. 5:28-35.
Krezoski, J. R., and Robbins, J. A., 1980, Radiotracer studies of solute and particle transport
in sediments by freshwater macrofauna, in: Abstracts, International Association for
Great Lakes Research Meeting, Kingston, Ontario.
Krezoski, J. R., Mozley, S. c., and Robbins, J. A., 1978, Influence of benthic macroinverte-
brates on mixing of profundal sediments in southeastern Lake Huron, Limnol. Oceanogr.
23:1011-1016.
Ladle, M., 1971, The biology of Oligochaeta from Dorset Chalk streams, Freshwater BioI.
1:83-79.
Lang, C., and Lang-Dobler, B., 1980, Structure of tubificid and lumbriculid worm commun-
ities, and three indices of trophy based on these communities, as descriptors of eutro-
phication level of Lake Geneva (Switzerland). in: Aquatic Oligochaete Biology (R. O.
Brinkhurst and D. G. Cook, eds.). pp. 457-470, Plenum Press, New York.
Lautenschlager, K. P., Kaushik, N. K., and Robinson, J. B., 1978, The peritrophic membrane
and fecal pellets of Gammarus lacustris limnoeus Smith, Freshwater BioI. 8:207-211.
LeFevre, G., and Curtis, W. C., 1912, Studies on the reproduction and artificial propagation
of freshwater mussels, Bull. Bur. Fish. 30:109-201.
Lellak, J., 1969, The regeneration rate of bottom fauna populations of fish ponds after win-
tering or summering, Verh. Int. Verein. Limnol. 17:560-569.
Lyman, F. E., 1943, Swimming and burrowing activities of mayfly nymphs of the genus
Hexagenia, Ann. Entomol. Soc. Am. 36:250-256.
Macan, T. T., and Mackereth, J. C., 1951, Notes on Gammarus pulex in the English lake
district, Hydrobiologia 9:1-12.
McCall, P. 1., 1977, Community patterns and adaptive strategies of the infaunal benthos of
Long Island Sound, J. Mar. Res. 35:221-266.
McCall, P. L., 1978, Spatial-temporal distributions of Long Island Sound infauna: The role
of bottom disturbance in a nearshore marine habitat, in: Estuarine Interactions (M. 1.
Wiley, ed.). pp. 191-219, Academic Press, New York.
McCall, P. 1., 1979, The effects of deposit feeding oligochaetes on particle size and settling
velocity of Lake Erie sediments, J. Sediment. Petrol. 49:813-818.
McCall, P. L., and Fisher, J. B., 1980, Effects of tubificid oligochaetes on physical and
chemical properties of Lake Erie sediments, in: Aquatic Oligochaete Biology (R. O.
Brinkhurst and D. G. Cook, eds.). pp. 253-317, Plenum Press, New York.
McCall, P. L., Tevesz, M. J. S., and Schwelgien, S. F., 1979, Sediment mixing by Lampsilis
radiata siliquoidea (Mollusca) from western Lake Erie, J. Great Lakes Res. 5:105-111.
Mackey, A. P., 1977, Growth and development of larval chironomidae, Oikos 28:270-
275.
McLachlan, A. J., 1974, Development of some small lake ecosystems in tropical Africa, with
special reference to the invertebrates, BioI. Rev. 49:365-397.
Effects of Benthos on Freshwater Sediment Physical Properties 173

McLachlan, A, J" 1977, Some effects of tube shape on the feeding of Chironomus plumosus
L. (Diptera: Chironomidae), J. Anim. Ecol. 46:139-146.
McLachlan, A. J., and Cantrell, M. A., 1976, Sediment development and its influence on the
distribution and tube structure of Chironomus plumosus L. (Chironomidae, Diptera) in
a new impoundment, Freshwater Biol. 6:437-443.
McLachlan, A. J., and McLachlan, J. M., 1976, Development of the mud habitat during the
filling of two new lakes, Freshwater Biol. 6:59-67.
Mackie, G. L., Qadri, S. U., and Clarke, A. H., 1976, Intraspecific variations in growth, birth
periods, and longevity of Musculium securis (Bivalvia: Sphaeriidae) near Ottawa, Can-
ada, Malacologia 15:433-446.
Margalef, R., 1968, Perspectives in Ecological Theory, University of Chicago Press, Chicago.
Martin, J. P., 1946, Microorganisms and soil aggregation. II. Influence of bacterial polysac-
charides on soil structure, Soil Sci. 61:157-166.
Martin, J. P., and Richards, S. J., 1963, Decomposition and binding action of a polysaccharide
from Chromobacterium violaceum in soil, J. Bacteriol. 85:1288-1294.
Marzolf, G. R., 1965, Substrate relations of the burrowing amphipod Pontoporeia affinis in
Lake Michigan, Ecology 46:579-591.
Matteson, M. R., 1948, Life history of Elliptia complanatus (Dillwyn, 1917), Am. Midl. Nat.
40:690-723.
Meier-Brook, C., 1969, Substrate relations in some Pisidium species (Eulamellibranchiata:
Sphaeriidae), Malacologia 9:121-125.
Milbrink, G., 1973, On the vertical distribution of oligochaetes in lake sediments, Rep. Inst.
Freshwater Res. Drottningholm 53:34-50.
Milbrink, G., 1980, Oligochaete communities in pollution biology: The European situation
with special reference to lakes in Scandinavia, in: Aquatic Oligochaete Biology (R. O.
Brinkhurst and D. G. Cook, eds.), pp. 433-455, Plenum Press, New York.
Mills, E., 1969, The community concept in marine zoology, with comments on continua and
instability in some marine communities: A review, J. Fish. Res. Board Can. 26:1415-1428.
Mitropolsky, V. 1., 1966, Notes on the life cycle and nutrition of Sphaerium corneum L.
(Mollusca, Lamellibranchia), Trans. Inst. Bioi. Inland Waters Acad. Sci. USSR 12:125-128.
Monakov, A. V .. 1972, Review of studies on feeding of aquatic invertebrates conducted at
the Institute of Biology of Inland Waters, Academy of Science, U.S.S.R., J. Fish. Res.
Board Can. 29:363-383.
Moore, D. G., and Scruton, P. C., 1957, Minor internal structures of some recent unconso-
lidated sediments, A.A.P.G. Bull. 41:2723-2751.
Moore, J. W., 1979, Ecology of a subarctic population of Pontoporeia affinis Lindstrom
(amphipoda), Crustaceana 36(3):267-276.
Mozley, S. c., and Alley, W. P., 1973, Distribution of benthic invertebrates in the south end
of Lake Michigan, in: Proceedings of the 16th Conference on Great Lakes Research, pp.
87-96, International Association for Great Lakes Research, Ann Arbor, Michigan.
Mozley, S. C., and Garcia, L. C., 1972, Benthic macrofauna in the coastal zone of southeastern
Lake Michigan, in: Proceedings of the 15th Conference on Great Lakes Research, pp.
102-116, International Association for Great Lakes Research, Ann Arbor, Michigan.
Mundie, J. H., 1959, The diurnal activity of the larger invertebrates at the surface of Lac la
Ronge, Saskatchewan, Can. J. Zool. 37:945-956.
Myers, A. C., 1977, Sediment processing in a marine subtidal sand bottom community. 1.
Physical aspects, J. Mar. Res. 35:609-632.
Nalepa, T. F., and Thomas, N. A., 1976, Distribution of macrobenthic species in Lake Ontario
in relation to sources of pollution and sediment parameters, J. Great Lakes Res.
2:150-163.
Negus, C. L., 1963, A quantitative study of growth and production of unionid mussels in
the river Thames at Reading, J. Anim. Ecol. 35:513-532.
174 Chapter 3

Nittrouer, C. A., and Sternberg, R. W., 1981, The formation of sedimentary strata in an
allochthonous shelf environment: The Washington continental shelf, Mar. Geol.
42:201-232.
Okland, J., 1963, Notes on the population density, age distribution, growth, and habitat of
Anodonta piscinalis rilss. (Moll. Lamellibr.) in a eutrophic Norwegian lake, Nytt Mag.
Zool. 11:19-43.
Oliver, D. R., 1971, Life history of the chironomidae, Annu. Rev. Entomol. 16:211-230.
Palmer, M. F., 1968, Aspects of the respiratory physiology of Tubifex tubifex in relation to
its ecology, J. Zool. 154:463-473.
Parthenaides, E., and Passwell, R., 1968, Erosion of cohesive soil and channel stabilization,
Civil Engineering Report No. 19, State University of New York, Buffalo, 119 pp.
Paterson, C. G., and Fernando, C. H., 1969, The macro-invertebrate colonization of a small
reservoir in eastern Canada, Verh. Int. Verein. Limnol. 17:126-136.
Pavoni, J. L., Tenney, M. W., and Echelberger, W. F., Jr., 1972, Bacterial exocellular polymers
and biological flocculation, J. Water Poll. Cont. Fed. 44:414-431.
Pennak, R. W., 1978, Freshwater Invertebrates of the United States, John Wiley and Sons,
New York.
Pillai, S. c., 1941, The function of Protozoa in the activated sludge process, Curro Sci.
10:84-92.
Poddubnaya, T. 1., 1961, Data on the nutrition of the prevalent species of tubificids in the
Rybinsk basin, Tr. Inst. BioI. Vodokhran Akad. Nauk SSSR 4:219-231.
Poddubnaya, T. L., 1980, Life cycles of mass species of tubificidae (OligochaetaJ, in: Aquatic
Oligochaete Biology (R. O. Brinkhurst and D. G. Cook, eds.J, pp. 175-184, Plenum Press,
New York.
Postma, H., 1967, Sediment transport and sedimentation in the estuarine environment, in:
Estuaries (G. H. Lauff, ed.J, pp. 158-179, American Association for the Advancement
of Science, Washington, D.C.
Potter, D. W. B., and Learner, M. A., 1974, A study of the benthic macro-invertebrates of a
shallow euthrophic reservoir in South Wales with emphasis on the Chironomidae (Dip-
tera); their life-histories and production, Arch. Hydrobiol. 74:186-226.
Pringle, C., 1980, Effects of chironomid tube-building activities and resultant floral changes
upon the feeding behavior of Baetis, in: Titles and Abstracts, North American Ben-
thological Society, 28th Annual Meeting, Savannah, Georgia, p. 36.
Prokopovich, N. P., 1969, Deposition of clastic sediment by clams, J. Sediment. Petrol.
39:891-901.
Ravera, 0., 1955, Amount of mud displaced by some freshwater Oligochaeta in relation to
depth, Mem. 1st. Ital. Idrobiol. 8(suppl.):247-264.
Rawson , D. S., 1953, The bottom fauna of Great Slave Lake, J. Fish. Res. Bd. Can. 10:486-520.
Reigle, N. J., 1967, An occurrence of Anodonta (Mollusca, Pelecypoda) in deep water, Am.
MidI. Nat. 78:530-531.
Rhoads, D. C., 1974, Organism-sediment relations on the muddy sea floor, Mar. BioI. Annu.
Rev. 12:263-300.
Rhoads, D. C., 1975, The paleoecologic and environmental significance of trace fossils, in:
The Study of Trace Fossils (R. W. Frey, ed.J, pp. 147-160, Springer-Verlag, New York.
Rhoads, D. c., and Young, D. K., 1970, The influence of deposit-feeding organisms on
sediment stability and community trophic structure, J. Mar. Res. 28:150-178.
Rhoads, D. c., Yingst, J. Y., and Ullman, W. J., 1978, Seafloor stability in central Long Island
Sound. Part 1. Temporal changes in erodibility of fine-grained sediment, in: Estuarine
Interactions (M. Wiley, ed.J, pp. 221-244, Academic Press, New York.
Robbins, J. A., Krezoski, J. R., and Mozley, S. C., 1977, Radioactivity in sediments of the
Great Lakes: Post-depositional redistribution by deposit-feeding organisms, Earth
Planet. Sci. Lett. 36:325-333.
Effects of Benthos on Freshwater Sediment Physical Properties 175

Robbins, J. A., Edgington, D. N., and Kemp, A. 1. W., 1978, Comparative lead-210, cesium-
137, and pollen geochronologies of sediments from Lakes Ontario and Erie, Quat. Res.
10:256-278.
Robbins, J. A., McCall, P. 1., Fisher, J. B., and Krezoski, J. R, 1979, Effect of deposit feeders
on migration of 137CS in lake sediments, Earth Planet. Sci. Lett. 42:277-287.
Sanders, H. L., 1956, Oceanography of Long Island Sound. X. The biology of marine bottom
communities, Bull. Bingham Oceanogr. ColI. 15:245-258.
Sanders, H. 1., 1968, Marine benthic diversity in a comparative study, Am. Nat. 102:224-
282.
Sanders, H. 1., and. Hessler, R R, 1969, Ecology of the deep-sea benthos, Science
162:1419-1424.
Scoffin, T. P., 1970, The trapping and binding of subtidal carbonate sediments by marine
vegetation in Bimini Lagoon, Bahamas, J. Sediment. Petrol. 40:249-273.
Segerstrale, S. G., 1967, Observations of summer breeding in populations of the glacial relict
Pontoporeia affinis Lindstr. (Crustacea, Amphipoda) living at greater depths in the
Baltic sea, with notes on the reproduction of P. femorata Kroyer, J. Exp. Mar. BioI. Ecol.
1:55-64.
Sephton, T. W., and Paterson, C. G., 1980, Bivalve-non-bivalve relationships in a small
reservoir in New Brunswick, Canada, in: Titles and Abstracts, North American Ben-
thological Society, 28th Annual Meeting, Savannah, Georgia, p. 39.
Sly, P. G., 1978, Sedimentary processes in Lakes, in: Lakes-Chemistry, Geology, Physics
(A. Lerman, ed.,J, pp. 65-89, Springer-Verlag, New York.
Smith, W. E., 1972, Culture, reproduction, and temperature tolerance of Pontoporeia affinis
in the laboratory, Tr. Am. Fish. Soc. 101:253-256.
Sorokin, J. I., 1966, Carbon-14 method in the study of the nutrition of aquatic animals, Int.
Rev. Geol. Hydrobiol. 51:209-224.
Southard, J. B., 1974, Erodibility of fine abyssal sediments, in: Deep-Sea Sediments (A. 1.
Inderbitzen, ed.J, pp. 367-379, Plenum Press, New York.
Stanczykowska, A., 1978, Occurrence and dynamics of Dreissena polymorpha, Verh. Inter-
nat. Verein. Limnol. 20:2431-2434.
Stansbery, D. H., 1967, Growth and longevity of naiads from Fishery Bay in western Lake
Erie, Am. Malacol. Union. Inc. Bull. 35:10-11.
Stansbery, D. H., 1970, A study of the growth rate and longevity of the naiad Amblema
plicata (Say, 1817) in Lake Erie (Bivalvia: Unionidae), Am. Malacol. Union. Inc. Bull.
37:78-79.
Stober, Q. J., 1972, Distribution and age of Margaritifera margaritifera (1.) in a Madison river
(Montana, U.S.A.) mussel bed, Malacologia 11:343-350.
Stockner, J. G., and Lund, J. W. G., 1970, Live algae in postglacial lake deposits, Limnol.
Oceanogr. 15:41-58.
Sundborg, A., 1956, The River Klavalven: A study of fluvial processes, Geogr. Ann.
38:127-316.
Sweeney, R, Foley, R, Merckel, C., and Wyeth, R., 1975, Impacts of the deposition of
dredged spoils on Lake Erie sediment quality and associated biota, J. Great Lakes Res.
1:162-170.
Tenore, K., 1968, Accumulation of detritus and phosphate from the sediments by the brackish
water bivalve, Rangia cuneata, U. S. Fish Wildl. Servo Circ. 289:7-8.
Tevesz, M. J. S., and McCall, P. L., 1979, Evolution of substratum preference in bivalves
(Mollusca), J. Paleontol. 53:112-120.
Tevesz, M. J. S., Soster, F. M., and McCall, P. L., 1980, The effects of size-selective feeding
by oligochaetes on the physical properties of river sediments, J. Sediment. Petrol.
50:561-568.
Thomas, G. J., 1965, Growth in one species of sphaerid clam, Nautilus 79:47-58.
176 Chapter 3

Thut, R. N., 1969, A study of the profundal bottom fauna of Lake Washington, Ecol. Monogr.
39:79-100.
Timm, T., 1975, Zoobenthos of Lake Vortsjarv in 1964-1972, Hydrobiol. Res. Tartu.
6:165-200.
Tudorancea, C., 1969, Comparison of the populations of Unio tumidus Philipsson from the
complex of Crapina-Dijila marshes, Ekol. Pol. Ser. A 27:185-204.
Tudorancea, C., and Gruia, 1., 1969, Observations on the Unio crassus Philipsson population
from the Nera River, Muz. Nat. Natur. Grig. Antipa. 8:381-394.
Unz, R. F., and Farrah, S. R., 1976, Exopolymer production and flocculation by Zoogloea
MP6, Appl. Environ. Microbiol. 31:623-626.
Van Cleave, H. J., 1940, Ten years observation on a fresh-water mussel population, Ecology
21:363-370.
Van Valen, 1. M., 1980, Patch selection, benefactors, and a revitalization of ecology, Evol.
Theory 4:231-233.
Wachs, B., 1967, Die Oligochaeten-Fauna der Fleissgewasser unter besonderer Beriicksichtigung
der Beziehungen zwischen der Tubificiden-Besiedlung und dem Substrat, Arch . Hy-
drobiol. 63:310-386 .
Walshe, B. M., 1947, Feeding mechanisms of Chironomus larvae, Nature 16:474.
Walshe, B. M., 1951, The feeding habits of certain chironomid larvae subfamily Tendipe-
dinae, Proc. Zool. Soc. 121:63-79.
Walters, 1. J., Wolery, T. J., and Myser, R. D., 1974, Occurrences of As, Cd, Co, Cr, Cu, Fe,
Hg, Ne, Sb, and Zn in Lake Erie sediments, in: Proceedings of the 17th Conference on
Great Lakes Research , pp. 219-234 , International Association for Great Lakes Research,
Ann Arbor, Michigan.
Wavre, M., and Brinkhurst, R. 0., 1971, Interactions between some tubificid oligochaetes
found in the sediments of Toronto Harbour, Ontario, f. Fish. Res. Board Can. 28:335-341.
Wetzel, R. G., 1975 , Limnology, W. B. Saunders, Philadelphia.
Wiederholm, T., 1978, Long term changes in profundal benthos of Lake Malaren, Verh. Int.
Verein. Limnol. 20:818-824.
Wilson, D. S., 1980, The Natural Selection of Populations and Communities, Benjamin/
Cummings, Menlo Park, California.
Wood, K., 1953, Distribution and ecology of certain bottom-living invertebrates of the western
basin of Lake Erie, Ph.D. dissertation, Ohio State University, Columbus, Ohio, 145 pp.
Yingst, J. Y., and Rhoads, D. C., 1978, Seafloor stability in central Long Island Sound. II.
Biological interactions and their importance for seafloor erodibility, in: Estuarine In-
teractions (M. Wiley, ed.), pp. 245-260, Academic Press, New York.
Yokley, P., 1972, Life history of Pleurobema cordatum (Rafinesque 1820) (Bivalvia: Union-
acea), Malacologia 11:351-364.
Young, R. A., and Southard, J. B., 1978, Erosion of fine-grained marine sediments: Sea-floor
and laboratory experiments, Geol. Soc. Am. Bull. 89:663-672.
Chapter 4
Effects of Macrobenthos on the
Chemical Diagenesis of Freshwater
Sediments
J. BERTON FISHER

1 . Introduction ............. .. ... ... ...... .. .. .. ... .... .. ... .. ..... ... ... .. .. .. ... ........ .. ..... .. ... .. .. ... ... ........ .. ... .. . 177
2. Freshwater Sediment System .... .. ...... .......... .. ... .. ... .. .. ... ... .. .. ... ... .. .. ... ... .. .. ... ... .. .. .... .. .. . 178
2.1. Macrobenthos ... .......... ....... .. ..... ................................................ .. ..... .. ................... 178
2.2. Chemical Diagenesis of Freshwater Sediments .................... .. ...... .. ................... 182
3. Mechanisms by Which Benthos Affect Chemical Diagenesis ........ ... .. .... .. ... .. .. ...... .. 186
3.1. Particle Transport .. .. .. .. ..... .... .. .. .... ........... .. ... .. ... ... ......... .. .... ....... .. ... ... .... ........ .. ... 186
3.2. Fluid Transport .. ....... .......... .... .... ... .. ... .. .. ....... ... ....... .. ...... .. .... .... .. .... ... .... .... ......... 189
3.3. Sediment Fabric Alteration ...... .. .. .. .. .... .. .. .. .. .... .. ................ .. .. .. ........... .. .... .. ........ 194
3.4. Addition of Reactive Materials ........................ .. ........................................ .... .. .. . 199
4. Observed Effects of Freshwater Macrobenthos on Chemical Diagenesis .......... .. .... . 200
4.1. Effects of Benthos on the Sedimentary Chemi cal Milieu .. .. .... .. .. ........ .. ... .... .. .. 200
4.2. Effects of Benthos on Diagenetic Rates .. ........ .. .... .. .. .. .... .. .. .. .... .. .. ...... .. .. .. .. ..... .. . 204
4 .3 . Effects of Benthos on Materials Flux ............ .. .. ..... .. .... .... .. .. ... .. ........ .... ...... ...... . 205
5. Conclusions .. .. ....... .. ... .. ... ...... .. .... ..... ... .. .. ..... .. ... ... .. ..... .... ... ... .. .. ... .... ... ... ... .. ... .. ... .. .. ..... 209
References .... ..... ....... ................. .. ....... ................. ... ...... .......... .... .. .. ...... .. ............... ... .... 211

1. Introduction
In this chapter the effects of macrobenthos on chemical changes in sed-
iments occurring during and after burial (Le., chemical diagenesis) will
be considered. The importance of sediments to the biogeochemical cycling
of materials is well known (Mortimer, 1941, 1942, 1971 ; Lee, 1970). Fresh-
water sediments act as both a source and a sink for biologically important
materials such as phosphorus, carbon, nitrogen, sulfur, and silicon. Fur-
thermore, sediments are known to play an active role in regulating cycles
of trace metals, radionuclides, and xenobiotics (Jones and Bowser, 1978).

J. BERTON FISHER • Department of Geological Sciences. Case Western Reserve Univer-

sity. Cleveland. Ohio 44106. Present address: Research Center. Amoco Production Company.
Tulsa. Oklahoma 74102.

177
178 Chapter 4

Because of this, knowledge of the chemical diagenesis of sediments is

essential to an understanding of materials cycling in freshwater
environments.
Historically, students of chemical diagenesis have viewed sediments
as a homogeneous medium dominated by one-dimensional vertical dif-
fusion. Reactive particles fall onto the sediment-water interface, move
into the sediments at the net rate of sedimentation, and react. Solutes
derived from these and subsequent reactions either move across the sed-
iment-water interface via molecular diffusion and/or participate in fur-
ther reactions (e.g., precipitation, adsorption).
Macrobenthos, through their burrowing, feeding, locomotive, respi-
ratory, and excremental activities play an important role in mediating
both physical and chemical processes near the sediment-water interface.
Sediment and interstitial water are vertically and laterally transported
(i.e., advected) by feeding, defecation, and respiratory pumping. Dwelling
and feeding burrows may act as conduits for material exchange via fluid
advection, solute diffusion, or sediment slumping. Secretion of mucus
and excretion of metabolites may create local "hot spots" of bacterial
activity (Aller, 1977, 1978; Aller and Yingst, 1978; McCall et a1., 1979).
Consequently, in the presence of macrobenthos, surfacial sediments are
not a homogeneous medium dominated by one-dimensional vertical dif-
fusion. Rather, they are vertically and laterally heterogeneous, strongly
mixed, and penetrated by solute exchange conduits.
In the following sections the mechanisms by which macrobenthos
affect chemical diagenesis will be examined. Specifically, we will be
concerned with how and to what extent the activities of macrobenthos
modify (1) sedimentary Eh and pH, (2) rates of chemical diagenesis, and
(3) materials flux across the sediment-water interface. Although this
chapter is concerned with an overview of how and to what extent mac-
robenthic organisms affect chemical diagenesis of freshwater sediments,
nearly all available information concerns animal-sediment interactions
in fine-grained (silt-clay), clastic (detrital), profundal (vegetation-free)
lake sediments . As a consequence, examples of animal-sediment inter-
actions in this environment will be used to illustrate the effects of mac-
robenthos on the diagenesis of freshwater sediments.

2. Freshwater Sediment System

2.1. Macrobenthos
Freshwater sediments are inhabited by a variety of macrobenthos,
(operationally defined as those organisms retained by a 500-f.\,m mesh
sieve)-principaUy arthropods (insects and amphipods), annelids (oli-
Effects of Macrobenthos on Freshwater Sediment Diagenesis 179

gochaetes and leeches), and molluscs (bivalves and gastropods) (Brink-

hurst, 1974)-all of which interact in some way with their substratum.
For any particular organism, the nature of this interaction is determined
by the manner in which food is obtained (trophic type), mobility of the
organism, and where, in relation to the sediment-water interface, life
activities are carried on (life position). The importance of any organism
or group of organisms in modifying sedimentary physical and/or chemical
properities is also dependent upon individual size (>10 cm for large
unionid b~valves), population density (up to >10 5 individuals m - Z), and
activity rates (which are typically temperature-dependent). Other biotic
and abiotic factors such as dissolved oxygen availability and ecological
interactions may also be important (McCall and Fisher, 1980).
Benthos obtain food in a number of ways. For marine benthos, these
methods of food acquisition have been categorized (Turpaeva, 1957; Sav-
ilov, 1957; Newell, 1970; Walker and Bambach, 1974). Although Cummins
(1973) has proposed such a classification for aquatic insects, no general
classification of trophic types exists for freshwater benthos. However, the
general scheme proposed by Walker and Bambach (1974) for marine ben-
thos is applicable to freshwater benthos. This scheme is presented in
Table I along with freshwater examples of each trophic type. The feeding
activity of an organism may embrace more than one trophic type. For
example, some chironomid species engage in browsing, suspension feed-
ing, and deposit feeding (Monakov, 1972; Cummins, 1973). The life habit

Table I. Classification and Terminology of Feeding in Benthic InvertebratesO

Trophic type Description Freshwater examples

Suspension feeding Removing food material from suspension Unionid bivalves.

in the water mass without the need to sphaerid bivalves
subdue or dismember particles.
Deposit feeding Removing food from sediment either Oligochaetes, amphipods.
selectively or nonselectively without chironomids
the need to subdue or dismember
particles
Browsing Acquiring food by scraping plant Gastropods, chironomids,
materials from environmental surfaces crayfish
or by chewing or rasping larger plants.
Carnivory Actively capturing and subduing prey. Leeches, chironomids,
crayfish
Scavenging Consuming large particles of dead Isopods. chironomids,
organisms. crayfish
Parasitism Obtaining nutrition from the fluids or Leeches
tissues of host organisms.

a After Walker and Bambach (1974).

 ~
00
Q

Table II. Trophic Type and Life Habit of Dominant Profundal Macrobenthos
Typical
Length population
scale density Primary trophic Life habit
Organism (cm) (m- 2 ) type Species referenced Mobility Activity zone reference

Chironomids 1-2 10-10 4 Multiple trophic Chironomus plumosus Sedentary (can Infaunal McLachlan and
types C. riparius be mobile)" McLachlan
C. tentaus (1976)
Oligochaetes 1-10 10-10 6 Subsurface Limnodrilus hoffmeisteri Sedentary (can Infaunal Alsterberg (1922)
deposit feeding Stylodrilus heringianus be mobile)
Tubifex tubifex
L. socialis
Branchyura sowerbyi
Unionide 3-10 100 _10 2 Suspension Anodonta radiata Sedentary (can Infaunal Coker et a1.
feeding silquoidae be mobile) (1921), Tevesz
Elliptio complanata and McCall
Anodonta sp. (1979)
Amphipods 0.5-1 10-10 3 Surface deposit Pontoporeia hoyi Mobile Epifaunal! Marzolf (1965)
feeding shallow
infaunal

o Capable of swimming.

n;:r
~

~
co
...,
>I>
Effects of Macrobenthos on Freshwater Sediment Diagenesis 181

of an organism is defined by its mobility and life position. Benthos are

mobile, sedentary, or attached, and epifaunal (most activity at or on the
sediment-water interface) or infaunal (most activity below the sedi-
ment-water interface).
The profundal macro benthic community of lakes is numerically dom-
inated by oligochaetes, amphipods, insect larvae (notably chironomid
larvae), and sphaerid and unionid clams. These organisms are classified
using trophic type/habit characteristics in Table II. Shallow infaunal de-
posit feeders, such as amphipods, and shallow infaunal suspension feed-
ers, such as sphaerid clams, affect only the upper 1-3 cm of deposit.
Infaunal forms that construct semipermanent dwelling and feeding bur-
rows, such as chironomid larvae, will not significantly contribute to sed-
iment mixing (at least for one generation of the larvae). Infaunal subsurface
deposit feeders, such as tubificid oligochaetes, and deep infaunal, mobile,
suspension feeders, such as unionid clams, are capable of thoroughly
mixing sediments within their life zone (~10 cm for unionids) (McCall
et aI., 1979). A generalized diagram showing the life positions of the
dominant freshwater benthos of the profundal zone is given in Fig. 1.

'---~ ......... ; ... ~---........... .. ~ ...... ' ... _........ .. , .... .,. ........... -....... - - ...... .
-

---~;L
-f-tj --
&.~_-b. ~ _~_:t:
CHR
.--..

-
..........
-.,. I AMP

I em
t--------l

....
• • _ ••~.:• •~ '..;1

Figure 1. Life positions of dominant profundal freshwater macrobenthos. AMP, Amphipods;

CHR.chironomids; TO, tubificid oligochaetes; LJC, unionid clams.
182 Chapter 4

Other ways of looking at macrobenthic communities (e.g., biomass,

relative sediment reworking activity, relative water pumping) may pro-
duce a somewhat different view of "dominance." For example, Fisher
and Tevesz (1975) report that, within its life zone, the unionid clam
Elliptio complanata accounts for only 0.2-2.5% of the total number of
benthos but comprises 79-98% of the living biomass (excluding shell
mass).

2.2. Chemical Diagenesis of Freshwater Sediments

The chemical diagenesis of sediments comprises those chemical re-

actions taking place during and after burial of sedimentary materials.
These reactions can be divided into two categories, biogenic and abio-
genic, based on whether or not the reactions are mediated by bacteria
or other microorganisms (Berner, 1976). The early diagenesis of both
marine and freshwater sediments is largely carried on by bacterially me-
diated reactions. Moreover, pore water chemistry is often dominated by
these reactions (Manheim and Sayles, 1974), which, in general, control
the Eh and pH of sedimentary environments (ZoBell, 1946; Baas Becking
et aI., 1960; Berner, 1963; Stumm and Morgan, 1970; Ben-Yaakov, 1973;
Gardner, 1973). Many geochemically important bacteria are heterotrophs.
These organisms gain energy for cell growth and maintenance by oxidizing
organic materials using a variety of electron acceptors (Kuznetsov, 1970).
A schematic outline of these reactions is given in Table III. For simplicity,
organic matter is represented here as CHzO. The sequence in which these
reactions occur closely parallels the order of decreasing free energy change
(LlGhN) of the reactions. Thus, the metabolic activities of microorganisms
tend to create a vertical zonation of biogenically mediated diagenetic
reactions (see Kuznetsov, 1970; Mortimer, 1971). A schematic diagram of
this vertical sequence is given in Fig. 2.

Table III. Free Energy Change (Relative to Aerobic Respiration) and Schematic
Chemistry of Important Bacterially Mediated Diagenetic Reactions"

Reaction Schematic chemistry C1G~XN/~G~Rb

Aerobic respiration !CHzO + !Oz ~ !COz + !HzO 1.00

Denitrification tCHzO + ~NO:; + m + ~ tco z + rr,N z + ioHzO 0.95
Nitrate reduction !CHzO + !NO:; + iH + !COz + ~NH~ + kHzO 0.66
Fermentation CHzO + m 2 0 ---> mcoo - + !CH 3 0H + H + 0.21
Sulfate reduction tCHzO + *CO~ + A H + ~ tCOz + iHS - + tHz)O 0.20
Methane fermentation tCHzO ~ !COz + ~CH4 0.19

a Based on data from Stumm and Morgan (1970).

b t>.Glb(N, free energy change of the reaction; t>.G9..R, free energy change of aerobic respiration.
Effects of Macrobenthos on Freshwater Sediment Diagenesis 183

...INCREASING DEPTH

Figure 2. Schematic diagram showing the relative vertical sequence of diagenetic reactions
in freshwater sediments.

Solutes produced by organic decay are free to react. These "second-

ary" reactions further change sediment chemistry. Hydrogen sulfide pro-
duced by sulfate reduction and the putrefaction of sulfur-containing or-
ganics reacts with iron and other metallics to form various sulfides
(Vallentyne, 1961, 1963; Swain, 1965; Doyle, 1967; Nriagu, 1968; Dell,
1971). Hydrogen ions produced by organic decay react with solid car-
bonates and sulfides to liberate metallics (e.g., Ca, Fe, Mn) and raise
alkalinity (Presley and Kaplan, 1968; Weiler, 1973). Phosphorus released
from decaying organics can precipitate as authigenic phosphate minerals
(D'Anglejan, 1967; Dell, 1973; Shapiro et aI., 1971; Otsuki and Wetzel,
1972; Muller and Forstner, 1973; Nriagu and Dell, 1974; Norvell, 1974;
Nriagu, 1976; Emerson and Widmer, 1978). Destruction of organic coat-
ings by bacterial decay may promote the abiogenic dissolution of diatoms
and other particles of biogenic opaline silica (Berner, 1976).
Diagenetic reactions that are not directly or indirectly controlled by
the metabolic activities of microorganisms are less numerous. Examples
are the dissolution of opaline silica, CaC0 3 dissolution at the
sediment-water interface, clay mineral adsorption reactions, and the au-
thigenesis of various clay minerals (Carmouze et al., 1977).
The chemical diagenesis of freshwater sediments has received con-
siderably less attention than that of marine sediments. In lacustrine sys-
tems, diagenetic processes can be complicated by seasonal patterns of
thermal stratification accompanied by oxygen depletion of hypolimnetic
184 Chapter 4

waters. Anoxia in the hypolimnion greatly facilitates the flux of several

constituents from sediments to overlying water (e.g., iron, manganese,
silica, ammonia, and phosphate) (Mortimer, 1941, 1942, 1971; Hutchin-
son, 1957; Gorham, 1964; Lee, 1970; Burns and Ross, 1972). For a review
of work on the chemistry of interstitial waters from marine and lacustrine
sediments, see Glasby (1973).
Is chemical diagenesis in freshwater sediments parallel to that in
marine sediments? Are there differences in the process between these
environments? Fresh water is, by definition, quite different from oceanic
water.
The chemistry of oceanic water represents a "titration" of basic crus-
tal materials (oxides, carbonates, and silicates) by strong acids derived
from volcanic gases (Lafon and Mackenzie, 1966), while the composition
of fresh water can be viewed as the result of interaction between atmos-
pheric CO 2 and crustal materials. Furthermore, oceanic waters represent
a long-time-scale interaction between water and the earth's crust while
fresh waters represent a much shorter time of interaction. While the com-
position of oceanic waters shows little areal variation, the composition
of fresh waters exhibits considerable variation among lakes and streams,
since it reflects local rock types and other local conditions (Hutchinson,
1957). Even so, it is possible to make some general statements about the
differences between oceanic and fresh waters.
In Table IV, the composition of present oceanic water is compared
to that of world average river water, median terrestrial water, and water

Table IV. Comparison of the Major Cation and Anion Chemistry and
Approximate Ionic Strengths of Present Ocean Water, World Average River
Water, Lake Erie Water, Lake Superior Water, and Median Terrestrial Watera

Present Median
oceanic World average Lake Erie Lake Superior terrestrial
Ion water!> ri ver water' waterd water d water"

Na' 480.6 0.27 0.52 0.04 1.21

Mgz+ 54.5 0.17 0.33 0.01 0.41
Ca 2 + 10.5 0.37 0.92 0.32 1.10
K+ 10.6 0.06 0.03 0.03 0.03

Cl 560.4 0.22 0.71 0.03 0.48

SO~- 28.9 0.12 0.27 0.04 0.31
HC03 2.3 0.96 1.85 0.84 3.77

Ionic strength! 0.71 0.002 0.005 0.001 0.006

a Values in mM. Ionic strength in M.

b Riley (1971). ' Wetzel (1975). d Upchurch (1975). B Davies and DeWiest (1955).

f Here, ionic strength, I, is calculated using only the major cations and anions listed. I = ~lc'Zf, where
C, is the concentration of the ith species and Z, is the charge of the ith species.
Effects of Macrobenthos on Freshwater Sediment Diagenesis 185

from two Laurentian Great Lakes, Lake Superior and Lake Erie. First, it
is immediately apparent that fresh waters are far more dilute than oceanic
water. The ionic strength of oceanic water is nearly three orders of mag-
nitude greater than that of fresh waters. Second, the patterns or relative
abundances of cations and anions are different. For ocean water the gen-
eral pattern is Na+ > Mgz+ > K+ > Ca z +, and Cl- > SO~- > HC0 3 .
For fresh water the general pattern is Ca z + > Na + > MgZ + > K +, and
HC0 3 > Cl- > SO~ +. With respect to the chemical diagenesis of sedi-
ments, the most remarkable difference between marine and fresh waters
is the paucity of SO~ - in fresh waters. In marine systems, the reduction
of sulfate largely controls the diagenetic milieu, and it has been the subject
of numerous studies (e.g., Berner, 1964a,b, 1970, 1972, 1974; Presley and
Kaplan, 1968; Gardner, 1973; Ben-Yaakov, 1973; Rees, 1973; Goldhaber,
1974; Goldhaber and Kaplan, 1975a,b). In freshwater systems (excluding
SO~- -rich waters), however, the reduction of sulfate is probably unim-
portant relative to aerobic respiration and methanogenesis (Hayes, 1964).
Bacterial reduction of sulfate is independent of sulfate concentration for
sulfate concentrations ::::;10- 3 M (Postgate, 1951; Harrison and Thode,
1958; Kaplan and Rittenberg, 1964; Ramm and Bella, 1974). In most fresh-
water environments, sulfate concentration (in the overlying water) is ap-
proximately two orders of magnitude less than this threshold value. Few
detailed measurements of sulfate concentration gradients exist for fresh-
water sediment. Recent data (T. Tisue, personal communication) from
Lake Erie and Lake Michigan indicate that sulfate concentrations fall off
quite rapidly with depth in sediments. In Lake Erie, no sulfate was found
below a depth of 1 cm, while in Lake Michigan sulfate was found only
to a depth of 4-10 cm.
One consequence of this lack of sulfur is high ferrous iron concen-
trations in the pore waters of freshwater sediments and the formation of
reduced authigenic iron minerals [e.g., vivianite, Fe 3 (P0 4Jz' 8H zO; greig-
ite, Fe3S4; siderite, FeC0 3] rarely found in marine sediments (Berner,
1980). For similar reasons, various authigenic manganese minerals [e.g.,
rhodochrosite, MnC0 3 ; reddingite, Mn(P0 4Jz' 3H zO; various forms of
MnS] are also found in freshwater sediments (Berner, 1980).
To be sure, sulfate-reducing bacteria are present in freshwater sedi-
ments (Kuznetsov, 1970; Dutka et al., 1974; Winfrey and Zeikus, 1979;
Kikuchi and Kurihara, 1977), and metallic sulfides, especially FeS, are
important constituents of the solid phase of freshwater sediments (Doyle,
1967; Nriagu, 1968, 1975; Jones and Bowser, 1978; Zinder and Brock,
1978; McCall and Fisher, 1980). It appears, however, that the principal
source of sulfur to many freshwater sediments is sulfur-containing organic
compounds (Nriagu, 1968; Wetzel, 1975; Zinder and Brock, 1978) rather
than sulfate. In fact, freshwater sediments with an unusually high content
186 Chapter 4

of organic matter (e.g., swamps, marshes, bogs) have an abundant source

of sulfide sulfur in their decomposing organic matter. In these sediments,
reactive iron is readily converted to pyrite (or marcasite) (Berner, 1980).
Except for the lack of a strong sulfate reduction reaction and its miner-
alogic consequences, diagenetic processes in freshwater sediments par-
allel those in marine sediments.

3. Mechanisms by Which Benthos Affect Chemical

Diagenesis
Benthos can affect the diagenetic environment in four ways: (1) trans-
portation of sediment particles, (2) transportation of interstitial and/or
overlying water, (3) alteration of sediment fabric, and (4) addition of
reactive materials. Transportation of sediment particles engendered by
the burrowing and feeding activity of benthos serves to mix surficial
sediments and results in the movement of materials between oxidizing
and reducing environments. The biogenic churning of surficial sediments
mixes freshly deposited organic matter with older material. Reduced
materials brought to the sediment-water interface can react with dis-
solved oxygen, thereby enhancing sediment oxygen demand. The net
effect of biogenic mixing is the alteration of the vertical profile of reactivity
in sediments. Advective transport of interstitial and/or overlying water
as a consequence of macro benthic respiration, feeding, and burrowing
will aid the movement of dissolved reactants (e.g., O2 , NO;- , NH.t) across
the sediment-water interface. Such enhanced transport may increase the
rate at which diagenesis proceeds. Alteration of sediment fabric through
the construction of burrows and the pelletization of sediments can also
affect solute transport by affecting the porosity and tortuosity of sedi-
ments. The addition of reactive materials through the excretion of urine,
mucus, and other materials by benthos may locally enrich sediments with
microbial nutrients and result in intense local bacterial activity.

3.1. Particle Transport

Movement of sediment particles by benthos, since it can significantly

confound stratigraphy, has received a good deal of attention. The particle
reworking activities of tubificid oligochaetes have been particularly well
studied (Alsterberg, 1922, 1925; Lundbeck, 1926; Ivlev, 1939; Ravera,
1955; Poddubnaya, 1961; Poddubnaya and Sorokin, 1961; Wachs, 1967;
Appleby and Brinkhurst, 1970; Davis, 1974a; Aller and Dodge, 1974; Kre-
zoski et a1., 1978; Robbins et a1., 1979; Fisher et a1., 1980; Tevesz et a1.,
Effects of Macrobenthos on Freshwater Sediment Diagenesis 187

1980), and the rate and nature of their particle movement activities have
been quantified and modeled (Robbins et aI., 1979; Fisher et aI., 1980).
Detailed quantitative data on particle reworking also exist for the lum-
briculid oligochaete Stylodrilus heringianus (Krezoski, 1981); the am-
phi pod Pantapareia hayi (Robbins et al., 1979; Krezoski, 1981); the chi-
ronomid Chironamus tentans (Krezoski, 1981); and the bivalve molluscs
Pisidium, Sphaerium, and Lampsilis radiata siliquoidea (McCall et aI.,
1979; Krezoski, 1981). Knowledge of the particle transport activities of
other freshwater benthos is qualitative, but other organisms, especially
mayflies, gastropods, crayfish, and various bivalves, are certain to have
some impact on sediment mixing (see Chamberlain, 1975).
Several authors have quantified particle mixing by macrobenthos in
terms of a particle diffusion coefficient, K, by analogy to eddy diffusion
(Goldberg and Koide, 1962; Guinasso and Schink, 1975; Aller and Coch-
ran, 1976; Nozaki et aI., 1977; Robbins, 1978; Robbins et al., 1979). Using
this approach, the distribution of a particle-associated property, such as
utilizable organic matter, microfossils, or particle-bound radio nuclides
(e.g., 137CS, 210Pb, 234Th), is determined by particle diffusion, net sedi-
mentation, and chemical reaction (if any) or radioactive decay. For ex-
ample, a simple model for the distribution of a nonexchangeable particle-
bound radionuclide whose source is the sediment-water interface (see
Aller, 1978) can be written as

aAat = K (aazA) az _ hA
(aA)
2
_ w (1)
2

Where A is the activity of the radionuclide, t is time, z is depth, h is the

decay constant of the radionuclide, and w is the sedimentation rate. The
particle diffusion coefficient, K, is calculated from observed vertical dis-
tributions of radionuclides such as 210Pb. This type of model is an ap-
propriate description of particle reworking by organisms that redistribute
particles without any pattern of directionality. Robbins et a1. (1979) have
shown particle diffusion to be a good representation of particle reworking
by a population (16,000 individuals m - 2) of the amphipod P. hayi. The
effect of P. hayi on the redistribution of 137CS in an experimental micro-
cosm is shown in Fig. 14 in Chapter 3 (p. 126). In this experiment there
was no sedimentation and the decay of 137CS could be neglected (t1l2 for
137CS ~ experimental time frame) so the distribution of 137CS by the am-
phipods was described as

aAat = K (aazA)
2

2
(2)
188 Chapter 4

This approach gave an excellent description of particle reworking by P.

hoyi. Robbins et a1. (1979) give a value of 0.162 x 10-6 cm 2 'sec- 1 for
K. This is similar to K values determined for coastal marine sediments
(Guinasso and Schink, 1975; Aller and Cochran, 1976) and is comparable
to the K inferred from 210Pb profiles reported for a core from Lake Huron,
where P. hoyi is the principal macrofaunal species (Robbins et a1., 1977).
The particle reworking activities of such benthos as amphipods, is-
opods, unionid and sphaerid clams, gastropods, mayfly larvae, and chi-
ronomids-which more or less randomly move particles-should be ad-
equately described by particle diffusion models. Such models, however,
do not adequately describe reworking processes that have directionality
(Aller, 1978; Robbins et a1., 1979; Fisher et a1., 1980).
In freshwater environments, tubificid oligochaetes are important
agents of bioturbation, by virtue of their life habit, widespread distribu-
tion, and high population densities (McCall and Fisher, 1980). The par-
ticle reworking activity of tubificid oligochaetes is highly directional.
Tubificids ingest sediment at depth in the substratum and expel this
material at the sediment-water interface as cylindrical, mucus-bound
fecal pellets. Thus tubificids provide a direct link between two distinctly
different biogeochemical regimes and can be functionally classified as
vertically oriented subsurface deposit feeders, or "conveyor belt" species
(Rhoads, 1974). Because of the directionality of tubificid reworking ac-
tivities, a model incorporating advection is needed to describe adequately
their particle transport activities. Such a model has been developed by
Fisher et a1. (1980) from data on the redistribution of a 137Cs-Iabeled
sediment layer by tubificids in laboratory microcosms (also see Robbins
et a1., 1979). The effect of tubificid reworking on the redistribution of
a 137CS layer initially at the sediment-water interface is given in Fig. 13
in Chapter 3 (p. 126). For another model of tubificid reworking, see Davis
(1974a). For other examples of advective-type bioturbation see Amiard-
Triquet (1975) and Aller and Dodge (1974).
As correctly pointed out by Aller (1978), particle transport activities
per se will not, in general, directly control diffusion rates of interstitial
solutes. Effective solute diffusion coefficients, which vary between ~6.75
x 10-6 cm 2 'sec- 1 for H+ and ~0.61 x 10-6 cm 2 'sec- 1 for S2- (Ro-
binson and Stokes, 1959; Chapman, 1967; Manheim, 1970; Li and Gregory,
1974; Reid et a1., 1977), are approximately one to two orders of magnitude
greater than particle diffusion coefficients. It should be mentioned, how-
ever, that the redistribution of reactivity owing to particle transport phe-
nomena can strongly affect the reactivity of the sediment-water interface.
Robbins (1978) showed that, under conditions of constant sedimentation
rate, constant 210 Pb supply to the sediment-water interface, and rapid
Effects of Macrobenthos on Freshwater Sediment Diagenesis 189

biological mixing, the activity of 21°Pb at the sediment surface, Af(O), is

given by

(3)

where As is the activity (i.e., concentration) of 21°Pb in material falling

to the sediment-water interface, w is the sedimentation rate, X. is the 21°Pb
decay constant, and S is the depth of the mixed layer. This concept can
be extended to other materials subject to decomposition (e.g., organic N,
C, P) that fall to the sediment-water interface. The amount of reactive
matter at the sediment-water interface is decreased by increased depth
of biological mixing or an increased decay constant (decreased half-life).
High sedimentation rates or changes in the supply of reactive materials
to the sediment-water interface may mask this effect.
Movement of materials from depth to the sediment-water interface
can also enhance the reactivity of the sediment surface. In a study of the
effects of tubificids on sediment oxygen consumption, McCall and Fisher
(1980) found that the oxygen consumption of tubificid-inhabited sedi-
ments (T. tubifex, simulated population density 100,000 individuals m -2)
exceeded the simple sum of tubificid respiration and the oxygen demand
of sediments lacking a tubificid population by a factor of -2. They found
that 50-70% of the enhanced demand could be accounted for by the
oxidation of FeS brought to the sediment-water interface by tubificid
feeding.

3.2. Fluid Transport

Advective transport of pore and/or overlying water across the sedi-

ment-water interface is potentially the most significant way in which
macrobenthos might affect chemical diagenesis and materials balance in
freshwater systems. Although nearly all benthic organisms are capable of
causing some fluid advection, infaunal organisms that engage in active
burrow irrigation, such as chironomids and burrowing mayflies (Lyman,
1943; Walshe, 1947), or that might inject water to aid burrowing, such as
unionid clams (McCall et a1., 1979), are probably the largest contributors
to biogenic fluid advection.
Despite the potential importance of biogenic fluid transport, few ex-
perimental data are available. Moreover, the published data are entirely
limited to chironomids. How can the effects of chironomid activities on
the transport of pore and overlying water and their associated solutes be
190 Chapter 4

viewed? The simplest approach would be to model the effects of chiron-

omids on solute transport as "enhanced diffusion." By way of example,
this interpretation was applied to Tessenow's (1964) experimental data
on interstitial silica concentrations in the presence of various chironomid
population densities. A very simple one-dimensional one-layer diffu-
sional transport model was used:

(4)

where D is the effective solute diffusion coefficient. An analytical solution

was derived for the following boundary conditions:

C(z = 0, t) = Co
c(z~ 00, t) =C

and the initial condition

C(z, t = 0) = Ci

where Co is the silica concentration in the overlying water (assumed

constant), and Ci is the initial interstitial water silica concentration. Both
Co and Ci were estimated from Tessenow's (1964) data. The results of
computations assuming various values for D are shown in Fig. 3. As can
be seen, there is reasonable agreement between the experimental data and
the theoretical computations. The calculated flux of silica using D = 10
x 10 - 6 cm 2 . sec -1 is approximately three times that calculated with D
= 1 x 10 - 6 cm 2 . sec -1. Of course, the model used here is not really
appropriate since silica is being produced in the sediments. Nevertheless,
it does appear that a diffusion analogue adequately describes the data,
and that the presence of a chironomid population density of 770 indi-
viduals m -2 produced roughly a tenfold increase in diffusivity over that
observed in sediments unaffected by chironomids.
Edwards and Rolley (1965) studied the migration of Li + from over-
lying water into chironomid-inhabited sediments (Chironomous riparius,
simulated population densities 0,25,50, and 100 x 10 3 individuals m - 2).
These workers found that in sediments containing chironomid popula-
tions Li + concentrations in the top ~2 cm of sediment were equivalent
to those of the overlying water after only 6 hr. Diffusion alone cannot
account for this finding (see Manheim, 1970). The data presented by
Edwards and Rolley (1965) are not extremely detailed, but at the two
highest chironomid population densities examined, it appears that Li + -
Effects of Macrobenthos on Freshwater Sediment Diagenesis 191

Cs/C j
o.ro______________~Or·5~------------~I. O
I
o

E
~
.c 5
0..
Q)
o

10
Figure 3. Comparison of silica concentration profiles calculated using a one-dimensional,
one-layer diffusion model for various values of D (A: D = 1 X 10 - 6 cm 2 sec-'; B: D = 2
X 10- 6 cm 2 sec - '; C: D = 5 X 10 - 6 cm 2 sec-'; D: D = 10 X 10 - 6 cm 2 sec-') to the silica

concentration profile data of Tessenow (1964) in the presence of various chironomid pop-
ulation densities (0 , 0 individuals m - 2 ; D, 193 individuals m - 2 ; 0, 385 individuals m - 2 ;
x, 580 individuals m - 2; /';, 770 individuals m - 2). Plotted concentrations are nondimen-
sionalized (Cs/Cd. Cs is the silica concentration at depth z. Ci is the silica concentration at
z = 00

laden overlying water is moving into the sediment at a velocity of -5

x 10 - 5 cm ' sec - 1. This translates to a "water exchange rate" of -4.3
X 10 4 cm 3 ·m - 2 ·day-1. Although this value is high compared to the
biopumping rate of 7 X 10 3 cm 3 ·m- 2 ·day-1 given by McCaffrey et a1.
(1980) for Narragansett Bay sediments, it is not unreasonable, since the
activities of each chironomid need result in only -1-0.5 cm 3 • day -1 of
water exchange.
In experiments conducted in our laboratory, the effect of chironomids
(Chironomus plumosus, simulated population densities 0, 1096, 3289,
and 6798 individuals m - Z) on pore water chemical profiles and materials
flux across the sediment-water interface was investigated. Pore water
concentration profiles for Fez + and H4 Si0 4 observed in this experiment
are given in Figs. 4 and 5. The presence of chironomids was found to
suppress the concentration of both Fe z + and H4 Si0 4 in the top 12 cm of
the sediment column. Soluble reactive phosphorus (SRP), NHt, and
192 Chapter 4

2+
Fe (}lM)
0 600 0 600
0

II
I
12

I
I
24
E control
1096
u

I 0

\
~
a..
w
0
I
I
12

24
3289
6798

Figure 4. Vertical profiles of Fe 2 + in control and experimental (with Chironomus plumosus)

sediments. Simulated population densities (individuals m - 2) indicated on figures.

HC0 3 concentrations were also strongly suppressed in this zone. The

suppression of Fez + and SRP concentrations is probably the result of
deeper Oz penetration into chironomid-inhabited sediments (evidenced
by their oxidized burrow walls). The suppression of HC0 3 , H4 Si0 4 , and
NHt concentrations is probably the result of chironomid water pumping
activities. By way of illustration, direct flux measurements (made by meas-
uring overlying water concentration change) and indirect flux estimates
(made by applying Fick's law to pore water concentration gradients) for
H4 Si0 4 are compared in Table V. Direct and indirect flux estimates for
H4 Si0 4 give comparable results when no chironomids are present, but
these two estimates deviate progressively as chironomid population den-
sity increases. At the highest density used (6798 individuals m - Z) the
Effects of Macrobenthos on Freshwater Sediment Diagenesis 193

Si ( )lM)
0 1000 0 1000
0
I)
II
I
I
12

24

E c on t ro l 1096
0

:I: 0
l- ) I
Cl. II
lLI
0 I! \
12

24
3289 6798

Figure 5. Vertical profiles of Si in control and experimental (with Chironomus plumosus)

sediments. Simulated population densities (individuals m -2) indicated on figures.

Table V. Direct and Indirect Estimates of Silica Flux from Control

and Experimental (Chironomid-Inhabited) Sediments
Direct flux estimate Indirect flux estimate
Chironomus plumosus (x 10 - 6 ( x 10 - 6

(individuals m- 2 ) mol.m - 2. day-l) mol.m - 2 .day - l)

0 1393 1651
1096 2644 855
3289 4191 150
6798 4335 76
194 Chapter 4

direct flux estimate is nearly 60 times larger than the indirect flux estimate
because the activities of the chironomids have greatly decreased the
H4 Si0 4 concentration gradient at the sediment-water interface. Further-
more, the presence of chironomids has increased silica production, since
the amount of silica transferred to the overlying water is approximately
four times greater than can be accounted for by simply flushing interstitial
water from the sediments. The activity of the chironomids, rather than
just the presence of their burrows, is necessary for enhanced materials
flux across the sediment-water interface. In recent experiments con-
ducted at the University of Michigan (M. A. Quigley, personal commu-
nication), it was found that silica flux from sediments was reduced by a
factor of 6.7 when chironomids were inactivated by physostygmine, a
neurotoxin.
More work is needed to describe adequately the effects of fluid-
advecting benthos on materials movement across the sediment-water in-
terface. Carefully controlled experiments using tracers with well-known
properties that can be followed nondestructively and with high resolution
(e.g., 22Na) need to be conducted. In particular, the assumption of one-
dimensionality needs to be critically examined, since the presence of
burrows creates an intrinsically two-dimensional situation (see Aller,
1977, 1978; Aller and Yingst, 1978; Berner, 1980).

3.3. Sediment Fabric Alteration

The feeding and burrowing activities of macrobenthos alter sediment

fabric. Their construction of feeding and dwelling burrows, movement
through the sediment, and production of fecal pellets radically alter the
textural properties of surface sediment (Rhoads, 1970; Rhoads and Young,
1970; McCall, 1979; McCall and Fisher, 1980). Collectively, these activ-
ities increase the porosity and decrease the tortuosity of the substratum.
Such modifications of porosity and tortuosity affect diffusive flux across
the sediment-water interface, since

Jz = Dw (:2) C~) , z =0 (5)

where <p is porosity, e is tortuosity, Dw is the Fick's law diffusion coef-

ficient in free solution, and Jz is the diffusional flux in mass' area ~ 1. time ~ 1.
Tortuosity is defined as

e =dl- (6)
dz
 Effects of Macrobenthos on Freshwater Sediment Diagenesis 195

where d1 is the length of the actual sinuous path over a depth interval dz.
Porosity is the ratio of the volume of contained water to the total volume:

<!> = Vwater (7)

V water + VsoJids

In general, porosity and tortuosity are not treated explicitly. Rather, their
effect on diffusivity is taken into account using an effective or apparent
diffusion coefficient, De. The maximum value of De is, of course, Dw.
Typically, values of De for chemical species that do not participate in
exchange reactions are on the order of 0.55Dw (Li and Gregory, 1974), but
values of 0.1Dw are commonly observed (Manheim, 1970). Thus the max-
imum effect of sediment fabric alteration on diffusivity is about a factor
of ten. McCall et a1. (1979) showed that the activities of unionid bivalves
significantly increased sediment water content in the upper 4 cm of their
experimental sediments. The change in porosity (and, consequently, dif-
fusional flux) owing to unionid activity calculated from their water con-
tent data is fairly modest, however, varying from 10% to 20%. Krezoski
(1981) found that the presence of Sty1odri1us heringianus (a lumbriculid
oligochaete, simulated population density 67,000 individuals m - 2) pro-
duced a zone of enhanced constant porosity 5.5 cm in thickness, while
the porosity of control cells exhibited an exponential decrease to a con-
stant value at 4 cm. His data are given in Fig. 6. In experiments with
Pontoporeia hoyi these workers found no difference in porosity between

POROSITY (cp)

0 .0

0. . 0
2
o Conlrol 0 eo-
o Control • S. heringianu5 0 0-·
• P. hoyi present . 0 •
E
.g
4
presenl
eID _
.s::; o· ~ O
0.
~ 6 O .~

«>-
8 0_
0
o • ..:I)
10
0-0
Figure 6. Effect of the presence of Stylodrilus heringianus and Pontoporeia hoyi on the
porosity of lake sediments. Data from Krezoski (1981).
196 Chapter 4

sediments with a P. hoyi population (simulated population density 47000

individuals m - 2 ) and control sediments (see Fig. 6).
McCall and Fisher (1980), in a study of the effects of Tubifex tubifex
on the movement of dissolved material across the sediment-water inter-
face, found that the movement of chloride from sediments to overlying
water was enhanced by the presence of tubificids (simulated population
density 100,000 individuals m -2), but found no evidence to support the
contention (Wood, 1975) that tubificids pump water. The effects of the
tubificid population on chloride movement were found to be well-
approximated by assuming that the diffusivity of chloride in the tubificid-
affected sediments decreased linearly from 11 x 10 - 6 cm 2 • sec -1 at the
sediment-water interface to 5.5 X 10 - 6 cm 2 'sec- 1 (the diffusivity of
chloride in control sediments) at a depth of 0.75 cm and remained at this
value to the base of the sediment column. McCall and Fisher (1980) at-
tributed this effect to the tubificids' generation of a porous fecal pellet
zone in the upper 1 cm of sediment. Chatarpaul et a1. (1980) observed a

Relative Activity (cps)

-CONTROL -

ao
-2 :
0.4 08

6!
..
12 n'-r-'T'-r-"""-.or--'1....
2 -, 00 04

.':..
08

,
I2

..
Co .6

r:-=::==~ ,

".
• ,'6
• " Co

• Cs -1 37 -'
'\'
• No-22 . ,'
I.
f

0.01r 627 3hr

"/ 103721'0'

20

. ..:.
- WITH WORMS-

".
oJ
,
/' 6

..i
OJ
.<: . f
Ci 10 ·f
'
8

20
0 .01r 66621>r V _.
Figure 7. Effect of the presence of Stylodrilus heringianus on the migration of 22Na and 137CS
through lake sediments. Data from Krezoski (1981).
Effects of Macrobenthos on Freshwater Sediment Diagenesis 197

Relative Activity (cps)

-CONTROL -

00 04 0.8 12 0 .0 04 08 12 0.0 0.4 08 1,2

E • Cs -137
s .6 No- 22
..c:
is. 10
~

00", 125,8", 333 3",

20
- WITH AMPHIPOOS-

-2 : .:II

°b===XA~•

..c:
is. 10
~

00 ... 104.0 ... 339.0 ...

20

Figure 8. Effect of the presence of Pontoporeia hoyi on the migration of 22Na and 137CS
through lake sediments, Data from Krezoski (1981),

pattern of chloride loss from overlying water to sediment in the presence

of tubificids, which could also be explained as the result of enhanced
diffusion in the upper 1 cm of sediment.
Krezoski (1981) studied the effects of S. heringianus and P. hoyi on
the migration of 22Na through sediments. Data were collected by non-
destructively scanning the experimental cells with a gamma scanner
(Robbins et a1., 1979). Krezoski found that the presence of both S. her-
ingianus (simulated population density 67,000 individuals m - 2) and P.
hoyi (simulated population density 47,000 individuals m- 2) enhanced
the apparent effective diffusivity, De, of 22Na. In control sediments the
De of 22Na was found to be 0.92 x 10 - 5 cm 2 . sec -1. In three replicate
experiments with S. heringianus populations the average De of 22Na was
found to be 1.33 x 10- 5 cm 2 ·sec- 1 • In two replicate experiments with
P. hoyi the average De of 22Na was 1.45 x 10 - 5 cm 2 . sec -1. Data from
this study are given in Figs. 7 and 8. Krezoski (1981) treats the sediment
as a one-layer system (Le., De is the same at all depths in the sediment).
198 Chapter 4

It appears (at least for S. heringianus) that a two-layer model would be

more appropriate. In the presence of S. heringianus the activity of 22Na
in the sediment is equivalent to that in the overlying water to a depth of
~3 cm after 1053.6 hr. P. hoyi does not produce a similar pattern of 22Na
activity. In the presence of the amphipod, the observed 22Na profiles
appear to be well-approximated by a one-layer diffusion model.
Despite the relatively modest increase in De that can be attributed to
macro benthic sediment fabric alteration, it should be remembered that
these alterations are made in the most biologically active zone of the
sediment column. As a result, the biogeochemical consequences of such
enhanced diffusion may be large. For example, such enhanced diffusion
could greatly modify sediment oxygen demand, speed the mineralization
of organic matter entering sediments, and facilitate silica release.
In the presence of lateral pressure gradients resulting from waves and
currents, changes in porosity and tortuosity may enhance passive (i.e.,
not the result of biological pumping) ventilation of sediments (Vogel and
Bretz, 1972), since porosity and tortuosity affect the hydraulic conduc-
tivity (Le., permeability) of sediments. Using falling head permeameters,
McCall and Fisher (1980) found that the presence of tubificids signifi-
cantly increased the permeability of sediments and that the permeability

10

]
"'"uE
~ 5
ID

Q
)(

o
o 50 100
Tubificids / m2 x 103
Figure 9. Coefficient of permeability, K, as a function of tubificid (Tubifex tubifex) population
density. Confidence intervals are 95% Scheffe-type intervals. Data from McCall and Fisher
(1980).
Effects of Macrobenthos on Freshwater Sediment Diagenesis 199

increased with increasing tubificid population density. It appeared that

the increased permeability was principally due to decreased tortuosity
(owing to tubificid burrows), since no significant difference in porosity
was observed between control and experimental sediments. Their data
are presented in Fig. 9. The significance of biogenic ally enhanced perme-
ability in affecting chemical diagenesis has not been evaluated. It is likely,
however, to have an impact similar in nature to that of fluid advection.
Aside from modifying tortuosity and porosity, the production of fecal
pellets may significantly affect the diagensis of their contained organic
matter. Iovino and Bradley (1969) showed that Chironomus spp. in a small
Florida lake produced strongly compacted fecal pellets rich in undigested
blue-green algae. They suggest that the copropel comprised of these pellets
appears to be a modern analogue to the precursor of oil shale in the Green
River Formation.

3.4. Addition of Reactive Materials

Particle transport, fluid transport, and sediment fabric alteration can

all change the distribution of reactivity in sediments and the flux of
materials across the sediment-water interface, but none of these processes
alters the amount of reactive material in sediments. Aquatic macrobenthos
can also add reactive materials to sediment through the excretion of urine
(Potts, 1954a,b), mucus, and various sediment-agglutinating materials
(Chamberlain, 1975). Nearly all aquatic invertebrates excrete copious
amounts of hypoosmotic urine (Potts, 1954a,b; Parry, 1960; Staddon, 1969;
Stobbart and Shaw, 1974). In addition to large quantities of ammonia and
bicarbonate, the urine of macrobenthos contains bacterial nutrients such
as sugars and amino acids. When excreted reactive materials are associ-
ated with burrows, dramatic local chemical changes can occur. Aller
(1977) and Aller and Yingst (1978), working in the marine environment,
showed that organic enrichment of Amphitrite burrow walls increased
microbiological activity in the burrow walls (particularly sulfate reduc-
tion); produced an increase in the concentration of iron sulfides imme-
diately around the burrows; and enriched the burrow walls in Fe, Mn,
and other metals. McCall et al. (1979) found that the burrows of the
freshwater mussel Lampsilis radiata siliquoidea were surrounded by a
zone enriched in iron sulfide. This enrichment was surmised to be a result
of intense microbiological activity owing to local nutrient enrichment and
enhanced availability of electron acceptors near the clam. Injection of
oxygen-rich water by the clams may result in a local SO~ - source through
the oxidation of FeS and other easily oxidized metallic sulfides.
200 Chapter 4

4. Observed Effects of Freshwater Macrobenthos on

Chemical Diagenesis
Data on the effects of freshwater macrobenthos on chemical diage-
nesis are limited to their effects on sedimentary Eh and pH, sediment
oxygen consumption, and the flux of various materials across the sedi-
ment-water interface. Since biogenically mediated diagenetic reactions
in general produce H + and reduce Eh, it is appropriate that this discussion
begin with a consideration of the effects of benthos on sedimentary Eh
and pH.

4.1. Effects of Benthos on the Sedimentary Chemical Milieu

Knowledge of the Eh and pH of an environment furnishes a synoptic

view of chemical processes taking place, but does not provide detailed
information about the specific chemistry of the environment (Baas Beck-
ing et aI., 1960). Eh data provide information about the relative abundance
of electron acceptors (e.g., O2 , NOi, SO~-) and electron donators (e.g.,
HS -, NO;, NHt), and pH data reveal the abundance of H+. Neither
parameter provides specific knowledge concerning the chemical reactions
that produced these conditions. Consequently, information on sedimen-
tary Eh and pH is useful in the examination of the effects of benthos on
chemical diagenesis but, taken alone, provides no knowledge concerning
the exact mechanisms whereby diagenesis is modified. Even so, it is useful
to begin our examination of the effects of benthos on sediment diagenesis
with a discussion of the patterns of Eh and pH modification noted in their
presence.
Numerous workers have observed that the presence of macrobenthos
is correlated with a thickened oxidized (i.e., light-colored) layer [e.g.,
Milbrink, 1969 (tubificids); Hargrave, 1972, 1975; Kirchner, 1975 (mixed
natural populations)]. Davis (1974b), in an experimental study, found that
the presence of a tubificid population of 800 individuals m - 2 increased
the redox potential of profundal lake sediment between depths of 1 and
4 cm and thickened the oxidized zone from 0.3 to 1.6 cm. In another
experimental study, Edwards (1958) found that the introduction of a Chi-
ronomus riparius population (simulated population densities of 12,500
and 50,000 individuals m - 2) into settled activated sludge displaced the
redox potential originally defining the redox color boundary downward
to 8 cm from its initial depth of 3 cm. Furthermore, he noted that the
higher density (50,000 individuals m -2) of chironomids produced more
extensive shifting of Eh and that redox conditions returned to their orig-
inal distribution after the chironomids pupated and emerged. Schumacher
Effects of Macrobenthos on Freshwater Sediment Diagenesis 201

(1963) observed that the oxidized zone extended to at least 15 cm with

a tubificid population of 50,000 individuals m - 2 and that the depth of
the oxidized zone was correlated with the population density of worms.
McCall and Fisher (1980) found that homogenized Lake Erie sediments
lacking macrobenthos developed an oxidized zone 0.1-0.3 cm in thick-
ness; the same sediments inhabited by tubificids (100,000 individuals
m - 2) developed an oxidized zone 1-3 cm thick. McCall et a1. (1979)
observed that sediment immediately adjacent to the unionid Lampsilis
radiata silquoidea were oxidized.
Simply stated, an oxidizing environment contains an excess of elec-
tron acceptors (e.g., O2 , N0 3 , SO~ - ), while a reducing environment con-
tains an excess of electron donators (e.g., Fe 2 +, HS-, NHi). From the
above data it is clear that the actions of chironomids, tubificids, and
unionids aid the transport of electron acceptors into sediments and thus
increase the depth over which more energetic diagenetic reactions occur
(see Table II). Chironomids more or less continuously pump overlying
water through their burrows, tubificids enhance porosity in the upper 1
cm of substratum, and unionids episodically pump water into sediments
to aid burrowing. It is likely that such activity increases the rate of organic
decomposition.
The actions of some macrobenthos can also affect sedimentary pH.
Davis (1974b) showed that the presence of a tubificid population de-
creased pH in the upper 1-2 cm and increased pH below 2 cm. Edwards
(1958) found that the presence of chironomids increased sedimentary pH
in settled activated sludge over a depth range of 8 cm. Fisher and Matisoff
(1981) found that the presence of tubificids (100,000 individuals m -2)
produced a local pH minimum at a depth of 4-5 cm (see Fig. 10). They
also noted that no pH gradient occurred below 4-6 cm in control cells,
while in cells containing worms a continuous decrease in pH occurred
below the local pH minimum. Profiles of pH in both with- and without-
worm cases did not vary substantially over the course of the experiment
(89 days) and were comparable to the pH profiles reported by Davis
(1974b). In further experiments, (see Fig. 10) a thin layer (0.2 cm) of
Milorganite (Milwaukee sewage sludge) was placed at a depth of 2 cm.
In Milorganite-treated sediments without worms, a highly systematic time
series of pH profiles evolved, while in similarly treated sediments with
worms (100,000 individuals m -2) no change in pH profile occurred during
the course of the experiment (89 days). In fact, pH profiles in Milorganite-
treated sediments with worms were equivalent to those observed in un-
treated sediments with worms. The activities of macrobenthos obviously
affect sedimentary pH. These modifications of pH have been attributed
to enhanced exchange (presumably of acid metabolites) across the sedi-
ment-water interface (Edwards, 1958; Davis, 1974b). This may well be
202 Chapter 4

pH
pH
6 7 8
6 7 8

------------7<r
SWI o
o ----------------~---
0
SWI
o. - ---
# 06'0

~
Io

<
o
o
o
o
o
5 o
o
o
o o
o
o o
o
o
o
o
10
o
o
o
o
o
o
15 o (A)
(B)

0'
o
o '~~ ----- - -~--------

o
5 o
o
o
E o
.8 o
.r:
ti. o
<I> o
o o
o
o
10 o
o
o 0 0 o
o
o
o [] 0 o
o
o [] 0 o
o
o0 0
o (e) (D)
15

Figure 10. Vertical profiles of pH observed in homogenized Lake Erie sediments. (A) Control;
(B) with tubificids; (C) without tubificids, but with a Milorganite layer (2 cm depth); (D)
with tubificids and a Milorganite layer (2 cm depth). SWI, sediment-water interface. Data
from Fisher and Matisoff (1981).
 NH4 (fLM) t%j
NH4 (~M) ~
Cl)
o 250 9.00 o 250 500 ;:l.
I I I en
CONTROL WITH WORMS g.,
o o ~
I C""l
...,'"
o
0-
Cl)
,'I ;:1
;-
5 I, 5 ) oen
o
;:1
'TI
...,
Cl)
on
II ::r'
:E
10 ~ 10
E E ...,'co"
2 2 (J)
Cl)
~
~ 0-
C. C. S·
Q)
Cl)
~
o 15 15 8-
o
0;.
OQ
Cl)
;:1
Cl)
~.
en
20 20

25 25

Figure 11. Comparison of ammonia concentration profiles observed in homogenized Lake Erie sediments with and without
tubificids . Population density: 10 5 individuals m - 2. N
o
w
204 Chapter 4

true in the case of chironomids, but the strong alteration of pH profiles

by tubificids in the Milorganite-treated sediments and their effects on pH
deep in sediments are difficult to explain in terms of solute exchange
alone, because aquatic oligochaetes do not pump water (McCall and
Fisher, 1980; Krezoski, 1981).
In experiments in our laboratory, it was found that the presence of
tubificids (100,000 individuals m - 2) resulted in a local ammonium max-
imum at a depth of -4 cm (see Fig. 11). This result indicates that an
additional source of ammonia is present in the upper 3-5 cm of worm-
inhabited sediments and suggests a possible explanation for the observed
effects of tubificids on interstitial water pH: The worms may be excreting
enough ammonia to "buffer" pore waters against changes in pH.

4.2. Effects of Benthos on Diagenetic Rates

The actions of freshwater macrobenthos affect the transport of reac-

tants and reaction products (e.g., NHt ,HCO:1 ) across the sediment-water
interface. It should follow, therefore, that their activities should also affect
the rate at which diagenetic reactions take place. One index of diagenetic
rate is the rate at which electron acceptors are consumed by sediments.
Edwards and Rolley (1965), Hargrave (1975), and Gramm (1982) all
noted a significant increase in oxygen consumption by sediments inhab-
ited by various chironomid species (Chironomus riparius, C. anthricinus,
and C. pulmosus, respectively) that could not be explained as the simple
sum of chironomid respiration and sediment oxygen demand in the ab-
sence of chironomids. Andersen (1977) found that the rate at which ox-
ygen was consumed by sediments was correlated with the wet weight of
C. plumosus present, and that the oxygen demand of sediment containing
chironomids was approximately twice the simple sum of chironomid
respiration and sediment oxygen demand in the absence of chironomids.
The relationship between the wet weight of chironomids and reported
sediment oxygen demand (SOD) (Andersen, 1977) was as follows:

SOD[mmol·m - 2 ·day-l) = (14.97 mmol·m - 2 ·day)

+ (0.18 mmol·m - 2 ·day-l· mg wet biomass- 1 )*B

where B is wet chironomid biomass in milligrams.

The actions of tubificids also affect the rate of electron acceptor uptake
by sediments. McCall and Fisher (1980) showed that the presence of
Tubifex tubifex (100,000 individuals m -2 ) enhanced sediment oxygen
consumption by a factor of 2. Approximately 20% of the enhanced demand
was the result of tubificid respiration, while 50-70% of the enhanced
Effects of Macrobenthos on Freshwater Sediment Diagenesis 205

demand was due to the oxidation of FeS brought to the sediment-water

interface by tubificid feeding activities. The remaining 10-30% of the
enhanced demand was apparently due to increased microbial activity.
Thus, tubificids enhance the rate of organic decay not only through aiding
the transport of dissolved oxygen, but also by recycling sulfur to produce
SO~ - at the sediment-water interface.
Macrobenthos also affect denitrification. Andersen (1977) reported
that the presence of C. plumosus increased the rate of nitrate consumption
by sediments. Nitrate consumption by sediments was strongly correlated
with biomass of C. plumosus present. The consumption of nitrate was
dependent on nitrate concentration in the overlying water and C. plu-
mosus enhanced nitrate consumption by 0.11 x 10 - 3 mol· m - 2. day-1 . mg
wet biomass -1 at nitrate concentrations of between 0.14 and 0.21 mM
and by 0.39 x 10 3 mol·m- 2 ·day-1· mg wet biomass- 1 at nitrate con-
centrations of between 0.42 and 0.71 mM.
Tubificids have also been shown to enhance the consumption of
nitrate by sediments. Chatarpaul et al. (1979, 1980) showed that the pres-
ence of tubificids (mixed population of T. tubifex and Limnodrilus hoff-
meisteri, 10,000 to 13,000 individuals m - 2) increased the rate of both
nitrification and denitrification in sediments. In the absence of worms,
the measured rates of nitrification and denitrification were respectively
1.9 and 3.3 x 10- 3 mol·m- 2 ·day-1. Chatarpaul et al. (1980) further
showed that, even when clean glass beads were used as a substratum for
the tubificids, the concentration of nitrate in overlying water decreased
at a rate of 0.5 mmol· day-1 (no data given to convert to mol· m - 2. day-1).
Furthermore, the decrease in nitrate concentration could be stopped by
removing the worms. This finding suggests that denitrifying bacteria were
associated with the oligochaetes. Indeed, Chatarpaul et al. (1980) isolated
nitrifying and denitrifying bacteria from the body walls and guts of
tubificids.

4.3. Effects of Benthos on Materials Flux

The effects of chironomids on the flux of materials across the sedi-

ment-water interface have received considerable attention. Ganapati
(1949) found that C. plumosus larvae increased the flux of both ammonia
and phosphorus from sand filters of the Madras waterworks. Edwards
(1958) observed that C. riparius enhanced ammonia flux from sediments
and that the observed flux was greater by a factor 5 than that expected
from chironomid ammonia excretion. Tessenow (1964) showed that the
presence of C. plumosus (100 individuals m -2) increased silica flux by
2.28-2.88 mmol· m - 2 . day-I, while Weissenbach (1974) found that the
206 Chapter 4

Table VI. Chironomid-Induced Flux Increase Factors for Various Materials in

O

Three Lake Sediments b

Lake Total P SiD z NH4-N ND3-N Fe Mn

Trummen
Aerobic conditions 2.27 2.86 0.61 1.97 3.43 2.20
Nitrogen atmosphere 0.92 1.29 1.18 1.00 0.81 0.71
Arungen
Aerobic conditions 2.08 2.07 154.90 1.10 13.71 4.01
Nitrogen atmosphere 1.08 1.14 0.94 1.10 0.67 0 .91
Vombsjon
Aerobic conditions 4.43 2.06 33.40 0.86 1.03 1.90
Nitrogen atmosphere 1.26 0.98 1.25 1.00 0.97 0.89

U Flux after chironomid additionlflux before chironomid addition.

b Based on data from Graneli (1979). Chironomid population density: 1000 individuals m -'.

presence of chironomids (species not given, ~1000 individuals m- Z )

increased silica flux by 2.52-2.88 mmol·m-z·day - l. In a study of sed-
iments from three Swedish entrophic lakes (Trummen, Arungen, and
Vombsjon), Gramm (1979) observed silica flux increases of 0.25 , 0.21, and
0.64 mmol· m -z. day-l (respectively) per 100 chironomids m - z, which
corresponds to a flux increase by a factor of 2-3 over sediments not having
a chironomid population (1000 individuals m- Z ). Graneli (1979) found
that the presence of chironomids affected the flux of total phosphorous
(x 2- x 4), ammonia (x 0.61- x 155), iron (x 1- x 14), and manganese
( x 2- x 4) when the overlying water was aerobic, but had virtually no
effect on flux under anoxic conditions (see Table VI). Gallepp (1978, 1979)
found that both C. riparius and C. tentans caused increased phosphorus
flux from sediments. The larger species, C. tentans, produced the greater
effect. Release of total phosphorus at 20°C increased in a linear manner
from 9.69 to 30.35 mmol· m Z • day-lover a range of six chironomid den-
sities from 0 to 6585 larvae m - z. Increasing temperature from 10° to 20°C
had little effect on phosphorus release rates from sediments without chi-
ronomids, but release from sediments with C. tentans populations was
increased tenfold at the higher temperature. Gallepp, however, could not
differentiate between phosphorus translocated from the sediments and
phosphorus excreted by the chironomids, and felt that the increased phos-
phorus flux might be almost entirely caused by chironomid excretion.
Chironmids may also transport materials, such as phosphorus and iron,
from lakes when the larvae emerge as adults (Neame, 1975, 1977).
Macrobenthos aside from chironomids are also known to enhance
materials flux across the sediment-water interface. Jernelbv (1970) noted
that the transfer of Hg from sediments was affected by both tubificids and
the unionid clam Anodonta sp. Tubificids (size and species not given,
Effects of Macrobenthos on Freshwater Sediment Diagenesis 207

820,000 individuals m ~Z) were found to cause Hg release from 0-2.5 cm,
while Anodonta sp. (126 individuals m ~ Z) caused release of Hg from a
depth of 0.9 cm. In the absence of macrobenthos, Hg was released only
from the upper 1 cm. Davis et a1. (1975) showed that the presence of
tubificids (mixed population, Tubifex spp. ~90%, 2500 individuals m ~ Z)
accelerated the removal of phosphorus from overlying water, increased
the depth of phosphorus penetration, and reduced the proportion of SRP
in the interstitial water of the uppermost sediments. These workers also
noted that the presence of tubificids did not have an effect on phosphorus
release from sediments under low-oxygen conditions in the overlying
water. Kikuchi and Kurihara (1977), in a study of the effects of tubificids
on the chemical and physical characteristics of submerged ricefield soil,
found that the presence of tubificids (L. socialis and Branchyura sowerbyi,
8842 individuals m~Z) kept the activity of Fe z + high in the upper 1 cm
of the sediment, increased the movement of Fez + into the overlying water,
enhanced ammonia concentration in the sediment and ammonia flux
across the sediment-water interface, decreased the number of aerobic
bacteria in the upper 1 cm of sediment, increased the amount of hexose
in the overlying water, and increased the number of sulfate-reducing
bacteria throughout the sediment column.
In experiments in our laboratory, control sediments and sediments
preconditioned by tubificid activity (mixed population of 1. hoffmeisteri
and Limnodrilus cervix, 100,000 individuals m ~z; 30 days of tubificid
activity; homogenized Lake Erie Sediment) were sealed and their over-
lying water permitted to go anoxic. Measurements of the concentration
changes in bicarbonate, ammonia, ferrous iron, SRP, and dissolved silica
were made in the water overlying both control and worm-affected sedi-
ments. In addition, control and experimental microcosms were sacrificed
at intervals to follow the concentration profiles of these materials in the
interstitial water. It was found that the presence of tubificids had no effect
on bicarbonate flux; enhanced ammonia flux (x ~1.4); and suppressed
the flux of silica (x ~1.4), phosphorus (x ~6), and ferrous iron (x ~2).
A tentative explanation for the difference in phosphorus flux between
bioturbated and nonbioturbated sediments is that in nonbioturbated sed-
iments a ferric-hydroxide-rich surface layer forms owing to upward dif-
fusion of FeZ + and its oxidation at the sediment-water interface. This
iron-hydroxide-rich layer is the site for the absorption of upward-diffusing
phosphorus. Phosphorus entering this surface layer forms a ferric hy-
droxide-orthophosphate complex [Fex(OHh(x~y)(P04)Y' 2H zO] (Stamm
and Kohlschutter, 1965; Williams et a1., 1976a,b). Phosphorus thus bound
will be readily available for release to the overlying water when the ferric
hydroxide dissolves following anoxia. In sediments bioturbated by tubi-
ficids such an iron-rich surface layer is prevented from forming because
 SRP {fLMl SRP (fLMl N
o
o ~ I~ oI 75 150 00
I I
CONTROL WITH WORMS
o 0, I

II
5 5 I I

I I
10 10
E E
~ ~
.£: £
Q Q
Q) Q)
0
15 o 15

20 20

25 25

n
::r-
oo
Figure 12. Phosphorus concentration profiles observed in homogenized Lake Erie sediments with and without tubificids. -0
Population density: 10 5 individuals m -2. SRP, soluble reactive phosphorus. ro...,
*"
Effects of Macrobenthos on Freshwater Sediment Diagenesis 209

the worms continuously subduct material from the sediment-water in-

terface. This action of tubificids during oxic conditions reduces the initial
flux of phosphorus (and iron) when the overlying water is driven anoxic.
Furthermore, examination of interstitial water SRP concentration profiles
from a strongly bioturbated experimental core and its companion control
core sacrificed near the end of the experiment (Fig. 12) shows that the
worms prevented a strong phosphorus concentration gradient from form-
ing at the sediment-water interface. Indeed, quite a long period of worm
inactivity would be required before enough phosphorus could diffuse to
establish a phosphorus gradient in the bioturbated core equivalent to that
in the nonbioturbated sediment. The enhancement of ammonia flux in
the presence of worms appears to be the result of tubificid-induced am-
monia production in the upper few centimeters of microcosm sediments
(see Fig. 11). The suppression of silica flux in the presence of worms is
not currently explicable.

5. Conclusions
Freshwater macrobenthos affect the chemical diagenesis of sediments
in a number of ways: advection of sediment particles, "diffusion" of
sediment particles, pumping of interstitial and overlying water, alteration
of sediment fabric, and injection of reactive materials. What is the relative
importance of these processes to the chemical diagenesis of freshwater
sediments? The pumping of water through sediments is almost certainly
the most significant of these activities. Data on the effects of chironomids
and tubificids on materials flux across the sediment-water interface in-
dicate that water pumping by chironomids is approximately twice as
important as sediment fabric alteration by tubificids. The advection of
sediment particles from depth to the sediment-water interface by tub i-
ficids links two distinctly different biogeochemical regimes and certainly
enhances sediment oxygen consumption. The advection of sediment par-
ticles does not appear to increase materials flux out of sediments appre-
ciably, but possibly enhances the movement of some redox-sensitive ma-
terials from the overlying water into sediments, preconditions sediments
so as to inhibit the release of materials such as iron and phosphorus, and
modifies the reactivity of the sediment-water interface. As mentioned
earlier, the alteration of sediment fabric through the construction of bur-
rows and the generation of surficial fecal pellet layers are inherently less
important than water pumping in enhancing the movement of material
across the sediment-water interface, but can, at least theoretically, en-
hance the diffusivity of sediments by as much as a factor of 10. Empirical
observations have shown an increase in diffusivity by approximately fac-
210 Chapter 4

tors of 2-10. The importance of particle "diffusion" to chemical diage-

nesis probably depends on the length scale (i.e., size of the particle-
diffusing organism) of the process, local sedimentation rate, and mixing
rate. For example, the actions of P. hoyi are confined to uppermost 1 cm
of sediment, while those of large unionid clams may extend to 30 cm
(Robbins et al., 1979; McCall et al., 1979). Consequently, the effects of
these two organisms on surface reactivity are vastly different. P. hoyi
could reduce surface reactivity by a factor of w/(w + A), while large union-
ids could reduce surface reactivity by w/(w + 30A). The significance of
reactant injection by macrobenthos has not yet been empirically evaluated
in freshwater environments. Indeed, basic data concerning the chemical
nature of materials excreted by freshwater macrobenthos are almost
wholly lacking. There is some evidence that suggests that the excretion
of ammonia by tubificids may aid in controlling interstitial water pH
(Fisher and Matisoff, 1981). Chatarpaul et al. (1980) suggest that the bodies
of at least some benthos (tubificid oligochaetes) may actually be sites of
certain diagenetic reactions (nitrification and denitrification). The follow-
ing ranking, in order of relative importance, of the above mechanisms is
suggested:
1. Water pumping
2. Particle advection and diffusion
3. Fabric alteration
? Reactant injection
? Reaction site
The above mechanisms are, of course, linked to tropic group/life habit
type. Water pumping through sediments is carried on by sedentary in-
faunal suspension feeders and may be continuous or episodic. Sediment
particle advection is carried on by sedentary infaunal subsurface deposit
feeders. Mobile epifaunal and infaunal deposit feeders will cause particle
diffusion. Nearly all tropic group/life habit types can alter sediment fabric
and inject reactants. Furthermore, the intensity of all of the mechanisms
is likely to be related to organism size, population density, temperature,
availability of dissolved oxygen, and other biotic and abiotic variables.
From the foregoing discussion, it is clear that the actions of macro-
benthos affect the chemical diagenesis of freshwater sediments. Current
knowledge of the mechanisms whereby these effects are generated is
sketchy. In particular, the significance of various transport-modifying
mechanisms should be investigated. For example, detailed time series
experiments should be conducted to examine the effects of water pumping
by chironomids on the migration of solutes through sediments. Ideally,
radiotracer experiments would be used to determine the behavior of var-
ious materials in sediments populated by chironomids. Particle and solute
Effects of Macrobenthos on Freshwater Sediment Diagenesis 211

transport measurements for organisms other than tubificids and chiron-

omids should be made. In particular, the effects of unionid bivalves and
burrowing mayflies on particle and solute transport should be investi-
gated. The significance of reactant injection to sediment diagenesis needs
investigation. What materials are excreted by various macrobenthos? How
much material is injected? How rapidly does this material decay? What
are the decomposition products? How does this process affect overall
chemical mass transfer in surficial sediments? Do macrobethos provide
reaction sites (as suggested by Chatarpaul et al., 1980)? What is the sig-
nificance of this phenomenon? Finally, the problem of dimensionality
needs to be considered in detail. For relatively large organisms, which
occur at lower population densities (e.g., chironomids, unionid clams),
two-dimensional models are probably necessary. Two-dimensional models
require two-dimensional data for development and verification. New and
far more advanced data-gathering techniques (e.g., multiple microelec-
trode probes, two-dimensional gamma scanning techniques) need to be
developed.
In addition to the above work, the relationships between particle
movement, water pumping, sediment fabric alteration, and reactant in-
jection by macrobenthos and various biotic and abiotic variables need
exploration. What are the effects of such factors as seasonal temperature
variations, episodic anoxic interludes, organism population density, inter-
and intraspecific interactions, and predation on interactions between ben-
thos and their substratum?
The work suggested here, because of the detail required, must largely
be carried out in the laboratory. How can the results of these suggested
investigations be extended to the field? What field experiments might be
conducted to answer the questions posed? Investigation of the effects of
animal-sediment interactions on the chemical diagenesis of freshwater
sediments should prove a fertile and productive area for future research.

ACKNOWLEDGMENTS. Funding from NOAA and EPA supported some of the

work described in this chapter.

References
Aller, R. C., 1977, The influence of deposit feeding benthos on chemical diagenesis of marine
sediments, Ph.D thesis, Yale University, New Haven, Connecticut.
Aller, R. c., 1978, The effects of animal-sediment interactions on geochemical processes
near the sediment-water interface, in: Estuarine Interactions (M. 1. Wiley, ed.J, pp.
157-172, Academic Press, New York.
Aller, R. c., and Cochran, J. K., 1976, 234-Th/238-U disequilibrium in near shore sediments:
Particle reworking and diagenetic time scales, Earth Planet. Sci. Lett. 20:37-50.
212 Chapter 4

Aller, R. c., and Dodge, R. E., 1974, Animal-sediment relations in a tropical lagoon-
Discovery Bay, Jamaica, J. Mar. Res. 32:209-232.
Aller, R. c., and Yingst, J. Y., 1978, Biogeochemistry of tube-dwellings: A study of the
sedentary polychaete Amphitrite ornata (Leidy), J. Mar. Res. 36:201-254.
Alsterberg, C., 1922, Die respiratorischen Mechanismen der Tubificiden, Lunds Univ. Arsskr.
18:1-176.
Alsterberg, G., 1925, Die Nahrungszirkulation einiger Binnensetypen, Archiv. Hydrobioi.
15:291-338.
Amiard-Triquet, c., 1975, Etude experimentale de la contamination par Ie cerium 144 et Ie
fer 59 d'un sediment a Arenicola marina L. (Annelide Polychete), Cah. Bioi. Mar.
15:483-494.
Andersen, J. M., 1977, Importance of the denitrification process for the rate of degradation
of organic matter in lake sediments, in: Interactions between Sediments and Fresh Water
(H. L. Colterman, ed.), pp. 357-362, Junk, The Hague.
Appleby, A. C., and Brinkhurst, R. 0., 1970, Defecation rate of three tubificid oligochaetes
found in the sediment of Toronto Harbour, Ontario, J. Fish. Res. Board Can. 27:1971-
1982.
Baas Becking, 1. C. M., Kaplan, I. R., and Moore, D., 1960, Limits of the natural environment
in terms of pH and oxidation-reduction potentials, J. Ceoi. 68:243-284.
Ben-Yaakov, S., 1973, pH buffering of pore water of recent anoxic marine sediments, Limnoi.
Oceanogr. 18:86-94.
Berner, R. A., 1963, Electrode studies of hydrogen sulfide in marine sediments, Ceochim.
Gosmochim. Acta 27:563-575.
Berner, R. A., 1964a, An idealized model of dissolved sulfate distribution in recent sedi-
ments, Geochim. Cosmochim. Acta 28:1497-1503.
Berner, R. A., 1964b, Distribution and diagenesis of sulfur in some sediments from the Gulf
of California, Mar. Ceoi. 1:117-140.
Berner, R. A., 1970, Sedimentary pyrite formation, Am. J. Sci. 268:1-23.
Berner, R. A., 1972, Sulfate reduction, pyrite formation, and the oceanic sulfur budget, in:
Nobel Symposium 20: The Changing Chemistry of the Oceans (D. Dyrssen and D. Jagner,
eds.), pp. 347-361, Almquist and Wiksell, Stockholm.
Berner, R. A., 1974, Kinetic models for the early diagenesis of nitrogen, sulfur, phosphorus,
and silicon in anoxic marine sediments, in: The Sea, Volume 5: Marine Chemistry (E.
D. Goldberg, ed.), pp. 427-450, John Wiley and Sons, New York.
Berner, R. A., 1976, The benthic boundary layer from the viewpoint of a geochemist, in: The
Benthic Boundary Layer (1. N. McCave, ed.J, pp. 33-55, Plenum Press, New York.
Berner, R. A., 1980, Early Diagenesis: A Theoretical Approach, Princeton University Press,
Princeton, N.J.
Brinkhurst, R. 0., 1974, The Benthos of Lakes, SI. Martin's Press, New York.
Burns, N. M. and Ross, C., 1972, Oxygen-nutrient relationships in the Central Basin of Lake
Erie, in: Project Hypo (N. M. Ross and C. Ross, eds.J, pp. 85-119, Canadian Center for
Inland Waters Paper 6, Canadian Center for Inland Waters, Hamilton, Ontario.
Carmouze, J. P., Golterman, H. 1., and Pedro, c., 1977, The neoformation of sediments in
Lake Chad; their influence on salinity control, in: Interactions between Sediments and
Fresh Water (H. 1. Golterman, ed.), pp. 33-39, Junk, The Hague.
Chamberlain, C. K., 1975, Recent lebensspuren in nonmarine aquatic environments, in: The
Study of Trace Fossils (R. W. Frey, ed.J, pp. 431-458, Springer-Verlag, New York.
Chapman, T. W., 1967, The transport properties of concentrated electrolytic solutions, Ph.D.
thesis, University of California, Berkeley.
Chatarpaul, 1., Robinson,). B., and Kaushik, N. K., 1979, Role of tubificid worms in nitrogen
transformations in stream sediment, J. Fish. Res. Board Can. 36:673-678.
Effects of Macrobenthos on Freshwater Sediment Diagenesis 213

Chatarpaul, L., Robinson, J. B., and Kaushik, N. K., 1980, Effects of tubificid worms on
denitrification and nitrification in stream sediment, Can. J. Fish. Aquat. Sci. 37:656-663.
Coker, R E., Shira, A. R, Clark, H. W., and Howard, A. D., 1921, Natural history and
propagation of fresh-water mussels, Bull. U.S. Bur. Fish. 37:77-181.
Cummins, K. W., 1973, Trophic relationships of aquatic insects, Annu. Rev. Entomol.
18:183-206.
D'Anglejan, B. F., 1967, Origin of marine phosphorites off Baja California, Mexico, Mar.
Ceol. 5:15-44.
Davies, S. N., and DeWiest, R C. M., 1966, Hydrogeology, John Wiley and Sons, New York.
Davis, R B., 1974a, Stratigraphic effects of tubificids in profundallake sediments, Limnol.
Oceanogr. 19:466-488.
Davis, R B., 1974b, Tubificids alter profiles of redox potential and pH in profundal lake
sediments, Limnol. Oceanogr. 19:342-346.
Davis, R B., Thurlow, D. L., and Brewster, R E., 1975, Effects of burrowing tubificids on
the exchange of phosphorus between lake sediment and overlying water, Verh. Int.
Verein. Limnol. 19:382-394.
Dell, C. I., 1971, Late quaternary sedimentation in Lake Superior, Ph.D. thesis, University
of Michigan, Ann Arbor.
Dell, C. I., 1973, Vinianite: An authigenic phosphate mineral in Great Lake sediments, in:
Proceedings of the 16th Conference on Great Lakes Research, pp. 1027-1028, Inter-
national Association for Great Lakes Research, Ann Arbor, Michigan.
Doyle, R W. S., 1967, Eh and thermodynamic equilibrium in environments containing
dissolved ferrous iron, Ph.D. thesis, Yale University, New Haven, Connecticut.
Dutka, B. J., Bell, J. B., and Liu, D. L. S., 1974, Microbiological examination of offshore Lake
Erie sediments, J. Fish. Res. Board Can. 31:299-308.
Edwards, R W., 1958, The effect of larvae of Chironomus riparius Meigen on the redox
potentials of settled activated sludge, Ann. Appl. BioI. 46:457-464.
Edwards, R W., and Rolley, H. L. J., 1965, Oxygen consumption of river muds, J. Ecol.
53:1-19.
Emerson, S., and Widmer, G., 1978, Early diagenesis in anaerobic lake sediments. II. Ther-
modynamic and kinetic factors controlling the formation of iron phosphate, Geochim.
Cosmochim. Acta 42:1307-1316.
Fisher, J. B., and Matisoff, G., 1981, High resolution vertical profiles of pH in recent sedi-
ments, Hybrobiologia 79:277-284.
Fisher, J. B., and Tevesz, M. J. S., 1975, Distribution and population density of Elliptio
complanata (Mollusca) in Lake Pocotopaug, Connecticut, Veliger 18:332-338.
Fisher, J. B., Lick, W., McCall, P. L., and Robbins, J. A., 1980, Vertical mixing of lake
sediments by tubificid oligochaetes, J. Geophys. Res. 85:3997-4006.
Gallepp, G. W., 1979, Chironomid influence on phosphorous release in sediment-water
microcosms, Ecology 60:547-556.
Gallepp, G. W., Kitchell, J. F., and Bartell, S. M., 1978, Phosphorus release from lake sed-
iments as affected by chironomids, Verh. Int. Verein. Limnol. 20:458-465.
Ganapati, S. V., 1949, The role of the bloodworm, Chironomus plurnosus, in accounting for
the presence of phosphorus and excessive free ammonia in the filtrates from the slow
sand filters of the Madras water works, Zoo 1. Soc. India J. 6:41-43.
Gardner, R L., 1973, Chemical models for sulfate reduction in closed anaerobic marine
environments, Geochirn. Cosrnochirn. Acta 37:53-68.
Glasby, G. P., 1973, Interstitial waters in marine and lacustrine sediments: A review, J. R.
Soc. N. Z. 3:43-59.
Goldberg, E. D., and Koide, M., 1962, Geochronological studies of deep sea sediments by
the ionium/thorium method, Geochirn. Cosrnochirn. Acta 26:417-450.
214 Chapter 4

Goldhaber, M, B., 1974, Equilibrium and dynamic aspects of the marine geochemistry of
sulfur, Ph.D. thesis, University of California, Los Angeles.
Goldhaber, M. B., and Kaplan, I. R., 1975a, Apparent dissociation constants of hydrogen
sulfide in chloride solutions, Mar. Chern. 3:83-104.
Goldhaber, M. B., and Kaplan, I. R., 1975b, Controls and consequences of sulfate reduction
rates in recent marine sediments, Soil Sci. 119:42-55.
Gorham, E., 1964, Molybdenum, manganese, and iron in lake muds, Verh. Int. Verein.
Limnol. 15:330-332.
Graneli, W., 1979, The influence of Chironomus plumosus larvae on the exchange of dis-
solved substances between sediment and water, Hydrobiologia 66:149-159.
Graneli, W., 1982, The influence of Chironomus plumosus on oxygen uptake of sediment,
Arch. Hydrobiol. (in press).
Guinasso, N. L., and Schink, D. R., 1975, Quantitative estimates of biological mixing rates
in abyssal sediments, 1. Geophys. Res. 80:3032-304:3.
Hargrave, B. T., 1972, Oxidation-reduction potentials, oxygen concentration and oxygen
uptake of profundal sediments in a eutrophic lake, Oikos 23:167-177.
Hargrave, B. T., 1975, Stability in structure and function of the mud-water interface, Verh.
Int. Verein. Limnol. 19:1073-1079.
Harrison, A. G., and Thode, 1958, The kinetic isotope effect in the chemical reduction of
sulfate, Tr. Faraday Soc. 53:1648-1651.
Hayes, F. R., 1964, The mud-water interface, Oceanogr. Mar. BioI. Annu. Rev. 2:121-145.
Hutchinson, G. E., 1957, A Treatise on Limnology, John Wiley and Sons, New York.
Iovino, A., and Bradley, W., 1969, The role of larval chironomidae in the production of
lacustrine copropel in Mud Lake, Marion County, Florida, Limnol. Oceanogr. 14:898-900.
Ivlev, V. S., 1939, Transformation of energy by aquatic animals. Coefficient of energy con-
sumption by Tubifex tubifex (OligochaetaJ, Int. Rev. Ges. Hydrobiol. Hydrogr. 38:449-458.
Jones, B. F., and Bowser, C. J., 1978, The mineralogy and related chemistry of lake sediments,
in: Lakes: Chemistry, Geology, Physics (A. Lerman, ed.J, pp. 179-235, Springer-Verlag,
New York.
Jernelov, A., 1970, Release of methyl mercury from sediments with layers containing in-
organic mercury at different levels, Limno!. Oceanogr. 15:958-960.
Kaplan, I. R., and Rittenberg, S. C., 1964, Microbiological fractionation of sulfur isotopes,
1. Gen. Microbiol. 34:195-212.
Kikuchi, E., and Kurihara, Y., 1977, In vitro studies on the effects of tubificids on the
biological, chemical, and physical characteristics of submerged ricefield soil and over-
lying water, Oikos 29:348-356.
Kirchner, W. B., 1975, The effect of oxidized material on the vertical distribution of fresh-
water benthic fauna, Freshwater Biol. 5:423-429.
Krezoski. J. R., 1981, The influence of zoobenthos on fjne-grained particle reworking and
benthic solute transport in Great Lakes sediments. Ph.D. thesis, University of Michigan,
Ann Arbor.
Krezoski, J. R., Mozley, S. C., and Robbins, J. A., 1978, Influence of benthic macroinverte-
brates on mixing of profundal sediments in southeastern Lake Huron, Limno!. Oceanogr.
23:1011-1016.
Kuznetsov, S. I., 1970, The Microflora of Lakes and its Geochemical Activity, University of
Texas Press, Austin.
Lafon, G. M., and Mackenzie, F. T., 1966, Early evolution of the oceans-A weathering
model, in: Studies in Paleo-Oceanography (W. W. Hay, ed.J, Soc. Econ. Paleonto!.
Mineral. Spec. Publ. 20:205-218.
Lee, G. F., 1970, Factors affecting the transfer of materials between water and sediments,
Effects of Macrobenthos on Freshwater Sediment Diagenesis 215

University of Wisconsin Water Resources Center, Eutrophication Information Program,

Literature Review 1, University of Wisconsin, Madison.
Li, Y.-H., and Gregory, S., 1974, Diffusion of ions in sea water and in deep sea sediments,
Geochim. Cosmochim Acta 38:703-714.
Lundbeck, J., 1926, Die Bodentierwelt Norddeutschen Seen, Arch. Hydrobiol. Suppl 7.
Lyman, F. E., 1943, Swimming and burrowing activities of mayfly nymphs of the genus
Hexigenia, Ann. Entomol. Soc. Am. 26:250-256.
McCaffrey, R J., Meyers, A. c., Davey, E., Morrison, G., Bender, M., Luedtke, N., Cullen, D.,
Froelich, P., and Klinkhammer, G., 1980, The relation between pore water chemistry
and benthic fluxes of nutrients and manganese in Narragansett Bay, Rhode Island,
Limnol. Oceanogr. 25:31-44.
McCall, P. L., 1979, Effects of deposit feeding oligochaetes on particle size and settling
velocity of Lake Erie sediments, J. Sediment. Petrol. 49:813-818.
McCall, P. L., and Fisher, J. B., 1980, Effects of tubificid oligochaetes on physical and
chemical properties of Lake Erie sediments, in: Aquatic Oligochaete Biology (R O.
Brinkhurst and D. G. Cook, eds.), pp. 253-318, Plenum Press, New York.
McCall, P. L., Tevesz, M. J. S. , and Schwelgien, S. F., 1979, Sediment mixing by Lampsilis
radiata siliquoidea (Mollusca) from western Lake Erie, J. Great Lakes Res. 5:105-111.
McLachlan, A. J., and McLachlan, S. M., 1976, Development of the mud habitat during the
filling of two new lakes, Freshwater BioI. 6:59-67.
Manheim, F. T., 1970, The diffusion of ions in unconsolidated sediments, Earth Planet. Sci.
Lett. 9:307-309.
Manheim, F. T., and Sayles, F. L., 1974, Composition and origin of interstitial waters of
marine sediments, in: The Sea, Volume 5: Marine Chemistry (E. D. Goldberg, ed .), pp.
527-568, John Wiley and Sons, New York.
Marzolf, G. R, 1965, Substrate relations of the burrowing amphipod Pontoporeia affinis in
Lake Michigan, Ecology 46:579-592.
Milbrink, G., 1969 , Microgradient at the mud-water interface, Rep. Inst. Freshwater Res.
Drottningholm 49:129-148.
Monakov, A. K., 1972, Review of studies on feeding of aquatic invertebrates conducted at
the Institute of Biology of Inland Waters, Academy of Sciences, USSR, J. Fish . Res.
Board Can. 29:363-383.
Mortimer, C. H., 1941, The exchange of dissolved substances between mud and water in
lakes, J. Ecol. 29:280-329.
Mortimer, C. H., 1942, The exchange of dissolved substances between mud and water in
lakes, J. Ecol. 30:147-201.
Mortimer, C. H., 1971, Chemical exchanges between sediments and water in the Great
Lakes-Speculations on probable regulatory mechanisms, Limnol. Oceanogr. 16:387-404.
Miiller, G., and Forstner, U., 1973, Recent iron ore formation in Lake Malawi, Africa, Miner.
Deposita 8:278-290.
Neame, P. A., 1975, Benthic oxygen and phosphorus dynamics in Castle Lake, California,
Ph.D. thesis, University of California, Davis.
Neame, P. A., 1977, Phosphorus flux across the sediment-water interface, in: Interactions
between Sediments and Fresh Water (H. L. Golterman, ed.), pp. 307-312, Junk, The
Hague.
Newell, R c., 1970, Biology of Intertidal Animals, Elsevier, New York.
Norvell, W. A., 1974, Insolubilization of inorganic phosphate by anoxic lake sediment, Soil
Sci. Soc. Am . Proc. 38:441-445.
Nozaki, Y., Cochran, J. K., Turekian, K. K., and Keller, G., 1977, Radiocarbon and (210) Pb
distribution in submersible taken deep-sea cores from project fAMOUS, Earth Planet. Sci.
Lett. 34:167-173.
216 Chapter 4

Nriagu, J, 0" 1968, Sulfur metabolism and sedimentary environment: Lake Mendota, Wis-
consin, Limnol. Oceanogr. 13:430-439.
Nriagu, J. 0., 1975, Sulphur isotopic variations in relation to sulphur pollution of Lake Erie,
in: Isotope Ratios as Pollutant Source and Behavior Indicators, pp. 77-93, IAEA-SM-
191/28, International Atomic Energy Agency, Vienna.
Nriagu, J. 0., 1976, Phosphate-clay mineral relations in soils and sediments, Can. J. Earth
Sci. 13:717-736.
Nriagu, J. 0., and Dell, C. I., 1974, Diagenetic formation of iron phosphates in recent lake
sediments, Am. Mineral. 59:934-946.
Otsuki, A., and Wetzel, R G., 1972, Coprecipitation of phosphate with carbonates in a marl
lake, Limnol. Oceanogr. 17:763-767.
Parry, G., 1960, Excretion, in: The Physiology of Crustacea (T. H. Waterman, ed.J, pp.
341-363, Academic Press, New York.
Poddubnaya, T. L., 1961, Concerning the feeding of the high density species of tubificids
in Rybinsk Reservoir, Tr. Inst. Biol. Vodokhran. Akad. Nauk SSSR 4:219-231.
Poddubnaya, T. L., and Sorokin, Yu. I., 1961, The depth of the layer of optimal feeding of
tubificids in connection with their movements in the sediment, Izv. Inst. Biol. Vo-
dokhran. AN SSSR 10:14-17.
Postgate, J. R., 1951, The reduction of sulphur compounds by Desulphovibrio desulphuri-
cans, J. Gen. Microbiol. 5:725-738.
Potts, W. T. W., 1954a, The energetics of osmotic regulation in brackish and freshwater
animals, J. Exp. Biol. 31:618-630.
Potts, W. T. W., 1954b, The rate of urine production of Anodonta cygnea, J. Exp. Bio1.
31:614-617.
Presley, B. J., and Kaplan, I. R., 1968, Changes in dissolved sulfate, calcium, and carbonate
from interstitial water of near-shore sediments, Geochim. Cosmochim. Acta 32:1037-1048.
Ramm, A. E., and Bella, D. A., 1974, Sulfide production in anaerobic microcosms, Limnol.
Oceanogr. 19:110-118. '
Ravera, 0., 1955, Amount of mud displaced by some freshwater oligochaeta in relation to
depth, Mem. 1st. Ital. Idrobiol. 8(suppl.):247-264.
Rees, C. E., 1973, A steady state model for sulfur isotope fractionation in bacterial reduction
process, Geochim. Cosmochim. Acta 37:1141.
Reid, R. c., Prausnitz, J. M., and Sherwood, T. K., 1977, The Properties of Gases and Liquids,
McGraw-Hill, New York.
Rhoads, D. C., 1970, Mass properties, stability, and ecology of marine muds related to
burrowing activity, in: Trace Fossils (T. P. Crimes and J. c. Harper, eds.J, pp. 391-406,
Geol. J. Spec. Issue 3.
Rhoads, D. C., 1974, Organism-sediment relations on the muddy sea floor, Oceanogr. Mar.
Bio1. Annu. Rev. 12:263-300.
Rhoads, D. C., and Young, D. K., 1970, The influence of deposit feeding organisms on
sediment stability and community trophic structure, J. Mar. Res. 28:150-178.
Riley, j. P., 1971, The major and minor elements in seawater, in: Introduction to Marine
Chemistry (J. P. Riley and R. Chester, eds.J, pp. 60-101, Academic Press, New York.
Robbins, J. A., 1978, Geochemical and geographical applications of radioactive lead, in:
Biogeochemistry of Lead (J. O. Nriagu, ed.J, pp. 285-337, Elsevier, Amsterdam.
Robbins, J. A., Krezoski, J. R, and Mozley, S. C., 1977, Radioactivity in sediments of the
Great Lakes: Post-depositional redistribution by deposit feeding organisms, Earth Planet.
Sci. Lett. 36:325-333.
Robbins, J. A., McCall, P. 1., Fisher, J. B., and Krezoski, J. R., 1979, Effects of deposit feeders
on migration of cesium-137 in lake sediments, Earth Planet. Sci. Lett. 42:277-287.
Robinson, R. A., and Stokes, R H., 1959, Electrolyte Solutions, Butterworths, London.
Effects of Macrobenthos on Freshwater Sediment Diagenesis 217

Savilov, A. I., 1957, Biolgical aspects of the bottom fauna grouping of the North Okhotsk
Sea, Tr. Inst. Okeano1. Mar. Bio1. 20:67-136.
Schumacher, A., 1963, Quantitative Aspekte der Beziehung zwischen Starke der Tubifici-
denbesiedlung und Schichtdicke der Oxydationzone in den Siisswasserwatten der Un-
terelbe, Arch. Fischwiss. 14:48-51.
Shapiro, J., Edmondson, W. T., and Allison, D. E., 1971, Changes in the chemical composition
of sediments of Lake Washington, Limno1. Oceanogr. 16:437-452.
Staddon, B. W., 1969, Water balance in Ilyocoris, Naucoridae, J. Exp. Bio1. 51:643-665.
Stamm, H. H., and Kohlschutter, H. W., 1965, Die Sorption von Phosphationene an Eisen
(IIIJ-Hydroxide, J. Inorg. Nuc1. Chern. 27:2103-2108.
Stobbart, R. H., and Shaw, J., 1974 , Salt and water balance: Excretion, in: The Physiology
of Insects (M. Rockstein, ed.), pp. 361-446, Academic Press, New York.
Stumm, W., and Morgan, J. J., 1970, Aquatic Chemistry, John Wiley and Sons, New York.
Swain, F. M., 1965, Geochemistry of some quaternary lake sediments of North America, in:
The Quaternary of the United States (H. E. Wright and D. G. Frey, eds.), pp. 765-781,
Princeton University Press, Princeton, New Jersey.
Tessenow, U. , 1964, Experimentaluntersuchungen zur Kieselsaureriickfiihrung aus dem
Schlamm der See durch Chironomidenlarven (Plumosus-Gruppe), Arch. Hydrobio1.
60:497 -504.
Tevesz, M. J. S., and McCall, P. L., 1979, Evolution of substratum preference in bivalves
(Mollusca), J. Paleonto1. 53:112-120.
Tevesz, M. J. S., Soster, F., and McCall, P. L., 1980, The effects of size selective feeding by
oligochaetes on the physical properties of river sediments, J. Sediment. Petro1.
50:561-568.
Turpaeva, E. P., 1957, Food interrelationships of dominant species in marine benthic bio-
coenoses, Tr. lnst. Okeano1. Mar. Bio1. 20:137-148.
Upchurch, S. B., 1976, Chemical characteristics of the Great Lakes, in: Great Lakes Basin
Framework Study, Appendix 4: Limnology of Lakes and Embayments, pp. 151-238,
Public Information Offices, Great Lakes Basin Commission, Ann Arbor, Michigan.
Vallentyne, J. R., 1961, On the rate of formation of black spheres in recent sediments, Verh.
lnt. Verein. Limn()1. 14:291-295.
Vallentyne, J. R, 1963, Isolation of pyrite spherules from recent sediments, Limno1. Ocean-
ogr. 8:16-30.
Vogel, S., and Bretz, W. L., 1972, Interfacial organisms: Passive ventilation in the velocity
gradients near surfaces, Science 175:210.
Wachs, B., 1967, Die Oligochaeten-Fauna der Fleissgewasser unter besonderer Beriicksichtigung
der Beziehungen zwischen der Tubificiden-Besiedlung und dem Substrat, Arch. Hy-
drobio1. 63:310-386.
Walker, K. R, and Bambach, R K., 1974, Feeding by benthic invertebrates: Classification
and terminology for paleoecological analysis, Lethaia 7:67-78.
Walshe, B. M., 1947, Feeding mechanisms of Chironomous larvae, Nature 160:474.
Weiler, R R, 1973, The interstitial water composition in the sediments of the Great Lakes.
I. Western Lake Ontario, Limnol. Oceanogr. 18:918-931.
Weissenbach, H., 1974, Untersuchungen zum Phophorhaushalt eines Hochgebirgsees (Vor-
der Finstertaler See, Kiihtai, Tirol) unter besonderer Beriicksichtigung der Sedimente,
Dissertation, Leopold Franzens-Universitat, Innsbruck.
Wetzel, R. G., 1975, Limnology, W. B. Saunders, Philadelphia.
Williams, J. D. H., Jaquet, J. M., and Thomas, R L., 1976a, Forms of phosphorus in the
surficial sediments of Lake Erie, J. Fish. Res. Board Can. 33:413-429.
Williams, J. D. H., Murphy, T. P., and Mayer, T., 1976b, Rates of accumulation of phosphorus
forms in Lake Erie sediments, J. Fish. Res. Board Can. 33:430-439.
218 Chapter 4

Winfrey, M. R, and Zeikus, J. G., 1979, Anaerobic metabolism of immediate methane pre-
cursors in Lake Mendota, Appl. Environ. Microbial. 37:244-253.
Wood, 1. W., 1975, Role of oligochaetes in the circulation of water and solutes across the
mud-water interface, Verh. Int. Verein. Limnol. 19:1530-1538.
Zinder, S. H., and Brock, T. D., 1978, Methane, carbon dioxide, and hydrogen sulfide pro-
duction from the terminal methiol group of methionine by anaerobic lake sediments,
Appl. Environ. Microbial. 35:344-352.
ZoBel!, C. E., 1946, Studies on redox potential of marine sediments, Bull. Am. Assoc. Petrol.
Geol. 30:477-513.
 III
Ancient Environments
Chapter 5
Geological Significance of Marine
Biogenic Sedimentary Structures
CHARLES W. BYERS

1. Introduction ................................................................................................................. 221

2. Traces and Sedimentology .......................................................................................... 222
2.1. Bioturbation in Recent Muds .............................................................................. 223
2.2. Rate and Continuity of Sedimentation ............................................................... 225
2.3. Substrate Consistency .............................................. .. ..... .. ... ................................ 227
2.4. Diagenesis ............................. .. ... ........................................................................... 230
3. Paleobathymetry .......... .. .................... .. .. .. ... ..... .. ... .................. .. .............. .. ............ .. ... .. 233
3.1. Introduction ............................................................. .. ........ .. ............ .................... 233
3.2. Problems with Seilacher's Model....................................................................... 234
3.3. Evidence from the Oceans .................... .. ............................................................ 238
3.4. The Graphoglyptid Controversy......................................................................... 241
4. Precambrian Traces ..................................................... .. ........... .. ............... .. ........... .. ... 243
4.1. The Precambrian-Cambrian Boundary.............................................................. 243
4.2. The Oldest Trace Fossil .......................................................................... .. .......... 248
References ...................................................... .. ........ .. ........ .. ........ .. ........ .. ........ .. .......... 252

1. Introduction
The "geological significance" of traces encompasses a huge range of top-
ics; to survey the field even briefly would require an entire book and
probably multiple authors. Fortunately a major review volume was pub-
lished within the last decade (Frey, 1975), and a diversity of topics were
addressed. In addition, the revised Volume W of the Treatise on Inver-
tebrate Paleontology (Hantzschel, 1975), which covers the systematics of
trace fossils, also appeared recently. These two books have served to
provide a solid base for research in trace fossils in terms of current sys-
tematics and concepts of classification, how traces are made and pre-

CHARLES W. BYERS • Department of Geology and Geophysics, University of Wisconsin,

Madison, Wisconsin 53706.
221
222 Chapter 5

served, and the relation of traces to sedimentology, biostratigraphy, pa-

leontology, and paleoecology.
My intent in this chapter is twofold: I will review some of the recent
literature on trace fossils, and I will focus on what seem to me to be the
most exciting and significant areas of trace fossil research. Gould (1980)
recently reviewed the status of paleontology in general and concluded
that the science has been overly concerned with performing as a "service
industry" for stratigraphic ordering and environmental reconstruction.
This has been especially true in the realm of trace fossils. The paramount
interest in traces has been for their utility as bathymetric indicators, and
the inductivist methodology Gould deplored has been pervasive. Nearly
all trace fossil studies have addressed one of these questions:
1. What traces are found in a particular depositional environment
(or in a particular inferred environment, as interpreted from the
rock record)?
2. What traces are found in rocks of a particular age?
3. What was the animal that made a particular trace, and what was
it doing?
Each of these questions focuses on the particular. I find Gould's thesis
quite applicable to ichnology-students of traces have hardly begun the
task of formulating and testing hypotheses. Although data have been gath-
ered and a few generalities have been formulated, the predictive and
explanatory aspects of science have been lacking. Perhaps this is simply
the result of the youthfulness of the discipline; not many years ago Sei-
lac her needed to assert "It is generally agreed that biogenic sedimentary
structures are true fossils and that their study ... is part of paleontology"
(Seilacher, 1964). I will discuss below some encouraging signs of new
approaches to ichnology.

2. Traces and Sedimentology

Trace fossils are a special kind of sedimentary structure. They are
formed, as a rule, within the depositional environment and soon after the
physical deposition of the sediment. Therefore they typically reflect con-
ditions in that environment that are of interest to sedimentologists: water
energy, rate of deposition, uniformity of deposition, consistency of sub-
strata. Traces may also be important in the interpretation of the postde-
positional history of a sedimentary rock. Because burrows usually have
fill that is chemically or physically different from the surrounding matrix,
or have no fill at all, they may react diagenetic ally in distinctive ways.
In the following brief survey of the interrelationships among physical
Marine Biogenic Sedimentary Structures 223

sedimentology, diagenesis, and trace fossils, I hope to indicate the inse-

parability of physical and biological processes.

2.1. Bioturbation in Recent Muds

Lamination in neritic sediments is rare because of the activities of

infauna. Although the effects of burrowing infauna on marine sediments
have been discussed theoretically for many years, only since the 1950s
have they been widely recognized in the field.
It was known in the last century that deposit-feeding infauna rework
marine muds (see Twenhofel, 1939). Twenhofel concluded that sediments
will be reworked by infauna unless bottom conditions are hostile to life,
and Dapples (1942) predicted that burrowing infauna such as annelids
and holothurians would strongly alter marine sediments, the greatest
change being found in areas of optimum living conditions. In 1950 Kuenen
discussed laminated muds and concluded that they would be rare at shelf
depth owing to both physical and biological reworking. He cited the clear-
cut laminae of the Black Sea sediments as being very uncommon in normal
shelf-depth seas. Pettijohn (1957) continued the theoretical discussion of
lamination by pointing out that laminae could only be preserved below
wave base in quiet water. He also postulated the role played by burrowing
infauna: "if such borings are numerous the original stratification planes
are largely destroyed and only vestiges remain. Complete eradication of
bedding is possible and has probably occurred in some sediments" (Pet-
tijohn, 1957, p. 193). This perspective was stratigraphic, an acceptance
of bedding as the normal situation, with a marginal note that biological
alteration of sediments was conceivable.
Moore and Scruton (1957) were the first to show that burrowing
deposit feeders are a major factor in the final product of sedimentation
of the Mississippi Delta. They found a decrease in the preservation of
original sedimentary structures moving away from the Delta. Nearshore,
where the sedimentation rate and physical reworking rates by waves and
currents are high, the original structures in the sediment are preserved.
Away from shore the physical sedimentary processes are slower owing
to increasing distance from the sediment source and increasing depth.
Here burrowing animals disrupt the regular bedding into irregular layers
and lenses. Further offshore, biogenic activity produces sediments of
mottled appearance, or even a homogeneous, structureless bioturbate.
Moore and Scruton demonstrated that only in the very shallow nearshore
region, where physical processes outstrip bioturbation, are regular layers
preserved. At depths of 40 m or more the sediments are homogeneous
muds, completely reworked by the infauna. Hence Moore and Scruton's
224 Chapter 5

conclusion: "the importance of animals in forming the final fabric of

shallow-water marine sediments cannot be overstated. Were it not for the
burrowing and crawling organisms, the appearance of many of our marine
sediments would be entirely different."
Van Straaten's (1959) observations of the sediments of the Rhone
Delta confirmed Moore and Scruton's conclusions. With increasing depth
the effects of burrowing infauna overwhelm the processes of physical
sedimentation. Van Straaten found that regular layering nearshore, a prod-
uct of seasonal changes in discharge, was progressively destroyed away
from the delta; below 80 m the muds were homogenized by infauna. Van
Straaten thought such homogeneous muds to be the rule for outer shelf
conditions worldwide, except for inclusions of relict Pleistocene sands
and ice-rafted material on polar shelves.
Other workers noted the same gradient in sedimentary structures
offshore from deltas. Allen (1965) described the sediments of the Niger
Delta, where the platform (less than 20 m depth) sediments are wave- and
current-winnowed and -layered, whereas the offshore prodelta and shelf
clays are completely reworked by infaunal benthos.
Deltas are not unique in showing a decrease in physical structures
away from shore. Moore and Scruton (1957J and Van Straaten (1959)
described similar sedimentation patterns offshore from barrier bar se-
quences. Nearshore layering is a product of wave and tidal current re-
working, rather than a rapid and fluctuating sediment supply. Reworking
weakens with increasing depth, and bioturbation destroys the bedding.
In 1967 Reineck discussed the patterns of sedimentary structures in the
North Sea. He found decreasing physical influence and increasing bio-
turbation correlated with deeper water. Wave effects diminish greatly at
depths of 20 m or more. Laminated sandy and silty deposits are confined
to these very shallow depths, where the sediment can be wave-winnowed;
bioturbation replaces physical structures completely in the seaward di-
rection. By 1970, perspective had shifted: "open shelf sediments, partic-
ularly of the transitional and mud belts, are subject to extensive reworking
by bottom dwelling and feeding animals, even to the extent that all traces
of primary laminations are destroyed" (Allen, 1970),
Modern normal-marine sediments are dominated by biological pro-
cesses except in areas of very rapid sedimentation influx or continual
current activity. Hence the degree to which sediments have been biotur-
bated is an index of the physical energy of the environment. Environ-
mental zonation of modern shelves based on the amount of bioturbation
provides models for interpretation of the sedimentary record.
The absence of bioturbation in quiet-water, slow-deposition environ-
ments implies chemical barriers to benthic habitation. Very high salinity
may restrict metazoans, but low oxygenation of the muds and/or bottom
Marine Biogenic Sedimentary Structures 225

waters is a more common cause. The relationships among depth, oxy-

genation, metazoan life, and trace fossils have been revie~ed by Rhoads
and Morse (1971) and Byers (1977a).

2.2. Rate and Continuity of Sedimentation

In addition to indicating the overall energy of the depositional en-

vironment, trace fossils can provide information on the timing of sedi-
mentary processes. The use of traces in this fashion has been pioneered
by Goldring (1964) and especially by Howard (1975, 1978). Howard has
identified patterns of sedimentation that can be distinguished on the basis
of trace fossil distribution.
Slow deposition in the marine realm will give rise to a homogeneous
sediment riddled with burrows; incipient stratification caused by changes
in sediment grain type or size, or by current lamination, will be mostly
obliterated. This is because the rate of bioturbation, especially by deposit-
feeding infauna, is much more rapid than the rate of deposition. For
example, Rhoads (1963) calculated that a single deposit-feeding species
in Long Island Sound was able to rework completely the annual depo-
sition twice in the course of a year. On average, each sedimentary grain
that accumulates is disturbed (moved, eaten, pelleted) several times before
it becomes part of the stratigraphic record. A thoroughly bioturbated tex-
ture in sediments or a sedimentary rock indicates continuity of deposition
at a rate outstripped by bioturbation.
It is difficult to determine an actual rate, for bioturbation depends on
population density. Abyssal sediments accumulate at rates of centimeters
per thousand years, but benthic biomass is very low. Shallow estuary or
marine shelf accumulation rates are two or three orders of magnitude
higher, but population sizes are enormous because of the abundant food
supply. Thus abyssal clays and shallow subtidal muds are both thoroughly
bioturbated.
Rapid deposition will leave a record of physical stratification. Dep-
osition of a turbidity current or a storm surge occurs much more rapidly
than the processes of bioturbation-even large populations of active de-
posit feeders take at least days to churn a sedimentary layer (Howard and
Elders, 1970; Elders, 1975). During the actual deposition of a graded bed
or storm stratum, biological processes are simply overwhelmed. This in-
flux of sediment will often destroy the existing benthic community, so
that the stratified layer remains intact even after rapid deposition has
ceased. If a period of slow deposition ensues, a new burrower population
can be recruited and bioturbation may rework the upper part of the rapidly
deposited layer. Burrowers typically forage within a few centimeters of
226 Chapter 5

the sediment-water interface, so a stratified layer thicker than this bio-

turbation zone should be preserved. A very thin (cm) stratum can be
obliterated; conversely, a sequence of several rapid depositional events
can form an amalgamated series of stratified units with no bioturbation.
An isolated rapid depositional event in a prevailing slow-deposition en-
vironment should produce an asymmetric stratum. The base will sharply
overlie bioturbated sediment; in fact, if the depositional event is preceded
by erosional scour, as in a turbidity current or storm surge, the underlying
burrows can be truncated. The top of the rapidly deposited stratum will
commonly be less well-defined; it will grade into the overlying slow-
deposition sediments.
These alternations in depositional rate are coming to be recognized
as common in the stratigraphic record. They have long been known in
turbidite basins, where the rapid depositional event produces graded beds
and Bouma sequences, and the intervening slow pelagic sedimentation
forms thin shale layers. Turbidites may have scoured bases and burrowed
tops (Seilacher, 1978). More recently, the importance of storm deposition
at shelf depth has been realized.
Howard (1972) noted that modern shoreface sediments consist of
alternations of laminated and burrowed sediment. Individual laminated
units are 10-50 cm thick and have basal scours. Burrows increase in
density upward through the unit; the upper sediments are thoroughly
bioturbated. A truncating erosional surface begins the next laminated
stratum. Howard interpreted these sequences as alternating periods of
storm and fairweather conditions in an environment at or just below
fairweather wave base. He recognized similar cycles, termed parallel-
laminated-to-burrowed, in the Cretaceous shoreline sediments of the
Western Interior.
Bowen et al. (1974) showed that episodic scour and deposition were
responsible for producing fossil coquinas in the Upper Devonian shelf
sediments of New York. Coquinas form the bases of laminated strata,
which grade upward into thoroughly bioturbated muds. The episodic
depositional events were postulated to be storms that swept across the
subaqueous delta platform west of the Catskill shoreline.
Bambach has been a strong advocate of the storm hypothesis to ex-
plain coquinas and laminated-to-burrowed strata. He and his co-workers
(Bambach et al., 1978) have proposed that much of the Phanerozoic fossil
record on the cratons consists of shell accumulations formed by storm
reworking. According to this hypothesis, storm events are recorded by
coquinas overlain by a hummocky-bedded stratum. The top of this stratum
will be burrowed by a reestablished benthic infauna.
Probable storm events have also been recognized in unfossiliferous
sandstones of the cratonic Cambrian (Porter and Byers, 1979; Byers and
Dott, 1980). The very fine sandstones of the Norwalk Formation (Fig. 1)
Marine Biogenic Sedimentary Structures 227

111111illlill
~
.,1 ..-., ~......... :;'1 .'~.":- J'
.. ~... -.....,,.,,,
Planar Lamination
/S coured Contact
Intense Bioturbation
,.;-, • .,.-<
-; " .,.., II ...... '.'

0."·
~ I •
" •
,
.1.
Scattered Burrows
c>
:I '. •
...
o
~

Q)
c:
o
t1
'a
c:
co Planar Lam in ation
<II

Q)
c:

...>-
Q)
>
lOomL
10 em

/ Scoured Contact

Figure 1. Field sketch of episodically burrowed sandstone in the Upper Cambrian Jordan
Formation of Wisconsin. Laminated beds with scoured bases are interpreted to be storm
surge deposits that were bioturbated by a reestablished benthic infauna.

display laminated-to-burrowed sequences similar in scale to those de-

scribed by Howard. The Cambrian deposits cannot be related to a pro-
grading clastic shoreline, unlike the Cretaceous or Recent examples How-
ard documented. Apparently, storm scour and deposition affected a broad
subtidal shelf area, a low-gradient "clastic bank" that covered the central
craton during the Late Cambrian.
In all the examples cited, the interpretation of episodicity of depo-
sition comes from recognition of the asymmetry of beds with a sharp basal
contact and gradational upper zone: "instants" of dominance by physical
processes followed by longer periods of "normalcy," dominated by bio-
logical processes.

2.3. Substrate Consistency

Biogenic structures can provide direct evidence of the physical prop-

erties of the substrate. Rhoads (1970) showed that water content in fluid
228 Chapter 5

muds controlled the way in which the sediment deformed under stress
by a burrower. The distinct burrows formed in a low-water, plastic mud
are frequently seen in modern sediments and preserved in ancient rocks.
High-water, thixotropic muds contain few distinct burrows; instead, both
modern and ancient sediments appear swirled or mottled.
Thorough bioturbation in very liquid muds obliterates any preexisting
stratification but does not impress a new biogenic architecture on the
sediment; such muds may even appear homogeneous. Cryptic bioturba-
tion can be inferred by the randomization of platy particles (organic car-
bon fragments, detrital micas) and by the lack of fissility in ancient mud-
rocks (Byers, 1974). There is also some evidence that bioturbation
permanently randomizes clay particle fabrics. Strongly parallel clay plates
are characteristic of black shales that have not undergone bioturbation;
these fabrics are visible in high-magnification scanning electron micro-
graphs taken perpendicular to stratification. My students and I have ob-
served that this parallelism of microfabric is absent in burrow fills in
black shale and in thoroughly bioturbated mudstones (Cluff, 1976; Larson,
1977). Apparently, disturbance of the mud microfabric is permanently
impressed into fine sediments, and even compaction does not fully realign
clay plates (ef. Moon, 1972; Bennett et al. , 1977).
A subject in need of further study is the timing and method of burrow
filling. A burrow that has been filled passively, by sedimentation, indi-
cates that an actual hole existed on the seafloor, whereas a burrow back-
filled by the organism was only "open" in small sequential segments, the
volume occupied by the organism as it moved (Byers, 1973). Backfill is
indicated where the burrow fill is meniscate (a series of nested crescent
shapes) or swirled. Active filling is characteristic of deposit feeders min-
ing the substrate. Backfill is derived from the matrix, with modifications
resulting from the feeding process. The fill may show grain segregation
(ingested versus rejected particles) (Fig. 2) or removal of the organic frac-
tion to produce a fill lighter in color than the matrix. Backfilled burrows
indicate that the sediment was plastic enough to record the passage of
the animal (Rhoads, 1970); they do not necessarily mean that the sediment
was solid enough to maintain an open tunnel.
Passive fill may be recognized by sediment obviously derived from
an overlying stratum and distinct from the surrounding matrix (Fig. 3).
An open burrow can be maintained in naturally strong (cohesive or
semiindurated) sediment or by burrow linings constructed by the animal
(Frey, 1978). Some linings are obvious and preservable, such as chitinous
tubes, sand grain walls in mudstones (Terebellina), and pelleted sand
walls (Ophiomorpha). Other linings are formed by mucus impregnation
of the matrix to form a solid wall; these are not preserved in ancient
burrows, but their existence is inferred where plainly passive fills or
Marine Biogenic Sedimentary Structures 229

Figure 2. Backfilled burrows in the Fox Hills Formation of South Dakota. Burrow consists
of a dark clay-rich fecal backfill surrounded by a halo of silt grains. The deposit-feeding
animal apparently exercised size selection, ingesting clay and organic detritus and rejecting
coarser particles. Scale bar: 1 cm.

actual open shafts are found in noncohesive sandy sediments (Skolithos).

Also, such linings may be selectively cemented or diagenetically en-
hanced (see Section 2.4). The presence of linings indicates a substrate of
little or no cohesion or induration.
Passive filling of burrows without linings would indicate sediments
strong enough to hold open the burrow shaft or tunnel. Unfortunately, it
is not always possible to distinguish between active and passive fill. A
case in point is Chondrites, which has been interpreted by various authors
as both actively filled and passively filled, and as lined and unlined (see
Ekdale, 1977, for review). Osgood (1970) employed a model Chondrites
burrow system to demonstrate experimentally that the burrow could be
filled passively. Further research along these lines should be carried out.
True burrows are defined as excavations in unconsolidated sedi-
ments; if the substrate is lithified or consolidated, then the excavation is
a boring. Borings are important in a sedimentologic context as agents of
erosion and clast generation and as indicators of substrate hardness and
its sequence of development (Warme and McHuron, 1978). Borings have
sharp wall boundaries, which may transect sedimentary grains, crystals,
or fossils.
Borings produced by macrofauna appeared by the Early Cambrian
230 Chapter 5

Figure 3. Deposit feeder burrows in the New Albany Shale of Kentucky. Burrows extend
downward from an intensely bioturbated greenish-gray shale layer into laminated black
shale. Scale bar: 1 cm.

(James et al., 1977) and are abundant in hard substrates today. During
Paleozoic time borings were mostly simple rod- or club-shaped holes in
reefs or hardgrounds. During the Mesozoic and Cenozoic a great diversity
and abundance of borings developed (Warme and McHuron, 1978).
Borings can be excavated in substrates of various degrees of consol-
idation, from stiff mud to igneous rock. Borings in calcareous substrata
are especially significant for their abundance and the evidence they pro-
vide for submarine lithification. Breaks in sedimentation can lead to the
development of an omission surface, dominated by a particular "omission
suite" of trace fossils (Bromley, 1975). The omission suite traces typically
extend down from a specific surface and may be filled by younger sedi-
ments after deposition resumes. Omission suite traces normally differ
from traces produced during active sedimentation. An unconsolidated
omission surface can become riddled with burrows that transect the older
preomission burrows in the underlying sediment.
In carbonate environments, an omission surface is subject to sub-
marine cementation (Purser, 1969); the omission suite on such a hardg-
round surface will comprise borings (Bromley, 1975). Identification of
borings is therefore crucial evidence in documenting hardgrounds in an-
cient carbonates (Byers and Stasko, 1978) (Fig. 4).

2.4. Diagenesis

Frey (1975) pointed out that the subject of diagenesis and minerali-
zation of burrows is an "open field." This assessment remains true today.
Marine Biogenic Sedimentary Structures 231

The following short discussion focuses on a few observations that have

recently been published and some hints of profitable research directions.
Differential diagenesis of burrows or burrow linings has been com-
mented on by several authors. Frey (1970), Simpson (1957), and Byers
and Stasko (1978) noted pyritization of burrow linings. Pyrite formation
was probably a consequence of bacterial reduction in a local organic-rich
concentration, the mucus lining of a burrow. Ekdale (1977) described two
forms of "reduction burrows" that occur in deep-sea sediments. Rind
burrows possess a whitened burrow wall; solid burrows are white
throughout their entire filling. Ekdale proposed that these two forms are
stages in a chemical reduction process that begins at the burrow margin
and extends inward. Reducing conditions apparently convert Fe 3 + , which
is dominant in the oceanic red clay matrix, to FeZ +. The soluble divalent
ion is leached away, leaving a zone colored only by white siliceous or

Figure 4. Borings (Trypanites) in a hardground surface in the Platteville Formation of Wis-

consin. Boring walls are sharp and mineralized by pyrite. Fill is derived from overlying unit.
Scale bar: 1 cm.
232 Chapter 5

calcareous microfossils. The reduction mechanism is not known, although

Ekdale suggested that infaunal respiration might increase the CO 2 con-
centration in interstitial waters. The bacterial reduction of dissolved sul-
fate to form sulfide has been shown by Berner (1971) to be dependent on
an adequate amount of metabolizable organic carbon in a nonoxidizing
environment. Sulfide ions then react with detrital hematite to produce
pyrite. In the deep-sea red clays, organic carbon content is very low, and
the bottom waters and sediments are well-oxygenated. It is not clear how
iron could be maintained in a divalent state and removed under these
conditions.
Ekdale observed that the development of color contrasts between
burrow fill and matrix may take millions of years. Burrow definition
increases with age and is best in sediments of the Miocene or older. This
slow diagenesis seems distinct from the bacterially mediated production
of iron sulfides, which occurs just below the sediment-water interface.
Differential diagenesis of burrows has been studied in terms of how
it affects burrow preservation. Shourd and Levin (1976), for example,
figured beautifully preserved specimens of Chondrites that had been dif-
ferentially dolomitized or chertified; the burrows were etched out by
outcrop weathering of the limestone matrix. Byers and Stasko (1978) sug-
gested that similar selective dolomitization of Chondrites was caused by
localization of diagenetic fluid flow through burrow systems in imper-
meable limestone. Recently, Delgado (1980) demonstrated that large
Ordovician Thalassinoides burrows were selectively dolomitized
syndeposi tionall y.
Finally, relating bioturbation to practical (financial) matters, it has
been shown in a few published studies that biogenic structures may in-
fluence the migration of hydrocarbons. Dawson (1978) studied the min-
eralogy, grain size, and fabric of an Upper Cretaceous sandstone contain-
ing a Cruziana assemblage of trace fossils. Burrow fillings and matrix
were analyzed separately for porosity. Stipsellus, Planolites, and small
Thalassinoides were filled with fine, well-sorted sand; elongate grains
paralleled the burrow walls. Initial porosity, determined by point count-
ing cement, was 20%. The matrix was poorly sorted, clayey-micaceous
sand with no apparent grain orientation. Matrix porosity was only 5-7%.
In contrast, Rhizocorallium, Pseudobilobites, Gyrolithes, and large Thal-
assinoides exhibited no differences between fill and matrix. Dawson sug-
gested that the increased porosity and probable increased permeability
in some of the burrows could provide permeable conduits for hydrocar-
bons. An unbioturbated sandy unit of this sort would not be porous
enough to form a reservoir; with the presence of the proper ichnogenera
it would be. The important conclusion to be drawn from Dawson's study
is that biogenic structures tend to contain their own sedimentary fabric,
Marine Biogenic Sedimentary Structures 233

which may differ from that of the matrix. A formation's physical prop-
erties may depend on macroscopic internal structures (burrows) as well
as on microscopic structures (average intergranular porosity).
Byers (1980) suggested that primary migration of petroleum from
source beds might be facilitated by the presence of a laminated fabric, the
laminae providing conduits for hydrocarbons. Conduits would be de-
stroyed by bioturbation. Hence source rock potential might also depend
on degree of bioturbation.

3. Paleobathymetry
3.1. Introduction
Sedimentary geologists have searched for years for a "magic wand,"
which, when knowledgeably waved, would conjure up the depositional
environment of the strata under study. More especially, one would like
to have a specific indicator of water depth in the ancient environment.
When Seilacher, in a series of papers (1958, 1964, 1967, 1978), proposed
that trace fossils might serve as such bathymetric indicators, the utility
and elegance of his concept combined with the intensity of desire for the
"magic wand" created a strong bandwagon effect. Strata were interpreted
or reinterpreted in light of Seilacher's model, which quickly assumed the
status of dogma. Seilacher's ichnofacies (Fig. 5) have been adopted uni-
versally. It would be superfluous to list here the large number of authors
who have employed Seilacher's method; summaries have recently ap-

REDBEDS SEA LEVEL

-~~--~-

-------------"'-'"
--~-------- ,

-- -----..----
--~-----........-
~~--
-------"
.... ~----~
..... .... ---.,..-~.
---------
~

--------
~--~--
~~:::~.:::::::::

----------------
~~------- .
---~---~-.

I/)
w
U
c:t
::; Scoyenia Skolithos Cruziana Zoophycos Nereltes
z
:r
u
Figure 5. Seilacher's bathymetric ichnofacies zonation. Each bathymetric zone is charac-
terized by a specific assemblage of trace fossils displaying similar behavior in response to
depth. After Seilacher (1967).
234 Chapter 5

peared in Frey (1975) and Howard (1978). Because the model has seemed
to work, it has been accepted rather uncritically, especially considering
its empirical basis. The Seilacher bathymetric model is not actually un-
iformitarian. It is a generalization taken from the stratigraphic record. As
with all such geological models, we may know that they work, but not
why they work, until we ground the model firmly in the Recent. Seilacher
has always attempted to do this, looking at modern ocean sediments for
traces similar to those in the record and attempting to rationalize the
ichnofacies patterns as responses to physicochemical conditions in the
ocean. However, nowhere does Seilacher clearly explain the dichotomy
between his empirical method and his deductive explanation of it. Indeed,
Seilacher's numerous papers on the topic are notable for the manner in
which these two aspects are interwoven. As time has passed and Sei-
lacher's model has become the paradigm, his theoretical explanation has
also been accepted with little comment. Now, however, it seems time for
a critical review of the basis of Seilacher's model. Evidence is mounting,
from the study of modern oceans, that some of Seilacher's deductions are
not correct. As the theoretical basis of the model has begun to be tested,
contradictions have appeared. The new evidence and its implications will
be discussed in Sections 3.2 and 3.3.
Before discussing the problems of Seilacher's model, it should be
pointed out that when he first proposed it, in the early 1960s, the turbidite
concept itself was new. Geologists were still adjusting to the very notion
of deepwater sands, and the criteria used to interpret shallow versus deep
were not yet codified. Note for instance that Seilacher's (1967) much-
copied bathymetric diagram lists only two physical sedimentary features,
"oscillation ripples" and "turbidites." Clearly this is a first-cut subdivi-
sion of the marine realm! Seilacher's ichnofacies are similar approximate
categories-he was concerned with recognizing major depth zones. In
recent years, resolution of physical criteria of depositional environments
has improved (e.g., detailed subdivision of deep-sea fans), but equal prog-
ress has not been made in ichnofossil criteria. A major question is whether
resolution can be improved. If enough effort is expended, will we be able
to refine the depth zonation? Do ichnofacies display slight changes in
response to bathymetry? Can we subdivide Seilacher's original depth
regions of the sea? Or has the model reached the limits of its resolution?
In order to answer these questions it is necessary to know why the bath-
ymetric model works in the first place.

3.2. Problems with Seilacher's Model

The basis of Seilacher's bathymetric model was his observation (1958)
that trace fossils do not occur in random mixtures. Rather, traces tend to
Marine Biogenic Sedimentary Structures 235

occur in associations or suites. A suite of traces is commonly found with

a particular assemblage of sedimentary structures as well; the inference
is that a given sedimentary environment is recorded in the sediment in
terms of the physical and biological processes acting therein. Process is
an important word, for it is implicit in Seilacher's model that traces are
to be taken mainly as records of behavior, and that behavior is dominated
by the physical environment. This mechanistic view colors much trace
fossil research (for obvious reasons-rarely do we have any idea what
actual animal made a particular trace). Still, I think it is a critical point
that the emphasis on behavior tends to relegate the organism to the status
of perpetrator; the result is to focus on the distribution of traces and
overlook the distribution of organisms. One has a sense that, given the
proper stimulus (from the physical environment), it really does not matter
much what kinds of animals are present, because surely at least one
species will exhibit the requisite behavior. This is a view of organisms
as readily moldable and programmable entities. Perhaps they are, but on
the other hand a trace may be the work of only one species, and if the
animal is not present for one of a variety of biological reasons, the trace
will be absent as well. The distribution of traces is inextricably linked to
the distribution of animals, and the latter is controlled by more than one
factor (i.e., depth).
If we adopt Seilacher's viewpoint and assume that behavior is
strongly controlled by physical conditions, then what is the evidence that
depth is the major physical factor? It happens that, although this is a
plausible assumption, Seilacher has simply asserted the point, not proved
it. For example, in his 1967 paper, he says "The [trace fossil) communities
are directly or indirectly related to depth no matter what factors caused
the differences between them" (p. 414); this statement comes early in the
paper, not as a conclusion, and is presented without qualification, evi-
dence, or explanation. Again, a decade later (1978): "Water depth, by
itself, is not a major ecological factor, but more important parameters such
as turbulence, sedimentation roles and processes, diagenetic processes,
productivity and evolutionary processes, are in one way or another depth-
related. Therefore, we are justified to arrange different ichnocoenoses in
a relative bathymetric sequence" (p. 194). This is not a very specific
causality.
The argument could be sharpened a bit by focusing on physical energy
level rather than depth per se. In Seilacher's original subdivision he was
interested in separating very shallow (littoral) from very deep (flysch)
environments. At this scale, depth and turbulence are inversely related.
However, within depth zones, or along a depth gradient, the negative
correlation may not hold. Several authors have pointed out that turbulence
can be quite variable within a given depth zone, and that ichnofacies may
236 Chapter 5

change in response to turbulence variations involving minimal change in

water depth.
Within a generalized shallow marine environment, the strict depth
zonation of the Skolithos and Cruziana facies may not apply. Skolithos
and other vertical burrows do seem to indicate more energetic conditions,
but these do not everywhere correspond to shallower water. Likewise the
horizontal burrows and trails of the Cruziana facies mean quieter but not
necessarily deeper water. Agel' and Wallace (1970), in a study of Jurassic
traces of the Boulonnais, recognized an offshore-to-onshore gradient from
horizontal to vertical burrows, but they also found horizontal Rhizocor-
allium and Thalassinoides in a nearshore setting protected by a barrier
bar. Fursich (1975) further developed the concept of separating depth and
energy in the shallow marine realm. In Upper Jurassic strata of England
and Normandy he identified three trace fossil associations, primarily re-
lated to hydrodynamic conditions: a high-energy Diplocraterion associ-
ation and low-energy Rhizocorallium and Teichichnus associations. The
Teichichnus association was found in quiet offshore sediments and in
shallow but protected lagoons. The reappearance of a quiet-water assem-
blage in a vertical sequence need not mean reversal of a transgressive or
regressive cycle. Crimes (1970) observed that the Middle Cambrian strata
of south Wales contain alternating beds of Skolithos-bearing sandstone
and siltstone-mudstone with Planolites and Chondrites. He ascribed
these alternations to temporary changes in environment (within a shallow
marine setting) and not to paleogeographic change. Similar conclu-
sions were advanced by Byers (1977b) and by Driese et al. (1981) for
Cambrian sandstones in the Upper Mississippi Valley. In short, every
Skolithos-Planolites couplet does not a transgression make.
Osgood (1970), in his monograph on Cincinnati trace fossils, reported
that the prevalent traces belonged to Seilacher's Cruziana facies, but he
found that the differences between the Cruziana and Nereites facies are
not always clear-cut. He pointed out that Seilacher had not rigorously
defined the depth zones, and he cited evidence that the facies might
overlap. The Zoophycos facies seemed especially suspect to Osgood. Zoo-
phycos is very common in shelf-depth rocks of the New York Devonian
but had also been found in Recent deep-sea cores (ef. Seilacher, 1967).
Osgood and Szmuc (1972) expanded the argument: they reported a wide-
spread occurrence of Zoophycos in Lower Mississippian strata in asso-
ciation with oscillation ripples, large cross-laminae sets, and Lingula. The
inferred shallow conditions prove that the range of Zoophycos extends
from above wave base, through Seilacher's "intermediate" depths, to
abyssal levels. Osgood and Szmuc concluded rather cautiously that "the
validity of Zoophycos as a reliable depth indicator is thereby somewhat
in doubt" (p. 1). Continuing study of DSDP cores has amply confirmed
Marine Biogenic Sedimentary Structures 237

the presence of Zoophycos (also Chondrites, Planolites, and Teichichnus)

at abyssal depths (Chamberlain, 1975; Ekdale, 1977, 1978; Ekdale and
Berger, 1978). In fact these traces dominate cores of deep-sea sediment,
even where the sediment surface is covered by traces of Nereites-facies.
Ekdale concluded that the cores preserve the last imprint of burrowers,
which are the infaunal deposit feeders rather than the surface grazers.
Thus the Nereites facies may be present in the deep sea but go unpres-
erved; only in a turbiditic regime are the grazing traces occasionally buried
by sandy layers and protected from destruction by bioturbation. The deep
sea is actually a combination of the Nereites and Zoophycos zones, one
on the surface and one infaunal, but under conditions of slow sedimen-
tation only the Zoophycos assemblage is preserved.
Traces belonging to the Cruziana assemblage have also been reported
in anomalously deep settings. Crimes (1977) described an Eocene deep-
sea fan from northern Spain in which the middle fan facies contains
typically a Nereites assemblage plus such "shallow-water" traces as Ar-
enicolites, Diplocraterion, Ophiomorpha, Pelycipodichnus, Rhizocoral-
lium, and Thalassinoides. Crimes interpreted this anomaly in terms of
water turbulence; the middle fan environment is characterized by rapidly
moving sand-laden currents, a sandy substrate, strong oxygenation, and
low levels of organic detritus-similar conditions to those in intertidal
and shoreface environments. The main controls on trace fossil distribution
in this instance appear to be substrate and water movement, not depth.
Similar anomalous occurrences were reported by Kern (1978), who
found ?Arthrophycus in the Nereites facies of an ancient Alpine deep-sea
fan, and by Armentrout (1980), who described Ophiomorpha from a series
of subdivisions of an Eocene fan. Armentrout estimated a paleobathymetry
of 200-1200 m on the basis of associated foraminifera.
Both Seilacher (1967) and Rhoads (1975) have shown how turbulence
may directly influence burrow morphology in the shallow marine realm.
Deep vertical burrows are characteristic of the shifting substrate and rapid
erosion or sedimentation events in a turbulent regime. Rhoads also dem-
onstrated that deep burrowing allows an animal to escape the rigors of
salinity or temperature changes, which may occur in a shallow environment.
It is in the switch from vertical to horizontal burrowing that turbu-
lence loses direct control of burrow morphology. Seilacher postulated
(1967) that horizontally directed burrows result from deposit feeding, that
animals mine the sediment following nutritious seams, and that these
seams form by settling of food particles in quiet water. Turbulence thus
acts to control burrow form through the agency of feeding behavior. Sei-
lacher expressed this postulate essentially as a conclusion (1967, p. 420),
and, once the primacy of the feeding postulate is assumed, then it follows
that the intricate patterns of the Zoophycos and Nereites facies could
238 Chapter 5

correspond to ever more careful mining techniques. This implies that

food was ever more scarce as water depth increased. Thus the "food
paradigm" was born. I review this didactically in order to point out that
Seilacher did not use data on food in the modern ocean as the theoretical
underpinning of his model.

3.3. Evidence from the Oceans

In a series of recent papers, Kitchell and her co-workers have at-

tempted to test the food paradigm by comparing the distribution of food
and trace types in the modern oceans (Kitchell, 1979; Kitchell et a1.,
1978a,b). The results of these studies cast doubt on the food paradigm
and suggest that the distribution of traces is controlled by variables that
are not depth-related.
Briefly stated, if the food paradigm were true, traces in the modern
ocean should show a bathymetric gradient of increasing organization and
complexity of form (Kitchell et a1., 1978a). Because complexity of form
can be expressed as complexity of commands needed to simulate the
trace, it is possible to rank traces on a scale of complexity, from simple
scribbles through spirals to meanders. Such a progression might be ex-
pected down a bathymetric gradient. In studying thousands of bottom
photographs from the Antarctic and South Pacific, Kitchell et a1. (1978a)
found no depth difference between spirals and meanders, although both
trace types occurred in a specific zone (3000-3500 m). This strongly
suggests that trace distribution is not being controlled by a food availa-
bility gradient. Even more significant is the total lack of either spiral or
meander traces in the Arctic Ocean at comparable depths; these trace
types are thus shown to be species-specific and not simply depth-
dependent.
Kitchell et a1. (1978b) made a more extensive comparison between
Arctic and Antarctic traces as a test of the hypothesis that productivity
controls feeding behavior. The productivities of Arctic and Antarctic
waters are approximately known; that of the Arctic is lower by a factor
of 104 • Food should be strongly limiting to benthic grazers in the Arctic
sediments and much less so in the Antarctic. The food paradigm would
predict intricate traces (Nereites facies) in the Arctic and scribble traces
in the Antarctic. The actual situation is just opposite-a highly diverse
assemblage of Nereites facies traces is present in the food-rich Antarctic,
whereas only a few scribble traces are to be seen in the impoverished
Arctic (Fig. 6). Kitchell concluded that the presence or absence of partic-
ular surface-grazing species exerts a greater influence on trace diversity
than depth or nutrient supply.
Marine Biogenic Sedimentary Structures 239

ARCT I C CANADA BASI N

ANTARCTIC BELLINGSHAUSEN BASIN

Figure 6. Traces from the Arctic and Antarctic deep-ocean floor. Diversity of complex feeding
trails is much higher in the Antarctic in spite of the relatively larger food supply. 1, Pinnate;
2, groove-mound; 3, ridge; 4, plow; 5, crenulated plow; 6, spiral; 7, meander; 8, decapod;
9, rosette; 10, tread, 1,2; 11, fecal; 12, tread, 3; 13, cone; 14, crater; 15, star; 16, railroad; 17,
circular ridge; 18, groove. Diagram and descriptive names from Kitchell et a1. (1978b).
240 Chapter 5

Kitchell's 1979 paper is a theoretical approach to the food paradigm;

trace fossil morphology is discussed in terms of optimal foraging theory.
The initial hypothesis reads thus: If trace morphology is controlled by
feeding behavior and if food is assumed to decrease with increasing depth,
then natural selection should produce optimal foraging. Feeding patterns
should maximize coverage for energy expended. Optimum foraging im-
plies nonrandom motion-down the bathymetric gradient, traces should
show less and less random movement. Kitchell used various components
of trace shape to quantify random motion in 302 traces (selected from
4806 deep-sea photos). Principal-components analysis extracted a single
measure of randomness for each trace. This measure was plotted against
depth (Fig. 7) and found to have a nonsignificant correlation (r = 0.26).
In other words, at depths from about 600 m to 6000 m traces do not

6300
6000
5700
5400

... . ...
5100
4800

E 4200
4500 • ~
•
.. ••
•••
:-.\. .. .
1- as- ••
;; 3900 .:. ..; . •
b:3600 ••: ::... • ••

. .. . ....
lJJ
a 3300 • • • • • • : ! 11 . . . . . . .! •

3000 t\ . . • et

2700
2400
2100 -.. .,
1800
1500
R-O.26
1200
900 • !f• ••
. .. . .
!- "n etS - ••
y- 626.9X + 2578

600
'-;;0:--::2;;;0:-;;.4"'0~60;:;--;8-::;'0;-;-;1.0~0'""'12"'0-C-17.40;;--;-;'6""0""18;;;0'-:2"""0·0
FIRST FACTOR AXIS

Figure 7. Trace fossil "morphological randomness" versus water depth. The horizontal axis
is a principal-components analysis measure of randomness, which decreases to the right.
Maximum randomness is exemplified by pinnate and plow traces (ef. Fig. 6), at values less
than 0.8. Minimum randomness is shown by spirals and meanders, at values of 0.8-1.9.
Multiple circles represent more than one score of the same value. No significant correlation
is present between water depth and degree of randomness. From Kitchell (1979).
Marine Biogenic Sedimentary Structures 241

become less random (more efficient) with depth. Clearly, trace shape and
the various ichnofacies are not controlled simply by depth.
If decreasing food supply is not the whole story, what is? Kitchell
(1979) suggested two possibilities, control of trace morphology by patchy
distribution of food and control by predation risk. The latter seems more
plausible: Although spirals and meanders might be efficient ways to mine
a rich food patch, Kitchell noted that the deep-sea food supply (for deposit
feeders, at least) is probably homogeneously distributed. Given a uniform
food distribution, an animal might just as well forage in a straight line;
in terms of efficiency alone, a straight line and a spiral are equivalent.
The spiral's advantage may be that it minimizes the risk of encountering
a predator. The prevalence of spirals and meanders in the Antarctic,
known to have a high population density, supports the hypothesis that
trace-makers are attempting to maximize food encounter per unit length
of travel and minimize total area covered. The overall import of Kitchell's
work is that it casts doubt on the simple process-response aspect of the
food paradigm and suggests that biological interactions are important in
controlling trace morphology and distribution. In a word, the animal has
been returned.

3.4. The Graphoglyptid Controversy

Another major problem in interpreting ancient bathymetry is that

posed by the graphoglyptid traces (Paleodiet yon and similar forms). These
extremely regular burrow systems (usually networks of hexagons) are
characteristic of many ancient flysch sequences and are a component of
Seilacher's Nereites facies, but they are unknown (with one exception-
see below) in the modern deep sea. Besides this enigma, the graphoglyptids
do not fit the optimal foraging hypothesis implied in the food paradigm.
Also, it is not even certain whether they were formed infaunally in muds
and later filled by sand or formed by burrowing along a sand-mud in-
terface. This latter point is actually most critical, for arguments regarding
the food paradigm ultimately hinge on whether the graphoglyptids were
pre- or postdepositional. The controversy is another instance of the tend-
ency of trace fossil theorizing to outrace the evidence.
Postdepositional burrows along sand-mud interfaces imply deposit
feeding, with the animal mining out a food-rich seam. Therefore gra-
phoglyptid morphology should be controlled by the same factors as other
Nereites traces, also thought to be feeding trails. Unfortunately, the open
network of Paleodietyon and its allies is not very efficient. Large areas
within the net are left unmined, so the burrow length/total area ratio is
small, compared to that of spiral and meander burrows. Why should the
242 Chapter 5

Paleodietyon animal go to the trouble of constructing such elaborate and

perfect nets if they are an inefficient method of deposit feeding? Seilacher
(1974, 1977a,b) recognized this paradox and sought to resolve it by pos-
tulating that graphoglyptids were not deposit feeding burrows at all, but
open tunnels in mud, in which animals survived by microbial cultivation
("mushroom gardens"). The nets are not optimal expressions of foraging
behavior because no foraging is occurring. Seilacher's hypothesis requires
that the nets be open tunnels in muds, preserved as trace fossils by in-
fillings of sand-in other words, the burrows must be predepositional,
relative to the overlying sandy bed.
Are the actual burrows pre- or postdepositional? Ksiazkiewicz (1970,
1977) and Roniewicz and Pienkowski (1977), in their studies of the Car-
pathian flysch, concluded that Paleodiet yon and some related graphog-
lyptids were produced by deposit feeders burrowing along sand-mud
interfaces (postdepositional burrows). The evidence for this origin is the
rare preservation of Paleodiet yon nets covering flutes or other traces
thought to be postdepositional and the fact that the basal parts of sand
units are apparently reworked by burrows (Ksiazkiewicz, 1970, Plates
4h,q). Roniewicz and Pienkowski (1977) provided no photographic evi-
dence, but simply asserted a postdepositional origin. Seilacher (1977a)
took the opposite view and claimed that the burrows were predepositional
tunnels in mud, which were scoured and filled by the overlying sandy
bed. Seilacher provided little evidence except to cite instances in which
burrow nets are cut by flutes. He published many drawings of recon-
structed burrow tunnel systems; these and the accompanying text appear
convincing, but fundamentally the burrow origin is simply not well-
documented from the fossil record. One wishes for more photographs of
traces showing their actual relationship to sole marks. Until the question
of timing and preservation is settled, all the theoretical superstructure
developed thereupon must be regarded cautiously.
For network burrows to be microbial farms, openings to the sedi-
ment-water interface would be essential for the introduction of organic
detritus. Seilacher has found instances of vertical shafts leading upward
from burrow networks, but it is impossible to prove whether these were
truly openings. If the tunnels were predepositional then they can only be
preserved where current scour cut down to the level of the network and
filled the tunnels with sand, thus obliterating the tunnel openings to the
original mud surface. Catch-22 is operating here: Without the scour-and-
fill, the burrow is not preserved; with scour-and-fill, the crucial upper
parts are destroyed. In his descriptions, Seilacher was forced to rely on
plausibility and analogy in postulating shaft openings (1977a, e.g., uni-
ramous burrows, p. 303; Helieolithus, p. 306; Punetoraphe, p. 307; Uro-
Marine Biogenic Sedimentary Structures 243

helminthoida, p. 307; Hormosiroidea, p. 309; Oseillorhaphe, p. 312;

Tuapseichnium, p. 314).
The aforementioned absence of graphoglyptids from modern deep-
sea bottom photos is not surprising if they were infaunally produced,
whether pre- or postdepositionally. Ekdale (1980) recently reported the
first instance of network burrows from the modern ocean: Spiroraphe,
Cosmoraphe, and Paleodietyon were recovered in the tops of the cores.
The burrows were open tunnels in soft sediment, a few millimeters below
the original sediment-water interface. Ironically, the burrows were visible
because the topmost millimeters of sediment had been washed away dur-
ing core recovery; vertical shafts, if they existed, would have been eroded
off. However, the presence of graphoglyptid tunnels in pelagic muds sup-
ports Seilacher's contention that the fossil nets are predepositional, and
Ekdale cited the smooth walls and uniform tunnel diameters as evidence
against a deposit-feeding origin.
Finally, it must be noted that Seilacher's (1977a) rather elaborate
graphoglyptid taxonomy, burrow network reconstructions, and "mush-
room garden" hypothesis all arise from the fact that network burrows fail
to support the food paradigm. Perhaps the network burrows are simply
additional evidence that the paradigm itself needs rethinking.

4. Precambrian Traces
Trace fossils have been utilized with some success in helping to
unravel the history of life in the Precambrian. Most work has focused on
two questions:
1. How do traces relate to the traditional Precambrian-Cambrian
boundary based on body fossils? (Can we use traces to define the
boundary in the absence of skeletal faunas?)
2. What is the oldest trace fossil, and what does it tell us about the
evolution of the metazoans?

4.1. The Precambrian-Cambrian Boundary

4.1.1. Evolution of the Concept

As with so many important aspects of ichnology, the concept of using
traces to define the basal Cambrian was enunciated by Seilacher (1956),
who pointed out that a rapid increase in trace types and abundances took
place across the Precambrian-Cambrian boundary. At the time Seilacher
244 Chapter 5

made this observation, Precambrian paleontology was steeped in mystery,

and the basal Cambrian was regarded as the beginning of the fossil record
of any consequence. Today, after the ensuing discoveries of Precambrian
microbiota and the Ediacaran-age fossils, the Precambrian-Cambrian
boundary is not regarded as such a fundamental break. Its definition is
now important for orderly stratigraphic procedure, rather than as a marker
of the origin of animal life.
The use of traces to define the basal Cambrian is made complex by
several factors. First, the definition of the traditional (body fossil distri-
bution) boundary has undergone revision during the last decade. In sev-
eral regions of the world, but not in most of North America, the lowest
trilobite zones are underlain by skeletal faunas dominated by archaeo-
cyathids, sponges, brachiopods, gastropods, hyolithids, and problematic
tubes. Soviet workers have established a new Lower Cambrian stage, the
Tommotian, based on this fauna; this stage conformably overlies Upper
Precambrian strata on the Siberian craton (Cowie and Glaessner, 1975).
Second, the discovery of the Ediacara fauna of soft-bodied metazoans
fortuitiously preserved, and the subsequent recognition of similar fossils
elsewhere, has extended the range of body fossils down below the tra-
ditional Cambrian. This raises the question of whether to define the Cam-
brian base at the lowest body fossil (the coelenterate-dominated Ediacara-
type fauna) or at the lowest shells (the Tommotian fauna) or at the lowest
trilobita (traditional in North America); see Stanley (1976) and Durham
(1978) for reviews. The "Ediacaran" period (or "Eocambrian" or "Ven-
dian") was approximately 100 m.y. [Basal Tommotian-570 m.y.; Basal
Vendian-680 m.y. (Cowie and Glaessner, 1975)] in duration, about the
same length as mid-Cretaceous to Recent; in trying to use trace fossils to
define the basal Cambrian, it clearly makes a difference whether we push
the Cambrian back to include the Ediacara faunas.
Young (1972) reported well-defined traces (e.g., Cruziana, Rusophy-
cus, Skolithos, Planolites) occurring below the lowest trilobite zones in
the Gog and Miette Groups of the Canadian Cordillera. The lowest trace
occurrence, Didymaulichnus, perhaps a mollusc trail, lies 2000 m stra-
tigraphically below the basal trilobite bed. Because the trace is rare and
not associated with other ichnogenera, Young relegated it to the Late
Precambrian. Similar situations were reported for other areas by Glaessner
(1969), Banks (1970), Bergstrom (1970), Cowie and Spencer (1970), Webby
(1970), and Crimes et al. (1977). The prevalent opinion has been that
there is an upward increase in number and complexity of trace types from
Precambrian into Cambrian strata (Crimes, 1975). In several of the cases
cited, this ichnofaunal transformation takes place a few hundred to a few
thousand meters below the overlying Cambrian body fossils, in apparently
Marine Biogenic Sedimentary Structures 245

continuous sequence. The problem is where to draw the line. Because

simple burrows and trails are known from rocks at least 100 m.y. older
than Tommotian, it is hard to tell whether the so-called upward diver-
sification is a synchronous, cosmopolitan progression (a 100-m.y. time
envelope is hardly an "event"). As long as the biostratigraphic criterion
is a sequential change or gradual diversification we will have little hope
of improving the degree of precision. Alpert (1977) clearly recognized this
problem and advanced a proposal to narrow the zone of confusion (see
Section 4.1.2).
The third major difficulty is that of facies control. The best biostra-
tigraphic guides are facies-independent. However, it is well understood
that most traces are facies-linked-the whole bathymetric utility is built
on that postulate. We should not be surprised to find that some burrows
and trails persist far back into the Precambrian, ignoring the Cambrian
threshold. Crimes et a1. (1977) sounded an overdue note of caution in
their analysis of Precambrian-Cambrian traces in Spain. Here again there
is an upward increase in trace diversity, but it correlates with an upward
facies change-a change only in detail, as all the rocks are cross-bedded
quartzites and interbedded sand-shale sequences of shallow marine or-
igin. The authors noted that the upward change through hundreds of
meters of section could reflect changing environments rather than world-
wide evolutionary change.
This latter point can be underlined by examining facies control in
rocks known to be Cambrian. For example, the basal cratonic quartz sand-
stone in the northern midwestern U.S., the Mt. Simon Formation (~65
m thick), directly and conformably underlies the finer-grained Eau Claire
Formation, which contains Upper Cambrian trilobites. There is essentially
no doubt that the Mt. Simon is also Upper Cambrian, but it contains no
diagnostic body fossils. Cruziana and allied traces are very common in
the shaly Eau Claire, rarer in the upper Mt. Simon, and essentially absent
in the very coarse sands of the basal Mt. Simon, where the only common
burrow is Skolithos (Driese et a1., 1981). Here is a situation in which trace
diversity increases upward into a zone containing trilobite traces, sur-
mounted by the lowest local body fossils. Because these lowest shells are
Upper Cambrian, we may be sure that the upward diversification is simply
facies-controlled, not evolutionary; for strata at the Precambrian-Cambrian
boundary level that surety is lost. The strong, demonstrable facies linkage
of traces makes them risky index fossils; the possibility of facies control
needs to be carefully considered before evolutionary explanations for
vertical changes are advanced. This is especially true for studies of local
extent; if similar trends can be documented on a cosmopolitan basis, and
in different sedimentologic settings, then traces may serve as guide fossils.
246 Chapter 5

4.1.2. Alpert's Proposal

Alpert (1977) surveyed the recent literature on trace fossil definition
of the Precambrian-Cambrian boundary and noted that the major problem
is the existence of a long trace-containing time-stratigraphic interval (the
lOO-m.y. envelope referred to previously), which is commonly labeled
"Cambrian or Precambrian?" This interval lies below beds containing
shelly fossils, which are unequivocally Cambrian, and above beds lacking
traces or containing rare, simple burrows; these latter strata may also
contain tillites and are usually called "Precambrian." From his own work
in California and from published reports of trace distributions in Canada,
Greenland, Norway, Sweden, Africa, and Australia, Alpert compiled a
range chart for trace fossils in the Upper Precambrian-Lower Cambrian
(Table I). He found that the upward increase in trace abundance, diversity,
and complexity is not gradual but occurs at specific horizons. Also, traces
do not appear singly, in random sequence, but as groups. Many ichno-
genera make their first appearance either at the levels of the lowest tri-
lobite body fossils (undoubted Cambrian) or below, at the level of the
lowest arthropod trace fossil. Alpert argued that trilobite traces (Ruso-
phycus, for example) clearly indicate an anthropod grade of evolution;
hence Rusophycus marks an evolutionary event. If the earliest Cambrian
has traditionally (in North America) meant the beginning of trilobites,
then the Cambrian should be extended downward to include the lowest
evidence of trilobite activity. Rusophycus, Cruziana, Diplichnites, and
other arthropod scratches would thus serve as index fossils, and non-
arthropod traces that appear coincidentally (e.g., Phycodes, Bergaueria)
could also serve, if arthropod traces were locally absent. In Alpert's
scheme, the upper part of the vague "Cambrian or Precambrian?" interval
would join the true Cambrian , on the basis of index ichnofossils.
Strata below the lowest Cambrian thus defined would be referred to
the Upper Precambrian. These beds may contain trace fossils. Alpert spe-
cifically noted that Skolithos, Planolites, Scolicia, and various unnamed
simple burrows are widespread in Upper Precambrian sediments.
Alpert's proposal has merit in that it does away with a cloudy interval
in the time-stratigraphic column. The "Cambrian or Precambrian?" clas-
sification was based not on any particular trace type but on the general
notion that metazoan life, as manifested by traces, was gradually ad-
vancing toward a "Cambrian grade." By keying his chronostratigraphy to
the appearance of specific traces, Alpert is employing the same kind of
zonation used for body fossils throughout the rest of the Phanerozoic. His
basal Cambrian is thus objective and is not dependent on a gradualistic
hypothesis regarding the diversification of life.
The major difficulty with Alpert's proposal is that it fails to address
Marine Biogenic Sedimentary Structures 247

Table I. Trace Fossil Zonation of the Late Precambrian-Early Cambrian°

GROUP 1 GROUP 2 GROUP 3

Trace fossils Trace fossils not useful Trace fossils known
indicative of early in delineating the basal only from the late
Cambrian age Cambrian boundary Precambrian

Dwelling burrows Diplocraterion Skolithos

Monocraterion
Laevicyclus
Dolopichnus
Bergaueria
Feeding burrows Phycodes Planolites Archaeichni urn
Rhizocorallium Unbranched horizontal
Teichichnus burrows
Arthrophycus Unbranched horizontal
Syringomorpha backfilled burrows
Zoophycos
Dictyodora
Trails or horizontal Plagiogmus Scolicia Bunyerichnus
feeding burrows Psammichnites Curvolithus Buchholzbrun-
Cochlichnus Didymaulichnus nichnus
Belorhaphe Torrowangea
Helminthopsis Helminthoidichnites
Astropolithon Unbranched horizontal
Dactyloidites trails
Oldhamia
Trilobite or Rusophycus
arthropod traces Cruziana
Diplichnites
Protichnites
Dimorphichn us
Monomorphichnus

a After Alpert (1977).

the problem of the Tommotian. As noted previously, the Tommotian cal-

careous body fossils form a pretrilobite assemblage in some parts of the
world and are recognized as being part of the Lower Cambrian. According
to Alpert, the Cambrian commenced with the first trilobite traces. These
commonly occur below the lowest trilobite body fossils, but Alpert is not
specific on whether these traces antedate or are synchronous with the
Tommotian. Apparently facies control returns to haunt us on this point.
The best trace fossils are from terrigenous clastic facies, whereas the
Tommotian faunas are found in carbonate facies. The basal Cambrian is
thus unfortunately defined by two different fossil types, and as yet there
is no tie point for correlation (Stanley, 1976). Almost certainly there is
time overlap between Alpert's Rusophycus-defined basal Cambrian and
248 Chapter 5

the Tommotian basal Cambrian, but the exact temporal relationship is

unknown. The problem is clearly exemplified by the Nama Group rocks
in South West Africa (Germs, 1972, 1973), which contain sprigginid
worms, probable Tommotian shelly fossils, and trace fossils that Alpert
considers Precambrian! Perhaps careful trace fossil studies in Tommotian
rocks will reveal traces common to both carbonate and clastic facies.
In summary, it appears that Seilacher was remarkably prescient in
his 1956 paper: Trace fossil assemblages do display an abrupt increase
in diversity, including the appearance of arthropod crawling traces, at or
just before the beginning of the Cambrian. In fact, the exact placement in
time of this diversification depends more on what is considered to mark
the Cambrian base than on our knowledge of the trace record itself.
Whether or not Alpert's proposal is accepted into formal stratigraphic
nomenclature, a radical narrowing of the 100-m.y. time envelope has been
accomplished.
A status report chronology:
1. Worldwide Late Precambrian glaciation at 680 m.y. BP (Cowie and
Glaessner, 1975).
2. Ediacaran (Eocambrian, Vendian) strata containing rare soft-
bodied fossil assemblages and simple burrows: 680-570 m.y. BP
(Stanley, 1976).
3. First appearance of arthropod traces (Rusophycus) and calcareous
faunas of Tommotian Stage (Alpert, 1977; Stanley, 1976): 570-560
m.y. BP. Trace fossils may be slightly older than earliest shelly
fossils (see Cowie and Glaessner, 1975, Fig. 13, Australian corre-
lation chart).
4. First trilobite body fossils (Soviet Atdabanian Stage, includes Fal-
lotaspis Zone, traditional base of Cambrian in North America).

4.2. The Oldest Trace Fossil

It is a matter of considerable interest to locate the "oldest" body fossil

or "oldest" trace; these Precambrian occurrences provide constraints on
models of animal phylogeny and rate of diversification. As implied in the
previous section, occurrences of both Ediacaran-type body fossils and
simple burrows such as Planolites seem to be restricted to sedimentary
rocks younger than about 680 m.y., postdating the Late Precambrian gla-
ciation. This implies that metazoan animal life was a relatively late de-
velopment, considering that microfossils and stromatolites are known
from Archaean times. The late-development veiw has recently been sum-
marized and championed by Stanley (1976), amplifying Cloud's (1968)
Marine Biogenic Sedimentary Structures 249

critical review of Precambrian animal fossils. Cloud's recognition of many

reported occurrences as pseudofossils cleared the decks for action; es-
pecially important was his rejection of numerous Proterozoic and Beltian
"fossils." No one can doubt the biogenicity of the Ediacaran faunas, but
supposed fossils or traces older than about 700 m.y. are regarded with
healthy skepticism now that Proterozoic "ancestral" brachiopods, worms,
or arthropods have been defrocked.
Traces from strata of Ediacaran age were divided by Alpert (1977)
into two groups, those long-ranging forms that persist up into the Cam-
brian and a few traces that seem to be known only from the Ediacaran.
All of these traces are rather simple burrows or surface trails. Vertical
burrows have been lumped into the genus Skolithos and horizontal bur-
rows into Planolites. It is extremely probable that many different biolog-
ical species have produced such simple structures over geological time.
These two ichnogenera are therefore even less useful as index fossils than
the average trace.
Of the traces known only from the Ediacaran, perhaps the oldest is
Bunyerichnus, a bilaterally symmetrical locomotion trace from the Bra-
china Formation in South Australia (Glaessner, 1969). The trace occurs
as a bedding plane structure, apparently produced by an animal without
discrete appendages; transverse ridging is reminiscent of traces made by
the foot undulations of molluscs (d. Scolicia, Plagiogmus). Webby (1970)
questioned Glaessner's interpretation of the fossil and suggested that it
might be a body fossil cast of an elongate organism, such as the Ediacaran
annelid Spriggina. Germs (1973) reported a similar trace, which he named
Buchholzbrunnichnus, from the Nama Group of South West Africa. A
possible sprigginid body fossil (mold) was also found in the unit.
In Australia, Bunyerichnus occurs a short distance above the upper-
most Late Precambrian tillite and 2000 m stratigraphically below the
Pound Quartzite, the stratum containing the classical Ediacaran fauna.
Glaessner's published photograph of Bunyerichnus shows a structure
more than twice as long (12 cm) as the well-known specimens of Sprig-
gina; in addition, Bunyerichnus has no hint of a head or tail, and the
specimen is broken at both ends. I find Glaessner's interpretation plau-
sible-Bunyerichnus does look like a trail. In terms of the origin of me-
tazoans, the trace-versus-body-fossil question is moot; Bunyerichnus rep-
resents a primitive soft-bodied, macroscopic animal that lived nearly 700
m.y. ago.
Supposed traces older than Bunyerichnus are more controversial;
their organic origin is questionable. Glaessner (1969) interpreted Brook-
sella canyonensis (1100-1300 m.y., from Grand Canyon Group) as a trace,
related to Asterosoma. Brooksella had also been considered a medusoid
impression (Bassler, 1941). Cloud (1968) rejected it as any sort of fossil
250 Chapter 5

and pronounced it a penecontemporaneous deformation structure. The

published photographs of Brooksella (Glaessner, 1969) are not entirely
convincing, and Glaessner's interpretation of this feature as a trace is both
tenous and convoluted. The adjacent bedding surfaces have apparent
raindrop imprints (Cloud, 1968), but Glaessner considers these to be algal
colonies. Suffice to say that Brooksella does not instantly cause the viewer
to exclaim, "That is a fossil!", whereas Bunyerichnus does.
Older so-called traces are even more problematic. Crimes et al. (1977)
gave a range for Planolites as Upper Precambrian-Recent*; they cited
Walcott's (1899) description of Planolites from the Belt Supergroup. Wal-
cott described two traces from Belt rocks, Helminthoidichnites and Plan-
olites. Cloud (1968) reexamined Walcott's specimens and described Hel-
minthoidichnites as "featureless spiral films"; he decided it was not a
trace but "probably algal." Walter et al. (1976) studied the microstructure
of Helminoidichnites and concluded that it was probably a macroscopic
eucaryotic algal filament, a body fossil preserved as a carbonized film
on bedding planes. Cloud consigned Walcott's Planolites specimens to
the category of "Algal?". Walter et al. rejected any organic origin for
these specimens and considered them to be isolated ripple marks and
concretions.
Byers (1976) examined Walcott's collecting localities in Montana and
found no trace fossils, nor even any hint of bioturbation in thousands of
meters of section. The Belt Supergroup spans an age of 1325-1100 m.y.
BP (Obradovich and Peterman, 1968); Walcott's specimens are from the
lower beds in the sequence, approximately twice as old as the Upper
Precambrian tillites and Bunyerichnus. Although Planolites is present in
rocks of Ediacaran age, the Belt occurrences can be discounted. The lack
of bioturbation in Belt shales suggests that even very small infaunal me-
tazoans, such as nematodes, had not yet evolved, much less a large coe-
lomic burrower capable of forming Planolites.
A few other trace occurrences somewhat older than the Late Precam-
brian glaciation have been reported in the literature. Palij (1974) described
Rugoinfractus, a bilobed trace 12 mm wide, from below the basal tillites
in the Ukraine. The sedimentary rocks containing this trace overlie ig-
neous rocks dated at 1100-1400 m.y. BP. This age would be a maximum;
the traces are in a sedimentary sequence containing the tillites, suggesting
an age closer to 700 m.y. than to the igneous date.
Faul (1950) described structures on bedding planes in the Ajibik

* Planolites is probably the biggest "wastebasket" ichnogenus extant, known from practi-
cally all environments throughout the Phanerozoic; this taxon is the convenient lodging
place of any nondescript horizontal burrow to which no more specific name can be attached.
Marine Biogenic Sedimentary Structures 251

Quartzite of Michigan, which he interpreted as burrows. The published

photographs show curving ropes of sand 3-5 mm wide and up to 15 cm
long. Faul reported a maximum length of 60 cm. No branching or internal
annulation was observable. The structures follow bedding planes but are
separated from the bed top by "a thin film of ferruginous material." Faul
gave a minimum age of 1200 m.y. for the age of the Ajibik Quartzite.
Similar curving marks and elongate sand tubes on bedding planes of
various rippled Precambrian quartzites were interpreted as burrows by
Frarey and McLaren (1963) and Hofmann (1967), but as mudcracks by
Wheeler and Quinlan (1951) and Barnes and Smith (1964). In 1971, Hof-
mann published a detailed analysis of this structure, including specimens
collected subsequent to his 1967 paper. One of the new specimens clearly
indicates a shrinkage crack origin, modified by compaction deformation
and removal (by dissolution?) of the clay layer. Although some specimens
retain patches or thin films of the mud layer that originally cracked, other
bedding planes contain only the curving sand tubes or spindles that filled
the cracks. It is the lack of preserved mud that gives these inorganic
structures their burrowlike appearance.
Clemmey (1976) reported burrows in sediments approximately 1000
m.y. old from Zambia. Cloud (1978) subsequently rejected Clemmey's
interpretation. The structures figured by Clemmey are undoubtedly bur-
rows, but their age is in question. Cloud suggested that the burrows are
Recent, produced by termites digging preferentially down strongly weath-
ered layers (in a steep-dipping section) from the lateritic soil zone.
Much older structures were described by Kauffman and Steidtmann
(1976) in a talk entitled "Are These the Oldest Metazoan Trace Fossils?".
The postulated traces are large, as big as Recent intertidal burrows. They
occur in a cross-bedded metaquartzite, 2 billion years old, in Wyoming.
Some of the various morphologies Kauffman and Steidtmann illustrated
do resemble burrows; others could be soft-sediment deformation features.
These structures deserve further study, but for the moment I would cer-
tainly retain the question mark in the title.
Durham (1978) recently reviewed the literature on Precambrian me-
tazoans and cited several of the cases described previously to support his
hypothesis of a very long history of metazoan evolution. Although geology
often depends on one bit of evidence to support a theoretical superstruc-
ture (an unconformity visible at a single locality implying regional uplift
and erosion; one unambiguous cross-cutting dike that gives the sequence
of intrusion of big plutons), we must be careful in applying this method
to the Precambrian fossil record. If one absolutely convincing occurrence
of Middle Precambrian burrows were found, it would radically constrain
phylogenetic models for the metazoa. I am convinced that we should
252 Chapter 5

follow Cloud's example and treat supposed traces of great age with skep-
ticism, requiring careful documentation that the structures are (1) biogenic
and (2) in situ, and that the rocks are (3) accurately dated.
The recurrent enigma inherent in the reports of trace fossils older
than 700 m.y. is why they are so scattered and so few in number. We are
dealing with fewer than ten possible occurrences over a time span of more
than a billion years (d. Durham, 1978).
It metazoan burrowers really did evolve in Middle Precambrian time,
why don't we see scads of traces? A glance at normal marine Phanerozoic
sediments will be rewarded with evidence of trace fossils or general bio-
turbation. Clemmey's (1976) comment is appropriate to the entire pre-
Ediacaran record: "The question remains of why, after being at such an
advanced stage of evolution ... the animals did not continue to flour-
ish .... " (p. 578).
As of this writing, it would appear that the oldest unambiguous trace
fossil is approximately coincident with the Upper Precambrian glacial
episode; metazoans began to leave their spoor about 700 m.y. ago.

References
Ager, D. v., and Wallace, P., 1970, The distribution and significance of trace fossils in the
uppermost jurassic rocks of the Boulonnais, northern France, in: Trace Fossils (T. P.
Crimes and j. C. Harper, eds.). pp. 1-18, Seel House Press, Liverpool.
Allen, j. R. L., 1965, Late Quaternary Niger Delta, and adjacent areas: Sedimentary environ-
ments and lithofacies, Am. Assoc. Pet. Ceol. Bull. 49:547-600.
Allen, j. R. 1., 1970, Physical Processes of Sedimentation, American Elsevier, New York,
248 pp.
Alpert, S. P., 1977, Trace fossils and the basal Cambrian boundary, in: Trace Fossils 2 (T.
P. Crimes and j. C. Harper, eds.). pp. 1-8, See I House Press, Liverpool.
Armentrout, j. M., 1980, Ophiomorpha from upper bathyal Eocene subsea fan facies, north-
western Washington (AbstL). Am. Assoc. Pet. Ceoi. Bull. 64:670-671.
Bambach, R. K., Kreisa, R. D., and Whitehurst, H. F., 1978, Storm reworked but untransported
faunal assemblages in Paleozoic age shelf environments, Ceoi. Soc. Am. Abstr. 10:362.
Banks, N. 1., 1970, Trace fossils from the late Precambrian and Lower Cambrian of Finnmark,
Norway, in: Trace Fossils (T. P. Crimes and j. C. Harper, eds.). pp. 19-32, Seel House
Press , Liverpool.
Barnes, A. G., and Smith, A. G., 1964, Some markings associated with ripple-marks from
the Proterozoic of North America, Nature 201:1018-1019.
Bassler, R. S., 1941, A supposed jellyfish from the Pre-Cambrian of the Grand Canyon, Proc.
U. S. Nat. Mus. 89:519-522.
Bennett, R. H., Bryant, W. R., and Keller, G. H., 1977, Clay fabric and geotechnical properties
of selected submarine sediment cores from the Mississippi Delta , Natl. Oceanic Atmos.
Adm. U. S. Prof. Pap. 9, 86 pp .
Bergstrom, j., 1970, Rusophycus as an indicator of early Cambrian age, in: Trace Fossils (T.
P. Crimes and j. C. Harper, eds.), pp. 35-42, Seel House Press, Liverpool.
Berner, R. A., 1971, Principles of Chemical Sedimentology, McGraw-Hili, New York, 240
pp.
Marine Biogenic Sedimentary Structures 253

Bowen, Z. P., Rhoads, D. C., and McAlester, A. L., 1974, Marine benthic communities in the
Upper Devonian of New York, Lethaia 7:93-120.
Bromley, R. G., 1975, Trace fossils at omission surfaces, in: The Study of Trace Fossils (R.
W. Frey, ed.), pp. 399-428, Springer-Verlag, New York.
Byers, C. W., 1973, Biogenic structures of black shale paleoenvironments, Ph.D. dissertation,
Yale University, New Haven, Connecticut, 134 pp.
Byers, C. W., 1974, Shale fissility: Relation to bioturbation, Sedimentology 21:479-484.
Byers, C. W., 1976, Bioturbation and the origin of the metazoans: Evidence from the Belt
Supergroup, Montana, Geology 4:565-567.
Byers, C. W., 1977a, Biofacies patterns in euxinic basins: A general model, in: Deep-water
Carbonate Environments (H. E. Cook and P. Enos, eds.), Soc. Econ. Paleontol. Mineral
Spec. Publ. 25:5-17.
Byers, C. W., 1977b, Ichnofacies and paleoenvironments in the Upper Cambrian sandstones
of Wisconsin (Abstr.), f. Paleontol. 51(Suppl. to No. 2):5.
Byers, C. W., 1980, Bioturbation as factor in hydrocarbon generation-Example from Mowry
Shale (Abstr.), Am. Assoc. Pet. Geol. Bull. 64:685.
Byers, C. W., and Dott, R. H., Jr., 1980, Depositional environments in Upper Cambrian Jordan
Sandstone in Wisconsin (Abstr.), Am. Assoc. Pet. Geol. Bull. 64:685.
Byers, C. W., and Stasko, L. E., 1978, Trace fossils and sedimentologic interpretation-
McGregor Member of Platteville Formation (Ordovician) of Wisconsin, f. Sediment Pe-
trol. 48:1303-1310.
Chamberlain, C. K., 1975, Trace fossils in DSDP cores of the Pacific, J. Paleontol.
49:1074-1096.
Clemmey, H., 1976, World's oldest animal traces, Nature 261:576-578.
Cloud, P. E., Jr., 1968, Pre-metazoan evolution and the origins of the metazoa, in: Evolution
and Environment (E. T. Drake, ed.), pp. 1-72, Yale University Press, New Haven.
Cloud, P. E., 1978, World's oldest animal traces, Nature 275:344.
Cluff, R. M., 1976, Paleoecology and depositional environment of the Mowry Shale (Albian),
Black Hills region, M.S. thesis, University of Wisconsin, Madison, 104 pp.
Cowie, J. W., and Glaessner, M. F., 1975, The Precambrian-Cambrian boundary: A sym-
posium, Earth Sci. Rev. 11:209-251.
Cowie, J. W., and Spencer, A. M., 1970, Trace fossils from the Late Precambrian/Lower
Cambrian of East Greenland, in: Trace Fossils (T. P. Crimes and J. C. Harper, eds.), pp.
91-100, Seel House Press, Liverpool.
Crimes, T. P., 1970, A facies analysis of the Cambrian of Wales, Palaeogeogr. Palaeoclimatol.
Palaeoecol. 7:113-170.
Crimes, T. P., 1975, The stratigraphical significance of trace fossils, in: The Study of Trace
Fossils (R. W. Frey, ed.), pp. 109-130, Springer-Verlag, New York.
Crimes, T. P., 1977, Trace fossils of an Eocene deep-sea sand fan, northern Spain, in: Trace
Fossils 2 (T. P. Crimes and J. C. Harper, eds.), pp. 71-90, Seel House Press, Liverpool.
Crimes, T. P., Legg, I., Marcos, A., and Arboleya, M., 1977, ?Late Precambrian-low Lower
Cambrian trace fossils from Spain, in: Trace Fossils 2 (T. P. Crimes and J. C. Harper,
eds.), pp. 91-138, Seel House Press, Liverpool.
.. Dapples, E. C., 1942, The effect of macro-organisms upon near-shore marine sediments, f.
Sediment. Petrol. 12:118-126.
Dawson, W. C., 1978, Improvement of sandstone porosity during bioturbation (Abstr.), Am.
Assoc. Pet. Geol. Bull. 62:508-509.
Delgado, D. J., 1980, Submarine diagenesis (aragonite dissolution, cementation by calcite,
and dolomitization in Ordovician Galena Group, Upper Mississippi Valley (Abstr.), Am.
Assoc. Pet. Geol. Bull. 64:697.
Driese, S. G., Byers, C. W., and Dott, R. H., Jr., 1981, Tidal deposition in the basal Upper
Cambrian Mt. Simon Formation in Wisconsin, f. Sediment. Petrol. 51:367-381.
254 Chapter 5

Durham, J. W., 1978, The probable metazoan biota of the Precambrian as indicated by the
subsequent record, Annu. Rev. Earth Planet. Sci. 6:21-42.
Ekdale, A. A., 1977, Abyssal trace fossils in worldwide Deep Sea Drilling Project cores, in:
Trace Fossils 2 (T. P. Crimes and J. C Harper, eds.). pp. 163-182, Seel House Press,
Liverpool.
Ekdale, A. A., 1978, Trace fossils in Leg 42A cores, Init. Rept. Deep Sea Drilling Project
XLII (Part 1):821-827.
Ekdale, A. A., 1980, Graphoglyptid burrows in modern deep-sea sediment, Science
207:304-306.
Ekdale, A. A., and Berger, W. H., 1978, Deep-sea ichnofacies: Modern organism traces on
and in pelagic carbonates of the western equatorial Pacific, Palaeogeogr. Palaeoclimatol.
Palaeoecol. 23:263-278.
Elders, C. A., 1975, Experimental approaches in neoichnology, in: The Study of Trace Fosslls
(R. W. Frey, ed.). pp. 513-536, Springer-Verlag, New York.
Faul, H., 1950, Fossil burrows from the Precambrian Ajibik Quartzite of Michigan, J. Pa-
leontol. 24:102-106.
Frarey, M. J., and McLaren, D. J., 1963, Possible metazoans from the Early Proterozoic of the
Canadian Shield, Nature 200:461-462.
Frey, R. W., 1970, Trace fossils of the Fort Hays limestone member of the Niobrara Chalk
(Upper Cretaceous). west-central Kansas Univ. Kans. Paleontol. Contrib. 53,41 pp.
Frey, R. W. (ed.). 1975, The Study of Trace Fossils, Springer-Verlag, New York, 562 pp.
Frey, R. W., 1978, Behavioral and ecological implications of trace fossils, in: Trace Fossil
Concepts (P. B. Basan, ed.). Soc. Econ. Paleontol. Mineral. Short Course 5:49-75.
Fursich, F. T., 1975, Trace fossils as environmental indicators in the Corallian of England
and Normandy, Lethaia 8:151-172.
Germs, G. J. B., 1972, New shelly fossils from Nama Group, South West Africa, Am. J. Sci.
272:752-761.
Germs, G. J. B., 1973, Possible sprigginid worm and a new trace fossil from the Nama Group,
South West Africa, Geology 1:69-70.
Glaessner, M. F., 1969, Trace fossils from the Precambrian and basal Cambrian, Lethaia
2:369-393.
Goldring, R., 1964, Trace-fossils and the sedimentary surface in shallow-water marine sed-
iments, in: Deltaic and Shallow Marine Deposits (1. M. J. U. van Straaten, ed.). pp.
136-143, Elsevier, New York.
Gould, S. J., 1980, The promise of paleobiology as a nomothetic, evolutionary discipline,
Paleobiology 6:96-118.
Hantzschel. W., 1975, Trace Fossils and Problematica, Treatise on Invertebrate Paleontology
(C. Teichert, ed.). Volume W, Miscellanea (Supplement 1). Kansas University Press,
Lawrence, 269 pp.
Hofmann, H. J., 1967, Precambrian fossils(?) near Elliot Lake, Ontario, Science 156:500-504.
Hofmann, H. J., 1971, Precambrian fossils, pseudofossils and problematica in Canada, Geol.
Surv. Can. Bull. 189, 146 p.
Howard, J. D., 1972, Trace fossils as criteria for recognizing shorelines in stratigraphic record,
in: Recognition of Ancient Sedimentary Environments (J. K. Rigby and W. K. Hamblin,
eds.). Soc. Econ. Paleontol. Mineral. Spec. Publ. 16:215-225.
Howard, J. D., 1975, The sedimentological significance of trace fossils, in: The Study of
Trace Fossils (R. W. Frey, ed.). pp. 131-146, Springer-Verlag, New York.
Howard, J. D., 1978, Sedimentology and trace fossils, in: Trace Fossil Concepts (P. B. Basan,
ed.). Soc. Econ. Paleontol. Mineral. Short Course 5:13-47.
Howard, J. D., and Elders, C. A., 1970, Burrowing patterns of haustoriid amphipods from
Sapelo Island, Georgia, in: Trace Fossils (T. P. Crimes and J. c. Harper, eds.). pp.
243-262, Seel House Press, Liverpool.
Marine Biogenic Sedimentary Structures 255

James, N. P., Kobluk, D. R, and Pemberton, S. G., 1977, The oldest macro borers: Lower
Cambrian of Labrador, Science 197:980-983.
Kauffman, E. G., and Steidtmann, J. R, 1976, Are these the oldest metazoan trace fossils?
Geo1. Soc. Am. Abstr. 8:947-948.
Kern, J. P., 1978, Trails from the Vienna Woods, Palaeogeogr. Palaeoclimato1. Palaeoecol.
23:231-262.
Kitchell, J. A., 1979, Deep-sea foraging pathways: An analysis of randomness and resource
exploitation, Paleobiology 5:107-125.
Kitchell, J. A., Kitchell, J. K., Clark, D. L., and Dangeard, L., 1978a, Deep-sea foraging behavior:
Its bathymetric potential in the fossil record, Science 200:1289-1291.
Kitchell, J. A., Kitchell, J. K., Johnson, G. L., and Hunkins, K. L., 1978b, Abyssal traces and
megafauna: Comparison of productivity, diversity and density in the Artic and Antarctic,
Paleobiology 4:171-180.
Ksiazkiewicz, M., 1970, Observations on the ichnofauna of the Polish Carpathians, in: Trace
Fossils (T. P. Crimes and J. C. Harper, eds.), pp. 283-322, Seel House Press, Liverpool.
Ksiazkiewicz, M., 1977, Trace fossils in the flysch of the Polish Carpathians, Paleonto1. Po1.
36:1-208.
Kuenen, Ph. H., 1950, Marine Geology, John Wiley and Sons, New York, 568 pp.
Larson, D. W., 1977, Paleoenvironment of the Mowry Shale Lower Cretaceous), western and
central Wyoming, as determined from biogenic structures, M.S. thesis, University of
Wisconsin, Madison, 121 pp.
Moon, C. F., 1972, The microstructure of clay sediments, Earth Sci. Rev. 8:303-321.
Moore, D. G., and Scruton, P. c., 1957, Minor internal structures of some Recent uncon-
solidated sediments, Am. Assoc. Pet. Geo1. Bull. 41:2723-2751.
Obradovich, J. D., and Peterman, Z. E., 1968, Geochronology of the Belt Series, Montana,
Can. J. Earth Sci. 5:737-747.
Osgood, R G., Jr., 1970, Trace fossils of the Cincinnati area, Palaeontogr. Am. VI(41):281-438.
Osgood, R G., Jr., and Szmuc, E. J., 1972, The trace fossil Zoophycos as an indicator of water
depth, Bul1. Am. Paleonto1. 62:1-22.
Palij, V. M., 1974, On finding of the trace fossil in the Riphean deposits of the Ovruch Ridge,
Rep. Acad. Sci. Ukr. SSR Ser. B. Geo1. Geophys. Chern. Bio1. Jahrg. 36:34-37 [in
Ukrainian].
Pettijohn, F. J., 1957, Sedimentary Rocks, Harper and Brothers, New York, 718 pp.
Porter, J., and Byers, C. W., 1979, Depositional environments of the Norwalk Member of the
Jordan Formation (Upper Cambrian), Southwest Wisconsin, Ceo1. Soc. Am. Abstr. 11:253
Purser, B. H., 1969, Syn-sedimentary marine lithification of Middle Jurassic limestones of
the Paris Basin, Sedimentology 12:205-230.
Reineck, H. E., 1967, Layered sediments of tidal flats, beaches, and shelf bottoms of the
North Sea, in: Estuaries (G. H. Lauff, ed.), pp. 191-206, American Association for the
Advancement of Science, Washington, D. C.
Rhoads, D. c., 1963, Rates of sediment reworking by Yoldia limatula in Buzzard's Bay,
Massachusetts and Long Island Sound, J. Sediment. Petro1. 33:723-727.
Rhoads, D. C., 1970, Mass properties, stability, and ecology of marine muds related to
burrowing activity, in: Trace Fossils (T. P. Crimes and J. C. Harper, eds.), pp. 391-406,
Seel House Press, Liverpool.
Rhoads, D. C., 1975, The paleoecologic and environmental significance of trace fossils, in:
The Study of Trace Fossils (R W. Frey, ed.), pp. 147-160, Springer-Verlag, New York.
Rhoads, D. C., and Morse, J. W., 1971, Evolutionary and ecologic significance of oxygen-
deficient marine basins, Lethaia 4:413-428.
Roniewicz, P., and Pienkowski, G., 1977, Trace fossils of the Podhale flysch basin, in: Trace
Fossils 2 (T. P. Crimes and J. C. Harper, eds.), pp. 273-288, Seel House Press,
Liverpool.
256 Chapter 5

Seilacher, A., 1956, Der Beginn des Kambriums als biologische Wende, Neues Jahrb. Geol.
Palaeontol. Abh. 103:155-180.
Seilacher, A., 1958, Zur okologischen Charakteristik von Flysch und Molasse, Eclogae Geol.
Helv. 51:1062-1078.
Seilacher, A., 1964, Biogenic sedimentary structures, in: Approaches to Paleoecology (J.
Imbrie and N. Newell, eds.). pp. 296-316, John Wiley and Sons, New York.
Seilacher, A., 1967, Bathymetry of trace fossils, Mar. Geol. 5:413-428.
Seilacher, A., 1974, Flysch trace fossils: Evolution of behavioural diversity in the deep-sea,
Neues Jahrb. Geol. Palaeontol. Monatsh. 4:233-245.
Seilacher, A., 1977a, Pattern analysis of Paleodictyon and related trace fossils, in: Trace
Fossils 2 (T. P. Crimes and J. C. Harper, eds.). pp. 289-334, Seel House Press, Liverpool.
Seilacher, A., 1977b, Evolution of trace fossil communities, in: Patterns of Evolution (A.
Hallam, ed.). pp. 359-376, Elsevier, New York.
Seilacher, A., 1978, Use of trace fossil assemblages for recognizing depositional environ-
ments, in: Trace Fossil Concepts (P. B. Basan, ed.). Soc. Econ. Paleontol. Mineral. Short
Course 5:185-201.
Shourd, M. 1., and Levin, H. 1., 1976, Chondrites in the upper Plattin Subgroup (Middle
Ordovician) of eastern Missouri, J. Paleontol. 50:260-268.
Simpson, S., 1957, On the trace fossil Chondrites, Q. J. Geol. Soc. London 112:475-499.
Stanley, S. M., 1976, Fossil data and the Precambrian-Cambrian evolutionary transition,
Am. J. Sci. 276:56-76.
Twenhofel, W. H., 1939, Environments of origin of black shales, Am. Assoc. Pet. Geol. Bull.
23:1178-1198.
Van Straaten, 1. M. J. U., 1959, Minor structures of some Recent littoral and neritic sediments,
Geol. Mijnbouw 21:197-216.
Walcott, C. D., 1899, Pre-Cambrian fossiliferous formations, Geol. Soc. Am. Bull. 10:199-244.
Walter, M. R., Oehler, J. H., and Oehler, D. Z., 1976, Megascopic algae 1,300 million years
old from the Belt Supergroup, Montana: A reinterpretation of Walcott's Helminthoid-
ichnites, J. Paleontol. 50:872-881.
Warme, J. E., and McHuron, E. J., 1978, Marine borers: Trace fossils and geological signif-
icance, in: Trace Fossil Concepts (P. B. Basan, ed.). Soc. Econ. Paleontol. Mineral. Short
Course 5:77-131.
Webby, B. D., 1970, Late Precambrian trace fossils from New South Wales, Lethaia 3:79-109.
Wheeler, H. E., and Quinlan, J. J., 1951, Pre-cambrian sinuous mud cracks from Idaho and
Montana, J. Sediment. Petrol. 21:141-146.
Young, F. G., 1972, Early Cambrian and older trace fossils from the Southern Cordillera of
Canada, Can. J. Earth Sci. 9:1-17.
Chapter 6
Geological Significance of Aquatic
Nonmarine Trace Fossils
MICHAEL]. S. TEVESZ and PETER 1. McCALL

1. Introduction ................................................................................................................. 257

2. Identification of Traces ............................................................................................... 259
3. Trophic Level Reconstruction .................................................................................... 270
4. Paleoenvironmental Reconstruction ........................................................................... 271
4.1. Environment of Deposition ................................................................................. 271
4.2. Vectorial Features ................................................................... .. ........................... 273
5. Discussion .................................................................................................................... 277
5.1. Usefulness of Systematics Information and Suggestions for Future Work ...... 277
5.2. Usefulness of Paleoenvironmental Information and Suggestions for Future
Work ..................................................................................................................... 278
6. Concluding Remarks .......................... .. ............... .................................................... .. .. 280
References ...................................................................................................... .. .. .......... 281

1. Introduction
In Chapters 3 and 4 it was shown that benthic invertebrates interact with
modern fluvial and lacustrine sediments and, through these interactions,
alter the physical, chemical, and biological properties of sediments. In
this chapter, the literature documenting invertebrate-sediment interac-
tions in ancient fluvial, lacustrine, and associated terrestrial environments
is reviewed in order to show that invertebrate activities influenced
benthic processes and properties in these environments in the past. The
literature concerning invertebrate traces of nonmarine origin is scant and
scattered and has not been the subject of recent review. Curran (1980)
recently emphasized the need for a review of this kind. We will attempt
to show that aquatic nonmarine trace fossils are abundant and widely
distributed (Table I) and are useful for a variety of geological purposes.

MICHAEL J. S. TEVESZ • Department of Geological Sciences, Cleveland State University,

Cleveland, Ohio 44115. PETER L. McCALL • Department of Geological Sciences. Case
Western Reserve University, Cleveland. Ohio 44106.
257
258 Chapter 6

Further research on these traces will probably improve the resolution of

paleoecological and paleoenvironmental reconstructions involving non-
marine rocks and increase the understanding of the origin and history of
freshwater life.

Table I. Examples of the Stratigraphic and Geographic Extent of Aquatic

Nonmarine and Associated Trace Fossils
Geologic Age North America Europe Other

Pleistocene Tarr (1935) Gibbard and Stuart

Banerjee (1973) (1974)
Ashley (1975)
Pliocene Link and Osborne
(1978)
Miocene Toots (1967) Riding (1979)
K. O. Stanley and
Fagerstrom (1974)
Oligocene Edwards (1975) Daley (1968)
Eocene Picard and High Truc (1978)
(1972a)
Ryder et 01. (1976)
(Early Tertiary)
Moussa (1966, 1968,
1970)
Brown (1934)
Surd am and Wolfbauer
(1975)
Eugster and Hardie
(1975)
Peterson (1976)
Paleocene Gilliland and LaRocque
(1952)
Cretaceous Hubert et 01. (1972)
Siemers (1970, 1971)
Jurassic Goldberg and
Friedman (1974)
(Israel)
Triassic Reinemund (1955) Klein (1962b) Turner (1978)
Sanders (1968) (South Africa)
Van Houten (1964) Bromley and
Klein (1962a) Asgaard (1972,
1979) (Greenland)
Webby (1970)
(Australia)
(Continued)
Aquatic Nonmarine Trace Fossils 259

Table I. (Continued)
Geologic Age North America Europe Other

Permian Hanley et a1. (1971) Barrett (1965)

Olson and Bolles (1975) (Antarctica)
White (1929) Van Dijk et a1.
(1978) (Permo-
Triassic, South
Africa)
Savage (1971) (Late
Carboniferous-
Early Permian,
South Africa)
Carboniferous Mutch (1968) R. E. Elliott (1968) Rattigan (1967)
Hesse and Reading T. Elliott (1976) (Australia)
(1978) Seilacher (1963) Glaessner (1957)
D. J. Stanley (1968) Eagar (1974) (Australia)
Belt (1968) Chisholm (1970)
Devonian Allen (1970) Allen (1964, 1970) Conolly (1965)
Cant and Walker (1976) Rayner (1963) (Australia)
Allen and Friend (1968) Read and Johnson Friend (1965)
Berg (1977) (1967) (Spitzbergen)
Miller (1979) Trewin (1976) Webby (1968)
(Antarctica)
Cambro-Ordovician Selley (1972)
(Jordan)
Selley (1970)
("Lower
Paleozoic,
Jordan")

2. Identification of Traces
One major goal of the study of aquatic nonmarine trace fossils is to
classify a trace in its proper ichnotaxon and then determine the identity
of the trace-making organism. This exercise is important because it often
provides information on the taxonomic compostion and richness of the
fauna, both of which may be useful tools for inferring benthic processes
and properties.
Table II shows that nonmarine trace fossils made by invertebrates are
taxonomically diverse and environmentally widespread. Common trace-
makers in these environments include molluscs, insect larvae and adults,
annelids, and nematodes. The trace-makers were identified by different
means. One commonly employed method was for workers to use direct
observations of the trace-making activities of modern organisms. For ex-
 Table II. Examples of Taxonomic, Lithologic, and Paleoenvironmental Diversity of Aquatic Nonmarine and Associated N
Trace Fossils C':)
Q
Author(s) Ichnogenus Trace-maker Lithology Paleoenvironment

Allen and Friend (1968) d. Chondrites "Root marks" (Barrell, Sandstone Deltaic
1913, p. 461), "worm
burrows" (Dyson, 1963,
p.25)
Arthropod tracks Deltaic
Belt (1968) d. Cruziana Arthropods larger than
ostracods and
conchostracans
Berg (1977) Archanodon (Mollusca: Sandstone Fluvial
Bivalvia)
Bromley and Asgaard See Bromley and Asgaard Notostracan branchiopods Siltstone
(1972) (1979) for discussion of (Arthropoda)
Cruziana taxonomy.
Merostomichnites Notostracan branchiopods Siltstone?
triassicus Linck (Arthropoda)
Planolites regulosus Insect larvae? Siltstone?
Reineck

Bromley and Asgaard Arenicolites sp . Annelida Sandstone intercalations Lacustrine

(1979)
Pelecypodichnus Bivalves (Mollusca) Sandstone Lacustrine
amygdaloides Seilacher
Skolithos spp. Terrestrial insect burrows Sandstone Dessicated fluvial
(environment)
?Margaritichnus Sandstone n
Scoyenia gracilis White Insects? Sandstone Very shallow lacustrine ::r-
oo
Annelids? ~
co
...,
Steinichnus corJsbergi Terrestrial insect Sandstone "Terrestrial suite" cr>
Bromley and Asgaard
 Fuersichnus communis Sandstone Lacustrine
Bromley and Asgaard >-
.c
~
Diplichnites triassicus Arthropod Sandstone Aquatic? ~
Linck C=;'
Rusophycus eutendorfensis Notostracan crustaceans Sandstone? " Aquatic suite" Z
0
;:I
(Linck) (Arthropoda) 3
QJ
Cruziana problematica Arthropod Sandstone? "Aquatic suite" ..,
(Schindewolf) 5'
ro
Cruziana spp , Arthropod Sandstone? "Aquatic suite" ...,
..,
QJ
n
Brown (1934) Celliforma spirifer Brown Larval chamber of mining Preserved as calcite with Lacustrine ro
bees or without greenish 'TI
0
Ul
clay ~,
Cii
Chisholm (1970) Planolites Siltstones and sandstones Nonmarine
Monocraterion Argillites Nonmarine
Eagar (1974) Pelecypodichnus Carbonicola (Mollusca : Sandstone. siltstone. Nonmarine
Bivalvia) flagstone. shaley
flagstones
Planolites Shaley flagstones Nonmarine
Arenicolites Crossbedded. ripple- Nonmarine
marked flagstones
Eugster and Hardie Insect larvae burrows Oil shale facies Lacustrine
(1975)
Gibbard and Stuart Isopod Laminated silty clays Proglacial lake
(1974)
Arthropod Laminated silty clays Proglacial lake
Crustacean Laminated silty clays Proglacial lake
Gastropod Laminated silty clays Proglacial lake
Wormlike animal Laminated silty clays Proglacial lake
Insect larvae Laminated silty clays Pro glacial lake
Gilliland and LaRocque Xenohelix Limestone Lacustrine? N
(1952) 0')
~
(Continued)
 Table II. (Continued) N
~
N
Author{s) Ichnogenus Trace-maker Lithology Paleoenvironment

Glaessner (1957) Isopodichnus osbornei Arthropod Shales Periglacial lake

Glaessner [but refer to
the taxonomic revision
of Bromley and Asgaard
(1979)]
Hanley et 01. (1971) Mole beetles? [but referred Sandstone Subenvironment of
to as "mole crickets" by coastal region? Fluvial
Ratcliffe and Fagerstrom sub environment?
(1980)]
Hesse and Reading (1978) Limulid arthropods Shale Lacustrine

Miller (1979) ?Chondrites Alluvial

Planolites sp. Alluvial
Skolithos sp. Alluvial
Spirophyton Alluvial

Moussa (1970) Nematodes (see also Tarr, Limestone Lacustrine

1935)
Di ptera larvae Limestone Lacustrine

Olson and Bolles (1975) Large crustaceans Sandstone? Lacustrine

Peterson (1976) d. Arenicolites Sandstone Freshwater delta

d. Thalassinoides Sandstone Freshwater delta
d. Bifungites Sandstone Freshwater delta

Rattigan (1967) Isopodichnus osbornei Arthropods Silts Aqueoglacial

Glaessner [but refer to n
::r
po
the taxonomic revision
~
of Bromley and Asgaard ...,CD
(1979)] C'l
Riding (1979) ?Hydrobia (gastropod) Algal cones, in micritic Lacustrine »
.0
fecal pellets crust ~
~
Savage (1971) Urnfolozia sinuosa Savage Syncarid or peracarid Very fine-grained varvite Periglacial lakes n'
crustaceans rocks Z
0
~
Diplichnites govenderi Syncarid or peracarid Very fine-grained varvite Periglacial lakes a
Savage crustaceans rocks ...,OJ
Diplichnites sp. Sync arid or peracarid Very fine-grained varvite Periglacial lakes S·
CD
crustaceans rocks ...,>-l
OJ
Protichnites spp. Syncarid or peracarid Very fine-grained varvite Periglacial lakes n
CD
crustaceans rocks >rj
0
Gyrochorte Sync arid or peracarid Very fine-grained varvite Periglacial lakes V>

crustaceans rocks V>

Isopodichnus Syncarid or peracarid Very fine-grained varvite Periglacial lakes =
crustaceans rocks
Kingella natalensis Savage Syncarid or peracarid Very fine-grained varvite Periglacial lakes
crustaceans rocks
Gluckstadtella coo peri Syncarid or peracarid Very fine-grained varvite Periglacial lakes
Savage crustaceans rocks
Selley (1970) Cruziana furcifera Shale Fluvial; channel
d'Orbignya infillings
Cruziana goldfussi Shale Fluvial; channel
(Roualt)" infillings
Two distinct types of Silts, fine sands Fluvial; channel floor
Cruziana
d. Diplichnites b Silts, fine sands Fluvial; channel floor
d. Ichnyspica b Silts, fine sands Fluvial; channel floor
d. Incisifex b Silts, fine sands Fluvial; channel floor
d. Tasrnanadia b Silts, fine sands Fluvial; channel floor
d. Merostornichnites b Silts, fine sands Fluvial; channel floor
Selley (1972) Cruziana Sandstone Abandoned channel
N
(Continued) CO)
W
 Table II. (Continued) N
0)
~
Author(s) Ichnogenus Trace-maker Lithology Paleoenvironment

Siemers (1970) Planolites Mudstone

D. J. Stanley and Beetles Sand, ash lenses Braided fluvial

Fagerstrom (1974)
Toots (1967) Taenidium Sandstone Shallow lacustrine or
paludal
Trewin (1976) Isopodichnus stromnessi Arthropod Various positions with Temporary lacustrine
Trewin respect to sand-mud
interface
Turner (1978) Scolicia Gastropods or bivalves Sandstone Fluvial
Van Dijk et a1. (1978) Paired arthropod tracks Siltstones, mudstones Lacustrine shelf facies
Scolicia Silty sands Fluvio-Iacustrine offlap
facies
Arthropod tracks Silty sands Fluvio-Iacustrine offlap
facies
Arthropod tracks Siltstones Bay facies
Gastropod trails Playa lake facies
Arthropod tracks Playa lake facies
Webby (1968) Skolithos Sandstone Shallow marine to
littoral. or fluviatile to
littoral
Tigillites Sandstone Shallow marine to
littoral. or fluviatile to
littoral
Cylindricum Sandstone Shallow marine to
n
::r-
oo
littoral. or fluviatile to ~
littoral ...,CD
rn
 " Scolicia" Sandstone Shallow marine to ::>
.0
littoral, or fluviatile to C
~
littoral C=;'
Beaconites Sandstone Shallow marine to Z
o
littoral . or fluviatile to i:l
littoral a
OJ
...
Webby (1970) Brookvalichnus obliquus Wormlike organisms or Silty shale Lacustrine 5'
CD
Webby insect larvae >-l
...OJ
C"l
CD
White (1929) Scoyenia gracilis White Worms? Fluviatile? 'Tj
Walpia hermitensis White d, Worms or crustaceans Fluviatile? oen
::
en
a See Bender (1963 . 1968),
b See Hiintzschel (1962) . per instructions of Selley (1970),

N
C)

'"
266 Chapter 6

ample, in order to determine the kind of organisms that produced the

Cruziana probJematica in Triassic freshwater sediments from Greenland
(Fig. 1), Bromley and Asgaard (1972) made in situ observations on the
trace-making activity of the Recent freshwater notostracan branchiopod
Lepidurus arcticus. Because of the similarities of the ancient and modern
traces, they concluded that the Cruziana were made by a notostracan like
modern Lepidurus.
Moussa (1966, 1968, 1970) published a series of papers in an attempt
to identify wavelike trails from Eocene lacustrine limestones from Utah.
He was finally able to identify the trace-maker when he made in situ
observations on flood-deposited sediments still covered by a thin film of
water and discovered modern nematodes making traces similar to the
fossil trails (Fig. 2).
Brown (1934) provided a detailed description of Celliforma spirifer,
a cylindrical mold capped by a low detral spiral (average length and width
2.7 x 1.2 cm) from Eocene lacustrine deposits of Wyoming (Fig. 3). After
dismissing a few possible explanations of its origin (e.g., bivalve burrows,
insect eggs), he demonstrated that the structure was similar to the larval

Figure 1. Cruziana problematica (Schindewolf, 1921). From Bromley and Asgaard (1979).
Reprinted with the permission of Elsevier Scientific Publishing Co. (Amsterdam). R. Brom-
ley, and U. Asgaard.
Aquatic Nonmarine Trace Fossils 267

Figure 2. Trails from the Green River Formation. Wavelike trails are nematode trails; insect
larvae may have made the irregular-line trails. From Moussa (1970). Reprinted with the
permission of the Society of Economic Paleontologists and Mineralogists.
268 Chapter 6

2cm

Figure 3. Celliforma spirifer Brown: fossil larval chambers of mining bees. Redrawn from
Brown (1934).

chamber of modern mining bees. Although mining bees are terrestrial, he

cited several possible mechanisms (p. 535: " ... faulting, land slipping,
subsidence, climatic change, collapse of a sink .... ") that could have
brought the chambers to their site of entombment in freshwater sediments.
Hanley et a1. (1971), K. O. Stanley and Fagerstrom (1974), and Turner
(1978) also made extensive use of in situ observations of living analogues
to identify trace-makers.
Other writers such as Glaessner (1957), Rattigan (1967), Trewin
(1976), and Bromley and Asgaard (1979) made their identifications of
trace-making organisms on the basis of available literature. Except for
Bromley and Asgaard (1979), the cited literature largely concerned marine
forms. Use of this marine-based literature provided identifications of non-
marine organisms only at very high taxonomic levels.
There are several examples of ichnotaxa whose origin is unknown.
For example, Gilliland and LaRocque (1952) described a helical "burrow"
from the flagstaff Limestone (Paleocene) of Utah that they tentatively
assigned to the new genus Xenohe1ix (Fig. 4). Their problem in un-being
able to identify the trace-maker was typical: they could not find a modern
analogue of the trace (ef. Olson and Bolles, 1975).
A few (mostly old) papers discuss at length the taxonomy of certain
nonmarine trace fossils. For example, Schindewolf (1928) and Linck
(1942) go into the interpretation of "Isopodichnus" in detail. Reineck
(1955) provides a sound discussion of Planolites rugu10sus (= Scoyenia
Aquatic Nonmarine Trace Fossils 269

,
'I

1cm

Figure 4. Xenohelix utahensis Gilliland and LaRocque 1952 (external view of holotype).
From Gilliland and LaRocque (1952). Reprinted with the permission of the Society of Eco-
nomic Paleontologists and Mineralogists.
270 Chapter 6

gracilis). Examples of more current discussions of the taxonomy of non-

marine traces include Trewin (1976) and Bromley and Asgaard (1979).

3. Trophic Level Reconstruction

Toots (1967) described cylindrical burrows, 11-13 mm in diameter
with meniscate internal structure, from the Miocene Sheep Creek for-
mation of Wyoming (Fig. 5). The burrows were associated with fluviatile
sediments and were referred to as "Taenidium." Toots suggested that the
meniscate internal structure of the burrows was the result of backfilling
of the burrows by sediment as a result of the deposit-feeding activities of
the organism. Several other workers, including Daley (1968), Hanley et

/:-.. ..'
~ "
,
.. .. "
"
. ,,+, '.

"

.
':.', -
.=!-:.
I.

,
.-...
• • •

2cm
Figure 5. Taenidium: burrow of a deposit-feeding organism as indicated by meniscate in-
filling. Redrawn from Toots (1967).
Aquatic Nonmarine Trace Fossils 271

1cm

Figure 6. Cross-section of a burrow made by a deposit feeder. Redrawn from Rayner (1963).

a1. (1971), K. O. Stanley and Fagerstrom (1974), Edwards (1975), and

Bromley and Asgaard (1979), also used meniscate burrow structure as
evidence of deposit feeding.
Rayner (1963) employed another aspect of burrow structure for in-
ferring former deposit-feeding activity in freshwater sediments. By ana-
lyzing burrow cross-sections in Devonian lacustrine deposits from Scot-
land, she concluded from grain-size differences (coarse exterior, fine
interior) (Fig. 6) that the burrows were filled by sediment that had passed
through the gut of the organism.

4. Paleoenvironmental Reconstruction
4.1. Environment of Deposition

Seilacher was one of the first researchers to use trace fossil assem-
blages as major clues to the nature of the depositional environment. In
a 1963 paper, he showed that a trace fossil assemblage, called the Scoyenia
association, was useful in recognizing nonmarine environments in gen-
eral. In a 1978 paper he summarized some of his thoughts on this asso-
ciation (p. 188):
The real problem, however, is the recognition of non-marine aquatic en-
vironments. In red-bed sequences of different ages we meet a lowly diverse
association of rather small trace fossils, for which the name Scoyenia has been
proposed. Unfortunately, none of the elements of this ichnofacies is by itself
a reliable indicator (see Chamberlain, 1975, for examples of modern fresh water
traces).
Scoyenia is a backstuffed, cylindrical.burrow probably produced by insect
larvae, but similar burrows (e.g., Planolites) may be produced by other orga-
nisms, including marine ones. The coffeebean trace Isopodichnus and asso-
ciated biserial tracks can be assigned to non-marine phyllopod shrimp, but they
are difficult to distinguish from the burrows and tracks of small trilobites
272 Chapter 6

(Trewin, 1976), which reduces their indicator value except in post-Paleozoic

sediments.
The third element is tracks of small Iimulid and merostome arthropods.
These animals are primarily marine, but tolerate low-salinity conditions. The
fact that their tracks are mainly found in non-marine deposits is a preservational
bias. Varved, silty sediments, which favor the formation and preservation of
recognizable undertracks, are deposited more frequently in smaller non-marine
basins than in the marine realm.
Nevertheless, the Scoyenia association can be used in conjunction with
other sedimentological criteria to define environments in the terrestrial realm.

Seilacher's Scoyenia association is illustrated here in Fig. 7. Another more

general way of recognizing nonmarine deposits was proposed by Frey
(1978), who suggested using trace fossil diversity gradients.
Bromley and Asgaard (1979) used trace fossils to help distinguish
between aquatic and terrestrial nonmarine deposits. They found that trace
fossils formed in sediments exposed to air have three major sets of char-
acteristics that are absent in subaqueously produced forms. In the terres-
trial forms, the burrow walls are sharply defined and striated as a result
of the organisms' burrowing into a stiff substratum. Second, the burrow
filling has a vuggy fabric that would collapse underwater. While this
vuggy material may often block the burrow mouth, the lower end of the
burrow may contain no sediment. Voids within the burrow may be filled
with calcitic cement, and sometimes the burrows are deformed by com-
paction. Finally, mudcracks are always associated with these traces, and

A B c
I
/ 0 I
J
- -
00
-- .'.//

-f
"
-/-
1·
oJ
I

Figure 7. Scoyenia ichnofacies. (A) Phyllopod tracks and burrows; (B) Merostome tracks;
(C) Scoyenia. Redrawn from Seilacher (1963).
Aquatic Nonmarine Trace Fossils 273

burrows commonly cross or follow the fill of the cracks. Where the crack
was empty when the animal entered, both the burrow and the crack will
contain similar fill. On the same general topic, Fagerstrom and Ratcliffe
(1975) observed that the density of Recent insect-produced sediment
traces appears to be considerably higher in moist floodplain substrates
than in uplands.
Trace fossils have also been used to distinguish among different
aquatic nonmarine environments. Some of the most successful of these
studies matched up trace fossil assemblages with analogous modern fresh-
water traces and used the degree of similarity to determine the environ-
ment of deposition. Two of the best examples of studies that used this
approach are now summarized.
Hanley et a1. (1971) used fluviatile trace fossils to determine the
environment of deposition of cross-stratified units from the Permian Cas-
per Sandstone, southern Laramie Basin, Wyoming and Colorado. Before
this study, most workers had supposed that the cross-stratified beds orig-
inated in a nonaquatic environment. The absence of body fossils was an
important reason for the uncertainty. Further searching by Hanley's group
uncovered trace fossils that had modern analogues on a beach of the
Seminole River, Wyoming (Fig. 8). The modern traces were made by mole
beetles when low water exposed the sand but still left it moist. Based on
this observation and the reasoning that the ridgelike trace fossils probably
would not have been preserved in loose sand, Hanley et a1. (1971) con-
cluded that the Casper Sandstone had formed in a similarly moist
environment.
K. O. Stanley and Fagerstrom (1974) did a study involving nonmarine
trace fossils in braided river deposits from the Miocene of Nebraska. They
described a trace fossil assemblage that they interpreted as consisting of
vertical shelter burrows, horizontal deposit-feeding burrows, bioturbated
layers, and vertical passageways between bioturbated layers. Then they
compared these Miocene burrows with burrow populations formed by
modern beetles and their larvae in Platte River, Nebraska, sediments. They
found that these modern burrows were only formed where the river sed-
iment periodically became subaerially exposed. Finally, they used this
information from the Platte River as a basis for inferring a similar genesis
for the Miocene burrows.

4.2. Vectorial Features

Bromley and Asgaard (1972) observed that the modern notostracan

branchiopod Lepidurus is rheotactic, and on the basis of this observation
Figure 8. Ribbonlike burrows (A) on beach at Seminole Reservoir. Wyoming; possible modern
counterparts for trace fossils pictured in B. From Hanley et oJ. (1971). Reprinted with the
permission of the Society of Economic Paleontologists and Mineralogists.
Aquatic Nonmarine Trace Fossils 275

" .
.
~ ..
.. :"o.~'"
.
.... ~

~. 3cm
",
.,

Figure 9. Current crescents developed around the infillings of vertical burrows. Redrawn
from Allen and Friend (1968).

suggested that the orientation of notostracan trace fossils (certain Cru-

ziana) could serve as clues to paleocurrent direction. Another way of
determining paleocurrent direction from nonmarine trace fossils is sug-
gested from observations by Allen and Friend (1968) . These authors il-
lustrated current crescents that formed around the resistant infillings of
vertical burrows from the Devonian Catskill facies (Fig. 9). Thus paleo-
current direction could be determined by plotting the orientation of the
convex portion of the crescent.
Thoms and Berg (1974) inferred that burrow structures from a deltaic
paleoenvironment from the Devonian of New York were made by upward
escape from burial by the nonmarine bivalve Archanodon. They needed
to find and study a Recent analogue of the burrow-producing organism
in order to use the traces in paleoenvironmental reconstruction, and, after
field and laboratory observations , decided that studies of modern Mar-
garitifera margaritifera would provide the needed information. Earlier,
Berg (1973) had used the bivalve Siliqua as an analogue. Because Siliqua
burrows in an oceanward direction, Berg concluded that burrow curvature
and hinge impressions found in the burrows could be useful vectorial
features in paleoenvironmental analysis. Based on more studies of Recent
analogues, Berg (1977) further advanced the idea of the usefulness of
Archanodon in determining paleocurrent direction. Eagar (1974) provides
good illustrations of Paleozoic nonmarine bivalve burrows (Fig. 10).
 N
"'-J
~

Scm
n
::r
OJ

Figure 10. Sectional views of Pelecypodichnus burrows. From Eagar (1974). ';:l.
co
....
Reprinted with the permission of the Editor, Lethaia.
'"
Aquatic Nonmarine Trace Fossils 277

5. Discussion
5.1. Usefulness of Systematics Information and Suggestions for
Future Work

Diverse ichnotaxa occur in nonmarine rocks; these fossils represent

a record of several major invertebrate phyla. It is evident that the available
information on the systematics of both nonmarine trace fossils and trace-
makers is limited, and that there is much room for more research. We
think that this research needs to be done on two complementary subjects:
the trace fossils themselves and modern traces in aquatic nonmarine
sediments.
Taxonomic resolution is poor when marine analogues are used to
identify nonmarine trace-makers, and identification of trace-makers is
greatly impeded when modern analogues of an ichnotaxon are unknown.
This all suggests that the ichnology of modern sediments should be much
more extensively studied. Such studies have recently begun to appear.
The most important include Chamberlain (1975), Wells (1977), and Rat-
cliffe and Fagerstrom (1980). Chamberlain provides a descriptive cata-
logue of traces from aquatic nonmarine environments. One important
aspect of the article is that it emphasizes the wealth of nonmarine traces
that are potentially preservable, and therefore serves as a stimulus for
paleo ecologists to look for them. The Wells thesis describes freshwater
invertebrate traces from the Mississippi alluvial valley and defines rela-
tionships between trace fossil assemblages and subenvironments, sug-
gesting their potential usefulness in paleoenvironmental reconstructions.
Like Chamberlain, Wells demonstrates that diverse invertebrate traces are
produced in modern freshwater sediments. Ratcliffe and Fagerstrom
(1980) also emphasize this wealth of potentially fossilizable information,
but point out some problems in interpreting it: for example, although the
taxonomic diversity of sediment-disturbing spiders and insects on flood-
plains is great, they found convergence in burrow form among taxon-
omically dissimilar groups (d. Wells, 1977). Also, within the same spe-
cies, ontogenetic stage and environmental factors such as substrate
conditioning and weather may further affect burrow form. Therefore, it
is difficult to infer taxonomic diversity from trace fossil diversity.
By comparing the literature on modern trace-makers with the liter-
ature on their trace fossil counterparts, it is obvious that the great majority
of traces identified in modern sediments have yet to be described from
the fossil record. Thus there is clearly a need for more careful ichnological
study of nonmarine rocks. The trace fossils are there to study. We agree
with the sentiments of Howard (1978, p. 15) that aquatic nonmarine ich-
ology has suffered more from lack of interest than from lack of fossils:
278 Chapter 6

"Traces are not always abundant in non-marine units. However, as more

and more geologists include trace fossils on their check-off lists, there are
more and more reports of fluvial and lacustrine traces."
Aquatic nonmarine environments are inhabited today by diverse and
abundant invertebrates. For example, Chamberlain (1975, p. 433, Table
19.1) lists approximately 30 groups (phyla, classes, or orders) that inhabit
freshwater environments. Nevertheless, the body fossil record of non-
marine invertebrates is not very impressive. The only invertebrate body
fossils that are usually referred to as being "abundant" or "well preserved"
in nonmarine deposits are bivalves, gastropods, and ostracods (see also
Picard and High, 1972b, Table 2). But several papers that mentioned the
paucity of body fossils also mentioned an abundance of trace fossils (K.
O. Stanley and Fagerstrom, 1974; also D. J. Stanley, 1968; Reinemund,
1955; Conolly, 1965; Bromley and Asgaard, 1979). Thus, in several cases,
trace fossils are both better preserved and more abundant than body fos-
sils. It would therefore appear that further studies of trace fossils will add
much new information concerning the taxonomic nature and distribution
of organisms inhabiting aquatic nonmarine environments.

5.2. Usefulness of Paleoenvironmental Information and

Suggestions for Future Work

It has been shown previously that some nonmarine trace fossil as-
semblages are facies-specific (see also Toots, 1975). This facies specificity
makes them a very useful tool for geologists because nonmarine, partic-
ularly freshwater, depositional environments often exhibit rapid lateral
facies changes. Correlation across facies boundaries may be difficult, par-
ticularly if the fossils are not diagnostic of environment (Picard and High,
1972a). Because most workers have relied on body fossils for these pur-
poses, it has often been extremely difficult to define particular paleoen-
vironments within nonmarine deposits because body fossils are frequently
rare or poorly preserved. But since trace fossils may be abundant where
body fossils are not, further study of these trace fossils may provide an
additional useful tool for more rapidly and accurately defining paleoen-
vironments in nonmarine deposits. While this would obviously help strat-
igraphers and paleo ecologists with their work, there would be benefits
for other groups of scientists as well. For example, paleoclimates and
climatic changes are often reconstructed based on the extent of ancient
lake deposits (Grove et aI., 1975). Since certain trace fossil assemblages
are well represented in ancient lakes (Table II), these fossils could be used
by paleoclimatologists to help them map the extent of these lakes.
The limits of resolution of aquatic nonmarine trace fossils in delim-
Aquatic Nonmarine Trace Fossils 279

iting depositional environments are suggested by Wells (1977). He found

that his Recent Mississippi Valley assemblages could not be used to de-
limit subenvironments on the scale of point bar, ridge and swale, levee
or backswamp. But he felt that the assemblages could at least be used to
distinguish marine from nonmarine environments on a regional scale.
Fine zonations were not preserved because of the limited temporal and
spatial extent of the distribution-controlling factors plus the fluctuation
of shorelines. This would lead to a fossil assemblage that would be time-
averaged throughout the whole of the alluvial deposits.
Most of the papers that have used aquatic nonmarine trace fossils in
environmental reconstructions employ them to define the general envi-
ronment of deposition, as indicated previously. Far less attention has
been paid to them as tools for reconstructing specific environmental pa-
rameters, such as mass properties of the bottom, sediment organic content,
and pore water chemistry, within a general environment. Judging from
studies based in the marine realm (see Chapter 5), it is clear that trace
fossils are useful for such purposes. Nevertheless, we found only one
paper (Moussa, 1970) that used nonmarine trace fossils in this context.
In it, Moussa showed that modern sinusoidal nematode trails form under
an extremely limited set of environmental conditions: they form only in
moist sediments covered by a film of water that is thinner than the animal.
Consequently, the sinusoidal trails he found in Eocene lacustrine deposits
precisely delimited these aspects of a portion of the depositional envi-
ronment. But Moussa is practically alone in his detailed work, and these
high-resolution reconstructions therefore are another promising subject
for aquatic nonmarine ichnological research.
But do distinctive trace fossil assemblages characterize a sufficiently
broad range of environments to be generally useful for the several pur-
poses discussed in this section? Not everyone thinks so. Turner (1978),
for one, believes that trace fossils are scarce in fluvial sediments of Pre-
Cretaceous age. Nevertheless, we found several references to trace fossils
being associated with Paleozoic fluvial deposits (e.g., D. J. Stanley, 1968;
Friend, 1955; Cant and Walker, 1976; Allen and Friend, 1968; Allen,
1970). Moreover, Turner says that trace fossils tend to be more common
in finer-grained sediments. While this is generally true, it should be
pointed out that they also occur in coarser deposits such as sands (e.g.,
Toots, 1967) and even conglomerates (e.g., Dineley and Williams, 1968)
(see also Table II). In addition, while workers like K. O. Stanley and
Fagerstrom (1974) and Hanley et a1. (1971) showed that some trace fossil
assemblages in nonmarine environments formed when the sediments be-
came subaerially exposed, works such as those by Chamberlain (1975),
McCall et a1. (1979), and Tevesz et a1. (1980) showed that extensive trace-
making activity by freshwater invertebrates occurs under subaqueous con-
280 Chapter 6

ditions in a variety of environments. These latter observations suggest

that the new paleontological information that trace fossils could provide
is potentially broad in its environmental scope.

6. Concluding Remarks
It is apparent that trace fossils produced by aquatic nonmarine in-
vertebrates are important sources of information that are still not widely
used by geologists. There is room for much further research on trace-fossil
and trace-maker systematics and the paleoecological significance of these
fossils. It is hoped that this review has served to organize a large part of
the interpretive literature for prospective researchers.
Even cursory notice of aquatic nonmarine trace fossils can be ex-
tremely useful to geologists. Truc (1978), for instance, studied evaporitic
sequences from an ancient saline lake from the Paleogene of the Mor-
moiron Basin, France, and noted that
Over the whole of the evaporitic area, well exposed bedding surfaces of
both the gypsum and green marls show mud-cracks and are covered with
footprints of birds and the trails of animals. The marls and gypsum beds are
also extensively burrowed. This can explain the abundance of bird footprints
since the birds (probably waders) would have been feeding on the infaunal
annelids and insect grubs responsible for the bioturbation. The presence of an
infauna might indicate that the formation of gypsum was not the result of
excessive salinities.

In this brief statement, Truc employs these trace fossils to infer the tax-
onomic identity of the organisms making them, reconstruct predator-prey
relationships, explain trace fossil abundance, and infer chemical condi-
tions that lead to the formation of gypsum.
Finally, it is our opinion that geologists effecting this research should
establish nonmarine ichnological studies on a basis that is separate and
distinct from marine ichnology. It is tempting to do otherwise, because
there is a great variety of interpretive literature concerning marine ich-
nology to serve as a guide, and, moreover, there are similar-appearing
traces common to both realms.
But major differences between the taxonomic composition and func-
tional associations in marine and nonmarine communities indicate that
nonmarine environments merit separate study. For example, insects are
perhaps the most common producers of nonmarine traces and trace fossils,
but are vastly different from marine arthropods and are virtually absent
in the marine realm. Tubificid oligochaetes are another group that com-
monly produce traces in freshwater sediment but are uncommon trace-
makers in the oceans, probably because marine tubificids are as a rule
Aquatic Nonmarine Trace Fossils 281

smaller than their freshwater relatives. Thus similar-appearing traces in

nonmarine and marine settings may have vastly different "meanings"
because of the different life habits and adaptive needs of the organisms
producing them.
There may also be differences in the functional associations of or-
ganisms that inhabit the two realms which would produce distinctive
sedimentary effects in each. For instance, filter-feeding and deposit-feed-
ing organisms have an important impact on sediments. Moreover, the net
effect of the activities of deposit-feeding organisms in muddy substrata
in the marine realm has been to exclude large suspension-feeding orga-
nisms. This concept is termed trophic group amensa1ism (e.g., Rhoads
and Young, 1971; Aller and Dodge, 1974). But in nonmarine aquatic en-
vironments, the net effect of deposit feeders on suspension feeders is less
clear. Oligochaetes, for example, co-occur with large suspension-feeding
bivalves in various environments (McCall et a1., 1979; Tevesz et a1., 1980),
including those with muddy substrata. This unexpected association may
be caused by freshwater suspension feeders having an expanded niche
owing to lower competition/predation pressure than that to which their
counterparts in the ocean are subjected. Thus this kind of amensalistic
interaction may not be as important in structuring communities in aquatic
nonmarine environments (Tevesz and McCall, 1978, 1979). Since deposit
feeders and large suspension feeders may regularly co-occur in nonmarine
sediments, the combined effects of the benthos on sediments are likely
to be qualitatively different from those in marine communities. Using
trace fossils, it may be possible to determine the origin and longevity of
this association.

ACKNOWLEDGMENTS. We thank J. A. Fagerstrom and R. Bromley for their

critical reviews of an earlier version of this chapter.

References
Allen, J. R. L., 1964, Studies in fluviatile sedimentation: Six cyclothems from the Lower
Old Red Sandstone, Anglo-Welsh Basin, Sedimentology 3:163-198.
Allen, J. R. L., 1970, Studies in fluviatile sedimentation: A comparison of fining-upwards
cyclothems, with special reference to coarse-member composition and interpretation,
/. Sediment. Petro1. 40:298-323.
Allen, J. R. L., and Friend, P. F., 1968, Deposition of the Catskill facies, Appalachian region:
With notes on some other Old Red Sandstone basins, in: Late Paleozoic and Mesozoic
Continental Sedimentation, Northeastern North America (G. deY. Klein, ed.), Geo1. Soc.
Am. Spec. Pap. 106:21-74.
Aller, R. C., and Dodge, R. E., 1974, Animal-sediment relations in a tropical lagoon-
Discovery Bay, Jamaica, J. Mar. Res. 32:209-232.
Ashley, G. M., 1975, Rhythmic sedimentation in glacial Lake Hitchcock,
282 Chapter 6

Massachusetts-Connecticut, in: Glaciofluvial and Glaciolacustrine Sedimentation (A.

V. Jopling and B. C. McDonald, eds.), Soc. Econ. Paleontol. Mineral. Spec. Publ.
23:304-332.
Banerjee, 1., 1973, Sedimentology of Pleistocene glacial varves in Ontario, Canada, Can.
Geol. Surv. Bull. 226,A:1-44.
Barrett, P. J., 1965, Geology of the area between Axel Heiberg and Shackleton glaciers, Queen
Maud Range, Antarctica. Part 2: Beacon Group, N. Z. J. Geoi. Geophys. 8:344-370.
Belt, E. S., 1968, Carboniferous continental sedimentation, Atlantic provinces, Canada, in:
Late Paleozoic and Mesozoic Continental Sedimentation, Northeastern North America
(G. deY. Klein, ed.), Geoi. Soc. Am. Spec. Pap. 106:127-176.
Bender, F., 1963, Stratigraphie der "Nubischen Sandsteine" in Sud-Jordanien, Geol. Jahrb.
81:237-276.
Bender, F., 1968, Geologie von Jordanien, Beitrage zur regional en Geologie der Erde 7,
Borntraeger, Berlin, 230 pp.
Berg, T. M., 1973, Pelecypod burrows in the basal sandstone member of the Catskill For-
mation, northeastern Pennsylvania, Geol. Soc. Am. Abstr. Progr. 5:137.
Berg, T. M., 1977, Bivalve burrow structures in the Bellvale Sandstone, New Jersey and New
York, Bull. N. J. Acad. Sci. 22:1-5.
Bromley, R. B., and Asgaard, U., 1972, Notes on Greenland trace fossils. L Freshwater
Cruziana ffrom the Upper Triassic of Jameson Land, East Greenland, Bull. Gr0nlands
Geoi. Unders. 49:7-13.
Bromley, R, and Asgaard, U., 1979, Triassic freshwater ichnocoenoses from Carsberg Fjord,
East Greenland, Palaeogeogr. Palaeoclimatol. Palaeoecol. 28:39-80.
Brown, R W., 1934, Celliforma spirifer, the fossil larval chamber of mining bees, Wash.
Acad. Sci. J. 24:532-539.
Cant, D. J., and Walker, R G., 1976, Development of a braided-fluvial facies model for the
Devonian Battery Point Sandstone, Quebec, Can. J. Earth Sci. 13:102-119.
Chamberlain, C. K., 1975, Recent lebensspuren in nonmarine aquatic environments, in: The
Study of Trace Fossils (R W. Frey, ed.), pp. 431-458, Springer-Verlag, Berlin.
Chisholm, J. L, 1970, Lower Carboniferous trace fossils from the Geological Survey boreholes
in west Fife (1965-66), Geol. Surv. G. B. Bull. 31:19-35.
Conolly, J. R, 1965, Petrology and origin of the Hervey Group, Upper Devonian, central New
South Wales, J. Geol. Soc. Aust. 12:123-166.
Curran, H. A., 1980, Ichnology, J. Sediment. Petrol. 50:312-314.
Daley, B., 1968, Sedimentary structures from a non-marine horizon in the Bembridge marls
(Oligocene) of the Isle of Wight, Hampshire, England, J. Sediment. Petrol. 38:114-127.
Dinely, D. 1., and Williams, B. P. J., 1968, Sedimentation and paleoecology of the Devonian
Escuminac Formation and related strata, Escuminac Bay, Quebec, in: Late Paleozoic
and Mesozoic Continental Sedimentation, Northeastern North America (G. deY. Klein,
ed.), Geol. Soc. Am. Spec. Pap. 106:241-264.
Eagar, R M. C., 1974, Shell shape of Carboni cola in relation to burrowing, Lethaia 7:219-238.
Edwards, P., 1975, Invertebrate burrows in an Oligocene freshwater limestone, Univ. Wyo.
Contrib. Geol. 14:7-8.
Elliott, R E., 1968, Facies, sedimentation successions and cyclothems in productive coal
measures in the East Midlands, Great Britain, Mercian Geol. 2:351-372.
Elliott, T., 1976, Upper Carboniferous sedimentary cycles produced by river-dominated,
elongate deltas, J. Geol. Soc. London 132:199-208.
Eugster, H. P., and Hardie, L. A., 1975, Sedimentation in an ancient playa-lake complex:
The Wilkins Peak Member of the Green River Formation of Wyoming, Geol. Soc. Am.
Bull. 86:319-334.
Aquatic Nonmarine Trace Fossils 283

Fagerstrom, J. A., and Ratcliffe, B. C., 1975, Synopsis of Recent burrowing insects as ana-
logues of non-marine trace fossils, Geol. Soc. Am. Abstr. Progr. 7:1070-1071.
Frey, R W., 1978, Behavioral and ecological implications of trace fossils, in: Trace Fossil
Concepts (P. B. Basan, ed.), Soc. Econ. Paleontol. Mineral. Short Course 5:49-75.
Friend, P. F., 1965, Fluviatile sedimentary structures in the Wood Bay Series (Devonian) of
Spitzbergen, Sedimentology 5:39-68.
Gibbard, P. L., and Stuart, A. J., 1974, Trace fossils from proglacial Lake sediments, Boreas
3(2):69-74.
Gilliland, W. N., and LaRocque, A., 1952, A new Xenohelix? from the Paleocene of Utah,
J. Paleontol. 26:501-504.
Glaessner, M. F., 1957, Paleozoic arthropod trails from Australia, Palaeontol. Z. 31:103-109.
Goldberg, M., and Friedman, G. M., 1974, Paleoenvironments and paleogeographic evolution
of the Jurassic system in southern Israel, Bull. Geol. Surv. Isr. 61, 44 pp.
Grove, A. T., Street, F. A., and Goudie, A. S., 1975, Former lake levels and climatic change
on the rift valley of southern Ethiopia, Geogr. J. 141(2):177-202.
Hanley, J. H., Steidtman, J. R, and Toots, H., 1971, Trace fossils from the Casper Sandstone
(Permian) southern Laramie Basin, Wyoming and Colorado, J. Sediment. Petrol.
41:1065-1068.
Hantzschel, W., 1962, Trace fossils and problematica, in: Treatise on Invertebrate Paleon-
tology (R C. Moore, ed.), Part W, pp. 177-245, Geological Society of America and
University of Kansas Press, Lawrence.
Hesse, R, and Reading, H. G., 1978, Subaqueous clastic fissure eruptions and other examples
of sedimentary transposition in the lacustrine Horton Bluff Formation (Mississippian),
Nova Scotia, Canada, in: Modern and Ancient Lake Sediments (A. Matter and M. E.
Tucker, eds.), Spec. Publ. Int. Assoc. Sedimentol. 2:241-257.
Howard, J. D., 1978, Sedimentology and trace fossils, in: Trace Fossil Concepts (P. B. Basan,
ed.J, Soc. Econ. Paleontol. Mineral. Short Course 5:13-47.
Hubert, J. F., Butera, J. G., and Rice, R F., 1972, Sedimentology of Upper Cretaceous Cody
Parkman Delta, southwestern Powder River Basin, Wyoming, Geol. Soc. Am. Bull.
83:1649-1670.
Klein, G. deV., 1962a, Triassic sedimentation, Maritime Provinces, Canada, Geol. Soc. Am.
Bull. 73:1127-1146.
Klein, G. deV., 1962b, Sedimentary structures in the Keuper Marl (Upper Triassic), Geol.
Mag. 99:137-144.
Linck, 0., 1942, Die Spur Isopodichnus, Senckenbergiana 25:232-255.
Link, M. H., and Osborne, R. H., 1978, Lacustrine facies in the Pliocene Ridge Basin Group:
Ridge Basin, California, in: Modern and Ancient Lake Sediments (A. Matter and M. E.
Tucker, eds.), Spec. Publ. Int. Assoc. Sedimentol. 2:169-187.
McCall, P. L., Tevesz, M. J. S., and Schwelgien, S. F., 1979, Sediment mixing by Lampsilis
radiata siliquoidea (Mollusca) from western Lake Erie, J. Great Lakes Res. 5:105-111.
Miller, M. F., 1979, Paleoenvironmental distribution of trace fossils in the Catskill deltaic
complex, New York state, Palaeogeogr. Palaeoclimatol. Palaeoecol. 28:117-141.
Moussa, M. T., 1966, Insect tracks? Entomol. Soc. Am. Bull. 12:377.
Moussa, M. T., 1968, Fossil tracks from the Green River Formation (Eocene) near Soldier
Summit, Utah, J. Paleontol. 42:1433-1438.
Moussa, M. T., 1970, Nematode fossil trails from the Green River Formation (Eocene) in the
Uinta Basin, Utah, J. Paleontol. 44:304-307.
Mutch, T. A., 1968, Pennsylvanian nonmarine sediments of the Narragansett Basin, Mas-
sachusetts-Rhode Island, in: Late Paleozoic and Mesozoic Continental Sedimentation,
Northeastern North America (G. deY. Klein, ed.J, Geol. Soc. Am. Spec. Pap. 106:177-209.
284 Chapter 6

Olson, E. C., and Bolles, K., 1975, Permo-Carboniferous freshwater burrows, Fieldiana Geol.
33:271-290.
Peterson, A. R, 1976, Paleoenvironments of the Colton Formation, Colton, Utah, Brigham
Young Univ. Geol. Stud. 23:3-35.
Picard, M. D., and High, 1. R, Jr., 1972a, Paleoenvironmental reconstructions in an area of
rapid facies change, Parachute Creek Member of Green River Formation (Eocene), Uinta
Basin, Utah, Geol. Soc. Am. Bull. 83:2689-2708.
Picard, M. D., and High, L. R., Jr., 1972b, Criteria for recognizing lacustrine rocks, in: Rec-
ognition of Ancient Sedimentary Environments 0. K Rigby and W. K Hamblin, eds.),
Soc. Econ. Paleontol. Mineral. Spec. Publ. 16:108-145.
Ratcliffe, B. C., and Fagerstrom, J. A., 1980, Invertebrate lebensspuren of Holocene flood-
plains: Their morphology, origin and paleoecological significance, J. Paleontol. 54:614-630.
Rattigan, J. H., 1967, Deposition, soft sediment and post-consolidation structures in Palaeo-
zoic aqueoglacial sequence, J. Geol. Soc. Aust. 14:5-18.
Rayner, D. H., 1963, The Achanarras Limestone of the Middle Old Red Sandstone, Caithness,
Scotland, Geol. Soc. Yorkshire Proc. 34:117-138.
Read, W. A., and Johnson, S. R H., 1967, The sedimentology of sandstone formations within
the Upper Old Red Sandstone and lowest Calciferous Sandstone Measures west of
Stirling Scotland, Scott. J. Geol. 3:242-267.
Reineck, H.-E., 1955, Marken, Spuren, und Fahrten in den Waderner Schichten (ro) bei
Martinstein/Nahe, Neues Jahrb. Geol. Palaeontol. (Abh.) 101:75-101.
Reinemund, J. A., 1955, Geology of the Deep River coal field, North Carolina, U. S. Geol.
Surv. Prof. Pap. 246, 159 pp.
Rhoads, D. C., and Young, D. K, 1971, Animal-sediment relations in Cape Cod Bay, Mas-
sachusetts. Part II: Reworking by Molpadia oolitica (Holothuroidea), Mar. Bioi. 11:255-261.
Riding, R, 1979, Origin and diagenesis of lacustrine algal bioherms at the margin of the Ries
crater, Upper Miocene, southern Germany, Sedimentology 29:645-680.
Ryder, R T., Fouch, T. D., and Elison, J. H., 1976, Early Tertiary sedimentation in the western
Uinta Basin, Utah, Geol. Soc. Am. Bull. 86:496-512.
Sanders, J. E., 1968, Stratigraphy and primary sedimentary structures of fine-grained, well-
bedded strata, inferred lake deposits, Upper Traissic, central and southern Connecticut,
in: Late Paleozoic and Mesozoic Continental Sedimentation, Northeastern North Amer-
ica (G. deV. Klein, ed.), Geol. Soc. Am. Spec. Pap. 106:265-305.
Savage, N. M., 1971, A varvite ichnocoenosis from the Syka Series of Natal, Lethaia
4:217-233.
Schindewolf, O. H., 1928, Studien aus dem Marburger Buntsandstein III-VII; IV. Isopodi-
chnus problematicus (Schdwf.) im Unteren und Mittleren Buntsandstein, Senckenber-
giana 10:27-37.
Seilacher, A., 1963. Lebensspuren und Salinitats-Fazies, Fortschr. Geol. Rheinland. Westf.
10:81-94.
Seilacher, A., 1978, Use of trace fossil assemblages for recognizing depositional environ-
ments, in: Trace Fossil Concepts (P. B. Basan, ed.), Soc. Econ. Paleontol. Mineral. Short
Course 5:185-201.
Selley, R C., 1970, Ichnology of Paleozoic sandstones in the Southern Desert of Jordan: A
study of trace fossils in their sedimentologic context, in: Trace Fossils (T. P. Crimes
and J. C. Harper, eds.), pp. 477-488, Seel House Press, Liverpool.
Selley, R c., 1972, Diagnosis of marine and non-marine environments from the Cambro-
Ordovician Sandstones of Jordan, J. Geol. Soc. Lond. 128:135-150.
Siemers, C. T., 1970, Facies distribution of trace fossils in a deltaic environmental complex:
Upper part of Dakota formation (Upper Cretaceous), central Kansas, Geol. Soc. Am.
Abstr. Progr. 2:683-684.
Aquatic Nonmarine Trace Fossils 285

Siemers, C. T., 1971, Deltaic deposits of upper part of Dakota formation (Upper Cretaceous),
central Kansas, Am. Assoc. Petrol. Geol. Bull. 55:364.
Stanley, D. J., 1968, Graded bedding-sole marking-greywacke assemblage and related sed-
imentary structures in some Carboniferous flood deposits, eastern Massachusetts, in:
Late Paleozoic and Mesozoic Continental Sedimentation, Northeastern North America
(G. deV. Klein, ed.), Geol. Soc. Am. Spec. Pap. 106:211-239.
Stanley, K 0., and Fagerstrom, J. A., 1974, Miocene invertebrate trace fossils from a braided
river environment, western Nebraska, U.S.A., Palaeogeogr. Palaeoc1imatol. Palaeoecol.
15:63-82.
Surdam, R C., and Wolfbauer, C. A., 1975, Green River Formation, Wyoming: A playa-lake
complex, Bull. Geol. Soc. Am. 86:335-345.
Tarr, W. A., 1935, Concretions in the Champlain Formation of the Connecticut River Valley,
Geol. Soc. Am. Bull. 46:1493-1533.
Tevesz, M. J. S., and McCall, P. 1., 1978, Niche width in freshwater bivalves and its pa-
leoecological implications, Geol. Soc. Am. Abstr. Progr. 10:504.
Tevesz, M. J. S., and McCall, P. 1., 1979, Evolution of substratum preference in bivalves, J.
Paleontol. 53:112-120.
Tevesz, M. J. S., Soster, F. M., and McCall, P. 1., 1980, The effects of size-selective feeding
by oligochaetes in the physical properties of river sediments, J. Sediment. Petrol.
50:561-568.
Thoms, R. E., and Berg, T. M., 1974, Comparison of the burrowing habits of a Devonian
pelecypod with those of a Recent analogue, Geol. Soc. Am. Abstr. Progr. 6:267.
Toots, H., 1967 , Invertebrate burrows in the non-marine Miocene of Wyoming, Univ. Wyo.
Contrib. Geol. 6:93-96.
Toots, H., 1975, Distribution of meniscate burrows in non-marine Tertiary sediments of the
western U.S., Univ. Wyo. Contrib. Geol. 14:9-10.
Trewin, N. K, 1976, Isopodichnus in a trace fossil assemblage from the Old Red Sandstone,
Lethaia 9:29-37.
Truc, G., 1978, Lacustrine sedimentation in an evaporitic environment: The Ludian (Pa-
laeogene) of the Mormoiron basin, southeastern France, in: Modern and Ancient Lake
Sediments (A. Matter and M. E. Tucker, eds.), Spec. Publ. Int. Assoc. Sedimentol.
2:189-203.
Turner, B. R, 1978, Trace fossils from the Upper Triassic fluviatile Molteno Formation of
the Karoo (Gondwana) supergroup, Lesotho, J. Paleontol. 52:959-963.
Van Dijk, D. E., Hobday, D. K, and Tankard, A. J., 1978, Permo-Triassic lacustrine deposits
in the eastern Karoo Basin, Natal, South Africa, in: Modern and Ancient Lake Sediments
(A. Matter and M. E. Tucker, eds.), Spec. Publ. Int. Assoc. Sedimentol. 2:225-239.
Van Houten, F. B., 1964, Cyclic lacustrine sedimentation, Upper Triassic Lockatong For-
mation, central New Jersey and Adjacent Pennsylvania, Kans. Geol. Surv. Bull.
169:497-531.
Webby, B. D., 1968, Devonian trace fossils from the Beacon Group of Antarctica, N. Z. J.
Geol. Geophys . 11:1001-1008.
Webby, B. D., 1970, Brookvalichnus, a new trace fossil from the Triassic of the Sydney
Basin, Australia, in: Trace Fossils (T. P. Crimes and J. C. Harper, eds.), pp. 527-530,
Seel House Press, Liverpool.
Wells, R F., 1977, Freshwater invertebrate living traces of the Mississippi alluvial valley
near Baton Rouge, Louisiana, M.S. thesis, Louisiana State University, Baton Rouge, 253
pp.
White, C. D., 1929, Flora of the Hermit Shale, Grand Canyon, Arizona, Carnegie Inst. Wash-
ington Publ. 405, 221 pp.
 IV
Models
Chapter 7

Mathematical Models of
Bioturbation
GERALD MATISOFF

1. Introduction .................................. .. ............................................................................. 289

1.1. Processes to be Modeled ..................................................................................... 289
1.2. Kinds of Models and Their Objectives ...... .. ....................................................... 291
2. Particle Transport Models ........................................................................................... 293
2.1. Diffusion Models .................................................................. .. ............................. 293
2.2. Box Models .......................................................................................................... 305
2.3. Signal Processing Models .................................................................................... 309
2.4. Markov Models .................................................................................................... 310
3. Fluid Transport Models .............................................................................................. 313
3.1. Diffusion-Reaction Models ................................................................................. 314
3.2. Advection Models ................................................................................................ 322
4. Conclusions .................................................................................................................. 325
References ......................... .. ..................................... .. ........... .. ............................ .. ....... 327

1. Introduction
The purpose of this chapter is to examine existing mathematical models
of important chemical, physical, and biological effects of organisms on
sediments. The objectives and nature of the models will be discussed, the
mathematical solution techniques will be identified, and the advantages
and disadvantages of each type of model will highlighted.

1.1. Processes to be Modeled

Let us consider first a broad and practical subdivision of the effects

of organisms on sediments. Sedimentary materials exist in three states-

GERALD MATISOFF • Department of Geological Sciences, Case Western Reserve Uni-

versity, Cleveland, Ohio 44106.

289
290 Chapter 7

solid particles, pore fluids, and gases-but the vast bulk is comprised of
particulates and interstitial water. It is the movement of these latter two
constituents by biological agents that most models of animal-sediment
relations address, and this is how we will divide our discussion. There
are good reasons for this division: particulates and fluids can move in
different ways at different rates and in different directions, and are acted
on by different organisms, as previous chapters have demonstrated.
The redistribution of sediment particulates by organisms can obscure
primary stratigraphic features and create other secondary structures
(Rhoads, 1974). Graded bedding, for example, may be either a primary
depositional feature or caused by size-selective feeding activities of ben-
thos. Mixing invariably leads to poorer resolution of microfossil dating
of deep-sea sediments by lengthening the sediment section in which index
fossils are found and by overlapping the horizons of fossils indicative of
distinct time intervals (Berger and Heath, 1968). Biological mixing of
particles may also affect the chemical diagenesis of sediments (Aller,
1977). For example, in the typical case, fine-grained sediments in marine
and freshwater environments are chemically layered, with an oxidizing
zone extending a few millimeters or centimeters from the sediment-water
interface overlying a much thicker reduced zone. Iron is present in re-
ducing sediments in the solids as ferrous carbonate, phosphate, and sul-
fides. Organisms may transport particles from a reduced zone to the sed-
iment-water interface and cause the iron in the solid phases to oxidize
and precipitate as iron oxides and hydroxides. This refluxing of reduced
sediment to the sediment-water interface enhances the sediment oxygen
demand and affects the oxygen budget of the environment. An under-
standing of the effects of bioturbation on particle redistribution is essential
to a proper interpretation of these problems, and mathematical models
that assist in deciphering the information are highly desirable.
Being able to model approximately the fluid motion (or its effects)
is also essential to the understanding of the sediment's chemical and
depositional dynamics. The exchanges of interstitial and overlying water
and solutes are important processes in many chemical balances (e.g.,
phosphorus in Lake Erie) and in sediment diagenesis (Aller, 1980b). For
example, in the Three-Basin Phosphorus Budget Model for Lake Erie
(United States Army Corps of Engineers, 1975) the flux of phosphorus
from sediments to the lakewater is slightly greater than the entire external
loading of phosphorus to the lake. Thus the postdepositional transfer of
phosphorus is thought to be an important part of the Lake Erie phosphorus
cycle. Benthos often alter sediment fabric (for example, by creating a fecal
pellet layer) and indirectly modify the exchange across the interface owing
to diffusion through the pelletal layer (McCall and Fisher, 1980). Some
organisms construct rigid tubes in the sediment that allow free exchange
Mathematical Models of Bioturbation 291

of water and solutes from several centimeters deep to the overlying water
(irrigation) (Aller, 1980a). Other organisms inject water directly into the
sediment or pump water through their habitats during feeding, burrowing,
and locomotory activities (McCall et al., 1979; McCaffrey et al., 1980).
These biological activities result in an increased exchange of overlying
water and solutes with those in the sediments, and mathematical models
are essential to interpret properly the varied effects of bioturbation on
fluid and solute exchange.

1.2. Kinds of Models and Their Objectives

The term model is a very general one and can be applied to any
process or mathematical expression capable of giving information about
the approximate behavior of a real system. In this wayan experiment can
be a good model. Similarly, a good understanding of one physical phe-
nomenon and the theory developed for it could be used as a model to
simulate other physical processes as a first approximation. For example,
heat transfer in fluids may be used to model dispersion in fluids. Defined
here, a mathematical model will be a mathematical expression or group
of equations that can simulate the processes that occur in the real world
and that may be used to predict approximately the reaction of the physical
system to perturbations or simply to predict its state at future times.
Mathematical models may be classified as either deterministic or
probabilistic. A model is deterministic if, for each given input to the
model, there is a unique output. Stochastic or probabilistic models assume
that changes in the physical state from one condition to another occur
instantaneously. They can then be used to predict the current or future
state of the system or the likelihood or probability of the existence of
different states at some other time. Inputs to a model may be prescribed
either as known functions or as random functions, which can be described
only in a statistical or probabilistic sense. In the formulation of complex
probabilistic models deterministic models are often used and the ran-
domness is introduced either in the input or somewhere in the system
itself, depending upon the nature of the physical problem (Kowal, 1971).
The construction of a mathematical model to aid in the deciphering
of mixed sediments is dependent upon the nature of the results desired.
In deterministic models, the physical process is represented mathemat-
ically and the results of the model compared to field and experimental
data to determine if the model output is not inconsistent with the effects
of the organism's activities and can approximate them adequately. Dif-
fusion models are of this type. The solid particles are clearly not diffusing,
but the random transport of the particulates on a macroscopic scale caused
292 Chapter 7

by the actions of organisms may be described by such a mechanism in

the appropriate spatial and temporal context. On the other hand, it may
be advantageous to know the likelihood of a particle infilling a burrow
hole or becoming incorporated into a fecal pellet. Such models usually
have an output of "yes or no or a degree of maybe" and are stochastic or
probabilistic by their nature. In practice, then, the selection of one type
of model over another is dependent upon the type of information available
as input data and the type of output desired.
Many models can be classified as either deterministic or probabilistic,
depending upon the scale of observation. For this reason, the bioturbation
models discussed in this chapter have not been formally labeled as either
"deterministic" or "probabilistic." Rather, the classification presented
here is a convenient framework for discussing models of similar objectives
and formulation. We will consider four classes of particle transport
models: diffusion models, box models, signal processing models, and
Markov models; and two classes of fluid transport models: diffusion
models and advection models.
Diffusion models are by far the most popular and make up the bulk
of the literature. These models have great appeal because of the degree
of sophistication that can be achieved with them. All of these models
assume that the transport may be mathematically described as a diffu-
sional process. This permits the differential equations describing diffu-
sion to be used, to which other effects can then be easily incorporated as
additional terms in the equations. In addition to the mixing described by
diffusion, processes that have been examined include sedimentation,
mixing depth, radioactive decay, directed vertical transport, adsorption,
and chemical reactions. These models will be discussed in more detail
in Sections 2.1 and 3.1.
Box models, first introduced by Berger and Heath (1968), assume that
the upper layer of sediment is mixed rapidly with respect to sedimen-
tation. This is a good assumption in many environments, so that box
models remain popular as a first approximation for a bioturbation model.
These models are discussed in Section 2.2.
Another model that deserves mention because of its originality and
simplicity was proposed by Goreau (1977). His signal-theory-based model
calculates the output (Le., ultimate distribution in the sediment) for any
given input (distribution in the sediment in the absence of mixing) by a
convolution integral once the mixing function is specified. The mixing
function mathematically represents the sediment redistribution pro-
cesses. The model may also be applied to solute transport where the
mixing function represents pore water advection caused by sediment re-
distribution and faunal water pumping. This model is discussed in detail
in Sections 2.3 and 3.2.
Mathematical Models of Bioturbation 293

Recently, Jumars et a1. (1981) presented a Markov model of particle

redistribution. Their probabilistic model is based on different assump-
tions and can have quite different structures from the other types of
models. In general, it assumes no memory of previous events. The like-
lihood of the occurrence of another event depends only upon the prob-
ability distribution within the system immediately before the event oc-
curs. Thus the history of preceding events is contained only in the
present state of the system, and exact calculations back in time are not
possible. On the other hand, the model can generate the probability of all
possible future states of the system. This model is discussed in Section
2.4.

2. Particle Transport Models

Redistribution of sediment particles by benthos can occur by a variety
of processes. They may be ingested and defecated, advected by flow,
cemented together by egested materials, or simply pushed aside.
The mixing may occur on a variety of time scales. For example, stir-
ring in the horizontal or planar dimensions may be very slow, while
vertical turbation of particles by advection and dispersion by smearing
may be quite rapid (Piper and Marshall, 1969). In this section we will
examine existing models of particle redistribution.

2.1. Diffusion Models

The diffusion equation has been frequently applied to the redistri-

bution of solid particles. Clearly, the particles themselves are not dif-
fusing, and neither the actions nor distribution of the organisms in sed-
iments is random. However, the redistribution of solid particles by large
numbers of organisms and over a large number of individual transport
events may be mathematically described by the diffusion analogy. This
model cannot be used for strongly directional motion or transient con-
ditions without the addition of terms to describe the other effects. The
ability to add an additional term to account for another process is a highly
desirable feature of any model and is probably the reason that diffusional
models are the most popular mathematical models of bioturbation.
Goldberg and Koide (1962) developed the first diffusional model to
explain observed homogeneity in the ionium-to-thorium ratio in the upper
portion of a pelagic sediment column. Since the publication of their
model, a large number of diffusion models have been utilized to explain
observed distributions of many radionuclides in sediments modified by
294 Chapter 7

biological mixing. The models have most frequently been employed to

"unravel" the effects of particle redistribution in a determination of se-
dimentation rates in mixed sediments. The model has been applied to
21°Pb distributions by Bruland (1974), Spencer (1975), Shokes (1976),
Turekian et 01. (1978), and Santchi et 01. (1980) for estuarine and coastal
marine sediments; by Nozaki et 01. (1977), Turekian et a1. (1978), and
Peng et 01. (1979) for deep-sea sediments; and by Robbins et 01. (1977)
and Robbins (1978) for lake sediments. Robbins (1978) discusses in detail
the application and interpretation of 210Pb data in mixed sediments, and
Imboden and Stiller (1982) consider the effects of radon diffusion on the
210Pb distribution. The model has also been applied to the distributions
of 239, 240 pU (Schink et 01., 1975; Schink and Guinasso, 1982) for deep-sea
sediments and (Benninger et 01., 1979; Santchi et 01., 1980; Olsen et 01.,
1981) for estuarine and coastal marine sediments; to 234Th (Turekian et
01., 1978; Aller et 01., 1980; Cochran and Aller, 1980; Demaster et 01.,
1980; Santchi et aI., 1980) for estuarine and coastal marine sediments and
(Kadko, 1980a) for deep-sea sediments; to 7Be (Krishnaswami et 01.,1980)
for estuarine and lake sediments; and to 137CS (Robbins et 01.,1977,1979;
Robbins , 1978; Fisher et 01. , 1980) for lake sediments and laboratory ex-
periments and (Olsen et 01., 1981) for estuarine sediments.
The distribution of a particle-bound radionuclide whose source is
constant in time and at the sediment-water interface is pictured in Fig.
1. The radionuclide reaches the sediment surface sorbed onto a sediment
particle. Stirring of particles occurs to a depth m in the sediment, and,
if the mixing is fast with respect to the decay constant and sedimentation,
then the activity is uniform in the mixed layer. It is important to note that

Activity. A -
Water
Sediment
Mixing
Layer 1
m~----~-,~-------

No Mixing
Layer 2 Figure 1. Distribution of a particle-bound
radionuclide whos e source is constant in
N time and at the sediment-water interface,

-
s=
0.
(1)
C
If mixing is fast with respect to radioac-
tive decay and sedimentation (solid line)
then the activity is uniform in the mixed
layer and decreases exponentially with

j
deeper burial. The distribution with no
mixing is given by the dashed line. The
area under the two curves is the same.
Mathematical Models of Bioturbation 295

mixing decreases the activity at the sediment-water interface and in-

creases the activity everywhere below the mixed zone. Slower rates of
mixing, greater sedimentation rates, or faster rates of radioactive decay
will generate profiles that show a decrease in the mixed zone.
The general equation for the distribution of the radionuclide depicted
in Fig. 1 is

aA = ~ [D B(aA)] _ w(aA) _ AA (1)

at az az az
where A is the activity of the radionuclide, t is time, DB is the particle
biodiffusion coefficient, z is the depth in the sediment (positive down-
ward), w is the sedimentation rate, and A is the decay constant of the
radionuclide. The first term on the right-hand side of equation (1) rep-
resents the rate of change of activity at depth z owing to the effects of
mixing. DB, the particle biodiffusion coefficient, is treated mathematically
like an eddy diffusion coefficient and represents the rate of biological
mixing. If DB = 0, then there is no mixing. DB can be assumed constant
within the mixed layer (most models) or can be assigned a functional
form such as a linear decrease with depth (McCall and Fisher, 1980); an
exponential decrease with depth (Schink and Guinasso, 1977; Peng et al.,
1979; Santchi et al., 1980; Olsen et al ., 1981); or a gaussian distribution
within the mixed layer (Christensen, 1982). The second term describes
the change of activity as a function of depth resulting from the accu-
mulation of sediment, and the third term accounts for radioactive decay.
Under steady-state conditions , aAlat = o. If it is assumed that there is no
compaction and DB is constant down to z = m and 0 below, i.e., the
organisms uniformly mix the sediment to a depth m, then equation (1)
may be expressed as

- -
aA I
- AAI = 0 z:S m
az (2)

w
aAz) - AAz
(-az- = 0 z >m

The boundary conditions for these equations specify the depositional flux
of the radionuclide, require continuity of concentrations and material
fluxes between the layers, and provide for the eventual decay of the
isotope. A l refers to the activity in the upper layer and Az to the activity
below the zone of mixing. If the particle-bound substance were not ra-
dioactive, then A = 0, and the general solution for the activity in the
296 Chapter 7

mixed layer is

Al = ae ~ (wI Dn)zlI + b (3)

The solution is an exponential like those derived from the homogeneous

mixing models in a subsequent section of this chapter. Thus, under these
restricted sets of conditions, the different models yield the same results.
The power of the diffusion model lies in the ease of examining the in-
dependent effects of sedimentation, mixing, and mixing depth, and of
including such factors as chemical reaction, adsorption, and radioactive
decay as necessary. A few examples of previous applications of this model
follow.
Guinasso and Schink (1975) adopted a model similar to that of Gold-
berg and Koide (1962), but with the concentration dependent upon time
as well as depth. They modeled the time-dependent redistribution of a
source layer deposited at the sediment surface and concluded that the
appearance of the historical record depends upon the relative rates of
biological mixing and sedimentation and the depth of the mixed layer.
They quantified this relationship in the form of a mixing parameter, G:

G = DB /mw (4)

where m is the thickness of the mixed layer. With little or no mixing (DB
small) and/or a high sedimentation rate (w large), the impulse layer is
buried below the zone of mixing before much reworking takes place (G
less than ~O.05) so that the final distribution is gaussian. On the other
hand, a high mixing rate or a low sedimentation rate will result in a
homogenized mixed layer in a time short compared to the residence time
in the mixed layer, and the final distribution will be an exponential (G
greater than ~1.0) like those in Berger and Heath (1968) (Fig. 4). Figure
2 illustrates this result. They also show that dimensional analysis may be
used to estimate the particle biodiffusion mixing coefficient DB:

(5)

Since Vc has units of length per time, it may be thought of as the apparent
sedimentation rate or apparent reworking rate. Thus, estimates of the
thickness of the mixed layer and the reworking rate may be used to eval-
uate the particle biodiffusion coefficients.
Benninger et a1. (1979) developed a three-layer mixing model in
which the top 2-3 cm were mixed about 50 times faster than the middle
layer (3-10 cm). No mixing occurred in the bottom layer (>10 cm). They
interpreted discontinuous profiles of excess 210Pb and 234Th as caused
Mathematical Models of Bioturbation 297

Relative Concentration
,.... o 2
E
......
•X ."x
--
~ x
Co
Q)
c :::>
(/)

'2
...
-
Q)
.~ Q)>.
CIS CIS
Qi ....I
a: "0
Q)
X G=O.03
3 L--_ _ _ _ _-""'
~
.......
Figure 2. Concentration profiles of identical impulse sources for different values of the
mixing parameter G = DB/mw. The elapsed time after deposition of the impulse source is
that required for sediments of two mixed layer thicknesses to accumulate (t* = wtlm = 2) .
The mixed layer is indicated by the shaded area. Note the similarity with the homogeneous
mixed layer case (Fig. 4) for G > 1. After Guinasso and Schink (1975).

by deep burrows filled by accumulating suspended solids rather than the

collapse of burrow walls. They found that, in an area of slow sediment
accumulation, a low rate of bioturbation below the surficial zone of rapid
mixing causes an increase of at least a factor of two in apparent accu-
mulation rate. This demonstrates the necessity of characterizing DB as a
function of depth.
Peng et 01. (1979) used a numerical extension of the mixing model
of Guinasso and Schink (1975) to calculate sedimentation rates from 14C
and 210Pb data. They quantified how a thicker mixed layer results in older
core-top ages and younger burial ages. They also observed and modeled
a marked decrease in sedimentation rate apparently associated with the
climatic transition from glacial to recent interglacial.
Robbins et 01. (1979) performed a laboratory experiment in which
137Cs-Iabeled surface sediment was redistributed by the oligochaete Tub-
ifex tubifex and the amphipod Pontoporeio hoyi. They found that a simple
diffusional model accurately described the results of the P. hoyi experi-
ments, but could not account for the mixing processes of the oligochaetes.
They concluded that particle redistribution caused by the life activities
of benthos that move particles in a more or less random order should be
adequately described by the particle diffusion model. This model, how-
ever, cannot be used to describe mixing processes that exhibit preferential
directionality, such as feeding by the tubificids ("conveyor belt" trans-
port-see previous chapters). Some types of bioturbation are distinctly
advective on short time scales (Amiard-Triquet, 1974; Aller and Dodge,
298 Chapter 7

1974). Fisher et al. (1980) modeled experimental data for the tubificid T.
tubifex by including advective transport:

aA = DB (a 2
at az 2
A) _a(wA)
az
+ S(z) (6)

where w = w(z) is the vertical velocity of a sediment layer at depth z and

S(z) is the rate of radioactive sediment loss from a layer owing to feeding.
In their model, they assumed a depth-dependent advective velocity to
account for the depth-dependent feeding activities. The feeding function
was assumed to be gaussian about the zone of maximum feeding and to
decrease linearly to 0 at the surface. For a population density of 10 5
organisms/m 2, the advective velocity was constant at 0.12 cm/day from
the surface to just above the zone of maximum feeding, decreased by about
a factor of two at the depth of maximum feeding, and equal to 0 when the
feeding function was equal to O. In order to evaluate the relative impor-
tance of tubificid particle mixing in Lake Erie, the authors calculated the
ratio of particle mixing rates (Vt) (based upon population densities of
tubificids) to sedimentation rates (\1 5 ) , The ratios ranged from 0.3 to 48.0.
This indicates that in most locations the biological turnover of the sedi-
ment is faster than the rate of burial, and the authors expected the sedi-
ment to be significantly mixed.
The experimental results of Fisher et al. (1980) may be substituted
into equation (4) in order to quantify G, the mixing parameter of Guinasso
and Schink (1975). In the Fisher et al. (1980) model, DB was taken to be
7.5 x 10~3 cm2/day, and the base of the mixed layer was set at z = 8 cm.
Robbins (personal communication) has identified the mixed layer in a
field core to be 4.5 cm in a Lake Erie central basin sediment that is
undergoing sedimentation at a rate of 0.168 cm/yr (4.60 x 1O~4 cm/day).
Substituting values, G = 2.04 for an 8-cm mixed layer and 3.62 for a 4 .5-
cm mixed layer. According to Guinasso and Schink's classification, this
sediment would not exhibit a completely homogeneous upper layer be-
cause G < 10, but it is sufficiently mixed so that an impulse source
deposited at the sediment surface would appear with an exponential
distribution upon burial in the historical layers.
There have been several models that couple the particulate and pore
water phases (Schink et aI., 1975; Schink and Guinasso, 1977, 1978, 1982;
Wong and Grosch, 1978; Cochran and Krishnaswami, 1980). Since these
models primarily address the effects of soluble and/or adsorbed parti-
culates on the pore water concentrations, they will be discussed with the
pore water models in Section 3.
All the diffusional models require the evaluation of DB, the particle
biodiffusion coefficient. A tabulation of reported values of DB is given in
Mathematical Models of Bioturbation 299

Table I. The values of DB vary about six orders of magnitude, from a low
of 1.6 x 10- 10 cm 2/sec to a high of 1.5 x 10- 4 cm 2/sec. There does
appear to be a decrease in biological reworking from shallow water (DB
- 10- 6 cm2/sec) to the deep sea (DB - 10- 8 cm 2/sec). The range in values
is probably attributable in part to real differences in the magnitude of
mixing of different organisms, to differences in biomass, and to temper-
ature dependencies. Some of the variation is probably also due to the fact
that not all of the diffusion models are the same, and hence other terms
in those models may account for some of the mixing. This would modify
the calculated values of DB' For example, the two entries from Fisher et
a1. (1980) are in significant disagreement. The first value, 8.68 x 10- 8 ,
comes from their model, which accounts for the majority of the sediment
reworking by an advective term. The second value, 1.4 x 10- 6 , was
calculated using equation (5). This approximation was presented by Gui-
nasso and Schink (1975) as a means to estimate DB in a model that con-
siders diffusion to be the only means of redistributing sediment particles.
In this case, then, it is not unreasonable to expect the diffusion coefficient
to be larger in the model that accounts for mixing only by diffusion.
The reported biodiffusivities should be thought of as a net biodiffu-
sivity, DB, which is the product of the biomass density biodiffusivity, DB,
and the biomass density, b (g dry weight organism/cm 2):

(7)

In this manner, a comparison of the relative magnitudes of DB between

organisms (and with different organism densities) becomes more mean-
ingful. Unfortunately, much of the data in Table I were collected without
accompanying community or population density data. In shallow-water
environments, the community at the time of sampling may be significantly
different in numbers and species from that which has caused the previous
mixing history (McCall, 1977). Laboratory studies provide a better control
on community distributions and population densities, but there are only
a few of these studies. Table II contains those studies from which a com-
parison of DB can be made. The organism biomass density, b, has been
calculated from the product of the population density of the benthos and
the dry weight per individual. For Yoldia limatula, the dry weight per
individual has been estimated as 100 mg and for Pectinaria gauldii, as
30 mg. Fisher (1978) gives a dry weight of 4 mg per tubificid, and Marzolf
(1965) gives 1 mg for the weight of the amphipod Pantopareia hayi.
Benthos biodiffusion coefficients are very strongly temperature-de-
pendent. Although the data for Y. limatula are significantly improved
when corrected for organism densities, the data also reflect the fact that
Buzzards Bay is cooler than Long Island Sound (Rhoads, 1963). The high-
 w
Q
Table I. Reported Values of DB, the Particle Biodiffusion Coefficient, and rn, the Depth of the Mixed Layer Q

DB m
(cm2/sec) (cm) Method Location Organism(s) Reference
10 - 6 Impulse sourceD Chesapeake Bay Duursma and Gross (1971)
5.4 x 10 - 8 210Pbb San Diego Basin Bruland (1974)
2.5 x 10 - 9 210Pbb San Clemente Basin Bruland (1974)
3.5 x 10- 8 2 Rhoads (1963)" Buzzards Bay Yoldia limatula Guinasso and Schink (1975)
1.5 x 10 - 7 2 Rhoads (1963)" Long Island Sound Yoldia limatula Guinasso and Schink (1975)
3.2 x 10 - 7 2 Rhoads (1963)" Long Island Sound Yoldia limatula Guinasso and Schink (1975)
7.6 x 10 - 8 6 Gordon (1966)" Barnstable Harbor Pectinaria gouldii Guinasso and Schink (1975)
4 .1 x 10 - 5 38 Davison (1891) C Holy Island sands Arenicola Guinasso and Schink (1975)
7.6 x 10 - 6 38 Davison (1891)" Caves Haven Arenicola Guinasso and Schink (1975)
1.5 x 10- 4 30 Fox ef al. (1948) C Intertidal sand Thoracophelia mucronata Guinasso and Schink (1975)
4.4 x 10 - 8 6 Davis (1974)C Freshwater lake Tubiflex Guinasso and Schink (1975)
1.0 x 10 - 9 23 Microtektite dataO Indian Ocean Schink et a1. (1975)
1.8 x 10- 10 34 Microtektite dataO Eastern Indian Ocean Schink et a1. (1975)
3 x 10 - 8 20 Microtektite dataO Gulf of Mexico Schink et a1. (1975)
7 x 10- 10 83 Microtektite data C Indian Ocean Schink ef a1. (1975)
1.6 x 10 - 10 U Southern Indian Ocean Schink et a1. (1975)
48 Microtektite data
4.1 x 10 - 8 17 Microtektite dataO Equatorial Atlantic Ocean Schink et a1. (1975)
1.6 x 10- 10 42 Microtektite dataO Northern equatorial Pacific Schink et a1. (1975)
Ocean
2.4 x 10 - 10 22 Microtektite dataO Indian Ocean Schink et 01. (1975)
1.2 x 10- 7 12 239.24OpUO Mediterranean Sea Schink ef a1. (1975)
7.0 x 10 - 8 8 239. 24OpU" North Atlantic Schink et a1. (1975)
8.0 x 10 - 8 9 239.24OpUO Northern equatorial Atlantic Schink et a1. (1975)
4.4 x 10 - 8 4 239.24OpUO South Atlantic Schink et a1. (1975)
3 .2 x 10 - 8 6 239.24OpUO Southern equatorial Atlantic Schink et a1. (1975) n
6 .3 x 10 - 8 210Pbb Buzzards Bay Spencer (1975) ::r
OJ
'0
1.2 x 10 - 6 4 Long Island Sound Yoldia limatula and Nucula Aller and Cochran (1976) §'
annulata
"
3.5 X 10 - 6 4 Long Island Sound Yoldia limatula and Nucula Aller and Cochran (1976) ~
annulata '&"
1.3 X 10 - 6 4 Long Island Sound Yoldia limatula and Nucula Aller and Cochran (1976) C!>
S
annulata ~
6 x 10 - " 8 210Pb b North Atlantic Nozaki et al. (1977) n·
1.0 x 10 - 7 3 210Pbb Lake Huron Tubificids and Pontoporeia Robbins et al. (1977) -'"
~
affinis 0
0-
1.8 x 10 - 7 6 210Pb b Lake Huron Tubificids and Pontoporeia Robbins et a1. (1977) C!>
0;-
affinis g.,
6.3 x 10 - 9 4 210Pb b Lake Ohrid Robbins (1978) ttl
9.5 x 10 - 6 Assumed Gulf of Mexico Schink and Guinasso (1978) o·
0.2 x 10 - 6 234Thb Long Island Sound Turekian et a1. (1978) ..,2"
234Thb cr
1.2 X 10 - 6 Long Island Sound Turekian et a1. (1978) ~
0.1 X 10 - 6 234Thb Long Island Sound Turekian et al. (1978) o·
~
0.5 X 10 - 6 210Pbb New York Bight Turekian et a1. (1978)
6 x 10 - 9 210Pbb North Atlantic Turekian et al. (1978)
2 x 10 - 9 21°Pbb South Atlantic Turekian et a1. (1978)
14 x 10 - " 210Pbb Northern equatorial Pacific Turekian et a1. (1978)
8 x 10 - " 210Pbb Northern equatorial Atlantic Turekian et al. (1978)
10 x 10 - 9 210Pbb Northern equatorial Pacific Turekian et al . (1978)
1 x 10 - 9 210Pbb Antarctic Turekian et al . (1978)
7 x 10 - 9 210Pbb Antarctic Turekian et a1. (1978)
8 x 10 - 9 210Pbb Antarctic Turekian et al. (1978)
2 x 10 - 9 210Pbb Antarctic Turekian et a1. (1978)
2 x 10 - 8 3 239.24O pU b Long Island Sound Squil1a sp. Benninger et a1. (1979)
3 x 10 - 8 2 239.24O pU b Long Island Sound Squilla sp. Benninger et a1. (1979)
3.8 x 10 - 9 8 210Pb 14Cb Western equatorial Pacific Peng et a1. (1979)
3.8 x 10 - 9 3 210Pb 14Cb Western equatorial Pacific Peng et a1 . (1979)
1.6 x 10 - 7 1.5 137CSb Laboratory Pontoporeia hoyi Robbins et a1. (1979)
5.4 x 10 - 6 9 Robbins et a1. (1979)" Laboratory Tubifex tubifex This work
1.3 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
(Continued) W
....Q
 Table I. (Continued) W
0
N
DB m
(cm2/sec) (cm) Method Location Organism(s) Reference

0.01 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)

1.6 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
1.3 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.59 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.99 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.47 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.33 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.21 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.25 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.24 x 10 - 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.43 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)
0.15 x 10- 6 5 234Thb Long Island Sound Aller et a1. (1980)
9.5 x 10 - 7 234Th Amazon Shelf Demaster et a1. (1980)
8.7 x 10 - 8 8 137CSb Laboratory Tubifex tubifex Fisher et a1. (1980)
1.4 x 10- 6 8 Fisher et a1. (1980)" Laboratory Tubifex tubifex This work
1 x 10 - 6 15 Assumed 234Th Central Pacific Kadko (1980a,b)
4.6 x 10 - 6 12 21OPb" Central Pacific Kadko (1980a)
2.9 x 10- 7 3 7Be b Long Island Sound Krishnaswami et a1. (1980)
1.2 x 10 - 7 2 7Be b Long Island Sound Krishnaswami et a1. (1980)
0.47 x 10- 7 3 7Be b Lake Whitney Krishnaswami et a1. (1980)
2.4 x 10 - 7 5 234Th, 21OPb, Narragansett Bay Santchi et a1. (1980)
239. 24O pUb
1.6-10 x 10- 7 9 234Th, 21OPb, Narragansett Bay Santchi et a1. (1980)
239. 24O pU b
3.4 x 10- 7 7.5 234Th,21OPb, New York Bight Santchi et a1. (1980)
239.24O pUb
7.5 234Th, 21OPb,
n
::r
1.2 x 10 - 7 New York Bight Santchi et a1. (1980)
239.24O pUb '0
'"
4.1 X 10- 8 234Th, 21OPb, New York Bight Santchi et a1. (1980)
..,co
239.24O pUb "
 3.2 X 10 - 9 234Th, 21OPb, New York Bight Santchi et 01. (1980) ~
OJ
239.24OpU b
&
C1>
8.2 X 10 - 9 8 239. 24O pU" Western Atlantic Schink and Guinasso (1982)
239.240pU" S
1.3 X 10 - 8 15 Western Atlantic Schink and Guinasso (1982) ~
1.6 X 10- 8 11 239, 24O pU" Western Atlantic Schink and Guinasso (1982) n'
E-
9 ,8 X 10 - 9 10 239,240pU" Western Atlantic Schink and Guinasso (1982)
~
6.3 X 10 - 9 10 239, 24OpU" Western Atlantic Schink and Guinasso (1982) a
0-
1.9 X 10- 8 239,240pU" C1>
8 Puerto Rican Trench Schink and Guinasso (1982) Cil
1.6 X 10 - 8 10 239,240pU" Eastern Atlantic Schink and Guinasso (1982) g,
1.0 X 10 - 8 9 239,240pU" Eastern Atlantic Schink and Guinasso (1982) ttl
3,2 X 10 - 8 10 239,240pU" Eastern equatorial Atlantic Schink and Guinasso (1982) o·
B'
...,
0"
a Standard deviation technique as employed by Guinasso and Schink (1975). ~
b Fit parameter in diffusion model. o·
C Dimensional analysis [equation (5) in text]. ::l

Q
""
""
304 Chapter 7

Table II. Comparison of Selected Values of DB from Table I with DB, the
Values Corrected for b, the Biomass Density
DB
b (cm4/g dry
DB (g dry weight per Temperature
(cm2/sec) weight/m 2) sec) (0C] Organism(s) Location
3.5 x 10- 8 2.2 1.6 x 10- 4 Yoldia limatula Buzzards Bay
1.5 x 10- 7 6.1 2.5 x 10- 4 Yoldia limatula Long Island Sound
3.2 x 10- 7 13.7 2.3 x 10- 4 Yoldia limatula Long Island Sound
1.6 x 10- 8 0.3 2.5 x 10- 3 19 Pectinaria gouldii Barnstable Harbor
1.3 x 10- 8 0.3 2.5 x 10- 4 13 Pectinaria gouldii Barnstable Harbor
4.4 x 10- 8 0.89 4.9 x 10- 4 10 Limnodrilus and Messalonskee Lake
Tubifex tibifex
1.6 x 10- 7 16 1.0 x 10- 4 7 Pontoporeia hoyi Laboratory
r,
5.4 x 10 200 2.7 X 10- 4 20 Tubifex tubifex Laboratory
8.7 x 10- 8 260 3.3 X 10-'; 15 Tubifex tubifex Laboratory
1.4 x 10- 6 260 5.4 x 10- 5 15 Tubifex tubifex Laboratory

est value of DB reported, 2.5 x 10~3 cm 2 /sec for P. gouldii, was calculated
at 19°C, far warmer than the mean annual temperature at any of the field
sites. The 13°C value for the polychaete is comparable to those of the
other organisms .
Except for the low value of DB from the advection-containing model
of Fisher et al. (1980), the variation between studies is improved by a
factor of about 8 when corrected for benthic biomass. The range in values
of DB is more than a factor of 400, and the range of DB is about a factor
of 50. The remaining variation in DB is probably due to temperature
effects, effects of organisms not reported, sampling and experimental
error, and processes besides biodiffusion that cause mixing.
The studies listed in Table II do not include any estimates of DB from
the deep sea. Schink et al. (1975) estimate oceanic values of DB to range
from about 1O~8 to 1O ~ 10 cm 2 /sec. Although they did not determine the
benthic community and population densities at these sites, an estimate
of the biomass may be obtained from the work of Rowe et al. (1974). They
derived a log-linear relationship between biomass density and water
depth for the Atlantic:

10glO biomass [mg/m 2 ]

= 3.03 - 0.0031(water depth in meters - 2104 m) (8)

Deep-sea sediment could be expected to contain a biomass of about 150

mg/m 2 (b = 0.15 g/m 2 ) . Substituting into equation (7) gives a range for
Mathematical Models of Bioturbation 305

DB of 6.7 X 10- 4 to 6.7 X cm4/g dry weight per sec. These values
10 - 6

are comparable to those in Table II and indicate that correcting the dif-
fusivities for biomass density can account for the majority of the enormous
variation in reported values of DB' This finding also suggests that equation
(7) may be readily utilized to estimate DB quantitatively.
Diffusion models provide a mechanistic interpretation of the biotur-
bation process. Techniques for quantitatively determining DB are well
established. In cases where the mixing processes are controlled by mech-
anisms other than just diffusion, the mathematics of diffusion models
readily permit the incorporation of these other processes. This feature has
made the diffusion model the most popular approach to describing
bioturbation.

2.2. Box Models

Figure 3 represents the simplest mixing model. This approach has

been classified as a "box model" (Berner, 1980). The model assumes
continuous sedimentation and incorporates mixing in a layer immediately
beneath the sediment-water interface. This layer is treated as being com-
pletely homogenized to a fixed depth (Berger and Heath, 1968; Ruddiman

Sedimentation

I
Sed iment - Water Interface

T T
Homogenized Heterogeneous
Mixed Mixed
Layer Layer

L ~
...L
Bottom of Maximum Depth
Homogenized of Present
Layer of Bioturbation
Thickness m

Figure 3. Schematic diagram of box mixing models. Particles deposited upon the sedi-
ment-water interface by sedimentation are rapidly and uniformly incorporated into the
mixed layer in the homogenous model and are redistributed with depth and spatial con-
straints in the heterogeneous model. The rate of deposition determines the rate of removal
of sediment from the mixed layer into buried sediment.
306 Chapter 7

and Glover, 1972; Davis, 1974; Peng et 01., 1977; Sundquist et 01., 1977;
Benoit et 01., 1979; Cochran, 1980) or as being heterogeneously mixed
(Piper and Marshall, 1969; Hanor and Marshall, 1971; Hakanson and
Kiillstrom, 1978). Homogeneous mixing implies complete disordering of
the particles within the mixed layer and hence is analogous to very rapid
"particle diffusion." The homogeneous models may be used to study only
steady-state mixing, because only two conditions are permitted, i.e., un-
mixed and totally homogenized. Heterogeneous mixing models account
for partial mixing in the mixed layer corresponding to burrows by having
small volumes of homogeneous mixing, or allowing the mixing to decrease
with depth in the sediment.
A box model is a mass balance model. When a particle lands on the
sediment surface, it is incorporated into the mixed layer. In the homo-
geneous model, the rate of mixing is large compared with the sedimen-
tation rate, while in the heterogeneous model it is not. A small increment
of sediment deposition results in a small increment of sediment removed
from the mixed zone and buried. Thus, there is a certain fraction of the
newly deposited sediment that is buried below the mixed zone. A few
examples of this model follow.
Berger and Heath (1968) expressed the mass balance as

dP -dL
(9)
P m

where P is the probability of finding the particle in the homogeneous

layer, dL is the thickness of sediment that has been deposited on the layer,
and m is the thickness of the mixed layer. Direct integration of this equa-
tion yields

P = Po exp( - Lim) (10)

where Po is the original probability. The mass balance of Sundquist et 01.

(1977) may be expressed as (ignoring dissolution)

dg F
- = - ( f - g) (11)
dt mp

where F is the total sediment mass flux to the mixed layer, g is the mass
fraction of the particles in the mixed layer, f is the mass fraction of the
particles in the sediment supply, m is the thickness of the mixed layer,
p is the dry density of sediment in the mixed layer, and t is time. The
solution to equation (11) if all parameters except g are assumed to be
Mathematical Models of Bioturbation 307

constant is

where go and to refer to initial conditions. If sedimentation occurs over

a time interval dt, then a sediment layer of thickness dL will have been
deposited. Hence, (F/p)(dt) = dL or (F/p)(t - to) = L. The first term in
equation (12) thus represents the fraction of the new particles in the
homogeneous layer and may be expressed as goexp( - Lim), which is
identical to the right-hand side of equation (10). This demonstrates that
the mass balance formulations of Berger and Heath (1968) and Sundquist
et 01. (1977) are the same. Furthermore, note that the exponential form
of the distribution in the mixed layer in these homogeneous box models
is the same as that derived from the Guinasso and Schink (1975) model
for G > 1 [equation (4); compare Figs. 2 and 4].
Berger and Heath (1968) developed the mixing model for the inter-
pretation of microfossil distributions in deep-sea sediments. Figure 4
shows a typical result of their homogeneous model. Application of this
model to a particle-bound radionuclide whose source is at the sedi-
ment-water interface would generate a profile similar to that of the solid
line in Fig. 1. Berger and Heath (1968) concluded that marked stratigraphic
errors can appear if the layers representing the time ranges are similar in
thickness to the mixed layer.
Davis (1974) reached a similar conclusion by calculating age fre-
quency distribution curves of pollen in sediments reworked by tubificids.
He measured pollen displacement rates in l-cm depth intervals in labo-
ratory experiments. His mass balance model was expressed as

Zj = PCo(F) + PC l (Yd + PCz(Y z) + ...

+ PCn(Yn) + ... + PCN(Y N) (13)

where Zj is the number of pollen grains present in the nth depth interval
at the beginning of the current year, Yn is the number of pollen grains in
the nth depth interval at the beginning of the previous year, N is the
number of depth intervals from which pollen grains in the ith interval
originated during the previous year, PC is percent, and F is the number
of pollen grains added from the water column during the previous year.
His model predicted that 36% of surface pollen would be older than 30
years and 5% older than 90. His age frequency distribution curves give
results that are similar to Fig. 4, but are plotted in a different manner.
308 Chapter 7

Fraction Component in Sediment

o 1

--------- ---- e T
m
.i..

I
I-
a..
w
o

~ T
m
..l

Figure 4. Vertical distribution of a component in the sediment resulting from homogeneous

mixing and sedimentation. a and e mark time-equivalent levels in the sediment of the
appearance and disappearance (extinction) of the component. Note that the component is
found to a depth of one mixed layer thickness below the time horizon of the first arrival of
the component. Similarly, note that the concentration of the component begins to decrease
at a depth of one mixed layer thickness deeper than the time associated with the extinction.
This is because of dilution by downward mixing of fresh sediment absent in this component.
Finally, note the similarity with the rapid particle diffusion case (Fig. 2). After Berger and
Heath (1968).

Peng et a1. (1977), Sundquist et a1. (1977), and Benoit et al. (1979)
used the box model to show that bioturbation causes systematic deviations
of apparent 14C ages from true ages of carbonate and humic-rich sedi-
ments. Sundquist et al. (1977) included a term for chemical reaction
(dissolution of CaC0 3 ) and concluded that sediment mixing may account
for discrepancies in correlating the high-carbonate top layer of sediments
with the last glacial period.
The heterogeneous models based on the same ideas were developed
by Piper and Marshall (1969) and Hanor and Marshall (1971) to examine
progressive mixing of a given volume of sediment. Piper and Marshall
(1969) examined the mixing of a simple two-component mud-over-sand
system. They estimated the amount of reworking at the junction of sand
Mathematical Models of Bioturbation 309

and mud beds by comparing observed distributions of sand-size materials

within the mud. They assumed that the sediment affected by a hypo-
thetical burrow is completely homogenized. They let the number of ver-
tical burrows decrease linearly with depth, while the number of horizontal
burrows remained the same at all depths. Otherwise a random distribution
of burrows was assumed. Their calculated redistribution curves are sim-
ilar to the fraction distribution curves of Berger and Heath (1968) (Fig. 4)
and the age frequency distribution curves of Davis (1974), but are plotted
in a different manner. They found that the relationship between the
amount of disturbance of a primary sedimentary structure and the total
amount of reworking is dependent on the spatial distribution of burrows,
and concluded that only where there is a very low degree of disturbance
can an estimate be made directly of the amount of reworking without making
assumptions as to the distribution of burrows. Thus, a high rate
of reworking of sediments makes a heterogeneous model reduce to a
homogeneous model, depending upon the scale of observation (Hanor
and Marshall, 1971).
The last model of this type mentioned here is that of Hakanson and
Kallstrom (1978), who presented a "biotransport" model of the porosity
in the top 20 cm of lake sediments, where the porosity is determined by
gravitational compaction of sediment and a net upward advection of sed-
iment by organisms. The model permits the calculation of age frequency
distribution for defined sediment layers and the results agree well with
those of Davis (1974).

2.3. Signal Processing Models

Goreau (1977) presented a signal-theory-based model that allows the

calculation of the effect of any series of time-varying mixing events on
any time-varying input to the sediment. The input function, I(x), is the
profile that would result if some time-varying input were delivered to the
sediments in the absence of mixing. The output of the model, O(x), is the
final mixed profile that results from the action of the mixing function,
F(x, x'):

O(x) = J: I(x ' ) F(x - x') dx ' (14)

Figure 5 schematically demonstrates the input-output nature of the

model.
In order to utilize the model, it is necessary that the mixing function
F(x), be known. F(x) can be determined from measurements of O(x) when
310 Chapter 7

I(x) O(x)

:r:
I-
a.
w
Q

Figure 5. Signal processing model of Goreau (1977). [(x) represents the profile that would
be observed in the complete absence of mixing of a sedimentary input that is increasing
exponentially with time and with a single source layer superimposed. F(X,X,) is the mixing
function, identified here at x = x" for a discrete input. This function is convolved with [(x)
at all depths to obtain the observed historical record, O(x). Note that the mixing function
F(X,X,) is similar to the profiles obtained with an intermediate value of G (Fig. 2). After
Goreau (1977).

the input to the sediment is known. However, the calculation of the mixing
function does not necessarily describe the actual mixing processes of the
organisms, only their net effect on the redistribution of the sediment
particles. Goreau suggests that when F(x) is known it can be used in
conjunction with measured mixed profiles, O(x), to deconvolve the effects
of mixing by inversion of equation (14) to derive the original input history,
I(x). The potential importance of the deconvolution technique to inter-
preting the historical record for detailed stratigraphic and pollution in-
formation is significant. Successful deconvolution of equation (14), how-
ever, remains to be demonstrated.
Application of the signal theory model appears to be straightforward
in the presence of a discrete input signal as seen in Fig. 5. The model
does not give a time history of the sediment redistribution, but rather the
net result of the mixing as defined by the input-output nature of the
model. Much work needs to be performed with this model to demonstrate
its full capabilities.

2.4. Markov Models

Recently, Jumars et ai. (1981) presented the formulation of a Markov

chain model of flow-sediment-organism interactions. A Markov process
Mathematical Models of Bioturbation 311

is a stochastic process in which the probability of a change in state of a

parameter depends only on the present state of the parameter and not on
the entire past evolution of the system. A Markov chain is a Markov
process with a finite number of states (Blanc-Lapierre and Fortet, 1965).
Figure 6 graphically illustrates the Jumars et al. (1981) model of se-
lective feeding and fecal pellet production coupled with sediment burial.
An individual sediment grain may be selected stochastically for ingestion
by a deposit feeder and move from the free state to the pelletal state.
Conversely, a pellet may break down into its constituent particles. Buried
sediment is considered removed from the activities of benthos. The re-
quired input data are the probabilities of transition from one compartment
to another. The probability of buried sediment becoming pelletized is
zero, by definition of buried sediment, and, for the sake of simplicity, the
probability of burial of an intact pellet was set equal to zero. Thus, the
model is built upon the assumption that the transition probabilities de-
pend only on the present state and not on the history of the sediment
particles. As Jumars et a1. (1981) point out, the model in absence of sed-
imentation is homologous with the infinite series model of deposit feed-
ing developed by Levinton and Lopez (1977).
The solution procedure of the Markov model is described by Kemeny
and Snell (1960) and Blanc-Lapierre and Fortet (1965). A matrix of tran-

remaining
pelletized Pellets
(rp)

disaggregation (d)

remaining ftFree"
free Sediments
(rt)

burial (b)

remaining Buried
buried Sediments
(rb)

Figure 6. Schematic diagram of the Markov model of deposit feeding and sediment burial.
A particle may exist in one of three compartments: (1) in a pellet, (2) as "free" sediment,
or (3) as buried sediment. Transfer between compartments is accomplished by assigning
probabilities to the transitions in particle state. After Jumars ef a1. (1981).
312 Chapter 7

sition probabilities is devised and constrained by the known transition

probabilities. For the model in Fig. 6, the transition matrix is

To

Buried Pellet Free

(15)

[~oo
~l
{ Buried 0
From Pellet rp
Free 0.01 s

In this case, as noted earlier, the probability of a buried particle remaining

buried is equal to 1, and that of a pellet becoming buried is equal to O.
Similarly, buried particles cannot become pellets or free particles. The
other "known" information is that the probability of a particle input to
the system as a free particle becoming buried is 0.01. Mass balance con-
straints are that all rows in the matrix add to 1:

rb = 1.00
rp + d = 1.00 (16)
s + rf = 0.99

In the above equations the dependent variables rp and rf can be calculated

as a function of the transitional probabilities, d, s, for breakdown and
selection, which will in general be variable- and problem-dependent (i.e.,
they depend on the type of particle, feeder, physical environments, etc.)
and should be determined experimentally.
The model predicts that particles that are more strongly selected will
have greater residence times in surficial sediments before they become
part of the stratigraphic record and will therefore achieve higher concen-
trations in surficial sediments than will less-preferred particles. Selective
feeding thus increases the average age of surficial sediments. The impli-
cations for deep-sea stratigraphy are enormous. The model perhaps pro-
vides a more biologically realistic alternative explanation of the frequently
occuring anomalously great ages of surficial sediments than core-top loss
or deep mixing. The authors extend the capabilities of the model to in-
clude lateral transport of surficial sediments. Since the transport is size-
dependent, it is possible to include feedback loops to the model in Fig.
6 to account for internal transport. Application of this model to a real
example is necessary to demonstrate its full potential. The authors have
indicated that they are pursuing that objective (see also Nowell et 01.,
1981).
Mathematical Models of Bioturbation 313

3. Fluid Transport Models

The activities of benthos not only redistribute sediment particles but
also influence the exchange of water and solutes across the sediment-water
interface. Burrows act as channels for the direct communication between
interstitial and overlying waters (irrigation) (Aller, 1977); fecal pellets
deposited on the sediment surface increase the porosity of the uppermost
sediment and therefore increase the diffusional exchange (Fisher et al.,
1980); and the injection of surface water for burrowing and feeding mod-
ifies the porosity and chemical conditions in the sediment and increases
the exchange of water and solutes across the sediment-water interface
(McCall et al., 1979; Aller, 1978). Each process may be caused by different
organisms, although some benthos may exhibit more than one exchange
technique and may exchange fluids and solutes to different degrees and
from different depths. The exchange may also occur on a variety of time
scales. For example, diffusional exchange of solutes may be very small
even in the presence of high organism densities, while exchange caused
by the turbulent mixing of surficial sediment during initial burrowing
may be many times larger (G. Matisoff and P. 1. McCall, unpublished
data). Understanding the exchange processes and the time scales on which
they occur is essential to a quantitative description of chemical mass
transfer between sediments and water.
Mathematical models of the exchange of pore fluids in the absence
of benthos are well developed (Berner, 1980). These are most commonly
one-dimensional diffusion models, where the ionic dispersion coefficient
represents the molecular diffusivity in the bulk sediment. These models
have been extrapolated for use in the presence of macroinfauna and are
similar to the particle diffusion models of Section 2.1. In these models
the apparent diffusion coefficient in the bioturbated zone is elevated over
the value in the bulk sediment (Hammond et a1., 1975; Schink et al., 1975;
Goldhaber et al., 1977; Schink and Guinasso, 1977, 1978, 1982; Vander-
borght et al., 1977; Aller, 1978; Wong and Grosch, 1978; Cochran and
Krishanaswami, 1980; Filipek and Owen, 1980; McCall and Fisher, 1980).
These models will be discussed in more detail in Section 3.1.1.
Permanent or semipermanent tube structures or significant directed
exchange of water cannot necessarily be described by simple one-dimen-
sional models or by quantitatively modifying a molecular diffusivity
(Aller, 1978). These conditions require diffusional models other than one-
dimensional vertical diffusion and/or other mathematical expressions to
describe the exchange. "Biopumping" models (Grundmanis and Murray,
1977; Hammond and Fuller, 1979; McCaffrey et a1., 1980) describe the
exchange of solutes between well-mixed and distinct reservoirs of sedi-
ment and overlying water as an advection velocity. The signal processing
314 Chapter 7

model (Goreau, 1977) formulates the volume of pore water exchanged

with overlying water owing to bioturbational mixing of sediment and
faunal water pumping. These models characterize pore water exchange
as flow and may be classified as "advection" models. They are discussed
in more detail in Section 3.2. Recently, Aller (1980a,b) has presented a
model that does not account for the effects of macrobenthos with empir-
ically elevated diffusion coefficients or advection velocities. His model
assumes that at any instant the sediment is a mosaic of micro environments
represented on the average by a central irrigated tube or burrow and its
immediately surrounding sediment. Solute exchange is described by
three-dimensional diffusion (cylindrical coordinates). This model is dis-
cussed in more detail in Section 3.1.2.

3.1. Diffusion-Reaction Models

Although the number of mathematical models of fluid transport is

small, models of the diffusion-reaction type have been selected in almost
all cases for applications involving benthos and fluid transport. The math-
ematics of one-dimensional vertical diffusion models are idential to those
of the particle diffusion models discussed in Section 2, although the
diffusion coefficient refers to the apparent diffusivity of the solute in the
sediment. Aller (1978) notes that sedentary infauna can alter sediments
by building tubes that are more accurately characterized by cylindrical
diffusion geometry in the sediment immediately surrounding the burrows.
He also demonstrates (Aller, 1977) that mobile infauna do not remain
stationary long enough to allow establishment of fixed diffusion-reaction
geometry. In this case, a one-dimensional vertical diffusion model with
an apparent diffusion coefficient may be used. The choice of a one-
dimensional vertical diffusion model or another diffusion model that more
accurately describes the benthic processes is dependent upon the spatial
and temporal scale of investigation.

3.1.1. One-Dimensional Vertical Diffusion Models

Particle reworking near the sediment-water interface can create a
mixed layer in which the pore water solutes migrate more rapidly than
can be accounted for by molecular diffusion. The solid particles would
be distributed as in Fig. 1. The solute transport within the sediment may
be described as a two-layer, one-dimensional system undergoing vertical
diffusion, sedimentation, compaction, adsorption, chemical reaction, and
radioactive decay. The upper, mixed layer has an apparent molecular
diffusivity higher than the molecular diffusivity in the lower layer. Math-
Mathematical Models of Bioturbation 315

ematically, this may be described as (modified from Berner, 1980)

aC 1 1 ] [a(<1>D 1 aCl!aZ)]
Zone 1: -
at = [ <1>(1 + K) az
_ w (aCl) _ AC 1 + (_1_) lR
az 1 + K

Zone 2:
aC 2 [ 1 ] [a(<1>D 2 aCl!aZ)]
at <1>(1 + K) az
_ w(aC2) _ AC
az
2 + (_1
1 +
) lR
K

(17)
where C is concentration, t is time, <1> is porosity, z is the depth below the
sediment-water interface, D is the apparent diffusion coefficient, K is the
adsorption coefficient, w is the sediment rate, A is the radioactive decay
constant, and lR represents all nonequilibrium slow reactions affecting
C. Ignoring compaction, adsorption, chemical reaction, and radioactive
decay, and assuming the sediment properties to be homogeneous within
each zone, the equations simplify to

Zone 1: aCl=D (a Cl) _ w (a~l)

az 2
2

at 1
(18)
Zone 2: aCat2= D 2
(aaz2C2)
2 _ w(a~2)
The two zones are coupled by requiring that the concentrations at the
boundary between the zones and the fluxes across the boundary be con-
tinuous. Assuming sedimentations to be small over the course of an in-
vestigation, the flux boundary condition becomes

(19)

Thus, if Dl > D2 , the concentration gradient in the upper zone is less than
the gradient in the lower zone and a break in curvature occurs at the
boundary (Fig. 7). Some of the models of this type are now discussed in
more detail.
Hammond et a1. (1975) were the first to utilize this approach. They
discovered that they could not explain observed fluxes of radon from
sediments with a semiinfinite (single-layer, Dl = D 2 ) model, but that the
316 Chapter 7

Concentration! C co
o 0.5 1.0

E
.....
.c
C.
C1)
c

Figure 7. Comparison of the effect on steady-state pore water profiles of increasing the
apparent diffusivity in the mixed layer (Zone I). No burrowing is described when D, = D L •
m is the depth of the mixed layer. Cx is the concentration observed at depth. when dC/dz
approaches zero. The profile shape is maintained at steady state by a source of the substance
in the sediment. After Aller (1978).

two-layer model, which incorporates an upper layer stirred by currents

and organisms, could account for the higher flux. They estimated that the
apparent diffusion coefficient in the mixed layer was 3-7 times the value
in the unmixed lower layer. An examination of the interstitial water data
alone would not have led them to this observation. Their results dem-
onstrate the importance of obtaining flux data in addition to interstitial
water data in order to identify the important processes occurring at the
sediment-water interface.
Schink et 01. (1975) presented a unique model in which the redis-
tribution of sediment solids by particle biodiffusion is coupled with the
pore water concentrations. In their model, silica is input to the sedi-
ment - water interface as particulate biogenic silica (diatom and radiolar-
ian frustules). In the steady state, the interstitial water concentration is
determined by a balance between the rate of dissolution of the particulate
silica and the rate of diffusional flux of dissolved silica from the sediment.
Mathematical Models of Bioturbation 317

They do not assign an enhanced value to the diffusion coefficient of the

solute in the mixed zone (Le., Dl = D2)' The dissolution rate depends
upon the degree of undersaturation (C s - C), which decreases with depth
in the sediment. Downward mixing will move particles from a zone of
rapid dissolution and rapid flux from the sediment to a level of slower
dissolution and less flux from the sediment. Subsequent models of theirs
(Schink and Guinasso, 1977, 1978, 1982) have included terms for ad-
sorption, sedimentation, water content, and sediment composition, and
the models have been applied also to calcium carbonate in sediments. In
order to illustrate the basics of their models, we will only examine their
early, simpler work on the effects of mixing of the solid particles on the
interstitial water concentrations (Schink et al., 1975).
Their coupling of interstitial water concentrations and solid particles
was expressed as:

For dissolved silica: D ( a

az 2
2
C) + (~)
'I' (C fC-f C) B-_ 0 (20)

(21)

where D is the aqueous molecular diffusion coefficient of silica modified

for the effects of tortuosity, C is the concentration of dissolved silica, z
is the depth below the sediment-water interface, KB is the first-order
dissolution rate constant of solid, reactive opaline silica, <!> is porosity,
Cf is the solubility concentration of solid, reactive opaline silica, B is the
concentration of solid, reactive opaline silica, and DB is the particle bio-
diffusion coefficient (= 0 below the mixed layer).
Figure 8, adapted from their work, illustrates some of their results.
For a given input flux of particulate biogenic silica, the concentration at
depth in the interstitial water (balance concentration, C b ) increases with
increasing biological mixing, other parameters (porosity, dissolution rate,
aqueous diffusion coefficient, concentration at the sediment-water in-
terface) being fixed. The downward mixing moves biogenic siliceous sed-
iment particles into a region where the supply of silica to the pore waters
by dissolution exceeds the removal by diffusional flux to the water col-
umn, and so the interstitial water concentrations increase.
The results in Fig. 8 demonstrate the important effects of biological
mixing on the calculated interstitial water concentrations. Schink and
Guinasso (1977, 1982) have subsequently extended the model to permit
the calculation of the fraction of reactive solid remaining as a function
of the other parameters. They do not believe, however, that variations in
bioturbation rates will significantly alter the CaC0 3 (and perhaps parti-
318 Chapter 7

10-6r-----~------~~------~

10-8~--~--
o __----------------
200 400 600

Figure 8. Effect of the biological particle mixing coefficient, DB, on the interstitial water
silica concentration at depth (C~ > 10-100 cm). After Schink et oJ. (1975).

culate silica) contents of accumulating sediments, and therefore have

fixed DB as constant in their newer models.
Aller (1978) applied the two-layer approach (with a source term) to
the pore water ammonium distribution in laboratory tanks with and with-
out Yoldia limatula. He estimated the apparent diffusion coefficient to
be 1 x 10- 5 cm 2 /sec in the mixed zone, compared to 7 X 10 - 6 cm 2 /sec
below the mixed zone. His calculated profiles did not agree with the
shapes expected (Fig. 7), i.e., the interstitial water concentrations below
the mixed zone should be lower in the presence of mixing by Yoldia
because of enhanced diffusive loss in the upper layer. He concluded that
microbial production rates of NHt are greater below the mixed zone in
the presence of Yoldia than in its absence, and suggested that this result
was caused by the provision of a nutrient source or a flushing from the
sediment of an inhibitor.
Mathematical Models of Bioturbation 319

Cochran and Krishnaswami (1980) examined concentration-depth

profiles of radium, thorium, uranium isotopes, and 21°Pb in northern
equatorial Pacific sediments and sediment pore waters. They divided the
sediment column into two zones: an upper zone, which is biologically
mixed, and deeper sediment. Throughout the sediment column 226Ra is
mobilized to the pore water following its production from radioactive
decay of 230Th. Radium in the pore water can be readsorbed onto sediment
particles, diffuse, and decay. Mixing redistributes the 230Th activity
(hence 226Ra production rate) such that a constant 230Th activity is present
in the mixed zone. They found good agreement between the deduced
parametric constants for the "good fits" of their data and those expedited
from independently available data.
McCall and Fisher (1980) adopted the two-layer approach in the ex-
amination of chloride transport across the sediment-water interface from
NaCI-spiked pore waters. They found that the transport could be described
in the absence of infauna by one-dimensional vertical diffusion in a semi-
infinite medium, with DC1 - = 5.5 X 10- 6 cm 2/sec. When tubificid oli-
gochaetes were present, the transport could not be explained by assuming
an apparent diffusion coefficient that was greater than the molecular dif-
fusion coefficient over the entire tubificid life zone. They found that best
agreement was obtained by applying the two-layer model of equations
(18) and (19). The upper layer was 0.75 cm thick and constituted only
the pelletal layer, even though the mixed layer was 6-8 cm thick. By
setting DC1 - equal to 11 x 10- 6 cm2/sec at the sediment-water interface
and decreasing it linearly to 5.5 x 10- 6 cm 2/sec (the value in the absence
of benthos) at the base of the pelletal layer, they obtained excellent agree-
ment with the observed data.
Table III compares the ratios of the apparent diffusion coefficients in

Table III. Comparison of the Ratios of the Apparent Molecular Diffusivities in

the Mixed and Unmixed Layers

Ratio m
Dl/Dz (cm) Parameter Organism(s )/process Reference

3 5 Radon-222 Polychaetes Hammond et a1. (1975)

7 2 Radon-222 Polychaetes Hammond et a1. (1975)
7 8 Sulfate Nephtys incisa Goldhaber et a1. (1977)
100 3.5 Silica Waves and currents Vanderborght et al. (1977)
1.4 3 Ammonium Yoldia limatula Aller (1978)
1.4-3.5" 7-12 Radium-226 Cochran and Krishnaswami
(1980)
6 20.75 Sodium Mixed assemblage McCaffrey et a1. (1980)
1-2 a Chloride Tubifex tubifex McCall and Fisher (1980)

a Calculated as (D + KDB]ID (Schink and Guinasso, 1978).

320 Chapter 7

the mixed layer with the values of the molecular diffusivity in the un-
derlying sediments. As in Table II, these numbers should be normalized
to the biomass density, but insufficient data do not permit this modifi-
cation. The experimental studies of Aller (1978) and McCall and Fisher
(1980) agree very well with D1 /D z = 1.5, but the results from the field
studies are a factor of 1-67 higher than this value. This may be the result
of additional stirring of the upper most sediments in natural conditions
by processes other than one-dimensional vertical diffusion, such as radial
diffusion, water pumping, and winds and waves. Vanderborght et al.
(1977) showed that mixing by waves and currents in the upper 3.5 cm of
North Sea sediments (20 m depth) requires a physical mixing coefficient
in the upper layer of 10- 4 cmz/sec. This value is about 100 times greater
than that of molecular diffusion and demonstrates the important effects
of physical processes on the sediments in shallow-water environments.

3.1.2. Three-Dimensional Diffusion Models

Aller (1977) observed that permanent tube dwellings formed by sed-
entary benthos produced three-dimensional chemical gradients in the
pore waters and solid phases of sediment and increased the flux of ma-
terial across the sediment-water interface. He conducted laboratory and
field studies of Amphitrite ornata, a sedentary, surface deposit-feeding
polychaete, and Clymenella torquata, a sedentary, subsurface deposit-
feeding polychaete. He found that chemical effects caused by the structure
of the tube are the result of the distribution of bacteria around the tube,
the way the burrow is maintained, and a change in the geometry of dif-
fusion because of the presence of a hollow tube maintained at seawater
concentrations.
For ammonium, Aller (1977, 1980a) considers the burrows to be ver-
tical cylinders that are uniformly spaced and constantly irrigated with
overlying oxygenated seawater (Fig. 9). He notes that the exact geometry
of the burrows is three-dimensionally complex and time-dependent, but
that the sediment may be approximated as a mosaic of micro environments
represented on the average by a central irrigated vertical burrow and its
immediately surrounding sediment. The ammonium produced by a mi-
crobial activity diffuses toward areas of low ammonium concentration
(burrows and overlying water). Mathematically, he expressed these pro-
cesses at steady state in terms of cylindrical and vertical diffusion:

(-D) [a(raClr)] (aZC)z

+D - + R(z) = 0 (22)
r ar az

where D is the aqueous molecular diffusion coefficient of ammonium in

Mathematical Models of Bioturbation 321

r----
Figure 9. Schematic representation of the cylindrical burrow irrigation model of Aller (1977,
1980a). Note that diffusion occurs down the greatest concentration gradient, i.e., toward
burrows and the sediment-water interface. Burrows are assumed to be continuously and
instantaneously mixed with overlying water. After Berner (1980).

the sediment, r is the radial distance from the nearest burrow axis, and
R(z) is the rate of microbial production of ammonium.
Aller (1980a) solved equation (22), calculated the average ammonium
concentrations for discrete depth intervals, and found good agreement
between calculated and observed ammonium profiles. He concluded that
his burrow irrigation model more closely simulates the processes affecting
the solute distributions in the bioturbated zone than do other pore water
models. His model is particularly significant because it represents a case
322 Chapter 7

in which both diffusion and faunal water pumping are considered. He

treats the pumping passively by assuming instantaneous communication
between the burrows and the overlying water, an assumption for which
he has supporting observations.

3.2. Advection Models

In all the diffusion-reaction models presented in the previous sec-

tion, mass transport has been examined only as a diffusional process. Not
all of the processes are diffusional in nature. Some organisms pump and
direct the flow of water as well as sediment. Overlying water that is drawn
into burrows during irrigation may be mixed with surrounding pore
waters and returned to the overlying water (Fig. 10). Models of this process
require that solutes be transported by advection between one or two well-
mixed sediment reservoirs (Grundmanis and Murray, 1977; Hammond
and Fuller, 1979; McCaffrey et a1., 1980) or as a function of depth (Goreau,
1977) and overlying water.
Hammond and Fuller (1979) observed radon deficiencies in surficial
sediments of San Francisco Bay. They suggested that this might be caused
by the pumping of overlying water into the sediments through worm

Overlying Water °1+°2

Co

Zone
h1
C1

h2
Zone 2
C2 GQ 2

Figure 10. Worm pumping model of Hammond and Fuller (1979). Two sediment layers
defined by the radon concentration profiles receive overlying water during irrigation. This
water is mixed uniformly with the pore waters within each zone and fluxed to the overlying
water.
Mathematical Models of Bioturbation 323

tubes. In their model (Fig. 10) the sediments were divided into irrigation
zones defined by the radon concentration profile. Each zone was assumed
to be well mixed. The upper zone receives flow directly from the overlying
water and from the underlying zone, whereas the lower layer receives
advected water only from the overlying water. They characterized the
radon balances as

(23)

where C1,2 are the radon concentrations in Zones 1 and 2, h 1,2 are the
thicknesses of Zones 1 and 2, A is the area, P is the production rate of
radon from radium, x. is the decay constant for radon, /1,2 is the net loss
by molecular diffusion from Zones 1 and 2 per unit area, and Q1,2 are the
net rates of inflow of water to Zones 1 and 2. They calculated an irrigation
velocity from the observed radon concentration profiles and estimated
fluxes of nutrients owing to irrigation. The irrigation velocities were =3
cm/day, and the calculated fluxes about an order of magnitude greater
than can be attributed to molecular diffusion.
McCaffrey et a1. (1980) examined Narrangansett Bay sediments that
were homogenized to a depth of 20 cm. They determined that infaunal
bivalves, gastropods, amphipods, and some polychaetes rapidly mix the
upper 5-6 cm of sediment, and that sediment from 15-20 cm was recycled
to the sediment surface by the "conveyor belt" (Rhoads, 1974) species
Maldanopsis elongata, a holothurian. The rate of exchange between over-
lying and pore waters was determined by measuring the rate at which
22Na introduced into the overlying water was removed by transport into
pore waters and the vertical distribution of 22Na in the sediment. They
found that the transfer of tagged overlying water was far greater than
would be predicted from vertical molecular diffusion alone. They mod-
eled the transport of dissolved species in two ways. In both models they
assumed that species released below 25 cm are transported into the zone
of bioturbation by simple diffusion. In the first model, they assumed that
the species are transported across the sediment-water interface by en-
hanced diffusion. This is identical to the approach of Hammond et al.
(1975), Aller (1978), and McCall and Fisher (1980), discussed earlier. The
second model assumes that dissolved substances are transported across
the sediment-water interface by both simple diffusion and advection.
In the advection model for 22Na, McCaffrey et a1. (1980) assumed that
organisms pumped water at a constant rate independent of depth in the
sediment between the sediment and the overlying water. They treated the
process as an exchange between two well-mixed reservoirs: pore water
324 Chapter 7

and overlying water. The decrease of 22Na activity in the overlying water
was then described by

d~t(t) = - KCs(t) + K (~:) Cs(O) - K (~:) Cs(t) (24)

where Cs(O) is the normalized 22Na concentration in the overlying water

at time t = 0, Cs(t) is the normalized 22Na concentration in the overlying
water at any time t, K is the exchange constant, Vs is the volume of the
overlying water, and Vp is the volume of pore water. Similarly, Cp(t), the
normalized 22Na concentration in the pore water at any time t, may be
calculated:

Cp(t) = - (~:) [d~s/t)J (25)

K was determined as the fraction of overlying water assumed to be pumped

across the sediment-water interface that is necessary to account for the
observed decrease in Cs with time. The advection rate is equal to K(VsI
AJ, where A is the cross-sectional area of the sediment, and was found
to be 0.7 mllcm 2 per day. The advective flux is equal to the product of
the average increase in the concentration of a species in the pore water
relative to bottom water times the pumping rate. A comparison of the
mass transport calculated from the diffusion and advection models is
given in Table IV. Note that the pumping model does not include transport
across the sediment-water interface by molecular diffusion, whereas the
Hammond and Fuller (1979) model does. These calculations indicate that
transport by diffusion and by advection are of the same order (if the
assumption that the irrigated water mixes with pore waters is valid).
Smethie et al. (1981) reached a similar conclusion by examining 222Rn
in Washington continental shelf sediments.

Table IV. Comparison of Fluxes across the Sediment-Water Interface

Calculated from the Diffusion and Advection Models of McCaffrey et 01. (1980)
and the Advection Model of Hammond and Fuller (1979)

Diffusional flux a Advective flux a Advective flux b

Species (fLmollcm z per day) (fLmollcm z per day) (fLmol/cm z per day)

H4 Si0 4 0.3 0.3 0.6

NH! 0.19 0.07 0.4
pol- 0.018 0.02 0.003
lCOz 0.09 0.6 4.0
Mnz+ 0.02 0.01

a McCaffrey et a1. (1980) (Narragansett Bay).

b Hammond and Fuller (1979) (San Francisco Bay).
Mathematical Models of Bioturbation 325

Goreau (1977) suggests that a signal theory approach to the study of

sediments may be used to extract this information in the presence of
biological mixing. He shows that pore waters may be analyzed in a manner
similar to the approach he used for the bulk sediment. He defines a depth-
dependent mixing function for the pore water, Fw(x), as the volume of
pore water exchanged with overlying water per unit area per unit time
(hence his model has been loosely classified as an advection model).
Goreau shows that the flux of a dissolved species across the sediment-water
interface is given by

Flux [g/cm2 per secl = LX max Fx(x) [C(x) - Cwldz (26)

where C(x) is the concentration of the dissolved species at depth x and

Cw is the concentration of dissolved species in the overlying water. When
[C(x) - Cwl is positive the flux will be from the sediment to the water
column, and when [C(x) - Cwl is negative the net flux will be into the
sediment.
McCaffrey et a1. (1980) essentially employed equation (26) but treated
Fw(x) as the advection rate and a constant in depth, and C(x) as the mean
concentration in the mixed zone. Carefully controlled laboratory exper-
iments are necessary to obtain Fw(x) and permit the use of equation (26).
The mass transfer of dissolved species across the sediment-water
interface can be a factor of two or more higher than that predicted by one-
dimensional vertical diffusion alone. The results of Hammond and Fuller
(1979) and McCaffrey et a1. (1980) demonstrate that the additional flux
may be the result of the direct exchange of pore water and overlying water
during mixing. Goreau (1977) has presented a general formula for this
exchange. Aller (1977, 1980a) has indicated that the nature of the ex-
change may be diffusion in three dimensions into burrows followed by
mixing of overlying and burrow waters by irrigation (pumping). Clearly,
the nature of the exchange is dependent upon the activities of the ma-
crobenthos. As McCall and Fisher (1980) show, Tubifex tubifex does not
pump water, so that, in a sediment inhabited only by them, it does not
make sense to employ a pumping model. Indeed, they were able to account
for chemical fluxing across the sediment-water interface by one-dimen-
sional vertical diffusion alone.

4. Conclusions
The purpose of constructing mathematical models of bioturbation has
been, so far, to describe the redistribution of sediment particles caused
326 Chapter 7

by biological mixing and the enhancement of the chemical mass transfer

between sediment pore fluids and overlying waters. Particle mixing
models are applicable to stratigraphic problems. They may provide an
absolute time framework for the accumulation of the sediments and may
be used to determine the sedimentologic and postdepositional history of
the deposit. Pore water models are usually applied to chemical cycling
problems. They permit estimates of chemical fluxing across the sedi-
ment-water interface and/or rates of mineralogical and biogeochemical
reactions occurring in the sediment.
As we acquire a more detailed knowledge of processes that occur
near the sediment-water interface we learn that the animal-sediment-water
system is very complex. There are many ways of explaining experimental
and field observations. We may need more complicated models, but the
utility of the models may actually increase, because they can be used to
rationalize anomalous observations or limit the number of possible pro-
cesses producing the observed pattern. McCaffrey et 01. (1980) attempted
this (Table IV), but they did not employ a very sophisticated diffusion
model. If they had developed the diffusion model in the same manner as
Aller (1980a), they might have found that diffusion with irrigation could
also account for the observed net flux. Nevertheless, a detailed comparison
of different models on the same data should permit a more accurate eval-
uation of the actual processes that are occurring.
The differences between the models are more pronounced in the
particle transport problem, and such a comparison is even more important
in these situations. Failure to recognize biological mixing in sediments
has been common in the historical geochemical literature because some
of the parameters analyzed do not readily indicate a mixed layer. For
example, 210Pb is commonly uniformly distributed in the mixed layer and
then decreases exponentially below, while organic carbon and nitrogen
usually decrease continuously from the surface. This apparent discrep-
ancy is of fundamental importance when attempting to interpret the sed-
iment record. Clearly, the relative rates of sedimentation, mixing, decom-
position/decay, and depositional flux, as well as selective feeding, must
result in a homogeneous profile of some parameters in the mixed layer
and a profile that appears unmixed for other parameters. This question
must be quantitatively resolved.

ACKNOWLEDGMENTS. Juan C. Heinrich (Department of Aerospace and Me-

chanical Engineering, University of Arizona, Tucson, Arizona 85721) as-
sisted in the discussions and early writing of this chapter. This work was
supported in part by NSF Grant OCE-8005103 and NOAA Grant
NA80RAD00036. This is Contribution No. 140 from the Department of
Geological Sciences, Case Western Reserve University.
Mathematical Models of Bioturbation 327

References
Aller, R. c., 1977, The influence of macrobenthos on chemical diagenesis of marine sedi-
ments, Ph.D. dissertation, Yale University, New Haven, Connecticut, 600 pp.
Aller, R. C., 1978, The effects of animal-sediment interactions on geochemical processes
near the sediment-water interface, in: Estuarine Interactions (M. L. Wiley, ed.), pp.
157-172, Academic Press, New York.
Aller, R. C., 1980a, Quantifying solute distributions in the bioturbated zone of marine sed-
iments by defining an average microenvironment, Geochim . Cosmochim. Acta
44:1955-1965.
Aller, R. c., 1980b, Diagenetic processes near the sediment-water interface of Long Island
Sound. 1. Decomposition and nutrient element geochemistry (S, N, P), in: Estuarine
Physics and Chemistry: Studies in Long Island Sound (B, Saltzman, ed,), pp, 237-350,
Advances in Geophysics, Volume 22, Academic Press, New York,
Aller, R. C" and Cochran, J, K., 1976, 234Th/238U disequilibrium in near-shore sediment:
Particle reworking and diagenetic time scales, Earth Planet. Sci. Lett. 29:37-50,
Aller, R. C" and Dodge, R. E., 1974, Animal-sediment relations in a tropical lagoon, Dis-
covery Bay, Jamaica, j, Mar. Res. 32:209-232.
Aller, R. C" Benninger, L, K., and Cochran, J. K., 1980, Tracking particle associated processes
in nearshore environments by use of 234Th/238U disequilibrium, Earth Planet. Sci. Lett.
47:161-175.
Amiard-Triquet, c., 1974, Etude experimentale de la contamination par Ie cerium 144 et Ie
fer 59 d'un sediment a Arenicola marina L. (Annelide Polychete), Cah. Bio!. Mar.
15:483-494,
Benninger, L, K., Aller, R. C" Cochran, J, K., and Turekian, K, K., 1979, Effects of biological
sediment mixing on the 21°Pb chronology and trace metal distribution in a Long Island
Sound sediment core, Earth Planet. Sci. Lett. 43:241-259,
Benoit, G. J" Turekian, K. K., and Benninger, L. K" 1979, Radiocarbon dating of a core from
Long Island Sound, Estuarine Coastal Mar. Sci. 9:171-180,
Berger, W. H., and Heath, G, R., 1968, Vertical mixing in pelagic sediments, J. Mar. Res.
26:134-143,
Berner, R. A., 1980, Early Diagenesis: A Theoretical Approach, Princeton University Press,
Princeton, New Jersey, 241 pp,
Blanc-Lapierre, A., and Forte!, R., 1965, Theory of Random Functions 0, Gani, trans!.),
Volume 1, Gordon and Breach, New York, 443 pp.
Bruland, K. W., 1974, Pb-210 geochronology in the coastal marine environment, Ph.D. thesis,
University of California, San Diego.
Christensen, E, R., 1982, A model for radionuclides in sediments influenced by mixing and
compaction, J. Geophys. Res, 87:566-572.
Cochran, J. K., 1980, The flux of Ra-226 from deep-sea sediments, Earth Planet. Sci. Lett.
49:381-392.
Cochran, J, K., and Aller, R. C., 1980, Particle reworking in sediments from the New York
Bight apex-Evidence from the Th-234-U-238 disequilibrium, Estuarine Coastal Mar.
Sci. 9:739.
Cochran, J. K., and Krishnaswami, S., 1980, Radium, thorium, uranium, and 210Pb in deep-
sea sediments and sediment pore waters from the north equatorial Pacific, Am, J. Sci.
280:849-889.
Davis, R. B., 1974, Stratigraphic effects of tubificids in profundallake sediments, LimnoI.
Oceanogr, 19:466-488.
Davison, C" 1891, On the amount of sand brought up by Lobworms to the surface, GeoI.
Mag. 8:489-493,
328 Chapter 7

Demaster, D. J., Nittrouer, C. A., Cutshall, N. H., Larsen, 1. 1., and Dion, E. P., 1980, Short
lived radionuclide profiles and inventories from Amazon continental shelf sediments,
Eos 61:1004.
Duursma, E. K., and Gross, M. G., 1971, Marine sediments and radioactivity, in: Radioactivity
in the Marine Environment (Committee on Oceanography, National Research Council).
pp. 147-160, National Academy of Sciences, Washington, D.C.
Filipek, 1. H., and Owen, R. M., 1980, Early diagenesis of organic carbon and sulfur in outer
shelf sediments from the Gulf of Mexico, Am. J. Sci. 280:1097-1112.
Fisher, j. B., 1978, Effects of tubificid oligochaetes on sediment movement and the movement
of materials across the sediment-water interface, Ph.D. dissertation, Case Western Re-
serve University, Cleveland, Ohio, 132 pp.
Fisher, j. B., Lick, W. J., McCall, P. 1., and Robbins, j. A., 1980, Vertical mixing of lake
sediments by tubificid oligochaetes, J. Geophys. Res. 85:3997-4006.
Fox, D. L., Crane, S. C., and McConnaughey, B. H., 1948, A biochemical study of the marine
annelid worm, Thoracophelia mucronata, J. Mar. Res. 7:567--585.
Goldberg, E. D., and Koide, M., 1962, Geochronological studies of deep sea sediments by
the ionium/thorium method, Geochim. Cosmochim. Acta 26:417-450.
Goldhaber, M. B., Aller, R. C., Cochran, J. K., Rosenfeld, J. K., Martens, C. S., and Berner,
R. A., 1977, Sulfate reduction, diffusion and bioturbation in Long Island Sound sedi-
ments: Report of the FOAM Group, Am. J. Sci. 277:193-237.
Gordon, D. C., Jr., 1966, The effects of the deposit feeding polychaete Pectinaria gouldii on
the intertidal sediments of Barnstable Harbor, Limnol. Oceanogr. 11:327-332.
Goreau, T. j., 1977, Quantitive effects of sediment mixing on stratigraphy and biogeochem-
istry: A signal theory approach, Nature 265:525-526.
Guinasso, N. L., and Schink, D. R., 1975, Quantitative estimates of biological mixing rates
in abyssal sediments, J. Geophys. Res. 80:3032-3043.
Grundmanis, V., and Murray, J. W., 1977, Nitrification and denitrification in marine sedi-
ments from Puget Sound, Limnol. Oceanogr. 22:804-813.
Hakanson, L., and Kallstrom, A., 1978, An equation of state for biologically active lake
sediments and its implications for interpretations of sediment data, Sedimentology
25:205-226.
Hammond, D. E., and Fuller, c., 1979, The use of radon-222 as a tracer in San Francisco
Bay, in: San Francisco Bay: The Urbanized Estuary (T. J. Comomos, ed.). pp. 213-230,
American Association for the Advancement of Science, San Francisco.
Hammond, D. E., Simpson, H. j., and Mathieu, G., 1975, Methane and radon-222 as tracers
for mechanisms of exchange across the sediment-water interface in the Hudson River
Estuary, in: Marine Chemistry in the Coastal Environment (T. M. Church, ed.). pp.
119-132, American Chemical Society Symposium Series 18, American Chemical Soci-
ety, Washington, D.C.
Hanor, J. S., and Marshall, N. F., 1971, Mixing of sediment by organisms, in: Trace Fossils:
A Field Guide to Selected Localities in Pennsylvanian, Permian, Cretaceous, and Ter-
tiary Rocks (B. F. Perkins, ed.). pp. 127-135, School of Geoscience Miscellaneous Pub-
lication 71-1, Louisiana State University, Baton Rouge.
Imboden, D. M., and Stiller, M., 1982, The influence of radon diffusion on the Pb-210
distribution in sediments, J. Geophys. Res. 87:557-565.
Jumars, P. A., Nowell, A. R. M., and Self, R. F. 1., 1981, A simple model of
flow-sediment-organism interaction, in: Sedimentary Dynamics of Continental Shelves
(C. A. Nittrouer, ed.), Mar. Geol. 42:155-172.
Kadko, D., 1980a, 230Th, 226Ra, and 222Rn in abyssal sediments, Earth Planet. Sci. Lett.
49:360-380.
Mathematical Models of Bioturbation 329

Kadko, D., 19BOb, A detailed study of some uranium series nuclides at an abyssal hill area
near the East Pacific Rise at B045'N, Earth Planet. Sci. Lett. 51:115-131.
Kemeny, 1. G., and Snell, J. L., 1960, Finite Markov Chains, D. Van Nostrand, Princeton,
New Jersey, 210 pp.
Kowal, N. E., 1971, A rationale for modeling dynamic ecological systems, in: Systems Anal-
ysis and Simulation in Ecology (B. C. Patten, ed.), pp. 123-194, Academic Press, New
York.
Krishnaswami, S., Benninger, L. K., Aller, R C., and Van Damm, K. L., 19BO, Atomospher-
ically-derived radionuclides as tracers of sediment mixing and accumulation in near-
shore marine and lake sediments: Evidence from 7Be, 21°Pb, and 239. 240 pU, Earth Planet.
Sci. Lett. 47:307-31B.
Levinton, J. S., and Lopez, G. R, 1977, A model of renewable resources and limitation of
deposit-feeding benthic populations, Oecologia (Berlin) 31:177-190.
McCaffrey, R J., Meyers, A. C. Davey, E., Morrison, G., Bender, M., Luedtke, N., Cullen, D.,
Froelich, P., and Klinkhammer, G., 19BO, The relation between pore water chemistry
and benthic fluxes of nutrients and manganese in Narragansett Bay, Rhode Island,
Limnol. Oceanogr. 25:31-44.
McCall, P. 1., 1977, Community patterns and adaptive strategies of the infaunal benthos of
Long Island Sound, J. Mar. Res. 3B:221-266.
McCall, P. L., and Fisher, J. B., 19BO, Effects of tubificid oligochaetes on physical and
chemical properties of Lake Erie sediments, in: Aquatic Oligochaete Biology (R O.
Brinkhurst and D. G. Cook, eds.), pp. 253-317, Plenum Press, New York.
McCall, P. 1., Tevesz, M. J. S., and Schwelgien, S. F., 1979, Sediment mixing by Lampsilis
radiata siliquoidea (Mollusca) from Western Lake Erie, J. Great Lakes Res. 5:105-111.
Marzolf, G. R, 1965, Substrate relations of the burrowing amphipod Pontoporeia affinis in
Lake Michigan, Ecology 46:579.
Nowell, A. R M., Jumars, P. A., and Eckman, J. E., 19B1, Effects of biological activity on the
entrainment of marine sediments, in: Sedimentary Dynamics of Continental Shelves
(C. A. Nittrouer, ed.), Mar. Geoi. 42:133-153.
Nozaki, Y., Cochran, J. K., Turekian, K. K., and Keller, G., 1977, Radiocarbon and 210Pb
distribution in submersible-taken deep sea cores from Project Famous, Earth Planet.
Sci. Lett. 34:167-173.
Olsen, C. R, Simpson, H. J., Peng, T.-H., Bopp, R F., and Trier, R. M., 19B1, Sediment
mixing and accumulation rate effects on radionuclide depth profiles in Hudson estuary
sediments, J. Geophys. Res. 86:11020-1102B.
Peng, T.-H., Broecker, W. S., Kipphut, G., and Shackleton, N., 1977, Benthic mixing in deep
sea cores as determined by 14C dating and its implications regarding climate stratigraphy
and the fate of fossil fuel CO 2, in: The Fate of Fossil Fuel CO 2 in the Oceans (N. R
Andersen and A. Malahoff, eds.). pp. 355-373, Plenum Press, New York.
Peng, T.-H., Broecker, W. S., and Berger, W. H., 1979, Rates of benthic mixing in deep-sea
sediment as determined by radioactive tracers, Quat. Res. 11:141-149.
Piper, D. J. W., and Marshall, N. F., 1969, Bioturbation of Holocene sediments on La Jolla
Deep Sea Fan, California, J. Sediment. Petrol. 39:601-606.
Rhoads, D. C., 1963, Rates of sediment reworking by Yoldia limatula in Buzzards Bay,
Massachusetts and Long Island Sound, J. Sediment. Petrol. 33:723-727.
Rhoads, D. C., 1974, Organism-sediment relations on the muddy sea floor, Oceanogr. Mar.
BioI. Annu. Rev. 12:263-300.
Robbins, J. A., 197B, Geochemical and geophysical applications of radioactive lead, in:
Biogeochemistry of Lead in the Environment 0. O. Nriagu, ed.). pp. 2B5-393, Elsevier,
Amsterdam.
330 Chapter 7

Robbins, J. A., Krezoski, J. R., and Mozley, S. C., 1977, Radioactivity in sediments of the
Great Lakes: Post-depositional redistribution by deposit-feeding organisms, Earth
Planet. Sci . Lett. 36:325-333.
Robbins , J. A., McCall, P. L., Fisher, J. B., and Krezoski, J. R., 1979, Effect of deposit feeders
on migration of 1 37 CS in lake sediments, Earth Planet. Sci. Lett. 42:277-287.
Rowe, G. T., Palloni, P. T., and Horner, S. G., 1974, Benthic biomass estimates from the
northwestern Atlantic Ocean and the Northern Gulf of Mexico, Deep Sea Res.
21:641-650.
Ruddiman, W. F., and Glover, L. K., 1972, Vertical mixing of ice-rafted volcanic ash in North
Atlantic sediments, Bull. Geol. Soc. Am. 83:2817-2836.
Santchi, P. H., Li, Y.-H., Bell, J. J., Trier, R. M., and Kawtaluk, K., 1980, Pu in coastal marine
environments, Earth Planet. Sci. Lett. 51:248-265.
Schink, D. R., and Guinasso, N. L., Jr., 1977, Modelling the influence of bioturbation and
other processes on calcium carbonate dissolution at the sea floor, in: The Fate of Fossil
Fuel CO 2 in the Oceans (N. R. Andersen and A. Malahoff, eds.), pp. 375-398 , Plenum
Press, New York.
Schink, D. R. , and Guinasso, N. L., 1978, Redistribution of dissolved and adsorbed materials
in abyssal marine sediments undergoing biological stirring, Am. ]. Sci. 278:687-702.
Schink, D. R., and Guinasso, N. L., 1982, Processes affecting silica at the abyssal sedi-
ment-water interface, in: Actes des Colloques du C.N.R.S.: Biogeochimie de la Matiere
Organique a l'Interface Eau-Sediment Marin (R. Daumas, ed .) (in press).
Schink, D. R., Guinasso, N. L., Jr., and Fanning, K. A., 1975, Processes affecting the con-
centration of silica at the sediment-water interface of the Atlantic Ocean , ]. Geophys.
Res. 80:3013-3031.
Shokes, R. F., 1976, Rate-dependent distributions of lead-210 and interstitial sulfate in
sediments of the Mississippi River delta, Ph.D. thesis, Texas A&M University, College
Station, Texas.
Smethie, W. M., Jr., Nittrouer, C. A., and Self, R. F. L., 1981, The use ofradon-222 as a tracer
of sediment irrigation and mixing on the Washington continental shelf. in: Sedimentary
Dynamics of Continental Shelves (C. A. Nittrouer, ed.), Mar. Geol. 42:173-200 .
Spencer, D. W., 1975, Distribution of lead-210 and polonium-210 between soluble and
particulate phases in seawater, USAEC Report, COD-3566-11, Woods Hole Oceano-
graphic Institute, Woods Holes, Massachusetts, 105 pp.
Sundquist, E., Richardson. D. K., Broecker, W. S., and Peng, T.-H., 1977, Sediment mixing
and carbonate dissolution in the southeast Pacific Ocean , in: The Fate of Fossil Fuel
CO 2 in the Oceans (N. R. Andersen and A. Malahoff, eds.), pp. 429-454, Plenum Press,
New York.
Turekian, K. K., Cochran, J. K., and DeMaster, D. J., 1978, Bioturbation in deep sea deposits:
Rates and consequences, Oceanus 21:34-41.
United States Army Corps of Engineers, 1975, Lake Erie wastewater management study:
Preliminary feasibility report, Corps of Engineers, Buffalo District, Buffalo, New York.
Vanderborght, J. P., Wollast, R., and Billen, G., 1977, Kinetic models of diagenesis in dis-
turbed sediments. Part 1. Mass transfer properties and silica diagenesis, Limnol. Ocean-
ogr. 22:787-793 .
Wong, G. T. F .. and Grosch, C. E., 1978. A mathematical model for the distribution of
dissolved silicon in interstitial waters~An analytical approach. J. Mar. Res. 36:735-750.
Index

Amphipods, 116-118 Benthos effects on physical properties of

Amphitrite burrow walls, 199 freshwater sediments, 105-176
Amphitrite ornata, 70-71, 90, 320 ecological interactions and sediment
Aquatic nonmarine trace fossils , geological properties, 161-166
significance, 257-285 freshwater sediments and macrobenthos,
conclusions, 280-281 106-113
depositional environment, 271-273 future work,.166-167
paleoenvironmental information, macrobenthos life-styles, 113-124
usefulness, 278-280 sediment mixing, 124-150
paleoenvironmental reconstruction, sediment transport, biogenic
271-276 modification, 150-161
systematic information, usefulness, Biological benthic boundary layer, 4
277-278 Bioturbation, 4
taxonomic, lithologic, and Bioturbation, mathematical models,
paleoenvironmental diversity, 260-265 289-330
traces, identification of, 259-270 advection models, 322-325
trophic level reconstruction, 270-271 box models, 292
vectorial features, 273-276 conclusions, 325-326
Archanodon, 275 diffusion models, 292, 293
Azoic values, 36 diffusion-reaction models, 314
equations, 295, 296, 298, 299, 306, 307,
Bacterial colonization of surfaces, 27 309
Bacterially mediated diagenetic reactions, fluid transport models, 313-325
free energy change and schematic kinds of models and objectives, 291-293
chemistry, 182 Markov models, 310-312
Bed roughness , 19-24 one-dimensional vertical diffusion
bioturbational activity, 19-20 models, 314-320
blocks, 21 particle transport models, 293-312
fluids literature, 20-21 processes to be modeled, 289-291
natural tube problem, 21 signal-theory-based model, 292, 309-310
projecting tubes, 20 three-dimensional diffusion models,
shell material, 23-24 320-322
Belt Supergroup, 250 Bivalves , 120-124
Benthic invertebrates, classification and Pisidiidae, 123-124
terminology of feeding, 179 Unionacea, 120-123

331
332 Index

Black Sea sediments, 223 Ecological succession and geotechnical

Bottom communities, 5-7 properties, 10-19
Box models, 292, 305-309 conclusions, 17-19
equations, 306, 307 experimental design, 11
Branchyura sowerbi, 164 FOAM sediments, 15-16
Buchholzbrunnichnus, 249 FOAM study sites, 10
Bunyerichnus , 249 methods , 13-15
Burrow habitat, chemistry of, 89-92 NWC sediments, 16-17
animal adaptations, 91-92 NWC station, 11
metabolic activity , 90 results, 15-17
model equations, 89, 91 water content profiles, 19
Burrowing in muds, Buzzards Bay, 30-31 Ecological succession and organism, 5-10
bottom communities, 5-7
Celliforma spirifer, 266 equilibrium stages, 9-10
Chironomids , 113-116 pioneering stages, 7-9
sediment mixing, depth and rate of, Ediacaran fauna, 244
135-137 Equilibrium stages, 9-10
Chironomus americanus, 31 coexisting species, 10
Chironomus anthracinus, 163 late successional stage, 9
Chironomus plumosus, 136 sedimentary effects, 9-10
Chironomus riparius, 206 Erosion resistance and sediment water
Chironomus tentans, 206 content, 34
Chitinous tubes, 228
Chondrities, 229 "Far-field" phenomena, 18, 149
Clymenella torquata, 70-71, 320 Filter-feeding clams, 143-144
Cohesive sediment erosion, 32 Fluid bioturbation, 37
Coquinas, 226 Fluid transport models, 313-325
Corbicula fluminea, 144 advection models, 322-325
Critical excitation velocity, 28 diffusion-reaction models, 314
Cruziana facies, 236 equations, 315, 317, 320 , 323, 324, 325
one-dimensional vertical diffusion
models, 314-320
Diagenetic reactions, 54-56 three-dimensional diffusion models,
biogeochemical successions , 55 320-322
macrobenthos influence, 56 FOAM study, see Ecological succession and
Decomposition reactions, idealized, 54 geotechnical properties
Detritus-feeding invertebrates, potential Freshwater sediment chemical diagenesis ,
food, 24-25 macrobenthos effects on, 177-218
Diffusion models, 292, 293-305 chemical diagenesis of sediments,
equations, 295, 296, 298, 299 182-186
Diplocraterion, 236 conclusions, 209-211
Dominant profundal macrobenthos, trophic diagenetic rates, 204-205
type and life habitat, 180 fluid transport, 189-194
Dreissena clam, 163 freshwater sediment system, 178-186
macrobenthos, 178-182
Eau Claire Formation, 245 materials flux, 205-209
Ecological interactions and sediment mechanism affecting chemical
properties, 161-166 diagenesis, 186-199
fauna , spatial-temporal variations, model equations, 187, 189, 190, 194
161-162 observed effects on chemical diagenesis ,
macrofauna interactions, 163-166 200-209
Index 333

Freshwater sediment chemical diagenesis , Lampsilis radiata siliquoidea, 199

macrobenthos effects on (cont.l Lepidurus, 273
particle transport, 186-189 Lepidurus arcticus, 266
reactive materials addition, 199 Limnodrilus cervix, 207
sediment fabric alteration, 194-199 Limnodrilus hoffmeisteri, 163, 164, 207
sedimentary chemical milieu, 200-204 Green Bay, Michigan, 130
Freshwater sediments and macrobenthos, Toronto Harbor, 163
106-113 Littoral community, 108
lacustrine and marine environments, 106 Lumbrineris-Alpheus-Diolodonta
salinity fluctuations and tidal motions, community, Kingston Harbor, Jamacia,
106 9
soft-bottom communities, 108
Freshwater sediment physical properties, Macoma baltica, 29
benthos effects on, 105-176 Macrobenthos effects on chemical
ecological interactions and sediment diagenesis in freshwater sediments,
properties, 161-166 177-218
freshwater sediments and macrobenthos, chemical diagenesis of sediments,
106-113 182-186
future work, 166-167 conclusions, 209-211
macrobenthos life-styles, 113-124 diagenetic rates, 204-205
sediment mixing, 124-150 fluid transport, 189-194
sediment transport, biogenic freshwater sediment system, 178-186
modification, 150-161 macrobenthos, 178-182
materials flux, 205-209
Gammarus, 117 mechanisms affecting chemical
Geotechnical mass properties, 29-36 diagenesis, 186-199
Glycoproteinaceous films, 27 model equations, 187, 189, 190, 194
Great Slave Lake, Canada, 111-112 observed effects on chemical diagenesis,
200-209
"Head-down" conveyor belt feeders, 9, particle transport, 186-189
125, 188, 297 reactive materials addition, 199
Helminthoidichnites, 250 sediment fabric alteration, 194-199
Heteromastus filiformis, 21 sedimentary chemical milieu, 200-204
Hexagenia mayfly nymph, 140 Macrobenthos effects on chemical
High surface-area-to-volume ratios, 25 properties of marine sediments and
"Hummocky" relief on surface, 24 overlying water, 53-101
Hyalella azteca, 116-117 burrow habitat, chemistry of, 89-92
Hydrobia sp. pellets, 28 diagenetic reactions, 54-56
macrofauna I influence on
Ichnogenera, 246 sediment-water exchange rates, 83-87
Ichnotaxa, 268 model equations, solutions, 94-95
diverse, 277 reaction rates, 87-89
Infaunal deposit feeders, 31 reactive particle redistribution, 56-71
Isopodichnus, 268 sediment chemistry, spatial and
temporal patterns, 92-93
Lacustrine and marine environments, 106 solute transport, 71-83
Lake Erie, Western Basin, 141-143 Macrobenthos life-styles, 113-124
box cores, 152, 155 amphipods, 116-118
Lake Esrom, Denmark, 108 bivalves, 120-124
Lake Miilaren, Sweden, 111-112 chironomids, 113-116
Lake V6rtsjiirv, Estonia, 110, 111 oligochaetes, 118-120
334 Index

Macrobenthos life-styles (con!.) Nearshore marine muds, geotechnical and

Pisidiidae, 123-124 transport properties, 38
Unionacea, 120-123 Nephtys incisa, 11
Maldane-Amphiura community, Ria de Nucula annulata, 11
Muros, Spain, 9 Nucula-nephtys community, Buzzards
Maldanid-Nucula-Syndosmya Bay, 9
community, Clyde Sea, 9
Maldanidae polychaete family, 9 Oceanic water chemistry, 184
Margaritifera margaritifera, 275 Oligochaetes, 118-120
Marine and lacustrine environments, 106 depth and rate of sediment mixing,
Marine benthos effects on sediment 127-135
physical properties, 3-52 feeding rate, 131
bioturbation, 4 Ophiomorpha, 228
equilibrium stages, 9-10 Organic-mineral aggregates, 27
future work recommendations, 40-43 Organism-sediment-fluid interactions, 37
invertebrate benthos, 4 Owenia filiformis, 21
laboratory experiments, 10-19 Owenia flume experiment, 22
ultrasound imaging, 41
Marine biogenic sedimentary structures,
geological significance, 221-256 Paleobathymetry, 233-243
paleobathymetry, 233-243 graphoglyptid controversy, 241-243
Precambrian traces, 243-252 ocean evidence, 238-241
traces and sedimentology, 222-233 Seilacher's model, 234
Marine sediment and overlying water, problems with, 234-238
chemical properties, macrobenthos Paleodictyoa, 241
effects on, 53-101 Pellet textures, 24-29
burrow habitat, chemistry of, 89-92 Peloscolex multisetous, 163
diagenetic reactions, 54-56 Physiochemical properties, interface,
macrofaunal influence on 32-34
sediment-water exchange rates, 83-87 Pioneering stages, 7-9
model equations, solutions, 94-95 adaptive effects, 8-9
reaction rates, 87-89 pore water chemistry, 8
reactive particle redistribution, 56-71 sedimentary effects, 8-9
sediment chemistry, spatial and Pisidiidae, 123-124
temporal patterns, 92-93 depth and rate of sediment mixing, 137
solute transport, 71-83 Planolites, 248 , 250
Markov models, 310-312 Planolites rugulosus, 268
Microbially mediated decomposition Pomatothrix, 163
reaction rates, 87-89 Pontoporeia, 117
Midges, 113 Pontoporeia hoyi, 137, 197
Mississippi Delta sedimentation, 223 Pore water irrigation, 31
Molluscs, Movice Lake, 163 Precambrian traces , 243-252
Molpadia-Euchone community, Cape Cod Alpert's proposal, 246-248
Bay,9 chronology, 248
Molpadia oolitica, 20 oldest trace fosil, 248-252
Mt. Simon foundation, 245 Precambrian-Cambrian boundary,
Mucopolysaccharides, 28-29 243-248
Mucus production and degradation, trace fossil zonation, 247
seasonal changes, 28 Profundal community, 109
Mud facies, Buzzards Bay, 20
Mulinia lateralis, 11 Radionuclide distribution, 295
Index 335

Reactive particle redistribution, 56-71 Sediment physical properties, effect of

biogenic reworking and decay rate, 64 marine benthos on (cont.)
homogeneous reworking, 56-67 organism-sediment-fluid interactions,
model equations, 57, 59, 60, 61 37
organic matter subduction, 66 pelletal textures, 24-29
particle eddies, 59 physical properties, 4, 151
particle motion, 58 pioneering stages, 7-9, 36
particle transport, isotopes, 60 qualitative predictive model, 36-40
representative particle reworking sediment relations, 5-10
coefficients, 62-63 sediment transport and biogenic activity,
resuspension and current transport, 64 19-36
selective reworking, 67-71 substratum stability, 20-23, 161
vertical properties distribution, 61 ultrasound imaging, 41
Redox potential discontinuity (RPD), 8-10 water content and erosion resistance, 34,
Rhizocorallium, 236 153
Rhoads-Cande Profile Camera, 41 Sediment transport and biogenic activity,
Rugoinfractus, 250 19-36
Rusophycus, 246 bed roughness, 19-24
cohesive sediment erosion, 32
fluids literature, 20-21
Scoyenia association, 271 geotechnical mass properties, 29-36,
Seaflume, in situ, 36 38-39
Sediment chemistry, spatial and temporal pellet textures, 24-29
patterns, 92-93 substratum stability, 20-23, 161
Sediment mixing, 124-150 tube fields, 20, 23
depth and rate, 127-137 Sediment transport, biogenic modification
feeding rate, 131 of, 150-167
kinds of and factors affecting, 124-127 anoxia effects, 159
particle size distribution effects, biological activities, importance in, 150
139-146 comparisons, 160-161
permeability, 149-150 field observations, 150-156
sediment stratigraphy, 124-139 future work, 166-167
sediments, mass property effects, laboratory experiments, 26-27, 157-160
146-150 macrofauna, spatial-temporal variations,
stratigraphic effects, 137-139 161-167
tubificid-induced subduction velocities, substratum properties, 161
132-133 summer-fall contrast, 156
water content, 146-149 Sediment water content, and erosion
Sediment physical properties, effect of resistance, 34, 153
marine benthos on, 3-52 oxygen content of water, 156
bed roughness, 19-24 Shrimp mounds, Discovery Bay, Jamaica,
biological benthic boundary layer, 4 20
bioturbation, 4, 34 Signal processing models, se Signal-theory-
cohesive sediment erosion, 32 based model
equilibrium stages, 9-10 Signal-theory-based model, 292, 309-310
fluid shear stress, 4-5 equations, 309
future work recommendations, 40-43 Skolithos-Planolites couplet, 236
geotechnical mass properties, 29-36, Solute transport, 71-83
38-39 apparent diffusion, 71-75
laboratory experiments, 10-19, 157-160 apparent transport coefficients, 75
macrofauna I effects, 5, 163-166 average diffusion geometry, 77-83
336 Index

Solute transport (cont.) Tubifex tubifex, 196

biogenic advection, 75-77 Toronto Harbor, 163
conclusions, 74 Tubifjcid life histories, 119-120
model equations, 72, 78 Tubifjcid oligochaetes
muddy sediments, 76 freshwater environment, 188
Species, identifying characterizing, 6 pelletizers, 141
Spriggina, 249
Stylodrilus heringianus, 195, 197 Upper Cambrian trilobites, 245
Subtidal mud erodibility, Long Island Upper Devonian shelf. New York, 226
Sound, 35 Unionacea, 120-123
Swedish entrophic lakes, 206 depth and rate of sediment mixing, 137

Taenidium, 270 Water content of sediment, and erosion

Taxonomic resolution, 277 reistance, 34, 153
Teichichnus, 236 oxygen content of water, 156
Temporal mosaics, taxonomic and Water exchange rates, macrofaunal
functional structure, 7 influence on sediment, 83-87
Terebellina, 228 model equations, 84, 86
Terrigenous sediments, Mud Bay, South solute flux ratio, 85
Carolina, 73 Western Basin, Lake Erie, 116
Traces and sedimentology, 222-233
bioturbation in recent muds, 223-225
Xenohelix, 268
diagenesis, 230-233
sedimentation, rate and continuity,
225-227 Yoldia limatula, 11
substrate consistency, 227-230
Transenella tantilla, 24 Zoophycos facies, 23G