Limnol. Oceanogr.

, 31(3), 1986, 478490

0 1986, by the American Society of Limnology and Oceanography, Inc.

Phytoplankton control by grazing zooplankton: A study on the

spring clear-water phase l
Winfried Lampert, Walter Fleckner, Hakumat Rai, and Barbara E. Taylor2
Department of Physiological Ecology, Max Planck Institute of Limnology,
Postfach 165, 2320 PlSn, W. Germany

Abstract
We tested the hypothesis that a clear-water period, regularly observed in many meso- and
eutrophic lakes, is caused by grazing herbivorous zooplankton. Such a clear-water phase occurs
during mid-May in the moderately eutrophic Schiihsee and involves a rapid increase in Secchi
transparency, and a drop in chlorophyll and particulate organic carbon in size fractions ~35 Nm.
Maxima of zooplankton biomass and community grazing rates (170% of volume cleared per day)
coincided with the greatest transparency. The algal decline was not related to nutrient depletion
or climatic events. Before the clear-water phase small phytoplankton contributed up to 88% of the
primary production, but the contribution of large particles was more important after the zooplank-
ton maximum. The effects of herbivory by zooplankton were examined in a series of time-over-
lapping enclosure experiments. Concentrations of small (< 35 pm) particles were always higher in
the bags lacking zooplankton than in the controls. A mass development of small algae occurred
in the zooplankton-free bags initiated during the clear-water phase, although the presence of zoo-
plankton stimulated the growth of large (> 35 pm) algae.

A distinct period of clear water is typical “spring bloom.” Subsequent increases in

of the spring algal succession in many me- biovolume occur first in the larger size
sotrophic and eutrophic lakes of temperate classes, while small particles remain low in
zones (Sommer et al. 1986). An early peak abundance.
of small, rapidly growing algae (flagellates, Although it is a widespread phenomenon
small diatoms), which often represents the (for a collection of examples see Lampert
highest biomass concentration of the annual 1985; Sommer et al. 1986), a clear-water
cycle, is followed by a short period of very phase is only apparent when sampling is
clear water and high Secchi transparency. A frequent (once or more a week) (H. Berner-
summer community dominated by large al- Fankhauser pers. comm.). A number of sea-
gae and blue-greens develops after the clear- sonal records of phytoplankton biomass
water phase. The period of high Secchi show clear spring depressions (e.g. Blaau-
transparency sometimes lasts only 2 weeks boer 1982; Bailey-Watts 1982; Riemann
or less. The increase in water clarity can be 1983; Barko et al. 1984; Meffert and Over-
spectacular. In Lake Constance for example beck 1985) but do not mention this explic-
transparency may increase from 1 to 10 m itly.
in a few days (Lampert 1978a; Schober 1980; The clear-water phase usually coincides
Tilzer 1983). The timing of the event can with a spring peak of the filter-feeding zoo-
be very predictable and relatively indepen- plankton and the increase in water clarity
dent of weather conditions (Geller 1980; has sometimes been attributed to grazing
Sommer et al. 1986). activity (Lampert and Schober 1978). Al-
The clear-water phase in Schohsee, oc- ternatively nutrient depletion, climatic
curring in mid-May, clearly demonstrates events, or parasitism have been suggested
the phenomenon (Fig. 1). The Coulter as causes for the algal crash (cf. Reynolds
Counter analysis of biovolume and particle 1984). Although there are several descrip-
distribution shows a sharp drop of the bio- tions of the time-course of phyto- and zoo-
volume of all particles present after the plankton abundance during this period of
the year (e.g. Krambeck et al. 1978; Oskam
1978; Petersen 1983; Kuparinen et al. 1984),
l This research was supported by the Deutsche For- the causes of the clear-water phase have not
schungsgemeinschaft through a grant to W.L.
2 Present address: Savannah River Ecology Labo- been analyzed directly. We here test the hy-
ratory, Drawer E, Aiken, South Carolina 2980 1. pothesis that zooplankton grazing causes the
478
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase

Fig. 1. Particle distribution as determined by Coulter Counter during spring 1983 in Schijhsee (2-m depth).

clear-water phase and initiates the succes- and B. Richter for technical assistance, and
sion toward larger algal species during sum- C. S. Reynolds, T. M. Frost, and W. R.
mer. We give a detailed description of the DeMott for comments on a first draft of the
clear-water phase in a mesotrophic lake, in- manuscript.
cluding information on abiotic factors, bio-
mass parameters and composition of phy- Met hods
toplankton, and on species composition and Study site-The study was carried out
biomass of the herbivorous zooplankton. We during spring 1983 in Schohsee, a small (80
discuss the possibility that the algal crash ha; max depth, 30 m; mean depth, 10.70
may be caused by nutrient depletion or cli- m), moderately eutrophic, hard-water Hol-
matic events. By measuring the patterns of stein lake near Plan. Details of its hydro-
primary production and grazing in situ, we graphic conditions, chemistry and produc-
try to balance the growth and loss param- tivity were given by Rai (1982). The
eters in the phytoplankton. Finally, we use sampling site was in the center of a 13-m
enclosures to provide a direct experimental deep bay close to the Max Planck Institute;
test of the effects of zooplankton. tests on several occasions had shown that
We thank W. Gabriel for help with the plankton at the sampling location did not
computer work, M. Volquardsen, H. Buhtz, differ significantly from that at the center of
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
480 Lampert et al.

the lake. Experimental enclosures were also cucullata, where we used our own length-
located in this bay, at a site where the water weight relationship:
was 6 m deep.
w = 9 . 12L3.03
Sampling-Water samples taken at 0, 1,
2, 3, 5, 7, and 11 m were analyzed for pH, ( W = dry weight in pg, L = length in mm
alkalinity, TIC, DOC, nitrite, nitrate, am- from the top of the eye to the base of the
monium, soluble reactive phosphorus, total tail spine). From depth profiles of daytime
phosphorus, and silica. Temperature and and nighttime zooplankton distributions, we
chlorophyll alpheopigments (Marker et al. estimated the amount of vertical migration
1980) were measured at 1-m intervals. Sam- in the zooplankton population and calcu-
ples for particulate carbon analysis collected lated the average zooplankton density by
at 0, 2, and 11 m (control determinations weighting the numbers by day length.
showed a gradual decrease of POC with Plastic bag experiments-Enclosure ex-
depth) were screened through meshes of 10, periments were carried out in plastic bags
35, and 250 pm, filtered onto precombusted (Kuiper 1977) 1 m in diameter and about
glass-fiber filters and burned in an infrared 3 m long, closed at the bottom and filled
carbon analyzer (Krambeck et al. 198 1). with 1.7 m3 of water. Four of these bags
Phytoplankton from the same depths as were hung in a rectangular aluminum frame
POC were preserved with Lugol’s solution moored over a water depth of 6 m. Two of
and documented photographically. the four bags were used as controls, two
In situ primary production and grazing- were treated. Preliminary experiments (W.
Size-fractionated primary production was Fleckner unpubl.) had shown that com-
estimated by the 14C method (Rai 1982). munities in all four bags when untreated
The incubation period was about 4 h. The developed similarly. In the long term (4-6
ratio of the global radiation integrated over weeks), however, they did not mimic the
the whole day to the sum of radiation during lake communities, because nutrient supply
the incubation period was used to estimate from external sources was prevented, re-
daily gross production (Wetzel and Likens cruitment from resting stages and immigra-
1979). tion was impossible, and fish predation was
In situ grazing rates were measured in lacking; therefore we ran these experiments
vertical series of nine depths (0, 1, 2, 3, 4, for only 10 days. Every week one treatment
.5, 7, 9, 11 m) by day and night. We mea- and one control were renewed, so that the
sured grazing rates on two radioactively la- experiments overlapped by 3 days.
beled particles, the green alga Scenedesmus Treatment and control bags were filled
acutus (10 pm) and the very small blue- with water pumped from 1-m depth. Before
green Synechococcus elongatus (1 pm) using the water entered the bags, it was filtered
a Haney (197 1) chamber. The grazing rates through 100~pm plankton gauze to remove
on 32P-labeled Synechococcus and 14C-la- all the larger zooplankton. The control bag
beled Scenedesmus estimate community was then inoculated with zooplankton
grazing rates on picoplankton and on nan- caught by vertical hauls of a 100~pm plank-
noplankton; techniques are described in ton net. It was not possible to stock the bags
more detail by Lampert and Taylor ( 198 5). with the same density of zooplankton as in
Zooplankton biomass -Before each graz- the lake, because the zooplankton abun-
ing experiment we took two samples of water dance in the epilimnion changes according
(6 liters total) with the Haney chamber, fil- to diel vertical migration. By simply filling
tered the zooplankters onto a 1OO-pm screen the control bags with unfiltered surface water
and preserved them in sucrose-Formalin we would have overstocked them at night
(Haney and Hall 1973). All zooplankton was but understocked them by day. We there-
counted under a dissecting scope; copepods fore used the same catching effort for all
were separated into stages, daphnids were control bags, so that the zooplankton den-
sized by an ocular micrometer. Literature sity in the bags changed in the same relative
values (Bottrell et al. 1976) were applied for fashion as in the lake.
biomass calculations except for Daphnia Bags were sampled immediately after fill-
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase 481

0 I I-

Fig. 3. Concentrations of Chl a (pg liter-‘) in

Schijhsee during spring 1983.

fractions exclusively, while the 35-250-pm

fraction increased during the clear-water
phase. Loss of algae was also indicated by
decreased concentrations of Chl a (Fig. 3).
Chlorophyll a decreased initially in the sur-
I- 8- face waters and reached concentrations of
3' 6- < 1 pg liter-’ during the time of maximum
7 - transparency.
3 4- The decrease of POC in the epilimnion
> _ 7

i5 2- C cannot be attributed to mixing in the water

.+:+
..:.:.: column; on the contrary, stratification start-
OL .+:.:.
APRIL MAY ed at the end of April and steadily became
stronger (Fig. 4). A strong wind disturbed
Fig. 2. Clear-water phase 1983 in SchShsee. A.
Concentration of particulate carbon (mg liter- ‘) in dif- the thermal structure near the surface on 28
ferent size fractions. Numbers indicate the mesh size May, but by then the clear-water phase was
&m) through which the water had been sieved before already coming to an end.
the carbon determination. B. Secchi transparency (m). The main nutrients are shown in Fig. 5.
C. Biomass of dominant zooplankton (g m-2). Open Soluble reactive phosphorus was low in ear-
area: Daphnia spp.; stippled area: Eudiaptomus spp.
ly April, since the SRP pool had been ex-
hausted by a diatom maximum after ice-
break in mid-March. It decreased slowly
ing, 3 days later, and at the end of the ex- during our sampling period. Total phos-
periment (day 9 or 10). Water was taken
from 1 m and subjected to the same ana-
0
lytical procedures as the lake water. Zoo- 6
plankton was collected by vertical tows of -2
a lOO-pm plankton net (25-cm diam) from E
-4
the bottom of the bag to the surface.
I i-r--- n””
Results
F
As in other years (Lampert unpubl.) the C
Secchi transparency in Schijhsee increased
;
in mid-May to >8 m (Fig. 2B). A clear- :
water phase lasted from about 10 May to 1 12
June. Increased transparency was caused by APRIL ’ MAY ’ JUNE
a rapid decline of particulate organic carbon Fig. 4. Isotherms (“C) in Schijhsee during spring
(Fig. 2A). Losses appeared in the small size 1983.
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
482 Lampert et al.

and cryptomonads. Large forms (Ceratium,

Closterium, Staurastrum, Sphaerocystis,
Fragillaria) became abundant in June.
The clear-water phase was accompanied
by significant shifts in the level and distri-
bution of primary production (Fig. 6). Dur-
ing mid-April production was high and the
shape of the curve was typical for a eutro-
phic lake. Production was lowest in mid-
May, but a considerable part of it took place
in deeper waters, in apparent response to
the high water transparency as in an oli-
gotrophic lake. Until 10 May small cells fixed
most of the carbon (about 88%); particles
between 35 and 250 pm contributed little.
This changed during the second half of May,
when with increasing total production large
particles accounted for an increasing por-
tion of it. Differences in the size structure
of primary production are especially clear
- APRIL’ MAY ’ JUNE between 13 April and 7 June.
Fig. 5. Nutrients in the epilimnion of SchGhsee(1
m) during spring 1983. Upper panel: nitrate and am- The low primary production during May,
monium (pg N liter-‘). Lower panel: total phosphorus however, was a consequence not only of the
and soluble reactive phosphorus (pg P liter-‘). relatively low biomass but also of lower light
intensity. The daily sum of global radiation
dropped from about 200 J cm-2 d- l at the
phorus (not including particles >250 pm) end of April to < 100 in mid-May. The
disappeared at a higher rate. Nitrate de- weather was particularly bad on 24 May (37
creased continuously until early June, with J cmm2 d-l), but improved afterward with
a particularly sharp drop in the first week about 250 J cm-2 d-l from 31 May until
of June; this has been repeated in Schijhsee the end of the study period (H.-J. Krambeck
for several years (Rai unpubl.). Concentra- unpubl. data).
tions of ammonium are usually very low in The change in the size structure of pri-
the lake, so that the short pulse during the mary production is clearly demonstrated
clear-water phase was especially noticeable. when production : biomass ratios of the dif-
At the same time, a short pulse could also ferent size classes calculated from fraction-
be seen in the soluble reactive phosphorus. ated primary production and particulate
This peak was very small and exceeded the carbon are compared (Fig. 7). Until the end
normal variability only slightly, but it was of the clear-water phase small particles had
reflected in the total phosphorus. The peaks the greater P : B ratios; subsequently the ac-
in both ammonium and phosphorus coin- tivity of the large fractions became higher.
cided with the maximum zooplankton One must, however, consider that this P : B
abundance, suggesting that they were caused ratio does not reflect the activity of live cells
by the grazing herbivores. There was no un- alone. It measures the rate of increase of the
usual shortage of algal nutrients during the particulate carbon pool and, thus, it de-
clear-water phase; the availability of nu- pends to a certain extent on the amount of
trients to the algae may even have been nonphotosynthesizing detritus present. It is
higher than before this period. likely that the relatively lower P : B ratios
Phytoplankton consisted mainly of small of the small fractions after the clear-water
cells (Rhodomonas, Stephanodiscus) during phase were caused by a high proportion of
April. During the clear-water phase the small detrital particles.
community was characterized by a few large Dominant species among herbivorous
cells or colonies (Ceratium, Sphaerocystis) zooplankton were the Calanoid copepods
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase 483

02468024602024024

PHOTOSYNTHETIC RATE ( mg C mB3h” )

Fig. 6. Profiles of primary production in different size fractions in Schiihsee during spring 1983. The outer
line indicates total carbon fixation. The stippled area represents the size fraction between 35 and 250 pm. The
lo-urn size fraction is indicated as the left line within the open field. Note the increasing share of the large
fraction.

Eudiaptomus gracilis and Eudiaptomus Grazing in the epilimnion of Schohsee

graciloides, and the cladocerans Daphnia shows pronounced diel fluctuations related
galeata and Daphnia hyalina. A third to the vertical migration of large zooplank-
Daphnia species, D. cucullata, contributed ton (cf. Lampert and Taylor 1984, 1985).
little to the biomass. Rotifers were not During the day, when the zooplankton is
abundant, although the smaller species may primarily in deep waters, grazing in the up-
have passed through the loo-pm mesh. Be- per 5 m was low, but very high grazing rates
cause their contribution to the biomass was were measured at night when zooplankton
negligible, they were excluded from the bio- was abundant in the epilimnion. Maximum
mass calculations as were nauplii and small values approached 10% of the water volume
copepodites of cyclopoid copepods (e.g. cleared per hour. We calculated the average
Thermocyclops oithonoides). Daphnids daily grazing rates for the upper 5 m and
(primarily D. galeata), accounted for most the average zooplankton density from day
of the steep rise in zooplankton biomass and night profiles as a function of daylength.
during the clear-water phase. Copepod bio- The daily in situ community grazing rates
mass showed little change (Fig. 2C). The tracked the zooplankton biomass (Fig. 8).
daphnid population grew exponentially and We recorded the highest grazing rates when
peaked on the day of maximum water trans- zooplankton biomass was at its peak. These
parency. The graphs of Secchi transparency highest rates were also concomitant with the
and zooplankton biomass nearly mimic each period of maximum transparency. They
other. amounted to 170% per day. Grazing pres-
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
484 Lampert et al.

n
35-250
o-‘~--‘-- o
1 - 0.8

- 0.6 E
z

_ Y
a
- 0.4 3

- :
-0.2 z

I
0 1 0
I
’ APRIL’ MAY ’ JUNE Fig. 8. Comparison of zooplankton biomass (mg
dry wt liter-l) (stars) and community grazing rate (%
Fig. 7. Change of the ratio primary production to d-l) (bars). The whole columns indicate grazing rate
particulate carbon (P : B) for small and large particles on Scenedesmus; the stippled parts represent grazing
(size in pm indicated) during spring 1983 in Schohsee. on the very small Synechococcus. Grazing rates and
Primary production calculated for midday incubation biomasses are daily averages for the upper 5 m of water,
period of 4-5 h, integrated over the upper 5 m of water. calculated from day and night measurements.

sure on very small particles (1 pm) was con-

sistently lower, due to the relatively low tial biomass declined as well as the relative
feeding efficiency of Eudiaptomus on small proportion of Daphnia. Once enclosed,
particles (Lampert and Taylor 1985). The zooplankton populations changed quanti-
period of extremely high grazing was rather tatively and qualitatively. In the first ex-
short. By mid-June it had approached typ- periment, when the initial biomass was
ical summer values of about 20% (cf. Lam- highest, both daphnids and diaptomids de-
pert and Taylor 198 5). creased during the 10 days of exposure. The
Population dynamics of the zooplankton second experiment started with lower den-
was strongly affected by the algal biomass. sities; the biomass increased first and then
During April egg numbers of the cladocer- decreased again. In all other experiments
ans were very high and, since the birth rate the biomass started from low levels and in-
was high, the population consisted of many creased continuously, mostly due to the
juvenile animals. Probably the population daphnids.
received some input of juveniles from rest- In all zooplankton-free bags, without ex-
ing stages as well. When the juveniles grew ception, the concentration of POC and Chl
up the biomass of the Cladocera increased a in the size fraction < 3 5 pm was higher at
(cf. Fig. 2C), grazing pressure increased, and the end of an experiment than in the bags
the algal biomass decreased. During the with grazers (Fig. 10). During the clear-water
clear-water phase daphnids were strongly phase the concentrations of small particles
food-limited. Adults carried very few eggs; increased in the bags without grazers, while
most had no eggs at all. As a consequence the concentration in the zooplankton bags
the birth rate dropped, the size structure of remained at the same level as in the lake.
the claphnid population shifted to large size Later in the year differences between bags
classes, and the population decreased. became less pronounced, as was to be ex-
Only the initial density and composition pected because the zooplankton densities
of the zooplankton population in each plas- were much lower (Fig. 9). The dynamics of
tic bag experiment reflected the lake pop- the particles in Fig. 10 are not directly com-
ulations (Fig. 9). We started during the parable to the values in Fig. 2, since the
clear-water phase with a high zooplankton samples were taken at a different location
biomass consisting mainly of daphnids. close to the bags in a shallower part of the
During the successive experiments the ini- lake.
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase 485

23 MAY 31 MAY Discussion

Periods of very clear water that were ev-
idently caused by mass developments of
Daphnia were first described for sewage
treatment ponds (Uhlmann 1971). Such
clear-water periods are typically of short du-
ration, although they may be prolonged in
low-stock fish ponds (Fott et al. 1980b).
However, while inverse relationships be-
tween phytoplankton and zooplankton are
commonly observed (e.g. Anderson et al.
1.955; Wright 1958; Edmondson 1979), the
I 1 I J 1 L ability of zooplankton to markedly depress
lilt 6 JUNE 13 JUNE 20 JUNE total phytoplankton biomass has often been
..
....*.*.*.*.*.*.......
...*.
.:.:.:.:.:.:.
*..*.:.:.: questioned (cf. Reynolds 1984).
A spring peak of zooplankton coinciding

E _
0.- 0 3 100 3 100 3 10
with a period of very clear water is fre-
quently reported in the literature (Lampert
1985; Sommer et al. 1986). The coincidence
TIME (days) of a clear-water phase and the zooplankton
Fig. 9. Relative densities of zooplankton in the en- maximum, however, is not proof that zoo-
closure experiments (mg dry wt caught per tow of the plankters are responsible for the removal of
plankton net). The date of the start of each experiment particles. The decline of algal populations
is indicated. The stippled areas represent daphnids; the could be a consequence of nutrient limita-
open areas show biomass of Eudiaptomus spp.
tion and sedimentation, and the zooplank-
ton maximum may occur during the clear-
water phase as a time-delayed response to
POC and Chl a did not always change in good food conditions. There is even a pos-
the same direction. In the series beginning sibility that the zooplankton maximum
on 6 June, POC increased in the zooplank- could be a consequence of algal breakdown,
ton-free bags while chlorophyll decreased in if the zooplankters feed on the increased
both bags as well as in the lake. This prob- numbers of bacteria decomposing the algal
ably reflects the increase of detritus or small material. On the other hand, calculations of
heterotrophs. Although it is clear that the the loss budget of algal populations (Reyn-
zooplankton in the bags had an influence olds et al. 1982; Sommer 1983; Tilzer 1984)
on the abundance of small particles, there and direct estimates of zooplankton grazing
seem to be additional factors affecting algal rates (Lampert and Schober 1978; Gulati et
growth or the Chl : C ratio. al. 1982) indicate that grazing losses can at
In the bags started during the clear-water times exceed primary production.
phase, removal of zooplankton permitted By monitoring biotic and abiotic factors,
increased growth of small algae, as indicated making in situ measurements of grazing, and
by the Chl values in the fraction <35 pm.
a manipulating zooplankton in bag experi-
However, just the opposite happened for the ments, we have provided multiple evidence
size fraction 35-250 pm (Fig. 11). Large al- that grazing by zooplankton causes the spring
gae showed enhanced growth in the bags clear water in Schijhsee. First, we have
containing zooplankton, but remained at shown that abiotic factors are unlikely to be
lake level in the zooplankton-free bags. No responsible for the clear-water phase. Lim-
consistent effect was to be seen in the series itation of phosphorus or nitrogen and low
that started later. Thus, when the zooplank- irradiance may reduce the algal growth rate
ton density in the bags was high, small algae but cannot alone be the reason for the rapid
were reduced while the growth of large algae declines in phytoplankton biomass. More-
was stimulated. over, in Schijhsee nutrient concentrations
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
486 Lampert et al.

POC Chl a

2
0
m
u
a
0 23 MAY

0.4 -31 MAY 20 JUNE

2 0.4
a
e 0.3 1 0.3
0.2
0 0.2
0 3 100 3 10
TIME (days)
Fig. 10. Results of the six enclosure experiments considering particles < 35 pm. O-Zooplankton removed;
O-enclosures with zooplankton; dotted lines-lake nearby the enclosure. Particulate carbon in mg liter-‘; Chl
a in pg liter-l.

are not unusually low at the beginning of nearly 80% of the zooplankton biomass was
the clear-water phase and even increase at Daphnia (cf. Fig. 2). Since specific ingestion
the time of the sharpest decline in phyto- rates of Eudiaptomus do not exceed those
plankton (Fig. 5). of Daphnia (Muck and Lampert 1984), we
Grazing rates measured in situ during the can assume that the main grazing activity
clear-water phase in Schiihsee are high was due to daphnids. Therefore, we con-
enough to exceed the rate of algal growth clude that our community grazing rates are
(see Reynolds 1984). Radiotracer measure- a reasonable estimate of the total grazing on
ments of grazing rates, however, have the nannoplankton.
disadvantage of assuming that all particles Zooplankton biomass is the governing
are ingested with the same efficiency as the factor determining the community grazing
tracer. This is certainly not valid for natural rate in Schiihsee (Lampert 1985) (cf. Fig. 8).
phytoplankton communities. We tried to The algal decline starts before the grazing
compensate for that by using particles of rate reaches its maximum. The decline be-
two different sizes. Daphnids showed only comes evident when zooplankton biomass
a slight preference for the larger Scenedes- exceeds 2.5 mg dry wt m-2, although max-
mus (Lampert and Taylor 1985), but Eu- imum zooplankton biomass is 8.8 mg dry
diaptomus was much more selective. In a wt m-*. This suggests that the filtering ca-
copepod-dominated community, consider- pacity of the zooplankton may be far higher
ing only the filtering rates on Scenedesmus than the rate of particle increase. A litera-
would result in overestimation of total graz- ture survey showed that zooplankton den-
ing. During the clear-water phase, however, sity in several lakes and reservoirs was be-
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase 487

X35 j4m 35-250 pm together with the bag results to provide in-
formation on the enclosure effects (Fig. 10).
Another complication occurs because the
grazer populations in the bags did not al-
ways mimic lake populations but increased
when the lake populations remained low (cf.
Fig. 9). Comparisons between the bags and
the lake are probably most reliable in the
experiment started at 16 May during the
clear-water phase. In this case zooplankton
biomass decreased in the bag and in the lake
simultaneously. Both particulate carbon and
c
23 MAY -I -I chlorophyll a in the bag with zooplankton
were very similar to those in the lake, in-
s dicating that we were probably not too far
I2 from reality.
0 The removal of zooplankton always re-
1 sulted in densities of small algae higher than
in the controls (Fig. 10). A significant ab-
I
solute increase of algal biomass in the zoo-
0I I I I I I I 1
plankton-free bags, however, was only to be
0 3 100 3 10 seen in the bags initiated during the clear-
T I M E (days) water phase (16 May and partly 23 May).
Fig. 11. Comparison of Chl a in small and large Later the removal of zooplankton only re-
particles in the enclosure experiments initiated during duced the rate of algal decrease. This strong-
the clear-water phase. Symbols as in Fig. 10. ly supports the conclusion that zooplank-
ton, and not physical and chemical factors,
tween 1 and 3 mg dry wt m-2 at the are responsible for the decline of small algae
beginning of a clear-water phase (Lampert during the clear-water phase. It also suggests
1985). that the algal turnover during this period is
The impact of grazers on phytoplankton considerably higher than later in the season.
can be tested directly by manipulation of Since small algae predominate during spring,
zooplankton densities in experimental en- grazing losses are very high during the zoo-
closures (e.g. Porter 1972; McCauley and plankton maximum and can only be com-
Briand 1979). Because we were interested pensated by algal species with very high
in the seasonal development of the impact growth rates (Fott et al. 1980a). When re-
of zooplankton, we started a new pair of leased from grazing pressure in experimen-
bags every week instead of running many tal enclosures, such rapidly growing algae
bags in parallel for a long period; thus our are able to build up dense populations very
bag experiments test the effect of zooplank- quickly. We can, therefore, conclude that
ton on phytoplankton through a series of grazing is the factor of overwhelming im-
comparisons between treatment bags (with- portance for algal control during the short
out zooplankton) and control bags (whole period around mid-May. When the zoo-
lake water). Since the physical and chemical plankton biomass declines in June there is
environment in a plastic bag is not identical still some effect, but other factors may be
with that in the surrounding lake (cf. Lund- equally important and thus make the bag
gren 1985), the controls cannot be expected results less convincing (cf. Reynolds et al.
to behave exactly like the lake. When four 1982).
bags were filled with untreated lake water, Although enclosure experiments may
their communities developed in a very simi demonstrate the effects of zooplankton
ilar way, but algal production was usually qualitatively (Fott et al. 1980b; DeCosta et
lower than in the lake (W. Fleckner un- al. 1983; Schoenberg and Carlson 1984),
publ.). We nevertheless present the lake data quantitative assessment of the role of graz-
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
488 Lampert et al.

ing in the lake is difficult because it is nearly summer, probably in response to grazing
impossible to create the same grazing pres- (Porter 1977). If a high grazer biomass can
sure in the bags as in the lake. For example be maintained during summer, transpar-
filter-feeders confined in the bags cannot ency may remain relatively high (Fott et al.
migrate vertically. Thus, the diel fluctua- 1980a). This will, however, not be such a
tions of grazing (Crumpton and Wetzel 1982; spectacular phenomenon as the spring clear-
Lampert and Taylor 198 5) are likely to be water phase, because algal biomass may be
reduced in the bags. Grazers in our bags may high but concentrated in large colonies,
have been active all day while those in the which cause less turbidity than small cells
lake entered the epilimnion only at night. (Benndorf et al. 1984).
Some large algae seem to be stimulated It is of considerable importance to un-
in the enclosures by the presence of zoo- derstand why filter-feeding zooplankters are
plankton. Similar results have been found usually confined to such a short period in
in small bags by Vyhnalek (1983) and Leh- spring. As in many other lakes (Sommer et
man and Sandgren (1985). This is not so al. 1986) the spring peak of filter-feeders in
easy to understand, since the biomass of net Schtihsee consists mainly of cladocerans.
plankton in the lake is increasing at a lower The timing of their increase is probably de-
rate (cf. Fig. 11). The change of physical termined by temperature and supported by
parameters in the enclosures (reduced ir- hatching of ephippia. In other lakes, how-
radiation, reduced turbulence) is unlikely to ever, there may also be a component of in-
stimulate the growth of large algae. One dif- vertebrate predation which delays the pop-
ference between the zooplankton bags and ulation increase (Lampert and Schober
the lake, however, is the timing of the zoo- 19 7 8). Exponentially growing cladoceran
plankton impact. Grazing and nutrient re- populations deplete their food resources by
generation in the bags take place during the causing the clear water and thus become
whole 24 h, but are restricted to the night seriously food-limited. In Schijhsee this is
hours in the lake (Lampert and Taylor 1985). clearly to be seen from the dramatic drop
Continuous nutrient regeneration may en- in fecundity of daphnids and from the shift
hance the growth of both small and large of their size distribution to large animals.
algae, but the stimulation will only be vis- Particulate carbon of ingestible size de-
ible in the large particles since the small creases to about 0.2 mg liter-’ (cf. Fig. 2),
ones are grazed upon. We do not know the threshold concentration for egg produc-
whether diel fluctuations of nutrient regen- tion in the field (Lampert 19783).
eration are important to the algal commu- The drop in fecundity causes the decline
nity, but if so the bag results will exaggerate of the grazers, so that the small particles can
the rate of change of the algal community then recover. The daphnid population,
with respect to the enrichment of large forms. however, does not build up another peak
The results, however, will not be qualita- before autumn but typically shows a sum-
tively wrong. mer decline (Threlkeld 198 5). Egg numbers
We think that our study clearly demon- of Daphnia increase in Schijhsee during
strates that even in relatively large lakes summer but do not reach maximum values,
zooplankton can cause phases of extremely probably because large algae interfere with
clear water. It also shows that the period of food collection (Gliwicz and Siedlar 1980).
high transparency due to grazer control is Although we do not have data on fish preda-
short. Two conditions must be met to pro- tion, it is also likely that predation contributes
duce a clear-water phase: the algal standing significantly to the control of daphnids when
stock must consist of small, “edible” cells, the new generation of planktivorous fishes
and the biomass of filter-feeding zooplank- starts actively feeding in June. Thus recov-
ton must reach high levels. ery of the daphnid population may be pre-
An algal community dominated by small vented by a combination of poor food qual-
cells is typical for spring (Sommer et al. ity and predation (cf. Larsson et al. 198 5).
1986), whereas large, bizarrely shaped, or Autogenic seasonal succession leaves only
gelatinous (“inedible”) algae develop during a small gap in time, during which filter-feed-
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Spring clear- water phase 489

tonsukzession des Bodensees (Uberlinger See).

ing zooplankton can reach extremely high Verh. Ges. Oekol. 8: 373-382.
densities, between unfavorable abiotic con- GLIWICZ, Z. M., AND E. SIEDLAR. 1980. Food size
ditions in winter and early spring and ex- limitation and algae interfering with food collec-
ternal control (interfering algae, predation) tion in Daphnia. Arch. Hydrobiol. 88: 155-177.
GULATI, R. D., K. SIEWERTSEN, AND G. POSTEMA. 1982.
in summer. Since zooplankton populations
The zooplankton: Its community structure, food
which are high enough to produce a clear- and feeding and role in the ecosystem of Lake
water phase cause their own food limitation, Vechten. Hydrobiologia 95: 127-l 64.
the rapid decrease of the population follows HANEY, J. F. 197 1. An in situ method for the mea-
automatically. This may explain why the surement of zooplankton grazing rates. Limnol.
Oceanogr. 16: 970-977.
duration of the clear-water phase is so brief -, AND D. J. HALL. 1973. Sugar coated Daphnia:
and why its timing is so predictable (Geller A preservation technique for Cladocera. Limnol.
1980). Oceanogr. 18: 33 l-333.
KRAMBECK, H.-J., B. HICKEL, W. HOFMANN, AND J.
References OVERBECK. 1978. Mathematische Modelle als in-
tegrierende Hilfsmittel des Limnologen, darge-
ANDERSON, G. C., G. W. COMITA, AND V. ENC;- stellt am Okosystem PluBsee. Verh. Ges. Oekol.
STRBM-HEG. 1955. A note on the phytoplank- 7: 137-144.
ton-zooplankton-relationship in two lakes in -, W. LAMPERT, AND H. BREDE. 198 1. Messung
Washington. Ecology 36: 757-759. geringer Mengen von partikulirem Kohlenstoff in
BAILEY-WATTS, A. 1982. The composition and abun- natiirlichen Gewtissern. GIT Fachz. Lab. 25: 1009-
dance of phytoplankton in Loch Leven (Scotland) 1012.
1977-l 979 and a comparison with the succession KUIPER, J. 1977. Development of North Sea coastal
in earlier years. Int. Rev. Gesamten Hydrobiol. plankton communities in separate plastic bags un-
67: l-25. der identical conditions. Mar. Biol. 44: 97-107.
BARKO, J. W., D. J. BATES, G. J. FILBIN, S. M. KUPARINEN, J., J. M. LEPPXNEN, J. SARVALA, A.
HENNINGTON, AND D. G. MCFARLAND. 1984. SUNDBERG,AND A. VIRTANEN. 1984. Production
Seasonal growth and community composition of and utilization of organic matter in a Baltic eco-
phytoplankton in a eutrophic Wisconsin im- system of Tvtirminne, southwest coast of Finland.
poundment. J. Freshwater Ecol. 2: 5 19-533. Rapp. P.-V. Reun. Cons. Int. Explor. Mer 183, p.
BENNDORF, J., H. KNESCHKE, K. KOSSATZ, AND E. PENZ. 180-192.
1984. Manipulation of the pelagic food web by LAMPERT, W. 1978a. Climatic conditions and plank-
stocking with predacious fishes. Int. Rev. Ge- tonic interactions as factors controlling the regular
samten Hydrobiol. 69: 407-428. succession of spring algal blooms and extremely
BLAAUBOER, M. C. 1982. The phytoplankton species clear water in Lake Constance. Int. Ver. Theor.
composition and the seasonal periodicity in Lake Angew. Limnol. Verh. 20: 969-974.
Vechten from 1956 to 1979. Hydrobiologia 95: - . 1978b. A field study on the dependence of
25-36. the fecundity of Daphnia spec. on food concen-
BOTTRELL, H. H., AND OTHERS. 1976. A review of tration. Oecologia 36: 363-369.
some problems in zooplankton production stud- -. 1985. The role of zooplankton: An attempt
ies. Norw. J. Zool. 24: 419-456. to quantify grazing, p. 54-62. In Lakes pollution
CRUMPTON, W. G., AND R. G. WETZEL. 1982. Effects and recovery. Proc. Int. Congr. Eur. Water Pollut.
of differential growth and mortality in the seasonal Contr. Assoc. Rome.
succession of phytoplankton populations in Law- -, AND U. SCHOBER. 1978. Das regelmil3ige
rence Lake, Michigan. Ecology 63: 1729-l 739. Auftreten von Friihjahrs-Algenmaximum und
DECOSTA, J., A. JANICKI, G. SHELLITO, AND G. WILCOX. “Klarwasserstadium” im Bodensee als Folge von
1983. The effect of phosphorus additions in en- klimatischen Bedingungen und Wechselwirkun-
closures on the phytoplankton and zooplankton gen zwischen Phyto- und Zooplankton. Arch. Hy-
of an acid lake. Oikos 40: 283-294. drobiol. 82: 364-386.
EDMONDSON, W. T. 1979. Lake Washington and the - AND B. E. TAYLOR. 1984. In situ grazing rates
predictability of limnological events. Ergeb. Lim- and particle selection by zooplankton: Effects of
nol. 13: 234-241. vertical migration. Int. Ver. Theor. Angew. Lim-
FOTT, J., B. DESORTOVA, AND J. HRBAEEK. 1980a. nol. Verh. 22: 943-946.
A comparison of the growth of flagellates under -,AND-. 1985. Zooplankton grazing in a
heavy grazing stress with a continuous culture, p. eutrophic lake: Implications of vertical migration.
39 5-40 1. In Continuous cultivation of microor- Ecology 66: 68-82.
ganisms. Proc. 7th Symp., Inst. Microbial. Czech- LARSSON, P., S. ANDERSEN, Y. BBRSIIEIM, P. JAKOBSEN,
oslovak Acad. Sci., Prague. AND G. JOHNSEN. 1985. Individual growth of
-, L. PECI-IAR, AND M. PRA~AKOVA. 1980b. Daphnia longispina in the summer decline phase
Fish as a factor controlling water quality in ponds, of the population. Ergeb. Limnol. 21: 34 l-350.
p. 255-26 1. In Hypertrophic ecosystems. Devel- LEHMAN, J.T., ANDC. D. SANDGREN. 1985. Species-
op. Hydrobiol. 2. Junk. specific rates of growth and grazing loss among
GELLER, W. 1980. Stabile Zeitmuster in der Plank- freshwater algae. Limnol. Oceanogr. 30: 34-46.
 19395590, 1986, 3, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.1986.31.3.0478 by Cochrane Chile, Wiley Online Library on [05/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
490 Lampert et al.
LUNDOREN, A. 1985. Model ecosystems as a tool in RIEMANN, B. 1983. Biomass and production of phy-
freshwater and marine research. Arch. Hydrobiol. to- and bacterioplankton in eutrophic Lake Tys-
Suppl. 70, p. 157-196. trup, Denmark. Freshwater Biol. 13: 389-398.
MCCAULEY, E., AND F. BRIAND. 1979. Zooplankton SCHOBER, U. 1980. Kausalanalytische Untersu-
grazing and phytoplankton species richness: Field chungen der Abundanzschwankungen des Crus-
test of the predation hypothesis. Limnol. Ocean- taceen-Planktons im Bodensee. Ph.D. thesis, Univ.
ogr. 24: 243-252. Freiburg/Breisgau. 162 p.
MARKER, A. F., E. A. NUSCH, H. RAI, AND B. RIEMANN. SCHOENBERG,~. A., AND R.E. CARLSON. 1984. Direct
1980. The measurement of photosynthetic pig- and indirect effects of zooplankton grazing on phy-
ments in freshwaters and standardization of meth- toplankton in a hypertrophic lake. Oikos 42: 29 l-
ods: Conclusions and recommendations. Ergeb. 302.
Limnol. 14: 91-106. SOMMER, U. 1983. Light, stratification and zooplank-
MEFFERT, M.-E., AND J. OVERBECK. 1985. Dynamics ton as controlling factors for the spring develop-
of chlorophyll and photosynthesis in natural phy- ment of phytoplankton in Lake Constance.
toplankton associations. 1. Seasonal and annual Schweiz. Z. Hydrol. 45: 394-404.
cycles of chlorophyll and phaeopigments, in re- -,Z. M. GLIWICZ, W. LAMPERT,AND A. DUNCAN.
lation to euphotic depth and their vertical distri- 1986. The PEG-model of seasonal succession of
bution in small north German lakes. Arch. Hy- plankton in fresh waters. Arch. Hydrobiol. In press.
drobiol. 104: 2 19-234. THRELKELD, S. 1985. Resource variation and the ini-
MUCK, P., AND W. LAMPERT. 1984. An experimental tiation of midsummer declines of cladoceran pop-
study on the importance of food conditions for the ulations. Ergeb. Limnol. 21: 333-340.
relative abundance of Calanoid copepods and TILZER, M. M. 1983. The importance of fractional
cladocerans. 1. Comparative feeding studies with light absorption by photosynthetic pigments for
Eudiaptomus gracilis and Daphnia longispina. phytoplankton productivity in Lake Constance.
Arch. Hydrobiol. Suppl. 66, p. 157-179. Limnol. Oceanogr. 28: 833-846.
OSKAM, G. 1978. Light and zooplankton as algae -. 1984. Estimation of phytoplankton loss rates
regulating factors in Biesbosch reservoirs. Int. Ver. from daily photosynthetic rates and observed bio-
Theor. Angew. Limnol. Verh. 20: 1612-1618. mass changes in Lake Constance. J. Plankton Res.
PETERSEN, F. 1983. Population dynamics and pro- 6: 309-324.
duction of Daphnia galeata (Crustacea, Cladocera) UHLMANN, D. 197 1. Influence of dilution, sinking
in Lake Esrom. Holarct. Ecol. 6: 285-294. and grazing rate on phytoplankton populations of
PORTER, K. G. 1972. A method for the in situ study hyperfertilized ponds and micro-ecosystems. Mitt.
of zooplankton grazing effects on algal species Int. Ver. Theor. Angew. Limnol. 19, p. 100-124.
composition and standing crop. Limnol. Ocean- VYHNALEK, V. 1983. Effect of filter-feeding zooplank-
ogr. 17: 913-917. ton on phytoplankton in fish ponds. Int. Rev. Ge-
-- 1977. The plant-animal interface in fresh- samten Hydrobiol. 68: 397-4 10.
water ecosystems. Am. Sci. 65: 159-170. WETZEL, R. G., AND G. E. LIKENS. 1979. Limnolog-
RAI, H. 1982. Primary production of various size ical analyses. Saunders.
fractions of natural phytoplankton communities WRIGHT, J. C. 1958. The limnology of Canyon Ferry
in a north German lake. Arch. Hydrobiol. 95: 395- Reservoir. 1. Phytoplankton-zooplankton rela-
412. tionships in the euphotic zone during September
REYNOLDS, C. S. 1984. The ecology of freshwater and October 1956. Limnol. Oceanogr. 3: 150-l 59.
phytoplankton. Cambridge.
-, J. M. THOMPSON, A. J. FERGUSON, AND S. W.
WISEMAN. 1982. Loss processes in the popula-
tion dynamics of phytoplankton maintained in Submitted: 26 July 1985
closed systems. J. Plankton Res. 4: 56 l-600. Accepted: 11 December 1985