Title 210 – National Engineering Handbook

Part 645 – Construction Inspection

Subpart H – Earthfill and Rockfill
Amended March 2025

645.80 Introduction

A. NRCS uses earthfills and rockfills in work projects for berms, dikes, levees,
embankments, and structural backfills. These projects’ primary structural
component is soil, rock, or a combination of the two.
B. The quality of available material and the condition of the foundation are major
factors that determine what type of structure can be constructed economically at a
given location. The variability inherent in earth and rock materials requires a
design and construction tolerance greater than for most other building materials.
Earth and rock materials range from fine-grained soils (50% or more passing a
No. 200 sieve) to rocky soils (predominantly gravel and cobbles) and every
combination thereof. The material’s onsite availability, their engineering
properties, and the structure’s intended purpose determine what structure can be
designed.
C. The proper use of earth and rock as construction materials requires special
attention to investigation, testing, and appropriate design procedures and
assumptions. Use of onsite materials for installation requires proper selection
from a borrow source, control of placement and processing, moisture control, and
adequate compaction. Inspection of earth and rock construction requires
knowledge, experience, and judgment to ensure compliance with the
specifications and a quality finished job.
D. Prior to earthfill and rockfill operations, the quality assurance (QA) inspector is
responsible to:
1. Review and understand the drawings and specifications related to earthfill and
rockfill.
2. Verify the diversion, dewatering, and drainage systems are in place.
3. Verify the construction of access and haul roads.
4. Verify the safety plans of earthfill and rockfill operations.
5. Verify the protection of buried utilities during required excavations.
6. Verify the acquisition of all required permits and the work’s adherence to
permit requirements.

210 H 645 Subpart H (amended March 2025) 1

645.81 Installation

A. Materials
1. Earthfill and Earth Backfill
a. Earthfill is composed of natural earth materials placed and compacted by
construction equipment operated in a conventional manner. Earth backfill
is composed of natural earth materials placed and compacted in confined
spaces or adjacent to structures (including pipes) by hand tamping,
manually directed power tampers or vibrating plates, or their equivalent.
b. Normally, builders construct earthfills from the most suitable materials
available at or near the site after removal of rock particles that are larger
than the maximum size specified. They must assess the materials’ physical
properties (permeability, strength, and compressibility), preferably from
prior laboratory and field test results. They must interpret these laboratory
and field test results to predict the suitability of the material for the type of
structure planned.
c. Earthfill and earth backfill details in the construction drawings and
specifications are based on the laboratory tests of samples taken from the
foundation and borrow areas at the site. These details identify the types of
materials used and their location in the installation according to their
Unified Soil Classification System (USCS) designation. The USCS,
developed by Casagrande in 1952 and adopted by most federal agencies as
a standard for all engineering work, is the recognized classification
standard for all engineering work in the NRCS. ASTM-International
(ASTM) D2487 publishes the USCS standard. ASTM D2488 provides the
field methods for classifying soils in accordance with the USCS.
d. All NRCS engineers and inspectors working on construction must be
familiar with the USCS and the procedures outlined in both ASTM D2487
and D2488. NRCS provides soil mechanics training courses and modules
for engineers and inspectors. Inspectors can take advantage of these and
other available resources to improve their knowledge, skills, and abilities
in understanding the engineering characteristics of various soil groups and
their use in construction of conservation measures. See Title 210 National
Engineering Handbook, Part 645 “Construction Inspection,” Subpart G
“Foundation Preparation, Removal of Water, and Excavation” for more
information. Appendix B contains Worksheet (WS) 7.1 to help inspectors
when performing a visual classification of the soil.
e. See figures 645H-1 to 645H-3 for examples of applying the USCS to
using earthfill for engineering. The four properties used to evaluate
earthfill are permeability, resistance to piping, compressibility, and
workability. A numerical rating from 1 (most suitable) to 14 (least
suitable) determines the relative suitability of soil materials in rolled earth
embankments, channel sections, and foundations.
f. Earth backfill may be the same materials as earthfill, but the specified
maximum particle size is usually smaller. Earth backfill may be composed

210 H 645 Subpart H (amended March 2025) 2

 of plastic materials with low permeability for sealing off the movement of
water. Common classifications of these types of earth backfill materials
are CL, ML, SC, SM, CH, and MH.
g. Earth backfill may also be nonplastic materials such as clean sands and
gravels, selected by builders over plastic soils because of the ease and
economy of placing (particularly in confined areas), good stability, and
low settlement potential. They cannot be used where impermeability is
required. These nonplastic, granular soils can be made to flow under a
pipe or other structure and be compacted with little effort if they do not
contain excessive amounts of fines (i.e., particles smaller than the No. 200
sieve size). Sands and gravels classified as GW, GP, SW, and SP make the
best granular backfill because, by definition, these soils contain less than 5
percent fines. Proper selection of granular materials is important for
successful results. Excessive amounts of fines tend to plug the voids
between the coarse-grained soils and inhibit drainage necessary for the
consolidation process.
h. Gravels and sands with up to 12 percent fines may be suitable for some
backfill applications. These include gravels and sands with a dual
classification: GW–GM, GW–GC, GP–GM, GP–GC, SW–SM, SP–SM,
SP–SC, and SW–SC. These gravels and sands contain 5 to 12 percent
fines. They are unsuitable for projects requiring free drainage. See
210 NEH Part 645 Subpart K “Drains and Filters” for more information.
i. In addition to being used for backfill, nonplastic granular materials are
often used for structure foundations to improve the bearing capacity by
removing and replacing soft soils with free draining coarse-grained soils.
j. Classifying soils according to the USCS alone does not always provide
enough information to predict soil performance. NRCS classified gravelly
and stony soils used in compacted embankments in three groups based on
the durability of the coarse rock and soil materials. Refer to Technical
Release 210-26, “The Use of Soils Containing More Than 5 Percent Rock
Larger Than the No. 4 Sieve” for more information.
k. Most earthfills and earth backfills are designed as homogeneous fills.
Larger structures, such as embankments for flood control dams, or sites
that have limited quantities of impermeable soils may be designed as
zoned fills. Zoned fills have less permeable material placed as a cutoff or
core zone and more permeable material placed at less strategic points such
as the outside sections of the embankment. The QA inspector must be
familiar with the quantity and quality of the borrow materials to verify
their selection and placement in the embankment to produce the zones
specified in the design. The design engineer must approve any changes to
zone locations or limits.
l. Borrow areas commonly contain layers or lenses of materials that vary in
texture and moisture. The contractor, by selecting the materials at the time
of loading and by proper routing and control of the haul equipment, places
the materials in the designated zones where they are suitable for use. On
some jobs, selecting borrow materials with a moisture content within the

210 H 645 Subpart H (amended March 2025) 3

 specified range controls compaction moisture. Thus, material suitability is
contingent on both texture and moisture content.
Figure 645H-1: Engineering of Earthfill - Important Properties of Soil Groups
Typical names of Group Permeability Shearing Compressibility Workability
soil groups symbols when strength when as a
compacted when compacted and construction
compacted saturated material
and
saturated
Well-graded GW Pervious Excellent Negligible Excellent
gravels, gravel-
sand mixtures,
little or no fines
Poorly graded GP Very Good Negligible Good
gravels, gravel- pervious
sand mixtures,
little or no fines
Silty gravels, GM Semipervious Good Negligible Good
poorly graded to
gravel-sand-silt impervious
mixtures
Clayey gravels, GC Impervious Good to Very low Good
poorly graded fair
clay-sand-silt
mixtures
Well-graded SW Pervious Excellent Negligible Excellent
sands, gravelly
sands, little or no
fines
Poorly graded SP Pervious Good Very low Fair
sands, gravelly
sands, little or no
fines
Silty sands, poorly SM Semipervious Good Low Fair
graded sand-silt to
mixtures impervious
Clayey sands, SC Impervious Good to Low Good
poorly graded fair
clay-silt mixtures
Inorganic silts and ML Semipervious Fair Medium Fair
very fine sands, to
rock flour, silty or impervious
clayey fine sands
with slight
plasticity

210 H 645 Subpart H (amended March 2025) 4

 Typical names of Group Permeability Shearing Compressibility Workability
soil groups symbols when strength when as a
compacted when compacted and construction
compacted saturated material
and
saturated
Inorganic clays of CL Impervious Fair Medium Good to fair
low to medium
plasticity, gravelly
clays, sandy clays,
lean clays
Organic silts and OL Semipervious Poor Medium Fair
organic silt-clays to
of low plasticity impervious
Inorganic silts, MH Semipervious Fair to poor High Poor
micaceous or to
diatomaceous fine impervious
sandy or silty soils,
elastic silts
Inorganic clays of CH Impervious Poor High Poor
high plasticity fat
clays
Organic clays of OH Impervious Poor High Poor
medium to high
plasticity
Peat and other P N/A N/A N/A N/A
highly organic soils

Figure 645H-2: Engineering of Earthfill–Relative Desirability for Various Uses–Rolled

Earth Dams and Canal Sections
Typical names of Group Homogeneous Core – Shell – Erosion Compacted
soil groups symbols Embankment – Rolled Rolled resistance Earth Lining –
Rolled earth earth earth – Canal Canal sections
dams dams dams sections
Well-graded GW N/A N/A 1 1 N/A
gravels, gravel-
sand mixtures,
little or no fines
Poorly graded GP N/A N/A 2 2 N/A
gravels, gravel-
sand mixtures,
little or no fines
Silty gravels, GM 2 4 N/A 4 4
poorly graded
gravel-sand-silt
mixtures

210 H 645 Subpart H (amended March 2025) 5

 Typical names of Group Homogeneous Core – Shell – Erosion Compacted
soil groups symbols Embankment – Rolled Rolled resistance Earth Lining –
Rolled earth earth earth – Canal Canal sections
dams dams dams sections
Clayey gravels, GC 1 1 N/A 3 1
poorly graded
clay-sand-silt
mixtures
Well-graded SW N/A N/A 3 if 6 N/A
sands, gravelly gravelly
sands, little or no
fines
Poorly graded SP N/A N/A 4 if 7 if N/A
sands, gravelly gravelly gravelly
sands, little or no
fines
Silty sands, SM 4 5 N/A 8 if 5 critical
poorly graded gravelly erosion
sand-silt
mixtures
Clayey sands, SC 3 2 N/A 5 2
poorly graded
clay-silt mixtures
Inorganic silts ML 6 6 N/A N/A 6 erosion
and very fine critical
sands, rock flour,
silty or clayey
fine sands with
slight plasticity
Inorganic clays CL 5 3 N/A 9 3
of low to
medium
plasticity,
gravelly clays,
sandy clays, lean
clays
Organic silts and OL 8 8 N/A N/A 7 erosion
organic silt-clays critical
of low plasticity
Inorganic silts, MH 9 9 N/A N/A N/A
micaceous or
diatomaceous
fine sandy or
silty soils, elastic
silts
Inorganic clays CH 7 7 N/A 10 8 volume
of high plasticity change
fat clays control

210 H 645 Subpart H (amended March 2025) 6

 Typical names of Group Homogeneous Core – Shell – Erosion Compacted
soil groups symbols Embankment – Rolled Rolled resistance Earth Lining –
Rolled earth earth earth – Canal Canal sections
dams dams dams sections
Organic clays of OH 10 10 N/A N/A N/A
medium to high
plasticity
Peat and other P N/A N/A N/A N/A N/A
highly organic
soils

Figure 645H-3: Engineering of Earthfill–Relative Desirability for Various Uses–

Foundations
Typical names of soil groups Group Seepage Seepage non-
symbols important - important -
Foundations Foundations
Well-graded gravels, gravel-sand mixtures, little or GW N/A 1
no fines
Poorly graded gravels, gravel-sand mixtures, little GP N/A 3
or no fines
Silty gravels, poorly graded gravel-sand-silt GM 1 4
mixtures
Clayey gravels, poorly graded clay-sand-silt GC 2 6
mixtures
Well-graded sands, gravelly sands, little or no fines SW N/A 2
Poorly graded sands, gravelly sands, little or no SP N/A 5
fines
Silty sands, poorly graded sand-silt mixtures SM 3 7
Clayey sands, poorly graded clay-silt mixtures SC 4 8
Inorganic silts and very fine sands, rock flour, silty ML 6 9
or clayey fine sands with slight plasticity
Inorganic clays of low to medium plasticity, CL 5 10
gravelly clays, sandy clays, lean clays
Organic silts and organic silt-clays of low plasticity OL 7 11
Inorganic silts, micaceous or diatomaceous fine MH 8 12
sandy or silty soils, elastic silts
Inorganic clays of high plasticity fat clays CH 9 13
Organic clays of medium to high plasticity OH 10 14
Peat and other highly organic soils P N/A N/A

m. With respect to earthfill and earth backfill materials, the QA inspector is

responsible for verifying:
(1) Adequate knowledge and competency in the field use of the USCS.
(2) Materials used are properly identified and do not differ significantly
from those materials specified in design.
(3) Materials are routed to specified locations in the earthfill.
(4) Frozen materials are not used in any earthfill, or earth backfill.

210 H 645 Subpart H (amended March 2025) 7

 (5) Moisture content of borrow materials is within the specified range or
will be adjusted to comply with specification requirements before
compaction.
(6) Significant changes in materials are promptly reported to the
responsible engineer.
2. Rockfill
a. Rockfill are natural materials that are hard, durable, and larger than gravel
size. Rockfill may contain only trace amounts of sand or fine-grained soil
materials; however, some specifications may allow various amounts of
sand and fine-grained materials. Riprap materials differ from rockfill in
that they are generally clean materials of a specified gradation. Rockfill
materials may be from a source on the construction site or from offsite
sources. On some occasions, rockfill materials may be of less durable
onsite materials that are a weathered but economical choice for zoned fill.
These types of materials will break down during processing and
compaction, so they require close attention to the specified methods and
field test results.
b. The drawings and specification must clearly identify borrow sources
(onsite or offsite), gradation, quality of rockfill materials, processing, and
compaction needs.
c. The QA inspector must verify the contractor only approves and installs
rockfill materials that meet the job specifications.
B. Foundation Preparation
1. The purpose of careful foundation preparation for earthfill placement is two-
fold. The first is to remove topsoil, loose materials, organic materials,
standing water, or any unsuitable materials that would compromise the
integrity of the contact area between the earthfill and the foundation. The
second is to prepare the foundation to allow good bonding of materials at the
contact area. The project specifications provide the necessary details for
preparing the foundation.
2. Prior to placing earthfill or rockfill, the foundations must be prepared as
specified by removing vegetation and other unsuitable material, excavation,
and dewatering to facilitate placement, processing, and compaction.
3. Rockfill foundations must be cleaned in a similar fashion to earthfill. Bonding
at the contact area is not a concern if the contact is clean and moist.
Overhangs and vertical surfaces may require some contouring.
4. Occasional rock outcrops in earth foundations for earthfill may require special
treatment if they interfere with foundation preparation, the initial layers of the
fill, or the bond between the foundation and the fill. Dams and other structures
designed to restrain water movement require specially treating rock outcrops.
Otherwise, occasional rock outcrops do not require special treatment.
5. Dams, and other structures designed to impound water, require their
foundation be prepared to achieve a good bond with the earthfill to prevent a

210 H 645 Subpart H (amended March 2025) 8

 future seepage path or a slippage surface due to weak shear strength
properties. Earth foundations must be compacted to the specified density with
the soil at the specified moisture content. Rock foundations, in the same
situation, must be clean, moist, and contoured to eliminate overhangs, vertical
surfaces, and excessively steep surfaces.
6. Some foundation soils, when left exposed to the sun and wind for any length
of time, will shrink and crack. These soils require careful inspection to ensure
their removal is needed to assure that these soils are removed, and the surface
reworked to restore a competent foundation.
7. Frozen material must not be placed on a foundation surface and no materials
may be placed on a frozen foundation surface. The fill operations may begin
only after all foundation surfaces are approved as meeting the specifications.
8. Abutment and sloping surfaces for both kinds of fill must be cleared of all
loose material by hand or other effective means and must be free of standing
water when fill is placed upon them. Foundation and abutment surface
steepness must be as specified for the job. Fill material may not be placed
until the specified foundation preparation has been approved.
9. See 210 NEH Part 645 G for more information about foundation preparation.
10. For foundation preparation, the QA inspector’s responsibilities include
verifying:
a. The foundation is prepared as specified prior to material placement.
b. The surface and subsurface drainage features are in place to control water
during earth or rockfill operations.
c. The clearing and foundation preparation work adheres to safety standards.
C. Placement and Processing
1. Earthfill
a. During the placement of earthfill, the inspection of borrow materials
coming to the work area is most important. Earthfill compaction, which
follows the placement process, is dependent on clean materials that allow
a consistent compaction effort to produce a uniformly dense material. Any
material that interferes with this compaction process is detrimental to the
operation. Debris, sticks, stones, roots, other organic material, soil too wet
to compact, and frozen soil all interfere with this process.
b. Good earthfill placement and processing that results in fill meeting the
quality defined by job specifications, requires the proper selection of
borrow materials, the use of proper equipment, and a systematic routing of
this same equipment. This scenario provides a repeatable process that will
produce a consistent quality fill. Inspectors must understand this process
and be alert to contractors taking shortcuts in any aspect of this sequence
of operations.
c. It is vital to ensure that the contractor understands the selection of borrow
materials and has the proper equipment to haul the fill to the work site and

210 H 645 Subpart H (amended March 2025) 9

 spread it in the required lift thicknesses. The placement of the earthfill to
the required pre-compaction lift thickness is an important first step.
d. After initial placement and spreading, all roots, vegetative growth,
oversize rocks, and any debris brought to the fill site from the designated
borrow sources must be separated and removed to the specified waste
areas. This may require a root rake or handpicking.
e. Large, hard lumps or masses of soil must be pulverized by disking,
harrowing, or other means. Any water added at the time of placement must
be thoroughly and uniformly mixed into the entire lift. The disk or plow
used to mix and incorporate water into the soil must penetrate completely
through the current lift and into the preceding (previously placed) lift as
depicted in figure 645H-4. This promotes good bonding to the lower lift.
When the specifications require disking, the disk must meet the size and
type specified and be capable of penetrating and mixing the full depth of
the lift. When the disk or plow cannot completely penetrate completely the
current lift, the alternatives are to get a bigger, heavier disk or reduce the
lift thickness until the equipment can meet the requirements.
Figure 645H-4: Disk Penetrating into Preceding Lift

f. During embankment construction, the top surface must be maintained

approximately level with just enough slope for drainage. Unless otherwise
specified, this slope must drain upstream toward the sediment pool area.
This drainage minimizes erosion on the back slope and transportation of
the resulting sedimentation downstream.
2. Earth Backfill
a. Earth backfill refers to fill placed adjacent to structures or in confined
spaces. This fill is placed in thinner lift thickness with more strict
requirements for removal of oversize particles. The QA inspector must
understand the specification requirements for this operation. The height of
the earth backfill adjacent to a structure must be maintained at the same
level on all sides of the structure. The idea is to prevent overloading of a
structure and allow the structure to assume the loads from the earth
backfill gradually and in a uniform manner. For more on earth backfill, see
210 NEH Part 642, Construction Specification (CS)-23 “Earthfill.”

210 H 645 Subpart H (amended March 2025) 10

 b. For earthfill and earth backfill placement and processing, the QA inspector
is responsible for verifying that:
(1) Earthfill zones are properly staked out.
(2) Earthfill is installed at locations designated on the drawings and
specifications.
(3) Foundation and embankment surfaces are conditioned for bonding and
comply with the specified grades and density.
(4) Oversize stones, roots, and debris are removed before compaction.
(5) The type of earthfill materials comply with specification requirements.
(6) Earthfill placed in an embankment has a top surface that is maintained
close to horizontal and properly sloped for drainage.
(7) The earthfill was placed, processed, and had its moisture adjusted
according to the job drawings and specifications.
(8) No frozen materials were placed on site and no materials were placed
on a frozen foundation or fill surface.
3. Rockfill
a. Rockfill must be staked out and placed in the designated zones or areas.
b. The preferred method to place rockfill is to dump it on the surface of the
layer being placed and then spread it to the desired thickness with a
crawler tractor by pushing the material over the advancing faces of the
layer. This process separates larger rocks into the bottom of the lift and
smaller rocks, spalls, and finer materials into the upper part. The main
advantage of this technique is the relatively smooth upper surface resulting
from pushing the dumped rock a short distance on top of each layer being
placed so that depressions and voids between larger rocks are filled with
the smaller materials. This approach also maintains the desired lift
thickness because the dozer operator is always advancing the lift ahead on
the smooth surface at its proper elevation. The smooth layer also reduces
tire wear, allows higher truck speeds, and provides a better surface for
compaction equipment operation.
c. The rockfill material’s maximum size controls the lift thickness.
d. The QA inspector must ensure the rock materials are of the size range
anticipated in the design. When this is not the case, the QA inspector must
notify the responsible engineer and see that changes are made in the lift
thickness to fit the size range of the rockfill.
e. With respect to rockfill placement, the QA inspector must verify that:
(1) The rockfill zones or areas are properly staked out.
(2) The rockfill is installed at locations designated on the drawings and
specifications.
(3) The prescribed placement methods are followed to produce a
competent rockfill.
(4) The lift thicknesses are appropriate for the material being placed.
D. Moisture Control
1. Earthfill and Earth Backfill

210 H 645 Subpart H (amended March 2025) 11

 a. Compaction moisture is the water content of soil at the time of compaction
and is very important. In 1933, R.R. Proctor, an engineer for the Bureau of
Waterworks and Supply in Los Angeles, California, published four articles
in the Engineering News Record describing how soil moisture is critical
when compacting soils. Proctor said, “For a given energy, if a given soil is
compacted at different water contents, the resulting dry density of the
compacted soil will vary.” And he emphasized the importance of
compaction moisture when he said, “The effect of the moisture content of
a soil upon the density to which it may be compacted is the most important
principle of soil compaction.”
b. Proctor was a field engineer involved in the design and construction of
earthen dams. He understood that geotechnical design of earthen structures
relied on knowing the engineering properties of soil such as
compressibility, shear strength, and permeability. But, since these
engineering properties are contingent on compaction moisture and soil
density, Proctor sought a way to relate laboratory compaction moisture
and soil density to that of soil compacted in the field. His search led him to
determine that a standardized laboratory procedure was necessary for
compacting soils at various compaction moisture contents. His work
developed what is called the standard Proctor curve. Figure 645H-5
illustrates Proctors’ principle and shows a typical standard Proctor curve.
c. ASTM D698 describes the standard Proctor test method used worldwide.
ASTM D1557 describes a modified Proctor test method with a compactive
effort approximately four times that of the standard Proctor method.
Conservation engineering measures rarely use the modified procedure.
Figure 645H-6 shows the energy application (effort) for the standard
Proctor method and 645H-7 shows the energy application for the modified
Proctor method. The rest of this subpart focuses on the more commonly
used standard Proctor method.
d. The Proctor method and resultant Proctor curve (figure 645H-5) yield two
important values. The first value, termed maximum dry density, is the
maximum dry density of the soil that can be obtained when compacted by
the Proctor method. The second value is the optimum moisture content.
Optimum moisture is the soil moisture at which the maximum dry density
can be attained.
e. As in Proctor’s time, geotechnical design of earthen structures requires
engineering properties of soils be known for soils compacted:
(1) By a specific compaction effort.
(2) At a specific moisture content.
(3) To a specific density.
f. Geotechnical design of earthen foundations and structures begins by
compacting the field-sampled soils in the laboratory with the standard or
modified Proctor method. The soils’ optimum moisture and maximum
density are determined from the resulting Proctor curve. The soils’
engineering properties can then be determined by testing the soil that has
been compacted within a specific moisture range to some degree relative

210 H 645 Subpart H (amended March 2025) 12

 to maximum density. By specifying that the density of in-place soil be a
value relative to maximum density, the earthen structure’s engineering
properties must match those considered by the design engineer. This was
Proctor’s goal when he sought a way to relate laboratory compaction
moisture and soil density to that of soil compacted in the field.
g. In addition to density, compaction moisture must be specified and must be
carefully controlled to gain efficient compaction and to meet the
specification requirements. Compaction moisture not only affects the
density to which it may be compacted, it also affects soil strength and
permeability.
Figure 645H-5: Proctor Curve

Figure 645H-6: Standard Proctor test method as specified in ASTM D698

210 H 645 Subpart H (amended March 2025) 13

Figure 645H-7: Modified Proctor test method as specified in ASTM D1557

h. Soil compacted at a moisture content below optimum tends to be stronger

but more permeable than if compacted at a moisture content above
optimum. Figure 645H-8 illustrates this concept, showing that soil
particles tend to reorient themselves into a less permeable structure when
compacted at a moisture content above optimum. The soil is also more
flexible when compacted at a moisture content above optimum. For many
conservation engineering measures, impermeability and flexibility are
more desirable than high strengths. Thus, most NRCS specifications are
written to specify compaction moisture above optimum. Conversely, for
high fill structures or other structures and foundations requiring high soil
strengths, compaction moisture below optimum is desirable.
i. Adding water in the borrow area is the best way to achieve large increases
in moisture content. This is accomplished by irrigation and aided by the
mixing that occurs when loading and transporting the material to the
placement area.
j. Reasonably small amounts of water can be added effectively on the fill
and incorporated into the soil with the proper equipment and consistent
processing. The size, weight, condition, and operating mode must be such
that the equipment can efficiently and uniformly stir and mix the full lift
depth of the soil that is being processed.

210 H 645 Subpart H (amended March 2025) 14

Figure 645H-8: Compaction Moisture’s Effect on Soil Strength and Permeability

k. One way to gauge the effectiveness of the equipment used is by the ribbon
test. Place a piece of survey ribbon on the ground where the soil will be
deposited. A piece approximately 10 feet long is sufficient. Place a small
rock on each end to hold it in place. If the equipment for incorporating
water is stirring and mixing the full depth of the lift, it will bring some of
the survey ribbon to the surface. Otherwise, a change in equipment size,
weight, condition, or mode or installing a thinner lift is necessary.
l. Earthfill that is too wet must be dried out to the specified moisture content
to assure success in attaining the desired density. Moisture can be removed
in the fill area or in the borrow area and is best accomplished by disking or
plowing. Use the graph in figure 645H-9 to estimate water needs for
specified compaction.
m. With respect to moisture control, the QA inspector must verify that:
(1) The foundation moisture is within the specified range.
(2) The compaction moisture is within the specified range throughout the
full depth of the lift.
(3) Any added water is properly incorporated into the earthfill to produce
a uniform moisture content throughout the lift thickness.
2. Rockfill
a. Moisture content in rockfill is usually not critical but may be specified.
b. Avoid excessive moisture as it may wet down adjacent earthfill and
adversely affect the placement, processing, or compaction.
E. Compaction
1. Earthfill
a. When the proper preparation and processing is complete, effective
compaction can begin.
b. Compaction is the consolidation of the loosely placed and processed
earthfill material. The objective is to reduce voids in the soil by removing
air and water, thus making the mass denser. This process lowers the soil
permeability and increases the strength properties. Compaction is achieved
by mechanical means such as heavy equipment designed for compacting

210 H 645 Subpart H (amended March 2025) 15

 soil. Examples of equipment designed specifically for compaction are
sheepsfoot rollers, static or vibratory tamping foot rollers, static or
vibratory smooth steel drum rollers, multi-tired pneumatic rollers, and grid
rollers. Equipment such as bulldozers, loaded dump trucks, and loaded
scrapers may be able to compact soils to the specified density but are not
as efficient as specifically designed compaction equipment.
c. Compaction equipment must be capable of compacting the entire lift of
earthfill to the minimum required density. The compaction equipment
must be suited to the type of material being placed. The QA inspector
must understand the different kinds of compactors and the types of
materials for which each is best suited. Figures 645H-10 and 645H-11
provide guidance for judging compaction equipment suitability for various
soils identified by the USCS.
d. If the compaction equipment is not able to compact the entire lift of soil to
the specified density, it may be necessary to get heavier equipment,
decrease the lift thickness, increase the number of passes of the
equipment, verify that the correct compaction curve is being referenced,
consider a different type of equipment, or check to make sure the moisture
content is appropriate.
e. When earthfill or earth backfill is being compacted, it is important that the
equipment establish a consistent pattern of travel to get complete coverage
of the area being compacted. Additionally, implementing a consistent
compaction process maximizes the probability that in-place testing of the
earthfill will be representative of the entire area being compacted.
f. As with most specifications, there are two ways to specify compaction.
One is to specify the compacted soil density meet or exceed a minimum
density or fall within a specified range of values. Specifications written in
this manner are known as performance-based specifications, or
performance specs. The other way to specify compaction is by method
specification, or method specs. In a method spec, the contractor must use
equipment that meets some specified criteria and must traverse the area
with this equipment a specified minimum number of passes. Performance
specs require the QA inspector to verify the compaction process achieves
the required density and water content results. Method specs require the
QA inspector to document that the equipment and number of passes
complies with the spec and that the moisture content at the time of
compaction is within the specified range.
g. Regardless of specification (performance or method), the QA inspector
must pay careful attention to verifying complete coverage over the entire
area requiring compaction, especially confined areas, or areas difficult for
the equipment to cover. Abutment interfaces or ties into existing earthfill
slopes are examples of areas that require special attention. The work may
require different, more maneuverable equipment, thinner lifts, or hand
compaction methods to achieve the desired results.

210 H 645 Subpart H (amended March 2025) 16

Figure 645H-9: Water Requirements for Compaction

210 H 645 Subpart H (amended March 2025) 17

Figure 645H-10: Compaction Characteristics of Coarse-grained Soils
Unified Relative Compacted Importance Preferred Number Typical Typical
class ease of lift of water type of of dry unit water
compaction thickness content equipment passes weights content
(in) (PCF) (%)
GW Very easy 10-12 Either dry Crawler 3-4 125- 9-12
or tractor 135
saturated vibratory
roller
GP Good to 10-12 Either dry Crawler 3-4 115- 12-16
excellent or tractor 125
saturated vibratory
roller
GM Good with 6-8 Fairly Rubber- 3-5 120- 6-13
close important tired or 135
control tamping
roller
GC Good 6 Very Rubber- 6-8 115- 9-14
important tired or 130
tamping
roller
SW Excellent 10-12 Either dry Crawler 3-4 110- 10-16
or tractor 130
saturated vibratory
roller
SP Fair 10-12 Either dry Crawler 3-4 100- 13-22
or tractor 120
saturated vibratory
roller
SM Fair 6-6 Important Rubber- 6-8 110- 10-16
tired or 125
tamping
roller
SC Good 6 Very Rubber- 4-6 105- 10-18
important tired or 125
tamping
roller

210 H 645 Subpart H (amended March 2025) 18

Figure 645H-11: Compaction Characteristics of Fine-grained and Highly Organic Soils
Unified Relative Compacted Importance Preferred Number Typical Typical
class ease of lift of water type of of dry unit water
compaction thickness content equipment passes weights content
(in) (PCF) (%)
ML Fair 6 Important Tamping 4-6 95-120 12-22
roller
CL Good to 6 Very Tamping 4-6 95-120 12-22
fair important roller
MH Poor 6 Very Tamping 4-6 70-95 22-40
important roller
CH Very poor 6 Very Tamping 4-6 70-105 20-40
important roller
OL Fair 6 Critical Tamping 4-6 80-100 20-32
roller
OH Very poor 6 Important Tamping 4-6 65-100 20-32
roller
Pt Not Not Not Not Not Not Not
suitable suitable suitable suitable suitable suitable suitable

h. After an earthfill operation’s temporary or seasonal shutdown, the water

content and density of the surface of the unfinished earthfill must be
checked for compliance with the job specifications. This material might
need to be reconditioned, removed, or replaced before resuming work.
i. CS-23 defines three classes of earthfill. See figure 645H-13 for more.
(1) Class A is compaction to a specified density for each borrow material
used. It is used extensively where a quality earthfill is required and
testing for field control will be carried out.
(2) Class B is a given mass density for all the material (only for rocky
material with varying rock amount in the soil matrix) and is rarely
used.
(3) Class C is compaction by a prescribed method. It is used where
compaction may not be critical, materials are difficult to define or
testing capabilities for field control are not available.
j. The QA inspector’s responsibilities for earthfill compaction include
verifying:
(1) The appropriate compaction equipment is properly used to compact
earthfill and earth backfill.
(2) The compaction equipment is properly controlled to provide a
systematic and complete coverage of entire area requiring compaction.
(3) For classes A and B compaction, the specified density is attained
throughout the full depth of each lift of earthfill.
(4) For class C compaction, the prescribed equipment and method are
consistently applied.
(5) For zoned embankments, all zones are located and configured as
specified or as shown on the drawings.

210 H 645 Subpart H (amended March 2025) 19

Figure 645H-12: Selection of Compactor Type

Figure 645H-13: Summary of Compaction Classes Defined in CS-23

Compaction Type Items specified Advantage or disadvantage
Class specification
A Performance Density test method Provides a way to verify the earthfill
and required percent density meets or exceeds the
of maximum density. minimum specified density required
for the earthfill to function as
designed. Requires the QA inspector
to use testing equipment and match
the soil to a Proctor curve developed
by the QA inspector or by a lab. Works
best for soil containing no rocks or
gravel or for soil containing uniformly
distributed gravel.
B Performance Required minimum Like Class A, provides a way to verify
mass density. the earthfill density meets or exceeds
the minimum specified density. Unlike
Class A, works for soil containing rocks
and gravel including nonuniform
amounts of rock or gravel. Requires
large equipment and several hours to
conduct one test. Rarely specified.
C Method Type of equipment Requires no density testing, but
including weight and constant surveillance by the QA
contact pressure, inspector to verify the specified
minimum vibrating methos is employed over the entire
force and frequency area. Used where compaction may not
for vibrating be critical, materials are difficult to
equipment, and define, and testing capabilities for
minimum number of field control are not available.
passes.

210 H 645 Subpart H (amended March 2025) 20

 2. Earth Backfill
a. Earth backfill is earthfill placed in confined spaces such as trenches,
around concrete structures, and around pipelines. Earth backfill adjacent to
structures or in confined spaces, is compacted by manually directed
equipment. Manually directed power tampers employ a foot of various
shapes and sizes that jump up and down to impact the soil and work best
on finer grained materials with plasticity. Manually directed plate
vibrators sit on the surface and compact by static weight and vibration.
They work best on coarse-grained nonplastic materials.
b. Earth backfill has two markedly different types of materials with
corresponding installation operations and compaction characteristics. The
first type is plastic soils and the second is sands and gravels.
c. Plastic backfill material is compacted with tamping rollers, when space is
available, or by hand-controlled power tampers in confined areas.
Tamping rollers must be operated with caution near pipes and concrete
structures to avoid damaging them. Plastic backfill materials are often
specified to be compacted to a density that is equal to or greater than the
surrounding earthfill. Some specifications may require the earth backfill
be compacted to a minimum density that is some percentage of standard
Proctor maximum density. For example, the specification may read
“compact to 95 percent of standard Proctor maximum density as
determined by ASTM D698.” Earth backfill compaction moisture is
typically specified as minus 1 percent to plus 2 percent of optimum
moisture. Compacted cohesive backfill is generally placed in 6-inch layers
prior to compaction. Plastic backfill material must be protected from
drying and cracking until covered by permanent cover.
d. Sands and gravels are consolidated with pneumatic rollers, tractors, and
surface or internal vibrators. Backfill generally uses manually directed
vibratory rollers or plate vibrators. Clean gravels do not require moisture
control. Sands are compacted dry or thoroughly wetted because they are
difficult to compact if the moisture content is within a range which causes
bulking. This moisture range varies depending on the soil, but for clean
sands the bulking moisture range is between 2 and 8 percent. One method
of compacting clean sand is to flood and drain it. The flooding breaks the
sand particle bonds that occur from bulking moisture and the movement of
water through the sand matrix causes the particles to settle into a denser
configuration. Additional movement caused by mechanical vibration can
make the particles settle even more.
e. When cohesionless, free-draining materials are placed as compacted
backfill, the density requirement is specified as a percentage of sand
relative density or as a percentage of the one-point standard Proctor
maximum density. As an alternative, a method specification may be
employed. Regardless of how compaction is specified, sand is best
compacted by flooding and applying some vibratory compaction as the

210 H 645 Subpart H (amended March 2025) 21

 material is draining. The uncompacted lift thickness is generally specified
to be a maximum of 12 inches and limited to 8 inches when only manually
directed compaction effort is applied. Gravels can be placed in thicker
layers, require no moisture control, and may require little compactive
effort beyond that incidental to placement. See 210 NEH Part 645 Subpart
K “Drains and Filters” for more information.
f. When placing backfill against a pipe and some small structures, the level
on each side of the pipe or structure must be the same to avoid damaging
or displacing the pipe. It is important to guard against over-compaction
near pipes or other structures to avoid similar damage and displacement.
g. The QA inspector’s duties for compacting earth backfill include verifying:
(1) The material and particle size for backfill meet the specification
requirements.
(2) The compaction requirements are met for the material involved.
(3) The finished backfill is protected from drying and cracking until
permanent cover is placed.
(4) Backfill is not over compacted.
3. Rockfill
a. Rockfill, once placed in the required lift thickness, is best compacted by
heavy vibratory compaction equipment. Heavy vibratory steel-wheeled
rollers can be very effective. Heavy bulldozers working to spread rockfill
may provide adequate compaction on some smaller sized materials.
b. Weathered rock is inclined to break down and produce smaller material,
including soil, is best compacted with a tamping roller or grid roller.
c. Rockfill compaction is specified by method specs rather than performance
specs. Inspectors must know what method is specified and plan how best
to verify specification compliance. CS-25 “Rockfill” lists three classes of
compaction: classes I, II, and III. These classes of compaction range from
the most dynamic method to the least dynamic method and produce results
from most to least dense, respectively.
d. When using a bedding layer under the rockfill or a transition zone adjacent
to the rockfill, these materials must be properly compacted. Bedding must
be placed and compacted prior to rockfill. Transition zones may be placed
and compacted before or after, depending on whether the transition zone is
below or above the rockfill. The QA inspector must verify the zones are
located and configured as specified or as shown on the drawings.
e. The QA inspector’s duties related to compaction of rockfill include
verifying:
(1) The specified equipment is used and specified method of compaction
is strictly followed.
(2) All compaction equipment is in good condition and operated properly.
(3) That placing and compacting transition zones is in the right sequence.
(4) All zones are located and configured as specified or as shown on the
drawings.

210 H 645 Subpart H (amended March 2025) 22

645.82 Sampling and Testing

A. When specified, in-place testing for moisture and density of compacted earthfill
must be conducted to verify compliance with contract specification requirements.
This testing is a significant part of contractor quality control and NRCS quality
assurance inspection. In-place testing involves the following steps:
1. Determine the frequency of testing.
a. The Quality Assurance Plan for any earthwork job must specify the
minimum frequency of compliance testing the QA inspector must perform.
CS-94 “Contractor Quality Control” specifies the frequency of inspection
required from the Contractor Quality Control (CQC). The actual field-
testing frequency for verifying the specification compliance can only be
determined on the job during the work performance.
b. In general, earthwork construction requires more field compliance testing
earlier rather than later in the job. Later, the contractor has usually
developed a process that consistently gets the specified results. CS-94 may
require the CQC to make in-place tests for density and moisture of about
one test for every 1,000 cubic yards of compacted earthfill. It may take the
contractor several days to get the process going, and to get production up
to 1,000 cubic yards per day. In this situation, the CQC must test once
daily, and the QA inspector must verify the daily CQC testing. This same
scenario could happen when the contractor brings in new borrow material
or establishes a new zone of fill. The QC testing frequency must be
whatever the contractor needs to verify the earthfill process achieves the
specified density and moisture results.
c. When the contractor is satisfied, the QA must verify the CQC results.
Once a reliable earthfill process has been established and consistent
density and moisture results are being obtained, the frequency of testing
and verification of these results can be reduced. Thus, the higher
frequency of testing is needed to establish a reliable process, and a lower
frequency needed for verifying and documenting compliance thereafter.
Where CQC testing is not specified, the entire above scenario falls on the
QA.
d. More frequent in-place testing is required in areas where a consistent
earthfill process cannot be established (i.e., areas susceptible of poor
compaction effort). This occurs where earthfill or earth backfill is to be
placed in confined or limited access areas where equipment and traffic
patterns are limited. Such areas include those adjacent to structures or
against steep foundations like the abutment of a dam. The QA inspector
must be aware of areas that are susceptible to poor compaction effort.
Since placing, processing, and compaction of earthfill in these areas are
generally accomplished in a variety of ways, it is not possible to rely on an
established method to achieve specified results. In these cases, more
frequent testing is the only way to check compliance to specification
requirements.

210 H 645 Subpart H (amended March 2025) 23

 2. Determine the test location.
a. Selecting the test location is based on two areas of concern. The first is a
random test of the recently placed and compacted material. This test needs
to be representative of all the material in that section or area. The second
is testing specific areas of concern. This may be a soft area, an overly
compacted area, or areas susceptible to poor compaction effort. Both tests
are required to adequately document the quality of the in-place fill.
b. It is also important to verify that the moisture and density at the test
location is representative of the earthfill in the specific lift being tested.
Close observation is needed to verify the consistency of processing and
compaction of the lift. If there is an observed difference in the processing
and compaction effort, the testing must be in the weaker section identified.
The elevation of the compacted lift surface and the actual elevation of the
test location must be recorded if the test is documented. The horizontal
location is usually referenced as an offset distance from a centerline or by
rectangular coordinates.
3. Obtain the sample.
a. Sampling and maintaining the integrity of the sample are critical steps to
the testing process. All the in-place density methods, except for the
nuclear method, take a soil sample from the test area. This removed soil
sample provides a weight and volume determination needed to compute a
density value and provide a portion of the sampled soil for the
determination of water content. Until these determinations can be made,
the sampled soil needs to be protected from moisture loss by being placed
in a plastic storage bag or an air-tight container. Sampling requirements
are typically specified in all ASTM test standards. Sampling according to
the specification is required for the test to be valid. When sampling is
required to conduct a test (generally required except for when the nuclear
gauge is employed), consider sampling the most critical part of the fill.
Proper sampling may also involve scrapping off the surface lift to be
tested to ensure all loose material is removed and not affect the results of
the test.
4. Perform the specified test.
a. The required test methods are identified in the performance specifications
where they call for end-result testing of in-place earthfill. CS-23 contains
two classes of compaction that fit this category, classes A and B. The QA
inspector must have an operational knowledge of each standard test
method.
b. For soil in-place density, the following test methods apply: sand-cone
method, ASTM D1556; drive cylinder method, ASTM D2937; and a
nuclear method, ASTM D6938 Nuclear Moisture/Density Method or
ASTM D8167 Low-Activity Nuclear Density Method. NRCS does not
currently have conventional nuclear density gauges used for ASTM
D6938, but contractor’s quality control personnel may be using them.

210 H 645 Subpart H (amended March 2025) 24

 NRCS does have several of the D8167 gauges for use in QA activities.
There are other methods, but these are the most used in NRCS field work.
c. For water content testing, the following tests apply: oven dry method,
ASTM D2216; microwave oven method, ASTM D4643; direct heating
method, ASTM D4959; calcium carbide gas pressure (Speedy Moisture
Tester) method, ASTM D4944; and nuclear method, ASTM D6938. These
are the primary methods used for NRCS field work. Moisture content
testing yields a value used to convert wet density to dry density. Moisture
content testing is also needed to verify that compaction moisture is within
the specified range. Other important uses are to assess the water content of
borrow materials and monitor drying efforts for fill materials that are too
wet for compaction.
d. The ASTM methods are well written and easy to follow once the format
becomes familiar. However, several significant precautions for these
methods are outlined below.
e. Density Testing
(1) Sand Cone Method, ASTM D1556
(a) A proven method that requires excellent technique. The procedure
for this test involves determining the volume of excavated soil
from a hole in the compacted earthfill. The excavation is carried
out through a base plate that is secured to the prepared soil surface
prior to any digging.
(b) Free-flowing sand of known unit weight is then poured into the
hole from a jug and a calibrated cone (figure 645H-14). The
volume of the hole is determined from this operation. Knowing the
volume, the density of the soil in-place can then be determined by
the weight of the soil removed and saved from the excavation. This
saved soil is then used to determine the in-place water content of
the test material. Care must be taken to place all the soil being
excavated from the hole in a suitable container to protect it from
losing moisture prior to performing the density and water content
determinations.
(c) The sand calibration required to determine the unit weight of the
sand must be accurate and performed frequently. Temperature and
relative humidity variations affect these calibrated values. Proper
storage and protection of calibrated sand will greatly reduce this
vulnerability. ASTM D1556 has a specific sand gradation
requirement, but NRCS experience has shown a need for even
tighter control on the sand. The NRCS recommends using Ottawa
Sand as described in ASTM C778. ASTM D1556 covers this
calibration procedure and all calculations in detail.
(d) This test is appropriate for all fine-grained soils from the USCS
(CL, ML, CH, MH), sands with fines (SM, SC), and gravels with
fines (GM, GC) with particles less than 1.5-inch diameter.

210 H 645 Subpart H (amended March 2025) 25

 (e) The test is generally not suitable for noncohesive soils or soft
saturated soils because it is difficult to maintain a stable hole in
these soils while performing the test.
(f) A primary advantage of this test method is the opportunity to
visually observe the in-place soil material as it is excavated out of
the hole. This observation helps to identify borrow materials and
evaluate fill material processing. Identifying borrow materials is
important for determining the standard reference curve to be used
for evaluating the density test results.
(g) As mentioned earlier, excellent technique is critical for obtaining
good test results. There are some cautions that deserve mention as
they are common causes of errors. The operator is cautioned to
ensure the hole from which the soil sample is obtained does not
deform when running the sand cone test. Placing a knee near the
hole may deform the hole thereby reducing the volume of the hole
and causing the reported density values to be erroneously high.
Also, vibrations might occur from heavy equipment operating near
the test site causing an increase in sand density, which would cause
reported density values to be erroneously low.
(h) Appendix B contains worksheet WS 8.8 to aid in the performance
of the sand cone moisture-density determination.
Figure 645H-14: Sand Cone Apparatus (NRCS-TX)

(2) Drive Cylinder Method, ASTM D2937

(a) A simple and accurate method, this test is accomplished by driving
a thin-walled metal cylinder of known volume into the prepared
surface of the compacted soil using a drop hammer. The cylinder is
then extracted from the soil by digging it out, and the sample is

210 H 645 Subpart H (amended March 2025) 26

 trimmed to determine the volume of soil in the cylinder. The
weight of the soil in the cylinder then allows the determination of
the density of the test sample. Soil removed from the cylinder is
used to determine the water content of the test sample. While the
density is determined with the soil still in the cylinder, the soil
must be removed and protected prior to determining the water
content. All test details and required calculations are covered in
ASTM D2937.
(b) Figures 645H-15 through 645H-17 show the cylinder and drop
hammer method.
(c) This method is appropriate for all fine-grained cohesive soils but
will not work for cohesionless soils. It also has limited use in
gravelly soils due to the likelihood of deformations and voids
along the sides of the cylinder caused by the coarse material as the
cylinder it is being driven in the ground.
(d) This method also provides a visual observation of the in-place
material that can be used to evaluate borrow materials and mixing
consistency. The cylinder must be driven into the soil deep enough
to get a complete sample. The excavation must extend below the
bottom of the cylinder to allow removal without damaging the
captured soil sample inside. The test method may be both laborious
and tricky in some soils.
(e) Appendix B contains worksheet WS 8.10 to aid in the performance
of the drive-cylinder moisture-density determination.
Figure 645H-15: Drive Cylinder Apparatus – Driving head placed on the cylinder
(NRCS-TX).

210 H 645 Subpart H (amended March 2025) 27

Figure 645H-16: Drive Cylinder Apparatus – Extracting Cylinder after being
driven into the ground (NRCS-TX).

Figure 645H-17: Drive Cylinder Apparatus – Trimming soil from cylinder after
excavation (NRCS-TX).

210 H 645 Subpart H (amended March 2025) 28

 (3) Nuclear Method ASTM D6938 and Low-Acidity Nuclear Method
ASTM D8167
(a) There are two nuclear methods used for density testing. The first is
the conventional moisture density gauge covered by ASTM
D6938. Although this method is the fastest and easiest to use, it
requires certification training for safe use, transportation, and
operation of the nuclear device. Also, regulations require periodic
inspections and calibration of the equipment, and equipment
storage is strictly regulated.
(b) The nuclear gauge can be used to determine the in-place density
and water content of most soil or aggregate mixtures used in
earthfills and structural backfill. The wet density is determined by
gamma rays either in a backscatter mode from the surface or by a
direct transmission mode from the tip of a probe inserted into the
material being tested. The direct transmission mode measures an
average density from the depth at which the source rod is
positioned back to the detectors located in the base of the gauge.
The direct transmission mode is generally more accurate and
reliable than the backscatter mode and preferred by most users.
(c) The water content is determined by a second source that emits
neutrons in a backscatter mode and provides a measurement that is
biased toward the surface. If the soil moisture content is uniform
throughout the soil profile, this measurement will not present a
problem. If the moisture in the soil profile is variable, the value
determined may not accurately represent the moisture content over
the full depth of the density measurement.
(d) Water content values obtained by the nuclear gauge may be
misleading in some soils. The gauge reports hydrogen ions in the
soil. If the only hydrogen ions present are in the soil moisture, the
measurement will be quite accurate. When other sources of
hydrogen are in the soil, the moisture reading will be erroneously
high making it necessary to calibrate the gauge to provide a true
moisture value. Although rare, there are some soils that absorb
neutrons and cause the reading to be low. Either way, a calibration
is usually required.
(e) Since the nuclear gauge measures only wet density and moisture, it
must compute the dry density. If the moisture measurement is
erroneously high, the computed dry density will be low. Some QA
inspectors claim that by not calibrating the gauge, the resulting dry
density value will be lower than the specified minimum and, thus,
be conservative. However, there may be concerns with
overcompaction that would not be addressed if the true value of
density is not known and therefore, verifying that compaction
moisture is within the specified range is important. Thus, it is
essential to have water content calibration corrections, especially

210 H 645 Subpart H (amended March 2025) 29

 for most fine-grained soils, as these generally contain no moisture
related hydrogen or contain elements that can absorb neutrons.
(f) The procedures and precautions for this test method are covered in
detail in ASTM D6938.
(g) ASTM D8167 covers the use of a low-activity nuclear gauge for
soil density. This equipment was developed recently and has the
advantage of not having the license requirements of the
conventional nuclear gauge. The testing process is very similar
except this gauge only measures soil moist density. The moisture
content of the soil must be determined with another test and the dry
density computed. The gauge may come furnished with a moisture
probe, or the water content can be determined by one of the
methods shown below.
(h) Appendix B contains worksheet WS 8.11 to aid in the performance
of the nuclear gauge moisture density determination.
f. Water Content Testing
(1) Oven Dry Method, ASTM D2216
(a) This test method is the standard to which all other water content
tests are calibrated and compared. The main drawback is the test
takes 24 hours to get a result.
(2) Microwave Oven Method, ASTM D4643
(a) Les reliable and less expensive than the oven dry method, the
readily available nature of equipment makes this method more
adaptable to field work.
(3) Direct Heating Method, ASTM D4959
(a) Simple equipment and field oriented. Be cautious not to burn off
organic material, thereby yielding an incorrect higher water
content value.
(4) Speedy Moisture Tester Method, ASTM D4944
(a) Handy and portable for field work. Needs calibration and has a
sample size limitation. Less accurate with the more plastic fine-
grained soils of higher water content.
(5) Nuclear Method, ASTM D6938
(a) This method requires good water content calibration for each soil
type involved in field testing. Water content generally reads high,
especially for fine-grained soils, and only rarely low with this test
method. The other limitation, previously addressed, is that the
water content is always determined in a backscatter mode and is
biased to the surface and, thus, may not be truly representative
throughout the density sample range.
(6) In all methods used for water content testing, it is extremely important
to protect the soil materials during sampling and transporting so that
they do not dry out prior to determining the water content.
(7) The QA inspector must have ready access to the applicable current
ASTM Standard specified for a given job. Onsite test results are only

210 H 645 Subpart H (amended March 2025) 30

 meaningful and valid when they are performed in accordance with the
appropriate ASTM Standard.
g. Correction for Oversized Particles
(1) When performing in-place moisture or density tests on soils containing
gravel, a correction for oversize particles may be necessary. Oversize
particles are particles that are larger than the maximum size particle in
the soil that was used to develop the Proctor curve. For example, if the
Proctor curve was made according to ASTM D698 Method A, the
curve represents a soil with a maximum particle size no greater than
the No. 4 sieve size (3/16 inch). Thus, if the field compacted soil
contains a significant number of oversize particles (an amount greater
than five (5) percent of the mass of the soil is generally considered
significant), a correction for oversize particles must be made so that
the measured field density and moisture values can be compared to the
Proctor curve values for moisture and density.
(2) The sand method may be used for testing soils with oversized
particles, but the drive cylinder is not used for soils containing gravels.
(3) The procedure described in ASTM D4718 Standard Practice for
Correction of Unit Weight and Water Content for Soils Containing
Oversize Particles must be employed whenever there are significant
amounts of oversize particles in soil that is being tested for moisture or
density. Worksheet WS 8.13 in appendix B can be used in conjunction
with D4718 to make the correction.
h. The QA inspector’s responsibilities related to test location, sampling, and
testing include verifying that:
(1) Selected in-place testing locations are truly representative of the
section or area being evaluated.
(2) Areas that have suspect compaction effort or may be too dry or too wet
are tested.
(3) Soil materials from the testing and sampling process are properly
protected to retain their integrity until the required determinations are
completed.
(4) Any test selected from those specified is performed in strict
accordance with the specified test standard and is appropriate for the
soils being tested.
(5) A moisture and density correction is made according to ASTM D4718
whenever the soil being tested contains significant amounts of oversize
particles.
(6) Test location and results are promptly recorded.
5. Evaluate the test results.
a. The evaluation of compaction and moisture test results includes selecting
the Proctor curve that best fits the tested soil and making sure the test
results are reasonable. Begin by choosing a Proctor curve.
b. Choosing a Proctor curve—A field density and water content
determination is compared to the chosen Proctor curve that has been
developed for the soil being compacted. Historically, Proctor curves for

210 H 645 Subpart H (amended March 2025) 31

 NRCS projects were developed in a laboratory from soils that were
sampled during the geologic investigation preceding the design. The
curves were available to CQC and QA who would be tasked with
determining which curves fit the various soils being used for the earthfill.
CQC and QA developed additional Proctor curves during the job if those
supplied by the lab did not represent all the soils or composite soils being
used for earthfill or earth backfill. The greatest source for error in
evaluating field test results for compaction was likely in determining
which curves fit the various soils being used to construct the earthfill.
c. To determine which Proctor curve fits the soil being tested, the QA
inspector must be able to field classify the soils by the method described
in ASTM D2488 for field classifying soils according to the USCS. Many
soils can be reliably matched with their corresponding Proctor curve by
field classification and visual appearance. Three methods are used to
verify which Proctor curve is appropriate to evaluate the field compaction
test values. These methods—the jar, one-point, and the family of curves or
one-point—are described below.
(1) Jar Method
(a) The jar method is a simple visual method and is most applicable
when only a few different soils having distinct visual appearances
are involved. Samples of soils used to develop the Proctor curves
are obtained from the laboratory and placed in jars that are labeled
with the corresponding Proctor curve designation. Soil being tested
for in-place density and water content can be compared to the jar
samples to select the Proctor curve that matches the test soil best.
(2) One-Point Method
(a) The one-point method is used to verify that a Proctor curve
matches the soil being tested. The soil being tested is compacted
according to ASTM D698 or D1557, as applicable, at a moisture
content judged to be about 2 percent dry of optimum. The results
are then plotted on the same sheet as the Proctor curve, and the
plotted point will fall on the Proctor curve if the curve and soil
match. The best match becomes the Proctor curve that will be used
for comparison to the field test results if the results do indeed plot
on or very near the Proctor curve.
(3) Family of Curves or One-point Method
(a) The jar and one-point methods work well if the soils that are being
placed and compacted correspond to the available Proctor curves.
However, most earthfills are constructed from a mixture of soils
that have different compaction characteristics than the individual
soils making up the mixture. These soil mixtures vary in field
classification and visual appearance, and these variations make it
difficult to identify which Proctor curve best matches the soil. The
family of curves or one point method can be utilized to fill in some
of the gaps between existing compaction curves. This method is
used by many of the State highway departments across the country,

210 H 645 Subpart H (amended March 2025) 32

 and there is an American Association of State Highway
Transportation Officials standard, AASHTO T 272, Standard
Method of Test for Family of Curves-One Point Method, that
covers this method.
(b) A family of curves can be developed specifically for each job by
plotting all the Proctor curves available for the job on the same
sheet. A smooth curved line is then drawn to best fit the apex of
each curve. The Proctor curves are then moved to the left or right
to align with this best-fit curved line. Do not include any curve that
must be moved so much that its moisture content values change
more than 2 percent in the family of curves. If large gaps exist
between the remaining plotted curves, additional curves may be
needed to make use of the family of curves. See the family of
curves in figure 645H-18.
(c) After the family of curves is developed, the results of a one-point
test made from any site-specific soil can be plotted on the family of
curves sheet. If the point falls on one of the curves, that curve is
selected to represent the soil being compacted and tested in the
field. If the point falls between two curves and these curves are
close together, a new Proctor curve can be free-hand drawn to pass
through the plotted point and the best-fit curved line. The shape of
the free-hand curve should follow that of the nearest existing
curves. This free-hand curve becomes the Proctor curve that will
be used for comparison to the field test results of the soil being
compacted and tested in the field. The dotted line in figure 645H-
18 depicts a freehand curve.
(d) Departments of Transportation (DOT) in several States have
developed a family of curves from Proctor curves in their States.
These curves have been successfully used in other areas outside the
State where they were developed. If one of these DOT family of
curves is to be used on any job, especially on a job located outside
of the state where it was developed, its applicability to the soils on
the jobsite must be verified. This can be done by plotting a few
site-specific Proctor curves on the same sheet as the DOT family
of curves to see if they align with the family of curves. Alignment
would verify that the DOT family of curves applies to these site-
specific soils and very likely would apply to other soils on the site.
d. Make sure test results are reasonable.
(1) Both the ASTM D698 and D1557 standards require a zero air voids
(ZAV) curve be plotted on the same sheet as the Proctor curve. The
ZAV curve will always plot on the wet side or to the right of the
Proctor curve. Since the absolute maximum density of any soil at a
specific moisture content is attained when all the air voids are filled
with water, the ZAV curve represents the maximum density of the soil
at various water contents.

210 H 645 Subpart H (amended March 2025) 33

 (2) Figure 645H-19 shows a ZAV curve (100% saturation line) plotted
with the Proctor curve. If the corresponding test results of field density
and water content were plotted on this same sheet, the point will plot
to the left of the ZAV curve. If it plots to the right of the ZAV curve,
something is wrong since water content cannot increase once the soil
is saturated.
(3) Furthermore, compacting soil to 100 percent saturation, although not
impossible, is very unlikely to occur with conventional compaction
equipment. Soils are usually compacted to a density that falls within
the 70 to 90 percent saturation range. It is good practice to plot a 70
percent ZAV curve and a 90 percent ZAV curve along with the Proctor
curve and ZAV curve. Field test results can then be plotted on the
same sheet to quickly determine if they fall within the reasonable
range of 70 to 90 percent saturation. Any field test results that fall
outside of this range are suspect and must be evaluated further. If
results plot outside of the 70 to 90 percent saturation range, there is a
high probability that either the Proctor curve does not correspond to
the tested soil or there was an error made when performing the test or
recording test results.
(4) ASTM D698 and D1557 include the formula for computing the ZAV
curve. The specific gravity of the soil must be known to compute the
ZAV curve. The formula is:

(𝛾𝛾𝑤𝑤 )(𝐺𝐺𝑠𝑠 )−𝛾𝛾𝑑𝑑

𝑤𝑤𝑠𝑠𝑠𝑠𝑠𝑠 = (𝛾𝛾𝑑𝑑 )(𝐺𝐺𝑠𝑠 )
𝑥𝑥 100 where:
wsat = the maximum water content at a given dry density in
percent
γd = dry density in lb/ft3
γw = unit weight of water (62.32 lbf/ft3)
Gs = the specific gravity of the soil

210 H 645 Subpart H (amended March 2025) 34

Figure 645H-18: Typical Family of Curves

Figure 645H-19: Proctor and ZAV Curves

(5) The 70 percent and 90 percent ZAV curves are computed by the
following formulas, which are the same formula multiplied by 0.7 and
0.9.
(𝛾𝛾𝑤𝑤 )(𝐺𝐺𝑠𝑠 ) − 𝛾𝛾𝑑𝑑
𝑤𝑤70% = 𝑥𝑥 70
(𝛾𝛾𝑑𝑑 )(𝐺𝐺𝑠𝑠 )

(𝛾𝛾𝑤𝑤 )(𝐺𝐺𝑠𝑠 ) − 𝛾𝛾𝑑𝑑

𝑤𝑤90% = 𝑥𝑥 90
(𝛾𝛾𝑑𝑑 )(𝐺𝐺𝑠𝑠 )

210 H 645 Subpart H (amended March 2025) 35

 where:
- w(70%) = 70 percent of maximum water content at a given dry
density
- w(90%) = 90 percent of maximum water content at a given dry
density
(6) See figure 645H-20 for these curves and the ZAV curve for a specific
soil. Several points are plotted along with the curves. The points were
plotted from real results obtained by testing compacted soil in the
field. If the specific gravity value used to develop the ZAV curves is
that of the tested soil, the following are true:
(a) The moisture and dry density values that fall within the 70 and 90
percent ZAV lines are reasonable values.
(b) Moisture and dry density values used for plotting points that fall
outside of these lines are suspect.
(c) The moisture or dry density value used to plot any point that lands
to the right of the ZAV curve is wrong.
(d) The selected Proctor curve represents the soil or composite soils
being tested.
Figure 645H-20: 70% ZAV Curve, 90% ZAV Curve, and ZAV Curve

6. Record the test results.

a. The designer specifies the tests for determining the in-place water content
and density of earthfill materials. The water content is specified as a range
above or below optimum water content. The density is specified as a
percentage of the maximum dry density. An example of an earthfill
specification might look like this: “the earthfill is to be compacted to 95
percent of maximum dry density as defined in ASTM D698. The water

210 H 645 Subpart H (amended March 2025) 36

 content at the time of compaction shall range from –1 to +2 percent of the
optimum water content as defined in ATSM D698.” The QA inspector is
responsible for verifying and documenting that the in-place processed and
compacted soil complies with specifications such as these, but this is only
a part of the QA inspector’s responsibilities. The other part is to verify that
the results represent the soil throughout the lift that is being compacted
and that enough test results are evaluated to verify that the entire lift is
compacted to the density and at the moisture content specified.
7. Assess the need for additional testing. There may also be action required to
correct noncompliant work identified from negative results.
a. If test results fail, action must be taken. The first step is to verify the test
was properly conducted and the correct Proctor curve was used to evaluate
the test results. This requires retesting near the same failed test location. If
the second test is properly conducted and fails, more tests may be needed
to determine the extent, both depth and lateral limits, of the noncompliant
earthfill or earth backfill. This is the CQC’s responsibility. They must
work with the contractor to fix any noncompliant earthfill or backfill. This
may require more processing, compaction, and even removal of some in-
place material. When failures occur, every effort must be made to
determine the cause of the failure and make necessary adjustments to the
earthfill process to limit reoccurring failure. The QA inspector must allow
this process to play out before getting too involved. When the CQC is
satisfied the noncompliant work is fixed and the process is adequately
adjusted to limit reoccurring failure, the QA inspector must verify
specification compliance and document the scenario including any
consultation with the CQC and contractor.
8. The QA inspector’s responsibilities related to evaluating test results include
verifying that:
a. Test results are reasonable.
b. Soils are retested if initial test results are unreasonable.
c. Action is taken to correct noncompliant work when reasonable test results
fail.
d. Any reworked areas are retested to verify and document compliance with
specifications.

645.83 Records and Reports

A. Accurate records of testing and test results are needed to document:

1. The method used to select the Proctor curve.
2. The horizontal and vertical location of the field density and moisture test and
the section or area represented by the test.
3. The standard test method applied and the test results.
4. Any other actions or testing done for verification of test results.

210 H 645 Subpart H (amended March 2025) 37

 5. Any actions taken to correct noncompliant work identified by negative tests.
B. The job diary must reference or further explain all the information listed.
1. These records and reports are related to earthfill and rockfill:
a. Daily diary—used to record the day-to-day activities of earthfill and
rockfill construction.
b. WS 8.1—Test Fill Report.
c. WS 8.2—Weekly Summary of Density Determinations.
d. WS 8.3—Determination of Volume of Compaction Mold ASTM D698
and D1557.
e. WS 8.4—Worksheet for Reference Density Compaction Data ASTM
D698 and D1557.
f. WS 8.5—Moisture Correction Determination.
g. WS 8.6—Earthfill Construction Report.
h. WS 8.7—Bulk Sand Density Determination and Calibration of Cone and
Base Plate for ASTM D1556.
i. WS 8.8—In-Place Moisture-Density Determination: Sand Cone Method:
ASTM D1556.
j. WS 8.9—In-Place Moisture-Density Determination: Rubber Balloon
Method: ASTM D2167.
k. WS 8.10—In-Place Moisture-Density Determination: Calibrated Cylinder
Method: ASTM D2937.
l. WS 8.11—Nuclear Compaction Test Data for ASTM D6938.
m. WS 8.12—In-Place Moisture-Density Determination: Template & Plastic
Liner Method: ASTM D5030.
n. WS 8.13—Correction of Unit Weight and Water Content for Soils
Containing Oversized Particles.
2. Appendix B includes the worksheets, guidance to complete them, examples
worksheets, and example entries. The QA inspector must reference the
applicable worksheet in their job diary entries and may include some test
results when they relate to the daily activity and need clarification. See
appendix C for sample job diary entries.

645.84 References

Association of State Highway Transportation Officials. 2004. Standard Method of

Test for Family of Curves One Point Method. T 272, Washington DC.
U.S. Army Corps of Engineers. 1994. Engineering and Design—Test Quarries and
Test Fills, EM 1110–2–2301. Washington, DC.
U.S. Department of Agriculture, Natural Resources Conservation Service. 2017.
National Engineering Manual, 210–533-M, 4th Ed., Geotechnical Engineering,
Subpart B, Engineering Classification of Soils. Washington, DC.

210 H 645 Subpart H (amended March 2025) 38

 U.S. Department of Agriculture, Natural Resources Conservation Service. 2019.
National Engineering Manual, 210-531-M, 4th Ed., Amend. 3, Geology, Washington,
DC.
U.S. Department of Agriculture, Natural Resources Conservation Service. National
Engineering Handbook, Part 642, Chapter 2, Construction Specification 25—
Rockfill, 2001
U.S. Department of Agriculture, Natural Resources Conservation Service. National
Engineering Handbook, Part 631, Geology, Chapter 4, Engineering Classification of
Rock Materials. 2022, Washington, DC.
U.S. Department of Agriculture, Natural Resources Conservation Service, National
Engineering Handbook, Part 642, Chapter 2, Construction Specification 23—
Earthfill, 2020, Washington, DC.
U.S. Department of Agriculture, Soil Conservation Service. Technical Release 26.
The Use of Soils Containing More than 5 Percent Rock Larger than the No. 4 Sieve.
1964. Washington, DC.
U.S. Department of Agriculture, Soil Conservation Service. 1984. Soil Mechanics
Note 8: Soil Mechanics Testing Standards. (Incorporated into National Engineering
Handbook, Part 633, Soil Engineering), Washington, DC.

210 H 645 Subpart H (amended March 2025) 39