Published 1965

30 Bulk Density 1

[Link]
University 0/ Minnesota
St. Paul, Minnesota

30-1 GENERAL INTRODUCTION

Soil bulk density Db is the ratio of the mass to the bulk or macroscopic
volume of soil particles plus pore spaces in a sample. The mass is deter-
mined after drying to constant weight at lO5°C., and the volume is that of
the sample as taken in the field.
Bulk density is a widely used value. It is needed for converting water
percentage by weight to content by volume, for calculating porosity when
the particle density is known, and for estimating the weight of a volume
of soil too large to weigh conveniently, such as the weight of a furrow
slice, or an acre-foot.
Bulk density is not an invariant quantity for a given soil. It varies with
structural condition of the soil, particularly that related to packing. For
this reason it is often used as a measure of soil structure.
The clod method, core method, and excavation method consist essen-
tially of drying and weighing a known volume of soil. These methods differ
principally in the way the sample of soil is obtained. A different principle
is employed with the radiation method. Transmitted or scattered gamma
radiation is measured; and, with suitable calibration, the density of the
combined liquid-solid components of a soil mass is determined. Correction
is then necessary to remove the component of density attributable to liquid
that is present. The radiation method is an in situ method.
Clod and core methods have been used for many years. Excavation
methods were developed in recent years, chiefly by soil engineers. Radia-
tion methods are relatively new, having been developed since 1950.
In most agricultural soils work, bulk density is expressed in grams per
cubic centimeter (g. cm. -3). In these units volume weight is equal, and
bulk specific gravity (or apparent specific gravity) is nearly equal, numeri-
cally, to bulk density. The term bulk density is preferred over the terms

1 Paper No. 4433 of the Scientific Journal Series, Minnesota Agr. Exp. Sta., St.
Paul.
374
30-2 CORE METHOD 375
volume weight, bulk specific gravity or apparent specific gravity. Because
of its definition, in terms of mass rather than weight, it has universal ap-
plicability, and conforms more closely to accepted physical terminology
than does volume weight. Specific gravity terms are relative terms, being
the ratio of the mass of a dry bulk volume of soil to the mass of an equal
volume of water. Since the mass of water per unit volume varies with tem-
perature, bulk specific gravity varies numerically with the temperature at
which the measurement is made.
In many engineering applications, bulk density is expressed in pounds
per cubic foot. One may convert from g. cm. -3 to lb. ft. -3 by multiplying
by 62.4, the mass, in pounds, of a cubic foot of water at the temperature
of its maximum density, i.e., 4 cC.

30-2 CORE METHOD

30-2.1 Introduction

With this method, a cylindrical metal sampler is pressed or driven into

the soil to the desired depth and is carefully removed to preserve a known
volume of sample as it existed in situ. The sample is dried to 105°C. and
weighed. Bulk density is the oven-dried mass divided by the field volume
of the sample.
The core method is usually unsatisfactory if more than an occasional
stone is present in the soil.

30-2.2 Method 2

30-2.2.1 SPECIAL APPARATUS

Core samplers vary in design from a thin-walled metal cylinder to a

cylindrical sleeve with removable sample cylinders that fit inside. A widely
used and very satisfactory sampler consists of two cylinders fitted one in-
side the other. The outer one extends above and below the inner to accept
a hammer or press at the upper end and to form a cutting edge at the
lower. The inside cylinder is the sample holder. The inside diameters of
the two cylinders when nested are essentially the same at the lower end,
the inner being fitted against a shoulder cut on the inner surface of the
outside cylinder. Figure 30-1 shows such a sampler (available in slightly
different design from the Utah Scientific Research Foundation, Utah State
University Campus, Logan, Utah).

"U. S. Dept. Agr. (1950, p. 121), Russell (1949), and Am. Soc. Testing Mater.
(1958, p. 442).
37'6 BULK DENSITY

Fig. 30-1. Typical double-cylinder, hammer-driven core sampler for obtaining soil
samples for bulk density.

Numerous samplers have been described in the literature. Some of the

more recent and accessible ones are described by Lutz (1947), Jamison
et aI. (1950), Baver (1956, p. 181), and U. S. Dept. Agr. (1954, p. 159).

30-2.2.2 PROCEDURE
The exact procedure for obtaining the samples depends on the kind of
sampler used. The following steps apply when the widely known double-
cylinder sampler is used.
Drive or press the sampler into either a vertical or horizontal soil sur-
face far enough to fill the sampler, but not so far as to compress the soil
in the confined space of the sampler. Carefully remove the sampler and its
contents so as to preserve the natural structure and packing of the soil as
nearly as possible. A shovel, alongside and under the sampler, may be
needed in some soils to remove the sample without disturbance. Separate
the two cylinders, retaining the undisturbed soil in .the inner cylinder. Trim
the soil extending beyond each end of the sample holder (inner cylinder)
flush with each end with a straight-edged knife or sharp spatula. The soil
sample volume is thus established to be the same as the volume of the
sample holder. In some sampler designs, the cutting edge of the sampler
has an inside diameter slightly less than the sample holder, so as to reduce
30-3 EXCAVATION METHOD 377
friction as the soil enters the holder. In these cases, determine the diameter
of the cutting head and use this to calculate the sample holder volume.
Transfer the soil to a container, place it in an oven at lOsoC. until con-
stant weight is reached, and weigh it. The bulk density is the oven-dry
mass of the sample divided by the sample volume.

30-2.2.3 COMMENTS
It is often desired to make other measurements on the same samples
taken for bulk density. Water content determinations, like bulk density, do
not require that the soil be kept undisturbed during transport to the labo-
ratory and drying. They do require a wet weight, however, so that the
samples must be transported from field to laboratory in containers that do
not permit loss of water. One-pint cylindrical waxed-paper cartons with
lids will receive the 3-inch-diameter, 3-inch-long cylinders, and serve very
well to transport the sample with almost no loss of water. Some studies,
such as pore-size distribution, require that the sample be maintained as
free of disturbance as possible.
Core samples should not be taken in wet or dry soils. In wet soils, fric-
tion along the sides of the sampler and vibrations due to hammering are
likely to result in viscous flow of the soil and thus in compression of the
sample. When this occurs the sample obtained is unrepresentative, being
more dense than the body of the soil. Compression may occur even in dry
soils if they are very loose. Whenever a sample is taken, one should care-
fully observe whether the soil elevation inside the sampler is the same as
the undisturbed surface outside the sampler. One can only roughly esti-
mate in this manner whether the density of the sample is changing because
of sampling.
In dry or hard soils, another problem arises. Hammering the sampler
into the soil often shatters the sample, and an actual loosening during sam-
pling may occur. Samplers pressed into the soil usually avoid the vibration
which causes this shattering. Close examination of the soil sample usually
allows one to estimate whether serious shattering occurs. And, as in the
case of wet soils, soil level inside and outside the sampler must remain the
same if the sample is to be considered satisfactory.

30-3 EXCAVATION METHOD

30-3.1 Introdudion

Bulk density is determined in this method by excavating a quantity of

soil, drying and weighing the soil, and determining the volume of the ex-
cavation. The volume is determined, in the sand-funnel method, by filling
the hole with sand, of which the volume per unit mass is known. The vol-
378 BULK DENSITY

Fig. 30-2. Apparatus for sand-funnel technique of determining soil bulk density in
place.

ume is determined, in the rubber-balloon method, by inserting a balloon

into the excavation and by filling it with water, or other fluid, until the
excavation is just full. The volume of the excavated soil sample is then
equal to the volume of the fluid dispensed.
Excavation methods were developed by soil engineers who required a
method suitable for use in gravelly soil. This is the chief advantage of the
sand-funnel and rubber-balloon methods over the core method. Soil that is
removed for drying is disturbed, precluding its use for other measurements,
such as pore-size distribution, where undisturbed structure is necessary.
Also, in contrast to the core method, determinations cannot be made ver-
tically into the walls of a pit, as is sometimes desirable where anisotropy
is to be measured.

30-3.2 Method 3

30-3.2.1 SPECIAL APPARATUS

30-[Link] Sand-Funnel Apparatus 4 (see Fig. 30-2).
1. A metal funnel 15 to 18 cm. at its largestdiameter, fitted with a valve
on the stem. Attached to the stem, when the funnel is inverted, is a sand
container.

" Am. Soc. Testing Mater. (1958. pp. 422-441).

< This apparatus can be purchased from Soiltest, Inc .. 2205 Lee Street, Evanston,
Ill.
30-3 EXCAVATION METHOD 379

2. A standard sand that is clean, dry, and free-flowing. Particle size should
be fairly uniform to avoid possible separation in the dispenser with con-
sequent error in calibration. Sand particles passing a No. 20 and re-
tained on a No. 60 sieve are recommended.
3. A template, consisting of a thin, flat, metal plate approximately 30-cm.
square, with a hole 10 to 12 cm. in diameter in its center.
4. Scales to weigh to 5 g.
30-[Link] Rubber-Balloon Apparatus.
1. A thin-walled rubber balloon (may be purchased from Barr Rubber Co.,
Sandusky, Ohio, and the Anderson Rubber Co., Akron, Ohio).
2. A I,OOO-cc. graduated cylinder, and a water container.
3. A template, described in 3 above.
Rubber-balloon density apparatus is available from several manufactur-
ers supplying soil testing equipment (one supplier is Soiltest, Inc., 2205
Lee Street, Evanston, IlL). The apparatus made commercially has the
convenience of a volumetrically calibrated water container-dispenser, with
suction facilities for returning the water to the container for re-use (see
Fig. 30-3).

Fig. 30-3. Apparatus for

determining soil bulk den-
sity in place by the rub-
ber-balloon technique.

30-3.2.2 PROCEDURE
Level the soil surfacc, and rcmove loose soil at the test site. Place the
template on the soil. Excavate a soil sample, through the center hole of
the template, leaving a hole with a diameter of approximately 12 cm. and
a depth of approximately 12 cm., or other value as desired. A large spoon
is convenient for excavating. Recover all excavated soil in a container, be-
ing careful to include any loose soil that has fallen in from the sides of the
excavation. Determine the oven-dry soil mass by drying it to 105 DC. and
weighing it.
380 BULK DENSITY

Determine the volume of the test hole in one of two ways:

30-[Link] Sand-Funnel Procedure. Fill the hole with the sand even
with the bottom of the template. Level the sand at the bottom of the tem-
plate with a spatula if necessary, but disturb it as little as possible to avoid
packing the free-flowing sand. (Dispensing the sand through a funnel
placed on the template, as is done with commercially available equipment,
avoids the problem of leveling the sand. The excavation, as well as the
funnel, is filled by free flow of sand, the predetermined weight required to
fill the funnel being subtracted as a tare.)
Determine the weight of sand required to fill the test excavation by
weighing it to the nearest 5 g. Precalibrate the mass to volume of sand
with sand faIling at a similar height and rate of flow as in the test pro-
cedure; Using the calibration curve, or values derived from it, determine
the volume of the excavation fr0111 the measured mass of sand dispensed.
30-[Link] Rubber-Balloon Procedure. Place the rubber balloon in
the test hole, and fill the balloon with water to the bottom of the template.
Determine the volume of water required to the nearest 2 cc. (A 1,000-cc.
graduate has markings to 10 cc., but one can estimate to 2 cc. if the gradu-
ate is placed on a horizontal surface.)
Calculate bulk density from the oven-dry mass of the excavated sample
and the volume of the test excavation.

30-3.2.3 Comments

Bulk density can be estimated accurately by excavation methods carried

out carefully. The relatively large sample (a cylinder of 12-cm. diameter
and of 12-cm. depth has a volume of 1,357 cc.) is at once an advantage
and a disadvantage. The disadvantage is the lack of discrimination to a
localized horizon. The advantage is that small errors in measuring water
volumes or sand weights result in insignificant errors.
An error of 5 cc. in liquid volume will give an error of 0.005 in a sam-
ple having a bulk density of 1.36 g. cm.- 3 • Though one determines the
water dispensed to perhaps 2 cc., a much greater source of error in water
measurement arises in determining when the water in the excavation is
level with the bottom of the template. Extreme care and judgment are
required in this. The volume of the balloon itself, being of the order of 2
cc., will give a considerably smaller error than the volume measurement,
and can be neglected.
An error in weighing the sand of 7 g. gives an error of 0.005 if the bulk
density is 1.36 g. cm. -3. As in the balloon technique, greater error is likely
to result in the precision with which one can determine the sand level at
the bottom of the template. An error of 1 mm. in this level will result in
an error of 0.01 in the bulk density. Extreme care is therefore required to
30-4 CLOD METHOD 381

obtain the sand level at the template bottom. The need for the dispensing
funnel in reducing this error is obvious.
A comparison of the sand funnel and radiation methods was made by
Mintzer (1961), and the results are summarized in Table 30-1.

30-4 CLOD METHOD

30-4.1 Introduction

The bulk density of clods, or coarse peds, can be calculated from their
mass and volume. The volume may be determined by coating the clod with
a water-repellent substance and by weighing it first in air, then again while
immersed in a liquid of known density, making use of Archimedes' principle.
The clod or ped must be sufficiently stable to cohere during coating, weigh-
ing and handling.

30-4.2 Method 5

30-4.2.r SPECIAL APPARATUS

1. A balance, modified to accept a container basket for the sample, the

basket being attached to the balance lever arm by a thin wire to allow
weighing a sample when it is immersed in a container of liquid.
2. A coarse-mesh wire pan or basket to contain the sample while it is
being dipped in a water-repelIent substance and while it is being
weighed. Thin wire mesh, 1 to 2 meshes per inch, will keep the amount
of repellent held by the basket to a low value, particularly along the
line of contact with the clod. Attach the basket to the balance arm with
a single strand of thin wire to keep at a small value the difference in
container buoyancy in case of a slightly variable immersion depth from
one sample to another.
3. Paraffin in a container kept at, or a few degrees above, 60°C. into
which the soil sample contained in the wire basket can be dipped.

30-4.2.2 PROCEDURE

Air-dry the clod or ped on which measurements are to be made. Place

it in the wire basket, and weigh it suspended in air. Dip the clod and con-
tainer momentarily in melted paraffin, and allow the excess to drain. When
the adhering paraffin solidifies, weigh the clod, paraffin, and container to-
gether. Suspend the system in water, and weigh it again. Determine the
tare weight of the basket in air and in water, since its weight will, of course,
differ in the two fluids. To obtain a correction for water content of the soil,
• Since the first printing of this Monograph, superior coating materials have been
developed which, if used, will alter the method somewhat. See Brasher et al. (1966)
and Abrol and Palta (1968).
382 BULK DENSITY

break open the clod, remove a sample of soil, and weigh the sample be-
fore and after drying it at lOsoC.
Calculate the oven-dry mass of the soil sample Wods from the subsample
that is removed from the clod after other weights are taken as follows:

W Wsa
ods = I + (PjlOO)
where
P = percent water, on an oven-dry basis, found in the subs ample, and
W. a = net weight of soil clod or ped in air.
Calculate bulk density as follows:

where
dw = density of water at temperature of determination,
Wod• = oven-dry weight of soil sample (clod or ped),
W. pw = net weight of soil sample plus paraffin in water,
Wpa = weight of paraffin coating in air, and
dp = density of paraffin (approximately 0.9).

30-4.2.3 COMMENTS
The clod method usually gives higher bulk-density values than do other
methods (see Tisdall, 1951). One reason is that the clod method does not
take the inter-clod spaces into account. A second reason is that the soil
volume is the air-dry volume, which is likely to be slightly less than the
volume of a field-moist sample used in other methods.
Extreme care should be exercised to get naturally occurring masses of
soil. Clods on or near the soil surface are likely to be unrepresentative, for
these are often formed by packing with tillage implements. Natural soil
masses, or coarse peds, that are more representative should be sought.
If bubbles appear on the paraffin when the sample is weighed in water,
or if the weight in water increases with time, water is penetrating the clod,
and the sample must be discarded.
Parafin is be&t used between 60° and 70°C., a few degrees above the
melting point, when clods are dipped. At this temperature, it quickly solidi-
fies upon removal of the clod and is less likely to penetrate the pores of
the clod than at higher temperatures. Furthermore, at higher temperatures,
pin-sized air leaks through the paraffin seal are more likely because of
penetration of heat into the clod with consequent expansion of soil gases
that continues after surface solidification of the paraffin.
Several other substances have been used as water seals, including col-
lodion, wax mixtures, oils, and synthetic resins. The claimed advantage of
these over paraffin is that a thin film that does not appreciably change the
30-5 RADIATION METHOD 383

clod volume can be applied, thus avoiding the uncertainty of a correction

(sec Russell and Balcerek, 1944).

30-5 RADIATION METHODS

30-5.1 Introduction

The transmission of gamma radiation through soil or scattering within

soil varies with soil properties, including bulk density. By suitable calibra-
tion, measurements of either transmission or scattering of gamma radiation
can be used to estimate bulk density.
In the transmission technique, two probes at a fixed spacing are lowered
into previously prepared openings in the soil. One probe contains a Geiger
tube, which detects the radiation transmitted through the soil from the
gamma source, located in the second probe. The scattering technique em-
ploys a single probe containing both gamma source and detector separated
by shielding in the probe. It can be used either at the soil surface or placed
in a hole, depending on design of the equipment.
Radiation methods have several advantages, among which are minimum
disturbance of the soil, short time required for sampling, accessibility to
subsoil measurement with minimum excavation, and the possibility of con-
tinuous or repeated measurements at the same point.
The transmission or double-probe technique has the advantage of con-
fining the sample to a soil horizon a few centimeters in vertical dimension
between the probes. The single probe used with the scattering technique
sees a somewhat spherical sample varying from about 20 to 75 cm. in di-
ameter, depending on characteristics of the sampler and the soil. Thus it
is not well suited to many soils studies. The scattering technique also re-
quires a greater source strength than the transmission technique in most
prototype designs described in the literature.
Both techniques [Link] the bulk density of all phases combined. The
densities of gaseous components are insignificant in comparison to those of
the solid or liquid components, and can therefore be ignored. It is neces-
sary, however, to determine the water content of the soil at sampling time
and to apply a correction to obtain bulk density on a dry-soil basis.

30-5.2 Methods6

30-5.2.1 SPECIAL APPARATUS

30-[Link] Transmission Apparatus. 7
1. Double-probe densitometer: The design may vary with the need. The
• Neville and Van Zelst (1961) and Vomocil (1954).
T This apparatus may be purchased from Troxler Laboratories, P.O. Box 5997,
Raleigh, North Carolina.
384 BULK DENSITY

COUNTER
CABLE

PIVOT P

GC'6
UBE
SOURC
CUP

SOURCE
BRASS
SPACER
DETECTOR
r-- 12 "--I CROSS SECTION - PROBES
DETECTOR SOURCE
Fig. 30-4. Double-probe densitometer (Vomocil, 1954).

densitometer described by Vomocil (1954) is shown in Fig. 30-4 and

is made as follows: Construct the probes using aluminum tubing with
an outside diameter of 1 inch and with a wall thickness of 0.064 inch.
Fasten the two probes together near one end by two 12-inch lengths of
2- by 3-inch oak wood.
As a source of radiation, use 1 millicurie of CoGO in the form of a
length of I-mm. wire contained in a hollow aluminum cylinder about
1.5 mm. in length and 2.1 mm. in diameter. As a lid, use a second
cylinder about 0.5 mm. by 2.1 mm. attached to the first by an off-center
rivet. Mount the source in one of the probes near the bottom between
2.5-inch plugs of lead. Provide shielding around the radiation source
by placing the probe with the CoGO source through a hole bored through
a lead sphere 5 inches in diameter. Fit the sphere so that it will slide
along the probe when the probe is lowered into the soil. Place a stop
at the bottom of the probe to prevent the lead sphere from being re-
moved completely and to hold it in place over the CoGO when the den-
sitometer is being transported.
In the other probe, mount a Geiger Mueller detector tube (a Tracer-
lab TGC-6 tube in a P-17 side-window shield has been found satisfac-
tory) at the same height as the source of radiation in the first probe.
30-5 RADIATION METHOD 385

f------13"·----i

BRASS WASHER--...;:::!:::pI
-I I- I 'la"
SILVER SOLDERED

2 STAINLESS
STEEL TUBES
O.D.-I.~O·; 1.0.-1.33"
I

*I
! -T-
I x 7 SEAMLESS
BRASS PIPE
0:0.-1.32" 1.0:-1.13"

Vi'GALVANIZED PIPE

r :;
I I

BRAZED O= = *'l-:I=~*,-=-:-: -" '-=:- =-=-'-:-=-=-=- =-=-=-=-" -'-;-~L-:l_:lJ~=== 1"" Vi'. f.'
CHANNEL
f - - - - - j - - - - 2 1 " ----+---~

Fig. 30-5. Guide for making holes for double-probe densitometer (Vomocil, 1954).

Fit a brass spacer to the end of the side-window shield opposite the
tube socket (Fig. 30-4) to provide a better fit in the probe. Attach a
10-foot length of coaxial cable to the detector tube, lead the cable out
the top of the probe, and attach a fitting to the end for connection to
the scaler.
2. Portable scaler.

3. Device for boring holes into soil: Prepare a guide frame as indicated in
Fig. 30-5. The guide frame consists of two stainless steel tubes with a
length of 10 inches, an inside diameter of 1.334 inch, and a wall thick-
ness of 0.083 inch, welded into a piece of channel iron 4 by 11/2 by ~
by 21 inches in size. In the channel iron, bore two holes with a diame-
ter just adequate to receive the steel tubes and with a distance of 13
inches between centers of the holes. Weld the guide tubes into these
holes in such a way that the tubes are perpendicular to the surface of
the channel iron and exactly parallel with each other. Allow about 1
inch of tubing to protrude in the concave side of the channel. At the
ends opposite the channel iron, weld a length of lh-inch galvanized
iron pipe between the tubes to aid in maintaining alignment and to
serve as a carrying handle. Prepare a 7-inch piece of brass pipe
with an outside diameter of 1. 315 inches and an inside diameter of
1.063 inches. To the top of the pipe, sol~er a brass washer with an
outside diameter of about 1VB inches and an inside diameter of 1.063
inches. This tubing fits inside of the stainless steel guide pipe and out-
side of the Vcihmeyer soil sampling tube (available from Robert Mc-
Nairn Machine Works, 1431 Second St., Sacramento, Calif.) and aids
in keeping the sampling tube aligned with the guide.
386 BULK DENSITY

4. Soil sample containers to remove samples to laboratory for water de-

termination.
30-[Link] Scattering Apparatus. 8
1. Single-probe, depth-density gauge, or surface-density gauge.
2. Access tube and hammer.
3. Items 2 and 4 described in section 30-[Link].

30-5.2.2 PROCEDURE
30-[Link] Transmission Procedure. Firm the parallel-guide device to
the soil surface. With the Veihmeyer sampler, bore parallel holes to the
desired depth in the soil. Retain a sample from 2 inches above to 2 inches
below the sampling depth, and determine the water. content by weight.
Lower the source and detector tubes in the holes to the depth desired.
Determine the count-rate. From the calibration curve prepared in the man-
ner described in the following paragraph, convert count rate to wet bulk
density D bw , i.e., bulk density of liquid plus solid soil components. Convert
wet bulk density to the commonly used dry bulk density, i.e., bulk density
of soil on a dry weight basis, as follows:
[1]
where
Dbw = wet bulk density, and
Dw = bulk density of water in sample.
Now
Dw = dwVw/V., and
P,I,jIOO = dlOVw / DbV••
Therefore,
[2]
where
dw = density of water,
V wand V. = volume of water and bulk volume of soil, respectively, and
P w = percent water by weight.
Substituting [2] into [1] and simplifying gives
Db = Dbw/(l + percent H 0 2 by weight/loo).
To calibrate the instrument, determine the count-rate obtained with a
number of soils selected to cover the range to be encountered in practice.
Immediately after recording the count-rate, measure the wet bulk density
of the soil between the source and the detector by means of the core method
in section 30-2. Plot the logarithm of the count-rate against the wet bulk
• This apparatus may be purchased from Nuclear Chicago Corp., 333 East Howard
St., Des Plaines, Ill., or from Troxler Laboratories, P.O. Box 5997, Raleigh, North
Carolina.
30-5 RADIATION METHOD 3 a7
density to obtain a calibration curve. Alternatively, make the calibration
using the instrument in a large tank of soil that has been compacted arti-
ficially to a desired wet bulk density, and calculate the wet bulk density
from the mass and volume of wet soil. Each such measurement provides
one point on the calibration curve. If a calibration curve has been supplied
with the instrument, check the validity of the calibration by one of the
techniques described here.
30-[Link] Scattering Procedure. Put either the surface-density gauge
or the single-probe depth gauge in operation according to the manufac-
turer's instructions.
If the surface gauge is used, remove the dried soil surface crust and
smooth an area 16 by 16 inches. Place the gauge on the soil, twisting it
gently to ensure intimate contact. Measure the count-rate on the scaler
according to the manufacturer's instructions. Rotate the gauge 90 degrees,
and take a second reading. Divide the average count-rate by the standard
count-rate obtained according to the manufacturer's instructions. If no pro-
vision is made for obtaining a standard count-rate, prepare a standard such
as a waterproofed I-foot cube of wood, and measure the count-rate when
the gauge is placed on it in a reproducible position. Make counts in the
standard position as frequently as needed to provide the appropriate value
for the standard count-rate. By means of a calibration curve described be-
low, find the wet bulk density. To find the dry bulk density, measure the
water percentage on a volume basis in the surface 6 inches of soil directly
below the gauge. Then correct the wet bulk density to bulk density on a
dry-soil basis as prcviously described in section 30-[Link] for the trans-
mission procedure.
To prepare a calibration curve, make measurements as described in the
preceding paragraph using selected soil materials compacted uniformly to
a depth of 12 inches or more. Measure the wet bulk density of each ma-
terial at the site of measurement using the sand-funnel method of section
30-[Link]. Plot the ratio of the count-rate with the given soil material to
the standard count-rate against the wet bulk density measured by the sand-
funnel method. The points plotted in this way form the calibration curve.
When using the single-probe depth gauge, drive the access tube into the
soil to the desired depth. Place the gauge in its carrying shield on top of
the access tube. Lower the probe to the desired depth. Measure the count-
rate according to the manufacturer's instructions. When measurelTlents are
complete, raise the gamma source into its carrying shield. Divide the count-
rate by the standard count-rate obtained according to the manufacturer's
instructions. If no instructions are given for obtaining a standard count-
rate, use the count-rate obtained where the probe has been retracted into
the carrying shield, unless this results in numerical values that differ by a
factor of five or more from the count rate obtained with the probe in soil.
Under the latter circumstances, prepare a standard source in the form of a
388 BULK DENSITY

cylinder of waterproofed wood about 1 foot in diameter and height. Pro-

vide an access pipe that will permit repeated positioning of the probe in
precisely the same place in approximately the center of the cylinder. By
means of a calibration curve described in the following paragraph, find the
wet bulk density. To find the dry bulk. density, measure the water percent-
age on a volume basis in the soil surrounding the site of measurement, and
make the calculation as described in section 30-[Link].
To prepare a calibration curve, use soil materials that have been packed
uniformly to the desired bulk density to a depth of 2 feet or more. Obtain
preliminary counts with the probe at different depths to make certain that
measurements will be taken within a zone that is uniform vertically, or use
depths in natural soils where the bulk density remains essentially the same
for distances of 4 inches or more above and below the source and detector
in the probe. Then measure the count-rate with the probe located in the
uniform portion. Following the counting, take say five core samples at the
depth of the probe and at a distance 2 to 4 inches from the probe using
the method of section 30-2. Weigh these samples in their initial moist con-
dition and calculate the wet bulk density from the weight and volume of
the cores. Average the values from all cores.

30-5.2.3 COMMENTS
There is radiation hazard with this method. Gamma photons are high-
energy radiation. Some will pass through several centimeters of lead shield-
ing. Commercially available equipment, as well as designs described in the
literature, reduce the hazard to safe levels. But it is important to adhere
strictly to time limits, distances, and other conditions described by the
manufacturers. One should be equipped and knowledgeable in means of
checking the equipment for radiation levels according to the way it is han-
dled in actual sampling. If there is doubt, the equipment should be checked
for safety by a competent testing laboratory. It is also recommended that
film badges be worn regularly by users as an added safety check.
The user must be somewhat familiar with electronic equipment and fac-
tors that might affect its normal performance. Checking the calibration
should be carried out at each sampling, or several times in a day if the
equipment is in continuous use, according to instructions of the manufac-
turer.
The transmittancy method, using a double probe, is superior in profile
discrimination, and is therefore more likely to be useful in most soils work.
The method suffers, however, from lack of a supplier. There are numerous
details to consider in design and selection Of equipment for this method.
These are discussed in a rather extensive literature, of which the most
pertinent references are those by Goldberg et al. (1955), Van Bavel et al.
(1957), and Vomocil (1954).
30-5 RADIATION METHOD 389

Table 30-1. Comparison of surface nuclear gauge and sand-cone method for de-
termining soil bulk density (Mintzer, 1961).

Soil material Number of Mean Extreme

comparisons difference difference

Wet bulk density, Dbw

Brown sand, trace silt 9 2.03 -0. 25 to 4. 28
Brown silt and clay 4 0.89 -0.64 to 1. 87
Brown sand and gravel
some silt, trace clay 6 1. 01 -2.27 to 7.75
Brown till 4 -1.19 -3.97 to 1.60
Dry bulk density, Db
Brown sand, trace silt 9 2.05 -1. 21 to 4. 98
Brown silt and clay 4 7.21 5.51 to 9.27
Brown sand and gravel
some silt, trace clay 6 1.48 -1. 71 to 8. 32
Brown till 4 -0.38 -4.53 t6 3.21
* A positive value indicates nuclear method gave higher bulk density value.
Since radiation transmitted from a source to a detector is dependent on
probe spacing, or sample thickness, care must be exercised with the two-
probe sampler to assure that access holes are parallel and spaced exactly
as in the calibration.
Ideally when radiation methods are used, soil water content will be de-
termined with a neutron meter (see section 7-3.3), obviating the need to
remove a sample for water. The neutron gauge is calibrated to percent
water on a volume basis, however, and so the correction to bulk density
on a dry soil basis is made a little differently:
Let

where
Dbw = wet bulk density, and
Dw = bulk density of water in sample.
But,

where
dw = density of water at temperature of observation (assuming a density
of I gives a negligible error at ordinary temperatures), and
Sw = fraction of soil volume occupied by water determined from a neutron
count-rate and a calibration curve of count-rate versus volume-
fraction of water.
Mintzer (1961) reported comparisons of the surface-density probe and
the sand-cone method on four engineering projects. He reported his com-
390 BULK DENSITY

parisons on both the wet and dry bulk-density bases. He used a surface
neutron meter for water content where the surface-density probe was used.
His results are summarized in Table 30-1.

30-6 LITERATURE CITED

Abrol, I. P., and Palta, I. P. 1968. Bulk density determination of soil clods using
rubber solution as a coating material. Soil Sci. 106:465-468.
American Society for Testing and Materials. 1958. Procedures for Testing Soils.
Am. Soc. Testing Mater., Philadelphia.
Baver, L. D. 1956. Soil Physics. Ed. 3. Iohn Wiley and Sons, Inc., New York.
Brasher, B. R., Franzmeier, D. P., Valassis, V., and Davidson, S. E. 1966. Use of
Saran Resin to coat natural soil clods for bulk density and moisture retention
measurements. Soil Sci. 101: 108.
Goldberg, I., Trescony, L. I., Campbell, 1. S., Ir., and Whyte, G. I. 1955. Measure-
ment of moisture content and density of soil masses using radioactivity methods.
Nat!. Conf. on Clays and Clay Minerals, Proc. 3:516-548.
lamison, V. C., Weaver, H. H., and Reed, I. F. 1950. A hammer-driven soil core
sampler. Soil Sci. 69:487-496.
Lutz, I. F. 1947. Apparatus for collecting undisturbed soil samples. Soil Sci. 64:399-
401.
Mintzer, S. 1961. Comparison of nuclear and sand-cone methods of density and
moisture determinations for four New York State soils. In Symposium on nuclear
methods for measuring soil density and moisture. Am. Soc. Testing Mater., Spec.
Tech. Pub!. 293:45-54.
Neville, O. K., and Van Zelst, T. W. 1961. Design and application of the Nuclear-
Chicago diM-Gauge. In Symposium on nuclear methods for measuring soil density
and moisture. Am. Soc. Testing Mater., Spec. Tech. Pub!. 293: 3-8.
Russell, E. W., and Balcerek, W. 1944. The determination of the volume and air
space of soil clods. I. Agr. Sci. 34: 123-132.
Russell, M. B. 1949. Methods of measuring soil structure and aeration. Soil Sci.
68:25-35.
Tisdall, A. L. 1951. Comparison of methods of determining apparent density of soils.
Australian I. Agr. Res. 2:349-354.
U. S. Department of Agriculture. 1954. Diagnosis and improvement of saline and
alkali soils. USDA Handbook 60.
Van Bavel, C. H. M., Underwood, N., and Ragar, S. R. 1957. Transmission of gamma
radiation by soils and soil densitometry. Soil Sci. Soc. Am. Proc. 21:588-591.
Vomocil, I. A. 1954. In situ measurement of soil bulk density. Agr. Eng. 35:651-
654.