Tải bản đầy đủ - 0trang
CHAPTER 7. ISOTOPES METHODS AND USES IN SOIL PHYSICS RESEARCH
DON KlRKHAM A h 9 RAYMOND J. W N Z E
also of nonradioactive isotopes. The isotopes which will be discussed are:
H2, H3, C14, N16, 0 l 8 , P32,S35, C136,K42, Ca45,Cow, Rbs6, SrgO,Il3I, C S I ~ ~ ,
Recent reviews on the use of radioisotopes, including some uses in
soil physics, have been prepared by Fried et al. (1958) and by Raney
and Thorne (1957). A recent general paper on radioisotopes and agronomy is that of Norman (1959). Most of the work reviewed on uses of
isotopes in soil physics has been concerned with moisture determination
by neutron scattering and with density determination by gamma-ray
transmission or scattering. Very little other work on nuclear methods in
soil physics has, at the time of this writing ( 196l), been reported. What
work is known to the authors, including recent work in neutron and
gamma ray scattering, will be reported here.
This paper will include, primarily at the end, suggestions for future
11. Soil Water
BY THE KEIJTRONMETHODOF WATERRETAINED
The most important information generally needed about soil water is
the amount of it contained in the soil rooting depth. This amount, in
humid regions where irrigation is not commonly practiced, will vary
with the following seven factors: (1) rainfall, ( 2 ) evaporation, ( 3 )
transpiration, ( 4 ) runoff, ( 5 ) deep seepage, ( 6 ) condensation, and ( 7 )
rising ground water level. In the above list, for arid regions, irrigation
must be considered with or in place of rainfall. With these many factors
and their interaction with various soils, and without prior moisture
tests of the soils, it is impossible to predict the soil moisture content; it
must be measured. The most useful method we know for measuring
soil moisture is the neutron-scattering method. We at Iowa State
University have also tried nuclear magnetic resonance, a method which,
with its present costly equipment ( $25,000 from the Varian Associates ) ,
does not seem practical. This technique was also tried by others ( Andreev
and Martens, 1960). The neutron method appears much more practical
and is becoming fairly well known to soil physicists.
1. How the Neutron Meter Works
Early work on the neutron meter for use in agriculture was done by
Gardner and Kirkham (1952). Even earlier Belcher et al. (1950) had
worked on the problem and had developed equipment for soil tests in
airport runway design. The Gardner-Kirkham equipment was not
portable. Stone et aE. (1955) and van Bavel et al. (1956) developed
portable meters. Commercial meters are now available. One commercial
ISOTOPES IN SOIL PHYSICS RESEARCH
model including surface and depth gauges is described by Kuranz
(1960). Other papers dealing with the calibration and the use of
commercial models are those of Burn ( 1961), Neville and van Zelst
( 1961), and Gnaedinger (1961).
The device works as follows: A source of fast neutrons, (about
1,000,000electron volts) usually from a mixture of radium and beryllium,
is lowered into a pipe (called an access pipe) driven into the soil. With
the neutron source at the desired depth the fast neutrons penetrate into
the soil, bouncing about from soil atom to soil atom, some of them
returning to the access pipe where a detector of slowed neutrons has also
been located with the neutron source. Essentially, the only slowing of the
neutrons is that resulting from collisions with hydrogen atoms. The other
soil atoms, all much heavier than hydrogen, do not slow the neutrons
appreciably. So the detector measures essentially the density of hydrogen
atoms in the soil; and since the hydrogen atoms in soil are almost all
associated with soil water, the number of slowed neutrons detected per
second is a measure of the soil water per unit volume of the soil.
2. W h y Neutrons Are Slowed Down by Hydrogen Atoms
To show the basic principle of the neutron moisture meter, which is
that hydrogen atoms slow up neutrons rapidly whereas other soil atoms
do not, a simple experiment may be performed. As a preliminary to such
an experiment, Table I is presented which gives the masses and relative
abundances of different soil atoms. The table shows that atoms comprising
Masses and Percentages of Elements in Soil@
0 (not in water form)
Per cent in soil
(Depends on water
content of soil)
Calculated from data of Thompson (1957, p. 7 ) .
Hydrogen of soil water is not considered as a part of the soil.
DON KIRKHAhL A S 9 FIAYMOND J. KUNZE
the dry soil substance are all heavy (massy) compared to hydrogen.
With the table in mind one can now perform the experiment.
Take two glass marbles, an orange one and a blue one, say, each
one weighing 5 g. These are standard-sized marbles used by boys. Take
also a third ball about the same size made of steel and weighing about
45 g., 9 times as much as the glass balls.
Now suppose that the blue glass ball represents a neutron, one such
as would be shot from the radium-beryllium neutron source. Suppose
also that the orange glass ball represents a hydrogen atom in soil. Let
the steel ball represent a heavy atom in the soil. Now, in a plastic
V-groove trough strike the ‘leavy atom” with the “neutron.” The
“neutron” will bounce back from the “heavy atom” (steel ball) with
very little loss of velocity, that is, energy. Now strike the “hydrogen
atom” (orange glass ball) with the “neutron” (blue glass ball). Observe
that the neutron stops; the hydrogen atom takes all the neutron’s velocity.
In soil, of course, the blows are not head-on as in the V-groove trough,
but the principle is the same. The hydrogen atoms slow up the neutrons.
The more hydrogen, and hence water, the more slowing down.
3. Adoantages and Disadvantages of the Neutron Meter
The neutron meter has some outstanding advantages. Some of these
1. The method is rapid, about one-half minute being required for a
determination at a single depth in the soil. The detector is easily moved
up and down in the access pipe to get the soil moisture at any desired
depth. Movement of the equipment between access pipes may be
facilitated by a gocart ( Ameniya and Namken, 1960).
2. Fewer samples are required than for most other methods. A
neutron determination is as good as six (Shaw et al., 1959) or seven
(Stone et al., 1960) gravity determinations.
3. The neutron meter equilibrates instantaneously with the soiI
moisture. Some other type instruments require 2 to 3 weeks to equilibrate.
4. One access pipe may be used the year round to determine
moisture content in the soil. A small test plot need not be dug up for
getting moisture samples. The method is nondestructive.
5. The meter gives the soil moisture on a soil bulk volume basis.
Moisture on a volume basis is what is generally desired. Soil moisture
detenninations by oven drying are on a weight basis. To get them on a
volume basis the bulk density of the soil sample must be determined,
Bulk density determinations may introduce large errors (Stone et al.,
6 . The meter gives the average moisture of about a 6-inch diameter
ISOTOPES IN SOIL PHYSICS RESEARCH
sphere. Ordinarily this is an advantage, as variations over very small
distances are not desired. In some cases a smaller distance is desired.
Mortier et al. (1959,1960) has shown how a paraffin sphere surrounding
the radium-beryllium source may be used to slow up the neutrons and
reduce the sphere of influence,
7 . The meter is insensitive to changes in salt concentration in the
soil. Different rates of fertilizer application will not upset the moisture
8. The type of material of which the access pipe is made is not
critical. Neutrons pass through steel tubes and aluminum tubes easily.
9. One calibration ordinarily serves for a number of soils (Stolzy
and Cahoon, 1957). If soils which contain large amounts of organic
matter are avoided (Gardner and Kirkham, 1952) and if certain clay
soils which bind up a great deal of water in the clay crystal lattice
(Mortier and De Boodt, 1956) are excluded, a high degree of accuracy
in determining soil moisture can be obtained. Stewart and Taylor (1957)
discuss the advantages of measuring soil moisture with a neutron meter
as opposed to resistance and tensiometer-type measurements. They point
out that the neutron meter is reliable in all types of inorganic soils of
medium texture regardless of salt content and soil type.
At the present time there are two marked disadvantages of the
neutron meter. The initial cost is high: the scalar costs $1500, and the
depth probe costs $1000. The other disadvantage is that skilled personnel
are needed to service the unit. Presently available commercial units
seem to get quickly out of repair.
For an article giving 91 references on the neutron meter and the
gamma-ray density meter (the latter to be discussed below) see a paper
by Kuranz (1960).
4. Uses of the Neutron Meter in Soil Moisture Studies
a. Field capacity measurements. Figure 1 shows curves of moisture
content versus time and field capacities, F.C., at a number of depths in
two soils. When the curves flatten, gravitational water is gone, and the
field capacities, as shown by arrows in the figure, become evident. The
neutron meter is especially useful in these field capacity determinations
since one can follow the moisture content with time at one spot;
variations in moisture content due to sampling at different sites do not
b. Determination of water stored termporarily above the jield capacity.
The neutron meter responds essentially instantaneously to the instantaneous moisture content in the soil. This characteristic has been used by
Nielsen et al. (1959) to measure the water stored temporarily above the
DON KIRKHAM A N D RAYMOND J. WNZE
field capacity. In Fig. 2 profiles of temporarily stored water are seen
for a Monona silt loam, a loess. The thickness of the profiles represents
moisture in excess of the field capacity at the time printed above the
profiles. Zero time is the time when the profile is near water saturation
to about the $foot depth and when water application to the profile has
Horiz. arrows indicote
scole o f ordinate
Vert. arrows give
. 4EC = 25.6
FIG. 1. Moisture vs. time curves for ( A ) Marshall silt loam and ( B ) Ida silt
loam obtained with a neutron meter. Field capacities are given at the right-hand
ends of the curves. (From Burrows and Kirkham, 1958.)
been discontinued. In Fig. 3 the storage profiles are for Webster clay
loam, a Wiesenboden soil formed on glacial till. The neutron meter
shows for the Webster, unlike the Monona, large amounts of water
stored temporarily (above the field capacity) in the surface 12 inches.
From these last two figures one sees that a considerable amount of soil
moisture may be stored above the field capacity even after 5 or more
ISOTOPFS IN SOIL PHYSICS RESEARCH
days’ time. In a later paper Nielsen et al. (1961) showed, using neutron
moisture meter data, that soil moisture profiles in the field can be
Two other recent papers on more conventional use of the neutron
meter may be cited here. Stolzy et al. (1959) calibrated the meter
against the soil suction and various moisture contents so that the meter
75 10 IS
HCB%XNTAL S & s r
I O S R C E N T WATER
FIG.2. Moisture profiles, obtained by the neutron meter, of water stored temporarily above the field moisture capacity for a Monona silt loam, a loess. (From
Nielsen et al. 1959.)
FIG.3. Same as Fig. 2, except that soil is a Webster silty clay loam, a Wiesenboden glacial till soil, with an obvious tight layer at 6 to 12 inches depth. (From
Nielsen et al., 1959.)
could be used for determining soil suction. Weeks and Stolzy (1958)
used the meter to measure soil moisture in columns of soil 3 feet in
diameter. Satisfactory agreement between the calculated changes of
total water content and the measured amounts of added water were
obtained for two soils, a sandy loam and a silt loam.
c. Water movement through soil profiles. Measurement of moving
water and stored water are closely related. Water which is moving may
be considered as stored instantaneously.
DON KIRKHAM A?;D
RAYMOND J. KU’NZE
In the paper by Nielsen et al. (1961) the movement of water in the
soil profile was determined by the neutron meter and the shape of the
moving moisture profile related to the theoretical profiles determined
on the basis of diffusion theory. In Monona-Ida loess soils the agreement
of the measured profiles with the theory was fairly good.
Kixon and Lawless (1960a, b ) measured soil moisture with a neutron
meter down to 20-foot depth under brush cover. They found that the
moisture content of various soil layers under their test plots varied with
time t according to the relation W = At - b.
5. Uses of the Neutron Meter in Studying Soil-Water-Plant
a. Water auailability for crops. Shaw (1959) covered some plots with
plastic film; other comparison plots he left uncovered. He pierced
neutron access pipes through the plastic-covered plots and installed
access pipes also in the uncovered plots. In both plastic-covered and
uncovered plots, corn (maize) was planted, holes for the corn seed
being punched in the plastic covers. Irrigation water could be added
under the plastic-covered plots and also onto the uncovered plots.
There was no seepage loss. Since no evaporation in the plastic-covered
plots could occur and since there was no seepage loss, a loss of moisture
in the plastic-covered plots was due to transpiration alone. In the uncovered plots the moisture loss was due to both evaporation and
transpiration. By subtracting values of the moisture content found in
the uncovered plots from the moisture content found in the covered
plots, evaporated moisture could be and was obtained. Evaporation and transpiration could be followed accurately and conveniently
from day to day. At high sunlight intensities, such as 5% g.-~al./cm.~/day,
the evaporation loss from the soil was a larger percentage of evapotranspiration than at lower sunlight intensities. Shaw found that on the
average for the season there was about 50 per cent of the water evaporated from the soil; this 50 per cent the plants did not get-it was wasted.
(Actual average vaIues were 46 per cent evaporation, 54 per cent
There was an interesting side effect in the experiment. Large amounts
of rain water were found to be intercepted by the corn plants. This
water, 2.21 inches between July 15 and September 16, was evaporated
from the leaves and probably served little useful purpose. The writers
of this review suggest that tritiated or deuteriated water might be
applied to corn leaves to see how much of “intercepted water,” as noted
by Shaw, is used by the plant.
In a later paper Fritschen and Shaw (1961) pointed out that it is
ISOTOPES IN SOIL PHYSICS RESEARCH
not accurate to substract transpiration on plastic-covered plots from
evapotranspiration on uncovered plots to obtain evaporation. A principal
reason is that the plastic cover changes the microclimate of the soil.
They gave some correction factors, which, when applied to their moisture
data of 1961, indicated that only 11 per cent of moisture loss was due to
evaporation, the other 89 per cent, to transpiration. Even without the
correction factors the 1961 work indicated 27 per cent loss by evaporation
(not about 50 as in 1959) and 73 per cent by transpiration. But the 1959
and 1961 data are not strictly comparable. The 1959 data were for
evaporation for the whole growing season; the 1961 data for the portion
of the growing season after the corn crop had reached its maximum
height. It appears that further work on measurement of evaporation and
transpiration on field plots is needed.
Denmead (1961)used the technique of the neutron meter and plastic
covers successfully in smaller “lysimeter” plots to study soil-plant-climate
relations, work which we will now discuss.
There has long been a controversy on whether or not water is equally
available to plants over the whole range of available water; that is, from
the field capacity to the wilting point. Veihmeyer and Henrickson (1955)
have supported the equal availability concept; Slatyer (1956) and
Richards and Wadleigh (1952) have disagreed. In his work Denmead
used the neutron meter to follow the moisture in 136 “lysimeters” in
which corn was grown. The lysimeters were 20-gallon garbage cans.
They were buried in the field to the surface of the soil and were
surrounded by growing corn. Evaporation was prevented by covering the
lysimeters with plastic covers through which the corn grew. It would
have been impracticable to attempt to weight this many lysimeters.
Denmead (1961) found that each of the two viewpoints on plant
available water were tenable. On overcast days in soils of high capillary
conductivity the corn plants would transpire at about the same rate at
moisture contents ranging all the way from the field capacity to the
wilting point; whereas, in soils of very low capillary conductivity and
on hot days, the soils would not supply enough water to the plant at
any moisture content between the field capacity and the “wilting point.”
Wilting would occur even at the field capacity.
The neutron meter already has been adapted in a variety of agronomic
moisture availability problems. Stolzy and Cahoon (1957) used the
neutron meter in measuring soil moisture under citrus orchards. Marston
(1958) used the meter in mountain soils for studying evapotranspiration.
Reproducibility of moisture values was a problem because of rocks.
Pits had to be dug for placing the access pipes. Merriam (1960) used the
neutron method in sampling “wild land soils.” He found that use of
DON KIRKHAM AND RAYMOND J. WNZE
herbicides resulted in a saving of 1 inch of water in 10-foot depth for
west-facing slopes and resulted in saving of about 3 inches of water in
10-foot depth on east-facing slopes. This was a sandy soil which contained
about 15 to 16 inches of water in 10 feet of profile.
b. Plant spacing and water utilization. Neutron meters have been used
by Yao and Shaw (unpublished data as of July, 1961) to test efficiency
of water use by corn at various rates of planting. Since Shaw’s work of
1959 (cited above) indicated that about 50 per cent of soil moisture
might be lost by evaporation, Yao and Shaw wondered if increased shade
at the soil surface would decrease this loss. The increased shade could
be obtained by planting the corn rows closer together or by planting
the rows in such directions as to obtain more shadow on the ground.
Table I1 shows the results for 21-inch spacing and 42-inch spacing. The
Water Use by Corn (Maize) for Different Rates of Planting and Different
Directions of ROWS=
Distance apart of
Number of corn plants
Direction of corn row
Yield, bushels corn per acre
Inches of water used
Bushels of corn produced
per inch of water
16.1 15.3 17.4 18.8
0 Yao and Shaw, Iowa State University, Ames, Iowa, unpublished data; moisture
data collected with neutron meter.
0 One corn plant per hill.
c Two corn plants per hill.
table indicates that 1inch of water produces 10 bushels of corn per acre
at 21-inch spacing of rows; and about 7 to 8 bushels per acres at 42-inch
spacing of rows. Whether there were 14,000 or 28,000 plants per acre at
a given spacing seemed to make little difference in the yield per inch
of water. Also, the direction of the rows seemed to make little difference
in the yield per inch of water.
Bahrani and Taylor ( 1961) investigated the influence of soil moisture
potential and evaporative demand on the actual evapotranspiration in
an alfalfa field. Soil water was measured to a depth of 9 feet using a
neutron meter. The actual evapotranspiration and its ratio to “potential
evapotranspiration” ( calculated by Penman’s formula ) showed a curvilinear relation with the average moisture potential.
ISOTOPES IN SOIL PHYSICS RESEARCH
In Section 11, A above, the papers reviewed were largely concerned
with the static condition of water in soil. In this Section (B) papers on
moving water and ions are reviewed. The movements are considered
first for unsaturated flow and then for saturated flow.
1. Unsaturated Flow
a. Miscible displacement. When one solution moves into another
solution, both solutions being soluble in each other, and when movement
of the solutions occurs in a porous medium such as soil, the original
solution is moved by a process called miscible displacement. This term
is used because when the added solution is washed through the soil it
mixes with the solution which is already there while displacing it. To
study how applied soil water displaces water already in the soil, Nielsen
and Biggar (1961) originally displaced the soil solution with water
containing a small concentration of chloride ion. They wondered how
the tracing ion might influence the miscible displacement process; therefore they used, in some later experiments ( Biggar and Nielsen, 1!36l),
tritium and chloride ion. The results showed (Fig. 4) that the chloride
ions came ahead of the tritium ions. The movement of the chloride ions
was faster in the soil column than was the tritiated water, even though
the latter had a higher diffusion coefficient. In the unsaturated state the
tritiated water could also diffuse in a vapor phase, where movement is
faster than in the liquid phase. Thus under some unsaturated conditions,
one might expect that the tritiated water would be detected first.
b. Digusion flow of water. Moisture movement in soil using deuterium
as a tracer was studied by Kunze and Kirkham ( 1961). They directed
their attention to the determination of the self-diffusion coefficient of
soil water because of the need for this coefficient in order to separate
mass flow and diffusion flow-types of flow they had observed in an
earlier experiment. In this earlier experiment they used a soil core,
initially at constant moisture content and having its middle portion
tagged with deuterium. Letting the core dry from one end, they found
that a large part of the tagged moisture moved in a direction the reverse
of that of the drying gradient. Since mass flow of water would, by
definition of mass flow, have to move in the direction of the drying
gradient, they reasoned that some of the water motion was due to
molecular self-diffusion. They therefore turned their attention to the
measurement of the self-diffusion coefficient of soil water. It is emphasized here that the self-diffusion coefficient of water as determined by
D O S KIRKHAhf A33l M Y M O N D J. KUNZE
Kunze and Kirkham does not refer to the type of diffusion of water
spoken of in connection with capillary soil-moisture flow. This capillary
flow diffusion coefficient is defined as the product of the capillary
conductivity of moist soil and the rate of change of the soil water tension
with the soil moisture content. The ( molecular) self-diffusion coefficient
is a measure of the rate at which molecules of a material diffuse into
other molecules of the same material.
COLUMBIA SILT LOAM
8 = 0.482
VOLUME OF EFFLUENT (ml.)
FIG. 4. Miscible displacement curves of soil solution being displaced by water
containing chloride ions and tritium, the displacing water moving into the soil
solution at a velocity V = 0.20 cm./hr. The moisture content 0 for the replicated
runs shown is about 0.48 cm."cm.3. C,, is the initial concentration of the ion or of
the tritium in the displacing solution; C is the concentration of the chloride ion or
tritium in the outflow solution (effluent) as measured in volumetric outflow incremants. (From Biggar and Nielsen, 1961.)
Figure 5 shows a curve used by Kunze and Kirkham in evaluating
the self-diffusion coefficient. In Fig. 5 on the y axis, a function F is
given which is essentially the concentration of deuterium atoms in a
soil core, the left half of which contained originally deuterium-tagged
soil water ( F = 1 ), and the right half, untagged water ( F = 0). The
test soil core in these experiments was kept completely enclosed to
prevent evaporation. The heavy line is a theoretical curve based on the
hypothesis that self-diffusion in soil water is a linear process. The circle