i'
.7
aC U' [ Of
pOC QOBa
f~ M0 o pip
OM -r~i
- - .- . , . -
DEPAR[MENI OF AGRONOMY AND SOILS
DEPARTMENTAL SERIES NO. 18
JULY 1974
AGRICULTURAL EXPERIMENT STATION/AUBURN UNIVERSITY
R. DENNIS ROUSE, Director AUBURN, ALABAMA

Leaching of Picloram and Nitrate in Two Alabama Soils!
/
A. E. Hiltbold, B. F. Hajek, G. A. Buchanan, and C. F. Scarshrook.
/
INTRODUCTION
Picloram
3/ 
(4-amino-3,5,6-trichloropicolinic acid) is an extremely
potent herbicide used for control of perennial broadleaved plants in non-
cropland areas. It is formulated as a water soluble potassium salt and
is registered for application in foliage spray mixtures with 2,4-D or
2,4,5-T. The potassium salt is available also as a granular product for
spot or broadcast treatment of soil over the roots of woody plants to be
controlled. Liquid formulations are usually applied at 0.5 to 1.5 lb acid
equivalent (ae)/A. The granular material is applied at 6 to 3 lb ae/A.
While grasses are relatively tolerant of picloram, most broadleaved
species are highly susceptible. Many legumes show injury symptoms at
picloram concentrations as low as 1 ppb in soil.
Picloram has the following structural formula:
NH
2
Cl , C1
N
1/ This study was supported in part Water Resources Research Institute,
Auburn University, Auburn, Alabama. OURR Project No. A-027-ALA and
Tennessee Valley Authority, Contract # TV-13436A.
2/ Professor, Associate Professors and Professor, respectively. Agronomy
and Soils Department, Auburn University Agricultural Experiment
Station, Auburn, Alabama.
3/ Tradenamed "Tordon" manufactured by Dow Chemical Co., Midland Michigan.
Mention of a trademark does not signify its approval by Auburn
University Agricultural Experiment Station.
2
The acid is soluble to the extent of 430 ppm in water at 25C, but the
potassium salt will dissolve to the extent of 40 g/10Q ml water. Picloram
has a pK of 3.6 and, therefore, the proportion of 
ionized to non-ionized
picloram decreases with decreasing soil pH. Minimum soil adsorption of
picloram has been observed in neutral, light textured soils low in organic
matter. Resistance to leaching is correlated with adsorption. The
mobility of picloram decreases with decreasing pH, increasing organic
matter, and increasing levels of hydrated iron and aluminum oxides (I).
The low vapor pressure of picloram (6.1 x 10
- 7 
mm at 35C) renders the
herbicide essentially non-volatile. It has a low acute oral toxicity
(LD
50 
= 8,200 mg/kg body weight of rats).
Picloram is degraded slowly by soil microorganisms. Its decomposition
in soil is at rates similar to the urea and triazine herbicides. The per-
cent decomposition of picloram decreases with increasing concentration
according to half-order kinetics. Estimated half-order constants, K
range from about 0.2 in colder, drier areas to 1.0 in hotter, wetter areas.
Starting with initial rates of 1 oz and 2 lb/A, the time required for the
herbicide to decompose to a negligible level (0.01 oz/A) 
varies from 4.5
months to 4.6 years for a half-order constant of 0.2 and from 0.9 to 11
months for a half-order constant of 1.0 (1).
The persistence of picloram, its mobility in surface and percolating
water, and its phytotoxicity at extremely low rates present 
an unusual
situation as to potential environmental impact of a herbicide 
(3). Research
reported here was undertaken with the objective of determining 
the leaching
and dispersion characteristics of picloram in two Alabama soils. 
Since
nitrate is a mobile anion of significance both in crop production 
and
ground water quality it was selected as a comparison for picloram 
in field
3
experiments. The subject of nitrate in soil, water, and plants has re-
cently been reviewed (4).
MATERIALS AND METHODS
Leaching of picloram was determined in field plots at Wiregrass and
Sand Mountain substations on Dothan loamy sand and Hartsells fine sandy
loam, respectively. The potassium salt of picloram was applied as broad-
cast spray at 10 pounds acid equivalent per acre on disked and smoothed
soil. Two nitrate treatments (200 and 800 lb N/A from NaNO
3
) and an
unamended check were included in complete randomized blocks with 4 repli-
cations. Experimental sites were flat. Furrows were turned in borders
and alleys between plots to prevent surface contamination. Plots were
not planted. Picloram and nitrate applications were made on September 14,
1971 and April 12, 1972 at the Wiregrass Substation and on October 12,
1971 at Sand Mountain Substation. Daily rainfall was recorded at each
location. Soil samples were collected at intervals over a 16-month period
at Wiregrass and over 9 months at Sand Mountain for determination of
picloram and nitrate. A hydraulic core sampler mounted on a truck pro-
vided a single core profile to 8 ft depth on each plot at each sampling
date. Bulk density of each depth interval was measured, then the samples
were screened and prepared for picloram and nitrate analysis. Nitrate
values reported are measured values from treated plots minus nitrate
measured in control plots.
Picloram was determined in a soil bioassav using Bragg soybeans.
Weighed portions of each soil sample were diluted with 400 g of picloram-
free soil to provide concentrations estimated to reduce the dry weight
of soybeans 50 percent relative to controls. Seven seeds probably pre-
ferred, were planted in each soil sample. At I week the plants were
4
thinned to the best four and fertilized with a N-P-K solution. At 3 weeks
the tops were cut at the level of the cotyledons and the plant material
dried and weighed. Soybean yields in unknowns were compared to those in
standards carried simultaneously to obtain picloram concentrations.
The standard response curve of soybeans grown at various concentra-
tions of picloram in soil is shown in Figure 1. The sensitivity of soybeans
to picloram (50 percent growth reduction with 1.1 ppb in soil) required
considerable dilution of field samples to reduce picloram levels to this
response range. For example, as little as 0.2 g of soil containing 2 ppm
of picloram was sufficient in a 400-g soil blank to establish 1 ppb in the
sample for bioassay. At the other extreme, a 40-g sample of field soil
diluted to 400 g was capable of detecting 0.01 ppm picloram. In bioassay-
ing unknowns, each sample was diluted 10-fold in a preliminary bioassay
to aid in selecting more appropriate dilutions. On the basis of injury
observed in the preliminary series, three dilutions were chosen to bracket
the suspected concentration.
Nitrate concentration was determined by extraction of 50-g soil
samples with water and distillation as NH
3 
after reduction with Devarda's
metal.
RESULTS AND DISCUSSION
Dothan loamy sand
Six samplings of field plots on Dothan loamy sand were made at the
Wiregrass Substation after application of picloram. Percentage recovery
of the applied picloram was generally low, ranging from about 20 percent
to 103 percent among the profiles sampled. This may result from factors
such as sampling variation within the plot, loss of picloram by degradation
and error in bioassay.
5
While the amounts of picloram recovered within profiles varied con-
siderably among replications, the vertical distribution was similar.
Averaging across the replications at each sampling date provided a view
of the progressive downward movement with time and cumulative rainfall
(Figure 2).
Distribution of nitrate with depth in Dothan loamy sand for the 200
and 800 lb N/A rates is shown in Figure 3. The 6 sampling dates extended
over a 16-month period. Comparison of cumulative rainfall and the depth
at which the bulk of nitrate occurred in the profile shows that the zone
of maximum nitrate moved downward about 1 inch for each 1 inch of rainfall
received during the fall and winter following application in September,
1971. During the summer and fall 1972, however, nitrate movement slowed
considerably despite accumulating rainfall. The penetration of nitrate
from the 200 lb N application proceeded at about the same rate as that
from the 800 lb N rate. Picloram movement was similar to that of nitrate,
except for the tendency of picloram to be retained in the surface foot
of soil while nitrate was thoroughly removed by leaching. This retention
of picloram in the surface soil is probably associated with organic matter.
The movement of picloram and nitrate applied in the spring (April 12,
1972) to Dothan loamy sand was measured in duplicate soil profiles taken
2 ft apart on July 20, 1972. Picloram distribution on this date is shown
in Figure 4 and nitrate in Figure 5 and Figure 6. The data show con-
siderable variation in amounts of picloram and nitrate recovered in soil
cores taken only 2 ft apart, however, the vertical zones of maximum con-
centration were similar. The penetration depth of nitrate averaged
slightly greater than the cumulative inches of rainfall. Picloram leach-
ing, on the other hand, was somewhat less than was nitrate. As observed
6
in samplings of the fall application, picloram remained in the surface
soil after nitrate had been completely leached.
Hartsells fine sandy loam
Picloram concentration with depth in Hartsells fine sandy loam at
Sand Mountain Substation is shown in Figure 7. The movement of picloram
in lHartsells soil appears more rapid than in Dothan soil with comparable
amounts of rainfall.
The average distribution of nitrate at each sampling date is plotted
in Figure 8 for both 800 lb N and 200 lb N treatments. The penetration of
nitrate into Hartsells soil was similar to Dothan, that is, at the rate of
about 1 inch depth for each 1 inch of rainfall received.
Nitrate leaching model
In both Dothan and Hartsells soils nitrate distribution is essential]
symmetrical around a peak concentration. This indicated that the equatiot
C/C
O 
= [erfc 2 Dt i] [1]
in which x = depth
v = ion velocity
t 
= 
time
D = dispersion coefficient
erfc = error function
C/C
o 
= relative concentrations
often used to predict anion distribution in laboratory soil columns (2),
may be used to model nitrate depth distribution patterns in the field.
The equation was modified to include factors for the presence of an initia
residual nitrate content before application. In addition, a factor was
included for incorporation of added nitrate into the surface soil layer
at time zero. When these factors are added the equation takes the follow-
ing form.
Co
C 
={
C
Co
CL
erfc
x
1
s
R
e
b
D
t
I R
- CL b Co b
2 erfe L 2 erfc
S 2(Dt)2 2(Dt) j
= concentration of NO3-N in the soil solution
= amount of NO3-N added
= concentration of residual NO3-N
= error function complement
= depth
= thickness of soil layer into which C
o 
is incorporated
= inches (or cm) of rain that is effective in leaching
= pore volume fraction
= adsorption factor equal to 1 if no adsorption occurs
= dispersion coefficient (.01 cm
2
/hr)
Stime since leaching began after C
o
was applied
[2]
Figures 9 and 10 show the calculated distribution for the 800 lb N/A
application rate on Dothan and Hartsells soils, respectively. The March
and April, 1972 sampling dates were used to compare measured values to
the calculated line. For these calculations the observed inch per inch
relationship for effective rainfall leaching was used. It was assumed
that the added nitrate was incorporated in the surface inch of soil. The
initial residual nitrate and bulk density was taken as the average in check
plots to 48 inches. The comparison of calculated values to concentrations
observed at depths from four replications shows that the calculated line
is within the variation from the field at most depths.
8
CONCLUSIONS
(1.) Nitrate moved downward in Dothan loamy sand and in Hartsells fine
sandy loam at the rate of about 1 inch per 1 inch of rainfall
received.
(2.) Some picloram was retained in surface soil whereas all nitrates
were eventually leached below the surface soil.
(3.) The picloram that was leached below the surface soil moved down-
ward at about the same rate as nitrate.
(4.) The rates of leaching 200- and 800-1b N/A applications are essen-
tially equal.
(5.) A mathematical model of nitrate leaching was developed that fit
the observed leaching in these soils within the range of field
sampling precision.
9
UITE'raTISRE CITED
(1.) Goring, C. A. I. and J. W. Hamaker. 1971. The degradation and
movement of picloram in soil and water. Down to Earth 27(l):12'-15,
(2.) Kirkham, D. and W. L. Powers. 1972. Solution for the 
dispersion
of a slug of fluid. Advanced Soil Physics. pp. 402-405.
(3.) Upchurch, R.
p. 443-512.
chemicals in
P. 1972. Herb icidesand plant growth regulators.
In C. A. I. Goring and J. W. Hamaker (ed.) Organic
the soil environment. Marcel Dekker, Inc., New York.
Viets, F. G., Jr. and R. H. Hageman. 1971. Factors
accumulation of nitrate In soil, water, and plants.
Handbook No. 413. U.S. Dept Agr . Washington, D). C.
affecting the
Agriculture
C
0
?OtI.0
a, H9
oV
Z7 
3
0.2
.2
O0 -I 0 3 
I I I a 3
0 .2 4 .6 8 1.0 1.2 14 1.6 1.82.0
ppb PICLORAM
Fig. 1. Dry Weight Yield of Soybean Plants Grown for 3 Weeks
in Soil Containing Picloran.
PICLORAM, ppm
0 I
PICLORAM, ppm
02
P I CLORAM, ppm
2 0 I
6
12
18
24
NOV. 12 11971
6
12
18
24
36
JAN. 20, 1972
1752" RAIN
2
6
12
18
24
36
48
0
6
12
18
24
U)
w
z36
72' 721
0 2 0
36
JULY 20,1972
45.46"IfRAIN
72,
2
JAN.17, -1973
76.89" RAIN
Fig. 2. Distribution of Picloram 
with Depth, Dothan
on September 14, 1971.
mean of 4 replications
l.s. after Application
6
12
18
24
36
2
0
z
f-
0
36
48
MAR.I , 1972
2.2 0" RAIN
APR. 1I, 1972
17.14 " RAIN
NITRATE, ppm
50 100 150 200 0
NITRATE , 
ppm
50
12
18- - - - 200 lb. N/A
800 Ib.N/A
36
NOV. 12, 1971
3.20" 'RAIN
50
48
0 50 0 50
JAN. 20, 1972
17.52" RAIN
0 50
MAR. I, 1972
22.20" RAIN
APR. II, 1972
27.14" RAIN
JULY 20,1972
45.46" RAIN
JAN. 17,1973
76.89" RAIN
Fig. 3. Distribution of Nitrate 
with Depth, Dothan l.s. after 
Application
of 200 and 800 lb N/A 
on September 14, 1971. 
mean of 4 replications
0
6
100
Cl)
w
0
Z
I-
L-
W
18
24
36
48 u
0
6
12
18
24
Ul)
w
z
I-
a-
W
0
36
48
72
PICLORAM, ppm
0 1I 2 3 45
6
12
18
24
36
48
Rep I
PICLORAM, ppm
0 I 2 3 4 5
6
12
18
24
36
6
12
18
36
Rep ]i
48
PICLORAM , ppm
0 I 23 45
Rep t
PICLORAM, ppm
0 1 2 34 5
Rep "I"
Fig. 4. Distribution of Picloram with Depth, Dothan l.s. 2 profiles
per plot taken 2 ft. apart on July 20, 1972. 18.32" rainfall
since application on April 12, 1972.
6
12
U,)
0w
a
z
I
H-
0:L
wU
0I
18
24
36
48
Cl)
LU
0
z
I
1.
0
NITRATE, ppm
0
61
12
18
24
36
48
0
18
24
36
48
25 50 75
6
36
Rep I
48
25 50 75
6
36
Rep I
48
0
0
25 50 75
Rep "I
25 50 75
Rep "I2
Fig. 5. Distribution of Nitrate with Depth, Dothan l.s. 2 profiles per
plot taken 2 ft. apart on July 20, 1972. 200 lb. N/A applied
April 12, 1972. 18.32" rainfall.
C,)
LUI
Z
I
a.
H
0a
w3
(l)
W
I
.)
w
0
NITRATE., ppm
NITRATE,
100
6
12
18
24
36
Rep -
48
100
6
12
36
Rep IT
48
NITRATE
100
Rep IE
Fig. 6. Distribution of Nitrate with Depth,
taken 2 ft. apart on July 20, 1972.
1972. 18.32" rainfall.
Dothan l.s. 2 profiles per plot
800 lb. N/A applied April 12,
0
ppm
200
6
300 0
, ppm
200 200
U)
V
I
Z
z
W
I
12
18
24
36
48
0
Rep m.
200 300 0
6
12
18
100 200
CW
U
Z
I
C)
z
0
U
36
48
PICLORAM, ppm
0 0.25 0.5 0.75 1.0
JAN. 27, 1972
18.19"
APR.6 , 1972
28.84"
Fig. 7. Distribution of Picloram with Depth,
Hartsells fsl. after application on
October 12, 1971.
mean of 4 replications
6
12
18
24
36
C f)
W
I
0
z
Z
I
Q.W
48
72
NITRATE , ppm
100 150 200 250
6
12
7- JAN. 27, 1972
/ 18. 19"
APR. 6 ,1972
28.84"
JULY 25, 1972
40.80"
800 lb. N
18
24
36
48
72
0 50 100
JAN. 27, 1972
18.19"
APR. 6,1972
28.84"
JULY 25, 1972
40.80"
200 lb. N
Fig. 8. Distribution of Nitrate with Depth, Hartsells fsl. after Application
of 800 and 200 lb. N/A on October 12, 1971.
mean of 4 replications
500
6
12
18
24
36
48
72
96
96
NITRATE , ppm
N0
3
-N (ppm)
120 160
10
ul N0
3
-7N "
sN, equation 2-
DO lb. N (NaNO
3
) per Ac-.
1972
Fig. 9. Calculated Nitrate Distribution Using a Dispersion Model (line)
Conpared to Measured Concentrations in Dothan loany sand on March 1,
1972 After Application of 800 lb N/A on September 14, 1971.
40 80
200 240
I I I I I I
00
4
E xpe r imentc
Calculated NO
3
Dothan 
Is, 
8 ' 
c
3
0
"" "
252
00
15
20
C
0
251
301
35
40
45
50
--
I
N0
3
- N (ppm)
80 100
5
I0
N0
3
-N 0
equation 2
lb. N(NaNO
3
)
72
""
22
Experimental
Calculated N
3
-N,
Hartsells fsl , 800
"" 
April 6, 19
20
" " 
"
cu"
r"
o2"
Fig. 10. Calculated Nitrate Distribution Using a Dispersion Model
(line) Compared to Measured Concentrations in Hartsells
fine sandy loan on April 6, 1972 After Application of
800 lb. N/A on October 12, 1971.
40
150
.
. 0
151
201
U
C.
a)
251
30
35
40
45
50
r A
C.._
-u- I I I ? I - - I I I I I