Spatial variability of soil redistribution processes in a small agricultural watershed
by John Cornelius Pings
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Earth Sciences
Montana State University
© Copyright by John Cornelius Pings (1990)
Abstract:
Physical soil redistribution processes were studied in a small (61 ha) watershed in a region of dryland
winter wheat agriculture in north-central Montana. Two approaches were used, a model approach using
the Universal Soil Loss Equation (USLE) and Wind Erosion Equation (WEE) soil erosion models, and
a field sampling approach using 137Cs. The 137Cs was used as a tracer of erosion and deposition from
upland sites (hilltops, midslopes and footslopes) to depositional zones (channels and a pond reservoir
bottom). Cesium-137, a fallout product of atmospheric nuclear testing, is strongly adsorbed to clay and
has been proven to trace sediment movement. A volumetric approach, developed by De Jong and
associates of the University of Saskatchewan, was used to estimate erosion rates of eroding sites and
deposition rates of the depositional sites. Landscape units, labelled topographic positions and
depositional zones, were defined from a 1:2500 scale plane table topographic contour map, and
analyzed for areal concentration of 137Cs to attain erosion rate estimates.
A 125 m random sample grid was used to generate USLE and WEE erosion rate estimates. USLE
estimates were calculated using the point method of Griffin et al. (1988). WEE estimates were
calculated using equations of Skidmore (1988) which were developed to fit the nomographs
conventionally used in WEE applications. The model approach yielded an erosion estimate of
approximately 9.0 Mg ha-1 yr-1; combining a USLE average estimate of 4.5 Mg ha-1 yr-1 with a WEE
average of 4.5 Mg ha-1 yr-1. A site by site comparison of combined model and 137Cs estimates for the
137Cs sample sites yielded a regression output of .07, possibly indicating poor model performance.
However, problems in assessing the spatial and temporal variability of soil redistribution indicate a
need to further refine the cesium method to reduce variances.
Using the 137Cs method, hilltops, midslopes and footslopes were found to be eroding at 22.9, 28.1, and
0.5 Mg ha-1 yr-1, respectively, for a total net erosion rate of 10.4 Mg ha-1 yr-1. Ponds and channels
were found to have deposition rates of 243.9 and 43.0 Mg-1 ha-1 yr, respectively, for a total net
deposition rate of 5.0 Mg-1 ha-1 yr. The USLE estimated 90 % of the measured value while the WEE
predicted only 44 % of the measured wind erosion. The poor model performance and low precision of
the cesium method suggests that the use of the models needs to be considered carefully, especially with
regard to watershed scale soil erosion assessments. SPATIAL VARIABILITY OF SOIL REDISTRIBUTION PROCESSES
IN A SMALL AGRICULTURAL WATERSHED
by
John Cornelius Pings
A th e s is submitted in p a r tia l f u lf illm e n t
o f the requirements fo r the degree
of
Master o f Science
in
Earth Sciences
MONTANA STATE UNIVERSITY
Bozeman, Montana
June, 1990
^7 ^
il
APPROVAL
o f a thesis submitted by
John Cornelius Pings
This th e s is has been read by each member o f the th e s is committee and
has been found to be s a tis fa c to ry regarding content, English usage,
form at, c ita tio n s , b ib lio g ra p h ic s ty le , and consistency and is ready fo r
submission to the College o f Graduate Studies.
^
Date
Chairperson, Graduate Committee
Approved fo r the Major Department
C
-/S '
Ia jo r Department
Date
Approved fo r the College o f Graduate Studies
Z jT 7
Date
^
/??&
Graduate Jean
iii
STATEMENT OF PERMISSION TO USE
In presenting th is thesis in p a rtia l fu lfillm e n t o f the
fo r a m aster's degree a t Montana State U n iv e rs ity ,
I agree
Library shall make i t a v a ila b le to borrowers under rules o f
B rie f
quotations
from
th is
thesis
are
allowable
requirements
th a t the
the Library.
without
special
permission, provided th a t accurate acknowledgement o f source is
Permission
fo r
extensive
quotation
from or
made.
reproduction
of
th is
thesis may be granted by my major professor, or in his absence, by the
Dean o f L ib ra rie s when, in the opinion of e ith e r, the proposed use
m aterial is fo r scholarly purposes.
Any copying or use o f the
in th is thesis fo r fin a n c ia l gain shall not be allowed without
permission.
Signature.
Date
of the
m aterial
my w ritte n
iv
I dedicate th is thesis to my w ife , Laura, to my daughter, Lauren
and to my great uncle Robert W. 0 ' lo ughlin .
M ichelle
V
ACKNOWLEDGEMENTS
I would lik e to thank the members o f my thesis committee fo r th e ir
assistance and guidance.
My committee included Dr.
Katherine Hansen-
B ristow, Dr. Joseph Ashley, Dr. W illiam Locke I I I and Dr. Gerald
I
would
e s p e c ia lly
lik e
to
thank my thesis
advisor
and
chairperson, Dr. John Wilson, fo r his guidance and fo r
me the p rin c ip le s of sound s c ie n tific research.
Drs.
Douglas
Sherman and Bernard
Bauer fo r
I
Nielsen.
committee
impressing upon
would lik e to thank
th e ir
help,
advice
and
frie n d s h ip .
I
would lik e
Montana
State
to
acknowldge the Department o f
U n iversity
research assistantships.
fo r
t h e ir
support
Earth
including
Sciences at
teaching
I also want to thank Dr. Wilson fo r
supporting
my research with funds from his National Science Foundation
would lik e
to thank my f ie ld
a ssistan ts,
grant.
Loretta Thomas and
Sanderson; Artem Vartanian and Lee Murray fo r th e ir d ra ftin g ;
fe llo w graduate students, in p a rtic u la r Jon Aspie,
and
I
Forrest
and my
Chet Clarke and Mike
Trombetta fo r t h e ir help and comradery.
I would lik e to express my appeciation and g ra titu d e to Mr. and Mrs.
Norman W. Jackson fo r allowing
would lik e to thank Mr.
the Choteau County
s u ita b le
study area
Nadwornick, SCS State
me to use th e ir farm fo r my research.
I
Raymond McPhail, SCS D is tr ic t Conservationist of
Conservation D is t r ic t ,
and fo r
fo r his help in fin ding
access to government data
and Mr.
a
Ron
Agronomist, fo r estim ating wheat residue values.
vi
TABLE OF CONTENTS
Page
LIST OF TABLES ..............................................................................................
LIST OF FIGURES ...........................................................................................
ABSTRACT .....................................................
1.
3.
x
x ii
INTRODUCTION .........................................................................................
I
Scope and Purpose ......................................................
Previous Watershed-scale Soil Erosion Studies
Model Studies and Applications ...................
Previous Cesium-137 Studies .........................
Description of Study Area ........................................................
Thesis Organization ..........................
12
16
METHODS AND DATA SOURCES ......................
r—I CO CO LO
2.
v iii
17
Topographic Map Generation ......................................................
Model Estimates .............................................................................
Universal Soil Loss Equation .................................
Wind Erosion Equation ........................................................
Cesium-137 Erosion Estimates ..................................................
Cesium-137 Sample S ite Selection .................................
Cesium-137 Sample C ollection .........................................
Cesium-137 Sample Increments .........................................
Cesium-137 Sample Preparation .......................................
Cesium-137 Laboratory Analysis ......................................
Bulk Density Sample C ollection ......................................
Method of Areal Cesium-137 Analysis ...........................
Erosion and Deposition Rate and Mass
Balance Estimation Method ............................................
17
20
20
25
33
33
35
37
38
38
39
39
RESULTS ....................................................................................................
42
USLE Erosion Estimates ..............................................................
WEE Erosion Estimates ................................................................
USLE and WEE Erosion Rate Estimates Combined .................
Cesium-137 Results ..............................
42
44
48
50
41
v ii
TABLE OF CONTENTS--Continued
Page
4.
DISCUSSION ...................................................................
USLE and WEE Soil Loss Estimates ............................................
V a lid a tio n o f USLE and WEE Soil Loss Estimates ...............
Soil Erosion/Deposition Rates In fe rre d from
Cesium-137 Gains/Losses ..........................................................
Conclusions .......................................................................................
60
60
64
66
70
REFERENCES C IT E D .........................................................................................
71
APPENDICES........................................................................................................
78
APPENDIX A. EROSION MODEL RESULTS ........................................
APPENDIX B. CESIUM-137 LABORATORY DATA AND
AREAL CONCENTRATIONS ..........................................
APPENDIX C. SIEVING RESULTS .....................................
79
86
91
v iii
x
LIST OF TABLES
Table
1.
Page
Erosive Wind Energy Occurring by Month
a t Great F a lls , MT .................................................................................
15
Confidence o f Table S tation B Elevation
Determined by D iffe re n t Sightings ..................................................
21
Computation o f Average Annual C Factor
fo r T ille d Soils .....................................................................................
24
4.
Soil E r o d ib ility by Soil Mapping Unit ..........................................
29
5.
Estimates of K Factor, Vegetation Weight and
Erosive Wind Energy fo r Cropstage Periods
and T illa g e Operations, Jackson Farm ............................................
30
6.
Sample WEE Computation fo r Sample Point ID ................................
33
7.
USLE Estimates .........................................................................................
43
8.
WEE Soil Loss Estimates .......................................................................
47
9.
Cesium-137 Erosion and Deposition Rate Estimates ...................
55
10.
Average Cesium-137 Erosion and Deposition Mass
Rates by Topographic Position ..........................................................
59
Comparison of USLE Factor Estimates Used by
Author and USDA-SCS .........................................................
61
Comparison of WEE Factor Estimates Used by Author
and WEQ Factor Estimates Used by USDA-SCS .................................
62
S ite by S ite Comparison of Model and 137Cs
Erosion/Deposition Rate Estimates ............................................
67
Summary o f Watershed Average 137Cs Erosion
and Deposition Mass Estimates ..........................................................
69
15.
USLE Factor and Soil Loss Point Estimates .................................
80
16.
WEE Factor and Soil Loss Point Estimates ....................... -...........
83
17.
Cesium-137 Laboratory Data and Areal Concentrations .............
87
2.
3.
11.
12.
13.
14.
ix
LIST OF TABLES--Continued
Table
,18.
Page
Sieving Results ............... .......................................................................
92
:x
LIST OF FIGURES
Figure
L
Page
Cesium-137 in a drainage basin
Tfrom Campbell e t a l . , 1982) ........................................
8
2.
Study area and lo c a tio n map . . . ................... .......... .............................
13
3.
Soil series map showing s o il mapping units as defined
by the s o il survey team, SCS Choteau County
Conservation D is t r ic t .............................................................................
15
Map showing lo c a tio n o f baseline AB and plane ta b le
s ta tio n s A, B, C, D and E w ith in study watershed ...................
18
Map showing lo c a tio n o f randomly selected 125 m g rid
,points and la b e ls used fo r USLE and WEE c a lc u la tio n s ...........
22
6.
Diagram showing topographic po sition s ...........................................
35
7.
Study area map showing lo catio ns o f 137Cs samples
w ith in study watershed and lo catio n s o f control samples
on the Jackson Farm ...................................................................
36
Areas o f the topographic positions used fo r
137Cs e x tra p o la tio n ...................................................................................
37
D is trib u tio n o f USLE s o il loss ra te s fo r C arter
watershed (n = 81) ........................
43
S p a tia l v a r i a b ili t y o f USLE s o il loss rates fo r
C arte r watershed .......................................................................................
45
D is trib u tio n o f s o il loss rates fo r C arter watershed
using WEE po int method and 125 m g rid (n = 80) ........................
47
S p a tia l v a r i a b ili t y o f WEE s o il loss rates fo r
C arte r watershed ......................................................................................
49
D is trib u tio n o f to ta l s o il loss ra te s fo r C arter
watershed using USLE/WEE point methods
and 125 m g rid (n = 81) ........................................................................
50
S p atial v a r i a b ili t y o f s o il loss rates fo r C arter
watershed using USLE and WEE point methods ..............................
51
4.
5.
8.
9.
10.
11.
12.
13.
14.
xi
LIST OF 'FISORES--CCTtanued
Figure
15.
16.
17.
Page
S ite areal 137Cs a c t iv it y fo r h illt o p s , midslopes,
foots!opes, t i l l e d channels, incised channel, pond
and control s ite s .....................................................................................
52
Range and average 137Cs areal concentrations
by topographic p o s itio n ........................................................................
54
S c a tte rp lo t diagram o f p redicted and measured values
generated fo r s it e by s ite comparison o f 137Cs
sample s ite s ................................................................................................
67
x ii
ABSTRACT
Physical s o il r e d is trib u tio n processes were studied in a small (61
ha) watershed in a region o f dryland w in te r wheat a g ric u ltu re in northc e n tra l Montana.
Two approaches were used, a model approach using the
Universal Soil Loss Equation (USLE) and Wind Erosion Equation (WEE) s o il
erosion models, and a f i e l d sampling approach using 137Cs. The 137ICs was used
as a tra c e r o f erosion and deposition from upland s ite s ( h illt o p s ,
midslopes and foots!opes) to depositional zones (channels and a pond
re s e rv o ir bottom ). Cesium-137, a f a llo u t product o f atmospheric nuclear
te s tin g , is stro n g ly adsorbed to c lay and has been proven to tra c e
sediment movement.
A volum etric approach, developed by De Jong and
associates o f the U n iv e rs ity o f Saskatchewan,
was used to estim ate
erosion ra te s o f eroding s ite s and deposition ra te s o f the depositional
s ite s . Landscape u n its , la b e lle d topographic positions and depositional
zones, were defined from a 1:2500 scale plane ta b le topographic contour
map, and analyzed fo r areal concentration o f 137Cs to a tta in erosion ra te
estim ates.
A 125 m random sample g rid was used to generate USLE and WEE erosion
ra te estim ates. USLE estim ates were c a lc u la te d using the po in t method o f
G r if f in e t a l . (1 9 8 8 ). WEE estim ates were c a lcu lated using equations o f
Skidmore (1988) which were developed to f i t the nomographs conventionally
used in WEE a p p lic a tio n s . The model approach y ie ld ed an erosion estim ate
o f approxim ately 9 .0 Mg ha'1 y r'1; combining a USLE average estim ate o f 4 .5
Mg ha"1 y r'1 w ith a WEE average o f 4 .5 Mg ha'1 yr"1.
A s it e by s ite
comparison o f combined model and 137Cs estim ates fo r the 137Cs sample s ite s
y ie ld e d a regression output o f .0 7 , possibly in d ic a tin g poor model
performance.
However, problems in assessing the s p a tia l and temporal
v a r i a b ili t y o f s o il r e d is trib u tio n in d ic a te a need to fu r th e r re fin e the
cesium method to reduce variances.
Using the 137Cs method, h illt o p s , midslopes and foots!opes were found
to be eroding a t 2 2 .9 , 2 8 .1 , and 0 .5 Mg ha'1 y r'1, re s p e c tiv e ly , fo r a to ta l
net erosion ra te o f 10.4 Mg ha"1 y r'1.
Ponds and channels were found to
have deposition ra te s o f 243.9 and 43.0 Mg'1 ha"1 y r , re s p e c tiv e ly , fo r a
to ta l net deposition ra te o f 5 .0 Mg"1 ha"1 y r . The USLE estim ated 90 % o f
the measured value w h ile the WEE predicted only 44 % o f the measured wind
erosion.
The poor model performance and low p recision o f the cesium
method suggests th a t the use o f the models needs to be considered
c a r e fu lly , e s p e c ia lly w ith regard to watershed scale s o il erosion
assessments.
I
CHAPTER I
INTRODUCTION
Scope and Purpose
The increasing scale of human impact in a g ric u ltu ra l
areas and a
desire to reduce soil erosion problems have provided the impetus fo r
erosion assessments in recent years.
(USDA) Soil Conservation
the Universal Soil Loss
Equation (WEE) to
and recent work
inputs to
State
soil
The U. S. Department of A gricultu re
Service (SCS) re g u la rly u t iliz e s models such as
Equation (USLE) and a version o f the Wind Erosion
guide the implementation of conservation procedures,
has focused on improving the methods o f estim ating model
produce erosion
estimates th a t are s p a tia lly v a ria b le .
In the
of Montana there is a need fo r large scale so il erosion studies to
assist in the development of soil erosion/crop p ro d u c tiv ity
S im ila rly , there is a need to c o lle c t so il erosion data
these modeling e ffo r ts so th a t the contribution of the
assessments.
independent of
modeling e ffo r ts ,
themselves, can be evaluated.
This
study addresses both needs by quantifying
deposition in a small a g ric u ltu ra l catchment.
used.
s o il
erosion
and
Two approaches have been
The f i r s t approach uses the USLE and WEE to estim ate soil losses
from water and wind, re sp ec tiv e ly , in the watershed.
The second approach
uses the s p a tia l v a r ia b ilit y of Cesium-137 ( 137Cs) detected in so il samples
2
to q u a n tify erosion from both water and wind
and deposition by w ater.
The USLE and WEE have emerged as the most w idely used s o il erosion
models
in
North
s ta te /p ro v in c ia l
America.
It
assessments
has
been
used
fo r
in the United States
many national
and
and Canada and fo r
watershed- and p lo t-s c a le studies in both countries (T rim b le, 1974, 1977,
1983; van V li e t and W all, 1979, 1981; Coleman, 1982; S n e ll, 1984, 1985;
Wilson, 1989).
SCS
to
USLE and WEE s o il loss estim ates are also used by the USDA
determine
q u a lific a tio n
and
maintenance
requirements
fo r
Conservation Reserve Programs (C .R .P .) in Montana.
The techniques used in th is study to estim ate erosion and deposition
ra te s
from
137Cs
areal
associates (1981a,
(1 9 8 3 ).
concentrations
were
developed
1981b) and la t e r re fin e d by De Jong and
The 137Cs isotope acts as a tra c e r o f physical s o il
processes by i t s
by
adhesion to fin e s o il
g ra in s .
It
Brown
and
associates
re d is trib u tio n
is a by-product o f
atmospheric nuclear te s tin g and is d e liv e re d through p r e c ip ita tio n and
w in d -c a rried
sediments.
depo sition al
processes
Pennington e t a l . ,
It s
has
use
been
as
an
in d ic a to r
widespread
1976; McHenry and R itc h ie ,
1980; Wise, 1980; Brown e t a l . ,
1984; Arnalds e t a l . ,
1981b; De Jong
of
(R itc h ie
erosional
et
a l.,
and
1974;
1977; McCallan e t a l . ,
et a l.,
1983; Arnalds,
1989; Dibb, 1989).
Using these approaches, the o b je ctiv es o f th is study were tw ofold.
The f i r s t o b je c tiv e was to estim ate s o il losses with the USLE and WEE fo r
a small a g ric u ltu ra l watershed.
The second o b je c tiv e was to q u an tify s o il
erosion and deposition using 137Cs and to use these re s u lts to evaluate
USLE and WEE performance in the same watershed.
The fo llo w in g kinds o f
data were generated and analyzed to answer these research questions: I )
3
analysis o f c lim a tic , s o il, topographic and vegetative cover factors to
produce USLE and WEE so il loss estimates at m u ltip le s ite s ; 2) analysis
o f 137Cs samples to define to ta l and incremental 137Cs areal concentrations
fo r several s ite s , and determine r e la tiv e erosion and deposition; and 3)
extra p o la tio n
of USLE and WEE estimates
and 137Cs areal
concentration
averages by landscape units to obtain erosion and deposition rates fo r
those units and the e n tire watershed.
Previous Watershed Scale Soil Erosion Studies
Model Studies and Applications
The USLE was derived from 10,000 plot-years o f data at locations
throughout the United States.
q u a n tific a tio n
of
fa c to r
This model estimates erosion through the
values
fo r
r a in f a ll- e r o s iv it y
(R ),
so il
c r e d ib ilit y (K ), topographic factors defined by slope length (L) and slope
steepness (S ), cover management (C) and supporting practices
(P ).
The
USLE is freq u en tly used by the SCS as a tool to determine conservation
p ractices
fo r
strip-cropping
the
control
and te rra c in g
of
flu v ia l
erosion,
(Wischmeier,
including
contour
1976; Wischmeier and Smith,
1978).
Some recent studies have tr ie d to improve the USLE by developing new
methods o f estim ating fa c to r values, p a r tic u la r ly the topographic fa c to rs ,
L and S.
W illiams and Berndt (1977), fo r example, proposed a method of
generating slope frequency data by using a th ird -o rd e r natural
spline
function developed by G reenville (1967) fo r points defined by horizontal
distances and elevations of contours th a t cross grid lin e s on topographic
maps.
Slope
in
the
d ire c tio n
of
the
g rid
lin e s
was determined
by
4
d if fe r e n t ia t in g
the
s p lin e
function
at
each g rid
in te rs e c tio n
p o in t.
Wilson (1986a, 1986b) developed a d if fe r e n t approach using topographic
map input data and GreenviT ie 's .(1967) s p lin e function to estim ate slope
le n g th , shape and steepness fo r slope p r o file s th a t cross the elevatio n
contours perpendicularly..
to
d iv id e
ir r e g u la r
computer
slope
generated
method o f
estim ate LS values.
o f estim ating
S t a t is t ic a l analysis o f slope segments was used
slope
Foster
p r o file s
in to
and Wischmeier
segments
(1974)
and
the
was used to
G r if f in and his associates (1988) developed a method
LS fa c to r values fo r a series o f random p o in ts.
T h eir
method used the distance downslope from the top o f the slope p r o f ile as
w e ll as cum ulative and slope segment gradients to estim ate LS values fo r
s p e c ific points in a landscape.
A more comprehensive re v is io n o f several fa c to rs w ill re s u lt in the
p u b lic a tio n o f a computer program and manual fo r the Revised Universal
Soil Loss Equation (RUSLE) in 1990 (Renard, 1989, personal communication).
The o rig in a l USLE and RUSLE w ill most l i k e l y be replaced by a completely
new physically-based modeling technology in the mid-1990s (F o s te r, 1989,
personal communication).
The Wind Erosion Equation (WEE) was developed a t about the same time
as the USLE to assess the s u s c e p tib ility o f f ie ld s o ils to wind erosion
and to
help
p ra c tic es
w ith
(Chepil
the
et
s e le c tio n
a l.,
1962;
and design
o f wind
erosion
Woodruff and Siddoway,
1965).
control
This
equation estim ates erosion as a function o f magnitude and d ire c tio n o f
wind as w ell as s o il c r e d ib i lit y ( I ) , s o il ridge roughness (K ), vegetation
o rie n ta tio n and cover ( V ) , and f ie ld fe tch length ( L ) .
was used in e a rly conservation a p p lic a tio n s .
The WEE equation
Skidmore and Woodruff (1968)
5
la t e r compiled p e rtin e n t c lim a tic data fo r many s ta tio n s from e x is tin g
sources to a s s is t in the a p p lic a tio n o f the WEE throughout the country.
The model and c lim a tic data were published fo r farmers and conservation
program workers.
In one o f the most innovative a p p lic a tio n s o f the WEE
to d a te , Bondy e t a ! . (1980) estimated wind erosion by cropstage period
as a function o f wind energy d is tr ib u tio n . T h e ir method used tem porally
v a ria b le v e g e ta tiv e residue e q u iva len ts, s o il t i l l a g e conditions and s o il
c r e d ib i lit y to estim ate wind erosion fo r 10 d if fe r e n t cropstage periods
in a w in te r wheat/summer fa llo w system in Kansas and a spring wheat/spring
w h e a t/fa llo w
system
in
North
"Dakota.
Skidmore
(1988)
substitu ted
equations to estim ate WEE fa c to r values fo r the tab les and nomographs o f
Woodruff and Siddoway (1 9 6 5 ).
These equations, f i r s t proposed by W illiam s
e t a l . (1 9 8 4 ), e lim in a te the need to in te rp o la te fa c to r inputs and reduce
the
time
and e f f o r t
needed to
apply the WEE.
The SCS A g ric u ltu ra l
Research Service developed a version o f the WEE, the WEQ, fo r use by SCS
personnel in f i e l d a p p lic a tio n s .
Previous Cesium-137 Studies
Cesium-137
is
a w idely
dispersed
ra d io a c tiv e
isotope
th a t
is
a
by-product o f atmospheric nuclear te s tin g which the United S ta te s, Soviet
Union and United Kingdom began on a frequent basis in 1954 (Campbell e t
a l . , 1982).
These te s ts are s t i l l c a rrie d out on a much sm aller scale by
France
China
and
(Anonymous,
1989).
The
period
of
most
intense
atmospheric dispersion occurred between 1962 and 1965 immediately before
the United S ta te s ,
Soviet Union and 41 other nations
signed a tr e a ty
suspending atmospheric te s tin g o f nuclear weapons in 1966 (Campbell e t
6
a l.,
1982).
Cesium-137 has a regional d is tr ib u tio n
p re c ip ita tio n sources and q u a n titie s .
the
stratosphere
d e liv e re d to s o ils
and troposphere
th a t is lin k e d to
The isotope is transported through
by global
through two methods.
c irc u la tio n
patterns
and
One method involves d e liv e ry
through p r e c ip ita tio n , and the second method involves dry deposition o f
137Cs w ith atmospheric p a r tic u la te m atter from the atmosphere.
I t s lo cal
concentration is dependent upon the a v a ila b le amount o f the isotope in the
atmosphere
at
the
tim e
of
p r e c ip ita tio n
events,
its
a lt it u d e ,
and
p re v a ilin g regional and lo cal m eteorological conditions (McCallan e t a l . ,
1980).
Several assumptions govern the use o f th is isotope as a tra c e r o f
s o il r e d is tr ib u tio n process a n a ly s is .
I t is assumed th a t: I ) the isotope
has been d e liv e re d to the watershed uniform ly; 2) the 137Cs becomes adsorbed
to the c la y - and s ilt - s iz e d s o il p a r tic le s when i t reaches e a rth ; 3) these
sediments have not undergone any s o rtin g ; and 4) the re d is tr ib u tio n o f 137Cs
by w in te r winds, p la n t and animal l i f e has been minor (< 5 percent) and/or
a t le a s t uniform throughout an in d iv id u a l watershed (Brown e t a l . , 1981a,
1981b; De Jong e t a l . , 1982, 1983; Arnalds, 1984).
Cesium-137 can be used to q u an tify s o il erosion and deposition rates
because
its
(dispersion
h a lf-life
(3 0 .2
since 1954)
years)
and
a d d itio n a l
v a ria b le s
topographic p o s itio n ,
period
of
allow estim ates o f average annual
deposition ra te s to be made (Brown e t a l . ,
known,
known
1981a).
such as in te n s ity
existence
erosion and
Once these rates are
o f areal
concentration,
and maximum depth o f 137Cs a c t iv it y
in
the s o il
p r o f ile f a c i l i t a t e the d e lin e a tio n o f erosional and depo sition al zones in
watersheds because the 137Cs moves only w ith sediment th a t is transported
7
(De Jong e t a l . ,
1982, 1983).
A conceptual model o f 137Cs in p u t, a c t iv it y and tra n s p o rt in a drainage
basin is reproduced in Figure I .
Most studies use th is type o f model to
q u a n tify erosion and deposition ra te s .
The bar diagrams o f ty p ic a l 137Cs
areal concentrations fo r various landscape un its in Figure I il lu s t r a t e
the v a ria b le nature o f 137Cs a c t iv it y a t d if fe r e n t lo ca tio n s in a watershed.
Several studies have examined 137Cs in te s t p lo ts and watersheds to
determine regional concentrations ( e . g . , Rogowski and Tamura, 1970; Lance
et a l.,
1986; Kachanoski,
average 137Cs areal
1987; Dibb,
1989).
concentrations between 3 .6
Arnalds
(1984) measured
and 20.2 pi cocuries per
square cm (pCi cm'2) a t 12 s ite s throughout Montana.
These concentrations
were stro n g ly re la te d to p re c ip ita tio n (R2 = 0 .9 2 ), in d ic a tin g th a t lo cal
p r e c ip ita tio n to ta ls w ill provide a good f i r s t estim ate o f expected 137Cs
le v e ls
(A rnalds,
1984;
Arnalds
et
a l.,
1989).
Given
the
previously
mentioned regional sources and controls and the assumptions noted above,
several researchers have constructed 137Cs mass balances
to in fe r s o il
erosion and deposition ra te s (De Jong e t a l . , 1983; Arnalds, 1984; Pennock
and De Jong, 1987).
Cesium-137 has been applied to erosion and sedimentation studies in
a v a rie ty o f contexts.
Several studies have shown th a t concentrations
have increased in areas o f sediment accumulation such as v a lle y flo o r s ,
la k e s , re s e rv o irs and s a lt marshes ( e . g . , McHenry e t a l . ,
et
a l.,
1974,
1975;
Pennington
et
a l.,
1976;
Delaune
1973; R itc h ie
et
a l.,
1978;
McCallan e t a l . , 1980; Brown e t a l . , 1981a, 1981b; Campbell e t a l . , 1982;
De Jong
et
a l.,
1983;
Arnalds,
1984;
Pennock
Cesium-137 has been used to determine erosion o f
and
De Jong,
natural
and
1987).
c le a rc u t
8
O
t a ’ Ce
AND
P R E C IP IT A T IO N
V
O
IN P U T
V
Figure I . Cesium-137 in a drainage basin (from Campbell e t a l . , 1982).
fo re s ts and f o r e s t / f ie ld systems at a v a rie ty o f s p a tia l scales (Brown e t
a l.,
1981a, 1981b; Campbell e t a l . ,
studies
have
a g ric u ltu ra l
et a l . ,
137Cs
to
evaluate
s o ils (Brown e t a l . ,
s o il
erosion
and
1981b; Campbell e t a l . ,
1988).
Other
deposition
on
1982; De Jong
1983; Arnalds, 1984; Pennock and De Jong, 1987).
Recent
e n tir e
used
1982; Lowrance e t a l . ,
studies
watersheds
have q u a n tifie d
by using weighted
d iffe r e n t landscape u n its
Arnalds,
and deposition
average areal
(Brown e t a l . ,
1984; Pennock and De Jong,
developed to d is tin g u is h
erosion
fo r
concentrations
fo r
1981b; De Jong e t a l . ,
1987).
landscape u n its .
rates
1983;
Several methods have been
For example,
(1981b) divided t h e ir watershed in to erosional
Brown e t
and depositional
a l.
zones,
9
w ith h illto p s and midslppes representing erosional zones and foots!opes
and an a llu v ia !
fan representing depositiona!
zones.
De Jong e t a ! .
(1 9 8 3 ) , on the other hand, distinguished three landscape u n its
m iddle, and lower slopes -
determined in the f ie ld by pacing.
(1984) added two fu rth e r un its to th is c la s s ific a tio n ,
between
h illto p s
-
and midslopes,
and
toes!opes
upper,
Arnalds
shoulder slopes
beneath
lower
slopes
(fo o ts lo p e s ) but defined his tran sects by s e le c tin g s o il samples fo r 137Cs
analysis 50 m a p a rt.
to De Jong e t a l .
Pennock and De Jong (1987) used a s im ila r approach
(1983)
in an attempt to decipher the
influence
p r o f ile and plan curvature on s o il r e d is tr ib u tio n processes.
of
A d ig it a l
te r r a in model was created from an e x is tin g d ig itiz e d topographic survey
and used to d is tin g u is h convergent and divergent backs!opes, convergent
and divergent shoulders, convergent and divergent foo tslo p es, and le v e l
areas (Pennock and De Jong, 1987).
Care must be exercised when using th is
term inology because the landscape units, are defined and labeled using
d iffe r e n t methods and d e fin itio n s
general
in d if fe r e n t s tu d ies .
approach o f e x tra p o la tin g average 137Cs areal
However, the
concentrations to
landscape u n its is now w idely used to estim ate average annual erosion and
deposition ra te s .
Most recent studies have undertaken 137Cs d etection w ith labo ratory
analysis
using
a
lith iu m -d r ifte d
germanium
semi-conductor
gamma ray
d e te c to r coupled to a nuclear data m ulti-channel analyzer (Cutshall and
Larsen, 1980; Larsen and C u ts h a ll, 1981; Brown e t a l . ,
Jong
et
a l.,
1983;
Arnalds,
1984).
Cesium-137
1981a, 1981b; De
concentrations
are
c a lc u la te d by m u ltip ly in g the 137Cs a c t iv it y in core samples (pCi g"1) by
the s o il mineral bulk density (g cm"3) and depth o f the s o il sample (cm).
10
Results are expressed as r a d io a c tiv ity
per u n it
surface area of soil
surface or the "areal concentration" ( e . g . , Brown e t a
l 1981b; De Jong
e t a l . , 1983; Arnalds, 1984; Pennock and De Jong, 1987).
Campbell e t a l . (1982) used a d iffe r e n t approach to determine 137Cs
areal concentrations. Their approach r e lie s on the fa c t th a t 137Cs adheres
to fin e s o il fra c tio n s , so th a t the areal concentrations are calculated
using only the s i l t and clay fra c tio n s .
S i l t and clay frac tio n s were
estimated using the hydrometer method and
th e ir bulk d e n s itie s were used
instead o f th a t o f the to ta l
sample.
This study used an experimental
watershed in New South Wales, A u s tra lia , which was found to have 137Cs
d is trib u te d
evenly
in
upper
so il
layers
due
to
plowing.
Areal
concentrations were q u an tifie d fo r a v a rie ty of topographic positions in
upland and lowland areas although no erosion or deposition rates were
estimated (Campbell e t a l . , 1982).
In co ntrast,
many North American studies have tr ie d
to estimate
erosion and deposition zones and rates from 137Cs an alysis.
Brown and
associates (1981b) computed erosion estimates ranging from 3 to 27 t
ha"1
yr"1 fo r erosional zones based upon detected 137Cs ranging from 3 .5 to 15.2
pCi
cm'2 in
two W illam ette
V a lle y ,
OR watersheds.
foots!opes tested were found to be eroding
Six
out of eight
(Brown e t a l . ,
1981b).
In
another study, De Jong e t a l . (1983) found upper slopes to have lo s t 200
to 600 t ha"1 o f s o il while lower slopes were found to have gained 250 to
800 t ha"1.
However, middle slopes were found to be both depositional and
erosional over 20 to 25 years in eight small Saskatchewan, Canada basins
with g la c ia l s o ils (De Jong et a l . , 1982; 1983).
41
Arnalds (1984) examined erosion in a small watershed near Power, MT
in which he detected 137Cs a rea l concentrations ranging from 2 .4 to 24.6
pCi cm"2.
H illto p s and foots!opes both were found to be eroding a t 16.5 Mg
ha'1 y r " \ w h ile shoulders and midslopes were eroding a t 20.9 and 45.1 Mg
ha"1 yr"1, re s p e c tiv e ly .
Xoeslopes were found to have deposition occurring
a t a ra te o f 9 .9 Mg ha"1 yr"1.
and foots!opes
The estim ates fo r the shoulders, mid slopes
compared favorab ly
w ith
USDA SCS s o il
loss
estimates
produced w ith the USLE and WEE models.
Lance
et
a l.
(1986)
c o lle c te d
data
on
137Cs
a c t iv it ie s
in
the
southwestern United States as w ell as in adjacent c u ltiv a te d and grassed
watersheds in Oklahoma.
A major conclusion o f th is study based upon the
re s u lts o f a s o il mass balance was th a t 137Cs a c t iv it y might be a more
s e n s itiv e in d ic a to r o f s o il p ro d u c tiv ity losses than measurements o f to ta l
mass o f s o il removed from a f ie ld caused by highly lo c a liz e d erosion and
d e p o sitio n .
M cIntyre e t a l . (1987) found a c le a rc u t fo re s t to be eroding
a t only 0 .2 t ha"1 yr"1 and a ttrib u te d the erosion to s o il compaction by
liv e s to c k grazing o f n a tiv e grasses.
In a re la te d study Lowrance e t a l .
(1988) estim ated erosion and deposition ra te s o f 63 and 256 Mg ha"1 y r'1,
re s p e c tiv e ly , fo r a rip a ria n f o r e s t / f ie ld system watershed on the southern
Georgia coastal p la in .
The unusually high deposition was a ttrib u te d to
sediment th a t was transported from upstream lo catio ns and deposited during
flo od
events.
Three
d if fe r e n t
c a lc u la tio n
methods
produced
n early
id e n tic a l re s u lts .
Pennock and De Jong (1987) examined more landform elements than the
other studies and found four o f seven to be erosive, from most to le a s t
erosive: convergent shoulders, divergent backs!opes, convergent backs!opes
12
and divergent shoulders.
The depositional units were, in order from le a s t
depositional to most: divergent foots!opes, level
foots!opes.
areas and convergent
Soil loss predictions fo r these same areas using the USLE
were two to nine times lower than those
indicated by the 137Cs method
(Pennock and De Jong, 1987).
Description of Study Area
The study watershed covers approximately 61 ha and is located in
Choteau County near C arter, Montana at 47° 53'30" N and 110° SZzOO" W
(Figure
2 ).
It
is
located
on the
U.S.G.S.
C arter
N.E.
7.5
minute
quadrangle in the SW and SE 1 /4 's of section 30, Township 25 North, Range
7 East.
The watershed is located w ithin a mapped u n it of the Colorado
Shale and is located approximately 40 km north of the southernmost extent
o f the Laurentide ice sheet.
The watershed consists o f hummocky te rra in o f g la c ia l o rig in .
Local
r e l i e f is 22 m, with the highest elevations along the western boundary
approaching 922 m.
The watershed is drained by an unnamed trib u ta ry of
the Frank G ilb e rt Coulee, which drains to the Teton River and eventually
to the Missouri R iver.
The area is a mixture of gentle slopes in the west
and steeper slopes along the eastern boundary.
A well defined ephemeral
channel system drains in to an incised channel and a 0.25 ha pond situated
behind an earth dam constructed in 1973.
The annual average p re c ip ita tio n measured in nearby Great F alls is
390 mm yr"1, with a May-June maximum.
Average annual r a in f a ll varied from
186 to 475 mm yr"1 between 1954 and 1986.
F a lls between
Analysis o f wind data fo r Great
1950 and 1955 indicates a predominant annual wind d ire c tio n
13
•
J a c k s o n Form,
C arter, M ontana
_ — Ephemeral
Channels
W atershed
Boundary
Pond
------- - I M eter
Contours
LOCATION MAP
NO SCALE
100
200 Meters
N orth
Figure 2. Study area and lo catio n map.
47° 53' 30" N and 110° 52' 00" W.
The Jackson Farm is located at
o f west-southwest (Skidmore and Woodruff, 1968).
Of p a rtic u la r in te re s t
to th is study is the percent of cumulative erosive wind energy (EWE) th at
occurs by a p a rtic u la r month (see Table I ) .
c r it ic a l
According to the data, the
period fo r erosive wind is between October and A p r il, during
which 83 % of the erosive wind occurs.
14
The
s o ils
A rg ib o ro lls
of
(E thridge
m ontm orillonit i c
(unpublished
the
study
area
s e rie s ,
are
a
f in e ,
s ilty
m ontm orillonit i c
clay
loam)
and
fin e ,
f r i g i d , Typic Albaqualfs (Nishon s e rie s , a s i l t y c la y )
s o il
survey
maps,
Mr.
Raymond
M cPhail,
C o n servatio n ist, Choteau County, MT, 1987) (Figure 3 ) .
p a rt o f the Mr.
Norman W. Jackson farm which is
c u ltiv a tio n o f w in te r wheat.
fa llo w
A rid ic
system.
SCS
D is t r ic t
The watershed is
used mostly fo r the
The farm is managed in a w in te r wheat/summer
The watershed
is
almost
completely
tille d
with
the
exception o f the incised channel and pond areas and t h e ir margins.
The
watershed
h a lf-c e n tu ry .
has
undergone
considerable
change
in
the
la s t
The middle portion o f the study area was form erly a sm all,
seasonal la k e . In the 1940's, a channel was incised from the lake to the
coulee bottom to drain the lake to allow t i l l a g e .
The incised channel
remains an a c tiv e component o f the f lu v ia l
system, channeling overland
flow from the upland areas o f the watershed.
An earth dam was constructed
across the mouth o f the coulee to create a pond in 1973.
The landowner
had planned to stock th is pond with fis h but the pond has always emptied
a fte r
p r e c ip ita tio n
(Norman Jackson,
events
C a rte r,
and spring
MT, personal
ru n o ff because o f a slow leak
communication,
1987).
This has
prevented permanent f i l l i n g o f the re s e rv o ir, but i t drains slowly enough
(a period o f months) to be an e f f ic ie n t sediment tr a p .
nature
o f the
pond was b e n e fic ia l
to
the
present
provided access to re s e rv o ir bottom sediments.
The
study
ephemeral
because i t
15
Table I .
Erosive Wind Energy Occurring by Month at Great F a lls , MT.3
J
F
M
A
M
J
J
A
S
O
N
D
Monthly
Total (%)
15
14
9
9
5
4
2
2
4
9
12
15
Cumulative
Total (%)
15
29
38
47
52
56
58
60
64
73
85
100
Month
aErosive wind g re a te r than 8 m sec"1.
Agronomy Manual,
Source : USDA SCS 1988 National
385B
LEGEND:
WATERSHED
BOUNDARY
200
M eters
North
Figure 3.
Soil series map showing s o il mapping un its as defined by the
s o il survey team, SCS Choteau County Conservation D i s t r i c t .
Ethridge
s o ils are 38A, 38B and 3858, and Nishon s o il is 28.
S tip p led area was
undefined by SCS.
16
Thesis Organization
This f i r s t chapter has described the scope and purpose o f the
study,
previous s o il erosion studies which use the USLE and WEE models and 137Cs
techniques, and the study watershed.
Chapter 2 describes the methods o f data c o lle c tio n , assumptions used
and the computations required to use both the so il erosion models and 137Cs
method.
I t describes the method used fo r the topographic survey and the
d iv is io n o f the watershed into landscape u n its .
Data sources and methods
used to estim ate USLE and WEE fa c to r values are also described.
sampling procedures and laboratory analysis fo r 137Cs detection
Soil
in the
watershed are outlined as are the methods of c a lc u la tio n of the areal
concentrations, erosion and deposition ra te s , and watershed mass balance.
Chapter 3 is a substantive review of a ll the re s u lts from both the
model and 137Cs methods.
I t includes average annual erosion ra te
estimates
from the USLE and WEE, as well as an analysis of the s p a tia l v a r ia b ilit y
of those estim ates.
each s ite ,
The 137Cs re su lts included are I ) the 137Cs a c tiv ity fo r
2) the so il
average erosion
average areal
erosion and deposition rates fo r each s it e , 3)
and deposition rates
by topographic p o sitio n ,
concentrations by topographic po sition s,
4)
the
and 5) a to ta l
watershed erosion and deposition mass balance.
Chapter 4 is a discussion of the re su lts and the im plications of the
performance o f these methods of estim ating so il erosion and deposition to
s o il
erosion
p ro d u c tiv ity
n o rth -cen tral Montana.
assessments of other s im ila r watersheds
in
Possible sources of e rro r are also discussed.
The fin a l p art of Chapter 4 is the conclusions section.
17
CHAPTER 2
METHODS AND DATA SOURCES
Topographic Map Generation
A
The study area was mapped using a plane ta b le , telesco pic alidade
and the beaman arc procedure described by Compton (1962). The topographic
map was drawn at a scale o f 1:2500 using a i m
contour
in t e r v a l.
The
s ta rtin g elevation was determined by lo catin g the highest point in the
watershed and assigning i t the elevation o f the point o f highest r e l i e f
on the U.S.G.S. C arter NE 15 minute quadrangle (3023 fe e t or 921.4 m)
(Table
Station
A in
measurements were
Figure
4 ).
A ll
elevations
and lin e a r distance
rounded to the nearest .1 m.
A baseline was established between two central h illto p s (Figure 4 ).
The baseline was hand measured twice (632.2 m and 632.6 m) using a 100 m
tape, giving an average distance of 632.4 m.
Plane ta b le measurements of
the base lin e f e l l w ith in ± 4 m of th is length, in d ic a tin g errors of less
than .7% when measuring lin e a r distances.
No check on the accuracy of
elevation measurements is possible using the method described by Compton
(1962).
Following establishment of the baseline, flags were set at
to f a c i l i t a t e telescopic sighting from anywhere in
each end
the watershed.
This
was not possible in several instances, so two additional flags were set
18
Baseline
North
200
LEGEND:
Meters
WATERSHED
BOUNDARY
Figure 4.
Map showing lo c a tio n o f baseline AB and plane ta b le stations
A, B, C, D and E w ith in study watershed.
up to perm it the sig h tin g
of
at le a s t
tr ia n g u la r resection from any point in the
three fla g s
watershed.
which allowed a
The elevations and
lo catio n s o f the fla g s ite s were established with the a lid ad e r e la tiv e to
the known e le v a tio n a t one end o f the baseline and it s le n g th .
points became plane ta b le
s ta tio n s A, B, C and D.
These fla g
19
One a d d itio n a l ta b le s ta tio n (E) was added to allow coverage o f
extreme northern p a rt o f the watershed.
map using a tr ia n g u la r
ta b le
s ta tio n s was used
-the
I t was located on the plane ta b le
resection method.
Hence, a to ta l o f fiv e plane
and those established a t fla g
s ta tio n s were
located e x a c tly I m east o f each fla g .
Four s ta tio n s could be relocated
from
sig h tin g
anywhere
with
the
The mapping was accomplished by sig h tin g the s ta d ia rod through
the
te le s c o p ic
in
the
watershed
by
t h e ir
fla g s
a lid a d e .
te le s c o p ic alid ade to reveal a s ta d ia in te rc e p t, d ire c tio n
h a ir reading.
Trigonom etric c a lc u la tio n s y ie ld e d distance and
d iffe re n c e s and d ire c tio n was established by marking the
w ith
a ray drawn along the fid u c ia l
edge o f the
Topographic fe atu res were id e n tifie d
s ta d ia rod d ir e c t ly on them.
in the
was drawn in the f ie ld to allow fo r
d e fin itio n o f lo ca l topography
points
were
The
cross
ele v a tio n
plane ta b le map
te le s c o p ic a lid ad e.
watershed by placing the
Two hundred and
on the plane ta b le map in th is manner.
Data
and
eigh ty points were marked
f i r s t d r a ft o f a contour map
c o lle c tio n o f more points wherever
became d i f f i c u l t .
elim inated
if
the
e levatio n s
could
not
be
reproduced using the mathematical formulae described in Compton (1962).
Four data points
(1.4% o f those mapped in the f i e l d )
were elim inated
because there was a d iffe re n c e between the elevations ca lc u la te d in the
f i e l d and those generated during subsequent rechecking o f computations.
D ifferences between the e le v a tio n computed fo r ta b le s ta tio n B from
ta b le s ta tio n A and from la t e r , shorter sightings o f data points
A28 from ta b le
s ta tio n
A27 and
B were discovered when the computations
rechecked. The e le v a tio n a t ta b le s ta tio n B was determined from
were
Table 2,
20
which shows the r e la tiv e confidence values associated w ith the individual
telesco pic alidade sigh ting s.
Hence, an elevation o f 918.7 m was chosen
fo r ta b le s ta tio n B due to the high confidence values associated with
shorter
sigh ting s.
The elevations
determined
a fte r
the
sighting
of
s ta tio n B were adjusted to r e fle c t th is change in elevation (since ta b le
stations
C,
D and
E re lie d
on
th is
elevation
e levatio n s) and a ll data points were rechecked.
were redrawn and a fin a l
to
determine
th e ir
A ll elevation contours
revision o f the topographic contour map was
drawn.
Model Estimates
Universal Soil Loss Equation
The modified version of the USLE developed by G r if f in e t a l . (1988)
to estim ate sheet and r i l l erosion at d iffe r e n t points in a landscape was
used in th is study.
This equation, which combines the
method of Foster
and Wischmeier (1974) fo r measuring the L and S
facto rs fo r irre g u la r
slopes with the o rig in a l
Smith,
model
(Wischmeier and
1978),
can be
w ritte n as follow s:
D = (m + I ) R K ( x / l u) m S C P
(I)
where D is so il loss in tonnes per hectare per year ( t ha"1 yr"1) , m is
slope length exponent, R is the r a in f a ll- e r o s iv it y fa c to r in
the
megajoules
per m illim e te r per hectare per hour per year (Md mm ha'1 hr"1 yr"1) , K is
the s o il c r e d ib ilit y fa c to r in tonnes per hectare per
per megajoules
per
hour per hectare
m illim e te r ( t ha hr ha"1 MJ"1 mm"1) ; x is the distance
o f the sample point from the top o f the slope
p r o file in meters (m), I u
21
Table 2.
Confidence of table station B elevation determined by
d iffe re n t sightings.
Known
Route
Unknown
Elevation
Confidence
Aa----------------------------► B
(921.4 m)
632 m
918.1 m
Moderate
------------------------ B
630 m
918.7 m
Moderate
--------- B
700 m
918.7 m
High
A ------—► A 28 -*--------- B
700 m
918.7 m
High
1,080 m
918.3 m
Low
A
A ------—► A27
Ir
X
t
A ------
CO
Distance
aElevation of highest h illto p from U.S.G .S. 15 minute Carter N.E.
quadrangle. Table station A is located on the highest point of the same
h i l l . See Figure 3 fo r location.
is the length of the USLE un it plot (22.13
m), S is the slope gradient
factor, C is the cropping management factor
practices fa cto r.
and P is the supporting
A grid with perpendicular
lines drawn 125 m apart was
placed randomly over the study watershed to
determine the locations at
which soil losses were to be estimated.
This grid was then shifted 62.5
m in both southerly and easterly directions to generate additional sample
points.
The sample grid
points generated and th e ir labels are shown in
Figure 5.
Although th is approach produced 81 locations marked by grid lin e
intersection points,
soil
losses from sheet and r i l l
estimated at only 60 of these locations.
were eliminated as follow s:
Twenty-one
erosion were
points (locations)
I) nine points located on or near h illto p s
were eliminated because slope lengths and related USLE erosion
estimates
22
15 G
15 F
13 G
I I O
I I C
10 O
10 N
12 Q
I I H
I I G
11 E
10 M
10 K
12 P
12 O
12 N
12 M
12 K
14 P
14 O
13 D
13 C
I I B
14 N
14 M
14 L
10 P
LEGEND:
SAMPLE POINT
WATERSHED
BOUNDARY
200
Meters
North
Figure 5. Map showing location of randomly selected 125 m g rid points and
labels used fo r USLE and WEE calculations.
are zero in these circumstances (Wischmeier
and Smith, 1978); 2) six
points located immediately adjacent to and/or in channels were eliminated
because the USLE does not apply to channel erosion (Wischmeier and Smith,
1978); and 3) six points located on concave slope segments with slope
gradients less than one h a lf o f the cumulative average slope gradient from
the drainage divide
to
th is
location
were eliminated
based on the
23
assumption th a t these areas represent depositiona-1 zones (to which the
USLE does not apply.) (Wilson,, lS86b.; JG riffin e t a ] . , 1 9 8 8 ).
o f s o il
losses a t the remaining 60 s ite s
,Determination
involved estim ation o f USLE
fa c to r values a t these lo c a tio n s .
The r a i n f a l l -e r o s iv it y and s o il e r o d ib ilit y values used by the
USDA
S oil Conservation Service in th is watershed (R = 4-76.6 MJ mm ha'1 hr"1 yr"1
and K = .049 tonnes ha hr ha"1 MJ"1 mm"1 fo r both the
s e rie s ) were used as w ell (M cPhail,
R value includes the Rs adjustment
Slope
length
(L)
and
slope
Ethridge and Nishon
personal communication, 1987). This
fo r snowmelt.
g ra d ie n t
(S)
fa c to r
values
were
determined from th e topographic map constructed fo r th is p ro je c t.
The
slope length (x ) was measured by extending a lin e from each sample
p o in t
(g rid in te rs e c tio n ) up the slope to the drainage d iv id e , perpendicular to
the contours.
The remaining inputs required to
taken from Wischmeier and Smith (1 9 78 ).
The
estim ated a t each po int using the method fo r
estim ate L ( I and m) were
slope g ra d ie n t fa c to r was
ir r e g u la r slopes f i r s t
proposed by Foster and Wischmeier (1 9 74 ).
The crop management (C) fa c to rs were computed using the procedure
of
Wischmeier
and
Smith
(1978,
2 8 -3 4 ).
Table
3
summarizes
methodology and in term ediate re s u lts o f computing an annual
the
average C
fa c to r value fo r the w in te r wheat/summer fa llo w system employed on the
Jackson farm.
Cropstage C values were estim ated fo r each tim e period
the two year w in te r wheat/summer fa llo w cycle by using a weighted
of
average
o f C according to the amount o f p lan t cover and EI ( r a i n f a ll- e r o s iv it y )
in each p erio d . EI in the period was taken from Table 7 and the s o il loss
r a tio s were taken from Table 5 o f Wischmeier and Smith
(1 9 78 ).
Time
24
Table 3.
Computation o f Average Annual C Factor fo r T ille d S o ils ."
Dateb A c t iv it y '
Year I
10/1
4/15
5/15
6/15
8/15
Cumulative
% o f E Id
EI in
Period"
98
4
20
63
79
.06
.16
.43
.16
.25
Pl
I
2
3
Harvest
Soil Loss
R atio’
.17
.14
.12
.07
.04
Cropstage
C Value
Crop
Year
Totals
.0102
.0224
.0516
.0112
.0100
.1054
Year 2
4/15 C u ltiv a tio n I
5/1 C u ltiv a tio n 2
10/1
Rotation Totals
4
6
98
.23
.39
.02
.92
.0046
.3588
.3634
.4688
2.00
Average C Value Year"1
.2344
"Annual C value c a lcu lated fo r residue - 2500 kg ha"1.
bDate by which cropstage growth or t i l l a g e operation is completed.
cCropstage a b b reviatio n s, Pl = plan t crop; I = 10 % crop canopy cover; 2
= 50 % crop canopy cover; 3 = 75 % crop canopy cover.
^Cumulative EI fo r Great F a lls , MT from Wischmeier and Smith, 1978,
Table 7.
"El fo r period ending at date in f i r s t column, from Wischmeier and Smith,
1978, Table 7.
’Soil loss r a tio fo r period ending a t dates shown in f i r s t column, from
Wischmeier and Smith, 1978, Table 5, 5-D.
periods were distingu ished by a c t iv it y and cropstage w ith
o f the Soil
Conservation Service and the landowner.
values were obtained by m u ltip ly in g
the
EI by the
the assistance
The
s o il
cropstage C
loss
Cropstage (C) values were then added together and divided by
to obtain the average annual C value fo r the w inter
r a t io .
two years
w h ea t/fa llo w system
25
o f 0.2344 reported in Table 3 .
A value o f 0.053
Table 10 o f Wischmeier and Smith (1978) fo r the
o f the watershed.
no
conservation
watershed.
was in te rp o la te d from
u n t ille d , grassed parts
The supporting p ra c tic es (P) fa c to r was ignored since
p ra c tic es
fo r
water
erosion
are
used
in
the
study
A ll USLE fa c to r estim ates were converted to m etric using the
method o f Foster e t a l .
(1981) and combined using the p o in t method o f
G r if f in e t a l . (1 9 88 ).
Wind Erosion Equation
The m odified
version
o f the
WEE proposed by Skidmore
(1988)
estim ate wind erosion a t d iffe r e n t points in a landscape was used in
study.
This
version
of
the
WEE s u b s titu tes
a
series
of
to
th is
equations
(c a lc u la te d in stages) fo r the o rig in a l nomographs and can be w ritte n :
El = r
(2)
where El is the f i r s t stage erosion estim ate in tonnes per hectare per
year ( t ha'1 yr"1) , and I '
is the s o il c r e d ib i lit y fa c to r value in
tonnes
per hectare per year ( t ha"1 yr"1) .
E2 = T K
(3)
where E2 is the second stage erosion estim ate in tonnes per hectare per
year ( t ha"1 y r'1) , and K is the s o il ridg e roughness
E3 = I'KC
fa c to r .
(4)
where E3 is the th ir d stage erosion estim ate in tonnes per hectare per
year ( t ha"1 yr"1) and C is the c lim a tic fa c to r.
26
E4 = (WF0-348 + E3°'348 - EZb34aF i87
(5)
where E4 is the fo u rth stage erosion estim ate in tonnes per hectare per
year ( t ha"1 yr"1) , and
WF is a f i e l d length weighting parameter
defined
by equation 5a.
WF = E 2 ( l.0 - 0.122(L/Lo)-°-383 e x p (-3 .3 3 L /L o ))
(5a)
where L is the s t r ip f i e l d length in meters (m) and Lo is a f ie ld
length
parameter defined by equation 5b.
Lo = 1.56 x 106(E2)"1'26 exp(-0 .0 0 15 6 E2)
(5b)
ES = P1 E4 (P2)
(6)
where ES is the f in a l erosion estim ate in tonnes per hectare per year
ha"1 y r'1) and P1 and P2 are v e g etative parameters defined by
(t
equations 6a
and 6b.
P1 = exp(- 0.759V - 4.74 x IO 2V2 + 2.95 x IQ-4V3)
(6a)
P2 = I + 8.93 x IO 2V + 8.51 x IO 3V2 - 1.5 x IO 5V3
(6b)
where V is the v e g e tative fa c to r input in tonnes per hectare ( t ha"1 yr"1)
defined by equation 7a.
The v e g e ta tiv e residue was estimated using the equations o f Armbrust
and Lyles (1985) reported in Skidmore (1988)
as a function o f small grain e q u ivalen ts.
which compute
The equations
the residue
used
fo r
27
w in te r wheat are:
V = 0.2533 (SGe) 1363
where V is
the
v e g e tative
kilograms per hectare (kg
fa c to r
ha:1) ,
input
(7a)
fo r
equations
6a
and 6b
in
and SGe is the small grain equivalent
c a lc u la te d using equations 7b, 7c and 7d fo r standing stubble,
fla tte n e d
stubb le, and growing crop in fla tte n e d stubble, re s p e c tiv e ly .
SGe= 4 .3
(Rws)"97
(7b)
where SGe is the small grain e q u iva len t, and Rws is the above-ground
weight o f the standing stubble residue in kilograms per hectare (kg
SGe= 7.3
(Rwf) 0
dry
ha'1) .
(7c)
where SGe is the small grain e q u iv a le n t, and Rwf is the above-ground
dry
weight o f the fla tte n e d
(kg
stubble residue in kilograms per hectare
ha'1) .
SGe= (8 .9 )-172(7 .3 )-82a(Rwg) (-9)(-172)+(-8)(-020)
(7d)
where SGe is the small grain e q u iva len t, and Rwg is the above-ground
weight o f the crop growing in fla tte n e d residue in kilograms (kg
The I , L, Lo and V fa c to r values were then determined fo r the
points generated fo r the USLE a p p lic a tio n
(see Figure 5)
dry
ha'1) .
same sample
and used in
equations I through 7d to estim ate WEE s o il losses.
Soil c r e d ib i lit y
( I ) was estimated by determining the non-erodible
fra c tio n o f the surface s o il.
The non-erodible fra c tio n ( i . e . ,
g re a te r than 0.84 mm in diam eter) was determined fo r 50
p a rtic le s
s u r f ic ia l s o il
28
samples b y standard dry sieving using
s ie v e .
c ir c u la r f l a t screen
, I n it i a l I values were c a lc u la te d fo r these samples using Table I
o f Woodruff and Siddoway (1965)
by s o il mapping u n it.
a
a 0.84 mm
s e ries
of
d iffe re n c e s
(see Table 18, Appendix C) and grouped
Average values were computed fo r each
d iffe re n c e
of
means
te s ts
was
used
between means were s t a t i s t i c a l l y
to
group and
determine
s ig n ific a n t
if
(Table 4 ) .
S t a t is t ic a lly
s ig n ific a n t d iffe re n c e s were found between the means fo r
the
(#38A)
Ethridge
Ethridge map u n its
s ig n ific a n c e .
map u n it
(#388 and #3858)
Nishon
model was used here to
series
(#28)
and other
a t the 5 % and 10 % le v e ls
The four means were used
s o il u n its were not sampled
losses.
and the
of
to estim ate I values because the
p ro p o rtio n a lly and because the wind erosion
assess s p a tia l v a r i a b ili t y o f wind-generated s o il
One po in t (2F) th a t f e l l w ith in the incised channel was excluded
from the WEE
analysis because wind does not erode th is area.
A fte r the i n i t i a l
computation o f I ,
the points located on windward
slopes s h o rter than 152 m were adjusted using Figure I o f Woodruff and
Siddoway
(1 9 6 5 ).
This
diagram
computes
knoll
c r e d ib i lit y
( I s)
function o f slope grad ien t fo r two d if fe r e n t landscape p o s itio n s .
c r e d ib i lit y ( I s) was expressed as a percentage (> 100 %) fo r
points and I '
(th e fa c to r combining I and I s) was estim ated
as
a
Knoll
q u a lify in g
by computing
the product o f I and I s fo r these points (Chepil e t a l . ,
1962; Woodruff
and Siddoway, 1965).
I ' was reduced
by 50 % u n til
Following periods o f c u ltiv a tio n ,
seeding follow ing
the advice o f Bondy and
associates
(1 9 8 0 ).
K fa c to r values were obtained from the equations o f W illiam s e t
a l.
(1984)
in
Skidmore (1 9 88 ).
K
values
were
estim ated
fo r each
29
Table 4. S oil E r o d ib ilit y by Soil Mapping U n it.*
Number
o f Samples
Average
I Valuec
28
Nishon(28)
3
280.9
-
E th r idge(38A)
7
180.2
LO
O
-
E th r idge(385B)
27
253.3
.#
.05
Ethridge(38B)
13
234.2
#
.10
Soi I sb
38A
385B
38B
#
d iffe r e n c e o f means te s t re s u lts in d ic a tin g le v e l o f s ig n ific a n c e o f
d iffe re n c e o f means in proportions. The # symbol in d ic a te s no s ig n ific a n t
d iffe re n c e .
bSoil numbers are s o il mapping un its used by the s o il survey team, SCS,
Fort Benton (see also Figure 3 ) .
0See Appendix C fo r ta b le showing map i d e n t i f i cation codes, s o il mapping
u n it, percentages o f s o il aggregates > 0.84 mm and s o il c r e d ib ilit y ( I )
index values fo r each o f the 50 s u r f ic ia l s o il samples.
cropstage period or t i l l a g e
op eration.
A ridg e height o f 42 mm and a
ridg e spacing o f 356 mm was used to estim ate K fo r periods o f
( .9 0 ) ;
w h ile a ridge height o f 102 mm and a ridge
used fo r periods immediately follow ing seeding
c u ltiv a tio n
spacing o f 356 mm was
( .4 9 ) ; and a ridge height
o f 25 mm and a ridg e spacing o f 610 mm was used fo r the periods follow ing
c u ltiv a tio n w ith a to o lb a r and duckfoot implements (.6 6 )
At th is stage, each p o in t's lo c a tio n
w ith in a s o il mapping u n it
recorded (see Figure 3 and Table 18, Appendix C).
average I '
( E l,
equation # 2 ),
(see Table 5 ).
Using the
was
appropriate
E2 (equation #3) was computed
fo r each
period o f the w in te r wheat/summer fa llo w system as the product o f I ' and
K.
Next, the annual C value, 0 .9 0 , was m u ltip lie d by E2 to
(equation # 4 ).
compute E3
This C value is used by the SCS to represent the v ic in it y
30
Estimates o f K Factor, Vegetation Weight and Erosive Wind
Energy fo r Cropstage Periods and T illa g e O perations, Jackson
Farm, C a rte r, MT.
A c tiv ity *
8 /1 5 -1 0 /1
Harvest
1 0 /1 -4 /1
Winter
4 /1 -5 /1
K
EWE in
Periodb
-
3035
10270
.49
2430
8270
.66
.74
1190
2110
.90
.09
950
1765
.90
.17
.30d
O
CXJ
Summer
Fallow
Vegetation Small Grain
Weight
Equivalent
kg ha'1
Seeding
.30d
760
1705
.49
.09
1 1 /1 -4 /1
Winter
180
520
.49
.65
390
980
.49
.09
1100
2310
.49
.12
4 /1 -5 /1
5 /1 -8 /1 5
3 - 4 weeks
growth
Mature
growth
-
°n
1 0 /1 -1 1 /1
ro
5 /1 -1 0 /1
C u ltiv a tio n (Z )
Loss of
Residue
%
O
CXJ
Period
O
CJl
Table 5.
“A c tiv ity or cropstage growth completed during period in Column I .
^Proportion o f erosive wind energy.
cLoss o f dead residue fo r summer and w in te r due to biomass decomposition.
dEstimates o f residue loss due to t i l l a g e operations from National
Agronomy Manual, Montana Supplement, Table I , (USDA-SCS, 1988); 60 kg ha"1
assumed to remain a f t e r deep furrow d r i l l seedings (Nadwornick, personal
communication, 1990).
o f C a rte r, MT and was in te rp o la te d from a statewide
contour map o f C
values (M cPhail, personal communication, 1987).
The next stage o f the computation
fa c to rs , L and Lo.
incorporated
the
f ie ld
length
The f ie ld length o f the study area was a function
the unsheltered distance ( i . e . , distance in which wind b a r r ie r
of
erosion
31
control measures were not used) o f the f ie ld s tr ip s . On the
th is was simply the width o f the f ie ld s trip s from the
Jackson farm
western edge to
the lo c a tio n o f the sample po int along the wind erosion
d ire ctio n
to the fa c t th a t no b a r r ie r and/or other wind control
p ractices were
employed.
F a lls ,
Data c h a ra c te rizin g wind
MT from the Montana
Manual (SCS, 1988) and WEE
used to determine the
,due
d ire c tio n and in te n s itie s a t Great
Supplement o f the 1988 National Agronomy
handbook (Skidmore and Woodruff, 1968) were
predominant wind erosion d ire c tio n .
The minimum windspeed necessary fo r wind erosion is approximately
29 km hr"1 (8 m sec"1)
d ire c tio n
( Bondy e t
a l.,
1980).
The eval uation
o f wind
and in te n s ity only considered winds above th is threshold and
revealed two predominant wind erosion d ire c tio n s o f SW and WSW.
d ire c tio n
applied
to
the periods
o f January to
December.
The SW d ire c tio n applied to the period o f A p ril to
The
WSW
March and October to
September.
The unsheltered distance fo r each po int was obtained by
measuring lin e s
drawn p a r a lle l to th is predominant wind d ire c tio n (WSW)
from the western
edge o f the f i e l d
s t r ip to the sample p o in t.
This
value was used fo r
January to March and also fo r October to December.
distance along the SW d ire c tio n was estim ated by
WSW by the cosine o f 22.5° (th e angle o f the
d ir e c tio n s ).
This L value was applied
These L values were then
d iv id in g the L along
d iffe re n c e between the two
between A p ril
used to compute the
fa c to rs WF and Lo (equations #5a and 5 b ).
The unsheltered
f ie ld
and September.
length weighting
E4 was then c a lc u la te d fo r each
period o f the cycle (equation # 5 ).
The fin a l
fa c to r o f the WEE, the v e g etative cover fa c to r ( V ) , was
needed to complete the computation o f erosion estim ates.
The V fa c to r
32
was a function o f cropstage period, t i l l a g e operation and stubble
above-ground weight
converted to
a small
(Table
5 ).
grain
The V fa c to r
equivalen t
fo r
each
using equations
residue
period
7b,
was
7c or 7d.
Equation 7a was then used to determine the fa c to r value in p u t, V, which
was used in equations #6a and #6b to estim ate s o il losses a t each
sample
p o in t.
Standing stubble weight was estim ated by m u ltip ly in g the average
y ie ld
fo r
the
Jackson
Farm,
25
bushels
acre'1
(Jackson,
personal
communication, 1987; M cPhail, SCS, personal communication, 1990), by
HO
pounds bushel"1 to obtain an above-ground weight estim ate o f residue o f
3025 kg ha'1 (USDA-SCS, 1988).
points th a t f e l l
(John Siddoway,
The small grain equivalen t o f
the s ix
in the grassed area was estimated to be 5600
SCS-Fort
Benton,
MT,
personal
kg ha'1
communication,
1990).
S ix ty kilograms per hectare o f dead residue was assumed to
remain a f t e r
seeding and fo u r weeks crop growth, weighing approxim ately
120 kg ha'1,
was assumed before dormancy in w in te r, fo r a to ta l o f
kg
ha'1
(Nadwornick,
SCS
S tate
Agronomist,
Bozeman,
approximately 180
MT,
personal
communication, 1990).
A s o il loss estim ate fo r each period o f the c ro p /fa llo w cycle was
c a lc u la te d by using ES (equation # 6 ).
The annual erosion ra te
estimates
fo r each period were then m u ltip lie d by a weighted average o f
wind energy
(see Table I
fo r monthly to t a ls )
estim ate period wind erosion ra te s .
in the same period
The period to ta ls were then
to estim ate annual wind erosion rates fo r the crop and fa llo w
the c y c le .
erosive
Table 6 shows these computations fo r sample po in t
to
added
years o f
ID.
33
Table 6. Sample WEE Computation fo r Sample Point 2M.a
Period
K
Residue
SGe
kg ha"1
Fallow year
8/15-10/1
10/1-4/1
4 /1 -5 /r
5/1-10/1°
Crop Year
10/1-11/1
11/1-4/1
4 /1- 5 /1
5/1-8/15
V
EWEb
Mg ha"1
ES Period
Erosion
t Ii a"1 yr"1
.49
.66
.90
.90
3035
2430
1190
950
10270
8270
2110
1765
74.40
55.39
8.60
6.74
.05
.74
.09
.17
0.0
0.0
0.1
0.1
0.0
0.0
0.0
0.0
.49
.49
.49
.49
759
180
390
1100
1705
520
985
2310
6.43
1.27
3.04
9.78
.09
.65
.09
.12
0.1
5.4
2.0
0.0
0.0
3.5
0.2
0.0
Total Cycle
2.0
3.7
Average
1.0
1.8
aWEE computed using I ' of 234.2 t ha1, C of .90, I. of 85 m.
Proportion of cumulative erosive wind energy (EWE) in period -in Column
cPeriods in which soil c re d ib ility ( I ')
plowing ( Bondy et aI . 1980).
is reduced by I50 % following
Cesium-137 Erosion Estimates
Cesium-137 Sample Site Selection
Slopes were paced in the fie ld to determine the location of 137Cs
transects.
The highest point on the transect was chosen
sample s ite and the lowest point was chosen as a
as a h illto p
channel sample s ite .
Following the method of Brown et a l. (1981b) slopes were segmented by
pacing u p h ill, following the path of
greatest slope, dividing slope
lengths in to th ird s with footslope and
midslope sample sites chosen at
.34
distances one th ir d and two th ird s up
6 il lu s t r a t e s how t h is
other h a lf selected
s ta rtin g a t the same
sample
s ite s
channel)
(two
were
the slope, re s p e c tiv e ly .
Figure
s e le c tio n was made fo r each tra n s e c t, with the
by
using the
channel s it e .
h illt o p s ,
d e lin e a ted
in
same method on the opposite slope
A to ta l o f s ix tran sects w ith seven
two midslopes,
the
fie ld .
two
Due to
footslopes
the
high
and one
cost
of
la b o ra to ry a n a ly s is , only a portion o f the samples were analyzed fo r 137Cs
a c t iv it y .
The samples c o lle c te d during the f i r s t sample c o lle c tio n were
analyzed f i r s t .
U n fo rtu n a te ly , enough funds were not a v a ila b le to analyze
the samples c o lle c te d during the second c o lle c tio n p e rio d , one year l a t e r .
T h erefore, the 137Cs re s u lts are biased towards the steeper sloping, eastern
po rtion o f the study area, where the f i r s t samples were c o lle c te d
(see
Figure 7 ).
Figure 7 also in d ic a te s the s ite s th a t were sampled in the incised
channel leading from the drained lake to the pond (n = 3 ) ,
bottom
(n
= 4 ),
and along
watershed (n = 4 ) .
(uneroded) control
a fence
lin e
250 meters
in the pond
from
the
study
The samples taken along the fe n c e lin e represent
s ite s because the fence acted as a b a r r ie r to wind
erosion and a t i l l a g e
berm
fe n c e lin e .
a to ta l
O v e ra ll,
prevented overland flow from crossing th is
o f 53 s ite s
was used and s o il
samples
representing 261 increments were c o lle c te d , although only 83 increments
from 25 s ite s were analyzed fo r 137Cs a c t iv it y .
Following the method used by Brown e t a l .
(1981b),
a strateg y to
defin e areas o f landscape u n its , termed topographic p o s itio n s , was
used.
These areas were computed using slope p r o file s which were
divided in to
segments by assigning boundaries to points halfway between
each 137Cs
35
H ILLTO P
SAM PLE
M ID S LO P E
S A M PLE
FO O TSLO PE
SAM PLE
CHANNEL
SAM PLE
Figure 6.
Diagram showing topographic positions.
sample lo ca tio n .
The p ro file s were drawn 2 cm (50 m) apart on a copy of
the topographic map and areas fo r each topographic position
by connecting each boundary point to it s counterpart on
p ro file s .
were defined
adjacent slope
These areas were calculated by adding up 2.5 m2 squares w ithin
the boundary lin e s using metric grid paper.
The
areas corresponding to
each o f the topographic positions are shown in Figure 8.
Cesium-137 Sample Collection
The 137Cs samples were gathered from I m3 hand excavated so il
p its .
Care was taken to preserve the soil pedon in a natural, uncompacted state.
A fte r the s o il p it was dug, the 137Cs samples were
removed in a la te ra l
method to minimize the effects of compaction during sampling.
A 5.1-cm
diameter PVC tube with a cap affixed to one end to prevent loss was used
to c o lle c t the
samples.
Holes were d r ille d in to the cap to allow a ir to
36
C O D E KEY
— 2T1
North
----4 T I
5 ----- 5 T I
QTI
q
7—7T1/7T2
8 ———- — QT2
9 -—— 5T 2/ /
----- ------
200
Meters
L O C A T IO N MAP
Control sites
□□
LEGEND;
EPHEMERAL
CHANNELS
Legend
Cs Sample Site
Not To Scalel
WATERSHED
BOUNDARY
Q
Buildings -H- Fence
Figure 7. Study area map showing locations o f 137Cs samples w ithin study
watershed and locations o f control samples on the Jackson Farm.
escape so i t
would not displace potential sample volume.
The tube was driven h o rizo n ta lly in to the wall of the so il p it
a m allet.
The tube was 7.6 cm long and was completely f i l l e d with
sample to obtain standard sample volumes.
p it wall w ith a large kn ife .
with
each
The tube was removed from the
Care was taken to
preserve the in te g rity
o f the sample by preventing the entry of topsoil
and organic matter into
the tube during extra ction .
The samples were then placed in p la s tic lined
37
Figure 8. Areas corresponding to each of the topographic positions used
fo r 137Cs extrapolation.
sample bags and transported to Bozeman.
Cesium-137 Sample Increments
The samples were taken in increments whose
depths corresponded to
the width of the sampling device, 5.1 cm, or m ultiples o f i t , depending
on t illa g e
practices at d iffe re n t sampling s ite s .
Plowed sites were
assumed to have uniform 137Cs a c tiv ity in th e ir plow la yer, or Ap horizons
38
(Brown e t a l . , 1981b; Campbell e t a
15
cm
(Jackson,
personal
communication, 1987).
l 1982) .
communication,
Below these increments, samples
bottom and the control s ite s )
in 5.1 cm increments.
was tre a te d as
were gathered as
30 cm fo r the
in the incised
were sampled from the top
A minimum o f f iv e samples were
personal
successive 5.1
depth o f about
The n o n -tille d s ite s ( i . e . ,
was set as
M cPhail,
From th a t point i t
s in g le increments o f 5 .1 cm to a to ta l
plowed s ite s .
1987;
This sample was c o lle c te d as three
cm deep increments and then bulked.
a homogeneous sample.
The plow la y e r
channel, pond
o f the s o il p it
c o lle c te d fo r each
n o n -tille d s it e .
Cesium-137 Sample Preparation
In the la b , the 137Cs samples were dried at 105° C fo r a minimum
of
72 hours, and then m echanically ground and sieved through a 2 mm screen.
Each sample was weighed and a r i f f l e
re p re s e n ta tiv e 100 g sample.
to
standardize
the
s p l i t t e r was used to
This uniform 100 g sample s ize
counting times
(D r.
Ingevar
Larsen,
remove a
was required
Environmental
Sciences D iv is io n , Oak Ridge National Laboratory, Oak Ridge, TM, personal
communication,
1988).
The samples were then tra n s fe rre d to
dry s o il
sample bags, closed and wrapped fo r shipment.
Cesium-137 Laboratory Analysis
The samples were analyzed fo r 137Cs concentrations
Larsen a t
by Dr.
Ingevar
Oak Ridge National Laboratory w ith a Canberra lith iu m
d r ifte d
germanium (G e (L i)) gamma ray detecto r coupled to a Nuclear Data
4096 channel data analyzer system (C utshall and Larsen, 1980;
CutshalI ,
1981).
The procedure required a counting tim e o f
No. 6700
Larsen and
60 to 100
39
minutes per sample.
137Cs a c t iv it y ,
c u rie
E ig h ty -th ree sample increments were
expressed in pi cocuries
analyzed fo r
per gram o f s o il
(pCi g"1) .
is a measure o f r a d io a c tiv ity such th a t one c u rie
A
is 3 .7 x IO10
d is in te g ra tio n s per second per gram o f radium and one pi cocurie is IO*12
curies
(C utshall
concentrations
and
Larsen,
are reported
1980;
in
A rnalds,
un its
o f pCi
1984).
g"1.
The
137Cs mass
The 137Cs data were
transformed from pCi g'1 to pCi cm"3 by m u ltip ly in g
mass concentrations by
average bulk d e n s itie s (obtained from a weighted
average bulk density by
depth o f the horizon in which the 137Cs sample
a c t iv it y was converted to areal
(pCi cm"3) by the depth o f
was c o lle c te d ).
This 137Cs
concentration by m u ltip ly in g the a c t iv it y
the sample la y e r (cm).
. Bulk Density Sample C o llec tio n
Bulk
density
measurements
concentration c a lc u la tio n s .
were
required
fo r
the
137Cs
areal
Bulk density samples were taken a t the
tim e as the 137Cs samples using s o il sampling tin s o f known volume.
s o il
horizon was sampled in each s o il
p it.
same
Every
The g e n e ra lly dry
s o il
samples were d rie d and weighed to determine dry bulk d e n s ity .
It
possible to determine water content or moist bulk d ensity due
to the lack
o f control o f evaporation in the storage area a t the f i e l d
long time between sampling c o lle c tio n
(two 4-week
s ite
periods,
was not
and the
one year
a p a r t).
Method o f Areal Cesium-137 Analysis
Areal analysis r e lie s on the fa c t th a t 137Cs loss has been found
be proportional to s o il loss and to vary a t depth ( Rogowski and
1970; McHenry e t a l . , 1973; McHenry and R itc h ie , 1977).
The
to
Tamura,
method used
40
fo r areal
analysis o f 137Cs in th is study was taken from
associates (,19.83).
This method computes s o il loss using
De Jong and
the follow ing
equations:
137Cs loss = [0 .9 5 X - Y] [0 .9 5 X ] 1
(8)
where 137Cs loss is the amount o f 137Cs lo s t (p e rc e n t), X is average
present a t control s it e (pCi cm'2) , Y is average 137Cs present a t
137Cs
eroded
s ite s (pCi cm"2) ; and
s o il loss = 137Cs loss x d x BD
(9)
where s o il loss is the amount o f s o il (g cm'2) , d is the thickness (cm)
the la y e r in which 137Cs is present, and BD is the average bulk
cm"3) o f the la y e r in which 137Cs is present (De Jong e t
a l.,
of
density (g
1983).
A value o f 95 % o f the 137Cs areal concentrations was used to make the
c a lc u la tio n s because previous studies had found th a t removal o f
137Cs by
crops, animals and snow d r if t in g
et a l.,
1982; Arnalds, 1984).
is approximately 5% (De Jong
The concentrations o f 137Cs were
cm'2 and were generated by m u ltip ly in g the la b o ra to ry
(pCi g'1) by the bulk density (g cm'3) and the
Jong e t a l . ,
expressed in pCi
concentration data
increment depth (cm) (De
1983).
In t i l l e d
s ite s , gains o f 137Cs by deposition were ca lc u la te d
137Cs d is tr ib u tio n a t depth.
The depth to which 137Cs was
areas was estim ated by subtracting 15 cm (plow la y e r
maximum depth in which 137Cs was present.
The
using
present in these
depth) from the
thickness o f s o il deposited
by s o il r e d is tr ib u tio n was found by:
D1 = D2 - 15 cm
(10)
41
where D1 is the thickness (cm) o f s o il deposited, D2 is the maximum depth
(cm) to which 13^Cs is present, and 15 cm is the depth o f
c u ltiv a tio n .
S oil gain was estim ated as
s o il gain =^= D1 x BD
where BD is s o il bulk density (g cm"3) ,
cm"2 (De Jong e t
a l.,
1983).
(11)
and s o il gain is expressed in g
For u n tille d
s ite s ,
gains o f 137Cs were
estim ated by m u ltip ly in g the maximum depth o f 137Cs a c t iv it y detected
the s it e 's
average bulk d e n s ity . Soil gains were only computed i f
s it e 's to ta l areal concentration was g re a te r than th a t o f the
by
the
average o f
the control s ite s (11.10 pCi cm"2) .
Erosion and Deposition Rate and Mass Balance Estimation Method
Annual
to ta l
s o il
erosion
and deposition
ra te s
were determined
by d iv id in g
lo s s /g a in by the time th a t 137Cs has been deposited in
watershed (since 1954) except fo r the pond s ite s .
Deposition a t
s ite s was computed using the time period since dam construction
1973 to 1987, or 14 y e a rs ).
and areas
of
landscape
Averages based upon the areal
un its
(topographic
determine the erosion or deposition ra te
p o s itio n .
topographic
yr"1.
These
values
were
then
p o sitio n s)
the
those
(i.e .,
concentrations
were
used to
associated w ith each topographic
m u ltip lie d
by
the
area
of
each
p o sitio n to obtain mass erosion and deposition rates in Mg
H illt o p , midslope, footslope and incised channel erosion rates were
combined to estim ate a net mass erosion ra te and t i l l e d channel and pond
deposition ra te s were combined to obtain a net deposition r a te .
42
CHAPTER 3
RESULTS
USLE Erosion Estimates
USLE so il loss estimates were computed at 81 points on a randomly
located 125 m grid using constant R, K and P values and s p a tia lly
variable C, L and S values in th is study.
Most of the v a r ia b ility in the
erosion estimates was due to variations in the topographic factors,
L and
S, because one C was used with the exception of six points which
f e ll
w ithin the grassed margin of the pond area.
The fin a l 60-sample set was
determined using the method of G riffin et a l. (1988) and Wilson (1986b)
to
eliminate
environments.
sample grid
points
near h illto p s
and in
depositional
The USLE factor values were combined using the point
method of G riffin et a l. (1988).
The topographic factor and soil loss estimates are summarized in
Table 7.
The slope gradients and slope lengths varied considerably
from
point to point and account fo r most of the spread in so il loss estimates.
Watershed so il losses averaged 4.5 t ha'1 yr"1.
A ll 81
are reproduced in Figure 9 as a frequency histogram
points where erosion equals 0.0 t ha"1 yr"1.
This
point estimates
(including the 21
diagram indicates how
the d is trib u tio n of point estimates is p o sitive ly
skewed with large
numbers of estimates concentrated in the 0 - 6 t ha"1 yr"1 range and smaller
43
Table 7.
USLE Soil Loss Estimates.®
Slope Length (m)
Slope Gradient (%)
Soil Loss ( t ha"1 yr"1)
Minimum
Maximum
Average
Standard
Deviation
19.0
0.3
0.0
373.0
16.0
14.9
92.2
4.2
4.5
71.3
3.0
3.6
aUSLE estimates using sample generated from a random 125 m g rid (n = 81).
EROSION ESTIMATES (T /H A /Y R )
Figure 9.
(n = 81).
D is trib u tio n of
USLE so il loss rates fo r Carter watershed
44
numbers spread throughout the 7 - 18 t ha"1 yr"1 range.
s ta tis tic a l analysis was conducted to determine
of the watershed USLE so il loss average.
Using
A nonparametric
the level of confidence
a sample size of 60, the
watershed so il loss average of 4.5 t ha"1 yr"1 and standard deviation of
3.6 t ha"1 yr"1, and a desired
2.13, corresponding to an
formation rates commonly
precision of + 1.0 t ha"1 yr"1, Z equaled
area under a normal curve o f
0.4834.
Soil
assumed fo r so ils developed in glacial t i l l s
exceed 1.0 t ha"1 yr"1 (Dr. Gerald Nielsen, Department of Plant and Soil
Science, Montana State University, Bozeman, MT, personal
1990).
communication,
This analysis indicates that the sample mean f e ll w ithin + 1.0 t
ha"1 yr"1 of the true mean with a confidence level
of 96.7 % (level of
significance of 0.033).
■ The spatial v a r ia b ility of USLE soil loss estimates is depicted by
Figure 10.
The squares,
triangles
and circle s
denote points that
represent h illto p s , ephemeral channels and concave slopes,
These points were excluded from the original 81 point
erosion was estimated to be 0.0 t ha"1 yr"1 fo r them.
diagram shows an area of high erosion due to
of the watershed.
respectively.
sample set and
Examination of th is
water in the eastern h a lf
This f lu v ia lIy erosive area corresponds to steeper
sloping areas of the eastern h a lf of the watershed (compare Figures 2 and
10) .
WEE Erosion Estimates
Eighty points were sampled fo r the WEE calculations.
calculations were based upon s p a tia lly variable I '
variable I ' , K and V, and constant C factor values.
The WEE
and L, temporally
The I ' and L factor
45
LEG END:
SAMPLE POINT
WATERSHED
BOUNDARY
200
Meters
North
Figure 10. Spatial v a r ia b ility of USLE s o il loss rates ( t ha"1 yr"1) fo r
Carter watershed. Squares, trian gles and c irc le s represent points where
the USLE estimated erosion as 0.0 t ha"1 y r'1. Shaded area represents area
o f high water erosion.
values
were obtained using topographic and so ils maps. L was
to re fle c t a seasonal s h ift in the wind erosion d ire c tio n .
Bondy et a l . (1980), K and V factors were estimated using
National Agronomy Manual and the equations reported in
The C value was held constant over time but erosive
I
adjusted
Sim ilar to
the 1988
Skidmore (1988).
wind energy (EWE)
46
values, which varied by period, were incorporated.
estim ated
fo r
each
cropstage
and
tilla g e
period
Wind erosion was
of
crop/summer fa llo w system fo r a ll 80 p o in ts . A two year
was computed based upon
period estim ates o f s o il
m u ltip ly in g the period s o il loss estim ate by the
the
two year
weighted average
lo ss ,
obtained by
cumulative erosive wind
energy in th e p e rio d .
Table
8
summarizes
standard deviatio n s
estim ates.
a ll
fo r
the
I'
watershed
minima,
and L fa c to rs
maxima,
averages
and watershed WEE s o il
The watershed WEE average was 4 .5 t ha"1 yr"1. Figure 11
80 s o il
loss
estim ates
as a frequency histogram.
This
reveals a p o s itiv e ly skewed d is tr ib u tio n , influenced by the
o f estim ates th a t were 0 .0 t ha"1 yr"1.
was used to determine the le v e l
average.
A nonparametric
yr"1) , and standard d e v ia tio n (4 .6 t ha"1 y r'1) and
+ 1.0 t ha'1 yr"1, Z equaled 0 .6 5 , corresponding
curve
of
0.2422.
in te rs e c tio n s
Therefore,
w ith
o f a 125 m g rid ,
a sample
loss
shows
diagram
la rg e number
s t a t is t ic a l te s t
o f confidence o f the
Using a sample s ize o f 80, the watershed
and
WEE watershed
average (4 .5 t ha"1
a desired precision o f
to an area under a normal
s ize
the watershed WEE
determined
by the
average estimates
erosion fo r the watershed w ith in ± 1.0 t ha"1 y r'1 o f
the tru e mean with a
confidence le v e l o f
o f 0 .5 1 6 ).
48.4 % (le v e l o f s ig n ific a n c e
This low
le v e l o f confidence occurred because o f the high standard d eviatio n th a t
is
due to
the
in fluence
estim ate (See Figure 1 1 ).
o f the o u tly in g
p o in t's
(2P)
extremely high
This po int had a high wind erosion estim ate due
to i t s high s o il c r e d ib i lit y caused by a high I s adjustment o f 671 %.
47
Table 8.
WEE Soil Loss Estimates.®
Minimum Maximum
I ' Factor Values ( t ha"1 yr"1)
Field Length L (m)
Soil Loss ( t ha'1 yr"1)
180.2
0
0
670.3
155
34.8
Average
256.0
41
4.5
Standard
Deviation
56.0
43
4.6
aWEE so il loss rate estimates using a sample set generated from a random
125 m g rid (n = 80).
EROSION ESTIMATES (T /H A /Y R )
Figure 11. D is trib u tio n of so il loss rates fo r Carter watershed
WEE point method and 125 m g rid (n = 80).
using
48
The spatial v a r ia b ility of the WEE results is depicted in
12.
Figure
The highest estimated soil losses were located on or near h illto p s
(compare Figures 2 and 12).
There was also a group of high estimates
the central portion, near the old lakebed.
Another trend is located
in
in
the grassed area near the pond, where the high small grain equivalent
assigned to crested wheatgrass resulted in estimates of 0.0
t ha"1 yr"1.
USLE and WEE Erosion Rate Estimates Combined
Though not customarily done, USLE and WEE results were combined to
determine i f any spatial patterns existed in the estimated soil losses.
This was done to permit a rough comparison to the 137Cs results, which
include erosion from both water and wind, even though the models do not
estimate g u lly erosion. This allowed fo r the id e n tific a tio n of highly
erosive areas and th e ir spatial relationship to the location of the 137Cs
sample s ite s.
erosion
Combining the USLE and WEE watershed averages, a predicted
rate of. 9.0 t ha"1 yr"1, based on a range from 0 to 36.5 t ha"1 yr"1,
was computed.
The 81 point estimates were combined and depicted as a
frequency histogram (Figure 13).
The spatial v a ria b ility of the USLE/WEE combined results is shown
in Figure 14.
The diagram shows the combined resu lt from the two models
fo r each grid point, with the squares, triangles and c irc le s
the points were water erosion equaled zero fo r the USLE.
representing
The point that
was excluded from the WEE analysis was assumed to have
erosion equal to
0.0 t ha"1 yr"1 due to its location in the bottom of
the 4-feet deep
incised channel (excluding i t from both the USLE and
WEE).
These data
indicate very high so il loss rates along the eastern
margin because of
49
LEGEND:
SAMPLE POINT
WATERSHED
BOUNDARY
200
Meters
North
Figure 12. Spatial v a r ia b ility of so il loss rates ( t ha"1 yr"1) fo r Carter
watershed using WEE point method and 125 m random g rid .
Shaded areas
represent areas of high wind erosion. Excluded point (40) not shown.
both high flu v ia l erosion rates and high rates fo r
h illto p s .
This re s u lt shows how the 137Cs transect
located in high so il loss areas.
wind erosion on the
sample sites were
50
EROSION ESTIMATES (T/H A /Y R )
Figure 13. D istrib u tio n of to ta l soil loss rates fo r Carter watershed
using USLE/WEE point methods and 125 m grid (n = 81).
Cesium-137 Results
The results of the laboratory analysis fo r 137Cs a c tiv ity are depicted
in Figure 15.
The 137Cs areal a c tiv ity at each sample s ite
depth and is reproduced in Figure 15 as bar graphs
increment's 137Cs a c tiv ity .
The 137Cs a c tiv itie s
calculated by m ultiplying the mass concentrations
varied with
representing each
(pCi cm"3) shown were
of 137Cs a c tiv ity (pCi
g"1 by the increment's bulk density (g cm3).
Note that 3 of 4 pond sites
and sites 3C and SC did not reach a depth
of zero detection, causing
estimates of deposition rates to be
underestimated.
The to ta l sample
51
1 3 .0
6 a0
4 .0
8 .2
1 4 .2
1 6 .4
9 .8
LEGEND:
SAMPLE POINT
200
WATERSHED
Meters
North
BOUNDARY
Figure 14. Spatial v a r ia b ility of so il loss rates fo r Carter watershed
using USLE and WEE point methods.
Squares, tria n g le s and c irc le s
represent points where the USLE estimated erosion as 0.0 t ha'1 yr'1.
Shaded areas represent areas of high net erosion.
depth was
inadequate
fo r
these
samples.
Site
concentration (pCi cm'2) was lower than the control
even though
it s
depth p ro file
is
s im ila r
depositional s ite , i t was treated as an
in
3C's
to ta l
areal
site s average, so,
shape to
eroding s ite .
that of a
The changes in
a c tiv ity fo r the con trol, pond and SC site s
are important to notice as
they re fle c t much greater 137Cs a c tiv ity than
the transects and three
of
52
ACTIVITY (pCI cm ~3)
DEPTH (cm)
137C3
O
LEGEND;
15
T
TT = Trace detection of
7T1 / 7T2
)ND = No detection of
137
Cs
13^Cs
30 J
Figure 15.
Site areal 137Cs a c tiv itie s (pCi cm"3) fo r Carter watershed.
53
the fo u r
incised channel
s ite s
( s it e
SC is
a depo sition al
s ite
and
th e re fo re an exception)..
The diagram o f s it e 7P stands out in comparison
to the other pond s ite s .
It s p r o f ile is ty p ic a l o f an eroding s it e , which
is unusual because i t was located on the margin o f the pond.
In g eneral,
however, the
diagrams fo r d iffe r e n t landscape po sition s are s im ila r to
those in the
general model reproduced in Figure I .
The 137Cs sample s ite s were grouped and averaged, w ith the re s u lts
shown in Figure 16.
This graphical illu s t r a t io n shows the re la tio n s h ip
o f the topographic po sition s to the control samples by d iv id in g the
in to two sections.
graph
Those topographic po sition s with averages to
the l e f t
o f the average areal concentration o f the control s ite s , 11.10
pCi cm"2,
represent erosional s ite s , w hile those topographic po sition s w ith averages
to the r ig h t o f the lin e represent depositional s ite s .
concentration is marked on the bar by a t ic k mark in
The average areal
the middle portion
o f the bar, w ith the whole bar representing the range o f values.
im portant to po int out th a t several bars
range o f d is trib u tio n s
I t is
overlap, in d ic a tin g the wide
and great s p a tia l
v a r i a b ili t y
o f the 137Cs in
sediments.
The
equation
p r e c ip ita tio n
based
o f Arnolds
on a strong
c o rre la tio n
(1984) and Arnolds e t a l .
predicted the amount o f 137Cs deposition a t C a rte r.
concentration,
of
137Cs and annual
(1989) was used to
This p re d icts an areal
7 .9 pCi cm'2, 71 % o f the amount found in the
control
average o f th is study, 11.1 pCi cm"2.
Table
9
shows
the
re s u lts
c a lc u la tio n s fo r each s it e .
of
the
erosion
and
deposition
ra te
These ra te s characterized the s o il erosion
or deposition r e la t iv e to the control s ite s .
The fo u rth column o f the
54
DEPOSITION
EROSION
INCISED
CHANNELS
•
■------------ H
MIDSLOPES
1— I------------ 1
HILLTOPS
.------------- 1------------------
FOOTSLOPES
,
1
I
I
CONTROL
TILLED
CHANNELS
.------------------ 1--------------------------------
■
POND
__________I_____________ ,
I
(
)
12
6
pCi
18
24
cm
Figure 16. Range and average 137Cs areal concentrations by topographic
po sitio n .
table li s t s the 137Cs depletion (a percent) fo r each
was made only i f a s ite 's to ta l areal
control s ite average.
s ite .
This estimate
concentration was less than the
The average 137Cs a c tiv itie s used fo r the control
s ite 's plow (0 to 15 cm) and 16 to 20 cm layers were 11.03 and 0.07 pCi
cm"2, respectively, or a to ta l
increments collected from
the 0.07 pCi cm"2 of the
comparison
of 11.10 pCi cm"2.
Some o f the 16 to 20 cm
transect sites had s lig h tly greater 137Cs than
control s ite average.
This would
cause the
to re s u lt in a negative 137Cs depletion value, although these
resu lts were disregarded
follow ing the advice of De Jong et a l. (1983).
Total s o il loss or
is
gain
shown
in
the
f if th
and
sixth
columns.
55
Table 9. Cesium-137 Erosion and Deposition Rate Estim ates.
S ite
#
Topographic
Position
Bulk
Cesium
Density D epletion6
g cm"3
Total
Total
Soil (o r) Soil
Loss
Gain
g cm"2
S ite
Rate3
S ite
E rror
(+)
Mg ha"1 yr"1
IT l
H illto p
7T1/7T2
1T2
1.27
1.23
1.25
0.08
0.76
0.42
1.61
13.44
7.67
--—
--
4 .9
40.7
23.3
0 .6
3 .2
4.1
2T1
6T1
2T2
6T2
Midslope
1.32
1.27
1.19
1.32
0.40
0.61
0.36
0.65
7.78
10.92
5.77
12.56
—-—
--
23.6
33.1
17.5
38.1
3 .6
6.6
2.1
7.4
3T1
5T1
3T2
5T2
Footslope
1.32
1.37
1.30
1.30
0.17
0.32
0.22
3.29
6.11
3 .98
———
12.70
10.0
18.5
12.1
-3 8 .5
1.1
2.5
1.8
3.4
13.73
12.50
16.35
-4 1 .6
-3 7 .9
-4 9 .6
3.3
4.2
3 .7
51.5
34.2
26.5
9 .0
5.3
6.2
-232.1
-2 5 7.5
-2 3 7.5
32.2
24.5
21.2
41.1
5.9
Il
Il
Il
Il
Il
M
Il
Il
-
-
"
4T1
T ille d
Channel
4T2
SC
Il
IC
Il
U n tille d
Channel
Il
2C
3C
IP
4P
6P
7P
Il
Pond
Bottom
H
Il
Pond
Margin
1.39
1.35
1.23
1.40
1.40
1.40
1.30
1.44
1.33
1.40
-
—
-
-
-
-
0.81
0.54
0.42
——
—
17.00
11.27
8.74
— -
-
—
— —
0.82
— —
-
-
5.76
— —
-
—
32.50
36.00
33.25
“P o s itiv e s ite rates in d ic a te erosion and negative rates
d epo sition .
bCesium-137 loss as proportion as computed using equation # 8.
in d ic a te
The s ite erosion ra te s are shown in the next column.
shows the counting e rro r.
The la s t column
For erosional s ite s the e rro r was
expressed as a proportion o f the 137Cs depletio n and then put
loss equations (#8 and #9) shown in Chapter 2.
For
the e rro r was c a lc u la te d as a s o il gain in equation
T n itia T ly
in to the s o il
depositional s ite s ,
11 and then expressed
as a proportion o f the s it e 's s o il gain.
The re s u lts in Table 9 show several in te re s tin g in sig h ts about the
s p a tia l v a r i a b ili t y o f 137Cs a c t iv it y and erosion/deposition ra te s .
tran s ec t s it e w ith the g re a tes t erosion
located on a h illt o p
The
was s ite 7T1 (4 0 .7 t ha"1 y f 1) ,
in the steeper eastern portion o f the
However, two other h illt o p s ite s have suffered much low er
- - 4 .9 t ha"1 yr"1 fo r IT l and 23.3 t ha"1 yr"1 fo r 1T2
watershed.
erosion rates
- - in d ic a tin g th a t as
a group the h illto p s vary g re a tly in ra te o f erosion.
The midslopes have erosion ra te s th a t are less v a ria b le .
midslopes th a t were on the north ends o f the
lower erosion ra te s (2 3 .6 and 17.5 t
ends, 6T1 and 6T2, (33.1 and
note th a t the midslope
same
tra n s e c t
d ir e c t ly
The two
tra n s e c ts , 2T1 and 2T2, had
ha'1 yr"1) , than those on the southern
38.1 t ha"1 yr"1) .
I t is also in te re s tin g to
w ith the highest erosion ra te was located on the
u p h ill
from
a
foots!ope
s it e
(5T2)
th a t
experienced 137Cs enrichment.
This s it e
w hile the
showed a net gain o f 137Cs a c t iv it y , in d ic a tin g deposition,
other foots!opes (3T1, 5T1 and 3T2) had 137Cs losses, r e fle c tin g
erosional cases.
A possible explanation fo r th is d if fe r e n t r e s u lt
be the fa c t th a t 5T2's 137Cs a c t iv it y r e fle c ts sediment
tra n s p o rt.
could
It s
p o sitio n on a sho rt, steep slope could r e la te to
sediment tran s p o rt from
s ite 6T2, and possibly from 7T2, which was also
on the same tran sect and
57
had the highest h illt o p erosion ra te .
The
were located on slopes th a t were not as
the sediment has already been
has not y e t reached them
fa c t th a t the other foots!opes
steep could in d ic a te th at e ith e r
transported through those s ite s or th a t i t
(see Figures 2 and 7 ).
The t i l l e d channel s ite s , 4T1, 4T2 and SC, a ll have deposition
th a t
are closely grouped,
ranging from 37.9 to 49.6 t
standard deviation fo r th is group, 4 .9 t ha'1 yr"1, was
e n tire watershed.
This indicated th a t th is s ite would lik e ly
those
sampled.
underestimated.
Therefore,
it s
This is very
already the highest in the
ha'1 yr'1.
The
the lowest in the
The other ephemeral channel s it e ,
amount o f 137Cs a c t iv it y in the deepest sample
rates
SC, has a large
increment (see Figure 15).
have more 137Cs at depths below
deposition
ra te
is
almost c e rta in ly
s ig n ific a n t since SC's deposition ra te was
watershed.
The s ite with the highest erosion in the watershed (51.5 t ha"1 yr"1)
was 1C, located in the incised channel.
The other incised channel
2C and SC, have erosion rates of 34.2 and 26.5 t ha"1 yr"1,
S ite 3C's re s u lt is misleading because i t did not
However,
it
was
trea te d
as
an
concentration was much less than
erosional
re sp ec tiv e ly .
reach "zero" a c t iv it y .
s ite
the c o n tro l.
s ite s ,
because
it s
to ta l
The p o s s ib ility th a t
rece n tly deposited sediments have buried the n a tu ra lly deposited 137Cs
remote because the sample was collected from between large boulders
were placed in the
incised channel
bottom to prevent channel
Despite th is p ro te c tiv e measure th is channel was a c tiv e ly
is
th a t
scour.
eroding at an
average erosion ra te o f 37.4 t ha"1 yr"1.
The re s u lts fo r the pond s ite s indicated the same problem with sample
depth.
Referring to Figure 15, s ite s IP , 4P and 6P a ll in d icate th a t the
58
sample depth did not reach an increment w ith "zero" 137Cs a c t iv it y , which
suggests th a t pond 137Cs deposition was almost c e r ta in ly underestimated.
Pond s ite 7P, located on the pond margin, exh ib ited 137Cs a c t iv it y th a t was
c h a r a c te r is tic o f an
erosional s it e .
This s ite was located on the margin
o f the pond near
the dam, which was b u ilt in 1973 w ith excavated s o ils
from other areas.
It s 137Cs a c t iv it y and erosion ra te probably in d ic a te a
s ite th a t was
s o il
pedon).
average areal
modified during dam construction ( i . e . i t is not a natural
This
s ite
was elim inated
from the
c a lc u la tio n
o f the
concentration fo r the pond s ite s .
Using the erosion and deposition ra te s , i t was possible to
determine
averages o f erosion or deposition ra te s fo r each topographic
po sition
(Table 1 0 ).
ra te was
The topographic p o sitio n w ith the highest erosion
the incised channel and the lowest was the foots!opes.
deposition were 243.0 and
re s p e c tiv e ly .
midslopes
The
Mass rates o f
61.0 Mg yr"1 fo r the t i l l e d channels and pond,
highest mass ra te o f erosion was a ttrib u te d to the
a t 609.8
Mg yr"1,
topographic p o s itio n s ,
le v e l o f confidence
which also
21.7 ha.
had the
la rg e s t
area o f the
This group o f samples also had a high
and, considering the problems p e rta in in g to sample
depth fo r the channel and pond s ite s , is li k e l y the most accurate o f a ll
the
topographic
p o s itio n s .
This
is
re in fo rced
by i t s
low standard
d e v ia tio n (8 .0 t ha"1 yr"1) .
The mass ra te
estim ates
in dicated
a c tiv e ly
eroding midslopes and
h illt o p s , both erosional and depositional footslopes, s ig n ific a n t
o f 137Cs laden sediments in the ephemeral channel system and
d e p o s itio n in the small pond.
storage
substantial
The watershed net mass erosion ra te is
approxim ately 631 Mg yr"1, which is the sum o f the mass ra te s in Table 10.
59
Table 10.
Average Cesium-137 Erosion and Deposition Mass Rates by
Topographic P ositio n .
Topographic
Position
Sample
Size
Area
ha
H illto p s
Midslopes
Footslopes
T ille d
Channels
Incised
Channels
Pond
3
4
4
13.5
21.7
19.6
Erosion/ Standard Confidence Total
Deposition D eviation
Level
Erosion/
Rate8
Deposition
t ha"1 y r'1
%
Mg year"1
23.0
28.1
0.5
14.6
8 .0
22.7
44.4
78.9
34.0
310.5
609.8
9 .8
4
5.65
-4 3 .0
4.9
95.5
-243.0
4
4
0.15
0.25
37.4
-243.9
10.5
9.3
65.8
72.9
5.6
-6 1 .0
“Erosion and deposition ra te s , negative rates in d ic a te depo sition .
Confidence values from nonparametric s t a t is t ic a l a n a ly s is .
60
CHAPTER 4
DISCUSSION
USLE and WEE Soil Loss Estimates
The USLE and WEE are the most widely used models fo r evaluating
erosion hazard in Montana.
The Soil
Conservation Service applies the
USLE and a version o f the WEE, the WEQ, to determine q u a lific a tio n fo r
and maintenance in the Conservation Reserve Program (CRP) program.
SCS
estimates fo r th is watershed are 0.7 and 10.10 t ha"1 yr"1 fo r the USLE and
WEQ, re s p e c tiv e ly .
The underlying ob jective of SCS a p p lic a tio n 's
these models is to determine average conditions.
This study sought
estimate average conditions as well as the spatial v a r ia b ilit y of
losses by using s p a tia lly v a riab le fa c to r values (Tables 11 and
in to the USLE while the SCS's application incorporated only one
to
so il
12).
The study incorporated s p a tia lly v a ria b le topographic fa c to r
the e n tire watershed.
of
values
value fo r
Both studies used the same R, K and P factors and
the C facto rs were s im ila r (0.234, th is study, and 0.242,
SCS) with the
exception o f six points in the grass area fo r which a d iffe r e n t value was
used (0 .0 5 3 ).
d iffe r e n t .
The topographic facto rs
of the two
The SCS's estimates r e fle c t a slope length
times the longest slope length measured in th is
gradient
studies
s lig h tly la rg e r than the
sm allest
study
are very
g reater than ten
and
gradient measured
a slope
in
th is
61
Table 11.
Comparison o f USLE Factor Estimates Used by Author and
USDA-SCS.
Factor
By Author
R a in fa ll
-e r o s iv ity , R
476.6
Soil
c r e d ib i lit y , K
By USDA-SCS
476.6
0.049
Slope le n g th , L
0.049
1,600 m
19.0 m to 92.2 m
0 .4 %
0.30 % to 4.17 %
Slope g ra d ie n t, S
0.12
0.01 to 0.56
LS
Cover
management, C
0.242
0.2344 or 0.053
Supporting
p ra c tic e s , P
1.0
1.0
Soil lo ss , A
4.47 t ha"1 yr"1
0.67 t ha"1 y r'1
study.
N eith er o f these values is close to watershed
from a series o f 81 points and are th e re fo re not
study watershed.
the
USLE g re a tly
The accuracy of the topographic
a ffe c ts
the
model's
a b il it y
re p res e n ta tiv e o f the
fa c to rs ' estim ation fo r
to
v a ria b le erosion rates in the watershed (Wischmeier,
Smith, 1978; Wilson, 1986b).
and deposition areas.
ap p lic a tio n s
are very
This study also
The average re s u lts
averages computed
p re d ic t
s p a tia lly
1976; Wischmeier and
distingu ishes net erosion
produced from these two USLE
d iffe r e n t and the point re s u lts produced in th is
study were highly v a ria b le .
62
Table 12.
Comparison o f WEE Factor Estimates
Factor Estimates Used by USDA-SCS.
Factor
Used
By Author
Soil
c r e d ib ility ,! '
107.6 t ha"1 yr"1
0 .4 9 , 0.66 and 0.90
1.00
0.90
0.90
0 to 155 m
142 m
C lim ate, C
F ie ld le n g th , L
Vegetation
Residue, V
By USDA-SCSa
180.2 to 670.3 t ha"1 yr"1
Soil ridge
roughness, K
by Author and WEQ
1200 kg ha"1
130 to 3025 kg ha"1
10.10 t ha"1 yr"1
4 .7 t ha"1 yr"1
Soil lo ss, E
aWEQ estim ates made by SCS f ie ld personnel, Fort Benton, MT, fo r Jackson
Farm fo r use in Conservation Reserve Program (CRP) co n tra c ts.
The
WEE
a p p lic a tio n
method
a p p lic a tio n
in
th is
study
d iffe r s
from
the
SCS
WEQ
because i t computes wind erosion by periods where the SCS
u t iliz e s sin g le fa c to r estimates th a t are n e ith e r tem porally nor
s p a tia lly v a ria b le (See Table 12).
study is
The mean o f the I ' values used in th is
more than twice as large as the SCS value.
This study also
adjusted I ' fo r periods follow ing c u ltiv a tio n , and also incorporates the
I s, knoll
varied
c r e d ib i lit y , fa c to r at fiv e p o in ts .
a t each point according to s t r ip
Eighty-one L facto rs were
fetch le n g th .
Many of the L
estim ates were less than the SCS L estim ate o f 142 m, accounting fo r many
low wind erosion estimates and a low average erosion estim ate (4 .7 t ha"1
yr"1) in comparison to the SCS (WEQ) estim ate (10.1 t ha1 y r
1) .
The C
63
fa c to rs used are the same but th is study determines period s o il losses as
a function o f the
year c ro p /fa llo w
maximum
as a
at
c y cle .
The SCS estim ate o f K was held constant a t it s
possible value, whereas, th is study c a lc u la te d period K values
function o f t i l l a g e operations.
varied
SCS.
d is tr ib u tio n o f erosive wind energy (EWE) over the two
The V fa c to r in th is study also
by period , as opposed to the s in g le , low estim ate p referred by
However, the SCS a p p lic a tio n suggested th a t the watershed is eroding
10.1 t ha"1 yr"1 w hile th is study found an average o f only 4 .5 t ha"1
yr"1.
Figure
confirmed by
12 shows th a t these re s u lts
also were hig h ly v a ria b le ,
the standard d eviatio n o f the WEE average (4 .6 t ha"1 yr"1j .
This is s im ila r to the v a r i a b ili t y o f th is study's USLE
ha"1 yr"1) , confirm ing the fa c t th a t modeled wind and
estim ate (3 .6 t
water erosion ra te s
in the watershed are highly v a ria b le .
It
is possible to make several general observations based on these
r e s u lts .
The SCS appears to apply these models a t too coarse a scale
(a t
in
le a s t
landscapes with
hummocky r e l i e f ) .
They also have used
n o n-rep resentative L and S fa c to rs in the USLE; where the L is too long
and the S is
near the gentle end o f the measured d is tr ib u tio n .
This
study d e lin e a ted erosional and depositional zones, w h ile the SCS made
attempt to do th is and consequently characterizes the e n tir e
as e ro s iv e .
This is obviously not tru e fo r the ephemeral
pond where deposition was indicated by the 137Cs method.
th a t the SCS's estim ates are not a d d itiv e because they
w ith a p a r tic u la r po int in mind and because o f the
erosion and deposition areas.
no
watershed
channel and
This suggests
were not produced
f a ilu r e to d efin e net
64
The model re su lts show a great deal o f v a r ia b ilit y (see Figures
12 and 14).
The model re su lts are combined in Figure
p o in t's
erosion.
net
depositional
This
zones were
step
was
id e n tifie d
based
14 to depict each
on the
by a rb itra ry
10,
fa c t
rules
th a t
fo r the
the
USLE
(w a te r), but th is was not done (nor could i t be done) fo r the WEE (w in d ).
This alludes to the question at issue, "How accurately do the
USLE and
WEE estim ate so il loss and sp atial v a r ia b ilit y ?".
V a lid a tio n o f USLE and WEE Soil Loss Estimates
A p o s s ib ility exists th a t the models or the 137Cs method, or both
are
simply wrong, at le a s t fo r estim ating so il re d is trib u tio n processes
in
the northern Great Plains and Montana.
v a lid it y
o f model
In the past, an assessment
re s u lts was d i f f i c u l t
a v a ila b le to answer th is (above) question.
because there were no
However in the la s t
years, 137Cs has offered an opportunity here and elsewhere to
data to address the question and assess s p a tia l
(De Jong e t
a l.,
1983;
Arnalds,
1984;
of the
several
a tta in f ie ld
v a r ia b ilit y in soil loss
Pennock
However, the app licatio n of th is method in th is
data
and De Jong,
1987).
study revealed two major
lim ita tio n s which warrant fu rth e r explanation.
The in a b ilit y o f the s t r a t if ie d
sampling scheme used in th is
and
other studies to capture the spatial v a r ia b ilit y of 137Cs a c tiv ity is one
lim ita tio n .
Although a la rg e r number o f data
points might improve the
confidence le v e ls of the erosion estimates d iffe r e n t landscape
inclusion o f samples collected from
u n its , the
the gently sloping, western portion
o f the watershed may make matters worse.
The transect samples used were
concentrated
portion
in
the
steeper,
eastern
of
the
watershed
and
65
e x tra p o la tin g t h e ir average 137Cs
whole watershed might have
occurred.
Even
if
in dicated more net erosion than has r e a lly
both
areas
p ro p o rtio n a lly , the s t r a t i f i e d
major lim it a t io n because i t
v a r i a b i l i t y o f 137Cs
approach could be
flow in to the
a c t iv it ie s to the eroding portion o f the
of
the
watershed
were
sampled
sample scheme employed would s t i l l
be a
might not capture a ll aspects o f the s p a tia l
a c t iv it y and s o il e ro s io n /d e p o s itio n .
An a lte r n a tiv e
to incorporate divergence and convergence o f overland
landscape un its such as th a t used by Pennock and De Jong
(1 9 87 ).
Temporal
v a r i a b ili t y
is
a second lim ita tio n
and may be ju s t
d i f f i c u l t to account fo r in s o il samples analyzed fo r 137Cs a c t iv it y .
frequency
and
magnitude
of
wind
and
water
erosion
im portant ro le in the dynamics o f s o il lo ss .
between October and March in th is region .
events
play
as
The
an
Most wind erosion occurs
Water erosion has freq u en t,
small magnitude events moving sediments to foots!opes and the ephemeral
channel system, and, in fre q u e n t, la rg e magnitude events moving
from these
s ite s
through
the
incised
channel
to
the
pond.
sediments
It
is
im portant to know the stage o f s o il re d is tr ib u tio n represented by the 137Cs
a c t iv it y .
This a c t iv it y depends on the recent h is to ry o f
f lu v ia l erosion
events ( i . e . whether or not a la rg e magnitude event has
occurred in the
short-term p a s t).
Th erefore, careful designation o f
landscape u n its is c r i t i c a l to the accurate
s o il g a in s/lo s s e s .
This study used sample
sample locations and
q u a n tific a tio n o f 137Cs and
lo catio ns based on equal areas
o f slope delin eated by f i e l d pacing and s o il samples c o lle c te d over a four
week period in mid-summer.
adequately c h a ra c te rize the
This sample d e lin e a tio n method f a ile d
to
tem porally v a ria b le dimensions o f 137Cs and
66
s o il re d is trib u tio n on the
Jackson Farm.
Despite these problems some data were collected th a t can be used
qu an tify and in te rp re t so il erosion/deposition rates a t s ite s from
so il
samples
were
c o lle c te d .
A s ite
by
s ite
comparison
which
of
so il
erosion/deposition rates estimated by the models and 137Cs approaches
performed to evaluate the model re s u lts .
to
was
The individual s ite s ' predicted
and measured rates were analyzed with lin e a r regression.
Predicted rates
were computed by adding together USLE and WEE estimates fo r each of the
eroding 137Cs sample points (Table 13).
The 137Cs depositional s ite s were
not used because the models do not apply to them.
A lin e a r regression of
ra te estimates fo r the 10 remaining s ite s yielded an R2 o f 0.07.
I f the
three s ite s fo r which the USLE did not apply ( i . e . , based on the h illto p s ,
channels,
and/or concave slope exclusion ru les)
are removed from the
sample set the re s u lt improves only m arginally (R2 =
s c a tte rp lo t is shown in Figure 17.
These re su lts
not work fo r estim ating to ta l erosion in the
0 .1 4 , n = 7 ).
A
in d ic a te the models may
C arter watershed.
Soil Erosion/Deposition Rates In ferre d from Cesium-137 Gains/Losses
Based on the comparison of measured versus predicted ra te s , the
USLE and WEE do not appear to work very well
However,
it
p a r tic u la r ly
appears
with
th a t
respect
the
to
137Cs method
understanding
is
in the study watershed.
not
very
sp a tia l
good e ith e r,
v a r ia b ilit y
and
estim ating a sediment budget unless many more data points are collected
and analyzed (le v e ls of confidence range from 34.0 to 95.5 % by landscape
u n it).
67
Table 13.
Site by Site Comparison of Model and 137Cs Erosion/Deposition
Rate Estimates.
S ite
USLE
IT l
2T1
3T1
4T1
5T1
6T1
7T1
1T2
2T2
3T2
4T2
5T2
6T2
0.0
10.4
15.0
0.0
45.6
9.3
0.0
0.0
13.2
1.9
0.0
11.5
5.1
Models
Total
WEE
t ha"1 yr"1
10.2
9.9
9.2
8.5
9.0
9.7
10.1
10.4
12.7
12.7
12.7
5.1
4.5
10.2*
20.3
24.2
8.5
44.6
19.0
10.I a
10.4*
25.9
14.6
12.7
16.6
9.6
137Cs
4.9
23.6
10.0
-41.6"
18.5
33.1
40.7
23.2
17.5
12.1
-37.9"
-38.5"
38.1
aSite predicted value attained from WEE estimate only (shown as empty
symbols in Figure 17).
6Depositional s ite excluded from regression analysis.
M EASUR ED ER O SIO N (T /H A /Y R )
Figure 17. S catterplot diagram of measured and predicted values generated
fo r s ite by s ite comparison of 137Cs sample sites (R2 = 0.07, n = 10).
Empty symbols are sites with WEE estimate only, with these sites removed,
R2 = 0.14 (n = 7).
68
However, i t is possible to compute a f i r s t estim ate o f the -watershed
sediment budget (Table 14) with the data th a t were collected,.
..Several
problems, including the Tow (le s s than 60 %) le v e ls o f confidence fo r
of
the
landscape
u n its ,
the
i n a b ili t y
to
separate
wind
and
water
processes, overlapping erosion and depo sition areas, the inadequate
sample depth and the fa c t th a t the 137Cs a c t iv it ie s
( p a r tic u la r ly the t i l l e d
two
pond
are time-dependent
channels.) in d ic a te the need to in te r p r e t such
re s u lts c a u tio u s ly .
The 137Cs a c t iv it ie s in dicated a net mass erosion r a te fo r the
portions
o f the watershed
(about
54 ha)
o f 9336 Mg y r'1..
The
deposition ra te a ttr ib u ta b le to f lu v ia l deposition (304 Mg yr"1)
in te rp re te d to represent f lu v ia l erosion.
erosion ra te from the to ta l erosion ra te
Subtracting th is
eroding
mass
could be
mass water
y ie ld ed a mass erosion estim ate
fo r wind (a net value not counting deposition) o f 632 Mg yr"1.
The mass
wind erosion ra te was used to . compute the net wind erosion ra te o f 10.4
t ha'1 yr"1 and the mass
erosion ra te o f 5 .0
water erosion ra te was used to compute a net water
t ha"1 yr"1 fo r a to ta l o f 15.4 t ha"1 yr"1.
The USLE and WEE estimates o f s o il loss
predicted
a net
s o il
(exclu sive o f several
loss
of
9 t
(4 .5 t ha"1 yr"1 fo r
ha"! yr"1 in
the
study
b o th ),
watershed
USLE points representing deposition and one WEE
po in t where wind erosion does not a p p ly ).
This r e s u lt suggests th a t the
models predicted only 58 % o f the t o ta l erosion estim ated w ith the 137Cs
method.
A comparison o f the net f lu v ia l erosion ra te (5 .0 t ha"1 yr"1)
and
the USLE watershed average showed th a t the USLE predicted 90 % o f
the
measured water erosion even though th is USLE average was known to
because i t did not include channel erosion.
The WEE estim ate
be low
(4 .5 t
ha"1
69
Table 14.
Summary of Watershed Average 137Cs Erosion and Deposition
Mass Estimates.
Topographic
P osition
Area
ha
Mass
Estimates3
Mg y r'1
13.5
21.7
19.6
0.15
311
610
10
6
Erosion
( f lu v i a l and a eo lian )
H illto p s
Midslopes
Footslopes
Incised Channel
Measured Erosion Mass Estimate
936
T ille d Channels
Pond
-.243
-61
Deposition
( f lu v ia l and aeolian )
5.65
0.25
Measured Deposition Mass Estimate
-304
aMass estim ates were c a lcu lated by m u ltip ly in g the topographic p o s itio n s '
average e rosion/deposition ra te by i t s ' area.
Negative values r e f le c t
dep o sitio n .
yr"1) was much less than the watershed net wind erosion ra te (1 0 .4 t ha"1
yr"1) and accounted fo r only 44 % o f the
WEQ estim ate (10.1 t
measured wind erosion.
The SCS's
ha"1 yr"1) is s im ila r to the net wind erosion r a te ,
but d i f f i c u l t to explain because they did not use re p re s e n ta tiv e s o il
c r e d ib i lit y
or f i e l d
length fa c to r estim ates.
T h erefore,
if
the 137Cs
re s u lts were t r u ly in d ic a tiv e o f the s o il re d is tr ib u tio n processes in the
study watershed, then the USLE has performed much b e tte r than the WEE in
p re d ic tin g s p a t ia lly v a ria b le erosion ra te s .
However, we probably cannot
t o t a l l y r e je c t the WEE re s u lts unless more 137Cs data are gathered in the
hope o f reducing the variances.
70
Conclusions
Both the USLE and WEE model and 137Cs methods indicated a substantial
erosion hazard and a great deal of sp a tia l v a r ia b ilit y .
is
p e rfe ct
as
applied
here
v a r ia b ilit y o f the hazard.
but
s u ffic ie n t
to
N either approach
show the
extent
and
C le a rly , the re su lts in d ic a te th a t there has
been s ig n ific a n t f lu v ia l and aeolian erosion of h illto p s , midslopes, some
foots!opes and an incised channel over 33
deposition on one footslope, in the pond
years, and s ig n ific a n t flu v ia l
and in the ephemeral channel
system over 14 years.
Despite
problems
associated with
capturing
temporal
and sp atial
v a r ia b ilit y , the 137Cs data suggests th a t the USLE point method used
much more e ffe c tiv e than the point WEE method used.
Measured
flu v ia l erosion rates were 10.4 and 5.0 t ha'1 yr"1,
the modeled erosion rates were 4.5 t ha"1 yr"1 fo r
was
aeolian and
re sp ec tiv e ly ; while
both wind and water.
These re su lts have important consequences fo r crop p ro d u c tiv ity
now and in the fu tu re (despite low p re c is io n ).
At present, 50
years of
erosion and deposition has occurred, ju s t one of many facto rs influencing
and explaining current crop p ro d u c tiv ity .
For the fu tu re , there is s t i l l
a need to compute s p a tia lly v a riab le s o il erosion rates in order to b e tte r
understand so il
scales.
erosion/crop p ro d u c tiv ity
re la tio n s h ip s
at watershed
REFERENCES CITED
72
References C ited
Anonymous., 1989, "Known Nuclear Tests: 1945 to December 31,
B u lle tin o f Atomic S c ie n tis ts , 4 5 (1 ), A p r il, 1989, 48.
1988",
Armbrust, D .V ., and Lyles, L ., 1985, "Equivalent Wind Erosion P rotection
from Selected Growing Crops", Agronomy Journal. 77, 703-707.
Arnolds, 0 . G. , 1984, "Radiocesium in Montana S oils and A pplications fo r
S oil Erosion Measurement", Unpublished M.Sc. Thesis, Montana S tate
U n iv e rs ity , 58p.
Arnalds, O .G ., C u ts h a ll, N .H ., and N ielsen, G .A ., 1989, "Cesium-137 In
Montana S o ils " . Health Physics. 5 7 (6 ), December, 1989, 955-958.
Bondy, £ . , Lyles, L.., and Hayes, W.A., 1980, "Computing Soil Erosion by
Periods Using Wind Energy D is trib u tio n " , Journal o f Soil and Water
Conservation, 3 5 (4 ), 173-176.
Brown, R .B ., C u ts h a ll, N .H ., and K lin g , E .F ., 1981a, " A g ric u ltu ra l Erosion
In dicated by Cesium-137 R e d is trib u tio n : I , Levels and D is trib u tio n o f
Cs-137 A c tiv ty in S o ils " , Soil Science Society o f America Journal, 45,
1184-1190.
Brown, R .B ., K lin g , E .F ., and C u ts h a ll, N .H ., 1981b, " A g ric u ltu ra l Erosion
In dicated by Cesium-137 R ed is trib u tio n : I I , Estimates Of Erosion Rates",
Soil Science Society o f America Journal. 45, 1191-1197.
Campbell, B .L ., Loughran, R .J ., and E l l i o t t , G .L ., 1982, "Cesium -137 as
an In d ic a to r o f Geomorphic Processes in a Drainage Basin System",
A u s tra lia n Geographical S tudies, 20, 49-64.
C h e p il, W .S., Siddoway, F .H ., and Armbrust, D .V ., 1962, "C lim atic Factor
fo r Estim ating Wind E r o d ib ility o f Farm F ie ld s ", Journal o f Soil and Water
Conservation, 1 7 (4 ), 162-165.
Coleman, D. E ., 1982, " Id e n tific a tio n o f P r io r it y Management Areas in the
Avon R iver Watershed", Upper Thames R iver Conserv. A u th ., London, O ntario.
Compton, R .R ., 1962, "Manual o f F ie ld Geology", John W iley and Sons, In c .,
New York, NY, 162p.
73
C utshalI , N .H ., and Larsen, 1. 1 . , 1980., "Peak-By-Peak C orrection o f G e(Li)
Gamma-Ray Spectra fo r Photopeaks from Background", Nuclear Instruments and
Methods. 178, 233-236.
De Jong., £ . ,
V illa r ., J i,,
and Bettany., J.R . ,
1982,
"P relim inary
In v e s tig a tio n s on the Use o f Cesium-137 to Estimate Erosion in
Saskatchewan", Canadian Journal o f Soil Science. 62, 673-683.
De JoPg, E ., Begg, C.B ;M ., and Kachanoski, R .G ., 1983, "Estimates Of Soil
Erosion and Deposition fo r Some Saskatchewan S o ils , Canadian Journal o f
Soil Science, 63, 607-617.
Delaune, R .D ., Patrick,, VLH. J r . , Buresh, R ..J., 1978, "Sedimentation Rates
Determined by Cs-137 Dating in a Rapidly Accreting S a lt Marsh", N ature,
275, 532-533.
D ibb, J . , 1989, "S c ie n tis ts Discover Problems Using Chernobyl F a llo u t fo r
Dating the Greenland Ice Sheet", Earth in Space. 2 ( 3 ) , November, 1989,
8 -9 .
Dyer, E .B ., 1977, "Use o f the Universal Soil Loss Equation: A River Basin
Experience", In "Soil Erosion: P redictio n and C o n tro l", Foster, G. R .,
e d ., S oil Conservation Society o f America, Ankeny, Iowa, 292-297.
Fo ster, G .R ., McKool, D .K ., Renard, K .G ., and Moldenhauer, W.C., 1981,
"Conversion o f the Universal Soil Loss Equation to SI M etric U n its ",
Journal o f Soil and Water Conservation, 3 6 (6 ), 355-359.
Foster, G .R ., and Wischmeier, W.H., 1974, "Evaluating Ir r e g u la r Slopes fo r
S oil Loss P re d ic tio n ", Transactions o f the American Society
of
A g ric u ltu ra l Engineers. 1 7 (2 ), 305-309.
G re e n v ille , T .N .E ., 1967, "Spline Functions, In te rp o la tio n , and Numerical
Quadrature", In "Mathematical Methods fo r D ig ita l Computers", Ralston, A .,
and W olf, H. S ., e d s ., John W iley, New York, NY, 156-168.
G r if f i n , M .L ., Beasley, D .B ., F le tc h e r, J . J . , and Fo ster, G .R ., 1988,
"Estim ating Soil Loss on Topographically Nonuniform F ie ld and Farm U n its ",
Journal o f Soil and Water Conservation, 4 3 (4 ), 326-331.
74
H i l l i a r d , R .L ., 1977, "The New York Erosion and Sediment In vento ry", In
"Soil Erosion: P red ictio n and C o n tro l" , Foster, 6. R. , e d ., S o il
Conservation Society o f America, Ankeny, Iowa, 273-276.
Kachanoski, R .G ., 1987, "Comparison o f Measured Soil 137-Cesium Losses and
Erosion Rates", Canadian Journal o f Soil Science. 67, 199-203.
Lance, J .C ., M cIntyre, S .C ., Nancy, J .W ., and Rousseva=, S .S ., 1986,
"Measuring Sediment Movement a t low Erosion Rates Using Cesium-137", Soil
Science Society o f America Journal. 50, 1303-1309.
Larsen, I . L . , and CutshalI , N .H ., 1981, "D ire c t Determ ination o f 7Be in
Sediments", Earth and P lanetary Science L e tte rs . 54, 379-384.
Lowrance, R ., M cIntyre, S ., la n c e , C ., 1988, "Erosion and Deposition in
a F ie ld /F o re s t System Estimated Using Cesium-137 A c t iv it y " , Journal o f
Soil and Water Conservation. 4 3 (2 ), 195-199.
McCall an, M .E ., O'Leary, B.M ., and Rose, C.W., 1980, "R e d istrib u tio n o f
Cesium-137 by Erosion and Deposition on an A u s tra lia n S o il" , A u stralian
Journal o f S oil Research. 18, 119-128.
McHenry, J .R ., R itc h ie , J .C ., and G i l l , A .C ., /1 9 7 3 , "Accumulation Of
F a llo u t Cesium-137 in S o ils and Sediments in Selected Watersheds", Water
Resources Research, 9 ( 3 ) , 676-686.
McHenry, J .R ., and R itc h ie , J .C ., 1977, "Estim ating F ie ld Erosion Losses
from F a llo u t Cesium-137 Measurements", In "Erosion and S olid M atter
Transport in Inland Waters. Proc. Symp.", J u ly , 1977. IAHS P ub lication
#122, 26-32.
M cIntyre, S .C ., Lance, J .C ., Campbell, B .L ., and M il le r , R .L ., 1987,
"Using Cesium-137 to Estimate Soil Erosion on a C learcut H ills id e " ,
Journal o f Soil and Water Conservation. 4 2 (2 ), 117-120.
Pennington. W., Cambray, R .S ., Eakins, J .D ., and Harkness, D .D ., 1976,
"Radionuclide Dating o f Recent Sediments o f Blelham Tarn", Freshwater
B iology. 6, 317-331.
Pennock, D .J ., and De Jong, E ., 1987, "The.Influence o f Slope Curvature
on Soil Erosion and Deposition in Hummock T e rra in " , S oil Science, 1 4 4 (3 ),
209-217.
75
R itc h ie , J .C ., Spraberry, J .A ., and McHenry, J .R ., 1974, "Estim ating Soil
Erosion from the R e d is trib u tio n o f FaTiout Cesium-137", Soil Science
Society o f America Proceedings. 38, 137-139.
R itc h ie , J .C ., Hawks, P .H ., and McHenry, J .R ., 1975, "Deposition Rates in
V alleys Determined Using F a llo u t Cesium-137", Geological Society o f
America B u lle t in . 86, 1128-1130.
R itc h ie , J .C ., and McHenry, J .R ., 1975,, "F a llo u t Cesiumr 137: A Tool in
Conservation Research", Journal o f Soil and Water Conservation. 30,
283-286.
Rogowski, A .S ., and Tamura, T ., 1970, "Erosional Behavior o f Cesium- 137",
Health Physics. 18, 467-477.
Skidmore, E .L ., and Woodruff. N .P ., 1968, "Wind Erosion Forces in the
United States and T h e ir Use in P redictin g Soil Loss", A g ric u ltu re Handbook
#346,
A g ric u ltu re
Research S ervice,
United
States
Dpartment
of
A g ric u ltu re , Washington, D. C.
Skidmore, E .L ., 1988, "Wind Erosion" In "Soil Erosion Research Methods",
L a l, e d ., Soil and Water Conservation S ociety, Ankey, Iowa, 244p.
S n e ll, E .A ., 1984, "A Manual fo r Regional Targeting o f A g ric u ltu ra l Soil
Erosion and Sediment Loading to Streams", Evironment Canada, Environment
Conservation S ervice, Lands D ir . Working Paper # 36, Ottowa, O n ta ria.
S n e ll, E .A ., 1985, "Regional Targeting o f P o ten tial Soil Erosion and
Nonpoint-source Sediment Loading", Journal o f Soil and Water Conservation,
40, 521-524.
Trim ble, S.W ., 1974, "Man-induced Soil Erosion on the Southern Piedmont,
1700-1970", Soil Conservation Society o f America, Ankeny, Iowa.
Trim ble, S.W ., 1977, "The Fallacy o f Stream E quilibrium in Contemporary
Denudation S tudies", American Journal o f Science. 277, 876-887.
Trim ble, S.W ., 1983, "A Sediment Budget fo r Coon Creek, the D r iftle s s
Area, Wisconsin, 1852-1977", American Journal o f Science, 283, 454-474.
76
United States Department o f A g ric u ltu re , S oil Conservation S ervice, 1988,
"1988 National Agronomy Manual", Part 502, "P redicting Wind Erosion", U.S.
Government P rin tin g O ffic e , Washington D. C.
United States Department o f A g ric u ltu re , Soil Conservation S ervice, 1988,
"1988 National Agronomy Manual", Montana Supplement, U.S. Government
P rin tin g O ffic e , Washington D. C.
Van V l i e t , L .J .P ., and W all, G .J ., 1979, "Comparison o f Predicted and
Measured Sheet and R ill Erosion Losses in Southern O n tario ", Canadian
Journal o f Soil Science. 59, 211-213.
Van V l i e t , L .J .P . and W all, G .J ., 1981, "Soil Erosion Losses from Winter
Runoff in Southern O n tario ", Canadian Journal o f S oil Science, 61,
451-454.
W a ll, G .J ., Van V l i e t , L .J .P ., and Dickinson, W .T., 1982, "A g ricu ltu re and
Water Q u a lity in the Canadian Great Lakes: 2, F lu v ia l Sediment", Journal
o f Environmental Q u a lity . 11, 482-486.
W illiam s , J .R ., and Berndt, H .D ., 1977, "Determining the Universal Loss
Equation's Length-slope Factor fo r Watersheds", In "Soil Erosion:
P red ictio n and C o n tro l", Foster, G. R ., e d ., Soil Conservation Society o f
America, Ankeny, Iowa, 217-225.
W illiam s, J .R ., Jones, C .A ., and Dyke, P .T ., 1984, "A Modeling Approach
to Determining the R elationship between Erosion and P ro d u c tiv ity ",
Transactions o f the American Society o f A g ric u ltu ra l Engineers, 27, 129144.
Wilson, J .P .,
1986a, "The Use o f S t a t is t ic a l Models to Document
Environmental Change in the Lake Simcoe-Couchiching Basin", Unpublished
Ph.D. D is s e rta tio n , Department o f Geography, U n iv e rs ity o f Toronto, 358p.
Wilson, J .P ., 1986b, "Estim ating the Topographic Factor in the Universal
S oil
Loss Equation fo r Watersheds",
Journal
o f Soil
and Water
Conservation. 4 1 (3 ), 179-184.
Wilson, J .P ., 1989, "Soil Erosion from A g ric u ltu ra l Land in the Lake
Simcoe-Couchiching Basin, 1800-1981", Canadian Journal o f Soil Science,
69, 137-151.
77
Wischmeier, W.H., and Smith, D .D ., 1978, "P redicting R a in fa ll Erosion
Losses- A Guide to Conservation Planning", A g ric u ltu re Handbook #537,
United States Department o f A g ric u ltu re , Washington, D. C ..
Wise, S .M ., 1980, "Caesium-137 and Lead-210: A reveiw o f the Techniques
and Some A pplications in Geomorphology", In "Timescales in Geomorphology",
C u llin g fo rd , R .A ., Davidson, D .A ., and Lewin, J . , e d s ., John Wiley & Sons
L t d ., London, England.
Woodruff, N .P ., and Siddoway, F .H ., 1965, A Wind Erosion Equation", Soil
Science Society Proceedings, 29, 602-609.
78
APPENDICES
APPENDIX A
EROSION MODEL RESULTS
80
Table 15.
USLE Factor and Soil Loss Point Estimates.®
Point
Slope
Gradient
Slope
Length
#
%
m
IF
IG
2M
2N
20
2P
3E
3F
3G
4N
40
4P
SE
5F
SG
SH
SM
SN
60
SP
7B
7C
70
7E
7F
7G
8J
8K
8L
SM
SN
80
8P
9A
9B
9C
90
9E
9F
9G
9H
91
10.7
6.7
0.2
1.7
16.0
8.0
0.8
6.7
7.3
0.0
0.0
0.0
0.9
1.8
7.3
4.3
0.3
2.4
7.3
4.4
0.3
0.2
2.1
0.0
5.7
8.0
2.5
1.4
1.1
2.8
1.7
6.3
6.7
0.7
3.6
1.8
3.1
1.3
4.7
0.8
7.3
10.0
105
120
0
118
55
26
33
43
50
10
0
18
33
60
19
50
25
13
135
53
45
2
80
0
24
90
80
108
20
123
83
38
63
13
123
213
55
153
168
190
60
28
USLE
USLE
C Valueb S Valuec
Exclusion
Ruled
Erosion
Estimate6
t ha"1 yr"1
0.053
0.053
0.234
0.234
0.053
0.053
0.234
0.234
0.234
0.234
0.053
0.234
0.234
0.234
0.053
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
1.295
0.661
0.074
0.161
2.455
0.844
0.105
0.661
0.743
0.065
0.065
0.065
0.111
0.168
0.743
0.380
0.079
0.211
0.743
0.391
0.079
0.074
0.189
0.065
0.535
0.844
0.219
0.141
0.123
0.243
0.161
0.609
0.661
0.100
0.313
0.168
0.268
0.135
0.422
0.105
0.743
1.169
0
0
I
0
0
0
0
0
0
I
I
I
2
0
0
0
0
I
0
0
0
I
0
2
0
0
0
2
0
0
0
0
0
I
0
0
0
0
0
3
0
0
5.20
2.84
N/A
1.88
7.14
1.69
0.74
7.50
9.09
N/A
N/A
N/A
N/A
1.60
1.27
4.01
0.53
N/A
14.94
4.21
0.59
N/A
1.96
N/A
4.54
13.87
2.27
N/A
0.84
2.87
1.69
6.50
9.08
N/A
4.72
2.34
2.93
1.70
7.22
N/A
9.96
10.71
81
Table 15.
USLE Factor and Soil Loss Point Estimates (Continued).a
Point
Slope
Gradient
Slope
Length
#
%
m
IOJ
IOK
IOL
IOM
ION
100
IOP
10Q
IlB
IlC
IlD
HE
HF
HG
IlH
111
12K
12L
12M
12N
120
12P
12Q
13C
130
13E
13F
13G
13H
14K
14L
14M
14N
140
14P
1.4
5.2
2.7
0.4
0.0
3.1
0.8
5.3
1.5
3.6
1.9
1.9
4.0
5.5
6.5
10.0
5.0
1.9
2.0
LI
3.6
8.0
0.0
3.1
3.0
1.4
1.4
1.2
1.2
1.6
3.1
2.0
1.3
1.4
1.8
30
120
200
380
0
112
80
90
30
125
285
373
60
83
165
40
38
163
288
153
58
25
0
44
80
198
138
125
23
5
104
38
95
125
68
USLE
USLE
C Valueb S Value0
Exclusion
Ruled
Erosion
Estimate'
t ha"1 y r
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.234
0.141
0.477
0.235
0.084
0.065
0.268
0.105
0.488
0.148
0.313
0.175
0.175
0.351
0.511
0.635
1.169
0.454
0.175
0.182
0.123
0.313
0.844
0.065
0.268
0.260
0.141
0.141
0.129
0.129
0.154
0.268
0.182
0.135
0.141
0.168
0
0
0
3
2
0
3
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
2
0
0
0
0
3
0
I
0
0
0
3
0
1.09
9.04
3.21
N/A
N/A
3.90
N/A
8.02
1.14
4.75
2.65
2.88
3.97
8.07
14.12
12.80
4.85
2.25
2.77
N/A
3.49
7.31
N/A
2.68
3.30
1.92
1.73
N/A
0.92
N/A
3.79
1.51
1.48
N/A
1.66
82
Table 15.
Point
USLE Factor and Soil Loss Point Estimates (Continued).a
Slope
Gradient
Slope
Length
#
%
m
15E
15F
15G
160
1.3
5.0
1.3
0.0
15
45
75
8
USLE
USLE
C Valueb S Valuec
Exclusion
Ruled
Erosion
Estimate®
t ha"1 yr"1
0.234
0.234
0.234
0.234
0.135
0.454
0.135
0.065
I
0
0
I
N/A
5.28
1.37
N/A
aUSLE soil loss estimates calculated using R = 476.6, K = .049, L as
calculated using equation # I, page 21 and P = 1.0.
bUSLE C factor from Table 2.
cUSLE S factor using formula of Foster and Wischmeier (1984).
dExclusion code; O = apply USLE, I = exclude due to h i ll crest rule,
2 = channel rule, and 3 = concave slope rule.
eN/A = USLE does not apply, erosion = 0.0 t ha"1 yr"1.
83
Table 16.
Site
WEE Factor and Soil Loss Point Estimates
Soil
Groupb
r
#
IF
IG
2M
2N
20
2P
SE
3F
SG
4N
40
4P
SE
SF
SG
SH
SM
SN
60
SP
7B
7C
70
7E
7F
7G
8J
8K
8L
SM
SN
80
8P
9A
9B
9C
90
9E
9F
9G
Is
rc
(%)
38B
38B
38B
38B
38B
38A
38B
38B
38A
38B
d
38A
38B
38B
38A
38A
38B
38B
38B
385B
38B
38B
38B
38B
38B
385B
38B
38B
38B
38B
38B
28
385B
385B
385B
38B
38B
38B
28
28
234.2
234.2
234.2
234.2
234.2
180.2
234.2
234.2
180.2
234.2
-
180.2
234.2
234.2
180.2
180.2
234.2
234.2
234.2
253.3
234.2
234.2
234.2
234.2
234.2
253.3
234.2
234.2
234.2
234.2
234.2
280.9
253.3
253.3
253.3
234.2
234.2
234.2
280.9
280.9
100
100
100
100
100
372
100
100
100
100
-
215
100
100
100
100
100
102
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
160
100
100
100
100
100
100
234.2
234.2
234.2
234.2
234.2
670.3
234.2
234.2
180.2
234.2
-
387.4
234.2
234.2
180.2
180.2
234.2
238.9
234.2
253.3
234.2
234.2
234.2
234.2
234.2
253.3
234.2
234.2
234.2
234.2
234.2
280.9
253.3
405.3
253.3
234.2
234.2
234.2
280.9
280.9
a
Fetch
Length
Soil Loss Estimates
Fallow
Crop
Total
Cycle
m
Mg ha"1 yr"1
38
113
28
5
58
80
43
S3
28
148
5
80
80
25
18
S3
13
140
8
70
50
88
38
73
55
35
30
25
25
8
133
28
155
5
130
83
85
68
50
90
0.0
0.0
0.0
0.0
0.0
1.2
0.0
0.0
0.0
0.0
-
0.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.1
5.8
13.0
3.7
0.0
8.7
68.4
6.6
9.2
1.0
14.6
2.9
6.5
1.9
0.0
4.4
34.8
3.3
4.6
0.5
7.3
-
-
27.4
10.8
2.9
0.0
5.0
0.0
14.8
0.0
11.7
7.6
11.4
5.8
10.2
8.3
6.5
4.1
2.9
2.9
0.0
14.0
6.7
17.0
0.0
15.9
11.0
11.2
9.7
11.6
16.2
13.8
5.4
1.5
0.0
2.5
0.0
7.4
0.0
5.9
3.8
5.7
2.9
5.1
4.2
3.3
2.1
1.5
1.5
0.0
7.0
3.4
8.6
0.0
8.0
5.5
5.6
4.9
5.8
8.2
84
Table 16.
Site
WEE Factor and Soil Loss Point Estimates
Soil
Groupb
r
#
9H
91
IO J
IOK
IOL
I OM
ION
100
IOP
IOQ
IlB
IlC
IlD
HE
H F
HG
IlH
111
12K
12L
12M
12N
120
12P
12Q
13C
13D
13E
13F
13G
13H
14K
14L
14M
14N
140
14P
Is
r
c
(%)
385B
385B
385B
385B
38B
38B
28
28
28
385B
385B
385B
385B
38B
28
28
28
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
385B
253.3
253.3
253.3
253.3
234.2
234.2
280.9
280.9
280.9
253.3
253.3
253.3
253.3
234.2
280.9
280.9
280.9
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
100
100
100
100
100
100
100
100
100
100
100
100
100
100
113
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Fetch
Length
Soil Loss Estimates
Fallow
Crop
Total
Cycle
m
253.3
253.3
253.3
253.3
234.2
234.2
280.9
280.9
280.9
253.3
253.3
253.3
253.3
234.2
317.4
280.9
280.9
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
253.3
45
68
65
18
20
I
128
23
150
15
123
78
80
60
45
83
40
63
10
15
140
25
86
143
8
73
73
55
40
8
33
5
8
133
115
10
138
a
Mg ha"1 yr"1
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.1
8.4
11.4
11.1
1.7
1.5
0.0
18.8
5.7
20.0
0.7
15.5
12.6
12.6
8.9
14.1
15.6
9.8
10.9
0.0
0.7
16.3
4.0
13.1
16.5
0.0
11.9
11.9
9.9
7.5
0.0
6.1
0.0
0.0
16.0
15.1
0.0
16.3
4.2
5.7
5.6
0.9
0.8
0.0
9.5
2.9
10.0
0.4
7.8
6.3
6.3
4.5
7.1
7.9
4.9
5.5
0.0
0.4
8.2
2.0
6.6
8.3
0.0
6.0
6.0
5.0
3.8
0.0
3.1
0.0
0.0
8.0
7.6
0.0
8.2
85
Table 16.
Site
WEE Factor and Soil Loss Point Estimates (Continued).a
Soil
Groupb
r
#
15E
15F
15G
160
Is
I'C
(%)
385B
385B
385B
385B
253.3
253.3
253.3
253.3
100
100
100
100
Fetch
Length
Soil Loss Estimates
Fallow
Crop
Total
Cycle
m
253.3
253.3
253.3
253.3
48
33
70
5
Mg ha"1 yr"1
0.0
0.0
0.0
0.0
8.9
6.1
11.7
0.0
4.5
3.1
5.9
0.0
aWEE soil loss estimates calculated using K = 0.49, 0.66 and 0.9, C =
0.90, and V varying by periods (see Table 5).
bSoil mapping un it numbers; 28, Nishon series, 38A, 38B and 385B, Ethridge
series (see Figure 3).
cSoil e ro d ib ility (I and I ' ) in t ha"1 yr"1.
dPoint excluded due to its location in the incised channel.
APPENDIX B
CESIUM-137 LABORATORY DATA AND AREAL CONCENTRATIONS
87
Table 17.
Site3
#
Cesium-137 Laboratory Data and Areal Concentrations
Depth
oi:
Increment
cm
ITl
0
16
-
15
20
Bul k
Density
g cm"3
1.27
1.28
Mass
Concerntra tio n
Counting
Error
(±)
Increment
Areal
A c tiv ity
pCi g-1
0.53
0.05
Error
(+)
Total
Site
A c tiv ity
pCi cm"2
0.05
0.01
10.10
0.32
0.95
0.06
10.42
2T1
0
15
16 - 20
1.30
1.34
0.34
0.05
0.04
0.01
6.63
0.34
0.78
0.07
6.97
3T1
0
16
-
15
20
1.30
1.34
0.47
0.00
0.04
0.00
9.17
0.00
0.78
0.00
9.17
4T1
0
16
21
26 -
-
15
20
25
30
1.35
1.44
1.44
1.44
0.51
0.33
0.07
0.02
0.04
0.03
0.03
0.03
10.33
2.38
0.50
0.14
0.81
0.22
0.22
0.22
13.35
5T1
0
16
-
15
20
1.29
1.41
0.39
0.03
0.04
0.01
7.55
0.21
0.77
0.07
7.76
6T1
7T1
0
16
15
20
1.19
1.36
0
15
16 - 20
1.18
1.27
-
0.24
0.02
0.04
0.01
4.28
0.14
0.71
0.07
4.42
0.15
0.05
0.01
0.01
2.66
0.32
0.18
0.06
2.98
1T2
0
16
-
15
20
1.22
1.27
0.35
0.00
0.05
0.00
6.41
0.00
0.92
0.00
6.41
2T2
0
16
21
-
15
20
25
1.07
1.39
1.39
0.44
0.25
0.00
0.05
0.01
0.00
7.06
1.74
0.00
0.80
0.07
0.00
8.80
88
Table 17.
S ite3
#
Cesium-137 Laboratory Data and Areal Concentrations
(Continued).
Depth
Bul k
of
Density
Increment
cm
3T2
0 - 15
16 - 20
Mass
Concentra tio n
Increment
Areal
A c tiv ity
pCi g'1
g cm"3
1.19
1.41
Counting
Error
(±)
0.48
0.00
0.06
0.00
Error
(+)
Total
Site
A c tiv ity
pCi cm"2
8.57
0.00
1.07
0.00
8.57
4T2
0 - 15
16 - 20
21 - 25
1.25
1.46
1.46
0.54
0.26
0.06
0.06
0.06
0.02
10.13
1.90
0.44
1.13
0.44
0.15
12.47
5T2
0
16
21
26
-
15
20
25
30
1.27
1.33
1.33
1.33
0.45
0.38
0.19
0.00
0.04
0.05
0.01
0.00
8.57
2.53
1.26
0.00
0.76
0.33
0.07
0.00
12.36
6T2
0 - 15
16 - 20
1.29
1.35
0.20
0.03
0.03
0.02
3.87
0.20
0.58
0.14
4.07
7T2
0 - 15
16 - 20
1.18
1.27
0.15
0.05
0.01
0.01
2.66
0.32
0.18
0.06
2.98
IC
2C
3C
0 - 5
6 - 10
11 - 15
1.40
1.40
1.40
0
6
11
16
- 5
- 10
- 15
- 20
1.40
1.40
1.40
1.40
0
6
11
16
21
-
1.40
1.40
1.40
1.40
1.40
5
10
15
20
25
0.24
0.06
0.00
0.03
0.01
0.00
1.68
0.42
0.00
0.21
0.07
0.00
2.10
0.27
0.28
0.18
0.02
0.03
0.03
0.04
0.04
1.89
1.96
1.26
0.14
0.21
0.21
0.28
0.28
5.25
0.30
0.12
0.08
0.26
0.16
0.05
0.01
0.03
0.04
0.16
2.10
0.84
0.56
1.82
1.12
0.35
0.07
0.21
0.28
0.07
6.44
89
Table 17.
Site"
#
Cesium-137 Laboratory Data and Areal Concentrations
(Continued).
Depth
Bul k
oiF
Density
Increment
cm
SC
0
16
21
26
-
15
20
25
30
Mass
Concentra tio n
g cm'3
1.09
1.36
1.36
1.36
pCi
0.54
0.59
0.65
0.93
Counting
Error
(±)
Increment
Areal
A c tiv ity
Error
(+)
pCi cm'2
g '1
0.04
0.03
0.05
0.06
8.83
4.01
4.42
6.32
0.65
0.20
0.34
0.41
23.58
'
IP
0
6
11
16
21
Total
Site
A c tiv ity
_
-
5
10
15
20
25
1.30
1.30
1.30
1.30
1.30
0.36
0.55
0.71
0.69
0.71
0.05
0.02
0.01
0.08
0.04
2.34
3.58
4.62
4.49
4.62
0.33
0.13
0.07
0.52
0.26
19.65
4P
0
6
11
16
21
-
5
10
15
20
25
1.44
1.44
1.44
1.44
1.44
0.42
0.38
1.01
0.59
0.32
0.04
0.01
0.09
0.06
0.02
3.02
2.74
7.27
4.25
2.30
0.29
0.07
0.65
0.43
0.14
19.58
6P
7P
IW
0
6
11
16
21
-
0
6
11
16
-
0
6
11
16
5
10
15
20
25
1.33
1.33
1.33
1.33
1.33
0.56
1.16
0.70
0.51
0.06
0.05
0.02
0.05
0.06
0.01
3.72
7.71
4.66
3.39
0.40
0.33
0.13
0.33
0.40
0.07
19.88
-
5
10
15
20
1.40
1.40
1.40
1.40
5
10
15
20
1.24
1.24
1.24
1.24
0.28
0.12
0.04
0.00
0.04
0.03
0.01
0.00
1.96
0.84
0.28
0.00
0.28
0.21
0.07
0.00
3.08
1.14
0.03
0.00
0.00
0.11
0.03
0.00
0.00
7.07
0.19
0.00
0.00
0.68
0.19
0.00
0.00
7.26
90
Table 17.
Site8
#
2W
Cesium-137 Laboratory Data and Areal Concentrations
(Continued).
Depth
Bul k
of
Density
Increment
0
6
11
16
cm
g cm"3
- 5
- 10
- 15
- 20
1.26
1.26
1.26
1.26
Mass
Concentra tio n
pCi
1.59
0.16
0.02
0.00
Counting
Error
(±)
Increment
Areal
A c tiv ity
Total
Site
A c tiv ity
pCi cm'2
g 1
0.10
0.04
0.04
0.00
Error
(±)
10.02
1.01
0.13
0.00
0.63
0.25
0.25
0.00
11.16
3W
0
6
11
16
- 5
- 10
- 15
- 20
1.13
1.13
1.13
1.13
2.07
0.29
0.06
0.00
0.12
0.06
0.04
0.00
11.70
1.64
0.34
0.00
0.68
0.34
0.23
0.00
13.68
4W
0
6
11
16
21
-
5
10
15
20
25
1.13
1.13
1.13
1.13
1.13
1.68
0.40
0.05
0.05
0.00
0.09
0.05
0.02
0.03
0.00
9.49
2.26
0.28
0.28
0.00
0.51
0.28
0.11
0.17
0.00
12.31
aSample s ite code; Tl = transect 1,T2 = transect 2, C = channel, P = pond,
W = c o n tro l.
m
APPENDIX C
SIEVING RESULTS
92
Table 18.
Sieving Results.
Soil
Mapping
Unit
Site
Non-erodible Particles
> 0.84 mm
%
WEE I Factor
t ha"1 yr"1
Nishon Series (28)
4T2
5T2
4C
36.5
43.5
40.5
314
253
275
53.0
45.0
53.1
48.2
37.4
52.6
36.5
49.0
41.5
28.5
52.2
49.5
51.2
32.3
40.0
40.2
44.8
50.8
48.8
35.5
44.8
39.5
40.0
33.2
35.5
41.5
157
242
155
Ethridge Series (3858)
ITl
2T1
3T1
4T1
5T1
7T1
1T2
6T2
ITA
2TA
3TA
4TA
5TA
6TA
7TA
ITB
2TB
3TB
4TB
5TB
6TB
2TC
2TD
3TD
2TE
3TE
212
307
161
314
206
267
392
166
195
177
354
282
280
244
182
204
321
244
379
282
345
320
267
93
Table 18.
Soil
Mapping
Unit
Sieving Results (Continued).
Site
Non-erodible Particles
> 0.84 mm
%
WEE I Factor
t ha"1 yr"1
Ethridge Series (38A)
19T3
25T4
26T4
27T4
28T4
29T4
30T4
51.0
62.0
52.8
52.2
54.8
44.8
44.8
179
96
170
188
139
244
244
45.6
49.9
52.2
39.0
49.5
50.8
35.8
51.7
40.4
48.0
40.6
46.7
39.5
238
193
166
291
202
182
318
170
278
215
276
229
287
Ethridge Series (388)
22T3
23T3
24T3
3TC
4TC
5TC
6TC
7TC
4TD
4TE
5TE
6TE
7TE
MONTANA STATE UNIVERSITY LIBRARIES
762
10069392
6