Effect of mechanical and biological enhancements on erosion at high elevation disturbed lands
by Susan Rhea Winking
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Land
Rehabilitation
Montana State University
© Copyright by Susan Rhea Winking (2002)
Abstract:
The objective of this study was to evaluate the effect of erosion control measures on sediment yields on
reclaimed steep slopes at the Treasure Mine, MT and compare measured sediment yields to values
predicted by Revised Universal Soil Loss Equation (RUSLE) Version 1.06.
Plots were constructed on a regraded waste rock area having a uniform 25 % slope. Five treatments
were replicated three times in a completely randomized design. Treatments applied were no coversoil,
30 cm coversoil, 30 cm coversoil/pitting of the soil surface, 30 cm coversoil/tree slash barriers, and 30
cm of coversoil/vesicular-arbuscular mycorrhizal (AM) fungi inoculum. Plots received identical
application of seed and fertilizer.
Total annual sediment yields for test plots were low during 2000 (mean 0.11 Mg/ha) and increased in
2001 (mean 1.17 Mg/ha). There were no differences in mean sediment yield by treatment in both years.
There was a trend on pitted slopes for lower sediment yield in 2000 and significantly lower rill severity
rating in 2001. Results suggest that pitting of the soil surface is potentially an effective erosion control
practice at the level of precipitation received during the study, preventing rill formation and reducing
sediment yields on steep slopes until vegetation can provides adequate slope stability.
Plant growth was significantly lower on the no coversoil treatment, but there were no differences
between those remaining treatments that received 30 cm of coversoil.
Prior to implementation of field treatments, Sorghum Sudanese grown in the greenhouse in coversoil
and waste rock material collected before application of AM inoculum had 39 % and 30 % AM root
colonization levels, respectively. After two growing seasons, there were no significant differences in
percent AM root colonization of Hordeum vulgare harvested from no coversoil (34 %), coversoil (34
%), and coversoil/AM inoculum (35 %) treatments. Agropyron trachycaulum harvested from AM
inoculum treated plots showed significantly higher AM colonization levels (53 %) compared to the
non-inoculated coversoil (46 %) and no coversoil treatments (44 %).
AM inoculation treatment did not enhance aboveground plant growth.
Although RUSLE version 1.06 overpredicted mean sediment yields by 0.2 ± 0.2 Mg/ha during 2000
and underpredicted by 1.0 ± 1.0 Mg/ha in 2001, estimates of sediment yields were close to actual
sediment yields. Rill formation factor constants were applied to the 2001 data when rilling was
moderate or greater, which improved RUSLE’s ability to predict sediment yield to within 97 % of
measured sediment yield.
The sediment-delivery ratio was 0.14 for the coversoil/pitting treatment and 0.60 for the coversoil/slash
barriers. EFFECT OF MECHANICAL AND BIOLOGICAL ENHANCEMENTS ON EROSION AT
HIGH ELEVATION DISTURBED LANDS
by
Susan Rhea Winking
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Land Rehabilitation
MONTANA STATE UNIVERSITY
Bozeman, Montana
July, 2002
V Jl^u
APPROVAL
of a thesis submitted by
Susan Rhea Winking
This thesis has been read by each member of the thesis committee and has
been found to be satisfactory regarding content, English usage, format, citations,
bibliographic style, and consistency, and is ready for submission to the College of
Graduate Studies.
Douglas J. DoIlhop/
?
)
/}. Q 6 ^ % % /
Ij a p o j.
Approved for the Department of Land Resources & Environmental Sciences
Jeffrey S. Jacobsen
ignature)
Date
Approved for the College of Graduate Studies
Bruce R. McLeo
(Signature)
Date
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a
master's degree at Montana State University, I agree that the Library shall make it
available to borrowers under rules of the Library.
If I have indicated my intention to copyright this thesis by including a
copyright notice page, copying is allowable only for scholarly purposes,
consistent with “fair use” as prescribed in the U.S. Copyright Law, Requests for
permission for extended quotation from or reproduction of this thesis in whole or
in parts may be granted only from the copyright holder.
Signature
iv
ACKNOWLEDGEMENT
I would like to thank Barretts Minerals, Inc. and PHC Reclamation for
providing funding for this research. I would also like to thank my committee
members, Rick Fasching, Paul Hook and especially Catherine Zabinski for her
guidance and expertise on the mycorrhizal fungi aspect of this research. Special
appreciation is extended to my major advisor, Douglas Dollhopf, for his
knowledge and experience in the development and refinement of reclamation
science. I dedicate this work to my daughter, Annabelle Winking.
TABLE OF CONTENTS
LIST OF TABLES...................................................................................................... vii
LIST OF FIGURES.................... .............................!................................. ................ xiii
ABSTRACT................................................................................................................ xiv
1. INTRODUCTION...................................................................................................... I
2. LITERATURE REVIEW........................................................................................... 3
<0. OO <1
Erosion Control Regulations....................................................................................... 3
Revised Universal Soil Loss Equation (RUSLE)........................
3
R - Rainfall/Runoff Erosivity.........
K —Soil Erodibility.........................
LS - Hillslope Length and Gradient
C - Cover-Management..................
P - Support Practices......................................................................................... 10
Improved Accuracy in Representing Effect of Slope Steepness....... '............... 11
RUSLE and Geomorphology...........................................................
12
Database Sensitivity of RUSLE......................................................................... 12
Use of RUSLE on High Elevation, Steep Slopes.............................................. 13
Commercial Mycorrhizal Fungi Inoculations in Mineland Reclamation..................14
3. METHODOLOGY................................................................................................... 16
Site Description ......
16
Experimental Design................................................................................................. 17
Sediment Collection Trough Design......................................................................... 18
Seeding Mix and Seeding Procedures....................................................................... 18
Commercial AM Inoculum Application Procedures ......................................., 20
Data Collection and Analysis.......... ..............
21
Spoil Backfill and Coversoil Physiochemical Analysis...........................
21
Precipitation and Evaporation Measurements...................................................21
Runoff Collection Methods................................
22
Sediment Collection Methods............................................................................22
Rill Classification Methods................................................................................24
Plant Measurements........................................................................................... 25
Pre-Inoculation AM Colonization Levels of Coversoil and
Spoil Backfill Material....................................................................................... 26
Percent Colonization by AM Fungi Procedures................................................ 26
TABLE OF CONTENTS continued
Mycorrhizal Infection Study.............................................................................. 27
Statistical Analysis............................................................................................. 27
RUSLE v.1.06 Computer Model Analysis........................................................ 28
Hillslope Area Calculations............................................................................... 29
4. RESULTS AND DISCUSSION............................................................................... 31
Coversoil and Spoil Backfill PhysiochemicalCharacteristics................................... 31
Sediment Yields on High Altitude Steep Slopes...................................................... 32
Effect of Mechanical and Biological Erosion Control Measures
on Sediment Yield.................................................................................. 33
Effect of Rill Severity Class on Sediment Yield...............................................34
Effect of Precipitation on Sediment Yield......................................................... 36
Effect of Rock Cover on Sediment Yield, Plant Biomass, Runoff and
Rill Severity........................................................................................... 39
Runqff on High Altitude Steep Slopes....................................
40
Effect of Runoff on Sediment Yield and Rill Severity Class............................ 40
Plant Growth on High Altitude, Steep Slopes.......................................................... 42
Effect of Erosion Control Measures on Plant Growth Characteristics..............42
Effect of Plant Growth on Sediment Yield, Runoff and Rill
Severity Class.....................................
43
Pre-Inoculation Mycorrhizal Infectivity Potential............................................. 44
Effect of AM inoculum on Colonization Levels m Agropyron
trachycaulum and Hordeum vulgare.................................................... 44
Ability of Measured Factors to Predict Sediment Yield and Runoff........................45
Ability of RUSLE v.1.06 to Predict Sediment Yields at High Elevation Sites........47
Measured Sediment-Delivery Ratios for Pitting and Slash Barrier
Treatments............................................................................................. 48
5. SUMMARY AND CONCLUSIONS....................................................................... 51
REFERENCES CITED.............................................................................................. ,...56
APPENDICES..........................
APPENDIX A APPENDIX B APPENDIX C APPENDIX D APPENDIX E -
60
Soil Physiochemical Data....................................................... 61
Precipitation Data............................................................... ,...63
Sediment Yield and RunoffData..............................
77
Vegetation Data...................................................................... 84
Statistical Analysis.............................................
100
vii
LIST OF TABLES
Table
Page
1. Examples of Montana legislation mandating erosion control......................................4
2. Seed mix and application rate on test plots................................................................20
3. Soil physiochemical analytical methods....................................................................21
4. Soil erosion condition classification for rills..............................................................24
5. Input variables for RUSLE version 1.06.................................................................... 28
6. Coversoil and spoil backfill physiochemical characteristics..................................... 31
7. Mean sediment yield by treatment during 2000 and 2001......... ............................... 34
8. Mean rill severity class by treatment during 2000 and 2001 ....... ............................. 34
t
9. Mean sediment yield and precipitation across all treatments for two-week
periods during 2000 and 2001....................................................................................36
10. Strength of relationship of total precipitation, maximum daily
precipitation, and maximum hourly precipitation to mean Sediment yield
across all treatments for tWo-week periods during 2000 and 2001............................ 37
11. Mean percent rock cover by treatment during 2001 ........................... ...................... 39
12. Strength of relationship of percent rock cover to sediment yield, biomass,
runoff and rill severity class during 2001 ..................................................................40
13. Mean annual runoff (m3/ha) by treatment during 2000 and 2001............................. 40
14. Strength of relationship of runoff to sediment yield and rill severity class
during 2000 and 2001 .............................. ................................................. ............... 42
15. Mean plant growth characteristics by treatment during 2001....................................42
16. Strength of relationship of plant growth characteristics to sediment yield, runoff
and rill severity class during 2001 .............................................................................43
17. Mean percent pre-inoculation AM colonization levels in the coversoil and spoil
backfill material during 2000........................ ............................................................ 44
V lll
LIST OF TABLES - continued
Table
-n
Page
18. Mean percent AM colonization levels of Hordeum vulgare and Agropyron
trachycaulum during 2001......................................................................................... 45
19. Multiple linear regression analysis results using measured factors to predict
sediment yield and runoff during 2000 and 2001..................................................... 46
20. RUSLE predicted and measured sediment yield values (Mg/ha) for all
test plot treatments during 2000 and 2001.................................................................47
21. Optimized RUSLE sediment yields (Mg/ha) using rill formation factors
during 2001................................................................................................................ 491
22. Measured sediment-delivery ratio values for Coversoil/Pitting and
Coversoil/Slash Barrier treatments.............. *............................................................ 50
23. Coversoil textural analyses.........................................................................................62
24. Coversoil pH and EC analyses....................................................................................62
25. Subsoil textural analyses..................................
62
26. Subsoil pH and EC analyses................................................
62
27. Monthly precipitation (cm) for Dillon, Montana (WMCE), 2000.................
64
28. Monthly precipitation (cm) for Dillon, Montana (WMCE), 2001..............................65
29. Precipitation (cm) at Treasure Mine, June 20 -30,2000.............................................66
30. Precipitation (cm) at Treasure Mine, July 2000..........................................................67
31. Precipitation (cm) at Treasure Mine, August 2000.................................................... 68
32. Precipitation (cm) at Treasure Mine, September 2000...............................................69
33. Precipitation (cm) at Treasure Mine, October I - October 10,2000..........................70
34. Precipitation (cm) at Treasure Mine, May 4 - May 31,2001.....................................71
V
ix
LIST OF TABLES - continued
Table
Page
35. Precipitation (cm) at Treasure Mine, June 2001..............................................
72
36. Precipitation (cm) at Treasure Mine, July 2001..........................................................73
37. Precipitation (cm) at Treasure Mine, August 2001.................................................... 74
38. Precipitation (cm) at Treasure Mine, September I -17,2001....................
75
39. Evaporation (cm) at Treasure Mine, 2000 and 2001 ............................ .................... 76
40. Sediment yield (kg) on test plots, 2000......................................................................78
41. Sediment yield (kg) on test plots, 2001......... ............... ........................ ;.................. 79
42. Depth (cm) of runoff in troughs at test plots, 2000.....................................................80
43. Depth (cm) of runoff in troughs at test plots, 2001.....................................................81
44. Rill severity (class) on test plots, 2000.......................................................................82
45. Rill severity (class) on test plots, 2001.......................................................................83
46. Perennial grass canopy cover (%) on test plots, 2001.............................................. 85
47. Forb canopy cover (%) on test plots, 2001 ................................................................86
48. Annual grass canopy cover (%) on test plots, 2001....................................................87
49. Perennial grass basal cover (%) on test plots, 2001....................................................88
50. Forb basal cover (%) on test plots, 2001................................................................. 89
51. Annual grass basal cover (%) on test plots, 2001............................. ........................ 90
52. Rock cover (%) on test plots, 2001......................................................................... 91
53 . Bare ground cover (%) on test plots, 2001..... ........................................................... 92
54. Perennial grass biomass (g) on test plots, 2001
93
LIST OF TABLES - continued
Table
Page
55. Forb biomass (g) on test plots, 2001..........................................................................94
56. Annual grass biomass (g) on test plots, 2001................. ........................................... 95
57. Vegetative litter cover (class) on test plots, 2001......................................................96
58. Percent colonization by vesicular arbuscular mycorrhizal fimgi in coversoil
and spoil backfill material, 2000.................................................................................97
59. Percent colonization by vesicular arbuscular mycorrhizal fungi in
Hordeum vulgare, 2001..................................................................................
...98
60. Percent colonization by vesicular arbuscular mycorrhizal fungi m Agropyron
trachycaulum, 2001.....................................................................................................99
61. Two way analysis of variance of sediment yield, 2000............................................ 101
62. Two way analysis of variance of sediment yield, 2001...............
103
63. Two way analysis of variance of rill severity class, 2000..........
104
64. Two way analysis of variance of rill severity class, 2001.............................
106
65. Linear regression of mean rill severity class on total annual sediment yield, 2000...108
66. Linear regression of mean rill severity class on total annual sediment yield, 2001 ...109
67. Linear regression of total precipitation on total annual sediment yield, 2000......... 110
68. Linear regression of maximum daily precipitation on total annual soil
loss,2000 ................................................................................................................. H I
69. Linear regression of maximum hourly precipitation on total annual soil
loss, 2000...........................................
112
70. Linear regression of total precipitation on total annual sediment yield, 2001...........113
71. Linear regression of maximum daily precipitation on total annual
sediment yield, 2001............................................................................................ ,....114
LIST OF TABLES - continued
Table
Page
72. Linear regression of maximum hourly precipitation on total annual
sediment yield, 2001 .............................................................................................. 115
73. Two way analysis of variance of percent rock cover during 2001........... ................ 116
74. Linear regressionof percent rock cover on sediment yield, 2001...............................117
75. Linear regressionof percent rock cover on biomass, 2001........................................ 118
76. Linear regressionof percent rock cover on total annual runoff, 2001........................119
77. Linear regression of percent rock cover on rill severity class, 2001..........................120
78. Two way analysis of variance of runoff (m3/ha), 2000............................................ 121
79. Two way analysis of variance of runoff (m3/ha), 2001............................................ 122
80. Linear regression of runoff on sediment yield, 2000................................................ 123
81. Linear regression of runoff on sediment yield, 2001................................................ 124
82. Linear regression of runoff on mean rill severity class, 2000...................................125
83. Linear regression of runoff to mean rill severity class, 2001...................................126
84. Two way analysis of variance on biomass (kg/ha), 2001................
127
85. Two way analysis of variance on percent canopy cover, 2001................................ 129
86. Two way analysis of variance on percent basal cover, 2001 ................................... 131
87. Linear regression of biomass (kg/ha) on sediment yield, 2001................................. 133
88. Linear regression of percent canopy cover on sediment yield, 2001 ....................... 134
89. Linear regression of percent basal cover on sediment yield, 2001 .......................... 135
90. Linear regression of biomass (kg/ha) on runoff, 2001
......................................... 136
91. Linear regression of percent canopy cover on runoff, 2001
137
\
xii
LIST OF TABLES - continued
Table
"
Page
92. Linear regression of percent basal cover on runoff, 2001........................................ 138
93. Linear regression of biomass (kg/ha) on rill severity, 2001..................................... 139
94. Linear regression of percent canopy cover on rill severity, 2001 ......... ................... 140
95. Linear regression of percent basal cover on rill severity, 2001................................. 141
96. Multiple linear regression of runoff, mean rill'severity class and slope area
on sediment yield, 2000............... ............................................................................ 142
97. Multiple linear regression of rill severity class, runoff, slope area, biomass,
canopy cover, basal cover and rock cover on sediment yield, 2001......................... 143
98. Multiple linear regression of rill severity class and slope area on runoff, 2000.......145
99. Multiple linear regression of mean rill severity class, slope area, biomass,
canopy cover, basal cover and rock cover on runoff, 2001....................................... 146
100. Multiple linear regression of maximum hourly precipitation, increase in
rill severity class and runoff on sediment yield, 2000 and 2001.............................148
101. T-test of mean percent AM colonization levels in the coversoil
and spoil backfill material during 2000................................................................... 149
102. Two way analysis of variance of percent AM colonization levels
VhHordeum vulgare, 2001....................................................................................... 150
103. Two way analysis of variance of percent AM colonization levels in
Agropyron trachycaulum, 2001........................................................................... :..152
Xiii
LIST OF FIGURES
Figure
Page
1. Location of study site in southwest Montana............................................................16
2. Design and layout of test plots at the Treasure Mine, MT........................................19
3. Calculation to convert water depth in trough into volumetric measure....................23
4. Schematic of dimensions used to calculate area contributing to runoff
and sediment yield on pitted treatment plots...........................................................30
5. Comparison of mean annual sediment yields for by treatment during 2000
and 2001................................ ....................................................... :.......................... 32
6. Strength of relationship of mean rill severity class to total annual sediment
yield during 2000 and 2001 ......................................................................;...... ....... 35
7. Comparison of total precipitation and mean sediment yield across all treatments
for two-week periods during 2000 and 2001........ ......................... ,........................ 38
8. Comparison of mean sediment yield and runoff by treatment during 2000
and 2001...................................................................................................................41
9. Comparison of measured, RUSLE predicted, and optimized RUSLE
sediment yields using rill formation factors on test plots during 2001.....................49
xiv
ABSTRACT
The objective of this study was to evaluate the effect of erosion control measures
on sediment yields on reclaimed steep slopes at the Treasure Mine, MT and compare
measured sediment yields to values predicted by Revised Universal Soil Loss Equation
(RUSLE) Version 1.06.
Plots were constructed on a regraded waste rock area having a uniform 25 %
slope. Five treatments were replicated three times in a completely randomized design.
Treatments applied were no coversoil, 30 cm coversoil, 30 cm coversoil/pitting of the soil
surface, 30 cm coversoil/tree slash barriers, and 30 cm of coversoil/vesicular-arbuscular
mycorrhizal (AM) fungi inoculum. Plots received identical application of seed and
fertilizer.
Total annual sediment yields for test plots were low during 2000 (mean 0.11
Mg/ha) and increased in 2001 (mean 1.17 Mg/ha). There were no differences in mean
sediment yield by treatment in both years. There was a trend on pitted slopes for lower
sediment yield in 2000 and significantly lower rill severity rating in 2001. Results
suggest that pitting of the soil surface is potentially an effective erosion control practice
at the level of precipitation received during the study, preventing rill formation and
reducing sediment yields on steep slopes until vegetation can provides adequate slope
stability.
Plant growth was significantly lower on the no coversoil treatment, but there were
no differences between those remaining treatments that received 30 cm of coversoil.
Prior to implementation of field treatments, Sorghum Sudanese grown in the
greenhouse in coversoil and waste rock material collected before application of AM
inoculum had 39 % and 30 % AM root colonization levels, respectively. After two
growing seasons, there were no significant differences in percent AM root colonization of
Hordeum vulgare harvested from no coversoil (34 %), coversoil (34 %), and
coversoil/AM inoculum (35 %) treatments. Agropyron trachycaulum harvested from AM .
inoculum treated plots showed significantly higher AM colonization levels (53 %)
compared to the non-inoculated coversoil (46 %) and no coversoil treatments (44 %).
AM inoculation treatment did not enhance aboveground plant growth.
Although RUSLE version 1.06 overpredicted mean sediment yields by 0.2 ± 0.2
Mg/ha during 2000 and underpredicted by 1.0 + 1.0 Mg/ha in 2001, estimates of
sediment yields were close to actual sediment yields. Rill formation factor constants
were applied to the 2001 data when rilling was moderate or greater, which improved
RUSLE's ability to predict sediment yield to within 97 % of measured sediment yield.
The sediment-delivery ratio was 0.14 for the coversoil/pitting treatment and 0.60 for the
coversoil/slash barriers.
I
I. INTRODUCTION
An integral part of land reclamation Best Management Plans (BMPs) is a
combination of management and structural practices to control erosion hazards resulting
from soil disturbance. BMPs can ihclude erosion control activities such as increasing
surface roughness of the soil, mulch incorporation, and terraces (Toy & Foster, 1998).
These mechanical erosion control measures are routinely used on many mined lands and
construction sites to provide short-term and long-term stability to disturbed areas and
minimize or eliminate of&site impacts (Toy & Foster, 1998).
Erosion hazards increase when vegetative cover is lost, soil permeability is low,
and the ground increasingly slopes, especially if soils are shallow (Brooks et al., 1997).
Good vegetative cover ideally reduces erosion hazards, but the development of adequate
plant cover may be difficult due to the short growing season, an often thin, nutrient-poor,
rocky soil resource, and mining practices. At higher elevations, soil erosion is dominated
by spring precipitation or runoff events including snowmelt, rain on snow, and thawing
soils. Soils are particularly susceptible to erosion when the frost layer recedes below
surface during spring thaw. The frost layer still prevents water infiltration and generates
ruhoff but leaves the thawed layer vulnerable to detachment and soil loss, which directly
affects vegetative establishment especially during initial years after reclamation begins
(Toy & Foster, 1998). This research focuses on mechanical practices that reduce the
energy of flowing water and biological enhancements to stimulate vegetative growth to
reduce soil erosion on high elevation steep slopes.
2
Investigative Objectives
The objectives of this study were i) to evaluate the effectiveness of twp
mechanical control measures in decreasing runoff volume and sediment yield from
coversoil erosion and the effect of these mechanical measures on plant growth, ii) to
determine the effectiveness of biological measures in enhancing plant growth and thereby
decreasing runoff volume and sediment yield from coversoil erosion, and iii) to
determine the effect of coversoil depth (0 cm and 30 cm) oh plant growth, and to evaluate
how coversoil affects runoff volume and sediment yield.
3
2. LITERATURE REVIEW
Erosion Control Regulations
Soil erosion control continues to be a significant challenge to agriculture as well
as industry resulting in poor water quality due to sediment (Renard & Ferreria, 1993),
geomorphological downstream impacts such as channel change (Knighton, 1998), loss of
soil resources. (Toy & Foster, 1998) and transport of adsorbed chemicals (Renard Sc
Ferreria, 1993). Because of this, there has been increasing demand by regulatory
authorities on industry to provide erosion analyses as a part of their land management
plans.
Since the 1970's, the United States and the State of Montana have passed
legislation that places more stringent environmental fcontrols on mining and mineral
extraction operations. For example, erosion control measures have been specifically
targeted under the U.S. Surface Mining Control and Reclamation Act and the Clean
Water Act (U.S. Congress, 1977; U.S. Congress, 1972). Montana environmental laws
place strict erosion control measures op the mining industry as well. Table I provides
examples of legislation concerning soil erosion on coal and metal mines in Montana.
Revised Universal Soil Loss Equation IRUSLEt
The main processes influencing soil erosion by water are raindrop irtipact arid
subsequent transport of soil by flowing water. Accelerated erosion is considered to be
occurring when soil erosion rates exceed 0.2-0.5 Mg/ha (0.1-0.2 tons/ac) annually, which
4
Table I. Examples of Montana legislation mandating erosion control.
M on tan a
M in in g L aw
C itation
In tent
Strip & Underground
Mine Siting Act
82-4-202 MCA*
Requires approved reclamation plan that
includes erosion control measures before
permit will be issued.
Strip & Underground
Mine Siting Act
82-4-231 MCA
Mining operation must take all measures
to prevent damages to the people and
property by soil erosion.
Strip & Underground
Mine Siting Act
82-4-231 (10)
(c) MCA
The mine operation must impound, drain,
or treat all runoff or underground mine
waters so as to reduce soil erosion.
Strip & Underground
Mine Siting Act
82-4-231 (11)
MCA
All stockpiled materials resulting from
land disturbances must be within permit
boundaries and cannot erode off-site.
Strip & Underground
Mine Siting Act
82-4-233 MCA
The vegetative cover must be capable of
preventing soil erosion to the extent
achieved prior to the operation.
Metal Mine
Reclamation Act
82-4-336 MCA
Reclamation plan must provide that
reclamation activities, particularly those
relating to control of erosion, to the
extent feasible, must be conducted
simultaneously with the operation.
Metal Mine
Reclamation Act
82-4-434 (2)
(g&l) MCA
The department may not approve a
reclamation plan or a plan of operations
unless the plans provide that:
■ all access, haul, and other support
roads will be located, constructed,
and maintained in such a manner as
to control and minimize channeling
and other erosion;
■ seeding and planting will be done in
a manner to achieve a permanent
vegetative cover that is suitable for
the postmine land use and that
retards erosion.
*MCA = Montana Code Annotated
5
is the rate at which soil lost to wind and water is replenished by weathering of parent
material on undisturbed lands, i.e., geological erosion (Brady & Weil, 1996). Erosion is a
two-fold process in which the soil particles are detached and then forces cause rolling,
dragging and splashing of the particles and induce transport by water (Brady & Weil,
1996). Detachment is initiated by such processes as freezing and thawing, overland
water flow, and raindrop splash applying shear stress upon soil particles. Raindrop
splash and flowing water transport the loosened soil particles (Brady & Weil, 1996;
Knighton, 1998).
The Revised Universal Soil Loss Equation (RUSLE) equation includes fiye
factofs; the erosivity potential of rainfall and runoff soil erodibility, hillslope length and
slope, plant cover and management, and erosion control support practices (Toy & Foster,
1998). This factorial approach to estimating annual average soil loss is the result of a set
of empirically-derived mathematical equations that have evolved from almost a century
of intense erosion research in the United States.
Up until the 1950's, soil and earth scientists were estimating soil loss based on
equations that were formulated in very specific geologic and climatic areas and were
therefore limited in their range of applicability (Renard et al., 1997). Thp United States
Department of Agriculture, Agricultural Research Service (USDA, ARS) formed the
National Runoff and Soil-Loss Data Center in 1954, The purpose df the Center was to
collect and assimilate soil loss data. Data used to develop the USLE and RUSLE
consisted of erosion-plot research collected from natural rainfall events and simulated
rainfall in which water was applied to erosion plots (Toy & Foster, 1998), The erosion
6
plots were 72.6 foot long by either 6.0 or 12.0 foot wide (0.01 or 0.02 acres). The
mathematical relationships betweeh each factor and soil loss were determined by
regression analysis, th is analysis led to the formulation of the Universal Soil Loss
Equation (USLE):
A - R o K o LS o C o P
Where:
A=
R=
K=
LS =
C=
P=
Average annual soil loss in tons/acre/year
Rainfall/runoff erosivity
Soil credibility
Hillslope length and steepness
Cover-management
Support practices
(Toy & Foster, 1998; Renard et al., 1997)
This equation should have universal validity because none of its factors utilized a
reference point that has direct geographic orientation (Toy & Foster, 1998).
The Revised Universal Soil Loss Equation uses.the same formula, but with
updated and improved calculations for the contributing factors (Renard et al., 1997).
Although there are many other models such as WEPP, ANSWERS, AGNPS, EPIC that
predict erosion, RUSLE is the most widely used prediction tool to date because of ease of
use, availability of parameter data, acceptable accuracy, and readily acceptable assistance
from USDA Natural Resources and Conservation Service (NRCS) personnel (Renard &
Ferreira, 1993; Toy & Foster, 1998; Renard et al., 1997; Yoder & Town, 1995). It is also
the only water erosion prediction model adopted by all states in the United States for
predicting sheet and rill erosion (USDA NRCS, 2000).
7
R - Rainfall/Runoff Erosivitv
The R factor in RUSLE represents the rainfall/runoff erosivity and reflects the
climatic contribution of precipitation to soil loss. The rainfall factor reflects the
relationship between total storm kinetic energy (E) times the maximum 30-minute
intensity (I30) (Toy & Foster, 1998). Volume of rainfall and runoff specify the storm
energy (E). Prolonged peak rates of detachment and runoff are accounted for in the (I30)
component. Total energy and peak intensity are combined in each particular storm to
produce the statistical interaction product term El, which is an abbreviation for energy
times intensity. Technically, the term indicates how particle detachment is combined
with transport capacity (Renard et al., 1997). R is equal to the average annual sum OfEI30
for storm events during a rainfall record of at least 22 years (Toy & Foster, 1998). When
erosion is dominated by spring thawing and snowmelt, an equivalent R value (Req) that
accounts for these processes is used (Toy & Foster, 1998;'Renard et al.', 1997).
K- Soil Erodibilitv
The soil credibility factor is a numerical value representing the average, long-term
susceptibility of soil and soil profile to a large number of erosive and hydrologic
processes (Renard et al., 1997). The K factor lumps the soil and soil profile reaction to
thebe processes into an integrated average annual value (Renard et al., 1997). The
RUSLE software can vary K values seasonally to account for temporal variability in the
processes (Renard & Ferreira, 1993). These processes consist of soil detachment and
transport by raindrop impact and surface flow, localized deposition due to topography
and tillage induced roughness, and rainwater infiltration into the soil profile (Renard et
8
al., 1997). If the soil is Undisturbed, K values can be obtained from published NRCS Soil
Survey data. Otherwise, RUSLE software will calculate an estimated K using a soilerodibility nomograph. This nomograph combines a series of equations that estimate K
based on texture (percent sand, silt, and clay), percent organic matter, soil structure, soil
permeability class and percent coarse fragments. Although RUSLE K factor was
specifically developed for soil properties equivalent to tilled agricultural soils, it is
appropriate on reclaimed soils because the handling and management of soil material on
disturbed sites often results in equivalent soil properties (Renard et ah, 1997).
v LS - Hillslope Length and Gradient
The LS factor in RUSLE is a combined parameter integrating length and gradient
of a hill. Soil loss increases as both slope or length increases because runoff accumulates
and accelerates downhill (Renard et al., 1997). The erosive force and velocity of water
increases with increasing slope (Toy & Foster, 1998). The ratio of rill (concentrated) to
interrill (diffuse) erosion on the hillslope is used to determine the effect of hillslqpe
length on soil loss, and is high for silty and recently disturbed soils and low for clayey
and sandy soils. It is a function of soil texture and general land use (Toy & Foster, 1998).
The hillslope length factor L has a value of I for a "unit plot" which is defined as
72.6 feet in length with a gradient of 9 perbent (Renard et al., 1997). The L value is less
than I for hillslope lengths less than 72.6 feet and greater than I foi; lengths greater than
72.6 feet. If soil loss results from interrill erosion, which is assumed to be uniform along
a hillslope, the L value will be I for all lengths. If rill erosion is the main process, the L
9
factor will increase linearly with length because rill erosion increases in the downslope
direction as runoff accumulates (Toy & Foster, 1998).
The hillslope gradient factor, S, reflects the effect of hillslope-profile gradient on
soil loss. For a unit plot, with a 9% gradient, the S value is equal to I. The S values vaiy
from above to below I, depending on whether the gradient is greater than or less than that
of the unit plot. Soil loss increases more rapidly as gradient increases than as length
increases. Also, rill erosion is affected more by hillslope gradient than is interrill erosion
(Toy & Foster, 1998).
Withip RUSLEi the hillslope length (L) and gradient (S) terms are combined into
a single topographic factor (LS) representing the ratio of soil loss from a given hillslope
length and gradient to soil loss from the unit plot (72.6 feet in length, 9% gradieht).
Thus, LS values are not absolute values but are based upon a value of I for unit plot
conditions. Because land use has a large impact on rill erosion, it is as important to select
the proper RUSLE land use categoiy as it is in determining hillslope length and gradient.
The RUSLE software will calculate the LS factor based upon assumptions that the rill to
interrill ratio is low, moderate or high for a given land use selected.
C - Cover-Management
The cover-management factor (C) represents the effects of vegetation and
management on soil loss. As with other RUSLfe factors, the C value is a ratio comparing
the existing surface conditions at a site to the standard conditions of the unit plot.
The C factor represehts the effect of plants, soil covers, foots, incorporated
residue, and soil-disturbing activities on soil loss. RUSLE offers a time-variant or time-
10
invariant option. The time-variant option is tp model situations where changes in soil and
vegetation are anticipated to greatly affect erosion. The time-invariant scenario is used to
estimate erosion on a stable landscape.
Fouf subfactors are normally used to estimate the C value: prior land use, canopy
cover, surface cover and surface roughness. In the Northwest Wheat and Range Region
of the United States, an additional subfactor representing antecedent moisture is added.
The C-factor is one of the most important factors in the RUSLE equation because
it represents surface soil conditions that can be manipulated by land managers to prevent
erosion, and tjie numerical C value calculated based on the above sub-factors can range
from almost Oto a little more than one, thus having a large Weighted value on the total
estimated annual soil loss (Toy et al., 1999).
P - Support Practices
The support practice factor (P) represents erosion control practices such as
contouring and terracing that reduce erosion (Renard et al., 1997). The P sub-factors that
are multiplied together to estimate an overall P value are based upon whether a timevariant or time-invariant option was selected when computing the C value. Ifa timeinvariant scenario is being modeled, the sub-factors used to calculate P are contouring
and other mechanical disturbances. If the time-variant option is chosen, P is calculated
based on contouring, permanent barrier strips, concave hillslope shapes, terracing and/or
sediment basins and subsurface drainage. The RUSLE model is able to assess the
effectiveness of the various support practices by weighing their effectiveness against
11
information listed in other factors. For example, contours are less effective where
rainfall/runoff erosivity is high (Toy & Foster, 1998).
The sediment-delivery ratio is associated with the terracing subfactor (Renard et
al., 1997 and Toy & Foster, 1998). When sediment productidn in the inter-terrace
interval exceeds the transport capacity of the flow in the terrace channel, deposition
occurs and the sediment-delivery ratio is less than I . When the transport capacity equals
or exceeds the soil loss, the sediment-delivery ratio equals I, indicating that all of the
sediment is removed from the hillslope by the channel flow. The soil loss estimated by
RUSLE can be multiplied by the sediment-delivery ratio to estimate the amount of
sediment leaving the hillslope. The same principles are used to estimate the sedimentdelivery ratio for concave hillslope profiles. If sediment ponds or basins are used to
retain sediment on-site, then the soil loss from hillslopes can be multiplied by the
sediment-delivery ratio to estimate the sediment discharged into a sediment pond (Toy et
al., 1999).
Improved Accuracy Representing Effect of Slone Steepness
RUSLE experts consider the slope steepness factor, S, to be moderately accurate
for slopes over 20%. In a study done by Nearing (1997), the RUSLE functions used for
the effect of slope steepness on soil logs by water were linear functions of the sine of the
slope anglfe. Two linear functions are used in RUSLE: one for slopes <9% and another
for >9%. By using the original data used to calculate the current linear functions used by
RUSLE along with what was considered to be the best data for steeper slopes, a single,
12
continuous logistic function.was derived that he contends is equivalent to the current
functions in RUSLE for slopes <25% and is better for slopes >25% (Nearing, 1997).
RUSLE and Oeomomholoev
Toy and Osterkamp (1995) investigated the applicability of RUSLE to
geomorphological studies because of anticipated use of the model for these purposes despite the fact that analysis at geomorphic scales are outside of the model's intended
scope. They indicate that soil loss estimates may be extended into the past as long as the
envirohmental conditions remain virtually the same as those used in the computations.
Soil loss estimates are likely to be of satisfactory accuracy at spatial scales ranging from
landscape (hillslope) profiles to small drainage basins where channel processes of
aggradation and degradation are insignificant (Toy & Osterkamp, 1995).
Database Sensitivity of RUSLE
Renard and Ferreira (1993) performed a sensitivity analysis of the three databases
used in the RUSLE modeling software. They compared the percent change in a
parameter to the resulting percent change in predicted soil loss. In their examination of
the CITY database, which represents geographic location and associated climate, the
RUSLE model was very sensitive to changes in the city codes. The CITY database
contains the rainfall/runoff (R) values for locations throughout the United States.
Temperature values were found to be significantly more important in estimating annual
soil loss values than precipitation. This was attributed to the effect of temperature on
residue decomposition. The authors stress that sensitivity will vary geographically and
13
the RUSLE may react in unpredictable ways. Sensitivity analysis should be performed
based on local modeling situations, and can be a very helpful tool when allocating
resources for field data collection (Renard & Ferreira, 1993).
Use of RUSLE on High Elevation. Steep Slones
Kapolka and Dollhopf (2001) calculated a rill formation factor using nonlinear
variable estimation (Kapolka & Dollhopf2001). An adjusted soil credibility factor (Kl)
is calculated by multiplying the RUSLE estimated K value by a rill formation factor (F
Factor) to obtain an optimized soil credibility factor, K l. The F Factor is; 1.0 if slopes
have stable to slight rilling, 8.4 for slight to moderate rilling, and 16.6 for moderate to
critical rill severity. Kl is then multiplied by the other RUSLE generated factors on a
spreadsheet to obtain the optimized sediment yield value.
Opportunities for research include calibrating the RUSLE model for use on
reclaimed lands, testing the model against applications for which it was not designed
(such as mine spoil piles), investigating process-based relationships between the factors
as opposed to the empirically-basfed relationships used now and creating methods that
standardize the measurement of variables. Additional areas that would assist disturbed
land reclamation efforts are furthdr testing of erosion control measures for input into the
P factor.
RUSLE is the most widely used model for estimating average annual soil loss
because of its accuracy and flexibility (Yoder & Lown, 1995). In Montana, many m in in g
companies and environmental consulting firms are using RUSLE version 1.06 software to
estimate potential erosion resulting from mining activities. RUSLE and other erosion
14
models are being used in a variety of settings by state and federal regulatory authorities
as well. The estimated annual soil loss estimated by the RUSLE model may be a part of
the information used for approving final reclamation plans, pre-and post-mining permits,
and reclamation bond release. Use of models requires expertise and a full dnderstanding
of inputs, outputs, and a sound background when science-based estimates are required
using variable data is needed or applying RUSLE to a new setting. It is important for
both industry and the regulatory authorities to understand the intricacies and limitations
of the erosion models.
Commercial Mycorrhizal Fungi Inoculations in Mineland Reclamation
Mycorrhizae are a plant-fungal symbiosis found in possibly 95 % of the world’s
plants (Smith & Read, 1999). The mycorrhizae fungus receives carbon from the plant
and in return enhances plant uptake of nutrients, particularly phosphorus and trace
minerals (Smith & Read, 1999).
Factors related to mining practices that decrease or eliminate a viable population
of mycorrhizal fungi are i) removal of vegetation; ii) topsoil storage; arid, iii) disturbance
of the soil (Jasper et al., 1987; Miller & Jastrow, 1992). Topsoil stockpiling can reduce
the density of mycorrhizal fungi in the soil (Rives et al., 1980; Gould & Liberia, 1981;
Liberia, 1981, Abdul-Kareem & McRae, 1984), depending on length of time soil is
stockpiled and soil moisture content. This reduction in mycorrhizal fungi can be
detrimental to not only the grass, forb, and shrub seedling establishment (Visser et al„
15
1984; Stark & Redente, 1987; Jasper et al., 1989), but also plant community function
^specially in nutrient or moisture limited environments (Miller & Jastrow, 1992).
Potential benefits of revegetation enhancement by the use of mycorrhizal
inoculants include the reduction of sediment pollution of surface waters, inasmuch as
surface mines are major non-point source contributors because of erosion due to plant
and soil removal. In soils containing non-ferrous metal contamination^, studies have
shown that mycorrhizal fungi inoculations can increase plant tolerance to heavy metals
and improve plant diversity (Shetty et al., 1995; Vangronsveld et al., 1996; Lambert et
al., 1979). Mycorrhizal fungi also make an important contribution to restoring lost soil
structure by the formation and stabilization of soil macroaggregates, which have an
influence on soil quality and erosion, mobility of hazardous chemicals and remediation of
contaminated sites (Jastrow et al., 1998).
16
3. METHODOLOGY
Site Description
The Treasure Mine open pit talc mine is located in southwestern Montana
approximately 24 kilometers northeast of Dillon, Montana, in the Ruby Mountain Range
(Figure I). Native vegetation is sagebrush grassland on south facing slopes and
lodgepole pine/Douglas fir on slopes facing north. Elevation is approximately 2590
meters. Average annual precipitation, based on 20 years of record, is 26.0 cm. Average
yearly maximum temperature is 14.8° C, minimum temperature is -1.3° C with 95
freeze-free days per year (Western Regional Climate Center, 2001).
Glendive
.Missoula
Helena
Billin1
Dillort ^
Miles City
■Ennis
Treasure Mine
Figure I. Location of study site in southwest Montana.
17
Experimental Design
The experiment was constructed on a regraded slope having a uniform slope
gradient of 25 %. The five following biological, mechanical, and soil depth treatments
were tested:
i)
Control: Spoil backfill material graded to 25 % slope gradient.
ii)
Coversoil: 30 cm thick coversoil application placed over spoil backfill material
graded to a 25 % slope gradient,
iii)
CoversoiIZPitting: 30 cm thick coversoil application placed over spoil backfill
material graded to a 25 % slope gradient, with pitting. Pit dimensions were
approximately 900 cm3 and dug by hand. Pits were staggered in a checkerboard
pattern with alternating rows of two and three pits. Distance between pits was
3.39 m.
iv)
Coversoil/Slash Barriers: 30 cm thick coversoil application placed over spoil
backfill material graded to a 25 % slope gradient, with four slash barriers (3.1 m
wide by 0.8 m long) installed every 6.8 m along the slope. The slash barriers
were constructed with lodgepole pine branches measuring 8 cm or less diameter
cut fresh from nearby trees and anchored tp the soil surface with wire and stakes.
Fine branches and needles were allowed to remain on the branches. Branches
were staked flush to the soil surface to prevent undercutting of the barriers by
water.
v)
30 cm coversoil application placed over spoil backfill material graded to 25 %
18
slope gradient, with commercial mycorrhizal fungi inoculation. Pelletized AM
inoculum provided by PHC Reclamation, Inc., was applied at 37.0 liters per
hectare.
Each treatment was replicated three times in a completely random experimental
design for a total of fifteen test plots (Figure 2). Each plot was 3.1 meters wide and 30.5
meters long and was bounded laterally by a silt fence to prevent sediment and runoff
from flowing onto adjacent plots. A diversion ditch was constructed above the test plots
to prevent upslope runoff from entering the plots. Troughs were installed at the base of
each plot to collect runoff and sediment.
Sediment Collection Trough Design
Sediment and runoff from eroding experimental plots was captured by a
collection trough, recessed into the ground so that the upper lip of the trough was level
with ground surface. Each trough was 3.1m long by 1.2 m wide by 0.6 m deep. Trough
capacity was designed to capture approximately 25 % of the runoff from a 24 hour - 100
year precipitation event, assuming no infiltration. The transitions between treatment
plots and collection troughs were packed with bentonite clay to-minimize, undercutting
between slope and trough.
SeedMix and Seeding Procedures
Test plots were seeded by a Barretts Minerals contractor on June 24,2000. Seed
was broadcast by hand and raked into the surface. Table 2 presents the seed mix and
19
Individual Plot Description (not to scale). All plots have a 25 % slope. Plots were
3.1m wide and 30.5 m long (10 ft x 100 ft). Sediment and water moving off plots
were captured in a collection trough buried in the ground at the toe of the slope.
Silt fences were installed on the top and sides of the plots.
Upslope
Silt fence
3.1 m
30.5 m
Downslope
Collection trough buried
beneath ground surface
Completely Random Experimental Design (not to scale). The fifteen test plots drawn
below represent the five treatments (A - E), each replicated three times.
Treatment Kev:
A = 30 cm Coversoil
B = 30 cm Coversoil/AM Inoculum
C = 30 cm Coversoil/Slash Barriers
D = 30 cm Coversoil/Pitting
E = No Coversoil
Runon diversion ditch
P \ B \ E \C \ E \ A \ I ^ B \ C \ D \ A \C \ A \ E \ B
Collection troughs
Figure 2. Design and layout of test plots at the Treasure Mine, MT.
20
Table 2. Seed mix and application rate on test plots.
Species
C om m on nam e
A g r o p y r o n d a systa ch yu m
A g ro p y ro n sp ica tu m
Thickspike wheatgrass
Bluebunch wheatgrass
Slender wheatgrass
Sheep fescue
Canada bluegrass
White yarrow
Alfalfa
Sanfoin
Barley
A g ro p y ro n trach ycau lu m
F e stu c a o vin a
P o a c o m p re ssa
A c h ille a m illefoliu m
M e d ic a g o s a tiv a
O n o b ryc h is v ic ia e fo lia
H o rd eu m vu lg a re
application rates. An annual barley species, H o rd eu m
P lant type
Pure L ive Seed
(PL S)
(kg/ha)
Grass
Grass
Grass
Grass
Grass
Forb
Forb
Forb
Grass
Total
vu lg a re,
10.5
11.6
9.9
2.7
2.9
0.6
2.9
8.5
24.2
73.9
was included with the
perennial seed mix to provide rapid cover. All plots were broadcast fertilized with 36kg/ha nitrogen and 36-kg/ha phosphorus after the test plots had been constructed.
Commercial AM Inoculum Application Procedures
The AM inoculum was the pelletized form and applied at 37.0 liters per hectare.
The commercial AM inoculum was provided by PHC Reclamation, Inc. Treatment plots
were inoculated by broadcasting the inoculum onto 6 m long x 3 m wide subsections and
then covering the pellets with 5 cm of coversoil. One hundred milliliters of AM
inoculum was applied to each subsection, for a total of 500 mL of AM inoculum per
treatment plot.
21
Data Collection and Analysis
Spoil Backfill and Coversoil Phvsiochemical Analysis
Soil physiochemical characteristics were determined by collecting composite
samples of the spoil backfill test plots (Control treatment) and plots with coversoil
(Coversoil, Coversoil/Pitting, Coversoil/Slash Barriers, Coversoil/AM Inoculum
treatments). Composite soil samples were oven dried at 41° C and disaggregated using a
mortar and pestle. Analytical procedures used are reported in Table 3.
Precipitation and Evaporation Measurements
An on-site precipitation-recording gage located next to experimental plots
monitored precipitation. A solar powered Campbell Scientific CR-IO datalogger
recorded precipitation data on an hourly basis and in one-millimeter increments. Hourly
measurements gave an indication of the intensity of various precipitation events.
Precipitation data are reported in Appendix B, Tables 27 - 38.
A Class I evaporation pan was installed adjacent to the precipitation gage to
measure evaporation at the site. A 200-liter capacity stilling well was connected to the
Table 3. Soil physiochemical analytical methods.
V ariab le
A nalytical T ech n iq u e
Particle size distribution
Day 1965. Hydrometer method.
Coarse fragment percentage
Sieved 2 mm fraction, measured weight and volume.
Electrical conductivity, pH, and Rhoades 1982. Water saturated paste extract.
sodium absorption ratio
Organic matter percentage
Nelson and Sommers 1982. Walkley-Black method.
22
evaporation pan by a hose and a constant water level in the evaporation pan was
maintained using a float and valve check. The volume of water evaporated each hour
was recorded using a Stevens recorder located in the stilling well and used to calculate
evaporation (cm). Evaporation data are reported in Appendix B, Table 39.
Runoff Collection Methods
Rundff accumulated in collection troughs was measured every two weeks from
late spring through early fall. Depth of water in the troughs was measured to the nearest
millimeter. To convert depth of water to runoff volume in liters, Equations 1-3 in Figure
3 were used to calculate the area of the cross section of wafer from the known variables
of trough diameter and water depth. Runoff data are reported in Appendix C, Tables 4243.
Precipitation falling into the trough and evaporation of water from the trough
were accounted for using data collected from the precipitation gage and the evaporation
pan (Appendix B, Tables 27-39). Total maximum amount of runoff was calculated using
Equation 4.
Total Runoff = (Initial Water in Trough) - (Precipitation) + (Evaporation)
Equation 4
Sediment Collection Methods
Accumulated sediment in collection troughs was measured and collected every
two weeks from late spring to early fall during 2000 and 2001. After water depth was
measured, a submersible pump powered by a gasoline generator was used tp remove the
23
Calculation for area of a segment of circle (the area of the trough occupied by water) of
depth y.
'd _
6
2 cos 1 2
d
< 2
Equation I
y
y4(m2) = — (<9-sin 6>)
8
Equation 2
where:
d = trough diameter (m)
y = water depth in trough (m)
A = area of water in trough (m2)
Multiply by trough length to obtain volume:
Volume (m3) = Area x length
Equation 3
Figure 3. Calculation to convert water depth in trough into volumetric measure.
24
water from the troughs. Sediment was collected in 19-liter buckets using spades and
shovels. Samples less than 22 kilograms were transported back to Montana State
University, oven-dried at 41° C and weighed. If there was greater than 22 kilograms of
sediment in the trough, the total volume of sediment was measured and the mass of dry
soil was determined using known mass values for a given volume of saturated sediment.
Sediment yields are reported in Appendix C, Tables 40-41.
Rill Classification Methods
Rill severity on experimental plots was classified during every monitoring cycle
of the study period using the Erosion Condition Classification, Montana Revised Method
(Clark, 1980). Depth, width, and frequency of rills are the criteria used to diagnose
severity (Table 4). Rill classification data for test plots are reported in Appendix C,
Tables 44-45.
Table 4. Soil erosion condition classification for rills.
Q u alitative R an k in g
D escription
R ank
Stable
Rills, if present, are less than 0.5 inch deep, and
generally at infrequent intervals over 10 feet.
0 or I
Slight
Rills are mostly 0.5 to I inch deep, and generally at
infrequent intervals over 10 feet.
2
Moderate
Rills are mostly I to 1.5 inches deep, and generally
at 10 feet intervals.
3
Critical
Rills are mostly 1.5 to 3 inches deep, and at intervals
of 5 to 10 feet.
4
Severe
Rills are mostly 3 to 6 inches deep, and at intervals
of less than 5 feet.
5
25
Plant Measurements .
During 2001, plant canopy cover, basal cover and aboveground plant biomass
(biomass) were measured during the peak of the growing season. All three variables
were analyzed by growth form (i.e., annual or perennial grasses, forbs). Plant growth
measurements are reported in Appendix D, Tables 46 ^ 57.
Transect locations were staked on all plots. These transects were along the
diagonal, running from the upper left-hand comer to the lower right hand comer of each
plot. Ten quadrats were sampled on each test plot to measure plant cover and biomass.
Quadrats were placed every 3 m beginning at the 3 m point along the transect.
Canopy and basal cover were estimated using a 20 x 50 centimeter Daubehmire
frame. Cover whs estimated by growth form, i.e., perennial grass, annual grass, or forb.
These cover values were averaged across quadrats to determine plant canopy and basal
cover. Rock fragments, litter, and bare ground were also measured in terms of cover. All
cover measurements were classified using a scale of 0 - 10 where 0 = no cover, 1 = 1 10 % cover, 2 = 11 —20 % cover and so. forth.
Biomass was measured using a 20x20 centimeter frame on the same quadrants
used to measure cover. Vegetation within the frame was clipped two centimeters above
the ground, sorted by growth form, and placed into paper bags. These samples were
oven-dried at 49° C to a constant weight. Vegetation was weighed and biomass for each
life form was calculated on a kg/ha basis.
26
Pre-Inoculation AM Colonization Levels of Goversoil and Spoil Rack-fill Material
A greenhouse study was conducted to determine whether propagules of
mycorrhizal fungi were present in the coversoil and spoil backfill material prior to
application of the commercial AM inoculum. Bulk composite samples were collected of
the coversoil and spoil backfill material with a hand trowel every 3 meters along the
slope. Hand trowels were thoroughly cleaned, rinsed and dried between plots. Each bulk
composite sample consisted of 10 scoops from each treatment replication. Soils were
brought to the Montana State University Plant Growth Cfenter and planted with Sorghum
Sudanese (Sudan grass). Fifty seeds per pot were planted 2 cm deep and kept moist with
a water mister until plants emerged. After germination, plants were culled to five per pot,
and allowed to grow for 90 days, with a 14-hour photoperiod, a daytime temperature of
2 10C and a nighttime temperature of 18 0C. Roots were harvested and analyzed for
presence o f-a n d percent colonization of - AM fungi (Appendix D, Table 58).
Percent Colonization by AM Fungi Procedures
Percent of root length colonized by AM fungi was determined after clearing and
staining root samples. Roots that had been washed free of soil were cut into 2 cm
segments, cleared with 1 5 % KOH solution for 48 hours, soaked in HCl for 12 hours and
stained with 0.05 % trypan blue stain in lactoglycerol. Two slides with twelve root
segments per slide were made for every plant sample. Transects recording presence or
absence of mycorrhizal structures were conducted across the roots. Ninety-six
observations of root segments per plant were observed with 200-x magnification. The
27
presence of AM hyphae, vesicles, arbuscules, or non-AM hyphae was recorded for each
intersection. Total colonization was calculated using Equation 5.
% AM Colonization
(intersections with hyphae, vesicles, or arbuscules)
(total intersections)
Equation 5
Mvcorrhizal Infection Study
Vegetation and the AM colonization levels were sampled during the second year
of plant growth (August, 2001). Four plants o f Agropyron trachycaulum and Hordeum
vulgare were collected to a 30 cm depth from the Coversoil, Coversoil/AM Inoculum and
Control treatment plots. Plants were transported in cold-storage to Montana State
University laboratory facilities for analysis. Roots were examined to determine if
mycorrhizal fungi were present. Percent AM colonization o f A. trachycaulum and H.
vulgare are reported in Appendix D, Tables 59-60.
Statistical Analysis
Analysis of variance techniques and mean separation tests were used to ascertain
whether significant differences were present at the 95 % level of confidence (P = 0.05).
Significant differences at P < 0.05 were separated using the Student-Newman-Keuls
method of pairwise multiple comparison for equal size data sets. Least-squares
regression was used to evaluate associations between independeht and dependent
variables. Multivariate associations were tested using multiple linear regression analysis.
These analyses were made using SigmaStat version 2.0 software (Jandel 1995).
28
RUSLE v.1.06 Computer Model Analysis
RUSLE version 1.06 is a DOS computer model (Galetovic, 1998). Various input
variables were required for each factor in the RUSLE model (Table 5). Using
mathematical equations, RUSLE estimated an average annual sediment yield. Input
values were either from field data, from Renard et al. (1997), or were provided by the
United States Department of Agriculture Natural Resource Conservation Service State
Agronomist (Fasching, 2000).
Table 5. Input variables for RUSLE version 1.06.
R U S L E F actor
In p u t V ariab le
R - Rainfall/Runoff
Erosivity Factor
City climate database*
D ata Source
Renard et al. (1997),
Field
Initial R value
Field
K - Soil Erodibility
Rock cover %
Field
Number of years to consolidate
Renard et al. (1997)
Hydrologic group
Renard et al. (1997)
Surface texture (% clay, % silt)
Field
Organic matter %
Field
Soil permeability class
Renard et al. (1997)
Coarse fragment %
Field
LS - Slope length and
Number of hill segments
Field
Gradient
General land use
Field
Slope gradient
Field
Slope length
Field
C - Cover Management
Effective root mass
Fasching (2000)
Canopy cover %
Field
Fall height of precipitation
Field
Roughness of field condition
Field
Ground cover %
Field
Rock cover %
Field
*City climate database included the following values: storm energy and intensity (EI)
curve value, 10 year EI value, an initial R value, number of freeze-free days per year, site
elevation, mean monthly temperature, and mean monthly precipitation. All data, except
for elevation and monthly precipitation, were obtained from Renard et al. (1987).
Elevation and monthly precipitation data were collected in the field.
29
Hillslope Area Calculations
Rill severity class and hillslope area were used as multiple linear regression
(MLR) independent variables to predict Year 2000 sediment yields and runoff volumes.
Plant growth was negligible during 2000 and thus plant growth effects were omitted from
the regression model. In Year 2001, rill severity class, hillslope area, plant biomass, and
percent plant cover were used as the MLR independent variables of sediment yields and
runoff volumes. The Coversoil/Slash Barrier and Coversoil/Pitting treatments provided
barriers to movement of water and sediment downslope into the collection trough, so the
surface area of these experimental plots contributing to sediment yield and runoff were
smaller than the Coversoil, Coversoil/AM inoculum and Control treatments. Because of
this, the hillslope area variable was added to the MLR model. The total area of the
Coversoil/Slash Barrier plots contributing to sediment yield and runoff was Palculated
using height as the distance from the bottom of the lowest barrier on the test plot down to
top of the collection trough multiplied by the plot width (6.8 m x3.1 m = 21.0 m2). Area
of Coversoil/Pitting treatment plots contributing to sediment yield was the total area of
the test plot between the bottom of the plot up to the nearest pit (Figure 4). Total area of
the pitted plots contributing to sediment yields was 18.9 square meters. The area of the
Coversoil, Coversoil/AM inoculum and Control plots were 94.55 square meters, or the
area of the entire plot.
30
▲
O
O
O
O
O
O
O
O
Total plot length = 30.5 m
O
a
O
O
O
O
O
O
O
O
P O
Q_
A
6.77 m
a
a
3.39 m
v
r
◄-------- ►
0.62 m
< —--------------------------------Total plot width = 3.1 m
Figure 4.
►
Schematic of dimensions used to calculate area contributing to runoff and
sediment yield on pitted treatment plots. Drawing not to scale.
31
4. RESULTS AND DISCUSSION
Coversoil and Spoil Backfill Phvsiochemical Characteristics
Coversoil applied to experimental plots was taken from a nearby stockpile of
native coversoil recovered during mining operations at the Treasure Mine. Coversoil
texture was sandy loam (Table 6). Chemical characteristics of the coversoil were all
within ranges that are not limiting to plant growth (pH = 6.8, electrical conductivity (EC)
= 0.9 mmhos/cm, sodium absorption ratio (SAR) = 0.1). Coversoil characteristics that
could impair vegetation establishment and growth were the high coarse fragment
percentage (49 % by weight and 38 % by volume) and low percentage (0.6) of organic
matter.
The spoil backfill texture was sandy loam. Coarse fragment percentage was 64 %
Table 6. Coversoil and spoil backfill physiochemical characteristics.
Soil P roperty
C oversoil
Spoil B ack fill
Sand percentage^
Silt percentage*
Clay percentage*
Textural class*
Coarse fragment percentage (weight)*
Coarse fragment percentage (volume)*
pH#
EC (mmhos/cm/
SAR*
Organic matter percentage*
* n=l
+ n=2
# n=3
1 mean ± I standard deviation
65.4 ± 0.0V
19.0+ O-Ov
16.6 ± 0.0v
sandy loam
49
38
6.8 ± 0.05 v
0.9 ± 0.02v
0.1
0.6
63.5 ± 0.0
15.7 ±0.0
20.8 ± 0.0
sandy loam
64
51
7.4 ± 0.03
1.2 ±0.01
0.8
0.1
32
by weight and 51 % by volume, and organic matter was 0.1 %. Chemical characteristics
of the spoil backfill (pH = 7.4, EC= 1.4 mmhos/cm, SAR = 0.8) were suitable for plant
growth.
Sediment Yields on High Altitude Steep Slopes
Total annual sediment yield during 2000 was low (mean = 0.11 Mg/ha, n = 15),
and increased during the second year (mean = 1.17 Mg/ha, n = 15). Figure 5 is a
comparison of mean annual sediment yield by treatment for the study period. The low
■Year 2000
□Year 2001
-s2
.y
>
S
E
•g
CA
23
I
§<L>
Coversoil
Coversoil/Pitting
Coversoi 1/Slash
Barriers
Coversoil/AM
Inoculum
*
Control
Mean of three replications.
Bars are one standard deviation
Figure 5. Comparison of mean annual sediment yields by treatment during 2000 and
2001.
33
sediment yield for the first year was a result of infrequent, low-intensity precipitation.
Total precipitation in 2000 was 24 % below normal annual rainfall at the Dillon WMCE,
Montana recording gage. Sedimeht yield increased during the second year although total
rainfall amounts in 2001 were 36 % below normal average annual rainfall at the Dillon
WMCE, Montana recording gage. Greater sediment yield the second year was the result
of two strong Summer storm events during July and August, 2001. These two
precipitation events generated more sediment yield than any other events during the twoyear study period. Due to below normal precipitation, vegetative growth was negligible
during the fust year of the study and developed to an average of 15 % plant canopy cover
in 2001. Although the soil surface was largely bare of vegetation during the first year,
sediment yield was higher during the second year of study due to more erosive
precipitation. The vegetative cover during the second year of the study was still not well
developed.
Effect of Mechanical and Biological Erosion Control Measures on Sediment Yield
Sediment yields were significantly lower on the CoversoilZPitting treatment at the
10 % probability level of confidence during 2000 (Table 7). There were no significant
differences between treatments during either year of study at the 5 % probability level.
Sediment yields measured were highly variable except on pitted slopes. This variation
could be attributed to rill formation that developed on some plots. Lower sediment yields
measured on pitted slopes were attributed to minimization of rill formation on the soil
surface. Rill formation provided a conduit for concentrated flow of water across the soil
surface, increasing the velocity and erosivity of water moving downslope.
34
Table 7. Mean* sediment yield by treatment during 2000 and 200 L+
T reatm ent
Y ear 2000
M ean* S ed im en t Y ield (M e/h a)
Y ear 2001
Pm
P..o~
a
b
a
a
a
Coversoil
0.15
1.68
a
Coversoil/Pitting
0.04
0.21
a
Coversoil/Slash Barriers
0.09
1.01
a
Coversoil/AM Inoculum
0.08
1.60
a
Control
0.21
1.34
a
* Mean of three replications.
Means followed by the same letter in the same column are not significantly different.
Test of significance is for P = 10 %. Treatment differences were not significant at
P = 5 %.
Effect of Rill Severity Class on Sediment Yield
In 2000, mean rill severity varied from stable to slight and there were no
significant differences between any of the treatments (Table 8). In the 2000 field season,
sediment yield was dominated by sheet erosion and slight rilling and sediment yields
were low. The Coversoil/Pitting treatment had significantly lower rill severity ratings
than all other treatments during 2001.
Rill severity class was strongly related to sediment yield both years (Figure 6). In
Table 8. Mean* rill severity class** by treatment during 2000 and 200L+
M ean* R ill S everity
T reatm ent_____________________________ Y ear 2000
Y ear 2001
Coversoil
1.7 a
3.5 a
Coversoil/Pitting
1.0 a
1.5 b
Coversoil/Slash Barriers
1.3 a
2.5 a
Coversoil/AM Inoculum
1.3 a
2.8 a
Control
1.3 a
3.2 a
* Mean of three replications.
** Class: I = stable, 2 = slight, 3 —moderate, 4 —critical, 5 = severe.
Means followed by the same letter in the same column are not significantly different.
35
2
Year 2000
<u
(Mg/ha)
P = <0.01
c
U
P
Mean* Annual Rill Severity Class
Year 2001
(Mg/ha)
P = <0.01
cC
<
I
Mean* Annual Rill Severity Class
* Mean of rill severity class individual plots received at every twoweekmonitoring cycle throughout the season (n = 15).
(Rill Severity Class: I=Stable, 2=slight, 3 = moderate, 4 = critical, 5 = severe)
Figure 6. Strength of relationship of mean rill severity class to total annual sediment
yield during 2000 and 2001.
36
2000, all test plots exhibited final rill severity classes of mostly stable or slight, which
was showed a strong linear relationship with total annual sediment yield (r = 0.76). In
2001, total annual sediment yield showed a strong exponential relationship with mean rill
severity class (r = 0.87). Test plots that were assigned with a rill severity class of
moderate to critical showed an exponential increase in sediment yield at the point when
rill severity class exceeded moderate levels. These data indicate that once rill formation
occurred beyond the moderate rill severity class, sediment yields were accelerated.
Effect of Precipitation on Sediment Yield
Total precipitation, maximum daily, and maximum hourly precipitation and mean
sediment yields measured for all plots during each two-week period are presented in
Table 9. These results were obtained by calculating the mean total sediment yield
collected at the end of each two-week monitoring period for comparison to on-site
precipitation gage records.
Table 9. Mean* sediment yield and precipitation across all treatments for two-week
periods during 2000 and 2001.+
D ate
S ed im en t
Y ield
(M g/ha)
T otal
P recipitation
(cm )
8/2/00
0.019
9/16/00
0.020
10/19/00
0.073
7/21/01
0.376
8/5/01
0.791
9/18/01
0.003
* Mean of fifteen replications.
Only sediment yield events are reported.
0.1
0.7
2.4
3.8
0.9
3.3
M axim um
D aily
P recipitation
(cm )
M axim um
H ourly
P recipitation
(cm )
0.0
0.2
2.3
0.9
0.7
2.0
0.0
0.1
1.7
0.7
0.4
0.9
37
Sediment yield increased with increasing amounts of total precipitation received
during the two-week monitoring cycle in 2000 (Figure I). In 2001, sediment yield did
not increase with increasing amounts of precipitation.
Duration of precipitation and mean sediment yield in the year 2000 was strongly
positively related to sediment yield for maximum daily and maximum hourly
precipitation but not in 2001 (Table 10).
In 2001, sufficient precipitation accelerated the rilling processes on test plots
(Table 8). Rainfall in July of 2001 generated considerable sediment yield and rilling of
the soil surface. During the two-week period ending July 21, 0.7 cm of precipitation fell
on July 8, 0.9 cm on July 14, and 0.7 cm fell on July 16. Marked increases in the severity
of existing rills along with the formation of new rills on plot soil surfaces were recorded
when the sediment was collected on July 21,2001, yielding a mean sediment yield of
0.376 Mg/ha. Intense precipitation occurred again in the August 5 two-week monitoring
cycle, which produced less total precipitation overall, but the maximum daily amount of
Table 10. Strength of relationship of total precipitation, maximum daily precipitation,
and maximum hourly precipitation to mean* sediment yield across all
treatments for two-week periods during 2000 and 2001.
In d ep en d en t V ariab le
D ep en dent
V ariab le
Total Precipitation
Sediment Yield
Maximum Daily
Sediment Yield
Precipitation
Maximum Hourly
Sediment Yield
Precipitation
* Mean of three replications.
C orrelation
C oefficien t
(r)
P
V alu e
C orrelation
C oefficien t
(r)
P
V alu e
Year 2000
0.97
0.17
0.99
0.04
Year 2001
-0.79
0.43
-0.91
0.19
1.00
-0.99
0.02
0.14
38
4.50
Year 2000
4.00
3.50
2.00
1.00
Total Precipitation (cm)
Mean- Sedimen1 Yield(Mgdla)
I---- 1Mean Sediment Yield (Mg/ha)
—♦—Total Precipitation (cm)
0.50
7/19/00 - 8/2/00
9/2/00 - 9/16/00
10/4/00 - 10/19/00
• M in ™ , YieId(MgZha)
I-----!Mean Sediment Yield (MgTia)
—♦—Total Precipitation (cm)
Year 2001
»
3.50
3.00
2.00 S
1.50 15
7/7/01 - 7/21/01
7/21/01-8/5/01
9/4/01 -9/18/01
Means are an average of 15 test plots.
r^i Bars are one standard deviation.
Figure 7. Comparison of total precipitation and mean sediment yield across all
treatments for two-week periods during 2000 and 2001.
39
rainfall received was very similar to that measured in July (Table 9). The August rainfall
event was similar in intensity to the one in July when 0.7 cm of rain fell on July 30,
yielding the largest sediment yield of 0.791 Mg/ha. This larger amount of sediment from
one precipitation event reflects the increased erosion of the soil surface during active rill
formation. The mean sediment yield during the September 18 two-week monitoring
cycle was lowest overall (0.003 Mg/ha), even though maximum daily (2.0 cm) and
maximum hourly precipitation (0.9 cm) received were the highest. Precipitation received
during September of 2001 generated little or no erosion indicating that the rills had
stabilized and were acting mainly as conduits of water rather than sediment.
Effect of Rock Cover on Sediment Yield. Plant Biomass. Runoffand Rill Severity
Mean percent rock cover varied between 36 and 58 % during 2001 and the
differences among treatments were not significant. Mean percent rock cover values are
presented in Table 11 for each treatment type. Percent rock cover was not significantly
related to sediment yield, runoff or rill severity class during 2001 (Table 12). Plant
biomass was negatively related to rock cover in 2001 (r = -0.68).
Table 11. Mean* percent rock cover by treatment during 2001.+
T reatm ent
M ean R ock C over (% )
Coversoil
36 a
Coversoil/Pitting
45 a
Coversoil/Slash Barriers
40 a
Coversoi 1/AM Inoculum
42 a
Control
58 a
* Mean of three replications.
Means followed by the same letter in the same column are not significantly different.
40
Table 12. Strength of relationship of percent rock to sediment yield, biomass, runoff and
rill severity class cover during 2001.
In d ep en d en t
V ariab le
D ep en d en t V ariab le
Rock Cover (%)
Rock Cover (%)
Rock Cover (%)
Rock Cover (%)
Sediment Yield (Mg/ha)
Biomass (kg/ha)
Runoff (m3/ha)
Rill Severity Class
C orrelation
C oefficien t (r)
P V alue
0.06
-0.68
0.41
0.23
0.85
<0.01
0.13
0.40
Results from this study indicate that the decreased rock cover in the treatments
with coversoil application (36 - 42 %) were associated with better plant growth and no
significant decreases in sediment yields were provided by the increased rock cover in the
control during this study.
Runoff on High Altitude Steep Slopes
Effect of Runoff on Sediment Yield and Rill Severity Class
Runoff was not significantly different between any of the treatments during 2000
and 2001 (Table 13). Figure 8 compares mean sediment yield and runoff measurements
by treatment. Runoff was not related to sediment yield in either year (Table 14).
Table 13. Mean* annual runoff (m3/ha) by treatment during 2000 and 2001.
T reatm en t
M ean* A nn u al R u n o ff
Y ea r 2000
Y ear 2001
Coversoil
49.2 a
123.5 a
Coversoil/Pitting
39.7 a
66.0 a
Coversoil/Slash Barriers
45.3 a
119.6 a
Coversoil/AM Inoculum
43.2 a
99.8 a
Control
71.5 a
202.7 a
* Mean of three replications.
Means followed by the same letter in the same column are not significantly different.
41
250
] Mean Sediment Yield
■Mean Runoff 2000
S 1.0
100
§<L>
&
/
3P
Cf
dp
I
200 &
o
150
2000
I
II
15
50
*
0
S
I
z
250 g
] Mean Sediment Yield
2001
- Mean Runoff 2001
200 &
to
150
100
§
y 1.0
15 0.5
*
I
Z
/ Z Z
z
0°
G^
G°
y
# V
G°
* Mean of three replications.
F 1H Bars are one standard deviation.
Figure 8. Comparison of mean sediment yield and runoff by treatment during 2000 and
2001.
42
Table 14. Strength of relationship of runoff to sediment yield and rill severity class
during 2000 and 2001.
C orrelation
C oefficien t (r)
In d ep en d en t V ariab le
D ep en d en t V ariab le
2000 Runoff (m3/ha)
2001 Runoff (m3/ha)
2000 Runoff (m3/ha)
2001 Runoff (m3/ha)
2000 Sediment Yield (Mg/ha)
2001 Sediment Yield (Mg/ha)
2000 Rill Severity Class
2001 Rill Severity Class
P V alue
0.04
0.03
0.83
0.21
0.89
0.93
<0.01
0.44
Runoff was strongly related to rill severity during 2000 but not during 2001 (Table 14).
A possible explanation for the lack of relationship between rill severity and runoff in
2001 is that rilling increased on all test plots during July and August. During September,
precipitation amounts similar to the previous storms generated runoff from the test plots
but little to no sediment yield was measured (Tables 7 and 9).
Plant Growth on High Altitude. Steep Slopes
Effect of Erosion Control Measures on Plant Growth Characfenstirs
Mean biomass, canopy cover and basal cover were significantly higher on plots
with a coversoil application during 2001 when compared to the Control (Table 15).
Table 15. Mean* plant growth characteristics by treatment during 200L+
B iom ass (kg/ha)
T reatm ent
C anop y cover
B asal cover
(%)
(%)
Coversoil
185 a
16 a
6 a
Coversoil/Pitting
129 a
15 a
4 a
Coversoil/ Slash Barriers
225 a
17 a
5 a
Coversoil/ AM Inoculum
130 a
15 a
4 a
Control
27 b
7 b
I b
* Means are an average of three replications.
+ Means followed by the same letter in the same row are not significantly different.
43
However, there were no significant differences among treatments that received a
coversoil application, indicating that the Coversoil/Pitting, Coversoil/Slash Barriers, and
Coversoil/AM Inoculum treatments during the period of this study neither enhanced nor
reduced plant growth when compared to the Coversoil treatment. Plant growth
developed very slowly during the year 2000 due to low precipitation following seeding.
Therefore, no plant measurements were made.
Effect of Plant Growth on Sediment Yield, Runoff and Rill Severity Class
Plant growth was not related to sediment yield, runoff, or rill severity class in
2001 (Table 16). Although there was significantly more plant growth on the plots with
coversoil compared to the control in 2001, the increased biomass and plant cover did not
result in significantly less measured sediment yield or runoff (Tables 7 and 13). Low
precipitation during 2001 contributed to slow plant development during the second
season. Canopy cover (7 - 17%) and basal cover (I - 6%) were small (Table 15) and did
not have a significant effect on sediment yield and runoff during 2001.
Table 16. Strength of relationship of plant growth characteristics to sediment yield,
runoff and rill severity class during 2001.
In d ep en d en t
V ariab le
D ep en d en t V ariab le
Biomass (kg/ha)
Canopy cover (%)
Basal cover (%)
Biomass (kg/ha)
Canopy cover (%)
Basal cover (%)
Biomass (kg/ha)
Canopy cover (%)
Basal cover (%)
Sediment Yield (Mg/ha)
Sediment Yield (Mg/ha)
Sediment Yield (Mg/ha)
Runoff (m3/ha)
Runoff (m3/ha)
Runoff (m3/ha)
Rill Severity Class
Rill Severity Class
Rill Severity Class
C orrelation
C oefficien t (r)
P V alue
-0.29
-0.01
-0.30
-0.43
-0.41
-0.37
-0.02
-0.29
-0.26
0.29
0.99
0.41
0.11
0.13
0.18
0.95
0.30
0.34
44
Pre-Inoculation Mvcorrhizal Infectivitv Potential
To determine whether propagules of mycorrhizal fungi were present in the
coversoil and soil backfill material prior to application of commercial AM inoculum,
S orgh u m Sudanese
(Sudan grass) was planted in the coversoil and spoil backfill material
sampled after experimental plot construction but prior to the application of the
commercial AM inoculum. Mycorrhizal fungi propagules were present in the coversoil
and spoil backfill material, as evidenced by AM formation in S orgh u m roots (Table 17).
Effect of AM Inoculum on Colonization Levels in A g ro v v ro n
tra ch yca u lu m
and
H o rd eu m v u ls a r e
There were no differences in percent colonization of the roots by mycorrhizal
fungi in H.
vu lg a re
(Table 18). In A.
trach ycau lu m ,
the AM inoculum treated plots
showed significantly higher mycorrhizal colonization levels, as compared to plant roots
from the coversoil and no coversoil treat plots (Table 18). However, there were no
significant differences between any of the above-ground plant growth characteristics on
plots that received a coversoil application (Table 15). Both plant species were colonized
by AM fungi at a relatively high level, and there is probably no measurable ecological
difference between 46 % and 53 % colonization rates. With such high colonization in
Table 17. Mean*' percent pre-inoculation AM colonization levels in the coversoil and
spoil backfill material during 2000.+
T reatm ent
P ercen t A M C olon ization L evels
Coversoil
39 a
Spoil Backfill Material
30 a
* Means are an average of nine replications.
Means followed by the same letter in the same row are not significantly different.
45
Table 18. Mean* percent AM colonization levels of H ordeu m
tra ch yca u lu m during 2001.+
vu lg a re
and A g ro p yro n
P ercen t A M C olon ization L evels
Hordeum vulgare
Agropvron trachycaulum
Coversoil
34 a
46 a
Coversoil/AM Inoculum
35 a
53 b
Control
34 a
44 a
* Means are an average of nine replications
+Means followed by the same letter in the same column are not significantly different.
both AM-treated and non-AM treated plots, we would expect to see no biomass
differences between those treatments. AM inoculation in this study seems to have been
unnecessary, since AM propagules were present in the coversoil and spoil prior to AM.
inoculation. This AM inoculation treatment was not found to enhance either AM
inoculation levels or plant growth.
Ability of Measured Factors to Predict Sediment Yield and Runoff
Sediment yield was significantly related to measured factors during both years of
study (Table 19). Runoff was not significantly related to measured factors either year.
Hillslope area was added as an independent variable to the regression model to represent
pitting and slash barrier treatment effects on sediment yields
Sediment yield during both years of study was highly related to maximum hourly
precipitation, increase in rill severity class and runoff (r = 0.99). These factors were
chosen because newly constructed slopes are most vulnerable to erosion during high
intensity storms. Sediment yields during this study did not always reflect a positive
46
Table 19. Multiple linear regression analysis results using measured factors to predict
sediment yield and runoff during 2000 and 2001.
D ep en d en t
V ariab le
In d ep en d en t V ariables
P V alues o f
In depend en t
V ariables
M u ltip le
C orrelation
C oefficien t
(r)
P V alu e
for
R egression
Sediment Yield
2000
Mean Rill Severity Class
Runoff
Slope Area
<0.01
0.73
0.43
0.88
<0.01
Sediment Yield
2001
Mean Rill Severity Class
Runoff
Slope Area
Biomass
Canopy Cover
Basal Cover
Rock Cover
0.81
0.22
0.14
0.09
0.54
0.18
0.15
0.93
0.02
Runoff 2000
Mean Rill Severity Class
Slope Area
0.21
0.12
0.49
0.20
Runoff 2001
Mean Rill Severity Class
Biomass
Canopy Cover
Basal Cover
Rock Cover
Slope Area
0.59
0.56
0.65
0.68
0.96
0.85
0.56
0.72
Sediment Yield
2000 and 2001
Maximum Hourly
Precipitation
Increase in Rill Severity
Class
Runoff
0.99
0.02
0.01
0.05
0.01
linear relationship with precipitation (Figure 7). This is due to increases in rill formation
during the summer of 2001. It is presumed that during July and August, the rills were
actively moving sediment as precipitation occurred. By September, the rills seem to have
stabilized and did not move appreciable amounts of sediment when compared to July and
47
August. For this reason an increase in rill severity between events would more
effectively model the behavior of sediment yields during this study. A zero would mean
that rill severity class on slope remained stable and increase in rill severity would reflect
plots that were actively rilling and give a relative value of the magnitude. Prediction of
sediment yield during this study appeared to be estimated by maximum hourly
precipitation, runoff and rilling activity on the plots
Ability of RUSLE to Predict Sediment Yield at High Elevation Sites
During the first year when sediment yields were low, RUSLE overpredicted mean
sediment yields by 0.2 ± 0.2 Mg/ha (Table 20). During the second year when larger
amounts of sediment were moved, RUSLE on average underpredicted sediment yields by
Table 20. RUSLE predicted and measured sediment yield values (Mg/ha) for all test plot
treatments during 2000 and 2001.
T reatm en t
M easured S ed im en t
Y ield*
R U S L E P redicted
S ed im en t Y ields
Year 2000
Coversoil
Coversoil/Pitting
Coversoil/Slash Barriers
Coversoil/AM Inoculum
Control
0.15
0.04
0.09
0.08
0.21
0.50
0.09
0.25
0.50
0.23
Year 2001
Coversoil
1.68
Coversoil/Pitting
0.21
Coversoil/Slash Barriers
1.01
Coversoil/AM Inoculum
1.60
Control
1.34
* Means are an average of three replications.
0.56
0.20
0.29
0.56
0.32
48
1.0 ± 1.0 Mg/ha. RUSLE underpredicted sediment yields by 1.6 + 0.9 Mg/ha on plots
that had an average annual rill severity rating of greater than 1.5.
The rill formation factor constants calculated by Kapolka and Dollhopf (2001)
were applied to the second year RUSLE analysis because rill fdrmation was active. The
optiriiized RUSLE qutputs are presented in Table 21. Appropriate F Factors were
selected for individual plots based upon mean annual rill severity for 2001.
The optimized RUSLE estimated sediment yields on average overpredicted
sediment yields only slightly by 0.03 ± 0.41 Mg/ha. Figure 9 is a comparison of the
actual sediment yields, original RUSLE estimated sedijnent yields, and the optimized
estimated sediment yields. The optimized sediment yields improved accuracy of the
RUSLE predictions to within 97 % of the measured sediment yields. Thfese results
provide evidence that the rill formation factor established by Kapolka and Dollhopf to
reflect the larger sediment yields during active slope rilling at the Treasure Mine retains
accuracy under different rainfall conditions. The rill formation factors may not apply to
soils that have different physiochemical characteristics that affect a soil’s potential for
rilling.
Measured Sediment-Delivery Ratios for Pitting and Slash Barrier Treatments
The influence of the mechanical treatments (Covfersoil/Pitting and Coversoil/Slash
Barrier) in decreasing sediment yields was evaluatfed by calculating the ratio of sediment
yield from the mechanical treatment and sediment yield from the Coversoil treatment to
obtain a sediment-yield delivery ratio (Table 22). The sediment-yield delivery ratio was
0.14 for the Coversoil/Pitting treatment and 0.60 for the Coversoil/Slash Barriers. These
49
Table 21. Optimized RUSLE sediment yields (Mg/ha) using rill formation factors
during 2001.
R U SL E Factors
T reatm en t
R
K
F
Kl
LS
C
P
Coversoil
Coversoil
Coversoil
Coversoil/Pitting
Coversoil/Pitting
Coversoil/Pitting
Coversoil/Slash Barriers
Coversoil/Slash Barriers
Coversoil/Slash Barriers
Coversoil/AM Inoculum
Coversoi I/AM Inoculum
Coversoil/AM Inoculum
Control
Control
Control
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.12
0.12
0.12
1.0
8.4
16.6
1.0
1.0
1.0
1.0
16.6
1.0
16.6
1.0
8.4
8.4
8.4
8.4
0.15
1.26
2.49
0.15
0.15
0.15
0.15
2.49
0.15
2.55
0.15
1.26
1.01
1.01
1.01
4.32
4.32
4.32
0.78
0.78
0.78
1.52
1.52
1.52
4.32
4.32
4.32
4.32
4.32
4.32
0.06
0.06
0.06
0.13
0.13
0.13
0.09
0.09
0.09
0.06
0.06
0.06
0.05
0.05
0.05
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
4.00
-
3.50
-
2.50
-
0.50
-
Covcrsoil
Pitting
Covers oil/
Slash Barriers
Coversoil/
AM Inoculum
O p tim ized
R U SL E
Sed im en t
Y ields
0.23
1.91
3.78
0.09
0.09
0.09
0.12
1.99
0.12
3.87
0.23
1.91
1.19
1.19
1.19
Control
♦ Measured Sediment Yield
□ RUSLE Predicted Sediment Yield
A Optimized RUSLE Sediment Yield
Figure 9. Comparison of measured, RUSLE predicted, and optimized RUSLE sediment
yields using rill formation factors on test plots during 2001.
50
Table 22. Measured sediment-delivery ratio values for Coversoil/Pitting and
Coversoil/Slash Barrier treatments.
T reatm ent
Coversoil/Pitting
Coversoil/Slash Barriers
Sed im en t D elivery R atio
0.14
0.60
mechanical treatments are intended as short-term, temporary erosion control measures to
provide adequate slope stability until vegetation is established. It is expected that the pits
will gradually fill in with eroded soil and the slash barriers will degrade over time,
and the resultant sediment-delivery ratios will correspondingly increase.
51
5. SUMMARY AND CONCLUSION
th e objective of this study was to evaluate the effect of several erosion control
measures on sediment yields on reclaimed steep slopes at the Treasure Mine and compare
measured sediment yields to values predicted by Revised Universal Soil Loss Equation
(RUSLE) Version 1.06.
Results indicate a trend towards pitting of the soil surface as a potentially
effective erosion control practice to prevent rill formation and reduce erosion rates. The
Coversoil/Pitting treatment consistently had decreased sediment yield on all plots and
within treatment variation was minimal. Sediment yields on all other treatments were
highly variable. This variation could be attributed to rill formation that developed on
some plots. Lower sediment yields measured on pitted slopes were attributed to the
increased surface roughness on the soil surface prohibiting the downslope linedr
trajectory of water flow and minimizing rill formation. Newly constructed coversoil and
spoil backfill are highly erosive materials and susceptible to rilling especially when
precipitation is intense. Pitting of the soil surface also increases slope storage of water.
Therefore, erosion control efforts that increase the surface roughness such as pitting or
gouging are ideal for mineland reclamation on steep slopes until vegetation is established.
A good vegetative cover may take two or more years to develop. TTie short duration of
the study (two years), limited vegetation development (< 17 % canopy cover), and low
replication (n = 3) all limit the ability to detect treatment differences. Therefore, these
results suggest that pitting of the soil surface is potentially an effective erosion control
practice at the level of precipitation received during the study, preventing rill formation
52
and reducing sediment yields on steep slopes until vegetation can provides adequate slope
stability.
Results from this study indicate that the decreased rock cover in the treatments
with coversoil application (36 —42 %) provided better plant growth than the Control.
The increased rock cover in the Control provided no significant decreases in sediment
yields.
Runoff was not significantly different between any of the treatments during 2000
and 2001. Sediment yield was not related to runoff in either year. Rill severity was
strongly related to runoff during 2000 when rilling was slight. Rill severity was not
related to runoff during 2001 when the slopes were actively rilling.
Mean biomass, canopy cover and basal cover were significantly higher on plots
with a coversoil application during 2001 wheh compared to the Control. However, there
were no significant differences in plant growth between treatments that received a
coversoil application, indicating that the Coversoil/Pitting, Coversoil/Slash Barriers, and
Coversoil/AM Inoculum treatment^ during the period of this study neither enhanced nor
reduced plant growth when compared to the Coversoil treatment.
Mycorrhizal fungi propagules were present in the coversoil and spoil backfill
material. There were no differences between AM inoculated, coversoil, and no coversoil
treatments in percent colonization of the roots by mycorrhizal fungi in H. vulgare. Inri.
trachycaulum, the AM inoculum treated plots showed significantly higher mycorrhizal
colonization levels, as compared to plant roots from the coversoil and no coversoil treat
plots. However, there were no significant differences between any Of the above-ground
53
plant growth characteristics dn plots that received a coversoil application. Both plant
species were colonized by AM fungi at a relatively high level, and there is probably no
measurable ecological difference between 46 % and 53 % colonization rates found in A.
trachycaulum. With such high colonization in both AM-treated and non-AM treated
plots, we would expect to see no biomass differences between those treatments. AM
inoculation in this study seems’to have been unnecessary, since AM propagules were
already present in the coversoil and spoil backfill material. This AM inoculation
treatment was not found to enhance either AM inoculation levels or plant growth.
Sediment yield, runoff, and rill severity were not related to plant growth in 2001.
The sandy loam texture, low organic matter content, and high coarse fragment percentage
of the coversoil and spoil backfill are not ideal for plant establishment and may have
impaired plant growth, especially during below normal precipitation years as experienced
in this study. Vegetative cover may be slow or difficult to develop on this site.
Vegetation reestablishment can be potentially limited by the effects of frequent
drought patterns experienced during this study and common in semi-arid, climates. In
addition, the sandy loam texture, low organic matter content, and loss of soil aggregation
during removal and regrading make the soil highly erodible and vulnerable to rilling
when bare of vegetation of vegetative cover is minimal. Erosion control measures such
as crimp mulches are not an option due to the stoniness of the soil.
Sediment yield was significantly related to measured factors during both years of
study. Runoff was not related to measured factofs either year. Sediment yield during
both years of study was highly related to maximum hourly precipitation, increase in rill
54
severity class and runoff (r = 0.99). Prediction of erosion during this study appeared to
be estimated by maximum hourly precipitation, runoff and rilling activity on the plots.
During the first year when erosion rates were small, RUSLE overpredicted mean
sediment yields by 0.2 ± 0.2 Mg/ha. During the second year when larger amounts of
sediment were moved, RUSLE on average underpredicted sediment yields by 1.0 ± 1.0
Mg/ha. RUSLE underpredicted sediment yields by 1.6 + 0.9 Mg/ha on plots that had an
average annual rill severity rating of greater than 1.5.
The rill formation factor constants calculated by Kapolka and Dollhopf (2001)
were applied to the second year data and the optimized RUSLE estimated sediment yields
on average overpredicted sediment yields only slightly by 0.03 ± 0.41 Mg/ha. The
optimized sediment yields improved accuracy of the RUSLE predictions to within 97 %
of the measured sediment yields. The rill formation factor calculated by Kapolka and
Dollhopf using sediment data from a previous erosion analysis at the Treasure Mine
during 1998 was highly effective in enhancing RUSLE’s ability to predict sediment
yields on plots where rilling was moderate or greater.
■ The sediment-yield delivery ratio was 0.14 for the Coversoil/Pitting treatment and
0.60 for the Coversoil/Slash Barriers. These mechanical treatments are intended as short-­
term, temporary erosion control measures to prbvide adequate slope stability until
vegetation is established. It is expected that the pits will gradually fill in with sedifnent
and the slash barriers will degrade and the sediment-delivery ratios will correspondingly
increase.
These results indicate that RUSLE is an effective tool for predicting sediment
yields on high elevation, steep slopes. The RUSLE successfully modeled the erosion
control measures used during this study.
56
REFERENCES CITED
Abdul-Kareem5A. W., and S.G. McRae. 1984. The effects on topsoil of long-term
storage in stockpiles. Plant and Soil 76, 357-363.
Brady5N. C. and R. R. Weil. 1996. The Nature and Properties of Soils. I Ith Edition.
Prentice Hall5Upper Saddle River5New Jersey. 740 pp.
Brooks5K. N., P.F. Folliott5H.M. Gregersen5and L.F. DeBano. 1997. Hydrology and
Management of Watersheds. Iowa State University Press. 502 pp.
Clark, R. 1980. Erosion Condition Classification System. Determination of Soil
Erosion Condition Montana Revised Method. U.S. Department of the Interior,
Bureau of Land Management Technical Note #346.
Day5P.R. 1965. Particle fractionation arid particle size analysis. In: Methods of Soil
Analysis - Part 2. American Society of Agronomy. Madison5WL 1159 p.
Fasching5R. A. USDA State Agronomist. Bozeman5MT. Personal communication,
August52000.
Galetovic5J.R. 1998. Guidelines for the use of the revised universal soil loss equation
(RUSLE) version 1.06 on mined lands, construction sites, and reclaimed lands.
Office of Surface Mining5Denver5CO.
Gould, A.B., and A.E. Liberta. 1981. Effects of topsoil storage during surface mining on
the viability of vesicular-arbuscualr mycorrhiza. Mycologia 73,914-922.
Jandel Corporation. 1995. SigmaStat for Windows. Version 2.0
Jasper5D.A., L.K. Abbott and A.D. Robson. 1989. Soil disturbance reduces the
infectivity of external hyphae of vesicular-arbuscular mycorrhizal fungi. New
Phytologist 112, 93-99.
Jasper5D.A., Robson5A.D., and Abbot5L.K. 1987. The effect of surface mining on the
infectivity of vesicular-arbuscular mycorrhizal fungi. Australian Journal of
Botany 35, 641-652.
Jastrow5J.D., R.M. Miller and J. Lussenhop. 1998. Contributions of interacting
biological mechanisms to soil aggregate stabilization in restored prairie. Soil
Biology and Biochemistry. 30, 905-916.
57
REFERENCES CITED continued
Kapolka, N.M. and D J. Dollhopf. 2001. Effect of slope gradient and plant growth on
Soil Loss on Reconstructed Steep Slopes. International Journal of Surface
Mining, Reclamation and Environment. 15, 86 - 99.
Knighton, D. 1998. Fluvial forms and propesses. A new perspective. Oxford University
Press, Inc., New York, NY. 383 pp.
Lambert, D.H., D.E. Baker, and H. Cole, Jr. 1979. The role of mycorrhizae in the
interactions of phosphorus with zinc, copper, and other elements. Soil Science
Society of America Journal, 43, 976-980.
Liberia, A.E. 1981. Effects of topsoil-storage duration on inoculum potential of
vesicular-arbuscular mycorrhizae. In: Symposium on Surface Mining
Hydrology, Sedimentology and Reclamation, University of Kentucky, Lexington,
KY. pp 45-48
Miller, R.M. and J.D. Jastrow. 1992. Mycorrhizal Functioning: An Integrative PlantFungal Process. Allen, M. F.[ed.]. Chapman & Hall, New York. 534 pp.
Montana. 1973. Metal Mine Reclamation Act. Title 82. Chapter 4. Part 3.
Montana. 1973. Strip and Underground Mine Siting Act. Title 82. Chapter 4. Part I.
Nearing, M.A. 1997. A single, continuous function for slope steepness influence on soil
loss. Soil Science Society Journal of America. 61, 917-919.
Nelson, D.W. and L.E. Sommers. 1982. Total carbon, organic carbon, and organic
matter. Methods of Soil Analysis - Part 2. American Society of Agronomy.
Madison, WL 1159 pp.
Osterkamp, W.R. and T.J. Toy. 1994. The healing of disturbed hillslopes by gully
gravure. Geological Society of America Bulletin. 106,1233-1241.
Renard, K.G. and V.A. Ferreira. 1993. RUSLE model description and database
sensitivity. Journal of Environmental Quality. 22,458-466.
58
REFERENCES CITED continued
Renard5K.G., G.R. Foster5GA. Weesies5D.K. McCool andD.C. Yoden [Coordinators],
1997. Predicting soil erosion by water: A guide to conservation planning with
the Revised Universal Soil Loss Equation (RUSLE)5U.S. Department of
Agriculture, Agriculture Handbook No. 703, 404 pp.
Rhoades5LD. 1982. Soluble salts. Methods of Soil Analysis - Part 2. American
Society of Agronomy. Madison5WL 1159 p.
Rives5C.S., MT. Bajwa5A.E. Liberia and R.M. Miller. 1980. Effects of topsoil storage
during surface mining qn the viability of VA mycorrhiza. Soil Science 129,253257.
Shetty5K.G., B A D. Hetrick and A.P. Schwab. 1995. Effects of mycorrhizae and
fertilizer amendments on zinc tolerance of plants. Environmental Pollution. 88,
pp 307-314.
Smith5S.E. and D.J. Read. 1997. Mycorrhizal Symbiosis. Academic Press. New York5
New York. 605 pp.
Stark5J.M. and E.F. Redente. 1987. Production potential of stockpiled topsoil. Soil
Science 144, 72-76.
Taiz5L. and E. Zeiger. 1998. Plant Physiology. 2nd Edition. Sihauer Associates, Inc.
Sunderland, MA. 792 pp
Tmker5P.B. and Gildon5A. 1983. Mycorrhizal fungi and ion uptake. In Metals and
Micronutrients: Uptake and Utilization by Plants, DA. Robb and W.S. Pierpoint5
eds., Academic Press5New York5pp. 21-32.
Toy, T. J., G.R. Foster and K. G. Renard. 1999. RUSLE for mining, construction and
reclamation lands. Journal of Soil and Water Conservation. 54,462-467.
Toy, T.J. and G.R. Foster (eds). 1998. Guidelines for the use of the Revised Universal
Soil Loss Equation on mined lands, construction sites, and reclaimed lands.
Office of Surface Mining5Reclamation, and Enforcement, 149 pp. + software.
Toy, T.J, and W.R. Osterkamp, 1995. The applicability of RUSLE to geomoiphic
studies. Journal of Soil & Water Conservation5 Special Issue: Water Research
and Management in Semiaiid Environments. 50,498-504.
U.S. Congress. 1972. Clean Water Act of 1972. Public Law 95-217.
59
REFERENCES CITED continued
U.S. Congress. 1977. Surface Mining Control and Reclamation Act of 1977. Public
Law 95-87.
USDA NRCS. June 2000. National Agronomy Manual. 3rd Edition. Part 501, 190-VNAM.'
Vangronsveld, J., J.V. Colpaert and K.K. Van Tichelen. 1996. Reclamation of a bare
industrial area contaminated by non-ferrous metals: physicochemical and
biological evaluation of the durability of soil treatment and revegetation.
Environmental Pollution. 94, 131-140.
Visser, S., C.L. Griffiths and D. Parkinson. 1984. Topsoil storage effects on primary
production and rates of vesicular-arbuscular mycorrhizal development in
Agropyron trachycaulum. Plant and Soil. 82,51-60.
Western Regional Climate Center. 2002. Climate and weather information.
www.wrcc.dri.edu/summarv.
Yoder, D. and J. Town. 1995. The future of RUSLE: inside the new Revised Universal
Soil Loss Equation. Journal of Soil & Water Conservation. Special Issue: Water
Research and Management in Semiarid Environments. 50,484-490. .
60
APPENDICES
APPENDIX A
SOIL PHYSIOCHEMICAL DATA
62
Table 23. Coversoil textural analyses.
S am p le
Sand
(% )
I
2
62.5
62.5
S ilt
C lay
(%)
20
20
(%)
17.5
17.5
Soil T extu re
Sandy loam
Sandy loam
Table 24. Coversoil pH and EC analyses.
EC
(m m hos/cm )
pH
6.63
6.56
6.64
Sam ple
I
2
3
0.94
0.84
0.78
Table 25. Subsoil textural analyses.
S am p le
Sand
(% )
I
2
60.0
62.5
Silt
C lay
(%)
17.5
17.5
(%)
22.5
22.5
Soil T exture
Sandy loam
Sandy loam
Table 26. Subsoil pH and EC analyses.
S am p le
I
2
3
pH
7.26
7.36
7.42
EC
(m m hos/cm )
1.19
1.18
1.10
APPENDIX B
PRECIPITATION DATA
64
Table 27. Monthly precipitation (cm) for Dillon, Montana (WMCE), 2000.
M onth
P recipitation
D eviation
January
1.14
-0.28
February
1.52
0.30
March
1.80
-0.41
April
0.23
-3.20
May
6.15
0.23
June
3.20
-2.46
July
1.32
-1.91
August
1.68
-1.19
September
0.99
-2.01
October
5.69
3.63
November
1.98
0.46
December
0.28
-1.02
Total
25.98
-8.20
65
Table 28. Monthly precipitation (cm) for Dillon, Montana (WMCE), 2001.
M onth
P recipitation
D eviation
January
0.13
-1.30
February
0.05
-1.17
March
0.36
-1.85
April
3.53
0.10
May
0.61
-5.31
June
6.78
1.12
July
2.72
-0.51
August
0.00
-2.87
September
2.44
-0.56
October
1.57
-0.48
November
0.74
-0.79
December
3.02
1.73
Total
21.95
-12.24
66
Table 29. Precipitation (cm) at Treasure Mine, June 20 -30, 2000.
D ate
T im e
Precipitation
27
1300
0.1
67
Table 30. Precipitation (cm) at Treasure Mine, July 2000.
Date
Time
Precipitation
I
1600
1700
0.3
0.1
2
1700
0.2
3
3
800
1200
0.1
0.1
4
1400
0.1
6
400
0.1
68
Table 31. Precipitation (cm) at Treasure Mine, August 2000.
D ate
T im e
Precipitation
10
2400
0.1
15
1100
0.1
26
2000
0.1
69
Table 32. Precipitation (cm) at Treasure Mine, September 2000.
D ate
T im e
P recipitation
4
200
0.1
5
1500
0.1
10
1900
0.1
11
600
0.1
16
1400
0.1
19
600
700
1500
0.1
0.2
0.1
23
1700
1800
1900
0.1
0.1
0.2
24
1100
0.1
30
1200
0.1
70
Table 33. Precipitation (cm) at Treasure Mine, October I - October 10, 2000.
D ate
T im e
P recipitation
I
100
900
1000
1100
1200
0.4
0.1
0.2
0.3
0.1
10
2300
2400
1.7
0.6
71
Table 34. Precipitation (cm) at Treasure Mine, May 4 - May 31, 2001._________
___________ D ate________________________ T im e____________________ Precipitation
27
1200
0.1
72
Table 35. Precipitation (cm) at Treasure Mine, June 2001.
D ate
T im e
P recipitation
2
1400
1600
1700
1800
0.1
0.1
0.1
0.1
3
1200
1300
1400
0.2
0.4
0.6
6
1700
0.2
10
2400
0.1
11
100
1300
1500
1700
1800
1900
2000
0.1
0.3
0.2
0.1
0.1
0.1
0.1
12
1500
1600
2100
2200
2300
2400
0.1
0.1
0.4
0.2
0.2
0.1
13
100
200
900
1100
1200
0.1
0.1
0.1
1.0
0.5
16
2200
0.2
17
100
0.1
23
1600
0.1
26
200
0.1
28
1800
0.1
73
Table 36. Precipitation (cm) at Treasure Mine, July 2001.
D ate
T im e
P recipitation
4
1300
2000
0.2
0.2
5
1700
0.2
8
1700
1800
1900
0.1
0.4
0.2
10
1800
0.1
11
1700
0.1
12
1600
0.2
14
1200
1600
1700
0.1
0.1
0.7
15
900
2100
0.2
0.1
16
400
1800
2200
2300
0.1
0.2
0.1
0.3
19
500
0.1
20
1200
1300
0.1
0.1
27
1300
0.1
30
200
400
500
0.1
0.4
0.2
74
Table 37. Precipitation (cm) at Treasure Mine, August 2001.
D ate
T im e
P recipitation
3
1400
0.1
75
Table 38. Precipitation (cm) at Treasure Mine, September 1 - 17, 2001.
D ate
T im e
Precipitation
4
2100
0.3
5
200
300
600
1700
0.2
0.3
0.7
0.9
6
1400
0.1
13
1400
0.6
17
1200
0.2
76
Table 39. Evaporation (cm) at Treasure Mine, 2000 and 2001.
D ates
E vap oration
Year 2000
6/20/00 - 7/12/00
5.7
7/12/00-7/27/00
5.8
7/27/00-8/15/00
6.1
8/15/00 - 8/29/00
3.2
8/29/00 - 9/16/00
1.2
9/16/00 - 9/30/00
1.9
9/30/00- 10/18/00
1.6
Year 2001
5/4/01-5/14/01
2.4
5/14/01-5/28/01
5.4
5/28/01-6/12/01
2.5
7/5/01-7/21/01
6.8
7/21/01 - 8/5/01
5.4
8/5/01 - 8/18/01
6.4
8/18/01-9/3/01
1.5
9/3/01-9/18/01
5.2
9/18/01 - 10/10/01
3.7
77
APPENDIX C
SEDIMENT YIELD AND RUNOFF DATA
78
Table 40. Sediment yield (kg) on test plots, 2000.
A ug 2
D ate
S ept 16
O ct 19
Coversoil
1.19
0.77
14.07
Coversoil
0.68
1.25
0.68
Coversoil
0.58
0.65
0.58
Coversoil/Pitting
0.28
1.22
0.13
Coversoil/Pitting
1.28
0.63
0.00
Coversoil/Pitting
0.62
0.90
0.68
Coversoil/Slash Barriers
0.37
0.58
0.23
Coversoil/Slash Barriers
1.97
1.30
5.65
Coversoil/Slash Barriers
0.60
0.75
0.60
CoversoiI/AM Inoculum
0.43
0.49
0.17
Coversoil/AM Inoculum
1.18
0.71
0.27
Coversoil/AM Inoculum
1.33
1.12
4.80
Control
0.16
0.86
17.14
Control
1.73
1.09
1.22
Control
0.91
1.45
4.68
T reatm en t
79
Table 41. Sediment yield (kg) on test plots, 2001.
Ju ly 21
D ate
A u gu st 5
Sep tem b er 18
Coversoil
45.66
13.56
94.55
Coversoil
13.11
56.18
47.33
Coversoil
4.65
99.77
0.00
Coversoil/Pitting
9.39
1.19
0.00
Coversoil/Pitting
4.26
7.26
0.00
Coversoil/Pitting
1.33
6.07
0.00
Coversoil/Slash Barriers
11.63
65.66
0.00
Coversoil/Slash Barriers
4.00
4.23
0.00
Coversoil/Slash Barriers
42.65
11.55
47.53
Coversoil/AM Inoculum
15.61
30.79
117.84
Coversoil/AM Inoculum
6.98
18.56
0.00
Coversoil/AM Inoculum
14.70
136.17
23.55
Control
35.87
40.24
23.66
Control
22.87
24.69
0.28
Control
28.60
33.62
23.98
T reatm en t
80
Table 42. Depth (cm) of runoff in troughs at test plots, 2000.
Ju ly
12
Ju ly
27
A ug
15
D ate
A ug
27
Sept
16
Sept
30
O ct
19
Coversoil
0
0
0
0
0
0
28
Coversoil
0
0
0
0
0
0
20
Coversoil
0
0
0
0
0
0
19.5
Coversoil/Pitting
0
0
0
0
0
0
18.5
Coversoil/Pitting
0
0
0
0
0
0
20
Coversoil/Pitting
0
0
0
0
0
0
20
Coversoil/Slash Barriers
0
0
0
0
0
0
20
Coversoil/Slash Barriers
0
0
0
0
0
0
24
Coversoil/Slash Barriers
0
0
0
0
0
0
20
Coversoil/AM Inoculum
0
0
0
0
0
0
19.5
Coversoil/AM Inoculum
0
0
0
0
0
0
20.5
Coversoil/AM Inoculum
0
0
0
0
0
0
22
Control
0
0
0
0
0
0
41.9
Control
0
0
0
0
0
0
24
Control
0
0
0
0
0
0
21
T reatm ent
81
Table 43. Depth (cm) of runoff in troughs at test plots, 2001.
T reatm ent
D ate
A u gu st
S ept 3
Ju n e
Ju ly 5
Ju ly
Coversoil
8
34.3
19.8
22
0
14.4
Coversoil
7.8
11.5
17.2
16.75
0
9.2
Coversoil
7.3
4.3
10.5
7
0
8.8
Coversoil/Pitting
7.3
7
18
12
0
7.4
Coversoil/Pitting
7.9
12.5
11.1
8.25
0
7.1
Coversoil/Pitting
7.7
10.7
7.9
2.3
0
6.2
Coversoil/Slash Barriers
8
6.9
12.5
8
0
7.7
Coversoil/Slash Barriers
8.6
14.2
9
5
0
7.9
Coversoil/Slash Barriers
8
10.5
28.8
34
0
22.3
Coversoil/AM Inoculum
7.3
8.3
21
30
0
7.4
Coversoil/AM Inoculum
7.9
9.2
13.2
7.5
0
7.3
CoversoiI/AM Inoculum
7.5
18.6
11.9
8
0
7.1
Control
7.6
8.1
25.4
15
0
25.6
Control
8.1
29
20.2
13.5
0
19.2
Control
9
32.2
30.8
15.5
0
26
S ept
82
Table 44. Rill severity (class) on test plots, 2000.
7/12
7/27
8/15
D ate
8/27
9/16
9/30
10/19
Coversoil
I
I
I
I
I
I
I
Coversoil
I
I
I
I
I
I
I
Coversoil
I
I
I
I
I
I
3
Coversoil/Pitting
I
I
I
I
I
I
I
Coversoil/Pitting
I
I
I
I
I
I
I
Coversoil/Pitting
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
2
Coversoil/AM Inoculum
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
2
Control
I
I
I
I
I
I
I
Control
I
I
I
I
I
I
I
I
I
2
T reatm en t
Control
I
I
I
I
Class: I = stable, 2 = slight, 3 = moderate, 4 = critical, 5 = severe
83
Table 45. Rill severity (class) on test plots, 2001.__________
T reatm ent
5/14
5/28
6/12
7/5
Coversoil
2
2
2
2
Coversoil
2
2
2
Coversoil
3
3
Coversoil/Pitting
I
Coversoil/Pitting
D ate
7/21
8/5
9/3
9/18
3
3
3
3
2
4
4
4
4
3
3
4
4
4
4
I
I
I
2
2
2
2
I
I
I
I
2
2
2
2
Coversoil/Pitting
I
I
I
I
2
2
2
2
Coversoil/Slash Barriers
2
2
2
2
3
3
3
3
Coversoil/Slash Barriers
2
2
2
2
3
4
4
4
Coversoil/Slash Barriers
2
2
2
2
2
2
2
2
Coversoil/AM Inoculum
2
2
2
2
3
5
5
5
Coversoil/AM Inoculum
2
2
2
2
3
3
3
3
Coversoil/AM Inoculum
2
2
2
2
3
3
3
3
Control
2
2
2
2
3
3
3
3
Control
3
3
3
3
4
4
4
4
4
4
4
Control
3
3
3
3
4
Class: I = stable, 2 = slight, 3 = moderate, 4 = critical, 5 = severe
84
APPENDIX D
VEGETATION DATA
I
85
Table 4 6 . Perennial grass canopy cover (%) on test plots, 2001.
Fram e N um b er
Plot
I
2
3
4
5
6
7
8
9
10
Coversoil
2
I
I
2
I
I
2
2
2
I
Coversoil
2
2
I
2
I
I
I
I
I
I
Coversoil
I
I
I
2
I
I
2
I
2
I
Coversoil/Pitting
0
I
I
2
I
I
0
0
I
0
Coversoil/Pitting
2
2
2
I
I
I
2
I
I
I
Coversoil/Pitting
I
2
I
2
I
2
I
2
I
2
Coversoil/Slash Barriers
I
I
I
2
I
I
I
2
I
I
Coversoil/Slash Barriers
I
2
I
I
I
I
2
2
I
I
Coversoil/Slash Barriers
2
I
2
I
I
I
I
I
I
I
Coversoil/AM Inoculum
2
2
I
I
I
I
I
I
I
2
Coversoil/AM Inoculum
I
I
I
3
2
2
I
I
2
I
Coversoil/AM Inoculum
2
I
I
I
2
I
3
2
I
I
Control
I
I
I
I
I
I
0
I
I
I
Control
I
I
I
I
I
I
I
I
I
I
Control
I
I
I
I
I
I
I
0
I
0
Cover class:
I = O- 10 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 5 1 - 60%
10 = 9 1 - 100%
3 = 2 1 - 30%
7 = 6 1 - 70%
4 = 3 1 - 40%
8 = 71 - 80%
86
Table 47. Forb canopy cover (%) on test plots, 2001.
Fram e N um b er
5
6
7
Plot
I
2
3
4
Coversoil
I
0
I
0
I
I
Coversoil
I
I
0
I
I
Coversoil
I
2
I
I
Coversoil/Pitting
I
I
I
Coversoil/Pitting
I
2
Coversoil/Pitting
I
Coversoil/Slash Barriers
8
9
10
I
I
I
I
0
0
I
0
0
I
2
I
2
0
I
I
I
I
I
0
I
I
I
I
2
0
0
I
0
I
I
0
I
I
0
2
2
I
I
I
I
I
I
2
2
I
I
I
I
Coversoil/Slash Barriers
0
I
I
I
I
0
I
I
0
I
Coversoil/Slash Barriers
I
I
I
2
I
0
I
I
I
2
Coversoil/AM Inoculum
I
I
0
I
I
I
I
I
2
I
Coversoil/AM Inoculum
2
I
0
I
I
I
I
2
I
I
Coversoil/AM Inoculum
I
I
2
2
I
2
2
0
I
I
Control
I
0
0
0
I
I
0
I
I
0
Control
0
I
I
I
I
I
0
I
I
0
Control
0
0
I
0
0
I
0
0
0
I
Cover class:
I = 0 -1 0 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 51 60%
10 = 91 100%
—
3 = 21 - 30%
7 = 6 1 - 70%
4 = 3 1 - 40%
8 = 7 1 - 80%
87
Table 48. Annual grass canopy cover (%) on test plots, 2001.
F ram e N um b er
Plot
I
2
3
4
5
6
7
8
9
10
Coversoil
0
I
0
0
0
0
0
0
0
0
Coversoil
I
0
0
0
0
I
0
0
I
0
Coversoil
I
0
2
0
0
0
0
I
0
0
Coversoil/Pitting
I
I
0
I
I
2
0
I
I
I
Coversoil/Pitting
0
0
0
0
0
0
0
0
0
I
Coversoil/Pitting
0
I
I
0
0
0
0
0
0
I
Coversoil/Slash Barriers
I
I
0
I
I
0
0
0
0
I
Coversoil/Slash Barriers
0
0
0
0
I
2
0
0
0
0
Coversoil/Slash Barriers
0
I
I
0
0
0
0
I
I
0
Coversoil/AM Inoculum
I
I
I
I
2
I
I
2
2
I
Coversoil/AM Inoculum
0
0
I
0
0
0
I
I
0
0
Coversoil/AM Inoculum
I
I
0
0
0
0
0
0
I
0
Control
I
0
I
I
0
0
I
0
0
0
Control
I
0
0
0
0
I
I
0
0
0
Control
0
0
0
0
0
0
0
0
0
0
Cover class:
I = 0 -1 0 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 5 1 -6 0 %
10 = 9 1 -1 0 0 %
3 = 21 - 30%
7 = 6 1 - 70%
4 = 3 1 - 40%
8 = 71 - 80%
88
Table 49. Perennial grass basal cover (%) on test plots, 2001.______
_____
I
2
Plot
3
4
F ram e N um b er
5
6
7
8
9
10
Coversoil
I
I
I
I
I
I
I
I
I
I
Coversoil
I
I
I
I
I
I
I
I
I
I
Coversoil
I
I
I
I
I
I
I
I
I
I
Coversoil/Pitting
0
I
I
I
I
I
0
0
I
0
Coversoil/Pitting
I
I
I
I
I
I
I
I
I
I
Coversoil/Pitting
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
I
I
I
Control
I
I
I
I
I
I
0
I
I
I
Control
I
I
I
I
I
I
I
I
I
I
Control
I
I
I
I
I
I
I
0
I
0
Cover class:
I = 0 -1 0 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 5 1 -6 0 %
10 = 91 - 100%
3 = 2 1 -3 0 %
7 = 6 1 -7 0 %
4 = 3 1 -4 0 %
8 = 7 1 -8 0 %
89
Table 50. Forb basal cover (%) on test plots. 2001.
Frame Number
5
6
7
Plot
I
2
3
4
Coversoil
I
0
I
0
I
I
Coversoil
I
I
0
I
I
Coversoil
I
I
I
I
Coversoil/Pitting
I
I
I
Coversoil/Pitting
I
I
Coversoil/Pitting
I
Coversoil/Slash Barriers
8
9
10
I
I
I
I
0
0
I
0
0
I
I
I
I
0
I
I
I
I
I
0
I
I
I
I
I
0
0
I
0
I
I
0
I
I
0
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
0
I
I
I
I
0
I
I
0
I
Coversoil/Slash Barriers
I
I
I
I
I
0
I
I
I
I
Coversoil/AM Inoculum
I
I
0
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
0
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
0
I
I
Control
I
0
0
0
I
I
0
I
I
0
no coveroil
0
I
I
I
I
I
0
I
I
0
Control
0
0
I
0
0
I
0
0
0
I
Cover class:
1= 0 - 1 0 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 5 1 -6 0 %
10 = 91 - 100 %
3 = 21 - 3 0 %
7 = 6 1 -7 0 %
4 = :3 1 -4 0 %
8 = '71 - 80 %
90
Table 51. Annual grass basal cover (%) on test plots, 2001.
Fram e N um b er
Plot
I
2
3
4
5
6
7
8
9
10
Coversoil
0
I
0
0
0
0
0
0
0
0
Coversoil
I
0
0
0
0
I
0
0
I
0
Coversoil
I
0
I
0
0
0
0
I
0
0
Coversoil/Pitting
I
I
0
I
I
I
0
I
I
I
Coversoil/Pitting
I
0
0
0
0
0
0
0
0
I
Coversoil/Pitting
0
I
I
0
0
0
I
I
0
I
Coversoil/Slash Barriers
I
I
0
I
I
I
I
I
I
I
Coversoil/Slash Barriers
0
0
0
0
I
I
0
0
0
0
Coversoil/Slash Barriers
0
I
I
0
I
0
0
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
0
0
I
0
0
0
I
I
0
0
Coversoil/AM Inoculum
I
I
I
0
0
0
0
0
I
0
Control
I
0
I
I
0
0
I
0
0
0
Control
I
0
0
0
0
I
I
0
0
0
Control
0
0
0
0
0
0
0
0
0
0
Cover class:
I = 0 - 10 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 51 - 60 %
10 = 91 - 100%
3 = 2 1 - 30%
7 = 6 1 - 70%
4 = 31 - 40%
8 = 71 - 80%
91
Table 52. Rock cover (%) on test plots, 2001.
F ram e N um b er
5
6
7
Plot
I
2
3
4
Coversoil
3
4
4
5
4
4
Coversoil
5
3
5
2
5
Coversoil
3
3
4
3
Coversoil/Pitting
7
7
6
Coversoil/Pitting
5
3
Coversoil/Pitting
4
Coversoil/Slash Barriers
8
9
10
4
5
4
7
3
6
7
3
3
3
4
5
4
3
4
5
6
7
6
7
8
4
5
5
5
5
6
6
4
3
4
4
3
4
4
5
5
4
2
6
5
5
6
4
4
6
4
5
5
Coversoil/Slash Barriers
5
6
4
4
5
6
3
4
7
4
Coversoil/Slash Barriers
3
2
3
5
4
4
3
5
4
3
Coversoil/AM Inoculum
5
5
4
4
5
6
6
5
5
6
Coversoil/AM Inoculum
4
4
6
4
4
5
4
5
3
4
Coversoil/AM Inoculum
4
5
5
5
6
5
4
4
5
3
Control
6
6
6
6
4
5
6
4
5
6
Control
6
7
6
6
5
6
7
5
5
5
Control
8
8
7
7
8
8
8
8
8
8
Cover class:
I = 0 - 10 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 5 1 -6 0 %
10 = 91 - 100 %
3 = 21 - 30%
7 = 61 - 70%
4 = 31 - 40%
8 = 7 1 - 80%
92
Table 53. Bare ground cover (%) on test plots, 2001.
F ram e N um b er
5
6
7
Plot
I
2
3
4
Coversoil
7
7
6
5
6
7
Coversoil
6
8
6
8
5
Coversoil
7
7
7
7
Coversoil/Pitting
2
2
3
Coversoil/Pitting
6
7
Coversoil/Pitting
6
Coversoil/Slash Barriers
8
9
10
6
5
6
4
7
4
4
7
7
7
6
6
6
7
6
5
4
3
4
3
3
6
6
6
6
6
5
5
7
7
6
7
7
6
7
6
5
6
8
5
7
5
4
6
6
4
6
6
5
Coversoil/Slash Barriers
5
4
6
7
5
4
7
6
3
6
Coversoil/Slash Barriers
7
8
7
5
6
7
7
5
6
7
Coversoil/AM Inoculum
5
5
6
6
5
4
4
5
5
4
Coversoil/AM Inoculum
6
7
4
6
6
5
6
5
7
7
Coversoil/AM Inoculum
6
6
5
5
4
5
6
7
6
7
Control
4
5
5
4
6
5
4
6
7
4
Control
5
4
5
5
5
5
4
5
5
5
Control
2
2
3
3
2
3
2
2
2
2
Cover class:
I = O - 10 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 51 - 60%
10 = 9 1 - 100%
3 = 2 1 - 30%
7 = 61 - 70%
4 = 31 - 40%
8 —71 - 80%
93
Table 54. Perennial grass biomass (g) on test plots, 2001.
Plot
I
Coversoil
0.54
Coversoil
0.39 0.94
Coversoil
2
0.4
4
3
0
F ram e N um b er
5
6
7
0 0.51
1.04
0.77
1.68
0.4 2.03
1.32
8
1.19
1.05
1.15 0.65 5.77
Coversoil/Pitting
0.19 0.64 0.52
1.02
0.2 0.29 0.47
Coversoil/Pitting
2.29 0.41
0 2.12 0.32 0.34
Coversoil/Pitting
Coversoil/Slash
Barriers
Coversoil/Slash
Barriers
Coversoil/Slash
Barriers
Coversoil/AM
Inoculum
Coversoil/AM
Inoculum
Coversoil/AM
Inoculum
0.29 0.27
0
4.56 0.78 0.55
0.45 0.11
0.2 0.32 2.81
1.11 0.41
0.68 0.22 0.62 0.31
1.75
0
0.75
1.33
1.49 1.48
0 1.16
0 9.02
1.41 0.02 0.89
1.55
1.25 2.22
0.17 0.26 0.46
1.67 0.73 0.36 0.16
Control
0.27 0.16 0.08 0.31
Control
0.01
0.1 0.03
0 0.19
0.2
1.25
1.97 0.92
1.15 0.35 0.55
0.39 0.13 0.25
1.26 0.93
0 0.03
2.35 0.18 0.83
Control
1.65 2.54
1.55
0.3 0.08 0.27
1.37 0.83 0.55 0.34
10
1.45 0.32 0.25
0 0.77
1.59 0.38
1.7 1.05
9
1.3 1.31
8.45
1.9 4.59 2.61
1.67 3.99 5.06
1.25 0.36
0
0
0
1.49
1.1
0 0.03
0 0.34
0
0.4 0.34 0.26 0.13 0.35 0.63
0 0.12
0.2 0.01
0.02 0.01
0.18
94
Table 55. Forb biomass (g) on test plots, 2001.
F ram e N um b er
5
6
7
Plot
I
2
3
4
Coversoil
2.23
0.91
0.14
0.00
0.09
0.10
Coversoil
0.53
1.58
4.12
0.00
1.88
Coversoil
2.09
1.64
0.56
0.14
Coversoil/Pitting 2.86
0.00
0.11
Coversoil/Pitting 0.43
0.00
Coversoil/Pitting
Coversoil/Slash
Barriers
Coversoil/Slash
Barriers
Coversoil/Slash
Barriers
Coversoi 1/AM
Inoculum
Coversoil/AM
Inoculum
Coversoil/AM
Inoculum
0.00
8
9
10
0.25
0.31
0.20
0.00
0.00
0.00
0.48
0.07
0.00
0.25
3.82
1.00
0.08
0.00
0.07
0.12
0.00
0.62
0.00
0.19
0.00
0.00
0.00
0.16
0.72
0.26
0.00
0.09
0.07
0.90
0.00
0.00
0.25
0.09
0.00
0.00
0.46
0.00
1.62
0.12
0.32
0.02
3.83
0.00
3.93
0.00
1.07
0.10
0.00
1.35
0.20
0.05
2.38
0.07
0.15
2.71
0.00
0.09
0.00
2.27
1.38
0.03
2.18
3.04
0.81
0.87
0.53
0.00
0.00
0.00
0.12
0.37
0.15
0.00
0.00
0.00
0.18
0.00
0.00
0.01
0.00
0.31
0.00
0.00
1.75
0.09
0.00
0.00
0.12
0.07
0.00
0.03
0.06
0.03
0.01
0.00
2.89
0.00
0.20
Control
0.00
0.00
0.00
0.00
0.00
0.24
0.00
0.00
0.21
0.00
Control
0.32
0.15
0.70
0.00
0.15
0.00
0.00
0.17
0.20
0.14
Control
0.00
0.00
0.00
0.01
0.03
0.07
0.08
0.00
0.00
0.00
95
Table 56. Annual grass biomass (g) on test plots, 2001.
Fram e N um b er
Plot
I
2
3
4
5
6
7
8
9
10
Coversoil
O
O
O
O
O
0
0
0
0
0
Coversoil
O
O
O
O
0.88 0
0
0
0
0
Coversoil
O
O
O
O
0
0
0
0
0
0
Coversoil/Pitting
O
O
O
O
0
0
0
0
0
0
Coversoil/Pitting
O
O
O
O
0
0
0
0
0
0
Coversoil/Pitting
O
O
O
O
0
0
0
0
0
0
Coversoil/Slash Barriers
O
O
O
O
0
0
0
0
0
0
Coversoil/Slash Barriers
O
O
O
O
0
0
0
0
0
0
Coversoil/Slash Barriers
O
O
O
O
0
0
0
0
0
0
Coversoil/AM Inoculum
O
O
O
O
0
0
0
0
0
0
Coversoil/AM Inoculum
O
O
O
O
0
0
0
0
0
0
Coversoil/AM Inoculum
O
O
O
O
0
0
0
0
0
0
Control
O
O
O
O
0
0.03 0
0
0.63 0
Control
O
O
O
O
0
0
0
0
0
0
Control
O
O
O
O
0
0
0
0
0
0
96
Table 57. Vegetative litter cover (class) on test plots, 2001.
F ram e N um b er
5
6
7
Plot
I
2
3
4
Coversoil
I
I
I
I
I
I
Coversoil
I
I
I
I
I
Coversoil
I
I
I
I
Coversoil/Pitting
I
I
I
Coversoil/Pitting
I
I
Coversoil/Pitting
I
Coversoil/Slash Barriers
8
9
10
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
0
0
I
I
I
I
I
I
I
I
I
0
I
I
I
I
0
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
I
I
I
Coversoil/Slash Barriers
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
I
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
0
I
I
I
I
I
I
I
I
I
Coversoil/AM Inoculum
0
I
I
I
I
I
I
I
I
0
Control
0
0
0
0
0
0
0
0
0
0
Control
I
I
I
I
I
I
I
I
I
I
Control
0
0
0
0
0
0
0
0
0
0
Cover class:
1= 0 - 1 0 %
5 = 4 1 -5 0 %
9 = 8 1 -9 0 %
2 = 11-20%
6 = 51 - 60 %
10 = 9 1 -1 0 0 %
3 = 21 - 30%
7 = 6 1 - 70%
4 = 31 - 40%
8 = 71 - 80%
97
Table 58. Percent colonization by vesicular arbuscular mycorrhizal fungi in coversoil
________ and spoil backfill material, 2000.__________________
T reatm ent
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
Spoil Backfill Material
* Arcsine transformation
N onAM
H yp h ae
H yp h ae
V esicles
A rb uscu les
I
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
35
38
44
37
51
18
35
29
21
15
42
17
23
35
15
27
38
31
2
3
2
I
0
0
0
4
0
4
3
5
5
0
3
0
0
0
0
0
0
0
0
0
I
0
0
0
0
0
0
0
0
0
0
0
N one
% AM
colon .*+
58
39%
55
43%
50
48%
58
40%
46
53%
30
38%
60
38%
63
34%
75
22%
80
19%
51
47%
74
23%
68
29%
61
36%
78
19%
69
28%
58
40%
65
32%
{hyphae + vesicles + arbuscules)
+% AM calculated as
(total observations)
98
Table 59. Percent colonization by vesicular arbuscular mycorrhizal fungi in Hordeum
vulgare, 2001.
T reatm en t
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
N onAM
H yp h ae
2
O
3
O
O
O
6
O
O
I
O
O
4
O
2
O
O
O
O
2
O
4
O
O
H yp h ae
V esicles
A rb uscules
N one
% AM
colon. *+
41
O
O
53
44%
39
O
O
57
42%
20
I
O
72
22%
31
O
O
65
33%
39
I
I
55
44%
43
2
O
51
49%
20
I
O
69
22%
30
3
O
63
35%
32
3
O
61
37%
17
O
O
78
18%
23
I
O
72
25%
19
O
O
29
41%
25
7
O
60
34%
32
O
O
64
34%
28
I
O
65
31%
32
O
O
64
34%
42
I
O
53
46%
8
O
O
88
8%
37
O
O
59
40%
42
O
I
51
46%
16
O
O
32
34%
31
O
O
61
33%
17
O
O
31
36%
34
O
O
62
36%
I
25
O
O
70
26%
O
20
5
O
71
26%
O
37
O
O
59
40%
O
32
O
O
64
34%
O
27
I
O
68
30%
2
19
6
O
69
26%
I
27
I
O
67
30%
3
12
O
O
33
25%
O
48
9
I
38
65%
2
35
O
O
59
37%
I
33
O
O
62
35%
I
40
I
O
54
44%
(intersections with hyphae, vesicles, or arbuscules)
* Arcsine transformation +% AM
(total intersections)
99
Table 60. Percent colonization by vesicular arbuscular mycorrhizal fungi in Agropyron
trachycaulum, 2001.
T reatm ent
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Coversoil
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Control
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
Coversoil/AM Inoc.
N onAM
H yp h ae
3
O
O
3
6
2
2
O
O
4
I
O
O
O
I
I
O
I
O
O
O
O
O
O
O
O
O
O
2
O
O
I
O
I
O
O
H yp h ae
V esicles
A rb uscu les
N one
% AM
colon.**
46
I
I
45
52%
41
2
O
53
46%
21
2
O
25
50%
20
I
I
23
48%
39
O
O
51
42%
38
O
I
55
42%
48
O
O
46
52%
35
3
O
41%
58
44
I
I
50
50%
37
O
O
55
40%
31
O
O
64
33%
47
I
O
48
52%
42
I
O
53
46%
41
O
2
53
46%
42
5
O
48
51%
22
5
O
20
60%
38
I
O
57
42%
41
4
O
50
49%
31
2
O
63
35%
43
2
O
51
49%
30
5
O
61
37%
31
2
O
63
35%
20
I
O
27
45%
16
I
O
31
36%
46
3
O
47
54%
38
12
4
42
60%
56
2
O
38
65%
27
O
O
21
60%
50
O
O
44
55%
47
2
O
47
54%
39
4
O
53
46%
37
O
O
58
40%
46
3
O
47
54%
49
O
I
45
55%
42
O
O
54
45%
23
O
O
25
50%
(intersections with hyphae, vesicles, or arbuscules)
* Arcsine transformation +% AM
(total intersections)
100
APPENDIX E
. STATISTICAL ANALYSIS
,
101
Table 61. Two way analysis of variance of sediment yield, 2000.
S tatistical A nalysis R esults_______________
General Linear Model (No Interactions)
Dependent Variable: 2000 Total Sediment Yield (Mg/ha)
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
0.0529
0.100
0.0375
0.191
MS
0.0132
0.0500
0.00469
0.0136
F
P
2.820 0.099
10.659 0.006
The difference in the mean values among the different levels of TRT is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in REP. There is not a statistically significant
difference (P = 0.099).
The difference in the mean values among the different levels of REP is greater than
would be expected by chance after allowing for effects of differences in TRT. There is a
statistically significant difference (P = 0.006). To isolate which group(s) differ from the
others use a multiple comparison procedure.
Power of performed test with alpha = 0.0500: for TRT : 0.335
Power of performed test with alpha = 0.0500: for REP : 0.911
Least square means for TRT :
Group
Mean
CoversoiI
0.150
Cvr/Pitting
.0433
Cvr/Sl.B.
0.0867
Cvr/AM inoc.
0.0767
Control
0.210
Std Err of LS Mean = 0.0396
Least square means for REP :
Group
Mean
1.000
0.0440
2.000
0.0680
3.000
0.228
Std Err of LS Mean = 0.0306
102
Table 61. Two way analysis of variance of sediment yield, 2000 continued.
S tatistical A n alysis R esults con tin u ed
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Diff of Means p
Control vs. CvrZPitting
0.167
5
Control vs. CvrZAM inoc.
0.133
4
Control vs. Cvr/Sl.B.
0.123
3
Control vs. Coversoil
0.0600
2
Coversoil vs. CvrZPitting
0.107
4
Coversoil vs. Cvr/AM inoc. 0.0733
3
Coversoil vs. CvrZSl.B.
0.0633
2
CvrZSl.B. vs. CvrZPitting
0.0433
3
Cvr/Sl.B. vs. Cvr/AM inoc. 0.01000
2
Cvr/AM inoc. vs. Cvr/Pitting 0.0333
2
Comparisons for factor: REP
Comparison
Diff of Means
3.000 vs. 1.000
0.184
3.000 vs. 2.000
0.160
2.000 vs. 1.000
0.0240
p
3
2
2
q
q
4.214
3.371
3.118
1.517
2.697
1.854
1.601
1.096
0.253
0.843
P
6.006 0.007
5.222 0.006
0.783 0.595
P
0.096
0.158
0.130
0.315
0.298
0.429
0.290
0.728
0.863
0.568
P 0.050
No
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
P0.050
Yes
Yes
No
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
103
Table 62. Two way analysis of variance of sediment yield, 2001.
S tatistical A n alysis R esults______
General Linear Model (No Interactions)
Dependent Variable: 2001 Total Sediment Yield (Mg/ha)
Normality Test:
Passed (P = 0.022)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
4.275
0.0385
11.062
15.375
MS
1.069
0.0192
1.383
1.098
F
0.773
0.0139
P
0.572
0.986
The difference in the mean values among the different levels of TRT is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in REP. There is not a statistically significant
difference (P = 0.572).
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.986).
Power of performed test with alpha = 0.0500: for TRT : 0.0502
Power of performed test with alpha = 0.0500: for REP : 0.0500
Least square means for TRT :
Group
Mean
Coversoil
1.680
Cvr/Pitting
0.210
Cvr/Sl.B.
1.010
Cvr/AM inoc.
1.603
Control
1.343
Std Err of LS Mean = 0.679
Least square means for REP :
Group
Mean
1.000
1.106
2.000
1.230
3.000
1.172
Std Err of LS Mean = 0.526
104
Table 63. Two way analysis of variance of rill severity class, 2000.
S tatistical A n alysis R esults______________
General Linear Model (No Interactions)
Dependent Variable: 2000 Rill Severity Class (rank transformed)
Normality Test:
Passed (P = 0.018)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
16.000
120.000
32.000
168.000
MS
4.000
60.000
4.000
12.000
F
1.000
15.000
P
0.461
0.002
The difference in the mean values among the different levels of TRT is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in REP. There is not a statistically significant
difference (P = 0.461).
The difference in the mean values among the different levels of REP is greater than
would be expected by chance after allowing for effects of differences in TRT. There is a
statistically significant difference (P = 0.002). To isolate which group(s) differ from the
others use a multiple comparison procedure.
Power of performed test with alpha = 0.0500: for TRT : 0.0502
Power of performed test with alpha = 0.0500: for REP : 0.980
Least square means for TRT :
Group
Mean
Coversoil
9.000
Cvr/Pitting
6.000
Cvr/Sl.B.
8.333
Cvr/AM inoc.
8.333
Control
8.333
Std Err of LS Mean = 1.155
Least square means for REP :
Group
Mean
1.000
6.000
2.000
6.000
3.000
12.000
Std Err of LS Mean = 0.894
105
Table 63. Two way analysis of variance of rill severity class, 2000 continued.
S tatistical A nalysis R esults con tin u ed
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Diff of Means p
Coversoil vs. CvrZPitting
3.000
5
Coversoil vs. Cvr/AM inoc. 0.667
4
Coversoil vs. Cvr/Sl.B.
0.667
3
Coversoil vs. Control
0.667
2
Control vs. Cvr/Pitting
2.333
4
Control vs. AM inoc.
1.776E-015 3
Control vs. Cvr/Sl.B.
1.776E-015 2
Cvr/Sl.B. vs. Cvr/Pitting
2.333
3
Cvr/Sl.B. vs. AM inoc.
0.000
2
Cvr/AM inoc. vs. Cvr/Pitting 2.333
2
Comparisons for factor: REP
Comparison
DifF of Means p
3.000 vs. 2.000
6.000
3
3.000 vs. 1.000
6.000
2
1.000 vs. 2.000
0.000
2
q
q
2.598
0.577
0.577
0.577
2.021
1.538E-015
1.538E-015
2.021
0.000
2.021
P
6.708 0.004
6.708 0.002
0.000 1.000
P
0.417
0.976
0.913
0.694
0.518
1.000
1.000
0.372
1.000
0.191
P0.050
No
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
Do Not Test
P<0.050
Yes
Yes
No
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
106
Table 64. Two way analysis of variance of rill severity class, 2001.
S tatistical A nalysis R esults______
General Linear Model (No Interactions)
Dependent Variable: 2001 Rill Severity Class
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
5.204
0.124
1.876
7.204
MS
1.301
0.0620
0.234
0.515
F
5.548
0.264
P
0.019
0.774
The difference in the mean values among the different levels of TRT is greater than
would be expected by chance after allowing for effects of differences in REP. There is a
statistically significant difference (P = 0.019). To isolate which group(s) differ from the
others use a multiple comparison procedure.
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.774).
Power of performed test with alpha = 0.0500: for TRT : 0.729
Power of performed test with alpha = 0.0500: for REP : 0.0500
Least square means for TRT :
Group
Mean
Coversoil
3.000
Cvr/Pitting
1.500
Cvr/Sl.B.
2.467
Cvr/AM inoc.
2.767
Control
3.167
Std Err of LS Mean = 0.280
Least square means for REP
Group
Mean
1.000
2.460
2.000
2.680
3.000
2.600
Std Err of LS Mean = 0.217
107
Table 64. Two way analysis of variance of rill severity class, 2001 continued.
S tatistical A n alysis R esults continued_________________________________
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Diff of Means p
Control vs. Cvr/Pitting
1.667
5
Control vs. Cvr/Sl.B.
0.700
4
Control vs. Cvr/AM inoc.
0.400
3
Control vs. Coversoil
0.167
2
Coversoil vs. Cvr/Pitting
1.500
4
Coversoil vs. Cvr/Sl.B.
0.533
3
Coversoil vs. Cvr/AM inoc. 0.233
2
Cvr/AM inoc. vs. Cvr/Pitting 1.267
3
Cvr/AM inoc.vs. Cvr/Sl.B. 0.300
2
Cvr/Sl.B. vs. Cvr/Pitting
0.967
2
Comparisons for factor: REP
Comparison Diff of Means
2.000 vs. 1.000
0.220
2.000 vs. 3.000
0.0800
3.000 vs. 1.000
0.140
P
3
2
2
q
q
5.961
2.504
1.431
0.596
5.365
1.908
0.835
4.531
1.073
3.458
P
1.016 0.760
0.369 0.801
0.646 0.660
P
0.018
0.352
0.591
0.685
0.022
0.410
0.572
0.030
0.470
0.040
P 0.050
Yes
No
Do Not Test
Do Not Test
Yes
Do Not Test
Do Not Test
Yes
Do Not Test
Yes
P0.050
No
Do Not Test
Do Not Test
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
108
Table 65. Linear regression of mean rill severity class on total annual sediment yield,
_________ 2000._______________________________________________________
S tatistical A nalysis R esults_________________________________
2000 LOG Sediment Yield = -4.876 + (3.606 * 2000 Mean Rill Severity Class)
N = 15.000
R = 0.756
Rsqr = 0.571 Adj Rsqr = 0.538
Standard Error of Estimate = 0.269
Constant
Mean Rill Severity
Coefficient
-4.876
3.606
Std. Error
0.904
0.867
t
-5.394
4.161
P
<0.001
0.001
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
1.248
0.937
2.186
MS
1.248
0.0721
0.156
F
17.312
P
0.001
Normality Test:
Passed (P = 0.452)
Constant Variance Test:
Passed (P = 0.944)
Power of performed test with alpha = 0.050: 0.927
109
Table 66. Linear regression of mean rill severity class on total annual sediment yield,
2001.____________________________________________________________
_______________
S tatistical A n alysis R esults________
2001 LOG Sediment Yield — 1.732 + (0.615 * 2001 Mean Rill Severity Class)
N = 15.000
R = 0.871
Rsqr = 0.758 Adj Rsqr = 0.740
Standard Error of Estimate = 0.259
Constant
Mean Rill Severity
Coefficient
-1.732
0.615
Std. Error
0.257
0.0964
t
-6.730
6.383
P
<0.001
<0.001
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
2.726
0.870
3.595
MS
2.726
0.0669
0.257
F
40.749
P
<0.001
Normality Test:
Passed (P = 0.326)
Constant Variance Test:
Passed (P = 0.050)
Power of performed test with alpha = 0.050: 0.996
no
Table 67. Linear regression of total precipitation on total annual sediment yield, 2000.
S tatistical A n alysis R esults
Data source: 2000 Sediment Yield to Total Precipitation
SedimentYield 2000 = 0.0120 + (0.0235 * Total Precipitation)
N =3.000
R = 0.966
Rsqr = 0.934 Adj Rsqr = 0.868
Standard Error of Estimate = 0.010
Coefficient
0.0120
0.0235
Constant
Total Precipitation
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
SS
0.00156
0.000110
0.00167
Std. Error
0.00893
0.00626
MS
0.00156
0.000110
0.000833
Normality Test:
Passed (P = 0.383)
Constant Variance Test:
Failed (P = <0.001)
F
14.119
t
1.349
3.758
P
0.406
0.166
P
0.166
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
Ill
Table 68. Linear regression of maximum daily precipitation on total annual sediment
________ yield, 2000.
__________________________________________
S tatistical A n alysis R esults_____________
Data source: 2000 Sediment Yield and Maximum Daily Precipitation
Sediment Yield 2000 = 0.0174 + (0.0224 * Maximum Daily Precipitation)
N =3.000
R = 0.998
Rsqr = 0.996 Adj Rsqr = 0.992
Standard Error of Estimate = 0.003
Constant
Max. Daily Precip.
Coefficient
0.0174
0.0224
Std. Error
0.00190
0.00140
t
9.148
16.021
P
0.069
0.040
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
SS
0.00166
0.00000647
0.00167
MS
0.00166
0.00000647
0.000833
F
256.688
P
0.040
Normality Test:
Passed (P = 0.504)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
112
Table 69. Linear regression of maximum hourly precipitation on total annual sediment
________ yield, 2000.____________________ ______________________________
S tatistical A n alysis R esults_________ __________________________
Data source: Sediment Yield 2000 to Maximum Hourly Precipitation
Sediment Yield 2000 = 0.0179 + (0.0301 * Maximum Hourly Precipitation)
N =3.000
R = 1.000
Rsqr = 0.999 Adj Rsqr = 0.998
Standard Error of Estimate = 0.001
Constant
Max. Hrly. Precip.
Coefficient
0.0179
0.0301
Std. Error
0.000943
0.000942
t
19.000
31.947
P
0.033
0.020
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
SS
0.00167
0.00000163
0.00167
MS
0.00167
0.00000163
0.000833
F
1020.593
P
0.020
Normality Test:
Passed (P = 0.518)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
113
Table 70. Linear regression of total precipitation on total annual sediment yield, 2001.
S tatistical A n alysis R esults___________
Sediment Yield 2001 = 0.932 - (0.200 * Total Precipitation)
N = 3.000
R = 0.786
Rsqr = 0.617 Adj Rsqr = 0.234
Standard Error of Estimate = 0.346
Constant
Tot. Precip.
Coefficient
0.932
-0.200
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
Std. Error
0.471
0.158
t
1.978
-1.270
P
0.298
0.425
SS
MS
F
P
0.193 0.193 1.612 0.425
0.120 0.120
0.312 0.156
Normality Test:
Passed (P = 0.447)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
114
Table 71. Linear regression of maximum daily precipitation on total annual sediment
________ yield, 2001.______________
S tatistical A nalysis R esults______________________________________
Sediment Yield 2001 = 0.989 - (0.498 * Maximum Daily Precipitation)
N =3.000
R = 0.907
Rsqr = 0.823 Adj Rsqr = 0.646
Standard Error of Estimate = 0.235
Coefficient
Constant
0.989
Max. Daily Precip. -0.498
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
SS
0.257
0.0553
0.312
Std. Error
0.309
0.231
t
3.199
-2.156
P
0.193
0.276
MS
0.257
0.0553
0.156
F
4.650
P
0.276
Normality Test:
Passed (P = 0.478)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
115
Table 72. Linear regression of maximum hourly precipitation on total annual sediment
_________ yield, 2001.___________________________________________________
S tatistical A nalysis R esults________________________________________________
Sediment Yield 2001 = 1.325 - (1.410 * Maximum Hourly Precipitation)
N =3.000
R = 0.975
Rsqr = 0.951 Adj Rsqr = 0.902
Standard Error of Estimate = 0.124
Coefficient
Constant
1.325
Max. Hourly Precip. -1.410
Std. Error
0.224
0.321
t
5.907
-4.398
P
0.107
0.142
Analysis of Variance:
DF
Regression
I
Residual
I
Total
2
MS
0.297
0.0153
0.156
F
19.343
P
0.142
SS
0.297
0.0153
0.312
Normality Test:
Passed (P = 0.370)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: <0.001
The power of the performed test (<0.001) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
116
Table 73. Two way analysis of variance of percent rock cover during 2001.
S tatistical A n alysis R esults__________
General Linear Model (No Interactions)
Dependent Variable: Percent Rock Cover
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
trt
rep
Residual
Total
DF
4
2
8
14
SS
831.204
87.481
753.732
1672.417
MS
207.801
43.741
94.216
119.458
F
2.206
0.464
P
0.158
0.645
The difference in the mean values among the different levels of trt is not great enough to
exclude the possibility that the difference is just due to random sampling variability after
allowing for the effects of differences in rep. There is not a statistically significant
difference (P = 0.158).
The difference in the mean values among the different levels of rep is not great enough to
exclude the possibility that the difference is just due to random sampling variability after
allowing for the effects of differences in trt. There is not a statistically significant
difference (P = 0.645).
Power of performed test with alpha = 0.0500: for t r t : 0.229
Power of performed test with alpha = 0.0500: for rep :0.0500
Least square means for tr t:
Group
Coversoil
Coversoil/Pitting
Coversoil/Slash Barriers
Coversoil/AM Inoculum
Control
Std Err of LS Mean = 5.604
Least square means for rep :
Group Mean
1.000 47.400
2.000 42.380
3.000 42.180
Std Err of LS Mean = 4.341
Mean
35.667
44.667
39.967
4 1.967
57.667
117
Table 74. Linear regression of percent rock cover on sediment yield, 2001.
S tatistical A n alysis R esults
2001 Sediment Yield (Mg/ha) —0.937 + (0.00528 * percent rock cover)
N = 15.000
R = 0.0552
Rsqr = 0.00305
Adj Rsqr = 0.000
Standard Error of Estimate = 1.084
Constant
% rock cover
Coefficient
0.937
0.00528
Std. Error
1.199
0.0265
t
0.781
0.199
P
0.449
0.845
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
0.0467
15.284
15.331
MS
0.0467
1.176
1.095
F
0.0397
P
0.845
Normality Test:
Passed (P = 0.260)
Constant Variance Test:
Passed (P = 0.154)
Power of performed test with alpha = 0.050: 0.038
The power of the performed test (0.038) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
118
Table 75. Linear regression of percent rock cover on biomass, 2001.
S tatistical A n alysis R esults______________________________________
Biomass 2001 = 356.054 —(4.930 * percent rock cover)
N = 15.000
R = 0.680
Rsqr = 0.463 Adj Rsqr = 0.422
Standard Error of Estimate = 60.208
Coefficient
Constant
356.054
% rock cover -4.930
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
Std. Error
66.600
1.472
SS
40640.428
47125.596
87766.024
t
5.346
-3.348
MS
40640.428
3625.046
6269.002
Normality Test:
Passed (P = 0.637)
Constant Variance Test:
Passed (P = 0.657)
Power of performed test with alpha = 0.050: 0.820
P
<0.001
0.005
F
11.211
P
0.005
119
Table 76. Linear regression of percent rock cover on total annual runoff, 2001.
S tatistical A nalysis R esults____________________
2001 Total Annual Runoff = 4.326 + (2.682 * percent rock cover)
N = 15.000
R = 0.408
Rsqr = 0.167 Adj Rsqr = 0.103
Standard Error of Estimate = 67.992
Coefficient
Constant
4.326
% rock cover 2.682
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
Std. Error
75.210
1.663
SS
12032.885
60098.433
72131.318
t
0.0575
1.613
MS
12032.885
4622.956
5152.237
Normality Test:
Passed (P = 0.080)
Constant Variance Test:
Passed (P = 0.863)
P
0.955
0.131
F
P
2.603 0.131
Power of performed test with alpha = 0.050: 0.324
The power of the performed test (0.324) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
120
Table 77. Linear regression of percent rock cover on rill severity class, 2001.
S tatistical A n alysis R esults______________________________________________
Rill Class Severity = 1.907 + (0.0153 * % rock cover)
N = 15.000
R = 0.233
Rsqr = 0.0543
Adj Rsqr = 0.000
Standard Error of Estimate = 0.724
Constant
MEAN
Coefficient
1.907
0.0153
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
Std. Error
0.801
0.0177
P
t
2.382 0.033
0.864 0.403
SS
MS
F
P
0.391 0.391 0.746 0.403
6.813 0.524
7.204 0.515
Normality Test:
Passed (P = 0.469)
Constant Variance Test:
Passed (P = 0.773)
Power of performed test with alpha = 0.050: 0.128
The power of the performed test (0.128) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
121
Table 78. Two way analysis of variance of runoff (mVha), 2000.
S tatistical A n alysis R esults_______
General Linear Model (No Interactions)
Dependent Variable: 2000 Total Annual Runoff (m3/ha)
Normality Test:
Passed (P = 0.093)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
1913.657
854.137
2875.583
5643.377
MS
478.414
427.069
359.448
403.098
F
P
1.331 0.338
1.188 0.353
The difference in the mean values among the different levels of TRT is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in REP. There is not a statistically significant
difference (P = 0.338).
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.353).
Power of performed test with alpha = 0.0500: for TRT : 0.0927
Power of performed test with alpha = 0.0500: for REP : 0.0696
Least square means for TRT :
Group
Mean
Coversoil
49.167
Coversoil/Pitting
39.733
Coversoil/Slash Barriers
45.300
Coversoil/AM inoculm
43.200
Control
71.533
Std Err of LS Mean = 10.946
Least square means for REP
Group
1.000
2.000
3.000
Std Err of LS Mean = 8.479
:
Mean
60.240
46.420
42.700
122
Table 79. Two way analysis of variance of runoff (nrVha), 2001.
S tatistical A n alysis R esults _________________________________
General Linear Model (No Interactions)
Dependent Variable: 2001Total Annual Runoff (m3/ha)
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
30451.777
4305.241
37386.959
72143.977
MS
7612.944
2152.621
4673.370
5153.141
F
P
1.629 0.258
0.461 0.647
The difference in the mean values among the different levels of TRT is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in REP. There is not a statistically significant
difference (P = 0.258).
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.647).
Power of performed test with alpha = 0.0500: for TRT : 0.136
Power of performed test with alpha = 0.0500: for REP : 0.0500
Least square means for TRT :
Group
Mean
Coversoil
123.533
Coversoil/Pitting
66.000
Coversoil/Slash Barriers
119.567
CoversoiEAM inoculum
99.767
Control
202.700
Std Err of LS Mean = 39.469
Least square means for REP :
Group
Mean
1.000
133.840
2.000
98.360
3.000
134.740
Std Err of LS Mean = 30.572
123
Table 80. Linear regression of runoff on sediment yield, 2000.
S tatistical A n alysis R esults________________________________
log I (!(Sediment Yield 2000) = -1.165 + (0.000789 * 2000 Runoff)
N = 15.000
R = 0.0401
Rsqr = 0.00161
Adj Rsqr = 0.000
Standard Error of Estimate = 0.410
Coefficient
Constant
-1.165
2000 Runoff 0.000789
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
Std. Error
0.291
0.00545
SS
0.00351
2.182
2.186
t
P
-3.999 0.002
0.145 0.887
MS
0.00351
0.168
0.156
Normality Test:
Passed (P = 0.138)
Constant Variance Test:
Passed (P = 0.388)
F
0.0209
P
0.887
Power of performed test with alpha = 0.050: 0.034
The power of the performed test (0.034) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
124
Table 81. Linear regression of runoff on sediment yield, 2001.
S tatistical A nalysis R esults_________________
2001 Sediment Yield = 1.216 - (0.000382 * 2001 Runoff)
N = 15.000
R = 0.0262
Rsqr = 0.000684
Adj Rsqr = 0.000
Standard Error of Estimate = 1.087
Coefficient
Constant
1.216
2001 Runoff -0.000382
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
0.0105
15.365
15.375
Std. Error
0.569
0.00405
t
2.137
-0.0943
MS
0.0105
1.182
1.098
Normality Test:
Passed (P = 0.157)
Constant Variance Test:
Passed (P = 0.176)
P
0.052
0.926
F
0.00890
P
0.926
Power of performed test with alpha = 0.050: 0.031
The power of the performed test (0.031) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
125
Table 82. Linear regression of runoff on mean rill severity class, 2000.
S tatistical A nalysis R esults________________________________________
Mean Rill Severity Class = 0.869 + (0.00000355 * 2000 Runoff)
N = 15.000
R = 0.833
Rsqr = 0.694 Adj Rsqr = 0.670
Standard Error of Estimate = 0.048
Constant
2000 Runoff
Coefficient
0.869
0.00000355
Analysis of Variance:
DF
SS
Regression
I
0.0666
Residual
13
0.0294
Total
14
0.0960
Normality Test:
Std. Error
t
0.0338
25.699
0.000000654 5.425
P
<0.001
<0.001
MS
0.0666
0.00226
0.00686
P
<0.001
Passed (P = 0.069)
Constant Variance Test:
Failed (P = <0.001)
Power of performed test with alpha = 0.050: 0.986
F
29.430
126
Table 83. Linear regression of runoff to mean rill severity class, 2001
S tatistical A n alysis R esults
2001 Mean Rill Severity Class = 2.733 - (0.0000000359 * 2001 Runoff)
N = 15.000
R = 0.214
Rsqr = 0.0457 Adj Rsqr = 0.000
Standard Error of Estimate = 0.723
Coefficient
Constant
2.733
2001 Runoff -0.0000000359
Std. Error
0.274
0.0000000456
Analysis of Variance:
DF
Regression I
Residual
13
Total
14
MS
0.325
0.523
0.508
SS
0.325
6.793
7.119
t
9.983
-0.789
F
0.623
P
<0.001
0.444
P
0.444
Normality Test:
Passed (P = 0.314)
Constant Variance Test:
Passed (P = 0.287)
Power of performed test with alpha = 0.050: 0.114
The power of the performed test (0.114) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
127
Table 84. Two way analysis of variance on biomass (kg/ha), 2001.
S tatistical A nalysis R esults_____________
General Linear Model (No Interactions)
Dependent Variable: 2001 Biomass (kg/ha)
Normality Test:
Passed (P = 0.079)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
66363.951
1443.316
19958.757
87766.024
MS
16590.988
721.658
2494.845
6269.002
F
P
6.650 0.012
0.289 0.756
The difference in the mean values among the different levels of TRT is greater than
would be expected by chance after allowing for effects of differences in REP. There is a
statistically significant difference (P = 0.012). To isolate which group(s) differ from the
others use a multiple comparison procedure.
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.756).
Power of performed test with alpha = 0.0500: for TRT : 0.826
Power of performed test with alpha = 0.0500: for REP : 0.0500
Least square means for TRT :
Group
Mean
Coversoil
185.467
Cvr/Pitting
128.967
Cvr/SIB
224.633
Cvr/AM inoc.
129.567
Control
27.467
Std Err of LS Mean = 28.838
Least square means for REP :
Group
Mean
1.000
126.280
2.000
150.020
3.000
141.360
Std Err of LS Mean = 22.338
128
Table 84. Two way analysis of variance on biomass (kg/ha), 2001 continued.
S tatistical A n alysis R esults con tin u ed
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Cvr/SIBvs. Control
Cvr/SIBvs. Cvr/Pitting
Cvr/SIBvs. Cvr/AM inoc.
Cvr/SIBvs. Coversoil
Coversoil vs. Control
Coversoil vs. Cvr/Pitting
Coversoil vs. Cvr/AM inoc.
Cvr/AM inoc. . vs. Control
Cvr/AM inoc. vs. Cvr/Pitting
Cvr/Pitting vs. Control
Difif of Means p
197.167
5
95.667
4
95.067
3
39.167
2
158.000
4
56.500
3
55.900
2
102.100
3
0.600
2
101.500
2
Comparisons for factor: REP
Comparison
Diff of Means
2.000 vs. 1.000
23.740
2.000 vs. 3.000
8.660
3.000 vs. 1.000
15.080
p
3
2
2
q
1.063
0.388
0.675
P
q
6.837 0.008
3.317 0.166
3.297 0.108
1.358 0.365
5.479 0.020
1.959 0.392
1.938 0.208
3.541 0.084
0.0208 0.989
3.520 0.038
P O .050
Yes
No
Do Not Test
Do Not Test
Yes
Do Not Test
Do Not Test
No
Do Not Test
Do Not Test
P
0.741
0.791
0.646
P<0.050
No
Do Not Test
Do Not Test
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
129
Table 85. Two way analysis of variance on percent canopy cover, 2001.
S tatistical A n alysis R esults________________________
General Linear Model (No Interactions)
Dependent Variable: 2001 Percent Canopy Cover
Normality Test:
Passed (P = 0.061)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
202.400
14.800
39.200
256.400
MS
50.600
7.400
4.900
18.314
F
10.327
1.510
P
0.003
0.278
The difference in the mean values among the different levels of TRT is greater than
would be expected by chance after allowing for effects of differences in REP. There is a
statistically significant difference (P = 0.003). To isolate which group(s) differ from the
others use a multiple comparison procedure.
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.278).
Power of performed test with alpha = 0.0500: for TRT : 0.968
Power of performed test with alpha = 0.0500: for REP : 0.106
Least square means for TRT :
Group
Mean
Coversoil
16.000
Cvr/Pitting
14.667
slash
17.000
Cvr/AM inoc.
14.667
Control
6.667
Std Err of LS Mean = 1.278
130
Table 85. Two way analysis of variance on percent canopy cover, 2001 continued.
S tatistical A n alysis R esults con tin u ed
Least square means for REP :
Group
Mean
1.000
14.400
2.000
14.600
3.000
12.400
Std Err of LS Mean = 0.990
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Diff of Means p
Cvr/SIB vs. Control
10.333
5
Cvr/SIB vs. Cvr/AM inoc.
2.333
4
Cvr/SIB vs. Cvr/Pitting
2.333
3
Cvr/SIB vs. Coversoil
1.000
2
Coversoil vs. Control
9.333
4
Coversoil vs. Cvr/AM inoc. 1.333
3
Coversoil vs. Cvr/Pitting
1.333
2
Cvr/Pitting vs. Control
8.000
3
Cvr/Pitting vs. Cvr/AM inoc. 0.000
2
Cvr/AM inoc. vs. Control
8.000
2
q
8.085
1.826
1.826
0.782
7.303
1.043
1.043
6.260
0.000
6.260
P
0.003
0.593
0.439
0.595
0.004
0.749
0.482
0.006
1.000
0.002
Comparisons for factor: REP
Comparison
Diff of Means
2.000 vs. 3.000
2.200
2.000 vs. 1.000
0.200
1.000 vs. 3.000
2.000
P
0.311
0.890
0.191
P<0.050
No
Do Not Test
Do Not Test
p
3
2
2
q
2.222
0.202
2.020
P<0.050
Yes
No
Do Not Test
Do Not Test
Yes
Do Not Test
Do Not Test
Yes
Do Not Test
Yes
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
131
Table 86. Two way analysis of variance on percent basal cover, 2001.
S tatistical A n alysis R esults______________________
General Linear Model (No Interactions)
Dependent Variable: % Basal Cover
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 1.000)
Source of Variation
TRT
REP
Residual
Total
DF
4
2
8
14
SS
46.649
2.505
19.715
68.869
MS
11.662
1.253
2.464
4.919
F
4.732
0.508
P
0.030
0.620
The difference in the mean values among the different levels of TRT is greater than
would be expected by chance after allowing for effects of differences in REP. There is a
statistically significant difference (P = 0.030). To isolate which group(s) differ from the
others use a multiple comparison procedure.
The difference in the mean values among the different levels of REP is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in TRT. There is not a statistically significant
difference (P = 0.620).
Power of performed test with alpha = 0.0500: for TRT : 0.633
Power of performed test with alpha = 0.0500: for REP : 0.0500
Least square means for TRT
Group
Coversoil
Coversoil/Pitting
Coversoil/Slash Barrier
Coversoil/AM inoculum
Control
Std Err of LS Mean = 0.906
:
Least square means for REP :
Group Mean
1.000 4.820
2.000 3.820
3.000 4.280
Std Err of LS Mean = 0.702
Mean
6.033
4.133
5.367
5.000
1.000
132
Table 86. Two way analysis of variance on percent basal cover, 2001 continuedsta tistic a l A nalysis R esults
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: TRT
Comparison
Diff of Means p
Coversoil vs. Control
5.033
5
Coversoil vs. Cvr/Pitting
1.900
4
Coversoil vs. Cvr/AM inoc. 1.033
3
Coversoil vs. Cvr/Sl.B
0.667
2
Cvr/Sl.B.vs. Control
4.367
4
Cvr/Sl.B.vs. Cvr/Pitting
1.233
3
Cvr/Sl.B.vs. Cvr/AM inoc. 0.367
2
Cvr/AM inoc. vs. Control
4.000
3
Cvr/AM inoc. vs. Cvr/Pitting 0.867
2
Cvr/Pitting vs. Control
3.133
2
Comparisons for factor: REP
Comparison
Diff of Means
1.000 vs. 2.000
1.000
1.000 vs. 3.000
0.540
3.000 vs. 2.000
0.460
P
3
2
2
q
P
5.553
2.096
1.140
0.736
4.818
1.361
0.405
4.413
0.956
3.457
0.027
0.489
0.710
0.617
0.038
0.619
0.782
0.034
0.518
0.040
P<0.050
Yes
No
Do Not Test
Do Not Test
Yes
Do Not Test
Do Not Test
Yes
Do Not Test
Yes
q
P
P 0.050
1.424 0.593 No
0.769 0.602 Do Not Test
0.655 0.656 Do Not Test
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
133
Table 87. Linear regression of biomass (kg/ha) on sediment yield, 2001.
S tatistical A nalysis R esults__________________
2001 Sediment Yield (Mg/ha) = 0.629 + (0.00388 * 2001 Biomass (kg/ha))
N = 15.000
R = 0.294
Rsqr = 0.0864 Adj Rsqr = 0.0161
Standard Error of Estimate = 1.038
Constant
Biomass (kg/ha)
Coefficient
0.629
0.00388
Std. Error
0.557
0.00350
t
1.130
1.109
P
0.279
0.288
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
1.324
14.006
15.331
MS
1.324
1.077
1.095
F
1.229
P
0.288
Normality Test:
Passed (P = 0.316)
Constant Variance Test:
Passed (P = 0.046)
Power of performed test with alpha = 0.050: 0.181
The power of the performed test (0.181) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
134
Table 88. Linear regression of percent canopy cover on sediment yield, 2001.
S tatistical A nalysis R esults_________________
2001 Sediment Yield (Mg/ha) = 1.186 - (0.00121 * percent canopy cover)
N = 15.000
R = 0.00497
Rsqr = 0.0000247
Adj Rsqr = 0.000
Standard Error of Estimate = 1.086
Constant
% canopy cover
Coefficient
1.186
-0.00121
Std. Error
0.977
0.0678
t
1.214
-0.0179
P
0.246
0.986
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
0.000378
15.330
15.331
MS
0.000378
1.179
1.095
F
0.000321
P
0.986
Normality Test:
Passed (P = 0.198)
Constant Variance Test:
Passed (P = 0.052)
Power of performed test with alpha = 0.050: 0.026
The power of the performed test (0.026) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
135
Table 89. Linear regression of percent basal cover on sediment yield, 2001.
S tatistical A nalysis R esults_____________________
2001 Sediment Yield (Mg/ha) = 1.635 - (0.108 * % basal cover)
N = 15.000
R = 0.229
Rsqr = 0.0524
Adj Rsqr = 0.000
Standard Error of Estimate = 1.057
Coefficient
Constant
1.635
%basal cover -0.108
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
0.804
14.527
15.331
Std. Error
0.613
0.127
t
2.668
-0.848
MS
0.804
1.117
1.095
Normality Test:
Passed (P = 0.018)
Constant Variance Test:
Passed (P = 0.753)
P
0.019
0.412
F
0.719
P
0.412
Power of performed test with alpha = 0.050: 0.125
The power of the performed test (0.125) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
136
Table 90. Linear regression of biomass (kg/ha) on runoff, 2001.
S tatistical A nalysis R esults____________________________________
2001 Runoff(m3/ha) = 176.081 - (0.386 * 2001 Biomass (kg/ha)
N = 15.000
R = 0.426
Rsqr = 0.181 Adj Rsqr = 0.118
Standard Error of Estimate = 67.399
Constant
OlPROD
Coefficient
Std. Error
176.081
36.139
-0.386 0.228 -1.698
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
13090.577
59053.400
72143.977
t
4.872
0.113
MS
13090.577
4542.569
5153.141
Normality Test:
Passed (P = 0.399)
Constant Variance Test:
Passed (P = 0.566)
P
<0.001
F
P
2.882 0.113
Power of performed test with alpha = 0.050: 0.351
The power of the performed test (0.351) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
137
Table 91. Linear regression of percent canopy cover on runoff, 2001.
S tatistical A n alysis R esults_______________________________________
2001 Runoff = 217.996 - (6.934 * Percent Canopy Cover)
N = 15.000
R = 0.413
Rsqr = 0.171 Adj Rsqr = 0.107
Standard Error of Estimate = 67.833
Constant
% Canopy
Coefficient
217.996
-6.934 4.236
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
12326.172
59817.805
72143.977
Std. Error
61.028
-1.637
t
3.572
0.126
MS
12326.172
4601.370
5153.141
F
P
2.679 0.126
Normality Test:
Passed (P = 0.071)
Constant Variance Test:
Passed (P = 0.985)
P
0.003
Power of performed test with alpha = 0.050: 0.331
The power of the performed test (0.331) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
138
Table 92. Linear regression of percent basal cover on runoff, 2001.
S tatistical A nalysis R esults______
2001 Runoff —173.771 - (11.948 * Percent Basal Cover)
N = 15.000
R = 0.369
Rsqr = 0.136 Adj Rsqr = 0.0698
Standard Error of Estimate = 69.233
Coefficient
Constant
173.771
%basal cover -11.948
Std. Error
40.130
8.343
Analysis of Variance:
DF
SS
Regression
I
9832.151
Residual
13
62311.826
Total
14
72143.977
t
4.330
-1.432
MS
9832.151
4793.217
5153.141
Normality Test:
Passed (P = 0.197)
Constant Variance Test:
Passed (P = 0.359)
P
<0.001
0.176
F
P
2.051 0.176
Power of performed test with alpha = 0.050: 0.268
The power of the performed test (0.268) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
139
Table 93. Linear regression of biomass (kg/ha) on rill severity, 2001.
S tatistical A nalysis R esults___________________
2001 Mean Rill Severity Class = 2.601 - (0.000152 * 2001 Biomass (kg/ha)
N = 15.000
R = 0.0168
Rsqr = 0.000281
Adj Rsqr = 0.000
Standard Error of Estimate = 0.744
Coefficient
Std. Error
t
P
Constant
2.601
0.399
6.518
<0.001
2001 Biomass -0.000152
0.00251
-0.0605
0.953
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
SS
0.00203
7.202
7.204
MS
0.00203
0.554
0.515
Normality Test:
Passed (P = 0.200)
Constant Variance Test:
Passed (P = 0.913)
F
0.00366
P
0.953
Power of performed test with alpha = 0.050: 0.029
The power of the performed test (0.029) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
140
Table 94. Linear regression of percent canopy cover on rill severity, 2001.
S tatistical A n alysis R esults
2001 Mean Rill Severity Class —3.240 - (0.0478 * 2001 Percent Canopy Cover)
N = 15.000
R = 0.285
Rsqr = 0.0814 Adj Rsqr = 0.0107
Standard Error of Estimate = 0.713
Constant
% Canopy Cover
Coefficient
3.240
-0.0478
Analysis of Variance:
DF
Regression I
Residual
13
Total
14
SS
0.586
6.618
7.204
Std. Error
0.642
0.0446
t
5.047
-1.073
P
< 0.001
0.303
MS
F
P
0.586 1.152 0.303
0.509
0.515
Normality Test:
Passed (P = 0.680)
Constant Variance Test:
Passed (P = 0.695)
Power of performed test with alpha = 0.050: 0.173
The power of the performed test (0.173) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
141
Table 95. Linear regression of percent basal cover on rill severity, 2001.
S tatistical A nalysis R esults___________________________
2001 Mean Rill Severity Class = 2.946 - (0.0851 * 2001 Percent Basal Cover)
N = 15.000
R = 0.263
Rsqr = 0.0692 Adj Rsqr = 0.000
Standard Error of Estimate = 0.718
Coefficient
Constant
2.946
%basal cover -0.0851
Analysis of Variance:
DF
Regression
I
Residual
13
Total
14
Std. Error
0.416
0.0865
t
7.077
-0.983
P
<0.001
0.344
SS
MS
F
P
0.498 0.498 0.966 0.344
6.706 0.516
7.204 0.515
Normality Test:
Passed (P = 0.328)
Constant Variance Test:
Passed (P = 0.120)
Power of performed test with alpha = 0.050: 0.152
The power of the performed test (0.152) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
142
Table 96. Multiple linear regression of mean rill severity class, runoff, and slope area on
________ sediment yield, 2000. ___________________________________________
S tatistical A n alysis R esults____________________________
2000 Sediment yield (Mg/ha) = -1.101 + (0.000335 * runoff (m3/ha)) + (1.122 * mean rill
severity class) + (0.00115 * slope area (m2))
N = 15.000
R = 0.881
Rsqr = 0.777 Adj Rsqr = 0.716
Standard Error of Estimate = 0.062
Constant
runoff (m3/ha)
mean rill sev. class
slope area (m2)
Coefficient
-1.101
0.000335
1.122
0.00115
Std. Error
0.230
0.000946
0.209
0.00140
t
-4.781
0.354
5.371
0.818
Analysis of Variance:
DF
Regression
3
Residual
11
Total
14
SS
0.148
0.0426
0.191
MS
0.0493
0.00387
0.0136
F
12.741
Column
runoff (m3/ha)
mean rill sev. class
slope area (m2)
SSIncr
0.00217
0.143
0.00259
SSMarg
0.000485
0.112
0.00259
P
<0.001
0.730
<0.001
0.431
VIF
1.306
1.227
1.305
P
<0.001
The dependent variable 2000 Sediment yield (Mg/ha) can be predicted from a linear
combination of the independent variables:
P
runoff (m3/ha) 0.730
mean rill sev. class <0.001
slope area (m2)
0.431
Not all of the independent variables appear necessary (or the multiple linear model may
be underspecified).
The following appear to account for the ability to predict 2000 Sediment yield (Mg/ha) (P
< 0.05): mean rill severity class
Normality Test:
Failed (P = 0.001)
Constant Variance Test:
Passed (P = 0.010)
Power of performed test with alpha = 0.050: 0.998
143
Table 97. Multiple linear regression of mean rill severity class, runoff, slope area,
________ biomass, canopy cover, basal cover and rock cover on sediment yield, 2001.
S tatistical A n alysis R esults______________
Sediment yield 2001 = -2.950 + (0.164 * mean rill severity class) - (0.00337 * runoff
(m3/ha)) + (0.0196 * slope area (m2)) + (0.0156 * biomass (kg/ha)) (0.0635 * canopy cover (%)) - (0.139 * basal cover (%)) + (0.0487 *
rock cover (%))
N = 15.000
R = 0.927
Rsqr = 0.859 Adj Rsqr = 0.718
Standard Error of Estimate = 0.556
Coefficient
Std. Error
t
P
VIE
Constant
-2.950
1.571
-1.878 0.102
Mean Rill Sev. Class 0.164
0.659
0.249 0.811 9.977
Runoff (m3/ha)
-0.00337
0.00250
-1.349 0.219 1.454
Slope Area (m2)
0.0196
0.0116
1.679 0.137 8.782
Biomass (kg/ha)
0.0156
0.00796
1.962 0.091 17.979
%Canopy Cover
-0.0635
0.0979
-0.649 0.537 7.936
% Basal Cover
-0.139
0.0916
-1.512 0.174 1.867
% Rock Cover
0.0487
0.0301
1.620 0.149 4.883
Warning: Multicollinearity is present among the independent variables. The variables
with the largest values of VIF are causing the problem. Consider getting more data or
eliminating one or more variables from the equation. The likely candidates for
elimination are: Mean Rill Severity Class, Slope Area, Biomass, % Canopy Cover, %
Rock Cover
Analysis of Variance:
DF
Regression
7
Residual
7
Total
14
SS
13.208
2.168
15.375
Column
Mean Rill Severity Class
Runoff (m /ha)
Slope Area (m2)
Biomass (kg/ha)
% Canopy Cover
% Basal Cover
% Rock Cover
MS
F
P
1.887 6.093 0.015
0.310
1.098
SSIncr
9.459
1.346
0.144
0.494
0.102
0.850
0.812
SSMarg
0.0192
0.564
0.873
1.192
0.130
0.708
0.812
144
Table 97. Multiple linear regression of mean rill severity class, runoff, slope area,
biomass, canopy cover, basal cover and rock cover on sediment yield, 2001
________ continued.__________________________________________________
S tatistical A nalysis R esults continued___________________
The dependent variable Sediment yield 2001 can be predicted from a linear combination
of the independent variables:
P
Mean Rill Severity Class
0.811
Runoff (m3/ha)
0.219
Slope Area (m2)
0.137
Biomass (kg/ha)
0.091
% Canopy Cover
0.537
% Basal Cover
0.174
% Rock Cover
0.149
Not all of the independent variables appear necessary (or the multiple linear model may
be underspecified). The following appear to account for the ability to predict Sediment
yield 2001 (P <005): [None]
Normality Test:
Constant Variance Test:
Passed (P = 0.400)
Passed (P = 0.050)
Power of performed test with alpha = 0.050: 1.000
145
Table 98. Multiple linear regression of mean rill severity class and slope area on runoff,
_________ 2000._________________________________________________________
S tatistical A nalysis R esults__________________________________________________________
Runoff 2000 (mVha) = 117.807 - (79.027 * mean rill severity class) + (0.228 * slope area
(m '))
N = 15.000
R = 0.486
Rsqr = 0.236 Adj Rsqr = 0.109
Standard Error of Estimate = 18.957
Constant
mean rill sev. class
slope area (m2)
Coefficient
117.807
-79.027
0.228
Std. Error
60.638
59.064
0.138
t
1.943
-1.338
1.656
Analysis of Variance:
DF
Regression
2
Residual
12
Total
14
SS
1331.619
4312.566
5644.185
MS
665.809
359.380
403.156
F
1.853
Column
mean rill sev. class
slope area(m2)
SSIncr
346.189
985.430
SSMarg
643.359
985.430
P
VIF
0.076
0.206 1.057
0.124 1.057
P
0.199
The dependent variable Runoff 2000 (m3/ha) can be predicted from a linear combination
of the independent variables:
P
mean rill sev. class 0.206
slope area (m2)
0.124
Not all of the independent variables appear necessary (or the multiple linear model may
be underspecified).
The following appear to account for the ability to predict Runoff 2000(m3/ha) (P < 0.05):
[ None ]
Normality Test:
Passed (P = 0.011)
Constant Variance Test:
Passed (P = 0.014)
146
Table 99. Multiple linear regression of mean rill severity class, slope area, biomass,
________ canopy cover, basal cover and rock cover on runoff, 2001.____________
S tatistical A n alysis R esults____________________________________________________
2001 Runoff (m3/ha) = 50.774 + (50.929 * Mean Rill Severity Class) - (0.332 * Slope
Area (m2)) - (0.677 * Biomass (kg/ha)) + (6.439 * % Canopy
Cover) - (5.528 * % Basal Cover) - (0.201 * % Rock Cover)
N = 15.000
R = 0.559
Rsqr = 0.312 Adj Rsqr = 0.000
Standard Error of Estimate = 78.750
Coefficient
Constant
50.774
Mean Rill Sev. Class 50.929
Slope Area (m2)
-0.332
Biomass (kg/ha)
-0.677
% Canopy Cover
6.439
% Basal Cover
-5.528
% Rock Cover
-0.201
Std. Error
221.551
91.473
1.644
1.101
13.666
12.819
4.254
t
0.229
0.557
-0.202
-0.615
0.471
-0.431
-0.0472
P
0.824
0.593
0.845
0.556
0.650
0.678
0.964
VIF
9.605
8.738
17.167
7.722
1.825
4.881
Warning: Multicollinearity is present among the independent variables. The variables
with the largest values of VIE are causing the problem. Consider getting more data or
eliminating one or more variables from the equation. The likely candidates for
elimination are: Mean Rill Severity Class, Slope Area, Biomass, % Canopy Cover, %
Rock Cover
Analysis of Variance:
DF
Regression
6
Residual
8
Total
14
SS
22531.146
49612.831
72143.977
Column
Mean Rill Severity Class
Slope Area (m2)
Biomass (kg/ha)
% Canopy Cover
% Basal Cover
% Rock Cover
MS
3755.191
6201.604
5153.141
SSIncr
7556.642
2152.658
10758.519
909.196
1140.326
13.805
F
P
0.606 0.721
SSMarg
1922.390
253.172
2346.584
1376.501
1153.387
13.805
147
Table 99. Multiple linear regression of mean rill severity class, slope area, biomass,
________ canopy cover, basal cover and rock cover on runoff, 2001 continued.
S tatistical A nalysis R esults con tin u ed ______________________________
The dependent variable 2001 Runoff (m3/ha) can be predicted from a linear combination
of the independent variables:
P
Mean Rill Severity Class
0.593
Slope Area (m2)
0.845
Biomass (kg/ha)
0.556
% Canopy Cover
0.650
% Basal Cover
0.678
0.964
% Rock Cover
Not all of the independent variables appear necessary (or the multiple linear model may
be underspecified).
The following appear to account for the ability to predict 2001 Runoff (m3/ha) (P < 0.05):
[ None ]
Normality Test:
Failed (P = 0.007)
Constant Variance Test:
Passed (P = 0.194)
Power of performed test with alpha = 0.050: 0.590
The power of the performed test (0.590) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
148
Table 100. Multiple linear regression of maximum hourly precipitation, increase in rill
__________ severity class and runoff on sediment yield, 2000 and 2001.____________
S tatistical A nalysis R esults
Sediment Yield 2000 and 20001 =0.120 - (1.271 * maximum hourly precipitation (cm))
(0.124 * increase in rill severity class) + (0.0445 *
runoff (m3/ha))
N = 6.000
R = 0.995
Rsqr = 0.990 Adj Rsqr 0.975
Standard Error of Estimate = 0.050
Std. Error
Coefficient
P
VIF
t
Constant
0.120
0.0328
0.067
3.663
Max. Hourly Ppt. (cm)-1.271
0.101
0.006 8.065
-12.583
Inc. Rill Sev. Class -0.124
-4.174
0.0296
0.053 1.938
Runoff (m3/ha)
0.0445
12.448
0.00357
0.006 9.654
Warning: Multicollinearity is present among the independent variables. The variables
with the largest values of VIF are causing the problem. Consider getting more data or
eliminating one or more variables from the equation. The likely candidates for
elimination are: Max. Hourly Ppt., Runoff
Analysis of Variance:
DF
Regression
3
Residual
2
Total
5
Column
Max. Hourly Ppt.(cm)
Inc. Rill Sev. Class
Runoff (rnVha)
SS
0.494
0.00506
0.499
SSIncr
0.00878
0.0932
0.392
MS
0.165
0.00253
0.0998
F
65.079
P
0.015
S
0.401
0.0441
0.392
The dependent variable Sediment Yield 2000 and 2001 can be predicted from a linear
combination of the independent variables:
P
Max. Hourly Ppt. (cm)
0.006
Inc. Rill Sev. Class
0.053
Runoff (m3/ha)
0.006
Not all of the independent variables appear necessary (or the multiple linear model may
be underspecified). The following appear to account for the ability to predict Sediment
Yield 2000 and 2001 (P < 0.05): Max. Hourly Ppt, Runoff
Normality Test:
Passed (P = 0.216)
Constant Variance Test:
Passed (P = 0.060)
Power of performed test with alpha = 0.050: 0.999______________________________
149
Table 101. T-test of mean percent AM colonization levels in the coversoil and spoil
_________ backfill material during 2000.__________________________________
S tatistical A n alysis R esults_____________________________________________________
t-test
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 0.554)
Group Name N
Coversoil
9
Control
9
Difference
Missing
0
0
Mean
39.444
30.333
Std Dev
8.719
9.513
SEM
2.906
3.171
9.111
t = 2.118 with 16 degrees of freedom. (P = 0.050)
95 percent confidence interval for difference of means: -0.00771 to 18.230
The difference in the mean values of the two groups is not great enough to reject the
possibility that the difference is due to random sampling variability. There is not a
statistically significant difference between the i nput groups (P = 0.050).
Power of performed test with alpha = 0.050: 0.408
The power of the performed test (0.408) is below the desired power of 0.800.
You should interpret the negative findings cautiously.
150
Table 102. Two way analysis of variance of percent AM colonization levels in Hordeum
__________vulgare, 2001.__________________________________________________
S tatistical A n alysis R esults__________________________________________________________
Two Way Analysis of Variance
Data source: % AM colonization levels in barley
Balanced Design
Dependent Variable: % AM colonization
Normality Test:
Passed (P = 0.059)
Equal Variance Test: Passed (P = 0.337)
Source of Variation
Treatment
REPS
Treatment x REPS
Residual
Total
DF
2
2
4
27
35
SS
0.000200
0.00912
0.0705
0.280
0.360
MS
0.000100
0.00456
0.0176
0.0104
0.0103
F
0.00965
0.440
1.702
P
0.990
0.649
0.179
The difference in the mean values among the different levels of Treatment is not great
enough to exclude the possibility that the difference is just due to random sampling
variability after allowing for the effects of differences in REPS. There is not a
statistically significant difference (P = 0.990).
The difference in the mean values among the different levels of REPS is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in Treatment. There is not a statistically
significant difference (P = 0.649).
The effect of different levels of Treatment does not depend on what level of REPS is
present. There is not a statistically significant interaction between Treatment and REPS.
(P = 0.179)
Power of performed test with alpha = 0.0500: for Treatment: 0.0500
Power of performed test with alpha = 0.0500: for REPS : 0.0500
Power of performed test with alpha = 0.0500: for Treatment x REPS : 0.194__________
151
Table 102. Two way analysis of variance of percent AM colonization levels in Hordeum
__________vulgare, 2001 continued._____________________________
S tatistical A nalysis R esults continued
Least square means for Treatment
Group
Mean
Coversoil
0.343
Control
0.343
Cvr/AM inoc.
0.348
Std Err of LS Mean ==0.0294
Least square means for REPS
Group
Mean
1.000
0.333
0.334
2.000
0.368
3.000
Std Err of LS Mean ==0.0294
Least square means for Treatment x REPS
Group
Mean
Coversoil x 1.000
0.353
Coversoil x 2.000
0.375
Coversoil x 3.000
0.302
Control x 1.000
0.333
Control x 2.000
0.350
Control x 3.000
0.348
Cvr/AM inoc. x 1.000
0.315
Cvr/AM inoc. x 2.000
0.278
Cvr/AM inoc. x 3.000
0.453
Std Err of LS Mean = 0.0509
152
Table 103. Two way analysis of variance of percent AM colonization levels in
_________ A gropyron trachycaulum , 2001.____________________________
S tatistical A n alysis R esults____________________________
Two Way Analysis of Variance
Data source: %AM colonization levels in slender wheatgrass
Balanced Design
Dependent Variable: % AM colonization
Normality Test:
Passed (P > 0.200)
Equal Variance Test: Passed (P = 0.240)
Source of Variation
Treatment
REPS
Treatment x REPS
Residual
Total
DF
2
2
4
27
35
SS
0.0551
0.0102
0.0137
0.135
0.214
MS
0.0276
0.00512
0.00342
0.00500
0.00612
F
5.505
1.023
0.684
P
0.010
0.373
0.609
The difference in the mean values among the different levels of Treatment is greater than
would be expected by chance after allowing for effects of differences in REPS. There is
a statistically significant difference (P = 0.010). To isolate which group(s) differ from the
others use a multiple comparison procedure.
The difference in the mean values among the different levels of REPS is not great enough
to exclude the possibility that the difference is just due to random sampling variability
after allowing for the effects of differences in Treatment. There is not a statistically
significant difference (P = 0.373).
The effect of different levels of Treatment does not depend on what level of REPS is
present. There is not a statistically significant interaction between Treatment and REPS.
(P - 0.609)
Power of performed test with alpha = 0.0500: for Treatment: 0.724
Power of performed test with alpha = 0.0500: for REPS : 0.0520
Power of performed test with alpha = 0.0500: for Treatment x REPS : 0.0500
Least square means for Treatment:
Group
Mean
Coversoil
0.457
Control
0.443
Cvr/AM inoc.
0.532
Std Err of ES Mean = 0.0204
153
Table 103. Two way analysis of variance of percent AM colonization levels in
_________ A gropyron trachycaulum , 2001 continued.___________________
S tatistical A n alysis R esults continued______________
Least square means for REPS :
Group
Mean
1.000
0.486
2.000
0.453
3.000
0.492
Std Err of ES Mean = 0.0204
Least square means for Treatment x REPS :
Group
Mean
Coversoil x 1.000
0.480
Coversoil x 2.000
0.405
Coversoil x 3.000
0.485
Control x 1.000
0.440
Control x 2.000
0.455
Control x 3.000
0.433
Cvr/AM inoc. x 1.000
0.538
Cvr/AM inoc. x 2.000
0.500
Cvr/AM inoc. x 3.000
0.558
Std Err of LS Mean =
0.0354
All Pairwise Multiple Comparison Procedures (Student-Newman-Keuls Method):
Comparisons for factor: Treatment
Comparison
Diff of Means
P
P0.050
P
q
Cvr/AM inoc. vs. Control
0.0892
3
4.366 0.013
Yes
Cvr/AM inoc. vs. Coversoil
0.0750
2
3.673 0.015
Yes
Cvr/AM inoc. vs. Control
0.0142
2
0.694 0.628
No
Comparisons for factor: REPS
Comparison
Diff of Means p
P
P 0.050
q
3.000 vs. 2.000
0.0383
3
1.877 0.393
No
3.000 vs. 1.000
0.00583
2
0.286 0.842
Do Not Test
1.000 vs. 2.000
0.0325
2
1.591 0.270
Do Not Test
A result of "Do Not Test" occurs for a comparison when no significant difference is
found between two means that enclose that comparison. For example, if you had four
means sorted in order, and found no difference between means 4 vs. 2, then you would
not test 4 vs. 3 and 3 vs. 2, but still test 4 vs. I and 3 vs. I (4 vs. 3 and 3 vs. 2 are
enclosed by 4 vs. 2: 4 3 2 I). Note that not testing the enclosed means is a procedural
rule, and a result of Do Not Test should be treated as if there is no significant difference
between the means, even though one may appear to exist.
MONTANA
state