United States Office of Water 841-R-93-006
Environmental Protection Washington, D.C. 20460 January 1993
Agency
oEPA Post-Audit Verification Of
The Model AGNPS In
Vermont Agricultural
Watersheds
-------
POST-AUDIT VERIFICATION OF THE MODEL AGNPS
IN VERMONT AGRICULTURAL WATERSHEDS
FINAL REPORT TO:
ENVIRONMENTAL PROTECTION AGENCY
NONPOINT SOURCE CONTROL BRANCH
SUBMITTED BY:
John C. Clausen
UNIVERSITY OF CONNECTICUT
DEPARTMENT OF NATURAL RESOURCES MANAGEMENT AND ENGINEERING
January 25, 1993
-------
ABSTRACT
The purpose of this project was to conduct a verification of the model AGNPS
(Agricultural Non-Point Source Pollution Model, Ver. 2.52, 1988). The observed runoff, and
concentration and mass export of sediment, nitrogen, and phosphorus from two Vermont
agricultural watershed were compared to simulated values for a total of 15 storms in one
watershed and 11 storms in another watershed. AGNPS underpredicted discharge except for the
largest (1-2 inch) storms. Sediment and nitrogen concentrations and exports were overpredicted
by AGNPS. Predicted phosphorus concentrations were of the same order of magnitude as
observed. Phosphorus exports were underpredicted. The differences between observed and
predicted values obtained in this verification of AGNPS were greater than previously reported.
The testing of the model in a completely different climatic region may explain the differences
obtained in the accuracy of the model.
Trade names are used in this publication solely for the purpose of providing specific information.
Mention of a trade name does not constitute a guarantee or warranty of the product by the U.S.
Environmental Protection Agency or an endorsement by the Agency over other products not
mentioned.
-------
ACKNOWLEDGEMENT
This project was initially funded by the U.S. Environmental Protection Agency (EPA)
Grant No. R-815362-01-0, and completed under EPA contract No. 68-CS-0013. Mr. Steven
A. Dressing, EPA Project Officer for the grant, is especially acknowledged for his guidance of
this project. Dr. Robert A. Young, USDA- Agricultural Research Service, Morris, MN and
model author, provided a great deal of advice and assistance in applying the model. Most of
the results presented in this report are based on work conducted by Mr. Michael Cassara, a
graduate student in the School of Natural Resources, University of Vermont. Mr. Donald W.
Meals, Jr. is gratefully acknowledged for his assistance in analyzing the climate data for the
LaPlatte watershed and for his overall assistance in the project. Mr. Jay Appleton assisted with
the geographic information system (CIS) programming. Ms. Bonnie Bradshaw, an
undergraduate student in the Department of Natural Resources Management and Engineering,
University of Connecticut assisted with the one-cell runs. Significant review comments were
provided by Mr. Steven Dressing, Mr. Tom Davenport, and Mr. Bruce Zander, all of EPA, and
Dr. Leslie Shoemaker of Tetra Tech, Inc. that greatly improved this report. Both the School
of Natural Resources, University of Vermont and the Department of Natural Resources
Management and Engineering, University of Vermont are acknowledged for general support of
this project.
11
-------
Table of Contents
Page
Abstract i
Acknowledgement ii
List of Figures iv
List of Tables vi
Introduction 1
Summary of AGNPS 3
Objectives 8
Study Areas 9
Approach 13
Model Verification 13
Methods 18
Results and Discussion 29
Subwatershed 3 29
Jewett Brook 45
Single cell . 56
Event extrapolations s. 61
Conclusions 62
Recommended Verification Procedure 63
References 65
Glossary 68
Appendices 70
Hydrographs 70
Example input files 97
111
-------
LIST OF FIGURES
Figure
1. Vermont map showing location of the two study watersheds . 10
2. Land use map of Laplatte River Subwatershed 3, 1989 11
3. Land use map of the Jewett Brook watershed, 1988 12
4. Steps in mathematical model development 14
5. Hydrograph for LaPlatte watershed 3 for September 21, 1983 storm 19
6. Recording precipitation chart for storm on September 21, 1983, Hannah gage... 20
7. LaPlatte subwatershed 3 with 10-acre grid cells and cell drainage paths 24
8. Jewett Brook watershed with 10-acre grid cells and cell drainage paths 25
9. Plot of discharge volume observed in LaPlatte River Subwatershed 3 and predicted
by AGNPS 36
10. Plot of peak discharge observed in LaPlatte River Subwatershed 3 and predicted
by AGNPS 36
11. Plot of sediment concentrations observed in runoff from LaPlatte River Subwater-
shed 3 and predicted by AGNPS 39
12. Plot of sediment export observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS 39
13. Plot of phosphorus concentrations observed in runoff from LaPlatte River Sub-
watershed 3 and predicted by AGNPS , 41
14. Plot of phosphorus export observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS 41
15. Plot of nitrogen concentrations observed in runoff from LaPlatte River Subwater-
shed 3 and predicted by AGNPS 44
16. Plot of nitrogen export observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS 44
IV
-------
LIST OF FIGURES (Continued)
Figure Page
17. Plot of discharge volume observed in Jewett Brook and predicted by AGNPS 50
18. Plot of peak discharge observed in Jewett Brook and predicted by AGNPS 50
19. Plot of sediment concentrations observed in Jewett Brook and predicted by AGNPS. 53
20. Plot of sediment export by Jewett Brook and predicted by AGNPS 53
21. Plot of phosphorus concentrations observed in Jewett Brook and predicted by
AGNPS . 55
22. Plot of phosphorus export by Jewett Brook and predicted by AGNPS 55
-------
LEST OF TABLES
Table
1. AGNPS input parameters
2. Rainfall energy intensity (El) calculation for September 21, 1983 storm, Laplatte
River Subwatershed 3 22
3. AGNPS default values and appropriate Vermont values for model adaption 28
4. Precipitation characteristics of modeled storms, LaPlatte River Subwatershed 3.... 30
5. Comparison of predicted discharge, P and N using default and computed AMC
and El values for two storms for LaPlatte Subwatershed 3 31
6. Mean observed values and values predicted by AGNPS with relative errors and
t-test between means for LaPlatte Subwatershed 3 33
7. Root MSE and significance of regressions between observed and AGNPS pre-
dicted values for the LaPlatte Subwatershed 3 34
8. Discharge observed from LaPlatte River Subwatershed 3 and predicted by
AGNPS 35
9. Sediment observed in runoff from LaPlatte River Subwatershed 3 and predicted by
AGNPS 38
10. Total phosphorus observed in runoff from LaPlatte River Subwatershed 3 and
predicted by AGNPS . 40
t
11. Nitrogen observed in runoff from LaPlatte River Subwatershed 3 and predicted
by AGNPS 43
12. Precipitation characteristics of modeled storms, Jewett Brook 46
13. Mean observed values (11 storms) and values predicted by AGNPS with relative
errors and t-test between means for Jewett Brook 47
14. Root MSE and significance of regressions between observed and AGNPS predicted
values for Jewett Brook 48
15. Discharge observed from Jewett Brook and predicted by AGNPS 49
VI
-------
LIST OF TABLES (Continued)
Table
16. Sediment observed in runoff from Jewett Brook and predicted by AGNPS 52
17. Total phosphorus observed in runoff from Jewett Brook and predicted by
AGNPS 54
18. Comparison of discharge observed from LaPlatte River Subwatershed 3 and pre-
dicted by AGNPS using 40 cells and 1 cell 57
19. Comparison of sediment observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS using 40 cells and 1 cell..... 58
20. Comparison of phosphorus observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS using 40 cells and 1 cell..... 59
21. Comparison of nitrogen observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS using 40 cells and 1 cell 60
vu
-------
INTRODUCTION
The purpose of this project was to perform a post-audit verification of the model AGNPS
(Agricultural Non-Point Source Pollution Model). This model was developed by USDA-
Agricultural Research Service in Morris, Minnesota to simulate sediment and nutrient export
from agricultural watersheds (Young et al., 1989). The version of the model used for this
verification (Ver. 2.52,1988), was single event-based and thus did not the prediction of periods
of no-flow or snowmelt. The model uses measured or estimated parameters as input and does
not require calibration or fitting. AGNPS was developed for watershed-scale applications and
can be applied to areas of up to 12, 000 ha (Onstad, et al., 1986). One of the unique
characteristics of AGNPS, as compared to other nonpoint source models, is that it uses
distributed watershed parameters and allows identification of up to 700 cells within a watershed.
However, rainfall information is lumped for a watershed. Model outputs include runoff volume
and peak rate, and the concentration and mass export of sediment, nitrogen (N), and phosphorus
(P) for the event.
AGNPS has received considerable attention nationally as a tool to detect and prioritize
nonpoint pollution problems in agricultural watersheds, and to assess the relative effects of
alternative conservation practices (Bartholic et al., 1987; Frevert and Crowder, 1987; Lee, 1987;
Young et al., 1987). However, comparisons of AGNPS predictions to observed data have been
limited. The model authors regressed predicted and observed peak flow for 20 watersheds in
the north central United States. The resulting R2 value was 0.81 (Young et. al., 1989). They
also compared observed and simulated sediment yield for two watersheds in Iowa (21 storms)
and one in Nebraska (8 storms). The Iowa results indicated that the model overpredicted
-------
sediment yield by two percent with a R2 value of 0.95; the Nebraska comparison resulted in a
R2 value of 0.76 (Bosch, et. al., 1983, Young, et. al., 1986). More recently, the model authors
reported observed and predicted concentrations of total N and total P for 20 locations in seven
Minnesota watersheds. The observed data were based on small (1-yr, 24-hr) storms. They
concluded that AGNPS gave realistic predictions of nutrient concentrations but did not provide
statistics on the goodness of fit between observed and predicted values (Young, et. al., 1989).
AGNPS was applied in the Highland Silver Lake watershed in Illinois at three sites (Lee,
1987, Lee and Comacho, 1987). Predicted and observed runoff volume and total suspended
solids exports, as a function of rainfall, were presented. They concluded that since model
predictions seemed to be an average of observed data, that the model simulations were
reasonably close to average field observations (Lee, 1987; Lee and Comacho, 1987). However,
these comparisons were not made statistically. They also compared average annual observed
exports (over 2.8 years) of sediment, N, and P to simulated annual values. Annualized
simulations were obtained from modeling seven storms representing certain precipitation
intervals. Modeled results were then multiplied by the frequency of storms per each interval and
'then summed to yield annual estimates (Lee, 1988, personnel communication). Lee (1987)
concluded that the model overpredicted total P load by five times, and total N load by 3.5 times.
Sediment loads differed by less than five percent.
D. German (1991, personal communication) compared observed and predicted values for
seven storms in 1989 in Loomis Creek, South Dakota. Loomis Creek monitoring is part of the
Oakwood Lakes-Poinsett Rural Clean Water Program (RCWP) project. He concluded that
AGNPS generally overpredicted discharge volume and peak, and the exports of sediment, N and
-------
p.
There have not been extensive tests of the model AGNPS, although usage of the model
is expanding. There has been a lack of verifications applying standard statistical tests as
suggested by Thomann (1982) and Reckhow et al. (1990). It is of benefit to the U.S.
Environmental Protection Agency (EPA) to assess whether AGNPS is suitable for assessing the
effectiveness of agricultural best management practices (BMPs). Both the EPA and the States
have multiple needs for such models, including: 1) comparing alternative pollution control plans,
2) developing total maximum daily loads (TMDLs), 3) locating critical areas in watersheds, and
4) estimating the water quality benefits to be gained from implementation of management
measures in the coastal zone under Section 6217 of the Coastal Zone Act Reauthorization
Amendments of 1990.
SUMMARY OF AGNPS
AGNPS (Agricultural Nonpoint Source) is an event-based, measured parameter,
watershed-scale, distributed model predicting discharge and the concentration and load of N, P,
-and sediment in runoff (Young et. al., 1987). The model has been described fully in Young et
al. (1987) and summarized in Bosch et. al., (1983), Onstad, et. al., (1986), and Young, et. al.,
(1989).
Flow
AGNPS predicts both runoff volume and peak runoff rate. Runoff volume from each cell
is estimated using the Soil Conservation Service (SCS) curve number (CN) technique (USDA-
SCS 1972):
-------
o = (P - 0.25)2 (1)
P + 0.85
where Q is the runoff volume, P is the precipitation, and S is a retention factor, all in uniform
units of length, such as inches or cm. The retention factor (S) is determined from:
5 = 2 - 10 (2)
CN
where CN is the curve number for the cell. The curve number is the percentage ratio between
stream discharge and precipitation, and varies with land use, hydrologic soil group, and
antecedent moisture content (AMC). Curve number are taken from standard tables provided in
the model documentation.
Peak runoff rate for each cell is determined from:
where Q, is the peak flow rate (cubic feet per second), A is the drainage area (acres), Sc is the
channel slope (ft/ft), RO is the runoff volume .(in), and LW is the watershed length-width ratio
which is determined from L2/A where L is the watershed length (ft). Although the units do not
cancel in this equation, they do cancel as used in the model (Young et al., 1986). This
procedure for estimating peak runoff was developed by Smith and Williams (1980) for the
CREAMS model.
-------
Erosion arid Sediment
Erosion is determined using the modified universal soil loss equation (USLE):
E = (EI)(K)(LS)(C)(PKSSF)
where E is the soil loss (tons/ac), El is the rainfall energy intensity, K is the soil credibility
factor, LS is the slope length and slope factor, C is the cover and management factor, P is the
practice factor, and SSF is the slope shape factor (Wischmeier and Smith, 1978).
Sediment is routed from cell to cell using a mass balance approach, and allowing for
deposition (Young et. al., 1989).
Nitrogen and Phosphorus
Both soluble and particulate forms of N and P are predicted using procedures found in
the CREAMS model (Frere, et. al., 1982). The concentration of soluble N or P is determined
from soil concentrations and extraction coefficients:
where Nut^ is the export (Ibs) of soluble N or P in runoff, C^ is the concentration (ppm) of
soluble N or P at the soil surface, Nut^ is an extraction coefficient for movement into runoff,
and Q is the runoff volume (in). The units cancel as used in the model (Young et al., 1986).
Nutrients transported in the sediment are based on soil nutrient concentrations, an
-------
enrichment ratio, and sediment yield from a cell:
where Nut^ is the N or P transported (Ibs) by the sediment in runoff, Nutf is the N or P content
(ppm) of the soil, and EH is the enrichment ratio determined from:
IT fAn-o^T
ER = 7.4QS Tf
where Q. the sediment yield (Ibs) and Tf is a correction factor for soil texture.
Model Input
Table 1 summarizes the input requirements for AGNPS ( Young et al., 1987). These
input parameters, as used in this study, are described in detail in the "Methods" section of this
report.
-------
Table 1. AGNPS Input Parameters.
Column
No.
Parameter
Source of
Input Data
Watershed
l
2
3
4
5
Cell
l
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
watershed identification
cell area (acres)
total number of cells
precipitation (inches)
energy - intensity value
cell number
number of cell into which it drains
SCS curve number
average land slope (%)
slope shape factor (uniform, convex, concave)
average field slope length (feet)
average channel slope (%) Se
average channel side slope (96)
manning's roughness coefficient (n) for channel
soil erodibility factor (K)
cropping factor (C)
practice factor (?)
surface condition constant based on land use
aspect of drainage from the cell
soil texture
fertilization level (zero, low, medium, high)
incorporation factor (% fertilizer, top 0.5 in. soil)
point source indicator
gully source level (estimate of gully erosion)
chemical oxygen demand factor
impoundment factor (terrace system)
channel indicator
User
User
CIS
Gage
Calculated
CIS
USGS topographic map
Young et al. (1987)
USGS topographic map
USGS topographic map
VT SCS, 1989
USGS topographic map
Young et al. (1987)
Young et al. (1987)
GIS - soil survey
Young et al. (1987)
Young et al. (1987)
GIS - Young et al. (1987)
USGS topographic map
GIS - Young et al. (1987)
GIS - land use monitoring
GIS - Young et al. (1987)
none
none
not applicable
none
GIS
-------
OBJECTIVES
The objectives of this study were to:
1) Perform a post-audit verification of the AGNPS model by comparing simulated runoff
and the concentration and mass export of sediment, nitrogen, and phosphorus to observed
values from two agricultural watersheds in Vermont to determine the accuracy of
AGNPS.
2) Assess methods of extrapolating event-based simulations to long-term findings so that
annualized information may be obtained from a series of event simulations.
8
-------
STUDY AREAS
Two watersheds in northwestern Vermont were used for the study (Figure 1). Both
watersheds were used predominantly for dairy agriculture and have been the sites of extensive
implementation of land treatment practices and comprehensive monitoring of water quality.
One study area, Subwatershed 3, was located in the LaPlatte River watershed
approximately 10 mi. south of Burlington, Vermont. Monitoring in this 400-ac watershed
occurred from 1979 to 1990 as part of the USD A Soil Conservation Service's small watershed
land treatment program (Public Law 566). The watershed monitoring program and results of
monitoring have been described in detail elsewhere (Cassell and Meals, 1981; Meals 1990).
Soils in the watershed were largely lacustrine silts and clays. Land uses within the watershed
were 77% agricultural, 19% forested, and 4% residential (Figure 2). Climate in the watershed
was continental. The normal precipitation was 34 inches and the mean annual temperature was
45°F (NOAA, 1983).
The second study area, the Jewett Brook watershed (Station 21), was located in the St.
Albans Bay watershed, Lake Champlain, approximately 35 miles north of Burlington.
.Monitoring in this 1,384-ha watershed occurred from 1981 to 1991 as part of a Comprehensive
Monitoring and Evaluation (CM & E) program associated with the St. Albans Bay Rural Clean
Water Program (RCWP) project. The monitoring program and results have been previously
described (Cassell et. al., 1983; Clausen, 1985; VT RCWP Coord. Com., 1991). Watershed
soils are predominantly lacustrine silts and clays. Land uses in the watershed were 83%
agricultural, 15% woodland, and 2% residential (Figure 3). Considering only the agricultural
land, 45% of the watershed was hayland, 23% was cornland, and 14% was pasture. Climate
-------
'*"( (/MCMPHICMAGOC ,
VERMONT
LOCATION
MAP
MONITORED AGRICULTURAL WATERSHEDS
Figure 1. Vermont map showing location of the two study watersheds.
10
-------
LEGEND
r7] Agriculture) land
Li-J ist unknown
ra Woodland
DUoo-ogt icullufol
loud
0
Miles
1
Figure 2. Land use map of Laplatte River Subwatershed 3, 1989.
11
-------
J liiles
LEGEND
7J Corn
f|] Hoylood
J] Posturelond
3 Formsteod
PT] Ajricullurol lond
^^ lit unknown
p;] Woodland
I1 Hoo-ootieuliufol
L-' loud
Figure 3. Land use map of the Jewett Brook Watershed, 1988.
12
-------
for the watershed was cool, humid, and continental. The mean annual temperature was 45° F
and the average annual precipitation was 33 inches (Cassell et al., 1983). Thunderstorms
occurred an average of 25 times per year.
APPROACH
Model Verification
Model verification refers to the testing of a model with new field data to determine
whether the model adequately predicts observed data (Thomann, 1982; Reckhow et al., 1991)
This process has also been called model testing, model validation, or model evaluation. This
is the final step in a series of stages in model development as outlined by Thomann (1982)
(Figure 4). The first step is problem identification. This step is needed to focus the modeling
effort. For example, the AGNPS model was originally developed to analyze and prioritize
agricultural watersheds in Minnesota in order to correctly direct public funds toward solving
pollution problems on a watershed basis. A method to systematically prioritize watersheds was
previously lacking. Concqpfiifll modeling refers to a description of the model components,
inputs, and outputs as are often described in flow charts. Next, the theoretical equations for the
t model are written, followed by the setting of appropriate quantities for default parameter values.
Model calibration is fitting the model output to observed data. Preferably, calibration is done
with a data set different from that used for the original model construction. Sometimes, when
several years of observed data are available, a portion of the data set has been used for model
construction, and the remaining data was used for model calibration. Model calibration usually
includes "tests of reason", including whether the model is predicting "reasonable" values with
reasonable input data.
13
-------
Observed
Data Set #1
Observed
Data Set #2
Observed
Data Set #3
Problem
Identification
Conceptual
Model
(flow chart)
Theoretical
Equations
Parameter
Values
Model
Calibration
Model
Verification
Simulated
Output
Figure 4. Steps in mathematical model development (after Thomann, 1982).
14
-------
Verification Methods
There are several measures used for model verification, most of which are statistical.
Each of the measures used in this verification are described briefly below. Readers that are
unfamiliar with the statistical terms used in this report should consult a statistics textbook.
1. Bivariate plot. A plot of simulated data as a function of observed data can provide a
good qualitative evaluation of model performance (Jamieson and Clausen, 1988; Reckhow, et.
al., 1990; Thomann, 1982). An example of a bivariate plot is shown in the "Results and
Discussion" section of this report (Figure 9).
2. Regression analysis. Simple linear regression can be used to determine if there is a
significant relationship between predicted and observed values (Thomann, 1982). The coefficient
of determination (r2) is used to describe the percent of variance accounted for by the regression.
Significant r2 values indicate good correlation but not necessarily accurate predictions.
Additional tests of the regression can be made that yield more information about the relationship
.between observed and predicted values. The students 't' test can be used to test the hypotheses
that the intercept is zero and the regression slope is one. Significant t-values for both tests
would indicate that the model simulations were accurate. Reckhow et. al. (1990) warn that
outlying values and data with little range can adversely influence the meaning derived from
hypothesis testing.
15
-------
3. Mean comparisons. The differences between the predicted and observed means can be
evaluated using the 't'-test (Reckhow et. al, 1990; Thomann, 1982). One advantage of the t-test
is that there is a wide variety of hypothesis testing that can be performed. For example, one
could test whether the difference between predicted and observed values is greater than some
acceptable error or threshold value. If the populations are not normally distributed, the
Wilcoxon test can be used (Reckhow et. al., 1990).
4. Relative error. The relative error (e) is the percent absolute value difference between
observed and predicted mean values (Thomann, 1982):
_ , observed - predicted^ 10Q (8)
observed
The maximum relative error can be 100% in cases where all values are positive. This statistic
may be misleading for very low values of observed data, and when the observed data are much
larger than the predicted values. James and Burges (1982) suggest a mean relative error of 5%
. with a standard deviation of 5-10% as criteria for model adequacy. However, the relative error
chosen should be a function of model use.
5. Root mean square error. The square root of the sum of the squares of the deviation
between observed and predicted values divided by the number of observations (n) is the root
mean square error (er) (Thomann, 1982). This term has not often been used in model testing.
However, it provides a measure of model error, and has been recommended (Thomann, 1982).
16
-------
T^observed - predicted?
n
6. Differences in distribution. The difference between the observed and predicted
cumulative frequency distributions can be assessed using the two-sample Kolmogorov-Smirnov
(KS) test (Reckhow et al., 1990). This test determines the maximum difference between all
quantiles of the two distributions. The calculated KS value is compared to values found in a table
to determine the significance of the difference.
17
-------
METHODS
Observed Data
For both watersheds, AGNPS predictions were compared to observed values of stormflow
depth (in), peak flow during the storm (cfs), and the concentrations (ppm) and mass exports
(Ibs/acre) of sediment, nitrogen (N) and phosphorus (P). Discharge was obtained from field-
determined stage-discharge relationships. Stage was continuously recorded using an ISCO*
bubble-type flow meter (Meals, 1990; Vermont RCWP Coordinating Committee, 1991). Water
samples in both watersheds were collected by refrigerated ISCO automatic samplers at 8-hr
intervals. Samples were analyzed for total suspended solids, total P, and total Kjeldanl N using
standard techniques (EPA, 1983). Analysis was conducted according to a QA/QC plan that
included standards analysis, duplicates, chemical recovery, and performance testing. Duplicate
results ranged from 5 to 12 %; chemical recovery ranged from 98 to 101%. Precipitation was
recorded at the Dunsmore station 0.6 miles from the Jewett Brook watershed and at the Hannah
gage within 0.6 miles of Subwatershed 3.
The duration of flow was determined by hydrograph separation (Wisler and Brater,
. 1967). The beginning of stormflow was defined as the rise in stream discharge. The end of
stormflow was determined as occurring at the inflection point on the falling limb of the
hydrograph. A hydrograph for the September 21, 1983 storm in the LaPlatte Subwatershed 3
is shown in Figure 5. The hydrographs for all storms are given in Appendix A. The accuracy
of discharge data was verified by comparison with runoff coefficients from USGS gaging stations
located in the Champlain valley. Runoff coefficients have agreed within the 20 percent
recommended by Winter (1981) for comparison with regional gages. Generally the error
18
-------
Hydrograph of LaPlatte Event
Date: 9/21/83
10
Time, hr
Figure 5. Hydrograph for LaPlatte Watershed 3 for September 21, 1983 storm.
19
-------
14 i«ion i 4 i i n>ir» 4 «t urn 14 i « HIT » 4 »t ionn 4 « » ior » < »t imn» < « »toy»«' 11 mm»tw»'»w 14«i urn « « «
111111 Breakpoint Minutes
Maximum 30 min. = 0.59 in.
11 i i wn i * tor r 4 10 xn 2 4 i o tor 2 4 ion i 4 « ior i 4 i tpia > 4 i i ior 14 < i to ID 2 4 i ior i < »iu >
-i ...ii.W«DMMOA* i^ ml mi" ^TMUHSOAV " - FHIOA.V * -^ATJmQAY i *li- UIXOAV i. .A. . MOMOAV
Figure 6. Recording precipitation chart for storm on September 21, 1983, Hannah gage.
20
-------
associated with discharge measurements are IS percent (Winter, 1981).
The energy intensity (El) of the storm (decimal units) was calculated from recording rain
gage records using the method described by Wischmeier and Smith (1978). Using this procedure
the rainfall intensity (in/hr) was calculated for each break point of the storm trace on the chart.
An example precipitation chart is shown in Figure 6; the El calculation is given in Table 2. The
corresponding kinetic energy per inch of rain was determined for each intensity from a table
provided by Wischmeier and Smith (1978). The total energy for the storm is the sum of the
energy for each breakpoint increment, adjusted for the proportion of total rainfall.
The antecedent moisture condition (AMC) was determined for each selected storm using
the five-day cumulative precipitation index (USDA-SCS, 1972). A 5-day growing season total
precipitation less than 1.4 inches was assigned to AMC group I (lowest runoff potential), and
precipitation of 1.4 to 2.1 inches was assigned to AMC group n (average condition).
Storm Selection
Storms between late March to late November of each year with three or more consecutive
8-hr composite samples were considered. Winter storms were ignored since the version of
'AGNPS used could not predict snowmelt runoff. The 8-hr composites indicated that the storm
was sampled intensively. Groups without precipitation events were dropped. Groups coinciding
with snowfall and with temperatures below freezing before, during, and following the event were
dropped since AGNPS does not simulate periods of no-flow in the winter. Storms with missing
precipitation data were dropped. Hydrographs for the remaining storms were plotted (Appendix
A). Hydrographs with complex, multi-peak were dropped because they indicated more than one
storm was influencing runoff. Such complex storms could not be easily modeled by AGNPS.
21
-------
Table 2. Rainfall energy intensity (El) calculation for September 21,1983 storm, LaPlatte River
Sub watershed 3.
Chart
Time
16:30
22:00
22:30
02:00
02:30
03:00
'From
+ From
Depth Duration
(in) (min)
0
3.35
3.35
3.65
3.65
3.67
Wischmeier
chart
El =
EI =
330
30
210
30
30
and Smith (1978)
Depth
(in)
3.35
0
0.30
0
0.02
Total energy * maximum hourlv
100
3011 * 1.18 = 35.53
Intensity
(in/hr)
0.61
0
0.09
0
0.04
intensity*
Energy*
(in-1) Total
845 2831
0 0
570 171
0 0
453 9
3011
100
22
-------
Watershed Data
Land use and soils information were obtained from data stored in a geographic
information system (GIS). Using the GIS, the cell area was set by creating a 10-acre grid cell
overlay on the watershed boundaries (Figures 7 and 8). This cell size approximates the average
field size found in both watersheds. There were 40 cells in the LaPlatte River Subwatershed 3
and 343 cells in the Jewett Brook watershed. A program within the GIS was used to determine
land use, soil types, and fertilization levels on a cell-by-cell basis.
Since each cell could contain several soil types and/or land uses, cell averages of
attributes were lumped by area weighting. Values for the cropping factor (C), practice factor
(P), manning's roughness coefficient (n), and the surface condition constant, which are functions
of land use/cover, were determined from tables in Young et al. (1987) based on the predominant
(percent of area) land use occurring in the cell. Land uses were determined using the GIS land
use data. The soil credibility factor (K), and soil texture, which are functions of soil type, were
determined from county soil surveys for each study area based on an areal, weighted average
value for the cell, as suggested by Young et al. (1987). Soil types for each cell were determined
from the GIS data files (Table 1). The SCS curve number (CN), which varies with hydrologic
soil group, antecedent moisture condition, and land use, was determined from tables in Young
et al. (1987) as an areal, weighted- average based on the different soil types and land uses in a
cell.
The number of the cell into which another cell drained was determined from the USGS
topographic map for the study areas by noting the direction of flow leaving the cell (e.g. Figure
7). This also allowed determining the aspect of the direction of the drainage from the cell.
23
-------
SUBWATERSHED 3
LAPLATTE RIVER WATERSHED, VERMONT
Figure 7. LaPlatte subwatershed 3 with 10-acre grid cells and cell drainage paths.
24
-------
JEWETT BROOK SUBWATERSHED
ST. ALBANS BAY WATERSHED, VERMONT
Figure 8. Jewett Brook watershed with 10-acre grid cells and cell drainage paths.
25
-------
Average land slope and average channel slope also were determined from the USGS topographic
maps for each cell as rise over run. The channel side slope was based on recommendations by
Young et al. (1987). The slope shape factor of either uniform, convex, or concave .was
determined for each cell from the USGS topographic maps. Field slope lengths were determined
from a table of soil type and average slope length (Vt SCS, 1989). The fertilizer availability
(incorporation factor) was based upon CIS data and recommendations of Young et al. (1987).
The presence of a channel within a cell (channel indicator) was assessed using GIS data.
Input files were modified for each year since land use and therefore the curve number
(CN), C factor, surface condition constant, and roughness coefficient could change each year.
Text editors were used to build the input data files rather than the AGNPS preprocessor because
it was quicker. AGNPS includes a separate computer program (DBDFL) for forming input data
files (Young et al., 1987). Examples of input files for both watersheds are given in Appendix
B. These files contain cell-by-cell information on soils, curve numbers, and practices.
Model Adaptation
Although AGNPS does not require calibration, the model was developed assuming
Minnesota conditions. Several factors were compared to conditions found in Vermont (Table
3). Although the rainfall concentrations were expected to be different, the impact on
precipitation loadings, relative to other loadings, was not considered to be significant. Since
these values were not substantially different, the default values were used. One change made
to the AGNPS code was to print smaller values of the runoff volume and peak runoff rate. The
original code rounded runoff volume to 0.01 in. and peak runoff rate to 0.01 cfs. These values
26
-------
were changed to 0.0001 for runoff volume and 0.001 for peak runoff. This change was made
because runoff values simulated by AGNPS were often <0.00 and we wanted to determine if
lower values were being calculated but truncated when printed out.
Young et al. (1987) have performed a sensitivity analysis to determine the relative
changes in model output associated with changes in model input and model parameters. The
most sensitive parameters affecting sediment and nutrient exports were land slope, soil
credibility, cropping factor, practice factor, and curve number. All of these factors vary with
local site conditions, and reflect the sensitivity of the curve number technique and the USLE.
Computer Resources
According to Robert et al. (undated), the following equipment is required to run AGNPS:
- IBM-PC or compatible
- Monochrome or color graphics adapter and monitor
- 512K memory
- Floppy disk system or hard disk system
- Dot Matrix printer
-DOS 2.1 or higher
Most AGNPS runs were made on an IBM* PS/2 Model 30, 286 personal computer.
Computation time for the 343-cell Jewett Brook watershed took approximately 30 sec. excluding
the time required to make input specifications. Additional single cell runs were made on a
Zenith* 386-SX personal computer. Average computational time for these runs was a few
seconds. The GIS used was ARC-INFO* which was maintained on a VAX* 11/750 computer.
27
-------
Table 3. AGNPS default values and appropriate Vermont values for model adaption.
Parameter [AGNPS Name] (Units)
AGNPS
Value
Vermont
Value
Source of
Information
Soil Concentrations
Soluble N [CN] (ppm)
Soluble P [CP] (ppm)
Sediment N [SOILN] (lb/lb)
Sediment P [SOILP] (lb/lb)
N
Application [NPPA]
Low (lb/ac)
Medium ab/ac)
High (lb/ac)
P Fertilizer Application [PPPA]
Low (lb/ac)
Medium (lb/ac)
High (lb/ac)
Rain Concentration
N [RCN] (mg/1)
5
2
0.001
0.0005
50
100
200
20
40
80
0.8
2-30
1-1.5
0.001-0.002
0.0005-0.001
50
100
150
20
40
60-80
Jokela, 1989
1.43
Likens et al.,
1977
28
-------
RESULTS AND DISCUSSION
LaPlatte Subwatershed 3
The precipitation characteristics of the IS storms that were modeled in Subwatershed 3
are summarized in Table 4. Rainfall amounts ranged from 0.2 to 3.67 inches. These storms
represented the full range of observed precipitation events that produced runoff in the LaPlatte
watershed, and therefore are appropriate storms to model for a verification of AGNPS. Based
on Weather Bureau intensity-duration-frequency maps (Hershfield, 1961), one storm had a 50-yr
return period, and one was an 8-yr storm. The remaining storms modeled had a return period
of less than one year. The precipitation amounts for all modeled storms were in the upper 50
percent observed in the watershed based on data collected at the Hannah gage in the LaPlatte
River watershed.
Prior to full testing of the model, default value assumptions and average state values
were checked against monitored results. Calculated El values using the Wischmeier and Smith
(1978) method were substantially lower than the statewide average value of 90 recommended for
use for the two study area counties in Vermont (USDA-SCS, 1987). Also, it is recommended
jn the model documentation that the antecedent moisture condition (AMC) value to use is II,
representing average conditions (Young et al., 1987). However, using the 5-day AMC index
(USDA-SCS, 1972), 13 out of the 15 storms had AMC values of I which indicates a dryer than
average condition. Using calculated, rather than average values for El and AMC, resulted in
lower predictions of discharges, concentrations, and mass export predictions (Table 5). Due to
these differences, values of El and AMC were calculated for each model run to meet the study
objectives.
29
-------
Table 4. Precipitation characteristics of modeled storms, LaPlatte River Subwatershed 3.
Date
MM/DD/YY
3/31/87
8/27/86
4/ 6/85
5/23/84
6/5/85
4/17/84
4/10/83
4/17/82
8/11/83
11/11/83
8/ 8/83
9/29/86
61 6/83
8/3/86
9/21/83
Precipitation
(in.) El
0.20
0.33
0.45
0.48
0.74
0.78
0.90
0.92
0.99
1.00
1.01
1.03
1.09
2.45
3.67
0.15
0.80
0.73
1.76
1.36
1.92
1.54
4.37
0.61
1.35
11.71
5.82
9.40
45.33
35.53
Percent of Storms Return
Less than or equal Period
AMC To observed (yr)
I
I
I
I
I
n
i
i
i
n
i
i
i
i
i
50 <1
75 <1
84 <1
84 <1
90 <1
90 <1
90 <1
90 <1
95 <1
95 <1
95 <1
95 <1
95 <1
99 8
99 50
30
-------
Table S. Comparison of predicted discharge, P, and N using default and computed AMC and
El values for two storms for LaPlatte subwatershed 3.
Date of Storm
Variable
Input
AMC
El
Default
II
90
8/ 03/86
Computed Observed
I
45.3
Default
n
90
8/27/86
Computed
I
0.8
Observed
Output
Runoff Volume (in) 0.68 0.13 0.5 0.0001 0.0 0.23
Runoff Peak (cfs) 196 43 15.83 0.064 0 7.88
P Concentration (mg/1) 0.8 0.1 0.1 0.1 0.1 0.2
P Mass Export Ob/ac) 8.90 3.12 0.01 1.19 0.03 0.01
N Concentration (mg/1) 4.3 1.6 0.74 4.5 4.5 0.88
N Mass Export (lb/ac) 17.57 6.30 0.08 2.39 0.06 0.05
31
-------
Observed and predicted mean values (IS storms), the relative error, and results from the
t-test of means for all modeled variables are summarized in Table 6. The hypothesis for the t-
test was that the mean observed and predicted values were not different. A low probability of
being greater than the Y value (Prob > t in Table 6) indicates that the means are significantly
different at that probability.
Tests of the goodness of fit of linear regressions between observed and predicted values
are summarized in Table 7. These include the Root MSE, the F-statistic for the significance of
the regression, and the coefficient of determination (R2) between observed and predicted values.
The significance of the F-statistic is given by the Prob > F. A probability of 0.05 would be
significant at the 95 percent level of confidence. A significant R2 value at a probability of 0.05
would be 0.72.
Discharge. Discharge volume (in.) was underpredicted by AGNPS for all storms less
than the 50 year event (Table 8). Mean discharge volumes were significantly different based
on the t-test, as indicated by a p value less than 0.01 (Table 6). The relative error in the
predicted mean was 87 percent. There was no significant relationship between observed and
.predicted volumes based on regression (Table 7, Figure 9). For the largest storm monitored,
AGNPS predicted discharge depth was of the same order of magnitude as that observed.
Peak discharge also was underpredicted by AGNPS except for storms greater than two
inches (Table 8). However, mean observed and predicted values were not different based on
the t-test (Table 6), and the relative error was small (9%). There were no significant
relationships between predicted and observed discharge based on regression analysis (Table 7,
Figure 10). Given the large discrepancy between predicted and observed discharges, the t-test
32
-------
Table 6. Mean observed values (15 storms) and values predicted by AGNPS with relative errors
and t-test between means for LaPlatte Subwatershed 3.
Means Relative
Variable Observed Predicted Error (%) t-value Prob >t
Discharge
Volume (in)
Peak (cfs)
Sediment
Concentration (mg/1)
Mass Export (lb/ac)
Phosphorus
Concentration (mg/1)
Mass Export (lb/ac)
Nitrogen
Concentration (mg/1)
Mass Export (lb/ac)
0.31
15.59
30.87
0.53
0.15
0.01
0.82
0.06
0.04
14.2
1,423,508
124.6
0.11
0.54
3.15
1.09
87
9
4,611,199
23,409
27
5,300
284
1,717
4.86"
0.13
-2.29*
-1.11
1.31
-1.64
-2.32*
. -1.65
0.003
0.896
0.043 .
0.290
0.212
0.125
0.040
0.128
** p < 0.01
* p < 0.05
33
-------
Table 7. Root MSE and significance of regressions between observed and AGNPS predicted
values for the LaPlatte Subwatershed 3.
Variable
Discharge
Volume (in)
Peak (cfs)
Sediment
Concentration (mg/1)
Mass Export (Ib/ac)
Phosphorus
Concentration (mg/1)
Mass Export (Ib/ac)
Nitrogen
Concentration (mg/1)
Mass Export (Ib/ac)
Root
MSE
0.196
13.670
31.087
0.540
0.066
0.008
0.273
0.050
F
0.335
0.175
0.360
2.472
0.904
0.311
0.407
1.567
Prob > F
0.57
0.68
0.56
0.15
0.36
0.59
0.54
0.24
R2
0.03
0.01
0.03
0.20
0.07
0.03
0.04
0.14
34
-------
Table 8. Discharge observed from LaPlatte River Subwatershed 3 and predicted by AGNPS.
Discharge
Date
MM/DD/YY
3/31/87
8/27/86
4/ 6/85
5/23/84
6/ 5/85
4/17/84
4/10/83
4/17/82
8/11/83
11/11/83
8/ 8/83
9/29/86
6/ 6/83
8/ 3/86
9/21/83
Precipitation
(in.)
0.20
0.33
0.45
0.48
0.74
0.78
0.90
0.92
0.99
1.00
1.01
1.03
1.09
2.45
3.67
Depth
Observed
0.11
0.23
0.28
0.15
0.05
0.41
0.43
0.54
0.06
0.47
0.01
0.45
0.55
0.50
0.38
(in.)
Predicted
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.04
0.00
0.00
0.00
0.13
0.51
Observed
4.38
7.88
11.57
3.94
1.27
27.24
17.20
25.87
1.61
36.68
0.29
18.46
43.53
15.83
18.93
Peak (cfs) '
Predicted
0.00
0.00
0.00
0.00
0.00
4.00
0.00
0.00
0.00
14.00
0.00
0.00
0.00
43.00
152.00
35
-------
LaPlatte Sufowatershed 3
Discharge volume
(In.)
O.1 O.2 O.3 O.4 O.6 O.6
Predicted (In.)
Figure 9. Plot of discharge volume observed in LaPlatte River Subwatershed 3 and predicted
byAGNPS.
LaPlatte Subwatershed 3
Peak Discharge
(cf s)
O 2O 4O OO 8O 1OO 12O 14O 16O
Pr*dlot«d (of*)
Figure 10. Plot of peak discharge observed in LaPlatte River Subwatershed 3 and predicted bv
AGNPS. y
36
-------
of means and the relative error of these means appear inappropriate measures for testing model
performance in this study.
Sediment. Suspended solids concentrations in runoff from Subwatershed 3 were
overpredicted by from two to six orders of magnitude (Table 9). AGNPS predicted values that
are unrealistic for streamflow. Mean concentrations were significantly different (Table 6).
There was no significant relationship between observed and predicted values (Table 7, Figure
11).
The mass export of sediment also was overpredicted by one or two orders of magnitude
(Table 9). Mean export values were not different based on the t-test (Table 6). Inspection of
Figure 12 reveals that the points are clustered together. There was no significant relationship
between observed and predicted exports (Table 7, Figure 12).
Phosphorus. Predicted phosphorus concentrations were of the same order of magnitude
as those observed (Table 10). However, the default concentration of 0.10 mg/1 predominated.
There was no significant difference between observed and predicted mean concentrations, and
their relative error was only 27 percent (Table 6). Also, there was no significant relationship
. between observed and predicted P concentration values (Table 7, Figure 13). However, it
appears that a low default value of 0.10 mg/1 was assumed by AGNPS for most cases.
The predicted mass export of phosphorus was of the same order of magnitude as
observed for storms less than one inch; larger storms were overpredicted (Table 10). Mean
observed and predicted exports were not different, but their relative error was 5,300 (Table 6).
This finding again questions the usefulness of the t-test in comparing model predictions to
observed values. There was no significant relationship between observed and predicted values
37
-------
Table 9. Sediment observed in runoff from LaPlatte River Subwatershed 3 and predicted by
AGNPS.
Total Suspended Solids
Date
MM/DD/YY
3/31/87
8/27/86
4/ 6/85
5/23/84
6/5/85
4/17/84
4/10/83
4/17/82
8/11/83
11/11/83
8/ 8/83
9/29/86
6/ 6/83
8/ 3/86
9/21/83
Precipitation
(in.)
0.20
0.33
0.45
0.48
0.74
0.78
0.90
0.92
0.99
1.00
1.01
1.03
1.09
2.45
3.67
Concentration (mg/1)
Observed Predicted
9.2
4.1
89.2
24.7
31.3
19.9
69.0
2.5
17.2
6.6
19.9
76.8
32,779.
473,498.
390,446.
988,888.
771,295.
31,899.
825,667.
2,403,695.
336,046.
9,579.
6,375,266.
3,410,522.
5,123,946.
136,222.
42,877.
Export (tons)
Observed Predicted
0.05
0.04
1.15
0.16
0.59
0.39
1.71
0.007
0.38
0.003
0.51
1.34
0.1
2.1
1.7
4.4
3.4
14.1
~ 3.7
10.7
1.5
15.5
28.3
14.9
22.7
770.3
975.7
indicates missing data
38
-------
LaPlatte Subwatershed 3
Sediment Concentration
(mg/l * 10)
10 100 1400 10400 100400 1400400
Predicted (mg/l » 1 O)
Figure 11. Plot of sediment concentration observed in runoff from LaPlatte River Subwatershed
3 and predicted by AGNPS.
LaPlatte Subwatershed 3
Sediment Export
(Tons)
1OOO
100
10
0.1
O.O1
O.OO1
O.OO1 O.O1 O.1 1 1O 1OO 1OOO
Pr*dlot*d (Ton*)
Figure 12. Plot of sediment export observed in runoff from Laplatte River Subwatershed 3 and
predicted by AGNPS.
39
-------
Table 10. Total phosphorus observed in runoff from LaPlatte River Subwatershed 3 and
predicted by AGNPS.
Total Phosphorus
Date
MM/DD/YY
3/31/87
8/27/86
4/ 6/85
5/23/84
6/ 5/85
4/17/84
4/10/83
4/17/82
8/11/83
11/11/83
8/ 8/83
9/29/86
6/ 6/83
8/ 3/86
9/21/83
Precipitation
(in.)
0.20
0.33
0.45
0.48
0.74
0.78
0.90
0.92
0.99
1.00
1.01
1.03
1.09
2.45
3.67
Concentration (mg/1)
Observed Predicted
0.08
0.20
0.18
0.17
0.09
0.19
0.14
0.13
0.05
0.14
0.29
0.08
0.10
0.22
0.00
0.10
0.10
0.10
0.10
0.40
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.70
0.10
Export
Observed
0.00
0.01
0.01
0.01
0.00
0.02
0.01
0.02
0.00
0.02
^__
0.03
0.01
0.01
0.02
(Ib/ac)
Predicted
0.00
0.03
0.02
0.01
0.04
0.13
0.04
0.10
0.02
0.14
0.22
0.13
0.19
3.12
3.78
indicates missing data
40
-------
LaPlatte Subwatershed 3
Phosphorus Concentration
(mg/l)
0.2
0.4 0.6
o.a
Predicted (mg/l)
Figure 13. Plot of phosphorus concentrations observed in runoff from LaPlatte River
Subwatershed 3 and predicted by AGNPS.
LaPlatte Subwatershed 3
Phosphorus Export
(Ib/ac)
10
i
O.1
0.01
0.001
0.001 0.01
0.1
10
Pr«dlot*d (Ib/aol
Figure 14. Plot of phosphorus export observed in runoff from LaPlatte River Subwatershed 3
and predicted by AGNPS.
41
-------
based on analysis of variance of regression (Table 7, Figure 14).
Nitrogen. Predicted nitrogen concentrations were generally two to three times observed
values (Table 11). Mean observed and predicted nitrogen concentrations were significantly
different based on a t-test (Table 6). There was no significant relationship between observed and
predicted nitrogen concentrations (Table 7, Figure 15). If the higher rainfall N concentrations
expected in Vermont had been modified in AGNPS, the difference between observed and
predicted values would have been greater than given in Table 11.
The mass export of nitrogen generally was overpredicted. The amount of overprediction
varied by three to eighty times (Table 11). There was no significant difference between mean
export values for nitrogen (Table 6), but the relative error was 1,717. There was no significant
relationship between observed and predicted values (Table 7, Figure 16).
42
-------
Table 11. Nitrogen observed in runoff from LaPlatte River Subwatershed 3 and predicted by
AGNPS.
Nitrogen
Date
MM/DD/YY
3/31/87
8/27/86
41 6/85
5/23/84
6/ 5/85
4/17/84
4/10/83
4/17/82
8/11/83
11/11/83
8/ 8/83
9/29/86
61 6/83
8/ 3/86
9/21/83
Precipitation
(in.)
0.20
0.33
0.45
0.48
0.74
0.78
0.90
0.92
0.99
1.00
1.01
1.03
1.09
2.45
3.67
Concentration (mg/1)
Observed Predicted
0.68
0.88
0.65
0.70
0.63
1.43
0.81
0.64
0.60
0.77
0.74
1.28
15.10
4.50
2.80
2.80
2.30
3.50
1.90
2.10
1.80
1.80
1.80
2.00
1.80
1.60
1.50
Export (Ib/ac)
Observed Predicted
0.02
0.05
0.02
0.01
0.06
~
0.17
0.01
<0.01
0.06
0.10
0.08
0.11
0.01
0.06
0.05
0.10
0.08
0.26
0.09
0.20
0.04
0.29
0.44
0.27
0.37
6.30
7.73
indicates missing data
43
-------
LaPlatte Subwatershed 3
Nitrogen Concentration
(mg/l)
s
14
12
1O
8
e
4
2
O 2 4 a 8 1O 12 14
Pr*dlot«d (mg/l)
Figure 15. Plot of nitrogen concentrations observed in runoff from LaPlatte River subwatershed
3 and predicted by AGNPS.
LaPlatte Subwatershed 3
Nitrogen Export
(Ib/ac)
10
0.1
O.O1
0.001
O.OO1
O.O1
O.I
10
Predicted (Ib/ac)
Figure 16. Plot of nitrogen export observed in runoff from LaPlatte River subwatershed 3 and
predicted by AGNPS.
44
-------
Jewett Brook
Eleven storms were modeled in the Jewett Brook watershed (Table 12). These storms
ranged in amount from 0.43 to 2.47 inches and occurred during dryer than average conditions
(AMC = I). Based on the Weather Bureau precipitation intensity-duration-frequency maps, one
storm had an 8-yr return period, and two were 2-year storms. The remaining storms modeled
had return periods of less than one year. The storms modeled represented the largest storms
monitored during the study; 86 percent of the storms that occurred were smaller than those used
for the testing of AGNPS. Comparisons could not be made with nitrogen concentrations or
exports due to an insufficient number of samples.
Discharge. As was observed for the LaPlatte Subwatershed 3, discharge volume (in.)
was underpredicted for all but the largest storms (Table 15). The mean discharge volumes were
significantly different based on the t-test (Table 13), and the relative error in the means was
96%. There was no significant regression relationship between observed and predicted values,
and the regression explained only 18% of the variation in values (Table 14, Figure 17).
Peak discharge also was underpredicted except for the three storms greater than 1.3
inches in amount (Table 15). The mean peak discharge predicted by AGNPS was significantly
lower than observed (Table 13). There was no significant regression between observed and
predicted values (Table 14). For six of the 11 storms modeled, a peak discharge of 0.136 cfs
was predicted (Table 15, Figure 18).
Sediment. The concentration of suspended solids was overpredicted by one to two orders
of magnitude (Table 16). The mean concentration predicted by AGNPS was significantly greater
45
-------
Table 12. Precipitation characteristics of modeled storms, Jewett Brook.
Date Precipitation
MM/DD/YY (in.)
4/ 6/85
4/16/85
3/10/83
10/ 5/83
4/17/84
5/29/84
4/18/85
11/26/86
10/28/87
5/23/84
9/21/83
0.43
0.49
0.55
0.69
0.77
0.94
0.94
1.03
1.32
1.72
2.47
El
1.61
0.96
1.18
2.56
1.73
0.95
4.21
1.81
1.86
17.17
9.41
Percent of Storms Return
Less than or equal Period
AMC To observed (yr)
I
I
I
I
I
I
I
I
I
I
I
86
88
92
95
96
99
99
99
99
99
99
<1
<1
<1
<1
<1
<1
<1
<1
2
2
8
46
-------
Table 13. Mean observed values (11 storms) and values predicted by AGNPS with relative
errors and t-test between mean for Jewett Brook.+
Means Relative
Variable Observed Predicted Error (%) t-value Prob >t
Discharge
Volume (in)
Peak (cfs)
Sediment
Concentration (mg/1)
Mass Export (Ib/ac)
Phosphorus
Concentration (mg/1)
Mass Export (Ib/ac)
+ insufficient data for
** p < 0.01
* p < 0.05
0.47
114.66
110.1
19.31
1.65
0.07
0.019
17.299
7,159
3.8
0.5
0.01
96
85
-6,402
80
70
86
4.42"
2.61*
-4.06*
2.42*
1.92
4.39"
0.001
0.026
0.004
0.042
0.087
0.002
nitrogen comparisons.
47
-------
Table 14. Root MSE and significance of regressions between observed and AGNPS predicted
values for Jewett Brook.
Root
Variable MSE F Prob > F R2
Discharge
Volume (in) 0.229 2.005 0.19 0.18
Peak(cfs) 89.645 2.357 0.16 0.21
Sediment
Concentration (mg/1) 56.515 3.716 0.09 0.35
Mass Export (lb/ac) ig.227 0.007 0.94 <0.01
Phosphorus
Concentration (mg/1) 1.829 0.016 0.90 <0.01
Mass Export flb/ac) 0.047 1.477 0.26 0.16
48
-------
Table IS. Discharge observed from Jewett Brook and predicted by AGNPS.
Discharge
Date
MM/DD/YY
4/ 6/85
4/16/84
3/10/83
10/ 5/83
4/17/84
5/29/84
4/18/85
11/26/86
10/28/87
5/23/84
9/21/83
Precipitation
(in.)
0.43
0.49
0.55
0.69
0.77
0.94
0.94
1.03
1.32
1.74
2.47
Volume
Observed
0.67
0.69
0.61
<0.00
0.71
0.39
0.56
0.77
<0.00
0.75
0.03
(in.)
Predicted
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0002
0.0044
0.0313
0.1722
Peak
Observed
229.59
180.01
127.56
0.72
237.45
86.04
197.16
177.52
1.38
17.17
6.71
(cfs)
Predicted
0.136
0.136
0.136
0.136
0.136
0.139
0.143
0.290
4.809
30.736
153.490
49
-------
Jewett Brook
Discharge Volume
(in.)
O.2 O.4 0.6 O.8
Predicted (In.)
Figure 17. Plot of discharge volume observed in Jewett Brook and predicted by AGNPS.
Jewett Brook
Peak Discharge
(cfs)
260
200A-
ioo
60
6O 1OO 16O 20O 26O
Predicted (cfs)
Figure 18. Plot of peak discharge observed in Jewett Brook and predicted by AGNPS.
50
-------
than the mean concentration of suspended solids observed (Table 13). There was no significant
relationship (p = 0.05) between observed and predicted sediment values (Table 14, Figure 19).
The mass export of sediment was generally underpredicted by AGNPS, except for the
largest storm (Table 16). This result is opposite of the findings for LaPlatte Sub watershed 3
where the export of sediment was overpredicted by AGNPS (Table 9). It is likely that modeling
the larger Jewett Brook watershed results in greater settling of sediment than for the smaller
Subwatershed 3. The mean predicted export was significantly lower than the mean observed
export of sediment (Table 13). There was no significant relationship between observed and
predicted sediment exports (Table 14, Figure 20).
Phosphorus. As was observed for simulations in LaPlatte Subwatershed 3, predicted
phosphorus concentrations were of the same order of magnitude as those observed (Table 17).
There was no significant difference between mean predicted and observed phosphorus
concentration values based on the t-test (Table 13). However, the differences were substantial
(Figure 19). There was not a significant relationship between observed and predicted
phosphorus concentration values (Table 14, Figure 21).
The mass export of phosphorus was underpredicted by AGNPS except for the largest
storm modeled (Table 17). This finding is different than observed in Subwatershed 3 where
mass export was generally overpredicted (Table 10). Again, the respective sizes of the two
watersheds may explain these differences. The mean predicted export of phosphorus was
significantly lower than that observed (Table 13). There was no significant relationship between
observed and predicted values of the mass export of phosphorus based on regression (Table 14,
Figure 22).
51
-------
Table 16. Sediment observed in runoff from Jewett Brook and predicted by AGNPS.
Sediment
Date Precipitation
MM/DD/YY (in.)
4/ 6/85
4/16/84
3/10/83
10/ 5/83
4/17/84
5/29/84
4/18/85
11/26/86
10/28/87
5/23/84
9/21/83
0.43
0.49
0.55
0.69
0.77
0.94
0.94
1.03
1.32
1.74
2.47
Concentration (mg/1)
Observed Predicted
92.7
70.9
74.9
98.9
116.2
34.7
247.5
181.4
73.5
--
9,302.
5,302.
5,716.
8,053.
6,820.
5,955.
19,483.
6,448.
1,355.
2,449.
313.
Export (tons)
Observed Predicted
24.22
19.03
17.63
0.01
31.96
5.28
54.03
0.29
21.37
---
0.4
0.2
0.2
0.3
0.3
0.2
0.8
0.5
2.3
29.2
20.5
indicates missing data
52
-------
Jewett Brook
Sediment Concentration
(mg/l)
100OO
1OOO
100
10
1O 100 1000 10000
Predicted (mg/l)
Figure 19. Plot of sediment concentrations observed in Jewett Brook and predicted by AGNPS.
Jewett Brook
Sediment Export
(Ib/ac)
10 20 so 40 BO eo
Predicted (tons)
Figure 20. Plot of sediment export by Jewett Brook and predicted by AGNPS.
53
-------
Table 17. Total phosphorus observed in Jewett Brook and predicted by AGNPS.
Total Phosphorus
Date
MM/DD/YY
41 6/85
4/16/84
3/10/83
10/ 5/83
4/17/84
5/29/84
4/18/85
11/26/86
10/28/87
9/21/83
Precipitation
(in.)
0.43
0.49
0.55
0.69
0.77
0.94
0.94
1.32
1.74
2.47
Concentration (mg/1)
Observed Predicted
0.60
0.45
0.52
2.95
0.63
0.74
0.65
5.51
1.03
3.43
1.4
0.2
0.8
1.5
0.1
0.2
0.7
0.3
0.1
0.1
Export
Observed
0.09
0.07
0.07
<0.01
0.10
0.07
0.08
0.01
0.17
0.02
(Ib/ac)
Predicted
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.04
0.03
54
-------
Jewett Brook
Phosphorus Concentration
(mg/l)
r
O1 23466
Pr*diot*d (mo/U
Figure 21. Plot of phosphorus concentrations observed in Jewett Brook and predicted by
AGNPS.
Jewett Brook
Phosphorus Export
(Ib/ac)
O.2
0.16 -
a OB ai O.16
Predicted (Ib/ac)
0.2
Figure 22. Plot of phosphorus export by Jewett Brook and predicted by AGNPS.
55
-------
Nitrogen. There was an insufficient number of nitrogen samples analyzed for the
modeled storms to perform an adequate evaluation of the ability of AGNPS to predict nitrogen
concentrations or mass export in Jewett Brook.
Overall, there was a poor relationship between observed and predicted values for this
verification in Vermont. The difference between observed and predicted values was greater than
previously reported Bosch et al., 1983; Young et al, 1986, 1989). However, this is the first
verification of AGNPS utilizing a full range of statistical testing as recommended by Thomann
(1982). The application of the model to a completely different climatic region than where the
model was developed may explain the differences obtained in the accuracy of the model. If
discharge is not accurately predicted, mass export predictions should be questioned since mass
is a function of discharge.
Single cell
For the 15 events observed in the LaPlatte Subwatershed 3, AGNPS simulations were
conducted a second time with lumped parameters. Thus, the data from the 40 cells were
combined into one cell. These additional simulations were conducted in order to determine if
the use of distributed parameters gave better results than lumped parameters. The results of
these simulations are presented in Tables 18-21 together with the results from the previous
simulations. In general, predictions with one-cell parameter values were worse than those using
the 40 cells. As compared to the original 40 cell simulations, the one-cell runs usually resulted
in higher values CTables 18 - 21).
56
-------
Table 18. Comparison of discharge observed from LaPlatte River Subwatershed 3 and predicted
by AGNPS using 40 cells versus 1 cell.
Depth (in.)
Observed
0.11
0.23
0.28
0.15
0.05
0.41
0.43
0.54
0:06
0.47
0.01
0.45
0.55
0.50
0.38
Predicted
40 -Cell
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.04
0.00
0.00
0.00
0.13
0.51
1-Cell
0.0001
0.0001
0.0001
0.0001
0.0001
0.0051
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.1529
0.4932
Observed
4.38
7.88
11.57
3.94
1.27
27.24
17.20
25.87
1.61
36.68
0.29
18.46
43.53
15.83
18.93
Predicted
40 - Cell
0.00
0.00
0.00
0.00
0.00
4.00
0.00
0.00
0.00
14.00
0.00
0.00
0.00
43.00
152.00
1 - Cell
0.098
0.098
0.098
0.098
0.098
3.522
0.098
0.098
0.098
0.098
0.098
0.098
0.098
77.424
224.774
Peak (cfs)
57
-------
Table 19. Comparison of sediment observed in runoff from LaPlatte River Subwatershed 3 and
predicted by AGNPS using 40 cells versus 1 cell.
Concentration (mg/1)
Observed
9.2
4.1
89.2
24.7
31.3
19.9
69.0
2.5
17.2
6.6
19.9
76.8
Predicted
40 - Cell
32,779.
473,498.
390,446.
988,888.
771,295.
31,899.
825,667.
2,403,695.
336,046.
9,579.
6,375,266.
3,410,522.
5,123,946.
136,222.
42,877.
1-Cell
12,751.
1,667,857.
1,459,488.
3,751,536.
2,918,064.
193,791.
3,126,432.
9,169,102.
1,125,121.
80,243,010.
24,379,960.
605,180.
19,587,500.
16,328.
92,167.
Export (tons)
Observed
0.05
0.04
1.15
0.16
0.59
0.39
1.71
0.007
0.38
0.003
0.51
1.34
Predicted
40 - Cell
0.1
2.1
1.7
4.4
3.4
14.1
3.7
10.7
1.5
15.5
28.3
14.9
22.7
770.3
975.7
1-Cell
0.1
7.6
6.6
17.0
13.2
45.0
14.2
41.5
5.7
363.5
110.4
2.7
88.7
113.1
2,059.4
58
-------
Table 20. Comparison of total phosphorus observed in runoff from LaPlatte River Subwatershed
3 and predicted by AGNPS using 40 cells versus 1 cell.
concentration (Mg/i)
Observed
0.08
0.20
0.18
0.17
0.09
0.19
0.14
0.13
0.05
0.14
___
0.29
0.08
0.10
0.22
Predicted
40 -Cell
0.00
0.10
0.10
0.10
0.10
0.40
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.70
0.10
1-Cell
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
0.1
tixpon (LDiac)
Observed
0.00
0.01
0.01
0.01
0.00
0.02
0.01
0.02
0.00
0.02
«.-
0.03
0.01
0.01
0.02
Predicted
40 - Cell
0.00
0.03
0.02
0.01
0.04
0.13
0.04
0.10
0.02
0.14
0.22
0.13
0.19
3.12
3.78
1 -Cell
0.00
0.08
0.07
0.15
0.12
0.32
0.13
0.30
0.06
1.69
0.65
0.03
0.55
0.66
6.76
indicates missing data
59
-------
Table 21. Comparison of nitrogen observed in runoff from LaPlatte River Sub watershed 3 and
predicted by AGNPS using 40 cells versus 1 cell.
concentration
-------
Event extrapolations
The second objective of the project was to assess methods of extrapolating event-based
simulations to annualized data. The lack of relationships between observed and predicted values
makes any test of a method of extrapolation impossible. However, the following was the method
that would have been used for the test.
1. Develop a frequency distribution of precipitation events from local data.
2. Perform simulations for the precipitation amounts that coincide with midpoints of
intervals on the cumulative frequency distribution, including the 5, 25, 50, 75, 90, 95,
and 99 percentiles.
3. Multiply the simulated mass export results for each frequency times the number of
events occurring for each interval and sum the results for a year.
4. Add a base flow component for periods with no storms. However, the version of
AGNPS being used in this verification does not predict base flow values.
This overall method assumes that there is a relationship between precipitation and mass export,
which may or not be true, depending on local conditions.
61
-------
CONCLUSIONS
AGNPS underpredicted discharge volume and peak flow except for the larger, rare
storms; that is, those storm with greater than an 8-year recurrence interval or 1-2 inches of
precipitation. Sediment concentrations were overpredicted by from one to six orders of
magnitude. Sediment export was overpredicted in one watershed and underpredicted in the other
»
watershed. Predicted phosphorus concentrations were of the same order of magnitude as
observed. Phosphorus exports were underpredicted in one watershed and of the same order of
magnitude in another watershed. Nitrogen concentrations and mass exports were overpredicted
in the one watershed where sufficient observed nitrogen data was available.
Lumping the parameters into one cell worsened the predictions by AGNPS in the one
watershed where the comparison was made with 40 cells for 15 storms.
Based on the results from testing in one watershed, it is recommended that El and AMC
be calculated rather than use "average" values as found in the AGNPS manual.
Some statistics that are often recommended for model verification were not useful in this
test of AGNPS, perhaps due to the poor relationship between observed and predicted values.
Both the students 't'-test of means and the root MSE were not meaningful and are not
recommended for use in applications similar to the one described herein.
62
-------
RECOMMENDED VERIFICATION METHODS
Based upon the experience obtained in this and other (Jamieson and Clausen, 1988)
verifications of water quality models the following steps are recommended as a method for
testing nonpoint source models:
1. Locations. Assuming that the model is intended to have application across the U.S., the
model should be tested in several locations other than where the model was developed. If
models are intended to have broad geographic application, it is especially important to test
models in different major climate zones such as those represented by areas where snowmelt or
no snowmelt would occur. In some cases, testing in each EPA Region might be appropriate,
although the method of testing should be centrally controlled to maintain uniformity in methods
of testing.
2. Method. Perform the test by comparing simulated to observed data. A sensitivity analysis,
which is determining the effect of varying a parameter value on the output, is not a verification
of a model but rather indicates important parameters. The test should display observed and
simulated data in a form where they are directly comparable, such as in a table or graph.
3. Fitting. The method of conducting the verification should be consistent with the type of
model. For example, a measured parameter model should not be fitted to local data before a
test since such a model is intended to be used without fitting. Often, a model will be fitted with
one year of data and tested with another year of data. This procedure is appropriate to calibrate
a model during development but should not replace a verification using independent data.
4. Statistics. Several statistical tests are recommended for verifications of NPS water quality
63
-------
models. Enough testing will be needed to achieve statistical significance. Generally, at least
IS pairs of observed and simulated data points are needed for a simple regression.
The following are recommended:
a. Linear regression of predicted and observed values. Analysis of variance of
regression should be used to test the significance of the regression equation.
b. The coefficient of determination (r2) should be determined to describe the percent of
variance explained by the regression.
c. Students Y to test the hypothesis that the intercept of the regression is zero.
d. Students t to test the hypothesis that the slope of the regression is one. A slope of
one is a perfect relationship between observed and predicted values.
5. Principal Investigator. The model testing should be conducted by individuals, other than the
model authors, who have the observed data in their possession. This independent test prevents
bias in the interpretation of the findings. However, it is equally important that a model author
or contact person assist in the verification effort. Model authors can notice inappropriate input
data quickly, and may understand unusual predictions.
, 6. Review. Verification results should receive peer review.
64
-------
REFERENCES
Bartholic, J.F., K.M. Kittleson, and B. Bill. 1987. Operational, spatial, and environmental water
planning and analysis. Michigan Institute of Water Research. E. Lansing, Technical
Report No. G1232-05.
Bosch, D.D., C.A. Onstad, and R.A. Young. 1983. A procedure for prioritizing water quality
problem areas. Paper 83-2156, ASAE Summer Meeting, June 26-29, Bozeman, Montana.
Cassell, E.A. and D.W. Meals. 1981. LaPlatte River watershed water quality monitoring and
analysis program: description of watershed and water quality monitoring program. Water
Resources Research Center, University of Vermont. Burlington. Report No. 1.
Cassell, E.A., J.C. Clausen, R.C. Furbush, M. Long, and D.W. Meals. 1983. St. Albans bay
watershed comprehensive water quality monitoring and evaluation: background
information. Vermont Water Resources Research Center, University of Vermont,
Burlington. Report No. 1.
Clausen, J.C. 1985. The St. Albans Bay watershed rcwp: a case study of monitoring and
assessment. In Perspectives on nonppint source pollution. USEPA 440/5-85-001.
Frere, M.H., J.D. Ross, and L.J. Lane. 1980. The nutrient submodel. In CREAMS, a field
scale model for chemicals, runoff, and erosion from agricultural management systems.
Cons. Res. Rpt. No. 26. Agric. Res. Serv. USDA. Washington, D.C. P. 65-85.
Frevert, K., and B.M. Crowder. 1987. Analysis of agricultural nbnpoint pollution control
options in the St. Albans Bay watershed. Natural Resource Economics Division,
Economic Research Service, USDA. Staff Report No. AGES870423. Washington, D.C.
German, D. 1991. Personal Communication. Research Associate. South Dakota Water Resources
Institute. Brookings, S.D.
Hershfield, D.M. 1961. Rainfall frequency atlas of the United States. U.S. Weather Bureau.
Technical Paper 40.
James, L.D. and S.J. Burges. 1982. Selection, calibration, and testing of hydrologic models. In
C.T. Haan, M.P. Johnson, D.L. Brakensick (eds.). Hydrologic modeling of small
watersheds. ASAE, St. Joseph, MI.
Jamieson, C.A. and J.C. Clausen. 1988. Tests of the CREAMS model on agricultural fields in
Vermont. Water Res. Bull. 24(6): 1219-1226.
Jokela, W. 1989. Personal Communication. Assistant Professor. University of Vermont,
Burlington.
65
-------
Lee, M.T. 1987. Verification and application of a nonpoint source pollution model. Proc. ASAE
National Engineering Hydrology Symp. New York, NY.
Lee, M.T. and R. Camacho. 1987. Geographic data base and watershed modeling for evaluation
of the Rural Clean Water Program in the Highland Silver Lake watershed. Illinois Dept.
of Energy and Natural Resources. Water Survey Division. SWS Contract Report'421.
Champaign, IL.
Likens, G.E., F.H. Bormann, R.S. Pierce, J.S. Eaton, N.M. Johnson. 1977. Bio-geo-chemistry
of a forested watershed. Springer-Verlag. New York.
Meals, D.W. Jr. 1990. LaPlatte River watershed water quality monitoring and analysis program.
Comprehensive final report 1979-1989. Vermont Water Resources Research Center.
Burlington, VT.
National Oceanic and Atmospheric Administration. 1983. Local climatological data, Burlington,
Vermont. Nat. Climatic Data Center, Asheville, N.C
Onstad, C.A., R.A. Young, D.D. Bosch, and W.P. Anderson. 1986. Prioritizing surface water
quality problems. USDA-ARS, Morris, MN. Unpublished report.
Reckhow, K.H., J.T. Clements, and R.C. Dodd. 1990. Statistical evaluation of mechanistic
water quality models. J. Env. Engr. 116(2):250-268.
Robert, P.C., J.L. Anderson, C.A. Bunn, R.A. Young, and W.P. Anderson, undated.
Agricultural nonpoint source pollution program user manual (AGNPS-PC). AG-CS-3016.
Minnesota Pollution Control Agency, USDA-Agricultural Research Service, University
of Minnesota.
Smith, R.E. and J.R. Williams. 1980. Simulation of surface water hydrology. In CREAMS, a
field scale model for chemicals, runoff, and erosion from agricultural management
systems. USDA Cons. Res. Report 26. Agr. Res. Serv. Washington, D.C.
Thomann, R.V. 1982. Verification of water quality models. J. Environ. Engr. Div. 108(5):923-
940.
USDA - Soil Conservation Service. 1972. Hydrology. Sec. 4, Chap. 10. SCS National
Engineering Handbook. Washington, D.C.
USDA - Soil Conservation Service. 1987. Universal soil loss equation (USLE). Guidelines for
estimating soil loss. VT Technical Guide, Section I-C. WinoosM, VT.
Vermont RCWP Coordinating Committee. 1991. Final Report St. Albans Bay Watershed Rural
Clean Water Program. Vermont Water Resources Research Center. Burlington, VT.
66
-------
Winter, T.C. 1981. Uncertainties in estimating the water balance of lakes. Water Resources
Bulletin 17(1):83-115.
Wischmeier, W.H. and D.D. Smith. 1958. Rainfall energy and its relationship to soil loss.
Trans. Amer. Geophysical Union. 39(2):285-291.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting rainfall erosion losses. U.S.D.A.
Handbook 537. Washington, D.C.
Wisler, and Brater, 1967. Hydrology. New York: John Wiley & Sons, Inc.
Young, R.A., C.A. Onstad, D.D. Bosch, and W.P. Anderson. 1986. AGNPS: an agricultural
nonpoint source pollution model. USDA-ARS, Morris, MN. Minnesota Pollution Control
Agency. St. Paul, MN. Unpublished report.
Young, R.A., C.A. Onstad, D.D. Bosch, and W.P. Anderson. 1987. AGNPS, agricultural non-
point-source pollution model. A watershed analysis tool. USDA-ARS Conservation
Research Report 35. NTIS, Springfield, VA.
Young, R.A., C.A. Onstad, D.D. Bosch, and W.P. Anderson. 1989. AGNPS: A nonpoint
source pollution model for evaluation of agricultural watersheds. J. Soil and Water Cons.
69-173.
67
-------
GLOSSARY
Antecedent moisture condition (AMC). An indication of the wetness of the soil, with II being
average, I being dryer than average, and HI being wetter than average.
Calibration. The process during model development of adjusting parameter values to match
observed values. Also, synonymous with fitting in some applications.
Distributed model. A model that defines spatial variations that are broken up into homogeneous
area.
Distributed watershed parameters. The variables within a model that change depending upon
location within the watershed.
Erodibility factor. A factor used in the Universal Soil Loss Equation that accounts for the ease
at which different soils may erode.
Event based. A model that simulates a single runoff event and does not simulate flow between
events.
Fitted model. A model that has parameter values obtained by fitting computed results to
observed results.
Lumped model. A model that assumes the watershed is homogeneous.
Manning's n. The roughness coefficient used in the Manning's Equation. Greater roughness
will result in lower stream velocity. Coefficients are available for various stream conditions.
Mass balance approach. A technique of determining all of the mass inputs, all of the mass
outputs, and the storage within a system.
Measured parameters model. A model where all the parameters are from known watershed
characteristics by either measurement or estimation.
Model calibration. See calibration above.
Model verification. Testing of a model with new field data to determine whether the model
adequately predicts observed data.
Nonpoint. With respect to water resources, nonpoint refers to runoff that would originate in
a diffuse manner from the landscape, rather than from a pipe.
68
-------
Rainfall energy intensity. A factor used in the Universal Soil Loss Equation that represents
the energy delivered to the ground to initiate soil detachment. The factor varies with rainfall
intensity, season, and location.
Slope shape. A factor used to adjust soil loss depending on whether the slope is uniform,
convex, or concave.
Surface condition constant. A value used in AGNPS that adjusts the time for overland flow
to become channelized based on the land use condition.
USLE. Represents the Universal Soil Loss Equation used in AGNPS.
Verification. See model verification above.
1 yr - 24 hr storm. The amount of a precipitation storm of a 24 hour duration expected to
occur, on the average, once a year.
69
-------
APPENDIX A
Hydrographs
70
-------
Hydrograph of LaPlatte Event
Date: 4/17/82
O
Time, hr
71
-------
Hydrograph of LaPlatte Event
Date: 4/10/83
Time, hr
72
-------
Hydrograph of LaPlatte Event
Date: 6/6/83
B
10
15
20
25
Time, hr
73
-------
Hydrograph of LaPlatte Event
Date: 8/8/83
025
0.15
01
0.05
0
74
-------
Hydrograph of LaPlatte Event
Date: 8/11/83
1.8
1.6
1.4
1.2
1
0.8
Q6
0.4
0.2
>
10
I
20
Time, hr
I
30
40
75
-------
Hydrograph of LaPlatte Event
Date: 9/21/83
20
15
10
Trne, hr
76
-------
Hydrograph of LaPlatte Event
Date: 11/11/83
o
40
35
30
25
20
15
10
5
10
~
15
Tims hr
20
25
30
77
-------
Hydrograph of LaPlatte Event
Date: 4/17/84
0
Trne. hr
78
-------
Hydrograph of LaPlatte Event
Date: 5/23/84
Tine.
79
-------
Hydrograph of LaPlatte Event
Date: 4/6/85
8
10
15
20
Time, hr
80
-------
Hydrograph of LaPlatte Event
Date: 6/5/85
1.4 rr
Time, hr
81
-------
Hydrograph of LaPlatte Event
Date: 8/3/86
40
Time, hr
82
-------
Hydrograph of LaPlatte Event
Date: 8/27/86
Tine, hr
83
-------
Hydrograph of LaPlatte Event
Date: 9/29/86
1.0 20
Tine, hr
84
-------
Hydrograph of LaPlatte Event
Date: 3/31/87
0
0
I
10
Time, hr
I
15
20
85
-------
Hydrograph of Jewett Brook Event
Date: 4/6/85
B
100
50
0
0
Tine, hr
86
-------
Hydrograph of Jewett Brook Event
Date: 4/16/84
200
150
8
100
50
0
0
10
I
15
20
Trne, hr
30 26
87
-------
Hydrograph of Jewett Brook Event
140
120
100
BO
60
40
20
0
Date: 3/10/83
~
10
I
20
30
40
Tine, hr
-------
Hydrograph of Jewett Brook Event
Date: 10/5/83
8
S
3
I
Q
QB
Q7
Q6
Q5
Q4
Q3
Q2
Q1
0
0
T
5
10
15
20
25
I
30
Time, hr
89
-------
Hydrograph of Jewett Brook Event
Date: 4/17/84
8
100
50
0
Tine, hr
90
-------
Hydrograph of Jewett Brook Event
Date: 5/29/84
8
o
Trne, hr
91
-------
Hydrograph of Jewett Brook Event
Date 4/18/85
8
10
T'me. hr
92
-------
Hydrograph-of Jewett Brook Event
Date: 11/26/86
0
0 10 20
Tine, hr
93
-------
Hydrograph of Jewett Brook Event
Date: 10/28/87
0
Tine, hr
94
-------
Hydrograph of Jewett Brook Event
Date: 5/23/84
50
0
0
95
-------
Hydrograph of Jewett Brook Event
Date: 9/21/83
8
250
Trie, hr
96
-------
APPENDIX B
EXAMPLE INPUT FILES
97
-------
LaPlatte River Subwatershed 3
LaP3 4-10-83 1-15-90 Rl
9.8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
40
4
6
7
9
41
5
6
13
13
9
6
11
12
20
14
22
18
12
18
19
20
27
17
18
19
20
20
27
28
23
24
38
34
27
28
37
31
37
38
37
.9 1.5
52
58
53
55
55
59
61
55
55
58
56
63
57
54
57
58
62
63
58
56
58
59
62
59
54
53
61
59
60
62
60
55
55
56
59
56
56
53
53
58
13.7
13.9
8.0
6.8
24.2
3.7
10.4
8.0
5.2
4.5
3.6
6.6
11.3
15.6
8.7
7.5
4.4
10.9
9.9
6.1
19.4
4.0
7.1
19.3
10.3
9.6
1.6
3.6
3.7
4.0
12.8
8.9
10.3
3.7
1.5
4.1
7.5
5.0
5.6
10.4
1
1
2
1
2
1
1
2
1
1
1
2
1
1
2
1
1
2
2
2
2
2
1
2
3
3
1
1
1
1
1
1
1
1
1
1
1
2
1
1
94
125
155
161
128
188
137
134
161
152
200
138
91
97
116
154
191
126
136
166
132
200
164
124
131
132
200
200
177
200
120
159
129
182
197
198
126
163
167
115
6.7
7.0
4.0
5.0
3.0
3.0
4.6
9.1
2.0
2.3
3.0
1.3
8.1
6.1
4.4
3.8
1.5
3.0
1.4
1.5
6.1
2.0
3.0
4.6
5.2
4.8
1.1
3.0
3.0
2.0
1.5
3.8
1.5
2.9
3.0
2.1
1.5
4.6
3.0
3.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
.100
.035
.100
.100
.030
.030
.080
.075
.080
.080
.075
.080
.035
.030
.030
.030
.035
.035
.035
.030
.035
.035
.035
.035
.100
.080
.080
.035
.080
.035
.035
.080
.080
.100
.080
.080
.080
.080
.080
.075
.35
.42
.43
.44
.49
.49
.40
.37
.47
.37
.49
.45
.37
.44
.32
.43
.49
.48
.45
.47
.39
.49
.47
.39
.40
.40
.49
.49
.46
.49
.43
.43
.39
.47
.49
.49
.46
.49
.48
.40
.01
.60
.01
.01
.01
.01
.40
.01
.40
.40
.40
.40
.60
.01
.01
.01
.60
.60
.60
.01
.60
.60
.60
.60
.01
.40
.40
.60
.40
.60
.60
.40
.40
.01
.40
.40
.40
.40
.40
.40
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
.29 4
.22 6
.29 6
.29 5
.15 7
.15 7
.29 7
.29 5
.29 5
.29 7
.05 1
.22 7
.22 7
.15 5
.15 7
.15 5
.22 3
.22 1
.22 7
.15 7
.22 7
.22 6
.22 1
.22 1
.29 1
.29 1
.29 8
.22 7
.29 7
.22 1
.22 1
.29 5
.29 3
.29 1
.29 1
.29 3
.29 1
.29 7
.29 7
.05 1
2 0
3 0
3 0
3 0
3 0
3 0
3 0
3 0
2 0
3 0
3 0
3 0
2 0
3 0
2 0
3 0
3 0
3 0
3 0
3 0
2 0
3 0
3 0
2 0
2 0
2 0
3 0
3 0
3 0
3 0
3 0
3 0
2 0
3 0
3 0
3 0
3 0
3 0
3 0
2 0
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
o
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
X)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
65
60
65
65
60
60
60
65
60
60
60
60
65
60
60
60
60
60
60
60
60
60
60
60
65
60
60
60
60
60
60
60
60
60
60
80
80
60
60
60
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
98
-------
Jewett Brook
Jewett Brook Event: 3/10/83
9.8 343 0.6 1.2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
3
4
5
3
8
9
9
12
13
14
15
17
17
15
19
21
18
25
25
26
20
28
24
25
26
34
26
36
28
31
32
33
34
42
44
44
46
37
49
50
51
43
53
43
56
56
46
47
61
62
63
53
65
53
54
67
56
70
48
42
51
46
55
56
58
63
55
36
55
59
54
57
56
56
48
54
61
66
50
55
61
55
52
60
51
54
61
64
59
61
63
66
64
57
59
58
66
63
64
58
61
63
67
65
69
57
66
64
65
62
64
66
60
60
53
45
2.8
1.7
1.9
1.5
1.1
1.4
1.2
1.0
1.2
1.5
1.3
1.0
1.1
1.9
1.5
1.5
1.6
1.2
1.0
1.4
1.6
1.2
2.0
1.2
1.0
1.0
1.5
1.2
1.1
1.7
3.8
1.6
1.0
1.0
1.0
1.2
1.2
1.7
1.1
1.7
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
1.1
1.2
1.3
1.0
1.0
1.0
1.0
1.0
1.4
1.7
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
314
328
351
354
398
374
437
423
382
351
404
430
391
320
349
348
340
380
400
374
362
400
300
381
400
441
373
394
442
352
257
371
403
450
444
417
403
339
396
364
400
413
450
450
450
450
449
381
390
388
384
433
450
450
450
450
387
334
1.4
.9
1.0
.8
.6
.7
.6
.5
.6
.8
.7
.5
.6
1.0
.8
.8
.8
.6
.5
.7
.8
.1
1.0
.6
.5
.5
.8
.1
.6
.9
1.9
.8
.5
.5
.5
.1
.6
.9
.6
.9
.5
.5
.1
.1
.5
.5
.5
.8
.6
.6
.7
.5
.1
.5
.5
.5
.7
.9
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
.130
.200
.200
.200
.200
.080
.200
.050
.080
.200
.200
.080
.080
.050
.080
.200
.080
.080
.080
.050
.130
.100
.050
.200
.080
.080
.100
.100
.100
.130
.050
.050
.050
.080
.050
.200
.100
.050
.130
.130
.130
.080
.080
.080
.080
.070
.080
.050
.080
.130
.130
.100
.080
.080
.200
.070
.100
.070
.30
.31
.32
.30
.33
.36
.47
.38
.28
.28
.41
.40
.28
.31
.29
.28
.28
.30
.32
.34
.29
.30
.25
.31
.32
.46
.30
.24
.47
.26
.31
.33
.33
.49
.48
.37
.33
.24
.31
.31
.35
.36
.49
.49
.49
.49
.49
.33
.33
.28
.33
.43
.49
.49
.49
.49
.35
.17
.01
.02
.02
.02
.02
.02
.02
.51
.51
.02
.02
.02
.02
.51
.02
.02
.02
.02
.51
.51
.01
.01
.51
.02
.02
.02
.01
.01
.01
.01
.51
.51
.51
.51
.51
.02
.01
.51
.01
.01
.01
.02
.02
.02
.51
.51
.01
.51
.01
.01
.01
.01
.01
.51
.02
.02
.01
.01
.50
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.50
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
1.00
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.22
.59
.59
.59
.59
.59
.59
.05
.05
.59
.59
.59
.59
.05
.59
.59
.59
.59
.05
.05
.22
.29
.05
.59
.59
.59
.29
.29
.29
.22
.05
.05
.05
.05
.05
.59
.29
.05
.22
.22
.22
.59
.59
.59
.05
.05
.01
.05
.22
.22
.22
.29
.29
.05
.59
.59
.29
.29
5 2
5 2
5 2
7 2
5 2
5 2
6 2
5 2
5 2
5 2
5 2
4 2
5 2
3 2
5 1
4 1
3 2
4 2
5 2
5 2
7 1
5 2
3 1
3 2
3 2
5 2
7 1
5 2
7 3
3 2
3 2
32
3 2
6 2
5 2
6 2
5 2
7 1
5 2
5 2
5 2
3 3
5 3
7 2
4 2
5 2
7 2
7 2
5 2
5 2
5 2
3 3
5 3
7 2
7 2
6 2
7 2
5 1
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
0
3
3
3
1
0
0
0
1
3
3
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
10
10
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
60
60.
60
60
60
60
60
115
115
60
60
60
60
115
60
60
60
60
115
115
60
65
115
60
60
60
65
65
65
60
115
115
115
115
115
60
65
115
60
60
60
60
60
60
115
115
80
115
60
60
60
65
65
115
60
60
65
65
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
99
-------
59 58
60 61
61 73
62 74
63 64
64 76
65 77
66 78
67 79
68 67
69 68
70 82
71 70
72 60
73 74
74 75
75 88
76 77
77 78
78 91
79 91
80 79
81 80
82 81
83 82
84 85
85 74
86 87
87 88
88 100
89 77
90 102
91 90
92 104
93 81
94 93
95 82
96 97
97 98
98 99
99 100
100 112
101 113
102 114
103 115
104 116
105 117
106 105
107 106
108 96
109 97
110 98
111 99
112 124
113 112
114 113
115 127
116 115
117 116
118 117
49
61
69
63
58
62
64
66
60
63
54
42
49
55
70
57
52
59
64
64
60
65
64
59
64
57
59
52
52
56
61
60
65
65
70
67
66
56
55
54
55
54
66
65
69
70
68
68
65
54
47
50
60
65
64
61
62
66
63
60
1.6
1.8
1.6
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
1.9
1.6
2.6
1.4
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
1.5
1.0
1.1
1.0
1.5
1.0
1.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.2
3.0
1.3
1.0
1.3
1.0
1.0
1.0
1.0
1.0
1.2
1.2
2.0
4.8
1.9
1.0
1.3
1.0
1.0
1.0
1.2
1.0
1.0
1 339
1 325
1 337
1 398
1 405
1 443
1 450
1 450
1 450
1 450
1 379
1 311
1 359
1 317
1 357
1 399
1 400
1 427
1 450
1 450
1 450
1 450
1 381
1 371
1 448
1 320
1 400
1 377
1 400
1 419
1 450
1 450
1 450
1 450
1 450
1 450
1 450
3 360
3 354
2 383
1 399
1 399
1 450
1 450
1 450
1 450
1 450
1 425
1 413
2 352
2 298
3 357
1 396
1 405
1 450
1 450
1 447
1 418
1 419
1 420
.8
.9
.8
.5
.5
.5
.1
.5
.5
.5
.8
1.0
.8
1.3
.7
.5
.5
.5
.1
.5
.5
.5
.8
.8
.5
.6
.5
.8
.5
.6
.1
.5
.5
.5
.5
.5
.5
.6
1.5
.7
.5
. .7
.1
.5
.5
.5
.5
.6
.6
1.0
2.4
1.0
.5
.1
.5
.5
.5
.6
.5
.5
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
.070
.080
.080
.080
.080
.080
.080
.050
.070
.080
.080
.070
.100
.080
.080
.080
.080
.100
.080
.080
.070
.070
.070
.070
.080
.130
.080
.200
.200
.080
.100
.070
.070
.080
.050
.080
.080
.200
.080
.080
.080
.080
.100
.070
.080
.080
.080
.080
. O80
.200
.080
.080
.080
.070
.070
.070
.070
.070
.070
.070
.17
.24
.43
.32
.35
.49
.49
.49
.49
.49
.33
.15
.28
.28
.47
.37
.32
.44
.49
.49
.49
.49
.33
.31
.48
.29
.41
.30
.32
.42
.49
.49
.49
.49
.49
.49
.49
.31
.33
.32
.32
.40
.49
.49
.49
.49
.49
.45
.43
.27
.28
.30
.31
.41
.49
.49
.48
.43
.41
.46
.01
.01
.51
.51
.02
.01
.01
.51
.02
.02
.01
.01
.01
.51
.51
.02
.02
.01
.01
.02
.02
.02
.51
.51
.01
.01
.02
.05
.02
.02
.01
.02
.02
.51
.51
.51
.01
.02
.01
.02
.02
.02
.01
.02
.51
.51
.51
.51
.01
.02
.02
.02
.51
.51
.02
.02
.02
.01
.01
.02
.60 .29 7 1 0 100 0
.60 .22 3 2 0 100 0
.60 .05 5 3 0 100 0
.60 .05 5 2 0 100 0
.60 .59 3 2 0 100 0
.60 .29 5 4 0 100 0
.60 .29 5 3 0 100 0
.60 .05 5 2 0 100 0
.60 .22 5 2 0 100 0
.60 .59 7 2 0 100 0
.60 .29 7 2 0 100 0
.60 .29 5 1 0 100 0
.60 .29 7 1 0 100 0
.50 .05 1 2 0 100 0
.60 .05 3 3 0 100 0
.60 .59 3 2 0 100 0
.60 .59 4 20 100 0
.60 .29 3 3 0 100 0
.60 .29 3.3 0 100 0
.60 .59 4 2 0 100 0
.60 .59 5 2 0 100 0
.60 .59 7 2 0 100 0
.60 .05 7 2 0 100 0
.60 .05 7 1 0 100 0
.60 .29 7 2 0 100 0
.60 .22 3 3 0 100 0
.60 .59 2 3 0 100 0
.60 .59 3 2 0 100 0
.60 .59 2 2 0 100 0
.60 .59 5 3 0 100 0
.60 .29 1 3 0 100 0
.60 .59 5 2 0 100 0
.60 .59 7 2 0 100 0
.60 .05 5 2 0^100 0
.60 .05 1 2 0 100 0
.60 .05 72 0 100 0
.60 .29 8 2 0 100 0
.60 .59 3 2 0 100 0
.50 .22 3 2 0 100 0
.60 .59 3 .2 0 100 0
.60 .59 3 2 0 100 0
.60 .59 5 2 0 100 0
.60 .29 5 3 0 100 0
.60 .59 5 2 0 100 0
.60 .05 4 2 0 100 0
.60 .05 5 2 0 100 0
.60 .05 5 2 0 100 0
.60 .05 7 2 0 100 0
.60 .29 720 100 0
.60 .59 12 0 100 0
.50 .59 1 2 0 100 0
.60 .59 1 2 0 100 0
.60 .05 1 2 0 100 0
.60 .05 520 100 0
.60 .59 7 2 0 100 0
.60 .59 7 2 0 100 0
.60 .59 5 2 0 100 0
.60 .22 7 2 0 100 0
.60 .22 7 2 0 100 0
.60 .59 7 2 0 100 0
0 65
0 60
0 115
0 115
0 60
0 65
0 65
0 115
0 60
0 60
0 65
0 65
0 65
0 115
0 115
0 60
0 60
0 65
0 65
0 60
0 60
0 60
0 115
0 115
0 65
0 60
0 60
0 60
0 60
0 60
0 65
0 60
0 60
0 115
0 115
0 115
0 65
0 60
0 60
0 60
0 60
0 60
0 65
0 60
0 115
0 115
0 115
0 115
0 65
0 60
0 60
0 60
0 115
0 115
0 60
0 60
0 60
0 60
0 60
0 60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
100
-------
119 106
120 108
121 109
122 111
123 111
124 139
125 124
126 140
127 126
128 116
129 117
130 117
131 130
132 147
133 149
134 149
135 136
136 153
137 154
138 139
139 155
140 155
141 140
142 141
143 127
144 128
145 130
146 131
147 163
148 149
149 166
150 166
151 135
152 170
153 154
154 155
155 172
156 155
157 140
158 157
159 142
160 144
161 162
162 146
163 162
164 163
165 182
166 182
167 184
168 169
169 170
170 171
171 188
172 189
173 172
174 156
175 157
176 175
177 176
178 179
65
55
51
57
59
58
63
62
59
57
59
56
60
54
57
51
55
49
62
64
66
63
57
60
57
60
54
61
62
61
54
47
60
62
61
64
65
53
52
54
56
64
65
60
61
61
61
53
55
51
52
65
65
58
59
53
52
56
65
69
1.2
2.9
3.0
1.1
1.0
1.1
1.0
1.0
1.0
1.5
1.0
1.0
1.0
1.0
1.4
2.4
2.3
4.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.2
1.0
1.0
1.0
1.0
2.1
2.0
1.3
3.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1.0
1.0
1.7
1.2
1.3
1.9
1.7
1.9
2.2
1.0
1.0
1.1
1.0
1.0
1.0
1.0
1.0
1.0
1 425
2 300
1 302
1 394
1 400
1 430
1 450
1 448
1 411
1 351
3 400
3 400
1 412
1 409
1 376
1 293
1 319
1 251
2 400
1 412
1 450
1 444
1 413
1 400
1 400
1 382
3 400
1 410
1 449
1 400
3 353
2 299
2 365
1 294
1 402
1 446
1 441
1 402
1 400
1 400
1 400
1 391
3 400
1 400
1 388
1 397
1 375
1 356
1 338
2 315
1 332
1 400
1 444
1 416
1 400
1 400
1 400
1 400
1 400
3 400
.6
1.5
1.5
.6
.5
.1
.5
.5
.5
.8
.5
.5
.5
.5
.7
1.2
1.2
2.0
.5
.5
.1
.5
.5
.5
.5
.6
.5
.5
.5
.5
1.1
1.0
.7
1.6
.5
.5
.1
.5
.5
.5
.5
.6
.5
.5
.9
.6
.6
1.0
.9
1.0
1.1
.5
.5
.1
.5
.5
.5
.5
.5
.5
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
.080
.130
.200
.130
.080
.070
.080
.070
.070
.130
.080
.070
.070
.070
.130
.130
.080
.080
.080
.080
.070
.070
.080
.100
.100
.080
.200
.070
.070
.130
.130
.130
.080
.080
.080
.080
.070
.080
.080
.080
.100
.130
.080
.070
.070
.100
.130
.080
.130
.130
.080
.080
.080
.070
.080
.080
.080
.080
.130
.080
.45
.25
.30
.28
.29
.45
.49
.48
.33
.31
.30
.39
.41
.23
.31
.24
.26
.31
.28
.33
.49
.47
.37
.36
.31
.29
.34
.47
.49
.35
.30
.24
.30
.31
.29
.47
.46
.32
.32
.30
.30
.28
.37
.35
.41
.10
.36
.29
.29
.28
.31
.28
.47
.41
.29
.30
.28
.28
.28
.43
.02
.01
.02
.01
.51
.02
.02
.02
.01
.01
.01
.02
.01
.02
.01
.01
.01
.01
.01
.01
.01
.02
.01
.01
.01
.01
.02
.01
.01
.01
.01
.01
.51
.51
.51
.02
.01
.02
.02
.02
.01
.01
.01
.01
.01
.01
.01
.01
.01
.01
.01
.51
.51
.02
.02
.02
.02
.01
.01
.01
.60 .59 8 2 0 100 0
.50 .22 1 2 0 100 0
.50 .59 1 2 0 100 0
.60 .22 2 2 0 100 0
.60 .05 1 2 0 .100 0
.60 .59 5 2 0 100 0
.60 .59 7 2 0 100 0
.60 .59 6 2 0 100 0
.60 .29 720 100 0
.60 .22 1 2 0 100 0
.60 .22 1 2 0 100 0
.60 .59 8 2 0 100 0
.60 .22 7 2 0 100 0
.60 .59 5 1 0 100 0
.60 .22 5 2 0 100 0
.50 .22 620 100 0
.50 .22 3 2 0 100 0
.50 .22 4 2 0 100 0
.60 .22 4 2 0 100 0
.60 .22 3 2 0 100 0
.60 .22 5 2 0 100 0
.60 .59 6 2 0 100 0
.60 .29 720 100 0
.60 .29 7 2 0 100 0
.60 .29 8 2 0 100 0
1.00 .01 8 2 0 100 0
.60 .59 1 2 0 100 0
.60 .22 1 3 0 100 0
.60 .29 5 4 0 100 0
.60 .22 320 100 0
.50 .22 521 100 0
.60 .22 6 2 0 100 0
.60 .05 1 2 0 100 0
.50 .05 4 2 0 100 0
.60 .05 3 2 0 100 0
.60 .59 3 2 0 100 0
.60 .22 520 100 0
.60 .59 7 2 0 100 0
.60 .59 8 2 0 100 0
.60 .59 7 2 0 100 0
.60 .29 8 2 0 100 0
.60 .22 1 2 1 100 0
.60 .22 3 20 100 0
.60 .22 1 2 0 100 0
.60 .29 7 3 0 100 0
.60 .29 740 100 0
.60 .22 5 2 0 100 0
.60 .22 6 2 1 100 0
.60 .22 5 2 0 100 0
.60 .22 3 2 0 100 0
.50 .22 320 100 0
.60 .05 3 2 0 100 0
.60 .05 5 2 0 100 0
.60 .59 520 100 0
.60 .59 7 2 0 100 0
.60 .59 8 2 0 100 0
.60 .59 8 2 0 100 0
.60 .22 7 2 1 100 0
.60 .22 7 2 1 100 0
.60 .22 330 100 0
0 60
0 60
0 60
0 60
0 115
0 6.0
0 60
0 60
0 65
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0 65
0 65
0 65
0 80
0 60
0 60
0 65
0 60
0 60
0 60
0 115
0 115
0 115
0 60
0 60
0 60
0 60
0 60
0 65
0 60
0 60
0 60
0 65
0 65
0 60
0 60
0 60
0 60
0 60
0 115
0 115
0 60
0 60
0 60
0 60
0 60
0 60
0 60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
101
-------
179 162
180 179
181 164
182 201
183 182
184 203
185 205
186 206
187 188
188 207
189 208
190 172
191 190
192 175
193 176
194 193
195 196
196 179
197 196
198 181
199 198
200 201
201 220
202 220
203 222
204 223
205 206
206 207
207 227
208 227
209 '208
210 192
211 192
212 193
213 194
214 195
215 196
216 197
217 198
218 217
219 220
220 239
221 240
222 221
223 224
224 225
225 245
226 227
227 246
228 227
229 228
230 249
231 211
232 231
233 214
234 215
235 216
236 217
237 238
238 256
70
68
61
63
61
61
46
50
62
65
57
66
65
53
53
59
63
68
68
64
62
60
56
54
51
45
52
50
58
55
60
60
57
53
54
66
57
58
61
62
56
58
56
56
52
54
58
61
63
58
56
67
61
53
57
58
58
58
58
57
1.0
1.0
1.0
1.2
1.0
1.3
2.6
1.5
1.1
1.3
1.2
1.0
1.0
1.0
1.0
1.0
1.4
1.7
1.5
1.2
.9
1.4
1.1
1.0
1.1
1.6
1.1
1.7
1.3
1.6
1.6
1.7
1.0
1.0
1.0
1.3
1.3
1.2
1.0
.7
1.2
1.0
1.6
1.9
2.0
2.3
1.6
1.2
2.0
1.7
1.0
1.2
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1 400
1 399
1 400
1 379
1 400
1 389
2 293
2 372
3 426
1 394
1 381
1 400
1 399
1 400
1 400
1 404
3 378
1 332
1 348
1 376
1 361
1 362
1 393
1 400
1 386
3 338
3 396
1 330
1 375
1 341
1 349
1 367
1 400
1 400
1 407
1 384
1 375
1 378
1 347
1 294
1 380
1 400
1 350
1 306
1 328
1 334
1 342
1 382
1 300
1 326
1 396
1 376
1 400
1 400
1 400
1 400
1 400
1 400
1 400
1 393
.5 10.0 .080 .49 .01
.5 10.0 .080 .42 .01
.5 10.0 .100 .14 .01
.6 10.0 .080 .41 .01
.5 10.0 .080 .29 .51
.7 10.0 .200 .33 .02
1.3 10.0 .130 .30 .01
.8 10.0 .200 .38 .02
.6 10.0 .080 .42 .02
.7 10.0 .080 .44 .51
.1 10.0 .070 .33 .51
.5 10.0 .080 .30 .51
.5 10.0 .050 .29 .51
.5 10.0 .080 .28 .02
.5 10.0 .080 .28 .02
.5 10.0 .080 .30 .02
.7 10.0 .080 .43 .01
.9 10.0 .080 .49 .01
.8 10.0 .080 .42 .01
.6 10.0 .100 .17 .01
.5 10.0 .100 .00 .01
.7 10.0 .130 .33 .01
.6 10.0 .080 .34 .02
.5 10.0 .080 .33 .02
.6 10.0 .200 .29 .02
.8 10.0 .200 .30 .02
.6 10.0 .200 .34 .02
.9 10.0 .080 .39 .02
.7 10.0 .080 .37 .02
.1 10.0 .070 .33 .02
.8 10.0 .080 .38 .02
.9 10.0 .080 .27 .51
.5 10.0 .080 .28 .02
.5 10.0 .080 .28 .02
.5 10.0 .080 .31 .02
.7 10.0 .080 .41 .01
.7 10.0 .130 .33 .01
.6 10.0 .080 .33 .01
.5 10.0 .100 .16 .01
.4 10.0 .100 .10 .01
.6 10.0 .080 .33 .01
.5 10.0 .080 .36 .02
.8 10.0 .080 .28 .02
1.0 10.0 .130 .30 .01
1.0 10.0 .130 .33 .01
1.2 10.0 .080 .31 .02
.8 10.0 .080 .43 .02
.6 10.0 .080 .40 .02
.1 10.0 .070 .49 .02
.9 10.0 .200 .45 .02
.5 10.0 .100 .29 .01
.6 10.0 .080 .33 .51
.5 10.0 .080 .28 .51
.5 10.0 .080 .28 .02
.5 10.0 .080 .28 .01
.5 10.0 .130 .28 .01
.5 10.0 .130 .28 .01
.5 10.0 .130 .18 .01
.5 10.0 .200 .42 .02
.6 10.0 .080 .40 .01
.60
.60
.60
.60
.60
.60
.50
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.65
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.65
.60
.60
.60
.60
.60
.50
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.22
.22
.29
.22
.05
.59
.22
.59
.59
.05
.05
.05
.05
.59
.59
.59
.22
.22
.22
.29
.29
.22
.59
.59
.59
.59
.59
.59
.59
.59
.59
.05
.59
.59
.59
.22
.22
.22
.29
.29
.22
.59
.59
.22
.22
.59
.59
.59
.59
.59
.29
.05
.05
.59
.22
.22
.22
.22
.59
.22
130
730
130
530
720
520
420
430
320
520
520
820
720
120
120
721
320
130
730
130
740
320
520
620
520
520
320
320
420
520
720
220
120
120
120
120
120
120
130
740
320
520
5 20
720
320
320
430
330
530
730
720
520
820
720
120
120
120
130
330
530
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
60
60
65
60
115
60
60
60
60
115
115
115
115
60
60
60
60
60
60
65
65
60
60
60
60
60
60
60
60
60
60
115
60
60
60
60
60
60
65
65
60
60
60
60
60
60
60
60
60
60
65
115
115
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
102
-------
239 238
240 239
241 259
242 243
243 261
244 245
245 263
246 263
247 246
248 247
249 267
250 249
251 250
252 233
253 252
254 271
255 256
256 273
257 256
258 239
259 276
260 278
261 278
262 279
263 280
264 263
265 264
266 247
267 283
268 249
269 268
270 252
271 270
272 273
273 274
274 290
275 291
276 292
277 294
278 294
279 278
280 296
281 264
282 265
283 282
284 283
285 267
286 285
287 270
288 271
289 290
290 291
291 292
292 305
293 307
294 307
295 294
296 295
297 280
298 282
54
57
61
50
59
58
59
65
65
56
64
66
62
57
64
65
62
61
65
63
64
53
60
65
63
59
65
59
66
68
63
64
65
62
57
58
65
63
48
50
59
58
55
69
68
66
61
65
59
64
64
68
66
57
53
58
60
57
60
69
1.0
2.0
1.8
1.9
1.8
1.4
1.5
1.9
1.8
1.3
1.7
1.0
1.0
1.0
1.0
1.6
1.0
1.2
1.2
1.5
1.2
1.3
1.7
1.7
1.6
1.5
2.0
1.7
1.4
1.0
1.0
1.0
1.5
1.0
1.2
1.5
1.4
1.6
3.6
1.9
3.2
1.1
1.0
1.4
1.3
2.3
4.6
1.0
1.0
1.0
1.0
1.1
1.3
2.1
3.0
3.3
2.8
1.2
1.1
1.3
1 400
1 326
1 320
1 332
1 336
1 362
1 352
1 311
1 316
1 372
1 330
1 400
1 400
1 400
1 398
1 371
1 400
1 383
1 382
1 350
1 377
1 368
1 334
1 327
1 342
1 346
1 300
1 333
1 385
1 430
1 400
1 400
1 375
1 400
1 378
1 350
1 355
1 368
1 289
1 312
1 303 .
1 394
1 398
1 363
1 367
3 343
1 331
1 409
1 400
1 400
1 402
1 391
1 365
1 337
1 294
1 258
1 312
1 391
1 389
1 366
.5 10.0 .080 .32 .02
1.0 10.0 .080 .28 .02
.9 10.0 .080 .42 .02
1.0 10.0 .080 .38 .02
.9 10.0 .080 .45 .02
.7 10.0 .080 .41 .02
.8 10.0 .080 .47 .02
1.0 10.0 .070 .47 .51
.9 10.0 .050 .46 .51
.1 10.0 .200 .34 .02
.9 10.0 .080 .43 .51
.5 10.0 .050 .28 .51
.5 10.0 .050 .28 .51
.5 10.0 .080 .28 .01
.5 10.0 .050 .29 .51
.8 10.0 .130 .27 .01
.5 10.0 .100 .49 .01
.6 10.0 .080 .44 .02
.6 10.0 .050 .32 .51
.8 10.0 .080 .43 .51
.6 10.0 .080 .48 .02
.7 10.0 .130 .35 .01
.9 10.0 .080 .44 .01
.9 10.0 .080 .41 .01
.1 10.0 .070 .45 .02
.3 10.0 .080 .37 .02
.3 10.0 .080 .49 .51
.9 10.0 .200 .47 .02
.7 10.0 .080 .48 .51
.5 10.0 .050 .41 .51
.5 10.0 .080 .28 .51
.5 10.0 .080 .37 .01
.8 10.0 .130 .32 .01
.5 10.0 .100 .49 .01
.6 10.0 .080 .32 .01
.8 10.0 .080 .38 .02
.7 10.0 .050 .48 .51
.8 10.0 .080 .46 .51
1.8 10.0 .130 .32 .01
1.0 10.0 .080 .40 .02
1.6 10.0 .080 .30 .01
.1 10.0 .070 .23 .01
,5 10.0 .130 .21 .01
.3 10.0 .080 .46 .51
.7 10.0 .080 .49 .51
1.2 10.0 .080 .36 .01
2.3 10.0 .080 .37 .51
.5 10.0 .050 .32 .51
.5 10.0 .130 .31 .01
.5 10.0 .130 .28 .01
.5 10.0 .080 .42 .01
.6 10.0 .130 .42 .01
.7 10.0 .050 .49 .51
1.1 10.0 .080 .44 .02
1.5 10.0 .130 .34 .01
.4 10.0 .070 .41 .01
.1 10.0 .070 .33 .01
.1 10.0 .070 .27 .01
.6 10.0 .130 .30 .01
.3 10.0 .080 .47 .51
.60
.60
.60
.60
.60
.60
. .60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.60
.50
.60
.50
.60
.60
.60
.60
.50
.50
.60
.60
.60
.60
.60
.60
.50
.50
.50
.50
.60
.60
.60
.59 7
.59 7
.59 5
.59 3
.59 5
.59 3
.59 5
.05 6
.05 7
.59 7
.05 5
.05 7
.05 7
.22 1
.05 7
.22 6
.29 3
.59 5
.05 7
.05 8
.59 5
.22 4
.22 5
.22 5
.59 5
.59 8
.05 7
.59 8
.05 6
.05 8
.05 7
.22 1
.22 7
.29 3
.29 3
.59 5
.05 5
.05 5
.22 4
.59 5
.22 7
.22 5
.22 1
.05 1
.05 7
.22 7
.05 8
.05 7
.22 1
.22 8
.29 3
.22 3
.05 3
.59 5
.22 4
.22 5
.22 7
.22 7
.22 8
.05 1
2 0
2 0
2 0
2 0
3 0
3 0
3 0
3 0
3 0
2 0
3 0
2 0
2 0
2 0
2 0
2 0
3 0
3 0
2 0
3 0
3 0
2 0
3 0
3 3
3 2
3 0
3 0
3 0
3 0
2 0
2 0
2 0
2 0
3 0
2 0
3 0
3 0
3 0
2 0
2 0
2 3
2 2
2 0
3 0
3 0
2 0
2 0
2 0
2 0
2 0
3 0
3 0
3 0
3 0
2 0
3 0
2 0
2 0
2 0
3 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
0
0
0
0
0
0
b
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
60
60
60
60
60
60
60
115
115
60
115
115
115
60
115
60
65
60
115
115
60
60
60
60
60
60
115
60
115
115
115
60
60
65
65
60
115
115
60
60
60
60
60
115
115
60
115
115
60
60
65
60
115
60
60
60
60
60
60
115
0
0
6
0
0
0
'0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
103
-------
299 298
300 284
301 285
302 286
303 304
304 305
305 317
306 318
307 306
308 295
309 296
310 297
311 298
312 299
313 300
314 301
315 302
316 317
317 318
318 327
319 318
320 307
321 309
322 311
323 311
324 311
325 313
326 314
327 344
328 321
329 322
330 323
331 324
332 325
333 328
334 329
335 330
336 335
337 336
338 334
339 334
340 339
341 338
342 339
343 342
65
56
42
54
67
70
66
59
57
63
57
59
63
63
48
48
46
68
68
49
58
61
48
51
57
63
51
49
49
46
48
60
53
46
52
59
61
44
44
53
56
48
45
56
48
1.5 1
1.3 2
9.8 1
1.5 1
1.0 1
1.4 1
2.1 1
4.0 1
4.6 1
3.2 1
1.9 1
1.0 1
1.2 1
1.7 1
3.0 2
5.7 1
3.2 1
1.5 1
2.1 1
7.3 1
2.1 1
1.5 1
4.2 1
2.6 1
2.9 1
1.0 1
2.1 3
2.6 3
6.0 1
2.6 1
1.8 1
2.7 1
2.5 3
6.2 3
2.3 1
1.1 1
1.3 1
4.3 3
5.5 2
1.4 1
1.0 1
4.8 2
3.6 1
1.1 1
2.6 2
345
388
180
376
400
362
318
265
244
301
351
400
387
330
299
286
289
350
318
218
303
360
267
341
305
398
344
327
223
318
332
316
327
214
335
386
373
246
235
379
400
273
268
395
339
.8 10.0 .080 .49 .02
.7 10.0 .080 .24 .01
4.9 10.0 .100 .23 .01
.8 10.0 .100 .23 .01
.5 10.0 .080 .32 .51
.7 10.0 .080 .49 .01
1.1 10.0 .080 .46 .01
.4 10.0 .070 .37 .01
.4 10.0 .070 .38 .01
1.6 10.0 .080 .33 .01
1.0 10.0 .080 .35 .01
.5 10.0 .130 .36 .01
.7 10.0 .080 .45 .02
.7 10.0 .200 .43 .02
1.5 10.0 .100 .26 .01
2.9 10.0 .100 .22 .01
1.6 10.0 .100 .24 .01
.8 10.0 .080 .49 .51
1.1 10.0 .080 .48 .51
.4 10.0 .070 .27 .01
1.1 10.0 .080 .46 .02
-.8 10.0 .080 .38 .02
2.1 10.0 .130 .28 .01
1.3 10.0 .130 .27 .01
.7 10.0 .080 .32 .01
.7 10.0 .080 .26 .01
1.1 10.0 .130 .22 .01
1.3 10.0 .130 .23 .01
.4 10.0 .070 .34 .02
1.3 10.0 .200 .30 .02
.9 10.0 .200 .30 .02
.7 10.0 .080 .30 .01
1.3 10.0 .080 .25 .01
3.1 10.0 .200 .29 .02
1.2 10.0 .200 .29 .02
.6 10.0 .130 .28 .01
.7 10.0 .080 .30 .01
2.2 10.0 .130 .28 .01
2.8 10.0 .200 .27 .02
.7 10.0 .100 .22 .01
.5 10.0 .100 .27 .01
2.4 10.0 .200 .22 .02
1.8 10.0 .100 .25 .01
.6 10.0 .100 .28 .01
1.3 10.0 .100 .22 .01
.60
.60
.50
.60
.60
.60
.50
.50
.50
.50
.60
.60
.60
.60
.50
.50
.50
.60
.50
.50
.50
.60
.50
.50
.50
.60
.50
.50
.50
.50
.60
.50
.50
.50
.50
.60
.60
.50
.50
.60
.60
.50
.50
.60
.50
.59
.29
.29
.29
.05
.22
.22
.22
.22
.22
.22
.22
.59
.59
.29
.29
.29
.05
.05
.22
.59
.59
.22
.22
.22
.22
.22
.22
.59
.59
.59
.22
.22
.59
.59
.22
.22
.22
.59
.29
.29
.59
.29
.29
.29
730
120
120
120
320
330
530
520
720
120
120
130
130
130
120
120
120
330
330
520
730
830
120
220
120
820
120
120
520
120
120
120
120
120
120
120
820
720
720
120
820
720
120
120
720
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
100 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
60
65
65
65
115
60.
60
60
60
60
60
60
60
60
65
65
65
115
115
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
65
65
60
65
65
65
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
104
------- |