Sabine Grunwald

GIS-Based Simulations of Non-Point Source Pollution Using AGNPSm

Collaborators
Time
Funding Source
AGNPS
AGNPSm - Modified AGNPS
Objectives
Study Areas
Linkage Between GIS input data and AGNPSm
Entropy / Sensitivity Analysis
AGNPSm Simulation Results
Conclusions
Publications

Collaborators

Ph.D. project - Sabine Grunwald, Department of Natural Resources, University of Giessen,

Germany

Major advisor: Professor H.G. Frede, Department of Natural Resources, University of Giessen,

Germany

Collaborators:

D.L. Norton; National Soil Erosion Research Laboratory, United States Department of

Agriculture (USDA) Agricultural Research Service (ARS), West Lafayette, IN, USA.

S. Haverkamp, M. Rode, and F. Lotz; Department of Natural Resources, University of Giessen,

Germany.

Chaubey I. and C.T. Haan; Department of Biosystems and Agricultural Engineering, Oklahoma

State University.

Time

01/1993 to 12/1996

Funding Source

Honors fellowship State Hessen, Germany

AGNPS

Agricultural Non-Point Source Pollution Model

AGNPS is an event-based, deterministic-analytical water quality model (Young et al, 1987; 1994).

Algorithms	References
Surface runoff - Curve Number (CN) method	(USDA-SCS, 1972)
Peak flow	Smith et al. (1980)
Runoff velocity	Manning's equation
Soil loss - Modified USLE (Universal Soil Loss Equation)	(Wischmeier et al., 1978)
Sediment transport capacity - modified stream power equation	Bagnold (1966)
Sediment transport - stationary continuity equation	Foster et al. (1981) and Lane (1982)
Nutrient transport (N, P)	Frere et al. (1980)

References

Bagnold R.A., 1966. An approach to the sediment transport problem from general physics.

U.S. Geological Survey Professional Paper 422-I, Washington.

Foster G.R., L.J. Lane, J.D. Nowlin, L.M. Laflen, and R.A. Young. 1981. Estimating erosion and

sediment yield on field-sized areas. Trans. of the ASAE, 24 (5): 1253-1262.

Frere M.H., J.D. Ross, and L.J. Lane. 1980. The nutrient submodel. In: Knisel W.G. (ed.).1980.

CREAMS: A field-scale model for chemicals, runoff, and erosion from agricultural management

systems. U.S. Department of Agriculture, Conservation Research Report, No. 26.

Lane L.J., 1982. Development of a procedure to estimate runoff and sediment transport in ephemeral

streams. In: Recent developments in the explanation and prediction of erosion and sediment yield.

Publ. No. 137, Int. Assoc. Hydrological Science, Wallingford, England: 275-282.

Smith R.E., and J.R. Williams. 1980. Simuation of the surface water hydrology. In: Knisel W.G.

(ed.): CREAMS: A field-scale model for chemicals, runoff, and erosion from agricultural management

systems. USDA, Conservation Research Report, 26: 13-35.

USDA-SCS. 1972. United States Department of Agriculture - Soil Conservation Service. National

Engineering Handbook, Sec. 4. Hydrology.

Wischmeier W., and D.D. Smith. 1978. Predicting rainfall erosion losses - A guide to conservation

planning. USDA, Handbook No.537.

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. United States Department

of Agriculture, Conservation Research Report 35.

Young R.A., C.A. Onstad, D.D. Bosch, W.P. Anderson. 1994. Agricultural Non-Point Source

Pollution Model, Version 4.03 - AGNPS User's Guide.

AGNPSm - Modified AGNPS

Modifications were integrated into the source code of AGNPS in order to adjust to Western

European climate and land use conditions (model transfer), as well as to improve simplified

AGNPS model routines. Changes made to the source code of AGNPS included the following:

(1) Replacement of the SCS Curve Number method by the Lutz method (Lutz, 1984) for

calculation of surface runoff

(2) Replacement of the LS factor algorithm of Wischmeier & Smith by the algorithm of

Moore et al. (1986) based on stream power theory

(3) Linkage of channel erosion by individual categories of particle size to runoff velocity

(4) Replacement of uniform precipitation input by grid-based precipitation input.

References

Lutz W., 1984. Berechnung von Hochwasserabfluessen unter Anwendung von Gebiets-

kenngroessen. Ph.D. Thesis, Karlsruhe University, Germany.

Moore I.D., and G.J. Burch. 1986: Physical basis of the Length-Slope Factor in the

Universal Soil Loss Equation. Soil Sci. Soc. Am. J., 50: 1294-1298.

Objectives

To simulate surface runoff, sediment and nutrient (N, P) yield using the event-based modified

AGNPS model (modified Agricultural Non-Point Source Pollution Model) and validate model

simulations in four watersheds showing contrasting landscape characteristics.

Study Areas

Glonn watersheds (G1 and G2) (drainage area: 1.2 and 1.6 km², respectively)

Weiherbach Watershed (drainage area: 3.5 km²)

Salzboede Watershed (size: 81.7 km²)

Location and size of study areas

Watershed characteristics

Linkage Between GIS input data and AGNPSm

Spatial data were stored and manipulated using SPANS geographic information system (GIS).

An interface was coded in C++ to link spatial input data (land use, soils, topography, climate)

to the AGNPSm.

Modeling framework

Entropy / Sensitivity Analysis

An entropy analysis was conducted in order to evaluate the degree of heterogeneity/homogeneity

of spatial natural resources. A moving window technique was used to calculate entropy for soils,

land use, elevation, and slope utilizing different grid sizes. We evaluated the loss of information

associated with data aggregation. Topographic data showed the largest entropy and therefore

the largest heterogeneity. Topographic data were most sensitive to information loss when aggregated.

A spatial sensitivity analysis was conducted to assess sensitivity of different grid sizes to simulation

output. Sediment yield showed large, peak flow moderate and surface runoff small sensitivity to

grid size variation.

AGNPSm Simulation Results

Simuation output was compared to measured data and evaluated using the Coefficient of Efficiency

(E) by Nash and Sutcliffe (1970). Simulation of the hydrological routines in watershed G1 (Glonn)

provided highly satisfactory results. Runoff volume showed an E of 0.96 for 19 flood events (validation).

Peak flow yielded an E of 0.84. Modifications to the sediment routine allowed the E of 0.26 (AGNPS)

to be raised to 0.90 (AGNPSm). Nutrients attached to sediment was satisfactorily calculated with an

E of 0.71 (P in sediment) and 0.79 (N in sediment), respectively. Simulations for dissolved nutrient

transport (an E of 0.40 for P and an E of 0.60 for N) were poor.

The results calculated for watershed G2 (Glonn) were similar to those obtained for watershed G1

(Glonn). Runoff volume was simulated with an E of 0.83, peak flow rate with an E of 0.82. Sediment

transport had an E of 0.72, and nutrients attached to sediment had E's of 0.64 (P) and 0.40 (N).

Simulation of dissolved nutrient transport was unsatisfactory, with Es of -1.92 (P) and 0.13 (N).

In Weiherbach watershed E for runoff volume was 0.88 and 0.36 for peak flow.

In Salzboede watershed, grid-based precipitation input (rN) was tested via sensitivity analysis with

varying synthetic precipitation fields. A comparison was also made between uniform and grid-based

precipitation inputs to the outputs surface runoff, peak flow, and sediment yield. Runoff volume yielded

an E of 0.87 (rN) and an E for peak flow of 0.57 (rN). The E for sediment delivery was 0.49.

References

Nash J.E., and J.V. Sutcliffe. 1970: River flow forecasting through conceptual models - Part I:

A discussion of principles. J. of Hydrology, 10: 282-290.

Validations results:

surface runoff & peak flow rate

sediment delivery

phosphorus and nitrogen

Simulation results for the rainfall-runoff event on Sept. 29, 1981 in Glonn G1 Watershed:

Simulation results for the rainfall-runoff event on Nov. 22, 1984 in Salzboede Watershed:

precipitation
Predicted surface runoff using uniform precipitation input	Predicted surface runoff using grid-based precipitation input
Predicted sediment delivery using grid-based precipitation input	Deviations in sediment delivery between grid-based and uniform precipitation input
Predicted particulate phosphorus using grid-based precipitation input	Deviations in particulate phosphorus between grid-based and uniform precipitation input

Conclusions

AGNPSm predictions for runoff volume, peak flow, and sediment yield were robust in four different

watersheds with contrasting landscape characteristics and size. Simulation of dissolved nutrient

transport was unsatisfactory.

Publications

Grunwald S. and H.-G. Frede. 2000. Application of modified AGNPS in German watersheds

(book chapter pp. 43-58). In: Schmidt J.‚ Application of Physically-Based Soil Erosion Models,

Springer, Berlin, New York.

Grunwald S. and L.D. Norton. 2000. Calibration and validation of a non-point source pollution

model. Agricultural Water Management, 45: 17-39.

Grunwald S. and L.D. Norton. 1999. An AGNPS-based runoff and sediment yield model for

two small watersheds in Germany. Transactions of the ASAE, 42(6):1723-1731.

Grunwald S., and H.-G. Frede. 1999. Using AGNPS in German watersheds. Catena, 37(3-4):

319-328.

Chaubey I., C.T. Haan, J.M. Salisbury, and S. Grunwald. 1999. Quantifying model output

uncertainty due to spatial variability of rainfall. J. of American Water Resources Association,

35(5):1113-1123.

Chaubey I., C.T. Haan, S. Grunwald, and J. M. Salisbury. 1999. Uncertainty in the model

parameters due to spatial variability of rainfall. J. of Hydrology, 220: 48-61.

Grunwald S. 1998. AGNPS (Agricultural Non-Point Source Pollution Model). Wiener Mitteilg.

Wasser - Abwasser - Gewaesser - Experiences with Soil Erosion Models, 151: 77-88.

Grunwald S. and H.-G. Frede. 1998. Application of AGNPSm in German Watersheds. Wiener

Mitteilg. Wasser - Abwasser - Gewaesser - Experiences with Soil Erosion Models, 151: 183-189.

Frede H.-G., S. Haverkamp, S. Grunwald, and N. Fohrer. 1998. Assessment of MEKA subsidies

for soil and water protection by AGNPS model. World Congress of Soil Science, Montpellier,

France, Aug. 20 - 26., Symposium 31, No. 2044.

Grunwald S. 1997. GIS based simulations of non-point source pollution using AGNPSm model.

Ph.D. thesis, Department of Natural Resources Management, University of Giessen, Germany.

Haverkamp S., S. Grunwald, and H.-G. Frede. 1997. Erhoehter Erosionsschutz und verminderter

Naehrstoffaustrag durch das MEKA-Foerderprogramm. Mitteilg. der Deutschen Bodenkundlichen

Gesellschaft (German Soil Science Society), 85, III: 1447-1452.

Grunwald S., S. Haverkamp, M. Bach, and H.-G. Frede. 1997. Assessment of MEKA subsidies

for soil and water protection by AGNPSm model J. of Rural Engineering and Development,

38(6): 260-265.

Chaubey I., C.T. Haan, J.M. Salisbury, and S. Grunwald. 1997. Effect of spatial variability of rainfall

on modeling hydrologic/water quality processes. ASAE Annual Int. Meeting, Minneapolis, MN,

Aug. 10-14, 1997. Paper No. 972099.

Grunwald S., and H.-G. Frede. 1996: GIS-based modeling of water quality using AGNPS. In:

Dollinger F., Strobl J. (ed.): Salzburger Geographic Materials - Applied GIS VIII, AGIT

Symposium, 3.-5. Juli 1996, 24: 231-236.

Rode M., S. Grunwald, and H.-G. Frede. 1995: Modeling of water quality using AGNPS and GIS.

J. of Rural Engineering and Development, 36(2): 63-68.