ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS’’, Bangkok, May 23-25, 2001 
130 
2.3 Execution of AGNPS 
The entire Dowagiac River Watershed is divided into 4803 cells, 
with the size of each cell equal to 40 acres (16 ha). Digital 
elevation data at a scale of 1: 250,000 were acquired from the 
U.S. Geological Survey. State Soil Geographic Data Base 
(STATSGO) was obtained from the U.S. Department of 
Agriculture Natural Resource Conservation Service. The 
watershed boundary was acquired from the Michigan 
Department of Environmental Quality Hydrologic Studies Unit. 
Land use/cover data of 1978-1981, the only data available for 
the study area, were obtained from the Michigan Resource 
Information System (MjRIS) for the entire Dowagiac River 
Watershed. 
The single storm event chosen in the model was a 24-hour 
rainfall of 4.5 inches with an exceedance probability of 4 
percent (recurrence interval of 25 years) from the Dowagiac 
Weather Station (Sorrel and Hamilton 1990). In the Dowagiac 
River Watershed, this type of storm usually occurs in the 
summer months (June through August). 
The AVNPSM was used to derive all the 22 input parameters 
and generate an input file for the AGNPS model. AGNPS 
Version 5.0 was run to produce estimates of surface runoff, soil 
erosion and sediment, and nutrient loadings. 
3.SIMULATION OF LAND USE CHANGE SCENARIO AND ITS 
IMPACT ON WATER QUALITY 
Residential and commercial developments are reportedly 
considered as the two most important land use issues in the 
Dowagiac River Watershed according to a joint report by the 
Cass County Conservation District, Michigan Department of 
Environmental Quality and Southwestern Michigan Commission 
(1998). Several townships are reported to .likely have 
population growth of more than 10 percent during the period of 
2000-2010. Accompanying the projected population growth 
comes increasing demands for residential housing and 
commercial services. These demands will likely lead to 
conversion of some agricultural, forest, and nonforest vegetation 
(grassland) to residential and commercial lands. To help 
resource planners and managers understand the potential 
impacts of these developments on water quality in the Dowagiac 
River, a land use change scenario is simulated and its water 
quality effect is analyzed by using the AGNPS model. 
The simulated land use change scenario is to convert the 
northwest corner of the La Grange Township to residential and 
commercial developments because it is adjacent to the City of 
Dowagiac and very likely to face residential and commercial 
expansion issues in the next decade. The selected area is about 
1,400 acres (560 ha) and includes urban, agricultural, nonforest 
vegetation, forest, and forested wetland. Conversion of this area 
to urban land represents an expansion of the current urban land 
by about 1240 acres. Assuming all other factors remaining 
same, the AGNPS model was run to simulate the effect of the 
changed land use on water quality and the result is summarized 
in Table 1. 
As shown in Table 1, conversion of about 1,240 acres of non- 
urban land to urban land is simulated to cause little increase in 
the peak flow rate and sediment yield at the mouth of the river 
but has almost no impact on runoff volume and nutrient yields 
(Fig.2). This is probably due to the fact hat the size of the 
proposed land use change was too small to have any impact at 
the watershed level. If more land in the watershed is converted 
to urban uses, it might lead to greater changes in discharge and 
sediment yield at the outlet of the watershed. In addition, the 
hydrologic effects are related to the types of land use change 
and the locations of the changes. The simulated land use 
change is in the northwest corner of the La Grange Township, 
adjacent to the City of Dowagiac. If the land use changes were 
to occur in areas adjacent to a waterbody such as a lake or 
river, 
Table 1. Summary of the Impact of the Simulated Land Use 
Change in La Grange Township on Runoff, Sediment, and 
Nutrients at the Outlet of the Dowagiac River Watershed from a 
Single Storm Event of 4.5 in. 
Variable 
Simulated 
Result 
Using the 
Baseline 
Data 
Simulated Result Using 
the Proposed Land Use 
Scenario 
25-Year Recurrence, 
24-hour Storm (inches) 
4.50 
4.50 
Surface Runoff (inches) 
1.82 
1.82 
Peak Flow Rate (cfs) 
3,231 
3,235 
Total Sediment Yield 
(tons) 
9,985 
9.992 
Total N in Sediment 
(Ibs/acre) 
0.85 
0.85 
Total P in Sediment 
(Ibs/acre) 
0.43 
0.43 
greater impact on water quality would have produced. In 
simulating the effect of urban land expansion on water quality, 
this study assumed the land cover is the only variable to be 
changed and all other factors such as soil, topography, and 
management practices are to remain unchanged. In reality, 
urban development often causes changes in some related 
factors such as topography and landscape irrigation. The 
simulated magnitude of the land use change impact on water 
quality should be evaluated using the field data in order to 
support the informed land use decision making process. 
Resource planners and decision makers should also realize that 
the location of the land use change such as proximity to a 
waterbody has significant impact on water quality and need to 
consider this factor in dealing with the land use issues in the 
Dowagiac River Watershed. 
4.SUMMARY AND CONCLUSIONS 
This study integrates AGNPS and digital databases of soil, land 
cover/use, topography, water resource features and 
management practices to simulate the impact of land use 
change scenario on runoff, sediment, and nutrient yields based 
on a 25-year, 24-hour period of single storm event of 4.5 inches 
in the entire Dowagiac River Watershed. The simulation results 
show that expansion of urban land will probably lead to an 
increase in surface runoff, peak flow, and soil erosion. The 
magnitude of the effect is related to the extent and proximity of 
proposed land use change to water bodies in the watershed. 
AGNPS requires 22 input parameters. Manual input of these 
input parameters for each of the 4,803 cells in the Dowagiac 
River Watershed would be time consuming, tedious, and 
problematic. This study develops AVNPSM, an interface 
between ArcView and AGNPS to derive, analyze, and visualize 
the required model parameters and simulated results from the 
databases of soil, topography, land cover, and water resource 
features. The interface consists of parameter generator, input 
file processor, model executor, output visualizer, statistical 
analyzer, and land use change simulator. Application of the 
interface to the study watershed indicates that it is user friendly, 
and robust, and significantly improves the efficiency of the 
nonpoint source pollution modeling process. With the interface, 
land use change scenarios can be readily explored in the model 
to help resource planners and decision makers develop 
watershed management plan to minimize sedimentation and 
nutrient loading to the receiving waters.