ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS”, Bangkok, May 23-25, 2001 
ISPRS 
128 
MODELING LAND USE EFFECT ON URBAN STORM RUNOFF AT THE WATERSHED SCALE 
Chansheng HE 
Department of Geography 
Western Michigan University 
Kalamazoo, Ml 49008-5053, U.S.A. 
KEYWORDS: AGNPS, GIS-interface, simulation, and nonpoint source pollution. 
ABSTRACT 
Assessment of land use impact on water quality at the watershed scale is essential to watershed management. This paper integrates 
GIS and AGNPS (Agricultural Nonpoint Source Pollution Model), a computer simulation model to analyze the effect of land use change 
on nonpoint source pollution in a study watershed. ArcView Nonpoint Source Pollution Modeling (AVNPSM), an interface between 
ArcView GIS and AGNPS is developed in support of agricultural watershed modeling. The interface is PC-based and operates in a 
Windows environment. It consists of five modules: a parameter generator, input file processor, model executor, output visualizer, and 
statistical analyzer. Basic input data to the interface include: soil, digital elevation model, land use/cover, water features, and 
management practices. Application of the AVNPSM to the study watershed indicates that it is user friendly, and robust, and 
significantly improves the efficiency of the nonpoint source pollution modeling process. 
1. INTRODUCTION 
The Dowagiac River is a major tributary of the St. Joseph River 
in southwestern Michigan, U.S.A. Its total drainage area is 
73,435 hectares. Agriculture is the major land use in the 
watershed, accounting for 61 percent of the total land. Corn, 
soybeans, hay, and wheat are the major crops, representing 98 
percent of the harvested crop acreage (Michigan Agricultural 
Statistics Service 1994; 1995; 1996). 
In recent years, local communities and the Michigan 
Departments of Environmental Quality and Natural Resources 
have formed a MEANDRS (Meeting Ecological and Agricultural 
Needs within the Dowagiac River System) group to restore and 
enhance the ecological and agricultural capacity of the 
Dowagiac River Watershed. Implementation of this goal requires 
assessment of land use and its impact on water quality and 
identification of critical area for targeting the management 
practices in the entire watershed. This paper describes the 
function of ArcView Nonpoint Source Modeling (AVNPSM), a 
Windows-based interface in facilitating watershed modeling and 
its application in analyzing the effect of the land use change on 
water quality of the Dowagiac River Watershed and identifying 
areas with high potential of surface runoff, sediment, nitrogen 
(N), phosphorus (P), and chemical oxygen demand (COD) yield. 
2. ANALYSIS PROCEDURE 
A watershed is a hydrologic system that consists of all the 
tributaries and drainage areas from upstream to its mouth. 
Analysis of a watershed must consider all the interacting factors 
in the entire watershed. In assessing the land use impact on 
nonpoint source pollution in the Dowagiac River Watershed, 
these factors include precipitation, topography, land cover, soil, 
hydrography, and management practices. In order to integrate 
all these factors in this project, a computer simulation model, 
Agricultural Nonpoint Source Pollution Model (AGNPS) (Young 
et al. 1987) is used to estimate soil erosion, sediment loading, 
and nutrient (N, and P) yields and to identify the critical areas in 
the Dowagiac River Watershed. A unique feature of the AGNPS 
is that it links the upper reaches to the downstream reaches in 
simulating the hydrologic and sediment transport processes and 
makes it possible to assess the effect of upstream land use 
practices on the water quality of the river downstream. 
2.1 Model Description 
AGNPS (Version 5.0) is a single storm-event based simulation 
model for evaluating sediment and nutrient transport from 
agricultural watersheds (Young et al. 1987; USDA 
Agricultural Research Service 1995). The model includes three 
basic components: hydrology, erosion and sediment, and 
nutrients (N, P, and Chemical Oxygen Demand). The hydrologic 
component calculates overland runoff (in inches) and peak flow 
rate (in cubic feet per second ) based on the SCS (Soil 
Conservation Service) curve number equation. 
Km, w • SS (1) 
(Sfornt Mfall + 0.8* MnicH Factor) 
Where storm rainfall represents total rainfall from a storm in 
inches, and retention factor is calculated by Eq. 2: 
Retention Factor 3 — - 10 (2) 
Curve Number 
Where the curve number is the SCS Curve Number, which is 
related to soil and land use factors. 
Upland erosion is computed based on the Universal Soil Loss 
Equation (USLE): 
Where, A is the computed average soil loss per unit area, 
expressed in tons/acre; R is the rainfall and runoff factor and is 
the number of rainfall erosion index (El) plus a factor for runoff 
from snowmelt or applied water; K is the inherent erodibility of a 
particular soil; L is the slope-length factor, S is the slope- 
steepness factor; C is the cover and management factor; P is 
the support practice factor; and slope shape factor represents 
the effect of slope shape on soil erosion (1.0 for uniform slope 
shape, 1.3 for convex slope shape, and 0.88 for concave slope 
shape) (Wischmeier and Smith 1978). 
In the nutrient component, AGNPS divides nutrient transport into 
the soluble nutrients, which are transported in the runoff, and 
the sediment nutrients, which are transported in the sediment. 
The soluble nutrients (N and P) are the amount of initial soluble 
N and P in the top 1 cm of soil prior to the rainfall event in 
Ibs/acre (or kg/ha). Sediment attached nutrient is the amount of 
nutrient (N or P) contained in the sediment. 
AGNPS works on a cell basis. The amount of nutrients (N and P 
and COD) is calculated for each cell. The nutrients from the 
cells flowing into the current cell are added to the amount that 
was generated within the current cell. The calculated amount is 
decayed as it runs through the channel to get the amount 
A = R*K*LS*C*P* Slope Shape Factor (3) 
rema 
routin 
and 
(measi 
compai 
within 
manag' 
AGNPi 
numbe 
cell 
slope 
Mann 
(12) 
factor 
it take 
impou 
(indicati 
1995) 
cons 
Outp 
(inch 
sedir 
sedir 
chemic 
viewed 
runoff. 
2.2 
Analys 
waters 
for eac 
is oftei 
issue 
such 
and 
1997) 
study 
(AVNI 
AGNF 
Analy: 
ArcV 
The 
datafc 
as w; 
crop 
STAI 
factor 
slope 
use/c< 
mana' 
fertiliz 
datab 
proce 
event 
the 
types 
files 
ArcVi< 
requir 
Once 
requir 
AGNF 
AGNF 
analy