Runoff loads of pollutant in agricultural watersheds were spatially analyze
d by using geographic information system(GIS) technology. The topological r
elationship between pollution sources in the watershed was, first of all, i
dentified by using the developed digital map of land use and then the pollu
tant loads generated from each source was estimated by applying a conventio
nal unit loading factor on the obtained digital information of pollution so
urces. To evaluate the loads delivered from spatially distributed pollution
sources to monitoring stations in down stream via surface of watershed, a
renovated empirical model incorporated with the information of pollutant di
scharge path was developed through introducing a digital terrain model(DTM)
technique. In this model, the function of degradation of pollution loads d
uring delivery process was simplified so that each watershed could have a b
asin-wide self-purification capacity which would be considered to be posses
sed inherently in each watershed and could retard the discharge of pollutan
ts from sources generated to stream water. Model credibility showed good co
nsistency with comparing the simulated values with observed data Monte Carl
o optimizing technique made it possible to estimate the basin-wide self-pur
ification coefficients. (C) 1998 Published by Elsevier Science Ltd. All rig
hts reserved.