183 
1SPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS”, Bangkok, May 23-25, 2001 
A GIS-BASED ENVIRONMENTAL DECISION SUPPORT SYSTEM FOR THE ERHAI LAKE WATERSHED 
MANAGEMENT 
Lei LIU \ Gordon HUANG 1 , and Jonathan LI 2 
1 Environmental System Engineering Program, Faculty of Engineering, University of Regina, Regina, Sask., S4S 0A2 Canada 
Tel: (306) 585-5631, Fax: (306) 585-4095, E-mail: {liulei, gordon.huang}@uregina.ca 
2 Department of Geography, University of Regina, Regina, Sask., S4S 0A2 Canada 
Tel: (306) 585-5273, Fax: (306) 585-4815, E-mail: jun.li@uregina.ca 
Keywords: decision support system, environmental modeling, GIS, watershed management. 
ABSTRACT 
In this paper the integration of environmental models and GIS is exploited. The focus is placed on the development of a GIS-based decision 
support system for the lake watershed management. The system consists of a GIS database, optimization models, and a user-friendly 
interface. ArcView GIS is used to enter and compare data from different sources and formats, assess data availability and quality, and 
identify potential data errors. A centralized database with a structure that allows efficient storage and retrieval of georeferenced time-series 
data is developed before the water quality simulation and system optimization processes. The database management focuses on the 
attributes and data necessary to run the simulation and optimization models. An inexact-fuzzy multiobjective linear programming (IFMOP) 
model is developed to form an environmental decision support system, in association with a number of simulation/evaluation tools. An 
interactive approach is proposed for conveniently obtaining indispensable intervention from decision-makers during the IFMOP modeling 
process. The IFMOP allows uncertainties to be directly communicated into the programming processes and resulting solutions. Its inexact 
solutions can be interpreted for generating decision alternatives and conducting further risk analyses. Also, the IFMOP solution approaches 
do not lead to complicated intermediate submodels, and thus have reasonable computational requirements. The integrated approach 
incorporates digital data, models, and system users into a general framework through developing a user-friendly interface. It can generate 
proper input data files automatically for each model. This component provides a two-way communication between the system and the user. 
Thus, the user may interactively delineate an area of interest, identify contamination sources to be considered, add additional data, or 
specify a particular planning objective. Meanwhile, the system explains to the user about each step in the modeling process, displays results 
from running simulation and optimization models. The system provides the user an evaluation of the quality of data, accuracy of the result, 
and level of uncertainty. If the user is not satisfied with the results from available data, the system can recommend to the user what data are 
needed to improve the modeling performance. Based on the integration of geospatial information technology and environmental models, this 
paper represents a unique contribution to regional planning and environmental system engineering for the methodology of integrated 
environmental decision support. The methodology allows more powerful manipulation of extensive data, more realistic expression and 
communication of system activities. It is directly useful for generating, presenting, and evaluating planning alternatives, which can provide 
general and consistent approaches for the generation of cost-effective and sustainable environmental management solutions. 
INTRODUCTION 
Human activities continuously have serious impacts on 
watershed systems due to economic development and 
population growth, leading to a series of environmental 
problems, such as deterioration of water quality, extinction of 
aquatic species, alternations of river flows, shortage of water 
resources, and so on (GAO, 1989; Master, 1990). The impact of 
urbanization is especially pernicious, as it is "cumulatively 
changing the dynamics of not just stream reaches over periods 
of years but entire watersheds and landscapes over many 
decades or centuries" (Dopplelt et al., 1993). 
Watershed management is related to a number of social, 
economical and environmental factors. Many of these factors 
have complicated interrelationships between each other, and 
may vary temporally and spatially with dynamic features 
(Dowlatabadi et al., 1993). For example, variations of 
environmental and socio-economic conditions may lead to 
conflicts between agricultural and industrial activities, and may 
need compromises among different stakeholders in order to 
obtain an overall optimal use of land and water for the entire 
basin; pollution from different human activities in water or on land 
may affect water quantity and quality, resulting in impacts on 
diversity/vitality of aquatic biota and ecological processes; prices 
of agricultural products may affect the planning of crop 
production levels; and expansion of agricultural production may 
have impacts on forest cover (and thus timbering production) 
due to land use conflicts. Consequently, integrated modeling 
approach that incorporates individual system components within 
a general framework instead of examining or presenting them in 
isolation may be useful for providing holistic and comprehensive 
analysis of a variety of system activities, as well as relevant 
policy responses for the sustainability of a water resources 
system (Huang 1995a and b). 
Since watershed management is related to a number of land use 
concerns, effective reflection and presentation of the spatial 
variations are critical for not only implementation of the modeling 
results but also procession of the related computational 
processes. Therefore, a systematic approach that can 
incorporate watershed modeling and geographic information 
system (GIS) technology within a general framework is desired 
for effectively reflecting the interactive and dynamic features of 
watershed systems. 
Previously, many studies of watershed modeling for individual 
system components have been reported (Haimes et al., 1980; 
Haimes, 1984; Haith 1987; Trezos et al., 1987; Gorelick, 1990; 
Kindler, 1992; Huang, 1996;). There are also some reports on 
the use of GIS for obtaining data and presenting modeling 
outputs (Brail, 1990; Grayman et al., 1993; Liao et al., 1994). 
This study is an extension of the previously efforts, emphasizing 
on the development of an integrated modeling-GIS approach that 
can (i) incorporate a variety of system components and the 
related submodels within a general framework, (ii) present and 
quantify system information dynamically with GIS throughout the 
modeling processes, and (iii) incorporate uncertainties in human 
decision-making processes regarding environmental 
management and planning. In detail, an integrated watershed 
decision support system - GISWMS (Hybrid GIS-Supported 
Watershed Modeling System) will be proposed. It will then be