Abstract Decision Support Systems (DSS) are defined
as computer-based information systems designed to support decision makers
interactively in thinking and making decisions about relatively unstructured
problems. Spatial Decision Support Systems (SDSS) , which are the integration of
DSS and GIS was initiated by Densham and Goodchild ( 1988) are emerging as
efficient tools for managing natural resources like land and water. AVSWAT ( Arc
View- SWAT) , a user- friendly PC based SDSS tool has been developed at the
Black Land Research Center , Temple, Texas, USA integrating Soil and Water
Analysis Tool (SWAT) and Arc View GIS version 3.0a software along with Spatial
Analyst version 1.1 extension. SWAT is a continuous time river basin or
watershed scale model operating on daily time step. In the present study, the
tool was applied in digitally delineating watersheds in a block of Bankura
district of West Bengal and then it was used for estimating potential water
,silt and crop yield from each of them. This would be helpful in prioritising
the watersheds and presenting the results spatially for the district level
decision makers.
Key Words:
Spatial Decision Support System,
AVSWAT, Watershed Management.
Decision Support Systems (DSS) are
defined as computer -based information systems designed to support decision
makers interactively in thinking and making decisions about relatively
unstructured problems. Traditionally, DSSs have three major components, a
database, a model base and a user interface as depicted in Fig. 1a. An extension
of the DSS concept, Spatial Decision Support Systems (SDSS), which are the
integration of DSS and GIS (Fig. 1b) was initiated by Densham and Goodchild (
1988).
GIS is a
general purpose technology for handling geographic data in digital form, with
the ability to preprocess data into a form suitable for analysis, to support
analysis and modelling directly, and to post-process results (Goodchild,
1993).
A significant capability of the SDSS is the ability to use spatial
analysis and display tools with the sectoral models and that would form the
model base of SDSSs. The modeling capability allows the user of the SDSS to
simulate changes in objects and attributes. The database component of the SDSS
can supply input data for the models. After the models are run, the resulting
output can be written to the database for later display via the user interface ,
in tabular, chart or map form. For planning purposes, this ability to
dynamically change information, forecast and perform sensitivity analysis is
essential.
Both GIS and DSS have been widely used in natural resources
management. Watkins and Mckinney (1995) presented a review on DSS in water
resources; and Goodchild et al. (1996) described a comprehensive study of GIS in
water resources and environmental engineering. Singh and Fiorentino (1996) gave
a comprehensive review of GIS in hydrology.
In summary, SDSS provide
unique advantages for land and water resources management in the following
aspects:
- spatial representation, that is representing the spatial relations of the
real world in a visual and analytical form;
- comprehensive database, which is the basis for the integration of
socioeconomic, environmental and physical components of the real world; and
- modeling capability, which can integrate simulation/optimization techniques
to solve complex natural resources management problems.
These advantages make SDSS a proactive tool for sustainable natural resources
management.In the present study, while selecting an appropriate model for
developing Spatial Decision Support System .
for land and water
management, the following considerations were made :
- the ability to model at scales ranging from watershed to basin (regional)
scales
- the ability to model water quantity and quality, silt production and crop
growth
- the ease of use of the model and ready availability of inputs data
- the ability to link to Geographical Information System (GIS)
- the documentation and degree of support available.
Based on
the above considerations, the Soil and Water Assessment Tool (SWAT) model was
selected. The SWAT model is a combination of the SWRRB, GLEAMS and ROTO models
and hence it is able to model both the hydrology and water quality of a
watershed (Arnold
et al., 1995). The model is reported to be able to
operate on both a raster and sub-watershed (hydrologic response unit) basis
(Arnold
et al., 1995).In addition, the model is linked to GIS packages
like GRASS (Geographic Resources Analysis Support
System) via the
SWAT-GRASS interface (Srinivasan and Arnold, 1993) and ARC-View through the
SWAT-ARC View interface (Diluzio
et al, 1997), thus easing the task of
data input and output display.