Using GIS-based multi criteria decision analysis for prioratizing drinking water supply needs: an application to southern region in Sri Lanka

Abstract

GIS-based Multi-Criteria Decision Analysis (MCDA) was carried out to prioritize drinking water supply need in the Southern Region, Sri Lanka. The degree of water demand is attributed to population density, water resources availability (Relative Water Balance), water quality (Equivalent Water Quality Index) and new development (Population Pressure) of the region. Spatial distribution of density of Grama Niladhari Division-wise (GND) population was obtained from census data of 2,131 GNDs; assigned to the spatial distribution domestic units available with Survey Department. Water resources availability of each GND was figured out by plotting relative water balance with the aid of average annual rainfall data from the Meteorological Department, Mean evaporation rate and inflow of rivers into the designated research area. Water quality of relevant GNDs was calculated and represented as Water Quality Index (WQI) with the aid of geochemical atlas produced by Dissanayake et al. (1985) and NWSDB treated water quality data. Moreover, the new development of the research area was symbolized with the development corridor stipulated under National Physical Plan for 2050. Four raster files were prepared. To weight criteria, Analytic hierarchy process (AHP) was adopted. To identify the prioritized patches, the weighted linear combination of raster shapefiles was derived as a method of compensatory aggregation. It was understood that a combination of water quality and population density was dominant in the prioritization of water supply needs. The model identified the values as 0.47, 0.39, 0.07 and 0.07 for the main criteria, respectively for relative water balance, equivalent Water Quality Index, population density and population pressure owing to new development. The output of the research was the preparation of an automated GIS tool for prioritization of water supply needs through the GIS model builder facility.