REMOTE SENSING AND GIS BASED DECISION SUPPORT SYSTEM FOR DISTRICT LEVEL PLANNING

Abstract

Planning has always been an integral part of national policy. District level planning specifically deals with the problems typical to its terrain, natural resources, socio economic conditions and demographic set-up. It requires timely, up-to-date and reliable information at spatial as well as non-spatial level. Satellite remote sensing has become one of the inevitable techniques to provide timely and reliable data for planning. Geographical Information System (GIS) provides a useful tool for the integration and analysis of multithematic information required for making decisions. A DSS expands the analytical powers of GIS through linking of planning models to a geographic database. In the present study, a methodology has been developed for district level planning using remote sensing and ancillary data in GIS. Dehradun district has been selected for the detailed investigations. It comprises of six community development blocks, i.e., Chakrata, Kalsi, Vikasnagar, Sahaspur, Raipur and Doiwala. The detailed objectives of the work are (i) design of an integrated geographic database for regional planning at district level, (ii) development of a GIS-based statistical model for the analysis of intra-district disparities, (iii) development of a rule-based model for spatial planning of educational and medical facilities in the least developed block, (iv) remote sensing based assessment of land utilisation and development of a GIS-based approach for agricultural land suitability, (v) development of a DSS by incorporating objectives (ii) to (iv). An integrated geographic database, consisting ofspatial as well as non-spatial data, has been created in ARC/INFO GIS at 1:250,000 scale by taking village as the basic spatial unit. The spatial elements ofthe database include various thematic maps describing administrative and village boundaries, drainage, transportation network, land use/land cover, soil types, contours, ground water potential and rainfall intensity. The non-spatial elements consist of data related to education, medical, post office, approach road to village, power supply, market/shopping centre, literacy rate and irrigated land area. The education facility has been considered at the levels of primary, middle and high schools, pre-university college and adult education centre. The medical facilities include Primary Health Sub-centre (PHS), Primary Health Centre (PHC), health centre, child welfare centre, maternity and child welfare centre, maternity house, family planning centre, n dispensary, hospital and registered medical practitioner. Aset of design elements has been adopted to maintain the accuracy and integrity of these databases. The intra-district disparities analysis has been carried out to identify the least developed block in Dehradun district by developing a GIS-based statistical model and several programs in Arc Macro Language (AML). To reflect the relative development of each block. Block Development Index (BDI) has been computed incorporating various existing facilities and assigning weight to each facility through median population threshold. Further, the relative weights for obtaining BDI have been computed using the statistical technique, based upon principal component analysis. This statistical model has been compared with a conventional model for assessing the intra-district disparities and Chakrata block is found to be the least developed block. A rule-based spatial model has been developed for the least developed Chakrata block through AML programming to identify the villages that lack the facilities and the villages where new facilities need to be provided. The planning has been carried out for educational (viz., primary, middle and high schools) and medical (viz., PHS, PHC and dispensary) facilities using Planning Commission, Government of India norms, which are suitably modified. The spatial planning maps for the development of these facilities have been generated for Chakrata block showing high priority, middle priority and low priority villages. The high priority villages need immediate attention by the planners and policy makers. Assessment ofland utilization has been carried out using land use/land cover maps prepared from digital IRS data of 1988 and 1997 and the changes identified. These land use/land cover maps have been prepared by adopting supervised classification technique and using maximum likelihood classifier in ERDAS Imagine system. Various classes identified include forest, agriculture, water, built-up, sand, scrub and rocky/barren. It is observed that agriculture and forest cover about 24% and 51% of the area in Dehradun district in 1997. Further, agriculture has decreased by about 35 km2 and forest area by about 57 km2 over aperiod of nearly 9years from 1988 to 1997. The rocky/barren area has increased by about 53 km2. Thus, there is an increased pressure on agricultural and forest m land in the district, and it is desirable that appropriate measures are taken for the conservation of productive land and resources. A GIS-based Composite Land Development Units (CLDU) approach through AML also has been developed for carrying out agricultural land suitability analysis. A CLDU is a derived map unit for a land parcel with a degree of homogeneity in selected physical characteristics. The physical characteristics selected include soil, slope, ground water potential and rainfall, which are arranged in hierarchical order, and weights are assigned based upon their relative importance for agriculture to derive agriculture suitability index. These index values are suitably categorised to identify five land suitability classes in Dehradun district. The areas covered by class I (most suitable), class II (suitable), class III (moderately suitable), class IV (marginally suitable) and class V (not suitable) are nearly 372, 819, 352, 501 and 1053 km2 respectively. A GIS-based prototype DSS (acronym ARCDIST) for Dehradun district has been developed using AML. Its basic structure has three components, viz., (i) integrated geographic database, (ii) model base, consisting of above developed planning models, and (iii) user interface. The data display and analysis in ARCDIST is performed through various modules. The main menu consists of MAPS, MAP-FEATURES and ANALYSIS modules. Each module consists of several sub-modules, which in turn is composed of various menus for specific work. A user can get the required information from the database interactively. The structure of ARCDIST DSS is modular with a menu-driven graphical user interface to increases its acceptability among planners and decision-makers. It is expected that ARCDIST may help in improving the efficacy of planning process. It has been found that the role of remote sensing, GIS and DSS in various planning activities at district level is indispensable for scientific decision-making. With the reduced cost of hardware and software and availability of data on Internet in future, it is hoped that these techniques will become a part of our life for planning even day to day activities. iv