Predicting Land Degradation Scenario Using Cellular Automata Model in Semi Arid Regions of Andhra Pradesh

Being a global phenomenon, land degradation needs to be assessed and its dynamics are to be studied for taking up any reclamative or preventive measures using state- of- the- art tools. Furthermore, equally important, is the projection of future land degradation scenario helps in formulating land use policies. Information on the spatial extent and land use pattern in conjunction with other driving factors like drainage pattern, soil texture, slope, extent of irrigated lands, aridity index, rainfall and population (both human as well as cattle) is of immense help in assessment of the vulnerability of a piece of land to degradation. Spaceborne spectral measurements hold a great promise in providing the information on land use pattern, and Geographic Information System (GIS) enables integrating it with other drivers of land degradation. The spatial-transition-based models facilitate projecting the future land degradation scenario. The study was taken up to project the land degradation scenario in part of Prakasham district of Andhra Pradesh, southern India. The approach involved studying the land use/land cover dynamics from multi-temporal Indian Remote Sensing satellite (IRS-1A/1B), Linear Imaging Self-scanning Sensor (LISS II) and IRS-1D LISS-III data for the period 1989 to 2002, vulnerability analysis of each land use/land cover category to land degradation, and projecting future land degradation scenario using cellular automata model. Validation of the model results from historical land degradation data indicated an accuracy of better than 77.6%.