DELINEATION OF SALT-AFFECTED SOILS AND WATERLOGGED AREAS IN THE INDO-GANGETIC PLAINS USING IRS-1C LISS-III DATA

Waterlogging and subsequent salinization and/alkalinization is the maj or land degradation problem in irrigated commands in the arid and semi -arid regions. The Linear Imaging Self-scanning Sensor (LISS-III) digi tal data acquired by the Indian Remote Sensing Satellite (IRS-1C) over part of the Indo-gangetic plains are digitally analysed for delineati ng salt-affected and waterlogged areas on a IBM RS-6000 system. A comp arison with the IRS-1B LISS-II data, with comparable spectral bands an d acquired very close to LISS-III data, reveals a relatively poor over all classification accuracy achieved from this data as compared to LIS S-II data. This may be attributed to higher intra-class spectral varia tions resulting from high spatial resolution LISS-III data. The classi fiers capable of exploiting the spectral variations in spatial domain, like segmentation, textural analysis, etc., need to be used for explo iting the full potential of IRS-1C LISS-III data for mapping salt-affe cted soils and other degraded lands.