부산대학교 도서관

Characterizing soil salinity at the regional scale remains challenging despite decades of effort in soil mapping. Using satellite remote sensing, an effort has been made to identify the coastal saline soils in India. The study made use of the OLI sensor of the Landsat-8 satellite. The images were downloaded from the USGS EarthExplorer website. For multi-temporal studies, absolute radiometric calibration was done to minimize the impacts of changing atmospheric conditions, solar inclination and sensor view angle. Images were categorized using unsupervised classification, while a ground survey and Google Earth data were used for ground truthing. Three indices, namely normalized difference vegetation index (NDVI), salinity index (SI), and canopy response salinity index (CRSI), were used to identify soil salinity regions. For testing the vegetation index with soil salinity, 192 georeferenced soil samples from the Indian Sundarbans were collected. A relationship was developed between NDVI, SI and square of CRSI (CRSISQR) with ECe (electrical conductivity of saturation paste extract) and EC1:2 (1: 2; soil: water). For the coastal region, soils with CRSISQR < 0.16 were considered to be influenced by salinity since the relationship between ECe and CRSISQR had a maximum R² (0.50). It has been estimated that India has 12.94 lakh ha of saline soil within arable lands in all the coastal districts, and Gujarat (5.28 lakh ha), West Bengal (5.08 lakh ha), and Andhra Pradesh (1.06 lakh ha) were identified as the top three-salinity affected coastal states in the country. [ABSTRACT FROM AUTHOR]