Development of Novel Soil Salinity Spectral Index Using Remotely Sensed Data: A Case Study on Balod District, Chhattisgarh, India

Abstract Soil salinity is a known phenomenon worldwide. It has a substantial influence on crop productivity and environmental well-being. Conventional approaches to evaluate soil salinity are laborious and expensive, which need efficient approaches such as geospatial. Geospatial approaches have led to the development of several indices for soil salinity estimation. Existing soil salinity indices are region specific and not verified for different regions. This study was conducted in the Balod district of Chhattisgarh, India. Landsat 9 imagery along with field electrical conductivity (EC) were used to evaluate the existing soil index and develop a new soil salinity index. A Soil multi-parameter recorder was used to collect 69 EC samples for April and May 2024. Sixteen spectral indices were evaluated to verify the applicability in the study area. The results showed that the existing spectral indices had a weak correlation with field EC values. Therefore, we have developed the new index by combining the Near Infrared surface reflectance, redsurface reflectance, and Shortwave Infrared-1 surface reflectance bands and using a linear regression analysis.The soil salinity classification was used to categorize the new index. The results showed that 78.40 % of the region is slightly saline, 16.50 % is moderately saline and 1.46 % is strongly saline. This study demonstrates a strong correlation between reflectance values and field EC data with an R 2 value of 0.83 and a mean relative error of 10 %. This study provides a reliable geospatial approach for soil salinity evaluation and sustainable land management techniques to improve agricultural productivity in semi-arid, arid regions with varying soil properties and salinity levels.