Investigating remote sensing properties for soil salinty mapping: A case study in Korat province of Thailand

Abstract

The objective of this study is to develop a technique of utilizing satellite remote sensing data and soil electrical conductivity (EC1.5) for assessing soil salinity in northeast Thailand. Total of 30 soil samples were collected for laboratory analysis through a field survey conducted in January 2014 to match the approximate timing of remote sensing Landsat 8 OLI data. From each site, samples were collected from the surface layers (0–30 cm), which were analyzed for texture, pH and organic matter content (OM) by using electrical conductivity method. Several variables of original and ratio remote sensing spectral bands and topographical variable were considered to examine their relationship with the observed EC1.5 values with correlation and regression analysis. The model results were validated using HJ-1A satellite data. Analysis showed that although there were number of bands with moderate to high correlation with observed EC1.5, NIR, visible and NDSI were the significant predictors of the EC1.5 variations. The generated models were able to 23 to 53% of the observed EC variation. The models were found to be satisfactory as the tested model during the validation gave a R-squared coefficient of determination value of 0.87 for observed and model calculated EC values. The study indicates the possibility of using remote sensing data in the area where salinity data are scarce and conducting ground survey is resource demanding.