Improving remote sensing of salinity on topsoil with crop residues using novel indices of optical and microwave bands

Remote sensing indices have been proposed to characterize soil salinity. However, the sensitivity of these indicators is unstable owing differences in geographic environment and vegetation type. This study investigated performance several existing estimate salinity topsoil with residues southern Xinjiang, China. The results showed that were not satisfactory. In order construct an index can be used directly indicate a specific area, novel calculated using optical bands (blue, green, red, red edge, shortwave infrared bands) derived from Sentinel-2 multispectral data Sentinel-1 radar (backscattering coefficient VV, VH). To enhance bands, five transformation methods (logarithmic, reciprocal, first-, second-, third-derivative) applied original spectra. Based on previous studies, statistical two-, three-, four-bands indices. One constructed three-bands second-derivative transformation, called Enhanced Residues Soil Salinity Index (ERSSI), highest correlation (r = 0.65 0.68 training testing). ERSSI establishes linear relationship estimation R2 0.53 LCCC dataset, 0.51 0.73 testing dataset. And it shows contribution random forest regression 0.80 0.86 0.77 0.81 consisted B, G, SWIR1 was sensitive variations remaining farmland soils. provides method for accurate robust assessment mapping covered by crop residues.