Abstract:
Fine-scale soil spatial variability mapping is one of the prerequisites for adopting precision agriculture. This study used a hyperspectral radiometer and satellite remote sensing data for fine-scale surface soil texture and organic matter. A total of 626 surface soil samples (7cm depth) were collected and analyzed for soil texture and organic matter in the Lab. Multiple linear regression (MLR) statistics were used to relate soil spectral data derived from multispectral LandSat-8 OLI imagery data and hyperspectral remote sensing data of ASD FieldSpec Spectroradiometer with sand, silt, clay, and organic matter data. The MLR analysis of Multispectral data of Landsat-8 OLI satellite showed a significant relationship (p < 0.05) with band-5, band-7, and band-11 with sand% (R2 = 0.558), clay% (R2 = 0.589) and O.M.% (R2 = 0.687). The MLR analysis of hyperspectral remote sensing data of ASD Field Spec spectroradiometer showed a relationship with a different significant level of wavelength X1362, X1366, X1843, and X1856 with sand% (R2 = 0.370), wavelength X1830, X1839, X1873 and X1882 with clay% (R2 = 0.317) and wavelength X 1363, X1833, X1886 and X1909 with O.M.% (R2 = 0.440). A soil texture map of the entire study area was developed in GIS using the USDA-ARS soil texture triangle. The findings imply that remote sensing and geographical information system approaches might be employed to map the soil surface texture and O.M. over a wider area at a fine scale.
