Satellite Multi-Sensor Data Fusion for Soil Clay Mapping Based on the Spectral Index and Spectral Bands Approaches

Integrating satellite data at different resolutions (i.e., spatial, spectral, and temporal) can be a helpful technique for acquiring soil information from synoptic point of view. This study aimed to evaluate the advantage using mono- multi-sensor image fusion based on either spectral indices or entire spectra predict topsoil clay content. To this end, multispectral images acquired by various sensors Landsat-5 Thematic Mapper (TM), Landsat-8 Operational Land Imager (OLI), Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER), Sentinel2-MultiSpectral Instrument (S2-MSI)) have been used assess their potential in identifying bare pixels over an area northeastern Tunisia, Lebna Chiba catchments. A index bands are generated each sensor TM, OLI, ASTER, S2-MSI). Then, two fusions generated, one other bands. The resulting band mono-and satellites compared through patterns ability content Multilayer Perceptron with backpropagation learning algorithm (MLP-BP) method. results suggest that prediction: (i) bands’ outperformed regardless sensor; (ii) fused derived provided best performances, 10% increase prediction accuracy; (iii) obtained many more beneficial than mono-sensor images. Soil maps elaborated via might become valuable tool survey, land planning, management, precision agriculture.