Digital Soil Mapping Using Multispectral Modeling with Landsat Time Series Cloud Computing Based

Geotechnologies allow natural resources to be surveyed more quickly and cheaply than traditional methods. This paper aimed produce a digital soil map (DSM) based on Landsat time series data. The study area, located in the eastern part of Brazilian Federal District (Rio Preto hydrographic basin), comprises representative basin Central Brazil plateau terms pedodiversity. A spectral library was produced spectroscopy (from visible shortwave infrared range) 42 samples from 0–15 cm depth using Fieldspec Pro equipment laboratory. Pearson’s correlation principal component analysis attributes revealed that dataset could grouped texture content. Hierarchical clustering allowed for extraction 13 reference spectra. We interpreted spectra morphologically resampled them 5 Thematic Mapper satellite bands. Afterward, we elaborated synthetic soil/rock image (SySI) frequency (number times bare captured) (1984–2020) Google Earth Engine platform. Multiple Endmember Spectral Mixture Analysis (MESMA) used model SySI, endmembers as input generating DSM, which validated by Kappa index confusion matrix. MESMA successfully modeled 9 endmembers: Dystric Rhodic Ferralsol (clayic); (very clayic); Haplic (loam-clayic); Petric Plinthosol Regosol Dystric, Cambisol (clayic). root mean squared error (RMSE) varied 0 1.3%. accuracy DSM achieved 0.74, describing methodology’s effectiveness differentiate studied soils.