Combining Proximal and Remote Sensors in Spatial Prediction of Five Micronutrients and Soil Texture in a Case Study at Farmland Scale in Southeastern Brazil

Despite the increasing adoption of proximal sensors worldwide, rare works have coupled with remotely sensed data to spatially predict soil properties. This study evaluated contribution and texture available contents micronutrients using portable X-ray fluorescence (pXRF) spectrometry, magnetic susceptibility (MS), terrain attributes (TA) via random forest algorithm. Samples were collected in Brazil from soils high, moderate, low weathering degrees (Oxisols, Ultisols, Inceptisols, respectively), analyzed by pXRF MS for micronutrients. Seventeen TA generated a digital elevation model 12.5 m spatial resolution. Predictions made via: (i) TA; (ii) + pXRF; (iii) MS; (iv) (v) (vi) validated root mean square error (RMSE) coefficient determination (R2). The best predictions achieved by: dataset alone Cu (R² = 0.80) clay (R2 0.67) content; Fe 0.68) sand 0.69) Mn 0.87) content. PXRF key predictions. Soil property maps created these supported sustainable management practices.