Testing Sentinel-2 spectral configurations for estimating relevant crop biophysical and biochemical parameters for precision agriculture using tree-based and kernel-based algorithms

Sentinel-2 spectral configurations, S2-10m and S2-20m, were evaluated for retrieving essential crop biophysical biochemical parameters their effect on the performance of three machine learning regression algorithms (MLRAs) in two African semi-arid sites. The results benchmarked against all bands (S2-All). show that S2-20m was more robust Leaf Area Index (LAI) (RMSEcv: 0.58 m2 m−2, 0.47 m−2), while provided optimal retrievals Chlorophyll a + b (LCab) 6.89 µg cm−2, 7.02 cm−2) sites, respectively. In contrast, performed better Canopy Content (CCC) Bothaville to an RMSEcv 35.65 yielded relatively lower uncertainties (RMSEcv 26.84 Harrismith. Moreover, various MLRAs sensitive varied by site. GPR XGBoost robust, thus have most potential parameter retrieval both Based benchmark results, configurations can be used independently. obtained here are relevant rapid development precision agriculture using Sentinel-2’s 10 m or 20 bands, without need resampling.