High-spatial-resolution gross primary production estimation from Sentinel-2 reflectance using hybrid Gaussian processes modeling

This publication presents a new hybrid Gaussian processes approach to estimate gross primary production (GPP)—the amount of carbon plants fix through photosynthesis—at high spatial resolution using Sentinel-2 reflectance data. By combining physical knowledge with data-driven modelling, the method produces more accurate and spatially detailed GPP maps, particularly in complex forest and vegetation landscapes. The results highlight the approach’s strong potential for improving carbon cycle assessments and supporting climate- and ecosystem-related monitoring efforts.