Toward a Unifying Framework for Biophysical Variable Retrieval From EUMETSAT Sensors: Application to Meteosat Data

The European Organization for the Exploitation of Meteorological Satellite (EUMETSAT) Land Surface Analysis Satellite Applications Facility (LSA SAF) generates and disseminates two suites of FVC, LAI, and FAPAR vegetation products (VEGA) based on Spinning Enhanced Visible and InfraRed Imager (SEVIRI)/Meteosat Second Generation (MSG) and Advanced Very High Resolution Radiometer (AVHRR)/EUMETSAT Polar System (EPS) satellite observations. These suites include leaf area index (LAI), fractional vegetation cover (FVC), and fraction of absorbed photosynthetically active radiation (FAPAR). Differences in algorithms and inputs lead to inconsistencies between the biophysical variables obtained from each sensor. This work addresses these inconsistencies by developing a unifying retrieval framework using a hybrid approach that combines radiative transfer models (RTMs) with machine learning. Specifically, a sparse multioutput Gaussian process regression (SMGPR) is proposed to retrieve multiple variables simultaneously with associated uncertainty estimates. Unlike current MSG algorithms that treat each variable independently, SMGPR jointly derives the three estimates, outperforming both single and multioutput Gaussian process (GP) models in accuracy and scalability. The SMGPR was prototyped using an 18-year data record (2004–2021) from MSG inputs to produce VEGA products at ten-day intervals. A quality assessment, comparing results with LSA SAF estimates and other validated products, showed realistic values and temporal variations across all biomes. The method improved spatiotemporal consistency between MSG and EPS VEGA suites, increasing target accuracy for LAI (73.0%–75.5%) and FVC (73.4%–73.8%), and substantially for FAPAR (52.7%–74.2%). Reduced performance was noted near the edges of the MSG disk, likely due to bidirectional reflectance distribution function (BRDF) input limitations at challenging geometries. This framework is expected to enhance the consistency of vegetation products from current and future EUMETSAT sensors.