GROSAT-GLOB: Synergetic Ground-Based and Satellite Retrievals for Global Surface Characterization and Validation

Accurate characterization of land surface reflectance remains a fundamental challenge in satellite remote sensing, particularly over heterogeneous and bright surfaces where surface and atmospheric contributions are strongly coupled. Ground-based AERONET direct-sun and sky-radiance observations provide robust constraints on aerosol optical and microphysical properties, while satellite measurements provide spatially resolved information on surface reflection. In the GROSAT-GLOB approach, these complementary measurement types are combined to enable a more reliable separation of surface and atmospheric signals for surface characterization.

The GROSAT-GLOB approach implements a synergetic retrieval framework based on the GRASP inversion algorithm, integrating ground-based AERONET observations with collocated satellite radiance measurements. Aerosol optical and microphysical properties are primarily constrained by AERONET direct-sun and sky-radiance data, while satellite observations are used to retrieve surface reflectance. To ensure robustness and computational efficiency at the global scale, a two-step retrieval strategy is employed: aerosol microphysical properties are first retrieved by combining spatially aggregated (10–30 km) satellite observations with temporally collocated AERONET almucantar and direct-sun measurements, and are subsequently introduced as constrained a priori information to retrieve surface reflectance at the native satellite resolution using satellite radiances and direct-sun AOD measurements only.

The GROSAT-GLOB retrieval framework has been applied to multiple satellite sensors, including Sentinel-3 OLCI-A/B, Sentinel-5P/TROPOMI, PACE/HARP2, and MTG-FCI, demonstrating its general applicability across instruments with differing spatial resolutions, spectral coverage, and viewing geometries. The retrieved surface reflectance products are validated against independent reference datasets, including MODIS white-sky albedo, HYPERNET reflectance, and RadCalNet bottom-of-atmosphere reflectance, while aerosol retrievals are assessed against AERONET products. These intercomparisons provide a quantitative assessment of retrieval consistency and stability, supporting the use of GROSAT-GLOB products as a surface reference dataset for cross-sensor harmonization and satellite surface reflectance validation.