Work

Discrepancies between Sentinel-2 and LiDAR-derived LAI in temperate deciduous forests: Influence of apparent optical depth, radiative transfer priors, and horizontal heterogeneity

Year: 2026

Type: article

Abstract: Accurate estimation of forests leaf area index (LAI) from satellite images remains challenging due to limited sunlight penetration and intricate scattering through structurally complex canopies. In this study, we compared LAI estimated from the inversion of the physical model PROSAIL applied to Sentinel2 (S2) images with LAI derived from airborne LiDAR scans across three temperate deciduous forests in France. We assessed three key factors to minimize discrepancies between both sources: (1) the optical signal attenuation caused by dense canopy, reducing the sensitivity of S2 reflectance to a limited canopy depth, (2) the impact of optimizing the parameterization of the PROSAIL inversion, and (3) the role of local horizontal heterogeneity in canopy height in driving residual errors. We found that S2 LAI estimates best matched LiDAR-derived LAI when the latter was restricted to the upper 48 m of canopy thickness. Optimizing PROSAIL parameters ranges reduced RMSE and bias, yet residual discrepancies remained largest in stands with strong horizontal heterogeneity within S2 pixels, where improvements in correlation and regression slope were limited. While the leaf mass area parameter configuration was robustly transferable across sites, other parameters (LAI, average leaf angle, brown pigments) required local tuning. These findings highlight the importance of accounting for canopy thickness and horizontal heterogeneity in canopy height when inferring LAI values from passive optical sensors, and suggest that future forest monitoring efforts using passive optical sensors may benefit from LiDAR-informed calibration strategies or hybrid inversion approaches that incorporate vertical and horizontal canopy information.

Source: Remote Sensing of Environment

DOI: 10.1016/j.rse.2026.115698

Authors: Nathan Corroyez, Jean-Baptiste Féret, Jérôme Ogée, Jonathan Lenoir, Sylvie Durrieu