Mapping Forest Height and Temporal Decorrelation With SAOCOM and GEDI Data

Polarimetric SAR interferometric (PolInSAR) methods have proven effective for estimating canopy height using airborne data, yet their application with spaceborne sensors remains limited. Addressing this gap is crucial for advancing operational canopy height estimation using spaceborne PolInSAR, particularly given the impact of temporal decorrelation in repeat-pass configurations. However, research on how these effects influence forest height estimation accuracy remains sparse. This study introduces a novel approach that integrates data from the GEDI LiDAR mission and the Argentine SAOCOM satellite constellation to generate canopy height and temporal decorrelation maps. Specifically, GEDI data is used to refine a multi-baseline inversion methodology for improved canopy height retrieval from SAOCOM. The three main contributions of this work are: 1) extending the use of GEDI’s sparse spatial data to enhance a baseline selection approach for PolInSAR inversion, improving canopy height mapping from SAOCOM; 2) introducing a joint exploitation of GEDI and SAOCOM data to produce spatially detailed maps of temporal decorrelation, adapting previous airborne-based approaches to a fully spaceborne context; and 3) demonstrating that incorporating temporal decorrelation into the inversion process leads to more accurate canopy height estimates. Validation against field measurements from forest owners reveals a strong relationship between stand age, height, and temporal decorrelation, with canopy height estimates achieving an RMSE of 2.47 m and an $R^{2}$ of 0.73. These findings highlight the potential for operational canopy height and temporal decorrelation mapping using spaceborne L-band SAR and LiDAR data, contributing to improved forest monitoring capabilities.