Answers are grounded in this paper and cite their sources.
The Global Ecosystem Dynamics Investigation (GEDI) is the first spaceborne Light Detection and Ranging (LiDAR) mission designed to capture near-global vegetation structure. Despite improvements in version 2, GEDI data still exhibit geolocation uncertainties of around 10 m. In this study, we developed a practical framework to correct GEDI footprint-level geolocation uncertainties by aligning GEDI terrain elevations with an ultra-high-resolution (~1 m) Digital Terrain Model (DTM). Using publicly available DTMs covering 153,670 km² of Japan’s land area, we applied both point-level and group-level correction methods to over 2.7 million forest footprints acquired between 2019 and 2023. Validation through comparison between GEDI RH98 and independent Airborne Laser Scanning (ALS)–derived maximum canopy height data demonstrated a notable improvement in accuracy after point-level correction, with the relative root mean squared error reduced from 35.55% to 21.07%. Our analysis revealed that the magnitude of geolocation offsets was affected by GEDI beam sensitivity across both correction levels, while algorithm settings exerted a stronger influence on group-level corrections. In Japan’s homogeneous needleleaf plantations, the accuracy of GEDI canopy top height (i.e. GEDI RH98) decreased progressively with increasing geolocation offset, with a particularly sharp decline beyond 12 m. Importantly, for footprints where GEDI terrain elevations closely matched ALS-derived DTMs (within 4 m), the original coordinates were sufficiently accurate, suggesting that such data can be retained without correction. All GEDI L2A footprints successfully corrected using the point-level method between April 2019 and March 2023 across forested regions of Japan have been made publicly accessible. Given the continued acquisition of GEDI data, this framework can be applied to future datasets to both mitigate geolocation uncertainty and identify footprints with inherently high positional accuracy. The corrected dataset produced in this study enriches the usability of GEDI-derived products and facilitates their application to fine-scale forest studies in Japan.
No takes yet. Share an insight, caveat, or question.
Li et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: