IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (Jan 2024)

Quantification of GEDI Geolocation Error and Its Influence on Elevation and Canopy Height

  • Cancan Yang,
  • Daoli Peng,
  • Kai Deng,
  • Ling Jiang,
  • Mingwei Zhao,
  • Weisheng Zeng,
  • Yakui Shao,
  • Ni Wang

DOI
https://doi.org/10.1109/JSTARS.2024.3461843
Journal volume & issue
Vol. 17
pp. 18012 – 18024

Abstract

Read online

The global ecosystem dynamics investigation (GEDI) mission aims to provide large-scale, high-precision, and high-frequency measurements of the Earth's three-dimensional structures. However, uncertain geolocation error may hinder or restrict the further application of GEDI products. Based on the error matrix, the laser spot center positioning method was employed to quickly evaluate and quantify the geolocation errors of GEDI L2A (version 2) data in the high-latitude region of China, thereby reducing the impact of systematic geolocation errors on elevation and canopy height detection performance. Combining with high-resolution airborne light detection and ranging data, we provided the geolocation offset characteristics at footprint, beam, and orbit scales, while monitoring their performance over nearly a year. Correcting geolocation errors at the footprint scale can significantly enhance the elevation accuracy, butit has the largest standard deviation (approximately 15 m). Conversely, at the orbit and beam scales, the standard deviations of along- and cross-track error are smaller and the ability to improve elevation accuracy is lower. At beam scale, the level of improvement in elevation accuracy increases with the slope but diminishes when the slope exceeds 20°. Unfortunately, the effect of geolocation correction on canopy height accuracy in this study area is not obvious. Over time, GEDI's elevation detection performance exhibits greater stability, whereas canopy height accuracy decreases as the leaf on season transitions to the leaf off season. The proposed approach provides a rapid solution for preliminary evaluation (beam and orbit scale) and detailed assessment (footprint scale) of GEDI's geolocation errors, thereby laying the groundwork for its future applications.

Keywords