Forest Canopy Height Estimation Using Satellite-Borne LiDAR and Optical Remote Sensing Data

Abstract

Forests are the main carbon sinks in terrestrial ecosystems, and accurately estimating forest canopy height is the basis for calculating terrestrial ecosystem carbon storage. Conducting large-scale forest canopy height research is crucial for monitoring changes in forest carbon stocks and assessing forest carbon sequestration capacity. This paper takes the forest in Maoershan, Heilongjiang Province, as the research object, using data from the Global Ecosystem Dynamics Investigation (GEDI) LiDAR satellite to perform forest canopy height inversion for Maoershan. The results show that the root mean square error (RMSE) value for terrain inversion accuracy in the Maoershan area is 4.49 m, and the mean absolute error (MAE) value is 3.33 m. Inversions based on GEDI data of forested terrain show that strong beams are superior to coverage beams, the daytime effect is better in humid areas, while the nighttime effect is better in arid areas. Furthermore, GEDI data are significantly more precise in determining understory terrain than in estimating canopy height. Increasing slope obviously causes a deterioration in the accuracy of GEDI data.