Optimizing Forest Canopy Height Estimation Through Varied Photon-Counting Characteristic Parameter Analysis, Window Size, and Forest Cover

Forests are an important component of the Earth’s ecosystems. Forest canopy height is an important fundamental indicator for quantifying forest ecosystems. The current spaceborne photon-counting Light Detection and Ranging (LiDAR) technique has photon cloud characteristic parameters to estimate forest canopy height, and factors such as the sampling window size have not been quantitatively studied. To better understand the precision for estimating canopy height using spaceborne photon-counting LiDAR ICESat-2/ATLAS (Ice, Cloud, and Land Elevation Satellite-2/Advanced Topographic Laser Altimeter System), this study quantified the impact of photon-counting characteristic parameters, sampling window size, and forest cover. Estimation accuracy was evaluated across nine study areas in North America. The findings revealed that when the photon-counting characteristic parameter was set to H70 (70% of canopy height) and the sampling window length was 20 m, the estimation results aligned more closely with the airborne validation data, yielding superior accuracy evaluation indicators with a root mean square error (RMSE) of 4.13 m. Under forest cover of 81%–100%, our algorithms exhibited high estimation accuracy. These study results offer novel perspectives for the application of spaceborne photon-counting LiDAR ICESat-2/ATLAS in forestry.