Estimation of Forest Canopy Height in Hilly Areas Using Lidar Waveform Data

Forest canopy height (FCH) is a key parameter in the estimation of forest biomass and productivity. However, areas with hilly or mountainous terrain present a genuine challenge to extract the vertical structural parameters by using the large footprint Lidar full waveform data. In this study, a mathematical method based on the inflection point of Lidar waveform is developed and applied to process geoscience laser altimeter system data. Furthermore, an improved model, the centroid-terrain index model (CTIM), is proposed to estimate FCH of different forest types in hilly areas. The accuracy of the CTIM model is evaluated by using different field measurement data collected from multiple forest districts in China. For conifer and broadleaf forests, the RMSE is 3.8 m in areas with slope angles larger than 5°. Compared to the ground-based Lidar data, the accuracy is satisfactory in hilly areas. The proposed approach makes a significant contribution toward improving the FCH estimation in hilly areas from large footprint full waveform data, and toward the forest ecosystem monitoring at the global scale.