Abstract:Aiming to solve the problem that it is difficult for spaceborne light laser detection and ranging (LiDAR) global ecosystem dynamics investigation (GEDI) emitted laser pulses to penetrate the forest canopy in dense forest areas to accurately obtain understory terrain information, and the overlap between GEDI stand canopy echo and understory terrain echo will be increased in highslope terrain, and it is difficult to estimate forest canopy height with high accuracy, combined with the characteristics of defoliation in winter broad-leaved forest and the strong penetration advantage of GEDI emitted wave laser pulse, a model of forest canopy overestimation in different seasons was constructed by using the GEDI waveform length parameters. The accuracy of forest canopy height estimation in summer and winter was analyzed by using different percentage waveform length parameters (rh_aN) of GEDI. Subsequently, terrain slope factor DTM data was introduced to correct the forest canopy height estimation model, segmenting forest canopy height estimation based on terrain slope, thereby addressing the problem of low accuracy in forest canopy height estimation caused by the overlap between canopy echoes and terrain echoes due to steep terrain. The research results showed that the coefficient of determination R2 of forest canopy height estimation accuracy in summer was 0.573, with root mean square error (RMSE) of 3.695m; in winter, the R2 was 0.633, with RMSE of 3.671m. After correcting the forest canopy height estimation model with terrain slope, the overall estimation accuracy in winter, with R2 of 0.709 and RMSE of 3.271m, was significantly improved. Accuracy of forest canopy height estimation in winter was markedly better than in summer, and the introduction of terrain slope factors effectively enhanced the accuracy of forest canopy height estimation under different terrain slope conditions.