Mapping forest above-ground biomass of coniferous forest in northern China by integrating simultaneous multi-angle spectral and full-waveform observations using the TECIS system

This is the authors' abstract. We don't add key points for this paper.

• Simultaneous acquisition of multi-angle and full-waveform data for AGB estimation. • First footprint-level forest AGB mapping using the Chinese TECIS system. • Forest background effects were accounted for in AGB estimation. • AGB values were directly retrieved from TECIS satellite observations. Natural forest above-ground biomass (AGB) accumulation alters both the three dimensional (3D) structural and spectral characteristics of forest stands. The laser scanning and five-angle multispectral systems mounted on the newly launched Chinese Terrestrial Ecosystem Carbon Inventory Satellite (TECIS, also known as “Goumang”) enable the simultaneous capture of both structural and spectral information of forest. In this study, we developed a novel method to estimate forest AGB at footprint level by incorporating full-waveform laser return intensity (LRI) and forest background reflectivity obtained from the TECIS system. Forest background reflectivity was derived using five-angle multispectral imaging from the TECIS system, which significantly influences both the bidirectional reflectivity factor and the full-waveform LRI of forest stands through a geometric-optical model. In addition, we retrieved vertical profiles of gap fraction and leaf area volume density from TECIS full-waveform LRI data. By doing so, we proposed a forest AGB estimation model that incorporates forest background reflectivity, 3D structural features, and tree shape factor information. Our results showed that incorporating forest background reflectivity improved forest AGB estimation accuracy, capturing 85 % (n = 20, p < 0.01) of the variance in field-based AGB, with the coefficient of determination (R 2 ) increasing from 0.39 to 0.66 and the root mean square error (RMSE) decreasing from 28.12 t/ha to 17.77 t/ha. Finally, we generated a continuous AGB map using a Random Forest model, achieving an R 2 of 0.63 and an RMSE of 14.21 t/ha. This study demonstrates the significant potential of the TECIS system for forest AGB estimation by leveraging simultaneous structural and spectral observations. The proposed approach provides a foundation for forest carbon monitoring in precision forestry.