The forest understory vegetation (UVEG)—including the erect woody shrubs and herbaceous layer play a critical role in forest ecosystem biodiversity, nutrient and carbon cycling, and even wildlife habitat. Despite the importance of UVEG in forest ecosystem, less research of understory biomass are available than the tree biomass. Allometric equation continue to be the widely used methods for understory woody shrubs and herbs biomass estimation, but these often rely on the measurement of allometric parameters and species-specific biomass models, which was time-consuming and ineffective. The Terrestrial laser scanning (TLS) can measure the understory structure in 3D with high detail, which can help to improve the methods of understory estimation in forest ecosystem. In this study, we develop an approach to estimate understory biomass in temperate forest from TLS point cloud data . We compare these estimates against destructively harvested biomass estimates and biomass estimation from allometric equations. We obtained 80 shrubs quadrats and 96 herbaceous quadrats destructive samplings in twenty-two 20m×20m forest plots in the east of National Park for Amur Tiger and Amur Leopard which based on the variation of forest type and topography conditions.
Results/Conclusions
The result shows that the understory biomass estimation derived from TLS voxel-based methods get a high agreement with the reference values from destructive samplings, with a concordance correlation coefficient (CCC) of 0.81(shrubs) and CCC=0.73(herbs), while the agreement between understory biomass estimates from allometric equations and the reference is lower with CCC=0.65(shrub) and CCC=0.43(herbs). The results indicate that the application of TLS could substantially reduce the costly destructive samplings and increase effective in understory estimation. Based on the great advantages of TLS data in the estimation of biomass in the understory of the forest, we can use the regression result to extrapolate a broader plot-scale biomass mapping.