Abstract: Timely and rapid monitoring of biodiversity is of great significance for forest management and diversity conservation. The traditional method of plot survey cannot meet the data needs of biodiversity monitoring and evaluation at landscape and regional scales. In this study, 27 Sentinel-2A image data in 2022 were used to construct a biodiversity estimation model at the small watershed scale using 12 vegetation indices. The spatial distribution pattern of biodiversity Caijiachuan Basin, Jixian County, Shanxi Province was explored with this model. The results showed that stand density of Robinia pseudoacacia was not the only factor controlling biodiversity in the forest. Simpson diversity index (D), Shannon diversity index (H) and Pielou evenness index (J) were significantly correlated with the percentage of infrared percentage vegetation index (IPVI) and red-edge chlorophyll index (CIre) (P<0.01). The mean square errors (MSE) of the estimated and measured values of D, H and J in the remote sensing model of biodiversity were 0.021, 0.192 and 0.016, respectively, indicating that the estimation model constructed by the vegetation indices can be used for biodiversity estimation and monitoring in the study area. The biodiversity level was best estimated on the shady slope areas with high altitude.

Key words: biological diversity, vegetation index, remote sensing monitoring, Sentinel-2A image