The Inversion of Forest Stock Volume in Huzhu Beishan Forest Farm Based on Sentinel-2 Images
Forest stock is a key indicator for assessing forest health and carbon sequestration capacity,and the use of remote sensing technology can achieve rapid estimation of regional forest stock.In this study,the Huzhu Beishan Forest Farm in Haidong City,Qinghai Province was taken as the study area.Based on the forest sample plot survey data and Sentinel-2 images,representative remote sensing features were screened out using extracted single-band reflectance,vegetation indices,texture factor and topography factor,and the Random Forest Algorithm was applied to realize the inversion of the forest stock in the study area.The main results are as follows:(1)Based on the Pearson correlation analysis,B11,B10,B6,B7,EVI,SAVI,NIRv,ExGR,MNLI,and B12 are identified as the remote sensing feature factors for forest storage inversion in this study.(2)The random forest algorithm can effectively invert the forest stock in the study area.The coefficient of determination R² of the inversion result was 0.73,and the root mean square error RMSE was 1.65m³/hm2.(3)The average storage volume of arbor forest in Huzhu Beishan Forest Farm was 134.4m³/hm2,and in the spatial distribution,the forest storage volume in the east-central,south-eastern,and part of the northern area was higher.
NETL
NSTL
万方数据
