Stand density estimation based on fractional vegetation coverage from Sentinel-2 satellite imagery

• DOI: 10.1016/j.jag.2022.102760
• Publication Date: 2022-04-05T14:53:42Z
• Authors: Zhichao Zhang; Xinyu Dong; Jia Tian; Qingjiu Tian; Yanbiao Xi; Dong He
• Keywords: Enhanced vegetation index

Given that forest stand density is an important parameter for studies of carbon, water, and energy cycles a core indicator management, it requires accurate mapping to better assess how impacts the eco-environment. Unfortunately, calculation has long relied on identification individual trees small-scale fine or empirical methods macro estimations large areas, making difficult balance cost accuracy. Thus, this work proposes more efficient method estimate absolute (n/ha) based fractional vegetation coverage retrieved from remote sensing image by establishing correlation between large- approaches perspective hectare scale. The study area covered planted evergreen coniferous forests featuring Pinus tabulaeformis sylvestris in Shaanxi province, China. Taking into account FVC made up contributions deciduous vegetation, phenological factors were considered minimize influence background forest. A Sentinel-2 satellite multispectral sensed at late November 2020 was selected when negligible. accuracy verified using WorldView-3 with high spatial resolution images. regression relationship established 22 sample plots, estimated other 108 plots. result shows mean error root square 41.69 n/ha 117 respectively, relative estimation total number entire reached 81.57%. We thus conclude proposed approach MSI data 10 m feasible way