文献类型：会议论文

英文题名：Estimation of Vegetation Biomass in an Alpine Marsh Using Multi-angle Hyperspectral Data CHRIS

作者：Wei Wei[1] Li Wei-na[2] Zhang Huai-qing[2]

第一作者：韦玮

通信作者：Wei, W[1]

机构：[1]Chinese Acad Forestry, Inst Wetland Res, Beijing 100091, Peoples R China;[2]Chinese Acad Forestry, Inst Forest Resources Informat Tech, Beijing 100091, Peoples R China

会议论文集：International Conference on Information Technology (ICIT)

会议日期：DEC 27-29, 2017

会议地点：Singapore, SINGAPORE

语种：英文

外文关键词：Hyperspectral; Multi-angle; Remote sensing; Biomass; Wetland vegetation

年份：2017

摘要：In this paper, it describes the estimation of vegetation biomass in an alpine marsh based on remote sensing data CHRIS. Vegetation biomass is an important index to evaluate the structure, function and health status of wetland ecosystems, directly reflecting the growth status of vegetation communities. Taking Longbaotan Wetland Nature Reserve as the research object, this study was conducted based on the ESA CHRIS/PROBA data. Remote sensing factors, including original spectral reflectance, narrow band Indices, red edge indices, and the newly established vegetation index - VInew, were extracted at the three angles of +36 degrees, 0 degrees and -36 degrees respectively. Correlation between the factors and vegetation biomass in the alpine marsh was analyzed, and sensitivity of the biomass to angle was discussed. The optimal biomass estimation model was established by using the regression analysis method, and then used to estimate the aboveground vegetation biomass in Longbaotan Wetland. The results showed that the exponential model established with VInew (-36 degrees) as the independent variable had the best fitting effect, with a determination coefficient of 0.613, and the inversion accuracy was improved obviously. The accuracy of reverse solving of biomass was obviously improved by using vegetation indices at different angles, which is very important for the optimization of remote sensing parameters used to retrieve wetland vegetation biomass.