Geographically Weighted Regression Modeling Using Optical and Lidar Data to Map Aboveground Biomass of Urban Trees

Abstract

Aboveground biomass (AGB) of trees refers to the dry weight of organic matter per unit area. Estimating the AGB of urban trees is helpful for urban ecosystem assessment. Due to the high heterogeneity of urban land use, the AGB of trees is greatly affected by geographical location, which brings difficulties to the estimation of AGB of urban trees. In this study, we use geographically weighted regression model(GWR) to take the geographical location into account in the regression model. In order to obtain a large number of samples, high-resolution optical data and airborne LiDAR data are used in combination. Taking Hengqin in Guangdong Province as an example, the experimental results show that the GWR model has good performance in estimation accuracy.