Abstract: Urban vegetation is one of the important factors in low-carbon governance, and its carbon sequestration capacity directly affects urban carbon emissions. In view of the time-consuming and labor-intensive existing technology, which has a certain impact on the environment, this paper proposes a study on the estimation of urban vegetation carbon reserves based on hyperspectral data. Taking the central urban area of Baiguan street, Cao'e street and Songxia street in Shangyu district, Shaoxing city as the study area, urban vegetation is divided into four types: arbor, shrub, bamboo forest and grassland. The proposed carbon sequestration index is used to qualitatively analyze the carbon sequestration capacity of vegetation, and the biomass empirical formula is used to quantitatively calculate the carbon storage. The hyperspectral vegetation carbon storage estimation model is obtained by combining the two, which provides a new idea for the rapid estimation of large-scale carbon storage. The results show that the classification accuracy of the four urban vegetation reaches 88.10%, and the carbon fixation capacity is ranked as bamboo forest>tree>shrub>grassland.

Key words: carbon storage, hyperspectral, carbon fixation, urban vegetation, low carbon