Estimation of Coniferous Forest Above-Ground Biomass Using LiDAR and SPOT-5 Data

Mapping the amount and geographic distribution of forest above-ground biomass (AGB) and its change with time is important for understanding the development of the carbon cycle. This study aimed to estimate forest AGB in coniferous tree species of Picea crassifolia stand in the Qilian Mountain, western China using LiDAR and SPOT-5 HRG data. For LiDAR data, the points were classified into ground points and vegetation points. The statistic of vegetation points including height quantiles, mean height, and fractional cover were calculated. The The statistic of SPOT image included spectrum, texture and topographic features. Then the stepwise multiple regression models were used to develop equations relating LiDAR and SPOT-5 HRG image statistic with field-based estimates of biomass for each sample plot. The variables that proved significant for predicting aboveground biomass were mean height, slope, canopy cover percent and the second principal component for principal component transform with R
²
of 0.784 and std of 17.148ton/ha. While when only the LiDAR data were used, the R
²
and std of forest biomass estimation was 0.736 and 18.64ton/ha. The result showed that estimation of forest above-ground biomass using LiDAR and SPOT-5 data could increase the biomass estimation accuracy comparing with only the LiDAR data, while the increase was little.