Abstract
Uncertainties in forest aboveground biomass (AGB) estimates resulting from over- and underestimations using remote sensing data have been widely studied. The uncertainties may occur due to the spatial effects of the plot data. In this study, we collected AGB data from a total of 147 Pinus densata forest sample plots in Yunnan of southwestern China and analyzed the spatial effects on the estimation of AGB. An ordinary least squares (OLS) and four spatial regression methods were compared for the estimation using Landsat 8-OLI images. Through the spatial analysis of AGB and residuals of model predictions, it was found that the spatial autocorrelation and heterogeneity of the plot data could not be ignored. Compared with the OLS, the impact of the spatial effects on AGB estimation could be reduced slightly by the spatial lag model (SLM) and the spatial error model (SEM) and greatly reduced by the linear mixed effects model (LMM) and geographically weighted regression (GWR) based on the distributions of prediction residuals, global Moran’s I, and Z score. The spatial regression models had better performance for model fitting and prediction because of the reduction in overestimations and underestimations for the forests with small and large AGB values, respectively. However, the reductions in the overestimations and underestimations varied depending on the spatial regression models. The GWR provided the most accurate predictions with the largest R2 (0.665), the smallest root mean square error (34.507), and mean relative error (−9.070%) by greatly reducing the AGB interval for overestimations occurring and significantly increasing the threshold of AGB from 150 Mg/ha to 200 Mg/ha for underestimations. Thus, GWR offered the greatest potential of improving the estimation of Pinus densata forest AGB in Yunnan of southwestern China.
Highlights
Trees grow by absorbing and using light, water, nutrients, and carbon dioxide, leading to accumulation of forest biomass
This study showed that all the spatial regression models had better performance for model fitting and prediction for Pinus densata forest aboveground biomass (AGB) compared with the non-spatial regression model ordinary least squares (OLS)
This study showed that the spatial effects of the plot AGB data that lead to overestimation and underestimation could not be ignored, and that the spatial regression models can improve the estimation accuracy of Pinus densata forest AGB
Summary
Trees grow by absorbing and using light, water, nutrients, and carbon dioxide, leading to accumulation of forest biomass. Estimating forest biomass is the basis of analyzing and understanding forest carbon dynamics and carbon cycling [1,2,3,4,5,6] and has been widely considered [7,8,9]. Saturation of spectral reflectance from highly dense and multi-layer canopy forests leads to underestimations of forest AGB, and mixtures of spectral reflectance from soil and vegetation result in overestimations of AGB for young forests. Radar and LiDAR data that characterize tree heights and forest canopy structures, to some extent, mitigate the effects of the data saturation and mixture of spectral reflectance from soil and vegetation [9,13,16,17,18,19,20]. Using optical images is still a good alternative for mapping forest AGB for large areas. Reducing the uncertainties due to the fact of overestimations and underestimations are still a challenge
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
