ALLOMETRIC EQUATION TO ESTIMATE ABOVEGROUND BIOMASS OF Pinus hartwegii LINDL. FROM LIDAR DATA

Abstract

Knowing the aboveground biomass content of individual trees is important in efforts to determine forests contribution to carbon sequestration. LiDAR (Light Detection and Ranging) data are an alternative to satellite images to estimate tree attributes. Considering the importance of forestland on the eastern slopes of the former Texcoco Lake basin for carbon capture, this study aimed to assess regression models for the estimation of above-ground biomass in individual Pinus hartwegii Lindl. trees from LiDAR data. In the field the breast-height diameter (Dn), total height (AT), stem height (AF) and crown diameter (DC) of 60 trees were measured. Those trees were identified on an air-borne LiDAR dataset and the same variables were measured (namely ATL, AFL and DCL, where L stands for LiDAR), except for Dn since it is not possible to measure it directly from LiDAR. Above-ground tree biomass was determined based on Dn and an allometric equation previously developed by other authors. Using the regression technique seven linear and five non-linear models were adjusted, choosing the one with the lowest mean square root of the error (RMSE), the largest R²_adj and the lowest value of the Akaike information criterion, in addition to the compliance of regression assumptions. The selected model is exponential with ATL and DCL variables to estimate biomass: RMSE = 406.70, R²_adj = 0.8107 and AIC = 723.88.