Estimating Forest Residual Biomass in Mediterranean Pinus Halepensis Forest Using Low Point Density ALS Data

The aim of this study is the estimation of forest residual biomass (FRB) in Mediterranean
Pinus halepensis
Miller forests. Two regression methods were compared in order to relate the FRB estimated in 45 field plots, to several independent variables derived from low-density Airborne Laser Scanning (ALS) data. The leave-one-out cross-validation technique demonstrated that Support Vector Machine (SVM) model with radial kernel, including the 40
^th
percentile of the return heights, the variance and the percentage of first returns above mean as the selected variables, provided the most accurate result (root mean square error of 2.73 tons/ha). This study confirms the usefulness of low-density ALS data to accurately estimate FRB to better assess the availability of organic renewable fuel energy in the Mediterranean environment.