Enhancement of bioenergy estimations within forests using airborne laser scanning and multispectral line scanner data

Abstract

Abstract Forest biomass is a substantial source of renewable energy and is becoming increasingly important due to environmental and economic reasons. In Germany, several studies have assessed the bioenergy potential for large areas, e.g. for an entire Federal state. However, in most cases it was not possible to provide detailed maps showing the biomass and the sustainable energy potential for individual forest stands. Thus, the aim of this study was to develop a new and robust method that provides detailed information regarding the spatial distribution of biomass and forest residues as a potential energy resource using a combination of remotely sensed and in situ data. A case study was carried out in a mixed forest in Southern Germany. First, regression analyses were applied to identify relationships between field measurements with several remote sensing metrics to estimate timber volume, mean stem diameter and age. Cross-validation yielded relative root mean square errors (RMSEs) of 30.20% for volume, 27.92% for diameter and 28.81% for the estimation of the age. The absolute RMSEs were smaller than the standard deviation of the observed variables. Next, the regression equations were used to compute attributes for individual forest stands. Stand attributes were then used to model forest residues. To estimate the sustainable annual potential, the actual harvest volume, as defined by forest management planning, was included in the model. Different model parameters were analyzed and an average potential from 0.993 to 1.181 t ha −1 a −1 was computed. The results were compared to previous studies in Germany.