Feature based tree sepcies classification using airborne hyperspectral and LiDAR data for the Bavarian Forerst National Park

The identification of tree species is a substantial part of forest inventory and necessary to efficiently manage a forest environment. However, for large forest areas, such as the Bavarian Forest National Park, conventional forest inventories are time-consuming and expensive. In recent years, studies showed that the use of hyperspectral data to map the spatial and structural patterns of individual tree species abundances under various forest conditions may benefit from the integration of LiDAR (Light Detection and Ranging) data. The aim of this thesis was to combine hyperspectral data with structural and topographic information on tree species derived from LiDAR data to improve tree species classification. Therefore, a feature database was developed including image based hyperspectral information and four derived vegetation indices as well as five LiDAR based parameters, namely tree height, aspect, slope, hill shade, and elevation. The study is based on airborne hyperspectral data acquired with the HySpex VNIR-1600 sensor (160 spectral bands, 400nm - 990 nm, 1.6m spatial resolution). Additional full waveform LiDAR data, including a Digital Surface Model, Digital Terrain Model and a Digital Canopy height Model, were available for the analysis. Individual tree crowns as well as clusters of tree crowns from 13 different tree species were located and identified during a field survey. The field-demarcated tree canopies were used as reference data for creating the feature database. Several preprocessing steps including atmospheric correction, spectral and spatial polishing, BRDF effect correction as well as ortho-rectification of the hyperspectral imagery were conducted before the analysis. A band selection procedure based on principal component analysis, band correlation, and band variance was performed to identify the most appropriate spectral bands for species discrimination, resulting in a set of 53 spectral bands. Seven different combinations of hyperspectral, structural and terrain specific parameters contained in the feature database were investigated in a Random Forest modelling approach to ascertain which variables enhance the overall classification accuracy. A classification model using all available parameters in the feature database yielded a 17% higher overall accuracy (94%) compared to using only the preselected spectral bands (77%). The most important predictors, which were found to play an important role in the different classification models, were elevation, the two vegetation indices simple ratio and photochemical reflectance index as well as the hill shade parameter. For most of the 13 tree species, the final classification model achieved individual class accuracies of more than 90%. The study showed that a tree species feature database consisting of hyperspectral signatures and relatively simple LiDAR derived features has high potential for a remote sensing based forest inventory. A model transferable to an area wide mapping of tree species based on the needs of the Bavarian Forest National Park was established