ALPINE FOREST BIOMASS LOCALIZATION BASED ON LIDAR DATA RESULTS OF THE NEWFOR PROJECT Hollaus Markus 1 , Eysn Lothar 1 and Frédéric Berger 2 1 Vienna University of Technology, Department of Geodesy and Geoinformation, Research Group for Photogrammetry (E120-7), Gußhausstraße 27-29, A-1040 Vienna, Austria Markus.Hollaus@geo.tuwien.ac.at; Lothar.Eysn@geo.tuwien.ac.at 2 Responsable de l'équipe PIER UR EMGR, Irstea, 2, rue de la papeterie BP76; 38 402 Saint Martin d'HèresCedex, Frederic.Berger@irstea.fr ABSTRACT: In the project NEWFOR, financed by the European Territorial Cooperation “Alpine Space”, new remote sensing technologies (LiDAR & UAV) for a better mountain forest timber mobilization were investigated. In this contribution possibilities and limitations for detecting forest area and forest biomass as well as the accessibility of forest are shown and discussed. The method for forest area delineation is fully automatically, can be applied to large areas and fulfils the requirements of an operational application. Different forest definitions can be considered by this method. Therefore, an application to different countries with different forest definitions is enabled. For the biomass estimation a semi-empirical regression model was used. The derived biomass maps have a very high spatial resolution and allow comprehensive forest management for large areas and serve as input data for various forest planning activities. It could also be shown that multi-temporal LiDAR data is an excellent data source for change detection of forest parameters (i.e. forest area, biomass).Finally an approach for deriving the forest road network including the road properties width, radius and inclination is presented. The investigations are done within several study areas distributed over the entire Alpine region. Several LiDAR data sets with different properties are analysed and demonstrate their practical relevance in forestry. Keywords: Harvesting, land use, renewable energies, timber, demonstration 1 Introduction The role which mountain forests play is extremely varied. Their contribution to the stability and overall development of life and economic factors in mountainous regions is highly significant. The production of renewable resources like timber has positive effects on climate change consequences attenuation, employment and makes for a strong regional value chain, which in turn has an enormous impact on rural development. The objective of preserving and improving the efficiency of mountain forests is a point of public interest and can only be guaranteed if the planning and implementation of all respective measures are integrated into an adequate and well known socio-economic context. Managing forests in mountain territories is significantly more cost intensive than in plain ones. This is due to the topographic conditions, climatic adversity and limited access which drive partly the economic context. A good knowledge of the forest biomass location, its characteristics and mobilization conditions (exploitability, service roads, and mobilization costs) is a prerequisite for an effective wood harvesting and transport and for a sustainable wood industry. This knowledge is currently insufficient to provide at reasonable costs, the required guarantees on the wood supply and on its sustainability. Improving an efficient and robust evaluation of the forest growing stocks (volume and quality) and its accessibility are the efficient measures to mobilise more wood from mountain forests in a sustainable way. As building forest roads and other infrastructures are often complex and expensive, the availability of financial resources is a key challenge. This could be achieved by providing technology and financial support. With such knowledge and tools it will be then possible to develop an active and sustainable cultivation of mountain forests and an efficient European mountain forest management policy. Recent developments in LiDAR technology – also called airborne laser scanning (ALS), combined to other available data sources (aerial photographs, aerial photo series by UAVs,…), are now allowing a precise and fine mountain forest resource quantification, qualification and mapping. Integrating this technology will provide an innovative response to the challenges of a precise and robust knowledge on the available growing stocks. Thus, the objective of this paper is to present ALS data based methods to derive objective and comprehensible forest maps, to estimate biomass maps and to extract the forest road network. 2 METHODS 2.1 Forest area A fundamental task in forest management is locating and analysing forested areas. The delineation of forested areas has a long tradition in forestry and therefore, worldwide different forest definitions exist to define, whether an area can be classified as forest or non-forest. The delineation task is critical, as a broad field of applications (i.e. obligatory reporting) and users (i.e. governmental authorities, forest community) rely on this information. The results determined from these applications are highly dependent on the fundamental input parameters size and position of the delineated forest areas. In the past mainly aerial images were used for a manual or semi-automated delineation. Shadow effects limit this task, particularly for detecting small forest clearings and the exact delineation of forest borders on a parcel level. Additionally, the quality of the results of a manual delineation is subjective and variable between interpreters and may lead to inhomogeneous, maybe even incorrect datasets. An automatic delineation of forested areas based on ALS data can overcome these limitations in most cases. Within the project NEWFOR the method of Eysn et al. [1] was applied to several NEWFOR pilot areas characterized with different forest structure and growing conditions to detect forested areas. The method relies on four clearly defined geometrical criterions (minimum area, minimum height, minimum width and minimum crown coverage) which are subsequently