Combining Lidar- and Gis Data for the extraction of forest inventory parameters

During the last decade the protection of nature has become a very important aspect for interfer-ence or structural alterations. It is necessary for all planning, which might have consequences for nature or impairments of nature, to get information about the actual state and the variations of landscape components. A promising field for the extraction of the needed information is the lidar (light detection an ranging) remote sensing technology. This technology provides vertical informa-tion of the land surface, especially of the vegetation. In this approach the automatic extraction of information is based on the combined analysis of lidar-and GIS-data. The remote sensing system used consists of a lidar scanner and a RGB/NIR digital line camera. The laser scanner device is a discrete-return lidar device. It is able to record the first and last pulse of each foot print simultaneously. The main objective of the current project, called NATSCAN, is to determine basic variables for forest inventories such as tree species, tree height and crown area. These parameters are used to estimate further variables like tree age, stem di-ameter, h/d-proportion, timber volume and tree species distribution. In order to achieve the above main objectives three important tasks are the delineation of trees, the determination of the tree species and the delineation of stand units. The paper will focus on the automatic delineation of the tree crown shapes, the automatic recognition of forest stand units and the calculation of the mean tree height of the highest trees in each stand unit and the calculation of the crown cover density in each unit. 1