Akademik Veri Yönetim Sistemi
20th International Symposium on Environmental Pollution and its Impact on Life in the Mediterranean Region, Athens, Yunanistan, 26 - 27 Ekim 2020, ss.236-237, (Özet Bildiri)
Due to the dynamic ecosystems of forests that often change as a result of continuous growth, expansion or some natural and unnatural external influences. Thus, an accurate and up-to-date inventory of forests is one of the most important features for establishing a forest management policy. Most appropriate methods should be chosen after a detailed investigation which used for eliminating the complexity of the ecosystem, developing forest management and shaping the environmental policy. In this study, photogrammetry and Light Detection and Ranging (LIDAR) data were evaluated both separately and together to obtain the most appropriate results related to forest inventories. Additionally, Digital Terrain Model (DTM), Digital Surface Model (DSM), Normalized Digital Surface Model (NDSM), Canopy Height Models (CHM), which are the well-known products were produced using point cloud data obtained separately from each technique. To evaluate the data, noise reduction which is a must important step was performed on the point clouds as the first step. Then, Digital Elevation Models was produced by means of the classification of point clouds by means of 4 different methods in the determined study region. In the first one, tree inventory was obtained by using only photogrammetric data. In the second, three points were obtained by combining ground points obtained from LIDAR data and photogrammetry data together. In the third one, gaps in ground points in the photogrammetric point cloud were filled by IDW interpolation to extract the tree inventory. In the fourth one, tree inventory was obtained by using only LIDAR data. Finally, although LIDAR data is more expensive and laborious than the photogrammetric method, it is the most successful method to produce tree inventory by producing more accurate results. The main reason one of the successes here was that the success in producing the Digital Terrain Model from the LIDAR data yielded better results than the photogrammetric method. Because of the LIDAR’s effectiveness in modeling vegetation three-dimensional structure, return information and the arrangement of parts of the vegetation can be predicted by the usage of this data. In the first model, the full inventories of the forest couldn’t have been realized. In the second method, better performance was realized than the first method, but worse results obtained than the 4th method. Unexpectedly a number of counted trees were obtained in the third method than the second one. The main reason for the differences obtained with different methods was evaluated as the 5 years' differences between obtaining LIDAR and photogrammetric data.