engleski

The UAV-based Photogrammetry for Estimation of Plot-level Structural Parameters of Pedunculate Oak Forests

Digital aerial photogrammetry by using an Unmanned Aerial Vehicle (UAV) has recently attracted great attention in forest inventory research. By photogrammetric processing of UAV images, dense point clouds and digital surface models (DSMs) can be derived. To obtain data that can be useful for forest inventory, i.e. normalized point cloud (height above ground) or rasterized canopy height model (CHM), a highly accurate digital terrain model (DTM) is required. In combination with field reference data and using established prediction models, the normalized point clouds and CHMs could serve to estimate various forest inventory attributes. Although, a number of recent studies emphasize the great potential of UAV- images for tree and forest attributes estimation, further research is required to prove its practical applicability over different forest types. Moreover, the UAV technology (hardware and software) rapidly progresses and constant validations of results are needed. This study aims to investigate the capability of UAV-based CHMs generated at different spatial resolutions (10 cm, 30 cm, 50 cm) for use in forest inventory, with a special focus on estimation of plot-level mean tree height and mean diameter at breast height (dbh) of even-aged pedunculate oak (Quercus robur L.) forests. Tree height and dbh are two of the more fundamental measurements in forest inventories and provide the basis for many other computations The study was conducted in the lowland forest complex of Pokupsko Basin located 35 km southwest of Zagreb, Central Croatia. Field measurements of tree structural parameters were collected from the systematic sample of 165 circular plots with radii of 8, 15 or 20 m, which depended on the stand age. Dbh was measured for all trees with dbh≥10 cm. Tree height was measured for at least 50% of trees per plot, while heights of other trees in the plots were estimated using the constructed height curves (height-dbh models). The mean dbh and mean height for each plot were calculated. The positioning (x, y, z) coordinates of the sample plot centres were recorded using the GNSS receiver Stonex S9IIIN connected with the Croatian Positioning System (CROPOS). The UAV images were acquired using the Trimble UX5 HP with Sony Alpha ILCE-R7 on 30 and 31 May 2017. The study area (≈1500 ha) was covered by 1441 images with the ground sampling distance (GSD) of ≈8 cm. The images were collected in 4 flights with endlap of 90% and sidelap of 80%. DSMs of different spatial resolution (10 cm, 30 cm, 50 cm) were generated using the Dense DSM (Semi- Global Matching) algorithm of PHOTOMOD UAV 6.3 software. Finally, raster CHMs images with spatial resolution of 10 cm, 30 cm and 50 cm were then generated by subtracting LiDAR DTM from corresponding DSMs. For each plot, various height (mean, standard deviation, max, min, mod, percentiles) and density metrics (ratio between area of canopy above certain height threshold and plot area) were extracted and calculated from each CHM. These metrics were then further considered in the statistical modelling (i.e., for development of plot-level models of mean height and mean dbh) as potential independent variables. The findings suggest the clear potential of UAV-based photogrammetry when used for plot-level forest inventory, with reasonable accuracy for all tested spatial resolutions. Acknowledgment: This research has been supported by the Croatian Science Foundation under the projects IP-2016-06-7686 “Retrieval of Information from Different Optical 3D Remote Sensing Sources for Use in Forest Inventory (3D-FORINVENT)” and IP- 2016-06-5621 “Geospatial Monitoring of Green Infrastructure by Means of Terrestrial, Airborne and Satellite Imagery (GEMINI)”. The authors wish to thank Hrvatske vode, Zagreb, Croatia, for providing LiDAR DTM data.

digital surface model ; canopy height model ; tree height ; diameter at breast height ; forest inventory

nije evidentirano