|M.Sc Student||Vidro Yulia|
|Subject||Shrub's Structure and Biomass Analysis along Climatic|
Gradient Based on Photogrammetric Point Cloud
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Maxim Shoshany|
|Full Thesis text|
Shrubs are dominant life forms in arid and semi-arid climatic areas. Shrubs are carbon dioxide sinks they also prevent soil erosion and contribute to water balance. Mapping shrubs’ biomass across transition zones between Mediterranean and arid climatic areas is important for better understanding their potential response to climate change (M. Shoshany, 2012). However, there is limited information regarding the structure and biomass of these shrubs and their change across climatic gradients, partly due to limitations on the use of destructive methods. This study concentrates on the study of the shape of shrubs, their volume and biomass in order to improve the accuracy of biomass allometric equations. During this research, a field survey of 28 shrubs and dwarf shrubs was conducted in three sites along the transition zone between sub-humid to arid climate areas in central Israel. The sites from north to south are: Ramat Avisur (Beit Guvrin), Amazya, Lahav. Field campaign included oblique video footage captured by smartphone camera, height and diameter by tape measurements and spectral data collected using portable spectroradiometer. Photogrammetric point cloud for each shrub was generated from the video footage using the PIX4D software (https://www.pix4d.com/).
Detailed analysis of cloud data was conducted in order to determine which volumetric shape is best to describe shrubs’ envelope. That was accomplished by (1) comparison of estimated volume of point clouds and volume of 3 common models: hemi ellipsoid, spherical cap and cylinder; and (2) analysis of distances and deviations of point clouds from these fitted forms. Afterwards, we tested statistical relationships between field allometric parameters versus the dimensions of the point clouds, and then applied two empirical biomass models based on previous studies. Finally, we assessed correlations between shrub biomass as determined through the allometric models and spectral vegetation indices.
Correlation analysis of cloud versus typical models’ volume, yielded high values of determination coefficient (0.95-0.99) and correlation was consistent for all sites and shrubs’ sizes. Using fine alignment between shrubs’ point clouds and the three model shapes, it was found that hemi ellipsoid is the most suitable to describe the shape of shrubs. The use of common allometric parameters such as maximal diameter and height resulted in distinctive overestimation of the shrubs’ volume, where the deviations varied between 33 and 125%. Fine alignment of shape relatively to the point cloud allowed decrease of the error level to between 20 and 29%. Low correlations were found between biomass values and the spectral indices of NDVI (Normalized Difference Vegetation Index) and VARI (Visible Atmospherically Resistant Index), thus, limiting their use for biomass mapping with satellite images.