Geospatial Analysis of Multi-Scale Topographic Roughness and the Morphological Characteristics of LiDAR Data

• Main Authors: Mon-ShiehYang, 楊孟學
• Other Authors: Ming-Chee Wu
• Format: Others
• Language: en_US
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/18532270224778277960

博士 === 國立成功大學 === 地球科學系 === 103 === Unpredictable weather condition and highly complex topography; in addition with frequent land mass disaster, which have made river management a substantial challenge in Taiwan. Especially, the sediment and incoming flow will directly affect the dynamic balance of the river. Even more, to induce the occurrence of secondary disaster. Namely, river management has currently become an important issue for today’s disaster prevention and mitigation. High-resolution topographic data are capable of describing the morphological features and evaluating the magnitude of a disaster; thus, providing more detailed and more complete information of the disaster on the land surface. Purposes of this study were to demonstrate the implementation of high resolution topographic data to show the multitemporal variability of river bed morphology through roughness mapping; before and after the disaster. In addition, due to the limitation of weather condition and abruption of transportation in the disaster region, the traditional investigation method would not be able to provide a real-time and complete information of the disaster. Therefore, this study has adopted the aids of remote sensing data to analyze the change of various geometric landform for a colluvium fan; results of the study may be extensively applicable for regional mapping of the vulnerable area. Besides, this study have also tried to analyze and compare the variability of roughness calculation and application between the commonly used regular grid raster data and the point cloud data. Thus, for a greater understanding of the spatial scale-dependent roughness at different scales and resolutions, semivariograms were adopted to determine the effectiveness of data in representing roughness in this study; the range of semivariograms can be a clear identification that raster data generate “rougher” results compared with point cloud data. In a smooth area, the results demonstrated low similarity among point cloud data, indicating that point cloud data are smoother than raster data.