Japan Geoscience Union Meeting 2023

Presentation information

[J] Online Poster

H (Human Geosciences ) » H-DS Disaster geosciences

[H-DS10] Geohazards in humid, tectonically active countries and their precursors

Fri. May 26, 2023 3:30 PM - 5:00 PM Online Poster Zoom Room (8) (Online Poster)

convener:Yoshihiko Kariya(Department of Environmental Geography, Senshu University), Taro Uchida(University of Tsukuba), Ryoko Nishii(Niigata University)

On-site poster schedule(2023/5/25 17:15-18:45)

3:30 PM - 5:00 PM

[HDS10-P07] Observation of areal details of collapsed areas using low-altitude images acquired by UAV and 3D terrain models

*Takumi Sahara1, Taro Uchida1, Shigetaka Takiguchi2 (1.University of Tsukuba , 2.National Institute for Land and Infrastructure Management)


Keywords:landslide, UAV, UAV-SfM, Soil layer, Bedrock, Spatial information

Since the survey of slope failure sites is the most fundamental step in elucidating the mechanisms of slope failure, many surveys have been conducted in various fields. For example, surveys and research to determine the surface profile of a collapsed area based on aerial photographs and satellite images have been widely conducted. In recent years, it has become possible to determine the three-dimensional shape of a collapsed site using aerial laser survey technology. In addition, one of the most representative surveys is field transecting, which provides detailed information on areal characteristics of the collapsed area, such as bedrock cracks, root system distribution, and soil layer thickness. The detailed areal information obtained from these surveys has led to research on the factors that influence slope failure, such as the effects of weathering on bedrock cracks and clay veins, groundwater in the soil and bedrock, and the distribution of root systems. However, there is a limit to the area that can be realistically surveyed due to the danger and time constraints of fieldwork. In addition, images taken from high altitudes, such as aerial photographs, do not provide information on overhangs that are shaded by trees, nor do they provide the detailed information that can be obtained by field survey. Therefore, it is difficult to say that sufficient detailed information on the top and margins of collapsed areas has been accumulated. On the other hand, with the recent commercialization of UAVs, it has become possible to take images from the sky at low altitudes.
In this study, a drone was flown over two collapsed areas in Nachikatsuura-cho, Wakayama Prefecture, and Bishamon, Osaki City, Miyagi Prefecture, at an altitude of 10 to 30 m above the ground to collect detailed areal information of the inaccessible areas by adjusting the camera angle to the collapsed areas, especially to the sides of the cliffs and collapsed areas, and to determine the spatial distribution in the depth direction using a three-dimensional model. We also attempted to understand the spatial distribution in the depth direction by analyzing a 3D model.
High-resolution information was obtained from aerial images with a pixel size of 0.86 to 1.76 mm/pixel on the ground and a 3D point cloud model with a resolution of 5.25 mm/pixel. As a result, the distribution of bedrock cracks and root systems could be determined areally. Furthermore, the created model was cut into small pieces along the circumference of the collapsed area, and by aligning the pieces in the depth direction from the edge, a cross-sectional view of the slip cliff and the side of the collapsed area was created. As a result, it was found that the thickness of the soil layer was 2.0~4.0m and 3.0~7.5m in the collapsed areas of Nachikatsuura-cho and Osaki-city, respectively, and that it was possible to obtain new information about the edge of the collapsed areas that had been difficult to obtain in the past.