[MTT49-02] Monitoring of topographic changes using RTK-UAV in landslide area caused by 2018 Hokkaido Eastern Iburi Earthquake
★Invited Papers
キーワード：北海道胆振東部地震、RTK-UAV、SfM-MVS、地すべり、モニタリング
The Hokkaido Eastern Iburi Earthquake occurred in September 2018, causing a large-scale landslides in about 4,300 ha of the forest area, which is currently in need of immediate forest restoration. However, forest restoration of large-scale landslide areas is a field lacking in knowledge of applicable methods. On the slopes of the landslide area, soil erosion due to rainfall, freezing, and thawing is a driving force moving the surface soil layer of several tens of centimeters depth, inhibiting seed settlement, germination, and growth of seedlings. Therefore, it is necessary to quickly and accurately assess the topographic changes of the landslide surface, which is the base for plant growth. We started monitoring topographic changes using real-time kinematic unmanned aerial vehicles (RTK-UAV) and structure-from-motion multi-view stereo (SfM-MVS) photogrammetry in April 2019.
First, we investigated topographic changes in a landslide area caused by the Hokkaido Eastern Iburi Earthquake in 2018. To verify our measurement accuracy, the position coordinates of 11 validation points in the 4-ha assessment area were obtained using a ZED-F9P, dual-frequency RTK global navigation satellite system (GNSS) on March 12th, 2019. The maximum height difference between verification points was approximately 28 m. We moreover extracted position coordinates of validation points from a digital surface model (DSM), and an orthomosaic image was created from aerial images obtained by the RTK-UAV. Subsequently, the position coordinates of validation points were compared. The analysis of topographic changes was performed on two DSMs created from the aerial images acquired on March 12th and April 23rd, 2019.
The average position between each validation point and the model was 0.060 m~0.064 m in both horizontal and vertical directions. The maximum vertical error was 0.108m. The result of topographical changes analysis indicates a change of -0.1 m to +0.1 in 86.86% of the total assessment area. A change of -0.5 m to -0.1 m was the next most frequent at 11.36%. In particular, the erosion area was confirmed at the boundary between the forest and the landslide area, as well as around the channels. These results demonstrate that monitoring using RTK-UAV the dynamics and stability of surface soil is important in the forest and the landslide area, where it is difficult to set ground control points.
First, we investigated topographic changes in a landslide area caused by the Hokkaido Eastern Iburi Earthquake in 2018. To verify our measurement accuracy, the position coordinates of 11 validation points in the 4-ha assessment area were obtained using a ZED-F9P, dual-frequency RTK global navigation satellite system (GNSS) on March 12th, 2019. The maximum height difference between verification points was approximately 28 m. We moreover extracted position coordinates of validation points from a digital surface model (DSM), and an orthomosaic image was created from aerial images obtained by the RTK-UAV. Subsequently, the position coordinates of validation points were compared. The analysis of topographic changes was performed on two DSMs created from the aerial images acquired on March 12th and April 23rd, 2019.
The average position between each validation point and the model was 0.060 m~0.064 m in both horizontal and vertical directions. The maximum vertical error was 0.108m. The result of topographical changes analysis indicates a change of -0.1 m to +0.1 in 86.86% of the total assessment area. A change of -0.5 m to -0.1 m was the next most frequent at 11.36%. In particular, the erosion area was confirmed at the boundary between the forest and the landslide area, as well as around the channels. These results demonstrate that monitoring using RTK-UAV the dynamics and stability of surface soil is important in the forest and the landslide area, where it is difficult to set ground control points.