[S-VC34] 火山の監視と活動評価

[SVC34-P12] An aerial survey using unoccupied aerial vehicles in response to the phreatic explosion of Aso volcano in October 2021

*横尾 亮彦¹、寺田 暁彦²、森田 雅明³、田中 良⁴、安田 裕紀⁴、梨元 昴¹、宇野 幸輝¹、宇津木 充¹、吉川 慎¹、井上 寛之¹、大倉 敬宏¹、森田 裕一⁵ (1.京都大学大学院理学研究科、2.東京工業大学火山流体研究センター、3.産業技術総合研究所地質調査総合センター、4.北海道大学理学研究院、5.防災科学技術研究所)

キーワード：無人航空機、阿蘇山、水蒸気爆発

A phreatic explosion with a plume height reaching 3.5 km altitude occurred at Aso volcano on October 20, 2021. At the eruption, large ballistic blocks were ejected and fell within a range of 900 m from the active crater. Associated pyroclastic flows traveled 1.6 km to the north and 1.0 km to the west from the crater. This eruption caused catastrophic damage to the proximal area around the crater. The Japan Meteorological Agency (JMA) raised the alert level from 2 to 3. The area of a 2 km distance from the crater was off-limits. Immediately after the eruption, we decided to conduct aerial photography using unoccupied aerial vehicles (UAVs) to understand the situation at the proximal area. We used four UAVs (one DJI M600 and three DJI M200), capable of flight stability even in a windy winter environment. The area to be surveyed was approximately 5 km², including areas whose surface turned to gray at the eruption (probably covered with pyroclastic-flow and ash-fall deposits). Two types of cameras, DJI Zenmuse X5S and DJI Zenmuse XT2, were used to image the area around the crater and the surrounding area, respectively. A ground resolution and an average relative flight height were set as 2cm/pix and 100 m, respectively. It provides a resolution capable of identifying blocks on the ground as small as 10 cm. The study area was divided into 462 sections in 0.001-degree latitude and longitude increments. Then, 22 integrated areas were set up such that the intra-area flight altitude was kept constant at higher than 25 m above the highest point on the topography of each section. Takeoff and landing points were located between 680 m and 1070 m from the crater. Considering the number of available batteries, takeoff and landing points, and flight paths, each area in charge for four groups (Kyoto and Hokkaido Universities, Tokyo Institute Technology, and The National Institute of Advanced Industrial Science and Technology), were assigned. After JMA lowered the alert level from 3 to 2 on November 18, we placed 19 thin metallic boards as ground control points in the surveyed area on November 24 and measured their locations using Leica SR520. We obtained approval and permission from the West Japan Civil Aviation Bureau and the Kansai Airport Office to use UAVs around the crater of Aso volcano. We also obtained permission for the flights from the Aso Volcano Disaster Prevention Council (AVDPC). A total of 22,157 visible images were consequently acquired over a three-day mission from November 25 to 27. We are currently creating a 3-D terrain model and orthoimages by programs using the Surface-from-Motion algorithm. In this presentation, we will report the preliminary results of this analysis and the spatial distribution of the ejected blocks around the crater. This survey was conducted with sufficient support from the "National project of mobile observations for volcano research" sponsored by the Ministry of Education, Culture, Sports, Science, and Technology Japan. We also appreciate the careful monitoring of the Aso volcano by AVDPC and JMA during the survey.