11:00 AM - 1:00 PM
[SVC31-P11] Thermal infrared observation using UAV at Izu Oshima Volcano
Keywords:UAV, Thermal Infrared observation, Izu-Oshima
The Izu-Oshima volcano has been reported to erupt every 36-38 years of 107 ton order by studies of eruption histories (Nakamura, 1964 and Endo et al., 1988), and more than 30 years have already passed since the last eruptive activity (1986-1990). The Japan Meteorological Agency (JMA) has been observing ground deformation for a long period of time, and has observed the expansion of the mountain, which suggests magma accumulation. However, just before the last eruption, the ground deformation was stagnant for several years (Watanabe, 1988), and some precursor phenomena related to heat and volcanic gases were detected (Kagiyama and tsuji, 1987, hirabayashi et al., 1988). For this reason, we have started new thermal observation in the "Research on Monitoring and Forecasting of Volcanic Activities" of the current five-year plan (FY2019-2023). For the former, we have been installed ground surface heat balance observation since FY2019 (Onizawa et al., Volcanological Society of Japan, 2020), and for the latter, we are working on airborne thermal infrared observations using UAVs starting in the second half of FY2020. In this presentation, we will report the outline and current status of our airborne thermal infrared observation in Izu-Oshima.
The UAVs used for the observations were DJI Matrice 210RTK and 300RTK, and the thermal camera was DJI Zenmuse XT2. The observation area was 850m square, including the crater at the top of Mt. Mihara. The UAVs took off from near the crater observatory and from the Meteorological Research Institute observation station near a hot spring hotel located about 3 km north-northeast of the crater. In addition, the observations were carried out after complete sunset to reduce the effects of solar radiation. All UAV flight operations and photography were outsourced to a contractor, and in order to create an orthomosaic image of the target area, an automatic navigation route was created in advance so that the image could be acquired at a resolution of 30 cm/pixel (at an altitude of about 150 m above ground level) with an overlap/sidelap ratio of 90% of the recommended value of the analysis software (Pix4Dmapper). The target area was observed by dividing the area in consideration of the UAV take-off and landing locations and the battery life of the aircraft used. The data was acquired in JPEG format (R_JPEG) with temperature information including EXIF of location information. In order to ensure the accuracy of the temperature data, the data were also observed from several altitudes above the ground using the heat balance monitoring equipment and the copper plate prepared at the time of the observation as known temperature points.
The main observation results are shown in the Appendix table. In January and December, there were clear location (altitude) acquisition errors on some routes in the target area. In August, weather conditions caused data loss in some of the target areas. We are still studying how to deal with these errors and how to correct the temperature data, we have created an orthomosaic image of the summit crater area for each observation day using about 3100 to 4000 data. As a result, the maximum temperatures at the floor of the summit crater and B crater were detected to be about 28-44°C and 36-42°C, respectably (each uncorrected). In the future, we will further examine the correction method of the obtained temperature data and calculate the surface heat heat discharge to confirm its usefulness for volcanic activity evaluations.
The UAVs used for the observations were DJI Matrice 210RTK and 300RTK, and the thermal camera was DJI Zenmuse XT2. The observation area was 850m square, including the crater at the top of Mt. Mihara. The UAVs took off from near the crater observatory and from the Meteorological Research Institute observation station near a hot spring hotel located about 3 km north-northeast of the crater. In addition, the observations were carried out after complete sunset to reduce the effects of solar radiation. All UAV flight operations and photography were outsourced to a contractor, and in order to create an orthomosaic image of the target area, an automatic navigation route was created in advance so that the image could be acquired at a resolution of 30 cm/pixel (at an altitude of about 150 m above ground level) with an overlap/sidelap ratio of 90% of the recommended value of the analysis software (Pix4Dmapper). The target area was observed by dividing the area in consideration of the UAV take-off and landing locations and the battery life of the aircraft used. The data was acquired in JPEG format (R_JPEG) with temperature information including EXIF of location information. In order to ensure the accuracy of the temperature data, the data were also observed from several altitudes above the ground using the heat balance monitoring equipment and the copper plate prepared at the time of the observation as known temperature points.
The main observation results are shown in the Appendix table. In January and December, there were clear location (altitude) acquisition errors on some routes in the target area. In August, weather conditions caused data loss in some of the target areas. We are still studying how to deal with these errors and how to correct the temperature data, we have created an orthomosaic image of the summit crater area for each observation day using about 3100 to 4000 data. As a result, the maximum temperatures at the floor of the summit crater and B crater were detected to be about 28-44°C and 36-42°C, respectably (each uncorrected). In the future, we will further examine the correction method of the obtained temperature data and calculate the surface heat heat discharge to confirm its usefulness for volcanic activity evaluations.