[S-TT40] Airborne surveys and monitoring of the Earth

[STT40-03] An aeromagnetic survey using an unmanned aerial vehicle over Azuma Volcano

*Hikari Yonakura^1,2, Masahiro Ichiki¹, Shigeo Okuma², Ayumu Miyakawa², Ryo Tanaka³, Toshiki Kaida¹, Ayuta Tsuge³, Toyonobu Ota⁴, Takeshi Hashimoto³ (1.Tohoku University, 2.Advanced Industrial Science and Technology, 3.Hokkaido University, 4.Tierra Tecnica)

Keywords:magnetic survey, alternation

Recent MT studies have shown that the location of volcanic alteration zone and volcanic earthquakes can be used to assess the imminence of an eruption. Therefore, a total intensity aeromagnetic survey using an unmanned aerial vehicle (UAV) was carried out over Oana Crater of Mt. Azuma, NE Japan on 7 and 9 September 2021, to infer the subsurface alteration zone beneath Mt Azuma. The observation equipment was the Tierra Technica GSMP35U-DR drone aeromagnetic measurement system composed of a GEM GSMP-35U potassium magnetometer and a DJI MATRICE 600 PRO. The survey area was a rectangle of 2 km by 1 km in the NNE-SSW and WNW-ESE directions, respectively, and the flight lines were arranged as 18 equidistant parallel lines in the WNW-ESE direction. The flight line altitudes varied, but the altitude of each line was kept constant. The wiggles from the set flight altitude within each flight lines was about ±0.15 m. The maximum altitude of sampling points was 1978 m above sea level. The sampling frequency was 20 Hz, and the flight speed was approximately 6 m/s. The geomagnetic reference station was installed about 600 m south of the Oana Crater, where the magnetic field was not affected by subsurface volcanic activity (e.g., Japan Meteorological Agency, 2020).

The raw data were resampled down to 10 Hz, and the main field and diurnal variations were subtracted from the down-sampled data using IGRF-13 and the reference station data. Magnetic anomaly distribution was obtained using the method of Nakatsuka and Okuma (2006). The average magnetization intensity was estimated using a statistical correlation method by Grauch (1987), and the average magnetization intensity was 2.06 A/m. We subtracted the contribution of the average magnetization intensity from the magnetic anomaly distribution, then used the obtained magnetic anomaly data as input for the following inversion.

We inverted the magnetic anomaly data into a three-dimensional subsurface magnetization perturbation from the average magnetization intensity using the effective source volume minimization method (Nakatsuka and Okuma, 2014). The horizontal model discretization width was 20 m*20 m. The vertical model block width was allotted based on the topography: The vertical block width was 20 m in 0-200m depth below the surface (bsf), 50 m in 200-700 m depth bsf, and 100 m in 700-1000 m depth bsf, respectively. The magnetization threshold to judge effective source in the inversion was fixed at 0.2 A/m. The inversion was iterated 20 times using the CG method with an initial model of the uniform zero magnetization perturbation. The result shows a weakly magnetized region below the Tsubakurozawa Crater. The weakly magnetized location coincides with the demagnetization source location modeled by using the repeating total magnetic intensity observations (Japan Meteorological Agency, 2020).