Hydrothermal Activity Inferred from Comprehensive Observation of Unrest in Adatara Volcano

Numano-Taira Crater in Adatara volcano has shown signs of volcanic unrest such as hot mud effusion since 1996, even though no obvious changes in seismicity have been observed. We carried out observations of magnetism, deformation, gravity, and self-potential around the crater to detect volcanic activity. We found remarkable variation of up to 110 nT in total magnetic intensity, from which we inferred that demagnetization caused by rising temperatures occurred under the southeastern part of the crater bottom from around 1997 to 2000, and that magnetization increased because of falling temperatures under the northeastern part after 2000. By geodetic observations with GPS, we found inflation of the crater before 2000 and deflation after 2000. We made gravity observations taking into account the effect of the height change by using the GPS to precisely position a benchmark. A gravity increase was observed in the crater from 2001 to 2005, and the amount of change was larger than that which could be explained by the height change. We speculated that the change in gravity reflected variations in the groundwater thickness. In the self-potential observations, we maintained long-term stability by calibration with a Cu/CuSO4 electrode, and ground temperature was measured to correct for the drift of the electrode potential. The results revealed a self-potential fluctuation probably caused by small-scale hydrothermal activity. The results of these comprehensive observations suggest that the Numano-Taira Crater was fairly active from 1996 to 2000, and that it became gradually calmer after 2000. Because the observed volcanic unrest was possibly related to the activity of hot groundwater, we used a hydrothermal simulation to study the variations in water flow and temperature distribution beneath the crater that would be expected if hot water was supplied from depth. Assuming intrinsic hydraulic permeability beneath the crater, we carried out simulations for various cases. The results showed that two hot water supplies, one in the north and the other in the south, were needed to explain the temperature changes estimated from the magnetic observations. We also calculated the changes in the magnetic and gravity fields from the results of the hydrothermal simulation, and found that they were consistent with the observed values. For volcanoes such as Adatara, it is not easy to monitor the volcanic activity by seismic observations, but comprehensive observations of magnetism, g