Hiromu Okada

Elastic models for the magma intrusion associated with the 2000 eruption of Usu Volcano, Hokkaido, Japan

Abstract After 23 years of dormancy, Usu Volcano (Hokkaido, Japan) erupted on March 31, 2000. Many observations (seismicity, deformation rates, gravity data, groundwater level monitoring) show that the period of intense activity was short, starting abruptly, and continuing for ca. 5 months with a decreasing rate. Uplift was observed at two successive and separate locations at the time of the eruption. We obtained GPS and microgravity data at Usu Volcano for two intervals, the first from August 1996 to July 1998, once every 2–4 months, and the second in November 2000, 2 months after the end of the eruption. Between July 1998 and November 2000, the displacements and gravity variations are among the largest ever recorded on an active volcano in association with an eruption. We review three different elastic models commonly used in volcano-geodesy (sphere, fault system, fissure zone) and invert the high-quality data using each of these models. The combined inversion of GPS and microgravity data leads to the best solution in the least-squares sense. It is compatible with the intrusion of approximately 5×10 11 kg of new magma into the western part of Usu Volcano. This appears to have occurred in a subvertical fracture zone (about 2.4 km length, 0.1 km width) aligned in the east–west direction. The fracture zone is between 0.4 and 3.3 km depth with an extension of about 30 m. The fractures are likely to be filled with material having a density slightly higher than the density of old products of Mount Usu, i.e. about 2400 kg m −3 . This model is consistent with the locations and magnitudes of the earthquakes recorded during the period of intense seismic activity in April and May 2000. These earthquakes correspond to the boundaries of the intruded magma body. The model suggests that the two locations of uplift are not independent.

Post-eruptive volcanic dome evolution as revealed by deformation and microgravity observations at Usu volcano (Hokkaido, Japan)

Abstract Usu volcano (Hokkaido, Japan) is a dacitic volcano, known for its high production rate of lava domes and crypto-domes. It is thus a good target to study processes of volcanic dome evolution (upheaval and/or relaxation). We carried out repeated GPS and microgravity surveys on the three most recent domes of Mt. Usu (1910: Meiji Shinzan; 1943–1945: Showa-Shinzan and 1977–1982: Usu-Shinzan). The repeat period was 1 to 2 months and extended from October 1996 to June 1997. We also compare new data with results from former studies. More than 20 years after the start of Usu-Shinzan dome growth, there is still subsidence at a maximum rate of about 7 to 8 cm/year. The reasons for this subsidence are discussed. Repeated gravity surveys revealed an increase of gravity on the domes (about 60±10 microgal/year for Usu-Shinzan, about 15 microgal at Showa-Shinzan and 10 to 20 microgal for Meiji-shinzan); this gravity increase exceeds that expected due to subsidence. We discuss and interpret the excess gravity change in terms of a density increase in the edifice, caused by a combination of processes (contraction of the edifice, water level change, devesiculisation, cooling and magma intrusion). Quantification of these processes at Usu volcano may help to understand the processes of evolution at domes on other volcanoes such as Merapi (Indonesia), Unzen (Japan) or Montserrat (West Indies).

Seismic experiments on Showa-Shinzan lava dome using firework shots

Seismic experiments were conducted on Showa-Shinzan, a parasitic lava dome of volcano Usu, Hokkaido, which was formed during 1943–1945 activity. Since we found that firework shots fired on the ground can effectively produce seismic waves, we placed many seismometers on and around the dome during the summer festivals in 1984 and 1985. The internal structure had been previously studied using a prospecting technique employing dynamite blasts in 1954. The measured interval velocity across the dome in 1984 ranges 1.8–2.2 km/s drastically low compared to the results (3.0–4.0 km/s) in 1954; in addition, the velocity is 0.3–0.5 km/s higher than that in the surrounding area. The variation of the observed first arrival amplitudes can be explained by geometrical spreading in the high velocity lava dome. These observations show a marked change in the internal physical state of the dome corresponding to a drop in the measured highest temperature at fumaroles on the dome from 800°C in 1947 to 310°C in 1986.