Resistivity and self-potential changes associated with volcanic activity: The July 8, 2000 Miyake-jima eruption (Japan)

Abstract

The Miyake-jima volcano abruptly erupted on July 8, 2000 after 17 years of quiet and gave birth to a crater, 1 km in diameter and 250 m deep. This expected unrest was monitored during the years 1995–2000 by electromagnetic methods including DC resistivity measurements and self-potential (SP) surveys. Beneath the 2500 yr old Hatcho-Taira summit caldera audio-magnetotelluric soundings made in 1997–98 identified a conductive medium, 200–500 m thick (within the 50 Ω m isoline) located at a few hundred metres depth. It was associated with the active steady-state hydrothermal system centred close to the 1940 cone and extending southward. A DC resistivity meter set in a Schlumberger array with 600, 1000 and 1400 m long injection lines evidenced strong resistivity changes between September 1999 and July 3, 2000 in the vicinity of the newly formed crater. The apparent resistivity has reached about three times its initial values on the 1400 m long line and has lowered to about 20% on the 600 m line. Just prior to the July 8, 2000 eruption SP mapping made inside the summit Hatcho-Taira caldera revealed negative anomalies where positive ones had occurred during the previous tens of years. The largest negative anomaly, −225 mV in amplitude, mainly took place above the 1940 cone which collapsed in the crater formation. A permanent 1 km long SP line across the caldera suggests accelerating changes during the 3 months preceding the eruption. On a larger scale, the comparison between 1995 and 2000 surveys has shown a global increase of the hydrothermal activity beneath the volcano. Its source could have been 250 m to the south of the crater. These observations suggest that the hydrothermal system was slowly disturbed in the months preceding the eruption while drastic changes have occurred during the 2 weeks before the summit collapse when tectonic and volcanic swarms have appeared.