HR: 0830h
AN: V11C-0511    [PDF]
TI: Temperature anomalies in the plumes of the August, 18 and August, 29, 2000 eruptions of Miyake Jima volcano (Japan) inferred from delays of GPS waves crossing them.
AU: * Houli\'e, N
EM: houlie@ipgp.jussieu.fr
AF: IPGP, Laboratoire de sismologie case 89 4, Place Jussieu cedex 05, PARIS, 75252 France
AU: Nercessian, A
EM: nerces@ipgp.jussieu.fr
AF: IPGP, Laboratoire de sismologie case 89 4, Place Jussieu cedex 05, PARIS, 75252 France
AU: Briole, P
EM: briole@ipgp.jussieu.fr
AF: IPGP, Laboratoire de sismologie case 89 4, Place Jussieu cedex 05, PARIS, 75252 France
AU: Murakami, M
EM: mccopy@gsi.go.jp
AF: GSI, 1, Kitasato, Ibaraki-ken 305, Tsukuba, 305 Japan
AB: Using the GAMIT software we processed seventy days of GPS data (30s sampling rate) collected by the GSI at four sites on Miyake Jima volcanic island (Japan) between June 27, 2000 and September 5, 2000. This period includes a large seismic swarm (June 27, 2000 - July 8, 2000) followed by several major paroxysms at the volcano crater (July 9, 10, 14, 15, August 29) producing a 1 km wide caldera. The medium term velocity of the stations coordinates, already published elsewhere, is maximum during the seismic swarm and corresponds to a large dyke intrusion mostly offshore west of the volcano. No anomalies are observed in the time series of the daily GPS coordinates for the days of the paroxysms. An epoch by epoch processing of those days, using a kinematic software shows that there is no deformation during the paroxysms themselves. We then examined epoch by epoch the path delay residuals of the GPS phases at each GPS station during the events. Those delays exceed 200 mm in some cases. As they cannot be explained by a temporal change of the stations coordinates, we conclude that the cause of these delays is the presence of the hot volcanic plume not modeled by the GPS data processing which assumes a homogenous troposphere. We used a classical seismic tomography algorithm (modified to handle 3D + time) to map the path delay anomaly in the plume as a function of time. We interpret the anomalous delays as temperature anomalies in the plume, assuming a normal pressure and a plume saturated in humidity. The maximum average temperature anomaly is 20$\deg$, a low value compared to what is currently proposed in the literature. Higher temperature should exist in the inner part of the plume, but the horizontal extension of this hot zone cannot be more than 50-100 m, otherwise the GPS data would detect it.
DE: 0370 Volcanic effects (8409)
DE: 7280 Volcano seismology (8419)
DE: 8404 Ash deposits
DE: 8414 Eruption mechanisms
DE: 8419 Eruption monitoring (7280)
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting