No abstract

higher than previous estimates based on low plate rate/low fault slip rate models and the relatively low rate of seismicity over the last century.

Abstract. Global Positioning System geodesy was used to measure surface deformation on Soufriere Hills volcano, Montserrat from October 6, 1995 to July 1, 1997 during initial dome growth and gravitational collapse. Our data from this period show non-axially symmetric horizontal displacements, and decreasing subsidence as a function of radial distance from the former topographic high of the volcanic edifice. Forward modeling suggests that surface deformation is caused by a shallow vertical dike (< 3 km), which expanded approximately I m, coupled with a deflating Mogi source at about 6 km depth. These inferred source parameters are in good agreement with independent observations of regional dike widths and preeruption magma storage depth.

A permanent Global Positioning System receiver at Casa Diablo Hot Springs, Long Valley Caldera, California was installed in January, 1993, and has operated almost continuously since then. The data have been transmitted daily to the Jet Propulsion Laboratory for routine analysis with data from the Fiducial Laboratories for an International Natural sciences Network (FLINN) by the JPL FLINN analysis center. Results from these analyses have been used to interpret the on going deformation at Long Valley, with data excluded from periods when the antenna was covered under 2.5 meters of snow and from some periods when Anti Spoofing was enforced on the GPS signal. The remaining time series suggests that uplift of the resurgent dome of Long Valley Caldera during 1993 has been 2.5 + 1.1 cm/yr and horizontal motion has been 3.0 + 0.7 cm/yr at S53W in a no-net-rotation global reference frame, or 1.5 _+ 0.7 cm/yr at S14W relative to the Sierra Nevada block.These rates are consistent with uplift predicted from frequent horizontal strain measurements. Spectral analysis of the observations suggests that tidal forcing of the magma chamber is not a source of the variability in the 3 dimensional station location. These results suggest that remotely operated, continuously recording GPS receivers could prove to be a reliable tool for volcano monitoring throughout the world.

No abstract