Magma Dynamics at Axial Seamount, Juan de Fuca Ridge, from Seafloor Deformation Data
Wednesday, 17 December 2014
Axial Seamount is located about 480 km west of the Oregon coast at the intersection of the Cobb hotspot and the Juan de Fuca Ridge. Two eruptions have been observed since routine observations began in the 1990’s, one in January 1998 and the other in April 2011. Precise bottom pressure measurements have documented an inflation/deflation cycle within Axial’s summit caldera. The slow inflation observed at the center of the caldera was punctuated by sudden rapid deflation of 3.2 m during the 1998 eruption and 2.4 m during the 2011 eruption. Pressure data collected in September 2013 from continuously recording bottom pressure recorders and campaign-style measurements with an ROV indicates that Axial Seamount inflated 1.57 m from April 2011 to September 2013 at an average inflation rate of 61 cm/yr, meaning it had already recovered more than 65% of the deflation from the 2011 eruption within just 2.4 years. The geometry and location of the deformation source is not well constrained by the spatially-sparse pressure data, particularly for the most recent co-eruption deflation and post-eruption inflation signals. Here, we use geodetic data collected in September 2013 to test the fit of multiple numerical models of increasing complexity. We show that for this time period (since April 2011) neither a simple point deformation source (Mogi model) nor an oblate spheroid (penny-shaped crack) provide a good fit to the data. We then use finite element models to build more complex inflation geometries, guided by recent seismically imaged magma reservoirs, in an attempt to understand the source(s) of the observed deformation pattern. The recent seismic data provide good constraints on magma reservoir geometry and show the most robust melt occurs under the southeast part of the caldera at Axial. However, previous geodetic measurements at Axial have consistently shown a deformation source near the caldera center. We use numerical modeling to attempt to reconcile these differences.