Modeling of 3D crustal shear structures from compliance measurements near East Pacific Rise 9°50'

Measurements of the seafloor deformation under ocean wave loading (seafloor compliance) in the infra-gravity band (0.002-0.03 Hz) reveal the distribution of melt in mid-ocean ridge magma bodies through the dependence of the compliance signal on shear modulus. We have studied the sensitivity of the compliance signal to crustal structures in three dimensions using 3-D finite element modeling. We model a series of compliance measurements near the ridge crest at 9°48'N along the East Pacific Rise from 1994, 1999, 2001 and 2007. This time series reveal significant variation in the distribution of melt in the crust and uppermost mantle beneath the site from 1994 to 2007 with most of the changes occurring following the eruption in 2005-2006. We show new crustal models from beneath the EPR 9°48 site and also from the 9°08 overlapping spreading center. The new 3-D modeling replaces previous 1-D modeling for inversion of the data. The new modeling suggests that the compliance results are primarily sensitive to large scale, low velocity zones, and not to thin melt lenses of dimensions comparable to the AMC reflector. This is different from our previous predictions using 1D modeling, that suggested a narrow, 100-meter-thick zone of pure melt in the lower crust could be identified from anomalous compliance values. Previous modeling identified the compliance peak at low frequencies (5-10 mHz) with a low velocity zone in the mid to lower crust (4-6km from the seafloor). The new modeling shows this peak is consistent with a deep low velocity zone but shows it must be at least 15km wide to be as large as detected in the compliance measurements. Therefore the measurements from before 1999 indicate a wide low velocity zone beneath 9°48. The 2007 measurements reflect a significant decrease in partial melt volume both shallow and deep in the crust. In addition, we have investigated the dependence of the compliance response to ocean waves from different directions (along and across ridge axis) and show there is a dependence on azimuth to the compliance response. Compliance response from across axis directed waves reflect more strongly the 2D across section structure, whereas the signal from an along axis directed waves is more sensitive to along axis structure. Thus we can explain the high compliance observed in 13-15 mHz at multiple sites as partly the result of the along-axis connectivity of melt region, rather than to a wide melt lens. The results suggest extended compliance measurements coupled with infra-gravity wave directional information (obtained from pressure arrays) could allow separation of along and across axis compliance signals making it possible to better invert for 3D crustal models using a small number of compliance sites.