The large 2001–02 fault slip event began within weeks after a M_w=6.1 intraplate normal faulting event (the Coyuca earthquake) near the center of the region of early aseismic slip. GPS campaign data suggest that significant slip also occurred throughout Guerrero immediately after the 1995 M_w=7.3 Copala earthquake. It is possible that static stress change during the Coyuca and Copala earthquakes excited stress waves which propagated the aseismic events. I am currently developing InSAR interferograms to more closely examine the connection between the Copala earthquake and the 1995–96 Guerrero slip event, and in collaboration with John Anderson at U. Nevada-Reno, improving estimates of locations and other parameters of seismicity for the period in which we have GPS measurements. These will be combined in a numerical model of rate- and state-dependent friction to examine the relationship of geodetically observed slip to seismic events. Interestingly, there are no significant earthquakes anywhere on the Cocos-North Americ plate boundary in the months preceding the 1998 slow slip event, so I am also examining other possible triggering mechanisms (including resonant response to climate-driven and other mass loading processes that vary stress on the fault at the 6- to 25-month periods of slow slip recurrence).

Aseismic moment release requires a slip deficit in the accommodation of relative motion across the fault. I have developed inverse models of GPS and other geodetic data in Guerrero to estimate simultaneously the steady-state slip rate and transient slip. The model indicates that the aseismic slow slip events release a slip deficit accumulated during the interval between events. The model can also be used to evaluate seismic hazard. If one assumes that slip during the 1992-2005 period of GPS measurement is representative of the entire (1911-2005) aseismic period, the Guerrero gap segment of the subduction megathrust has accumulated sufficient potential energy to generate a M_w=7.9 earthquake. I also find that, if oblique Cocos-North America plate motion is partitioned into upper-plate strike-slip faulting, most of the ~8-10 mm yr^-1 strike-slip movement is accommodated on the Chapala-Oaxaca fault zone with a small contribution of 0-3 mm yr^-1 across the Atoyac fault.

Slip during aseismic events loads the up-dip, seismogenic portion of the fault by increasing the static shear stress. An important question that I hope to resolve via numerical modeling of dynamic friction relates to the conditions necessary for an aseismic slip event to generate a major earthquake rupture (instead of healing as the Guerrero events have done thus far). A second issue relates to why large fault slip events have been seen in Guerrero, Cascadia, Alaska and Japan but not in other subduction zones or on faults in other tectonic regimes. These locations are unusual in that they are all characterized by near-horizontal slabs at depths of 35-60 km: I.e., at transitional temperatures between pure stick-slip behavior and low-stress stable sliding. In Guerrero, nearly 100 km of down-dip fault surface is in the transitional slip regime, resulting in the potential for slip deficit over a broad area, and correspondingly, very large moment release. Faults that are more nearly vertical will likely have less than 10 km of surface in the transitional regime, so potential moment release would be proportionately smaller, and possibly too small to be seen with GPS (but still large enough to be observed with more sensitive instruments such as borehole strainmeters and long baseline tiltmeters).

This material is based upon work supported by the National Science Foundation under grant numbers 9725712, 0125618, and 0207820. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.