GPS measurements of crustal deformation in Guerrero, southern Mexico, include surveys collected between 1992 and 2001 as well as continuous GPS measurements at a few sites. These geodetic observations are used to calculate interseismic deformation rates and assess the presence and possible location of transient deformation during the period encompassing 1992.25 to 2001.75. The data are used to examine transient deformation in 1998 previously described from data at a single site by Lowry et al. [2001]. Survey measurements and continuous data from a site near Popocatépetl volcano confirm the 1998 transient, and survey data also suggest another transient occurred following the 14 September 1995 (Mw = 7.3) Copala earthquake. All of the available GPS position estimates have been inverted for a combined model of slip during each event plus the steady state slip on the plate interface. Modeling of the steady state deformation rates confirms that the Guerrero seismic gap is partially frictionally locked at depths shallower than about 25 km and accumulating strain that may eventually be released in a great earthquake. The data also suggest that there is frictional coupling to much greater (>40 km) depths, which releases more frequently in aseismic slip events. The locations and sizes of the transient events are only partially constrained by the available data. However, the transient models which best fit the GPS coordinate time series suggest that aseismic slip was centered downdip of the seismogenic portion of the plate‐bounding thrust in both events, and the moment release had equivalent magnitudes Mw = 7.1 + 1.3/−1.0 in 1995–1996 and 7.1 + 0.4/−0.1 in 1998.

SUMMARY We use horizontal and vertical crustal displacements derived from GPS measurements at 26 sites in western Mexico to study the coseismic and post‐seismic kinematics and dynamics of the 1995 October 9 (Mw=8.0) Colima–Jalisco earthquake along the Middle America Trench. The measurements bracket the entire landward edge of the approximately 150 km long rupture zone and span a 4 yr period for most sites. We solve for the temporal evolution of slip along the subduction interface by inverting GPS displacements for the coseismic and four post‐seismic intervals (March 1995–March 1999), subject to the assumption that the crust responds elastically to slip along a shallow‐dipping, curved subduction interface. Coseismic rupture of up to 5 m was largely focused above depths of 20 km and was limited to a 120–140 km long segment of the subduction zone. Within one week of the earthquake, post‐seismic slip migrated downdip to depths of 16–35 km, where it has since decayed logarithmically. We also find evidence for shallow aseismic slip during 1996 or early 1997 northwest of the coseismic rupture zone and increasingly widespread relocking of shallow regions of the subduction interface after early 1997. The relative lack of afterslip in shallow regions of the subduction interface suggests that the interface lies in the unstable frictional regime and hence is strongly coupled between earthquakes. By 1999, the cumulative slip moment associated with post‐seismic slip equaled ∼70 per cent of the coseismic moment, with nearly all of this slip occurring downdip from the coseismic rupture zone. The migration of slip after the earthquake to a deeper and presumably velocity‐strengthening area of the subduction interface and the logarithmic decay of afterslip conform to the qualitative and quantitative predictions of a model in which the fault kinematics are prescribed by rate‐ and state‐variable frictional laws. However, misfits to the geodetic displacements exceed the average displacement uncertainties for all epochs, implying one or more of the following: (1) the elastic response is heterogeneous due to slip along unmodelled upper crustal faults or variations in the elastic properties of the crust; (2) other post‐seismic mechanisms such as viscoelastic or poroelastic effects contribute to or possibly dominate the post‐seismic response; (3) we have underestimated the uncertainties in the GPS displacements.

Modeled and measured velocities at coastal sites in Baja California south of the Agua Blanca fault, a region that most previous models consider Pacific plate, differ by 3-8 mm/yr, with coastal sites moving slower that the Pacific plate. We interpret these discrepancies in terms of strain accumulation on known on-shore faults, combined with right lateral slip at a rate of 3-4 mm/yr on additional faults offshore peninsular Baja California in the Pacific. Offshore seismicity, offset Quaternary features along the west coast of Baja California, and a discrepancy between the magnetically determined spreading rate in the Gulf Rise and the total plate rate from a geological model provide independent evidence for a "Baja California shear zone."

Abstract. A finite-element model (FEM) incorporating ge-ologic properties characteristic of a subduction zone is compared with FEMs approximating homogeneous elastic halfspaces (HEHS)s to investigate the effect of heterogeneity on coseismic and postseismic deformation predictions for the 1995 Colima-Jalisco M•o=8.0 earthquake. The FEMs are used to compute a coefficient matrix relating displacements at observation points due to unit dislocations of contactnode pairs on the fault surface. The Green's function responses are used to solve the inverse problem of estimating dislocation distributions from coseismic GPS displacements. Predictions from the FEM with heterogeneous material properties, loaded with either of the HEHS dislocation distributions, significantly overestimate coseismic displacements. Postseismic deformation predictions are also sensitive to the coseismic dislocation distribution, which drives poroelastic and viscoelastic relaxation. FEM-generated Green's functions, which allow for spatial variations in material properties, are thus preferable to those that assume a simple HEHS because the latter leads to dislocation distributions unsuitable for predicting the postseismic response.

Abstract. The October, 1995 Mw 8.0 Jalisco subduction earthquake has provided a thorough geodetic observation of the coseismic subduction process. An 11 station regional GPS network located directly onshore of the rupture demonstrates consistent vertical subsidence verified by tide gauge data and southwest-directed extension, with measured displacements reaching I meter. Unusually shallow and non-uniform faulting is required to explain the displacements. We determine that up to 5 meters of slip occurred within the upper 15 km of the thrust fault zone and 2 meters possibly as shallow as 8 kin, and that slip was likely distributed in two main patches. The paucity of continental sediments in this subduction zone could be responsible for the anomalously shallow faulting.