Forearc kinematics in obliquely convergent margins: Examples from Nicaragua and the northern Lesser Antilles.

摘要

In this study, I use surface velocities derived from GPS geodesy, elastic half-space dislocation models, and modeled Coulomb stress changes to investigate deformation in the over-riding plate at obliquely convergent margins at the leading and trailing edges of the Caribbean plate. The two principal study areas are western Nicaragua, where the Cocos plate subducts beneath the Caribbean plate, and the northern Lesser Antilles, where the North American plate subducts beneath the Caribbean plate.In Nicaragua, plate convergence is rapid at 84 mm yr1 with a small angle of obliquity of 10° along a slightly concave portion of the Middle America Trench. GPS velocities for the period from 2000 to 2004 from sites located in the Nicaraguan forearc confirmed forearc sliver motion on the order of &sim14 mm yr1 in close agreement with the value predicted by DeMets (2001). These results are presented here in Chapter 3 and were reported in Geophysical Research Letters (Turner et al., 2007). GPS observations made on sites located in the interior and on the eastern coast of Nicaragua during the same time period were combined with new data from eastern Honduras to help better constrain estimates of rigid Caribbean plate motion (DeMets et al., 2007). Slip approaching the plate convergence rate along the Nicaraguan and El Salvadoran sections of the Middle America Trench was quantitatively demonstrated by finite element modeling of this section of the plate interface using GPS velocities from our Nicaraguan network together with velocities from El Salvador and Honduras as model constraints (Correa-Mora, 2009).The MW 6.9 earthquake that ruptured the seismogenic zone offshore Nicaragua on October 9, 2004 resulted in coseismic displacements and post-seismic motion at GPS sites in the central part of the Nicaraguan forearc that currently prevent extension of interseismic time-series in this region. An elastic half-space dislocation model was used to estimate coseismic displacements at these sites and to qualitatively examine the observed post-seismic motion. Coseismic and post-seismic motion in this portion of the forearc indicate that long-term motion of the forearc across the earthquake cycle may proceed in a zig-zag pattern, which may contribute to east-west extension as observed in the Managua graben. Sites to the northwest and southeast were not substantially effected by the earthquake, and longer duration time-series (&sim7 yrs) from these areas support the earlier estimates of forearc sliver motion. Results from our analysis of Nicaraguan GPS time-series from 2004--2008 and our earthquake modeling efforts are discussed in Chapter 4 and will form the basis of an article to be submitted for publication.In the northern Lesser Antilles, plate convergence is slow at &sim2 cm yr1, and obliquity varies substantially along the convex Lesser Antilles Subduction Zone. In chapter 5, I present GPS velocities derived from a decade of observations on sites in the northern Lesser Antilles and Virgin Islands. The velocities support forearc sliver motion on the order of &sim2--3 mm yr1 consistent with the lower value estimated by Lopez et al. (2006), indicating convergence in the northern region is only partially partitioned. GPS velocities in the northern Lesser Antilles show considerable variation between islands, suggesting possible independent block motion and internal deformation within the forearc region, however, velocity uncertainties for some sites remain high and may reflect the low signal-to-noise ratio of our residual velocities for the region. The lack of a substantial arc-normal component of shortening is similar to that seen in Nicaragua and may indicate a small amount of locking along the subduction interface with forearc sliver motion being driven from a more fully locked region south of Guadeloupe where the Barracuda and Tiburon aseismic ridges impinge on the subduction zone. (Abstract shortened by UMI.)