Abstract
In western Mexico, the subduction of the Rivera and Cocos plates beneath the North America plate has deformed and fragmented the overriding plate, forming several structural rifts and crustal blocks. To obtain a reliable subsurface image of the continental crust and uppermost mantle in this complex area, we used P-wave arrivals of local earthquakes along with the Fast Marching Method tomography technique. We followed an inversion scheme consisting of (1) the use of a high-quality earthquake catalog and corrected phase picks, (2) the selection of earthquakes using a maximum location error threshold, (3) the estimation of an improved 1-D reference velocity model, and (4) the use of checkerboard testing to determine the optimum configuration of the velocity nodes and inversion parameters. Surprisingly, the tomography results show a very simple δVp distribution that can be described as being controlled by geologic structures formed during two stages of the separation of the Rivera and Cocos plates. The earlier period represents the initial stages of the separation of the Rivera and Cocos plates beneath western Mexico; the later period represents the more advanced stage of rifting where the Rivera and Cocos plates had separated sufficiently to allow melt to accumulate below the Colima Volcanic complex. During the earlier period (14 or 10–1.6 Ma), NE–SW-oriented structures/lineaments (such as the Southern Colima Rift) were formed as the two plates separated. During the second period (1.6 Ma to the present), the deformation is attributed to magma, generated within and above the tear zone between the Rivera and Cocos plates, rising beneath the region of the Colima Volcanic Complex. The rising magma fractured the overlying crust, forming a classic triple-rift junction geometry. This triple-rift system is confined to the mid- to lower crust perhaps indicating that this rifting process is still in an early stage. This fracturing, along with fluid circulation and associated heat advection within the fractures, can easily explain the observed distribution of δVp, as well as many of the results of previous seismological studies. Also surprisingly, we find no evidence at deep crustal depths to support either a trenchward migration of the volcanic arc or toroidal asthenospheric flow through the slab tears bounding the Jalisco Block to the NW and SE.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have