S43B-01
Seismicity and State of Stress in the Subduction Zone of Guerrero, Mexico.
Until recently it was thought that moderate and large inslab earthquakes below Mexico occurred downdip from the strongly-coupled part of the interface. In the last 15 years, however, there have been several moderate/large inslab earthquakes with normal or thrust mechanism with one nodal plane characterized by steep dip. These earthquakes are located below or close to the downdip edge of the coupled plate interface. In order to understand the relationship between these earthquakes, and the state of stress and the geometry of the subducted Cocos plate, we study small to moderate events (M≥4) along the Guerrero seismic zone. Good station coverage in this region permits an accurate determination of depth and focal mechanism. Broadband seismograms and accelerograms (1995-present) were used to relocate earthquakes. Focal mechanisms were obtained from waveform modeling using local or regional data. Apart from the usual shallow thrust events located along the plate interface, we recorded three types of events: (1) shallow normal-faulting earthquakes in the upper continental crust, (2) steeply-dipping normal or reverse-faulting earthquakes, and (3) intermediate-depth normal- faulting earthquakes. Shallow normal events are related to a seismic sequence that started with an Mw 5.8 earthquake in 2001. Steeply-dipping events are located at a depth between 25 and 40 km, landward of the shallow thrust events. Normal-faulting earthquakes cluster around 180 km from the trench and occur at a depth between 50 and 60 km depth. Shallow normal-faulting events indicate a trench-normal tension in the upper plate. This state of stress in the upper plate near a convergent margin may be explained by tectonic erosion and/or backward retreat of the trench. The steeply-dipping mechanisms seem related to the unbending of the Cocos plate. Normal- faulting events, at 50-60 km depth, are located close to where other studies identify the commencement of steep bending of the Cocos plate and its penetration into the continental mantle. There is a complete lack of seismicity (even at a low-magnitude level) in the region between the steeply-dipping and the intermediate-depth normal-faulting events where the slab is subhorizontal (L~100 km). Seismic activity on the subducted Cocos plate is related to bending or unbending of the plate.
S43B-02
Geometry of Cocos Plate Beneath Guerero Using Receiver Functions
The MesoAmerican Subduction Experiment (MASE) began on December 2004, and its goal is to model the dynamic behavior of the subducting Cocos Plate beneath the North American Plate. One hundred broadband seismic stations were installed along a line that starts in Acapulco and ends close to the Gulf of Mexico. Our particular objective is to find the geometry along its southern segment (38 stations) by means of a receiver function analysis. Receiver functions for each station were ordered by back azimuth, distance, and ray parameter. Stacked profiles in the back azimuth ranges (90°-180°, 180°-270°, 270°-360°) were obtained to get arrival times of the converted Ps corresponding to the oceanic crust boundaries (top and bottom) and the continental Moho. Results show that the slab has three dip variations: from coast to ~70 km inland, the slab dips at ~16°, then up to ~130 km, it dips at ~3°; after that point, the downgoing slab becomes almost horizontal until ~210 km from the coast; finally, it shows an increases to ~30°, sinking into the mantle. In addition, with the slab geometry we observe, receiver functions pulses from earthquakes coming from a back azimuth between 270° to 360° show arrival time delays. Among other possibilities, this suggests a possible lateral dip component to the slab.
S43B-03
Determination of a three-dimensional velocity structure for the Southeastern of Mexico, by means of seismic ray tracing.
The objective of the present study is to obtain a three-dimensional velocity structure for three different tectonic provinces (Oaxaca, Chiapas and north of Guatemala). The Southeastern of Mexico is a seismic active region, in which several geologic structures of great importance are located: the Tehuantepec ridge, the Central-American volcanic arc, the Chiapas batholit, the extension of the Motagua-Polochic fault system, and also the existance of complex tectonostratigrafic terrenes at cortical level. In this area of study, there are a considerable number of tectonic studies and cortical velocity models (1D). For this reason is desired to obtain a three-dimensional realistic velocity model that agrees with the results obtained in previous studies. To make it posible, a preliminary velocity model has been proposed and has been discretized, and is now at the test stage. Also the geometry of the Cocos plate is determined (variation of the subduction angle) and we will try to obtain the interaction between the Motagua-Polochic fault system and the previously described subduction provinces. We will use P and S waves, from local and regional earthquakes from 1994 to 2004 reported by National Seismological Service (SSN) in eight broadband seismic stations in the Southeastern of Mexico (CCIG, CMIG, EVV, HUIG, OXIG, SCX TPX, TUIG). In the study earthquake relocalizations with the DD method will be performed, and a 3-D ray tracing will be used to test the seismic model. The 3-D velocity model will allow us to better understand the wave propagation characteristics, and apply them to the mitigation of the seismic risk in the region. 1. Earth Science Graduate Program. UNAM. 2. Institute of Geophysics. UNAM.
S43B-04
The 2003 Armeria, Colima, Mexico, Earthquake.
The January 22, 2003 Mw 7.4 Armeria earthquake occurred off the Pacific coast of the state of Colima, México, near of the town of Armería. Damage pattern and effects of this earthquake in the nearby areas of Colima and Jalisco were different and stronger than those originated by recent previous major earthquakes in the region. This earthquake and its aftershocks were registered the two local telemetric seismic networks, the Red Sísmica Telemétrica de Colima (RESCO) and the Red Sísmica Digital Telemétrica de Jalisco (RESJAL). After 72 hours of the mainshock a portable digital seismic network with seven 3D seismographs was deployed in the epicentral area to study the aftershock sequence. In the first 72 hours after the mainshock no important seismicity was located on the plates interphase, or in the Rivera Plate, with epicenter located west of Armeria River, which is the western border of the Colima graben, outside of the Colima Gap, nevertheless our results for focal mechanism for the mainshock and aftershock distribution for the first 72 hours do not agree with the various teleseismic solutions proposed. After the first 72 hours more than 200 aftershocks were located, most of them with local magnitude less than 3.0. The seismicity near the epicentral region remained unchanged constant for this period. However, the depths of aftershocks increase and most of them are located on the hypothetical interface between Riviera and North America Plates assuming a dip angle of 12°. A composite focal mechanism for the aftershocks located in both periods shows a inverse solution which change in time. The regional data analyzed indicates that the earthquake is a shallow intraplate event.
S43B-05
Evaluation of noise and site effect at MASE stations
The MesoAmerican Subduction Experiment installed 100 broadband seismic stations along a line between Acapulco and Tampico, crossing Mexico City at its middle point. Its main objective is the modelling of the subduction of the Cocos plate beneath North America. Several studies are being conducted, such as those regarding seismic source, but our objective is to identify site effects and noise levels and its sources at MASE stations, in order to evaluate the site and signal quality. We observe that most of the sites have noise levels within those from Peteterson (1993). At all stations the tide noise is clear. For those stations closer to the coast, the noise from the ocean waves is clear. Those installed within cities, large towns or close to the freeway have a difference in noise levels between day and night, either at a specific frequency or within a frequency range. Almost all of the MASE stations show some site effect, which is mainly a consequence of the criteria used in the site selection, since its location with respect to the line, security of the equipment, access to the station and line of sight (for those that transmit in real time) had a higher priority than the site quality itself. Therefore, most of the stations show an amplification at high frequencies (>1 Hz), and some have a resonant frequency. Despite the large site effect, few of them show poor noise levels, and on the contrary, a few with little site effect, show large noise levels, particularly during the day. Our results will benefit other studies that need to be able to take into account site effects and signal to noise ratio for their particular objectives, in order to select optimal stations and/or signals.