Mexico (MEX)


J. Garcia-Pelaez, R. Gee, R. Styron, V. Poggi


The covering the Mexico (MEX) was developed created by the GEM hazard team and within a project funded by Suramericana (Sura). The model was originally implemented in the OpenQuake (OQ) engine.

Information about the OQ model versions and input files can be found on the Results and Dissemination page.

Interactive Viewer

The viewer below depicts the seismic sources and hazard results in terms of PGA for a return period of 475 years. Click on the menu in the upper right corner to select the layer.


Tectonic overview

Mexico occupies the southern end of the North American plate, and includes the active plate boundaries on its western and southern margins. In the west, the Pacific plate translates to the NW relative to North America along an oblique (right-lateral and normal) transform and spreading ridge system located in the Gulf of California between the Baja Peninsula and the mainland. This system, which is the southern extension of the San Andreas fault in the US, is the source of frequent moderate to large magnitude earthquakes, particularly along the transform segments of the plate boundary. Its southern terminus is offshore of Guadalajara, where a triple junction exists, and to the south the Pacific Plate shares an extensional boundary with the Rivera and Nazca oceanic plates, which subduct to the northeast under the Mexican mainland. This subduction is characterized as a 'flat slab' system, as below ~50 km the subducting slab flattens out and moves at this depth for hundreds of kilometers below central Mexico before diving farther into the asthenosphere. As a consequence, the frequent in-slab earthquakes (which may be quite large) may be located a few tens of kilometers below major metropolitan areas in central Mexico. Coupled with the soft lacustrine sediments that many of these cities are built on (particularly the Mexico City metro region), ground shaking from these in-slab events as well as events on the subduction interface to the southeast can produce severe ground shaking and often result in great losses. Additional faults are present in the Trans-Mexican Volcanic Belt and farther north, though the faults are for the most part fairly small and the slip rates are quite low, which limits the hazard from them though they could be damaging locally in the event of an earthquake. The Chiapas region at the southern tip of Mexico is quite close to the Motagua-Polochic Fault System, the sinistral transform boundary between the North American and Caribbean plates; there is some component of distributed deformation in this region, and intraplate faulting in the Chiapas Fold and Thrust Belt could produce appreciable shallow seismicity.

Basic Datasets

Earthquake Catalogue

A harmonized catalogue to be used in PSHA calculations was created for Mexico using a wide collection of earthquake databases. The procedure performed to obtain this catalogue is similar to those used in other GEM studies by Weatherill et al. (2016). The resulting catalogue contains 40485 events with 3.0 ≤ M W ≤ 8.6 from 1502 to 2016 The catalogue was purged from fore- and aftershock sequences and possible seismic swarms, using the Gardner and Knopoff (1974) declustering algorithm and a space-time window (Uhrhammer, 1985) with the OpenQuake Hazard Modeller’s Toolkit (Weatherill, 2014).

Fault Database

A database of ~600 active shallow faults was compiled by GEM in the framework of this project. The dataset containing fault trace locations and attributes describing geometry and kinematics (i.e. slip rates, dip angle, etc.) of Mexican faults in a vector GIS format. This is the major input for the fault-based modelling during the PSHA analysis.

Ground Motion Database

A strong motion database was created for Mexico for the purpose of ground motion prediction equation (GMPE) selection. Data was initially collected from the following networks: UNAM, CICESE, and CIRES. We include only the data from the UNAM and CICESE networks in the strong motion database since the CIRES network is mostly limited to stations in Mexico City, strongly affected by site response and therefore not suitable for selecting GMPEs at a national scale. A total of 3057 and 161 3-component recordings were collected from the UNAM and CICESE networks, respectively (Figure X). Events were classified into different tectonic regions based on their locations. Event locations and magnitudes were taken from the catalogue developed for Mexico within this project (ccara_mexico_201711.hmtk). The stations were assigned Vs30 values based on topography and also by using geological conditions, if provided in the metadata.


Figure X - The GEM strong ground motion database for MEX.

Hazard Model

Seismic Source Characterisation

The source model component includes faults sources with a 3d geometry modelling shallow seismicity and subduction interface earthquakes/seismicity, gridded point sources accounting for shallow distributed seismicity in active and stable crust and 3d ruptures constrained within the volume of the slab describing the in-slab subduction seismicity.

Ground Motion Characterisation

The GMPE selection process for MEX involved three main steps. First, we pre-selected a set of about 10 candidate GMPEs from the literature for each tectonic region considered in the SSM. The pre-selection was performed using a subset of the well-established exclusion criteria proposed by Cotton et al (2006) and Bommer et al. (2010). This was followed by a comparison of the ground motion scaling of the pre-selected GMPEs using a suite of rupture scenarios consistent with the ruptures modelled in the SSM (e.g. consistent in terms of magnitude, distance, style of faulting, etc). Such comparisons (referred to hereinafter as trellis plots) allowed for identifying and excluding GMPEs that behave unfavourably, for example during extrapolation outside the suggested applicability range. The final step of the selection process involved comparison between the ground motions computed by the pre-selected GMPEs and the ground motions observed in the region. Data-to-model comparisons were performed by analysing the ground motion residuals (e.g. Scherbaum et al., 2004; Stafford et al., 2008) using the OpenQuake strong motion toolkit (Weatherill, 2014).

For the final selection we tried to achieve balance by selecting models that both over and underpredict the observed ground motions in each of the tectonic regions when possible, according to the results of the residual analysis. A notable result of the residual analysis was the observation of different ground motions for crustal events than expected for data recorded by the UNAM network (central and southern Mexico) compared to the CICESE network (northwestern Mexico). The final GMPE logic tree is shown in Table X. The GMPEs selected for active shallow crust are different for northwestern Mexico. Hence the logic tree distinguishes between five tectonic regions: Active Shallow Crust, Active Shallow Crust Ridge (northwestern Mexico), Subduction Interface , Subduction IntraSlab, and Stable Shallow Crust (northeastern Mexico).

Epistemic Uncertainties

For every tectonic region, epistemic uncertainty is considered by using multiple GMPEs, each with an associated logic tree weight.

Active Shallow Crust Ridge Weight
AbrahamsonEtAl2014 0.34
CauzziEtAl2014 0.33
ZhaoEtAl2006Asc 0.33
Active Shallow Crust Weight
BindiEtAl2014Rjb 0.34
CauzziEtAl2014 0.33
BooreEtAl2014 0.33
Stable Shallow Crust Weight
Campbell2003MwNSHMP2008 0.34
PezeshkEtAl2011NEHRPBC 0.33
Atkinson2008prime 0.33
Subduction Interface Weight
YoungsEtAl1997SInter 0.33
ZhaoEtAl2006SInter 0.33
MontalvaEtAl2016SInter 0.34
Subduction IntraSlab Weight
YoungsEtAl1997SSlab 0.34
AbrahamsonEtAl2015SSlab 0.33
Kanno2006Deep 0.33

Table X - GMPEs used in the MEX model.


Hazard curves were computed with the OQ engine for peak ground acceleration (PGA) and spectral acceleration (SA) at 0.2s, 0.5s, 1.0s, and 2s. The computation was performed on a grid of 32755 sites (spaced at approximately 10 km) with reference site conditions with shear wave velocity in the upper 30 meters (Vs30) of 760-800 m/s.

The hazard map for PGA corresponding to a 10% probability of exceedance in 50 years (475 year return period), can be seen using the interactive viewer. For a more comprehensive set of hazard and risk results, please see the GEM Visualization Tools.


Bommer JJ, Douglas J, Scherbaum F, Cotton F, Bungum H, Fäh D (2010) On the selection of ground-motion prediction equations for seismic hazard analysis. Seismol Res Lett 81:783–793

Cotton, F., Scherbaum, F., Bommer, J. J. and Bungum, H., (2006). Criteria for selecting and adjusting ground-motion models for specific target regions: Application to central Europe and rock sites, J. Seism., 10:2, 137-156.

Scherbaum, F., E. Delavaud, and C. Riggelsen (2009). Model selection in seismic hazard analysis: An information-theoretic perspective, Bull. Seismol. Soc. Am. 99, no. 6, 3234–3247.

Scherbaum F, Cotton F, Smit P (2004). On the use of response spectral-reference data for the selection and ranking of ground-motion models for seismic-hazard analysis in regions of moderate seismicity: the case of rock motion. Bull Seism Soc Am 94(6): 2164–2185

Weatherill, G. A. (2014). OpenQuake Ground Motion Toolkit - User Guide. Global EarthquakeModel (GEM). Technical Report.