Magnitude 6.4 Earthquake off the coast of the Mexico/Guatemala border.

The United States Geological Survey recorded a Magnitude 6.4 Earthquake at a depth of 75.4 km, approximately 17 km off the south coast of the border between Mexico and Guatemala, slightly before 5.40 am local time (slightly before 11.40 am GMT) on Sunday 12 May 2024. This even was felt across much of southern Guatemala, and Chiapas State, Mexico, and triggered a number of small landslides as well as minor damage to some buildings, but there are no reports of any injuries.

The approximate location of the 12 May 2024 Mexico/Guatemala Earthquake. Contour lines show rates of movement during the quake, the red line is the Middle American Trench. USGS.

Mexico is located on the southernmost part of the North American Plate. To the south, along the Middle American Trench, which lies off the southern coast off Mexico, the Cocos Plate is being subducted under the North American Plate, passing under southern Mexico as it sinks into the Earth. Guatemala is located on the southern part of the Caribbean Plate, close to its boundary with the Cocos Plate, which underlies part of the east Pacific. The Cocos Plate is being pushed northwards by expansion of the crust along the East Pacific Rise, and is subducted beneath the Caribbean Plate along the Middle American Trench. This is not a smooth process, and the plates frequently stick together then break apart as the pressure builds up, causing Earthquakes on the process. 

The position of the Cocos, Nazca and Rivera Plates. MCEER/University at Buffalo.

The Cocos Plate is thought to have formed about 23 million years ago, when the Farallon Plate, an ancient tectonic plate underlying the East Pacific, split in two, forming the Cocos Plate to the north and the Nazca Plate to the south. Then, roughly 10 million years ago, the northwesternmost part of the Cocos Plate split of to form the Rivera Plate, south of Beja California.

In a paper published in the Journal of Geophysical Research, in 2012, a team led by Igor Stubailo of the Department of Earth and Space Sciences at the University of California Los Angeles, published a model of the subduction zone beneath Mexico using data from seismic monitoring stations belonging to the Mesoamerican Seismic Experiment, the Network of Autonomously Recording Seismographs, the USArray, Mapping the Rivera Subduction Zone and the Mexican Servicio Sismologico Nacional.

The seismic monitoring stations were able to monitor not just Earthquakes in Mexico, but also Earthquakes in other parts of the world, monitoring the rate at which compression waves from these quakes moved through the rocks beneath Mexico, and how the structure of the rocks altered the movement of these waves.

Based upon the results from these monitoring stations, Stubailo et al. came to the conclusion that the Cocos Plate was split into two beneath Mexico, and that the two plates are subducting at different angles, one steep and one shallow. Since the rate at which a plate melts reflects its depth within the Earth, the steeper angled plate melts much closer to the subduction zone than the shallower angled plate, splitting the Trans-Mexican Volcanic Belt into sections above the different segments of the Cocos Plate, and causing it to apparently curve away from the subduction zone.

Top the model of the Cocos Plate beneath Mexico, split into two sections (A & B) subducting at differing angles. (C) Represents the Rivera Plate, subducting at a steeper angle than either section of the Cocos Plate. The Split between the two has been named the Orozco Fracture Zone (OFZ) which is shown extended across the Cocos Plate; in theory this might in future split the Cocos Plate into two segments (though not on any human timescale). Bottom Left, the position of the segments on a map of Mexico. Darker area is the Trans-Mexican Volcanic Belt, orange circles are volcanoes, brown triangles are seismic monitoring stations, yellow stars are major cities. Bottom Right, an alternative model showing the subducting plate twisted but not split. This did not fit the data. Stubailo et al. (2012).

See also...