A REVIEW OF RECORDED TSUNAMIGENIC EARTHQUAKES IMPACTING MANZANILLO, MEXICO AND EL SALVADOR
Most large, shallow earthquakes in subduction zones are capable of generating tsunamis. While truly devastating tsunamis are relatively rare, events such as the 2004 Sumatra-Andaman and 2011 Tohoku earthquakes have shown their potential to cause massive destruction and result in tens or even hundreds of thousands of casualties. Tsunamis typically result from shallow undersea earthquakes that displace large volumes of water. Tsunami earthquakes (Kanamori, 1972) are characterized by the generation of disproportionately large tsunamis relative to their seismic magnitude as inferred from strong ground motion. These events exhibit several distinct seismological features: slip occurs very close to the trench, rupture propagates at unusually slow speeds, high-frequency seismic radiation is deficient, and the moment-scaled radiated energy is low (7×10^(-7)to 3×10^(-6), Venkataraman and Kanamori, 2004). Collectively, these characteristics suggest a rupture process fundamentally different from typical megathrust earthquakes.
El Salvador is located in a highly seismically active region and has experienced 15 tsunamis between 1859 and 2012, 9 of which occurred during the 20th century. All of these tsunamis were triggered by earthquakes, with two particularly destructive events: one in 1902 that impacted the eastern coast and another in 1957 that affected Acajutla (Gonzalez-Riancho et al., 2014). The earthquake on August 27, 2012, exhibited a prolonged source time function, low moment-scaled radiated energy, and a depleted short-period spectral content—characteristics consistent with tsunami earthquake behavior (Ye, Lay, and Kanamori, 2013). Borrero et al. (2014) confirmed that this event generated a surprisingly large local tsunami, reinforcing its classification as a tsunami earthquake.
The Pacific coast of Mexico, particularly along the Mesoamerican Trench from Jalisco to Chiapas, is also highly susceptible to tsunamis due to subduction-related seismic activity. One notable case is the 1932 Colima earthquake sequence in western Mexico. Okal and Borrero (2011) conducted a detailed seismological analysis of this sequence, which began with a mainshock on June 3 (MPAS8.1), followed by aftershocks on June 18 (MPAS7.9) and June 22 (MPAS6.9). Interestingly, the June 22 aftershock generated a more destructive tsunami than the mainshock, despite having a significantly lower magnitude. Okal and Borrero (2011) concluded that this aftershock exhibited key characteristics of a tsunami earthquake.
This study aims to propose realistic tsunamigenic earthquake scenarios for risk mitigation efforts in both Mexico and El Salvador. To this end, we reviewed available data to identify events that have generated tsunamis affecting Manzanillo and El Salvador. We then analyzed the relationship between radiated energy and seismic magnitude, using the concept of scaled energy to identify events with anomalously low values.