2017 멕시코시티 지진: 1985 재해 기념일에 발생한 M7.1

13:14 CDT, September 19: The Eerie Anniversary

September 19, 2017 began in Mexico City with an unusual ritual. At 11:00 AM, the city conducted its annual earthquake drill — the same drill held every September 19 since 1985, the anniversary of the catastrophic earthquake that killed at least 5,000 people and perhaps many more in the Mexican capital. Sirens wailed, residents filed into the streets from their offices and apartments, and for a few minutes the megacity of 21 million people practised the choreography of earthquake response.

Two hours and fourteen minutes later, the choreography became real.

At 1:14 PM local time, a magnitude 7.1 earthquake struck with its EpicenterThe point on the Earth's surface directly above the hypocenter (focus) where an earthquake originates underground. Often reported as the earthquake's location in news reports. approximately 120 kilometres southeast of Mexico City, beneath the Puebla-Morelos border in central Mexico. The shaking in Mexico City lasted approximately 20 seconds. When it stopped, at least 369 people were dead across the region, more than 6,000 were injured, and dozens of buildings had collapsed in the Mexican capital.

The coincidence of the date with the anniversary of the 1985 disaster was eerie enough that initial social media reports suggested the earthquake had been triggered by the drill sirens — a factually incorrect claim that spread rapidly before being corrected. But the symbolism was undeniable: thirty-two years to the day after their most traumatic urban earthquake, Mexico City found itself once again pulling survivors from rubble.

Use Earthquake Energy Calculator to compare the 2017 M7.1 event with the 1985 M8.1 earthquake in terms of energy release. Use Distance from Epicenter to understand how the unusual wave propagation path amplified shaking in the city centre.

Intraslab Rupture: A Different Mechanism Than 1985

To understand the 2017 earthquake, it is essential to understand what it was not: it was not a repeat of the 1985 disaster in any seismological sense. The two earthquakes occurred by entirely different mechanisms.

The 1985 Mexico City earthquake (M8.1) was a classic Subduction ZoneA region where one tectonic plate dives beneath another into the mantle. Subduction zones produce the world's largest earthquakes (M8.5+) and are associated with deep ocean trenches and volcanic arcs. interface earthquake — a rupture on the megathrust boundary between the Cocos Plate and the North American Plate, approximately 350 kilometres west of Mexico City off the Pacific coast near Michoacán. The rupture was extensive, covering a large area of the plate interface, and the Seismic WaveAn elastic wave generated by an earthquake or explosion that propagates through the Earth. Seismic waves carry the energy released at the earthquake source to distant locations.s it generated travelled inland from the coast over several hundred kilometres before arriving in Mexico City, where Soil Amplification (Site Effect)The increase in shaking intensity caused by soft soil or sediment layers amplifying seismic waves. Structures built on soft soil can experience 2-10 times stronger shaking than those on bedrock. of the old lake bed produced catastrophic shaking.

The 2017 earthquake, by contrast, was an "intraslab" event — a rupture within the body of the subducting Cocos Plate itself, rather than on the interface between the Cocos and North American plates. As the Cocos Plate descends into the mantle beneath central Mexico, it becomes subject to internal stresses from the bending forces as it flexes downward, and from the gravitational pull on its dense leading edge. These internal stresses can cause the plate to crack — a process known as intraplate fracturing. The 2017 rupture occurred on a Normal FaultA fault where the rock above the fault plane (hanging wall) moves downward relative to the rock below. Associated with extensional forces in rift zones and divergent boundaries. within the Cocos Plate, at a depth of approximately 57 kilometres beneath the surface.

The implications of this mechanism for the pattern of ground shaking were significant. An intraslab earthquake at this depth beneath the Puebla-Morelos region produces a very different radiation pattern than a distant interface earthquake off the Guerrero coast. The EpicenterThe point on the Earth's surface directly above the hypocenter (focus) where an earthquake originates underground. Often reported as the earthquake's location in news reports. was relatively close to Mexico City — much closer than the 1985 source — and the waves arrived with different frequency content, different directional characteristics, and different interaction with the amplifying sediments of the Mexico City lake bed. Intraslab earthquakes also tend to produce higher-frequency ground motion than interface earthquakes of comparable magnitude, which affects which building heights are most vulnerable: higher frequencies are more damaging to shorter, stiffer structures.

SASMEX Failure: When the Epicenter Is Too Close for Warning

Mexico operates one of the world's longest-running seismic Earthquake Early Warning (EEW)A system that detects an earthquake and sends alerts to people and systems before strong shaking arrives. Can provide seconds to tens of seconds of warning, enough to take protective action. systems: the Seismic Alert System of Mexico, known by its Spanish acronym SASMEX. Established in 1993 in direct response to the 1985 catastrophe, SASMEX detects P-waves from earthquakes along the Guerrero coast using a network of seismographs, and transmits a warning signal to Mexico City approximately 60 to 120 seconds before the more destructive S-waves and surface waves arrive. This warning time has allowed residents to evacuate buildings, stop trains, halt surgeries, and take other protective actions in the years since its deployment.

SASMEX worked as designed on September 19, 2017 — in the sense that its sensors detected the earthquake, its algorithms computed the source parameters, and its transmitters broadcast the alert. The problem was that the EpicenterThe point on the Earth's surface directly above the hypocenter (focus) where an earthquake originates underground. Often reported as the earthquake's location in news reports. was too close to Mexico City for the system to provide meaningful warning.

SASMEX was designed primarily to detect and warn against earthquakes from the Guerrero seismic gap — a section of the subduction zone off the Pacific coast that seismologists have long identified as likely to produce a M8+ earthquake. For such a distant source, the P-wave and S-wave separation gives ample warning time. But the 2017 EpicenterThe point on the Earth's surface directly above the hypocenter (focus) where an earthquake originates underground. Often reported as the earthquake's location in news reports. was inland, beneath Puebla, more than 100 kilometres closer to Mexico City than the Guerrero coast. The warning signal and the destructive waves arrived almost simultaneously. In many parts of the city, residents heard the SASMEX alarm while the building was already shaking.

This "blind zone" phenomenon — the geographic radius around any EpicenterThe point on the Earth's surface directly above the hypocenter (focus) where an earthquake originates underground. Often reported as the earthquake's location in news reports. within which warning is impossible because the waves arrive before any warning can be transmitted and acted upon — is a fundamental physical limitation of Seismic Alert SystemMexico's SASMEX, one of the world's first public earthquake early warning systems, operational since 1991. Provides up to 60 seconds of warning for Mexico City from coastal earthquakes. technology that the 2017 earthquake brought into sharp public relief. Subsequent improvements to SASMEX have included sensors placed closer to Mexico City and computational improvements that reduce processing latency, but the physics of wave propagation sets an absolute limit that no technology can overcome for very close sources. The 2017 experience accelerated discussion of complementary warning approaches — including on-site sensing systems that detect P-waves at the building itself rather than relying on distant sensors — as a partial solution to the blind zone problem.

Structural Performance: 1985 Retrofits Tested

One of the most consequential tests of any earthquake is the performance of structures that were built, strengthened, or regulated based on the lessons of a previous disaster. The 1985 Mexico City earthquake generated a massive regulatory and engineering response: new Building Code (Seismic)A set of legal requirements governing the design and construction of buildings to ensure minimum levels of earthquake safety. Updated after major earthquakes reveal new vulnerabilities.s with stricter seismic design requirements, government programmes to Seismic RetrofitStrengthening an existing building to improve its earthquake resistance. Common methods include adding steel bracing, reinforcing foundations, and bolting structures to foundations. existing buildings, and systematic evaluation of the city's building stock.

The 2017 earthquake provided a partial but sobering assessment of how well this 32-year engineering project had succeeded.

The overall result was mixed. In general, buildings constructed after 1985 to the new codes performed significantly better than older structures, with far fewer collapses per building in the modern stock. The most dramatic structural failures in 2017 involved buildings constructed in the 1950s through 1970s, before rigorous seismic design was required. Many of these buildings had not received Seismic RetrofitStrengthening an existing building to improve its earthquake resistance. Common methods include adding steel bracing, reinforcing foundations, and bolting structures to foundations. despite the post-1985 programmes, either because the owners had not been required to strengthen them, because the retrofits had been inadequately executed, or because the buildings had been inspected and passed as adequate when they were not.

The building type that failed most consistently was the "pilotis" configuration — reinforced concrete buildings whose ground floor consisted of an open colonnade of columns without shear walls, often used to accommodate shops or parking. The ground-floor columns proved vulnerable to soft-story collapse, where the weak open floor buckles and the upper floors pancake onto it. This configuration was known to be vulnerable before 1985 and had been the subject of post-1985 retrofitting programmes that proved insufficient for many buildings.

The pattern of Soil Amplification (Site Effect)The increase in shaking intensity caused by soft soil or sediment layers amplifying seismic waves. Structures built on soft soil can experience 2-10 times stronger shaking than those on bedrock. beneath Mexico City also played its familiar role. The ancient lake bed of Lake Texcoco — on which much of Mexico City was built after the Aztec capital was flooded and drained by Spanish colonisers — consists of extraordinarily soft clay sediments that amplify seismic waves dramatically at particular frequencies. The Structural ResonanceThe amplification of building motion when earthquake wave frequency matches the building's natural frequency. Low-rise buildings resonate with high-frequency waves; tall buildings with low-frequency. between these amplified frequencies and the natural resonance of multi-story buildings was the mechanism that made the 1985 earthquake so devastating, and it operated again in 2017. The precise frequency content of the 2017 intraslab source differed from 1985, but the fundamental amplification problem of building on lake sediments was unchanged.

The 2017 earthquake also revealed a troubling pattern of construction outside the formal regulatory system. Investigation of collapsed buildings found some structures where construction records were incomplete, where approved designs had been modified during construction, or where oversight had been cursory. Addressing this gap — between what codes require and what is actually built — is perhaps the most difficult challenge in urban earthquake risk reduction, requiring not just better regulations but better institutional capacity to enforce them.

Social Media and Crowdsourced Rescue Coordination

The 2017 Mexico City earthquake occurred in an era of ubiquitous smartphone use and social media connectivity that had not existed during the 1985 disaster. The difference in information availability and coordination capacity was dramatic and instructive.

Within minutes of the earthquake, residents throughout the city were documenting collapsed buildings, injured survivors, and blocked roads with smartphone cameras and uploading images to Twitter, Instagram, and WhatsApp. Volunteer rescue coordinators — often young professionals working from undamaged apartments — began aggregating these reports, creating crowd-sourced maps of collapse locations, and matching volunteer rescuers with sites where survivors might still be trapped. The hashtag #FuerzaMexico (Strength Mexico) became a coordinating mechanism for volunteer activity.

The results were genuinely impressive in some cases: volunteer rescue teams equipped with simple tools and organized via smartphone messaging reached collapse sites before official rescue brigades and extracted survivors. The social media coordination also facilitated the rapid delivery of supplies, the organization of blood donation, and the matching of displaced residents with emergency housing.

The experience also exposed the limitations of uncoordinated voluntary response. Some social media posts about collapsed buildings turned out to be inaccurate, sending rescue volunteers on futile missions while genuine sites waited for help. The rumour of a girl named "Frida Sofia" trapped in a collapsed school — reported live on national television for hours — proved to be entirely fabricated, a story constructed from a misunderstanding that consumed enormous media and rescue resources before it was retracted.

Official rescue services, accustomed to managing response through formal command structures, found the integration of self-organized volunteer networks both helpful and chaotic. The 2017 earthquake thus became a case study in the emerging field of disaster informatics — the systematic study of how information technology, social media, and crowd-sourcing interact with traditional emergency management — and subsequent research has examined what types of information sharing accelerated effective rescue and where coordination failures cost time that might have saved lives.

1985 vs. 2017: Measuring 32 Years of Progress

Comparing the 1985 and 2017 Mexico City earthquakes provides a direct, if imperfect, measure of three decades of investment in seismic safety.

The most favourable comparison involves the death tolls. The 1985 earthquake, with a MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released. of 8.1, killed a minimum of 5,000 people in a city of approximately 14 million. The 2017 earthquake, with a MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released. of 7.1 — roughly 250 times less energy — killed 369 people in a city of 21 million. While the two earthquakes are not directly comparable because of differences in source distance, mechanism, and shaking characteristics, the disparity in fatality rates per unit of ground shaking suggests that structural improvements made a genuine difference.

The less favourable comparison involves the persistence of vulnerable building stock. Despite 32 years of code improvements, retrofit programmes, and mandatory inspections, the 2017 earthquake still found buildings that collapsed and killed people — buildings that should, under existing regulations, have been identified and strengthened. This gap between what regulations require and what actually exists in a city's building stock is a universal challenge in earthquake engineering, not unique to Mexico.

The earthquake highlighted a geographic shift in vulnerability. In 1985, the worst damage was concentrated in the soft lake bed sediments of the historical city centre. By 2017, some of the worst-performing buildings were in areas that had developed rapidly since 1985 — newer colonias where construction quality varied enormously and oversight had been inconsistent. Progress in one domain can create false confidence about overall safety if newer construction has its own problems.

The 2017 earthquake thus occupies a complex place in Mexico City's history. It is simultaneously a demonstration of progress — the relatively low death toll for an earthquake striking the world's fifth-largest city is a genuine achievement — and a reminder of persistent failure, in the form of the specific buildings that collapsed and the specific people who died in them. Holding both of these truths simultaneously, without allowing the progress to breed complacency or the failures to breed despair, is the challenge that Mexico City's earthquake preparedness community faces in the decade after the disaster. The Seismic Risk AssessmentThe process of evaluating earthquake hazard, building vulnerability, and potential losses for a specific area or structure. Combines hazard maps, building inventory, and damage models. framework continues to evolve, incorporating the 2017 data into improved ground motion models, updated Seismic Hazard MapA map showing the probability of earthquake shaking exceeding specified levels over a given time period. Used by engineers, planners, and insurers to assess earthquake risk.s, and revised Building Code (Seismic)A set of legal requirements governing the design and construction of buildings to ensure minimum levels of earthquake safety. Updated after major earthquakes reveal new vulnerabilities. provisions. The work of making the city safer proceeds incrementally, earthquake by earthquake, lesson by lesson.

The date of September 19 will continue to be Mexico City's day of reflection on seismic risk. The annual earthquake drill, now observed for four decades, represents an institutional commitment to preparedness that few other cities have matched. The 2017 earthquake tested that commitment and found it imperfect — as any realistic test would — but also found genuine improvement from the darkest days of 1985. The challenge for the coming decades is to close the gap between the engineering knowledge that exists about how to make buildings safe and the messy, politically and economically constrained reality of actually making a megacity's built environment match that knowledge.

The Lake Bed Problem: Texcoco's Enduring Influence on Seismic Risk

The ancient lake bed beneath Mexico City represents one of the most extensively studied and most persistently dangerous geological settings for any major city in the world. Lake Texcoco — the lake on which the Aztec capital Tenochtitlan was built and which was subsequently drained by Spanish colonisers to create more urban land — left behind deposits of extraordinarily soft clay that continue to shape Mexico City's seismic risk nearly 500 years later.

The clay of the former lake bed has a water content of up to 400 percent by weight — a value that civil engineers find almost incredible, as it means the material is almost entirely water with clay particles floating within it. This material amplifies earthquake shaking by factors of up to 50 times compared with rock sites at the same distance from the source. The amplification is strongly frequency-selective, concentrating energy in the frequency band corresponding to the natural period of the lake bed — approximately two seconds — which overlaps dangerously with the natural period of many mid-rise buildings.

This amplification mechanism was identified by Mexican seismologists before the 1985 earthquake, and the 1985 data confirmed and quantified it in unprecedented detail. The 2017 earthquake provided additional recordings that extended understanding of how the amplification varies across the lake bed — it is not uniform, but varies with local sediment thickness and characteristics. Areas where the lake bed is thicker experience longer-period amplification; areas where it is thinner experience shorter-period amplification. This spatial variation in amplification means that different building height ranges are most vulnerable in different parts of the city, and that Seismic DesignThe practice of designing structures to withstand earthquake forces. Modern seismic design aims to prevent collapse and protect life, while accepting some structural damage in major earthquakes. requirements should ideally be site-specific rather than zone-wide.

The ongoing settlement and subsidence of Mexico City — which is slowly sinking as groundwater is extracted from the aquifer beneath the lake bed, causing the clay to compact — adds a further complication. As the depth to bedrock changes, the amplification characteristics of the lake bed change as well, meaning that Seismic Hazard MapA map showing the probability of earthquake shaking exceeding specified levels over a given time period. Used by engineers, planners, and insurers to assess earthquake risk.s based on measurements made decades ago may progressively become less accurate. Keeping Mexico City's seismic hazard characterization current as the geological setting evolves beneath it is one of the ongoing challenges of urban earthquake risk management in the city.

Intraslab Earthquakes: An Underestimated Hazard Class

The 2017 Puebla earthquake also focused scientific attention on the hazard posed by intraslab earthquakes more broadly. Intraslab events — ruptures within the body of the subducting plate rather than on its interface with the overriding plate — occur throughout subduction zone settings worldwide, but they have historically received less attention in hazard assessments than the larger interface events. The 2017 earthquake, by causing major damage in a setting where the hazard was assumed to be primarily from distant interface earthquakes, demonstrated that intraslab events can pose serious urban risk even when they are smaller in MagnitudeA single number that quantifies the total energy released by an earthquake. Each whole number increase represents roughly 31.6 times more energy released. than the largest possible interface events.

Mexico is particularly exposed to intraslab earthquakes because of the shallow angle at which the Cocos Plate subducts beneath central Mexico — a configuration that keeps the plate at relatively shallow depths beneath Mexico City even as it extends far inland from the coast. This "flat slab" configuration means that the seismically active interior of the plate is geographically close to major population centres, producing a class of earthquake threat that differs in character from the Guerrero interface scenario for which SASMEX was primarily designed.

Subsequent revisions to Mexico's seismic Seismic Hazard MapA map showing the probability of earthquake shaking exceeding specified levels over a given time period. Used by engineers, planners, and insurers to assess earthquake risk. have incorporated more detailed modelling of intraslab sources, accounting for their specific frequency characteristics, radiation patterns, and the depth-distance combinations that produce the strongest shaking in Mexico City. The 2017 earthquake's data — including dozens of high-quality strong-motion recordings from the city's dense accelerograph network — have been used to improve ground motion prediction equations for intraslab sources, which differ systematically from interface sources in ways that affect Seismic DesignThe practice of designing structures to withstand earthquake forces. Modern seismic design aims to prevent collapse and protect life, while accepting some structural damage in major earthquakes. demands.

The Volunteer Culture: Topos and Spontaneous Response

One of the enduring cultural legacies of the 1985 Mexico City earthquake was the emergence of the "topos" — volunteer rescuers who became experts in urban search and rescue and organized into formal brigades. The topos (named for the moles they compared themselves to as they dug through rubble) emerged spontaneously from the 1985 disaster, when the government's response was overwhelmed and ordinary citizens organized their own rescue efforts. In the three decades between the disasters, the topos institutionalized their expertise and began deploying internationally to earthquake disasters around the world.

In 2017, the topos were among the first organized responders at collapse sites, drawing on their institutional knowledge and equipment to work more effectively than purely spontaneous volunteers. Their presence exemplified the way the 1985 disaster had permanently altered Mexican civil society's relationship with earthquake response — creating a culture of prepared, skilled voluntary action that has no direct equivalent in most other countries.

The 2017 earthquake also demonstrated that this culture of voluntary preparedness could work synergistically with modern technology. Topos using social media coordination were more effective than either technology alone or volunteer action alone would have been. The combination of experienced human judgment, institutional memory of how to navigate collapse sites, and real-time information sharing via smartphones represented a model of disaster response that subsequent researchers have studied as a potential template for other megacities with large, technically capable civil societies.