Puerto Rico's Earthquake Crisis: What Happened

It started quietly, the way earthquake swarms often do. On December 28, 2019, a M4.7 shook the southwestern corner of Puerto Rico — noticeable, briefly alarming, then apparently over. But the events that followed were not over. Within days, more earthquakes arrived: a M5.0 on December 29, a M5.6 on January 6, 2020, and then, at 4:24 AM on January 7, the M6.4 — the mainshock, striking when most of the island's southwestern communities were still in bed. The shaking lasted approximately 20 seconds. In that time, whole neighborhoods of concrete block construction in the towns of Guánica, Guayanilla, Yauco, and Ponce cracked, buckled, and in some cases collapsed. A 73-year-old man died when a wall fell on him — the only direct earthquake fatality in the sequence. Eight thousand homes were damaged or destroyed. More than 8,000 people camped in parking lots and on baseball fields rather than return to buildings they no longer trusted. The Costa Sur power plant — Puerto Rico's largest electricity generating facility, supplying roughly one-third of the island's power — was knocked offline by a structural failure in its switching yard, cutting power to approximately 1.1 million customers across the island.

What made the 2020 Puerto Rico earthquake crisis more than a single event — and what makes it scientifically and humanistically instructive beyond its immediate statistics — is the compounding context in which it occurred. Puerto Rico in January 2020 was still rebuilding from Hurricane María, which had made landfall as a Category 4 storm in September 2017, killed an estimated 2,975 people, and left the island without electricity for months. When the earthquake swarm began, thousands of Puerto Ricans were living in homes that had been patched but not fully repaired from María — roofs covered with blue FEMA tarps that were still in place more than two years after the hurricane. The earthquake shook buildings that were already compromised. It stripped the credibility of informal repairs. It took an island community that had been recovering, slowly and painfully, from the worst disaster in its modern history and delivered a second blow to its most vulnerable residents — those in the southwest whose homes had been most damaged by the quake, who had the least resources to absorb additional loss, and who faced the prospect of months of aftershocks in buildings they could no longer inhabit.

This post covers the earthquake science in depth: the tectonic setting that makes southwestern Puerto Rico seismically hazardous, the specific fault source of the 2020 sequence, why the building stock performed as it did, and what the extended aftershock sequence revealed about the seismic system beneath the island. It also treats the crisis as what it was — not just a geophysical event but a compound disaster that played out in a specific social and infrastructure context, and whose consequences cannot be understood without that context.

The Tectonic Setting: Puerto Rico at the Plate Boundary Junction

Puerto Rico occupies one of the most tectonically complex positions in the Western Hemisphere — a narrow island sitting on a microplate at the junction between the Caribbean plate, the North American plate, and the subducting Atlantic oceanic lithosphere, with active fault systems on virtually every flank. The Caribbean post covered this tectonic setting in the broader regional context; here we focus on the specific fault architecture directly beneath Puerto Rico and what it means for the island's seismic hazard.

The Puerto Rico Trench and Northern Subduction

To the north of Puerto Rico, the Puerto Rico Trench — at 8,376 meters the deepest point in the Atlantic Ocean — marks the subduction of the North American plate westward beneath the Caribbean. The trench and its associated thrust fault system represent the highest-magnitude potential source for Puerto Rico: megathrust earthquakes comparable to M8+ are theoretically possible, and the 1946 M8.0 Dominican Republic earthquake, which ruptured the northern Hispaniola fault system to the west and generated a tsunami that killed 1,700 people, provides a calibration of what the northern plate boundary can produce. For Puerto Rico's north coast, a major Puerto Rico Trench event is the most consequential long-recurrence-interval hazard — but it is not what struck in 2020.

The Muertos Trough and Southern Convergence

South of Puerto Rico, a different and less well-characterized tectonic process is underway. The Muertos Trough — a bathymetric depression running east-west along Puerto Rico's southern continental margin — is the surface expression of a zone of south-dipping subduction or underthrusting where the Caribbean plate itself is being consumed southward beneath a microplate or crustal block in what some researchers characterize as back-arc compression. The Muertos Trough convergence is slower and less clearly defined than the Puerto Rico Trench subduction to the north, but it generates a distinct population of seismicity on the south-facing faults of the Muertos Trough system — and it is this southern fault system that was the source of the 2020 earthquake sequence.

The 2020 Source Fault

The January 7, 2020 M6.4 mainshock occurred at a shallow depth of approximately 8 km on a northwest-dipping reverse (thrust) fault approximately 10 km offshore of the southwestern coast, near the town of Guánica. The focal mechanism — determined from the pattern of first-motion P-wave arrivals at seismic stations throughout the Caribbean — showed predominantly reverse faulting with a component of left-lateral strike-slip, consistent with the compressional stress regime of the Muertos Trough convergence zone. The rupture area is estimated at roughly 15 × 10 km based on the spatial extent of the aftershock cloud, releasing a seismic moment equivalent to approximately 10¹⁸ newton-meters.

The shallow depth of the mainshock — 8 km — was the primary driver of the intensity of near-field shaking. At 8 km depth, the closest surface points are within 10–15 km of the rupture surface, producing peak ground accelerations in the range of 0.3–0.6g in the towns most directly above the fault. These accelerations exceeded the design ground motions for unreinforced concrete masonry construction built to the building codes prevailing in the southwestern Puerto Rico corridor during the 1960s through 1990s — the period when most of the damaged housing stock was constructed.

The Swarm: Months Before and Months After

One of the defining characteristics of the 2020 Puerto Rico crisis — distinguishing it from most single-mainshock earthquake disasters — was the extraordinary duration of the earthquake sequence. The USGS and Puerto Rico Seismic Network catalogued more than 11,000 earthquakes in the southwestern Puerto Rico region between December 28, 2019 and the end of 2020 — an unprecedented level of seismic activity for Puerto Rico in the modern instrumental record, producing a prolonged period of public anxiety, disrupted sleep, and persistent structural concern for residents who were trying to assess whether buildings damaged in the January 7 mainshock were safe to reoccupy.

The Precursor Sequence

Seismologists looking at the December 2019 activity in retrospect identified it as a classic foreshock-mainshock pattern — but of a type that is only clearly recognizable as foreshock activity after the mainshock has occurred. The December 28 M4.7 and subsequent events were consistent with either a standalone swarm or the beginning of a foreshock sequence; distinguishing between these possibilities in real time is one of the fundamental unsolved problems in operational earthquake forecasting. The Puerto Rico Seismic Network did issue public notifications of the elevated seismicity and general public guidance on preparedness as the activity continued into early January, but no specific short-term forecast of the January 7 mainshock was possible — nor is such a forecast currently achievable with any operational earthquake prediction methodology.

The Aftershock Sequence

Following the January 7 mainshock, Puerto Rico experienced one of the most prolonged and widely felt aftershock sequences in the island's modern history. Several M5+ aftershocks occurred in the days and weeks following the mainshock, each causing additional damage to already-weakened structures and prolonging the period during which residents were reluctant to return to their homes. The M5.9 aftershock on January 11, 2020 — the largest event in the sequence after the mainshock — caused additional damage in Guánica and Guayanilla and further delayed any resumption of normal occupancy in the most affected communities.

The aftershock sequence decayed following the modified Omori law — the empirical relationship describing how aftershock rates decrease proportionally with time after the mainshock — but more slowly than the median expectation for events of this magnitude. Felt aftershocks continued at a rate of several per week through the spring and summer of 2020, and the elevated background seismicity in the southwestern Puerto Rico region persisted well into 2021 — a pattern consistent with either a slow redistribution of stress in the fault zone following the mainshock or with the triggering of additional fault segments adjacent to the main rupture.

The Buildings: Why They Failed

The dominant building type damaged in the 2020 earthquake was unreinforced or poorly reinforced concrete masonry unit (CMU) construction — concrete block walls supporting concrete or light frame roofs, built without engineered seismic detailing. This construction type is the predominant residential building technology throughout Puerto Rico, the Caribbean, and much of the tropical developing world — adopted because it resists hurricane winds better than wood frame and because concrete block is locally available and affordable. But CMU construction without proper reinforcing and connection detailing is highly susceptible to seismic damage: the heavy, stiff walls attract large lateral forces in an earthquake but lack the ductility to deform without cracking, and the connection between wall and roof is often inadequate to prevent out-of-plane wall collapse.

The Double Failure Mode

Post-earthquake reconnaissance by engineering teams from EERI (Earthquake Engineering Research Institute) and Puerto Rico's University of Puerto Rico documented two dominant failure modes in the 2020 damaged building stock. The first was diagonal shear cracking in CMU walls — the X-pattern cracking visible throughout the affected area that indicates the wall has been pushed laterally to its shear capacity and the mortar joints have failed in tension and compression. Buildings with this failure mode are typically repairable but require significant structural intervention before re-occupancy. The second was partial or complete collapse — most commonly the out-of-plane failure of unreinforced CMU walls that were not adequately tied to the roof or floor diaphragm, allowing the wall to topple outward when the lateral seismic force exceeded the wall's out-of-plane capacity.

A particularly troubling finding was the failure of structures that had been informally repaired after Hurricane María. Many homes in the southwestern corridor had sustained wind damage in 2017 and had been patched with new CMU walls, repairs using different mortar mixes, or non-structural repairs that did not address the underlying structural connections. These hybrid structures — partially original, partially repaired — sometimes performed worse than either all-original or properly retrofitted buildings, because the repair introduced discontinuities in the load path that the seismic forces exploited selectively.

Code-Compliant Buildings and Their Performance

Buildings designed and constructed to the International Building Code (IBC) provisions applicable to Puerto Rico's Seismic Design Category — primarily post-2000 engineered construction with properly detailed reinforced CMU or reinforced concrete frames — performed significantly better than the older unengineered stock. Modern reinforced concrete frame buildings with masonry infill, properly tied to seismic force resisting systems, sustained minor cracking in the most severely shaken areas but remained structurally intact and occupiable. The damage distribution across the affected area traced almost precisely to the age distribution of the building stock: pre-1970s construction in the town centers suffered the most severe damage; post-2000 construction at the urban periphery was largely undamaged.

The Compound Disaster: Earthquake on a Hurricane-Wounded Island

Any honest accounting of the 2020 Puerto Rico earthquake crisis must confront the context that made its consequences as severe as they were. Puerto Rico in January 2020 was not a normal island community struck by an unexpected earthquake. It was a community still in a state of profound disruption from Hurricane María — a storm that had not merely damaged buildings and cut power but had caused a demographic crisis (estimated net population decline of 130,000–200,000 people between 2017 and 2020 as residents emigrated to the mainland), a fiscal crisis (Puerto Rico was in a federally supervised bankruptcy process before María and the storm made it catastrophically worse), an infrastructure crisis (the electrical grid was partially rebuilt but remained fragile), and a governance crisis (FEMA's disaster response to María had been widely criticized as inadequate, and the relationship between the federal government and Puerto Rico's territorial government was acrimonious).

The earthquake arrived into this context like a second punch to a patient still recovering from the first. The power outage — 1.1 million customers without electricity in the immediate aftermath, caused by the structural failure of the Costa Sur plant — was not merely an inconvenience. Puerto Rico's water distribution system depends on electrically powered pumping. Without power, water pressure dropped or failed in many municipalities. Medical equipment in homes and clinics required for the island's elderly and chronically ill population went offline. The hospitals and dialysis centers that had been rebuilt or repaired after María faced a second stress event.

The displacement picture was similarly compounded. The 8,000+ people who moved to outdoor camps rather than occupy damaged homes were not doing so on the basis of unfamiliarity with earthquake damage assessment — many had lived through months of blue-tarp existence after María and had learned, empirically, that the formal governmental assessment and repair process was slow, underfunded, and unreliable. The mistrust of official "safe to occupy" determinations was not irrational given that experience; it was a rational response to an institutional track record that had taught these communities not to trust official certifications about building safety.

The PRSN: Puerto Rico Seismic Network and Its Response

The Puerto Rico Seismic Network (PRSN), based at the University of Puerto Rico at Mayagüez, operates the island's primary earthquake monitoring infrastructure — a network of approximately 50 seismograph and accelerometer stations that detected and catalogued the 2020 swarm in real time. The PRSN's 24-hour operations during the peak of the crisis — issuing public magnitude and location updates for each significant event, providing technical briefings to government officials, and communicating with the public through social media in both English and Spanish — represented exactly the kind of science-to-public communication that earthquake early warning and response depends on.

The PRSN's accelerograph network — strong motion sensors specifically designed to remain on-scale during damaging shaking — recorded the mainshock and major aftershocks at several stations in the southwestern corridor, providing ground motion records that engineering researchers subsequently used to calibrate ground motion prediction models, assess building performance against recorded shaking levels, and refine the seismic hazard assessment for the region. These records are now part of the NGA-West2 and related strong motion databases used globally to develop ground motion prediction equations — Puerto Rico's 2020 earthquake has become a calibration event for seismic hazard models worldwide.

The Recovery: What Came After

The formal recovery from the 2020 earthquake — housing repair, infrastructure restoration, and economic rehabilitation of the affected southwestern municipalities — proceeded on a timeline measured in years, not months. FEMA's Individual Assistance program and the Community Development Block Grant Disaster Recovery (CDBG-DR) funds appropriated by Congress provided the primary financial mechanism for housing repair, but the bureaucratic complexity of accessing those funds — compounded by the ongoing processing of María-related claims that had not yet been resolved — meant that many displaced families waited a year or more for formal assistance to arrive.

The housing repair work that eventually proceeded under federal funding programs included, for the first time in any systematic Puerto Rican disaster recovery, a seismic mitigation component — funding provisions requiring that rebuilt homes in the highest-hazard zones meet current seismic code provisions rather than simply replicating the pre-damage construction. This requirement — which was not present in the María recovery programs — represents a genuine policy advance: the recognition that rebuilding identical buildings in the same locations to the same non-seismic standards is not recovery but rather the preparation of the next disaster.

Puerto Rico's Ongoing Seismic Hazard

The 2020 earthquake sequence was not a once-in-a-generation anomaly for Puerto Rico — it was a demonstration, at moderate magnitude, of the seismic hazard that the island's tectonic position guarantees. The USGS hazard maps classify Puerto Rico in one of the highest seismic design categories applicable to US territory outside of coastal Alaska and the Pacific Northwest, with 2% probability of exceedance in 50-year PGA values of 0.4–0.8g for the highest-hazard zones in the southwest — comparable to the most active fault zones in California.

The Muertos Trough fault system that sourced the 2020 sequence is capable of significantly larger events than M6.4. The full convergence zone has not been comprehensively characterized for maximum magnitude, and paleoseismic evidence for prehistoric large events on the Muertos Trough system is limited — the offshore geometry of the fault and the absence of exposed onshore fault traces make conventional paleoseismic investigation impractical. GPS coupling models suggest that the Muertos Trough interface may be partially locked over a broad area, capable of accumulating strain toward events in the M7–7.5 range that would dwarf the 2020 mainshock in both energy and consequences.

The Puerto Rico Trench to the north — capable of M8+ megathrust events — adds a longer-recurrence but higher-magnitude hazard tier that any comprehensive assessment of Puerto Rico's future seismic risk must include. A Puerto Rico Trench megathrust event would generate not only intense shaking throughout the island but a major tsunami affecting the north coast within 5–10 minutes — essentially no time for organized evacuation and requiring the same ingrained "shaking is the warning" behavioral response that characterizes coastal preparedness in Cascadia and the Pacific Northwest.

What Puerto Rico Needs: The Path Forward

The 2020 earthquake identified four specific gaps in Puerto Rico's seismic resilience that the recovery process must address if future earthquakes are to produce fewer casualties and less displacement.

The first is the building stock. The majority of Puerto Rico's residential buildings — particularly in the older towns and lower-income communities that have the highest concentration of pre-code CMU construction — have never been seismically assessed and are not subject to any retrofit requirement. A systematic building vulnerability assessment program, combined with incentive or mandatory retrofit provisions for the highest-risk structures, is the single most impactful preparedness investment available to Puerto Rico's government but has not yet been fully implemented.

The second is the power system. The Costa Sur plant failure in 2020 — a single generating facility taking out power for 1.1 million customers — reflects a grid design with critical single points of failure that the Puerto Rico Electric Power Authority (PREPA) was in the process of restructuring after María but had not completed. Seismic hardening of the most critical generating and transmission infrastructure, and distribution of power generation to reduce single-point-of-failure concentration, would significantly reduce the cascading consequences of any future major earthquake.

The third is early warning. The absence of ShakeAlert coverage for Puerto Rico means that residents receive no seconds of advance notification that violent shaking is imminent — the kind of warning that allows people to drop, cover, and hold on before the worst of the shaking arrives, and that can automatically trigger protective actions in hospitals, utilities, and transportation systems. Extending ShakeAlert to Puerto Rico should be a federal priority given the island's documented seismic exposure and its status as US territory.

The fourth is compound hazard planning. The 2020 earthquake demonstrated definitively that Puerto Rico cannot plan for a single hazard in isolation. The island's earthquake preparedness must be integrated with its hurricane preparedness, its infrastructure recovery planning, and its public health emergency planning — because the next major earthquake will almost certainly arrive at a moment when the island is already managing some other crisis, in a building stock that reflects the incomplete recovery from the last one.

Conclusion

Puerto Rico's 2020 earthquake crisis was, in one sense, a moderate event — M6.4, one death, no tsunami, infrastructure damage measured in billions rather than tens of billions. In another sense it was a profound demonstration of what happens when a geologically active island encounters a vulnerable building stock in the context of pre-existing economic and infrastructure stress. The earthquake itself was the geological process doing what it does on a convergent plate boundary. The crisis — the 8,000 damaged homes, the tens of thousands camping outdoors, the power loss to 1.1 million customers, the months of aftershock anxiety — was the human and institutional response to an event that a better-prepared community would have absorbed at significantly lower cost.

Puerto Rico's 3.3 million people live on an island with major seismic hazard sources on every side — the Puerto Rico Trench to the north, the Muertos Trough to the south, the Mona Rift to the west. The 2020 M6.4 was a warning shot, not a worst case. Whether the warning translates into the building assessments, the code enforcement improvements, the power grid hardening, and the early warning infrastructure that would reduce the consequences of the next major event — events that the physics of the Caribbean plate boundary guarantee will come — is the central earthquake preparedness question facing Puerto Rico and the federal government responsible for the island's safety and welfare.