Introduction
When Americans think about earthquake hazards, their minds jump to California — the San Andreas Fault, the shaking skyline of San Francisco, the Hollywood disaster movie. But buried beneath the farmland of the central United States lies a seismic zone that has produced some of the most powerful earthquakes in North American history. The New Madrid Seismic Zone (NMSZ), centered where Missouri, Arkansas, Tennessee, Kentucky, and Illinois converge, is a ticking geological clock that most Americans have never heard of.
During the winter of 1811–1812, the NMSZ unleashed three major earthquakes estimated at M7.0 to 7.5 or higher, along with thousands of aftershocks over several months. The Mississippi River appeared to flow backwards. A new lake was created overnight. Church bells rang as far away as Boston, more than 1,600 kilometers from the epicenters. These were not California-scale events — they were larger, and they struck a region that remains almost entirely unprepared for a repeat.
The challenge is one of awareness and probability. Major earthquakes in the NMSZ occur on timescales of hundreds to thousands of years, making it easy for residents, builders, and policymakers to dismiss the risk. But the consequences of a major earthquake here would be staggering — not because the shaking would be unprecedented, but because the built environment and population centers sitting atop the zone were never designed for it.
Geology of the New Madrid Seismic Zone
The Reelfoot Rift: An Ancient Wound in the Earth's Crust
The New Madrid Seismic Zone owes its existence to a geological structure far older than the earthquakes it produces. The Reelfoot Rift is a failed continental rift — an aborted attempt by the North American continent to split apart — that formed approximately 500 to 750 million years ago during the breakup of the supercontinent Rodinia.
When continental rifting occurs, the crust thins and weakens as tectonic forces pull it apart. In most cases, rifting continues until a new ocean basin forms — this is what happened with the Atlantic Ocean. But in the case of the Reelfoot Rift, the process stalled. The rift failed, and the weakened crust was buried under kilometers of sediment deposited by the Mississippi River and its predecessors over hundreds of millions of years.
According to the USGS, the Reelfoot Rift extends roughly 300 kilometers from northeast Arkansas through the Missouri Bootheel and into southern Illinois. The rift is defined by a series of faults, including the Reelfoot Fault, the Cottonwood Grove Fault, and the New Madrid North Fault. These structures remain seismically active because the ancient zone of crustal weakness concentrates tectonic stress transmitted from distant plate boundaries.
Why Earthquakes Happen So Far from Plate Boundaries
The NMSZ sits more than 2,000 kilometers from the nearest active plate boundary. This makes it an intraplate seismic zone — one of the most challenging types of earthquake hazard for geologists to understand and forecast.
The prevailing explanation, according to research published by the Center for Earthquake Research and Information (CERI) at the University of Memphis, is that far-field tectonic stresses — generated primarily by the slow push of the Mid-Atlantic Ridge and the pull of subducting plates along the Pacific margin — are concentrated along the weakened crust of the Reelfoot Rift. The deep, buried faults act as stress guides, channeling strain into the zone despite its great distance from any plate boundary.
Some researchers, including Seth Stein at Northwestern University, have argued that the current rate of strain accumulation in the NMSZ is too low to sustain the recurrence of large earthquakes at the rate suggested by the 1811–1812 sequence. GPS measurements show less than 1 mm/year of deformation across the zone — far below what would be expected for a fault system producing M7+ earthquakes every few hundred years. This has led to an ongoing scientific debate about whether the NMSZ is winding down or simply in a quiet period between major clusters of activity.
Sediment Amplification
One factor that makes the NMSZ particularly hazardous is the thick blanket of soft sediment deposited by the Mississippi River and its tributaries. These unconsolidated sediments — sand, silt, and clay — can amplify seismic waves significantly compared to hard bedrock. According to USGS research, seismic waves in the central U.S. travel farther and lose energy more slowly than in the fractured, geologically complex crust of the western U.S. This means that a M7.0 earthquake in the NMSZ would produce damaging shaking over a much larger area than a comparable earthquake in California.
The 1811–1812 earthquakes were felt over an area of approximately 5 million square kilometers — roughly 10 to 15 times the felt area of a comparable earthquake on the San Andreas Fault.
The 1811–1812 Earthquake Sequence
Timeline of Destruction
The 1811–1812 New Madrid earthquake sequence remains the most powerful series of earthquakes in the recorded history of eastern North America. Three mainshocks, each estimated at M7.0 to 7.5 or higher (with some researchers estimating the largest exceeded M7.7), struck within an 82-day window, accompanied by hundreds of damaging aftershocks.
| Date | Estimated Magnitude | Epicentral Area | Key Effects |
|---|---|---|---|
| December 16, 1811 (2:15 AM) | M7.2–7.5 | Northeast Arkansas (near Blytheville) | Violent shaking across the Mississippi Valley; chimneys toppled as far as Cincinnati; large landslides along river bluffs; ground fissures ejected sand and water |
| January 23, 1812 | M7.0–7.3 | Northeast Arkansas / Missouri Bootheel | Renewed destruction to already-damaged settlements; further landsliding; continued sand blows and ground deformation |
| February 7, 1812 | M7.4–7.7 | New Madrid, Missouri | Most powerful event of the sequence; created Reelfoot Lake in Tennessee by warping the ground and damming streams; Mississippi River appeared to flow backwards temporarily; the town of New Madrid largely destroyed |
Source: USGS, "Historic Earthquakes: New Madrid Seismic Zone"; Johnston & Schweig, 1996
Eyewitness Accounts and Physical Effects
The 1811–1812 events produced a range of physical effects that seemed almost apocalyptic to frontier settlers. Contemporary accounts describe the Mississippi River heaving and churning, with waterfalls forming temporarily where the riverbed was uplifted. Entire islands in the river disappeared. Vast tracts of forest were knocked down. Sand blows — geysers of sand and water erupting from the ground through a process called liquefaction — covered fields across the Mississippi Valley.
Reelfoot Lake, now a 73-square-kilometer lake in northwest Tennessee, was created when the February 7, 1812 earthquake caused the ground to subside and the Reelfoot Creek drainage to be dammed by uplift along the Reelfoot Fault. The lake remains one of the most visible geological legacies of the sequence.
The shaking was so intense that church bells rang in Boston, more than 1,600 kilometers from the epicenter. Pendulum clocks stopped in Washington, D.C. The felt area extended across approximately 50,000 square miles of intense shaking, with perceptible shaking across nearly the entire eastern United States.
Because the region was sparsely populated in 1811–1812, direct fatalities were relatively low — probably in the dozens to low hundreds, though precise counts do not exist. The same earthquakes striking today would produce a fundamentally different outcome.
Modern Risk: Who Is in Danger?
Population and Infrastructure Exposure
The NMSZ directly underlies or closely borders major population centers that did not exist — or were tiny settlements — in 1811. Memphis, Tennessee (metropolitan population approximately 1.3 million), sits at the southern end of the zone and is by far the most vulnerable major city. Other at-risk cities include St. Louis, Missouri; Little Rock, Arkansas; Evansville, Indiana; and Paducah, Kentucky.
| State | Major Cities at Risk | Approximate Population in Hazard Zone | Building Vulnerability |
|---|---|---|---|
| Tennessee | Memphis, Jackson | ~2.2 million | High — extensive URM, limited seismic code adoption until recently |
| Missouri | St. Louis, Cape Girardeau, Poplar Bluff | ~3.5 million | High in older construction; St. Louis has significant URM inventory |
| Arkansas | Little Rock, Jonesboro, Blytheville | ~1.8 million | Moderate to high — limited seismic code enforcement in rural areas |
| Kentucky | Paducah, Owensboro | ~1.2 million | High — minimal seismic design requirements historically |
| Illinois | Carbondale, Cairo | ~1.4 million | Moderate — southern Illinois lacks seismic construction standards |
| Indiana | Evansville | ~0.8 million | Moderate — some seismic provisions in building codes |
| Mississippi | Tupelo, northern counties | ~0.5 million | High — minimal seismic provisions |
| Alabama | Northern counties | ~0.2 million | High — minimal seismic provisions |
Source: FEMA, USGS, U.S. Census Bureau
According to FEMA, approximately 11.6 million people live within the NMSZ impact area across eight states. The built environment is the critical factor: unreinforced masonry (URM) buildings — brick buildings without steel reinforcement — are extremely vulnerable to earthquake shaking and are common throughout the region. Memphis alone has thousands of URM structures, including schools, churches, hospitals, and commercial buildings.
The Preparedness Gap
California has spent decades and billions of dollars retrofitting bridges, highways, hospitals, and schools for earthquake resilience. The central U.S. has not. According to the USGS, the fundamental challenge is that large NMSZ earthquakes occur infrequently enough that the hazard is easy to ignore, but the consequences are severe enough that they cannot be safely ignored.
Key vulnerabilities include:
The Interstate highway system — particularly I-55 and I-40 — crosses the NMSZ on bridges and elevated roadways that were not designed for significant seismic loading. The Memphis-Arkansas Bridge (I-55) and the Hernando de Soto Bridge (I-40), both critical Mississippi River crossings, could be severely damaged or destroyed.
Natural gas pipelines crossing the NMSZ serve millions of customers across the Midwest. Ruptures could cause fires and cut heating fuel during winter months.
The region's power grid, water systems, and wastewater treatment infrastructure have received limited seismic hardening. Unlike California, where utilities are required to plan for earthquake scenarios, central U.S. utilities have historically given minimal attention to seismic resilience.
Earthquake insurance penetration is extremely low. According to insurance industry data, fewer than 3% of homeowners in the NMSZ carry earthquake insurance, compared to roughly 10–13% in California.
Probability and Scientific Debate
USGS Hazard Assessment
The USGS National Seismic Hazard Model incorporates the NMSZ as a significant source of earthquake hazard. According to the most recent USGS assessment, the probability of a M6.0+ earthquake in the NMSZ within the next 50 years is estimated at 25–40%. The probability of a repeat of the 1811–1812 sequence (M7.5+) within 50 years is much lower — on the order of 7–10%.
These probabilities are significantly lower than the earthquake probabilities in California. The Uniform California Earthquake Rupture Forecast (UCERF3) estimates a 93% probability of a M6.7+ earthquake in Southern California and a 72% probability in the San Francisco Bay Area within a 30-year window (2014–2043). The NMSZ operates on a fundamentally different timescale.
The Stein-USGS Debate
One of the most prominent scientific debates in American seismology centers on whether the NMSZ is being overestimated as a hazard. Seth Stein and colleagues at Northwestern University have published extensively arguing that GPS data shows the NMSZ is not currently accumulating strain at a rate consistent with frequent large earthquakes. Their research suggests that the 1811–1812 sequence may have been a cluster event — a temporary burst of activity that released accumulated strain and may not repeat for thousands of years.
The USGS and CERI researchers counter that the paleoseismic record — evidence of prehistoric earthquakes preserved in sand blows and fault offsets — shows that large earthquakes have occurred in the NMSZ at intervals of roughly 500 years over the past several thousand years, suggesting the hazard is real and ongoing. Events have been identified around AD 900 and AD 1450, in addition to the 1811–1812 sequence.
This debate has direct policy implications. If the NMSZ is winding down, then the billions of dollars it would cost to retrofit buildings and infrastructure in the region may not be justified. If the USGS assessment is correct, the region remains dangerously underprepared.
FEMA Impact Scenario
What a Repeat Would Look Like Today
FEMA has modeled the impact of a repeat of the 1811–1812 earthquake sequence on the modern central United States. The results are sobering.
A M7.7 earthquake centered in the NMSZ would produce approximately 86,000 casualties (deaths and injuries combined), displace roughly 7.2 million people, damage or destroy 715,000 buildings, and cause direct economic losses exceeding $300 billion. These figures come from FEMA's 2008 catastrophic planning initiative for the NMSZ.
[CHART: Bar Chart — Estimated Impact Comparison: M7.7 New Madrid vs. M7.8 Southern San Andreas] Data: New Madrid M7.7 — 86,000 casualties, 7.2M displaced, $300B+ losses, 715,000 buildings damaged; Southern San Andreas M7.8 (ShakeOut scenario) — 1,800 deaths, 268,000 displaced, $213B losses, 300,000 buildings damaged. Source: FEMA, USGS ShakeOut Scenario. Note: New Madrid's much higher casualty and displacement figures reflect the lack of seismic building codes, not necessarily stronger shaking.
The comparison is striking. Despite the San Andreas Fault producing comparable or larger earthquakes, the projected impact in the NMSZ is dramatically worse — primarily because California has invested heavily in seismic building codes, retrofitting, and emergency preparedness, while the central U.S. has not.
For Memphis specifically, the FEMA scenario projects that 60% or more of the city's URM buildings could collapse or be rendered uninhabitable. Critical lifelines — water, power, transportation — could be disrupted for weeks to months. The economic ripple effects would extend far beyond the immediate damage zone, disrupting national supply chains that pass through the Memphis logistics hub (FedEx's global hub is located at Memphis International Airport).
Current Seismicity
The NMSZ is not silent. According to CERI, which operates a regional seismic network, approximately 200 earthquakes per year are detected in the zone. The vast majority are too small to be felt — below M2.0 — but occasional M3.0 to M4.0 events remind residents that the faults are still active.
The USGS locates approximately 20,000 earthquakes per year across the United States (roughly 55 per day). The NMSZ accounts for a small but consistent fraction of this activity. The persistent low-level seismicity confirms that the fault system is not dormant, even if the rate of large earthquakes remains debated.
[MAP: New Madrid Seismic Zone] Data source: USGS, CERI (University of Memphis). Features: Reelfoot Rift boundary, Reelfoot Fault, Cottonwood Grove Fault, New Madrid North Fault traces, historical epicenters of 1811–1812 mainshocks, modern seismicity (M2.0+ events from CERI catalog), and major cities (Memphis, St. Louis, Little Rock, Paducah, Evansville). Inset showing NMSZ location relative to tectonic plate boundaries.
Preparedness and Mitigation
Building Codes and Retrofitting
The adoption of modern seismic building codes in the NMSZ region has been slow and inconsistent. Tennessee adopted the International Building Code (IBC) with seismic provisions statewide in 2004, but enforcement varies by jurisdiction. Missouri, Arkansas, Kentucky, and Illinois have adopted seismic provisions in some areas but with significant gaps in rural counties.
Retrofitting existing URM buildings is expensive — typically $10 to $50 per square foot depending on the building and the level of retrofit — and there is no equivalent of California's mandatory retrofit ordinances for the NMSZ. The result is that thousands of vulnerable buildings remain unaddressed.
The Great Central U.S. ShakeOut
Modeled after California's successful ShakeOut drill, the Great Central U.S. ShakeOut has been conducted annually since 2011 to raise earthquake awareness in the NMSZ region. Participation has grown from roughly 2.4 million in the first year to several million annually, but awareness levels remain far below those in California.
What Individuals Can Do
Residents of the NMSZ face the same fundamental preparedness steps as those in any earthquake zone. Securing heavy furniture, maintaining emergency supplies (water, food, first aid, medications for at least 72 hours), and developing a family communication plan are the foundations of earthquake readiness. Because emergency response in the NMSZ would be complicated by limited transportation infrastructure and the wide geographic area of damage, self-sufficiency for at least seven days is advisable.
What to Do During an Earthquake — Step-by-Step Guide
Is Earthquake Insurance Worth It?
The New Madrid Seismic Zone Compared to Other U.S. Fault Systems
The NMSZ is often compared to the San Andreas Fault in California, but the two systems are fundamentally different. The San Andreas is a plate boundary fault with a well-documented slip rate of 20–28 mm/year and a recurrence interval for major earthquakes of roughly 100–300 years depending on the segment. The NMSZ has no measurable surface slip rate, and its recurrence interval for major events is measured in centuries to millennia.
The Hayward Fault in the San Francisco Bay Area and the Wasatch Fault in Utah share one important characteristic with the NMSZ: they all threaten major population centers that may not be fully prepared. But each operates through different geological mechanisms and on different timescales.
What makes the NMSZ unique is the combination of infrequency and severity. Large earthquakes are rare enough that institutional memory fades and preparedness lapses, but powerful enough that a repeat event would overwhelm the region's capacity to respond.
What Causes Earthquakes? The Science Behind Seismic Activity
Tennessee Earthquake History and Risk
Missouri Earthquake History and Risk
Sources
- USGS, "New Madrid Seismic Zone" — USGS New Madrid Seismic Zone
- Center for Earthquake Research and Information (CERI), University of Memphis — CERI Homepage
- FEMA, New Madrid Seismic Zone Catastrophic Planning (2008)
- Johnston, A.C. & Schweig, E.S. (1996), "The Enigma of the New Madrid Earthquakes of 1811–1812," Annual Review of Earth and Planetary Sciences
- Stein, S. (2010), "Disaster Deferred: A New View of Earthquake Hazards in the New Madrid Seismic Zone," Columbia University Press
- Tuttle, M.P. et al. (2002), "The Earthquake Potential of the New Madrid Seismic Zone," Bulletin of the Seismological Society of America
- USGS National Seismic Hazard Model — USGS NSHM
- FEMA — FEMA