ON SHAKY GROUND

The earthquake hazard here may be greater than you think. And planning for it is less than you'd expect.

By C.D. Stelzer

published: December 15, 1999

There are mysteries in southeast Missouri that have drawn state geologist David Hoffman here time and again. What makes this place different, sets it apart from the rest of the planet, is an upheaval that occurred eons ago, when the earth was younger and more rambunctious. The theory is that hundreds of millions of years in the past, the North American continent tried to rip itself apart here, creating a rift in the earth's crust. The resulting depression allowed the Gulf of Mexico to extend hundreds of miles northward into what is now southern Illinois. As the sea receded, the Mississippi and Ohio rivers formed. Driving south on Interstate 55 past Cape Girardeau, Hoffman tells how the course of the Mississippi subsequently changed, for unknown reasons. "Back there where we just passed the airport, (the Mississippi) used to go to the west," says Hoffman. "Now it cuts through here just south of Cape Girardeau at a place called Thebes Gap and joins the Ohio down at Cairo, Ill." He has by now entered the Benton Hills, steering his Ford Taurus across a five-mile-wide forested incline reminiscent of the Ozark Plateau. Enigmatic in their own right, the gentle slopes seem to have become somehow separated from the pack, leaving them marooned as a remote series of elevations in the otherwise flat expanse of alluvial plain.

The Benton Hills, Hoffman explains, are an isolated segment of Crowley's Ridge, which runs down to Helena, Ark., a distance of about 200 miles. Why the river abandoned its original path more than 11,000 years ago and carved its way through 100 feet of bedrock so it could flow to the east of Crowley's Ridge is the subject of scientific speculation, as are the origins of the ridge itself.

"It's a very strange formation," Hoffman says. "The traditional explanation is that it's just an erosion element that the river left there. I find that hard to accept. It may have been formed by earthquake-related activity."

For nearly half of his 20-year stint with the Missouri Department of Natural Resources (DNR), Hoffman has been an earthquake specialist in the Division of Geology and Land Survey. On this day, he is carrying the tools of his trade in the trunk of his car: geological charts, work boots and a hoe. With his graying flattop and aviator glasses, Hoffman looks more like a high-school teacher than a haughty scientist, and when he talks about the Benton Hills, he sounds more like an awestruck student.

Hoffman's hypothesis that earthquakes helped form these hills is supported by his own field research. Beginning in 1993, he supervised the excavation of more than a dozen 15-foot-deep trenches in the uplands near Commerce, Mo. The digs uncovered folds in the soil strata showing that seismic activity occurred here in the past. This evidence leads him to believe that the faults in the area remain active.

Hoffman's opinion is bolstered by other seismic authorities, who acknowledge that much remains to be learned about the area's subterranean history. Despite these affirmations, the minimal amount of research funding for the Benton Hills project has been curtailed.

Hoffman's pursuit raises the probability of earthquake activity at an obscure location. But the one recognized cataclysmic source in the state is the New Madrid seismic zone, a 120-mile-long system of faults that traverses the Missouri Bootheel and cuts across five adjacent state lines. And, indications are, that fault system may be more restless than previously thought.

On the basis of its latest mapping efforts, the United States Geological Survey (USGS) issued projections that increase the estimated earthquake hazard for parts of the St. Louis area by 30-40 percent. That dire caveat has been followed by an insurance-industry request to the state to raise earthquake-coverage premiums by as much as 266 percent. The proposed rate hike is based on the industry's own analysis of the latest scientific data -- which point to a significantly increased probability of a damaging earthquake's striking the region.

In the face of this clear evidence, government spending on seismic research and emergency preparedness has nonetheless remained stagnant and, in some instances, has been drastically slashed. The one exception to the rule came last week, when the Missouri Department of Transportation asked for a $32 million increase in the cost of seismically retrofitting the elevated sections of Highway 40 (Interstate 64) downtown, raising the project's cost to more than $65 million. By contrast, the total federal allocation for all types of disaster preparedness in the state of Missouri, the city of St. Louis and St. Louis County amounts to about $300,000 this year. As for the official recommendations of the Missouri Seismic Safety Commission, they are buried under an avalanche of indifference.

Meanwhile, development of the vulnerable floodplains -- Chesterfield being a prime example -- continues unabated.

Geology students call it "the Beach." Located in Pemiscot County, Mo., near the town of Deering, the 136-acre strand is purported to be the largest sand boil in the world, a vestige of the New Madrid earthquakes of 1811-1812. Aerial photographs of the Missouri Bootheel show similar geologic formations throughout the fault zone. Underground pressure associated with the seismic disturbance created tiny volcanoes, which belched up a mixture of sand, mud and water.

The zone -- which zigzags northeasterly from Marked Tree, Ark., to Cairo, Ill. -- is considered the second most active earthquake region in the United States. It is named after the small town of New Madrid, Mo., which bore the brunt of the earthquakes that beset the region early in the last century. Scientists reckon the New Madrid quakes to be among the largest seismic bursts in the history of the United States.

The first of a series of earthquakes struck the region in the early-morning hours of Dec. 16, 1811. Major shocks also occurred on Jan. 23 and Feb 7, 1812. In each instance, the earth's movement was felt as far away as the East Coast. On the basis of eyewitness accounts from the time, scientists now estimate that the quakes reached 7.5 or more on the Richter scale. Historical accounts tell how the quake created at least two waterfalls in the Mississippi River and caused the current to temporarily reverse its flow. The uplifting and subsidence of ground in nearby western Tennessee resulted in the creation of Reelfoot Lake, which today is 16 miles long and four miles wide.

Though the seismic event irrevocably changed the face of nature, destruction of human life and habitat was minimal simply because few people then lived in the region. St. Louis was a small pioneer village in 1811-12, and Memphis didn't even exist. If a temblor of the same magnitude occurred today, the effects would be far more devastating.

It has been more than a century since St. Louisans experienced quivering originating from the New Madrid area. That quake, which was centered near Charleston, Mo., occurred at about 5:10 a.m. on Oct. 31, 1895. The shaking lasted almost a minute. About 150 miles from the epicenter, ground movement did structural damage to Holy Ghost German Lutheran Church at Eighth and Walnut streets in downtown St. Louis.

The long lapse since the last major earthquake in the New Madrid area has led many scientists to speculate that a moderate-to-severe earthquake is overdue. Other experts downplay the possibility of a large-scale natural disaster.

The late Iben Browning, a business consultant and climatologist, had predicted a 50-50 chance of an earthquake along the New Madrid Fault between Dec. 1 and 5, 1990. Browning based his prognostication on the gravitational pull of the moon. Most scientists didn't take the prediction seriously, but intense media coverage of the forecast fueled an overreaction. Concerned citizens flooded government agencies with calls. Fear led many people to stockpile food and water. Meetings were held and emergency drills initiated. Retailers capitalized on the expected disaster by advertising specials on everything from guns to water-purification systems. As doomsday approached, schools and businesses closed.

Despite the absurdity of Browning's prediction, it temporarily galvanized popular awareness of the hazard and spurred public officials to address a long-ignored issue.

Since then, the scientific debate has continued, reaching a new crescendo earlier this year, when a group of scientists from Northwestern and other universities published research based on a seven-year study. The project involved monitoring ground movement through satellite observations. Using a network of portable Global Positioning System (GPS) receivers, the scientists sought to ascertain the amount of strain on the fault zone. Their findings led them to conclude that the earthquake hazard in the New Madrid seismic zone has been overestimated.

The study, published in the journal Science in April, prompted a flood of letters from experts in the field who disagreed with the conclusions, but the periodical refused to publish them. One of the harshest critics is Arch Johnston, who directs the Center for Earthquake Research and Information at the University of Memphis. "The conclusions that they drew were very premature because of the uncertainties," Johnston says. "The uncertainties are big enough that they allow zero strain or they allow strain comparable to current estimates of the hazard -- either one." In other words, the margin of error, as defined by the researchers, was larger than the detected ground motion. Within the next few years, Johnston expects to have result from his own research, which will include data from a dozen permanent GPS monitoring stations.

Another article in the Nov. 5 issue of Science contradicted the Northwestern study. In this study, scientists dug trenches near Reelfoot Lake to determine the amount of ground movement. Folds in the soil strata indicated a shift of 5 or 6 mm a year, which equates to about a 3-foot change in position since the great earthquakes of 1811-12.

Johnston says the gradual movement of the fault is analogous to the stretching of a rubber band: Eventually something must give. "The whole rubber band is stressed," says Johnston, "but it snaps only in one place." Once the earth's elasticity stretches beyond its breaking point, its reverberations will cause catastrophic consequences. Another major New Madrid quake might come tomorrow, or it could happen decades from now.

Nobody knows.

But MoDOT isn't taking any chances. The agency, which asked for $32 million to retrofit Highway 40 downtown to withstand an earthquake, says it based its call on the work of Johnston and his cohorts at the University of Memphis. On Tuesday, the agency's request was approved by the East West Gateway Coordinating Council, which oversees all federal highway spending for the St. Louis area.

The Missouri State Emergency Management Agency (SEMA) anticipates that a 6.7 shock would significantly damage more than 32,000 buildings in eastern Missouri. Many buildings would totally or partially collapse. Some houses could shift on their foundations. Towers and chimneys would fall. The estimated damage represents 6 percent of the total number of structures in the affected area and would be comparable to the 1994 quake in Northridge, Calif., which killed 57 people, injured 1,500 and caused $15 billion in property damage.

In contrast to California, however, the consequences here could be more far-reaching because faults in the Mississippi Valley are buried under sedimentary deposits up to a mile deep. These conditions allow seismic waves to travel as much as 20 times farther than they do in California. As a result, a moderate New Madrid quake would shake a seven-state region -- Missouri, Illinois, Arkansas, Tennessee, Kentucky, Mississippi and Indiana -- like a bowl of jelly.

"If you have an earthquake in New Madrid, I can't guarantee you that any of those bridges from Vicksburg, Miss., to Louisiana, Mo., are still going to be functioning," says Ed Gray of SEMA. Under such circumstances, all rail shipments would, out of necessity, be routed through Chicago, resulting in massive delays in the delivery of goods. It could also stop highway traffic along Interstate 55, impede barge traffic on the Mississippi and Ohio rivers, and disrupt air service to the hub cities of St. Louis and Memphis. Electrical-power outages would almost inevitably occur.

But what might be the most serious danger from the next New Madrid earthquake pertains to another valuable resource that courses through the region. "You have four of the five major natural-gas pipelines come right through the soup in New Madrid, the soft alluvial soil," says Gray. "They carry gas all the way to Detroit, Chicago, Indianapolis and Pittsburgh. If (the earthquake) happened during the winter, you're going to have major-league problems on your hands. Try to explain to somebody why you cannot heat a nursing home or keep a hospital warm."

Unfortunately, the prospect of future New Madrid rumblings is not the state's only seismic worry. Recent jolts to the region have stemmed from disturbances outside what is normally defined as New Madrid seismic zone. The fault that is not even known could wreak the greatest havoc. "If any one of these goes off periodically, and any of them happen to be near a metropolitan area, (it) can do tremendous damage," says Buddy Schweig of the USGS in Memphis. Seismologists weren't even aware of the fault that caused the 1994 Northridge earthquake until after the ground started moving.

Nothing precludes a similar freak occurrence from happening here. Indeed, there is ample precedent for such an anomaly. On Nov. 9, 1968, for example, the seismographs in the St. Louis University geophysics department recorded an earthquake that registered 5.5 on the Richter scale. The epicenter was located 120 miles southeast of St. Louis, near the Illinois-Indiana state line. According to various reports, the quake made the Gateway Arch bounce, caused cars parked near the St. Louis County Courthouse to skip a few inches and cracked the rear wall of the synagogue on Skinker Boulevard that now houses the Missouri Historical Society Library. An 11-year-old boy sustained a head injury when he was hit by falling debris outside his home on South 11th Street. Brick chimneys and parapets rained down into the streets across the city. The tremor was felt as far away as Milwaukee.

Martin J. Walsh, then deputy building commissioner for the city of St. Louis, recalls helping coordinate the emergency response. Building inspectors, with the aid of the fire department, were dispatched and initially determined that 16 buildings had sustained structural damage. Fortunately for one city official, the earthquake took place on a Saturday. "After we thought we had cleaned it all up, we went to work Monday morning and found the plaster had fallen in the safety director's office in City Hall," says Walsh. "It would have killed him if that would have happened when he was there."

Ten years later, on Sept. 20, 1978, another twitch of the earth -- this one registering 3.5 on the Richter scale -- rattled the St. Louis area. As earthquakes go, it amounted to barely a squiggle on the seismograph. No damage or injuries were reported, but the tiny tic nonetheless remains noteworthy because it was centered virtually under somebody's backyard in South St. Louis. The source of the spasm was the St. Louis Fault, which was first discovered running through a quarry on South Broadway in 1948. The geologist who discovered the fault subsequently traced its path from Valmeyer to Alton, Ill., a distance of 45 miles.

In the past decade, the strongest tremor in the region has come out of the Benton Hills south of Cape Girardeau. It registered 4.6 on the Richter scale and was felt over a seven-state region on Sept. 26, 1990.

Seismographic data indicated that the epicenter was located near New Hamburg in Scott County, Mo., which is outside the New Madrid seismic zone. After the shaking stopped, the DNR and the USGS decided to take a closer look at the Benton Hills, and this is when Hoffman and his team began their trenching project on a farm a few miles south of New Hamburg. The excavations showed folds in the soil layers suggesting that seismic activity had occurred within the last 10,000 years.

"That's kind of a wink of the eye in geologic time," says Hoffman. "There are lots of faults in the area, and it might (indicate) the potential for earthquakes in the future. In the past, I don't think they were even considered. It's just the way people thought at that time. They didn't think there was a problem with earthquakes in the Midwest. It takes studies of lots of these kinds of places and features and compiling all the data and analyzing it together to get the whole picture. We just kind of have one piece of the puzzle with this spot. We're the primary ones that have even looked here. Then there's the rest of the state that hasn't been looked at."

Hoffman's tectonic revelation has garnered no interest, though, outside the scientific community. More significantly, there are no plans to investigate his findings further. "Because of funding considerations, we had to stop doing this basic field research," he says. "We don't have the time or the people to do it, and it's an expensive type of activity."

In the preface to its 1997 report, the Missouri Seismic Safety Commission warned that the threat of an earthquake should not be ignored. The commission, established by the state Legislature in 1993, comprises 15 members appointed by the governor, including volunteers from the architectural and engineering professions. The group, which holds no statutory authority, advises and makes recommendations to the state government. But to date, few resources have been allocated to carry out the commission's suggestions.

The Missouri DNR and SEMA are the two agencies that deal with earthquake-related issues in the state. One percent of the DNR's Division of Geology and Land Survey annual budget -- a mere $71,000 -- is devoted to the subject. No field research is now being conducted by the agency, and Hoffman is the only full-time employee assigned to earthquake duties.

As for SEMA, that agency receives only about $65,000 a year for earthquake preparedness from the state. By comparison, Gov. Mel Carnahan's annual salary is $112,755.

And even that feeble effort by SEMA may be of dubious value. For example, the Center for Earthquake Studies at Southeast Missouri State University doesn't really study anything. The staff is made up of two students who are employed part-time through the school's work-study program. They answer the phone when a tremor shakes the region. In addition to that courtesy service, the center publishes brochures and gives occasional seminars at elementary schools.

SEMA also gets about $65,000 from the Federal Emergency Management Agency. The vast majority of federal earthquake assistance goes for programs on the West Coast, including more than half of the $4 million in FEMA dollars allocated to the states. "California gets the main meal and we get the table scraps," says Johnston, the University of Memphis scientist.

In the city, the entire budget of the St. Louis Emergency Management Agency has remained at $90,000, all of which comes from FEMA. That total goes toward planning not only for earthquakes but for every other imaginable disaster. "We survive a lot on the goodwill of the people that we work with," says Ken Walk, the spokesman for the city agency. "We have police officers who teach classes for us and help us prepare the plans that we have. We work real closely with the fire department. We save a lot of money in that respect. Is more money the answer all the time?" asks Walk. "Not always. Sometimes it's a matter of focus."

If that be the case, myopia afflicts St. Louis County, where emergency planners must try to coordinate preparedness within a jurisdictional quagmire that includes 91 municipalities and nearly a million people. "The county is an absolute nightmare," says one public official who requests anonymity. "Those guys are scrambling all the time. They do a relatively good job, but it's like an extended family that doesn't really get along."

The St. Louis County Emergency Management Agency, which is part of the county police department, has a 12-member staff and a budget of $600,000 (including $150,000 from FEMA) to cover the full gamut of emergencies -- not just earthquakes. This year's allocation -- approximately 60 cents per person -- is expected to ensure the safety of a 524-square-mile area that contains more than 400,000 households, or 20 percent of the state's population. In other words, the annual per capita appropriation for emergency management in St. Louis County doesn't even equal half the fare for a single ride on a Bi-State bus.

In the aftermath of an earthquake, a group of professional volunteers will be called upon to inspect buildings. The program, which is coordinated by SEMA, enlists the help of architects and engineers. But in the absence of a statewide building code and other enforcement mechanisms, the state emergency officials have no regulatory authority to enforce recommendations before a disaster strikes. Instead, Gray of SEMA relies on what he calls the "jawbone" method of persuasion. "You try to reason with people that they should be doing the right things by including seismic design in their new buildings," he says.

The state emergency-management official would like to see the state budget increased for earthquake readiness, but he also realizes that the problem spans political boundaries. Gray notes that FEMA has cut back the funding of the Central U.S. Earthquake Consortium in Memphis, which is charged with coordinating preparations for disaster relief in the seven-state region. A spokesman for that agency says its budget has been cut by 80 percent in the last two years. The reduction reduces the agency's budget to about $350,000 a year.

Another federal agency, the USGS, spends a few hundred thousand dollars monitoring ground movement over the entire length of the New Madrid seismic zone, splitting the stipend between the University of Memphis and St. Louis University. In 1996, the same agency issued projections that increase the estimated earthquake hazard for parts of the St. Louis area by 30-40 percent.

Early last year, the Insurance Services Office (ISO) of New York, a firm that represents 500 insurance companies doing business in Missouri, filed for rate increases with the state Department of Insurance. The requested rate hike would raise the price of earthquake coverage between 113 and 266 percent in portions of the St. Louis metropolitan area. The highest increases are planned for parts of St. Charles County, sections of north St. Louis County and the northern and southern tips of the city of St. Louis.

ISO relied on data analysis purchased from Risk Management Solution Inc., which specializes in computer modeling. "We were pretty unfamiliar with the way that works," says Mark Doerner, senior counsel for the Missouri Department of Insurance. The state regulatory agency is allowing the insurance companies to phase in the commercial rate hikes over a three-year period. It has placed a moratorium, however, on the requested increases to homeowners insurance. Doerner has used the delay to study esoteric earthquake research to see whether the insurance companies' data analysis holds up under scrutiny. He expects to complete his investigation soon.

With billions of dollars of property damage estimated as the result of a future Midwestern earthquake, the insurance companies are seeking to hedge their bets. "How do you make coverage available and how do you encourage the insurance companies to write it, if there is a potential for them taking such a hit?" asks Doerner.

There is very little chance of the state Legislature's establishing a reinsurance program such as the one in California, because only a handful of the state's 110 counties would likely be affected by a quake. As a result, the price of private insurance in high-risk areas is becoming unaffordable. "Some of the numbers from the Bootheel indicate that you would have to spend a thousand dollars (annually) for a $100,000 structure," says Doerner.

That leaves the possibility of rate relief up to federal legislation now being considered by Congress. The proposed law would authorize the Treasury Department to sell contracts to back up, or reinsure, homeowners' insurance issued by a state program or private firms to cover damages brought on by earthquakes and hurricanes. The bill would limit the Treasury Department's liability to $25 billion. Conservatives decry the proposal for interfering with the free-market economy. Liberals oppose the measure because they view it as corporate welfare. Environmentalists don't like the idea because it would spur development in ecologically sensitive areas. The bill is expected to be voted on sometime next year.

Even though insurance executives, government bureaucrats and academic experts disagree about the timing or magnitude of an earthquake, none questions that at least a moderate one will rock the area sooner or later. When that happens, chunks of the St. Louis infrastructure will fall apart. Building codes in the city of St. Louis and St. Louis County have only included seismic-safety standards since 1987. Structures built before that date are not required to adhere to the current guidelines unless the building undergoes major renovation. Even then, the building does not have to be seismically retrofitted unless there is a change in its usage, according to the St. Louis building code.

"Unreinforced masonry buildings are probably going to have the most damage and sometimes complete collapse," says John C. Theiss, a structural engineer for EQE-Theiss in St. Louis. EQE specializes in designing buildings to withstand earthquakes. In the city of St. Louis, an estimated 66,000 residences -- 75 percent of the housing stock -- are unreinforced brick structures, according to the Missouri Seismic Safety Commission. Almost all of the two- and four-family flats lining the city's residential streets were constructed long before the seismic-safety standards came into effect. Some of these structures were built more than a century ago.

St. Louis Building Commissioner Ron Smith downplays the effect that a quake would have on the city but admits that risks exist. "We've got a whole host of different types of buildings in the city, from frame residential all the way to high-rise masonry and heavy-timber warehouses downtown," says Smith. "Just about any structure could be vulnerable, depending on the intensity of the earthquake and the condition of the soils the building is sitting on."

In its 1997 report, the Missouri Seismic Safety Commission says that mandating structural improvements of the city's brick residences would be prohibitively expensive and could result in the abandonment of many such structures. The commission recommended incentives such as tax and insurance reductions to entice homeowners into retrofitting their houses to withstand an earthquake. No such programs have been established yet.

Although the St. Louis County Department of Public Works has assessed the hazards to some extent, mitigating the damage to schools, public buildings and emergency response facilities has not been mandated, either. The commission lists scores of hospitals, police stations, fire stations and ambulance services in the St. Louis area that would be vulnerable to the effects of an earthquake.

Nowhere are the economic risks higher, however, than areas such as Chesterfield, where river bottoms are undergoing rampant development. Soils in the floodplain are typically unstable and subject to liquefaction in the event of an earthquake. The sand boils in southeast Missouri, which remain evident from the New Madrid earthquakes of 1811-12, were caused by liquefaction. The phenomenon often occurs when seismic activity transforms loosely packed sand and silt into a fluid mass as a result of extreme pressure in the groundwater. The water table in the Missouri River bottom in St. Louis County is sometimes only a few feet from the surface. Whereas levees can redirect floodwaters elsewhere, nothing can stop a major earthquake from shaking the alluvial plain and turning parts of it into a giant cache of quicksand.

"Good firm soil will not amplify the ground motion as much as a weak soil," says Theiss, who is a member of the Building Seismic Safety Council, a professional group that oversees building standards nationwide. "If you're near a river and you build on soil that was previously on a floodplain, that's the worst type of soil, because it magnifies the ground motion." Design techniques exist that can compensate for building on unstable soil, but the latest upgrade in the USGS recommended standards are not currently being applied, Theiss says.

"The conclusion that they (USGS) have reached, as far as the St. Louis area is concerned, is that the ground motion that we should be designing for is greater than the current ground motion that's specified in the building code," Theiss says. Developers, however, are still only meeting the minimum code specifications, Theiss says.

"There's nothing that prohibits you from going beyond the code," says Michael Werner, St. Louis County deputy director of code enforcement. "It makes sense to construct just about anywhere, so long as you take the proper precautions," he adds. "That's why you have engineers who not only comply with the law but also use what's called sound engineering judgment in the design."

But sound judgment often bows to the bottom line. "We tell our clients that there are new maps out there, and we recommend designing to them," says Theiss. "(But) they say: "We're just going to go by the code, period.' That's the prevailing attitude." Those code standards are subject to change again next year when a new nationwide building-code standard code takes effect, Theiss says.

The situation outside the St. Louis area is much worse because of a lack of a statewide building code. State law actually prohibits smaller counties from establishing an office to enforce building codes. At the same time, another Missouri law, enacted in 1991, mandates that towns and counties where earthquake damage is anticipated require new buildings to meet national seismic safety standards. The statute applies to more than three dozen counties in the eastern part of the state. It's a classic Catch-22, because there is no means of enforcing the rule.

The laws of nature are not bound by human contradictions, though, and in the Benton Hills the earth itself is deliberating its next move.

Hoffman, the state geologist, is standing in a 15-foot-deep trench that was excavated on the farm of Leonard Weber in Scott County. The excavation is one of a pair that have not been refilled since he completed his research a few years ago. He scrapes the walls of the trench with a hoe. Turning the implement around, he uses its handle as a lecture tool, pointing out a distinct soil layer that juts up toward the top of the ditch.

"The Peoria loess started to be deposited about 25,000 years ago and ended deposition about 10,000 or 12,000 years ago -- and it's offset," says Hoffman. "So we know the fault had to happen after that was deposited. That tells us that it happened within the last 10,000 years. There's been two or probably three periods of movement in the last 10,000 years. That suggests a recurrence interval somewhere on the order of 3,000-5,000 years. The fact that these faults go up to the surface suggests moderate to large-scale earthquakes."

For now, the secrets of the Benton Hills defy further explanation. "There are many lifetimes' worth of study here," says Hoffman. "It's had a long, complicated history. We've just sort of scratched the surface on it."