The Anna (Shelby County) earthquake on March 9, 1937 was the strongest earthquake to strike Ohio. Few seismographs were available at that time; however, based on the felt area of the earthquake and the damage that occurred in Anna and surrounding communities, the U.S. Geological Survey assigned a magnitude of 5.4 to this earthquake.

As a general rule the size (magnitude) of an earthquake is related to the length of rupture of a fault. That is, big faults generate big earthquakes and small faults generate small earthquakes. We can speculate on the potential length of rupture on known active faults in Ohio but inadequate information is available to make such predictions with a high degree of credibility. Most faults in Ohio are poorly known and are not visible at the surface. Seismologists have speculated that active seismic zones in Ohio could theoretically generate an earthquake at least an order of magnitude larger than the largest historic event. Conservatively, they have suggested that the western Ohio seismic zone could generate an earthquake with a magnitude of between 6.5 and 7.0 and the northeastern seismic zone could generate an earthquake with a magnitude of between 6.0 and 6.5.
 
At this time, there is no evidence that such large earthquakes have occurred in the last 10,000 years in Ohio. However, the historic record of earthquakes in the state is a little more than 200 years, which is short, geologically speaking, and large earthquakes in the eastern United States typically have long recurrence intervals, on the order of centuries or millennia. If Ohio has the potential for an earthquake above 6.0 magnitude, it might occur tomorrow or perhaps thousands of years from now.

About 2.0 magnitude is generally the lower range of an earthquake being felt, although there are documented cases of people feeling earthquakes in the high 1+ magnitude range. Many people close to the epicenter of such small events may feel them but the vibrations are so slight that they commonly do not recognize them as being from an earthquake.

Historically, there have been 15 earthquakes in the state that have caused minor to moderate damage. However, the probability of such damage in any given year is low, but this varies in different parts of the state. Seismically active zones, such as Allen, Auglaize, Mercer, and Shelby Counties in western Ohio and Ashtabula, Cuyahoga, Geauga, and Lake Counties in northeastern Ohio have a higher risk than does central and eastern Ohio, which have had almost no earthquakes. The epicenter map of Ohio [107 KB PDF] provides a depiction of historical seismicity. Southwestern Ohio has earthquake risk from its proximity to the New Madrid, Missouri seismic zone. Building construction and the type of geologic material upon which the home is built are significant factors as well. Un-reinforced masonry buildings are more susceptible to earthquake damage than are buildings of frame construction. Buildings built on thick, unconsolidated sediments (such as fill along river valleys) are likely to experience greater shaking than buildings built on or close to bedrock.

In making the decision to purchase an earthquake rider on building insurance, one should evaluate the factors noted above and determine the deductible amount on the rider. Then one should consider the probability that damage from an earthquake will exceed the deductible during the projected time of ownership of the building. However, this is not an exact or predictable science and individuals should make this decision on both facts and level of comfort in having, or not having, earthquake insurance.

Most small earthquakes in Ohio (2 or 3 magnitude) are felt throughout a part of a county or several counties, and most strongly near the epicenter. Many people describe a booming sound followed by a sharp jolt and a few seconds of shaking. Some describe the sound as decreasing to a low rumble as it quickly passes into the distance. Commonly, the booming sound is at first interpreted as a sonic boom from an aircraft or an explosion ("I thought the furnace had exploded"). Although military aircraft do occasionally and inadvertently break the sound barrier, this is uncommon. The booming sound many people report is thought to be from P waves (the fastest moving seismic waves) exiting the ground and in audible frequencies. Some of the noise is probably related to the shaking of the building and to the brittle bedrock as the seismic wave pass through. Many people in favorable positions (sitting or resting) feel vibrations that they interpret as possible earthquakes. In most cases these felt events are not attributable to earthquakes. It is probable that they are local atmospheric or cultural in origin.

The Ohio Seismic Network encourages everyone to report shaking that they think might be an earthquake. These reports should be filed on the U.S. Geological Survey's Community Internet Intensity Map web site http://pasadena.wr.usgs.gov/shake/cus/ (link also on the OhioSeis Web site). The reports from Ohio and adjacent areas are immediately forwarded via e-mail to the Ohio Seismic Network, where records from the network's 29 seismograph stations are evaluated to see if it was indeed an earthquake. Telephone reports should be directed to the toll-free Ohio Earthquake Hotline at 1-855-QuakeOH (782-5364), which is staffed 24/7.

A glance at the Catalog of Ohio Earthquakes would suggest that earthquakes are much more frequent, beginning in the 1980's. This statistic is misleading, however. The comparatively small number of earthquakes in the 1800's can be attributed to a small population, lack of awareness of brief jolts or vibrations being earthquakes, and lack of rapid and widespread communications. Most earthquakes noted in the historic record prior to the late 1970's are greater than 3.0 magnitude, suggesting that many felt earthquakes in the 2.0 - 2.9 range were not reported by newspapers or not attributed to earthquakes. In addition to greater public awareness of earthquakes, the installation of numerous seismic stations in Ohio and a central reporting center (Ohio Earthquake Information Center) has resulted in the confirmation of many earthquakes that formerly may have escaped notice. The record of earthquakes in Ohio since the 1980's is skewed from the normal distribution because of the large number of felt earthquakes since 1987 at Ashtabula, Ohio. These earthquakes (at least 40) are thought to be associated with a now-abandoned deep injection well at Ashtabula. It is interesting to note that the second-most active decade of earthquakes was in the 1930's, when the Western Ohio Seismic Zone (Anna Zone) was very active.

It is well documented that earthquakes can have significant effects on water wells. The shaking associated with an earthquake may cause sand to plug a well screen, therefore reducing the volume of water that can be pumped. Conversely, the shaking can dislodge sand plugging a well screen and cause an increase in the volume of water that can be pumped from the well. Both of these phenomena have been widely reported from the epicentral areas of Ohio's larger earthquakes, including the 1937 Anna earthquakes and the 1986 Chardon earthquake. In some cases the well returns to its normal state but in others the well needs to be serviced to restore former production volume. Production spikes from oil and gas wells have been reported following a local earthquake. Very large earthquakes at great distances can also cause the water table to temporarily rise and fall when the long-period surface waves pass through the state. The 7.9-magnitude Denali, Alaska earthquake on November 3, 2002 caused water-level changes in some Ohio wells.

Many seismologists and emergency planners consider the New Madrid, Missouri seismic zone to pose the greatest threat to Ohio, particularly the greater Cincinnati area of southwestern Ohio. The great earthquakes in the New Madrid zone, from December, 1811 through February, 1812, are thought to have had magnitudes in the 7-8 range. They were strongly felt throughout most of Ohio and caused chimneys to fall in Cincinnati. It is difficult to evaluate the effects that a recurrence of these earthquakes would have on Ohio today. In 1811, much of Ohio was still a wilderness and the population was about 40,000. Cincinnati was the largest town in Ohio, with about 2,500 residents. The dense population, infrastructure, and abundance of older, non-reinforced masonry buildings in southwestern Ohio would contribute to the vulnerability of the area. Structures built on unconsolidated valley-fill sediments would be particularly vulnerable because of the tendency of these sediments to amplify ground motion.

The threat from seismic zones in Ohio is difficult to assess because the historical record is so short. We know that the Western Ohio (Anna) Seismic Zone and the Northeastern Ohio Seismic Zone can produce earthquakes in the 5 magnitude range and some seismologists suggest that each of these zones could produce earthquakes at least one whole magnitude or more larger. Such earthquakes would probably produce significant damage.

It is an interesting speculation to consider our perception of the New Madrid Seismic Zone if the great events of the early 1800's had occurred a century earlier, when there was no significant European settlement in the Midwest. It is probable that eastern settlements would have noted a slight shaking and Native Americans in the area may have had oral accounts of great earthquakes in the past. Geologists would have noted surviving topographic features suggesting great upheaval; however, without the many first-person accounts recorded during the New Madrid sequence, we would probably not interpret these earthquakes to have been so large and the public would not perceive them as indicative of a significant threat.

In general, earthquakes in the eastern US have a felt area about ten times larger than a comparable size earthquake in the western US. This is a phenomenon that seismologists call attenuation. In Ohio, the rocks are nearly flat-lying, brittle, and cold. These characteristics are favorable for seismic waves to travel longer distances without losing significant energy. In contrast, California earthquakes, for example, travel through less consolidated, warm rocks that are interrupted by mountain ranges and other features that tend to absorb (attenuate) seismic energy. Although larger earthquakes are much less common in the eastern US than in the west, there is concern that much more damage, over a larger area, would result from an eastern earthquake of moderate size.

The 5-kilometer depth assigned to most Ohio earthquakes is a predetermined or fixed depth used in the location model. In most cases it is very difficult to determine the true depth to the point of initiation (hypocenter) of the earthquake unless the distance from a seismic station to the epicenter is equal to or less than the depth of the earthquake. We do know that Ohio earthquakes for which depth determinations have been made, are usually at depths of 10 kilometers or less and commonly at or near the 5 kilometer fixed depth. These depths are in the upper part of the Precambrian crystalline rocks that underlie the state.

It is fortunate that no one has died from injuries from an Ohio earthquake. Minor injuries were reported from the 1986 Chardon earthquake in northeastern Ohio. These consisted primarily of minor cuts and scrapes from falling ceiling tiles. It is perhaps amazing that no deaths or serious injuries were reported for the 1937 earthquakes that struck Anna, in western Ohio. Nearly every chimney in town was damaged and many chimneys toppled, some crashing through roofs into rooms below. Anyone unfortunate enough to have been struck by falling bricks would certainly have suffered severe consequences. Falling objects pose the greatest threat to people during moderate-to-strong earthquakes.

The fastest-moving seismic waves, P waves, travel at about 6 kilometers per second. The Cincinnati area is about 500 kilometers from the New Madrid zone; therefore, P waves from a large New Madrid event would reach southwestern Ohio in a little less than a minute and a half and the more damaging S waves would arrive about a minute later. This would not be enough time to warn people to take cover, but automatic electronic warning systems could be triggered to shut down some critical facilities.

The type of material upon which a building is constructed greatly influences the amount of shaking that will occur from an earthquake. Thick, unconsolidated sediments, such as sand and gravel deposits or alluvium along river valleys tend to cause earthquake waves to slow down, but increase in amplitude. This, in turn, causes greater shaking and potential damage. This is known as site amplification. Buildings constructed on bedrock tend to experience much less ground motion and, therefore, less damage.

This phenomenon was clearly noted by Daniel Drake in Cincinnati during the great New Madrid, Missouri earthquakes of 1811 and 1812. Drake indicated that the shocks frightened everyone and knocked down chimneys in Cincinnati, which at that time was situated in the valley of the Ohio River. However, Drake pointed out that the shocks did not awaken residents with homes built on bedrock hills across the river in Kentucky.

It has been suggested that the Shelby County community of Anna, which suffered much damage from the earthquakes on March 2 and March 9, 1937, owes this severe shaking to the fact that the town lies above a 400-foot deep preglacial valley (Teays River) that was filled with sediment by glaciers of the Pleistocene Ice Age.

About two-thirds of Ohio is covered by unconsolidated sediments deposited by the Pleistocene glaciers. Some of these sediments consist of sand and gravel along river valleys and lake clays from post-glacial lakes and in the Lake Erie area, both of which are prone to amplification of seismic waves and, therefore, greater shaking. The Division of Geological Survey is mapping surficial sediments in Ohio and plans to use these data to produce maps that will depict areas that are more prone to shaking, if an earthquake of sufficient size would occur.

It is a myth that a great, multistate fault cuts across Ohio. However, Ohio is certainly not "faultless." However, very few faults are visible in surface rocks in Ohio. In large part, this is because a thick blanket of sediment deposited by glaciers of the Pleistocene Ice Age covers bedrock in about two-thirds of the state (all but southeastern Ohio) and thick vegetation obscures other exposures of bedrock. Most known faults in the state originate in crystalline Precambrian rocks that lie deep beneath the sedimentary bedrock. Many of these deep faults are suspected to be blind faults; that is, they do not reach the surface. These deep faults have been discovered primarily through deep drilling for oil and gas and remote-imaging techniques such as seismic reflection, and gravity and magnetic surveys. Some of these faults appear to be seismogenic (earthquake generating) whereas others seem to be inactive, at least during the short span of time that earthquakes have been recorded in Ohio. It is probable that there are many more faults in the basement rocks of Ohio than the few that have been mapped. The map of deep structures in Ohio shows the distribution of these known faults. One objective of the Ohio Seismic Network is to detect and precisely locate earthquakes in the state so that seismogenic faults can be identified.

Earthquakes in Ohio and adjacent areas are known as Intraplate Earthquakes; that is, they occur in the middle of a large crustal plate (North American plate). We know, of course, that the vast majority of earthquakes occur at plate boundaries, where two plates are crashing together or sliding past one another, as in the case of the San Andreas fault in California. However, intraplate earthquakes might be viewed as remnants of ancient plate boundaries where events of long ago have left persistent zones of weakness.

About a billion years ago, Ohio was a tectonically active area. An episode of crustal extension (pulling apart) created an area of down-dropped crust, known as a rift zone, in western Ohio. Crustal extension stopped before the crust was completely split, creating a failed rift. Numerous faults marked the boundary of the failed rift, now known as the Anna or Fort Wayne rift, and these faults form the zone of weakness where many earthquakes occur today.

After the rifting event, the eastern margin of the North American continent collided with a continental mass to the east, creating a chain of mountains that ran north-south from Canada to Tennessee and beyond. These mountains, known as the Grenville Mountains for exposures in Canada, were eventually worn down by erosion and buried beneath sediments deposited by continental seas during the Paleozoic Era. Many deeply buried faults were created during this mountain-building event.

The faults associated with the Grenville Mountains and the Anna rift are now ancient zones of weakness. As the North American continent is now pushed slowly westward as the Atlantic Ocean widens, stresses are created in these crustal rocks. The ancient fractures are zones of weakness along which stress is relieved by occasional earthquakes.