Sometimes the convection that causes the plates to move will cause them to rip apart. If a continent happens to be sitting over the tear, then it too will be torn, or rifted, apart. Studies have revealed that such rifting apparently began about 600 million years ago beneath the present Mississippi River valley. Although the rift, called the Reelfoot Rift, failed to completely rupture the crust, the pulling apart of the rocks caused the formation of a zone of weakness in the Earth's crust (see pictures to the right). As the North American plate pushes against the Pacific plate, compressional forces are probably causing stress on these weakened rocks, which is occasionally relieved when the rocks slip past each other during an earthquake. Many earthquakes have occurred in the Reelfoot Rift in a region known as the New Madrid Seismic Zone. These earthquakes are shown in Figure 1 in the area around New Madrid, Missouri.
Indiana has several faults, but, unlike California's famous San Andreas Fault, nearly all of our faults are buried and can't be seen at the surface. Researchers have been able to map some faults in Indiana using evidence found in rocks from oil and gas wells and by employing a method called reflection seismic profiling. This method involves recording man-made vibrations reflected off layers of rock below the surface. Most of the faults that have been mapped in Indiana are located in the southwestern corner of the state. These faults extend into Illinois and are collectively known as the Wabash Valley Fault System. All of the mapped faults in Indiana are normal faults. Normal faults form when the crust is being pulled apart instead of being compressed as it is now. These mapped faults are unlikely candidates for future movement. The earthquakes that have occurred in Indiana during the last 200 years are the result of movement along faults at great depth (10 or more kilometers) below the surface. This depth and the nature of the rock layers at that depth have limited the ability of seismologists to successfully map earthquake-generating faults using reflection seismic profiling. Unfortunately, the best method available for mapping these faults is to wait patiently for the next large earthquake, then determine the precise location of the aftershocks using very sensitive seismometers (see figure 8 on the next page). Much more research is needed before we will know the full extent of faulting beneath Indiana and the potential for movement along those faults.