Capitalizing on the success of a pilot study conducted in 2008, which tested the feasibility of marine seismic data acquisition along the Mississippi River, University of Memphis and University of Texas researchers are planning two field programs to acquire approximately 600 km of high-resolution, marine seismic reflection and sub-bottom profile data along the Mississippi River. The two campaigns will image with unprecedented resolution the New Madrid Seismic Zone active fault system and a series of inferred faults suspected to be responsible for earthquake induced liquefaction features predating the seismic activity of the New Madrid Seismic Zone. Exploiting the advantages of marine acquisition (time effective, low cost), the new data will identify the location and timing of deformation within the active fault system (which generates more than 200 magnitude 1.5-4.0 earthquakes a year), and outside the New Madrid Seismic Zone, and contribute to the understanding of how long-term deformation is partitioned in time and space among faults within the Mississippi Embayment. This information is critical to solve the apparent paradox raised by the contrasting evidence of small geodetic vectors, a puzzling lack of substantial deformation, and the high level of instrumental and historical seismicity.
One of the pillars of plate tectonics is the assumption of the rigidity of major plates. Indeed GPS studies show that most plate interiors behave rigidly and, as a result, the majority of earthquakes occur along narrow or diffuse belts at plate boundaries. Located in the heart of the North American continent and 2,000 km away from the nearest plate boundary, the New Madrid Seismic Zone is one of the most active intraplate seismic areas in the world, and one of the most notorious deviations from plate rigidity. Paleoseismological observations show that the area experienced a series of catastrophic earthquakes in historic and prehistoric time and evidence is mounting that the long-term seismicity is not limited to the presently active fault system, but that large earthquakes have occurred in other areas within the Mississippi Embayment, for which the responsible fault (or system of faults) are yet to be identified. This study will provide observational constraints on theoretical models proposed to explain the timing, magnitude and location of intraplate seismicity in the Central United States. The results of this project will also have a crucial impact for earthquake hazard assessments by localizing and characterizing seismogenic faults, and improving the quality and usefulness of seismic hazard maps.
Most earthquakes occur at the edges of tectonic plates as they move toward, away, or past each other. The New Madrid Seismic Zone (NMSZ) is an exception to that rule because it is a persistent zone of small earthquakes that occur in the "middle" of the North American plate. We care about this zone because earthquakes here have not always been small. In fact, in the winter of 1811-1812, there was a series of three large earthquakes in this area, now the southeast corner of Missouri. Each of these "events" is estimated to have had a magnitude between 6.8 and 8.0, capable of causing catastrophic damage to urban areas. At the time there were no large cities anywhere close, but now cities such as Memphis, Tennessee could be at risk. One of these earthquakes caused a fault to rupture and lift the ground surface; the uplift was enough to cause the Mississippi River to flow upstream for many hours. For the last 200 years there have been no large earthquakes in the NMSZ, but the persistent small earthquakes remind us that larger events are possible. Our project was developed to acquire seismic reflection images across the NMSZ and other parts of the Mississippi embayment. The seismic reflection technique produces sound waves that travel downward into the Earth and bounce off of rock layers below the surface and get recorded back at the surface like an echo. With a series of these echoes we can produce an image of the Earth’s layers below the surface, similar to a cross section or a road-cut where a highway has cut through a hill or ridge. Our goal was to see where the nearly flat-lying rock layers are offset or shifted—this is a very good sign of a fault. Furthermore, we were looking for offsets in rock layers deep in the Earth but also in layers just below the surface. Rock layers form when new material (soil, sand, mud) is piled on top, so the oldest rock is on the bottom of the pile and the newest is on top. With offsets both shallow and deep, we can expect that the fault has moved recently (cutting the new material) and may still be active. We successfully acquired seismic data through the NMSZ and created good subsurface images of the rock layers. As expected, we found offsets in the rock layers and were able to identify where these earthquake faults come close to the surface. While it was an important goal to confirm the location of these faults, it was also very important just to identify what active faults would look like in this area. With this information were able to investigate another question: Is the NMSZ the only part of the Mississippi embayment with active, or recently faulting? If so then it could be the only source of significant seismic hazard; however, if recent faulting is apparent throughout the embayment, then there may be unrecognized hazard in other areas. In answer to the question above, as we imaged more of the embayment, we found several areas where offset rock layers indicated faults with an ongoing history of movement, including the very shallow (young) layers. For example, there was a fault that had previously been identified in one location north of Memphis, Tennessee. Our new data show that this fault extends from this location at least 28 miles along the river to well south of Memphis. By confirming the length of this fault and its recent movements we have identified a potentially important seismic hazard to Memphis. Farther south along the river we have identified several more zones of recent deformation. One area is located at the confluence of the Mississippi and Arkansas rivers and is likely to mark the location of the Arkansas River fault zone. This fault had been suggested by surface features, but our project was the first data to show images of this fault zone in the subsurface. This is in an area where there are other indications of seismic shaking, so this may help focus the seismic hazard assessment on the fault we identified. Finally, we unexpectedly located a significant fault near the northeast corner of Louisiana. Here, we can see that rock layers over one kilometer to less than 100 meters deep are offset or strongly tilted by movement on a fault. In summary, this project has produced useful images in the active New Madrid Seismic Zone, but it has also located several other zones that currently produce few earthquakes but show that significant deformation has occurred in the recent past. Most importantly, this project indicates that at least three fault zones outside of the NMSZ are potentially active and should be further investigated to see if they pose a seismic hazard to local and regional inhabitants.
