The New Madrid Fault system is leading researchers to question the relevance of traditional plate-boundary hazard models for intraplate earthquake zones like New Madrid, writes Sarah Day
Geoscientist Online 16 March 2009
Rather than maintaining a regular pattern of movement and earthquakes, the research shows that the fault system is moving at a rate of less than 0.2 millimetres per year, and there is likely no movement. This suggests the fault may be closing down.
‘The slower the ground moves, the longer it takes until the next earthquake, and if it stops moving, the fault could be shutting down’, explains Seth Stein, co-author of the paper. ‘We can’t tell whether the recent cluster of earthquakes in the New Madrid is coming to an end. But the longer the GPS data keeps showing no motion, the more likely it seems’.
The New Madrid fault system is a major seismic zone in the Southern and Midwestern United States. It was responsible for a series of earthquakes in 1811/12, one of the largest succession of earthquakes in the US, and has previously been considered highly likely to cause similar quakes in the near future. (See Geoscientist 15.7, July 2005 pp4-7)
The team of researchers from Purdue and Northwestern Universities used GPS measurements to analyse the fault motion over eight years. Given the 500 to 1000-year repeat cycle for major earthquakes on the fault, they found that there was much less movement than models have predicted.
The research has wider implications for the modelling of earthquake hazards on intra-continental fault lines, which may not follow the pattern seen at plate boundaries.
‘Our findings suggest the steady-state model of quasi-cyclical earthquakes that works well for faults at the boundaries of tectonic plates, such as the San Andreas fault, does not apply to the New Madrid fault’, says Eric Calais, the lead researcher from Purdue. ‘At plate boundaries, faults move at a rate that is consistent with the rate of earthquakes so that past events are a reliable guide to the future. In continents, this does not work. The past is not necessarily a key to the future, which makes estimating earthquake hazard particularly difficult’.
Calais and Stein are investigating the possible explanation s for the behaviour of faults like the New Madrid. One suggestion is that earthquakes in such areas occur in clusters, before migrating to a nearby fault.
‘There is the possibility that seismicity migrates with time as earthquakes trigger earthquakes on nearby faults’ Calais says. ‘Geologists studying the seismic history of faults have found that there have been several faults in the central and eastern US and that they seem to produce bursts of earthquakes and then turn off’.
Being able to estimate an accurate earthquake threat is crucial for the area, which includes parts of Illinois, Indiana, Tennessee, Arkansas and Kentucky. For Calais, the study suggests that a new approach to studying earthquakes may be needed: ‘We need to develop a new paradigm for how earthquakes happen at faults that are inside continents’.
- Calais et al. Time-Variable Deformation in the New Madrid Seismic Zone. Science, 2009; 323 (5920)