2002 Denver Annual Meeting (October 27-30, 2002)

Paper No. 7
Presentation Time: 3:20 PM


HOLBROOK, John M., Geosciences, Southeast Missouri State Univ, 1 University Plz, Cape Girardeau, MO 63701-4710, AUTIN, Whitney J., Earth Sciences, SUNY College at Brockport, Brockport, NY 14420, RITTENOUR, Tammy M., Dept of Geosciences, Univ of Nebraska, 214 Bessey Hall, Lincoln, NE 68588 and MARSHAK, Stephen, Dept. of Geology, Univ of Illinois, 1301 W. Green St, Urbana, IL 61801, jholbrook@semovm.semo.edu

The New Madrid seismic zone (NMSZ) hosted three large (M>7) intraplate earthquakes during the winter of 1811-12, and is currently the main focus of seismic hazard assessment in the U.S. Midcontinent. NMSZ seismicity concentrates along two blind NE-oriented right-lateral strike-slip faults linked by the co-slipping N-S-trending Reelfoot blind thrust. Published trenching and liquefaction studies reveal two prior co-seismic slip events on these faults (AD900 and AD1450). Though suspected, no evidence for discrete slip episodes prior to AD900 has been confirmed. We employed an alternative paleoseismic approach which assumes Mississippi River floodplain sediments over the Reelfoot fault record channel adjustments reflective of tectonic deformation and associated river gradient adjustments.

Reconstruction of Mississippi River channel courses using allostratigraphic techniques yields evidence for previously unrecognized millennial-scale fault-slip trends in the NMSZ, and extends the current record of deformation into the early Holocene. Two, and possibly three, Holocene channel straightening events are proxies for co-seismic fault-slip episodes occurring around AD900, 2000BC, and 5000BC, respectively. This result, added to previous studies, suggests ~1000-yr-duration episodes of relatively rapid slip in the NMSZ punctuate ~2000-yr-duration intervals of minimal deformation. If so, the current (post AD900) episode of rapid displacement in the NMSZ may be coming to a close.

Our data has implications for seismic-hazard assessment in the U.S. Midcontinent. Specifically, if Midcontinent fault zones, in general, exhibit a pattern of punctuated activity, then the current lack of seismicity on a given zone does not mean that the zone is risk free on the long term. In fact, stress release may oscillate regionally among several different fault zones in the Midcontinent on a time scale of several thousand years. Candidates for a currently interacting, and similarly punctuated, fault system are the southeast Reelfoot rift margin, the Commerce lineament, and the Wabash Valley fault zone. Each of these fault systems is currently placid, but has a record of NMSZ-scale deformation during the Holocene. The potential hazard of these currently quiescent faults may be underestimated.