ESTIMATION OF THE
EFFECT OF THE MISSISSIPPI EMBAYMENT SEDIMENTS ON MAGNITUDE DETERMINATION USING
SYNTHETIC SEISMOGRAMS
PUJOL, J.,
Dept. of Earth Sciences, The Univ. of Memphis, Memphis, TN 38152,
jpujol@memphis.edu; and PEZESHK, S., Dept. of Civil Engineering, The Univ. of
Memphis, Memphis, TN 38152, jpujol@memphis.edu,
spezeshk@memphis.edu.
The magnitudes of the largest 1811-1812 New Madrid earthquakes are not well known, with estimated values ranging between the low 7's and 8.1. The current national hazard maps are consistent with a M=7.7 earthquake and predict higher hazard than in parts of California, which has generated a lively debate. Uncertainty in the determination of the magnitudes originates in a poor understanding of the effect of unconsolidated sediments on intensities, from which magnitudes are derived. As low velocity sediments amplify the ground motion (relative to hard-rock sites), their effect on people and structures are enhanced, thus increasing the estimated values of intensities and magnitudes. To address this issue we used synthetic seismograms generated for layered media with and without a low Vp (2 km/s) layer with Vp/Vs=2, and Q=30,50. This layer is underlain by high-velocity rocks (Vp=6 km/s) Event depths and epicentral distances are in the 5-20 km and 10-300 km ranges. The largest amplitudes of the synthetic seismograms were used to determine "local magnitudes" (Langston et al., BSSA, 1998, Pujol, BSSA, 2003). The effect of the sediments is to increase the computed magnitudes by up to 0.6 units when a Q=50 is used, which is consistent with results derived by other researchers. The sediments also increase the duration of the events markedly, which is significant because damage to buildings and the possibility of liquefaction increase with the number of stress cycles (proportional to duration). Therefore, increased durations result in larger estimated intensities and magnitudes. The national hazard maps do not account for site effects and the predicted hazards may be even higher when they are accounted for, but as our results show, site effects may have been taken into account in the national maps already.