LIQUEFACTION EVIDENCE FOR LARGE MAGNITUDES OF THE 1811-1812 NEW MADRID EARTHQUAKES

The magnitudes of the 1811-1812 New Madrid, MO, earthquakes are an important consideration for seismic hazard assessment in the central and eastern United States. Published estimates of the moment magnitudes of the three main shocks of the 1811-1812 earthquakes range from approximately M6.7 to 8.1. These estimates are primarily based on intensities inferred from written historical accounts. In this study we used observations of liquefaction in the epicentral region to conclude that magnitudes of the 1811-1812 main shocks were greater than or equal to approximately M7.5. Three independent lines of argument support our conclusion. First, numerous eyewitness accounts describe liquefaction effects in the epicentral regions that typically are associated with both long shaking duration and very strong shaking, which are characteristic of earthquakes of large (M>7.5) magnitude. These effects include both large lateral spreads with tens of feet of displacement and large settlements associated with post-liquefaction consolidation. Second, modern mapping of sand boils, lateral spreads, and ground settlement reveals that liquefaction affected an area of more than 11,380 km². Such extensive liquefaction is observed only in larger magnitude earthquakes. We also prepared maps of liquefaction probability, based on cone penetration testing conducted in susceptible surficial geologic units in the Mississippi River valley, and compared these maps to the observed areal pattern of the liquefaction intensity in 1811-1812. Liquefaction probability maps for earthquakes with assumed magnitudes of M7.5 compare more favorably to the observed liquefaction than do the maps based on assumed magnitudes of seven or less. And third, the maximum distance of 1811-1812 liquefaction from the assumed seismic sources suggests larger magnitudes. These maximum distance curves are less reliable for older and incompletely documented earthquakes where isolated and more distant liquefaction may have gone undetected. Thus, we also considered only paleoseismology sites mapped by Tuttle et al. (2002), where they observed multiple liquefaction events that they assumed came from all three of the main shocks. Collective consideration of these sites suggests the magnitudes of the main shocks were approximately M7.5 or greater.