SUMMARY The sizes of three major or great historical earthquakes are reassessed using the isoseismal‐area‐regression tools developed in Parts I and II of this study of stable continental region (SCR) seismicity. The earthquakes are 1811 New Madrid, central United States, and its following sequence; 1886 Charleston, coastal South Carolina; and 1755 Lisbon, oceanic intraplate off the continental shelf of Portugal. The analysis confirms the large size of these events and for the first time places constraints on the uncertainty of their seismic moment release. Because of the exceptionally low seismic‐wave attenuation of eastern North America (ENA), a separate North American regression of seismic moment on isoseismal area was developed. Additionally, the unknown western extents of the New Madrid isoseismal areas were calibrated with the patterns of the M 6.3‐6.6 1843 and 1895 earthquakes. Application of Part II analysis procedures with these corrections yields New Madrid size estimates, expressed as moment magnitude, of M 8.1±0.31 for the 1811 December 16, M 7.8±0.33 for the 1812 January 23, and M 8.0±0.33 for the 1812 February 7 principal events. The Charleston earthquake's magnitude decreases from M≳7.4 to M 7.3±0.26 after compensation for the effect of coastal plain sediments on its inner isoseismals. Intensity regressions for Lisbon are calibrated against the isoseismal pattern of the nearly co‐located M 7.8 1969 St Vincent earthquake, which in this case increases the predicted size of Lisbon from M 8.4 to M 8.7±0.39. These size estimates are supported by data from independent phenomena: extent and severity of liquefaction, the maximum distance of induced landslides, and for Lisbon, tsunami wave amplitudes. Estimated source parameters are controlled by crustal or lithospheric temperature, which governs the depth extent of brittle faulting. Using estimated continental and oceanic geotherms, viable fault lengths are 30–80 km for Charleston, 120–180 km for 1811 New Madrid, and 180–280 km for Lisbon for average displacements of 2–4 m, 8–11 m, and 10–14 m, respectively, and for average static stress and strain drops. At the estimated seismic moments of this study, the 1811 New Madrid and the 1755 Lisbon events are, respectively, the largest known SCR and oceanic lithosphere earthquakes.

Their described impacts on the land and the river were so dramatic as to produce widespread modem disbelief. However, geological, geophysical, and historical research, carried out mostly in the past two decades, has verified much in the historical accounts. The sequence included at least six (possibly nine) events of estimated moment magnitude M. 7 and two of M ≅ 8. The faulting was in the intruded crust of a failed intracontinental rift, beneath the saturated alluvium of the river valley, and its violent shaking resulted in massive and extensive liquefaction. The largest earthquakes ruptured at least six (and possibly more than seven) intersecting fault segments, one of which broke the surface as a thrust fault that disrupted the bed of the Mississippi River in at least 2 (and possibly four) places. ... it is a riddle wrapped in a mystery inside an enigma.

No abstract

S U M M A R YSeismic intensity observations contain sufficient information about the earthquake source to quantitatively constrain its scalar seismic moment, M,, and hence moment magnitude, M, within useful limits. This is valuable, especially in the pre-instrumental and early instrumental seismic eras, but also in the modern era. This study is limited to stable continental regions where intensity data are especially important for seismic hazard assessment, but the methodology is generic and can be applied to other tectonically active iegions. It builds on regression techniques developed in the Part I analysis (Johnston 1996) and applies them to isoseismal area (Ai) data. Derived regression relations for modified Mercalli isoseismal areas for levels felt to VIII yieid a predicted log(M,) or M value within specified la uncertainty bounds. Standard linear and polynomial regressions are tested against a new functional regression form proposed by Frankel ( 1994), which contains both geometrical spreading and anelastic attenuation terms. Goodness-of-fit statistics are similar, but the Frankel regression form is preferred because it is derived from physical principles of wave propagation.The Afel,-AIV regressions, with most data at epicentral distance r > 100 km, are controlled by surface-wave (L,) geometrical spreading and attenuation characteristics. For the A,,, and A,,,, regressions, r is mainly less than 100 km and body-wave propagation dominates, although near-source site and path effects are significant. The A, and A,, regressions are transitional between the L, and body-wave domains. With either the Frankel or quadratic regression, individual isoseismal areas can constrain the source event's moment magnitude within an estimated &0.30-0.45 M units except at the regression extremes. If a suite of isoseismal areas is available for the same earthquake, these uncertainties can be approximately halved by weighted averaging. This means that when good-quality isoseismal data exist for an event, they are capable of constraining its seismic moment estimate within the same order of uncertainty as individual instrumental magnitude readings. Hence, historical seismicity data may usefully be combined with instrumental data in seismic hazard analyses. A hierarchy of methods to recover an earthquake's M , or M combines the instrumental results of Part I with the isoseismal area results of this study. Finally, regressions on I, , , and number of recording stations provide the means to estimate M to within --+0.45-0.55 M units when no other data are available.

More than 700 earthquakes have been located in the central New Madrid seismic zone during a two-year deployment of the PANDA array. Magnitudes range from < 0.0 to the mblg 4.6 Risco, Missouri earthquake of 4 May 1991. The entire data set is digital, three-component and on-scale. These data were inverted to obtain a new shallow crustal velocity model of the upper Mississippi embayment for both P- and S-waves. Initially, inversion convergence was hindered by extreme velocity contrasts between the soft, low-velocity surficial alluvial sediments and the underlying Paleozoic carbonate and clastic high-velocity rock. However, constraints from extensive well log data for the embayment, secondary phases (Sp and Ps), and abundant, high-quality shear-wave data have yielded a relatively robust inversion. This in turn has led to a hypocentral data set of unprecedented quality for the central New Madrid seismic zone. Contrary to previous studies that utilized more restricted data, the PANDA data clearly delineate planar concentrations of hypocenters that compel an interpretation as active faults. Our results corroborate the vertical (strike-slip) faulting of the the southwest (axial), north-northeast, and western arms and define two new dipping planes in the central segment. The seismicity of the left-step zone between the NE-trending vertical segments is concentrated about a plane that dips at ∼31°SW; a separate zone to the SE of the axial zone defines a plane that dips at ∼48°SW. The reason for this difference in dip, possibly defining segmentation of an active fault, is not dear. When these planes are projected up dip, they intersect the surface along the eastern boundary of the Lake County uplift (LCU) and the western portion of Reelfoot Lake. If these SW-dipping planes are thrust faults, then the LCU would be on the upthrown hanging wall and Reelfoot Lake on the downthrown footwall. If in turn these inferred thrust faults were involved in the 1811–12 and/or pre-1811 large earthquakes, they provide an internally consistent explanation for (1) the existence and location of the LCU, (2) the wide-to-the-north, narrow-to-the-south shape of the LCU, and (3) the subsidence and/or impoundment of Reelfoot Lake.