The 1886 Charleston, South Carolina earthquake and the Appalachian Detachment

Abstract

Several lines of evidence from new and existing data are consistent with an extended fault slip associated with the 1886 earthquake in South Carolina. Back slip over a portion of the Appalachian detachment approximately corresponding to the Coastal Plain of South Carolina fits some of the data including coseismic and post seismic strain indicators. A small seismic source for the August 31, 1886, main shock cannot be ruled out, and a large, but only partially seismic fault slip is possible. During this event intensities of MM ≥ VIII occurred over a wide area of the Coastal Plain and are not limited to the Charleston‐Summerville area, where the traditional small source for this earthquake is located. Near‐source effects such as ground fissuring, liquefaction, changes in line‐of‐sight, foreshocks, and aftershocks occurred over a wide area of the Coastal Plain and neighboring Inner Piedmont, which we tentatively refer to as the meizoseismal area. This area is sharply bounded by the Savannah River toward the southwest and approximately by the North Carolina‐South Carolina state border toward the northeast. These boundaries are not associated with any known structure transverse to the strike of the Appalachians. During the 14 years prior to 1886 seismicity was concentrated around the extended meizoseismal area while the seismicity within that area was unusually low, suggesting a premonitory doughnutlike pattern.The intensity distribution in the far and near field of the 1886 event and of the great Indian detachment earthquakes, e.g., the 1905 M = 8 and the 1934 M= 8.3 events, are qualitatively and quantitatively similar. This is consistent with comparable source dimensions, since well‐known intermediate magnitude earthquakes in India and in eastern North America exhibit similar intensity patterns and indicate similar seismic attenuation in the two cratonic environments. Several meters of coseismic (1886) northwest compressive strain near Summerville are indicated by buckling of rails. A large postseismic normal displacement is suggested by releveling data (1935–1968) across the Brevard fault zone updip (northwest) of the presumed 1886 detachment slip. These strains are consistent with a southeastward slip of the detached crystalline sheet associated with the 1886 earthquake. Available data indicate that in the southern Appalachians the updip transition from maximum horizontal to least horizontal compressive stress in the dip direction (perpendicular to structural trends)occurs somewhere in the Piedmont and may coincide with the steepest portion of the detachment. The stress field may also contain a time‐dependent component. In the sediments of the Coastal Plain, orthogonal sets of subvertical clastic dikes oriented along and across the Appalachian strike direction suggest one or more interchange of the principal stress axes. We infer from coseismic dry fissuring and liquefaction that clastic dikes filled from above and from below were also generated during the 1886 event. Prehistoric clastic dikes along with other structures in the sediments may have been generated during prehistoric earthquakes similar to the 1886 event. Stress reversals corresponding to earthquake cycles may characterize the stress field within detached sheets or wedges that move primarily by great detachment earthquake ruptures.Appendices are available with entire article on microfiche. Order from American Geophysical Union, 2000 Florida Avenue, N.W., Washington, D.C. 20009. Document J81‐008; $1.00. Payment must accompany order.