POSSIBLE SURFACE DEFORMATION ALONG A LINEAMENT IN THE NEW MADRID SEISMIC ZONE, YARBRO, ARKANSAS

In northeastern Arkansas, an anomalous northeast-oriented, topographic lineament may be the surface expression of the axial fault of the New Madrid seismic zone. As seen on historical aerial photographs and recently acquired LiDAR imagery, the lineament is very straight, extends 7.0 km across the landscape, crossing Holocene point-bar deposits of several meander loops of the Pemiscot Bayou as well as Holocene floodplain deposits of the Mississippi River, and appears to make several en echelon steps; all characteristics suggestive of surface faulting. Two trenches excavated across the western side of the lineament revealed both a large, linear graben and a large linear sand blow with multiple sand units. The graben is 6.5 m wide and 1.0 m deep, and is filled by an organic-rich clay plug with many fractures along which root casts have formed and sand dikes have intruded. The liquefaction features include a main feeder dike, several subsidiary dikes, and an overlying and related compound sand blow composed of five depositional units. The ~1.8 m wide main feeder dike crosscut a silty loam paleosol in which sand-tempered potsherds of the Woodland cultural period (600 – 900 A.D.) occur. The very long and linear ridge can most easily be explained by the formation of a flower structure above a strike-slip fault. The linear graben may have formed due to extension in the dilating mass above the fault and the linear sand blow may have formed as the result of venting of water and entrained sediment along a fault surface. If so, the two features both reveal and mask the presence of an active fault. An alternative interpretation is that the linear graben and sand blow and dike may be related to lateral spreading related to earthquake-induced liquefaction. We currently prefer the fault interpretation because the ridge is exceptionally long and straight, continues across different landscape elements, and makes several en echelon steps. If the features were related to lateral spreading, they more likely would be curvilinear in form and more limited in extent. Additional investigations of the ridge, including trenching and geophysical imaging below the depth of trenching is needed to evaluate the relationship of the graben and sand blows to one another and to subsurface faults.