Fault plane solutions and tectonics of the South Atlantic and Scotia Sea

Abstract

Focal mechanisms are presented for 46 earthquakes that occurred in the South Atlantic Ocean, in the Scotia Sea, and in southern Chile during the period 1963–1973. The slip vectors of shallow earthquakes indicate that the South American plate is moving directly west with respect to the Antarctic plate at the ridge-fault-fault triple junction in the South Atlantic. The directions of motion of Africa with respect to the South American (SA) and Antarctic (ANT) plates at the triple junction are N70°E and N47°E, respectively. The SA-ANT relative motion between the triple junction and the South Sandwich trench is best described by a pole of rotation at 80°S, 166°W, with an angular rotation rate of 0.24 deg/m.y. Shallow earthquakes along the South Sandwich trench indicate that the oceanic portion of the South American plate is being thrust under the South Sandwich arc in an east-west direction. Most of the earthquakes at the northern end of the arc are due to hinge faulting or bending stresses within the underthrust oceanic plate. The focal mechanisms of intermediate depth events beneath the arc indicate down-dip extension in the northern end of the downgoing slab and down-dip compression in the southern end. This change in the stress pattern may be caused by reduced negative buoyancy forces in the younger, southern half of the subducted plate. The seismicity and focal mechanisms suggest that the SA-ANT relative motion in the Scotia Sea region is taken up on both the north and the south Scotia ridges. However, the plate boundaries in this area and in southern Chile are not well defined. There does appear to be a consistent pattern of horizontal compressive stress directed ENE-WSW throughout the Scotia Sea region, probably induced by the convergence of the South American and Antarctic plates. The SA-ANT relative motion observed in this study is not consistent with the motion predicted from the summation of motions observed on other plate boundaries. This discrepancy may be due to (1) systematic errors in the data, (2) a recent change in plate motions, or (3) minor nonrigid plate behavior. The third explanation is preferred because internal deformation of the plates at very slow strain rates can explain other examples of seemingly inconsistent plate motions and may also account for the existence of diffuse intraplate seismicity. The apparent relative drift of hot spots may also be due to internal deformation of the plates after the seamount chains are created.