Intraplate seismicity, reactivation of preexisting zones of weakness, alkaline magmatism, and other tectonism postdating continental fragmentation

Abstract

The distribution of intraplate earthquakes and of igneous rocks postdating continental rifting is summarized and placed into a plate tectonic framework for the following continental areas: eastern and central North America, Africa, Australia, Brazil, Greenland, Antarctica, Norway, Spitsbergen, India, and the margins of the Red Sea and Gulf of Aden. In continents, intraplate earthquakes tend to be concentrated along preexisting zones of weakness within areas affected by the youngest major orogenesis that predates the opening of the present oceans. Many preexisting zones of weakness (including fault zones, suture zones, failed rifts, and other tectonic boundaries), particularly those near continental margins, were reactivated during the early stages of continental separation. In contrast, intraplate shocks rarely occur within the older oceanic lithosphere or within the interiors of ancient cratonic blocks of the continents. In several continental areas, rocks and tectonic features postdating the opening of present‐day oceans, including carbonatites, kimberlites, other alkalic rocks, mafic dikes, and ring dikes, as well as some of the largest intraplate shocks, seem to be located preferentially along old zones of weakness near the ends of major oceanic transform faults that were active in the early opening of adjacent oceans. In several places, alkaline magmatism and earthquakes extend several hundred kilometers inland from the ends of oceanic transform faults (but not necessarily with the same strike as the transform fault). Major preexisting zones of weakness that are oriented subparallel to the directions of relative continental separation appear to control the locations of transform faults that develop in a new ocean. In some instances, alkaline magmatism persisted along reactivated features of this type for as long as 100 m.y. after the initial stages of continental fragmentation. Most kimberlites in South Africa seem to have been emplaced along preexisting zones of weakness that were reactivated during the early opening of the South Atlantic. The type of intraplate magmatism appears to be related to the thickness of the lithosphere. Unlike oceanic transform faults where large horizontal movements have occurred, reactivated zones of weakness in continents usually appear to have been the sites of only relatively small displacement. Seismic activity and alkaline magmatism may be controlled by deep fractures that penetrated the entire lithosphere to tap asthenospheric sources of magma. Seismic activity along these zones seems to occur in response to the present‐day stress regime, which is not necessarily the same as that which was active during the emplacement of the alkaline rocks. Other intraplate shocks are concentrated along old zones of weakness that are subparallel to continental margins. Such shocks are found in the Appalachians, northeastern and northern Greenland, Norway, Great Britain, Spitsbergen, northern Canada, and Australia. These zones of weakness were also reactivated during continental separation in either the Mesozoic or the Cenozoic. Evidence is now mounting for Cretaceous and Cenozoic deformation along some of these features. Although not many focal mechanism solutions or in situ measurements of stress are available for intraplate areas, horizontal compressive stresses appear to be present today in many of the pre‐Mesozoic orogenic belts that were reactivated by continental rifting. This evidence, as well as examples of Cenozoic thrust faulting, indicates that the stress field has changed since rifting commenced. High compressive stresses, the absence of earthquakes in Antarctica, their near absence along the margins of the Gulf of Mexico, and the much lower levels of activity in the oceanic lithosphere adjacent to most continents argue against mere sedimentary loading and the cooling of the oceanic lithosphere as the main source of stress that is reactivating faults of these older fold belts. The large compressive stresses and the uplift found in many continental areas adjacent to continental margins may be caused by a deep‐seated source in the mantle of long wavelength or by stresses transmitted in the lithosphere. These effects may be related to either the cooling and underplating of the continental lithosphere adjacent to continental margins, large tractions on the base of the lithosphere in shield areas, stress concentrations related to marked changes in the age and thickness of the lithosphere, convective motions of the mantle beneath these areas, or those regions acting like broad zones of weakness that are being compressed between adjacent areas of greater strength. During the fragmentation of a supercontinent, multibranched rifting usually follows the youngest zone of previous orogenesis and as much as possible avoids passing through old cratonic areas where the lithosphere is thick, cold, and strong. Rift junctions seem to be related to the preexisting mosaic of cratons and younger belts of deformation rather than to a motive force involving mantle plumes. Likewise, many zones of unusually high magmatic activity, i.e., hot spots, appear to be related to nodes or junctions in this mosaic pattern. Thus these hot spots appear to be passive features rather than the surficial expression of mantle plumes. Major transform faults that are active during the early opening of an ocean also tend to develop where the margins of the older cratons undergo an abrupt change in strike. During the early development of an ocean the preexisting mosaic of structural elements within the thick continental lithosphere may result in large normal forces across some plate margins, leaky transform faulting, and localized stress concentrations. The early directions of sea floor spreading and of transforming faulting may be altered by these boundary forces and by the geometrical constraints imposed in separating old cratonic blocks. These constraints are relaxed once old, thick lithosphere is no longer in contact across long transform faults. Since these early directions are strongly influenced by the preexisting tectonic framework and may not coincide with the direction of the forces driving the plates apart, early transform faults may have components of extension (or compression) along them in addition to strike slip motion. A small component of extension may be responsible for the formation of volcanic ridges and seamount chains such as the Walvis ridge, Rio Grande rise, and New England seamount chain. These features predate the marked change in the strike of transform faulting that occurred in the North and South Atlantic about 80 m.y. ago as thin oceanic lithosphere finally came in contact across large oceanic transform faults. Several zones of intraplate magmatism in the surrounding continents also ceased at that time.