Indian Ocean Actively Deforms

See allHide authors and affiliations

Science  08 Jun 2001:
Vol. 292, Issue 5523, pp. 1850-1851
DOI: 10.1126/science.1061082

According to plate tectonics, plates are rigid and deform only at their boundaries. There is, however, ample evidence of intraplate deformation in the Equatorial Indian Ocean. The deformation results from exceptionally high stresses in the oceanic lithosphere caused by the ongoing collision of India with Eurasia. Such intraplate deformation is commonly observed on the continents but not in the oceans, where deformation is usually localized at a narrow plate boundary rather than distributed over a wide area. The Indian Ocean thus offers a rare opportunity to study intraplate deformation of oceanic lithosphere.

Deformation in the Indian Ocean was first proposed in the 1970s (1) to resolve the difficulties in fitting global plate motion using a single rigid Indo-Australian plate. This traditionally defined plate is now considered as a composite plate, which includes three rigid or nearly rigid component plates and several deforming zones (2). The deforming zones may be described as diffuse plate boundaries, thus relaxing the fundamental assumption of plate tectonics that all oceanic plate boundaries are narrow (2, 3). In the following, the term “intraplate” refers to earthquakes or deformation occurring within these large diffuse boundaries, with reference to the “classic” plate boundaries of the Indo-Australian plate.

The Equatorial Indian Ocean is known for its intraplate seismic activity and long-wavelength undulations in satellite-derived gravity data. Many earthquakes, with magnitude as large as 6 or 7, have occurred here during the last century (see the figure) (4). On 18 June 2000, an earthquake of magnitude 7.8 was registered in the Wharton Basin south of Cocos island (5). Earthquakes in the Central Indian Basin generally follow mechanisms different from those in the Wharton Basin. Their mechanisms indicate that the main compressive stress rotates from north-south in the Central Indian Basin to northwest-southeast in the Wharton Basin (4). In addition, the long-wavelength gravity undulations strike east-west in the Central Indian Basin and northeast-southwest in the Wharton Basin, in both cases roughly perpendicular to the compression axis. Numerical modeling of the Indo-Australian plate stress field (6, 7) suggests that the rotation of the main compressive stress can be largely explained by the change in boundary conditions north of the Indo-Australian plate. The northward motion of India is resisted by the India-Asia collision, whereas the Wharton Basin freely subducts under the Java-Sumatra trench.

A large area of intraplate deformation.

The compression axis in the Equatorial Indian Ocean (orange arrows) rotates from north-south in the Central Indian Basin to northwest-southeast in the Wharton Basin, yielding a different pattern of deformation in the two basins. Many large earthquakes have occurred during the last century [blue, Harvard CMT solutions (13); purple, from (4); pink, 18 June 2000 earthquake (5)]. In both basins, large-scale deformation may occur through buckling/folding perpendicular to the compression axis (long wavelength gravity undulations in light red). Brittle failure seems to occur along preexisting weakness directions (black), namely the north-south fracture zones and the east-west abyssal hill fabric.

The stress directions can thus be predicted reasonably well. The strain pattern is more difficult to assess, however, and requires field data from marine surveys or detailed source mechanism of recent earthquakes.

Most studies have focused on the Central Indian Basin. Here, tectonic deformation is characterized by long-wavelength (100 to 300 km) undulations of the oceanic basement—associated with the gravity undulations—and superimposed small-scale reverse faulting and folding of the crust and overlying sediments (8, 9). We do not yet have enough data to determine the definitive cause of the long-wavelength undulations, but analog and numerical modeling suggests that it is caused by folding and buckling of the lithosphere (9, 10).

Until recently, little was known about the deformation pattern in the Wharton Basin. The satellite-derived gravity undulations strike northeast-southwest and have wavelengths and amplitudes similar to those in the Central Indian Basin. They may again indicate folding and buckling of the lithosphere to accommodate northwest-southeast shortening. It remains unclear, however, whether the undulations are again associated with reverse faulting in the crust.

Some Wharton Basin earthquakes exhibit thrust mechanisms (indicative of reverse faulting), but most are strike slip. A marine survey in the northern part of the basin has provided direct evidence of active strike-slip faulting (12) at several north-south active faults. These faults have a present-day left-lateral movement and are reactivated fracture zones of a fossil spreading center. According to the seismicity farther north and south, the faults must be at least 1000 km long, reaching the Sumatra trench to the north.

The large magnitude of the 18 June 2000 Wharton Basin earthquake and the availability of high-quality digital data with good station coverage allowed Robinson et al. (5) to model the details of the source. The earthquake turns out to have an unusual mechanism: Two subevents simultaneously ruptured a nearly north-south plane and a plane nearly conjugate to the first. Rupture along the north-south plane was similar to the movement along the surveyed strike-slip faults 800 km farther northwest. Therefore, all the fracture zones in the northern Wharton Basin are probably reactivated by strike-slip faulting between the Nynety East ridge and the Investigator ridge (see the figure). The northern Wharton Basin thus appears to be cut into north-south slivers that subduct more and more easily the further east one goes (12). Rupture along the east-west plane introduces some northwest-southeast compressional deformation (5).

The east-west plane is consistent with the orientation of the abyssal hills of the oceanic lithosphere. The lithosphere thus deforms along preexisting weakness directions: the north-south fracture zones and the east-west abyssal hill fabric, both of which originate at the mid-ocean spreading centers. Note that in the Central Indian Basin, most of the reverse faults result from the reactivation of the abyssal hill fabric (8). In both basins, the lithosphere may deform at large scale by buckling and folding perpendicular to the compression axis, but brittle failure of its upper part occurs along preexisting weakness directions (see the figure).

Some questions are still open. Is rupture along north-south and east-west directions specific to this earthquake or not? The June 18, 2000 earthquake is located in a broad area covered by numerous ridges and seamounts. Is there an influence of volcanism in the deformation process in this area? And how can southwest-northeast folding of the Wharton Basin lithosphere be compatible with brittle failure along north-south and east-west directions? To resolve these questions, we need to better understand the role of preexisting features in the mechanical response of an oceanic lithosphere when it is subjected to high compressive stress.

References and Notes

Navigate This Article