Why does near ridge extensional seismicity occur primarily in the Indian Ocean?

The possibility that thermoelastic stresses due to plate cooling contribute significantly to the stress field and seismicity in young oceanic lithosphere has been a subject of considerable recent interest. This effect is suggested by three key observations: a decrease in seismicity with lithospheric age, the fact that focal mechanisms show extension perpendicular to the spreading direction, and a depth stratification of mechanism types. A difficulty with this idea is that although thermoelastic stresses should be comparable in different regions, the intraplate seismicity seems to occur in local concentrations. In particular, the ridge-parallel extensional seismicity occurs preferentially in the Central Indian Ocean region. We explore the possibility that much of the data favoring thermoelastic stresses can be interpreted in terms of stresses resulting from individual plate geometry and local boundary effects. In particular, the dramatic concentration of extensional seismicity in the Central Indian Ocean region is consistent with finite element results for the intraplate stress incorporating the effects of the Himalayan collision and the various subduction zones. The ridge parallel extensional stresses show a decrease with age similar to that of the seismicity. As earthquakes in this area provide a major portion of the data for both ridge-parallel extension and depth stratification, these effects may be due more to the regional stress. We thus propose that thermoelastic stresses provide a low level "background" in all plates, but that the dominant effect is that of individual plate geometry and local boundary effects.