Kinematic constraints on distributed lithospheric deformation in the equatorial Indian Ocean from present motion between the Australian and Indian Plates

Publisher Website
Google Scholar
Abstract
From an expanded, accurate, and up‐to‐date data set comprising 110 spreading rates, 46 transform azimuths, and 151 earthquake slip vectors from the Indian Ocean and Gulf of Aden, we determine a new rigid plate model describing the motion since 3 Ma between India and Australia. The Euler vector (ω = 0.313°/m.y. about 5°S, 78°E) lies near the middle of the equatorial, diffuse plate boundary dividing the Indian from the Australian plate and predicts a rate of north‐south shortening along 85°E of 4±3 mm/yr, only 30% as fast as predicted by our prior model. The new model also predicts north‐south extension of 6±2 mm/yr (at 68°E) along the western segment of the diffuse plate boundary, where our prior model predicted north‐south contraction. Using data only along the Carlsberg and Central Indian ridges and no other plate boundaries, we show that plate motion data cannot be fit by a single Euler vector. However, the data are well fit by two Euler vectors when an east‐west striking India‐Australia plate boundary is assumed to intersect the Central Indian Ridge near the equator. The best location along the Central Indian Ridge for this triple junction is 6°S–3°S, with 95% confidence limits of 9°S–4°N, just west of a region of intense seismicity. The sense of deformation, as recorded in earthquake focal mechanisms, reverse faults mapped with reflection seismic data, and undulations in basement topography, surface gravity, and the geoid, agrees well with the north‐south extension predicted in the western part and the north‐south shortening predicted in the eastern part of the India‐Australia boundary. The predicted rate of north‐south shortening between 79°E and 86°E is consistent with the rate of shortening inferred from observed faulting and folding; shortening taken up by faulting is ∼6 to 100 times greater than that taken up by the spectacular basement folds. Little of the rapid north‐south shortening predicted east of ∼86°E is taken up by crustal thickening. Instead, lithosphere is transported northeastward toward the Sumatra trench through strike‐slip faulting and possible clockwise rotation. The failure of the convergent segment of the diffuse plate boundary to form a subducting trench provides some new observations that must be satisfied by models for the initiation of subduction. A model of separate, rigid Indian and Australian plates divided by a diffuse plate boundary appears to be valid and useful because the predictions of the model agree with independent data and because the predicted velocity of relative motion is statistically significant and comparable with that across other plate boundaries.