Motion between the Indian, Capricorn and Somalian plates since 20 Ma: implications for the timing and magnitude of distributed lithospheric deformation in the equatorial Indian ocean

SUMMARY Approximately 2200 magnetic anomaly crossings and 800 fracture zone crossings flanking the Carlsberg ridge and Central Indian ridge are used to estimate the rotations of the Indian and Capricorn plates relative to the Somalian Plate for 20 distinct points in time since 20 Ma. The data are further used to place limits on the locations of the northern edge of the rigid Capricorn Plate and of the southern edge of the rigid Indian Plate along the Central Indian ridge. Data south of and including fracture zone N (the fracture zone immediately south of the Vema fracture zone), which intersects the Central Indian ridge near 10 ◦ S, are well fit assuming rigid Capricorn and Somalian plates, while data north of fracture zone N are not, in agreement with prior results. Data north of fracture zone H, which intersects the Central Indian ridge near 3.2 ◦ S, are well fit assuming rigid Indian and Somalian plates, while data south of and including fracture zone H are not, resulting in a smaller rigid Indian Plate and a wider diffuse oceanic plate boundary than found before. The data are consistent with Capricorn‐Somalia motion about a fixed pole since ≈8 Ma, but require rotation about a pole 15 ◦ farther away from the Central Indian ridge from 20 to ≈8 Ma. The post-8-Ma pole also indicates Capricorn‐ Somalia displacement directions that are 7 ◦ clockwise of those indicated by the pre-8-Ma stage pole. In contrast, India‐Somalia anomaly and fracture crossings are well fit by a single fixed pole of rotation for the past 20 Ma. India‐Somalia motion has changed little during the past 20 Myr. Nonetheless, astronomically calibrated ages for reversals younger than 12.9 Ma allow resolution of the following small but significant changes in spreading rate: India‐Somalia spreading slowed from 31 to 28 mm yr −1 near 7.9 Ma and later sped up to 31 mm yr −1 near 3.6 Ma; Capricorn‐Somalia spreading slowed from 40 to 36 mm yr −1 near 11.0 Ma, later sped up to 38 mm yr −1 near 5.1 Ma and further sped up to 40 mm yr −1 near 2.6 Ma. The motion between the Indian and Capricorn plates is estimated by differencing India‐Somalia and Capricorn‐Somalia rotations, which differ significantly for all 20 pairs of reconstructions. India has rotated relative to the Capricorn Plate since at least ≈20 Ma. If about a pole located near 4 ◦ S, 75 ◦ E, the rate of rotation was slow, 0.11 ◦ ± 0.01 ◦ Myr −1 (95 per cent confidence limits), from 20 to 8 Ma, but increased to 0.28 ◦ ± 0.01 ◦ Myr −1 (95 per cent confidence limits) at ≈8 Ma. The onset of more rapid rotation coincides, within uncertainty, with the inferred onset at 7‐8 Ma of widespread thrust faulting in the Central Indian basin, and with the hypothesized attainment of maximum elevation and initiation of collapse of the Tibetan plateau at ≈8 Ma. The plate kinematic data are consistent with steady India‐Capricorn motion since 8 Ma and provide no evidence for previously hypothesized episodic motions during that interval. The convergence since 8 Ma between the Indian and Capricorn plates significantly exceeds (by 13 to 20 km) the convergence estimated from three north‐south marine seismic profiles in the Central Indian basin. Where and how the additional convergence was accommodated is unclear.