Variations in the velocities of the major plates since the late cretaceous

Abstract

We present models for the relative and absolute velocities of the major plates from 80 m.y. ago to the present as tests of several alternative plate reconstructions and as a guide in the evaluation of the hypothesis that pronounced changes in plate speeds or directions arise directly from changes in the configuration of plate boundary forces. Digitized plate boundaries were rotated backwards in time using published rotations to determine positions at 5 m.y. intervals back to the late Cretaceous. The oceanic plate boundaries were then corrected so as to properly match magnetic anomalies. Relative plate velocities were obtained by differencing finite rotations. “Absolute” velocities were determined by assuming that no net torque is exerted on the lithosphere. Rms “absolute” speeds of each plate were calculated as functions of time as a conceptually simple measure of plate velocities and accelerations. We have tested several possible reconstructions for the Antarctic and Pacific plates proposed to explain the failure of simple rigid Antarctic and/or Pacific plate models to match paleomagnetic data, hotspot tracks, and equatorial crossings defined by pelagic sediment deposition. We find that: 1. (1) Plate reconstruction models with either two Antarctic or two Pacific plates in the early Tertiary are distinguishable from models with single Antarctic and Pacific plates principally on the basis of faster speeds for the Kula plate and the northern Pacific, greater areas for the Kula and Farallon plates, and faster subduction at the Aleutians and along the west coast of South America for that time. 2. (2) Oceanic plates as a group have moved faster than those with continental lithosphere only since 50 m.y. ago; active subduction appears to be a sufficient condition for rapid plate motion regardless of the fractional plate area comprised of continental lithosphere. 3. (3) The absolute motion of India in the late Cretaceous was anomalously fast (15–20 cm/yr) compared to the speeds of the present plates. 4. (4) The rms velocities of a number of plates and of the lithosphere as a whole are much higher in the late Cretaceous than through most of the Cenozoic. 5. (5) The predicted apparent polar wander for North America is in fairly good agreement with paleomagnetic data, with the slight discrepancy attributable to a net contribution to the torque balance from plate boundary forces.