Global plate boundary evolution and kinematics since the late Paleozoic

Many aspects of deep-time Earth System models, including mantle convection, paleoclimatology, paleobiogeography and the deep Earth carbon cycle, require high-resolution plate motion models that include the evolution of the mosaic of plate boundaries through time. We present the first continuous late Paleozoic to present-day global plate model with evolving plate boundaries, building on and extending two previously published models for the late Paleozoic (410–250 Ma) and Mesozoic-Cenozoic (230–0 Ma). We ensure continuity during the 250–230 Ma transition period between the two models, update the absolute reference frame of the Mesozoic-Cenozoic model and add a new Paleozoic reconstruction for the Baltica-derived Alexander Terrane, now accreted to western North America. This 410–0 Ma open access model provides a framework for deep-time whole Earth modelling and acts as a base for future extensions and refinement. We analyse the model in terms of the number of plates, predicted plate size distribution, plate and continental root mean square (RMS) speeds, plate velocities and trench migration through time. Overall model trends share many similarities to those for recent times, which we use as a first order benchmark against which to compare the model and identify targets for future model refinement. Except for during the period ~ 260–160 Ma, the number of plates (16–46) and ratio of “large” plates (≥ 107.5 km2) to smaller plates (~ 2.7–6.6) are fairly similar to present-day values (46 and 6.6, respectively), with lower values occurring during late Paleozoic assembly and growth of Pangea. This temporal pattern may also reflect difficulties in reconstructing small, now subducted oceanic plates further back in time, as well as whether a supercontinent is assembling or breaking up. During the ~ 260–160 Ma timeframe the model reaches a minima in the number of plates, in contrast to what we would expect during initial Pangea breakup and thus highlighting the need for refinement of the relative plate motion model. Continental and plate RMS speeds show an overall increase backwards through time from ~ 200 to 365 Ma, reaching a peak at 365 Ma of > 14 and > 16 cm/yr, respectively, compared to ~ 3 and ~ 5 cm/yr, respectively, at present-day. The median value of trench motion remains close to, yet above 0 cm/yr for most of the model timeframe, with a dominance in positive values reflecting a prevalence of trench retreat over advance. Trench advance speeds are excessive during the 370–160 Ma period, reaching more than four times that observed at present-day. Extended periods of trench advance and global continental and plate RMS speeds that far exceed present-day values warrant further investigation. Future work should test whether alternative absolute reference frames or relative motions would mitigate these high speeds, while still being consistent with geologic and geophysical observations, or alternatively focus on identifying potential driving mechanisms to account for such rapid motions. © 2016 Elsevier B.V.Science and Industry Endowment Fund, Australian Research Council, M.S.I. Foundatio