Scotia Sea regional tectonic evolution: implications for mantle flow and palaeocirculation

The Scotia Sea and surrounding Scotia Are have evolved over the past 40 Ma, by extension behind an east-migrating subduction zone, at the boundary bet ween the South American (SAM) and Antarctic (ANT) plates. The considerable data set now available (regional geology and geophysics, earthquake seismol ogy, satellite altimetry, global plate analyses) suggest why east-migrating subduction began, what has been the driving force that has sustained it, a nd what other processes have controlled the made of back-arc extension in t he Scotia Sea. A suite of six reconstructions has been developed, based on this data set. The reconstruction to 40 Ma creates a compact, cuspate conti nental connection between South America and the Antarctic Peninsula at the subducting Pacific margin, with fragments (now dispersed around the Scotia Arc) occupying positions within it compatible with their known geology. The driving force has been subduction of South American ocean floor, which beg an as a result of southward migration of the pole of South American-Antarct ic plate rotation, and a key modulator of back-arc extension has been colli sion of ridge crest sections of the South American-Antarctic plate boundary with the east-advancing trench. Cenozoic regional tectonic evolution has t wo other likely consequences which greatly increase its importance. Firstly , this region saw the tectonic disruption of the final barrier to complete circum-Antarctic deep water flow, that may have had a profound effect on pa laeoclimate. Secondly, it is possible that the rapid roll-back of the hinge of subduction is related to shallow eastward flow in the sub-lithospheric mantle. Both of these consequences are explored. The reconstructions show t hat rapid roll-back of the subduction hinge (averaging 50 mm/a over the las t 40 Ma with respect to the South American plate) has been a feature of all of Scotia Sea evolution, and provide a history of motion of several oceani c microplates, most of which are now welded together within the Scotia Sea. This will guide the location of seismometers and/or dredge hauls to test t he hypothesis of shallow mantle flow, and help interpret the results. The r econstructions also allow an assessment of the creation of deep-water pathw ays that would have permitted the development of the present-day Antarctic Circumpolar Current (ACC). An early Miocene onset (within the period 22-17 Ma) seems likely for the ACC, depending on the structure and palaeo-elevati on of Davis Bank and Aurora Bank, sections of the North Scotia Ridge. Howev er, the study shows there was a delay (of one or more million years) betwee n initial provision of a deep-water pathway and the major rr:mid-Miocene ch ange in global climate (involving the general level of Antarctic glaciation ) that may have been related. If these changes were related, then the delay suggests that other factors, possibly rough elevated ocean floor but also non-tectonic factors (such as atmospheric CO2), were important in determini ng palaeoclimate. (C) 2001 Elsevier Science B.V. All rights reserved.