Melting Processes in Infant Subduction Zones: HFSE fractionation in boninites

How subduction of one plate under another begins is a fundamental problem in plate tectonics. Unfortunately, there are no active examples of subduction initiation that we can study. So the primary constraints come from looking at the preserved magmatic products of infant subduction zones. Such lavas are often found in the fore-arc of subduction zones, the area between the main arc-related volcanic chain and the trench. They can also be found in ophiolites, slivers of oceanic crust that have been thrust onto continents or otherwise brought to the surface. The magmas produced in infant arcs often have a distinctive major and trace element composition and are called boninites. The major element composition of boninites records the pressures and temperatures of the mantle from which they were produced. By looking at the temporal evolution of these mantle P-T conditions, one can constrain how the mantle beneath the early arc was flowing. This in turn provides invaluable clues to why subduction started in the first place. While complicated, the P-T conditions recorded by boninites are by far the most direct record of early-arc mantle flow, and are key to understanding the initiation of subduction. A number of studies have constrained the melting conditions of boninites using high-pressure experiments to mimic melting conditions. The estimates vary over a wide-range of P-T conditions, largely depending upon the particular boninite composition chosen for the experiments. One source of the confusion may come from interaction of boninite melts with the lithosphere, which may be recorded in the trace element composition of boninites. In addition to their odd major element composition, boninites also have unique trace element compositions. One notable feature is that the elements Hf, Ta and Ti are often depleted, while Zr and Hf are enriched. Usually these elements behave coherently (all depleted or all enriched) and so are usually grouped together as the high field-strength elements (HFSE). A number of processes have been proposed to explain the HFSE fractionation but the most current data suggest it originates in mantle by partitioning between boninite melts and mantle minerals. The different hypotheses have distinct implications for the melting process and for estimated P-T conditions. However, current partitioning data is insufficient to decide between the various hypotheses. The proposed research will measure the partitioning of HFSE between melt and mantle minerals using high-pressure experiments. The results will be used to test the existing hypothesis and will lead to more confident and accurate estimates of mantle conditions in early arcs. Because of their distinctive chemistry, boninites are relatively easy to identify, even in ancient terranes that have been heavily metamorphosed. Some of the oldest magmas known (3.7 Ga) are boninites. As boninites are only formed in subduction zones, they are one of a very few types of sample that can give an unambiguous tectonic setting from just the geochemistry of the rocks. Thus boninites represent a fairly unique chance to trace one element of plate tectonics (subduction) from the present, back to the Earth's earliest history and address issues such as secular cooling and the initiation of plate tectonics. Thus, an improved understanding of boninite formation in the modern Earth will provide key clues to the Earth's earliest history.