Experimental determination of the melting phase relations of subducted sediment - a case study in the Lesser Antilles

One of the most striking pieces of evidence for modern plate tectonic was the discovery that the ocean floor is spreading, forming new oceanic lithospheric plate and giving rise to drifting continents. The fact that oceanic lithosphere forms at mid-ocean ridges requires that elsewhere oceanic lithosphere must be transferred back into the deep Earth. This happens at subduction zones, where dense oceanic lithosphere sinks below over-riding oceanic or continental lithosphere. One process that is ultimately related to subduction zones is the formation of volcanic arcs. The most prominent example is subduction of the Pacific Oceanic plate below the north American plate in the east, below the Aleutians in the north and below Japan in the northwest, forming the so called 'Ring of Fire'. Volcanoes above subduction zones are characteristically explosive, as exemplified by the 1980 eruption of Mount St. Helens (USA) or the 1883 eruption of Krakatoa (Indonesia). Such large eruptions can have a profound effect on local populations and global climate. Subduction zones give rise to volcanism because fluids and melts are released from the subducting lithospheric slab as it becomes gets heated up while sinking through Earth's mantle. These fluids and melts are released from different portions of the slab, namely from the part that represents the oceanic crust, from its sedimentary cover (e.g. shales, deep marine oozes, clays) and from the underlying lithospheric mantle, variably serpentinised by hydrothermal activity at mid-ocean ridges. These chemically buoyant fluids and melts interact with the overlying column of mantle peridotite, eventually triggering melting by lowering the melting point. Water-bearing basaltic magmas so-produced ascend into the crust, differentiate to more silicic compositions and eventually give rise to explosive volcanism on the over-riding plate. While geologists broadly know that such processes must happen in subduction zones, its details remain poorly understood. Geochemical data on volcanic rocks has proven to be a useful tracer of plate tectonic processes in general. Magmas erupted in different plate tectonic settings have characteristic geochemical 'flavours'. For example, enrichment in light rare earth elements, uranium and thorium, coupled with depletion in the high field-strength elements characterises lavas from subduction zones, supporting the involvement of fluids from the subducting slab. Although the process is conceptually simple, the details remain elusive, most notably the temperature to which the subducting slab is subjected at depth, the nature of the extracted fluids and the chemistry of the residual materials recycled into the deep mantle. In order to be able to study subduction zone processes in more detail, the conditions where fluids and melts are generated in subduction zones must be reproduced in laboratory experiments. Traditionally such experiments focus on the volumetrically dominant basaltic and serpentinised portions of the slab, with scant experimental data on the diverse (and trace element-rich) subducted sediment. Our pilot study on high-pressure melting of red clay with variable amounts of water highlights the important role that accessory phases rich in certain trace elements play in controlling the chemistry of the fluids and melts released from the slab. The temperature dependence of the stability of these minerals (notably rutile, monazite, ilmenite and apatite) means that the chemistry of erupted arc magmas has unrealised potential as a precise geothermometer of conditions in the underlying subduction zone. We aim to conduct further experiments on red clay and other oceanic sediments. Chemical data from the West Indies will be used as a field example against which geochemical characteristics of the experimental results will be compared. Involvement of the Project Partner will enable our results to be extended to the Tonga-Kermedec arc in the SW Pacific.