CHEMISTRY OF SPRINGS ACROSS THE MARIANA FOREARC SHOWS PROGRESSIVE DEVOLATILIZATION OF THE SUBDUCTING PLATE

Abstract

Cold springs upwelling through large serpentinite mud volcanoes in the outer half of the Mariana forearc provide a unique window into processes of devolatilization of the subducting Pacific Plate. We have sampled upwelling pore waters with lower chlorinity than seawater from six sites on five serpentinite mud volcanoes, by conventional gravity and piston coring, by push coring from the ROV Jason, by drilling on ODP Legs 125 and 195, and by manned submersible. The sites range from 13°47'N to 19°33'N and 52 to 90 km from the Mariana trench axis, corresponding to approximate depths to the top of the downgoing plate of 16 to 29 km. The composition of the springs varies systematically over this distance: nearer the trench the upwelling waters have much higher Ca and Sr than seawater and much lower carbonate alkalinity, sulfate, Na/Cl, K, Rb, and B. Farther from the trench the waters show the opposite trends relative to seawater. Chlorinity is consistently lower than in seawater and shows large variations that are not systematic with distance from the trench. Cs is consistently higher than in seawater and increases with distance from the trench. All of the waters have high pH and are heavily depleted in Mg, Si, Li, F, and 87Sr/86Sr relative to seawater. They tend to be enriched in O18/O16. Except for ODP drilling, none of the cores was long enough to produce an asymptotic compositional trend with depth. We have inferred the end-member compositions of the upwelling waters by extrapolation against Mg. At two sites we were able to compare data from gravity cores with data from drill cores or push cores collected at springs to estimate the effects of reactions that occur at shallow depth below the seafloor, on mixing of the upwelling waters with seawater. These effects are different for sites high in dissolved Ca, nearer the trench, vs. those high in alkalinity, farther from the trench. Common to both are large losses from solution of 1) Ca, as CaCO3 and in exchange for Na; 2) Mg, in exchange for Na or Ca and as brucite; 3) sulfate, probably reduced by microbes or possibly precipitated as gypsum; 4) Sr, Ba, Si, and F. Na is consistently leached from the solids into solution, whereas K and O18/O16 are relatively unreactive. We infer that the upwelling waters are uniformly saturated with CaCO3 and that the excess H2O and the trends in Ca, Sr, alkalinity, and sulfate with distance from the trench result from introduction of H2O and dissolved carbonate and sulfate from an external source, the sediment and altered basalt at the top of the subducting plate. The concurrent trends in Na/Cl, B, Cs, and especially K and Rb indicate that these species originate from the top of the subducting plate in response to increasing temperature. These systematic variations across the outer forearc imply that the solutions ascend more or less vertically from the source region and do not travel long distances laterally along the décollement before ascending. Based on leaching of K, the 150°C isotherm is crossed approximately beneath Big Blue Seamount at a depth of ˜22 km below the seafloor, 70 km behind the trench. By this point it appears that carbonate dissolution has joined dehydration as a significant process at the top of the subducting plate.