We present trace element analyses of basaltic and picritic lavas recovered from the Ocean Drilling Program Leg 152 transect on the southeast Greenland volcanic rifted margin.These lavas span the stratigraphic interval from continental to oceanic magmatism.Incompatible element patterns define two geochemical groups within the Site 917 upper series.Group 1 flows are characterized by (La/Sm) N = 0.3-0.8,whereas group 2 flows have (La/Sm) N = 0.7-1.2 and higher mantle-normalized Th, Ba, and Pb, and lower Nb and U concentrations.The transition upsection from group 1 to group 2 at Site 917 occurs between 114 and 65 meters below seafloor, and is represented by interfingering of flows belonging to the two groups.(La/Sm) N , Ba/Zr, and Th/Pb ratios and mantle-normalized incompatible element concentrations decrease systematically with stratigraphic height within each group, and the Site 915 and Site 918 units are generally continuous with group 2. These variations imply an increase in the extent of partial melting with time.Lu/Hf ratios vary from 0.14 to 0.25 through the upper series and into the Site 915 and Site 918 flows.This relationship suggests the importance of residual garnet in the mantle source melting decreased with time.We develop a quantitative model for mantle melting to investigate melting systematics responsible for these relationships.Comparison between observed data and model results suggests a progressive increase in the extent of partial melting (from 4% to 12%) and decrease in mean pressure of melting with time.This temporal evolution of primary magma compositions is explicable by rapid thinning of the continental lithosphere during eruption of the upper series.We conclude that group 1 units were derived from mantle with normal mid-ocean-ridge basalt source characteristics, whereas units from group 2 and from Sites 915 and 918 were derived from a source similar to depleted Icelandic mantle.We infer that the thermal anomaly associated with the ancestral Iceland plume pre-dated the transition in mantle source compositions.
Coastal oceans are vital to world health and sustenance. Technology that enables new observations has always been the driver of discovery in ocean sciences. In this context, we describe the first at sea deployment and operation of an inductively coupled plasma mass spectrometer (ICPMS) for continuous measurement of trace elements in seawater. The purpose of these experiments was to demonstrate that an ICPMS could be operated in a corrosive and high vibration environment with no degradation in performance. Significant advances occurred this past year due to ship time provided by Scripps Institution of Oceanography (UCSD), as well as that funded through this project. Evaluation at sea involved performance testing and characterization of several real-time seawater analysis modes. We show that mass spectrometers can rapidly, precisely and accurately determine ultratrace metal concentrations in seawater, thus allowing high-resolution mapping of large areas of surface seawater. This analytical capability represents a significant advance toward real-time observation and understanding of water mass chemistry in dynamic coastal environments. In addition, a joint LLNL-SIO workshop was convened to define and design new technologies for ocean observation. Finally, collaborative efforts were initiated with atmospheric scientists at LLNL to identify realistic coastal ocean and river simulation models to support real-time analysis and modeling of hazardous material releases in coastal waterways.
