The effects of variable sources, processes and contaminants on the composition of northern EPR MORB (8–10°N and 12–14°N): Evidence from volatiles (H2O, CO2, S) and halogens (F, Cl)

New volatile (H2O, CO2, S), halogen (F, Cl) and trace-element data for selected fresh MORB glasses are reported from two geologically and geophysically well-studied regions on the East Pacific Rise (8–10°N and 12–14°N) with distinct differences in spreading rate and magma supply. Sample locations include on-axis and young off-axis eruptions, as well as off-axis fissures, abyssal hills and pillow mounds. H2O, F, S and trace-element concentrations increase with decreasing MgO content, displaying over-enriched liquid lines of descent consistent with combined fractional and in-situ crystallization. A negative correlation between CO2/Nb and MgO indicates simultaneous degassing and magma crystallization, while broadening of this correlation to lower CO2/Nb at constant MgO indicates shallow degassing and CO2 loss during magma transport to the seafloor. Cl concentrations and associated high Cl/Nb and Cl/K ratios of some northern EPR MORB result from variable pre-eruption contamination by high-salinity brines derived from supercritical phase separation of seawater within deeply-rooted hydrothermal circulation systems. In the faster-spreading 9–10°N and the Siqueiros Transform region magma-brine contamination is limited to off-axis samples, while on-axis samples contain no excess Cl. In the slower-spreading 12–14°N region, on-axis and off-axis samples exhibit similar extents of Cl addition and magma-brine contamination. This contamination appears to affect only Cl, and does not appear to impact concentrations of H2O or highly incompatible trace elements such as K and Ba. ved H2O and CO2 concentrations provide pressure estimates of eruption initiation, and indicate the majority of on-axis samples in the 9–10°N region (including Siqueiros Transform) and 12–14°N region were extracted from crustal depths equivalent to the top of the local axial magma chamber without significant degassing. Samples recovered from off-axis locations in both regions were in magma-vapour equilibrium at eruption depths or at best extracted undegassed from within layer 2B. These results indicate vertical transport of magma leading to eruption is generally faster on-axis than off-axis. Pressure estimates show that Cl addition mainly occurs within or below layer 2B, and rarely at the top of the axial magma chamber. Off-axis Cl contamination occurred similarly in both regions, independent of spreading rate. aminated samples from both regions yield similar estimates for H2O and halogen abundances in the northern EPR mantle. Combined estimates (128 ± 5 ppm H2O, 178 ± 1 ppm F, 1298 ± 6 ppm S, and 5.38 ± 1.1 ppm Cl) are similar to previous estimates based on volumetrically and geographically more-restricted host glass and melt inclusions from Siqueiros Transform samples.