Chemistry of hydrothermal solutions from Pele's Vents, Loihi Seamount, Hawaii

Abstract

Hydrothermal fluids were sampled from Pele's Vents on the summit of Loihi Seamount, an intraplate, hotspot volcano, on four occasions from February 1987 to September 1990. The warm (≤ 31°C) vent solutions are enriched in dissolved Si, CO 2, H 2S, alkalinity, K +, Li +, Rb +, Ca 2+, Sr 2+, Ba 2+, Fe 2+, Mn 2+, NH 4 +, and possibly Ni 2+, and depleted in SO 4 2−, O 2, Mg 2+, 87Sr 86Sr , NO 3 −, and sometimes Cl − and Na + (calculated), relative to ambient seawater. Dissolved Si correlates linearly with sample temperature, suggesting that the solutions sampled from numerous vents in the ~ 20 m diameter field have a common source and that Si can be used as a conservative tracer for mixing of the vent fluids with ambient seawater. There are general similarities of the vent waters with ridge-axis warm springs on the Galapagos Rift and Axial Seamount, but also striking differences: very high total dissolved CO 2 (> 200 mmol/kg), high alkalinity (> 8 meq/kg) and dissolved Fe 2+ (almost 1 mmol/kg), and relatively low pH (~ 4.2–4.4 estimated, in situ, sws) and dissolved H 2S (several μmol/kg). The Mg 2+ and SO 4 2− data are inconsistent with the “Galapagos model” proposed for the warm springs at 86°W, Galapagos Rift, whereby the warm fluids result from sub-seafloor mixing of a high-temperature (~ 350°C) hydrothermal endmember with essentially unaltered seawater. The variable Cl − depletions in the vent fluids, however, suggest that the warm vent fluids do contain a high-temperature (> 200°C) component. The fluid history can be qualitatively described by a modified “Galapagos model” which includes the overprint of reactions resulting from the addition of juvenile CO 2 and SO 2 to the circulating fluids; the CO 2 attacks the basalt releasing metal cations, HCO 3 − and possibly Si into solution, and the SO 2 is hydrolyzed to SO 4 2−. These juvenile inputs likely reflect the shallow, hotspot setting of this hydrothermal system. A simple quantitative fluidhistory model is considered and shown to be consistent with mass-balance constraints and saturationstate calculations, which suggest that the Si concentration of the fluids may be controlled by amorphous silica saturation at ~ 31°C. Observed temporal variations in fluid composition between expeditions—specifically, in Cl −, A T , C T , Na + (calculated), Mg 2+, Ca 2+, Sr 2+, 87Sr 86Sr , Fe 2+, Mn 2+ and perhaps NH 4 +, relative to Si—are, excepting Mg 2+, 87Sr 86Sr , and Mn 2+, consistent with the effects of variable phase segregation at the proposed high-temperature endmember.