From seawater Li isotope records to carbon cycle in deep time-revisiting the hydrothermal control on oceanic Li isotope mass balance

Abstract

The deep-time seawater Li isotope records have essential implications for carbon cycle and climate change because the oceanic Li isotope mass balance is controlled by the chemical weathering input which consumes atmospheric CO2 and reverse weathering sink that generates CO2. However, hydrothermal-seafloor interaction also contributes major oceanic Li budgets yet the controls on long-term evolution of seawater Li isotope composition are not well constrained. Venting fluids sampled from the mid-ocean-ridge (MOR), arc and back-arc hydrothermal sites show large variations in both Li concentrations (20 to 6000 mmol/kg) and δ7Li values (1.3‰ to 15.4‰). The mechanisms controlling the Li mobility between seafloor and vent fluids as well as the isotope fractionation are not well understood.  Additionally, the arc and back-arc hydrothermal systems need to be reassessed for their contribution to the oceanic Li budgets and Li isotope composition because their venting fluids show overall higher Li concentrations and lower δ7Li values than those from the MOR. Here we systematically investigate the Li isotope behaviors in two drilling cores of the submarine volcano Brothers (IODP 376) where two distinct types of hydrothermal systems are developed. U1528 is located at the volcano summit whose hydrothermal fluid is influenced by magmatic degassing, and U1530 at the NW caldera wall develops a seawater-dominated hydrothermal system that resembles the one at MORs. Relative to the unaltered precursor, the extreme Li loss of 95% to 99% in samples from U1528 are attributed to intensive mineral dissolution under low pH. Compared to U1528, samples from U1530 show lower extents of Li depletion between 50% and 95% and are explained by chlorite precipitation and possible seawater Li addition. The decreasing Li concentration in altered rocks is accompanied by elevating δ7Li values, implying preferential removal of light Li isotope from rock to the hydrothermal fluid. The kinetic (Raleigh) and equilibrium isotope fittings suggest the Li isotope fractionation between oceanic crust and hydrothermal input endmember is smaller than 3‰. For the inclusion of Li isotopes into biogeochemical and climate models, we propose a refined mass balance for the oceanic Li isotopes. We argue that previously proposed hydrothermal Li influx of 13 x 109 mol/yr and δ7Li values of either 8.3‰ or 6.5‰ are overestimated.  The higher δ7Li values reported in venting fluids likely reflect seawater mixing signal.