Interactions between deep formation fluid and gas hydrate dynamics inferred from pore fluid geochemistry at active pockmarks of the Vestnesa Ridge, west Svalbard margin

Hong, W.L.1,2,3, Pape, T.4, Schmidt, C.5, Yao, H.2, Wallmann, K.5, Plaza-Faverola, A.2, Rae, J.W.B.6, Lepland, A.1,2, Bünz, S.2, and Bohrmann, G.4

1Geological Survey of Norway (NGU), Trondheim, Norway
2Center for Arctic Gas Hydrate, Environment, and Climate (CAGE), the Arctic University of Norway (UiT), Tromsø, Norway
3Department of Geological Sciences, Stockholm University, Stockholm, Sweden
4MARUM–Centre for Marine Environmental Sciences and Faculty of Geosciences, University of Bremen, Bremen, Germany
5GEOMAR Helmholtz Centre for Ocean Research, Kiel, Germany
6School of Earth and Environmental Sciences, University of St. Andrews, St. Andrews, UK

Seafloor seepage sites along the Vestnesa Ridge off west-Svalbard have been, for decades, a natural laboratory for the studies of fluid flow and gas hydrate dynamics at passive continental margins. The lack of ground truth evidence for fluid composition and gas hydrate abundance deep in the sediment sequence however prohibits us from further assessing the current model of pockmark evolution from the region. A MARUM-MeBo 70 drilling cruise in 2016 aims to advance our understanding of the system by recovering sediments tens of meters below seafloor from two active pockmarks along Vestnesa Ridge. We report pore fluid composition data focusing on dissolved chloride, stable isotopes of water (δ18O and δD), and the isotopic composition of dissolved boron (δ11B). We detect a saline formation water around two layers where gas hydrates were recovered from one of the seepage sites. This saline formation pore fluid is characterized by elevated chloride concentrations (up to 616 mM), high B/Cl ratios (9×10-4 mol/mol), high δ18O and δD isotopic signatures (+0.6 ‰ and +3.8 ‰, respectively) and low δ11B signatures (+35.0 ‰), which collectively hint to a high temperature modification at great depths. Based on the dissolved chloride concentration profiles, we estimated up to 47 % of pore space occupied by gas hydrate in the sediments shallower than 11.5 mbsf. The observation of bubble fabric in the recovered gas hydrates suggests formation during past periods of intensive gaseous methane seepage. The presence of these gas hydrates without associated positive anomalies in dissolved chloride concentrations however suggests that the decomposition of gas hydrate is as fast as its formation. Such a state of gas hydrates can be attributed to a relatively low methane supply transported by the saline formation water at present. Our findings based on pore fluid composition corroborate previous inferences along Vestnesa Ridge that fluids sustaining seepage have migrated from great depths and that the variable gaseous and aqueous phases through the gas hydrate stability zone controls the distributions of authigenic carbonates and gas hydrates.