MODELING PORE FLUID CHEMISTRY IN ACCRETIONARY WEDGE SEDIMENTS - NEW INTERPRETATION OF ODP DATA

The chemistry of accretionary wedge sediments and their associated pore fluids is important in understanding the cycling of chemical elements in subduction zones. Interstitial pore fluid chemistry is used to assist in the interpretation of exchanges within the sediment column, as fluid migration and diagenetic processes occur. Data have been collected during several ODP cruises, and initial interpretations made by the onboard scientists. These interpretations are generally qualitative, based on knowledge of conditions in the sediments and likely reactions.

This study takes a different approach, in that it models the data mathematically before looking to make interpretations. This allows us to find relationships among the chemical elements that hold true from site to site, and then interpret the processes that vary from location to location that are not accounted for by the model. The model was developed using data from Site 671 in Barbados, where the geology is relatively simple. Ca and Mg concentrations were fitted based on the concentrations of other ions in the pore fluids. All data collected by ODP were entered into the statistical regression, but not all were found to have a statistical correlation with the Ca and Mg. The ions that showed a statistical correlation were Na, Cl, Mg, Ca and SO₄. This model was then used to predict Ca and Mg concentrations at other sites from Barbados and the Nankai Trough. The other sites in Barbados were a close fit to the actual results at depth, but showed offsets at the surface, down to depths between 100 and 300 mbsf. Further investigation showed that these offsets were directly correlated with variations in sulfate in the pore fluids. The sulfate variations close to the surface likely result from organic activity, so it seems that the model accounts for inorganic chemical reactions, but not organic processes. When the model was applied to the Nankai Trough site, the fit was good down to the depth where illite-smectite transformation has been clearly documented. This would suggest that the model does not account for processes that alter pore fluid chemistry during clay mineral diagenesis, and would therefore imply that there is no such diagenesis at the Barbados sites. This is an important finding, as previous authors have disagreed about the influence of diagenesis on pore fluids in Barbados.