The pore fluids at Hydrate Ridge show the strongest enrichment in iodine observed at any gas hydrate site, consistent with the unusually high concentration of gas hydrates there. The biophilic behavior of iodine and the close association between iodine and methane observed here and at other gas hydrate areas suggest that methane and iodine are derived from the same organic material and travel together in aqueous fluids until they reach areas close to the seafloor (e.g., Fehn et al., in press). The ages found for the iodine in the pore water can thus be used to identify formations responsible for the release of the methane found in the gas hydrate.
All iodine ages found in pore water at Hydrate Ridge are older than 15 Ma. Because this age is considerably older than the ages of the shallow sediments hosting the gas hydrates, the pore water must contain iodine that was not locally derived. The majority of the samples have ratios between 150 and 250 x 10–15, corresponding to ages between 40 and 52 Ma. These ages are minimum values because the calculation does not take into account the potential addition of fissiogenic 129I, which is a function mainly of age and uranium concentration in the host formation. A correction for the presence of this component in the age range observed here and with uranium concentrations typical for marine sediments would increase the ages by ~5 m.y. (Fehn et al., 2000), shifting the proposed ages for the old component in the pore fluids into a range between 45 and 57 Ma. This age range is beyond the ages of the currently subducting sediments (<10 Ma) and the formations underlying Hydrate Ridge and suggests that fluids contributing to the current pore water come from formations of early Eocene age or older in the overriding wedge. Profiles presented by Snavely (1987) suggest the presence of formations with ages of early Eocene just to the west of the Siletz terrane, the crystalline backstop in the area. Because the closest formations of this age are found at distances of 40 km or more, fluids must have traveled considerable distances through the overriding wedge in order to reach their present location. Fluid flow models of active margins suggest that fluids can move along the décollement or fractures, with actual velocities depending strongly on the permeability and the mode of flow (e.g., Saffer and Bekins, 1998, 1999). Rates between 0.0007 and 0.043 m/yr were calculated for steady-state flow in subduction zones, reaching values up to 54 m/yr for transient periods (Saffer and Bekins, 1998). These rates suggest that transport of fluids from the Eocene formations in the accretionary complex to the Hydrate Ridge is possible within time frames of the age of the current subduction configuration. The new suggestion here is that the origin of iodine and methane is not from the subducting sediments but from the overriding wedge. Although it seems likely that the waters are also derived from this source, it is conceivable that waters expelled from the subducting sediments have come into contact with formations of Eocene age, where they picked up iodine and methane. The ubiquitous presence of fractures in the overriding wedge suggested by seismologic surveys (Snavely, 1987; Snavely and Wells, 1996) supports the suggestion that fluid flow can occur over great distances in this area, which might also involve transport from the suggested source region into the décollement, a preferred conduit for fluids. Transport of hydrocarbons over distances of this magnitude is common in oil reservoirs (e.g., Moran et al., 1995) and sedimentary basins (e.g., Person et al., 1996). Improved information on age distribution and the presence of deep fractures would be needed in order to be more specific on the flow paths taken by these fluids and the role played by the décollement in this transport model.
The presence of a distinct maximum in iodine concentrations in the depth profile for Site 1245 (and perhaps at some other sites) is unique as far as gas hydrate sites and warrants further discussion. The younger ages coincide with the maximum in iodine concentrations, although the widths of the two maxima are somewhat different. If ratios are plotted vs. the reciprocal of concentration (Fig. F7), two trends are visible in the data set. Trend A indicates dilution of a component with high concentrations of iodine and low 129I/I ratios with fluids of low concentration and high 129I/I ratios. If this trend is extrapolated to the surface, mixing is indicated between an old, iodine-rich source (C1 = 2.2 mM; R1 = 150 x 10–15) with waters in shallow sediments (C2 = 0.5 mM; R2 = 1500 x 10–15). The concentration found for the Holocene component in this mixing pattern is within the range observed for fluids in shallow sediments (e.g., Martin et al., 1993; Egeberg and Dickens, 1999) and thus represents the contribution from recently deposited marine sediments. Although the "old" end-member used for this calculation is perhaps also affected by the presence of "young" material, the iodine concentration of the fluids with young material are relatively low so that their influence on the old end-member is negligible at the level of precision here. Trends similar to line A are observed also at other gas hydrate sites, as, for example, at Site 1230 (ODP Leg 201, Peru margin) (Fehn et al., in press).
Trend B describes a line almost parallel to the y-axis, indicating mixing between two fluids with different iodine ages but quite similar iodine concentrations. Whereas the minimum age for the older source is close to that of trend A, the end point in the mixing plot for line B is at a sample with a ratio of 650 x 10–15. The essentially vertical shape of the mixing line between the two end-members suggests that the mixing ratio here is close to 1:1 (Faure and Mensing, 2005). Assuming concentrations of 2.0 mM and equal contributions between the old and young member, we find a ratio of 1140 x 10–15 (corresponding to 6.2 Ma) for the young end-member. The actual age for the latter component is probably somewhat younger, however, because we cannot be sure that we found the exact maximum in the distribution of isotope ratios. Likely sources for the "young" material are the formations of Pliocene age underlying Hydrate Ridge (see Fig. F1B), although derivation from currently subducting sediments cannot be ruled out either.
I/Br ratios may also be used to indicate the presence of more than one source at several of the sites, specifically at the flank sites. A comparison to other gas hydrate sites (Egeberg and Dickens, 1999; Fehn et al., 2003) indicates that this ratio is always considerably elevated compared to seawater but is site specific, reaching quite constant values within a given site at depths beyond the influence of recent seawater. These ratios probably reflect the characteristics of the organic sources responsible for the addition of iodine and bromine to pore water. In the case of Hydrate Ridge, maxima in the depth distribution of iodine concentrations are accompanied by simultaneous maxima in I/Br ratios. This observation is in good agreement with the interpretation of 129I/I results, indicating that the young high-I source also has higher I/Br ratios than the old high-I source.
The results of the iodine isotopes suggest that a major part of the iodine and methane in the pore water is not derived from local sources but from older formations at considerable distance from Hydrate Ridge. Indirect evidence for large-scale transport comes also from the freshening with depth of pore water observed at several sites, most clearly at the slope basin sites, which has been interpreted at sites such as the Nankai Trough as evidence for the addition of deep fluids (Kastner et al., 1993). A concurrent study of strontium isotopes at Hydrate Ridge might also show the presence of waters of Eocene age in several of the cores, most clearly discernable at the slope basin sites (R. Matsumoto, pers. comm., 2004), although these results have also been interpreted as the result of the interaction of waters with the basement (Torres et al., 2004; Teichert et al., in press).