A total of 56 samples of mud and mudstone were analyzed during this investigation. Figures F5, F6, and F7 illustrate how relative abundances of smectite, illite, and chlorite (+ kaolinite) change as a function of depth and stratigraphic unit in the vicinity of Hydrate Ridge. All of the X-ray diffraction data are tabulated in Table T1. For the most part, the results are unremarkable. At Sites 1244, 1245, and 1251, we note a modest increase in the abundance of smectite within deeper stratigraphic intervals that have been interpreted to be part of the Cascadia accretionary complex. On average, the early Pleistocene to Holocene trench slope and slope basin deposits contain 29% smectite (standard deviation = 5.3%), 31% illite (standard deviation = 3.7%), and 40% chlorite (+ kaolinite) (standard deviation = 3.6%). These percentages are entirely consistent with the results of Karlin (1980), who mapped clay mineral abundances using surface sediments from across the continental margin of Oregon and southern Washington.
Late Pliocene to early Pleistocene strata from the underlying accretionary prism show moderate enrichments of smectite with average values of 38% smectite (standard deviation = 3.9%), 27% illite (standard deviation = 2.9%), and 35% chlorite (+ kaolinite) (standard deviation = 2.6%). We attribute this enrichment of expandable clay minerals to detrital point sources associated with the ancestral Columbia River, combined with south-directed transport of hemipelagic suspensions on the floor of Cascadia Basin. As supporting evidence for this interpretation, Karlin (1980) and Knebel et al. (1968) showed that Holocene clays emanating from the mouth of the Columbia River contain more than 50% smectite. Oceanographic currents generally push the Columbia River mud plume toward the north over the continental shelf and slope, but turbidity currents on the floor of Cascadia Basin move sand and smectite-rich suspended sediment toward the south with local funneling through channel-levee complexes (Duncan et al., 1970). DSDP Site 174 is located on the distal edge of Astoria Fan, west of Hydrate Ridge (Fig. F1). The range and average of the relative clay mineral proportions in mud and mudstone samples from that site (Underwood, 2002, in press) are similar to the values reported here for the inferred accretionary complex beneath Hydrate Ridge. Smectite values for older (Pliocene) and finer-grained abyssal plain deposits beneath Astoria Fan are slightly higher with an average abundance of 42% and a maximum of 64% (Underwood, in press). Thus, there seems to have been a temporal change in clay discharge onto the floor of Cascadia Basin in addition to the spatial distribution of point sources.
Among the significant results of shipboard pore water analyses during Leg 204 are the chloride concentration profiles along an east-to-west transect from Site 1252 to Site 1245, which show progressive depletions at depth relative to ocean bottom water (Shipboard Scientific Party, 2003; Torres et al., 2004). This geochemical pattern is consistent with diffusion from a fluid source within deeper-seated sediments of the accretionary complex. In this postulated source area of fluid, a progressive increase in smectite dehydration is driven by an increase in temperature and reaction time with distance from and depth beneath the prism toe. Any effect of changing clay composition seems to be minimal.
In further support of the geochemical interpretations cited above, the XRD data show no evidence of in situ smectite-to-illite diagenesis at Hydrate Ridge. The expandability of the expandable clay ranges from 70% to 52% with a mean value of 64% and a standard deviation of 4% (Table T1). There are no trends in the expandability values as a function of depth or depositional age in the sampled lithologies (Table T1). Thus, the smectite appears to be mostly detrital in origin with no diagenetic overprint. This result is not unexpected given the shallow burial depths of most samples. Linear geothermal gradients range from a maximum of 0.061°C/m at Sites 1244 and 1246 to a minimum of 0.052°C/m at Site 1247 (Shipboard Scientific Party, 2003). If one assumes that the gradients remain linear to total depth at each site, this means that the maximum burial temperatures are approximately 24°C at Site 1244, 33°C at Site 1245, and 27°C at Site 1251. To initiate smectite-to-illite diagenesis, burial temperatures need to be within the range of 58°C to 92°C (Freed and Peacor, 1989). Thus, the likelihood of in situ smectite dehydration is remote.
On the other hand, the data presented here document the presence of enough smectite in the accreted sediments beneath Hydrate Ridge to support significant interlayer water release at depths >1000 to 1150 meters below seafloor (mbsf). Bulk-powder XRD data from DSDP Site 174 indicate that the average content of total clay minerals is ~42% by weight (relative to quartz + feldspar + calcite), so the amount of smectite in bulk mudstone averages ~16% with a maximum of ~26%. Given the regional geothermal gradient, the freshened fluids must be migrating from deeper than 1 km. A source at that depth is also consistent with data from analyses of hydrocarbons (Claypool et al., this volume) and strontium isotopes (Teichert et al., 2005).