Heat flow data from Leg 204 are summarized in Table T3. Detailed discussion of data analysis and uncertainties is presented by Tréhu (this volume). The uncertainties in heat flow shown here include an estimated uncertainty of ±0.1 W/(m·K) in thermal conductivity, ±0.3°C in temperature, and ±2 km in depth. The data indicate that the thermal gradient is dominated by conduction and that the base of the GHSZ predicted by the thermal gradient and pore water composition is consistent with that indicated by seismic observations of BSR depth, although an offset of up to 15 m is allowable at several sites. Lack of curvature in the temperature vs. depth profiles indicates that heat transport through advection of pore fluid is below the thermal detection threshold of 1–10 cm/yr (Wang et al., 1993; Tréhu, Bohrmann, Rack, Torres, et al., 2003; Torres et al., 2004b). The apparent paradox of very low rates of fluid advection combined with observations of venting of gas bubbles, geochemical signatures of gas migration, and formation of abundant gas hydrate near the seafloor can be reconciled by models in which methane is transported through the sediments as free gas (e.g., Torres et al., 2004b; Milkov et al., 2005; Tréhu et al., 2004a; Liu and Flemings, 2006).
One of the striking results is the small variation in heat flow from site to site, including sites near the active vents near the summit of SHR as well as sites on the flanks and in the adjacent slope basin. The data are generally consistent with the model of Hyndman et al. (1993) and Wang et al. (1993), in which tectonically induced sediment thickening depresses the thermal gradient near the deformation front. This model predicts a thermal gradient in shallow sediments near the deformation front that is lower than the gradient predicted by a simple conductive model of a thick, sediment pile overlying young oceanic crust; farther landward, the thermal gradient will approach the gradient predicted by the conductive model as fluid is expelled from overpressured sediments. The thermal gradient can therefore be used to constrain large-scale fluid flow in the accretionary complex. At SHR, the apparent heat flow across the entire region studied during Leg 204 is only ~60% of that predicted by a conductive thermal model (Oleskevich et al., 1999). This suggests that fluid expulsion continues to the east beneath the mid and upper slope and continental shelf, consistent with observations of pockmarks and other vent structures in this region (e.g., Johnson et al., 2003).
Leg 204 also provided an opportunity to successfully test a new instrument that continuously recorded the temperature, pressure and electrical conductivity at the top of an APC core during the entire coring and core recovery process (Ussler et al., this volume). The data have the potential to lead to a better understanding of gas hydrate dissociation during core recovery.