To test the hypotheses discussed above, we originally proposed three primary drill sites extending to depths of 400–700 mbsf. Preliminary analysis of the 3-D seismic data confirmed the rationale behind these three sites but led to minor modifications of site locations. It also led to the addition of seven shallow-penetration (<260 mbsf) sites to sample the massive hydrate at the summit and to determine updip changes along subsurface horizons that appeared to be fluid pathways feeding the summit vents. Four of these additional sites were primary sites and three were alternates as described in the Leg 204 Scientific Prospectus; all were approved for drilling by Pollution Prevention and Safety Panel (PPSP). We drilled all sites except for alternate site HR1c, for a total of nine sites. Locations of the sites drilled are overlain on a map showing seafloor bathymetry in Figure F1C.

Leg 204 started with 3 weeks dedicated to LWD. For safety reasons, Site 1244 (proposed Site HR1a) was cored to 350 mbsf depth prior to LWD, allowing for sampling of all seismic facies that were to be drilled. We had approval from the PPSP to proceed with LWD prior to coring at all additional sites, or until the time allocated to LWD was expended, if no safety issues were encountered while drilling at Site 1244. Our objective was to use the LWD data to determine where to use time-consuming special downhole tools prior to coring. The LWD data turned out to be of excellent quality, and this strategy proved to be very useful.

Another novel aspect of this leg was the use of IR thermal imaging to systematically scan each core (at least those from within and near the hydrate stability zone) as soon as it was brought on board. Because hydrate dissociation is a strongly endothermic process, cold spots thus detected permitted us to quickly confirm what portions of the core contained significant amounts of gas hydrate. Experiments were designed to determine whether the original amount of hydrate present could be determined based on the change in temperature with time by calibrating the temperature record obtained by the IR scanner with concentration estimates obtained from closely spaced pore water chloride concentration sampling.