Gas hydrates are concentrated in sediments of the continental rise off the southeastern United States (Paull and Dillon, 1981), particularly in the broad Blake Ridge, which extends seaward approximately perpendicular to the general trend of the continental margin and continental rise (Fig. 1). The Blake Ridge is believed to be a giant sediment drift deposit (Tucholke et al., 1977) that has accumulated since about Oligocene time (Dillon and Popenoe, 1988). Seismic profiles collected over the Blake Ridge area and the continental rise are marked by a large-amplitude reflection, the bottom-simulating reflector (BSR), which parallels the seafloor. Most studies explained the BSR by the significant impedance contrast between sediments containing gas hydrates lying above sediments devoid of gas hydrates and/or filled with free gas. Results from previous Deep Sea Drilling Program surveys in this area (Leg 11) only inferred that the BSR corresponds to the base of the gas hydrate stability zone (Shipboard Scientific Party, 1972). Gas hydrate-bearing sediments were first observed during Leg 76 of DSDP Site 533 on the outer part of the Blake Ridge, confirming the existence of gas hydrates in this area (Kvenvolden and Barnard, 1982).
Leg 164 purposely located a transect of sites to penetrate the Blake Ridge BSR, with the major objectives to investigate its nature, to explain its noticeable changes in reflection strength in terms of lateral variations in hydrate or free gas concentration, and to determine its relationship to the presence and amount of gas hydrate and/or free gas in the sediment pores. The nature of the sedimentary section in this area is very uniform, as demonstrated by sedimentary and mineralogical analysis performed on board; only small lithologic variations occur within the uppermost 150 m (Paull, Matsumoto, Wallace, et al., 1996). Indirect techniques for estimating gas hydrate concentrations employed the chloride ion content (Cl-) of pore water as indicator. During gas hydrate formation, water and methane are removed from the pore waters, leaving residual pore waters increasingly saline. Over time, these locally elevated chloride concentrations diffuse away. When gas hydrates decompose in sediments during drilling, they release water into the pore space, freshening the pore waters. Pore-water profiles at the three sites revealed that more than 8% of volume of some samples, and on average more than 1% of the samples from a 200- to 250-m-thick zone above the BSR, were filled with gas hydrate. Beneath 450 mbsf, chloride values were nearly constant (Paull, Matsumoto, Wallace, et al., 1996), interpreted to signify that gas hydrate is not present. Gas volume determinations (deploying a Pressure Core Sampler) at 17 depths at Sites 995 and 997, revealed that the pore space contains more than 12% gas bubbles in the free gas zone beneath the BSR, assuming gas exists in oversaturated pore water. This result suggests that the volume of gas in the free gas zone has been underestimated in the literature for this area (Dickens et al., 1997).
In this paper, we applied tomographic seismic techniques on multichannel seismic reflection data (MCS) to reconstruct the acoustic velocity distribution associated with the Blake Ridge BSR along Sites 994, 995, and 997. This technique provided a qualitative approach to determine the presence of gas hydrates and free gas in the surveyed area. Moreover, we propose a theoretical method for estimating the amount of gas hydrates and free gas in marine sediments. The results of these two independent methods are then compared to those obtained using direct (downhole logging velocities, VSP profiles, and core measured index properties) and indirect methods (chloride contents) carried out during Leg 164 on the three Blake Ridge sites. The consistency of the obtained values shows the validity of the method that can be used to extrapolate elastic parameters for velocity estimations in porous media.