Site 995 was the first site on Leg 164 at which drilling penetrated below the base of the gas hydrate stability zone and through a strong BSR (Fig. 5). The site is located on the southern flank of the Blake Ridge, 3.0 km northeast of Site 994, and within the same stratigraphic interval as Site 994. However, a strong BSR is present at Site 995 at 0.53-s sub-bottom, which is not observed at Site 994 (Fig. 5). Sites 994 and 995 are coupled sites that were intended to establish the nature of the BSR and to understand the causes of profound differences in acoustic characteristics (Fig. 5) within essentially the same sedimentary sequence.
At Site 995, we recovered a 700 m-thick sedimentary sequence that is dominantly composed of clay and nannofossils (Fig. 6). Three major lithologic units were identified, primarily based on downhole variations in carbonate contents and diatom and nannofossil abundances. Unit I (0-13.4 mbsf; Holocene to upper Pleistocene), comprises foraminifer-bearing nannofossil-rich clay and nannofossil clay, with carbonate contents as great as 50 wt% CaCO3. The sediments in Unit II (13.4-131.9 mbsf; upper Pleistocene to upper Pliocene) contain abundant diatoms and have lower nannofossil contents (average 10%) than those in Unit I. The upper part of Unit II (Subunit IIA) contains foraminifer-rich winnowed layers that indicate deposition by contour currents. Unit III (131.9-704.6 mbsf) extends to the bottom of the hole and comprises an upper Pliocene to upper Miocene sequence of monotonous diatom-bearing nannofossil-rich clay, and nannofossil-bearing clay and claystone, with average CaCO3 contents of 15 wt%.
Nannofossil biostratigraphy indicates that the sequence recovered at Site 995 is mostly continuous except for a short hiatus or gap detected within the uppermost Miocene (Fig. 7). The sedimentation rates for the Quaternary and lower Pliocene sequences are almost identical to those at Site 994. However, the rate for the upper Pliocene (110 m/m.y.) is about 10% higher than at Site 994. The upper Miocene sequence is estimated to have been deposited at a rate of approximately 260 m/m.y. The age of the oldest sediments cored at Site 995 (704.6 mbsf) is estimated to be 6.1 Ma. A magnetostratigraphy was determined for Hole 995A, despite remanences of <0.5 milliampere/m (mA/m), and major chron boundaries were recognized as follows: C2An/C2r (Gauss/Matuyama), 125-145 mbsf; C2Ar/C2An (Gilbert/Gauss), 295-320 mbsf; C3n/C2Ar, 320-350 mbsf; C3r/C3n, 545-580 mbsf; and C3An/C3r (Anomaly 5/Gilbert), 620-640 mbsf.
Concentrations of methane in headspace gas samples increase from 0.021 ml/kg of sediment near the sediment-water interface to a maximum of 114 ml/kg at a depth of 42.7 mbsf. Headspace methane contents then steadily decline to about 10 ml/kg near 240 mbsf and remain at 1 to 10 ml/kg throughout the rest of the hole. Methane- to-ethane ratios decrease with depth and reach a minimum of 146 at 699.4 mbsf. Higher molecular weight hydrocarbons are present in <10 ppm concentrations from 171.2-699.4 mbsf, and heptane occurs in the lower sections of Hole 995A, suggesting a small amount of thermally mature gas has migrated to the site. The total organic carbon contents of the sediments are near 1%, and the organic matter is immature, containing both terrestrial and marine components.
Interstitial-water geochemical data from Site 995 are remarkably similar to those found at Site 994 (Fig. 8). In particular, anomalously low values of interstitial-water chloride concentrations occur at the same depths at both sites. The low chloride values (as low as 466 mM) occur between 195 and 450 mbsf, coincident with the zone of anomalously low sediment temperatures measured on the catwalk after recovery, indicating that gas hydrate recently decomposed in these cores. These results imply that sites 3 km apart possess similar vertical distributions and amounts of gas hydrate.
Eleven WSTP runs from 78.7 to 200.2 mbsf were made at Site 995. The recovered samples contain 0.1%-94% interstitial water, with the proportion of interstitial water decreasing with increasing depth. After correction for the effects of dilution by borehole water, the WSTP water samples indicate that in situ chloride concentrations generally are comparable to those measured for water squeezed from whole-round core samples recovered from corresponding depths.
The PCS was successfully deployed 11 times at Site 995. Gas samples taken from the PCS are methane, with approximately 1% CO2, that evolves from the tool at pressures below 500 psi (at 0°C). The amount of gas recovered from certain cores exceeds that expected from methane saturation of the interstitial waters at in situ pressures. Because some of these same cores were recovered from above the base of gas hydrate stability in the zone associated with the erratic interstitial-water chloride concentrations, the "excess" gas is probably derived from decomposition of methane hydrate.
Physical properties data for sediments from Hole 995A are nearly coincident with those found at Site 994. The data do not reveal major differences that could account for the remarkable lateral variability in the strength of the BSR between the sites. The data also show a 390-m-thick interval (220-610 mbsf) in which the wet-bulk density values are constant as a function of depth, an unusual observation in sediments undergoing normal compaction.
Rock magnetism defines a trend of magnetite-greigite-pyrite conversion in which greigite develops within the first 20 mbsf in response to bacterial oxidation of organic material and is progressively reduced to pyrite downhole. Below this a second generation of greigite extends downward to ~260-300 mbsf, approximately coinciding with the zone of high gas hydrate concentration inferred from interstitial-water chloride values. This sequence of reduction steps is similar to that documented at Site 994.
Twenty in situ temperature measurements were attempted at Site 995 using the Adara and WSTP tools and a new prototype temperature probe , the Davis-Villinger temperature probe (DVTP) (Fig. 9). Based on 15 successful deployments of the temperature tools, the geothermal gradient is estimated at 33.5 ± 0.9 °C/km between 0 and 381 mbsf. Taking into account vertical variations in thermal conductivity, the average heat flow from 0 to 381 mbsf is 34.2 ± 1.7 milliwatt/m2 (mW/m2), which is 35% lower than previous measurements from this region. Preliminary extrapolation of the thermal gradient to 440 mbsf, the depth of the BSR at Site 995, yields a temperature of 18.3°C. This is well within the experimentally determined pressure-temperature stability field for methane hydrate.
A complete suite of wireline logs (natural gamma, resistivity, sonic, neutron porosity, lithodensity, geochemistry, Formation MicroScanner [FMS], and the experimental shear wave tool) was run in Hole 995B from 130 to 630 mbsf. Preliminary analysis of the acoustic and resistivity logs (Fig. 10 and 11) shows a pattern similar to that at Site 994, with low acoustic velocities (~1600 m/s) in the top of the hole that begin to increase at 220 mbsf to a maximum of 1900 m/s at 450 mbsf before decreasing again to 1600 m/s at 600 mbsf. The zone from 220 to 450 mbsf has higher resistivities than are found either above or below. The resistivity and acoustic velocity measurements are consistent with the presence of hydrate in the zone from 220 to 450 mbsf and with the presence of gas bubbles in the section below 450 mbsf.
VSPs were conducted at depths of 144-664 mbsf during two lowerings of the three-component WHOI borehole seismometer in Hole 995B. Walkaway VSPs were shot by the Cape Hatteras and recorded at eight depths from 176 to 680 mbsf. A stacked record section shows clear downgoing first arrivals and upgoing reflections, including a strong reflection from the BSR. The intersection of the downgoing first arrival and the upgoing BSR reflection indicates that the BSR is located at 440 ± 10 mbsf. A preliminary P-wave velocity model (Fig. 12) shows that compressional velocity reaches a maximum of 1850 m/s at 400 mbsf. Velocities decrease below this depth, reaching a minimum of about 1550 m/s at 590 mbsf, suggesting the presence of gas bubbles.