Accurate determination of the total rate of heat loss from oceanic lithosphere has long been a goal of marine heat-flow surveys. "Calibration" of the theoretical heat flow vs. age relationship would improve estimations of deep-rock properties and of global heat loss using the known distribution of seafloor age and would provide accurate background reference values so that quantitative heat-flow anomalies can be calculated. And, of course, the lithospheric heat-flow budget allows estimation of local and regional hydrothermal budgets. But obtaining high quality data has been extremely difficult over young crust, because of the perturbing effects of hydrothermal circulation.
Shallow-seafloor heat-flow measurements were attempted across the Buried Basement Transect, but stiff seafloor sediments prohibited probe penetration. Temperature and thermal conductivity data from holes along this part of the overall ODP Leg 168 transect (Sites 1028, 1029, and 1032) were intended to provide accurate estimates of regional basement temperatures. This information, combined with the estimates of depth to basement based on seismic data, was intended to allow calculation of the regional lithospheric heat flux.
Another goal of drilling along the Buried Basement Transect, at shallow basement ridge Sites 1030 and 1031, was to investigate the nature of a high-chloride pore-water anomaly and fluid upflow through the sediments, documented with piston cores in this area. A final goal of operations along the Buried Basement Transect was to obtain a set of "reference" logs through the sediment section. These data would be of interest to Leg 168 scientists, as well as to scientists involved in other regional studies (e.g., Middle Valley, Cascadia Margin).
Scientific Results
The sedimentary succession within the Buried Basement Transect area includes three lithologic units (Fig. 12). Subunit IA is Quaternary in age and is composed of hemipelagic mud (clayey silt to silty clay), thin-bedded turbidites (silt to sandy silt), and thin- to thick-bedded sand turbidites. The base of Subunit IA occurs at approximately 81 mbsf in Hole 1028A and 96 mbsf in Hole 1029A. Sand turbidites are much more common in Hole 1029A. Hole 1028A, conversely, contains more thin-bedded, fine-grained turbidites. The total number of inferred gravity-flow deposits within Subunit IA decreases from 437 at Site 1028 to 388 at Site 1029. Subunit IB is Quaternary in age and is characterized by thin beds of silt and sandy silt intercalated with hemipelagic mud deposits. Erratic increases in the content of calcareous nannofossils generally result in a subtle lightening of color. The top of Subunit IB was not cored at Site 1032. The base of Subunit IB occurs at depths of 107.87 mbsf (Hole 1028A), 197.39 mbsf (Hole 1029A), and 272.47 mbsf (Hole 1032A). The hemipelagic mud deposits of Unit II are Quaternary in age. The thickness of Unit II is 24.6 m in Hole 1028A, 22.7 m in Hole 1029A, and 17.8 m in Hole 1032A. The sediment/basalt contact occurs at curatorial depths of 132.48 and 220.07 mbsf in Holes 1028A and 1029A, respectively. The basement contact is considerably deeper (290.29 mbsf) at Site 1032. Sediments immediately above the basement contact display irregular color variations because of fluctuations in primary clay mineral content and biogenic carbonate content and/or the hydrothermal formation of clay minerals and Fe-Mn oxides.
We did not subdivide sedimentary deposits from Holes 1030B and 1031A into lithologic units. Each hole contains approximately 41 m of interbedded hemipelagic mud, carbonate-rich mud, and silt to sandy silt turbidites. The hemipelagic deposits at both sites appear to contain unusually high contents of clay-sized material. Although both sites are perched above a prominent basement high, turbidites are significantly more abundant at Hole 1030B. Basalt was recovered only from Hole 1031A. The drillers defined basement contact at approximately 46.9 mbsf in Hole 1030B; the curatorial basement depth in Hole 1031A is 41.30 mbsf. The lowermost deposits at Site 1030 contain abundant nannofossil-rich intervals, but there are no obvious indications of hydrothermal alteration. Basal deposits in Hole 1031A, in contrast, include variegated carbonate-rich claystone, clay-rich siliciclastic mud, and two poorly sorted silt layers with irregular top and bottom contacts and considerable internal disruption. Alteration of these sediments by fluid flow near the basement contact seems likely.
Nannofossils are common to abundant in all Buried Basement sites. They are mostly well preserved, except in Hole 1032A, where they exhibit moderate to strong dissolution. In Hole 1028A, the top of G. lumina (130.60 mbsf) and the absence of Helicosphaera sellii in the sediment section indicates a basal sediment age younger than 1.55 Ma. In Hole 1029A, the determination of the top of C. macintyrei in the basal sediment (220.02 mbsf) suggests an age of 1.58 Ma. The occurrence of Gephyrocapsa caribbeanica within the section of Hole 1030B indicates the basal sediment of this hole is younger than 0.76 Ma. In Hole 1031A, the top of Reticulofenestra asanoi was recognized at 37.93 mbsf; however, the base of this species (1.15 Ma) was not observed in sediments below that depth. Thus, the age of the basal sediment should be younger than 1.15 Ma. The occurrence of P. lacunosa through the section of Hole 1032A suggests that the top of the recovered section (184.52 mbsf) is older than 0.46 Ma, and the absence of H. sellii in this section implies the basal sediment is younger than 1.55 Ma. The ages of the basal sediments of these holes are several thousand years younger than basement ages, indicating a hiatus between the formation of basalt and initial sedimentation. Sedimentation rates vary greatly between these holes and show a reduction from the eastern holes to the western holes.
Basement was recovered from Holes 1028A, 1029A, 1031A, and 1032A and consisted predominantly of aphyric, sparsely to moderately phyric, or moderately phyric plagioclase ± olivine ± pyroxene and olivine ± plagioclase basalt. All the rocks recovered are lithologically similar, consisting of pillow basalt. Subunits were identified on the basis of the presence of chilled margins and grain-size changes within the cored sequence, allowing the recognition and logging of individual cooling units. The pillow basalts at Holes 1028A and 1029A are sparsely to moderately phyric, containing 2%-5% plagioclase, 1%-2% olivine, and trace amounts of pyroxene, whereas at Holes 1031A and 1032A, the pillow basalts are aphyric, containing is less than or equal to 1% to a trace amount of olivine, plagioclase ± pyroxene phenocrysts. All basalts contain ~1%-3% vesicles with diameters is less than or equal to 1 mm.
Intense secondary alteration affects all rocks recovered from Sites 1028, 1029, 1031, and 1032. Secondary minerals occur as (1) vesicle or cavity linings or fillings; (2) coatings, fracture fillings, and veins; (3) replacement of phenocrysts and microphenocrysts; and (4) patches within mesostasis. The following secondary minerals were identified: clay minerals (saponite and celadonite), iddingsite, calcium carbonate, sulfides, talc, and zeolites. The alteration intensity varies from about 1% to 40% secondary phases. The intensity of glass alteration is relatively high at the Buried Basement sites, particularly at Hole 1032A. Alteration halos, 1-15 mm wide, occur commonly as dark borders on rock pieces or along clay veins. Haloes are distinguished by the presence of completely filled vesicles, in contrast to empty or saponite-lined and filled vesicles in the gray portion of the rock without halos. The vesicle filling material is clay, commonly iddingsite and/or celadonite. Hydrothermal veins, typically is less than or equal to 1 mm wide, include varieties of celadonite, saponite, aragonite, and zeolite minerals.
Alteration temperatures were low, probably less than 100°C, and possibly no higher than the present basement temperatures of 40°-58°C. All of the sites exhibit varying degrees of alteration, including varying degrees of oxidative alteration that required significant open seawater circulation. A subsequent stage characterized by carbonate, saponite, and sulfide alteration may represent relatively closed hydrothermal circulation.
Sediments from Holes 1028A and 1029A exhibit physical properties trends similar to those documented within sediments from the Hydrothermal Transition Transect. The mud-rich layers have lower magnetic susceptibility, thermal conductivity, bulk density, and P-wave velocity and higher porosity than the sand-rich layers. Porosity within muddy intervals decreases with depth, following standard compactional trends, whereas P-wave velocity increases. There is little difference between the grain density and natural gamma radiation of the mud layers from those of the sand layers. Cores from Holes 1030B and 1031A contain relatively little sand, yet do not appear to show any reduction in porosity with depth. These sediments also have lower magnetic susceptibilities and bulk densities than observed elsewhere along the transect at comparable depths. The physical properties of sediments from Hole 1032A are quite similar to those at the same depths along the Rough Basement Transect.
The concentrations of organic carbon in sediments from the Buried Basement Transect are relatively low and vary broadly within the same range as at the Hydrothermal Transition and Rough Basement Transects. Calculated C/N ratios indicate a predominantly marine origin for organic matter. The hydrocarbon gas content is low and dominated by methane at all sites. Concentrations of methane are significant only where dissolved sulfate is depleted, and concentrations fall to background levels near the seafloor and near basement.
Pore waters from sediments at both Sites 1030 and 1031 provide clear evidence of upward fluid flow. For example, alkalinity normally increases significantly with depth because of organic diagenesis; at Site 1030 it increases slightly before decreasing, and at Site 1031 alkalinity decreases monotonically with depth (Fig. 13). Upward advection seems to be slightly faster at Site 1031 than at Site 1030, as indicated by a contrast between sulfate and ammonia concentrations. It appears that there is a balance between the input of ammonia by bacterial breakdown of organic matter and the rate at which pore-water advection removes ammonia from the sediment pores. Chloride concentrations increase with depth, approaching or slightly exceeding values believed to have been present in Pleistocene glacial seawater. Concentrations remain elevated at depth and are associated with low Mg, K, and Si concentrations. The elevated chloride values are significantly greater than any others seen during Leg 168. The fluid moving upward from basement over the ridge at Sites 1030 and 1031 is geochemically more mature than the fluid found at adjacent sites, showing effects of extensive hydration reactions. These observations may indicate a deeper reaction zone and/or a longer residence time in basement.
Sites 1028, 1029, and 1032 form a west-to-east transect east of the buried ridge where Sites 1030 and 1031 were drilled. The pore waters from these sites show features typical of organic diagenesis (e.g., rapid decrease in sulfate and increase in alkalinity, ammonia, and phosphate with depth) and inferred mineral reactivity (calcium and magnesium minima) seen elsewhere, as well as evidence for partially evolved seawater in basement (as is apparent from the increase in sulfate and calcium and decrease in alkalinity with depth).
There are complications to this simple picture. Sulfate in near-basement pore waters is low at Site 1028 and forms a natural extension to the sequence seen at the Hydrothermal Transition sites (Fig. 14) (apart from the Sites 1030 and 1031 data). However, near-basement sulfate is higher at Site 1029 than at Site 1028 and this seems to be a break in the overall geochemical pattern. In contrast, calcium increases systematically, and values near basement at Site 1032 are considerably greater than those measured at a nearby basement outcrop. This observation most likely indicates that there is a reaction zone where Ca and Na are exchanged very close to the sediment/basement interface.
Sediment temperatures were measured at all sites along the Buried Basement transect. Along a transect from west to east, Sites 1030/31, 1028, 1029, and 1032 have extrapolated temperatures at the sediment/basement contact of approximately 40°, 51°, 59°, and 57°C, respectively. The 40°C basement temperature at shallow-ridge Sites 1030 and 1031 correspond to heat flow close to 1 W/m2. These high heat-flow values are consistent with the ridge being a fluid upflow zone, as indicated by the inorganic pore-water data. The remaining basement temperatures are considerably lower than would be expected if the crust was indeed fully sealed from exchange of hydrothermal fluids with the overlying ocean. In fact, upper basement temperatures at Sites 1029 and 1032 are remarkably close to the upper basement temperature of 61°-64°C at Rough Basement Sites 1026 and 1027, about 28 km east of Site 1032. Because of uncertainties in the actual location of the most hydrothermally active part of upper basement, these similar temperature values may indicate relative thermal homogeneity over a remarkable distance of nearly 50 km. If fluid flow actually occurs on such a crustal scale, this will require a substantial revision of many conceptual models of heat and mass transfer in the upper oceanic crust.
Logging was completed in Hole 1032A during the last days of scientific operations. Although the hole was drilled well into basement with the hope of obtaining high-quality measurements across the sediment/basement contact, hole fill and bridging problems prevented passage of the tools to this depth. Three tool strings (triple-combination, Formation MicroScanner [FMS]/sonic, and geochemical) were run successfully over the upper 275-280 m of sediment in the hole.