The question naturally arises whether the dolomite layers and similar lithified beds are recognizable in seismic profiles. Judging from the data available for Site 1082, the answer is ambiguous: some are, some are not, and many reflectors seen are not from lithified layers (Fig. 12). Reflectors arise where there are sudden changes in density or velocity, or both, and such changes can be provided by different changes in sedimentary facies. Dolomitic intervals represent thin layers with regard to the seismic wavelengths. For this conditions, reflection amplitudes are not directly proportional to the change in acoustic impedance but depend on geometry and continuity of the layers. Some of the dolomitic layers seem clearly recognizable as continuous, high-amplitude reflectors in the seismic profiles. How-ever, a dependency on layer thickness (reconstructed from logging data) could not be observed. Also, it is surprising that the band of lithified layers between 300 and 320 mbsf on the Walvis Ridges is not associated with more pronounced amplitudes.
We expected to see more phosphorite minerals than we found, based on previous reports of phosphorite abundance off southwest Africa. During Leg 175, phosphorites were noted only in sediments from the Congo Fan sites, as well as in the two southernmost sites (1086 and 1087). In both cases phosphorites may not be forming in situ, but rather were brought in from the shelf. Phosphorites are abundant on the shelf of Namibia, where sporadic winnowing of deposits is an important factor in concentrating the material (Calvert and Price, 1983).
It is unlikely that phosphatic minerals are absent in most of the sites drilled during Leg 175. The intensity of sulfate reduction in many of these sites indicates active oxidation of organic matter, which releases phosphorus. At every site drilled during Leg 175, concentrations of dissolved phosphate increase through the uppermost 50 mbsf, which is the zone of intense organic matter remineralization (Murray et al., Chap. 20, this volume). Below these relatively shallow maxima in dissolved phosphate, the concentration dramatically decreases, indicating the precipitation of authigenic phosphate minerals. The finely disseminated apatite is physically difficult to observe.
Pyrite (and other iron sulfides) were observed in smear slides at all sites, but most commonly at Sites 1075 and 1076 on the Congo Fan (Fig. 13). Because organic matter (for the reduction of sulfate) is not particularly abundant here compared with other sites (Fig. 4), we assume that the availability of iron is the decisive factor in generating high abundances of pyrite at these sites. Some of this iron may reach the site of deposition bound within clay minerals; much may be in the form of iron (hydr)oxide coatings, or flocs. Low iron sulfide abundances are typical for sites (e.g., 1083 and 1085) where organic matter contents are low, carbonate values are high, and influx of terrigenous sediments is reduced.
The growth of iron sulfide aggregates proceeds from finely disseminated iron monosulfides (Siesser, 1978), which take up sulfur to form microscopic crystals of pyrite, marcasite, and other sulfides. In the uppermost cores of a hole, the microscopic crystals within aggregates smear over some distance when the core is cut into sections, producing long streaks emanating from a pocket source (e.g., a burrow). Deeper in the holes, pyrite is present as small nodules and tubes. At Site 1085, the transitions are remarkably well expressed from disseminated pyrite to isolated silt-sized and sand-sized grains (starting at ~130 mbsf) and from grains to nodules as much as 1 cm in diameter (starting at ~430 mbsf). At Site 1086 (with a facies similar to that at Site 1085 but with even higher carbonate content), small, fine sand-sized pyrite grains are present below ~150 mbsf.
Large pyrite concretions were found at Site 1087, the southernmost site and the one with the highest carbonate content and with comparatively low sedimentation rates. Sand-sized pyrite grains are common throughout the sediment column. In Core 175-1087C-50X (453.5 mbsf) an ~3-cm-thick pyritic aggregate is present, as well as several other centimeter-sized pieces. Pyrite aggregates at this horizon are rimmed with iron (hydr)oxides, which suggests a return to a more oxidizing environment after pyrite formation. Biostratigraphic analysis indicates the presence of a substantial disconformity near this horizon, presumably a result of erosion of the nannofossil ooze during a time of increased dissolution.
It is not clear why carbonate ooze provides a good environment for making large pyrite aggregates. Perhaps differences in the chemistry of microenvironments are more pronounced here than in typical hemipelagic sediments, and sites of iron sulfide precipitation are less abundant. A burrow within carbonate ooze, with relatively high sulfate reduction potential from organic matter within, may draw iron from a larger space than a burrow within hemipelagic sediments where reducing sites are more common throughout the sediment. Large pyritic aggregates are well known from both mudstones and limestones in the Jurassic sediments of southern Germany. Such aggregates are commonly associated with microfossils (the organic matter associated with these microfossils was responsible for sulfate reduction during early diagenesis).
Greenish small grains, apparently of authigenic origin, were observed at several sites and were reported as "glauconite" (Fig. 14). The highest values were observed at Sites 1076 and 1077 on the Congo Fan. This mineral may also be present at Site 1075, but perhaps less conspicuously. Glauconite is reported from shelf environments in regions of high productivity (to account for reduction of iron), of availability of (iron-rich) terrigenous sediments, and of low sedimentation rates. Glauconite-cemented fecal pellets are common in shelf and margin surface sediments off the Congo and Angola (Wefer and Shipboard Scientific Party, 1988). The glauconite observed at the Congo Fan sites may represent redeposition, rather than in situ growth, which may be true for the other sites as well. If so, one would expect increased abundance of glauconite during times of low sea level. At this point, no data are available to test this hypothesis.
The seismic profiles off Angola show numerous indications of vertical tectonics caused by salt diapirism and associated vertical migration and trapping of fluids and gas. Because of the fine overall grain size of the hemipelagic sediments and their low permeability, fluid/gas migration is probably restricted to areas of tectonic fracturing. In these areas, zones of high reflection and scattering amplitudes (Fig. 15) seem to indicate the presence of trapped gas at depths as shallow as 250 m. A typical continuous bottom-simulating reflector (BSR) with reversed polarity has been observed only within a restricted interval between 200 and 300 mbsf. The upper boundary generally follows the topography. BSRs off Peru and elsewhere in methane-rich regions are commonly interpreted as denoting the lower boundary of clathrate stability (Shipley et al., 1979; Kvenvolden and Kastner, 1990); that is, methane-rich water ice (with a structure different from that of regular ice). Because the region off Angola is known to be rich in hydrocarbons (it is being actively explored by oil companies), we assumed that the observations regarding an apparent BSR indicated varying abundances of clathrates within their stability field. With a gradient near 45°C/km, we should expect instability below 600 mbsf at Site 1077, for example. As previously mentioned, the depth in the section with the strong reflectors is generally shallower than 600 mbsf. However, nothing is known about lateral variability in heat flow caused by fluid/gas migration in the area.
Because of the strong reflectors and their possible association with hydrocarbons, the safety panel quite generally restricted drilling to the uppermost 200 m of the sediment column. The panel took particular care to have us avoid sites where the reflectors are especially strong, because gas pockets might be found trapped below clathrate-cemented (and therefore relatively impermeable) deposits. We were surprised, however, not to find any indication of the presence of clathrates at any of the sites drilled during Leg 175.
During the entire expedition, we looked for unusually strong gas development from the release of gas from clathrates. Such release can be rapid and pose serious problems during handling of cores on deck (Suess, von Huene, et al., 1988). Although gas release was vigorous in many places (Meyers et al., Chap. 21, this volume), it did not seem to exceed what might be expected from normal degassing of depressurized, warming, gas-rich sediments. When clathrates melt, freshwater is added to interstitial waters (Hesse and Harrison, 1981). We found no evidence for decreases in salinity or chlorinity corresponding to an addition of freshwater from melting of clathrates in locations where clathrates might be expected (Murray et al., Chap. 20, this volume).
Our negative findings do not preclude the presence of clathrates on the continental slope off Angola, or elsewhere off southwest Africa. We have to take into account that we did not drill too deeply because of safety concerns. There could be clathrates deeper in the sediment. Certainly, methane is present in great abundance in the sediments recovered during Leg 175, and the conditions for formation of clathrates would seem to be favorable (unless dominance of CO2 in the gases within the sediment prevents or slows the process). As discussed earlier, we expressly avoided areas near "cloud structures" and higher reflection amplitudes where clathrate might be expected to be most abundant.