The second working area of Ocean Drilling Program (ODP) Leg 175 was located on the marginal plateau off the coast of Angola at 12°S. The area lies between the Congo River and the Angola-Namibia upwelling cell. Major scientific targets are based on the assumption that the location represents a (hemi-) pelagic depositional regime with respect to biogenic sediment input, which is still under the influence of the Benguela Current, but is typical for open-ocean conditions. The sites are suitable to monitor variations in the Angola-Benguela Front. Sedimentation rates are expected to be generally lower than in the Congo Fan area because of reduced coastal upwelling and the lack of large river systems in the area. To explain the strategy of the site survey (Bleil et al., 1995), a short summary of the evidence regarding the expected sedimentary structures at this part of the continental margin is given.
The regional morphology is controlled by a rift basin, which developed during the early opening of the South Atlantic. In Aptian-Albian times, a shallow-water basin was repeatedly flooded, and evaporitic layers of altogether several hundred meters in thickness accumulated (Baumgartner and van Andel, 1971; Leyden et al., 1972; Emery et al., 1975; Emery and Uchupi, 1984). The subsequent phase of seafloor spreading disrupted this basin. A major portion between the equator and ~13°S remained at the African continental margin, forming a plateau in water depths shallower than 2500 to 3000 m. Its seaward boundary, the Angola Escarpment, can be traced as a prominent morphological and structural feature (Fig. 3).
Soon after deposition of the salt layers, buoyant forces initiated vertical movements, which are also observed in the coastal basins (Brice et al., 1982; Emery et al., 1975). They continued during the Late Cretaceous and Cenozoic, when several kilometers of sediments were accumulated on the continental margin and the plateau. This salt tectonism has shaped the morphology of the area and dominates the regional sedimentation pattern. Numerous diapiric structures in the relict rift basin were imaged and analyzed in some detail by the seismic surveys of Baumgartner and van Andel (1971), Leyden et al. (1972), and Brice et al. (1982). Intensive salt tectonism is associated with the diapirism. Because of the short average distance between diapirs, the identification of suitable potential drill sites with undisturbed sedimentary sequences was rather complicated. Also, because of the hydrocarbon potential associated with black shales and salt, the area became of great interest to the oil industry. In recent years, intensive seismic surveying has started; however, given that concessions are still being negotiated, this information is not yet accessible to the public.
The preliminary site proposals in this region were based on seismic Line 44 of Emery et al. (1975), but the data quality was not sufficient to identify fine-scale sedimentary structures and to prove the absence of disturbances. Based on recent bathymetric data, the relatively complex topographic structures around this profile were avoided, and the survey area was moved south to ~12°S. All drill sites proposed here (Fig. 3) are positioned on Line GeoB/AWI 93-015 (Fig. 4), for which six crossings are available. Additional high-resolution seismic profiles were recorded across Deep Sea Drilling Project (DSDP) Sites 364 and 365 (Bolli, Ryan, et al., 1978) to establish a seismostratigraphic correlation with the former drilling results.
Bathymetric data acquired with the Hydrosweep swath sounder system show that water depths in the selected area are <2000 m. West of ~12°40'E on the outer plateau, and less pronounced on the inner plateau to ~12°50'E, the morphology is irregular with slightly elongated north-northwest/south-southeast trending features of 50- to 100-m amplitude. Salt diapirism, which reaches close to the surface and thereby also affects the topography, has created this sequence of peaks and troughs. As the basins between the diapiric structures are only ~10 km wide, appropriate locations for first priority paleoceano-graphic drill sites could not be identified. East of 12°50'E, the upper continental slope is smooth.
Altogether, 16 seismic lines were recorded in the Mid-Angola Basin area, with six crossings at potential drill sites in water depths between 550 and 1460 m along Line GeoB/AWI 93-015 (Fig. 3). Line GeoB/AWI 93-014 was shot to link the drilling results of DSDP Leg 40 (Sites 364 and 365) with the new findings of this leg. It turns out, however, that seismic characteristics at the DSDP sites and in the working area were different. Accordingly, it must be concluded that the depositional pattern is different, probably because of significantly enhanced terrigenous input in the Leg 175 working area. Recordings at Line GeoB/AWI 93-017 revealed a mostly disturbed sedimentary section, indicating a mass flow probably originating from the shelf area in the south. On Line GeoB/AWI 93-029, it could be confirmed that these mass-flow deposits extend into the Angola Basin, and no hemipelagic sequences suitable for drilling were identified.
Vertical salt movements (the predominant tectonic process in the area) have directly affected the regional morphology and sediment deposition. In the western part of the plateau, diapiric structures nearly reach the surface. The sediment structures on top and on the flanks are deformed, indicating postdepositional salt tectonics. Figure 4 shows the eastern portion of Line GeoB/AWI 93-015, which lies east of the diapiric zone (Baumgartner and van Andel, 1971) where the influence of salt tectonism in the surface sediments is of minor importance. Only smooth changes in near-surface reflection patterns are observed on this line, indicating the absence of significant mass-flow deposition.
For safety and pollution considerations, only two of six proposed sites were approved, and penetration was limited to 200 and 120 m, respectively. This restriction was based on the analyses of reflection amplitudes, which appear to be very high and sometimes reveal some scattering in the surface sediments. Furthermore, numerous pockmarks downslope at proposed sites MAB 5A and 5B strongly indicated fluid and/or gas migration. Besides a salt diapir around common depth point (CDP) 7500 and further evidence for diapirs on crossing lines north and south of Line GeoB/AWI 93-015, indications for shallow gas were found between CDPs 7800 and 8050 and for deeper gas accumulation between CDPs 6300 and 6600 and CDPs 6800 and 7200. Seismic reflectors of pronounced lateral variability in amplitude exist in the vicinity of all proposed sites, but were less disturbed for the uppermost two sites.
A subdivision into seismostratigraphic units here is less straightforward than in the Lower Congo Basin. Reflectors in the uppermost 300 ms two-way traveltime (TWT) at Site 1079 and 500 ms TWT at Site 1078 appear to be very coherent and can be easily correlated between the sites. Beneath this zone, which could not be reached by drilling, a nearly transparent interval from 200 ms TWT (CDP 6300) to 500 ms TWT (Site 1078) appears.
The uppermost band of reflectors can be subdivided into two seismostratigraphic units, which were reached by drilling and will be described for the vicinity of the two drill sites.
The uppermost Unit 1 of 120 to 160 ms TWT reveals a coherent reflection pattern with a high-amplitude reflector at the surface and the base, which, however, decreases in amplitude upslope. Within this unit, amplitudes are generally weak with minor lateral changes between the sites, except for a stronger reflector at 100 ms TWT at Site 1078, which is less pronounced downslope at Site 1079. The distance between individual reflectors varies, indicating minor slumping units.
Beneath the base reflector of Unit 1, amplitudes are generally weaker. Some reflectors are interrupted by transparent or disturbed zones. The thickness of the unit varies from 60 to 100 ms TWT, and its base is again defined by a stronger, continuous reflector. Beneath it, a local amplitude anomaly, which is associated with a thickening, is found between the two sites. Beneath seismostratigraphic Unit 2, scatter increases significantly and is interpreted as a potential indicator for shallow gas leading to the restriction in drilling depth. Alternatively, the scatter could be caused by irregular surfaces of slump units.
The basic assumption that the sedimentary sequences at the margin are not dominated by mass-flow deposits and that deposition is generally hemipelagic, as was found for the Lower Congo Basin, could be confirmed by the seismic survey. Although a sufficient number of sites were proposed to study a transect in water depth and distance from the shelf, safety considerations limited the drilling program to the shallowest sites at 432 and 738 m water depth, respectively.
Site 1078 was located in 426 m water depth at the eastern end of the survey area on Line GeoB/AWI 93-015 (CDP 8550). Figure 5 shows a 10-km-long seismic section of Line GeoB/AWI 93-015 across Site 1078. Weak reflectors were enhanced by normalization of amplitudes to a constant value. Therefore, the general characteristics described for the seismic units do not appear as pronounced as they would in a color-coded section. But the figure confirms the separation into two units within the proposed drilling depth range. Regarding distance from the shelf, the site is located where most of the sedimentary units are thickest, providing the highest sedimentation rate in the area within the upper 200 m. However, it cannot be excluded that the thickening is at least partially attributed to thin mass-flow deposits.
Significantly disturbed and scattering intervals are found beneath 200 m, and a sharp transition in reflectivity at a depth of 530 ms TWT. The site was moved from the originally proposed location to avoid the amplitude anomaly between CDPs 8370 and 8500 at a depth of about 300 ms TWT.
Figure 6 shows a close-up of the seismic section, plotted against sub-bottom depth for a sound velocity of 1500 m/s, for a 1-km-long interval in the vicinity of the drill site. Seismic reflectors are compared with the smoothed GRAPE density core log for Hole 1078C. Seismograms are plotted as wiggle traces with gray-scaled amplitudes as background. Laminated intervals are indicated by a thick arrow, and "D" indicates documented dolomite layers or dolomitic clays. Small arrows are plotted at depth where one or a few anomalously high GRAPE readings were found in any of the four holes (1800 to 2400 kg/m3). These density anomalies have different origins: stones at 8-9 mbsf, silty bands at 38.7 or 43.8 mbsf, an (ero-sional) contact at 55.1 mbsf, or dolomite at 82.9 mbsf. Depending on their thickness and distribution, they are potential candidates for seismic reflectors, but some of them might only be visible in ultra-high-resolution seismic data from echosounders. Selected reflectors can be directly correlated with sharp changes in the density log, but the seismic wavelength is not short enough to uniquely distinguish individual reflectors from interference patterns. The probably compaction-related density gradient in the upper 10 m results in a significant decrease of the surface reflection amplitude, which, in turn, would affect the evaluation of deeper reflector amplitudes. Such a gradient was not observed in sediments from the Lower Congo Basin.
Because of the lack of logging data, it is not possible to state at this point whether high amplitudes are generally associated with levels of dolomitization. There is no direct correlation between the presence of dolomitized layers in the sediment and reflection amplitudes. This would imply that either the distribution of dolomites is nodular and does not affect the seismic wave significantly, or the thickness is too small to generate a high-amplitude echo. A regional analysis of amplitudes in the vicinity of the drill sites is intended to further elucidate the presence and distribution of dolomitization. Where reflectors are coherent and of constant amplitude, dolomitized layers must be assumed to have a major impact on migratory processes of gas and fluid in the sediment, as well as the trapping of shallow gas and hydrocarbons. The low reflectivity beneath 530 ms TWT can be explained only by the absence or dissolution of dolomitized layers, a drastic overall change in lithology, or an overprint of reflection amplitude caused by the presence of gas and/or fluids. So far, no evidence can be derived from the shallow drill holes to explain the observed variations.
Laminated intervals appear as low-amplitude, but consistent, reflectors in the multichannel seismic data and might be used as stratigraphic markers. The digital echosounder can be used in shallow depths to analyze the regional distribution of this sedimentation event. Calculation of synthetic seismograms from calibrated and spliced GRAPE data sets is required to carry out more detailed analyses. The required thorough editing and quality control can be carried out only on shore.
Site 1079 is located downslope of Site 1078 in a water depth of 738 m and at CDP 8150 of Line GeoB/AWI 93-015. Figure 7 shows a 10-km-long section of seismic Line GeoB/AWI 93-015 close to the drill site. Drilling was approved down to 120 mbsf, as discussed above. The transition between the two seismostratigraphic units can be reached at ~120 ms TWT. The upper sedimentary sequence appears ~25% thinner than that at Site 1078. At Site 1079, disturbances in greater depth are generally more pronounced and reveal hyperbolic echoes. The strong reflector denoting the transition between higher and lower reflective zones is overprinted by seismic scatter down-slope, which could suggest shallow gas accumulation.
The 1-km-long close-up section near the drill site is shown in Figure 8. Seismic reflectors are compared with the wet bulk density log, derived from index properties measurements (see "Physical Properties" section, "Site 1079" chapter, this volume). Numerous reflectors could be assigned to sharp changes in physical properties, but additional reflectors might derive from the interference of seismic waves at closely spaced variations in acoustic impedance. In contrast to Site 1078, evidence for a lithified dolomitic layer was not found at Site 1079. An accumulation of silty bands was reported at a depth of ~85 mbsf, which might be assigned to a regionally correlatable reflector. The transition between seismostratigraphic units can be explained by a generally decreased scatter in the data, which leads to subdued reflection amplitudes and reduced interference.
A correlation of core data with seismograms awaits further editing of MST measurements and calculation of synthetic seismograms based on cleaned and spliced data sets.