The third working area of Ocean Drilling Program (ODP) Leg 175 was in the northern part of the Angola-Namibia coastal upwelling cell (Schell, 1970; Nelson and Hutchins, 1983; Stramma and Peterson, 1989), which also represents the northernmost extent of the modern trade-wind-driven upwelling in this area. The region is located at ~17°S between the Mid-Angola Basin Sites 1078 and 1079 and the Walvis Ridge.
The early opening of the South Atlantic in the Walvis Ridge area is supposed to be of Late Jurassic age according to Sibuet et al. (1984). It was accompanied by crustal thinning, development of basaltic wedges of seaward-dipping reflectors, and blockfaulting. In Aptian–Albian times, an eastward ridge jump occurred (Cande and Rabinowitz, 1978; Kumar, 1979), which placed part of the original African continental margin in this area onto the South American Plate. Accordingly, the continental margin closest to Site 1080 is rather narrow with steep slopes, resulting in downslope sediment transport by slides, debris flows, and turbidity currents (Embley and Morley, 1980). The morphology is further complicated by sedimentary tectonism and giant sediment slides. Thus, a well-defined pelagic depositional pattern could be identified only in a few places at this site.
As multiple major and minor slides are documented in the survey region, highest priority was given to the identification of areas with an apparently undisturbed sedimentation. Because of the generally steep slopes, the most promising depth ranges were expected in the mid and lower stretches of the continental margin. The initial seismic line was oriented along an earlier Parasound profile (Meteor Cruise M20/2), which had revealed undisturbed near-surface sediments in a number of places. Only two suitable drill sites could be identified in water depths between 2200 and 3000 m and were crossed by additional seismic lines. All areas shallower than 2200 m and deeper than 3000 m were found to be affected by intensive slumping.
The seismic characteristics are summarized here for the cored interval at Site 1080. The bathymetric records confirm the complex nature of the depositional environment, and it became clear from the combined Hydrosweep and Parasound echosounder data sets that very few potential drill sites may be found in the region. Some locations are apparently better protected from mass movements by the local topography, perhaps featuring an underlying basement high.
The selection of drill sites in the Angola-Namibia upwelling region also was based on digital Parasound data. They allowed a clear distinction between chaotic and regular sediment structures on a meter scale. Seismic data commonly showed a continuous layering, whereas the Parasound system imaged a microtopography with rough, scattering surfaces. The signal penetration, on the order of 50 to 75 m, was generally lower than in the survey areas farther north, possibly because of lower clay and higher silt and sand concentrations, which may indicate higher proportions of coarse-grained shelf materials transported downslope. The selected drill sites showed regular seismic patterns of layered sediment structures, which should represent a continuous and primarily hemipelagic sedimentation, possibly intercalated with turbidites and thin mass-flow deposits. Occasionally, small hyperbolic echoes suggest undulating surfaces or a rough microtopography that cannot be distinguished because of the limited resolution of the seismic data. Because the frequency band of the seismic source, as well as the recording system, extends to several hundred hertz, a comparatively detailed image of the sedimentary structures is provided. If only the frequency band below 100 Hz had been considered—as for usual multichannel seismic surveys—the reflection pattern would appear much more regular.
Altogether, six seismic lines were shot in the working area (Fig. 2; Bleil et al., 1995), with Line GeoB 93-031 lying slightly north of Line 40 of Emery et al. (1975), where the originally proposed site was located. It was obvious from these old data that new locations with less disturbed or undisturbed sedimentary sequences had to be found to carry out high-resolution paleoceanographic studies. In addition to these lines, seven seismic lines were shot across Deep Sea Drilling Project (DSDP) Site 530 (Hay, Sibuet, et al., 1984) in the southernmost Angola Basin and DSDP Site 532 (Hay, Sibuet, et al., 1984) on the Walvis Ridge with equally high resolution for a direct comparison with the working area (Fig. 2). These data provide a basis for a seismostratigraphic framework making use of pronounced similarities of the seismic characteristics, which is under development. The two proposed sites SAB-1 and SAB-2 lie on Line GeoB/AWI 93-030 (Fig. 3), which also revealed large vertical scarps from slope failures (e.g., at common depth points [CDP] 3250, 3750, or 4000), or slump deposits at the surface from CDP 1600 to 1750.
Penetration of the seismic signal was generally much higher here than at the northern sites, indicating the absence of attenuating gas charges and allowing the development of a seismostratigraphic concept down to 8 s two-way traveltime (TWT). Coring was originally planned for 600-m penetration, but only 52 m of the sediment column were recovered from Site 1080 because of the presence of lithified dolomitic layers. Therefore, sediments were retrieved only from the uppermost seismic unit, which is characterized by numerous continuous, distinct parallel reflectors of high amplitude, partly disrupted by minor vertical faulting (Fig. 3). The base of the unit is marked by a prominent, discontinuous high-amplitude reflector, which is associated with a significant decrease in overall reflection amplitudes and signal frequency at depth. In contrast to all deeper lying units down to 3 s TWT sub-bottom depth, this unit shows a remarkable thickening toward the coast from ~100 m at the western end of Line GeoB/AWI 93-030 to ~450 m. This is not typical for a depositional style dominated by turbiditic sediments, but for hemipelagic sediment input with a source close to the shore. Therefore, biogenic sediments resulting from coastal upwelling and terrigenous sediment from the shelf (and ultimately from the Kunene River) are tentatively identified as the dominating sediment sources.
Compared with proposed site SAB-1 (Fig. 3), a sediment package of ~50 m thickness is missing at Site 1080. It could not be determined from the seismic data whether this package at SAB-1 was derived from a slumped deposit or overlying normal sediments, or whether it is truly missing at Site 1080. The local and small-scale disturbances in the sedimentary sequences seen in the record will presumably complicate interpretation of the drilling results. Nevertheless, the selected sites are among the very few places in the area with a reasonably clear, parallel reflector pattern, which should greatly facilitate the reconstruction of the regional sedimentation history.
Site 1080 is located in 2766 m water depth at the northeastern end of the survey area on Line GeoB/AWI 93-030 (CDP 2156). Figure 4 shows a 10-km-long seismic section of Line GeoB/AWI 93-030 across Site 1080. The sedimentary sequence appears largely undisturbed, with some minor faulting. Some horizons with diffractions indicate rough interfaces (e.g., top of slump units). Most reflectors show a pronounced continuity, as is known from hemipelagic or turbiditic sequences. Reflectivity decreases beneath 300 ms TWT, with a gradual transition to a mostly transparent unit. The origin of the contrast between this transparent unit and the high-amplitude reflections in the upper unit is unknown, but it must represent significant differences in physical properties.
Figure 5 shows a close-up of the seismic section over a 1-km-long interval near the drill site, plotted against sub-bottom depth for a sound velocity of 1500 m/s. Seismic reflectors are compared with the wet bulk density data for Hole 1080A, derived from index properties measurements (see "Physical Properties" section, this chapter), which show some pronounced variations. Seismograms are plotted as wiggle traces with gray-scaled amplitudes as background. Density, which mostly controls changes in acoustic impedance at shallow depth, reveals marked physical and lithologic changes that should be visible in the seismic records. All changes identified could be assigned to individual seismic reflectors; however, verification will come from postcruise modeling.
Dolomitic layers were found at 52 mbsf at Hole 1080A and at 38 mbsf at Hole 1080B, where drilling was stopped. At both holes, a continuous, high-amplitude reflector is observed, which could result from the presence of dolomite as a continuous layer. A nodular distribution could be generating the reflector, but reflection amplitudes would be reduced (depending on the density of the nodules). Because thin layers have seismic attributes different from those of normal reflectors, further shore-based analysis could shed some light on the areal pattern of dolomitization.