Given the present deep open marine setting of the holes drilled during Leg 161 and the likelihood that similar conditions have existed throughout the late Pliocene-Holocene (as indicated by the similarity of the fossil assemblage and lithology throughout the sequence), three possible interpretations can be used to explain the paucity of primary sedimentary structures, poor sorting, abundance of trace fossils, and the mixed siliciclastic/carbonate provenance of most of the sediment. They are dependent in part on the grain size of the detrital siliciclastic material. These scenarios, discussed further later in this paper, are the following:
One unexpected result of this study is the possibility of climatic cycles being represented by variation in the sedimentary section. The presence of diffuse ORLs in the Alboran Basin, many of which can be correlated with more characteristic sapropels (sensu Hilgen, 1991) further to the east (see Murat, Chap. 41, this volume), suggests that climate/ocean circulation cyclicity has affected part of the sedimentary record (i.e., the organic content) in the past, at a frequency that is an order of magnitude greater for the ORLs than that discussed below for the carbonate sediments. The latter are mineralogically indistinguishable from sediments adjacent to the ORLs, which supports an argument for at least two levels of regional control operating in the basin.
Hole 976B was drilled on a basement high, but seismic data (Shipboard Scientific Party, 1996b, figs 3 and 4) show the upper Pliocene-Pleistocene section to thicken toward the southeast, in the direction of the current depocenter of the West Alboran sub-basin. In other words, during the late Pliocene to Holocene, Site 976 was located at a zone of marine slope deposition, rather than at the base of the slope. The location is physiographically separated from the South (Site 979) and from the East (Sites 977 and 978) Alboran sub-basins by the northeast-trending volcanic Alboran Ridge (Fig. 1).
Sites 977 and 978 are located with the East Alboran Basin, near the center of two east-dipping in-filled submarine depressions, to the south and north of the Al-Mansour Seamount, respectively. The seamount is a prominent bathymetric high, which could have influenced the nature and distribution of sedimentation throughout the late Pliocene-Holocene. Site 979 is located on the southeastern flank of the Alboran Ridge, and from late Pliocene to Holocene times, it has been located near the distal end of a deep submarine ramp (Shipboard Scientific Party, 1996e, figs. 3 and 4).
The Al-Mansour Seamount and the Alboran Ridge are believed to have formed during the early Miocene (Comas et al., 1992; Woodside and Maldonado, 1992) during a period of extension that continued into the late Miocene (Watts et al., 1993). Both the volcanics and adjacent Miocene sediments have undergone compressional structural modification since then (Shipboard Scientific Party, 1996e). The four sites occupied during Leg 161 provided widely spaced coverage of the main physiographic provinces of the Alboran Basin, some of which have been isolated from each other throughout the late Pliocene-Holocene. One result of this is that correlatable sedimentation events must invoke regional control mechanisms rather than site-specific ones in their explanation.
The four sites represent a variety of deep marine depositional environments including slope, ramp, and basin depocenter settings, and the minor lithologies present at each site may represent these differences. For example, slumping is more common in the depocenter settings represented by Holes 977A and 978A.
Graded beds of sand and coarse silt (lithostratigraphic Unit II, Hole 976B; lithostratigraphic Subunit IA, Hole 978A; Unit I, Hole 979A), packstone (Hole 978A, Subunit IA), and mass-flow units, including slumps and breccias (Subunits IA and IB, Hole 977A; Subunit IC, Hole 978A), all suggest that both gravity currents and mass-flow mechanisms have been operating in the basin throughout the last 4 m.y.
The presence of sand-sized quartz throughout the sequence in all holes necessitates transport by sediment gravity currents, either to the site of original deposition (in the case of mass-flow emplaced units) or at the current locations of the drill holes. Localized gravity flows are further supported by the lack of regional correlation between quartz cycles (Fig. 10, Fig. 11, Fig. 12).
The trace-fossil assemblage comprising mainly Zoophycus and Chondrites ichnogenera and belonging to the Zoophycus to Nereites associations (Collinson and Thompson, 1982, fig. 9.41) also supports a deep marine depositional setting for both the sandy and finer material. Graded sand and coarse silt beds 2-10 cm thick are uncommonly present in the sequence. If these are representative of average deposition of material of this grade, then it is likely that the benthic fauna would have little difficulty in reworking and homogenizing this material into the hemipelagic sediment, especially if the turbidite events are relatively infrequent as is suggested by the overall low percentage of sand and coarse silt in the sequence (everywhere <10%). In contrast, mass-flow units of thickness exceeding 1 m are unlikely to be completely reworked, and so their relative paucity suggests a low frequency of occurrence. None of the sand material comprising lithostratigraphic Unit I is considered to have been deposited as part of base-of-slope or inner (upper) fan complexes because of the lack of thick, coarse-grained units. Even in Unit II, Hole 976B, sand beds appear to have been of similar dimension to those preserved rarely in Unit I, the principal difference being their more common occurrence.
At this stage it is not possible to support the contention of shipboard scientists (Shipboard Scientific Party, 1996b, fig. 25) about cycles of upward increasing detrital content being present in Unit I at Site 976. It is more likely that the sand- and coarse silt-sized sediment recovered from the sites represents deposition onto the deep marine basin floor into which the distal parts of infrequent gravity currents flowed.
The origin of fine silt- and clay-sized detrital quartz (together with feldspar and presumably muscovite) is more equivocal. This material could easily have been incorporated in localized turbidity current flows as part of the grain-size continuum. However, the possible correlation of <20-µm quartz cycle peaks between 2 and 3 Ma (Fig. 12) points to a regional control that may be climatic in origin. Possibilities include (1) that the cycles relate to arid periods occurring on the north Africa continent and a resultant increase in eolian transport, or (2) fluctuations in atmospheric circulation causing changes in prevailing wind direction (Moulin et al., 1997).
The 9-m spacing of samples (~1 per core) has been sufficient to define a basin-wide correlation of carbonate cycle peaks on a frequency of ~500 ka for the distribution of calcareous nannofossils, and this suggests a control mechanism that is at least of regional extent. The fact that similar synchroneity is not present in the detrital (quartz) distribution indicates the control is probably not a tectonic one. Cycles of this order of magnitude (~400 ka) fall at the upper end of fourth- and fifth-order sea-level cycles (Fischer, 1986) and, hence, have been interpreted to result from climatic variation associated with Milankovitch orbital cycles (Cotillon, 1991). Similar periodicities were reported by Moore et al. (1982) for upper Miocene and Quaternary deep marine carbonate sediments in the Pacific Ocean. In that example, the variation is attributed to changes in the carbonate compensation depth brought about by climatic factors. In the Alboran Sea, climate appears to have affected microfossil abundance rather than preservation. The fact that cycles in the >63-µm carbonate fraction (mainly foraminifers) do not appear to be synchronous with those in the <20-µm carbonate fraction (mainly nannofossils) suggests that the climatic effect may also be species selective.