One of the more noteworthy results from studies of Site 976 samples was to prove that extensional tectonics in the Alboran Sea basin continued up until the late middle Miocene and that the structural basement high was exposed to the sea floor by about 12 Ma (late Serravallian; Shipboard Scientific Party, 1996). However, direct drilling gives no further information on the structural behavior of this basement high.
Regional seismic data in the Western Alboran Basin are worth considering to maximize drilling data and better understand the brittle fracturing in the top of the basement high. With this aim, we discuss below seismic images from the two high-resolution seismic lines taken by the JOIDES Resolution across Site 976 (Fig. 5, Fig. 6) and those from a multichannel seismic (MSC) composite-line (5 s, TWT penetration) located about 25 km northeast of Site 976, traversing the entire depocenter of the northern branch of the Western Alboran Basin (Fig. 7). Among several options, we chosen this MCS line to support our conclusion as ages and thickness of the lower sedimentary units in the basin are constrained from Alboran Alb-A1 well data. Figure 8 shows a basement contour map of the same depocenter, including the northwest slope of the Site 976 basement high.
Normal faulting bounding the basement high is apparent from seismic Line 2S-1 (Fig. 6), and from this image the structural high may be considered as a simple horst (Shipboard Scientific Party, 1996). However, as examined below, shallow faulting in the top of the high does not reveal its structural behavior.
Noteworthy structural features can be deduced from the high-resolution seismic picture in Figure 6. The top of the basement is very irregular, similar to the basement top imaged along the perpendicular seismic line 2S-II (Fig. 5). In places this morphology is clearly related to high-angle normal faulting, which is consistent with data from brittle deformation of the basement. Although no indication of emersion has been encountered in basement breccias, concurrent erosional topography, perhaps submarine, cannot be discounted given this widespread basement-top morphology. In the northwest flank, individual reflectors within middle Miocene sediments (Reflector R4 and those within seismic Unit IV; Fig. 6) are normal faulted, indicating that extension continued until this time (Shipboard Scientific Party, 1996). However, small reverse faults are seen in the same northwest flank affecting upper Miocene (Reflector R3 and those within seismic Unit III) deposits, as well as an anticline attitude of late Miocene to Pleistocene sediments (seismic Unit III/Tortonian to Subunit Ic) capping the basement. We believe these seismic data indicate that the basement high has also been restructured by compressional tectonics during the contractive reorganization of the Alboran Sea basin (Comas et al., 1992). In addition, the pre-early Pliocene (pre-Subunit Ic; Fig. 6) erosion of the Messinian Unit II just above the basement high and even some steep drag structures in reflectors from Subunit Ic (early Pliocene) near the basement hinge seem to point to post-Messinian folding superimposed on the extensional structures. Accordingly, we suggest that the influence of later compressional structures uplifting the basement must be considered for any reliable conclusion on the Miocene extensional uplift of this high.
Around Site 976, the basement high trends northeast-southwest, turning to northwest-southeast toward the south (Fig. 2). To the northwest, it limits a complex graben filled with up to 5-6 km of lower Miocene (late Aquitanian(?)-early Burdigalian, 20-19 Ma) to Pleistocene sequences (Fig. 7). The complex geometry of the basement top beneath the basin (Fig. 8) probably results from the interference of different surfaces formed largely by low-angle and high-angle normal faults, oblique-slip faults, and the tops of roll-over structures related to major extensional detachments (de la Linde et al., 1996).
The seismic section in Figure 7 shows the asymmetric character of the graben. The northern slope corresponds to a very low-angle faulted surface formed most likely by horsetail faults or a listric fan rooted on deeper listric detachment faults within the basement (apparent direction of extension toward the south-southeast). Individual reflectors (i.e., Reflectors R5 and R4) imaged in the northwest half of the MCS profile are tilted or rooted by low-angle normal faults within the sedimentary cover. This low-angle listric system is postdated by Reflector R3, which corresponds to about 11-10 Ma, intra-lower Tortonian sediments (Comas et al., 1992). In contrast, the steeper southern flank is cut by higher angle antithetic normal faults. From northwest to southeast, reflectors within the basin first diverge eastward and then progressively onlap or pinch out (or drag) against the basement (i.e., Reflectors R5, R4, and R3; Fig. 7). Reflector attitudes on the southern flank of the graben suggest that the basement scarpment developed from a sequence of faults, antithetic to the main low-angle listric fault, cutting back toward the top of the high, which developed from at least the Burdigalian-Langhian (Reflector R5, about 18-19 Ma), to the early Tortonian (Reflector R3, about 10-11 Ma). Given this set of structures in the graben, we interpret the basement high in this region as corresponding to a faulted rollover anticline in the hanging wall of a south-southeast-dipping major extensional detachment system. The rollover appears to be progressively extended by successive counter faults, therefore representing a central high (see Gibbs, 1984) in the Alboran Sea basin by the middle Miocene. The shallow crustal image presented here does not allow an interpretation of the extensional detachment system to deeper levels in the crust. However, single high-reflectivity reflections (intracrustal reflectors in Fig. 7) imaged at the western end of Figure 7 (at 3 to 5 s, TWT), seemingly ubiquitous in the Western Alboran Basin (Watts et al., 1993), suggest that the shallow extensional system is rooted in deeper and probably more complex intracrustal extensional detachments that caused major thinning of the Alboran Crustal Domain.
As similar asymmetric structures are observed the length of the graben southward to Site 976 (compare MCS profile in Fig. 7 with MCS profile ALB-39 in Fig. 3 of Shipboard Scientific Party, 1996), we determine that Site 976 was drilled in the same basement rollover as shown in Figure 7. The brittle deformation reported here may have formed largely because of middle Miocene antithetic counter faults and associated tensional conditions, developing near the top of a rollover structure. High-angle dip reported from the fault-gouge zone penetrated at Hole 976B (Fig. 5) as well as the steep main foliation in metamorphic rocks sampled at Site 976 may have originated from rotations of originally lower angle structures in this rollover.
Further implications from the above inference is that the exhumation of metamorphic basement rocks in the Western Alboran Basin took place up to the early Miocene (before deposition of seismic Unit VI; Fig. 7), and that the vertical relief of the basement high was uplifted later on and by steps, as the subsequent middle Miocene extension progressed in the basin. The basement high at Site 976 must have been exposed to the seafloor not only by the late Serravallian, as demonstrated by data from drilling, but has probably also been a submarine (or occasionally emergent?) high since the early Miocene.