The shipboard early Pliocene to middle Miocene biostratigraphy of Site 1006 has been refined for this paper, and the main results are described here. These results include (1) the planktonic foraminiferal biostratigraphy, (2) nannofossil biostratigraphy, and (3) cyclostratigraphy. Unfortunately, we could not use the cyclostratigraphy for redating the biohorizons because of the level of uncertainty in the picked beds, the absence of eccentricity control on the amplitude of the resistivity-log cycle peaks, and the lack of magnetostratigraphy. Thus only a 'floating' cyclostratigraphy was produced. The 'floating' stratigraphy represents a period of time that needs to be attached to a certain biostratigraphic datum level with a well-known absolute age to obtain a time frame for the entire 'floating' section.
Planktonic and benthic foraminifers of the middle Miocene-early Pliocene assemblages are very well preserved and abundant throughout this interval. Reworking and contamination of older specimens and/or shallow-water benthic species are very rare or absent. Reworking generally consists of rare specimens of shallow-water benthic foraminifers and of older planktonic foraminifer species such as Eocene turborotaliids or Oligocene Paragloborotalia pseudokugleri.
Planktonic foraminiferal faunas from Hole 1006A are very rich in tropical warm-water species; therefore, the low-latitude standard zonation of Blow (1969, 1979), with slight modification by Kennett and Srinivasan (1983) and Curry, Shackleton, Richter, et al. (1995), was applied for the biostratigraphy of Site 1006, following the criteria used in Eberli, Swart, Malone, et al. (1997; see also Wright and Kroon, Chap. 1, this volume). The datum levels at Site 1006 are presented in Figure 2. Assigned ages for the datum levels that we used in the first instance to calculate sedimentation rates are from Berggren et al. (1995a, 1995b), except for the first occurrence of Globigerinoides conglobatus at 6.2 Ma (Chaisson and Pearson, 1997). The major bioevents used in this study for the identification of the zonal boundaries and zones are presented in Figure 2, and the planktonic foraminiferal datum levels are summarized in Table 1 and Table 2. The following zones have been recognized.
The base of this zone was not observed in the studied samples because F. fohsi together with Fohsella praefohsi and Fohsella peripheroacuta are present from the base of the sequence. Fohsella fohsi lobata first occurs in Sample 166-1006A-73X-3, 137-140 cm, at 673.57 mbsf. This bioevent is equated to be 12.5 Ma (Berggren et al., 1995a, 1995b) The last samples attributed to this zone contain the FO of Fohsella fohsi robusta which is equated to be 12.10 Ma; therefore, the top of the zone (which is equated to be 11.9 Ma) may be condensed at Hole 1006A, or nonrecovery (Fig. 2) may account for the absence of the upper boundary of this zone.
Planktonic foraminiferal assemblages of Zone N12 are also characterized by common Orbulina universa, Dentoglobigerina baroemoenensis, Dentoglobigerina altispira, Globigerinoides ruber, Globigerinoides sacculifer, Globorotalia scitula, Globorotalia archeomenardii, Globoquadrina dehiscens, Zeaglobigerina woodi, Para-globorotalia siakensis/Paragloborotalia mayeri group, and Sphaeroidinellopsis group. Rarer components are Dentoglobigerina larmeui, Globorotalia lenguaensis, Globigerina falconensis, Globigerina bulbosa, Globigerina bulloides, and Globigerinoides obliquus.
Within this zone, Globorotalia praemenardii evolves from its ancestor G. archeomenardii.
Except for the absence of the Fohsella group, planktonic foraminiferal assemblages of this zone are similar to those observed for Zone N12.
In this zone, Globorotalia menardii evolves from its ancestor G. praemenardii. The top of Zone N14 is identified based on the LO of the P. siakensis/P. mayeri group instead of P. mayeri only. These two species belong to the same lineage and sometimes are considered as morphotypes of the same species (Bolli and Saunders, 1985; Berggren et al., 1995b). However, they are considered here also as single species based on their peculiar differences (Spezzaferri, 1994), and informally grouped under the P. siakensis/P. mayeri group. Berggren et al. (1995b) equate the LO of P. mayeri to be 11.4 Ma.
At Hole 1006A, Zones N15 and N16 are considered as a single interval. The lower boundary of Zone N15 is commonly identified with the LO of P. mayeri and the upper boundary with the FO of Neogloboquadrina acostaensis. This latter species was also traditionally used to separate the upper and middle Miocene (N15/N16 boundary) (Blow, 1979; Berggren et al., 1995b). At Hole 1006A, the two species occur together in Sample 166-1006A-65X-5, 85-90 cm, belonging to Zone N14; therefore, Zone N15 as traditionally described was not observed at Hole 1006A. The co-occurrence of P. mayeri and N. acostaensis was observed by Spezzaferri (1998) in the North Atlantic, Spezzaferri et al. (unpubl. data) in South Italy, and Foresi et al. (1998) in the Mediterranean basin. Moreover, the FO of N. acostaensis is reported from middle Miocene sediments by Spiegler and Müller (1992), Müller and Spiegler (1993), and Rio (pers. comm., 1992). Therefore, it is reasonable to assume that the sedimentary sequence at Hole 1006A is continuous and the hiatus, which spans Zone N15 and was identified by shipboard analysis (Eberli, Swart, Malone, et al., 1997), is not present.
Three additional bioevents are identified in this interval. In Sample 166-1006A-60X-4, 85-90 cm, the FO of Polipheribola christiani is observed. This poorly known species seems to be related to Candeina nitida and is regarded as its ancestor (Bolli and Saunders, 1985). It is reported from Zone N16 by Bolli and Saunders (1985). Neogloboquadrina humerosa occurs in Sample 166-1006A-60X-3, 85-90 cm, and is equated to be 8.5 Ma. Finally, Globorotalia merotumida first occurs in Sample 166-1006A-57X-2, 85-90 cm.
The accompanying assemblage includes, among others, O. universa, Sphaeroidinellopsis group, D. baroemoenensis, D. larmeui, G. scitula, G. lenguaensis, G. menardii, Globigerinella obesa, Globigerinella siphonifera, G. sacculifer, Globorotaloides hexagonus and rarer Streptochilus spp.
Several additional bioevents are identified from this interval. The FO of Globigerinoides extremus coincides with the FO of G. plesiotumida, which is equated to be 8.3 Ma according to Berggren et al. (1995b) and Eberli, Swart, Malone, et al. (1997). At the same level, the first occurrence of Globorotalia sphericomiozea can be found. The FO of this species is reported by Berggren et al. (1995b) as young as 5.6 Ma and, therefore, much younger than in Hole 1006A. However, Barbieri (1998) identified this species in older sediments from the Atlantic coast northwest of Morocco.
The FO of Globorotalia mediterranea is observed in Sample 166-1006A-56X-1, 85-90 cm, and is very rare. The first occurrence of Globorotalia suterae is equated to be 7.8 Ma as calibrated in the Mediterranean area. In Hole 1006A this species is first observed in Sample 166-1006A-55X-6, 85-90 cm, below the FO of C. nitida (8.1 Ma). However, this discrepancy may be only apparent. Poore (1979) and Spezzaferri (1998) identified this species from early middle Miocene sediments in the North Atlantic and, therefore, the calibrated age may be the age of the migration of this species in the Mediterranean rather than its real first occurrence as observed in Hole 1006A.
The LO of P. christiani is observed in Sample 166-1006A-55X-4. 85-90 cm. Bolli and Saunders (1985) describe this species from Zone N16 only; however, since no evidence of displacement is found in Hole 1006A, its range should be extended to the base of Zone N17. The FO of Globorotalia juanai is observed in Sample 166-1006A-55X-3, 85-90 and is equated to be 8.1 Ma. This datum level is not applied here, because of the scarcity of this species in Hole 1006A. The FO of C. nitida in Sample 166-1006A-54X-6, 85-90 cm, is considered a more reliable bioevent for Hole 1006A because of the more continuous and abundant presence of this species throughout the sequence. Globorotalia cibaoensis first occurs in Sample 166-1006A-54X-3, 85-90 cm. Its FO is equated to be 7.8 Ma. The taxonomic interpretation of G. cibaoensis is very different from author to author and often it does not correspond to the original description of Bermudez (1949). Therefore, its datum levels are not used for the evaluation of the sedimentation rate, and their stratigraphic positions are only tentatively indicated in Figure 2.
The FO of Globorotalia conomiozea is observed in Sample 166-1006A-51X-1, 85-90 cm, and is equated to be 7.12 Ma. In Hole 1006A, this species is very rare and discontinuous and its FO may not be real; therefore, it is only tentatively indicated in Figure 2.
Globigerinoides conglobatus first occurs in Sample 166-1006A-48X-2, 135-137 cm, and is equated to be 6.2 Ma, whereas Globorotalia margaritae first occurs in 166-1006A-50X-6, 85-90 cm, and is equated to be 6.4 Ma. The LOs of both G. lenguaensis and G. dehiscens are in Sample 166-1006A-44X-1, 35-37 cm. These two bioevents are equated to be 6 and 5.8 Ma, respectively. The co-occurrence of the two species at the same level in Hole 1006A is probably due to the low recovery of Core 166-1006A-43X. The position of the two datums may lie within the missing 9.5 m of sediment between Sample 166-1006A-44X-1, 35-37 cm, and Sample 166-1006A-43X-1, 35-37 cm, according to the time scale of Berggren et al. (1995b).
With the exception of FO and LO of the above-mentioned species, the accompanying assemblages are similar to those described for the other zones and are typical of Atlantic low latitudes. The relatively abundant and continuous presence of Streptochilus spp. is remarkable.
Definition: Interval from the FO of Globorotalia tumida to the FO of Globorotalia miocenica
At Hole 1006A, Zones N18 and N19 are considered as a single interval. The base of Zone N18 is placed with the FO of G. tumida and its top is not observed at Hole 1006A. The boundary between Zone N18 and N19 is based on the FO of Sphaeroidinella dehiscens, which has not been identified in the examined samples from Hole 1006A.
In this interval, the FO of Globorotalia aemiliana and the LO of G. cibaoensis are identified in Sample 166-1006A-40X-1, 85-90 cm, and 37X-2, 85-90 cm, respectively. The latter bioevent is equated to be 4.6 Ma. Globorotalia aemiliana evolves within this zone into Globorotalia crassaformis following the lineage proposed by Colalongo and Sartoni (1967); Globorotalia hirsuta (= G. margaritae)-G. aemiliana-G. crassaformis. Finally, also Neogloboquadrina pseudopima was first observed in Sample 166-1006A-40X-5, 85-90 cm. This species is not used here for the identification of Zone N20 because this species is also reported from the upper Zone N18 (Brönniman & Resig, 1971; Fleisher, 1974).
Planktonic foraminifer assemblages in this zone are very rich and diversified and include, among others, O. universa, Sphaeroidinellopsis group, D. baroemoenensis, D. altispira, G. scitula, G. menardii, G. margaritae, G. suterae, G. dehiscens, Z. woodi, Z. nepenthes, G. falconensis, Gg. bulloides, G. obliquus, G. ruber, G. sacculifer, Globigerinoides trilobus, G. conglobatus, N. acostaensis sinistral and dextral, N. humerosa, Neogloboquadrina dutertrei, and G. siphonifera. Minor components are D. larmeui, G. obesa, G. bulbosa, Neogloboquadrina pachyderma and Streptochilus spp.
The Miocene/Pliocene boundary is currently defined by the age of the base of the Trubi Formation at Capo Rossello in Sicily (Cita and Gartner, 1973; Cita, 1975). Traditionally, the beginning of the Pliocene (the base of the Zanclean stage) is equated with evidence in the sedimentary record of the recovery of the Mediterranean Sea to full marine conditions after the "Messinian Salinity Crisis," when the basin became completely isolated (Ryan, Hsü, et al., 1973; Hsü, Montadert, et al., 1978; McKenzie at al., 1990; Spezzaferri et al., 1998). Sedimentary expression of this major event is a sharp lithologic change from an evaporative sequence topped with the "Lago Mare Facies" to pelagic sediments rich in planktonic foraminiferal faunas.
A sharp lithologic change at the Miocene/Pliocene boundary is not known to be present in geological records from the open ocean. In addition, this boundary does not correspond to any major biological event. Rather, it is characterized by a relatively low degree of biotic turnover (extinction plus origination) within all the microfossil groups and especially in planktonic foraminifers.
Blow (1969, 1979) placed the Miocene/Pliocene boundary within Zone N18, between the FOs of G. tumida and S. dehiscens. The FO of G. tumida is the bioevent most commonly accepted to identify this boundary in open ocean sequences (Premoli Silva et al., 1993; Berggren, 1973; Fleisher, 1974; Vincent, 1975; Berggren et al., 1995a, 1995b). However, based on the absence of marker species, the correlation between bioevents from the Mediterranean Sea and those from the open ocean remains difficult.
Since the Miocene/Pliocene boundary in the sequences from the Mediterranean Sea and the open ocean cannot be directly correlated using only bioevents, McKenzie et al. (in press) proposed to make an indirect correlation using the biostratigraphically derived sedimentation rates from the Bahamian sequence recovered at Hole 1006A to locate the age of the Miocene/Pliocene boundary as determined using the astronomically calibrated time scale as applied in the Mediterranean (Hilgen, 1991). Based on this correlation, we place the Miocene/Pliocene boundary at ~370-375 mbsf (tuned age of 5.33 Ma according to Hilgen, 1991) and the Zones MPL1/MPL2 boundary at ~361 mbsf (5.1 Ma according to Hilgen, 1991).
Postcruise analysis of the calcareous nannofossil events was conducted to obtain a higher biostratigraphic resolution in the lower Pliocene-middle Miocene interval in Hole 1006A. The sampling distance for the lower Pliocene-middle Miocene interval is ~1.5-2 m for the nannofossil bioevents. The ages for the nannofossil events are derived from Berggren et al. (1995b). The zonal scheme established by Martini (1971) was used for the taxonomic interpretation. The datum levels are presented in Table 3.
The lower Pliocene is divided into three zones: Zones NN15, NN14-NN13, and NN12. The interval between the last appearances of Sphenolithus abies and Amaurolithus spp. defines NN15 (3.66-4.5 Ma). At Site 1006, the lower boundary of this interval occurs between Samples 166-1006A-28H-3, 114-115 cm, and 28H-5, 14-15 cm. Zones NN14 and NN13 are combined and are bounded by the LO of Amaurolithus spp. above and the first appearance of Ceratolithus rugosus between Samples 166-1006A-36X-1, 114-115 cm, and 36X-2, 14-15 cm.
At Site 1006, well-preserved specimens of Discoaster quinqueramus and Discoaster berggrenii can be found, marker species for upper Miocene Zone NN11. The total concurrent range of both species (5.6-8.6 Ma) defines Zone NN11. The highest occurrence of specimens of D. quinqueramus is present in Sample 166-1006A-38X-4, 64-65 cm. However, this occurrence is too high with respect to planktonic foraminiferal datum events; thus, the specimens in this sample are considered to have been reworked. Also, this occurrence is incompatible with the sedimentation rate changes based on the cyclostratigraphy (see "Early Pliocene-middle Miocene cyclostratigraphy" section). The basal occurrence of D. berggrenii is in Sample 166-1006A-54X, CC. In addition, useful datum levels such as Amaurolithus amplificus and Amaurolithus primus were found (Table 4). Zone NN10 ranges from the base of D. quinqueramus (8.6 Ma) to the last appearance of Discoaster hamatus. The base of Zone NN10 is placed between Samples 166-1006A-59X-3, 14-15 cm, and 59X-CC (538.05 mbsf). The total range of D. hamatus defines Zone NN9. The base of Zone NN9 was placed between Samples 166-1006A-52X-6, 14-15 cm, and 62X, CC.
Middle Miocene Zones NN6 through NN8 are recognized at Site 1006. The first appearance of Catinaster coalitus occurs at the NN7/NN8 boundary. This level is identified between Samples 166-1006A-66X-2, 14-15 cm, and 66X-CC. The base of Zone NN7 (13.2 Ma) is marked by the last appearance of Cyclicargolithus floridanus. This level is placed between Samples 166-1006A-75X-CC and 76X-1, 12-13 cm.