Lower Cretaceous nannofossil biostratigraphy was first established in low-latitude, Tethyan, onshore sections (Thierstein, 1971, 1973; Sissingh, 1977, with modifications by Perch-Nielsen, 1979, 1985; Applegate and Bergen, 1988) and later DSDP oceanic sections, mainly from the low-latitude central Atlantic (Roth, 1978, 1983, with modifications by Bralower, 1987; Bralower et al., 1993). Perch-Nielsen (1979) recognized the (bio)geographic limitations of these schemes and incorporated data from high-latitude Boreal sections (northern Europe and the North Sea Basin), although more recently separate high-resolution schemes have been developed for these areas (see reviews in Bown et al., 1998; Jeremiah, 2001). More recent studies on Tethyan sections have yielded particularly good correlations with paleomagnetic and stable isotope stratigraphies (e.g., Bralower et al., 1989; Channell et al., 1993, 1995) and have improved resolution and integration with ammonite biostratigraphy (Bergen, 1994; Gardin et al., 2000). The Jurassic/Cretaceous boundary interval is particularly problematic because of the paucity of stratigraphically complete nannofossil-productive sections with good preservation. Boreal sections are largely in nonmarine or transitional facies with stratigraphic gaps (Bown and Cooper, 1998) and Tethyan Mediterranean–Caribbean sections in massive carbonates with poor nannofossil preservation (Bralower et al., 1989). High-quality data are almost entirely restricted to two DSDP sites from the Central Atlantic Ocean (Sites 391 and 534) (Roth, 1983; Bralower et al., 1989; Bergen, 1994; but see also de Kaenel and Bergen, 1996).
The NK and NC zones and subzones of Roth (1978, 1983), Bralower et al. (1989), and Bralower (1987) were used to provide the biostratigraphic framework for Sites 1207, 1213, and 1214, but the Shatsky region is far removed from the source areas of these schemes and thus the differences observed are not surprising and will be discussed further below.
The zones and zonal boundaries were identified as follows.
The Jurassic/Cretaceous boundary interval zonation of Bralower et al. (1989) is based on a distinctive succession of nannolith appearances, notably Conusphaera and Nannoconus; however, these taxa were absent in this part of the section and the former was absent throughout. In addition, a number of important marker species of the family Cretarhabdaceae (C. cuvillieri, R. angustiforata, and Retecapsa octofenestrata) and genus Eiffellithus (E. primus, E. windii, and E. striatus), although present, are rare and restricted to a small number of samples, and their first and last occurrences may not be biostratigraphically reliable. The lowest Cretaceous zones are thus identified using marker species, where present, together with alternative datum events and aspects of the entire assemblages.
The lowermost productive samples (Core 198-1213B-27R) yielded H. chiastia, L. carniolensis, Tubodiscus bellii, and R. laffittei, indicating Subzone NJKc or younger. The nannofossils do not unambiguously indicate a Cretaceous age, but correlation with Zone NK1 is inferred based on the presence of the genus Tubodiscus and absence of R. angustiforata, P. fenestrata, and R. wisei (Bralower et al., 1989). Support for this interpretation also comes from radiolarian fauna that also indicate a Berriasian age for the lowermost cores (H. Kano, pers. comm., 2003).
The subzone base is marked by the FO of R. angustiforata (Sample 198-1213B-26R-1, 17–18 cm), although this is an isolated trace occurrence and thus of questionable reliability. The FO of A. infracretacea is also recorded in this sample (FO indicates Subzone NJKd [lower Berriasian] after Bralower et al., 1989), along with a number of unambiguously Cretaceous species (e.g., S. silvaradius and Diadorhombus rectus). The FO of R. wisei, an intra-Subzone NK2a event (upper Berriasian) (Bralower et al., 1989), is recorded in Sample 198-1213B-22R-1, 58–60 cm. An interesting feature of these lowermost Cretaceous assemblages is the consistent presence of a number of taxa that were previously thought to have had LOs in the Upper Jurassic, notably P. grassei, A. cylindratus, and Biscutum dorsetensis. It is unlikely that these are reworked specimens, as no other Upper Jurassic taxa are present, and, in any case, a Jurassic sediment source may not have been present in the Shatsky area. These species may have been overlooked due to low numbers in previously described sections or, more likely, may have had a distinctive Pacific paleobiogeography, and survived longer in this region.
The FO of the zonal marker P. fenestrata occurs in Sample 198-1213B-18R-1, 1 cm, although this is an isolated trace occurrence and thus of questionable reliability. The LOs of Helenea quadrata and U. granulosa were also recorded in this interval and have been previously noted in Zone NK2 (Bralower et al., 1989; Bergen, 1994).
The zonal marker C. oblongata is absent, but the FO of R. dekaenelii has been recorded at a similar stratigraphic level (lower Valanginian) (Bergen, 1994) and is used here as a proxy marker for the NK3 zonal boundary. R. dekaenelii is distinctive, relatively common, and consistently present throughout its range at Site 1213.
The LO of the subzonal marker R. wisei occurs in Sample 198-1213B-14R-1, 12 cm, and the co-occurrence of E. windii in Core 14R indicates a well-constrained age of mid-Valanginian (Bralower et al., 1989; Bergen, 1994; Gardin et al., 2000).
The stratigraphy of this interval cannot be further subdivided because of the absence of T. verenae, L. bollii, and nannoconids and only a trace occurrence of E. striatus. The LO of Helenea conus, recorded in Core 198-1213B-13R, has previously been observed in the upper Valanginian (Bergen, 1994).
The upper part of this interval in Hole 1213B (lower part of Section 198-1213B-9R-1) cannot be younger than lower Subzone NC4b (lower Hauterivian) due to the continued presence of C. cuvillieri and R. dekaenelii (Bergen, 1994).
The co-occurrence of Z. scutula, C. cuvillieri, C. oblongata, and L. bollii in Sample 198-1214-24R-1, 72 cm, infers correlation with a short interval within the upper part of Subzone NC4b (upper Hauterivian) (Bergen, 1994; Gardin et al., 2000). The latter two species were not recorded at Sites 1207 and 1213, suggesting this interval was only recovered at Site 1214. This is further indicated by the absence of R. dekaenelii, the rare but consistent presence of Nannoconus (three samples, the only persistent occurrence of the genus observed during Leg 198), and abundant Micrantholithus fragments, although the latter was also observed in Subzones NC5d–NC5e at Site 1207 (see below).
The FO of Assipetra terebrodentarius, recorded at Site 1214 within this interval, has previously been recorded at a higher stratigraphic level (Zone NC5), close to the Hauterivian/Barremian boundary, and not overlapping with the range of C. cuvillieri (Bralower, 1987; Bergen, 1994; Channell et al., 1995; Gardin et al., 2000).
Subzone NC5d is defined by the LO of C. oblongata and is a widely used event. Subzone NC5e has been variously defined by the LO of Nannoconus steinmannii (Bralower, 1987) and FO of Flabellites oblongus (Bralower et al., 1995). The former event is unreliable or stratigraphically higher, and the latter event usually lies very close to the base of the overlying zone, defined by the FO of H. irregularis. Thus, the two subzones are merged in this study.
This interval was recovered in Sites 1207 and 1213 but lacks age-diagnostic taxa. The absence of C. cuvillieri, C. oblongata, and H. irregularis and presence of A. terebrodentarius and Z. scutula were used to infer a correlation with Subzones NC5d–NC5e (Barremian). The common occurrence of Z. scutula was utilized as a Barremian acme subzone by Bown et al. (1998) and Jeremiah (2001), and this may be a widespread, correlatable feature.
This zone is defined by the FO of H. irregularis. The FO of F. oblongus is also recorded at or just above this level. All three sites recovered noncalcareous black shale intervals bracketed by the FOs of H. irregularis and E. floralis, defining the extent of the lower Aptian Zone NC6. This strongly indicates that the black shales are stratigraphically equivalent to the Selli event and OAE1a.
This zone is defined by the FO of E. floralis. Zone NC7 could not be further subdivided due to the absence of Micrantholithus and Rhagodiscus achlyostaurion. The latter species does occur higher in the sections, in the lower Albian, and has previously been recorded at similar levels (Erba, 1988).
This zone is defined by the FO of small, rare P. columnata. Subzone NC8b could not be reliably recognized due to absence of Hayesites albiensis (see "Appendix A," and discussion in Kennedy et al., 2000). The FO of P. columnata is recorded in the uppermost Aptian (plesiotypica/jacobi ammonite zone) (Bown in Kennedy et al., 2000) but is often used as a proxy marker for the Aptian/Albian boundary. The LO of large A. terebrodentarius (ssp. youngii) was recorded just above the FO of P. columnata by Tremolada and Erba (2002), and although this subspecies extends to higher stratigraphic levels, its last common, consistent occurrence may provide an alternative approximation for this stage boundary level.
Robinson et al. (2004) inferred the presence of OAE1b within this interval, expressed as more silica-rich sediments, identified using ODP downhole logging data.
The middle and upper Albian marker species T. orionatus, A. albianus, and E. turriseiffelii were all identified at each site, but their stratigraphic spacing and order is atypical. T. orionatus and A. albianus are recorded anomalously high at all three sites (i.e., close to the FO of E. turriseiffelii), and A. albianus is recorded out of sequence, above the FO of E. turriseiffelii, at Sites 1213 and 1214. The clustering of these events (see age-depth plots in Fig. F3) perhaps indicate low sedimentation rates and/or a possible mid-Albian hiatus, but the out-of-sequence ordering indicates problems with the range of at least A. albianus in this area. The problem may also be exacerbated by datum/timescale correlation problems. The extent of the mismatch can be most clearly seen on the age-depth plots (Fig. F3), where no single straight line can be fitted to the mid-Albian datum points.
The NC9b subzonal marker, Eiffellithus monechiae, is present in correct stratigraphic position at Site 1207 but is rare at Site 1214 and is not recorded at Site 1213.
This subzone is defined by the FO of E. turriseiffelii; this, and the overlying Subzone NC10b, are relatively thick at Sites 1207, 1213, and 1214, indicating a relatively complete section across the Albian/Cenomanian boundary.
Due to the absence of Corollithion kennedyi, this subzone cannot be recognized by its primary marker species; however, a number of other boundary interval events are present, including the LO of G. stenostaurion, the LO of Watznaueria britannica, the FO and LO of Gartnerago chiasta, the FO of Gartnerago ponticulus (= nanum of many authors), the FO of Gartnerago theta, and the LO of Staurolithites glaber. According to Burnett (1999), the LO of W. britannica lies closest to the Albian/Cenomanian boundary, and we have used this event as a proxy for the base of Subzone NC10b (Zone UC1). The younger event, the LO of S. glaber, occurs in upper Zone UC1 (Burnett, 1999), and we use this as an approximation for the base of Zone UC2 in the absence of Gartnerago segmentatum.
This zone is marked by the FO of L. acutus.