High-resolution biotratigraphy and chemostratigraphy of the Cenomanian/Turonian boundary event in the Vocontian Basin, southeast France

Since the late 1970s, Oceanic Anoxic Event 2 (OAE2) (Schlanger and Jenkyns, 1976, Arthur and Schlanger, 1979, Jenkyns, 1980, Jenkyns, 1997) has drawn continuous attention because of its short duration and because of the superimposition of three crises during the event: a sedimentary one with the wide scale occurrence of black shales (Schlanger and Jenkyns, 1976, Arthur and Schlanger, 1979, Hart and Bigg, 1981, Arthur and Premoli-Silva, 1982, Amédro and Robaszynski, 1993, Arthur and Sageman, 1994); a planktic biologic one with the temporary disappearance of keeled planktic foraminifera in the world's oceans (Eicher, 1969, Eicher and Worstell, 1970, Sigal, 1977, Hart and Bigg, 1981, Caron and Homewood, 1982, Eicher and Diner, 1985, Leckie, 1985, Hart and Ball, 1986, Hart, 1996, Lamolda et al., 1997, Grosheny and Malartre, 1997, Keller et al., 2001, Caron et al., in press); and a geochemical one with a positive δ¹³C excursion, also recorded in many basins both on the carbonate and organic-matter fractions (Pratt and Threlkeld, 1984, Pratt, 1985, Hilbrecht and Hoefs, 1986, Schlanger et al., 1987, Jarvis et al., 1988, Jenkyns et al., 1994, Accarie et al., 1996, Luderer and Kuhnt, 1997, Morel, 1998, Paul et al., 1999). A number of sections have been studied worldwide, some of them eventually becoming reference sections, such as the Eastbourne section in the Anglo-Paris Basin (Jarvis et al., 1988, Gale et al., 1993, Hart et al., 1993, Paul et al., 1999), the Pueblo section in the US Western Interior (Pratt and Threlkeld, 1984; Pratt, 1985; Eicher and Diner, 1985, Leckie, 1985, Pratt et al., 1993, Leckie et al., 1998, Morel, 1998), the Gubbio section in the northern Tethys (Arthur and Premoli-Silva, 1982, Corfield et al., 1991, Premoli Silva and Sliter, 1995, Luciani and Cobianchi, 1999), the Wunstorf section in boreal German basins (Weiss, 1982, Ernst et al., 1984, Hilbrecht, 1986), and the Tarfaya section on the Atlantic margin (Kuhnt et al., 1990, Luderer and Kuhnt, 1997). The event has been also recognized in the deep ocean at many DSDP and ODP sites. Regional syntheses have been attempted like that of Lüning et al. (2004) for the North African craton. Basinal deposits, in both the deep ocean and the deepest parts of epicratonic basins, usually include one or several black shale layers close to the Cenomanian/Turonian boundary, but shallower deposits may be devoid of them. These nevertheless show the δ¹³C anomaly (work in progress in Saharan and Middle East carbonate sections devoid of black shales), which appears to be the best way of correlating sections worldwide in different depositional environments.

The event is also recorded in the French subalpine basin where the “Thomel level” (Crumière, 1990, Crumière et al., 1990) of the Vergons section has been termed as a local equivalent of the “Bonarelli level” (Arthur and Premoli-Silva, 1982) of the Gubbio section in Italy. The Vocontian Basin is interesting because many sections can be studied and correlated on a bed by bed basis in both basinal and slope deposits. In this paper we deal only with deep basinal sections that include the black shales.

The Cenomanian/Turonian boundary is difficult to define purely from a palaeontological viewpoint. Using ammonites, the boundary lies between the last occurrence of Neocardioceras juddii and the first occurrence of Watinoceras coloradoense or W. devonense (Bengston, 1996). In the thin Pueblo section (US Western Interior) the boundary is between two successive limestone beds, 84 and 86 (numbering system of Cobban and Scott, 1972, Kennedy et al., 1999, Kennedy et al., 2000). The transitional interval is as thin as an interbed, which is among the reasons why this section was proposed for the boundary reference section during the Second International Symposium on Cretaceous Stage Boundaries (Bengston, 1996). In other basins, the transitional interval is usually thicker owing to either the lack of ammonites or to a greater accumulation rate. With planktic foraminifera, the boundary is more difficult to define. The W. archaeocretacea Partial Range Zone (or “zone à grosses globigérines” of Sigal, 1977) has been defined (Robaszynski et al., 1979, Robaszynski and Caron, 1995) between the last occurrence of the upper Cenomanian index species Rotalipora cushmani and the first occurrence of the lower Turonian index species Helvetoglobotruncana helvetica, within which is the Cenomanian/Turonian stage boundary. This W. archaeocretacea Zone has been proven recently to be extremely reduced in the Pueblo section (Beaudoin et al., 1995, Morel, 1998, Desmares et al., 2004). Beaudoin et al. (1996) have shown that in the lower Cenomanian of Italy, the vertical range of planktic foraminifera may vary markedly between closely spaced sections within the same basin. We have used the closely spaced basinal sections of the Vergons area similarly to investigate the subalpine basin at the Cenomanian/Turonian boundary.

It has been shown that several sections in distant basins (Eastbourne, Wunstorf, Tarfaya) show the same three-peaked δ¹³C positive excursion pattern as that defined in the Pueblo section (Pratt and Threlkeld, 1984). We have used this particular pattern to correlate our sections in the subalpine basin. High-resolution correlations based on both bed bundles and δ¹³C curves thus enabled us to check the vertical ranges of planktic foraminifera against anoxic layers, as demonstrated in this paper. We also show that the Vergons section proposed by Crumière et al. (1991) as the local reference section is unsuitable because it is affected by synsedimentary sliding.

Geological setting

The study area is located in the Castellane tectonic arc area of the southern subalpine ranges in southeast France (Fig. 1A). The northern part of the arc belongs palaeogeographically to the eastern reaches of a relatively deep basin known as the Vocontian Trough. The southern part of the arc corresponds to the outer part of the Provence Platform situated further to the south. The slope area in between was relatively stable throughout the Cretaceous Period (see facies maps in Debrand-Passart et

Material and methods

Results

Lithologic, isotopic and gross biostratigraphic data obtained in the Ondres, Pont d'Issole and Vergons sections are illustrated on Fig. 2.

Correlations

There is no problem in correlating the Pont d'Issole and Ondres sections (Fig. 5). The vertical ranges of planktic foraminiferid index species supports the correlations between lithological units M1 to M3, the base of true Turonian deposits being defined either by the first occurrence of marginotruncanids or by the index species H. helvetica. The last occurrence of R. cushmani is at the base (Ondres) or the top (Pont d'Issole) of the upper, calcareous part of the M1 unit. This result suggests

Conclusions

The Cenomanian/Turonian boundary black shales have been correlated over several sections at the base of the northern Provence platform slope in the French subalpine basin. The local δ¹³C isotope anomaly shows three events (1, 2, 3 or A, B, C) defined in other basins by others workers. The Cenomanian/Turonian boundary spans roughly the duration of the black shale anomaly. The setting of black shale deposition is well defined and equates with profound changes in both benthic and planktic

Acknowledgements

We are deeply indebted to our reviewers, Christopher Wood and especially Haydon Bailey who spent much time carefully annotating the manuscript to make our sentences clearer. David Batten improved the final version. This paper is a contribution of the Ecole et Observatoire des Sciences de la Terre, Université Louis Pasteur n° 2005-503-UMR7517.