Biomagnetostratigraphic data on the Pollenzo section: new insight on the onset of the Messinian Salinity Crisis

During the Messinian (7.2 to 5.3 Ma) the Mediterranean area experienced fast and prominent climatic and structural changes, culminating in the so-called Messinian Salinity Crisis, starting in shallow water settings with the deposition of the Primary Lower Gypsum (PLG sensu Roveri et al., 2008). The stratigraphic framework for this interval is relatively well constrained and the base of the PLG is dated at approximately 5.96 Ma and thought to be coeval all over the Mediterranean (Krijgsmann et al., 1999). Nevertheless, recent studies have demonstrated that in the deep basins, a lateral equivalent of the PLG, mainly consisting of fine grained euxinic sediments, occur (Manzi et al., 2007; Roveri et al., 2008). In this work we present an integrated study based on microbiostratigraphy and magnetostratigraphy of the fine-grained Messinian sediments of the Sant’Agata Fossili marls (SAF) underlying the PLG cropping out at Pollenzo (nearby Alba, Cuneo). In the studied section, the pre and syn-evaporitic sediments (Dela Pierre et al., this volume) consist of: 1) cyclic alternation of homogeneous marly and laminated mudstones (Marne di Sant’Agata Fossili, SAF) up to 125 m thick, interrupted by a 80 m thick slumped interval; 2) two 1.5 m thick selenitic gypsum beds separated by laminated mudstones; 3) a 7 m thick bed of microcrystalline laminated gypsum (MLGB); 4) an alternation of thin gypsum and euxinic mudstones. The integrated study addresses the SAF above the slumped interval, consisting of seven lithologic cycles characterized by the alternation of prominent thin dolomitic layers (10-15 cm thick), and thick homogeneous marls grading into laminated mudstones. The dolomitic layers of the lower cycles show a higher carbonate content with respect to the upper cycles; the homogeneous marls predominate over the laminated mudstones in the lower cycles, whereas in the upper ones the laminated mudstones overcome the homogeneous marls. Also CN assemblages show a cyclic pattern: they are dominated by “small” Reticulofenestra in the mudstones and by “normal” assemblages in the homogeneous marls of each cycle; this correlation has already been described elsewhere in the Mediterranean Basin (Flores, 2005) and attributed to orbital control (precession, 21 ky period). The fossil association recorded in the SAF below the slump contains Globorotalia nicolae and G. conomiozea, and is thus older than 6.72 MA (Hilgen & Krijgsmann, 1999); the sediments above the slump belong to the non distinctive Zone and to the MNN11b/c Zone. The presence of Bulimina echinata in the sediments above the slump demonstrates an age younger then 6.29 Ma (Kouwenhoven et al., 2006); on the other hand, the lack of Turborotalita multiloba, whose last occurrence is younger then 6.09 MA (Sierro et al., 2001; Hilgen & Krijgsmann, 1999), suggests an age younger than the base of Chron C3r for the sediments above the slump. More then two samples per cycles were sampled and proved to carry a reliable magnetic polarity; thermal demagnetization revealed that all the stratigraphic interval comprised between the slump and the first gypsum bed has reverse polarity, thus, according to its micropaleontological content, is correlated with the lower part of chron C3r. Calcareous nannofossil (CN) and foraminifer assemblages revealed that several paleoecological events punctuated the seven cycles preceding the onset of gypsum deposition. They are, from bottom to top: 1) the disappearance of planktonic foraminifers (PF; upper part of cycle 2); 2) the disappearance of benthic foraminifers and the peak abundance of Pontosphaera japonica, coeval with the last occurrence of warm water taxa such as Discoaster spp. and Amaurolithus spp. (upper part of cycle 3); 3) the co-occurring peak abundances of Helicosphaera carteri and Sphenolithus abies (mudstone layer in cycle 4); 4) the final disappearance of calcareous nannofossils and the occurrence of euryhaline fish Aphanius crassicaudus (top of cycle 5, 4.5 m below the first selenitic gypsum bed). According to previous studies carried out in a number of sites all over the western and eastern Mediterranean Basin, the disappearence of PF usually occur one to two cycles below the PLG (Manzi et al. 2007), whereas the striking peak abundance of S. abies slightly postdate this event and occurs just below the gypsum deposition (Manzi et al. 2007) or in the mudstones intercalated to gypsum beds (Cyprus, Wade & Bown, 2006). At Pollenzo, these events occur respectively in the fifth and in the fourth (precessional) cycle below the onset of gypsum deposition. The paleomagnetic analysis does not allow to precisely locate the base of the reverse chron recorded at Pollenzo but, on the base of micropaleontological data, it is interpreted as chron C3r; unfortunately, the absolute age calibration of these paleoecological events is hindered by the absence of the base of Chron C3r in the studied section. Nevertheless our biomagnetostratigraphic data suggest that the base of the selenitic gypsum at Pollenzo is younger than 5.96 MA (as inferred for the base of PLG in the Mediterranean Basin; Krijgsmann et al., 1999) and supports the hypothesis that a lateral equivalent of the PLG, represented by the euxinic pelitic sediments of the upper part of the SAF, is recorded at Pollenzo.