Refinement of the Messinian APTS from sedimentary cycle patterns in the lacustrine Lava section (Servia Basin, NW Greece)

Abstract

A high-resolution cyclostratigraphy and magnetostratigraphy is presented for the Messinian lacustrine Lava section from the Servia Basin in NW Greece, constraining more precisely the absolute ages of magnetic polarity subchrons C3An.1n and C3An.2n. The section contains 15 distinct sedimentary cycles of alternating dark- and light-coloured marls, while the gamma-ray attenuation record reveals an additional five to six cycles. The cycles in the lower half of the section are on average 5.3 m thick, as opposed to the cycles in the upper part, which have an average thickness of 3.1 m. Palynological results define the lithological alternations in both the lower and upper cycles in terms of periodic changes in humidity, where the light marls represent the humid periods and the dark marls the relatively dry periods. Changes in cycle thickness and shifts in average gamma-ray values suggest a rather abrupt decrease in sedimentation rate at ∼60 m in the section. This is confirmed by the magnetostratigraphy, which recorded four reversals, which – given the biostratigraphic constraints from the Lava locality – could be correlated unambiguously to subchrons C3An.1n and C3An.2n of the geomagnetic polarity time scale. With this magnetostratigraphic time control, the average duration of the cycles can be calculated to be constant in the entire section, and similar to precession. The astronomical origin of the cycles is confirmed by the results of spectral analyses of gamma-ray and susceptibility time series. The sedimentary cycles in the upper part of the Lava section are unambiguously tuned to insolation using the typical clustering of the cycles that follows the eccentricity cycle. The filtered gamma-ray record centred at 41 kyr confirms the tuning in the upper part and allows tuning of the lower part. The tuning results in accurate ages for the sedimentary cycles and polarity reversals that confirm the astronomical tuning of [Krijgsman et al., Nature 400 (1999) 652–655], but define more precisely the astronomical polarity time scale.