David De Vleeschouwer , Donald Penman, Simon D'haenens, Fei Wu, Thomas Westerhold, Maximilian Vahlenkamp, Carlotta Cappelli, Claudia Agnini, Wendy Kordesch, Daniel King, Robin van der Ploeg, Heiko Pälike, Sandra Kirtland Turner, Paul Wilson, Richard Norris, Steven Bohaty, Pincelli Hull

Cyclostratigraphy and astrochronology are now at the forefront of geologic timekeeping. While this technique heavily relies on the accuracy of astronomical calculations, solar system chaos limits how far back astronomical calculations can be performed with confidence. High-resolution paleoclimate records with Milankovitch imprints now allow reversing the traditional cyclostratigraphic approach: Middle Eocene drift sediments from Newfoundland Ridge are exceptionally well-suited for this purpose, thanks to high sedimentation rates and distinct lithological cycles. Per contra, the stratigraphies of Integrated Ocean Drilling Program Sites U1408-U1410 are highly complex with several hiatuses. Here, we build a two-site composite and construct a conservative age-depth model to provide a reliable chronology for this rhythmic, highly-resolved (<1 kyr) sedimentary archive. Astronomical components (g-terms and precession constant) are extracted from proxy time-series using two different techniques, nevertheless producing similar results. We find astronomical frequencies up to 4% lower than reported in astronomical solution “La04”. This solution, however, was smoothed over 20-Myr intervals, and our results therefore provide constraints on g-term variability on shorter, million-year timescales. We also report first evidence that the g4-g3 “grand eccentricity cycle” may have had a 1.2-Myr period around 41 Ma, contrary to its 2.4-Myr periodicity today. Our median precession constant estimate (51.28 ± 0.56”/year) confirms earlier indicators of a relatively low rate of tidal dissipation in the Paleogene. Newfoundland Ridge drift sediments thus enable a reliable reconstruction of astronomical components at the limit of validity of current astronomical calculations, extracted from geologic data, providing a new target for the next generation of astronomical calculations.