Bypassing the Suess-effect: Age determination of charcoal kiln remains using OSL dating

Ancient forests all over Europe often preserve remains of (pre)historic charcoal production (kilns), informing on past forest composition and anthropogenic woodland exploitation. So far, the chronology of these features has been entirely derived from 14C-dating of associated charcoal. Though generally successful, 14C-dating cannot provide meaningful age information for post-1650 CE features, while most of the archaeological remains of charcoal kilns date to this period. Here, we investigate the potential of quartz-based optically stimulated luminescence (OSL) dating of sediments that were heated during charcoal production as a more precise alternative tool for age determination of these kilns. Using a set of 14 samples collected from both relict charcoal kilns and the underlying parent material, we first document the quartz OSL characteristics in terms of the procedural tests commonly used to assess the appropriateness of the single-aliquot regenerative-dose (SAR) protocol. The results suggest that the measurement procedure allows determining equivalent doses both accurately and precisely. We then demonstrate how the linear part of the dose-response curve can be exploited to examine completeness of resetting and/or post-depositional disturbance without compromising instrument time, precision and laboratory accuracy. Optical ages derived from these distributions are consistent with the available independent (14C) age information for the charcoal kilns. We conclude that OSL can provide an accurate and viable alternative to 14C and will be especially useful to constrain post-1650 CE features in time. Our results also suggest that the approach may allow establishing relative chronologies with a time resolution of 2–4%. •Quartz-OSL dating is applicable to heated sediments associated with relict charcoal kilns.•Good agreement between OSL and 14C dates for pre-1650 CE features confirms accuracy.•Post-1650 CE features can be dated with significantly higher precision.•Large datasets can be generated using a simplified procedure, not affecting precision or accuracy.•Potential pitfalls and possibilities are highlighted.