Significance of uncertain lags between the transition onsets in different Greenland ice-core proxies around Dansgaard-Oeschger events

For previous glacial intervals, concomitant shifts in different proxy records from Greenland ice cores indicate repeated abrupt climate transitions comprising among others abrupt warming, sudden reorganization of the atmospheric circulation, retreat of perannial sea ice, and increase in local precipitation. The physical mechanism underlying these so-called Dansgaard-Oeschger (DO) events remains debated. In order to constrain potential triggering mechanisms former studies have deduced the progression of the aforementioned changes at the onset of DO events by investigating the phasing of respective 5 transitions in paleoclimate proxy records. A recent analysis of DO events evidenced in the 60 10kyr BP period of the NGRIP and NEEM ice core proxy records found delayed transitions in Na concentrations and δO values compared to corresponding transitions in Ca concentrations and the annual layer thickness by about one decade. This is interpreted as a temporal lag of sea ice retreat and Greenland warming with respect to a synopticand hemispheric-scale atmospheric reorganization at the onset of DO-events. However, the explanatory power of these results is limited by the uncertainty of the transition onset detec10 tion in noisy proxy records. Here, we extend previous work by testing the significance of the reported lags with respect to the null hypothesis of simultaneous transitions in the different proxy variables. For this aim, we regard the identified transition lags between each pair of proxies as outcomes of a repeated random experiment. If detection uncertainties are averaged out at the level of individual transitions lags, temporal delays in the δO and Na transitions with respect to Ca and the annual layer thickness are indeed pairwise statistically significant. In contrast, under rigorous propagation of uncertainties the samples are 15 not significant across all considered pairs of proxies. We thus confirm the previously reported tendency of delayed transitions in the δO and Na concentration records. Yet, substantial uncertainties in the determination of the transition onsets prevent to rule out that this tendency arises by chance. Hence, under the assumption that all DO events followed the same physical mechanism and that the interpretations of the different climate proxies hold true, the analyzed set of DO events cannot serve as evidence for systematic lead-lag relationships. Our results highlight the importance of rigorous uncertainty treatment in the 20 analysis of paleoclimate records. Copyright statement. TEXT