TiPES webinar by Keno Riechers

Title: Changes in stability and jumps in Dansgaard–Oeschger events: a data analysis aided by the Kramers–Moyal equation

Abstract:
Dansgaard–Oeschger (DO) events are sudden climatic shifts from cold to substantially milder conditions in the arctic region that occurred during previous glacial intervals. They can be most clearly identified in paleoclimate records of δ18O and dust concentrations from Greenland ice cores, which serve as proxies for temperature and atmospheric circulation patterns, respectively. The existence of stadial (cold) and interstadial (milder) phases is typically attributed to a bistability of the North Atlantic climate system allowing for rapid transitions from the first to the latter and a more gentle yet still fairly abrupt reverse shift from the latter to the first. However, the underlying physical mechanisms causing these transitions remain debated. Here, we conduct a data-driven analysis of the Greenland temperature and atmospheric circulation proxies under the purview of stochastic processes. Based on the Kramers–Moyal equation we present a one-dimensional and two-dimensional derivation of the proxies' drift and diffusion terms, which unravels the features of the climate system's stability landscape. Our results show that: (1) in contrast to common assumptions, the δ18O proxy results from a monostable process, and transitions occur in the record only due to the coupling to other variables; (2) conditioned on δ18O the dust concentrations exhibit both mono and bistable states, transitioning between them via a double-fold bifurcation; (3) the δ18O record is discontinuous in nature, and mathematically requires an interpretation beyond the classical Langevin equation. These findings can help understand candidate mechanisms underlying these archetypal examples of abrupt climate changes

Presenter: Keno Riechers