Juliane Mariesvej 30, 2100 København Ø
PhD defense by Hannah Kleppin
An unforced pre-industrial control simulation of the Community Climate System Model 4 (CCSM4) is found to have Greenland warming and cooling events that resemble Dansgaard-Oeschger-cycles in pattern and magnitude. With the caveat that only 3 transitions were available to be analyzed, we find that the transitions are triggered by stochastic atmospheric forcing. The atmospheric anomalies change the strength of the subpolar gyre, leading to a change in Labrador Sea sea-ice concentration and meridional heat transport. The changed climate state is maintained over centuries through the feedback between sea-ice and sea-level pressure in the North Atlantic. The full evolution of the anomalous climate state depends crucially on the climatic background state.
We present evidence for an El Niño Southern Oscillation (ENSO) -like mode that varies in tandem with the Greenland cooling and warming phases. Greenland cold phases correspond to dominant El Niño-like conditions, increased variance of ENSO and a displacement of El Niños (La Niñas) farther to the east (west). This ENSO-like mode is correlated with sea-level pressure anomalies over the Labrador Sea. ENSO-like variability with said characteristics on centennial scales is also present in a different pre-industrial control simulation. The absence of Greenland climate transitions in this simulation suggest that the centennial scale climate variability presented here, originates in the tropics.
In the southern hemisphere the only significant changes in association with the Green land cooling and warming phases are out-of-phase sea level pressure anomalies over the Weddell and Ross Sea. We explore why a strong interhemispheric coupling is absent and demonstrate that increased atmospheric CO 2 – as observed for the cold phases of Dansgaard-Oeschger-cycles – improves the southern hemisphere temperature response.
Net snow accumulation changes between the Greenland cold and warm phases are comparable in pattern and magnitude to those induced by orbital changes. This suggests that internal climate variability has the potential to favour a glacial inception even though orbital conditions might not yet be in optimal conditions – based on Milankovitch theory. We discuss potential effects of glacial boundary conditions for the above described climate changes.