The two major warming phases of the last deglaciation at ∼14.7 and ∼11.5 ka cal BP in Europe: climate reconstructions and AGCM experiments

Abstract

During the last deglaciation two distinct warming phases occurred in the N Atlantic region at ∼14.7 and ∼11.5 ka cal BP. These two shifts are the transitions from (1) GS-2a (Greenland Stadial 2a) to GI-1e (Greenland Interstadial 1e) and (2) GS-1 to the Preboreal. In this study we characterise these two important climate transitions by comparing maps of January and July temperatures for Europe acquired with two independent methods: (1) simulations with the ECHAM4 atmospheric general circulation model in T42 resolution and (2) temperature reconstructions based on geological and palaeoecological data. We also compare estimated lake level changes with simulated P-E (effective precipitation) values. These comparisons enable quantification of the climate change during the two phases. January temperatures increased by as much as 20°C in NW Europe from values between −25°C and −15°C in both GS-2a and GS-1 to temperatures between −5°C and 5°C in both GI-1e and the Preboreal. During July the changes were smaller, as the July temperatures increased in NW Europe by 3–5°C from about 10°C to 15°C in both GS-2a and GS-1 to values of 13°C to 17°C in both GI-1e and the Preboreal. In S Europe the increase in July temperature was less intense. Our analysis suggests that the effective precipitation remained at the same level during the 14.7 ka cal BP transition, whereas a small increase is inferred for some regions for the 11.5 ka cal BP shift. This small effect in effective precipitation is explained by comparable increases in precipitation and evaporation during both transitions. We infer that the strong increase in January temperatures was forced by changes in the N Atlantic Ocean, as the variations in sea surface temperatures and the position of the sea ice margin determined the temperature change over land. The increase in July temperatures was mainly driven by two factors: the increase in insolation and the deglaciation in Scotland and Scandinavia. The insolation changes were gradual (2 to 3 W/m 2) compared to the changes in the N Atlantic Ocean, explaining the relatively small temperature increase during July compared to January. In regions that were deglaciated during the two climate transitions, July temperatures appeared to have increased by up to 10°C. Our results suggest that the registration of the magnitude of the two climate shifts in terrestrial proxy records was geographically different due to the changing environmental conditions; variations in the N Atlantic sea ice limit appear to be the most important. This implies that reconstructed temperature curves from different places in Europe should show different magnitudes. Moreover, it is to be expected that the timing of the major warming phases is spatially different, as this timing is mainly determined by the position of the sea ice and land ice margins relative to the place of interest.