Simulating the Holocene climate evolution at northern high latitudes using a coupled atmosphere-sea ice-ocean-vegetation model

The response of the climate at high northern latitudes to slowly changing external forcings was studied in a 9,000-year long simulation with the coupled atmosphere-sea ice-ocean-vegetation model ECBilt-CLIO-VECODE. Only long-term changes in insolation and atmospheric CO2 and CH4 content were prescribed. The experiment reveals an early optimum (9–8 kyr BP) in most regions, followed by a 1–3°C decrease in mean annual temperatures, a reduction in summer precipitation and an expansion of sea-ice cover. These results are in general agreement with proxy data. Over the continents, the timing of the largest temperature response in summer coincides with the maximum insolation difference, while over the oceans, the maximum response is delayed by a few months due to the thermal inertia of the oceans, placing the strongest cooling in the winter half year. Sea ice is involved in two positive feedbacks (ice-albedo and sea-ice insulation) that lead regionally to an amplification of the thermal response in our model (7°C cooling in Canadian Arctic). In some areas, the tundra-taiga feedback results in intensified cooling during summer, most notably in northern North America. The simulated sea-ice expansion leads in the Nordic Seas to less deep convection and local weakening of the overturning circulation, producing a maximum winter temperature reduction of 7°C. The enhanced interaction between sea ice and deep convection is accompanied by increasing interannual variability, including two marked decadal-scale cooling events. Deep convection intensifies in the Labrador Sea, keeping the overall strength of the thermohaline circulation stable throughout the experiment.