资源环境科技发展态势分析平台

Two coupled climate models, CESM1 and CanESM2, are used to isolate the climate response to Arctic sea ice loss (high-latitude warming). The sea ice loss and radiative forcing protocols differ between the sets of experiments. This response is compared to the remaining climate change signal, which is dominated by low-latitude warming. Some aspects of the wintertime circulation response to sea ice loss are remarkably robust: warming over much of the high latitude and midlatitude; weak cooling over eastern Eurasia, strengthening of the Aleutian Low and Siberian High, equatorward intensification of the lower tropospheric winds, and increased precipitation over high latitudes. Pattern scaling separates the parts of the response that scale with low-latitude warming and with sea ice loss. The thermal response patterns, for both sea ice loss and low-latitude warming, are similar between the models. However, the circulation response patterns that scale with low-latitude warming differ between the models. Preliminary evidence shows that these conclusions apply to other models driven by distinctive sea ice loss protocols.

Plain Language Summary Arctic sea ice is melted in two climate models in such a way as to isolate its impact on the atmosphere from the general climate change response. We find that sea ice loss alone generates changes to atmospheric circulation patterns that are consistent between different climate models, but that the remaining signal, dominated by warming in the tropics, depends on the model used. As a result, the full circulation response to climate change remains uncertain. In contrast, the temperature and precipitation responses to both Arctic sea ice loss and tropical warming are consistent between the models. Preliminary results show that these conclusions apply to other models and are not sensitive to the method used to melt the sea ice.