Memory effects of Eurasian land processes cause enhanced cooling in response to sea ice loss
- 1Faculty of Environmental Earth Science, Hokkaido University, Sapporo, JAPAN
- 2Faculty of Science, Niigata University, Niigata, Japan
Amplified Arctic warming and its relevance to mid-latitude cooling in winter have been intensively studied. Observational evidence has shown strong connections between decreasing sea ice and cooling over the Siberian/East Asian regions. However, the robustness of such connections remains a matter of discussion because modeling studies have shown divergent and controversial results. Here, we report a set of general circulation model experiments specifically designed to extract memory effects of land processes that can amplify sea ice–climate impacts. The results show that sea ice–induced cooling anomalies over the Eurasian continent are memorized in the snow amount and soil temperature fields, and they reemerge in the following winters to enhance negative Arctic Oscillation-like anomalies. The contribution from this memory effect is similar in magnitude to the direct effect of sea ice loss. The results emphasize the essential role of land processes in understanding and evaluating the Arctic–mid-latitude climate linkage.
How to cite: Nakamura, T., Yamazaki, K., Sato, T., and Ukita, J.: Memory effects of Eurasian land processes cause enhanced cooling in response to sea ice loss, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2284, https://doi.org/10.5194/egusphere-egu2020-2284, 2020
Comments on the display material
AC: Author Comment | CC: Community Comment | Report abuse
Hi Tetsu, interesting results!
Have you tested how sensentive your results are to the ensemble size? LICE and LICE* will differ due to internal varaibility (non-identical initial conditions). Do you think 100 members is sufficient to separate "memory" effects from internal variability? Could another way to interpret your results be that the precise magnitude of sea-ice-induced Eurasian cooling is difficult to quantify with 100 members, and can appear to be larger/smaller due to the presenece of internal varaibility?
It would be informative to subsample the simulations and see if the subsamples deviate by as much as the diagonsed "memory" effect. This would provide some measure of whether or not the memory effect is larger than could arise due to sampling error.
Cheers, James
Hi James, thank you for your comment.
At least based on Mori et al 2014 NatGeo, sample size of 100 will be enough to minimize the effect of internal vliability.
However, Mori-san's estimation is based on direct response to the sea ice change (i.e., similar to LICE* - HICE in our study), which is not involving memory effect longer than annual time scale.
Therefore, if the memory effect would be additional factor of non-linearity through magnifying the atmospheric anomaly, which could be due to both sea ice impacts and internal variability, it might require more sample size. I have not yet examine it.
Thank you for your suggestions. I will test sample size issue.
Thanks for the quick reply. We know from some of the emerging PAMIP results that aspects of the response to sea ice can vary in magnitude (and in some cases, in sign) between 100-member ensembles of the same model. It surprised us to be honest; we thought 100 members would be sufficient, but unfortunately, that is not always the case!
Thank you, James.
So, it suggests that impacts of sea ice changes are smaller than we expected from the studies so far?
I suppose that some feedback mechanisms like land process, stratopshere, and of course air-sea interaction amplify the direrct response to the sea ice loss.
It is an interesting thought. Although the PAMIP models include a coupled land surface, the experiments are (mostly) 1 year long, so might miss any longer-than-annual "memory" effects.