Climate change is occurring rapidly in the Arctic, and climate models project further extensive changes later in this century. Global warming is enhanced at high northern latitudes where the Arctic surface air temperature has warmed at twice the rate of the global average in recent decades - a feature called Arctic amplification. The rapidly shrinking Arctic cryosphere has become the most visible aspect of climate change. On the other hand, North America, Europe, and East Asia have experienced anomalously cold winters with record high snowfalls during recent winters, associated with a negative phase of the North Atlantic Oscillation. These cold winters had increased occurrences of cold air outbreaks from high latitudes. These rapid changes in both the snow and sea ice extent observed in the northern regions may affect the atmospheric circulation and extreme weather events down to middle latitudes. There a need to better understand the connections between the decline in summer Arctic sea ice and the build-up of the Eurasian snowpack in the following autumn, and their impacts on the winter circulation. The overall goal of the SNOWGLACE project is to improve our understanding of the complex interconnections and feedbacks in the Arctic region, their role in the rapid and unprecedented climate changes in the north, their impacts on extreme weather events in mid- to high-latitudes. It also aims to assess the impact of changing sea ice and snow cover on the predictability of extreme weather events on the seasonal time scale.
New results from the project are that climate re-analyses over both the recent Arctic warming and the early 20th century Arctic warming show similar linkages between sea ice reduction over the Barents Sea and build-up of snow over parts of Siberia. The Barents Sea has also been identified as a key region where the June sea-ice variability exerts the most significant impacts on the East Asian summer rainfall. A reduction in June sea ice excites persistent anomalous upward air motion due to strong near-surface thermal forcing, which further triggers an east-west Rossby wave train known as the Silk Road pattern, extending to East Asia.
To demonstrate that snow initialisation influences surface temperature over Eurasia and can improve model prediction skill, we carried out a large ensemble seasonal simulations with the Norwegian Climate Prediction Model. NorCPM consists of the Norwegian Earth System Model (NorESM) and advanced initialisation and assimilation techniques to make initialised predictions. The results reveal that snow initialisation improved the prediction skill for wintertime surface temperature over Eurasia, up to 2 months in advance.
The benefits and key practical implementations for the assimilation of sea ice concentration in the NorCPM have been investigated in a perfect model framework. First, it is found that a flow-dependent, strongly coupled ocean?sea ice assimilation method outperforms weakly coupled (sea ice only) assimilation. Extending the ocean updates below the mixed layer is slightly beneficial for the Arctic hydrography. Second, using a multicategory of sea ice greatly reduces the errors in the ice state.
The role of snow depth as a predictor of temperature one month ahead in the Northern Hemisphere has been elucidated for different seasons, in comparison to other predictors like soil temperature or soil moisture. Snow depth is an important mediator when, and where, its interannual variability is large: in winter at midlatitudes and in spring and autumn at high latitudes.
We found an important role of the Atlantic Multidecadal Oscillation (AMO) in modulating the atmospheric response to the declining Arctic sea ice, with Eurasian snow cover and the stratosphere playing a role in the linkages. We discovered that sea surface temperatures over the Atlantic modulate the atmospheric response to the sea ice decline. During a cold AMO phase, increased Ural blocking activity and associated northerly cold air advection and moisture transport from the Arctic leads to an extended snowpack and a cold continent anomaly over Eurasia in December. The enhanced upward propagation of planetary waves into the stratosphere over the Siberian?Pacific sector leads to a weakened stratospheric polar vortex and a negative Arctic Oscillation (AO) phase at the surface in February.
We also found that the North American snow cover influences the Eurasian surface temperature downstream with a 1-month lag, through a mechanism involving sea surface temperatures and synoptic weather systems over the North Atlantic. Another result is that the stratospheric sudden warming of February 2018, linked to cold air outbreaks over Europe, Asia and North America was associated with a precursory ridge over the Urals and high snow cover over Eastern Eurasia, the latter contributing to intensifying upward planetary wave propagation.?
We demonstrated the impact of snow on surface temperature persistence and forecasts, and the benefits of the assimilation of sea ice concentration in the Norwegian Climate Prediction Model. The project has served to further develop the capabilities of NorCPM, most notably the snow and sea ice initialisation. We found an important role of the sea surface temperatures in modulating the atmospheric response to the declining Arctic sea ice.
Thanks to the support of this project, the project leader has been able to initiate a first international SNOWGLACE initiative about the impact of snow initialisation on seasonal forecasts, sponsored by the WMO WCRP working group on Seasonal-to-Interannual Prediction (WG-SIP) (http://www.wcrp-climate.org/index.php/wgsip-overview), and is one of the three scientific initiatives from WG-SIP.
The project has cemented a strong scientific collaboration between NILU and several institutes in China, in particular with the Nansen-Zhu Centre.
Arctic summer sea-ice extent exhibits a sharp declining trend, and the induced atmospheric warming at high latitudes in autumn has potentially important consequences for the climate of Europe and Asia. On the other hand, North America, Europe, and East Asia have experienced anomalously cold winters with record high snowfalls during some recent winters. The autumn snow cover over Eurasia is increasing, while in spring, the snow decline at high northern latitudes is the largest cryospheric change in terms of spatial extent.
To examine the impact of the changing Arctic cryosphere, both land and sea ice, as a predictor of Eurasian climate at the seasonal time scale using advanced dynamical prediction systems is the central theme of this proposal. A key geographical focus will be Europe and Asia, with a strong involvement of partners in Japan, Korea and China. We will focus on the sea ice retreat influence onto the Eurasian snowpack in autumn as a link to winter weather anomalies, and address the cryospheric influence on mid-latitude extreme weather events, such as cold air outbreaks, blockings or heat waves.
To this end, we will carry out ensemble seasonal simulations with NorCPM and coordinate an international initiative to investigate the effect of snow on sub-seasonal to seasonal forecasts, and other simulations with realistically initialised sea ice and snow. The results will be examined in the context of a multi-model framework.
The implications of projected accelerating Arctic sea ice cover disappearance in a warming world are wide ranging. Hence, a secondary objective is to allow Norwegian scientists to participate and play a leading role in international programmes on cryosphere-climate interactions, for example by promoting the use of the Norwegian prediction and climate models in multi-model assessments