A stratospheric pathway linking a colder Siberia to Barents-Kara Sea sea ice loss

Abstract

Previous studies have extensively investigated the impact of Arctic sea ice anomalies on the midlatitude circulation and associated surface climate in winter. However, there is an ongoing scientific debate regarding whether and how sea ice retreat results in the observed cold anomaly over the adjacent continents. We present a robust "cold Siberia" pattern in the winter following sea ice loss over the Barents-Kara seas in late autumn in an advanced atmospheric general circulation model, with a well-resolved stratosphere. Additional targeted experiments reveal that the stratospheric response to sea ice forcing is crucial in the development of cold conditions over Siberia, indicating the dominant role of the stratospheric pathway compared with the direct response within the troposphere. In particular, the downward influence of the stratospheric circulation anomaly significantly intensifies the ridge near the Ural Mountains and the trough over East Asia. The persistently intensified ridge and trough favor more frequent cold air outbreaks and colder winters over Siberia. This finding has important implications for improving seasonal climate prediction of midlatitude cold events. The results also suggest that the model performance in representing the stratosphere-troposphere coupling could be an important source of the discrepancy between recent studies.

Fig. 1. Observed linkage between cold Siberia and BKS sea ice loss with and without the stratospheric circulation activity.

(A) Regression of SAT (in Kelvin) in December-January-February (DJF) on the normalized early winter [November-December (ND)] BKS sea ice concentration (SIC). (B) Same as (A) after first regressing out the component of SAT variability related to DJF stratospheric vortex anomalies. Both the long-term trend and El Niño–Southern Oscillation (ENSO)–related component in all fields are removed before the regression to emphasize the interannual variability. See Materials and Methods for the data and calculation details. The sign of SIC is reversed here to emphasize the variation related to sea ice loss. Stippling denotes regions that are statistically significant at the 95% confidence level using a two-sided t test for the regression. Red sector in (A) highlights the Siberian cold anomaly region.

Fig. 2. SIC anomalies (%) in the BKS_FL and BKS_TP runs compared to those in the CTRL run.

(A) Monthly SIC changes averaged over the BKS region [highlighted by the black sector in (B)]. (B) Spatial distribution of BKS SIC changes in November. The orange and red lines in (B) denote the ice edge (15% contour of SIC) in the CTRL and BKS_FL (BKS_TP) experiments, respectively.

Fig. 3. SAT, surface wind, and 500-hPa geopotential height anomalies.

(A) SAT (color shadings; in Kelvin) and surface wind (vectors; in m s⁻¹) anomalies in the BKS_FL experiments during DJF. (B) Geopotential height anomalies (500 hPa) [color shadings; in geopotential meters (gpm)] in the BKS_FL experiments during DJF. (C) Same as Fig. 1A but for regression of geopotential height on the BKS SIC index. The contours in (B) and (C) show the climatological DJF 500-hPa geopotential height in the CTRL and the observations, respectively. The contour interval in (B) and (C) is 120 gpm. Stippling denotes regions that are significantly different from zero at the 95% confidence level using a two-sided t test for the color-shading variables. The surface wind anomaly that is significant at the 95% confidence level is shown. Red sector in (A) highlights the Siberian cold anomaly region similar to that in Fig. 1A.

Fig. 4. Contributions of tropospheric and stratospheric pathways.

(A, C, and E) Same as Fig. 3A but for the SAT and surface wind anomalies in the BKS_TP (A) and BKS_SP (C) and the sum of the BKS_TP and BKS_SP (E) runs. (B, D, and F) Same as Fig. 3B but for the geopotential height anomaly in the BKS_TP (B) and BKS_SP (D) and the sum of the BKS_TP and BKS_SP (F) runs. Stippling denotes regions that are significantly different from zero at the 95% confidence level using a two-sided t test for the color-shading variables. Red sector in (E) highlights the Siberian cold anomaly region similar to that in Fig. 1A.

Fig. 5. Siberian daily SAT distribution and CAO events.

(A) PDF of the daily SAT over Siberia in the CTRL, BKS_FL, BKS_TP, and BKS_SP experiments. (B) Same as (A) but for the observations in which the red line is that in the 10 lowest BKS SIC years (referred to as LowSIC) and the gray line is that in all other years (regarded as normal years and referred to as NorSIC) (see Materials and Methods for details). (C) Frequency of CAOs per winter in each experiment. (D) Same as (C) but for the duration of the CAO events (days). The vertical line in (A) and (B) shows the 1σ of daily Siberian SAT below the DJF mean in the CTRL and NorSIC years. In (B), the DJF SAT component related to long-time trend and ENSO (calculated as linear regression on Niño 3.4 index) are first removed in the daily time series of Siberian SAT. The bars in (C) and (D) show the 2.5th to 97.5th percentile interval of the mean values based on 10,000 bootstrap samples. The asterisks in (C) and (D) denote that the difference between the perturbation runs and the CTRL are statistically significant at 90% (*) and 95% (**) confidence levels based on 10,000 times bootstrap samples.