| 11 Mar 2026
Teleconnection processes linking snow cover variability over western Siberia to enhanced early-winter cold waves in South Korea
Abstract. This study examines thermodynamic teleconnection processes to identify the drivers of the increasing frequency of December cold waves in South Korea over the past 46 years. The identified teleconnection is associated with land–atmosphere interactions triggered by reduced October snow cover (SC) over western Siberia. By refining the cold wave advisory criteria of Korea Meteorological Administration, we calculated cold day frequencies. We applied k-means clustering and principal component analysis to extract dominant time series representing nationwide cold day variability. Analyses based on automated synoptic observing system and ERA5 reanalysis data revealed that October SC reduction over western Siberia initiates a positive snow–ice albedo feedback, leading to enhanced surface solar radiation absorption and increased turbulent heat fluxes. These thermodynamic anomalies lead to atmospheric thermal expansion during October through December, thereby strengthening a December anticyclonic anomaly centered over western Siberia. The intensified northerly winds along the eastern flank of this anomaly established a teleconnection pattern that amplifies early-winter cold waves in South Korea. Overall, these results indicate that October SC may serve as a critical precursor to the recent intensification of early-winter cold waves, providing a scientific foundation for policies to adapt and mitigate the socioeconomic impacts of climate change.
Review comments on Kim and Lee “Teleconnection processes linking snow cover variability over western Siberia to enhanced early-winter cold waves in South Korea”
General comment:
This manuscript examines the relationship between early-winter cold waves in South Korea and antecedent snow-cover variability over western Siberia. Using statistical analyses of station observations and reanalysis data, the authors identify reduced October snow cover over western Siberia as a precursor to enhanced cold-wave activity in December. They propose that associated land-atmosphere interactions and resulting circulation changes provide the physical mechanism linking western Siberian snow anomalies to cold conditions over South Korea.
Overall, the manuscript is well organized and generally easy to read. The logical flow is clear, and the discussion proceeds in a coherent and straightforward manner. The study may also have some novelty, particularly in its use of a revised cold-wave indicator for South Korea and in its focus on western Siberian snow cover, a factor that has received comparatively limited attention in previous studies of East Asian winter climate.
However, I also have important concerns precisely regarding these two aspects. The first concern is the scientific and meteorological interpretation of the newly introduced cold-wave definition. The second is the independence of the western Siberian snow-cover variability from the sea-ice variability in the nearby region. I provide more detailed comments below, but in my view, publication in this journal would be difficult to recommend unless these issues are properly addressed and the discussion is strengthened accordingly.
Major comment 1.
The authors have pointed out limitations in both the KMA cold day index and the cold wave advisory criteria, and then developed a revised cold day index for use in this study. I appreciate the attempt to improve the existing index. However, I think more careful consideration is needed regarding the interpretation of combining conditions (1) and (2) within a single index. As I understsnd it, condition (1) is intended to capture cases with a rapid temperature drop, even when the cold air itself is not particularly intense, whereas condition (2) is intended to capture cases with persistent and strong cold air. Since both types of events may have substanital social impacts, defining a warning criterion in terms of satisfying either one of the two conditions may be reasonable from an operational or public-advisory perspective.
However, from a scientific and meteorological viewpont, the two types of events may be governed by rather different processes. For example, a day when the minimum temperature drops rapidly from 13°C to 3°C and a day when a very low temperature of around −15°C persists from the previous day are likely associated with substantially different meteorological conditions. In the present study, however, this distinction is not explicitly made. Instead, the authors construct a cold-day-based index (through the PC time series) and use it for regression and composite analyses without separating these two categories.
I therefore think that the revised cold day index requires a more careful discussion. For example, among the cold days identified by the revised criteria, what are the relative contributions of conditions (1) and (2)? My impression is that condition (2) may dominate, but if that is the case, the rationale for including condition (1) should be more clearly discussed. One possible justification might be that condition (2)-type events are relatively well represented in climate/forecast models, whereas condition (1)-type events are more difficult to capture, so including condition (1) could potentially improve prediction or forecasting. Even in that case, however, it may still be preferable to analyze conditions (1) and (2) separately.
Major comment 2.
I find it interesting that this study focuses on western Siberian snow-cover variability, since relatively few previous studies have examined its relationship with East Asian winter climate. In fact, the analysis of the surface energy budget presented here clearly shows enhanced upward heat transfer from the land surface to the atmosphere over regions of reduced snow cover in northern western Siberia during October, which strongly suggests an active role of land-surface processes.
At the same time, however, this active region is located close to the Barents–Kara Seas (BKS), where sea-ice reduction from autumn to winter is among the largest in the Arctic. For this reason, I think the manuscript needs a more careful discussion of the extent to which the proposed role of western Siberian land processes is independent of, or coherent with, variability in nearby sea ice. In fact, a large number of previous studies have suggested that sea-ice reduction in the BKS can lead to cold conditions over East Asia. In addition to the studies already cited by the authors, several mechanisms have been proposed, such as the northward shift of cyclone tracks (Inoue et al., 2012), a staitionary Rossby wave response to surface heating (Honda et al., 2009), and a shift toward the negative phase of the AO regime (Nakamura et al., 2015). Although these mechanisms differ in detail, they all involve enhanced anticyclonic circulation over Siberia (i.e., strengthened Siberian High) as a pathway toward East Asian cooling. In this respect, the mechanism proposed in the present manuscript appears analogous to those earlier interpretations.
Therefore, the authors should discuss much more carefully whether the snow-cover variability over western Siberia acts independently of sea-ice variability, or whether the two vary coherently and jointly contribute to the atmospheric response. For example, in Fig. 5, it may be useful to calculate regression anomalies with respect to one standard deviation of the National-PC1 time series, using ERA5 also over the ocean, so that the magnitudes of surface flux anomalies over land and ocean can be directly compared. If this is done, how large is the land contribution relative to the oceanic flux anomalies that may be associated with sea-ice loss?
More directly, it may also be useful to add to Fig. 3b a time series of October snow-cover area over western Siberia, together with an index of BKS sea-ice area in September or October, and then examine their correlations with the National-PC1 and cluster time series. If the authors can demonstrate either that western Siberian snow-cover variability is largely independent of sea-ice variability, or that even if they vary coherently, the land-atmosphere interaction proposed here amplifies the impact in a physically meaningful way, then the value of this study would become much clearer.
In this context, previous approaches may be useful as references. For example, statistical removal of the component coherent with sea-ice variability (McCusker et al., 2016) or climate-model experiment designs that isolate the sea-ice impact (Nakamura et al., 2019) may provide helpful guidance. At the very least, the manuscript should discuss its results in light of such previous studies.
Minor comments
References
Honda, M., Inoue, J. & Yamane, S. Influence of low Arctic sea-ice minima on anomalously cold Eurasian winters. Geophys. Res. Lett. 36, L08707 (2009).
Inoue, J., Hori, M. E. & Takaya, K. The role of Barents sea ice in the wintertime cyclone track and emergence of a Warm-Arctic Cold-Siberian anomaly. J. Clim. 25, 2561–2568 (2012).
Nakamura, T. et al. A negative phase shift of the winter AO/NAO due to the recent Arctic sea-ice reduction in late autumn. J. Geophys. Res. Atmos. 120, 3209–3227 (2015).
McCusker, K. E., Fyfe, J. C. & Sigmond, M. Twenty-five winters of unexpected Eurasian cooling unlikely due to Arctic sea-ice loss. Nat. Geosci. 9, 838–842 (2016).
Nakamura, T., Yamazaki, K., Sato, T. et al. Memory effects of Eurasian land processes cause enhanced cooling in response to sea ice loss. Nat Commun 10, 5111 (2019).