This study investigates how different types of downward-propagating sudden stratospheric warming (SSW) events are associated with distinct regional surface cold extremes and classifies them based on their surface impacts, using reanalysis data and CESM-WACCM simulations. According to the abstract, introduction and conclusion, the paper aims not only to classify cases of surface cold extremes associated with SSWs, but also to explain the mechanisms behind how different types of surface cold extremes are caused by difference in SSWs. To this end, the study presents a comprehensive analysis using both observations and CESM-WACCM. One of the main contributions of the paper is the classification of downward-propagating SSW events based on the regional characteristics of the associated surface cold extremes, followed by a global analysis of the related dynamical fields. However, the main analyses are mostly descriptive, such as “this type tends to have these features,” without sufficient efforts to clarify the causal connections between processes. This creates a mismatch between the intended positioning of the paper and the actual description of the results, which may confuse readers about the central message of the paper. Moreover, even if the main purpose was the classification of SSW types, there are places where the descriptions overinterpret differences between the types that may naturally result from the way they are classified.

Nevertheless, I believe the paper has potential, since the analyses are extensive and well-organized. It would be publishable as a descriptive classification study if the authors revise the manuscript to focus on the meaningful dynamical differences revealed by the classification, while excluding the differences that naturally arise from the way the types are defined.

Major Comments

Unclear objective: classification or mechanism?

If the goal is to understand the cause of different SSW surface impacts, the paper needs to move beyond descriptive comparisons and provide a clearer explanation of how variables interact across steps. If the goal is to classify the types, the paper should clarify up front that it is intended as a classification study based on observed differences.

Insufficient interpretation and unclear physical connections

The authors should avoid wording that may be misinterpreted as implying a causal relationship between features that simply co-occur. It would be helpful to clarify which parts of the discussion are supported by physical reasoning and which are more descriptive.

1. Some differences reported in the results seem to directly stem from how the cases were defined, but this point is not acknowledged. For example, the EA-type cold extremes are defined to cover a broader longitudinal range. Since the NAM index more reflects zonal-mean changes, it is not surprising that EA cases show a stronger and persistent NAM signal than NA cases. This might be a natural result of the classification.
2. The paper tends to overinterpret co-occurring patterns as physical links. The paper lists many local differences across types but does not sufficiently explain whether these patterns are meaningfully connected.
3. the paper lists a number of detailed local differences between SSW types and presents them as if they are physically meaningful, and this seems to depend on the statistical significance of the composite. However, it is not always evident whether these differences are statistically significant. There are figures which show that composite for NDW show widespread dotted areas of statistically significant values (e.g. Fig. 6d), and the statistical testing method is not clearly explained. This raises the concern that some of these apparent differences may not be meaningful physical distinctions, especially in relation to the difference in SSW downward impact.
4. The anomalies suggested as precursors are not clearly distinguished as being part of the SSW-related signal or independent tropospheric variability.

Methodology

1. Even if the main objective of the study was the type classification, more careful methodological design would be necessary. For example, EA DW cases already show strong cold anomalies before the SSW occurred, which raises questions about whether the cold extremes are truly caused by SSW. If the anomalies are computed based on a fixed climatology, long-term signals may be included. However, there is also no clear method for determining whether post-SSW cold events are actually caused by the SSW. This makes it difficult to rule out the possibility that some tropospheric anomalies (or anomalies of longer timescale) developed independently of the SSW.
2. How CESM-WACCM contributes to the overall interpretation is not clear. Is it simply to provide more cases, or to test mechanisms? For example, when CESM-WACCM and observational results differ, the physical meaning of the differences is not discussed. Although the study analyzes both ERA5 and CESM-WACCM, the interpretation is mostly based on ERA5.

Lack of structural and editorial refinement

1. Although the analysis is limited to the Northern Hemisphere, this is not explicitly stated in the methodology.
2. The structure of the methodology section also lacks consistency. For example, the CESM-WACCM model description is under the subsection “2.1 Reanalysis Data.”
3. While the study defines four geographical domains (Europe, Asia, the United States, and Canada) the classification later becomes just two types (NA and EA), without explanation of the criteria used to group these regions. Furthermore, there is no explanation for why those particular domains were chosen for each region.
4. A few grammatical errors are present in the manuscript.