Review of “21st-Century Strong Wind and Heavy Precipitation Hazards in the Asian Monsoon Region Driven by Mesoscale Convective Systems: Climatology, Variability, and Trends” by Feng et al.

This paper presents a valuable and comprehensive study on the climatology, variability, and trends of Mesoscale Convective Systems (MCSs) over the Asian Monsoon Region (AMR) from 2001 to 2023. The manuscript is well-written and concise, making it a pleasure to read. There remain a few points that require further clarification and polish.

Major comments

1. The authors use a simplified spatial-temporal overlap method to associate hazards with MCSs: if a hazard event occurs within the “impact area” of an MCS on the same day, it is considered MCS-associated. While operationally straightforward, this approach introduces significant uncertainties and potential overestimation of MCS contributions, particularly for hazards with different spatial and temporal scales.
   1. The MCS impact area is defined via a circumscribed circle of the MCS’s bounding rectangle (Fig. 2), which may not accurately represent the true spatial extent of convective hazards.
   2. The analysis uses daily resolution for all hazards. MCSs evolve rapidly (lifetimes of hours), and associating a daily-aggregated hazard with an MCS that may have occurred only briefly within that day can lead to false positives. This is especially problematic for short-lived events like tornadoes.
2. The identification of MCSs relies heavily on Tb thresholds (233 K for CCAs, 225 K for cold cores) and an area threshold of 3000 km². While the authors justify these thresholds based on precipitation efficiency and computational trade-offs (Fig. 3), the sensitivity of the results to these choices is not thoroughly tested. The authors are not advised to pick “the best” set of thresholds, but at least they need to provide the error bars of their results.

Minor comments

1. The delineation of the "AMR" domain warrants clarification. As currently defined (Line 185), it encompasses South Asia while omitting Japan and the Korean Peninsula in East AMR. To avoid confusion and improve geographical precision, refining the terminology for this study region would be beneficial.
2. Line 21. Change “upward” to “increasing”.
3. Line 62. This statement is true maybe only for China.
4. Figures 1 and 2. The two figures can be merged.
5. Line 83. The precipitation attribution does not make sense. Assuming a squall line passing through, the diameter of influence would be the length of the linear system.
6. Line 107. Please keep consistent in putting space between number and unit.
7. Line 107. Just say “CCAs with cold cores”.
8. Line 111. Any sensitivity tests on varying overlapping thresholds?
9. Line 112. Same, any sensitivity in duration thresholds?
10. Line 155. The authors may list the possible data sources of the reports.
11. Line 188. How are TCs removed, both spatially and temporally?
12. Line 191. Then how is this product matched to the hourly IMERG rainfall data?
13. Line 196. The accumulated MCS-hour would be interesting too.
14. Lines 189-214. The authors may summarize all the MCS metrics into a table.
15. Line 232. Where are the “some areas” located?
16. Line 296. How about “fraction” instead of “proportion” to be consistent with MCSPF.
17. Line 328. There is an inherent misalignment in the temporal resolution of the two key datasets: daily hazard reports versus half-hourly MCS tracks. This leads to an approximation where any hazard on the same day is linked to an MCS, even if it preceded the MCS initiation (e.g., hail before MCS onset). It is suggested that the authors more clearly articulate this as a methodological limitation, discussing its potential to dilute the specificity of the association between MCSs and convective hazards.
18. Line 362. What about the seasonality?