Development Processes of the East Asian Cyclones over the Korean Peninsula

The development processes of the extratropical cyclones passing the Korean Peninsula during the period of 1979– 2017 are quantitatively evaluated in the potential vorticity (PV) perspective. A feature tracking algorithm is applied to the ERA-Interim reanalysis data to objectively identify the distinct northernand southern-track (NT and ST) cyclones affecting the region in the cold season. The dynamic and thermodynamic contributions to the development of these two categories of cyclones are then comparatively assessed in terms of the relative vorticity tendency resulting from the PV tendency inversion. 10 It is quantified through inversion that the NT cyclones develop 87.9% dynamically and 6.2% thermodynamically. In contrast, the respective contributions to the ST cyclones are 71.8% and 43.5% for the ST cyclones, with negative effects from nonexplicit processes. In both NT and ST cyclones, the zonal PV advection in the upper troposphere is the most influential for the dynamic development, while nonlinear advection being more important in the former. The larger thermodynamic contribution of the latter is attributed to more latent heating being involved in the development, which produces more lower15 level PV and reduces damping from vertical PV advection. These results indicate that East Asian cyclones passing the Korean Peninsula have different development processes depending on their tracks.


May). The dynamic and thermodynamic contributions to the development of these two categories of cyclones are then comparatively assessed from the PV tendency equation. The results
suggested that East Asian cyclones passing the Korean Peninsula had different development processes depending on their tracks. Generally, the structure of this manuscript is fine, and its English language is better being non-native authors, as most sentences can be understood easily without any difficulties. However, it is necessary to indicate that, this manuscript at the present stage, didn't supply new and sharp insights into to deeply understand extratropical cyclones, and it is lack of sufficient explanation of physics for those two types of extratropical cyclones passing over the Korean Peninsula. Thus, it is very hard for referee to recommend this manuscript to be acceptable before the authors make clearer clarifications to the following key questions. It strongly suggests that this manuscript needs MAJOR REVISION before its potential publication in WCD.

Major Comments
(1) Reviewer: The background and methods of tracking ETCs were not introduced adequately.
There were a great number of methods of tracking ETCs. As indicated by Neu et al. (2013) "Identifying and tracking extratropical cyclones might seem, superficially, to be a straightforward activity, but in reality it is very challenging" (line 1-3 in right half, P529). The use of vorticity at 850 hPa for cyclone tracking is only one of 22 methods to identify ETCs as indicated by Neu et al. (2013)  Response: The selection of the tracking variable and pressure level follows Lee et al. (2019) and Kang et al. (2020). The advantages and disadvantages of using relative vorticity at 850 hPa can be found in the Sections 1 and 2 of Lee et al. (2019). In the revised manuscript, we note that we are using one of the various tracking algorithms, and that more details can be found in Lee et al.

Reviewer:
The data and methods used in the present study is strongly argued. The present authors employed the six-hourly ERA-Interim data during the period from 1979 to 2017 which were interpolated onto 1.5 o×1.5 o (same data with K20), and the ETCs under investigation were identified on the 850-hPa relative vorticity field "Note that these ETCs fall into the categories of rapidly intensifying cyclones in K20" (line 89 Response: The PV on an isobaric surface is expressed as follows. In the present study, Eq. (R1) is computed by second-order finite differencing with 1.5 o ×1.5 o data.
It turns out that synoptic scale PV is not very sensitive to the data resolution in ERA-Interim. Figure

(4)
Reviewer: It is not convincing that two important references supporting WCD-2020-65 properly.
The authors indicated at "In East Asia, cyclogenesis is remarked over Mongolia and East China (Chen et al., 1991;Adachi and Kimura, 2007"(line 29). This citation plays quite important role for WCD-2020-65. We examined carefully the detailed information from words and figures in the aforementioned two papers (named as "Chen1991" and "AK2007", respectively), and found a great discrepancy between Fig. 2 of "Chen1991" (see Figure A) and Fig.1a of WCD-2020-65 (see Figure B). It is OK that "N region" is Figure A can be seen in Figure B correspondingly.
But it is very strange that "S region" (the high cyclogenesis density region ) in Figure A can not be found in Figure B. This may be perhaps explained that they used different period data with different resolution. "Chen1991" used the twelve-hourly historic weather maps from 1958 to 1987, and 2.5 o × 2.5 o latitude-longitude (coarse-resolution data) for cyclogenesis frequency counting. WCD-2020-65 used six-hourly 1.5 o × 1.5 o ERA-Interim data (fine-resolution data) from 1979 to 2017. The overlapping period of these two data is INDEED 9 years (from 1979 to 1987). It is very hard for referee to understand the important feature of "high cyclogenesis density region" in "Chen1991" (coarse-resolution data) disappeared in WCD-2020-65 (fineresolution data). Is it suggested this great discrepancy hint that the method of using vorticity at 850 hPa for cyclone tracking is questionable? Moreover, the present referee failed to find "cyclogenesis is remarked over East China". In "Chen1991" paper, we didn't find the term "East China" but "East China Sea". In AK2007 paper, they mentioned "East China Sea" several times, but not "East China". It is strongly expected that the authors could clarify this issue.
Response: While AK2007 utilizes sea-level pressure (SLP) for objective ETC tracking, Chen1991 relies on manual tracking on surface weather maps. In other words, both of them are based on SLP field. Since the relative vorticity at 850-hPa pressure level is used in this study, the resulting ETC properties could be different from those reported in AK2007 and Chen1991.
The difference between AK2007/Chen1991 and the present study mostly results from the difference in reference variables. As described in Hoskins and Hodges (2002) and Lee et al.
(2019), the relative vorticity field, compared to SLP field, captures a relatively smaller-scale perturbation and allows and early detection of the surface cyclone. In fact, a strong cyclonic vorticity (momentum field) is often followed by a minimum SLP (mass field) with a time lag.
Their relationship is well documented in Hoskins and Hodges (2002) who compared the cyclogenesis statistics achieved from SLP and relative vorticity tracking. Figure R2 shows their Asia. The cyclones generated at "S region", which is located southeast of the Korean Peninsula, are unlikely to pass through the target domain. We clarified this point in the revised manuscript as follows.
Lines 80-83: From the tracking algorithm, the ETCs passing the Korean Peninsula (120-135 o E, 33-48 o N; yellow box in Fig. 1) are selected. As in K20, only the ETCs that are generated west of 120 o E are considered to focus on their development processes while traveling eastward through the target region. The track density of the selected ETCs in the cool season (October-May) is illustrated in Fig. 1a.
Reviewer: In the present study, the PV tendency equation is a very important tool for analysis.