Comparison of the Morrison and WDM6 Microphysics Schemes in the WRF Model for a Convective Precipitation Event in Guangdong, China, Through the Analysis of Polarimetric Radar Data
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study discussed the representation of ice-phase hydrometeors in the Morrison and WDM6 microphysics schemes for a convective event that occurred in Guangdong, China, in May 2017. Additionally, they examined predefined N0 of snow in WDM6 through a sensitivity test with DSD parameters. The results present several intriguing findings. However, the following points suggest that substantial revisions are required prior to publication.
The title of this manuscript, highlighting the use of a polarimetric radar simulator, initially captured my interest. As an enthusiastic supporter of integrating radar simulators in the evaluation of NWP microphysics schemes, I was eager to explore this study. Unfortunately, the analysis provided in this manuscript did not align with the expectations set by its interesting title. I had anticipated a more in-depth and comprehensive analysis (e.g., Wu et al. 2021 - doi: 10.1175/MWR-D-19-0378.1; Shrestha et al. 2022 - doi: 10.5194/gmd-15-291-2022; Köcher et al. 2022 - doi: 10.5194/amt-15-1033-2022), but the paper merely provides a qualitative comparison of radar variable distributions across different altitudes. In my opinion, it does not offer a unique contribution worthy of the emphasis suggested by the title.
I personally required a significant amount of time to read through this manuscript. The main points get lost in the text in the current form. To help readers better understand the content, I strongly recommend the following four improvements:
1. The paper is quite lengthy relative to the implications it seeks to convey. Many sections could be shortened or removed to emphasize the main points, preventing the reader from missing the main messages.
2. I strongly suggest restructuring the paragraphs, particularly in the results section, by placing the main points at the beginning of each paragraph, followed by supporting statements.
3. Sentences that are excessively long (e.g., L299-302, L482-486, L489-493) could be revised for clarity and readability.
4. The manuscript's English could be improved, particularly in the results section, which contains several unnatural sentences and numerous grammatical errors.
These issues significantly slow down the reading process.
The methods section of this study should provide a clearer description of the two approaches used to represent rimed particles (graupel and hail). Although these are discussed more specifically in the final paragraphs of the appendix for each scheme, I believe this information is at least as important as the other parts in the methods section. I urge the authors to move this information into the main methods section rather than leaving it in the appendix. If this content is removed from the appendix, the remaining description would largely mirror content from previously published papers, making it questionable whether such repetition is necessary in this manuscript.
The introduction needs improvement. Some parts are too similar to existing literature. For instance, L44-57 are highly similar to the fifth paragraph in Morrison and Milbrandt (2015; doi: 10.1175/JAS-D-14-0065.1). I do not intend to imply plagiarism, but I would encourage the authors to construct their introduction using their own logical flow and language, which better supports their motivation and the implications of their results. The MY15 paper provides a brief history of microphysics schemes, even extending to the triple moment, to introduce the new P3 scheme, but such an approach is unnecessary for this manuscript. These elements make the introduction less concise and dilute the focus on motivation. Furthermore, I see little need for three lengthy paragraphs before L80 to introduce ice phase processes. These are just examples, and I recommend revisiting the introduction as a whole to revise it for conciseness.
Regarding the comparison with radar variables from the forward simulator, it is unclear whether the comparisons of Zh, Kdp, and Zdr are performed under the same or at least similar conditions as the NWP models, particularly concerning grid resolution. The observations appear to have a coarse resolution, possibly due to averaging or smoothing, which would narrow the distribution and eliminate extreme values. Furthermore, this study links KDP exclusively to drop size. However, it overlooks the significant contribution of number concentration to KDP in addition to drop size. In addition, the observed ZDR (Fig. 6) contains an unusually high number of negative values. It is not clear if bias correction has been applied. Without this correction, a quantitative analysis would not be possible. In other words, it is challenging to confirm if the statement in L319, "the ZDR values at lower levels are overestimated by these two schemes," is correct.
The paper relies on a single case study for all the presented validations. It is uncertain if these findings can be generalized. The conclusion should highlight this limitation more.
Additional specific comments are as follows:
L58-79: The term "better" is used four times here. I recommend using this term more carefully. Something may be better or worse depending on the specific aspect. Please clarify the specific improvements or differences, or use alternative phrasing.
L131: Wrong statement. Scattering calculations do not necessarily have to be performed using the T-matrix method.
L168: I do not see mechanism study in this manuscript.
L195: The table caption is missing.
Table 1: Describe somewhere what subscripts (i.e., c, r, I, s, g, and h) stand for.
L201: What about the vertical coordinate? Is it also at a 1 km resolution?
L211: How can a spherical hydrometeor can have a variable axis ratio?
L219: How was the observational data obtained? This information appears to be missing. Radar does not directly measure the rainfall rate.
L220: What is GB/T 28592-2012?
L250: The sentence is unclear to me.
L268: Is the term "strongest hourly precipitation" based on the maximum value within the domain, the domain-averaged hourly precipitation, or a specific point?
L279: When you refer to "distributions", are you referring to the distribution of reflectivity values or the spatial distribution?
L281: The figure does not show this characteristic for WDM6.
L287: Not clear to me. A more quantitative evaluation is required.
L347: How did the authors conclude that this is related to raindrop sorting?
L388: "quicker" -> "a higher"
L389: It should be "a longer distance" rather than "more time to melt."
L389: hail can "fall" to a lower height -> hail can "survive" to a lower height
L389: contributing to radar reflectivity -> contributing to "stronger" radar reflectivity
L390: Check grammar carefully.
L444: I find it difficult to follow how the authors determined that the Morrison schemes have 1) a slower fall speed and 2) a faster generation rate, based on the statements in lines 437-443.
L445: "quicker" -> "more rapid"
L449-L466: The authors seem to be only comparing the changes in PSD parameters and their magnitudes. It is unclear what interpretation they intend to draw from this.
Figures 10 and 11: Why does the N0 of WDM6 have a broad distribution if they are constant? Why N0s is decreasing in a stepwise manner?
L476: Should this be associated with the parameterization of melting rather than just the melting process?
L483: Need to improve English for clarity. N0 is not a formula.
L550: Check grammar.
Figures 14 and 15: Are they showing the differences between the two simulations? The caption is unclear and needs to be revised to prevent any misunderstanding.
Figure 14: Is the PDF for all grid points and all time steps?
L554: "PDF distribution" is redundant. Consider using "PDF" alone.
Figure 15: The figure and caption both include two numbers next to N0g. Is the figure meant to show the differences between the two simulations? If so, this notation is misleading and needs correction.
L564: What criteria are used to define "severe rainfall level?"
L592: I do not agree. There have been numerous efforts to understand the discrepancies in DSD.
L600: Making a statement that the predefined N0 in WDM6 is "unreasonable" based on only one case study is too assertive and potentially misleading.
Figures 17 and 18: There is no description of these figures in the main text. The large number of arrows makes it difficult to discern if they represent microphysical processes, causal connections, or merely guide the visual flow. The meaning of the thin right arrow is not indicated in the legend.
Comments on the Quality of English LanguageAs mentioned in the review report, improvements to the English are needed.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsPlease find the pdf .
Comments for author File: Comments.pdf
The English language needs to be polished. It should be improved to increase readability.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsReview for “comparison of the Morrison and WDM6 microphysics schemes in the WRF model for a convective precipitation event in Guangdong, China, using a polarimeteric radar data simulator”
The authors used a polarimetric radar data simulator to deeply discuss the difference among Morrison and WDM6 in the WRF model for a convective precipitaton event in Guangdong, China. although some interesting finding occurred using the new radar technology; however, some description improvement should be addressed.
1. There are lack of some literature on the description of study advancement. For instance, L65~67; L67~69; L83~86
2. L70~73: What the relationship of this sentence to previous or after sentence? This sentence described the ice-phase description complexity and existed uncertainties; however, how it is related to the advantage of the double-moment scheme
3. The introduction has too heavy description, it seems to scratch the important information, the authors should simplify it.
4. Section 2.2: how to used the dual-polarization radar data. Or, which observation it can provide
5. L206: which is the polarimetirc vairables from the WRF outputs, please point out
6. L219~221: There are some description on rainfall observation, if it is come from the radar, please explain it more clearly
7. The title of Section 3.2 seems missing
8. Fig16: what is the default parameter value for fig16(a) and fig16(f). what is the difference on critical parameter values among them and other subfigure
9. There are required subtitles in the discussion section, which make the structure more clearly.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsI appreciate the authors' efforts in revising the manuscript and addressing the points raised in the earlier review. The revisions effectively address most of the concerns and suggestions. However, I still have a few major concerns.
Major concerns:
1. According to the authors' response, if the observed radar reflectivity in Fig. 4 has the same horizontal resolution as the simulation, it is plausible that the observed reflectivity has been somehow smoothed. The figure shows a significantly more smeared spatial pattern compared to the simulated reflectivity. For instance, the 10-20 dBZ region near 23.5°N, 114.5°E in Fig. 4(e1-5) supports this hypothesis. This could result in extreme values being reduced. Therefore, it cannot be ruled out that these values might have been similar to the larger values seen in WDM6. The authors should carefully investigate whether any additional data processing was performed beyond aligning the resolution with the simulation.
2. It appears that the authors performed a bias correction for observed ZDR during the revision process. However, the manuscript lacks an explanation of the correction process. Since the manuscript includes a quantitative comparison with simulated ZDR, it is important to detail the method used for the bias correction. It is also advisable to include the applied Zdr bias value.
3. The authors stated in their responses, “I have analyzed the horizontal spatial distributions of raindrop mass concentration and number concentration, although these plots are not included in this paper. The analysis indicates that the contribution of raindrop number concentration to KDP is significantly greater than that of mass concentration.” This conclusion is also reflected in the manuscript. Nevertheless, to properly evaluate this claim, it is important that these figures be transparently shared with reviewers and the editor, even if they are not presented in the manuscript.
Minor concerns:
I failed to mention in the initial review that the abstract should stand alone, meaning terms such as Mor_G and Mor_H must be understandable to readers based solely on the abstract.
The authors should consider re-including the statement “using the 88D2ARPS program from the Advanced Regional Prediction System (ARPS) model,“ which was omitted from the original manuscript. Being specific about the methods used for Cartesian conversion will enhance reproducibility.
It is inferred that the observations used in Figs. 2-4 are based on hourly precipitation data from automatic surface stations, but this is not clear from the text. Please clarify in the methods, results, captions, or within the figures whether the observations are obtained from radar or automatic surface stations.
L167: Please add a reference for CMORPH
L311: Perhaps the authors mean “due to better representation of raindrop size sorting.”
L535: “unrealistic” is still too assertive and dangerous to be concluded from one case study
Figs 6, 12, and 13: Typo (WDM6_G)
Comments on the Quality of English LanguageWhile there are still areas for improvement in the English, I believe that they will not be a significant issue if addressed during the proofreading stage.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 3 Report
Comments and Suggestions for Authorsthe authors have addressed all my concerned, and I suggest the manuscript should be accepted as current format.
Author Response
Thank you for offering your valuable review of the paper amidst your busy schedule.
