Dynamic Field Retrieval and Analysis of Structural Evolution in Offshore Core Area of Typhoon Higos Based on Ground-Based Radar Observation

Round 1

Reviewer 1 Report (Previous Reviewer 1)

After reviewing the manuscript, I thought this manuscript present this time is much better than the previous one. But only one thing I would ask the authors to do a minor revision.

1. From lines 228 - 248, please separate the formulas or equations with texts. I mean, put the equation in a single line and mark the equation with the number (Ex: eq(1)...etc); it will help readers to get your DDA ideas. 

Author Response

Minor comments:

1. From lines 228 - 248, please separate the formulas or equations with texts. I mean, put the equation in a single line and mark the equation with the number (Ex: eq(1)...etc); it will help readers to get your DDA ideas. 
Response: Thanks for the reviewer’s recommendation. We have separated formulas with texts as your suggestion. Indeed, the new presentation form is helpful for readers to get the DDA method.

Thanks again sincerely for your two rounds review. The suggestions have greatly improved the quality of our manuscirpt.

Author Response File: Author Response.docx

Reviewer 2 Report (Previous Reviewer 3)

I would like to thank for the authors' substantial revision based on my previous comments. However, in my opinion while reading the current manuscript, it is difficult to understand the content, including the novelty. I have no choice but to return the manuscript before a comprehensive review.

First, I believe that an observing system simulation experiment (OSSE) is a modeling experiment used to evaluate the value of a new observing system when actual observational data are not available. In this manuscript, I do not know whether actual observational data may be 3-dimensional winds or reflectivity.

So, I imagined the following storyline.

1. 3-dimentional wind velocities are determined from Rader data using direct data assimilation (DDA) method and then compare the result with the output simulated by WRF

2. Reflectivity is calculated with CR-SIM from the output simulated by WRF and compare it with radar observations.

3. Reflectivity is determined with CR-SIM from the output simulated by WRF and then 3-dimensional wind velocities are determined by DDA.  The assimilated wind velocities are compared to the WRF output.

Only the third storyline is regarded as OSSE. However, the benefit of this paper is not OSSE but the radar observations near the typhoon. Reader would be confused when reading the manuscript for the first time. Nevertheless, it seems that the mechanism in the inner core of the typhoon is the main focus, and the result of OSSE is not clear in the conclusion.

Which method is used to simulate the radial wind, tangential wind, and vertical velocity for explaining the inner-core mechanism of the typhoon in section 4?

The abstract is needlessly long on the inner core mechanism of the typhoon. A schematic diagram on the inner core mechanism is desired. However, the terminology needs to be confirmed before preparing the schematic diagram.

・The definition of 'eyewall' in this study

・The definition of 'rainband stratus cloud' in this study

In general, eyewall is the area of extreme turbulence immediately surrounding the eye of a　cyclone or is the mass of clouds that whirls around the eye of a cyclone. I cannot confirm 'surrounding' or 'whirl around' in the typhoon case. It is a broad rainband rather than an eyewall. In addition, I cannot understand 'rainband stratus cloud' shown in Figure 7. Reflectivity is not always connected with clouds. Can data used in this study show the existence of such thinly spread cloud outside the 'rainband'? In any ways, the authors need to define each of these words in the text as readers can easily understand them.

The mechanism analyzed in this study is the process of changes in the inner core of the typhoon during the mature phase, so it is questionable to relate the process to maximum intensity/intensification mechanism. To begin with, the predominance of wavenumber-1 structure may be related to the sea surface temperature distribution and monsoon activity. In addition, the calculation area of the vertical shear, a radius of 300 km, gives the impression of being too narrow.

As for the figures, the gray arrows are difficult to see. Not only the figures, but the paper itself is not complete. I really regret it. It appears as if the first author rushed to submit the manuscript in order to meet the deadline. Compared to the previous manuscript, this presentation is certainly improved to some extent. However, there are many problems that should be further improved such as detail explanations on the experiment and terminology.

I had intended to review this manuscript more carefully, but since it is too incomplete, I will return it at once. Please check references and layout.

Author Response

Response to Reviewer #2

First, thanks sincerely for your careful review of two rounds. The suggestions have greatly improved the quality of our manuscript. The following are the specific modifications for your opinions and suggestions of round 2.

Major comments:

1. First, I believe that an observing system simulation experiment (OSSE) is a modeling experiment used to evaluate the value of a new observing system when actual observational data are not available. In this manuscript, I do not know whether actual observational data may be 3-dimensional winds or reflectivity. So, I imagined the following storyline.                                                 (1) 3-dimentional wind velocities are determined from Rader data using direct data assimilation (DDA) method and then compare the result with the output simulated by WRF. (2) Reflectivity is calculated with CR-SIM from the output simulated by WRF and compare it with radar observations.(3) Reflectivity is determined with CR-SIM from the output simulated by WRF and then 3-dimensional wind velocities are determined by DDA. The assimilated wind velocities are compared to the WRF output.                                                            Only the third storyline is regarded as OSSE. However, the benefit of this paper is not OSSE but the radar observations near the typhoon. Reader would be confused when reading the manuscript for the first time. Nevertheless, it seems that the mechanism in the inner core of the typhoon is the main focus, and the result of OSSE is not clear in the conclusion.

Response: Thanks for the reviewer’s recommendation. In our manuscript, the OSSE indeed is your third storyline as you imagined. Maybe the description of OSSE was not clear so that you and other readers be confused and misled. We have added a flow chart of OSSE as the following Fig.1 (see the format word file) to make the OSSE more clear. The OSSE steps are that. First, WRF model simulates the Typhoon Higos. Then, the wind u, v, w and cloud precipitation microphysics data of WRF-out will input into the radar emulation software CR-SIM. After setting radar band, beam width, observation mode and other radar parameters for the CR-SIM, the radar radial velocity and reflectivity will be emulated out, which are used to calculate emulated wind field by DDA method. Last, the emulated vertical velocity will be compared with the WRF-out w (as truth), so as to verify the accuracy of DDA method to retrieve the vertical velocity of Typhoon Higos. The purpose of OSSE in this paper is to validate the accuracy of DDA method for retrieving the vertical velocity, and does not provide the dynamic information to subsequent analysis. The using of words of OSSE part in original manuscript are improper, such as simulating the radar observation, which also misled the readers. So, we use the words 'emulate radar' replacing 'simulate radar' to make the expression more accurate. And, the words 'radar observation' in OSSE part are deleted. The other contents of OSSE also amended a lot. Please see the detail in the revised manuscripts 3.2.

2. The terminology needs to be confirmed before preparing the schematic diagram.

・The definition of 'eyewall' in this study

・The definition of 'rainband stratus cloud' in this study

In general, eyewall is the area of extreme turbulence immediately surrounding the eye of a　cyclone or is the mass of clouds that whirls around the eye of a cyclone. I cannot confirm 'surrounding' or 'whirl around' in the typhoon case. It is a broad rainband rather than an eyewall. In addition, I cannot understand 'rainband stratus cloud' shown in Figure 7. Reflectivity is not always connected with clouds. Can data used in this study show the existence of such thinly spread cloud outside the 'rainband'? In any ways, the authors need to define each of these words in the text as readers can easily understand them.

Response: Thanks for the reviewer’s recommendation. The terminology of 'eyewall' and 'rainband stratus cloud' in original manuscript indeed are confused. After modification, we confirmed the eyewall, inner rainband and outer rainband of Typhoon Higos in this study. The eyewall is the unclosed and surrounding the eye. The broad echo located on the east-by-south side of the eyewall was vigorously active, which is the inner rainband. Another rainband developed in the northeast and northwest sides is the outer rainband. The following Fig.2 (see the word format file) is the screenshot of Figure 2 in the revised manuscript, which indicates the eyewall and rainbands. We are very sorry for the words 'rainband stratus cloud' confusing you. We have replaced the words with 'stratiform sector of outer rainband' as other scientists' paper description. The focus of our paper is the dynamic analysis of the stratiform sector of outer rainband. We divided the stratiform into two part, one part is the upwind of stratiform, and another part is the downwind of stratiform.

3. The mechanism analyzed in this study is the process of changes in the inner core of the typhoon during the mature phase, so it is questionable to relate the process to maximum intensity/intensification mechanism. To begin with, the predominance of wavenumber-1 structure may be related to the sea surface temperature distribution and monsoon activity. The abstract is needlessly long on the inner core mechanism of the typhoon. A schematic diagram on the inner core mechanism is desired. In addition, the calculation area of the vertical shear, a radius of 300 km, gives the impression of being too narrow.

Response: Thanks for the reviewer’s recommendation. Indeed, the analyzed periods in our study is mature phase of Typhoon Higos, and the intensity in this periods did not change obviously. Your question is right. Our goal in our paper is to analyze the dynamic structure evolution of typhoon core area. So, we modified the first section introduction and second section description of radar observation to focus on the structural evolution rather than the intensity. After confirmed the terminology, we diagramed a schematic about the understanding of the stratiform dynamic of outer rainband in Typhoon Higos. It is shown in the following Fig.3 (see the word format file), which indicates the obvious dynamic difference between the upwind of stratiform rainband and the downwind of stratiform rainband. The mesoscale secondary circulation, vortex-scale motion associated with the overall storm, updraft and downdraft are drawn on the schematic chart. The upwind dynamic of stratiform rainband is in line with the research results of other scientists. But the downwind dynamic of stratiform rainband is different with upwind in Typhoon Higos, which affected by inflow promoted by southwest monsoon and outflow from inner side rainband. Please see the detail mechanism description in the revised manuscripts. We have reduced the words of abstract. And the calculation radius of the vertical shear has expanded to 500 km.

4. Which method is used to simulate the radial wind, tangential wind, and vertical velocity for explaining the inner-core mechanism of the typhoon in section 4?

Response: Thanks for the reviewer’s recommendation. Our fuzzy representation leads to your misunderstand. The radial wind, tangential wind and vertical velocity for explaining the inner-core mechanism are calculated or obtained from the wind field retrieved by DDA method based on three radars observation. We have modified the expression in the beginning of fourth section, and the expression of simulated part in the third section also been modified as the above response 1.

5. As for the figures, the gray arrows are difficult to see. Not only the figures, but the paper itself is not complete.

Response: Thanks for the reviewer’s recommendation. The gray arrows have been replaced by red and larger arrows in all Figures. As for your opinion about the paper completeness, we revised and modified the whole content of our manuscript. Please see the detail modification in the revised manuscript.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (Previous Reviewer 3)

Thanks to the authors' substantial revision, I can easily understand the novelty of this study, although the presentation becomes poor. In addition, I suspect whether the simulation results by WRF are comparable to the assimilated results presented in this study. 

The followings are simple advice for the subsequent substantial revision.

1. How to describe the abstract

The words should be equal to or less than 300.

The logic should be the fact, purpose, method, and results. The results should be included in the novelty.

In my impression, the structure of the abstract in the current manuscript should be divided into the following four parts.

[Fact] Three ground-based radars observed Higos

[Purpose] Detail understanding of asymmetry of the mechanism of stratiform rainbands

[Methods] The three-dimensional variational direct data assimilation Dual-Doppler analysis technique, the Cloud Resolving Model Radar SIMulator (CR-SIM) software, and numerical simulations by WRF

[Results] Summary of Figure 14

2. Purpose of this study that should be described in the introduction

The review of the introduction is clear, but the relation between previous study and the research purpose of this study is not clear. This problem is the same as that in the abstract.

3. Figures

Figures 1, 2, 4, 5b, 6, 8, 9, 10, 11, 12, 13, and 14 are poor. I cannot find Figure 5b and 7. 

The authors did not mention the reason why 190,240,270,340 degrees (dashed lines) are selected in Fig. 8. It seems to be subjective.

The authors should add the explanation on the horizontal axis of Figure 11. I am confusing between angles of Figure 8 and Figure 11 partly because of poor resolution of these figures.  

4. Evidence on 'high reliability'

First, the time series of simulated central pressures by WRF should be shown as another figure that is different from Figure 4. 

Second, in addition to Figure 6, the correlation of the assimilated vertical wind velocity should be verified with the simulation results by WRF and should show the significance.  

5. Minor comments

CMA, OISST, NOAA, NCDC in section 2, WRF and ERA5 in section 3 should be spelled out.

Line 68: typhoons over the ocean

Figure 2a: Please add the explanations regarding three red characters in the caption.

Line 210 regions where (add a space)

Line 492: 'There is no obvious latent heat cooling' why? This means heating occurs? How did the authors show the value of heating/cooling?

 

Author Response

First, thanks again sincerely for your careful review of three rounds. The suggestions have greatly improved the quality of our manuscript. The following are the specific modifications for your opinions and suggestions of round 3.

Major comments:

1.How to describe the abstract

The words should be equal to or less than 300.

The logic should be the fact, purpose, method, and results. The results should be included in the novelty.

In my impression, the structure of the abstract in the current manuscript should be divided into the following four parts.

[Fact] Three ground-based radars observed Higos

[Purpose] Detail understanding of asymmetry of the mechanism of stratiform rainbands

[Methods] The three-dimensional variational direct data assimilation Dual-Doppler analysis technique, the Cloud Resolving Model Radar SIMulator (CR-SIM) software, and numerical simulations by WRF

[Results] Summary of Figure 14

Response: Thanks for the reviewer’s recommendation. Your careful instruction about abstract indeed makes our abstract of the manuscript concise and clear. We have modified the abstract as following. And the words are equal to 252.

[Fact] Three ground-based radars in the Pearl River Delta observed the August 2020 Typhoon Higos during the offshore. The stratiform sector of outer rainband of Higos was active, which whirled inwardly, merged with the unclosed eyewall and broaded, while the structures of upwind and downwind of stratiform rainband were asymmetrical. [Purpose] The densely distributed radars observation provide good conditions for acquiring the refined wind field in Higos’ inner-core area to detailly understand the asymmetry of dynamic mechanism of stratiform rainband. [Methods] The three-dimensional variational direct data assimilation Dual-Doppler analysis (DDA) is adopted to retrieve wind field. Prior to the retrieved, the Observing System Simulation Experiment (OSSE) based on the numerical simulations by Weather Research and Forecasting (WRF) and Cloud Resolving Model Radar SIMulator (CR-SIM) software were operated to have validated the reliability. [Results] The analysis result shows that dynamic mechanism between the upwind of stratiform rainband and the downwind are significantly different. The former has sinking inflow in the middle layer and also inflow in the lower layer with outflow in the upper part of the inflow. However, the latter are primarily affected by the outflow from the inner side of the rainband and the inflow from the outer side of southwesterly wind related to the monsoon. The vertical velocity characteristics of the downwind are also distinct from the upwind in Higos. The height of the strong updraft is even lower, which are caused by the convergence of the outflow from the inner side of rainband with the southwesterly inflow, promoting the enhanced reflectivity line developed.

2. Purpose of this study that should be described in the introduction

The review of the introduction is clear, but the relation between previous study and the research purpose of this study is not clear. This problem is the same as that in the abstract.

Response: Thanks for the reviewer’s recommendation. We have modified the last paragraph of introduction section, added the relation between previous study and the research purpose of this study. Please see the detail modifications in the revised manuscripts.

3.Figures

Figures 1, 2, 4, 5b, 6, 8, 9, 10, 11, 12, 13, and 14 are poor. I cannot find Figure 5b and 7. 

The authors did not mention the reason why 190,240,270,340 degrees (dashed lines) are selected in Fig. 8. It seems to be subjective.

The authors should add the explanation on the horizontal axis of Figure 11. I am confusing between angles of Figure 8 and Figure 11 partly because of poor resolution of these figures.  

Response: Thanks for the reviewer’s recommendation. We are very sorry for ignoring the lost of the Figures 5b and 7 in the last revised manuscript, which have been added in our this revised version. At the same time, we have redrawn the all Figures as your suggestion. The axis labels, colorbar labels, and the numbers of contour line have been enlarged, and some other contents also been modified to make the Figures are clear.

As for the degrees selected in Fig.8, we have refined our description to select the azimuths in revised manuscript from line 393 to line 397. The following is the content.

The azimuths selection comprehensively consider the reflectivity evolution positions of the upwind and downwind from 17:00 to 20:00 in Figure 2. At the same time for avoiding mutual influence during calculation, the azimuth of upwind and downwind are separated by 30°.

According to the horizontal axis of Figure 8 and Figure 11 making you confused, we have supplemented the explanations in revised manuscript from line 399 to line 404. The following is the content.

   The radial winds and vertical velocity corresponding to the upwind and downwind are respectively averaged by azimuth at the following text to analyze the variance of dynamic characteristics with the radius range varying. The tangential winds are averaged by radius range to analyze the variance of dynamic characteristics azimuthally. The radius range means the horizontal range to typhoon center.

4. Evidence on 'high reliability'

First, the time series of simulated central pressures by WRF should be shown as another figure that is different from Figure 4. 

Second, in addition to Figure 6, the correlation of the assimilated vertical wind velocity should be verified with the simulation results by WRF and should show the significance.  

Response: Thanks for the reviewer’s recommendation. You had suggested to draw the time series of simulated central pressures by WRF. Indeed, we have drawn it shown in the following Figure 1 (see in word format file). But, we think that it is not necessary to put it in the manuscripts. Because the Fig.4 has showed the intensity of simulated Typhoon by WRF is weaker than best track, and the influence of which to validation the reliability of DDA method has been described in lines 266 to 273. If our understanding is wrong, we are pleased to accept your guidance again.

    The correlation of the emulated w by DDA method and the WRF-out w has been calculated shown in following Figure 2 （see in word format file）. The correlation coefficient (CC) of total point is only 0.49 in Figure 2a, and CC is obviously increasing to 0.63 when filtering the points which are emulated w value below 2 m/s or WRF-out w value below 2 m/s in Figure 2b. Although the value is still not high, we can see that the emulated w retrieved by DDA method has relatively good consistency with reference value (WRF-out w), especially the reference value is lager. Vertical velocity is a physical quantity that always difficult to be measured or retrieved, but its’ characteristic is quite necessary to be recognized. So OSSE result shows that the vertical velocity retrieved by DDA method can be provided to qualitatively understand the Typhoon Higos’ dynamic evolution. But the 'high reliability' words indeed are not proper to be described as the validation results, we have removed it. Please see the detail modification in 3.2 paragraph of revised manuscript.

Minor comments:

1. CMA, OISST, NOAA, NCDC in section 2, WRF and ERA5 in section 3 should be spelled out.

Response: Thanks for the reviewer’s recommendation. The CMA, OISST, NOAA, NCDC have been spelled out in section 2. The ERA5 has been spelled out in section 3. The WRF has been spelled out in abstract. Please see the detail modification in the revised manuscript.

2. Line 68: typhoons over the ocean

Response: Thanks for the reviewer’s recommendation. We have modified the expression of Line 68 as your suggestion.

3. Figure 2a: Please add the explanations regarding three red characters in the caption.

Response: Thanks for the reviewer’s recommendation. We have added the explanations in Figure 2 caption. Please see the detail modification in the revised manuscript.

4. Line 210 regions where (add a space)

Response: Thanks for the reviewer’s recommendation. We have added a space in line 210 regions and where. And we have check out the whole manuscript to avoid the same fault.

5. Line 492: 'There is no obvious latent heat cooling' why? This means heating occurs? How did the authors show the value of heating/cooling?

Response: Thanks for the reviewer’s recommendation. The description that there is no obvious latent heat cooling is arbitrary. In this paper, we did not research the heating and cooling of downwind of stratiform, so we should not draw that conclusion. We think that it is necessary to perform more thermodynamic researches through numerical simulation in future. Please see the detail modification in the revised manuscript.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report (Previous Reviewer 3)

Thank you for your quick revision. I have one proposal and minor comments.

Minor comments:

1. L254,270 reliability of (need a space)

2. Figure 7: The analysis time is the same as Figure 6? Please show the analysis time.

3. Figure 7 and related descriptions: Not only the value of correlation coefficient but also the significance should be shown. t-test or any test is required.

4. Figure 9a: The range 270-340 degree seems to be wide when looking at Figure 9b-d. In contrast, the range 190-240 seems to be reasonable. Therefore, I pointed out that the decision seems to be subjective.

5. Conclusion (not only the conclusion but also all text): The descriptions should be modified if the authors accept my proposal on the abstract below.

Proposal: 

Reference

Houze, R. A., Jr. (2010). Clouds in Tropical Cyclones, Monthly Weather Review, 138, 293-344. https://doi.org/10.1175/2009MWR2989.1

For the abstract,

Three ground-based radars in the Pearl River Delta successfully observed Typhoon Higos (2020), which traveled over offshore area in the South China Sea. During the observation period, the stratiform region of the outer rainband of HIGOS became active while swirling inward, merging into an unclosed eyewall and spreading outward, but its structure was asymmetric between upwind and downwind. To understand the dynamic mechanism of the asymmetry of the stratiform region in detail, refined wind speed distributions in the inner core of Higos was retrieved by using the radar observation data and a three-dimensional variational direct data assimilation Dual-Doppler analysis (DDA) . In addition, the Observing System Simulation Experiment (OSSE) was conducted with the numerical simulations by Weather Research and Forecasting (WRF) model and numerical emulations by Cloud Resolving Model Radar SIMulator (CR-SIM) software to validate the retrieved data. From the OSSE, the retrieved data was comparable with the emulated data. The analysis shows that the dynamic mechanisms are different between upwind and downwind in the stratiform rainband. In the former, the inflow sinks in the middle troposphere. In addition, there is an inflow in the lower troposphere with an outflow aloft the inflow. In the latter, however, the stratiform rainband is primarily influenced by outflow from inside the rainband and inflow from outside the monsoon-related southwesterly winds. The vertical velocity characteristics in the stratiform rainband downwind also differ from those upwind. The upwind updraft was distinct in the middle troposphere, whereas the downwind updraft was caused by convergence of outflow from inside the stratiform rainband and the monsoon-related southwesterly inflow in the lower troposphere.

Author Response

Response to Reviewer #2

    All our authors sincerely thank you for your fine and careful review of four rounds. The all suggestions have greatly improved the quality of our manuscript. The following are the specific modifications for your proposal and minor comments of round 4.

Proposal:

1. Reference

Houze, R. A., Jr. (2010). Clouds in Tropical Cyclones, Monthly Weather Review, 138, 293-344. https://doi.org/10.1175/2009MWR2989.1

Response: Thanks for the reviewer’s proposal. We have added the doi of Houze, R. A., Jr. (2010) paper in our reference.

2. Abstract

Three ground-based radars in the Pearl River Delta successfully observed Typhoon Higos (2020), which traveled over offshore area in the South China Sea. During the observation period, the stratiform region of the outer rainband of HIGOS became active while swirling inward, merging into an unclosed eyewall and spreading outward, but its structure was asymmetric between upwind and downwind. To understand the dynamic mechanism of the asymmetry of the stratiform region in detail, refined wind speed distributions in the inner core of Higos was retrieved by using the radar observation data and a three-dimensional variational direct data assimilation Dual-Doppler analysis (DDA). In addition, the Observing System Simulation Experiment (OSSE) was conducted with the numerical simulations by Weather Research and Forecasting (WRF) model and numerical emulations by Cloud Resolving Model Radar SIMulator (CR-SIM) software to validate the retrieved data. From the OSSE, the retrieved data was comparable with the emulated data. The analysis shows that the dynamic mechanisms are different between upwind and downwind in the stratiform rainband. In the former, the inflow sinks in the middle troposphere. In addition, there is an inflow in the lower troposphere with an outflow aloft the inflow. In the latter, however, the stratiform rainband is primarily influenced by outflow from inside the rainband and inflow from outside the monsoon-related southwesterly winds. The vertical velocity characteristics in the stratiform rainband downwind also differ from those upwind. The upwind updraft was distinct in the middle troposphere, whereas the downwind updraft was caused by convergence of outflow from inside the stratiform rainband and the monsoon-related southwesterly inflow in the lower troposphere.

Response: Thank you very much for the reviewer’s proposal. We are grateful for you writing the abstract yourself. New abstract summarized our manuscript more concisely and the words more appropriately. We have fully accepted the new abstract, except one sentence slightly modified. It is the sentence 'From the OSSE, the retrieved data was comparable with the emulated data.'. We have modified it as 'From the OSSE, the emulated retrieved data was comparable with the WRF-out data.'.

Minor comments:

1. L254,270 reliability of (need a space)

Response: Thanks for the reviewer’s recommendation. We have added a space in line 254,270 regions and where. And we have check out the whole manuscript to avoid the same fault.

2. Figure 7: The analysis time is the same as Figure 6? Please show the analysis time.

Response: Thanks for the reviewer’s recommendation. We have added the analysis time in Figure 7 caption. Please see the detail modifications in the revised manuscripts.

3. Figure 7 and related descriptions: Not only the value of correlation coefficient but also the significance should be shown. t-test or any test is required.

Response: Thanks for the reviewer’s recommendation. We have added the t-test in Figure 7 and related descriptions to show the significance. In addition, we are sorry for the wrong calculation of correlation coefficient of previous version, we have corrected it. Please see the detail modifications in the revised manuscripts.

4. Figure 9a: The range 270-340 degree seems to be wide when looking at Figure 9b-d. In contrast, the range 190-240 seems to be reasonable. Therefore, I pointed out that the decision seems to be subjective.

Response: Thanks for the reviewer’s recommendation. We already know what you mean. We will further analyze the influence of azimuth selection to the results in detail. Now, we believe that the conclusion is credible based on the azimuth selection of our manuscript. Thanks again for your recommendation.

5. Conclusion (not only the conclusion but also all text): The descriptions should be modified if the authors accept my proposal on the abstract below.

Response: Thanks for the reviewer’s recommendation. The conclusion and all text have been modified according to the new abstract. Please see the detail modifications in the revised manuscripts.

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Thanks, the authors work on this article. But I found some fundamental issues in this research work.

1. All figures are presented with low quality and are hard for the reader. Please improve all of them (EX: x-y axis labels, text on the figures)

2. Because the authors mention the "eyewall replacement cycle" several times in the draft. I checked the case used here called "Typhoon Higos (2020). This event was only marked as a "tropical storm or Cat.1," Are you sure the results presented in the draft could be treated as "ERC"? If you are sure that, please provide more pieces of evidence. Typically, ERC has a significant definition for a powerful tropical cyclone. Usually should be greater than Cat. 3. If you believe this case is under such intensity with ERC (Cat 1 or TS), you need to provide more solid results. Otherwise, I would suggest the authors remove "ERC."

Reviewer 2 Report

 

The authors examined the fast strengthening Typhoon Higos in 2020 August using CR-SIM and 3D-Var DDA. They found that the strengthening eyewall evolves from a loose asymmetric structure to a symmetrical structure, with vertical velocities developed following an eyewall replacement cycle. They further showed that a strong vertical upward velocity is still developing in the downshear area of the eyewall after Higos makes landfall and the maximum vertical velocity occurs in the left area of the downshear.

 

This is an interesting that contributes to our understanding of the development of typhoon eyewall. The paper structure is clear and the analysis is comprehensive. I thereby would like to recommend publication of the manuscript pending on minor revisions.

 

1. The authors may want to add space between numbers and units, for example, change 10m/s to 10 m/s (Line 28), 17-18km to 17-18 km (Line 66), 970hPa to 970 hPa (Line123), and many other places in the manuscript.

2. Line 76: Please change “in” to “for”.

3. Line 98: Please change “in” to “to”.

4. Line 211: “y” is missing in the denominator of the last term of the equation defining Jv, i.e., should be \partial(v^a) / \partial(y).

5. Figure 4: The authors may want to elaborate a bit more on the cause of the difference between the simulated and observed tracks of Typhoon Higos.   

Reviewer 3 Report

General comments:

This paper proposes a method to retrieve vertical winds around Typhoon Higos (2020) during the mature and decaying phases by Obsering System Simulation Experiment (OSSE) with radar observations in Pearl River Delta and the Weather Research and Forecasting (WRF) model. The authors try to show the validity of the retrieval results by conducting an analysis focusing on structural changes in the inner core region of the typhoon.

In the abstract, about a half is the authors' interpretation on the structural changes in the inner core region bases on the observations and retrieval analysis results. I am skeptical on the description of the eyewall replacement cycle, possibly resulting in misleading for readers.

This submission is not sufficiently complete. As for the presentation, the followings should be improved:
1. Need OSSE information in the title and abstract of the paper.
2. The equations are difficult to follow. There is no explanation of the variables. The format should be improved.
3. The paper uses terminology (e.g. strength, strengthening instead of intensity, intensification) that is not often used by tropical cyclone researchers.
4. The paper lacks descriptions of basic information necessary to understand the content, such as the source of the best track, definition of vertical shear, and location information on the horizontal plane of the cross-section (Fig. 10).
Therefore, there is still room for improvement.

Actually, the results simulated by the WRF do not match the best-track data. However, there is no information to what extent the difference does affect the OSSE procedure and numerical results.

I do not understand what is a new finding included in this paper. Since the scientific description of this paper is questionable, I cannot propose a revision.

Based on the above, I recommend this paper should be rejected for publication.

Major comments:

http://tropic.ssec.wisc.edu/real-time/archerOnline/cyclones/2020_08W/web/summaryTableERC.html

According to the information on the above website, I can confirm that primary and secondary rainbands exist from 09 UTC to 21 UTC on 18 August in 2018. However, the website says that the probability of eyewall replacement is zero.

Higos has not intensified after a slight intensification during the period when the innermost reflectivity increases. This observational results shown in Figure 3 was obtained in a phase in which little intensification was taking place.

It is understandable that there is a banded precipitation area upstream of the moving direction of Higos, which consists of two spiral rainbands. On the other hand, the banded precipitation area becomes ring-shaped only at 23 UTC when Higos made landfall. I feel a strong sense of discomfort regarding the eyewall replacement cycle (ERC) process described on the basis that the authors' interpretation is consistent with the previous famous paper.

With respect to Figure 4, the simulated track and intensity of Higos are presented so that readers are easy to understand the difference particularly after 12 UTC on 18 August. The authors should show the distribution of reflectivity at 3 km altitude by time and the distribution of vertical velocity at 3 km altitude at the same time, as in Figure 2. I feel uncomfortable with the use of simulation results for OSSE without validation.

Minor comments:

1. The font size in figures should be larger.

2. What is the source of the SST product?

3. The Jo term is always zero in line 209.

4. Figure 11: Please calculate convergence and divergence at least.

 

Reviewer 4 Report

This manuscript presented a study to retrieve the vertical velocity for a tropical storm, Typhoon Higos, using the ground-based multiple radar measurements. The topic of this study fits well in the scope of Remote Sensing, but the section of data and methodology will need significant improvements before it can be published. The results presented in Section 4 (Results Analysis and Validation) are not that meaningful without those clarifications. 

 

Firstly, as the authors have pointed out in the manuscript, a high-density radar observational platform was available during Typhoon Higos. However, I failed to see a detailed description of the development of this storm using the four identified radars. Figure 2 did provide the storm’s evolution using radar reflectivity, but it was only based on one radar, Hong Kong Tate’s Cairn Radar. In addition, I suggest that the authors add a table summarizing the key parameters (wavelength/frequency, PRFs, etc.) for those radars and discuss how those parameters would impact the conducted retrievals. 

 

Secondly, the WRF simulations and the associated results need to be expanded. Based on the results shown, the simulations clearly had trouble reproducing the observed storm track and intensity. Quite a few things related to the simulations are not clear to me. Below are just some of those.

·      What is the model domain? 

·      What boundary layer scheme, radiation scheme, and surface layer scheme have been adopted? 

·      Why those parameterization schemes were chosen? 

·      Did the simulations capture the intensity of the storm? If not, why?

·      Were simulated radar reflectivity and radial velocity using the CR-SIM able to reproduce radar observations? 

In my opinion, these questions need to be addressed before any retrievals using radar measurements are performed. 

 

Moreover, previous studies have shown that CR-SIM tends to overestimate the mean radar reflectivity in the middle and high parts of convective clouds. Have the authors considered how that bias would impact the retrieved vertical velocity?

 

Below please find my other comments.

 

-       L89: Correct the format of citations.

-       L62-114: This paragraph is too long to follow. 

-       L121: I suggest to change the title of section 2 to ‘Data and Methods’

-       L122-127: Add references and data sources to support the case description.

-       L132-136 (Figure 1): the texts on the figure are difficult to read. This is also the case for Figure 4. What data were used for the sea surface temperature, central pressure and wind speed?

-       L203-215: What ‘Radx’ refers to in the observation term of the cost function? Both reflectivity factor and radial velocity are the inputs of the retrieval method, but I only see radial velocity in the equations shown. How were the weights of different cost functions chosen?

-       L216-227: Vague words have been used to describe the observational error including ‘highly reliable’, ‘improve obviously’, and so on. Please provide quantitative values for those uncertainties. 

-       L261-262: What can be possible reasons to cause this underestimation of storm intensity? How would this bias impact the retrievals of vertical velocity? Could this be improved by modifying the WRF configurations?

-       L268-271: add reference(s) for the CR-SIM software. 

-       L277-279: add units for Zhh and DV in Figure 5. Also, I suggest that the authors add a figure comparing the simulated radar reflectivity and radial velocity with the observations. 

-       L295-306: I am confused here about the 0.1 elevation data. Was it from the weather radar observations? If yes, how the model simulations and radar observations were combined? 

-       L307-309: The wind barbs are missing in the eastern part of the study domain in Figure 6a. What’s the possible reason?

-       L310-315: I suggest that the authors reproduce Figure 7 by adding panels of difference plots (e.g., WRF vertical velocity – vertical velocity retrievals from the DDA method). It could help readers identify the limitations/uncertainties of the retrievals. This should also apply for Figure 8.