Review Reports

Round 1

Reviewer 1 Report

This study describes the use of atmospheric data assimilation to predict a typhoon's track using numerical modeling. This kind of work is important to know what methods that can be more useful when it comes to predicting where typhoons will landfall, in order to minimize potential damage. Despite this importance, I have some concerns about the work, especially on the general lack of bibliographic references and the weak explanation of how some results were produced. Of course the second aspect is related with the first (lack of scientific contextualization). In this sense, I return my review, suggesting "major revision" for this manuscript. The authors need to work on these issues  before the approval of the work for publication. Below I do some specific comments.

 

LINE 33- Rewrite the sentence. What the authors saying by “are firstly assimilated”?

LINE 45- Rewrite “by collocating” to “by setting”.

LINE 46- Rewrite “in this study” to cite the authors.

LINE 64- Rewrite Philipines’

LINE 66- The authors cite the impact of Typhon Lekima in Zhejiang province and others, but Figure 1 lacks to point the location of those provinces. To better clarify the locations, I suggest to redo the figure locating the provinces.

LINE 67- Rewrite “serve” to “severe”

LINE 74- Figure 1 legend lacks to clarify what TD, TS, STS, TY and L means. It is important to write it.

LINE 79- WRF is, currently, in its 4.2.2 version, but authors used the very old 3.8 version. Is there any reason for that?

LINE 79- The 2.2 section fails to reference the models, data and the physical parameterization. It is needed to reference everything (i. e., WRF, GSI, GFS, MetOp-A/B, AMSU-A) that is used to make easy to be reproducible by others.

 

LINE 81- The authors used a 9km horizontal grid, but it is very well know that WRF has a cumulus parameterization grey-zone between 1~10km which can led to errors in those kinds of study which small scale convection systems are important. If the authors are aware about this problem, I suggest to add the following references in the text to explain it:

 

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/grl.50347

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2013JD020887

LINE 84- I suggest the authors to put the WRF model parameterization setup in a table. This way can led to better understating about the used parameterizations. I suggest to create a collunm to write down the physical parameterization references. They can be found in : https://www2.mmm.ucar.edu/wrf/users/physics/phys_references.html

LINE 92- Rewrite “at 0600 UTC and 1800 UTC” to “from 0600 UTC to 18 UTC”.

LINE 109- What the authors are saying about the data thinning by 120 km? Please rewrite this setence.

LINE 145- I would like to know more about how and why the authors used the equation (1) to calculate U and V mean flow. It seems to me that the authors horizontally interpolated each layer of the atmosphere instead of vertically integrating from, for example, 300 to 850 hPa.

LINE 160- The Figure 3 shows the vertical RMSE of several variables in Eta Levels. How many vertical levels were used for the simulations? How they are spacialized? The authors can include this information in the section 2.2.

LINE 199- Figure 5 shows the track errors between the simulations. It is needed to write how the authors calculated it. It is a simple bias? It can be write in the sentence in the line 187.

LINE 203- Clarify how the mean track errors are calculated.

LINE 215- There are studies that also points to the role of the steering flow in the movement of typhoons. The authors can reference some of then.

LINE 230- The authors point to future works that they will develop from the submitted study. What future work can others 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Overview: this manuscript investigated the impact of assimilating FY-3D microwave upper-air sounding channels on the initialization and track forecasts of a typhoon case. The manuscript is nicely organized and relatively well written. Particularly, the figures are very clear and the visualization is excellent. However, there are several flaws I would like to point out (see below), most of which are the experiment set-up, missing references, and confused captions.  Overall, I would recommend a major revision.

Major comments:

1. The introduction is a bit weak to me. Please add more discussions on the context of radiance assimilation, challenges in radiance assimilation, and data impact for microwave sounding assimiaitons, and the advantage of using FY-3D MW soundings. There are tons of references on radiance assimilation with regional models, but the authors overlooked many of them. 
2. The data assimilation method is not described in Section 2. Additionally, channels 11-14 might be above the model top. Please explain why you use all 15 AMSUA channels. I would recommend to remove (some of ) them and re-do the experiments.
3. Please check all the captions and make sure the model initialization dates, track valid times, and clarify the lead time vs. forecast time.

Minor comments: 

L52: on -> into

L67: serve -> severe?

L63-69: the description provided little information to readers. Please re-write and improve it by adding lifecycle of the typhoon and detailed impact it caused.

Figure 1: explain TD, TS, STS, TY, L in the main text. The caption did not explain the track information such as starting time and ending time.

L97: some of the AMSUA channels have max pressure above 10 hPa. I am doubting why all 15 channels are used while your model top is only 10 hPa.  I would suggest that removing channels from 11-14 or a couple of them (check relevant references) to get more appropriate simulations.

Section 2: please include description of what assimilation method you used.

L104-106: please add relevant references.

Figure 3: the authors did not mention how RMSE is calculated such as w.r.t. what verification data.

Also, could you please add a bit discussions on why EXP_AMP slightly increased the RMSE?

L136: why not add pressure axis? 

Figure 5: the caption is confusing to me. Are these tracks valid for similar time period? What are the model initialization time? Are the x-axis lead time or forecast hour? 

Figure 6: again, please clarify lead time or forecast time, for instance, are they for the fixed track from day xxx to day xxx?

Figures 5-6: I doubt about the significance of the differences across three experiments.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I think the authors have dealt with my previous comments satisfactorily. I no longer have important scientific comments, but I believe that a complete revision in the language and editing is appropriate to improve English. The meaning of the information contained in the text is generally clear, but after this review, it would be even clearer and easier to read. I do not need to read the manuscript again and leave the final decision to the editor. I suggest to "accept" after this last writing review.

Author Response

We thank the editors and anonymous reviewers for their comments and suggestions, which are very helpful for us to further improve the quality of the manuscript. And we further improved ours languages of manuscripts to avoid the spell and grammar errors.

Reviewer 2 Report

The authors have addressed most of my comments. They did a great job in revising the manuscript! I don't have other additional comments except for final checks for spelling and grammar errors, and would recommend an acceptance after text editing. 

Author Response

We thank the editors and anonymous reviewers for their comments and suggestions, which are very helpful for us to further improve the quality of the manuscript. And we further improved ours languages of manuscripts to avoid the spelling and grammar errors.