Assessment of vertical air motion among reanalyses and qualitative comparison with very-high-frequency radar measurements over two tropical stations

Uma, Kizhathur Narasimhan; Das, Siddarth Shankar; Ratnam, Madineni Venkat; Suneeth, Kuniyil Viswanathan

doi:https://doi.org/10.5194/acp-21-2083-2021

Articles | Volume 21, issue 3

https://doi.org/10.5194/acp-21-2083-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

The SPARC Reanalysis Intercomparison Project (S-RIP) (ACP/ESSD...

https://doi.org/10.5194/acp-21-2083-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 21, issue 3

Research article

|

11 Feb 2021

Research article |

| 11 Feb 2021

Assessment of vertical air motion among reanalyses and qualitative comparison with very-high-frequency radar measurements over two tropical stations

Kizhathur Narasimhan Uma, Siddarth Shankar Das, Madineni Venkat Ratnam, and Kuniyil Viswanathan Suneeth

Download

Final revised paper (published on 11 Feb 2021)
Preprint (discussion started on 27 Mar 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

SC1: 'Vertical wind in other data', Ulrich Schumann, 28 Mar 2020
- AC1: 'Reply to the comment', Uma K N, 29 Mar 2020
RC1: 'A review report', Anonymous Referee #1, 20 Apr 2020
RC2: 'Comment on Uma et al: Assessment of reanalysis vertical velocity', Anonymous Referee #2, 01 Jun 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Uma K N on behalf of the Authors (13 Jul 2020) Manuscript

ED: Referee Nomination & Report Request started (30 Jul 2020) by Gabriele Stiller

RR by Anonymous Referee #1 (13 Aug 2020)

Suggestions for revision or reasons for rejection

Second review report
Assessment of vertical air motion among reanalyses and qualitative comparison with VHF radar measurements over the two tropical stations
By K. N. Uma et al.

The authors have revised the manuscript reasonably in response to my first review. Below, I list several minor suggestions, which the authors may consider when they prepare the final manuscript.

(The following line numbers refer to those in acp-2020-18-author_response-version2.pdf, i.e., the changes-tracked version.)

Lines 14: add “, India”

Lines 15: add “, Indonesia”

Lines 21-22: I think this sentence needs clarification. One possibility is “. . . in the w when testing different spatial sampling for reanalysis data around the Gadanki station.”

Line 168: add “, India” and “, Indonesia”

Line 188, Table 1: add the information on the horizontal extent (in km or km^2) of the radar sampling volume at e.g., 10 km and 20 km altitudes.

Line 214: Change “daily mean profiles” to “daily 16:30-17:30 IST (11:00-12:00 UTC) averaged profiles”

Line 254: “This wind shear” – add the information on the altitude

Lines 255-256: Do you mean “above 6-7 km altitudes” ? If so, please explicitly write so.

Line 415: “long-period” – please explicitly write the temporal scale.

Line 416: “long-term” – again, please explicitly write the temporal scale.

Line 480: “long-term” – same as above. “diurnal” – is this “day-to-day”? (“diurnal” may mean 1-day (and 0.5-day) periodicity)

Line 715: “any” – did you do the same analysis for other reanalyses? Or, only ERA-Interim?

Lines 717-718: Change this sentence to: “We therefore show the directional tendencies of reanalysis data relative to the radar measurements.” Here, I assume that the authors use the term “tendencies” as the ratio that reanalysis reproduces (i.e., agree with) radar measurements in terms of vertical wind direction. Also, I think we need one more sentence, right after this, explaining/defining what is “directional tendencies” more clearly, or please add something like: “The directional tendencies would be 100% when all radar measurements at certain height range are reproduced by a reanalysis in terms of vertical wind direction.”

Lines 726-951: All “produce/producing/produced” should be changed to “reproduce/reproducing/reproduced”.

Line 1004: “any”: again, did you do the same analysis for other reanalyses?

Line 1010: change “care” to “caution”

Figure 2, caption: change “MST Radar” to “IMSTR”

Figure 4, caption: change “MST Radar” to “IMSTR”

Figure 10, caption: add the following sentence (between the two sentences), “The reference is the reanalysis ensemble mean.”

Figure 12, caption: add the information that these are the ERAi cases.

Table 2: Change “Kobayachi” to “Kobayashi”. Also, commas after the author names are unnecessary if “(year)” is used.

Hide

RR by Anonymous Referee #3 (05 Oct 2020)

Suggestions for revision or reasons for rejection

Overall the manuscript provides a novel intercomparison of reanalysis vertical velocities with ground-based measurements from 2 subtropical VHF radar locations. The study illustrates how widely vertical velocities can vary among reanalyses and observations, meaning that caution should be used when interpreting results from studies using reanalysis-based vertical velocities in the troposphere and lower stratosphere.

First, throughout the manuscript the authors should emphasize that this study focuses on a very limited geographical area, so the results do not necessarily apply for reanalyses in general. I think this is important, especially given the conclusion (lines 440-442) that the results somehow provide an initial basis to improve calculation of w in reanalyses. Instead, I would say that the results demonstrate that how approaches to generating global reanalysis products (encompassing different models, assimilation methods, spatial resolution, etc) can impact estimates of w. I think this is important since providing uncertainty estimates for derived meteorological products like w is currently needed by the SPARC community.

In that spirit, I think one thing that is currently missing from this paper, which is needed before I can recommend publication, is a quantitative discussion of the uncertainties in the retrieved vertical velocities from the radar. As other referees have commented, the authors do mention sources of uncertainty but I don't have an idea for what a typical error bar would be. For example, can the vertical velocities from an individual profile be determined with an accuracy of a cm/s or less? Perhaps this is described in other papers, but it needs to be discussed here so we can make sense of the comparisons with w from reanalyses, and perhaps also included in Table 1?

Another item that must be addressed is the process for computing vertical velocities from reanalyses on an altitude grid. Again, earlier referee comments requested clarification, particularly because vertical velocity is essentially a model-produced variable and so is subject to the details of each system (model vertical coordinate, vertical resolution especially in the TTL region, in addition to model physical parameterizations, etc). I only saw a brief mention in one of the author responses that the conversion to altitude is done using the hypsometric equation. Details are needed, and I'm not sure that is the best way to do things, if the authors are saying the performed some kind of integration themselves using temperature profiles to determine geometrical altitude (z). The most straightforward method would seem to be using the reanalysis Geopotential height fields to specify the altitude of each pressure level at each grid point where pressure velocity is evaluated. Can the authors please provide specific details about how geometric altitude conversion was performed. This could potentially clear up any underlying biases or disagreements among the observations and reanalyses w profiles.

Finally, the manuscript does not mention that one very large source of observations in the troposphere and lower stratosphere come from radiosondes (in fact, a search of the manuscript finds no mention of radiosondes at all, which I find quite surprising). First, I believe radiosondes can provide vertical wind information that is directly assimilated into these analysis systems. I would assume there have been comparisons between the VHR radar vertical velocities and nearby radiosonde observations (a very brief search provides many results, including some early studies by some of the coauthors of the present manuscript). If so, these should be mentioned, and how do they compare? This would further support the validity of the radar observations and also further highlight possible issues with reanalyses (i.e, if they assimilate radiosonde vertical velocities and still giving different results, that’s an issue).

Specific items:

1. Throughout the manuscript please use UTC consistently instead of UTS and GMT.

2. Make sure all figure captions clearly indicate the time period of the results. This is missing from, e.g., caption of Figs. 4, 5, 8, 9, 10, 11, 12, 13.

3. Lines 51-52: This statement clearly is not true for aircraft measurements. Please revise to more accurately capture what the relevant point is (e.g., that independent ground based observations are limited in their geographic distribution).

4. lines 89-91: Meteor radar is a technique for the mesosphere, and is not applicable in the present study, this should be removed.

5. The abstract and introduction should clearly state that this study is focusing on the troposphere and lower stratosphere. I had to read quite far into the manuscript to determine the scope of the study. Since this is submitted to the SRIP special issue, revisions that specifically address the relevance of this study to SRIP would be greatly beneficial.

6. Lines 101-102 and elsewhere: Throughout the manuscript the authors use the term tendency to describe the sign of the vertical velocity (positive or negative). This is in opposition to the usage I am most familiar with, i.e., a time tendency, or specifically, a time derivative. I think it is better to say here that the best way (in the authors’ opinion) is to evaluate time-mean profiles, which is what they are doing in figures 4, 5, 6, etc.

Hide

ED: Publish subject to minor revisions (review by editor) (19 Oct 2020) by Gabriele Stiller

AR by Uma K N on behalf of the Authors (05 Nov 2020)

ED: Reconsider after major revisions (20 Nov 2020) by Gabriele Stiller

AR by Uma K N on behalf of the Authors (23 Nov 2020) Manuscript

ED: Referee Nomination & Report Request started (23 Nov 2020) by Gabriele Stiller

RR by Anonymous Referee #3 (02 Dec 2020)

ED: Publish subject to minor revisions (review by editor) (02 Dec 2020) by Gabriele Stiller

AR by Uma K N on behalf of the Authors (09 Dec 2020) Author's response Author's tracked changes Manuscript

ED: Publish as is (14 Dec 2020) by Gabriele Stiller

AR by Uma K N on behalf of the Authors (17 Dec 2020) Manuscript

Short summary

Reanalysis data of vertical wind (w) are widely used by the atmospheric community to determine various calculations of atmospheric circulations, diabatic heating, convection, etc. There are no studies that assess the available reanalysis data with respect to observations. The present study assesses for the first time all the reanalysis w by comparing it with 20 years of radar data from Gadanki and Kototabang and shows that downdrafts and peaks in the updrafts are not produced in the reanalyses.