Method for Determining Neutral Wind Velocity Vectors Using Measurements of Internal Gravity Wave Group and Phase Velocities
, Roman V. Vasilyev and Maxim F. ArtamonovRound 1
Reviewer 1 Report
See the attached file.
Comments for author File: 
Comments.pdf
Author Response
Reply in pdf file.
Author Response File: Author Response.pdf
Reviewer 2 Report
The paper deals with the problem of wind measurements in the thermosphere. The authors claim that they developed a new method for determining the neutral wind velocity vector from the incoherent scattering radar and ionosonde data. In principle, the paper contains new and valuable information, which can be interesting for publication in the Atmosphere journal. However, the new method is not described adequately, and the paper practically does not contain any analysis of errors of the obtained results. This makes difficulties in understanding the applicability of the method and results obtained in the paper. Therefore, I think that the paper could be published only after substantial revision.
More specific comments are the following:
Page 1, line 19. “… demonstrated that two independent methods…” Fabri-Perot method is not practically described in the paper. It is difficult to understand that the Fabri-Perot and radar/ionosonde methods are really independent.
Page 3, eq. 4. There is no references, where the eq. 4 is taken from? What is ω in eq. (4)? – previously only ω’ and ωobs were described.
Page 4, lines 112-113. How the k-spectrum was obtained from the measurements? What is ∆ω? How big is it in your measurements?
Page 4, eq. 8. Why group velocity is the same at different frequencies? Generally IGW group velocity might be different at different frequencies.
Page 4, line 120. It is not clear, how frequencies ω1, ω2, ω3 were chosen from the specified set.
Page 5, lines 135. What is the Hines dispersion relation? What is the difference? Does the Hines equation take into account dissipative and ion-drag processes, which should be important in the thermosphere and ionosphere?
Page 5, line 141-142. The least square method allows estimation not only wind parameters, but also there variances, which are essential for estimations of measurement errors.
Page 6, lines 173. There is no information how many data points were used for comparisons. Lines in Figure 3 do not contain error bars.
Page 6, lines 174-191. The paper does not contain evidences that vertical velocity values obtained both form Fabri-Perot and radar/ionosonde observations are statistically confident.
Author Response
Reply in pdf file
Author Response File: Author Response.pdf
Reviewer 3 Report
Review of the paper by A. Medvedev et al.
In this paper the authors describe a new method of measuring neutral wind velocity in the ionosphere using the group velocities of internal gravity waves. In the Introduction they refer to their previous work where the network is described: “… Ionosonde DPS-4 and two beams of the Irkutsk Incoherent Scatter Radar (IISR) [1] form a triangle with ~ 100 km side, convenient for investigating medium and large-scale traveling ionospheric disturbances (TID).” As mentioned in the Introduction the measuring of three electron density profiles at three spaced sites and applying to these profiles cross-correlation and phase difference analysis enables estimate the parameters of propagation of TID (phase velocities and wave numbers). However, the description of the method of measuring components of the wind velocity (including vertical component) using formulas (5)-(8) can’t explain the capability of the method to estimate wind components in a way the authors claim. I do not recommend publishing this paper in the journal “Atmosphere”. Below I explain the drawbacks of the paper.
1)First of all, the method is based on the assumption that the variations of wind components are caused by linear internal gravity waves (IGWs) with the frequencies ω1, ω2, ω3 connected to the wave numbers by linear dispersion relations (4) that allow to calculate the components of group velocity components (5).
However, the authors do not explain how one can filter the linear IGWs from the moving eddies or nonlinear gravity waves (which always exist at high altitudes) that contribute to the wind variations, but do not obey linear dispersion relations (4) or Hines dispersion.
2) The finite differences of wave number vectors and frequencies are supposed to be measured with some measurement error, therefore the values of the group velocities as derivatives (frequency difference divided by wave number difference) will randomly scatter as well. Such scatter may significantly affect the stability of the estimates of the group velocities and wind velocity components, causing the uncertainty of these velocities. The authors do not take into account such uncertainty in their wind estimates.
3) Actually the authors do not explain how they obtain a set of frequencies (k-spectrum) and do not prove that these frequencies and wave numbers belong to linear IGWs.
4) There is a high deviation in Figure 3 (up to 80 m/s for months 9-10, left panel ) between the obtained wind values (black line) from the wind values obtained by the model HWM2007 (blue). What is the cause of such deviation?
5) Lines 107-109. The symbols for the wind velocity and wave number vectors are missed.
Author Response
Reply in pdf file
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Second review of the paper by A. Medvedev et al.
The authors substantially revised their paper in accordance with the answers to my comments, which made the presentation of its results more clear. I believe that the paper can be published after explaining in the text (not only in the answer to question 3 of my first review) the possibility of using the linear dispersion relation for internal waves and the source of the errors in the method for estimating the components of the wind velocity in the ionosphere described.
Author Response
The authors substantially revised their paper in accordance with the answers to my comments, which made the presentation of its results more clear. I believe that the paper can be published after explaining in the text (not only in the answer to question 3 of my first review) the possibility of using the linear dispersion relation for internal waves and the source of the errors in the method for estimating the components of the wind velocity in the ionosphere described.
We are grateful to the Reviewer for the careful perusal of our paper and for valuable remarks, which help us to had made the presentation of our results more clear. We have added explaining the possibility of using the linear dispersion relation for internal waves and the source of the method errors to the manuscript(from answers to review)
Author Response File: Author Response.pdf
