Round 1

Reviewer 1 Report

The study performed in this paper is of interest to the OTH radar community and should be published.  It is well written and informative and I found no errors in the methodology or presentation.  There is one typo in line 456: Figure) were accounted ...

Author Response

Thank you for taking the time to assess out manuscript. 

 

Response to Reviewer 1 Comments

The study performed in this paper is of interest to the OTH radar community and should be published.  It is well written and informative, and I found no errors in the methodology or presentation. 

Point 1. There is one typo in line 456: Figure) were accounted ...

This typo has been corrected.

Reviewer 2 Report

This paper discusses the calculation of sea surface backscattering coefficient of over the horizon radar under the condition of large incident angle, and focuses on the comparison of the calculation results of Barrick and ionogram methods. I think the calculation process and result analysis in this paper are correct, and some interesting phenomena have been found. The results have certain reference significance for one to deepen the understanding of sea surface scattering mechanism and the application of sea state information detection by a radar.

My comments or concerns are as follows:

1. It is suggested that the "sea backscatter coefficient" in this paper had better be changed to the "sea surface backscatter coefficient"  because the radar echo is mainly generated by the sea surface.
2. Lines 224-226 in the manuscript, "The spatial median backscatter coefficient for each range cell was 224 calculated from the Barrick method results using a radar frequency closest to the mean 225 frequency of the rays reaching that location". Why is the "median" backscattering coefficient calculated here instead of the "mean" backscattering coefficient? What is the reason?
3. Line 253: (-9˚N, 120˚E) is to express (9˚S, 120˚E)?
4. In equation (1) : Does K0 in S (− 2mk0) refer to radar wave vector or ocean wave number? In fact, the wave height spectrum is a function of ocean wave number and wave direction. Please check and clarify.
5. I wonder if the authors have considered the shadowing or occlusion effect between waves when calculating the backscatter coefficient under the condition of grazing incidence? How is it considered?
6. If possible, I hope to see the variation of backscatter coefficient with relative wind direction( the angle between radar observation azimuth and wind direction) in the calculation results.

Author Response

Thank you for taking the time to assess our manuscript. Please find our responses to your comments attached.

Author Response File: Author Response.pdf

Reviewer 3 Report

This study compares theoretical and observational sea surface scattering radar cross sections for skywave radar.
This study is significant because it contributes to the development of remote sensing technology using long-range ocean radar.

The radio frequencies covered are 6MHz-26MHz with Bragg frequencies ranging from 0.25Hz to 0.52Hz.
These are mainly wind-wave frequencies and are not considered to be swell frequencies.
In fact, the JOWSWAP-type spectra employed in this study are wind wave spectra.
However, Equation 7 shows that the JOWSWAP spectra of wind waves and swells are fitted from their wave heights and peak periods, respectively, and that the wave spectrum is the sum of them.
The JOWSWAP spectrum is proportional to k^{-3} at high wavenumbers (Equation 12), but it is doubtful that the swell spectrum is also proportional to k^{-3} at higher wavenumbers, as is the case with the wind-wave spectrum.
The reason that the radar cross-section is correlated with the observations is that the radar cross-section is highly dependent on the wind direction.
As a problem, what is the validity of adopting a JOWSWAP-type spectrum for the swell spectrum?
Is it valid to use the same spectrum for wind waves, especially at the higher 
frequencies corresponding to the Bragg frequencies?
You mention that you used significant wave heights (line 70), but were these used in the calculations for wind-wave heights and swell wave heights, respectively?
Was the peak period of the wind-wave used?
For the reasons discussed above, the validity of the wave spectra used in the calculations needs to be discussed, although the radar scattering cross-section may not be very dependent on the correlation of the observations.

Other comments

(1)Equation 5: From the definition of f, -2k_0 is not correct.
(2) Line 223: Did you use the radar cross-sections for lower radio wave frequencies at closer distances and higher radio wave frequencies at farther distances 
from Figure 4 for comparison?
(3) Figure 6: Why are there missing values in the right panel calculations?
If wave estimation can be done, can't radar cross-sections be calculated?
(4) Figure 9: Please indicate the number of data for comparison.
(5)Line 319: For each of the calculated and observed radar cross-sections, there are autocorrelations for time and space.
That is, the calculated (or observed) radar cross-sections are not statistically independent of each other. Therefore, it is complicated to evaluate the degrees of freedom (effective degrees of freedom).
In this case, I think the significance test is not necessary.
(6) Line 330: The "wave spectrum method" is only used in this section.
(7) Section 4: 
Issues related to modeling backscatter ionograms seem to be less discussed.
 Can we assume that this is accurate?

Author Response

Thank you for taking the time to assess our manuscript. Please find our responses to your comments attached.

Author Response File: Author Response.pdf