A Comprehensive Evaluation of Key Tropospheric Parameters from ERA5 and MERRA-2 Reanalysis Products Using Radiosonde Data and GNSS Measurements

Round 1

Reviewer 1 Report

The resubmitted manuscript was revised accepting many of the reviewers' advices and comments. Now text, equations, tables and graphs are much more homogeneous.
It seems to me that it is still necessary to double check the English. There are errors, in some cases the verb seems to be missing in the sentence:
e.g. Lines 176-177: The detailed introduction for calculating the temperature, pressure and specific humidity of grid points at the station height using pressure-level data in Huang’ paper;
or Lines 182-183: Therefore, the grid points of water vapor pressure values need to calculated firstly by using formula (5);
or the term higher is used (e.g. Line 30) without saying with respect to what.
Or other errors such as at Line 38: in my opinion it should be: The Saasmoninen model can calculate with high accuracy the zenith hydrostatic delay or The Saasmoninen model can calculate high accuracy zenith hydrostatic delay.

Also there are still some repetitions, although I understand that it is often better to repeat several times the names of vriables or models for clarity. A balance must be found.

If the proposed changes are made, the manuscipt may be suitable for publication.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

See attached ascii text file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors made the necessary corrections to the issues I raised. And most importantly, they improved the figures.
In my opinion, the paper can now be published.
Good luck

Author Response

Please see the attachment.

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

The manuscript “A Comprehensive Evaluation of Key Tropospheric Parameters 2 from ERA5 and MERRA-2 Reanalysis Products Using Radio-sonde Data and GNSS Measurements” presents an interesting study on the new reanalysis products available to overcome spatial and temporal resolution problems present in the preceding products. The paper follows other studies, enhancing the assessment of characteristics of the tropospheric parameters worldwide.

Introduction and Data and Methodology sections are quite clear, the Result section needs some improvements in the text. There are many repetitions when the different results are presented. The text should be carefully revised in order to have a smoother reading.

Conclusions section: The results are comparisons of the same parameter series obtained through different estimates. If there is a seasonal variations, one should also figure out which series induces this behaviour. Since GNSS and Radiosondes are taken as references, it is therefore the reanalysis products that are seasonally dependent. The proposed explanation is due to the complex climate and the high variability of water vapor. However, I wonder if the estimate is global, shouldn't the seasons in the two hemispheres balance each other? Does the Northern hemisphere weigh more due to the much greater number of GNSS and RS stations there? Have you tried to select a subset of stations homogeneously distributed worldwide and check if the trends are the same?

I wonder if the abbreviations (MERRA-2_pl, MERRA-2_sfc, ERA5_pl, ERA5_sfc) could be used in the text in some cases, as well as in the figures, for better smoothness.

The term “obvious” is used many times in the manuscript. I wonder if in some cases other terms could be used instead: clear, evident, expected….

Figure captions: In all the figures where bias and RMS are displayed, biases are shown first and then the RMS, but in the caption, RMS are presented first, why? For linearity, I would match graphs and descriptions.

Also in the text, the parameters should be listed in the order of appearance. As an example: Line 394 Figures 15 and 16 represent ZTD and ZWD and not the opposite as written there.

In section 2 Radiosonde data are presented before GNSS data. On the contrary, in section 3 the comparison between reanalysis products and GNSS are presented first.

Figure colors: it would be better that the 4 sets of solutions were always represented with the same 4 colors in all figures from 15 to 24. In some cases there are different colors or exchanged between pl and sfc.

Line 133-134: The sentence seems incomplete.

Equation 1 and 2 should have the same dimension.

Table 1 and 2 should have the same format. In Table 1 max, min and mean values are presented, in Table 2 mean, min and max values are listed. Furthermore, horizontal lines should be added to divide the different parameters (ZTD, ZHD ...), for clarity.

Line 392: Figures 15-18 (instead of Figure 15).

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Review of "A Comprehensive Evaluation of Key Tropospheric Parameteres from ERA5
and MERRA-2 Reanalysis Products Using Radiosonde data and GNSS measurements", by
Lijie Guo et al.

This manuscript assess the quality of ERA5 and MERRA-2 data as regards ZTD, ZHD,
ZWD and Tm. As such something that deserves publication.

However, the manuscript lacks precission. In part this is due do not being 
careful with the sciencetic description, in part du to the level of English.

For that reason I have to require a major revision of  the manuscript. I urge
the authors to produce a more precise manuscript. It is not difficult.

In all friendliness, there is too much lack of precision for a manuscript
with 9 authors.

These are harsh words, but let me give a few examples, which will hopefully
both convince you there are problems, and make you realise they are easy to solve.

1) The very first sentence of the abstract is not meaningful in English.
Scientifically positioning and derivation of PWV is something quite different. PWV
estimation requires Tm, but GNSS positioning does not. But that sentence does not
make such a distinction.

2) In line 359 you speak about the offset of ERA5 and MERRA-2, saying they are
within 2 cm. I think you mean with with respect to the radiosondes!? But your
sentence means ERA5 with respect to MERRA-2. To write a scientific paper, allways
write with respect to what an offset or bias is measured.

There are several such examples in the manuscript.

Additional recommendations for the revision.

Put less attention on the surface level results.
Just show a table with statistics to demonstrate the lack of quality.

If you want to exploit the seasons by means of monthly statistics, remember 
to to split up into hemispheres - for example use Southern, Equator, Northern - as latitude
divisions. Otherwise we learn very little, except perhaps that there are more
GNSS sites in the north than in the south.

You show results for altitudes below 0 m. There are not many of those locations
with radiosondes or GNSS stations in the real world I think. Do you really have
data enough for a plot like figure 10? Or are you using the ERA5 altitude, which
might be slightly below 0 when the real altitude is above 0 
because ERA5 is a spectral model?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Paper review: A Comprehensive Evaluation of Key Tropospheric Parameters from ERA5 and MERRA-2 Reanalysis Products Using Radiosonde Data and GNSS Measurements

This paper makes an assessment of tropospheric parameters (ZHD, ZWD, ZTD, Tm) using data from two different atmospheric models (ERA5 and MERRA-2). In general, it is well written, although there is a significant repetition of the same terms throughout the text, making reading difficult. The authors find some statistical differences but do not point to their cause. Could this difference arise due to the different resolutions between the models? Does it have to do with the fact that ERA5 has more data assimilated than MERRA-2?

This study will only be well justified if the authors explain the main differences between the models, which eventually explains the results obtained. If authors make an effort to explain small details, I believe that the paper deserves to be published.

The abstract is too long, should be more concise, with less discussion and conclusions in this section.

The introduction could be better referenced. For example, it was enough to do a simple search in the MDPI site to find a global model based on the ERA5 (An ERA5-Based Hourly Global Pressure and Temperature (HGPT) Model) that should be referenced in this paper.

Figure1, It would be much better to put a legend instead of reading the meaning of the circles and color in the caption.

Equations 1 and 2 have different symbols (size of the sum symbol), etc.

Line 168, one reference is required for the refractivity coefficients.

In equation 5 there must be an approximation sign and not an equal sign.

Line 171, 172, what are the input units of these variables.

The first coefficient of Equation 8 is not consistent with that found in the Saastamoinen paper; see formula 56b. The signs of the denominator also do not correspond to the formulation found in the same paper. See the last formula in the paper.

Equations 10, 11, and 12 are applied only to the surface-level data? If yes, could the results in table 1 be influenced by these equations? Please explain this point further.

Do you make any corrections to the ellipsoidal altitude? To convert this to the orthometric (used in the atmospheric models).

It is very strange the differences between surface- and pressure-level products in terms of rmse. All products at the surface level are obtained by vertical interpolation of the pressure level products. Can you explain a little more about this topic? Where does this difference come from?

Line 495, 15.3% can we attribute this difference between models to the different spatial resolution?

Author Response

Please see the attachment.

Author Response File: Author Response.docx