A Global Conversion Factor Model for Mapping Zenith Total Delay onto Precipitable Water

Round 1

Reviewer 1 Report

Overall comments:

Thanks to the authors for proposing a new model for the conversion factor between ZWD and PWV, which is indeed a critical parameter for GNSS remote sensing. As we lack meteo-data at many GNSS sites, the proposed modelis is of huge interest for the community. The refinement  for different periodic signals is for sure a benefical approach.

However, I have to a few points which could/should still be addressed by the authors.Some suggestions in the following:

• Section 2: The outliers seen in the relation between height and conversion factor might also be introduced by the fact that the majority of elevations above 4000-5000 m (mainly Himalayan region) are located in the latitude band which shows the highest values (see Figure 5).
• Section 3: The pattern of largest errors is interesting, maybe you can check the GGOS data you obtained in these regions (around the equator) shows anomalies linked to El Nino.
• Case study: This would be my main point: Including more stations (also on the global scale) into your case study would be good in order to have more convincing statistics. Moreover, it would be really benefical for the analysis if you could compare your results with an external source of PVW (radiosondes, ERA-reanalyis, satellite observations, ...) as e.g. Yao et. al (2015) did. This would really represent an independent validation of your model in terms of the paramters of interest (PWV). Best would be a site with radiosonde data which is not included/used in GGOS. If this is possible for you, I think the paper would really benefit from it. Or a grid-wise validation using reanalysis data.
• Conclusion: In my opinion, the conclusion should be a bit longer and some relation to other studies/models (e.g. Yao et al. (2015)) should be given in terms of performance and benefits your approach has to offer.  In these last sentences, the reader should be (once more) convinced why to prefer your model over any other for converting ZWD to PWV.

Smaller comments:

1. Line 51/52: this statement is not true, you cannot calculate precipitation from GNSS observations. GNSS observations (like ZTD or PWV) can aid numerical weather models (through data assimilation) to improve forecasts of precipitation. Please correct
2. Line 58: I would add here that ZWD is typically estimated as an unknown paramter in GNSS processing (as you do it for the case study later)
3. Line 277-280: maybe add an explaination why you chose these two stations? Do you have meteo-data on site there? This I could not figure out completely from the text

Author Response

Thanks for the reviewer's valuable suggestions, which has a significant improvement for the manuscript. The point-by-point response has been given in the PDF.

Author Response File: Author Response.pdf

Reviewer 2 Report

REVISION MANUSCRIPT Remote Sensing- 1587989: “A global conversion factor model for mapping zenith total delay onto precipitable water”

 General comments:

The authors overcome the dependence on high-precision atmospheric weighted average temperature for mapping ZWD onto PWV, and build a global non-meteorological parametric model for a given conversion factor. The results show that the  model has a high accuracy, which can be used to calculate the PWV value where no observed meteorological parameters are available. Therefore, I found the research interesting. However, some suggestions and revisions need to be applied in order to consider for publication.

Specific comments:

Line 61:  I recommend moving the PWV equation as well as its respective description into Section 2.

Also, in the discussion to obtain the PWV value from the Zenith Total Delay (ZTD), you should include the following references:

https://doi.org/10.1029/2004JD005715

https://doi.org/10.1007/s10291-012-0258-8

Line 122: The authors used a temporal resolution of ~6 hours between samples from the GGOS atmosphere products. The question is, it is possible to obtain data with better temporal resolution? If so, an analysis comparing the results should be interesting.

Author Response

Thanks for the reviewer's valuable suggestions, which has a significant improvement for the manuscript. The point-by-point response has been given in the PDF.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All my comments have been adressed, thank you.

However some minor suggestions from my side still:

• Station selection: While it is clearly beneficial to have meteorological measuremnts at the same site, I do not really understand the second point:  why having high PW values brings a benefit for the vaildation?
• El-Nino influence: it is still not as clear as you write in the discussion if those outliers really are linked to El-Nino so I would be cautious to state that here
• Final point: please go through all the new passages in the manuscript and have another look at the English style and grammar. There are a  number of new typos and grammar errors.

Author Response

Thanks for the reviewer's comments, and the manuscript has been carefully revised according to the reviewer's suggestion.

Author Response File: Author Response.pdf