Retrieval of Daytime Total Column Water Vapour from OLCI Measurements over Land Surfaces

Round 1

Reviewer 1 Report

The manuscript includes a report on activities aimed at creating the possibility of assessing the water vapor content in the atmosphere. It is an important parameter for weather forecasting as well as for water balance assessment.

L121/ It seems to me that definition (1) represents “radiance reflectance” for define wavelength band. “Normalised radiance” is unlikely to fit here. But I don't insist.   

Tab. 2/ What do the authors want to express with blue?

L216 & 219/ Lambert-Beer or Beer-Lambert?

L255-259/ Check punctuation marks.

Equation 12/ When I saw the dots, I thought it was the dot product of vectors ...

L498/ -0.34 kg/m2 is it correct?

I like the web addresses in the text, but please be prepared that the Editorial Board will ask you to remove web addresses from the text. Is there any advice on how to easily return to the text after clicking on the reference number?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper announces an important improvement in the retrieval of total column water vapor (TCWV) over land using satellite-based visible and near-infrared observations of the Earth's surface. The authors highlight the advantage of continuity with the Envisat MERIS mission, which also had a TCWV product. The MERIS TCWV algorithm used only one nearby window band, whereas the improved algorithm for Sentinel-3 OLCI introduced in this study uses two nearby window bands. Does this fundamental change in TCWV retrieval introduce a bias difference between MERIS TCWV and OLCI TCWV? The authors should at least comment on this possibility; ideally, they should make a formal comparison of the MERIS TCWV time series and OLCI TCWV time series. Or perhaps this is not an issue because longitudinal TCWV records are meaningless, i.e., TCWV is a geophysical parameter of interest only in the short term?

Overall, the article runs a little long. I think the authors could either shorten some of Section 3 by referring to the literature or shorten Section 4's discussion of the uncertainty evaluation, which is better placed in an external Algorithm Theoretical Basis Document (or Supplemental Materials).

More specific comments are presented below.

Line 5: "passive" may not be a useful distinction here, as OLCI is also a passive electro-optical sensor?

Lines 43-45: I'm assuming the alternative terms for TCWV ("IWV" or "TPW") are terms used in the wider literature and not used interchangeably in this paper, specifically. Please clarify.

Lines 64-65: This sentence could be made more clear, for example: "However, MODIS and MERIS-based TCWV observations require clear-sky conditions, and data dropouts from this requirement can lead to a significant difference of up to..."

Line 65: Please re-phrase "For a German domain..." Spatial domain? Topical domain? Are you talking about a German study or is there something particular about TCWV retrievals over Germany?

Lines 65-67: Speaking of the "German domain..." Are you referring to the climate of Germany--that acquiring clear-sky observations over Germany is not as difficult as over, e.g., tropical rainforests?

Lines 71-73: Somewhat awkward phrasing here... I would suggest instead, for example: "The launch of the Copernicus Sentinel-3a and Sentinel-3b satellites...provides for continuity with the Envisat MERIS mission..." Or, e.g., "continues the legacy of the Envisat MERIS mission..." Or, e.g., "enables a continuation of the Envisat MERIS instrument's visible and thermal imaging..."

Line 92: Again, "German domain" might be better phrased as "for the region of Germany"

ca. Line 121: How is F_i (in-band solar irradiance) established? Is this a fixed value (e.g., from ground-based spectral measurements)?

ca. Line 122 and Equation 2: Please provide a reference for the equation and, more importantly, the units. Are p_sur and p_sea in Pascals? Is h in meters?

Lines 126-137 and Table 2: It is not clear how band Oa17 (865 nm) is used in the existing OLCI TCWV product.

Lines 218-219: You write: "the absorption measured by a medium resolution satellite instrument cannot be quantified by the simple Lambert Beer law." Why is that? Because the satellite bands are too broad? Some additional references in this paragraph would help the reader.

Lines 305-306: The wording is a bit confusing here... I would recommend changing (and shortening) this sentence to: "The stopping criteria for the optimization are as follows." Or, combining the first two sentences (and acknowledging there is only one one stopping criteria, hence, criterion): "The stopping criterion [or "stopping condition"] for the optimization is based on the step width relative to..."

Lines 312-313: I would recommend using a single letter/ symbol for the air mass factor rather than "amf" as the latter could be construted as the product of three different terms: a, m, and f. Alternatively, all-capitals "AMF" and not italicized (e.g., with TeX declaration \mathrm) would make this more clear.

Lines 316-317: It would be helpful if you could remind the reader why the optical thickness here corresponds to the TWVC; i.e., why other species (e.g., N2O, aerosols) do not contribute to this optical thickness. Is that solely because of the bands selected (with the correction factors of Section 3.3)?

Lines 335-336 and Equation 12: Again, recommend against using "amf" for air mass factor.

Lne 354: What is the significance of the 10-km radius? Is this an approximation of the region of influence on ARM observations?

Figure 7 caption: You write: "The circles and colours within show the GPS stations with valid TCWV retrievals at the OLCI overpass time." I think the circles show the TCWV measurement from the German GNSS station? If that is the case, the caption would read more accurately as: "The circles and colours within show the GPS stations with and GNSS TCWV retrievals at the OLCI overpass time for GPS stations with valid retrievals."

Figure 7: Perhaps the left panel should be masked so it only shows TCWV retrievals over the land?

Lines 482-483: "has been used" for what? I think you're referring to the correction described in Section 3.3. You conclude that "[uncorrected] the deviations would lead to a wet bias" and, indeed, this seems to be the case from
Figure 4. However, as you acknowledged, the ARM data cannot be used to validate this improvement because they are not independent. I would recommend contextualizing this claim--a wet bias correction is probable, but not certain, given the prior literature and your experiment.

Lines 484-490: This would be a good place to discuss the compatibility of the new OLCI TCWV observational record with the previous MERIS and MODIS records.

**Minor typographical/ grammatical edits:**

Line 144: "are performed" should be "is performed"

Line 147: "Netwerk" should be "Network"

Line 151: "are given" should be "is given"

Line 201: "independent on" should probably be "independent of"

Line 246: "and fixed, additionally assumed parameter" should probably be "and additional parameters assumed to be fixed"

ca. Line 260: "radiometer based" should be "radiometer-based"

ca. Line 260: "camera dependent" should be "camera-dependent"

ca. Line 260: "description water vapor absorption" should probably be "description of water vapor absorption"

Line 294: "a priory" should be "a priori"

Line 296: "bias free" should be "bias-free"

Line 300: "gradient based" should be "gradient-based"

Line 391: "miss-classified" should be "mis-classified"

Line 450: "and possibly be related" should either be "and possibly related" or "and could possibly be related"

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. The manuscript presents retrieval of daytime total column water vapour from OLCI measurements over land surfaces, which is interesting. The subject addressed is within the scope of the journal.
2. However, the manuscript, in its present form, contains several weaknesses. Appropriate revisions to the following points should be undertaken in order to justify recommendation for publication.
3. Full names should be shown for all abbreviations in their first occurrence in texts. For example, MERIS in p.1, MODIS in p.1, AERONET in p.1, ARM in p.1, etc.
4. For readers to quickly catch your contribution, it would be better to highlight major difficulties and challenges, and your original achievements to overcome them, in a clearer way in abstract and introduction.
5. It is shown in the reference list that the authors have several publications in this field. This raises some concerns regarding the potential overlap with their previous works. The authors should explicitly state the novel contribution of this work, the similarities, and the differences of this work with their previous publications.
6. 1 - a retrieval algorithm based on the differential absorption technique is adopted for retrieval of daytime total column water vapour from OLCI measurements over land surfaces. What are other feasible alternatives? What are the advantages of adopting this algorithm over others in this case? How will this affect the results? The authors should provide more details on this.
7. 1 - AERONET, ARM and U.S. Suominet, and a German GNSS TCWV data set are adopted as reference data sets for the validation. What are the other feasible alternatives? What are the advantages of adopting these data sets over others in this case? How will this affect the results? More details should be furnished.
8. 1 - absolute bias and root mean square error are adopted to test the accuracy of the models. What are the other feasible alternatives? What are the advantages of adopting these evaluation metrics over others in this case? How will this affect the results? More details should be furnished.
9. 5 - Radiative Transfer based simulation is adopted to estimate the water vapour column above a pixel. What are other feasible alternatives? What are the advantages of adopting this approach over others in this case? How will this affect the results? The authors should provide more details on this.
10. 5 - a one-dimensional gradient descent (Newton secant method) is adopted to perform the optimisation with respect to the total column water vapour. What are other feasible alternatives? What are the advantages of adopting this method over others in this case? How will this affect the results? The authors should provide more details on this.
11. 6 - a 1D-Var scheme is adopted to optimise the difference between simulated and measured radiances. What are other feasible alternatives? What are the advantages of adopting this scheme over others in this case? How will this affect the results? The authors should provide more details on this.
12. 7 - an advanced k-distribution routine is adopted to calculate the needed gaseous absorption coefficients. What are other feasible alternatives? What are the advantages of adopting this approach over others in this case? How will this affect the results? The authors should provide more details on this.
13. 7 - the radiative transfer Matrix Operator Model is adopted to simulate interactions with aerosol particles and molecules. What are other feasible alternatives? What are the advantages of adopting this model over others in this case? How will this affect the results? The authors should provide more details on this.
14. 9 - the Gauss-Newton update step as shown in Eq.(6) is adopted in the algorithm. What are other feasible alternatives? What are the advantages of adopting this equation over others in this case? How will this affect the results? The authors should provide more details on this.
15. 10 – Eq.(7) following Rodgers [48] is adopted as a criterion. What are other feasible alternatives? What are the advantages of adopting this criterion over others in this case? How will this affect the results? The authors should provide more details on this.
16. 11 - historical records of 2017 to 2018 are taken. Why are more recent data not included in the study? Is there any difficulty in obtaining more recent data? Are there any changes to situation in recent years? What are its effects on the result?
17. 16 - an evaluation framework as described in Sayer et al. [60] is adopted to assess the quality of the TCWV uncertainty estimates provided by the optimal estimation retrieval scheme. What are other feasible alternatives? What are the advantages of adopting this framework over others in this case? How will this affect the results? The authors should provide more details on this.
18. 17 - “…Alternatively, it could mean that the AERONET uncertainties are overestimated. We tend to lean to the latter, since.…” More justification should be furnished on this issue.
19. Some key parameters are not mentioned. The rationale on the choice of the set of parameters should be explained with more details. Have the authors experimented with other sets of values? What are the sensitivities of these parameters on the results?
20. Some assumptions are stated in various sections. Justifications should be provided on these assumptions. Evaluation on how they will affect the results should be made.
21. The discussion section in the present form is relatively weak and should be strengthened with more details and justifications.
22. Moreover, the manuscript could be substantially improved by relying and citing more on recent literatures about contemporary real-life case studies of remote sensing such as the followings:

• Xu, X., et al., “Regime shifts in shallow lakes observed by remote sensing and the implications for management,” Ecological Indicators 113: 106285 2020.
• Faroughi, M. et al., “Computational modeling of land surface temperature using remote sensing data to investigate the spatial arrangement of buildings and energy consumption relationship,” Engineering Applications of Computational Fluid Mechanics 14 (1): 254-270 2020.
• Kothari, N.S., et al., “Semisupervised classification of remote sensing images using efficient neighborhood learning method,” Engineering Applications of Artificial Intelligence 90: 103520 2020.

23. Some inconsistencies and minor errors that needed attention are:

• Replace “…trough the years…” with “…through the years…” in line 38 of p.2
• Replace “…Additionally, to numerical criteria…” with “…Additional to numerical criteria…” in line 305 of p.10
• Replace “…the following criteria as is used…” with “…the following criterion is used…” in line 306 of p.10
• Replace “…The criteria is based on…” with “…The criterion is based on…” in line 307 of p.10

24. The conclusion section is missing. In the conclusion section, the limitations of this study, suggested improvements of this work and future directions should be highlighted.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

No more comments.