Sensitivity Analysis of the Differential Atmospheric Transmission in Water Vapour Mixing Ratio Retrieval from Raman Lidar Measurements

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript by Díaz-Zurita et al. covers an important topic of accurate water vapor mixing ratio detection. They discussed the errors of water vapor retrievals under two configurations (354+408 and 387+408), considering different atmospheric transmission terms. Error analysis is clear and comprehensive. The results is convincing and supported by observations. 

 

I suggest acceptance of this mansucript, after the authors can correct very few minor issues.

 

1. In Abstract, line 18, "systematic uncertainties" is confusing. It's better to say "systematic error".

2. In Abstract, line 19, it's better to explain "the first and second configuration".

3. In Introduction, line 39, the latest IPCC report should be cited. IPCC report (2007) is an old story...

4. Page 3, line 113, "200~km" and "50~km" should be corrected.

5. Page6, line 201, details of reference 42 should be added.

Author Response

Response to Reviewer 1 Comments

Comments and Suggestions for Authors

The manuscript by Díaz-Zurita et al. covers an important topic of accurate water vapor mixing ratio detection. They discussed the errors of water vapor retrievals under two configurations (354+408 and 387+408), considering different atmospheric transmission terms. Error analysis is clear and comprehensive. The results is convincing and supported by observations. 

 I suggest acceptance of this manuscript, after the authors can correct very few minor issues.

Response

We sincerely thank the referee for his/her time and suggestions, which have greatly contributed to improving the quality of this study. All the points raised have been carefully considered, and the corresponding revisions have been incorporated into the manuscript. Below we provide detailed, point-by-point responses to each comment (in blue).

Specific comments:

1. In Abstract, line 18, "systematic uncertainties" is confusing. It's better to say "systematic error".

Response

Thank you very much for this suggestion. Initially, we considered using the term systematic error, as it is more commonly known. However, according to the recommendations of the Joint Committee for Guides in Metrology in their Guide to the Expression of Uncertainty in Measurement (GUM, 2020) the correct term is systematic uncertainties (or systematic effects). This is because when carrying out experimental measurements in atmospheric sciences, the true value is generally unknown, so we cannot define a true “error.” Instead, what can be quantified are the deviations from the measured value in terms of estimated uncertainties, which reflect the combined effect of all known and potentially unknown sources of bias in the measurement process. Therefore, we believe that the use of systematic uncertainties is more appropriate in the context of lidar measurements, where the true is not known.

2. In Abstract, line 19, it's better to explain "the first and second configuration".

Response

We appreciate this valuable suggestion. To clarify what is meant by “the first and second configuration”, we have replaced the original sentence (lines 11–13):

"Such issue is evaluated for a vibrational–rotational Raman nitrogen (~387 nm) configuration and for a pure–rotational Raman molecular reference near 354 nm (nitrogen and oxygen). "

with the revised version:

"Such issue is evaluated for two optical configurations: the first is a vibrational–rotational Raman nitrogen (~387 nm) configuration, and the second is a pure–rotational Raman molecular reference near 354 nm (nitrogen and oxygen). "

We appreciate this suggestion. To clarify what is meant by the first and second configuration, we have replaced the original sentence (lines 11–13):

3. In Introduction, line 29, the latest IPCC report should be cited. IPCC report (2007) is an old story...

Response

We thank the referee for this important suggestion. The previous reference to the IPCC 2007 report has been updated to the IPCC 2021 report:

Douville, H.; Raghavan, K.; Renwick, J.; Allan, R.; Arias, P.; Barlow, M.; Cerezo-Mota, R.; Cherchi, A.; Gan, T.; Gergis, J.; et al. Water Cycle Changes. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Masson-Delmotte, V.; Zhai, P.; Pirani, A.; Connors, S.; Péan, C.; Berger, S.; Caud, N.; Chen, Y.; Goldfarb, L.; Gomis, M.; et al., Eds.; Cambridge University Press: Cambridge, United Kingdom and New York, NY, USA, 2021; Chapter 8, pp. 1055–1210. https://doi.org/10.1017/9781009157896.010

This IPCC chapter update highlights the importance of atmospheric water vapor as a key greenhouse gas, influencing energy transport, cloud formation, and climate feedback

4. Page 3, line 113, "200~km" and "50~km" should be corrected.

Response

The terms were corrected by 200 km and 50 km.

5. Page6, line 201, details of reference 42 should be added.

Response

Thank you very much for this suggestion. We apologize because the missing details in reference 42 were due to a mistake during LaTeX compilation. The citation has now been updated following the recommendation of the ACTRIS Centre:

O’Connor, E. Model data from Granada on 13 February 2025, 2025. ACTRIS Cloud Remote Sensing Data Centre Unit (CLU). License: https://creativecommons.org/licenses/by/4.0/

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Review: Sensitivity Analysis of the Differential Atmospheric Transmission in Water Vapour Mixing Ratio Retrieval from Raman Lidar Measurements

The article is well written, clear, well-structured, and proposes a practical approach to improving water vapour profiles.  The authors propose a new technique for correction of the differential atmospheric transmission, yielding an improved accuracy for retrieved water vapour profiles.

The Introduction is sufficient and well cited, the lidar system is well described, the methods are sufficiently detailed and the Raman technique is well established in the literature, the improvements to the water vapour profiles are clearly demonstrated in the results, and the conclusions are well-supported.

In my view, the paper could be published in its current form.  I will make two comments (which the authors can consider or not at their discretion).

1. Consider defining MULHACEN.  Not really needed, but I am curious what it stands for
2. Figure 4, interpolated AOD values during nighttime.  It’s a reasonable approach but could be improved in future work.   Is it possible to try a hybrid approach?  Perhaps using a global reanalysis and weighting by the most recent local AOD measurements?  Or perhaps AOD could be estimated from an all sky camera at night.

Author Response

Response to Reviewer 2 Comments

Comments and Suggestions for Authors

Review: Sensitivity Analysis of the Differential Atmospheric Transmission in Water Vapour Mixing Ratio Retrieval from Raman Lidar Measurements

The article is well written, clear, well-structured, and proposes a practical approach to improving water vapour profiles.  The authors propose a new technique for correction of the differential atmospheric transmission, yielding an improved accuracy for retrieved water vapour profiles.

The Introduction is sufficient and well cited, the lidar system is well described, the methods are sufficiently detailed and the Raman technique is well established in the literature, the improvements to the water vapour profiles are clearly demonstrated in the results, and the conclusions are well-supported.

In my view, the paper could be published in its current form.  I will make two comments (which the authors can consider or not at their discretion).

Response

Thank you very much for taking the time to review this manuscript. We sincerely appreciate your positive and constructive comments. Your feedback on the clarity, structure, and methodology of the study is greatly valued. Our detailed, point-by-point responses to your specific comments are provided below (in blue)

Specific comments:

1.Consider defining MULHACEN.  Not really needed, but I am curious what it stands for.

Response

Thank you very much for your curiosity about the name MULHACEN. The acronym does not correspond to any technical aspect of the lidar technique. Instead, it is the given name of first multiwavelength Raman lidar operated at the UGR urban station (Spain). This name was chosen because MULHACEN is the highest and most important mountain in the Iberian Peninsula. This peak is in Sierra Nevada (Granada), very close to our urban station, where we also operate a high-altitude site.

2. Figure 4, interpolated AOD values during nighttime. It’s a reasonable approach but could be improved in future work. Is it possible to try a hybrid approach?  Perhaps using a global reanalysis and weighting by the most recent local AOD measurements?  Or perhaps AOD could be estimated from an all sky camera at night

Response

We appreciate this suggestion, which highlights an important point for improving the estimation of aerosol extinction profiles from AOD measurements. The discrepancy between AOD values estimated from sun-photometer and those retrieved by lidar could be significantly reduced if actual nocturnal measurements were available, for example, using star or moon photometry. Although these approaches would be ideal, such measurements are not widely available, and unfortunately, our star photometer at Granada was not operating during MULHACEN lidar Raman measurements and is no longer operational.

For this study, we relied on daytime sun photometer measurements, which are continuously acquired and have proven to provide reliable results. By interpolating between the last daytime measurement and the following day, we consider that this approach provides a more accurate estimate of nighttime AOD than would be possible using model-based averages. This is particularly relevant for Granada, which is located in a complex orographic region where global or regional models are generally less precise and often struggle to reproduce aerosol local variability.

Despite this limitation, we did explore the potential use of models during the preliminary stages of the study, in particular CAMS. However, model data were not available for the entire study period, and after contacting with the model developers, they indicated that data was still being reprocessed. Moreover, if these data were published, CAMS AOD should undergo a validation for the Granada location before being able of considering them in our methodology. For the latter, we would still have the issue of no nighttime AOD measurement availability for the validation.

We fully agree that, in future work, hybrid approaches combining model reanalyses with local observations, or using nocturnal sky or star photometry when available, could further improve the estimation of aerosol extinction profiles at night. This hybrid method could be of great interest in places where Raman lidar systems operated alone without a collocated sun photometer.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper evaluates the effect of the differential atmospheric transmission term (ΔT) on the retrieval of the water vapour mixing ratio from Raman lidar measurements, and proposes an operational estimation method based on sun photometer AOD data. The study's results clearly quantify the systematic bias caused by neglecting this term, especially when using the configuration with a larger spectral separation. The conclusions and the proposed operational methodology are of significant importance for improving the accuracy of Raman lidar water vapour products and expanding their scope of application. I recommend this paper for publication after minor revisions.

I appreciate the authors' decision to prioritize operational simplicity by selecting the exponential decay model. However, for modern computing systems, the incremental computational load of an AOD-constrained Klett/Fernald inversion is arguably no longer a prohibitive factor. I suggest the authors discuss the potential for utilizing an AOD-constrained Klett/Fernald or generalized inversion method to estimate the aerosol extinction profile. This technique yields a profile whose column integral is strictly bound by the sun photometer data, offering significantly higher vertical structure accuracy than the exponential decay model. Could the authors discuss this method as a potential future direction for enhancing the ΔT correction precision, even if it entails a slight increase in computational complexity?

Besides, there are some format issues. For example, the citation in Line 61, [e.g., 11,13,19–23], seems to deviate from standard guidelines.

Author Response

Response to Reviewer 3 Comments

Comments and Suggestions for Authors

This paper evaluates the effect of the differential atmospheric transmission term (ΔT) on the retrieval of the water vapour mixing ratio from Raman lidar measurements, and proposes an operational estimation method based on sun photometer AOD data. The study's results clearly quantify the systematic bias caused by neglecting this term, especially when using the configuration with a larger spectral separation. The conclusions and the proposed operational methodology are of significant importance for improving the accuracy of Raman lidar water vapour products and expanding their scope of application. I recommend this paper for publication after minor revisions.

Response

Thank you very much for taking the time to review this manuscript. We sincerely appreciate your comments and suggestions. Our detailed, point-by-point responses to your specific comments are provided below (in blue), and the corresponding revisions have been incorporated into the manuscript.

Specific comments:

1. I appreciate the authors' decision to prioritize operational simplicity by selecting the exponential decay model. However, for modern computing systems, the incremental computational load of an AOD-constrained Klett/Fernald inversion is arguably no longer a prohibitive factor. I suggest the authors discuss the potential for utilizing an AOD-constrained Klett/Fernald or generalized inversion method to estimate the aerosol extinction profile. This technique yields a profile whose column integral is strictly bound by the sun photometer data, offering significantly higher vertical structure accuracy than the exponential decay model. Could the authors discuss this method as a potential future direction for enhancing the ΔT correction precision, even if it entails a slight increase in computational complexity?

Response

We greatly appreciate this pertinent suggestion. We would like to highlight that the ideal scenario would be to have aerosol extinction measurements obtained using the signal at 355 nm. The second ideal choice would be to implement Klett/Fernald inversions. However, there are cases where misalignments between elastic and Raman signals make this impossible. Apart from that, the lack of complete overlap in a lidar system such as MULHACEN introduces additional limitations to the use of Klett/Fernald or Raman methods. These facts make it difficult to have a complete database of inversions for the entire study period, which is a must if the methodology is to be applied correctly.

Because in many previous studies the differential atmospheric transmission term was often neglected, the main goal of this manuscript was to exploit this fact to demonstrate the impact of the differential transmission term using long-term robust statistics. The exponential decay of aerosol with height, although not ideal, serves to translate this systematic uncertainty into random uncertainty, which we believe is of great interest for the scientific community. This simple method is also ideal for very large databases such as the one used in our study. It supported the use of long-term reliable AOD measurements like AERONET sun photometer Level 2 data.

We tested our approach both against Klett/Fernald-style inversions and Raman-based retrievals, demonstrating that it significantly reduces uncertainties in the water vapor mixing ratio while maintaining operational simplicity. In particular, the method avoids the need for explicit inversions and additional aerosol channels, representing a practical advantage in terms of computational efficiency, human resources, supervision, and systematic consistency.

To clarify all these points, we have modified the manuscript accordingly. In Section 3.2 in lines 213 it is given as:

Ideally, the impact of aerosols on water vapor mixing ratio retrievals could be addressed if measurements at 355 nm were available. However, possible misalignments between backscatter and Raman measurements make this impossible. Additionally, the differences in the incomplete overlap region for the backscatter signal at 355 nm led us to look for an alternative to minimize this effect. Under such circumstances, many previous studies have neglected ΔT (z, λMR, λH2O), particularly in the analyses of very large databases due to computational effort. But the calculation of …

And also, in the conclusion section (line X)

This study presents a sensitivity analysis of ΔT (z, λMR, λH2O) with respect to wavelength and aerosol loading, using lidar inversions and an operationally feasible approach based on sun photometer (SP) aerosol optical depth (AOD) measurements, where aerosol extinction is modeled using an exponential decay function with altitude. Although this methodology cannot replace more robust retrievals based on the 355 nm backscatter signal, it effectively mitigates the limitations arising from misalignments between the backscatter and Raman channels, from incomplete-overlap effects, and even enables an approximate correction of aerosol effects for lidar systems lacking dedicated aerosol channels.

2. there are some format issues. For example, the citation in Line 61, [e.g., 11,13,19–23], seems to deviate from standard guidelines.

Response

We thank the referee for detecting these typos. We have updated the manuscript to ensure that all references now follow the correct format.

Author Response File: Author Response.pdf