the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 May 2025
First results of the XBAER aerosol optical depth algorithm with EnMAP data
Abstract. New high-resolution hyper and multispectral satellite instruments enable the retrieval of aerosol optical depth (AOD) at spatial resolutions of tens of meters. The eXtensible Bremen AErosol Retrieval (XBAER) AOD retrieval algorithm has previously been developed for use with Ocean and Land Colour Instrument (OLCI) and MEdium Resolution Imaging Spectrometer (MERIS) radiance data. With the intention of later modifying XBAER to use the full 30 m spatial resolution data from the Hyper-Spectral Imager (HSI) on board the Environmental Mapping and Analysis Program (EnMAP) satellite, the present study investigates how HSI data compare to OLCI data. For the bands of interest, top of atmosphere reflectances generally compare well (R > 0.9), the intercept of the best fit line is less than 0.05 from the origin, and the slope is less than 0.1 from 1. However exceptions exist and these are explained as the result of differences in the spectral response functions of the instruments in the region of the spectrum around the O2 A-Band absorption feature, or as a result of differences in the viewing geometry of the satellites which produces differing bidirectional reflectance distribution function (BRDF) effects. XBAER is then used to retrieve OLCI and HSI surface reflectance (SRF) and AOD. For SRF the comparison between OLCI and HSI yields R = 0.953, best fit intercept = 0.003 and best fit slope = 1.082. The respective comparison for AOD yields R = 0.809, best fit intercept = 0.153 and best fit slope = 0.785. These comparisons are then separated by surface type and insights are gained into the performance of the algorithm. Finally, the unmodified XBAER algorithm is run using the full spatial resolution HSI data. Plumes from biomass-burning are identified in a single scene, and a comparison with AErosol RObotic NETwork (AERONET) AOD is performed for multiple scenes, achieving R = 0.631. Future modifications to XBAER that would allow it to produce more accurate retrievals at HSI's spatial resolution are discussed.
Competing interests: Linlu Mei is an editor of AMT
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(5931 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 17 Sep 2025)
-
RC1: 'Comment on egusphere-2025-1282', Anonymous Referee #1, 28 May 2025
reply
Review of the manuscript titled “First results of the XBAER aerosol optical depth algorithm with EnMAP data” by Laffoy et al.,
In this study, the authors apply the eXtensible Bremen AErosol Retrieval (XBAER) AOD retrieval algorithm to data from the Environmental Mapping and Analysis Program (EnMAP) satellite. The retrieved aerosol optical depth (AOD) values are compared with observations from the Ocean and Land Colour Instrument (OLCI). In addition, biomass-burning plumes are identified within a single scene, and a comparison with AErosol RObotic NETwork (AERONET) AOD measurements is conducted across multiple scenes. The results demonstrate strong agreement among these datasets.
The study also highlights that discrepancies between measurements can occur due to differences in the spectral response functions of the instruments, particularly in the region around the O₂ A-band absorption feature or due to varying viewing geometries, which influence the bidirectional reflectance distribution function (BRDF) effects.
It is worth noting that working with EnMAP satellite data poses significant challenges, particularly due to the lack of geo-referencing, which likely required considerable effort and time during data processing for the authors.
This manuscript presents valuable findings at meter-scale resolution by leveraging the XBAER algorithm and EnMAP data. The work provides important scientific insights that contribute to the global monitoring of high-resolution AOD and biomass-burning plume detection.
The manuscript is well-written, clearly structured, and aligns with the aims and scope of the Atmospheric Measurement Techniques (AMT) journal. I recommend its publication, subject to minor revisions and a few general suggestions.
I will start with general suggestions:
1. The title is “First results of the XBAER…..”, my question if this is the first result then what is/will be the second result, third result so on….?
In general the title of the manuscript should reflect the content as well as should be catchy for the scientific audience to read it and so some applications as well. So, my suggestion would be something like (Application of XBAER aerosol optical depth retrieval algorithm to EnMAP satellite observations).
I feel like this title would reflect the content as well as it will develop the interest among the scientific community to read it. However the are authors free to think about this.
Minor comments:
1. Line 16: “Future modifications to XBAER that would allow it to produce more accurate retrievals at HSI’s spatial resolution are discussed”. This sentence is not needed in the abstract, remove it.
2. In introduction please cite some literatures, in line 19- 21.
3. Line 51: “This paper investigates the possibility of adapting the XBAER algorithm for use with HSI data”, Mention that the combinations of XBAER and EnMAP data would be valuable to detect fine resolution small scale air pollution and bio-mass burning plumes over different regions.
4. At methods and Data section, at line 85-89, please mention advantage of using EnMAP satellite observations over all other sensors mentioned above, and its further alignment with the aim/objective of the manuscript.
5. The results sections are very well written and easy to follow.
6. In figure 1, these stations are used for global validation, so is it possible to use present the AOD retrieval from the combination of XBAER and EnMAP for the global region? If yes, then present the figure and discuss it. If no, then mention it why?
7. As Arctic is warming much faster than global average and the new method of retrieval of AOD is very much needed. This would be very nice to create a separate section after conclusion, may be discussion section, and mention how this entire framework of retrieval would be helpful for different regions undergoing rapid climate change particularly Arctic region, and the future application aspects of this study.
8. Is it possible to improve the plot quality, particularly the spatial plots?
I hope these suggestions would be helpful for the further improvement of this study and would be easier for the potential readers to follow. At the end, I would like to mention that this manuscript very well written and well structured and I enjoyed reading it.
Citation: https://doi.org/10.5194/egusphere-2025-1282-RC1 -
RC2: 'Comment on egusphere-2025-1282', Anonymous Referee #2, 26 Aug 2025
reply
In this paper presents an extensive analysis of the eXtensible Bremen AErosol Retrieval (XBAER) algorithm for aerosol optical depths (AOD) retrievals at very low spatial resolution using passive satellite data. The paper is based on look-up tables with is a well-known technique in the retrieval of AODs from space radiometry. Authors present a very interesting comparison of reflected radiance at the top of the atmosphere for OLCI and the hyper-spectral imager on board the Environmental Mapping and Analysis Program. Results of this comparison are very well discussed and presented, being easy to follow the main points remarked by the authors. The comparison is extended to retrieved AODs product with XBAER algorithm, and authors have a great discussion about the agreements between both products and the causes of disagreement when it happens. The XBAER algorithm is later applied to hyper-spectral imager with 30 m resolution, and coherent retrievals are observed for the study cases presented.
Overall, the paper looks very good and the topic is very interesting, and therefore I recommend the publication in AMT. Finally, the 30 m resolution products are validated versus reference ground-based AERONET measurements. This might be the weakest part of the manuscript as it was not clear for me to see the difficulties remarked for evaluating the 30 m resolution retrievals. I highly recommend re-making this part making clearer how the matchups are made. Another point I want to remark on is that the title is confusing as not everybody is familiar with the term ‘EnMAP’, and thus I highly recommend finding a more suitable title. Apart of that, I have some minor comments that might help authors to improve their manuscript.
Introduction: My general comment is that authors do not discuss the advantages of an algorithm based on look-up tables versus that retrieve aerosol and surface properties using algorithms based on least-squares minimization methods. That should be acknowledged in the introduction.
Lines 19-21: I think that a more clear and concise definition of AOD is needed. References are also required.
Lines 22-25: AERONET is based on sun-photometry measurements from the ground and is usually the reference versus space measurements. This should be highlighted. Also, I recommend adding in the discussions other airborne field campaigns.
Methodology: The filtering of cloud-affected data is something that I could not understand after reading the manuscript. Can the authors give a better explanation?
Line 72: What are the different aerosol types and their properties used in the look-tables computations?
Line 99: I think there are typos. I miss the verb in ‘pixel fully within’. Also, typo in 1,600 30 m HIS
Section 3 Comparisons of OLCI and HIS RTOA: If I understand well, one of the main findings is the influence of surfaces blocking to explain the differences. Why not is mentioned in the abstract?
Line 225: What is SAVI? It is not defined.
Line 237 – 239: I think that references are needed because this is known from other studies for other sensors and algorithms.
Figures 8 and Figure 9: They need improvement. I recommend adding the coordinates to the figure. Also, I guess that color-bars refer to AOD but it is not indicated. Wavelength of AODs retrieval is not indicated.
Line 269: How do you explain negative values in AODs retrievals with XBAER ?
Lines 282-283: It is not mentioned which variable presents the good retrieval.
Citation: https://doi.org/10.5194/egusphere-2025-1282-RC2 -
RC3: 'Comment on egusphere-2025-1282', Anonymous Referee #3, 04 Sep 2025
reply
The manuscript presents a timely and relevant comparison between data from the EnMAP HSI and Sentinel-3 OLCI sensors, with a focus on extending the XBAER aerosol retrieval algorithm to leverage HSI’s higher spatial resolution. The study is well-motivated, particularly in light of the increasing availability of high-resolution hyperspectral satellite data for atmospheric applications.
The comparison of top-of-atmosphere reflectances shows strong agreement (R > 0.9), indicating good consistency between sensors, though the identified deviations near the O₂ A-band and due to BRDF effects are important and well explained. The retrieval comparisons for surface reflectance (R = 0.953) and AOD (R = 0.809) are encouraging, suggesting that HSI data are promising for future AOD retrievals using XBAER. However, the reduced performance in AOD retrievals using unmodified XBAER on full-resolution HSI data (R = 0.631 vs. AERONET) highlights the need for targeted algorithm adaptation.
Minor revisions are recommended. To improve the manuscript's clarity and impact, please address the following critical points:
- A more detailed explanation of the causes and implications of the differences in AOD retrieval performance between surface types would be valuable.
- The discussion of future algorithm modifications would benefit from a clearer outline of specific technical challenges and proposed solutions.
- Including more scenes or seasons in the AERONET comparison could help generalize the findings.
Overall, this is a valuable contribution to the field of aerosol remote sensing and highlights the potential of EnMAP HSI for atmospheric applications when appropriate retrieval adaptations are made.
Line by line comments:
- At line 26, cite some articles to support this statement “Several decades of AOD measurements from satellites are now available that offer increased spatial coverage over both surface-based and airborne measurements.”
- For the method section is it possible to show a graphical flow chart?
- At line 199, please mention here as well why the authors used XBEAR algorithm, so that the reader gets the flow while reading the manuscript.
- Please shorten the conclusion and introduction part a bit for easy reading of the manuscript for the potential readers.
- Is it possible to present the spatial global view of the retrieval? If not then please discuss why, as this is bit important to get the overview that the combination of XBEAR and EnMAP is effective to capture the AOD distribution globally regardless of the background pollution scenarios.
Citation: https://doi.org/10.5194/egusphere-2025-1282-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 339 | 37 | 15 | 391 | 16 | 28 |
- HTML: 339
- PDF: 37
- XML: 15
- Total: 391
- BibTeX: 16
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1