the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Mar 2023
Aeolus wind lidar observations of the 2019/2020 Quasi-Biennial Oscillation disruption with comparison to radiosondes and reanalysis
Abstract. The quasi-biennial oscillation (QBO) was unexpectedly disrupted for only the second time in the historical record during the 2019/20 boreal winter. As the dominant mode of atmospheric variability in the tropical stratosphere, and a significant source of seasonal predictability globally, understanding the drivers behind this unusual behaviour is very important. Here, novel data from Aeolus, the first Doppler wind lidar in space, is used to observe the 2019/20 QBO disruption. Aeolus is the first satellite able to observe winds at high resolution on a global scale, and is therefore a uniquely capable platform for studying the evolution of the disruption and the broader circulation changes triggered by it. This study therefore contains the first direct wind observations of the QBO from space, and exploits measurements from a special Aeolus scanning mode, implemented to observe this disruption as it happened. Aeolus observes easterly winds of up to 20 ms−1 in the core of the disruption jet during July 2020. By co-locating with radiosonde measurements from Singapore and ERA5 reanalysis, like-for-like comparisons of the observed wind structures in the tropical stratosphere are produced, showing equatorial Kelvin wave activity and key parts of the Walker Circulation during the disruption period. The onset of the disruption easterly jet occurs 5 days earlier in Aeolus observations compared with the reanalysis. This analysis highlights how Aeolus and future Doppler wind lidar satellites can deepen our understanding of the QBO, its disruptions, and the tropical upper-troposphere lower-stratosphere region more generally.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(44257 KB)
-
Supplement
(12221 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(44257 KB) - Metadata XML
-
Supplement
(12221 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-285', Anonymous Referee #1, 11 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Timothy Banyard, 22 Aug 2023
-
RC2: 'Comment on egusphere-2023-285', Anonymous Referee #2, 18 Apr 2023
In their manuscript “Aeolus wind lidar observations of the 2019/2020 Quasi-Biennial Oscillation disruption with comparison to radiosondes and reanalysis”, the authors observe the evolution of the second ever Quasi-Biennial Oscillation (QBO) disruption in novel Aeolus satellite measurements. The observations are validated with radiosonde and reanalysis data. In addition, the reanalysis and satellite data are analyzed with respect to Kelvin wave occurrence before the disruption.
This study is an excellent show case for the use of novel Aeolus wind measurements for the investigation of dynamic phenomena in the upper troposphere / lower stratosphere (UTLS) region. It clearly demonstrates the capabilities and restrictions of this unprecedented dataset for atmospheric research. With this the article is in general of high scientific significance. Nevertheless, I see a major need for improvement with respect to the scope of the journal and the scientific questions answered by the manuscript.
In its current state the paper is more a validation of the new Aeolus dataset and with this more in the scope of Atmospheric Measurement Techniques. However, by applying some changes and extending the discussion the paper could help to address important scientific aspects of QBO research. This would make it very suitable to Atmospheric Chemistry and Physics.
In detail, the paper would strongly benefit from a more detailed discussion on what can be learnt from the differences between ERA5 and Aeolus for the generation of the disruption. Which processes are different (e.g. Kelvin wave activity) and what does this tell us about our current understanding of the physical processes behind?
A detailed list specific comments is given below.
Specific comments:
L27: Any reference to support this sentence? Maybe Smith et al. (2019)?
L29: Also, here a reference would provide additional information to the reader, e.g. ESA (2020a)
L36ff: This sentence is really long, contains a lot of information and is hard to read. Maybe better split into two or more sentences.
L39: What is your study adding to the current knowledge / understanding of the QBO? What are the research questions you are trying to answer? For the reader it usually helps to briefly address the “why exactly this research” in the introduction to easier follow the manuscript.
L46f: Global and local are antonyms, but how does reanalysis fit into the picture? Is a reanalysis perspective really a measurement perspective? I would suggest to reformulate this sentence.
Section 2.1: Maybe it is worth mentioning that the Singapore Radiosonde Station is commonly used as the gold standard when it comes to QBO analysis as it provides the longest available data record in the tropical stratosphere. Okay, this is somewhere later in the paper. Maybe you could move it forward to the dataset description.
L66ff: Is Banyard et al. (2021) really an appropriate reference here? Maybe remove this part of the sentence. The references before already support this sentence sufficiently.
L71: I would suggest to better reference ESA (2020b) here.
L77f: This is in general correct, but the sentence is confusing here. 35° off-nadir (as mentioned in the sentence before) per se gives mainly the vertical wind component. Only because the vertical wind component is much smaller than the horizontal component, the zonal wind can be derived, which you actually mention, but only one sentence afterwards. So, the order of the sentences here confused me. Maybe just remove this sentence on the advantage for QBO observation?
L108: ERA5 is available at a temporal resolution of 1 hour: https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5
Figure 1: How did you fill the data gaps in this plot? E.g. between the weekly QBO settings or during instrument down times (e.g. spring 2021).
L116f: This would probably be better visible if the Aeolus dataset would be extended until launch in 2018.
L161: Why only for reanalysis?
L185: …, as *mentioned / described* on Kawatani et al. (2016), …
L185ff: I had to read these sentences a couple of times, before understanding their meaning. Maybe rephrase?
L193: Why 3D?
L204f: This is a courageous statement. Weiler et al. (2021) and Abdalla et al. (2020) only look into this on global average. According to ESA (2021), there could be higher regional biases. In addition, there are small bias differences between ascending and descending orbits in certain months (also ESA, 2021).
L206: I fully agree to use negative HLOS in figure 3, but please also state this in the figure or caption.
L211ff: You nicely describe the evolution of easterly winds in the profiles. The weakening of westerly winds is also nicely visible in profiles b-e. Maybe you could also guide the reader through this point first, before coming to the evolution of the easterlies. Otherwise, why do you show the profiles b-e?
L219: Aeolus measurements are performed separately for each range bin and, unlike the derivation of temperature and aerosol backscatter and extinction, the derivation of wind from lidar measurements does not rely on an iterative profile reconstruction. Thus, cloud contamination in higher atmospheric bins cannot originate from clouds in lower atmospheric bins. To me it seems more that your limit of 250km is quite a large range (in validation studies often 100km is used as colocation criterion) and the atmosphere varies within this distance. There are many points in your profile comparisons where two or more Aeolus winds in the same range gate at the same time are quite far from each other (not only in figure 3g & 3h). This is probably due to the variability of the wind in latitudinal direction. Maybe the altitude offset in figure 3f and 3i are also due to a not perfect match in location.
Figure 4: Which colocation method is used for the comparison of these data points? Aeolus vs radiosonde, probably the same 200km, but what did you do for the comparison to ERA5? Interpolate the model onto the Aeolus and radiosonde location? This might explain the better agreement between ERA5 and Aeolus vs radiosonde and Aeolus.
L251ff: Maybe a good idea for a next study. ECMWF forecast / analysis data with and without Aeolus assimilation is available at ECMWF.
L255: What exactly do you mean with like-for-like comparison? Are you interpolating ERA5 data to Aeolus measurement locations?
Figure 5: Maybe you could show the ERA5 contour lines in the whole plot.
L257: You not only see this difference in onset time, you also clearly see stronger winds and wind gradients in the Aeolus data in the troposphere before the disruption (e.g. -5 m/s line is at a higher altitude in Aeolus data). These stronger wind gradients might have an impact on the Kelvin wave propagation (you discuss afterwards), so I think they should be mentioned here.
L270: … highlight only symmetric wind structures … (antisymmetric waves with respect to the equator are removed due to your averaging from -5° to +5° latitude; for a detailed analysis of equatorial waves in the Aeolus dataset, you could have a look at Ern et al. 2023)
L291 – 296: This paragraph mixes different things and draws conclusion which I either do not understand correctly or a very far-fetched. I would suggest to remove the whole paragraph.
animation S2: Title of plot and caption are not in line with each other. It should be +-5° latitude as everywhere in the manuscript.
Animation S2 and Figure 7: Why don’t you apply the same temporal filtering as in the Hovmöller plots? You want to show the Kelvin waves, but these are barely visible due to the dominating feature of the Walker circulation. By applying a similar filter as in Figure 6, this strong dipole should vanish and the Kelvin waves should become clearly visible.
L328ff: You could perhaps reformulate the sentence to stress the importance of a future wind lidar measuring up to at least 30km or higher for QBO research.
L336: … this change in random error … (the high random error itself is a problem especially for short analysis periods and perturbation analysis, for these a bias would be less of a problem)
L348: Why only reanalysis model?
L349: Why is your analysis not spanning the whole Aeolus measurement period, so why is the data before summer 2019 missing?
Discussion: This is not a general discussion of the results of the study. What you describe here are drawbacks of the Aeolus mission. Thus, I would suggest to either rename the section or revise its content.
L379f: Why is this in agreement with the data validation?
L385: Maybe better: … have been discussed.
L389: Maybe good to stress here the importance of measurements up to at least 30km.
Conclusions: You nicely describe the data, but what have we learned from a scientific point of view (except that Aeolus is a well-suited dataset for observing wind related phenomena in the UTLS)? Why did it come to this QBO disruption? Where is the difference between ERA5 and Aeolus and what can we learn from this difference for our understanding of the underlying physical processes? These are questions I would expect to be answered in an ACP manuscript.
References:
Ern, M., Diallo, M. A., Khordakova, D., Krisch, I., Preusse, P., Reitebuch, O., Ungermann, J., and Riese, M.: The QBO and global-scale tropical waves in Aeolus wind observations, radiosonde data, and reanalyses, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-408, 2023.
ESA (2020a) https://earth.esa.int/eogateway/news/aeolus-activates-new-range-bin-setting
ESA (2020b) https://earth.esa.int/eogateway/news/a-guide-to-aeolus-range-bin-settings
ESA (2021) https://earth.esa.int/eogateway/documents/20142/0/Aeolus-Summary-Reprocessing-2-DISC.pdf
Smith, A.K., Holt, L.A., Garcia, R.R., Anstey, J.A., Serva, F., Butchart, N., et al. (2022) The equatorial stratospheric semiannual oscillation and time-mean winds in QBOi models. Q J R Meteorol Soc, 148( 744), 1593– 1609. Available from: https://doi.org/10.1002/qj.3690
Citation: https://doi.org/10.5194/egusphere-2023-285-RC2 -
AC2: 'Reply on RC2', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Timothy Banyard, 22 Aug 2023
-
RC3: 'Comment on egusphere-2023-285', Anonymous Referee #3, 20 Apr 2023
The manuscript by Banyard et al. presents wind measurements of the 2019-2020 QBO disruption from the first spaceborne Doppler wind lidar ADM-Aeolus. The general topic is relevant for publication in ACP and the paper is interesting, demonstrating that the QBO is generally well captured in Aeolus data. However, I find the manuscript very focused. It would benefit from including a broader analysis and discussion of the new information brought by Aeolus on tropical lower stratospheric dynamics compared to a state-of-the-art reanalysis. Some differences between the ERA5 and Aeolus are already mentioned but not much commented.
For these reasons, major revisions (additions) are required before the paper can be considered for publication. My recommandation to the authors would be to make more quantitative statements and expand the comparison between Aeolus and ERA5 to other modes of variability than the QBO (e.g., equatorial waves) which are only alluded to in the present manuscript. The authors should also consider the recent preprint by Ern et al. (ACPD) on a related topic.
Main comments :
1) Lack of significance test : in a few instances the author claim that a bias/difference is significant, but do not include a statistical significance test. For instance the near-tropopause wind bias at Singapore (l238-240) is not clear to me, since it is also present in the ERA5-Singapore radiosonde comparison.
2) The authors touch the topic of equatorial wave representation in ERA5, but they could make more quantitative statements. They might also wish to change the colormap used for Figure 6. Given that one of the advantages of Aeolus is global sampling, the authors could without much effort quantify the differences as a function of longitude, beyond the location of Singapore. This is of interest, in particular since earlier studies (e.g., Baker et al., 2013 ; Podglajen et al., 2014) found that reanalysis uncertainties and wind errors in the tropical UTLS were larger in regions without assimilated radiosonde observations, resulting in a pronounced zonal structure. Note that the need for wind observations to better constrain the flow in tropical regions is what motivated Aeolus in the first place.
3) Regarding the last sentence of the abstract : ‘‘This analysis highlights how Aeolus and future Doppler wind lidar satellites can deepen our understanding of the QBO, its disruptions, and the tropical upper-troposphere lower-stratosphere region more generally.’’ , it is not shown in the paper that Aeolus can help clarify the mechanisms of the disruption, at least directly. I imagine spaceborne lidars would help deepen the understanding of the QBO by putting observational constraints on equatorial waves and tropical gravity waves, but this is not really the focus of the paper.
Line 1 : The abstract could mention some key number (bias, etc.)
Line 102 : please describe the method briefly here Line 238-240 : Is this a significant bias ? It does seem smaller than the difference between Aeolus and the radiosonde (panel a), which by the way, also maximizes near the tropopause.
Line 264-265 it is suggested that the sampling by aeolus does not induce much bias. You could easily prove this with a figure comparing ERA5 with aeolus sampling and the actual zonal mean
Fig 4 : the pressure scale is wrong in this figure. What do the dots correspond to? Some of the numbers in the legend would find their place in the main text.
Fig 5 : Could you rather plot the mean wind in contour and the difference in colors ? This would be more quantitative. There seems to be a delay, does it hold for the later period or is this just a feature of the disruption?
Fig. 6 : Is there mean a zonal structure in the error (see main comment 2) ? Also, I do not understand the need for time filtering. Eastward propagation appears clearly in your unfiltered plot.
lines 271-277 : The phrasing is a bit odd, suggesting that this long period was not an optimal choice for Kelvin waves. You cite a few papers which show that the typical stratospheric Kelvin waves commonly seen in Hovmoller diagrams indeed have planetary wavenumber 1-2, periods of 10-20 days or more, as first described by Wallace and Kousky and reported in many papers and textbooks. Convectively-coupled Kelvin waves in the OLR are higher frequency but this is not what dominates at tropopause altitude where free-travelling waves are prominentl. I would shorten this discussion.
References :
Baker, W. E., et al. (2013), Lidar-measured wind profiles: The missing link in the global observing system, Bull. Am. Meteorol. Soc., 94, 543–564, doi:10.1175/BAMS-D-12-00164.1
Podglajen, A., Hertzog, A., Plougonven, R., and Žagar, N. (2014), Assessment of the accuracy of (re)analyses in the equatorial lower stratosphere, J. Geophys. Res. Atmos., 119, 11,166– 11,188, doi:10.1002/2014JD021849
Ern, M., Diallo, M. A., Khordakova, D., Krisch, I., Preusse, P., Reitebuch, O., Ungermann, J., and Riese, M.: The QBO and global-scale tropical waves in Aeolus wind observations, radiosonde data, and reanalyses, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-408, 2023.
Citation: https://doi.org/10.5194/egusphere-2023-285-RC3 -
AC3: 'Reply on RC3', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC3-supplement.pdf
-
AC3: 'Reply on RC3', Timothy Banyard, 22 Aug 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-285', Anonymous Referee #1, 11 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Timothy Banyard, 22 Aug 2023
-
RC2: 'Comment on egusphere-2023-285', Anonymous Referee #2, 18 Apr 2023
In their manuscript “Aeolus wind lidar observations of the 2019/2020 Quasi-Biennial Oscillation disruption with comparison to radiosondes and reanalysis”, the authors observe the evolution of the second ever Quasi-Biennial Oscillation (QBO) disruption in novel Aeolus satellite measurements. The observations are validated with radiosonde and reanalysis data. In addition, the reanalysis and satellite data are analyzed with respect to Kelvin wave occurrence before the disruption.
This study is an excellent show case for the use of novel Aeolus wind measurements for the investigation of dynamic phenomena in the upper troposphere / lower stratosphere (UTLS) region. It clearly demonstrates the capabilities and restrictions of this unprecedented dataset for atmospheric research. With this the article is in general of high scientific significance. Nevertheless, I see a major need for improvement with respect to the scope of the journal and the scientific questions answered by the manuscript.
In its current state the paper is more a validation of the new Aeolus dataset and with this more in the scope of Atmospheric Measurement Techniques. However, by applying some changes and extending the discussion the paper could help to address important scientific aspects of QBO research. This would make it very suitable to Atmospheric Chemistry and Physics.
In detail, the paper would strongly benefit from a more detailed discussion on what can be learnt from the differences between ERA5 and Aeolus for the generation of the disruption. Which processes are different (e.g. Kelvin wave activity) and what does this tell us about our current understanding of the physical processes behind?
A detailed list specific comments is given below.
Specific comments:
L27: Any reference to support this sentence? Maybe Smith et al. (2019)?
L29: Also, here a reference would provide additional information to the reader, e.g. ESA (2020a)
L36ff: This sentence is really long, contains a lot of information and is hard to read. Maybe better split into two or more sentences.
L39: What is your study adding to the current knowledge / understanding of the QBO? What are the research questions you are trying to answer? For the reader it usually helps to briefly address the “why exactly this research” in the introduction to easier follow the manuscript.
L46f: Global and local are antonyms, but how does reanalysis fit into the picture? Is a reanalysis perspective really a measurement perspective? I would suggest to reformulate this sentence.
Section 2.1: Maybe it is worth mentioning that the Singapore Radiosonde Station is commonly used as the gold standard when it comes to QBO analysis as it provides the longest available data record in the tropical stratosphere. Okay, this is somewhere later in the paper. Maybe you could move it forward to the dataset description.
L66ff: Is Banyard et al. (2021) really an appropriate reference here? Maybe remove this part of the sentence. The references before already support this sentence sufficiently.
L71: I would suggest to better reference ESA (2020b) here.
L77f: This is in general correct, but the sentence is confusing here. 35° off-nadir (as mentioned in the sentence before) per se gives mainly the vertical wind component. Only because the vertical wind component is much smaller than the horizontal component, the zonal wind can be derived, which you actually mention, but only one sentence afterwards. So, the order of the sentences here confused me. Maybe just remove this sentence on the advantage for QBO observation?
L108: ERA5 is available at a temporal resolution of 1 hour: https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5
Figure 1: How did you fill the data gaps in this plot? E.g. between the weekly QBO settings or during instrument down times (e.g. spring 2021).
L116f: This would probably be better visible if the Aeolus dataset would be extended until launch in 2018.
L161: Why only for reanalysis?
L185: …, as *mentioned / described* on Kawatani et al. (2016), …
L185ff: I had to read these sentences a couple of times, before understanding their meaning. Maybe rephrase?
L193: Why 3D?
L204f: This is a courageous statement. Weiler et al. (2021) and Abdalla et al. (2020) only look into this on global average. According to ESA (2021), there could be higher regional biases. In addition, there are small bias differences between ascending and descending orbits in certain months (also ESA, 2021).
L206: I fully agree to use negative HLOS in figure 3, but please also state this in the figure or caption.
L211ff: You nicely describe the evolution of easterly winds in the profiles. The weakening of westerly winds is also nicely visible in profiles b-e. Maybe you could also guide the reader through this point first, before coming to the evolution of the easterlies. Otherwise, why do you show the profiles b-e?
L219: Aeolus measurements are performed separately for each range bin and, unlike the derivation of temperature and aerosol backscatter and extinction, the derivation of wind from lidar measurements does not rely on an iterative profile reconstruction. Thus, cloud contamination in higher atmospheric bins cannot originate from clouds in lower atmospheric bins. To me it seems more that your limit of 250km is quite a large range (in validation studies often 100km is used as colocation criterion) and the atmosphere varies within this distance. There are many points in your profile comparisons where two or more Aeolus winds in the same range gate at the same time are quite far from each other (not only in figure 3g & 3h). This is probably due to the variability of the wind in latitudinal direction. Maybe the altitude offset in figure 3f and 3i are also due to a not perfect match in location.
Figure 4: Which colocation method is used for the comparison of these data points? Aeolus vs radiosonde, probably the same 200km, but what did you do for the comparison to ERA5? Interpolate the model onto the Aeolus and radiosonde location? This might explain the better agreement between ERA5 and Aeolus vs radiosonde and Aeolus.
L251ff: Maybe a good idea for a next study. ECMWF forecast / analysis data with and without Aeolus assimilation is available at ECMWF.
L255: What exactly do you mean with like-for-like comparison? Are you interpolating ERA5 data to Aeolus measurement locations?
Figure 5: Maybe you could show the ERA5 contour lines in the whole plot.
L257: You not only see this difference in onset time, you also clearly see stronger winds and wind gradients in the Aeolus data in the troposphere before the disruption (e.g. -5 m/s line is at a higher altitude in Aeolus data). These stronger wind gradients might have an impact on the Kelvin wave propagation (you discuss afterwards), so I think they should be mentioned here.
L270: … highlight only symmetric wind structures … (antisymmetric waves with respect to the equator are removed due to your averaging from -5° to +5° latitude; for a detailed analysis of equatorial waves in the Aeolus dataset, you could have a look at Ern et al. 2023)
L291 – 296: This paragraph mixes different things and draws conclusion which I either do not understand correctly or a very far-fetched. I would suggest to remove the whole paragraph.
animation S2: Title of plot and caption are not in line with each other. It should be +-5° latitude as everywhere in the manuscript.
Animation S2 and Figure 7: Why don’t you apply the same temporal filtering as in the Hovmöller plots? You want to show the Kelvin waves, but these are barely visible due to the dominating feature of the Walker circulation. By applying a similar filter as in Figure 6, this strong dipole should vanish and the Kelvin waves should become clearly visible.
L328ff: You could perhaps reformulate the sentence to stress the importance of a future wind lidar measuring up to at least 30km or higher for QBO research.
L336: … this change in random error … (the high random error itself is a problem especially for short analysis periods and perturbation analysis, for these a bias would be less of a problem)
L348: Why only reanalysis model?
L349: Why is your analysis not spanning the whole Aeolus measurement period, so why is the data before summer 2019 missing?
Discussion: This is not a general discussion of the results of the study. What you describe here are drawbacks of the Aeolus mission. Thus, I would suggest to either rename the section or revise its content.
L379f: Why is this in agreement with the data validation?
L385: Maybe better: … have been discussed.
L389: Maybe good to stress here the importance of measurements up to at least 30km.
Conclusions: You nicely describe the data, but what have we learned from a scientific point of view (except that Aeolus is a well-suited dataset for observing wind related phenomena in the UTLS)? Why did it come to this QBO disruption? Where is the difference between ERA5 and Aeolus and what can we learn from this difference for our understanding of the underlying physical processes? These are questions I would expect to be answered in an ACP manuscript.
References:
Ern, M., Diallo, M. A., Khordakova, D., Krisch, I., Preusse, P., Reitebuch, O., Ungermann, J., and Riese, M.: The QBO and global-scale tropical waves in Aeolus wind observations, radiosonde data, and reanalyses, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-408, 2023.
ESA (2020a) https://earth.esa.int/eogateway/news/aeolus-activates-new-range-bin-setting
ESA (2020b) https://earth.esa.int/eogateway/news/a-guide-to-aeolus-range-bin-settings
ESA (2021) https://earth.esa.int/eogateway/documents/20142/0/Aeolus-Summary-Reprocessing-2-DISC.pdf
Smith, A.K., Holt, L.A., Garcia, R.R., Anstey, J.A., Serva, F., Butchart, N., et al. (2022) The equatorial stratospheric semiannual oscillation and time-mean winds in QBOi models. Q J R Meteorol Soc, 148( 744), 1593– 1609. Available from: https://doi.org/10.1002/qj.3690
Citation: https://doi.org/10.5194/egusphere-2023-285-RC2 -
AC2: 'Reply on RC2', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Timothy Banyard, 22 Aug 2023
-
RC3: 'Comment on egusphere-2023-285', Anonymous Referee #3, 20 Apr 2023
The manuscript by Banyard et al. presents wind measurements of the 2019-2020 QBO disruption from the first spaceborne Doppler wind lidar ADM-Aeolus. The general topic is relevant for publication in ACP and the paper is interesting, demonstrating that the QBO is generally well captured in Aeolus data. However, I find the manuscript very focused. It would benefit from including a broader analysis and discussion of the new information brought by Aeolus on tropical lower stratospheric dynamics compared to a state-of-the-art reanalysis. Some differences between the ERA5 and Aeolus are already mentioned but not much commented.
For these reasons, major revisions (additions) are required before the paper can be considered for publication. My recommandation to the authors would be to make more quantitative statements and expand the comparison between Aeolus and ERA5 to other modes of variability than the QBO (e.g., equatorial waves) which are only alluded to in the present manuscript. The authors should also consider the recent preprint by Ern et al. (ACPD) on a related topic.
Main comments :
1) Lack of significance test : in a few instances the author claim that a bias/difference is significant, but do not include a statistical significance test. For instance the near-tropopause wind bias at Singapore (l238-240) is not clear to me, since it is also present in the ERA5-Singapore radiosonde comparison.
2) The authors touch the topic of equatorial wave representation in ERA5, but they could make more quantitative statements. They might also wish to change the colormap used for Figure 6. Given that one of the advantages of Aeolus is global sampling, the authors could without much effort quantify the differences as a function of longitude, beyond the location of Singapore. This is of interest, in particular since earlier studies (e.g., Baker et al., 2013 ; Podglajen et al., 2014) found that reanalysis uncertainties and wind errors in the tropical UTLS were larger in regions without assimilated radiosonde observations, resulting in a pronounced zonal structure. Note that the need for wind observations to better constrain the flow in tropical regions is what motivated Aeolus in the first place.
3) Regarding the last sentence of the abstract : ‘‘This analysis highlights how Aeolus and future Doppler wind lidar satellites can deepen our understanding of the QBO, its disruptions, and the tropical upper-troposphere lower-stratosphere region more generally.’’ , it is not shown in the paper that Aeolus can help clarify the mechanisms of the disruption, at least directly. I imagine spaceborne lidars would help deepen the understanding of the QBO by putting observational constraints on equatorial waves and tropical gravity waves, but this is not really the focus of the paper.
Line 1 : The abstract could mention some key number (bias, etc.)
Line 102 : please describe the method briefly here Line 238-240 : Is this a significant bias ? It does seem smaller than the difference between Aeolus and the radiosonde (panel a), which by the way, also maximizes near the tropopause.
Line 264-265 it is suggested that the sampling by aeolus does not induce much bias. You could easily prove this with a figure comparing ERA5 with aeolus sampling and the actual zonal mean
Fig 4 : the pressure scale is wrong in this figure. What do the dots correspond to? Some of the numbers in the legend would find their place in the main text.
Fig 5 : Could you rather plot the mean wind in contour and the difference in colors ? This would be more quantitative. There seems to be a delay, does it hold for the later period or is this just a feature of the disruption?
Fig. 6 : Is there mean a zonal structure in the error (see main comment 2) ? Also, I do not understand the need for time filtering. Eastward propagation appears clearly in your unfiltered plot.
lines 271-277 : The phrasing is a bit odd, suggesting that this long period was not an optimal choice for Kelvin waves. You cite a few papers which show that the typical stratospheric Kelvin waves commonly seen in Hovmoller diagrams indeed have planetary wavenumber 1-2, periods of 10-20 days or more, as first described by Wallace and Kousky and reported in many papers and textbooks. Convectively-coupled Kelvin waves in the OLR are higher frequency but this is not what dominates at tropopause altitude where free-travelling waves are prominentl. I would shorten this discussion.
References :
Baker, W. E., et al. (2013), Lidar-measured wind profiles: The missing link in the global observing system, Bull. Am. Meteorol. Soc., 94, 543–564, doi:10.1175/BAMS-D-12-00164.1
Podglajen, A., Hertzog, A., Plougonven, R., and Žagar, N. (2014), Assessment of the accuracy of (re)analyses in the equatorial lower stratosphere, J. Geophys. Res. Atmos., 119, 11,166– 11,188, doi:10.1002/2014JD021849
Ern, M., Diallo, M. A., Khordakova, D., Krisch, I., Preusse, P., Reitebuch, O., Ungermann, J., and Riese, M.: The QBO and global-scale tropical waves in Aeolus wind observations, radiosonde data, and reanalyses, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-408, 2023.
Citation: https://doi.org/10.5194/egusphere-2023-285-RC3 -
AC3: 'Reply on RC3', Timothy Banyard, 22 Aug 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-285/egusphere-2023-285-AC3-supplement.pdf
-
AC3: 'Reply on RC3', Timothy Banyard, 22 Aug 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 542 | 228 | 35 | 805 | 68 | 20 | 18 |
- HTML: 542
- PDF: 228
- XML: 35
- Total: 805
- Supplement: 68
- BibTeX: 20
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 223 | 27 |
| United Kingdom | 2 | 154 | 19 |
| Germany | 3 | 110 | 13 |
| France | 4 | 57 | 7 |
| China | 5 | 45 | 5 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 223
Corwin J. Wright
Scott M. Osprey
Neil P. Hindley
Gemma Halloran
Lawrence Coy
Paul A. Newman
Neal Butchart
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(44257 KB) - Metadata XML
-
Supplement
(12221 KB) - BibTeX
- EndNote
- Final revised paper