A case study of a ducted gravity wave event over northern Germany using simultaneous airglow imaging and wind-field observations

Sarkhel, Sumanta; Stober, Gunter; Chau, Jorge L.; Smith, Steven M.; Jacobi, Christoph; Mondal, Subarna; Mlynczak, Martin G.; Russell III, James M.

doi:https://doi.org/10.5194/angeo-40-179-2022

Articles | Volume 40, issue 2

https://doi.org/10.5194/angeo-40-179-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/angeo-40-179-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 40, issue 2

Regular paper

|

22 Mar 2022

Regular paper |

| 22 Mar 2022

A case study of a ducted gravity wave event over northern Germany using simultaneous airglow imaging and wind-field observations

Sumanta Sarkhel, Gunter Stober, Jorge L. Chau, Steven M. Smith, Christoph Jacobi, Subarna Mondal, Martin G. Mlynczak, and James M. Russell III

Download

Final revised paper (published on 22 Mar 2022)
Preprint (discussion started on 25 Aug 2021)

Interactive discussion

Status: closed

RC1:
'Comment on angeo-2021-48', Anonymous Referee #1, 07 Oct 2021

The work combines Airglow imager and Meteor radar observations from Germany and uses SABER measurements to bring out/discuss a special gravity wave event that leaves its footprint in oxygen green line, O2 emissions but not in OH emission. The authors propose the formation of a leaky duct to explain the observations. While this is a realistic possibility, other factors need to be ruled out to strengthen the argument. The observation of this wave activity in Na emission is also enigmatic and require more attention. The observations reported merit publication but the authors need to critically address a few issues described below.

On dataset and techniques: Does this imager consist of 512 × 512 pixels ANDOR back-illuminated CCD camera or 1k × 1k camera as mentioned by Vargas et al. (2021)? For completeness, provide details on the bandwidth of the interference filters. Is there a possibility of contamination of OH broadband emission from other mesospheric lines? How do the authors rule that this contamination is not responsible for the lack of wave signatures in the OH band? What are the integration times used for these observations?

Figures 2-5: Why does the central dark spot appear in all the images? Detector issue? How does the 2D FFT filtering remove this feature?

Science issues: Figures 2-5: The wind fields are widely different even within the “ducted” layer. Why? This is contrary to the argument (used later for Figure 7a) that the profile does not show any drastic change within the ducted layer.

Figure 6: Why is the downward phase progression not seen in the Na airglow intensity? The authors have mentioned 91 km as the emission height of this emission and 85-91 km as the altitude extent of the thermal duct. The intensity variation at this wavelength also seems to be different at larger horizontal distance.

I feel that the sub-units of Figure 7 are wrongly described in the text. For example, 7c and 7e (instead of 7d and 7e) show variations in m^2. Same for Figures 7b and 7c.

Is m² negative at 85 km? How important is the negative m²? If it is less but positive (at 85 km), how efficient is the ducting? This is an important issue as the authors propose a leaky duct here. If there is leakage each time the wave is reflected from the edges of the duct, how does the Na intensity variation show similar intensity variation but different phase progression (at least at shorter horizontal distances) in Figure 6? Will the original and reflected waves inside the duct not interfere? How is the spatial coherence maintained? This is also an important issue given the different wind fields at the airglow emission altitudes. The authors need to discuss these issues.

Citation: https://doi.org/10.5194/angeo-2021-48-RC1
- AC1: 'Reply on RC1', Sumanta Sarkhel, 27 Oct 2021
  
  Reply to the Reviewer 1
  
  Manuscript #: angeo-2021-48
  Title: A case study of a ducted gravity wave event over northern Germany using simultaneous airglow imaging and wind-field observations
  
  The work combines Airglow imager and Meteor radar observations from Germany and uses SABER measurements to bring out/discuss a special gravity wave event that leaves its footprint in oxygen green line, O2 emissions but not in OH emission. The authors propose the formation of a leaky duct to explain the observations. While this is a realistic possibility, other factors need to be ruled out to strengthen the argument. The observation of this wave activity in Na emission is also enigmatic and require more attention. The observations reported merit publication but the authors need to critically address a few issues described below.
  
  Reply: We thank the reviewer for the appreciation and critical comments that will improve the content of the manuscript significantly. The responses to the reviewer’s comments are provided below in the same sequence.
  
  On dataset and techniques: Does this imager consist of 512 × 512 pixels ANDOR back-illuminated CCD camera or 1k × 1k camera as mentioned by Vargas et al. (2021)? For completeness, provide details on the bandwidth of the interference filters. Is there a possibility of contamination of OH broadband emission from other mesospheric lines? How do the authors rule that this contamination is not responsible for the lack of wave signatures in the OH band? What are the integration times used for these observations?
  
  Reply: Yes, it is the same 1k × 1k CCD camera mentioned in Vargas et al. (2021). On the night of 25 April 2017, the images were binned 2 × 2 (in real time) in order to achieve better signal-to-noise ratio. The bandwidth of the 589.3 nm, 866.0 nm, and 557.7 nm filters is 2.0 nm whereas, the OH filter is a broadband (695–1050 nm). Hence, there is no contamination of OH broadband emission from other mesospheric lines (557.7 and 589.3 nm).
  The nearest lines of OH (X²П) are at P1(6) and R(7,3) (854.9 and 877.1 nm respectively) that are quite far-off from O₂ emission line (866.0 nm) (Chamberlain, 1961). In addition, the wave signatures in the OH (when detected) and O₂ images differed in both morphology and phase. Hence, we are confident that no significant broadband OH occurred within the 866.0 nm filter bandwidth. The integration times for the 589.3 nm, 866.0 nm, and 557.7 nm images were 120s and 15s for the OH filter. We will include this information in the revised manuscript.
  
  Figures 2-5: Why does the central dark spot appear in all the images? Detector issue? How does the 2D FFT filtering remove this feature?
  
  Reply: The central dark spot appearing in every all-sky airglow image is due to the van Rhijn effect; a combination of the finite width of the emission layer and the variation of the line of sight through the layer with increasing zenith distance. There was no issue with the CCD detector. We will mention this in the revised manuscript.
              We have used 2D FFT filtering techniques adopted from Mondal et al. (2019). In this filtering technique, the Gaussian filter window behaves as a low-pass filter and therefore chosen to remove high frequency noises. In the spatial domain, the filtered image is the convolution of an unfiltered image and the Gaussian filter. However, in the power spectral domain, we can simply multiply the 2D FFT power spectra of the unfiltered image with the 2D power spectra of Gaussian filter window by positioning it at the peak-power position of the 2D FFT spectra of the image and then perform the inverse 2D FFT. The airglow image contains range of spatial scale sizes instead of monochromatic of gravity waves. Hence choosing the proper window size of Gaussian filter, we can select the desired scale sizes of gravity waves and the filtered images get enhanced with those features. In this process, the central dark spot is automatically removed.
  
  Science issues: Figures 2-5: The wind fields are widely different even within the “ducted” layer. Why? This is contrary to the argument (used later for Figure 7a) that the profile does not show any drastic change within the ducted layer.
  
  Reply: We wanted to convey that the thermal gradient dominates in the m² profile whereas the wind shear doesn’t play any significant role. We will remove the sentence “The wind profile does not show any drastic change within the duct layer (Figure 7a).” in order to avoid any confusion.
  
  Figure 6: Why is the downward phase progression not seen in the Na airglow intensity? The authors have mentioned 91 km as the emission height of this emission and 85-91 km as the altitude extent of the thermal duct. The intensity variation at this wavelength also seems to be different at larger horizontal distance.
  
  Reply: The upper half of the Na airglow emission is situated above the ducted layer which is the free upward propagating region of waves. However, the bottom half of layer lies in the ducted region (as sketched in Figure 8) wherein the wave is confined to propagate horizontally and not in the upward direction. Therefore, the difference in the vertical phase progression in the Na airglow intensity is caused by the combined effect of wave structure within and beyond the ducted layer as the airglow imager captured vertically integrated Na airglow emission intensity. Whereas, the O(¹S) and O₂ airglow emission layers are beyond the ducted layer and hence clearly shows the downward phase progression.
  Yes, we agree with the reviewer that the intensity variation at this wavelength also seems to be different at larger horizontal distances. However, we are unable find out the cause based on the present dataset. We will mention this in the revised manuscript.
  
  I feel that the sub-units of Figure 7 are wrongly described in the text. For example, 7c and 7e (instead of 7d and 7e) show variations in m^2. Same for Figures 7b and 7c.
  
  Reply: We thank the reviewer for pointing out these typos. We will correct it in the revised manuscript.
  
  Is m² negative at 85 km? How important is the negative m²? If it is less but positive (at 85 km), how efficient is the ducting? This is an important issue as the authors propose a leaky duct here. If there is leakage each time the wave is reflected from the edges of the duct, how does the Na intensity variation show similar intensity variation but different phase progression (at least at shorter horizontal distances) in Figure 6? Will the original and reflected waves inside the duct not interfere? How is the spatial coherence maintained? This is also an important issue given the different wind fields at the airglow emission altitudes. The authors need to discuss these issues.
  
  Reply: The m² profile at SABER 2 reveals that a duct region exists in the altitude range of 85-91 km where positive m² is vertically sandwiched between a negative m² value (at 91 km) and a less positive value (at 85 km). The m² profile indicates that the existence of a weak evanescent region at 85 km (as m² is not negative). Therefore, the duct layer observed in the present case is a “leaky duct” from the bottom side. Hence, some of the energy of the wave could penetrate through the bottom of the duct layer and other part of the energy will be reflected back downward. The penetrated wave got trapped inside the duct layer and could only travel horizontally.
              As we have already discussed above that the upper half of the Na airglow emission is situated above the ducted layer which is the free upward propagating region waves. However, the bottom half of layer is lies in the ducted region wherein the wave is trapped and don’t any propagate in the upward direction. Therefore, the difference in the vertical phase progression in the Na airglow intensity is caused by the combined effect of wave structure within and beyond the ducted layer as the airglow imager captured vertically integrated Na airglow emission intensity.
              The penetrated wave only entered the leaky duct layer from the bottom side. The reflected wave from the edge of the bottom of the duct layer propagated downward below and didn’t enter the duct layer. Therefore, there won’t be any interference between the original and the reflected waves inside the duct layer.
  The winds are analyzed as an average along the propagation path of the wave fronts. The spatial wind retrievals imply a large spatial coherence of about 60 km (the correlation length is set to include the next grid cell) and a temporal correlation of about 30 minutes. Thus, a certain degree of coherence is an essential part of the wind analysis for this case. Furthermore, due to the thermal wind balance, changes in the vertical temperature profile led to corresponding changes of the winds at the different altitudes. The spatial variability of the wind field is also indicated by the bending of the wave fronts found in Figure 2 and 3, whereas the wind fields remain stable at the altitude of the OH emission line presented in Figure 5. However, due to the missing spatial information of temperature observations, it is not possible to understand how this spatial variability affects also the leakage of the duct.
  We will discuss this in the revised manuscript.
  
  Citation: https://doi.org/10.5194/angeo-2021-48-AC1
RC2:
'Comment on angeo-2021-48', Anonymous Referee #2, 11 Oct 2021

Overview

This paper is about multi-wavelength airglow observations of a gravity wave event above northern Germany. The peculiar finding is a strong wave signature visible in the O(¹S) (97 km altitude) and O₂ (94 km altitude) bands, a fainter signature in the Na band (91 km altitude), and no wave signature in the OH band (86 km altitude). Auxiliary data used in the study are wind measurements by a meteor radar network and temperature profiles acquired by the SABER instrument on the TIMED satellite. To my knowledge, this is the first report on such a wave event with co-located local measurements of horizontal wind. Information on horizontal wind is of particular importance for the estimation of intrinsic wave parameters. Using the available measurements, the authors derive a vertical wavenumber profile from the gravity wave dispersion relation and conclude that a thermal duct is responsible for the non-detection of wave signatures in the OH band.

The paper is well written and presents some novel results. I recommend it for publication in Annales Geophysicae subject to the authors addressing the comments below. My main concern is the lack of a plausible explanation for the non-detection of wave signatures in the OH band (see major comment below). Without an explanation, the paper is merely a compilation of observations. Although the uniqueness of the presented observations may justify their publication (I leave this to the editor), a proper description, discussion and evaluation of potential mechanisms responsible for the diminishing modulation in OH airglow brightness will greatly increase the scientific value of this paper.

Major comment

I am surprised that the authors did not use Fig. 6 to estimate the vertical wavelength. From the slopes of the dashed lines I get values ranging between 8 and 12 km. On the other hand, m^2 = 0.5 (taken from Fig. 7e) leads to lambda_z = 8.8 km. This value is consistent with the previous estimate and thus increases confidence in the derived vertical wavelengths. Also, from Fig. 7e I estimate the width of the duct h=5.5 km. According to lambda_z = 2h/n (see e.g., Dong et al. 2021; reference id given below) we can conclude that the wave propagating within the duct is likely of 0^th order (n=1) assuming that the vertical wavelength inside the duct is approximately the same as above the duct. Indeed, based on Fig. 7e, m^2 values at 88 km (center of the duct) and at 95 km (above the duct) are similar. However, the 0^th mode is a symmetric mode and, because the center of the duct (~88 km) is approximately aligned with the center of the OH layer (~86 km), the ducted wave should result in detectable modulations in airglow brightness. Some pieces of the explanation of the non-detection of wave signatures in OH airglow are clearly missing here. On the other hand, the authors provide no explanation either. In my opinion the sentence “The coincidental appearance of the duct layer caused the wave amplitudes to diminish” (l.327) is unsupported speculation. Before making such a statement, the authors should at least discuss conditions which result in cancellation of wave signatures in OH airglow due to the viewing geometry or otherwise.

Minor comments

l. 50: It would be helpful if you could define “large-scale waves”

l. 55: “Instabilities present in the atmosphere do not support free propagation of GWs.” This sentence is not clear to me. The assertion implies that instabilities are a property of the (background) atmosphere. However, the question whether an instability arises also very much depends on the properties of a particular wave propagating through the atmosphere. For instance, depending on the wind profile, an eastward propagating wave may propagate freely whereas a westward propagating wave is filtered due to a critical level.

l. 56: This may be a discussion about terminology, but it is my understanding that wave breaking is a process during which a wave breaks down and energy is transferred to smaller eddies. I wouldn’t call these eddies waves because the word ‘waves’ implies some form of coherent structure. The generation of secondary waves is a separate process.

l. 65: Please explain “inhomogeneities in the background medium”. Do you mean variations in density?

l. 75: “according to whether m^2 arises”. I believe you meant “imaginary m^2”?

l. 124: What is the horizontal averaging of the SABER measurements?

l. 142: What window sizes did you use?

Equation 1: What are the values used for H and k_x?

l. 185: What are the time offsets between acquisitions with different filters?

l. 185: “mean phase velocities”. Are these observed phase velocities?

l. 216: “there will not be any significant differences in the temperature”. Well, I agree that significant differences between the SABER measurement and the temperature one hour later are not very likely, but the possibility can’t be ruled out.

l. 255: “In fact, clear signatures of wave activities were observed on other nights in OH airglow images over northern Germany.” What about before and after the event you described in this paper? Was there any coherent structure observable in all four airglow bands?

l. 264: “The SABER measurements indicated that the centroid height of OH airglow emission occurred near 87 km.” Can you also provide estimates of the thickness of the OH layer (and potentially shape)? Because the vertical wavelength of the gravity wave is likely comparable to the thickness of the OH layer, even small changes in the thickness may have a large impact on the observability of the wave in OH airglow.

l. 273: “We believe that the perturbation wave amplitude will finally decide whether the structure observed in the OH airglow image is faint or not.” Please explain. Are you arguing that the strength of the OH airglow emission (“faint or not”) solely depends on the amplitude of the gravity wave?

l. 294: conductive

l 302: “It is well-known that the MIL tends to be quite stable within a few hours’ time scale.” Well, it depends on how you define “a few hours’ time scale”. For instance, tides can have a huge impact on the MLT . Depending on the phase of the tide, mountain waves may propagate well into to the OH airglow layer or experience critical levels below.

l. 309: delete “that” -> indicates the existence

l. 311: delete first “of” -> some energy

l. 341: Please check the grammar of this and the following sentences.

Literature

Dong, W., Hickey, M. P., & Zhang, S. (2021). A numerical study of gravity waves propagation characteristics in the mesospheric Doppler duct. Journal of Geophysical Research: Atmospheres, 126, e2021JD034680. https://doi.org/10.1029/2021JD034680

Citation: https://doi.org/10.5194/angeo-2021-48-RC2
- AC2: 'Reply on RC2', Sumanta Sarkhel, 23 Nov 2021
  
  We have uploaded the reply as a supplement file.
  
  Citation: https://doi.org/10.5194/angeo-2021-48-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to revisions (further review by editor and referees) (10 Dec 2021) by Dalia Buresova

AR by Sumanta Sarkhel on behalf of the Authors (13 Dec 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (28 Dec 2021) by Dalia Buresova

RR by Anonymous Referee #1 (02 Jan 2022)

RR by Anonymous Referee #2 (31 Jan 2022)

Suggestions for revision or reasons for rejection

Thank you for your explanations and clarifications. I am sorry that I could not reply earlier.

I am still having difficulties following your arguments. In your comments you write:

“We suggest that the ducting layer (85-91 km) observed in the present case (Figure 7c & e) was weak and so inhibited free propagation within that particular altitude region. Only part of the energy of the wave could penetrate through the bottom of the duct layer and the other part of the energy will be reflected back downward. Being localized, the part of the wave continued to propagate upwards into the O2 and O(1S) emission layer at higher altitudes (represented schematically in Figure 8). Since only a part of the wave energy could enter in the duct region (85-91 km), the wave structure was observed to be faint in the OH airglow images.”

Please forgive me my ignorance. According to your above explanation, upon reaching the bottom of the wave duct, the wave packet splits into three parts: one part being reflected and propagating downward (this part does not show up in Fig. 8), another part being trapped inside the duct, and the remaining part continuing propagating upward. Inside the duct, the waves are guided towards north-east and eventually cross your observation volume. But how do upward propagating waves which are detected in the O(1S) and O2 bands get there? If they crossed the thermal duct in the south-west where the ducting was supposedly weaker, the waves would have to propagate at a very shallow angle (nearly horizontally) in order to reach the observation volume at 94 km altitude (O2 emission). Given the large vertical wavelength, this doesn’t seem to be likely. The other possibility I can think of – waves crossing the thermal duct within the observation volume – is not likely either, because these waves should show up in the OH band with approximately the same intensity as in the O(1S) band. Is there any obvious piece of information that I am missing?

Based on the data at hand there may not be a fully convincing explanation. But in the interest of scholarship you may want to acknowledge any inconsistencies in the explanation presented in the manuscript. In particular, the depiction of waves in Fig. 8 is misleading. I don’t think waves can be refracted around the duct layer in this way.

Hide

ED: Publish subject to minor revisions (review by editor) (09 Feb 2022) by Dalia Buresova

Dear Dr. Samanta Sarkhel, dear co-authors,

It is my pleasure to inform you that both referees are in principle satisfied with your response to their comments and suggestions. However, one of the opponents would like to clarify the explanation you provided (please, see below). Please consider this request. Currently, the article requires this minor revision.
Kindest regards

Yours cordially
Dalia Buresova

Referee #1

Thank you for your explanations and clarifications. I am sorry that I could not reply earlier.

I am still having difficulties following your arguments. In your comments you write:

“We suggest that the ducting layer (85-91 km) observed in the present case (Figure 7c & e) was weak and so inhibited free propagation within that particular altitude region. Only part of the energy of the wave could penetrate through the bottom of the duct layer and the other part of the energy will be reflected back downward. Being localized, the part of the wave continued to propagate upwards into the O2 and O(1S) emission layer at higher altitudes (represented schematically in Figure 8). Since only a part of the wave energy could enter in the duct region (85-91 km), the wave structure was observed to be faint in the OH airglow images.”

Please forgive me my ignorance. According to your above explanation, upon reaching the bottom of the wave duct, the wave packet splits into three parts: one part being reflected and propagating downward (this part does not show up in Fig. 8), another part being trapped inside the duct, and the remaining part continuing propagating upward. Inside the duct, the waves are guided towards north-east and eventually cross your observation volume. But how do upward propagating waves which are detected in the O(1S) and O2 bands get there? If they crossed the thermal duct in the south-west where the ducting was supposedly weaker, the waves would have to propagate at a very shallow angle (nearly horizontally) in order to reach the observation volume at 94 km altitude (O2 emission). Given the large vertical wavelength, this doesn’t seem to be likely. The other possibility I can think of – waves crossing the thermal duct within the observation volume – is not likely either, because these waves should show up in the OH band with approximately the same intensity as in the O(1S) band. Is there any obvious piece of information that I am missing?

Based on the data at hand there may not be a fully convincing explanation. But in the interest of scholarship you may want to acknowledge any inconsistencies in the explanation presented in the manuscript. In particular, the depiction of waves in Fig. 8 is misleading. I don’t think waves can be refracted around the duct layer in this way.

Hide

AR by Sumanta Sarkhel on behalf of the Authors (28 Feb 2022) Author's response Author's tracked changes Manuscript

ED: Publish as is (01 Mar 2022) by Dalia Buresova

AR by Sumanta Sarkhel on behalf of the Authors (01 Mar 2022)

Short summary

A rare gravity wave event was observed on the night of 25 April 2017 over northern Germany. An all-sky airglow imager recorded an upward-propagating wave at different altitudes in mesosphere with a prominent wave front above 91 km and faintly observed below. Based on wind and satellite-borne temperature profiles close to the event location, we have found the presence of a leaky thermal duct layer in 85–91 km. The appearance of this duct layer caused the wave amplitudes to diminish below 91 km.