the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Stratospheric gravity waves excited by Hurricane Joaquin in 2015: 3-D characteristics and the correlation with hurricane intensification
Abstract. Despite progress, accurately forecasting tropical cyclone (TC) intensity, especially rapid intensification, remains a significant challenge. The correlations between the stratospheric gravity waves (GWs) excited by TCs and TC intensity have been recognized. However, partly due to the limitations of conventional analysis methods and observational filters of current satellite instruments, the characteristics of stratospheric GWs that indicate TC intensification remain unclear. This study examined the specific characteristics of GWs and their linkage to hurricane intensification by high-resolution, realistic model simulations and 3-D wave analysis method. First, the stratospheric GWs excited by Hurricane Joaquin in 2015 were simulated using the Advanced Weather Research and Forecasting (WRF) model. Then, the GW characteristics were analyzed using the novel 3-D Stockwell transform method. The GWs excited by Hurricane Joaquin are in the mid-frequency range and propagate outward from the hurricane center counterclockwise while moving upward in a spiral. A high-level time-lagged correlation exists between the intensities of the hurricane and stratospheric GWs during hurricane intensification, making it possible to detect an increase in hurricane intensity by observing an increase in stratospheric GW intensities. Compared to the weakening period, the stratospheric GWs excited during hurricane intensification exhibit relatively higher frequencies, shorter horizontal wavelengths, and longer vertical wavelengths, with this contrast particularly evident near the center of the hurricane. This study provides further knowledge for potentially monitoring hurricane intensification by observing stratospheric GWs using satellite instruments in the infrared and microwave bands when it is difficult to use other measurement techniques.
- Preprint
(5125 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-3008', Anonymous Referee #1, 24 Jan 2024
Stratospheric gravity waves excited by Hurricane Joaquin in 2015: 3-D characteristics and the correlation with hurricane intensification
General comments
In the paper “Stratospheric gravity waves excited by Hurricane Joaquin in 2015: 3-D characteristics and the correlation with hurricane intensification” by Xue Wu et al., properties of gravity waves excited by Hurricane Joaquin are studied, using a WRF model simulation. The study is a continuation of work published in the paper Wu et al. 2022, mainly discussing the previous analysis, with focus on division to tropical cyclone intensification and weakening and on the analysis of wavelengths by 3D Stockwell transform. In general, this topic is important to increase our knowledge about gravity waves and tropical cyclones, with the perspective of improving predictions of rapid intensification of the cyclones.
My major concern about the study is the similarity to the publication Wu et al. 2022. First, texts in the Data and WRF Model Configuration sections are very similar between the papers. This is, of course, understandable, since the data and model configuration are the same. Still, I believe that a few summarizing sentences with a reference to the old paper might be preferable over creating a slightly reformulated duplicate. Section 5.1 introducing the simulation is mildly extended by showing also the results from the coarse-scale subdomain. However, Fig. 5 presented here as the principal result of Section 5.2 was already published as Fig. S2 of Wu et al. 2022, which is not even mentioned in the submitted paper. And finally, it is not true that this is the first time when 3D Stockwell transform was used for analysis of convective gravity waves, since the method was already used to compute wavelengths in the Wu et al. 2022 paper (although with shorter discussion).
With this in mind, the paper gives the impression of being very short. The main, so-far unpublished, contributions are the plots of the distribution of wavelengths that support the properties already discussed in Wu et al. 2022.
Also, results in the paper are often presented as general properties of tropical cyclones, although they are based on a single simulation of a single tropical cyclone. More work in this direction (either adding simulations of other cases or a more-detailed discussion of connection to other studies) would increase the validity of the results.
From the positive point of view, the study is comprehensible, with illustrative figures and without extensive number of typos or similar technical problems. Still, I believe the presented work might be considered for publication only after it has been sufficiently extended by more original results. Some specific and technical comments are listed below.
Specific comments
L157: “we find the peak spectral amplitude in the localized spectrum”: What does the spectrum look like? Is the Peak amplitude well defined?
L163: Are the wavenumbers fx, fy and fz really taken to be wavenumbers k, l and m? In my understanding, this would lead to a different definition of the wavenumbers than e.g. in Fritts and Alexander, 2003 (2 pi factor), so equations 7 and 8 would need some modifications.
L220: “Generally, GWs exhibit higher intensity during the intensification period compared to the weakening period”. After 10-3, this does not seem to be so clear. Do you have some idea why this could be?
L237: Again, is there always a unique “large” time lag? Some ambiguity in the time lag might align with the large spreads in the plots.
Technical corrections
L24 and further: For me, it is very questionable if the 3D Stockwell transform can be considered as “novel”, when the original 1D variant was introduced in 1996, with the straight-forward 2D generalisation being developed (for example in the papers of Stockwell) a few years later. Although the first applications on the GWs were using the 1D method only (e.g. around 2010), the use of the 2D transform in a very similar way to the use of the 3D method in the presented study was applied already in Hindley et al. 2016, closely followed by studies using 3D Stockwell transform. Similar time axes can be constructed for many other GW separation methods.
Hindley, N. P., Smith, N. D., Wright, C. J., Rees, D. A. S., & Mitchell, N. J. (2016). A two-dimensional Stockwell transform for gravity wave analysis of AIRS measurements. Atmospheric Measurement Techniques, 9(6), 2545-2565.
L34: Statement about the past decade is referenced by a 10 years old paper – either add a newer source or reformulate.
L92: I suggest switching the order of chapters 2 and 3: Chapter 2 currently refers to the simulation that was not mentioned before. (Although merging and significant shortening of the chapters, as discussed above, might be even better.)
L138, L143, L147: Dots after the equations.
L145 and further: The multi-dimensional formulas would be more transparent with a notation for vectors (arrows/bold letters).
L148: Missing space after “Here”.
L164: fx twice.
L193: Until 1st October, it seems to me that the D01 domain is even closer to the observation, so I do not see the discussed contrast in this period.
L304: “This phenomenon is expected and aligns with the characteristics observed in a realistic hurricane case.” Do you have a reference to support this statement?
L307 – L308: Math symbols should be in math style.
L339 – L340: Possibly put the wavenumbers to a bracket, so that it does not look like new information?
L378 – L379 (and elsewhere): Described one simulation only, so use “simulation” instead of “simulations”.
L387: a specific hurricane case -> the specific hurricane case.
Citation: https://doi.org/10.5194/egusphere-2023-3008-RC1 -
AC1: 'Reply on RC1', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xue Wu, 10 Oct 2024
-
RC2: 'Comment on egusphere-2023-3008', Anonymous Referee #2, 25 Jan 2024
-
AC2: 'Reply on RC2', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Xue Wu, 10 Oct 2024
-
RC3: 'Comment on egusphere-2023-3008', Anonymous Referee #3, 26 Jan 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-RC3-supplement.pdf
-
AC3: 'Reply on RC3', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC3-supplement.pdf
-
AC3: 'Reply on RC3', Xue Wu, 10 Oct 2024
Status: closed
-
RC1: 'Comment on egusphere-2023-3008', Anonymous Referee #1, 24 Jan 2024
Stratospheric gravity waves excited by Hurricane Joaquin in 2015: 3-D characteristics and the correlation with hurricane intensification
General comments
In the paper “Stratospheric gravity waves excited by Hurricane Joaquin in 2015: 3-D characteristics and the correlation with hurricane intensification” by Xue Wu et al., properties of gravity waves excited by Hurricane Joaquin are studied, using a WRF model simulation. The study is a continuation of work published in the paper Wu et al. 2022, mainly discussing the previous analysis, with focus on division to tropical cyclone intensification and weakening and on the analysis of wavelengths by 3D Stockwell transform. In general, this topic is important to increase our knowledge about gravity waves and tropical cyclones, with the perspective of improving predictions of rapid intensification of the cyclones.
My major concern about the study is the similarity to the publication Wu et al. 2022. First, texts in the Data and WRF Model Configuration sections are very similar between the papers. This is, of course, understandable, since the data and model configuration are the same. Still, I believe that a few summarizing sentences with a reference to the old paper might be preferable over creating a slightly reformulated duplicate. Section 5.1 introducing the simulation is mildly extended by showing also the results from the coarse-scale subdomain. However, Fig. 5 presented here as the principal result of Section 5.2 was already published as Fig. S2 of Wu et al. 2022, which is not even mentioned in the submitted paper. And finally, it is not true that this is the first time when 3D Stockwell transform was used for analysis of convective gravity waves, since the method was already used to compute wavelengths in the Wu et al. 2022 paper (although with shorter discussion).
With this in mind, the paper gives the impression of being very short. The main, so-far unpublished, contributions are the plots of the distribution of wavelengths that support the properties already discussed in Wu et al. 2022.
Also, results in the paper are often presented as general properties of tropical cyclones, although they are based on a single simulation of a single tropical cyclone. More work in this direction (either adding simulations of other cases or a more-detailed discussion of connection to other studies) would increase the validity of the results.
From the positive point of view, the study is comprehensible, with illustrative figures and without extensive number of typos or similar technical problems. Still, I believe the presented work might be considered for publication only after it has been sufficiently extended by more original results. Some specific and technical comments are listed below.
Specific comments
L157: “we find the peak spectral amplitude in the localized spectrum”: What does the spectrum look like? Is the Peak amplitude well defined?
L163: Are the wavenumbers fx, fy and fz really taken to be wavenumbers k, l and m? In my understanding, this would lead to a different definition of the wavenumbers than e.g. in Fritts and Alexander, 2003 (2 pi factor), so equations 7 and 8 would need some modifications.
L220: “Generally, GWs exhibit higher intensity during the intensification period compared to the weakening period”. After 10-3, this does not seem to be so clear. Do you have some idea why this could be?
L237: Again, is there always a unique “large” time lag? Some ambiguity in the time lag might align with the large spreads in the plots.
Technical corrections
L24 and further: For me, it is very questionable if the 3D Stockwell transform can be considered as “novel”, when the original 1D variant was introduced in 1996, with the straight-forward 2D generalisation being developed (for example in the papers of Stockwell) a few years later. Although the first applications on the GWs were using the 1D method only (e.g. around 2010), the use of the 2D transform in a very similar way to the use of the 3D method in the presented study was applied already in Hindley et al. 2016, closely followed by studies using 3D Stockwell transform. Similar time axes can be constructed for many other GW separation methods.
Hindley, N. P., Smith, N. D., Wright, C. J., Rees, D. A. S., & Mitchell, N. J. (2016). A two-dimensional Stockwell transform for gravity wave analysis of AIRS measurements. Atmospheric Measurement Techniques, 9(6), 2545-2565.
L34: Statement about the past decade is referenced by a 10 years old paper – either add a newer source or reformulate.
L92: I suggest switching the order of chapters 2 and 3: Chapter 2 currently refers to the simulation that was not mentioned before. (Although merging and significant shortening of the chapters, as discussed above, might be even better.)
L138, L143, L147: Dots after the equations.
L145 and further: The multi-dimensional formulas would be more transparent with a notation for vectors (arrows/bold letters).
L148: Missing space after “Here”.
L164: fx twice.
L193: Until 1st October, it seems to me that the D01 domain is even closer to the observation, so I do not see the discussed contrast in this period.
L304: “This phenomenon is expected and aligns with the characteristics observed in a realistic hurricane case.” Do you have a reference to support this statement?
L307 – L308: Math symbols should be in math style.
L339 – L340: Possibly put the wavenumbers to a bracket, so that it does not look like new information?
L378 – L379 (and elsewhere): Described one simulation only, so use “simulation” instead of “simulations”.
L387: a specific hurricane case -> the specific hurricane case.
Citation: https://doi.org/10.5194/egusphere-2023-3008-RC1 -
AC1: 'Reply on RC1', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xue Wu, 10 Oct 2024
-
RC2: 'Comment on egusphere-2023-3008', Anonymous Referee #2, 25 Jan 2024
-
AC2: 'Reply on RC2', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Xue Wu, 10 Oct 2024
-
RC3: 'Comment on egusphere-2023-3008', Anonymous Referee #3, 26 Jan 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-RC3-supplement.pdf
-
AC3: 'Reply on RC3', Xue Wu, 10 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2023-3008/egusphere-2023-3008-AC3-supplement.pdf
-
AC3: 'Reply on RC3', Xue Wu, 10 Oct 2024
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
483 | 135 | 47 | 665 | 17 | 15 |
- HTML: 483
- PDF: 135
- XML: 47
- Total: 665
- BibTeX: 17
- EndNote: 15
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1