the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Jan 2025
A New-Generation Internal Tide Model Based on 30 Years of Satellite Sea Surface Height Measurements
Abstract. An internal tide model ZHAO30yr is developed using 30 years of satellite altimetry sea surface height measurements from 1993 to 2022 by a newly improved mapping technique that consists of two rounds of plane wave analysis with a spatial bandpass filter in between. Prerequisite wavelengths are obtained using climatological annual-mean hydrographic profiles in the World Ocean Atlas 2018. The model contains 12 internal tide constituents: 8 mode-1 constituents (M2, S2, N2, K2, K1, O1, P1, and Q1) and 4 mode-2 constituents (M2, S2, K1, and O1). Model errors are estimated to be lower than 1 mm in amplitude, thanks to the new mapping technique and long data record. The model is evaluated by making internal tide correction to independent altimetry data in 2023. Ten constituents (but for K2 and Q1) can reduce variance on global average; K2 and Q1 can cause variance reductions in their source regions. The model decomposes the multiconstituent multimodal multidirectional internal tide field into a series of plane waves at each grid point. The decomposition reveals unprecedented features previously masked by multiwave interference. The model divides each internal tide constituent into components by propagation direction. The directionally-decomposed components reveal numerous long-range internal tidal beams associated with notable topographic features. The semidiurnal beams off the Amazon shelf and the diurnal beams in the Arabian Sea are examined in detail. ZHAO30yr is available for download at http://doi.org/10.6084/m9.figshare.28078523 (Zhao, 2024).
- Preprint
(12023 KB) - Metadata XML
-
Supplement
(9946 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on essd-2024-611', Anonymous Referee #1, 29 Jan 2025
The comment was uploaded in the form of a supplement: https://essd.copernicus.org/preprints/essd-2024-611/essd-2024-611-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Zhongxiang Zhao, 16 Mar 2025
The comment was uploaded in the form of a supplement: https://essd.copernicus.org/preprints/essd-2024-611/essd-2024-611-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Zhongxiang Zhao, 16 Mar 2025
-
RC2: 'Comment on essd-2024-611', Anonymous Referee #2, 25 Feb 2025
Summary:
I consider this journal focused toward documenting data sets that are useful to the earth science community. As the new 30-year internal tide model is likely to be quite useful, I am happy to see it so thoroughly documented. I have some suggestions about items that appear confused or confusing, or might improve the presentation.1. I think it is too much to show all 12 derived constituents in large 3-panel figures. This covers Figures 7 through 18, a lot of figures and a lot of page space. Each figure is described in the text and most readers, if like me, will eventually find this tedious. Cannot some of this go into the Supplement? Are they all so important to justify so much page space? Â "A picture is worth a thousand words" (Line 296), but it can still be too many, unless there is good reason to show all of these.
2. In each figure there are two panels (b and c) that decompose the wave field into two directions. Â These are interesting to see, but I am less clear about what they mean and what Dr Zhao thinks they mean. At Line 58 it is claimed that this decomposition "reveals numerous long-range internal tidal beams." But so does the original (not decomposed) wave field. Â It is my understanding that the apparent beams (horizontal, not vertical!) represent mostly interference patterns (e.g., Rainville & Pinkel, 2006), and this must still be true in all three of the figure panels. I still think the decompositions are useful, because they (presumably) reflect in some way how much wave energy at a location is arriving from which direction. But the justification for these decompositions and the assertions about "beams" needs some more careful thinking.
3. There is some apparent confusion about "sun-synchronous" measurements and how these impact the S2 waves - Lines 169-172. The statement that "signals caused by solar radiance have longer spatial scales" is not the most important point. That is not the problem with sun-synchronous measurements. The problem is that all the altimeter data measure the tide at the same phase, so you are trying to solve for a sine wave when you have measurements at only one phase. I think the reason S2 is recovered here is because there is a lot of CryoSat data. If the fitting were tried with only the sun-synch data, it would be less good and maybe fail, I suspect. Â Also, (Line 170), Ubelmann et al. solved for S2, but they did not discuss the results, and it is not clear how successful their S2 solution was.
One item that does potentially point to small S2 errors (caused by lack of phase sampling in sun-synchronous measurements) can be seen by comparing M2, N2, and S2 results - Figures 7, 9, and 11. Patterns in N2 look more like M2 than do patterns in S2, even though S2 forcing is closer to M2 forcing. For example, look at the southward components in panels (c) in the North Pacific. Â S2 appears different. I suspect this is because of the problem of so many altimeters that do not sample S2 well enough.
4. Related to that, I suspect that another reason Zhao20 is inferior to Zhao30 for S2 is because there was far less CryoSat data in the old solution. Â This affects the explanation Lines 287-291.
5. The Abstract needs to acknowledge that the mapping is only the "phase-locked" component. This is clear from the Intro, but it also should be stated in the Abstract.
6. Since non-repeat altimeter data is used, it is important in Section 2.1 to state what Mean Sea Surface model was used. Â Also, Line 86, polar tide should be pole tide.
7. Table 1, column "Bandpass width" has units?
8. Line 123. I think this statement should be removed unless it can be backed up with evidence.
9. Line 139: For plane-wave fitting, perhaps Ray and Cartwright (2001, GRL) should be mentioned here?
10. Section 4.1 on model errors describes an interesting approach to determine errors, with interesting results. Also, I notice that mode-2 errors are larger than mode-1 errors, even in absolute terms, not just relative. Â I like this section.
11. In Figure 3 caption, please state what the numbers in upper left mean. In Figure 4 caption, please state units (yes, the figure gives mm^2, but it is tiny in upper right and easy to miss). In Figure 8 caption, what is "too many circles"?
12. Line 437: Answering WHAT question?
13. Figure 19 caption. I do not understand the sentence "Note that the largest..." Â Also, fix apparent typo on "flurier".
More typos:
78 - ??
Figure 19h, "moe"
504 - should "model" be "data"?
728 - "using"Â
Citation: https://doi.org/10.5194/essd-2024-611-RC2 -
AC2: 'Reply on RC2', Zhongxiang Zhao, 16 Mar 2025
The comment was uploaded in the form of a supplement: https://essd.copernicus.org/preprints/essd-2024-611/essd-2024-611-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Zhongxiang Zhao, 16 Mar 2025
Data sets
ZHAO30yr: An internal tide model based on 30 years of satellite sea surface height measurements Zhongxiang Zhao https://doi.org/10.6084/m9.figshare.28078523.v1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 212 | 37 | 8 | 257 | 29 | 6 | 4 |
- HTML: 212
- PDF: 37
- XML: 8
- Total: 257
- Supplement: 29
- BibTeX: 6
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1