Predicting ocean waves along the US east coast during energetic winter storms: sensitivity to whitecapping parameterizations

Allahdadi, Mohammad Nabi; He, Ruoying; Neary, Vincent S.

doi:https://doi.org/10.5194/os-15-691-2019

Articles | Volume 15, issue 3

https://doi.org/10.5194/os-15-691-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/os-15-691-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 3

Research article

|

06 Jun 2019

Research article |

| 06 Jun 2019

Predicting ocean waves along the US east coast during energetic winter storms: sensitivity to whitecapping parameterizations

Mohammad Nabi Allahdadi, Ruoying He, and Vincent S. Neary

Download

Final revised paper (published on 06 Jun 2019)
Preprint (discussion started on 01 Nov 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Peer review of OS-2018-112', Anonymous Referee #1, 24 Nov 2018
- AC3: 'Response to Comments from Reviewer#1', M.Nabi Allahdadi, 27 Dec 2018
RC2: 'Predicting Ocean Waves along the U.S. East Coast During Energetic Winter Storms: sensitivity to Whitecapping parameterizations', Anonymous Referee #2, 28 Nov 2018
- AC1: 'Reply to comments from Reviewer#2 on os-2018-112', M.Nabi Allahdadi, 16 Dec 2018
  - RC4: 'Recommended:Predicting Ocean Waves along the U.S. East Coast During Energetic Winter Storms:...', Anonymous Referee #2, 25 Dec 2018
    - AC4: 'Final response to Reviewer#2', M.Nabi Allahdadi, 02 Jan 2019
- AC4: 'Final response to Reviewer#2', M.Nabi Allahdadi, 02 Jan 2019
- AC2: 'Reply to comments from Reviewer#3 on os-2018-112', M.Nabi Allahdadi, 17 Dec 2018
RC3: 'Review: Predicting ocean waves along the US East Coast during energetic winter storms: Sensitivity to white-capping parameterizations', Anonymous Referee #3, 06 Dec 2018
- AC2: 'Reply to comments from Reviewer#3 on os-2018-112', M.Nabi Allahdadi, 17 Dec 2018

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by M.Nabi Allahdadi on behalf of the Authors (29 Jan 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Jan 2019) by Neil Wells

RR by Anonymous Referee #2 (10 Feb 2019)

RR by Anonymous Referee #1 (10 Feb 2019)

Suggestions for revision or reasons for rejection

Having seen earlier versions of this manuscript, the revised version has improved considerably. I appreciated the marked sections which highlight the changes in view of previous reviewer’s comments. The new figures are a good addition. My advice is a now a minor revision.

The authors present the results of a sensitivity study of applying two wave model settings concerning the whitecapping parameterisations. As the results of the different settings show (as expected) different results, the authors performed an analysis to explain the differences and determined the range of applicability.

The problem can be pinpointed further. As we all know the present source terms are not yet truly third-generation and they still contain a lot of assumptions and shortcomings. As a consequence, tuning of models is often the way to proceed. But tuning is mostly done on total Hsig and some period measure. Due to this emphasis on the total wave height, the low-frequency wave components may perform poorer, especially when Komen type whitecapping formulations are used. Although Rogers et al. (2003) recommend a small fix by choosing delta=1 in this formulation. Reasons for tuning a model have been discussed in Bingolbali et al. (2019).

#26 It’s rather the shortcomings of various source terms, not only the balance

#57 Correct reference to Van Vledder et al., 2016.

#65 As discussed in W007 his new source term was primarily developed for coastal and inland waters with relatively short fetches.

#78-#81 This is incorrect. The OMP version of SWAN4120 contains ST6. Please drop this fake argument. Implementation issues can never be a proper argument for not doing something. Please concentrate on scientific/physical arguments. A stronger argument is twofold. There is limited experience with ST6 in coastal waters, and the range of applicability of W007 for larger fetches with inhomogeneous wind fields is poorly known. As said before, the crucial difference between Komen type and W007 whitecapping source term is the use (or not) of a mean wave steepness, which gives modelling errors in multi-peaked seas. Note that W007 did not really test his source term for multi-peaked seas.

#82 please state which complications may arise. This is now too vaguely formulated.

#87 The apparent contradiction is just another argument that no generally applicable source term setting exists requiring tuning. In the North Sea also shallow water effects like bottom friction are a significant component of the total source term balance.

#104 If you know about the gustiness effect, why did you not use it to ‘correct’ the CFSR winds?

#108 Now it becomes clear why you used W007. When you stick to the available SWAN-ADCIRC executable you run into unnecessary problems.

#112 what are these?: … each Komen method.. This is not yet introduced. I can imagine choosing delta=0 or delta=1 may be choices.

#116 You should add also the relatively large fetches and instationary large wind fields as an argument. No reason to state that no new form is suggested. You may put that in the discussion.

#124 Only using January, although representative for the winter months, is still a bit meager. You can also reverse the argument, because you want to study some features of the wind fields, you choose this month for studying. Moreover, this also fits better in view of the limitations of a fetch-limited approach. This is a stronger argument and better fits in the purpose of this manuscript.

#134 It is better to discuss the physical aspects, viz. the impact on lower or higher frequencies, see Rogers et al. (2003) for a discussion

#186 typo: July 2009

#195 typo: wind roses (2 words)

#208 …was developed… suggest that a new model setup was created. This seems to contradict the statement that you were forced to use an existing SWAN-ADCIRC approach.

#216 unit of bias is wrong, should be m/s. Lower panels in Figure 4 shows systematic trend of under-prediction with higher wind speeds. This should be noted, as well as the question whether you corrected for this or not.

#218 before -> around

#218 Make the buoy position markers and text larger for visibility

#222 timestep is the wrong word here. A time step is an interval, whereas here you mean a moment in time. Please rephrase.

#236 which two approaches? From earlier remarks I count 3. One W007 and two Komen-type formulations?. If only one Komen is used, what delta is used?

#238 as defaults may change over time (as they did), you can better explicitly state the settings used.

#242 It is still unclear which delta is used in the Komen formulation. Please note that the choice of delta may have significant effects.

#252 Crucial pieces of information are missing here. Which type of buoys are used? Over which frequency range have the buoy spectra been integrated? Now it seems that different frequency intervals have been used. In case SWAN table output for Tm02 has been used, the integration is up to 10 Hz (using the prognostic and parametric spectrum). Buoys usually deliver reliable spectra up to 0.5 Hz. This mismatch may cause significant differences for especially Tm02, see section 4.3 Akpinar et al. (2012) for a discussion on this topic. In such a comparison, one should always use the same frequency interval to derive parameters. For Hm0 the effect is often insignificant, but the higher the frequency moment, the larger the effect. This effect may partly explain the under-prediction. Please check carefully, the consequences.

#271 Note that this under-prediction may also be due to the fact that no calibration was performed.

#300 Can you find a reason why Komen spectra are larger than those of W007. Is this due to the use of a mean wave steepness which in the presence of low-frequency wave components results in an over-prediction of the higher frequencies?

#309 Author Clyson does not exist. Check reference

#326 Can you estimate how much wind strength errors contribute to the total prediction error. As noted, there is an inherent prediction error in each source term, but here also the numerical determination of Tm02 may contribute to the total error. Without such a quantitative consideration, the discussion is a bit pointless.

Figure 9 Add proper name and symbol of wave parameters in legend ‘significant’ wave height Hm0 , .. mean wave period Tm02

#332 Specify the kind of source terms shown in Figure 11. I guess they are integrated source term magnitudes. This should be explained.

Figure 11 This figure seems inconsistent with what I expect in relation to Figure 9.
1) The shape of the spatial distributions seems inconsistent with those in Figure 9.
2) The upper panels have a color band in yellow, whereas this is not present in the lower panels.
3) Using the unit w/m2 is wrong, as these are integrated source terms of the rate of change of wave variance. No multiplication with rho g has to be done.
4)
#336 As mentioned earlier, the source terms are different because the underlying variance density spectra are also different. Still, I agree that for the same spectrum, the trend is similar. This was easily checked for a JONSWAP spectrum.

#344 This choice is related to keeping the total balance appropriate.

#357 The chosen spectrum and related wcap source terms do not really show some effect for small frequencies. I did a test for a JONSWAP spectrum and then subtle for significant differences popup.

#360 A comment on the strange shape of Snl4 is missing

#368 How can a model be more than reality? Please rephrase

#393 are resulted -> result

#430 result -> results

#442 Such a distinction will be difficult in practice. This is another indication that these formulations are not generally applicable. This is an interesting conclusion. As noted in Bingolbali, tuning for a large basin may give different results when tuning is done for a local area.

#447 rephrase … too much…, -> occur often?

#475 directional spectra cannot be observed, only reconstructed from the Fourier coefficients that can be distilled from the buoys time series. So, state how the buoy spectra were reconstructed

#489 How do you quantify directional spreading?

#468 Note that the quadruplet term here is the crude DIA. See Ardhuin et al. (2007) and Bottema and Van Vledder (2008) for possible effects on slanting fetch situations when using the accurate Snl4.

#515 This is a welcome addition to the manuscript, leading to a nice conclusion about the inclusion of stability effects in wave modelling.

#557 calibrated by whom and in what way? This is an important notion.

#571 The logic here is a bit flawed. Variabilities in the wind field itself are no cause for slanting fetch effects. Please rephrase.

#585 What are ‘real values’?

#588 technical -> physical ?

#590 omit the distinction between serial and parallel, that is not relevant

References
Akpinar et al. (2012), Continental Shelf Research 50-51.
Bingolbali et al. (2019), Ocean Engineering, Vol 172
Bottema and Van Vledder, Coastal Engineering, 55.

Hide

ED: Publish subject to minor revisions (review by editor) (11 Feb 2019) by Neil Wells

AR by M.Nabi Allahdadi on behalf of the Authors (05 Mar 2019) Author's response Manuscript

ED: Publish subject to minor revisions (review by editor) (25 Mar 2019) by Neil Wells

AR by M.Nabi Allahdadi on behalf of the Authors (25 Mar 2019) Author's response Manuscript

ED: Publish as is (29 Apr 2019) by Neil Wells

AR by M.Nabi Allahdadi on behalf of the Authors (01 May 2019) Manuscript

Short summary

Dissipation of ocean waves due to whitecapping is one of the most important processes that affect generation of gravity waves by wind. Different behavior of traditional approaches used for quantifying whitecapping dissipation under different wave conditions has always been a challenge to choose the most appropriate approach for a given area. The present paper examines the performance of two popular whitecapping approaches incorporated in SWAN during the winter storms along the US east coast.