Review of the paper “Tropical cyclone storm surge probabilities for the east coast of the United States: a cyclone-based perspective” by Towey et al, 2021.

Here the Authors analyze how much of surges characteristics in selected locations along the East US coast may explain by TCs producing them or crossing the areas nearby. The work is quite very interesting and absolutely worthwhile to be published. However sometimes the text and above all, the methodology are not very clear and the use of terms such as “noisy” is somehow misleading. Thus I think that text can be published after some major revisions.

Line 20-22: please reformulate

Lines 25-29 : I would merge these two sentences together

Line 35 : I would avoid to use the term “scenario” because as far as I understand you are talking about “atmospheric configuration”

Line 45-47 : Could you please explain what you mean with “noisy”?

Line 49: Again…what do you mean with “noisy”? how is this relationship? Linear? Please explain better

Line 62 : As the sentence is formulated, it looks like that storm surges affect TC characteristics. Is it like that? Please explain.

Line 86-87: Could you please explain why it is not important in the statistics neglecting the wave setup? Do you have a reference for that?

2.2 Methods section

A1) maybe I miss the point in the text but do you explain somewhere in text the choice of the radius for the search of TC in the circle centered on the location of the surge? In Figure 2 you talked about 250 km, why do not you choice 400 or 500 km?

A2) I do not understand well the method. You consider all the TCs crossing a radius in a certain distance from the location and you associated the closest one to the event in the day of each maximum daily storm surge, do you? If so it is not very clear from text. If you have two closed systems crossing the area (probably something rare or impossible) how do you find that one responsible for the event?

A2) Why do you choose a liner fit as best fit and not for example an exponential fit as done in the references in the introduction?

Line 154 : I think that the use of term noise is misleading. I would just state that the propagation speed is less important than other variables

Line 161-164 : not very clear this sentence, please rephrase

Section 3.1

I would avoid to say “strong enough predictor” or “better predictor”. What I see is that the variables that you consider are not able to explain completely the variability observed for the storm surges . Please reformulate

Line 215-17 : not very clear… do you mean TC stronger that the climatology of the systems crossing the area?

Line 303-305: Not very clear..as far as I see, you correlate storm surges and TCs characteristics not the opposite. Again the adjective noisy here is not correct in the sense that relationship between surge feature and TC, I think, is not linear not noisy. Please rephrase

Review “Tropical cyclone storm surge probabilities for the east coast of the United States: A cyclone-based perspective”

General comments

In their manuscript “Tropical cyclone storm surge probabilities for the East Coast of the United States: A cyclone-based perspective”, the authors seek to identify relationships between tropical cyclone (TC) characteristics and storm surge heights along the US coastline. While I believe this is a relevant topic to study, I recommend additional analyses to improve on the novelty of the research. Please find below my reasoning:

1) The authors consider the TC distance to tide gauge station, TC intensity, and TC angle at landfall in their analysis. I feel it’s debatable whether the distance to a tide gauge station is something that can be truly attributed as a TC characteristics (why not consider TC size?). In addition, various past studies have (extensively) discussed similar characteristics. Many of them are already cited in the text (Lines 44 – 52) so I will not repeat them here, but these could be added for a more comprehensive overview of what’s already been done:

-  Needham & Keim (2014) (https://doi.org/10.1175/2013EI000558.1) who assessed the influence of storm size on hurricane surge;

- Ramos-Valle et al (2020) (https://doi.org/10.1029/2019JD031796) who extensively studied the influence of TC landfall angle on storm surges along the Mid-Atlantic Bight;

- Bloemendaal et al (2019) (https://doi.org/10.1007/s00382-018-4430-x) who also assessed the influence of various different TC and geographical characteristics on storm surges;

- Peng et al (2006) (https://doi.org/10.1016/j.ocemod.2006.03.004) on the asymmetry of storm surges and TC wind fields;

- Akbar et al (2017) (https://doi.org/10.3390/jmse5030038) on the influence of wind drag coefficients and bottom friction on Hurricane Rita’s storm surge height

While I welcome research seeking additional answers to explain storm surge heights, I would strongly recommend the authors to improve on the novelty of the research to make this research truly stand out compared to the literature that’s already out there. This can be achieved through (for example) 1) including more TC and landfall (coastal slope/coastal complexity/terrain features near the tide gauge station) characteristics 2) extensively seeking for multivariate relationships 3) and to also trying different types of relationships rather than just a linear one.

2) Throughout the manuscript, it seems like the authors are solely looking at TCs in their analysis. However, in the Methods-section, they say that they also include TCs that have undergone extratropical transition. These systems can no longer be considered tropical by nature (rather, extratropical), hereby having different characteristics than TCs and they should thus be excluded from the analysis.

3) The authors use daily maximum storm surge heights and couple this with 6-hourly TC data. I don’t see the added value of using daily maximum storm surge heights when the tide gauge data is provided in hourly data (see line 85) and TCs are characterized by strong spatial and temporal gradients that can strongly vary within hours.

4) The results-section could benefit from some in-depth discussion of why the spatial differences emerge in relation to typical TC behavior/patterns.

Introduction – specific comments

Nowhere in the introduction is there any mention of the TC characteristics that will be under consideration in this manuscript. Please add this description.

The introduction (more specifically, the second paragraph, lines 33 – 44) makes it seem like there will also be a focus on ETCs. Please make it explicitly clear you will solely focus on the TCs.

Line 24-25: exposure is not the same as vulnerability

Line 29-30: The amount of destruction is also influenced by changes in exposure and vulnerability

Line 30-31: What do you mean with this sentence?

Line 35: Could you please elaborate on what these differences exactly are?

Line 36 – 38: Please explain to the reader why these cities have less TC-related storm surge extremes (along the lines of ocean waters are colder + more wind shear, so less favorable for TCs), that will also help the reader understand why this Boston example is noteworthy.

Line 38 – 44: Please fill in the gaps that are left in this paragraph: 1) what are the differences in storm dynamics? 2) What are the different characteristics of the flood exceedance curves? 3) What exactly is the frequency of TCs compared to ETCs?  4) How can they cause more damage?

Line 45 – 54: Could you please summarize this in a few sentences? Also: the term “noisy” is very vague. I also feel like the literature is very tailored towards US case studies and misses some other relevant studies (see my earlier comment)

Line 54 – 56: I strongly disagree with the wording here. The authors make it seem like they will overcome the regional scale, but they still perform a regional assessment (namely, the US East Coast).

Line 59 – 60: Perhaps good to also mention coastal complexity here (Bloemendaal et al 2019)

Methods – specific comments

Line 86 – 87: This is quite a bold statement to make, without any additional clarification. How big is the contribution of TC waves to total water levels?

Table 1: Please round the pressure to one decimal place

Line 104: What do you mean with wind intensity? Wind speed? What is the time reference for this wind speed? (1-min, 10-min, 3-sec?) Please also add units with every TC characteristic listed here.

Line 105: Please state the exact dimensions of the “specified distance”

Line 123: Why are you solely testing for linear relationships?

Line 130: Please explain to the reader what these results are

Figure 2: Why are you differentiating between a radius of 250 km (in Figure 2) and 500 km (in the text)? Also, how did you derive the track angle? To me it seems like one of the green tracks for Charleston has a N-NNW angle upon approaching the landfall location, but it is listed as SW.  

Results – specific comments

To me, a 0.5-yr return level of 0.8m seems like a lot. This implies that (assuming the authors correctly identified individual storm surge events) Sandy Hook is affected by TC storm surges of this height on average twice per year! Could you please validate these results against other studies?

Please quantify the statistical significance and correlation throughout the results-section.

A lot of results aren’t shown (indicated by “not shown” in the text). Could you please add these results to the supplementary materials, so that the reader can have a look at these results?

The continuous switching between a 500 km and 250 km radius is highly confusing – please re-read this section carefully and try to homogenize this usage of radii.

Conclusions

Line 329 – 333: Please check the work of Ramos-Valle et al (2020); they synthetically changed details of various storms in the Mid-Atlantic Bight.