we'll have to add the sensitivity analysis

should these also be broken into single and repeat flooder panels?

a way to estimate?

we don't take into account severity. same comment as using the word intensity. You could have had a parcel that got carpet-wetting 5 times vs one that got severe damage once or even twice and these would not be comparable motivators to apply for funding.

I actually wonder if this is a caveat we should point out somewhere earlier in the paper. That in this study flood exposure DOES NOT consider flood severity and that many of our once-flooded properties could have experienced very minor flooding.

not sure i like using the word intensity. it came up once before and i flagged it there, too. intensity feels like a measure of depth or velocity and not just exposure. I think exposure count is sufficient context here.

parcels that flooded two or more times (i.e., repeatedly flooded parcels)

for both (instead of separately)

four?

were these centered around the 2010s? We don't really provide much context for when the 18 declarations were.

obtained at and joined to the parcels?

revisit this sentence. I'm not sure about the word "flow"

label every parcel by its?

attribute feels like a weird word choice

Who do we need to include in acknowledgements? Helen since she worked on the mit/app assignments? NC DPS for providing the data?

We should also think about funding acknowledgements. DHS CRC and NOAA RISA. Anything else?

canned statement for NOAA:

This study was supported in part by a grant from the National Oceanic and Atmospheric Administration (NOAA) Climate Program Office Regional Integrated Sciences and Assessments program [NA21OAR4310312]. NOAA had no role in the design or conduct of the study; data analysis or interpretation; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

A bunch of work went into geolocating these and tying them to parcels. Is that part of this paper or a different one?

maybe we should be more explicit about the date range of the apps and mitigations and the PDDs they likely represent?

remove?

are we removing this stage?

insert residential?

?

add fraction in parenthesis? Its like 89% or something

generated by Garcia et al. (2025) using a random-forest model trained on...

Do we need to say anything here about how stacked parcels are handled.

The last highlighted sentence about " a small number of residential parcels are recorded..." how were they identified as residential? because they have a mitigation or application on them or because NC OneMap had them labeled as residential?

I assume we need to include links or citations for each of these that is publicly available.

I wonder if we should just make the statement about study area in the previous section when it is first introduced and stick to unit of analysis for this section

here we are keeping all 5. we may need to figure out how to say in the results that we only plot 4/5

across XX counties in Eastern North Carolina <-- I think we should be more explicit because it looks like we also limited ourselves to complete counties from Helena's flood database (which was watershed boundaries). Just double check.

I feel like this might be worth expanding on and I don't know if we want to use the word marginalized (with a race implication). Maybe here the wealth piece would be more noteworthy given that the funding may have a relatively larger benefit to the applying households than the (non-applying) wealthier households who might be able to afford some level of mitigation themselves, demonstrating the importance of these programs.

Do we want to put in any current context about the FEMA review panel and on-going efforts to modernize/reform FEMA? Might make the piece more timely.

I also wonder if its worth noting that there are other kinds of mitigations that flow to communities (non-household) that we don't consider but might have household-level risk reduction benefits. In a comprehensive flood risk reduction/resilience portfolio, these might also be worth taking into consideration in future work.

but would likely have little impact on the demographics of mitigations that are funded (since applications at the neighborhood level would increase but in-neighborhood differences in funding are negligible) <--- is this true? do i read this correctly. this goes back to my question earlier about "to what end"

reach subsequent stages (in the pipeline)

would this potentially lead to an over count of repetitively flooded properties given that we don't consider if a property was mitigated before a subsequent flood occurred? or is no longer exposed (e.g., were it to have been elevated). Does this need to be articulated?

Prior to this do we want to have any discussion about where we think think these results might differ/whether they can be extrapolated to other places? We've had a lot of conversation about New Jersey (and matching) and I also wonder about Texas. Are there any maps in the other national scale studies that might suggest we would find something different in another state? a la caveats?

do we have a citation for this statement?

property or applicant

what kind of returns are we referring to here? If we have more applications, our award rates might go down. What kind of returns are we hoping for? Just pointing that out here as vague.

never entered? or do not appear in?

Perhaps add in... While application rates are higher for repetitively flooded properties, they are still low (5.6%). Success rates among both single-flooded and repetitive-flooded properties are similar (~60%).

delete?

Do we need to test whether these are statistically significant shifts?

This feels like a discussion sentence rather than a result.

Revisit phrasing.

?

word choice: reside? are within?

I'd make this sentence just about the property value comparison and not the number of surviving properties since that was related to Fig 2.

sometimes nice to include a key takeaway from the figure at the start of the caption

what about the results for the repeat-flooders. We also considered whether the outcomes were different for repetitively flooded properties. Approximately 40%? of flooded properties, flooded more than once. Of these, X% were in communities that had at least one application, X% applied and X% were funded, a rate nearly double that of properties that had only flooded once... (Or something along these lines)

four

We've also heard that matching funds are a big piece here, such that wealthier homes are more likely to be able to provide the match to elevate (in places where local/state government does not), like new jersey. I wonder if there are any citations we could use to support the point. miyuki?

should we try to avoid opinion? remove "and more troubling"

do we need to highlight examples of these programs (perhaps pulling the sentence out of the abstract and putting it here)

HUD?

are we still including all 5 perhaps in supplement? I think your later figures only have 4.

how about... past researchers have observed only who receives funding, not who is eligible or who applies.

since "we" can actually see these things :)

1% flood hazard areas

We should use language that highlights that we created this dataset (and perhaps how) rather than just that we are using it. A large part of the work was the creation of a novel dataset.

Perhaps we should also report the absolute numbers in these datasets. 90,000 flooded properties, X000 applications resulting in x000 buyouts and x000 elevations

We see slightly higher success rates for repeat flooders?

This is a pretty long lead in for an abstract.