General comment:

This is the first time that such detailed, micro-scale flood damage data has been made publicly available. The dataset is accessible through the provided links and is clearly described and well-organized in the attached Excel sheets, with separate documentation for both commercial and residential buildings. In its current form, the data can be used in a variety of contexts, including cross-validation of flood damage models, both residential and commercial, for other regions and countries, improving existing models, and identifying overlooked damage mechanisms.

It is noteworthy that several Italian universities have collaborated to develop a common survey methodology and a standardized dataset for post-event flood damage data collection. I hope that the aim is to maintain this effort over time, thereby creating a longitudinal dataset that supports the continuous improvement and adaptation of damage models to reflect the evolving physical and economic vulnerabilities of exposed assets, as well as to enable their validation.

Specific comments:

• Section 2.2: Data collection

Data collection is reported to have started immediately after the event. In contrast, previous studies that relied on post-event data collection typically began 6–8 months later, or even beyond that. This delay was intentional, allowing people time to reconstruct their buildings so that when surveys were conducted, most or all buildings would have been fully reconstructed, enabling a more accurate assessment of the original damage.

This difference in timing should be considered a limitation of the current dataset. Since data were collected immediately, the reported damage may not capture the full cost of damages that become apparent only during or after reconstruction efforts.

• Section 2.4: Technical validation

It is very good that the collected data is also reviewed by an external team, in addition to the original data collection team. However, it is still unclear whether this review is conducted only on the paper forms before they are entered into KoboToolbox, or if it also includes forms that have already been digitized in the platform. If the review is limited to the paper forms, it is important to also double-check the digital entries. In previous studies, reviewing the digitized data has proven useful in identifying additional typing errors.

Minor comments:

1. Introduction and case study 

• Figure 1. Avoid using a yellow dot to represent the municipality of Catarino, as it is very similar to the one representing the economic activities. Please change for another colour.
• Figure 1. We are missing the representation of the three surveyed municipalities within the Misa River basin. Would it be possible to include them in the figure showing the basin? This would help illustrate which part of the basin was surveyed.
• Figure 1. In the legend, specify as in the caption that the economic and residential buildings in the municipalities are the surveyed ones.

2. Methods

In the excel data dictionary:

Economic activities

Form A:

• It is not very clear to me the differences among the 4 variables representing building elevation (ΔQ, hg, h1 and h2).
• It would be valuable for future work to include an additional sediment variable representing large objects (e.g., tanks, cars, rubble from other buildings), as these objects could cause additional damage to building structures upon impact.

Residential buildings

• In the dictionary of the database, it is not very clear the distinction of B, C and D forms, specify there too that B is for the housing unit, C is for the common areas and D for attached buildings.

Form B, C, D:

• This form in the floor section includes a variable ‘damage due to high velocity’ how is this collected? Based on people perception? How do you double check this assumption?

3. Data records

• Figure 3: Since all the forms are connected to Form A in a 1:1 relationship, please indicate the 1:1 connection for Forms C and D in the sketch as well.

After the flood of 2022 in the Marche region, Italy, colleagues from several Italian universities collected data on 123 damaged residential buildings, as well as on 133 affected commercial or industrial premises using standardized forms that have already been proven fit for such a purpose earlier. The paper describes the survey forms and the two datasets that are available online at: https://doi.org/10.5281/zenodo.15591850. Data can be downloaded and processed in Excel.

The joint effort of collecting damage data after such an extreme event and of providing these two unique datasets to the scientific community is noteworthy and much appreciated since it allows further development or validation of flood damage models.

The paper itself is well structured, clearly presented and well written. I  have a few minor suggestions for further improvement:

- Throughout the paper, the authors use the term "business activities". I my view "premises" instead of "activities" would better describe that mostly the physcial items (stocks) at the place of operation (i.e. buildings, equipment, vehicles) were surveyed, not the economic activities in terms of processes or flows.

- Another term that needs some clarification, i.e. a proper definition, and some more explanation on what was collected in the field, is the term "indirect damage". For example, business interruption is not always seen as indirect damage, but as a separate category (e.g. Meyer et al, 2013: https://doi.org/10.5194/nhess-13-1351-2013).

- Line 56/57: "The flood caused extensive damage to buildings and infrastructure and [...]": Do you have some official numbers on the amount of damage? Please add.

- Line 89: "which is tailored for masonry and wooden buildings" should be "... tailored to..."

- Line 100: refer explicitly to Fig. 2.

- Line 112: The weekly discussions and decisions are a bit unclear to me. I thought that the three municipalities were agreed upon at the beginning of the campaign. Later in the paper (e.g. on line 120 and in line 142) it is mentioned that measures were taken to avoid duplications etc. Why was that a risk at all? And were the measures undertaken successful? Please clarify.

- Line 128/129: Obviously, each building was surveyed not by an individual team member alone, but by a whole sub-team with several members having different roles. How many people were involved in one sub-team and would you recommend to keep this?

- Line 259-261: Please add an example.

- Line 285: Is this correct? Rodriguez Castro et al (2025) only report data from Belgium.

- Line 289: What does "a significant portion" mean in terms of numbers?

- Line 295-297: This is an interesting aspect. Do most of the residents only live in the affected area during the spring and summer season? If yes, please add that information earlier.

- At the end of the paper, a brief outlook on future uses of the data and the forms would be nice. Consider shifting lines 281 to 285.

- Figure 1: Since the Figure consists of three parts, all parts should have a number or a letter (A, B, C) and a brief description in the figure caption.

- Figure 2: Check figure caption. Do you mean "tools used" instead of "tool used"?

- Figure 3: The figure is clear and the structure can be found in the datasets. I was wondering whether this nested structure is feasible and ready/easy to use for data analysis. Please comment (in the discussion).

- Figure 5: Please sort items in a descending/ascending order in both figures. And in 5b: The item "Damage due to high velocity" does not match with the other items since it describes a causing process, not a damaged item or activity. Please comment. The answer option "other damage" was chosen very often. Please briefly describe in the text what is included in this category. Would you recommend adding further answer options in future surveys?

- Figure 6b: Only one dataset is shown in Fig. 6b, but two are mentioned in the figure caption. Please clarify.

Thank you for this valuable effort and data.