Dear Referee

We sincerely thank you for your constructive feedback and insightful questions, which have significantly improved the clarity, transparency, and methodological rigor of our manuscript.

We have carefully considered and addressed each of your comments. 

Below, we provide a point-by-point answer to your questions and suggestions. 

We hope our replies meet your expectations and help clarify the issues.

Please do not hesitate to let us know if further clarification is required.

With kind regards,

Thiago

1) This paper compares and discusses five methods: PMRR, GRS, GAUC, SS, and GHS. At the very least, a brief explanation of each method should be written in order to allow the reader to follow the arguments of this paper. Unless this point is clearly stated, it is difficult to properly review and comment on the results and discussion of this paper. So that I believe that a re-review is necessary.

Thank you for your valuable comments and suggestions. We agree that providing a brief explanation of each federal risk assessment method is essential for ensuring clarity and helping the reader follow the comparisons and discussions presented throughout the manuscript. To address this, we added a new section 2. This section outlines the main objectives, scope, responsible institutions, and typical applications of five federal surveys in Brazil. (L93 - L152 in the revised manuscript).

Other points I noticed are listed below.

2) Introduction: The past knowledge or information on natural disasters in general and those on landslides disasters are mixed together. So that some organization is necessary. For example, it would be possible to first state those on natural disasters in general, and then on specific landslides disasters.

Thank you for your valuable suggestion. We agree that the initial version of the introduction mixed general disaster concepts with specific information on sediment disasters. So, in the first four paragraphs of the revised manuscript, we argue about previous studies and reports on general disaster concepts (L24 - L61 in the revised manuscript). Then, we introduce previous studies and reports on sediment disasters in the fifth to seventh paragraphs (L62 - L92 in the revised manuscript).

3) Line 31: What is “sediment disaster”? Write the definition.

According to the reviewer’s comments, we added the definition of sediment disasters (L62 - L68 in the revised manuscript). As follows: In this study, sediment disasters refer to hazardous natural phenomena resulting from the movement, accumulation, or erosion of soil, rock, or debris materials, typically triggered by gravitational forces and/or hydrometeorological conditions (Uchida et al., 2009). Typical processes that cause sediment disasters include landslides, debris flows, mudslides, rockfalls, and severe soil erosion (Dai et al., 2002; Hungr et al., 2014). Sediment disasters are subject to the complex effects of two factors: natural factors, such as terrain morphology, hydrological regimes, and vegetation cover, and anthropogenic activities, such as road excavations, cut-and-fill operations, unregulated urban sprawl on unstable slopes, and the presence of informal settlements in high-risk zones.

4) Line 76: What is “the core elements”? What kind of elements did the previous studies deal with?

We revised the text to clearly express which core elements are lacking in the current literature (L87 - L90 in the revised manuscript). The final paragraph was rewritten as follows:

“In Brazil, a comparative evaluation of the five federal risk assessment methodologies initiated after 2004, including the PMRR, Geological Risk Survey (GRS), Susceptibility Survey (SS), Geotechnical Aptitude for Urbanization Charts (GAUC), and Geological Hazard Survey (GHS), was conducted. Information was collected and reviewed from official guidelines and their updates (Alheiros, 2006; Brasil, 2007; Bittar, 2014; Pimentel & Dutra, 2018; Lana et al., 2021). Recently, Mendonça et al. (2023) focused exclusively on evaluating the effectiveness of the PMRR. Dias et al. (2021) conducted technical comparisons of various landslide susceptibility mapping methods, including the official SS, and several academic approaches. Rocha et al. (2021) argued the effectiveness of SS, GAUC, and GHS based on the case studies in Nova Friburgo, Rio de Janeiro state. However, no previous studies have undertaken a systematic and comparative analysis encompassing all five federal risk assessment methodologies currently implemented in Brazil. Moreover, the existing literature has not thoroughly examined the methodological components, national coverage, their suitability to inform and support DRR initiatives, and cost per beneficiary. This study seeks to bridge these critical gaps by offering a comprehensive evaluation of each federal survey, identifying methodological deficiencies, and proposing evidence-based improvements to enhance the Brazilian DRR strategies for a more resilient society.”

5) Line 120: …disaster reduction…  --> …disaster risk reduction… ?

Yes. Update to enhance clarity and standardization. Disaster Risk Reduction

6) Line 125: There are two “Local Scale”s.

Removed the repeated local scale. And included the corrected Site-specific scale (< 1: 5,000).

7) Line 126-127: This sentence seems to be very difficult to understand. I would like it to be written more clearly.

Thank you for your observation. We agree that the original sentence required greater clarity and have reformulated the passage to enhance its readability and explicability. The revised section now clearly outlines the categorization of the surveys and the rationale behind the assessment scales (L204–L210 in the revised manuscript). However, we are not entirely sure if we have fully captured the intent of your comment. If the updated version still does not address your concern adequately, we would be grateful if you could kindly provide further clarification so we can make the necessary improvements.

Specifically, the text now reads:

“This study adopted a structured approach to evaluate the applicability of federal surveys across DRR initiatives. The classification framework was based on guidelines from Fell et al. (2008) and Corominas et al. (2013), emphasizing scale-appropriate zoning and risk management practices. Surveys were first grouped into four operational categories—National Scale (< 1: 250,000), Regional Scale (1: 250,000 to 1: 25,000), Local Scale (1: 25,000 to 1: 1,000), and Site-specific scale (< 1: 5,000)—according to their spatial resolution, data availability, intended outputs, and use cases. In parallel, DRR initiatives were assessed based on their functional purpose, governance level, and spatial requirements, using relevance indicators adapted from Corominas et al. (2013) across four scales of analysis. Then, we applied a cross-referential matching process to align initiative demands with survey capabilities. Applicability scores were assigned on a 0–4 ordinal scale, representing levels of relevance from “not applicable” to “fully applicable.” These scores reflect thematic alignment and technical-spatial compatibility. Intermediate values were used when surveys could partially support initiatives beyond their initial scope, capturing contextual flexibility.”

8) Line 143-144: Is “structural measure (SC)" included in the “six non-structural initiatives”? It seems obviously contradictory.

Corrected the sentence. Five non-structural counter measurements (MP, LULOL, SL, EWS) and one structural counter measurement (SC). We revised it as follows:

“These comprise five nonstructural initiatives promoting appropriate land-use policies (MP, LULOL, and SL), management (EWS, ERP), and one structural measure (SC).” (L225 -L226 in the revised manuscript).

9) Line 177: “cost-benefit ratio” --> “cost per beneficiary” ?

Systematically adjusted the terminology related to cost per beneficiary to ensure consistency and standardization.

10) Figure 1.: At the top are figures comparing the number of disasters per state with total area, urban area, population and so on. I would like a clear explanation of how the number of disasters is counted. If 10 landslides occur in one heavy rainfall event, should each be counted as one, or should they be counted as ten? This would likely change the interpretation of the figures.

It is essential to clarify that the disaster counts in this database (Brasil, 2023) are based on officially recognized disaster declarations, not on individual occurrences of phenomena. In other words, a single event recorded in the Atlas may encompass multiple landslides triggered by the same meteorological episode. Therefore, the number of disasters reflects the number of declared emergencies rather than the total number of individual landslide occurrences.

The methods section includes the enhanced version (3.1 Data collection and analysis). (L163 – L166 in the revised manuscript).

“In this database, disasters are recorded based on the issuance of official emergency or disaster declarations, rather than on the count of individual physical phenomena. For example, a single entry may represent one or several landslides that occurred during the same rainfall event. Therefore, the disaster count in this study reflects the number of formally recognized events at the municipal level, not the total number of landslide occurrences.”

11) What is “critical municipalities”? It means the 286 municipalities? Now that the explanation appears in the latter part, you should add some explanation on it before readers see this figure.

We have clarified the definition of "critical municipalities" in Section 3.1 (Data Collection and Analysis), where we explain that this designation is based on federal risk classifications and specify that the total number of municipalities included in our study is 821, based on the most recent available data. (L166 - L171 in the revised manuscript).

“The municipalities highly susceptible to sediment disasters were retrieved from the Ministry of Regional Development—National Secretariat of Civil Protection and Defense (MDR, 2012), which designates these locations as 'critical municipalities' due to their elevated risk levels. Initially, 286 cities were identified under this classification. The number was later expanded to 821 based on updated federal reports. In this study, we utilized the most recent data, comprising 821 critical municipalities, for our analysis. These areas have been prioritized for the implementation of DRR assessment surveys. “

12) Figure 2: Why is this figure just for PMRR, GRS, and SS? Why are not GHS and GAUC shown?

Thank you for raising this important point. The decision to include only PMRR, GRS, and SS in Figure 2 was based on the extent of their municipal coverage across Brazil. While GAUC and GHS are indeed part of our assessment framework, they were excluded from this specific figure due to their limited representativeness. According to Table 1, GAUC and GHS assessments have been conducted in only 17 and 12 municipalities, respectively—a notably small sample size compared to the broader implementation of the other methods.　So, we add this in the text: “GAUC and GHS were excluded from this figure due to the small number of municipalities implemented (17 and 12 respectively).” (L323 – L325 in the revised manuscript).

13) Line 195: “rho” should be written in Greek letter.

The term “rho” has been updated to the corresponding Greek symbol (ρ) in the revised manuscript to ensure proper formatting and consistency with academic standards. (L277 in the revised manuscript).

14) Line 292-293: “plot scale” and “partial plot” How large are they?

We agree that “plot scale” and “partial plot” require further clarification. We added the size of each scale as follows:

“The topographic units used to assess risk vary depending on the purpose of each survey. The SS, GAUC, and GHS conduct catchment (> 10 ha) analyses (Table 1; Fig. 3). In some instances, GHS also conducts plot scale (1 – 100 m2) analyses. On the other hand, the PMRR and GRS employ partial plot (100 – 500 m2) and hillslope (> 500 m2 – 10 ha) examinations.” (L355 – L357 in the revised manuscript).

15) Line 345-347:  If it is written in the literature, the accuracy of the prediction should be evaluated not only in terms of the hit rate but also in terms of the miss rate. The GHS method may have determined in advance that 95% of the collapsed areas were dangerous, but it would be appropriate to also indicate how many slopes were determined to be dangerous but did not collapse.

According to the reviewer’s comment, we revised the original version about the missed rate, which was not directly shown in the previous study. On the other hand, we did find out what range was considered dangerous, so we have added that information as follows. “However, the authors also reported that 47% of areas were classified as hazardous by the GHS, suggesting that there were also many slopes that were deemed dangerous but did not collapse.” (L411 - L413 in the revised manuscript)

16) Line 376: Where is Figure 7?

Thank you for catching this error. The reference to Figure 7 was a mistake introduced during the final editing stage, and it should be referred to as Figure 4.

17) Line 384 – 402: The percentage values in the text cannot be found in Figure 4. Please either add a figure or discuss only the values that can be found in the figure.

Thank you for pointing out this incredible mistake. We have revised the text to ensure that all percentage values correspond directly to those in Figure 4. Additionally, the graphs have been improved to visually emphasize regional divisions through color coding, enhancing interpretability and regional comparison.” (L440 - L475 in the revised manuscript)

“Risk assessment surveys are vital resources for various risk-management initiatives. Therefore, the effectiveness of these surveys can be evaluated by examining the activities and initiatives developed from the basic information provided by them. Figure 4 illustrates the distribution of municipalities across the Brazilian states that have adopted various DRR initiatives, such as master plans (MP), landslide–specific laws (SL), land-use and land-occupation laws (LULOL), early warning systems (EWS), emergency response plans (ERP), and structural countermeasures (SC). The regional distribution of DRR initiatives across Brazilian states reveals notable implementation-level contrasts. First, excluding the Federal District, the implementation of landslide-specific laws (LSL) remains notably low across all states. Only Rio de Janeiro and Pará exceed the 5% threshold, standing out as the exceptions in this category. In the Northern region, most states exhibit relatively low adoption of disaster risk reduction measures. However, Amazonas, Pará, and Amapá present higher percentages in specific regional indicators. Amazonas, for instance, shows considerable efforts in implementing emergency response plans (23%) and early warning systems (11%). Pará demonstrates moderate values across all initiatives, particularly in master plans (13%) and LULOL (12%). Amapá also stands out with 31% of municipalities having ERP and 19% implementing LULOL. In contrast, Roraima and Tocantins register the lowest levels in the region, with most indicators below 5%, and complete absence of early warning systems, emergency plans, and structural countermeasures in Roraima. In the Northeastern region, the implementation pattern is more heterogeneous. States such as Pernambuco and Alagoas lead in most indicators. Pernambuco exhibits significant adoption of master plans (20%), emergency response plans (25%), and structural countermeasures (11%), while Alagoas shows high percentages in early warning systems (14%) and ERP (18%). Other states like Ceará and Bahia demonstrate moderate values across all initiatives. In contrast, Piauí and Paraíba appear among the least engaged in the region, with consistently low percentages for specific plans, early warning systems, and structural countermeasures. In the Midwestern region, results vary significantly. The Federal District represents a clear outlier, reporting 100% implementation for all DRR categories except for structural countermeasures. Mato Grosso do Sul follows with a comparatively high adoption of master plans (13%), LULOL (14%), and EWS (8%). Meanwhile, Mato Grosso and Goiás exhibit limited implementation, with most indicators—particularly EWS, ERP, and SC—remaining below 5%.

The Southeastern region stands out as the most advanced in DRR implementation. Rio de Janeiro and Espírito Santo lead the country, with exceptionally high percentages across nearly all indicators. Rio de Janeiro, for example, reports that 77% of municipalities have ERP, 41% have EWS, and 30% have SC. Espírito Santo shows similar results, including 63% ERP and 26% EWS. São Paulo and Minas Gerais also demonstrate widespread adoption, with São Paulo exceeding 10% in all indicators and Minas Gerais registering 20% for ERP and 18% for SC. In the Southern region, DRR measures are generally well adopted. Paraná shows the highest percentages for master plans (31%) and LULOL (31%) among all regional states. Santa Catarina also performs well, particularly in EWS (16%) and ERP (32%). While displaying lower values than its southern counterpart, Rio Grande do Sul still achieves notable implementation for ERP (30%). Overall, the Southeast and South regions exhibit the highest concentration of municipalities with DRR measures, while the North and Midwestern—excluding the Federal District—tend to lag, with considerable disparities within and between regions.”

18) Line 398: “Santa Catarina leads to… EWS implementations.” What is this sentence based on? I cannot find any evidences in Figure 4 or others.

Thank you for pointing this out. The sentence referring to Santa Catarina as a leader in EWS implementation has been removed from the revised manuscript. This adjustment was made to ensure consistency with the data presented in Figure 4. As explained in our response to Comment 17, the entire paragraph was updated to reflect only the values available in Figure 4 accurately.

19) Figure 6: The correspondence with the six initiatives written in section 2.3.3 is unclear.

Based on the reviewer's comments, we have reconsidered Figure 6. Disaster prevention initiatives are highly diverse, so covering them all in one figure is difficult. In this study, we believe that following the reviewer's comments and organizing the six initiatives used in the previous section will improve the consistency of this manuscript and facilitate readers' understanding. We will, therefore, revise the figure and the contents of the main text accordingly.

20) Line 443: cost-benefit ratio --> cost per beneficiary?

We corrected it. Cost per beneficiary (L527in the revised manuscript)

21) Line 447: $0.0004 per beneficiary …  $0.009 in Table.5 - Which is correct?

Thank you for pointing this out. The correct value is $0.009 per beneficiary, as indicated in Table 5. The discrepancy in the text has been corrected accordingly. (L531 in the revised manuscript)

Dear Reviewer #2,

Thank you very much for your evaluation and for dedicating your time to review our manuscript.

We sincerely appreciate your kind comments and your recognition of the revisions we made in response to Referee #1’s suggestions. These changes have strengthened the manuscript, making it clearer and more robust.

We are grateful for your support and encouraged by your positive feedback.

Finally, we hope that this article encourages a critical and constructive dialogue within Brazil’s scientific and technical community, aiming to strengthen methodological approaches to quantitative risk assessment. By reflecting on existing federal frameworks, our intention is to support their refinement and enhance their effectiveness in disaster risk prevention strategies.

Best regards,

On behalf of all co-authors

Thiago Dutra dos Santos