Dear Referee,

Thank you very much for your thoughtful and detailed comment. We would like to take this opportunity to address some of your concerns. To begin with, we would like to clarify some overarching issues that are fundamental to the research presented in this paper:

The main objective of this paper is to present a database that serves as a foundational step towards a multi-hazard assessment. We understand the term "multi-hazard" as it is defined by UNDRR (2017): “the selection of multiple major hazards that the country faces, and the specific contexts where hazardous events may occur simultaneously, cascadingly or cumulatively over time, and taking into account the potential interrelated effects”. This definition is also broadly adopted by the Myriad-EU Project, despite some of the limitations it presents—such as the use of terms like “major” and “country” (Gill et al., 2022, p. 26).

It is true that our paper does not yet include an in-depth analysis of interactions between different natural hazards. Such analyses are planned for future stages of our research, using geospatial techniques, Bayesian statistics, and other advanced computational tools, all applied to the dataset we have compiled. 

Nevertheless, the database has been designed with these future analyses in mind. The standardized template already incorporates variables specifically selected to support the study of spatial and temporal relationships between natural hazards—for example, geographical coordinates for each event, as well as precipitation and wind records for meteorologically driven hazards. 

We chose the Garrotxa region as a case study due to its geological and socio-economic context. Being a mountainous area, it is particularly vulnerable to the effects of climate change and the resulting natural hazards—especially meteorological ones. Additionally, this research supports the GarMultiRisk project, in which we are actively involved. One of the project’s main goals is the development of a Natural Risk Reduction and Management Plan targeted at technicians, first responders, and policymakers. For this reason, there is a strong practical interest in conducting this research in that region. It also allows us some flexibility to address not only data analysis but also practical improvements, such as enhancements to the monitoring network, which will help generate more relevant data for assessing hazard interrelationships.

We recognize that various methodological approaches to multi-hazard assessment exist in the current scientific literature. Our ongoing work aims to contribute to this field by offering one such approach and testing its implementation. We deeply value the opportunity for public discussion, as it provides valuable feedback and helps us improve our research.

With these general points outlined, we now respond to your specific comments in the same order as received. We hope our responses clarify our approach and demonstrate our willingness to incorporate improvements in how we communicate our work.

• Use of Terms: “Multi-hazard”, “Multi-risk”, “Natural hazards”

We understand your concerns regarding the use of these terms. There is indeed considerable ambiguity in the scientific literature concerning their definitions and application. Furthermore, one of the objectives of our previous research has been to help clarify these concepts (see López-Saavedra & Martí, 2023). Therefore we will carefully revise our manuscript to ensure consistent and accurate terminology throughout.

As noted earlier, the database presented in this paper is structured around different natural hazards. This structure was chosen to simplify data collection; however, the variables gathered for each event are specifically selected with future analysis of multi-hazard interrelationships in mind. 

In our work, we use the term "multi-risk" with the understanding that multi-hazard assessment is a key component within the broader process of multi-risk assessment—which also includes the evaluation of exposure and vulnerability. In this context, our approach is aligned with the objectives of the GarMultiRisk project, where we aim to provide recommendations and strategies for multi-risk management, grounded in the outcomes of our multi-hazard assessments. As such, the multi-risk perspective represents a future step in the development of our research and is not yet included as a result in this paper. 

• Title

Regarding the title, we chose "Methodological Approach to Multi-Hazard Analysis" as a broader title, as you correctly noted. We view the presented dataset as a foundational step within a wider methodological framework for multi-hazard assessment. Although this particular paper focuses primarily on the development of the database, we felt that the broader title reflects the context and future direction of our work. That said, we are currently considering alternative title formulations to better align with the content of the paper and to improve clarity.

• Abstract

We agree that the abstract would benefit from a more defined structure, with greater focus on the results obtained. As part of the final revision of our manuscript, we will revise the abstract accordingly to ensure it is clearer and more aligned with the content of the paper.

• Introduction

In the introduction, we felt it necessary to contextualize the need for a multi-risk perspective (with multi-hazard assessments being a component of it) by referencing various international frameworks, in alignment with the broader title of our manuscript. However, we agree that this section would benefit from additional scientific citations that specifically address methodological approaches to multi-hazard assessment and the development of datasets that can serve as foundational steps in this process. We are currently working on incorporating relevant literature into the revised version of the manuscript.

The same applies to the section in which we introduce the case of Garrotxa. We intend to include evidence showing that no similar databases currently exist for this region, which would further emphasize the originality and relevance of our work.

Additionally, we recognize the need to clarify our use of the terms extrinsic, intrinsic, and evolutionary variables. In our standardized database, different variables are selected for each natural hazard to support future analyses of their interrelationships. We define intrinsic variables as those inherent to the process itself, extrinsic variables as those external to the process but potentially influential, and evolutionary variables as those that affect the process once it has begun. Although data for some of these variables is currently limited, we believe our approach provides a strong foundation. Moreover, it allows us to offer practical recommendations to policymakers. For instance, if a relevant variable is identified but lacks adequate data, this finding can support the case for expanding or enhancing the monitoring network.

• Section 3 “Natural hazards in the Garrotxa region”, lacks citations

We agree with this point and will include additional scientific support with relevant citations. However, we would like to highlight that a central contribution of our research is addressing the lack of existing studies on natural hazards in the Garrotxa region. As a result, much of the information presented in this section of the paper is based on local knowledge provided by members of our team who are from the area, as well as by local government agencies and environmental organizations—stakeholders with whom we are working closely as part of the GarMultiRisk project, the broader framework within which this research is conducted.

We recognize that this may raise concerns within the scientific community, as some of the information cannot be easily cross-referenced with published literature. We also acknowledge that incorporating data from non-academic sources into scientific research can be challenging, and we will make a clear effort to define the scope and context of our work in the final version of the manuscript.

That said, and as noted at the beginning of this paragraph, we will thoroughly review the section and, where possible, include relevant literature to further substantiate our claims.

• Methodology

To improve transparency in data collection and validation, we are working on the following enhancements to this section:

• Including a summary table of the variables considered in the standardized template.
  
  
  
  
• Adding a summary table of the data sources and data validation workflow used to construct the multi-hazard database for the Garrotxa region.
  
  
  
  
• Providing a more detailed explanation of the citizen science workshops conducted, including how the data was collected and incorporated into the database.
  
  
  
  
• Including more scientific literature on ground subsidence data collection and expert elicitation methods.
  
  
  
  

Where appropriate, we are also open to adding further links or references to scientific literature.

In this regard, we would like to highlight that the paper already includes relevant information in section "4.2 Data Sources and Collection Strategy". This section outlines how information was retrieved from public institutions via formal requests, how geospatial techniques were applied to select wildfire events, and how historical press sources were consulted. We also encourage readers to refer to the "Discussion" section, where some of the challenges encountered during data collection are addressed.

Regarding the acronyms used (e.g., ICGC, AEMET), we believe that we have provided their full names when first mentioned in the text. That said, we will review this section to assess whether additional context would be helpful—such as links to their official websites, more detailed descriptions of their roles, or clarification of their institutional status. Our intention was to avoid excessive detail that might distract from the core content, and we initially considered acronym clarification to be sufficient.

6) Discussion and Challenges

Regarding the sections “Discussion” and “Challenges and Future Steps”, we are committed to reviewing the structure of these parts to ensure clearer distinctions and improve the organization of information. We will also consider the bullet points you mentioned to better highlight the key outcomes of our work.

7) Final comments

Thank you once again for your valuable insights. We look forward to continuing this discussion and improving how we communicate our research so that it can be more useful for the scientific community. 

8) References

Gill, J.C., Duncan, M., Ciurean, R., Smale, L., Stuparu, D., Schlumberger, J, de Ruiter M., Tiggeloven, T., Torresan, S., Gottardo, S., Mysiak, J., Harris, R., Petrescu, E. C., Girard, T., Khazai, B., Claassen, J., Dai, R., Champion, A., Daloz, A. S., … Ward, P. (2022). MYRIAD-EU D1.2 Handbook of Multi-hazard, Multi-Risk Definitions and Concepts. https://doi.org/10.5281/zenodo.7135138 

López-Saavedra, M. & Martí, J. (2023). Reviewing the multi-hazard concept. Application to volcanic islands. Earth-Science Reviews, 236, 104286.  https://doi.org/10.1016/j.earscirev.2022.104286 

United Nations Office for Disaster Risk Reduction (UNDRR). (2017). The Sendai Framework Terminology on Disaster Risk Reduction. "Hazard". Accessed 13 May 2025.  https://www.undrr.org/terminology/hazard 

Dear Referee,

Thank you very much for your extended and detailed review, and for allowing the discussion on our preprint to continue.

In general terms, we would like to emphasize that the construction of the database presented in our work is conceived as a first step towards multi-hazard assessment. This, in turn, is a key component of a broader multi-risk assessment, where exposure and vulnerability come into play.

It is true that our paper does not yet provide a complete workflow for the full implementation of a multi-hazard assessment framework. However, we would kindly invite you to refer to some of our previous work, particularly López-Saavedra & Martí (2023), in which we analyzed the concept of multi-hazard interactions and proposed a multi-risk management framework. While that earlier work was more theoretical in nature, we are now moving into a more practical phase, actively exploring and testing tools that can help us operationalize these ideas. 

The current paper is intended to address this first step, by focusing on the construction of a standardized multi-hazard database, which can serve as a foundation for future analysis of spatial and temporal interrelationships between hazards. Our work is structured around two main contributions:

• To build a standardized database for each of the natural hazards considered in this study—landslides, ground subsidence, floods, wildfires, earthquakes, and rockfalls. To do so, we identified a set of key variables for each hazard type that would allow us, once the data were fully collected, to analyze spatial and temporal relationships.
• To apply this standardized database to the specific case of the Garrotxa region, by collecting all available data on natural hazard events that occurred there over a period of 123 years, in accordance with the variables previously defined.

We would like to highlight that we are already actively engaged in processing and analyzing the data collected for the Garrotxa region. This work employs techniques such as geospatial analysis, Bayesian networks, and graph theory to better understand the spatial and temporal patterns among different hazards. Our intention is to present these results in a second manuscript, which is currently in preparation.

We deliberately separated the development and structuring of the database into its own dedicated paper because we recognize that multiple analytical methods and applications may be used to explore hazard interrelationships. This separation allows us to emphasize the foundational importance of the database itself before addressing subsequent analytical steps in a follow-up publication.

We also acknowledge that the use of the terms "multi-hazard" and "multi-risk" may have caused some confusion in the context of our paper, as we do not explicitly state in the "Introduction" section that this research is part of the GarMultiRisk project. The ultimate aim of this project is to develop a Natural Risk Reduction and Management Plan for the Garrotxa Region, which will be delivered to local technicians, emergency services, and policymakers. At present, our approach to risk assessment remains qualitative, as our team does not yet include specialists in quantitative risk analysis.

That said, we will revise the manuscript carefully to ensure that the terms "multi-hazard" and "multi-risk" are used appropriately and consistently throughout the paper.

We would also like to clarify that our work is part of a broader effort to contribute to the ongoing discussions within the multi-hazard and multi-risk assessment community, particularly in the development of methodological frameworks for such assessments. In this regard, we appreciate the scientific literature you have shared with us. We are familiar with the referenced works and will revise the manuscript to incorporate additional citations where appropriate.

With all this in mind, we now turn to address your specific comments point by point:

1) The title: We are currently exploring alternative title options that explicitly include the keyword "database". The title we originally chose was intended to reflect the future direction and broader objectives of our work.

2) The abstract: We agree that its structure can be improved to better highlight the main results and contributions of this specific paper. Once we complete the revisions to the manuscript, we will rewrite the abstract to provide a clearer and more detailed summary.

3) At the end of the abstract (lines 23-24), the possibility of confronting “similar potential hazard scenarios” is mentioned. This statement is not clear. What does it mean? Moreover, what do you mean exactly by the expression “hazard scenarios”? This expression is not mentioned at all throughout the whole manuscript, except for two brief mentions in chapter 7: We use the phrase “similar potential multi-hazard scenarios” to refer to the context of Garrotxa—a mountainous region that is particularly vulnerable to the effects of climate change and therefore more likely to experience frequent and intense natural hazards, especially those related to meteorological phenomena. In our case, the six natural hazards identified as most relevant are: earthquakes, landslides, floods, wildfires, ground subsidences, and rockfalls. This is why we highlight that our standardised template may be most applicable to other regions with similar characteristics—namely, mountainous areas where these specific hazards are also prevalent—since the structure of our database has been tailored with these six natural hazards in mind.

Regarding the use of the term “scenarios” in Section 7, it refers to risk scenarios, as we begin to outline future steps in our research. In this context, risk scenarios involve assessing where and how natural hazard impacts might occur—for example, by identifying vulnerable populations and critical infrastructure. This type of information is essential for the development and application of targeted mitigation strategies (see line 399).

If the use of the term “scenarios” remains unclear after these explanations, we are fully open to improving the wording in the final revised version of the manuscript.

4) The expressions “multi-hazard” and “multi-risk”: We have already addressed this issue earlier, and we will conduct a thorough review of the manuscript to ensure the correct use of the terms multi-hazard and multi-risk throughout. For example, in line 53, we currently state: “lack of an integrated multirisk database,” and in line 63: “first multirisk database for the Garrotxa region.” The use of the term multi-risk in these instances is incorrect, as the database we present is a multi-hazard one—it does not yet integrate data on impacts to infrastructure, human lives, or other components required for a full risk assessment. We apologize for this oversight. 

5) Introduction, line 35: “In this context, scientific knowledge about the hazards that affect a region, both temporally and spatially, is crucial.” This sentence lacks references: We agree that this sentence would benefit from additional scientific references, and we are familiar with the works of the authors you mention. As noted above, we will revise the manuscript with particular attention to where further citations can strengthen the text. This includes the specific sections you pointed out, such as lines 37–39 and 44–46.

6) Introduction, lines 64-65: What does “intrinsic, extrinsic, and evolutionary variables” mean? We acknowledge that these terms are not fully explained in the current version of the manuscript, and we agree that using alternative wording may have made their meaning clearer.

• By intrinsic variables, we were referring to internal and constitutive characteristics that describe the physical nature of the hazard itself. For example, in the case of an earthquake, the epicenter location.
• By extrinsic variables, we meant external or environmental factors that may influence the impact of the hazard. Continuing with the earthquake example, this could include the local stratigraphy of the area, which—depending on its geomechanical characteristics—can significantly increase the intensity of the ground shaking and associated damage.
• Finally, by evolutionary variables, we were referring to process-related, time-dependent characteristics. In the context of earthquakes, an example would be the distance traveled by the seismic wave, which evolves as the event unfolds.

We are fully willing to include a clearer explanation of these terms in the revised version of the manuscript, particularly in Section 4.3, as you suggested.

7) Section 3 (Natural hazards in the Garrotxa region) completely lacks references: We agree with this comment and will make a concerted effort to provide relevant background scientific literature to support the section. In this regard, we would also like to emphasize that one of the novel aspects of our work lies in the fact that very few studies or analyses related to natural hazards have been conducted in the Garrotxa region.

This section was primarily written based on the local knowledge of geologists and hydrogeologists who live in the area and collaborate with our research group. Although they have accumulated valuable insights through years of observation and photographic documentation, this topic does not fall within their primary research focus, and thus, they have not yet published their findings.

Thanks to the GarMultiRisk project—which has provided funding and broader access to regional experts, particularly those with a more technical background—we have been able to incorporate further observations about hazard dynamics in the Garrotxa region. However, we recognize that integrating such experiential and technical knowledge into a scientific paper poses challenges, especially when it lacks formal publication.

For this reason, we are considering restructuring the manuscript to include a dedicated section that explains how this type of information was gathered. This section could be relocated to the results part of the paper, as it represents a descriptive, non-database-based contribution that still offers valuable context for understanding the region.

That said, we will also make a dedicated effort to include additional references—potentially including grey literature, as you suggested—so that the scientific community can access and assess some of the foundational work carried out in this area.

8) Section 4 (Methodology), lines 175-181: What has been the criterion for the selection of these hazards? This selection needs to be explained and supported better: We understand your concerns. The selection of natural hazards in our study was primarily guided by those identified as major concerns by the Regional City Council, with whom we are closely collaborating within the framework of the GarMultiRisk Project. Initially, we also considered other hazards such as snow avalanches, snowstorms, tornadoes, and downbursts. However, these were ultimately excluded due to their rarity in the region and the practical need to establish clear boundaries for the scope of our study.

Regarding volcanic eruptions, none have occurred in the Garrotxa region within the last 123 years. As a result, it would be difficult to identify meaningful temporal correlations or patterns with the other hazards included in our database, given the disparity in temporal scale.

We are currently working on expanding the manuscript to include more detail on the selection criteria for the hazards considered. Additionally, we plan to include a new “Limitations” section that will clearly outline the scope and boundaries of our work, as well as areas where further research is needed.

9) Section 4 (Methodology), lines 185-188: Which is the source of this information?: As mentioned previously, much of the contextual information on natural hazard occurrences has been gathered through local knowledge and direct observations. In the paper, we also note that for certain hazards—such as landslides and ground subsidences—there was very limited formal data available. To address this, we consulted historical press archives and conducted a citizen science workshop, which helped us gather additional, otherwise unavailable information that has been incorporated into the study.

As stated earlier, we recognize the need to present this type of information more transparently. We will revise the manuscript to make the sources and limitations of this data clearer to the reader, particularly regarding the lack of available scientific literature and the alternative methods used to compensate for that gap.

10) Section 4 (Methodology), lines 254-257: Which stakeholders have been involved? How have the workshops been organised? I invite the authors to provide more details. Yes, we would be happy to include more information about the development of these workshops in the final revised version of the manuscript. The citizen science workshops were open to the entire Garrotxa population, inviting anyone willing to share their memories or experiences related to natural hazard events in the region.

The primary goal was not to collect large volumes of data, but rather to present the existing database—displayed as a map at that stage—and to see if participants could provide more precise or refined information. The workshops naturally evolved into open discussions about the dynamics of wildfires, floods, and other hazards.

Additionally, we invited specific individuals with direct experience or observational expertise, such as local forest rangers and amateur meteorologists who have been independently recording data. Their contributions significantly enriched the workshop outcomes.

11) General comments about “Methodology”: We acknowledge that our paper would benefit from greater transparency and clarification regarding the methodology employed. To address this, we propose the following enhancements:

• Provide a summary table listing the variables considered for each natural hazard in the standardized database template.
• Explain how the database architecture was developed.
• Add a summary table outlining the workflow and data collection process for the various natural hazard events of the Garrotxa region.
• Incorporate relevant references where appropriate to support the methodology.

12)  “Discussion” and “Challenges and Future steps” sections: We agree that the current organization is somewhat confusing. Therefore, we propose the following changes:

• Integrate the challenges within the "Discussion" section.
• Restructure the "Discussion" section to follow the narrative you propose. 
• Add a new “Limitations” section to clearly and accurately outline the various limitations of our approach.
• Develop a “Future Steps” section that details the upcoming objectives and methodologies planned for performing multi-hazard analyses based on the data we have gathered.

13) Final comments:

We believe we have carefully addressed all your concerns and would like to highlight that, based on your feedback, we are actively revising several sections to improve the clarity and accuracy of the paper’s narrative.

As a final comment, we consider our work valuable as it contributes to advancing in a trial-and-error process towards multi-hazard assessment—and, looking ahead, towards multi-risk assessment when viewed in the broader context. We also recognize that the communication of our work can be improved, and we are committed to making those enhancements based on your comments and those of previous reviewers.

We hope our responses have helped clarify the scientific aspects and methodology of our study. Once again, we thank you for your insightful comments and appreciate the opportunity to participate in this open discussion, which we believe enriches the peer-review process and fosters learning.