the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Aug 2025
Recovery under consecutive disasters: how recovery dynamics shape societal resilience
Abstract. Consecutive disasters, where two or more disasters occur in succession before recovery from the first event has been completed, can have non-linear impacts on societies that can surpass the effects isolated events. Drawing on empirical examples and insights from scientific literature, we explore how consecutive disasters affect societal recovery across four interconnected pillars of society: human settlements, human health, economic systems, and socio-political systems. We identify pathways through which repeated disasters can either erode a community's ability to effectively respond to and recover from disasters or provide opportunities for social learning and positive change. By examining both immediate and long-term effects, we show how societies might be pushed towards critical tipping points, resulting in either a systemic breakdown of societal resilience, or transformative adaptation and improved capacity to manage future risks. Recognising these dynamics underscores the need for a long-term, multi-hazard approach to disaster risk reduction. Recovery planning must move beyond short-term, reactive measures toward integrated, forward-looking strategies, supported by reliable, proactive, and equitable financing mechanisms. Addressing the complexity of recovery under consecutive disasters is essential for both research and policy to enable adaptive, resilient societies in a future of increasingly frequent and intense hazards.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(3143 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3143 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
We explored how disasters that occur before recovery from earlier events is complete can complicate recovery, using examples from science and the real world across different societal domains. We then reflect on why recovery under repeated disasters is more complex than simply returning to normal, and how the way we define recovery and the system we study shapes our understanding. These insights can help us better understand, study, and manage recovery from disasters.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-3200', Anonymous Referee #1, 05 Nov 2025
-
AC1: 'Reply on RC1', Sophie L. Buijs, 09 Mar 2026
Response to RC1
Dear authors,
The paper addresses an interesting topic and is generally well written. I agree with the authors that the recovery process has received comparatively little attention in the literature, especially when it comes to consecutive disasters.
[...]The paper is extremely broad[...]
[...]Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
[...]Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations.
We thank the reviewer for their thoughtful and constructive comments. We found the review especially valuable in highlighting that the manuscript’s current framing may create expectations that are misaligned with its intended contribution, and that both its added scientific value and its conceptual nature could be more clearly conveyed to the reader.
In response, we have carefully reassessed the manuscript’s framing. We agree with the reviewer that the paper is best understood as a perspective, conceptual contribution, rather than an analytical study. Accordingly, this paper does not aim to determine at what stage tipping points occur or which pathways are most prevalent. As the reviewer rightly observes, addressing these questions would necessitate more detailed analytical and quantitative investigation beyond the scope of the present study.
We acknowledge that, if the paper is read with the expectation of providing analytical results, the broad range of topics covered may appear problematic in terms of analytical depth. However, we believe that the value of the paper lies in illustrating the complexity of recovery processes in the context of consecutive disasters, and in showing how recovery can be affected in many different ways within and across societal domains, spatial and temporal scales.
To clarify the framing and purpose of the paper, we propose several adjustments to the manuscript, which we will outline below and address in more detail under the respective comments.
- First, we will move away from our framing around the Spaiser et al. (2024) tipping-point framework, as this may have suggested causal identification or quantitative analysis beyond the scope of the manuscript. Instead, more in line with the intended purpose of the empirical examples, we will present the different societal systems as a cross-section of society that illustrates how consecutive disasters can fundamentally increase the complexity of the recovery process across spatial and temporal scales, with potentially interrelated consequences across societal domains. We will clarify that these domains do not represent an exhaustive coverage and that additional (societal) domains could be considered. We believe that the strength of this paper lies in bridging the disciplinary silos. While people might already be aware of some of the complexities, we show how far-reaching the effects can be, within one aspect of society, but also connecting between the different societal domains.
- Second, we propose to revise and restructure Chapters 6-7 into a single reflection chapter that provides a clear conceptual synthesis of recovery complexity in the context of consecutive disasters.
We understand how chapter 6 might have been interpreted as vague if the expectation was that it would provide analytical insights into e.g., which pathways are most likely to lead to tipping points, when tipping points are reached, etc. However, this chapter was intended as a conceptual discussion, highlighting that recovery is not always linear and that a system does necessarily return to a pre-disaster state, particularly under consecutive disasters.
Reassessing the discussion in light of the reviewers’ comments, we also acknowledge that several of the recommendations included in the original manuscript, particularly those related to scientific research, were based on conceptual reflections of the concept of recovery in a consecutive disaster context. However, we see that these implicit conceptual reflections were not made sufficiently explicit, making the recommendations feel insufficiently grounded in relation to the rest of the manuscript
Therefore, rather than focusing on recommendations, the revised chapter will explicitly reflect on how recovery, consecutiveness, and outcomes are shaped by system definition and analytical choices. Integrating Chapters 6 and 7 into one combined new Chapter titled: “Reflecting on the complexity of recovery and consecutive disasters.” This chapter will be structured around three interrelated reflection points that we will outline below.
Reflection point 1: System definition
- This reflection point will address how system definition (i.e., the choice of indicators, societal subsystems, spatial and temporal scales, and the interconnections considered) directly shapes how recovery is defined, how consecutiveness is interpreted, and what recovery outcomes are observed.
- We will reflect on how recovery may appear complete in some subsystems (e.g., physical reconstruction) while remaining incomplete in others (e.g., mental health, governance capacity, or financial buffers), thereby affecting whether subsequent events are classified as consecutive. We will further discuss how incorporating spatial, sectoral, and functional interconnections can reveal forms of consecutiveness that are missed when recovery is assessed for a system or region in isolation. Finally, we will reflect on practical challenges in defining recovery timescales and event boundaries in contexts of prolonged or incomplete recovery, highlighting limitations of existing definitions of consecutiveness and the need for transparent system boundary choices tailored to the research question.
Reflection point 2: Recovery trajectories and “the recovered state”
- This reflection point will integrate the conceptual reflections that were originally embedded in Chapter 6, focusing on the non-linearity in recovery trajectories under consecutive disasters, tipping points, and the definition of “the recovered state”.
- We will discuss how consecutive disasters can increase the likelihood of non-linear responses and the occurrence of threshold effects leading to complex recovery trajectories . Such complex trajectories potentially pushing systems toward alternative, new, or harder to reverse states, a process which may be conceptualized as tipping. Specifically, in this context, recovery does not necessarily imply a return to pre-disaster conditions. We will therefore reflect on how the definition of a “recovered state” is inherently system-dependent, and should encompass non-linear and complex responses, as well as alternative system states. Additionally, we will discuss how what constitutes recovery may differ substantially across systems, such as physical assets, financial systems, governance structures, or social well-being.
- We will also address that while recovery can, in principle, create opportunities for positive transformation, consecutive disasters often inhibit such change by shifting attention and resources from long-term development to short-term response. These reflections highlight why simplified or static recovery assessments are inadequate in consecutive-disaster contexts and motivate the need for more dynamic recovery and risk assessment approaches.
Point 3: Conceptualization of “disasters”
- This reflection point will address conceptual ambiguities surrounding what constitutes a disaster versus an impact or contextual/vulnerability factors, and how this becomes more difficult to separate when adopting an all-hazard, consecutive disaster perspective. This point is added in response to feedback from the second reviewer regarding the framing of disasters stemming from natural hazard events, in relation to e.g., health crises, political instability, and socio-economic disruptions. In addition to this reflection point, a clear definition of how we consider these concepts throughout the paper will be added to the introduction.
Rather than prescribing a single definition or framing, all of these reflections highlight the need for transparency and reflexivity in analytical framing, recognizing that multiple valid interpretations may exist depending on the research question.
We believe these changes will result in a more coherent manuscript with a clearer conceptual contribution, better aligned with the reviewer’s expectations and concerns.
The paper is extremely broad and examines four different societal subsystems: human settlements, human health, the economic system, and the socio-political system. In my view, each of these subsystems is already extremely complex in terms of recovery processes and challenging to address in a single paper, let alone when combining all of them. Moreover, the paper aims to explore the interactions between these systems, considering how they may lead to a deterioration of resilience or to transformative learning. This broad focus inevitably results in a superficial conceptual discussion with rather ad-hoc assembled examples.
We fully acknowledge that each of the four societal subsystems addressed in the paper entail complex recovery dynamics that would require in-depth analysis in dedicated studies to fully untangle them. As explained in the overarching response, the aim of this paper is not to provide detailed subsystem-specific analyses, nor to comprehensively explain recovery processes within each domain. Instead, the purpose of the paper is to illustrate how the consecutiveness of disasters adds multiple layers of complexity to recovery processes across societal domains, and how recovery dynamics in different subsystems can interact across spatial and temporal scales. We hope that the revised framing of the manuscript, which explicitly clarifies its conceptual nature and intended contribution, will help to better understand the added value of this high-level perspective and to address concerns regarding the paper’s broad focus.
This issue is also reflected in Figure 1, which essentially indicates that recovery trajectories are diverse, unknown, and involve interactions. Overall, I question whether this approach truly provides many new insights.
In line with the observation of the reviewer, the figure indeed aims to convey the complexity in recovery trajectories across societal domains, and that interactions are involved. We believe that through the adjusted framing of the paper, clarifying its conceptual nature, the added value of the figure becomes clearer to the reader. It is a visual aid that helps the reader to understand the different dimensions of complexity that are involved and that will be illustrated through examples in the paper. We further will in addition improve the figure based on the concrete suggestion from RC2.
In this regard, while the authors accurately describe the examples as empirical, I would consider them anecdotal evidence to support the conceptual points the authors wish to convey. Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
We agree with the reviewer that the paper has a conceptual nature that should be conveyed more clearly to the reader. As outlined in the overarching response, we will therefore 1) adjust the framing of the paper in the abstract and introduction to clarify that the purpose of the paper is to illustrate and discuss the different dimensions of complexities of the recovery process involved and highlighted in the context of consecutive disasters, and 2) rewrite the discussion section to be a more conceptual discussion of the complexities involved in recovery dynamics in a consecutive disaster context.
In line with this reframing, we will also revise the title of the manuscript to avoid misleading expectations, to “A perspective on the complexity of recovery dynamics in the context of consecutive disasters” or something similar.
With respect to the reviewer’s comment on the empirical examples, we acknowledge that these should be understood as illustrative rather than systematic evidence. The examples in Chapters 2–5 are used non-exhaustively to support and ground the conceptual discussion, illustrating the diversity of recovery dynamics and pathways that can arise under consecutive disasters. The pathways presented were developed through iterative interdisciplinary discussions among the co-authors, informed by both the empirical examples and the existing literature, rather than being pre-defined. We expect that the revised framing will make the role of these examples clearer and align expectations accordingly.
Although the paper largely focuses on the potential negative impacts of consecutive disasters, considerable evidence suggests that disasters can serve as "windows of opportunity" leading to transformative learning (see e.g., Kreibich et al. 2017). This is rather briefly addressed in sections 5.3 and 4.2, but many more examples could be provided for each of the subsystems discussed. For instance, in Section 2, I am sure you can find numerous examples where disasters have led to revised spatial planning policies and/or building codes, significantly reducing the impact of subsequent disasters.
We agree with the reviewer that the paper includes more negative rather than positive examples and that there are more positive effects on transformative learning that can take place after a disaster. However, based on a comment provided by the second reviewer, we have decided to take the opportunity to strengthen the consecutive event focus of our paper. After careful consideration, we recognise that some of the pathways and examples that were provided in the first version of the manuscript, particularly those related to positive change, are actually beyond the scope of consecutive disasters, addressing instances where recovery was able to finish between events (e.g., Section 5.3 on societal learning and adaptation action).
With this refined focus, we find limited evidence of positive transformation occurring during unfinished recovery phases (e.g., in the example of spatial planning policies and/or building codes, while the measures are designed and implemented already during the recovery process, they only become relevant in practice once you have rebuilt in a new location or using improved building guidelines, so after recovery is done). Instead, consecutive disasters often appear to inhibit positive change by redirecting attention and resources from long-term planning to short-term response.
We will go over the text in detail to sharpen our consecutive focus, which will result in the removal of some examples of positive change that are now present in Chapters 2-5, such as section 5.3, and integrate “consecutive disasters as inhibitors of positive change” into reflection point 2 as described above.
Section 6 also remains vague, offering little insight into what leads to tipping points in terms of deterioration or enhanced learning. The conclusion that these are linked to disaster intensity and frequency seems quite evident based on existing literature and cannot be derived directly from the empirical approach provided in the paper. We gain limited understanding of which discussed pathways are important or common.
This concern is addressed under Revision II in the overarching response. In short, identifying which pathways are most important or common, or determining the conditions under which tipping points occur, was not the intended aim of this paper, as such questions would require dedicated analytical and quantitative studies beyond the scope of this conceptual contribution. In the revision, Section 6 will be integrated into the new reflection chapter and explicitly reframed as a conceptual discussion of non-linear recovery trajectories and alternative recovered states under consecutive disasters, clarifying what insights this section intended to provide.
Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations. For example, the authors state, “…to support efficient and equitable recovery after disasters, countries need more reliable and proactive financing solutions. These may also include pre-arranged recovery financing mechanisms such as forecast-based financing, where funds are automatically released for humanitarian actions that are agreed upon in advance (IFRC & RCCC, 2020) or parametric insurance, which offers rapid, flexible payouts based on pre-defined parameters, such as certain rainfall or wind speed, ensuring rapid payments in post-disaster settings (Ocampo & Moreira, 2024)." (lines 631ff). While I do not disagree with the recommendation itself, I question how this particular suggestion is derived from section 4, and why it was chosen as one of four key recommendations among potentially many others?
We thank the reviewer for this comment and agree that several recommendations in the discussion section may appear insufficiently grounded in the preceding sections. These recommendations initially aimed to provide potential avenues forward that could address the challenges we identified across societal domains in Chapters 2‑5. However, we acknowledge that selecting specific policy instruments (such as particular financing mechanisms) can give the impression of prioritization that goes beyond the conceptual scope of the paper.
As outlined in the overarching response, to improve coherence and better align the manuscript with its conceptual and perspective-oriented contribution, we will move away from a recommendation-focused discussion. Instead, Chapters 6 and 7 will be integrated and revised into one reflection chapter that explicitly synthesizes conceptual insights on recovery complexity under consecutive disasters. In this process, policy-oriented recommendations will be removed, as they do not fit well within the revised framing.
Citation: https://doi.org/10.5194/egusphere-2025-3200-AC1 -
AC3: 'Reply on RC1 - pdf version', Sophie L. Buijs, 09 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3200/egusphere-2025-3200-AC3-supplement.pdf
-
AC1: 'Reply on RC1', Sophie L. Buijs, 09 Mar 2026
-
RC2: 'Comment on egusphere-2025-3200', Anonymous Referee #2, 06 Jan 2026
General Comments
Dear Authors,
Thank you for this very interesting and timely manuscript. The paper addresses an important and still underexplored gap in the multi-hazard and multi-risk literature, an area that is rapidly expanding. I believe this work represents a valuable contribution and has the potential to become an influential reference in the field.
My main overarching suggestion is to strengthen and clarify the paper’s focus on consecutive disaster recovery. While I provide a detailed set of comments below, these are intended to be constructive rather than discouraging. On the contrary, I believe the authors have conducted a valuable and insightful study, and that further clarification and refinement would significantly enhance its impact and readability.
- Introduction
The introduction opens with a real-world example, and the manuscript as a whole makes extensive and effective use of empirical illustrations. Given this, I would suggest that the introduction focus less on additional examples and instead prioritize clearly framing the problem, introducing key definitions, and establishing the conceptual focus of the paper.
In particular, the scope of the paper would benefit from greater clarification. For example, in line 74 the authors state: “Even when recovery is completed between events, decisions made during the recovery process… can fundamentally shape a society’s resilience to subsequent events.” At this point, it is unclear whether the paper focuses specifically on consecutive disaster recovery, recovery processes more generally, or whether the concept of “consecutive disaster” itself still requires clearer definition. I encourage the authors to explicitly clarify this distinction early in the paper.
Following this, the subsystems proposed by Spaiser et al. (2024) are introduced, and a systemic definition is developed. However, the rationale behind adopting and adapting this framework remains unclear. I recommend explaining in greater detail why this particular framework was selected, how it relates to other existing approaches, and how the authors arrived at the proposed system definition.
Additionally, the introduction refers to “health crises” alongside “disasters.” It would be helpful to clarify whether health crises are conceptualized as disasters within the framework of this paper, or whether they are treated as distinct phenomena.
Regarding Figure 1, since uncertainty in recovery outcomes is a central theme of the paper, it is unclear why uncertainty is not visually represented in the recovery phase following the first event. As currently presented, the figure seems to associate uncertainty only with the recovery from the second event.
Between lines 90 and 100, the authors introduce the concept of societal recovery. It would be beneficial to clearly define this term and explain how it differs from “recovery” more generally. Furthermore, I recommend clarifying the criteria and process used for selecting the recovery events and cases discussed in the paper. Although the authors note that this is not a systematic review, a more transparent methodological explanation would add rigor and clarity.
Finally, briefly outlining the key issues addressed in each section at the end of the introduction with their related section numbers could greatly improve readability and guide the reader through the manuscript.
- Impacts within Human Settlements
In the opening of this section, human settlements and critical systems are defined; however, health is not included among the listed critical systems unlike in section 2.3. This omission raises questions and should be either justified or clarified.
I also suggest shortening the sentence in lines 105–107 for clarity and readability.
In the first pathway, the authors discuss increased exposure and vulnerability through displacement and relocation. However, the connection between this argument and the discussion in lines 114–117 is not entirely clear and could be strengthened.
In lines 139–144, the manuscript refers to the lack of critical services such as WASH infrastructure. Here, it would be useful to clarify whether health-related impacts are considered disasters in their own right, and whether this example is intended to illustrate a consecutive disaster. If WASH is considered part of the health subsystem, human settlement subsystem, or both, this should be explicitly stated.
In Section 2.3, healthcare is identified as a critical system. This highlights the importance of clearly defining what constitutes a “critical system” and how these systems relate to the societal subsystems introduced earlier. Such clarification would also help explain why food systems are treated as critical systems rather than as considering ecosystems as critical system or why the issue of the food is not addressed economy societal subsystem section.
In Section 2.4, it is unclear whether the focus is on resource shortages, unworkable environments, or response capacity. I suggest maintaining a clear focus on one dimension and addressing others in relevant subsections. For example, maybe resource depletion could be discussed more comprehensively within the economic system section.
The statement in lines 184–185 regarding emergency systems remaining functional also requires clarification. If emergency workers are unable to perform their duties, it is difficult to consider the system functional.
Sections 2.5 and 2.6 appear to address similar concepts related to “Build Back Better” and could potentially be merged to reduce redundancy.
In line 198, the discussion refers to situations where recovery is completed, which may fall outside the paper’s stated focus on consecutive disasters. Some examples in Section 2.6 similarly refer to completed recovery and may introduce ambiguity.
- Health-Related Pathways
Section 3 presents four pathways, three of which focus on outbreaks, epidemics, and pandemics, and one on mental health impacts. In my view, these topics could be more effectively integrated into earlier sections. For example, health-related impacts could be framed as disasters shaping a consecutive disaster environment, with health systems treated consistently as critical systems. Changes in working environments due to outbreaks could also be discussed within Section 2.4.
Please consider this as a suggestion. In my view, this section could be structured differently and, rather than being framed around the health system, it could be presented with a stronger focus on human health as shown in figure 1. I recognize that this would require substantial changes to the overall structure of the paper; however, I encourage the authors to reflect on this possibility and to consider alternative conceptual frameworks beyond the current organization around four societal subsystems. There is also an easy solution to partiality address this confusion: just change the title of the section to impact on the human health! and keep it consistent through the section.
- Economic System
Section 4 is, in my opinion, one of the strongest parts of the paper. One aspect that could be further strengthened is a clearer definition of economic recovery. Specifically, what does it mean for an economic system to be considered “recovered”?
Section 4.4 raises important questions about the scale of analysis. I suggest explicitly incorporating scale into the definition of consecutive disasters, clarifying at which spatial and temporal scales events are considered consecutive.
In Section 4.5, the relevance to consecutive disaster recovery is less evident and may require further justification or reframing.
- Socio-Political System
A key issue in this section is whether political instability is conceptualized as a disaster. This point should be clarified explicitly. If political instability is considered a disaster, then it would be useful to apply a similar logic across other societal subsystems (e.g., economic instability as a disaster interacting with other hazards).
I recommend revisiting page 18 with a strong focus on consecutive disaster recovery dynamics. Additionally, the issue mentioned in line 474 would benefit from a concrete example.
Finally, the term “socio-political system” may be too broad to capture the concepts discussed in this section. Alternative framings such as disaster risk governance, institutional capacity, or disaster risk management may be more precise.
Section 5.3, in particular, appears only loosely connected to consecutive disaster recovery dynamics. The ideas presented here may be more effectively integrated into the next sections like in section ragrding the tipping point that leads to increase societal resilience.
- Implications
Section 6 has the potential to be a core contribution of the paper; however, in its current form it reads more as a summary of previous sections. I encourage the authors to deepen the analytical discussion, particularly by strengthening the link between their findings and the framework proposed by Spaiser et al. (2024).
Specifically, clarifying what constitutes tipping points within each societal subsystem and what is meant by “system transformation” would greatly enhance the paper’s theoretical contribution. Examples discussed earlier in the paper, or new illustrative cases, could be used to demonstrate how societal resilience may be improved.
Finally, in line 557 migration is described as a negative feedback loop. I would caution that migration is not necessarily negative and can also be understood as an adaptive strategy, particularly for populations facing adverse or uninhabitable conditions.
Citation: https://doi.org/10.5194/egusphere-2025-3200-RC2 -
AC2: 'Reply on RC2', Sophie L. Buijs, 09 Mar 2026
Response to RC2
- General Comments
Dear Authors,
Thank you for this very interesting and timely manuscript. The paper addresses an important and still underexplored gap in the multi-hazard and multi-risk literature, an area that is rapidly expanding. I believe this work represents a valuable contribution and has the potential to become an influential reference in the field.
My main overarching suggestion is to strengthen and clarify the paper’s focus on consecutive disaster recovery. While I provide a detailed set of comments below, these are intended to be constructive rather than discouraging. On the contrary, I believe the authors have conducted a valuable and insightful study, and that further clarification and refinement would significantly enhance its impact and readability.
We would like to thank the reviewer for the thorough review and constructive feedback. We greatly appreciate the encouraging words and the recognition of the manuscript’s relevance. Based on the detailed comments provided by the reviewer, we have carefully reassessed the manuscript and propose several substantial revisions, as well as smaller changes, that, together, will help to refine the framing, scope, clarity, and conceptual contribution of the paper. Overarching adjustments are outlined below, with individual comments addressed in more detail under the respective points.
I. We agree that the manuscript would benefit from a more explicit and consistent focus on consecutive disasters, avoiding references to more general recurrent disaster contexts. We will therefore revise the manuscript to strengthen the consecutive-disaster focus throughout, removing or reframing examples that assume completed recovery between events and establishing a clear framing from the start. We will take specific care to reevaluate the sections explicitly highlighted by the reviewer.
II. After reassessing the conceptual focus of the paper, based on the reviewers feedback, we recognise that the main value of the paper lies in illustrating the complexity of recovery processes in the context of consecutive disasters, and in showing how recovery can be affected in many different ways within and across societal domains, spatial and temporal scales. The paper therefore is best understood as a perspective, conceptual contribution.
In this context, we chose to frame our manuscript around the tipping point framework of Spaiser et al. (2024), as we recognised the tipping point literature and framing to be relevant in the context of consecutive disasters specifically. The tipping literature was intended to be a conceptual lens to reflect on the possibility that recovery under consecutive disasters may not return to a pre-disaster state and may instead lead to alternative system trajectories. We see how the strong framing around the tipping-point framework has created the suggestion of more quantitative and concrete statements regarding a definition of the identification of tipping thresholds within each subsystem, which tipping points are common, etc. that cannot be provided using the illustrative examples presented in Chapters 2 through 5.
Therefore, more in line with the intended purpose of the empirical examples, we will move away from the Spaiser et al. (2024) framework and present the different societal systems as a cross-section of society that illustrates how consecutive disasters can fundamentally increase the complexity of the recovery process across spatial and temporal scales, with potentially interrelated consequences across societal domains. We will also clarify that these domains do not represent an exhaustive coverage and that additional (societal) domains could be considered.
III. Related to this, we propose to revise and restructure Chapters 6-7 into a single reflection chapter that provides a clear conceptual synthesis of recovery complexity in the context of consecutive disasters. Rather than focusing on recommendations, the revised chapter will explicitly reflect on how recovery, consecutiveness and outcomes are shaped by system definition and analytical choices. This chapter will be structured around three interrelated reflection points that we will outline below.
Reflection point 1: System definition
- This reflection point will address how system definition (i.e., the choice of indicators, societal subsystems, spatial and temporal scales, and the interconnections considered) directly shapes how recovery is defined, how consecutiveness is interpreted, and what recovery outcomes are observed.
- We will reflect on how recovery may appear complete in some subsystems (e.g., physical reconstruction) while remaining incomplete in others (e.g., mental health, governance capacity, or financial buffers), thereby affecting whether subsequent events are classified as consecutive. We will further discuss how incorporating spatial, sectoral, and functional interconnections can reveal forms of consecutiveness that are missed when recovery is assessed for a system or region in isolation. Finally, we will reflect on practical challenges in defining recovery timescales and event boundaries in contexts of prolonged or incomplete recovery, highlighting limitations of existing definitions of consecutiveness and the need for transparent system boundary choices tailored to the research question.
Reflection point 2: Recovery trajectories and “the recovered state”
- This reflection point will integrate the conceptual reflections that were originally embedded in Chapter 6, focusing on the non-linearity in recovery trajectories under consecutive disasters, tipping points, and the definition of “the recovered state”.
- We will discuss how consecutive disasters can increase the likelihood of non-linear responses and the occurrence of threshold effects leading to complex recovery trajectories . Such complex trajectories potentially pushing systems toward alternative, new, or harder to reverse states, a process which may be conceptualized as tipping. Specifically, in this context, recovery does not necessarily imply a return to pre-disaster conditions. We will therefore reflect on how the definition of a “recovered state” is inherently system-dependent, and should encompass non-linear and complex responses, as well as alternative system states. Additionally, we will discuss how what constitutes recovery may differ substantially across systems, such as physical assets, financial systems, governance structures, or social well-being.
- We will also address that while recovery can, in principle, create opportunities for positive transformation, consecutive disasters often inhibit such change by shifting attention and resources from long-term development to short-term response. These reflections highlight why simplified or static recovery assessments are inadequate in consecutive-disaster contexts and motivate the need for more dynamic recovery and risk assessment approaches.
Reflection point 3: Conceptualization of “disasters”
- This reflection point will address conceptual ambiguities surrounding what constitutes a disaster versus an impact or contextual/vulnerability factors, and how this becomes more difficult to separate when adopting an all-hazard, consecutive disaster perspective.
Rather than prescribing a single definition or framing, all of these reflections highlight the need for transparency and reflexivity in analytical framing, recognizing that multiple valid interpretations may exist depending on the research question.
IV. Lastly, we will address the reviewers observations regarding unclarity of some of the terminology used within the paper.
- As described, we will provide more consistent use and framing around the provided definition for consecutive disasters.
- We will add a clear definition of critical systems and services early in the paper, making sure that these definitions are then applied consistently throughout the manuscript.
- At the start of the paper, we will add a section clarifying how economic shocks, health crises and political instability are conceptualised in relation to the concept of “disaster.” Following the all-hazards approach adopted in the UNDRR Hazard Information Profiles (HIPs), phenomena such as political and economic instability can be classified as societal hazards, while infectious diseases are classified as biological hazards. Disasters can then result from the interaction of any hazard type with exposure and vulnerability (H × E × V). In principle, therefore, large-scale health crises, financial shocks or political crises may constitute disasters in their own right. At the same time, we recognize that in much of the disaster risk reduction literature, the term “disaster” is commonly used to refer primarily to events triggered by natural hazards (e.g., meteorological, climatological or geological hazards). To discuss the interaction between natural-hazard-induced disasters in this more traditional sense and health, economic or political crises in a consistent and transparent manner, we will use the term “disaster” in this paper primarily to refer to natural-hazard-induced disasters, while explicitly referring to economic crises, political crises and health crises as distinct phenomena. This while acknowledging that these phenomena can, under an all-hazards perspective, constitute disasters in their own right and may therefore create consecutive-disaster dynamics when their impacts overlap with those of natural-hazard-induced events.
- We will also add a clarifying statement regarding the definition of “societal recovery” in the introduction, explaining that the addition of the word “societal” meant to limit the scope to human systems, which also clarifies the question as to why ecosystem recovery was not included in the review.
- We will add some lines on how the “recovered state” can be defined or conceptualised within each of the different societal domains, yet we do not aim to provide one binding definition for this (as is addressed in reflection point 2).
We believe these changes, made based on the reviewers' elaborate feedback, will result in a more coherent manuscript with a clearer conceptual contribution.
Introduction
The introduction opens with a real-world example, and the manuscript as a whole makes extensive and effective use of empirical illustrations. Given this, I would suggest that the introduction focus less on additional examples and instead prioritize clearly framing the problem, introducing key definitions, and establishing the conceptual focus of the paper.
We thank the reviewer for this suggestion and agree that it would be better to take out the empirical example from the introduction, to instead prioritize a clear framing and scope of the paper, introduction of definitions, and clarification of the conceptual focus as outlined in the overarching response.
In particular, the scope of the paper would benefit from greater clarification. For example, in line 74 the authors state: “Even when recovery is completed between events, decisions made during the recovery process… can fundamentally shape a society’s resilience to subsequent events.” At this point, it is unclear whether the paper focuses specifically on consecutive disaster recovery, recovery processes more generally, or whether the concept of “consecutive disaster” itself still requires clearer definition. I encourage the authors to explicitly clarify this distinction early in the paper.
We agree with the reviewer that the statement in line 74 is not sufficiently specific for consecutive disasters. As outlined in revision I, we will refine the manuscript to limit its focus more strictly to consecutive events, following the original definition provided in the first version of the manuscript.
Following this, the subsystems proposed by Spaiser et al. (2024) are introduced, and a systemic definition is developed. However, the rationale behind adopting and adapting this framework remains unclear. I recommend explaining in greater detail why this particular framework was selected, how it relates to other existing approaches, and how the authors arrived at the proposed system definition.
We thank the reviewer for pointing out this ambiguity. As outlined in revision II, we have reconsidered the role of the Spaiser et al. (2024) framework in the manuscript. While it was initially introduced because we considered tipping-point theory conceptually relevant to recovery under consecutive disasters, we recognize that its inclusion as an organizing structure was not sufficiently justified and may have created expectations that extend beyond the conceptual and illustrative contribution of this work.
We will therefore move away from this framework and instead present the selected societal domains as a cross-section of societally relevant subsystems, in line with the examples’ intended purpose, which was to illustrate how consecutive disasters add complexity to recovery processes across interconnected societal domains.
Additionally, the introduction refers to “health crises” alongside “disasters.” It would be helpful to clarify whether health crises are conceptualized as disasters within the framework of this paper, or whether they are treated as distinct phenomena.
We thank the reviewer for raising this point. As outlined in revision IV, we will add a section to the introduction to address this conceptualisation explicitly. In the HIPS, infectious diseases like cholera and tuberculosis are conceptualised as biological hazards, which means that a large disease outbreak can be conceptualised as a disaster.
At the same time, we emphasize that such classifications are not always clear-cut, particularly in the context of consecutive disasters. Health crises, political instability, and economic shocks may be framed as independent disasters, as impacts resulting from prior disasters, or as contextual conditions or vulnerability factors shaping subsequent recovery trajectories. We will address these blurred boundaries in reflection point 3 of the new reflection chapter.
Regarding Figure 1, since uncertainty in recovery outcomes is a central theme of the paper, it is unclear why uncertainty is not visually represented in the recovery phase following the first event. As currently presented, the figure seems to associate uncertainty only with the recovery from the second event.
We suggest adjusting figure 1 and/or its caption to address the concerns about the additional dimensions of recovery that are currently not visually represented. While we initially chose to limit the visual representation of uncertainty to the second recovery phase, as this is what the paper focuses on specifically, we will clarify either visually or through text that the recovery uncertainty is also relevant to the initial recovery process.
Additionally, we will add a fifth grey box to represent additional societal, or non-human systems, as we show only a cross-section of possible domains that can be included. This will help clarify to the readers that the chosen domains are non-exhaustive.
Between lines 90 and 100, the authors introduce the concept of societal recovery. It would be beneficial to clearly define this term and explain how it differs from “recovery” more generally.
We thank the reviewer for this helpful suggestion. As outlined in revision IV, we will explicitly explain in the introduction that the term “societal” is added before recovery simply to delimit the scope of the paper to various recovery processes within human systems, as similar dynamics are relevant in non-human systems as well but outside the scope of this paper.
Furthermore, I recommend clarifying the criteria and process used for selecting the recovery events and cases discussed in the paper. Although the authors note that this is not a systematic review, a more transparent methodological explanation would add rigor and clarity.
We agree with the reviewer that it would be good to clarify our methodological approach. The initial pathways were identified based on expert input through multiple rounds of discussion among the authors, all of whom have research experience relevant to the topic. These preliminary pathways were subsequently refined through iterative rounds of targeted literature searches and collective discussion. As new insights emerged from the literature, additional iterations were undertaken, leading to the final set of illustrative pathways. We will describe this expert-driven and literature-informed iterative process more explicitly in the revised manuscript to enhance methodological clarity.
Finally, briefly outlining the key issues addressed in each section at the end of the introduction with their related section numbers could greatly improve readability and guide the reader through the manuscript.
We thank the referee for this suggestion and will add a brief description of each section at the of the introduction that will help the reader to navigate the dense content of our manuscript.
- Impacts within Human Settlements
In the opening of this section, human settlements and critical systems are defined; however, health is not included among the listed critical systems unlike in section 2.3. This omission raises questions and should be either justified or clarified.
We thank the reviewer for noting this point. As outlined in revision IV, we will introduce a clearer and more consistent definition of “critical systems” early in the manuscript, including how these systems relate to the concept of “critical services” and to the societal domains analysed in the paper. In this context, healthcare systems (e.g., hospitals) and WASH infrastructure are recognised as critical systems that are relevant both within the human settlements domain and for human health.
While the list of critical systems at the start of Section 2 was not intended to be exhaustive, we agree that the omission of healthcare systems may create unnecessary confusion, particularly as human health is explicitly treated as a societal domain in the paper. We will therefore revise the opening of the human settlements section to explicitly include healthcare systems among the relevant critical systems, thereby clarifying how these concepts are integrated and interconnected across domains.
I also suggest shortening the sentence in lines 105–107 for clarity and readability.
We thank the reviewer for this suggestion and will revise the sentence in lines 105-107.
In the first pathway, the authors discuss increased exposure and vulnerability through displacement and relocation. However, the connection between this argument and the discussion in lines 114–117 is not entirely clear and could be strengthened.
We appreciate this suggestion and will strengthen our argumentation in this section.
In lines 139–144, the manuscript refers to the lack of critical services such as WASH infrastructure. Here, it would be useful to clarify whether health-related impacts are considered disasters in their own right, and whether this example is intended to illustrate a consecutive disaster. If WASH is considered part of the health subsystem, human settlement subsystem, or both, this should be explicitly stated.
We thank the reviewer for this important clarification request. As outlined in revision IV.2, we will introduce a clearer definition of critical systems and their relation to the societal domains discussed in the paper. In this context, WASH infrastructure will be defined as a critical system that is of relevance both within the human settlements and human health domains.
Regarding the conceptualisation of health-related impacts as potential disasters and whether the example intends to illustrate a consecutive disaster, we will revisit the text to ensure consistency with the definitions of disasters as we will introduce as outlined in revision IV.3. Following this conceptualisation and the hazard definitions as provided in the HIPS, we believe that a lack of access to critical services is not to be considered a hazard, but rather a vulnerability factor, increasing the vulnerability to biological hazards such as cholera. An outbreak resulting from this interaction (between the presence of a hazard and heightened vulnerability, for an exposed population) could, within the conceptualisation as defined in IV.3 then be considered a disaster (in the manuscript explicitly referred to as health emergency, to distinguish from natural-hazard caused disasters). We will revise the paragraph accordingly to make explicit that in this case absence of WASH services is conceptualised as a vulnerability driver within the recovery process, while a subsequent disease outbreak may constitute a consecutive disaster. As we discuss also in the reflection chapter we believe that conceptualisations like this are not set in stone and can be done differently depending on your research goal, but we agree that given the purpose, scale, and boundaries of the system of the study it is important that provided definitions are coherent.
In Section 2.3, healthcare is identified as a critical system. This highlights the importance of clearly defining what constitutes a “critical system” and how these systems relate to the societal subsystems introduced earlier. Such clarification would also help explain why food systems are treated as critical systems rather than as considering ecosystems as critical system or why the issue of the food is not addressed economy societal subsystem section.
As outlined in revision IV.2, we will introduce an explicit definition of critical systems early in the paper. In the revised manuscript, critical systems will be defined as functional systems that provide essential services necessary for the maintenance of societal functioning and well-being (critical services such as healthcare, WASH, food, energy, transport). These systems are not conceptualised as separate societal subsystems, but rather as cross-cutting infrastructures that are essential for the functioning of different societal domains.
The critical systems discussed in the manuscript are not an exhaustive list, but rather serve as illustrative examples that help demonstrate how disruptions to essential services can influence multiple societal domains simultaneously. Other systems may equally qualify as critical depending on the specific research context, scale, or hazard type under consideration. While we address the issue of breadbasket failures in the economy social subsystem section, we will make sure to clarify the link to the food system.
With regard to the ecosystems mentioned by the reviewer, we will clarify that the scope of the paper is explicitly focusing on societal recovery within human systems. While ecosystems evidently constitute a network of critical systems in themselves for society, the study of their recovery processes in regards to consecutive events represents an entire distinct field of study. We will explicitly discuss this crucial complexity in the outlook of the paper.
In Section 2.4, it is unclear whether the focus is on resource shortages, unworkable environments, or response capacity. I suggest maintaining a clear focus on one dimension and addressing others in relevant subsections. For example, maybe resource depletion could be discussed more comprehensively within the economic system section.
We thank the reviewer for this suggestion and we will revisit this section to ensure a clear focus and logical structure for the discussed pathways.
The statement in lines 184–185 regarding emergency systems remaining functional also requires clarification. If emergency workers are unable to perform their duties, it is difficult to consider the system functional.
We thank the reviewer for this helpful observation and will revisit the relevant section, adding a more clear explanation of what we mean with system in this regard and ensuring a coherent reasoning in the section.
Sections 2.5 and 2.6 appear to address similar concepts related to “Build Back Better” and could potentially be merged to reduce redundancy.
In line 198, the discussion refers to situations where recovery is completed, which may fall outside the paper’s stated focus on consecutive disasters. Some examples in Section 2.6 similarly refer to completed recovery and may introduce ambiguity.
We thank the reviewer for this helpful observation and for identifying instances where the discussion extends to recovery dynamics more generally, including cases where recovery is completed, rather than focusing explicitly on consecutive-disaster dynamics. In line with revision I, we will carefully reassess Sections 2.5 and 2.6 to determine whether their content aligns with the refined consecutive-disaster scope of the manuscript, reformulating or removing where necessary. Additionally, we will consider merging overlapping components to reduce redundancy.
- Health-Related Pathways
Section 3 presents four pathways, three of which focus on outbreaks, epidemics, and pandemics, and one on mental health impacts. In my view, these topics could be more effectively integrated into earlier sections. For example, health-related impacts could be framed as disasters shaping a consecutive disaster environment, with health systems treated consistently as critical systems. Changes in working environments due to outbreaks could also be discussed within Section 2.4.
Please consider this as a suggestion. In my view, this section could be structured differently and, rather than being framed around the health system, it could be presented with a stronger focus on human health as shown in figure 1. I recognize that this would require substantial changes to the overall structure of the paper; however, I encourage the authors to reflect on this possibility and to consider alternative conceptual frameworks beyond the current organization around four societal subsystems. There is also an easy solution to partiality address this confusion: just change the title of the section to impact on the human health! and keep it consistent through the section.
We thank the reviewer for recognising the possibility of organising the sections in a different manner. We acknowledge some overlap between the pathways discussed in Section 2 and 3, particularly given the cross-cutting nature of health systems as critical systems. However, we argue for maintaining the current structure, as the section on human settlements primarily focuses on the physical and built environment (e.g., housing and critical infrastructure systems, which also includes health-related infrastructure such as hospitals). In contrast, Section 3 discusses human health in a broader sense, including not only disruptions to health-related critical systems, but also other health outcomes such as mental health effects.
We agree with the reviewer that health-related impacts can be conceptualised as disasters that shape a consecutive-disaster environment, and that health systems can be treated consistently as critical systems, as was addressed in earlier comments as well as the overarching response (IV.2 and IV.3).
At the same time, we recognise that the current title of the chapter may create confusion between “health systems” as critical systems and “human health” as a societal domain. To reduce ambiguity, we will revise the title of Section 3 to “Impacts on Human Health”, in line with the reviewers suggestion.
- Economic System
Section 4 is, in my opinion, one of the strongest parts of the paper. One aspect that could be further strengthened is a clearer definition of economic recovery. Specifically, what does it mean for an economic system to be considered “recovered”?
We thank the reviewer for posing this important question. We agree that it’s not inherently clear for each subsystem what it means for that system to be “recovered”, and that this ambiguity deserves an explicit reflection.
Simultaneously, we don’t consider it within the scope of this paper to formalize how recovery should be defined within each societal subsystem. As outlined in the overarching response, we will discuss the “recovered” state, and how this is also indicator and system-dependent, in reflection point 2 of the new reflection chapter.
To highlight this better in the paper we will add a brief remark at the beginning of each subsystem chapter that reflects on how recovery might be operationalised within that specific context. For economic recovery one could consider for instance sales for a store, financial reserves for a government, debt, GDP.
Section 4.4 raises important questions about the scale of analysis. I suggest explicitly incorporating scale into the definition of consecutive disasters, clarifying at which spatial and temporal scales events are considered consecutive.
We thank the reviewer for highlighting the importance of scale in defining consecutiveness. Similarly to the issue of defining the recovered state, we believe that it’s beyond the scope of this paper to prescribe a specific definition. The intended contribution, rather, is to show how decisions in system definition and the analytical choices made directly affect recovery outcomes as well as your definition of consecutiveness. A reflection on scale specifically will be added to reflection point 1, as outlined in the overarching response.
In Section 4.5, the relevance to consecutive disaster recovery is less evident and may require further justification or reframing.
We thank the reviewer for this observation. In line with revision I, in which we sharpen the manuscript’s focus on consecutive disaster recovery specifically, we will revisit Section 4.5 to ensure that its relevance to consecutive-disaster dynamics is made explicit. Where necessary, we will either strengthen the connection to incomplete recovery and subsequent events or reframe (or remove) elements that do not clearly align with the refined consecutive scope of the paper.
- Socio-Political System
A key issue in this section is whether political instability is conceptualized as a disaster. This point should be clarified explicitly. If political instability is considered a disaster, then it would be useful to apply a similar logic across other societal subsystems (e.g., economic instability as a disaster interacting with other hazards).
We thank the reviewer for raising this important point. This concern is closely related to earlier comments regarding the conceptualisation of health crises and economic shocks as potential disasters. We will address this point as outlined in earlier comments as well as the overarching response (explicitly revision IV.3 and reflection point 2). We will ensure that the conceptualisation regarding these concepts is clarified and kept consistently throughout the paper.
I recommend revisiting page 18 with a strong focus on consecutive disaster recovery dynamics.
We thank the reviewer for this comment. This point closely relates to earlier remarks on the scope of the paper (Revision I) and our conceptualisation of disasters (Revision IV.3 and Reflection Point 2).
We will revisit page 18 to ensure a consistent framing, with a clearer focus on consecutive events and a transparent conceptualisation of political instability as outlined in the overarching response.
Additionally, the issue mentioned in line 474 would benefit from a concrete example.
We thank the reviewer for this suggestion and we will revisit this point and complement it with a concrete empirical example, consulting the referenced literature as well as additional relevant sources.
Finally, the term “socio-political system” may be too broad to capture the concepts discussed in this section. Alternative framings such as disaster risk governance, institutional capacity, or disaster risk management may be more precise.
We thank the reviewer for this observation, and we will consider alternative framings to better cover the content of this chapter.
Section 5.3, in particular, appears only loosely connected to consecutive disaster recovery dynamics. The ideas presented here may be more effectively integrated into the next sections like in section regarding the tipping point that leads to increase societal resilience.
We thank the reviewer for this observation. In line with revision I, we will narrow the scope of the main chapters more explicitly to consecutive disasters, excluding examples and pathways, such as the one in Section 5.3, that are only loosely connected to consecutive disasters.
Importantly, and in response both to this comment and to earlier remarks from RC1 regarding the limited discussion of positive examples, we have reassessed how we frame transformative change in the context of consecutive disasters. While disaster recovery may in principle create windows of opportunity for positive transformation, we recognise that the consecutiveness of disasters can itself act as an inhibitor of such change.
Therefore, we will use the examples described in section 5.3 to demonstrate that recovery processes have, in some contexts, created windows of opportunity for positive change. This will be integrated into reflection point 2, as outlined in the overarching response, where we also integrate former Section 6. Here we then critically reflect on how opportunities for positive transformation are affected by consecutive disasters.
Implications
Section 6 has the potential to be a core contribution of the paper; however, in its current form it reads more as a summary of previous sections. I encourage the authors to deepen the analytical discussion, particularly by strengthening the link between their findings and the framework proposed by Spaiser et al. (2024).
Specifically, clarifying what constitutes tipping points within each societal subsystem and what is meant by “system transformation” would greatly enhance the paper’s theoretical contribution. Examples discussed earlier in the paper, or new illustrative cases, could be used to demonstrate how societal resilience may be improved.
We appreciate the suggestions of the reviewer regarding Section 6. Based on the combined feedback from both reviewers, we have reassessed the role of the tipping point discussion as well as the role of the Spaiser et al. (2024) framework, as outlined in revision II and III.
As these revisions explain, we have decided to integrate the former Section 6 into the new reflection chapter (in reflection point 2). This aligns better with the original intention of the tipping-point framing, which was not to introduce a fully operationalised tipping-point framework. Rather, we use tipping dynamics as a conceptual lens, illustrating that recovery processes under consecutive disasters do not necessarily return to pre-disaster conditions but may instead lead to alternative system states.
Moveover, in light of the reviewers’ broader feedback regarding the scope and purpose of the paper and the observation of RC1 that the recommendations in Section 7 were insufficiently connected to Chapters 2 through 5, we have reconsidered the inclusion of policy-oriented recommendations. While we agree that providing suggestions on how to improve societal resilience is valuable, we recognise that particularly the policy-oriented recommendations previously included could not be directly and systematically derived from the empirical and conceptual analysis presented in the manuscript. Additionally, as addressed in the previous reply as well as reflection point 2, we question if with the narrowed consecutive focus positive change within recovery is possible, recognising consecutiveness of disasters as an inhibitor of positive change.
Finally, in line 557 migration is described as a negative feedback loop. I would caution that migration is not necessarily negative and can also be understood as an adaptive strategy, particularly for populations facing adverse or uninhabitable conditions.
We thank the reviewer for this important point and agree that migration should not be framed as inherently negative. We will revise the respective section to clarify that migration can also function as an adaptive strategy.
Peer review completion
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-3200', Anonymous Referee #1, 05 Nov 2025
Dear authors,
The paper addresses an interesting topic and is generally well written. I agree with the authors that the recovery process has received comparatively little attention in the literature, especially when it comes to consecutive disasters.
The paper is extremely broad and examines four different societal subsystems: human settlements, human health, the economic system, and the socio-political system. In my view, each of these subsystems is already extremely complex in terms of recovery processes and challenging to address in a single paper, let alone when combining all of them. Moreover, the paper aims to explore the interactions between these systems, considering how they may lead to a deterioration of resilience or to transformative learning. This broad focus inevitably results in a superficial conceptual discussion with rather ad-hoc assembled examples. This issue is also reflected in Figure 1, which essentially indicates that recovery trajectories are diverse, unknown, and involve interactions. Overall, I question whether this approach truly provides many new insights.
In this regard, while the authors accurately describe the examples as empirical, I would consider them anecdotal evidence to support the conceptual points the authors wish to convey. Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
Although the paper largely focuses on the potential negative impacts of consecutive disasters, considerable evidence suggests that disasters can serve as "windows of opportunity" leading to transformative learning (see e.g., Kreibich et al. 2017). This is rather briefly addressed in sections 5.3 and 4.2, but many more examples could be provided for each of the subsystems discussed. For instance, in Section 2, I am sure you can find numerous examples where disasters have led to revised spatial planning policies and/or building codes, significantly reducing the impact of subsequent disasters.
Section 6 also remains vague, offering little insight into what leads to tipping points in terms of deterioration or enhanced learning. The conclusion that these are linked to disaster intensity and frequency seems quite evident based on existing literature and cannot be derived directly from the empirical approach provided in the paper. We gain limited understanding of which discussed pathways are important or common.
Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations. For example, the authors state, “…to support efficient and equitable recovery after disasters, countries need more reliable and proactive financing solutions. These may also include pre-arranged recovery financing mechanisms such as forecast-based financing, where funds are automatically released for humanitarian actions that are agreed upon in advance (IFRC & RCCC, 2020) or parametric insurance, which offers rapid, flexible payouts based on pre-defined parameters such as certain rainfall or wind speed, ensuring rapid payments in post-disaster settings (Ocampo & Moreira, 2024)." (lines 631ff). While I do not disagree with the recommendation itself, I question how this particular suggestion is derived from section 4, and why it was chosen as one of four key recommendations among potentially many others?
Citation: https://doi.org/10.5194/egusphere-2025-3200-RC1 -
AC1: 'Reply on RC1', Sophie L. Buijs, 09 Mar 2026
Response to RC1
Dear authors,
The paper addresses an interesting topic and is generally well written. I agree with the authors that the recovery process has received comparatively little attention in the literature, especially when it comes to consecutive disasters.
[...]The paper is extremely broad[...]
[...]Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
[...]Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations.
We thank the reviewer for their thoughtful and constructive comments. We found the review especially valuable in highlighting that the manuscript’s current framing may create expectations that are misaligned with its intended contribution, and that both its added scientific value and its conceptual nature could be more clearly conveyed to the reader.
In response, we have carefully reassessed the manuscript’s framing. We agree with the reviewer that the paper is best understood as a perspective, conceptual contribution, rather than an analytical study. Accordingly, this paper does not aim to determine at what stage tipping points occur or which pathways are most prevalent. As the reviewer rightly observes, addressing these questions would necessitate more detailed analytical and quantitative investigation beyond the scope of the present study.
We acknowledge that, if the paper is read with the expectation of providing analytical results, the broad range of topics covered may appear problematic in terms of analytical depth. However, we believe that the value of the paper lies in illustrating the complexity of recovery processes in the context of consecutive disasters, and in showing how recovery can be affected in many different ways within and across societal domains, spatial and temporal scales.
To clarify the framing and purpose of the paper, we propose several adjustments to the manuscript, which we will outline below and address in more detail under the respective comments.
- First, we will move away from our framing around the Spaiser et al. (2024) tipping-point framework, as this may have suggested causal identification or quantitative analysis beyond the scope of the manuscript. Instead, more in line with the intended purpose of the empirical examples, we will present the different societal systems as a cross-section of society that illustrates how consecutive disasters can fundamentally increase the complexity of the recovery process across spatial and temporal scales, with potentially interrelated consequences across societal domains. We will clarify that these domains do not represent an exhaustive coverage and that additional (societal) domains could be considered. We believe that the strength of this paper lies in bridging the disciplinary silos. While people might already be aware of some of the complexities, we show how far-reaching the effects can be, within one aspect of society, but also connecting between the different societal domains.
- Second, we propose to revise and restructure Chapters 6-7 into a single reflection chapter that provides a clear conceptual synthesis of recovery complexity in the context of consecutive disasters.
We understand how chapter 6 might have been interpreted as vague if the expectation was that it would provide analytical insights into e.g., which pathways are most likely to lead to tipping points, when tipping points are reached, etc. However, this chapter was intended as a conceptual discussion, highlighting that recovery is not always linear and that a system does necessarily return to a pre-disaster state, particularly under consecutive disasters.
Reassessing the discussion in light of the reviewers’ comments, we also acknowledge that several of the recommendations included in the original manuscript, particularly those related to scientific research, were based on conceptual reflections of the concept of recovery in a consecutive disaster context. However, we see that these implicit conceptual reflections were not made sufficiently explicit, making the recommendations feel insufficiently grounded in relation to the rest of the manuscript
Therefore, rather than focusing on recommendations, the revised chapter will explicitly reflect on how recovery, consecutiveness, and outcomes are shaped by system definition and analytical choices. Integrating Chapters 6 and 7 into one combined new Chapter titled: “Reflecting on the complexity of recovery and consecutive disasters.” This chapter will be structured around three interrelated reflection points that we will outline below.
Reflection point 1: System definition
- This reflection point will address how system definition (i.e., the choice of indicators, societal subsystems, spatial and temporal scales, and the interconnections considered) directly shapes how recovery is defined, how consecutiveness is interpreted, and what recovery outcomes are observed.
- We will reflect on how recovery may appear complete in some subsystems (e.g., physical reconstruction) while remaining incomplete in others (e.g., mental health, governance capacity, or financial buffers), thereby affecting whether subsequent events are classified as consecutive. We will further discuss how incorporating spatial, sectoral, and functional interconnections can reveal forms of consecutiveness that are missed when recovery is assessed for a system or region in isolation. Finally, we will reflect on practical challenges in defining recovery timescales and event boundaries in contexts of prolonged or incomplete recovery, highlighting limitations of existing definitions of consecutiveness and the need for transparent system boundary choices tailored to the research question.
Reflection point 2: Recovery trajectories and “the recovered state”
- This reflection point will integrate the conceptual reflections that were originally embedded in Chapter 6, focusing on the non-linearity in recovery trajectories under consecutive disasters, tipping points, and the definition of “the recovered state”.
- We will discuss how consecutive disasters can increase the likelihood of non-linear responses and the occurrence of threshold effects leading to complex recovery trajectories . Such complex trajectories potentially pushing systems toward alternative, new, or harder to reverse states, a process which may be conceptualized as tipping. Specifically, in this context, recovery does not necessarily imply a return to pre-disaster conditions. We will therefore reflect on how the definition of a “recovered state” is inherently system-dependent, and should encompass non-linear and complex responses, as well as alternative system states. Additionally, we will discuss how what constitutes recovery may differ substantially across systems, such as physical assets, financial systems, governance structures, or social well-being.
- We will also address that while recovery can, in principle, create opportunities for positive transformation, consecutive disasters often inhibit such change by shifting attention and resources from long-term development to short-term response. These reflections highlight why simplified or static recovery assessments are inadequate in consecutive-disaster contexts and motivate the need for more dynamic recovery and risk assessment approaches.
Point 3: Conceptualization of “disasters”
- This reflection point will address conceptual ambiguities surrounding what constitutes a disaster versus an impact or contextual/vulnerability factors, and how this becomes more difficult to separate when adopting an all-hazard, consecutive disaster perspective. This point is added in response to feedback from the second reviewer regarding the framing of disasters stemming from natural hazard events, in relation to e.g., health crises, political instability, and socio-economic disruptions. In addition to this reflection point, a clear definition of how we consider these concepts throughout the paper will be added to the introduction.
Rather than prescribing a single definition or framing, all of these reflections highlight the need for transparency and reflexivity in analytical framing, recognizing that multiple valid interpretations may exist depending on the research question.
We believe these changes will result in a more coherent manuscript with a clearer conceptual contribution, better aligned with the reviewer’s expectations and concerns.
The paper is extremely broad and examines four different societal subsystems: human settlements, human health, the economic system, and the socio-political system. In my view, each of these subsystems is already extremely complex in terms of recovery processes and challenging to address in a single paper, let alone when combining all of them. Moreover, the paper aims to explore the interactions between these systems, considering how they may lead to a deterioration of resilience or to transformative learning. This broad focus inevitably results in a superficial conceptual discussion with rather ad-hoc assembled examples.
We fully acknowledge that each of the four societal subsystems addressed in the paper entail complex recovery dynamics that would require in-depth analysis in dedicated studies to fully untangle them. As explained in the overarching response, the aim of this paper is not to provide detailed subsystem-specific analyses, nor to comprehensively explain recovery processes within each domain. Instead, the purpose of the paper is to illustrate how the consecutiveness of disasters adds multiple layers of complexity to recovery processes across societal domains, and how recovery dynamics in different subsystems can interact across spatial and temporal scales. We hope that the revised framing of the manuscript, which explicitly clarifies its conceptual nature and intended contribution, will help to better understand the added value of this high-level perspective and to address concerns regarding the paper’s broad focus.
This issue is also reflected in Figure 1, which essentially indicates that recovery trajectories are diverse, unknown, and involve interactions. Overall, I question whether this approach truly provides many new insights.
In line with the observation of the reviewer, the figure indeed aims to convey the complexity in recovery trajectories across societal domains, and that interactions are involved. We believe that through the adjusted framing of the paper, clarifying its conceptual nature, the added value of the figure becomes clearer to the reader. It is a visual aid that helps the reader to understand the different dimensions of complexity that are involved and that will be illustrated through examples in the paper. We further will in addition improve the figure based on the concrete suggestion from RC2.
In this regard, while the authors accurately describe the examples as empirical, I would consider them anecdotal evidence to support the conceptual points the authors wish to convey. Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
We agree with the reviewer that the paper has a conceptual nature that should be conveyed more clearly to the reader. As outlined in the overarching response, we will therefore 1) adjust the framing of the paper in the abstract and introduction to clarify that the purpose of the paper is to illustrate and discuss the different dimensions of complexities of the recovery process involved and highlighted in the context of consecutive disasters, and 2) rewrite the discussion section to be a more conceptual discussion of the complexities involved in recovery dynamics in a consecutive disaster context.
In line with this reframing, we will also revise the title of the manuscript to avoid misleading expectations, to “A perspective on the complexity of recovery dynamics in the context of consecutive disasters” or something similar.
With respect to the reviewer’s comment on the empirical examples, we acknowledge that these should be understood as illustrative rather than systematic evidence. The examples in Chapters 2–5 are used non-exhaustively to support and ground the conceptual discussion, illustrating the diversity of recovery dynamics and pathways that can arise under consecutive disasters. The pathways presented were developed through iterative interdisciplinary discussions among the co-authors, informed by both the empirical examples and the existing literature, rather than being pre-defined. We expect that the revised framing will make the role of these examples clearer and align expectations accordingly.
Although the paper largely focuses on the potential negative impacts of consecutive disasters, considerable evidence suggests that disasters can serve as "windows of opportunity" leading to transformative learning (see e.g., Kreibich et al. 2017). This is rather briefly addressed in sections 5.3 and 4.2, but many more examples could be provided for each of the subsystems discussed. For instance, in Section 2, I am sure you can find numerous examples where disasters have led to revised spatial planning policies and/or building codes, significantly reducing the impact of subsequent disasters.
We agree with the reviewer that the paper includes more negative rather than positive examples and that there are more positive effects on transformative learning that can take place after a disaster. However, based on a comment provided by the second reviewer, we have decided to take the opportunity to strengthen the consecutive event focus of our paper. After careful consideration, we recognise that some of the pathways and examples that were provided in the first version of the manuscript, particularly those related to positive change, are actually beyond the scope of consecutive disasters, addressing instances where recovery was able to finish between events (e.g., Section 5.3 on societal learning and adaptation action).
With this refined focus, we find limited evidence of positive transformation occurring during unfinished recovery phases (e.g., in the example of spatial planning policies and/or building codes, while the measures are designed and implemented already during the recovery process, they only become relevant in practice once you have rebuilt in a new location or using improved building guidelines, so after recovery is done). Instead, consecutive disasters often appear to inhibit positive change by redirecting attention and resources from long-term planning to short-term response.
We will go over the text in detail to sharpen our consecutive focus, which will result in the removal of some examples of positive change that are now present in Chapters 2-5, such as section 5.3, and integrate “consecutive disasters as inhibitors of positive change” into reflection point 2 as described above.
Section 6 also remains vague, offering little insight into what leads to tipping points in terms of deterioration or enhanced learning. The conclusion that these are linked to disaster intensity and frequency seems quite evident based on existing literature and cannot be derived directly from the empirical approach provided in the paper. We gain limited understanding of which discussed pathways are important or common.
This concern is addressed under Revision II in the overarching response. In short, identifying which pathways are most important or common, or determining the conditions under which tipping points occur, was not the intended aim of this paper, as such questions would require dedicated analytical and quantitative studies beyond the scope of this conceptual contribution. In the revision, Section 6 will be integrated into the new reflection chapter and explicitly reframed as a conceptual discussion of non-linear recovery trajectories and alternative recovered states under consecutive disasters, clarifying what insights this section intended to provide.
Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations. For example, the authors state, “…to support efficient and equitable recovery after disasters, countries need more reliable and proactive financing solutions. These may also include pre-arranged recovery financing mechanisms such as forecast-based financing, where funds are automatically released for humanitarian actions that are agreed upon in advance (IFRC & RCCC, 2020) or parametric insurance, which offers rapid, flexible payouts based on pre-defined parameters, such as certain rainfall or wind speed, ensuring rapid payments in post-disaster settings (Ocampo & Moreira, 2024)." (lines 631ff). While I do not disagree with the recommendation itself, I question how this particular suggestion is derived from section 4, and why it was chosen as one of four key recommendations among potentially many others?
We thank the reviewer for this comment and agree that several recommendations in the discussion section may appear insufficiently grounded in the preceding sections. These recommendations initially aimed to provide potential avenues forward that could address the challenges we identified across societal domains in Chapters 2‑5. However, we acknowledge that selecting specific policy instruments (such as particular financing mechanisms) can give the impression of prioritization that goes beyond the conceptual scope of the paper.
As outlined in the overarching response, to improve coherence and better align the manuscript with its conceptual and perspective-oriented contribution, we will move away from a recommendation-focused discussion. Instead, Chapters 6 and 7 will be integrated and revised into one reflection chapter that explicitly synthesizes conceptual insights on recovery complexity under consecutive disasters. In this process, policy-oriented recommendations will be removed, as they do not fit well within the revised framing.
Citation: https://doi.org/10.5194/egusphere-2025-3200-AC1 -
AC3: 'Reply on RC1 - pdf version', Sophie L. Buijs, 09 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3200/egusphere-2025-3200-AC3-supplement.pdf
-
AC1: 'Reply on RC1', Sophie L. Buijs, 09 Mar 2026
-
RC2: 'Comment on egusphere-2025-3200', Anonymous Referee #2, 06 Jan 2026
General Comments
Dear Authors,
Thank you for this very interesting and timely manuscript. The paper addresses an important and still underexplored gap in the multi-hazard and multi-risk literature, an area that is rapidly expanding. I believe this work represents a valuable contribution and has the potential to become an influential reference in the field.
My main overarching suggestion is to strengthen and clarify the paper’s focus on consecutive disaster recovery. While I provide a detailed set of comments below, these are intended to be constructive rather than discouraging. On the contrary, I believe the authors have conducted a valuable and insightful study, and that further clarification and refinement would significantly enhance its impact and readability.
- Introduction
The introduction opens with a real-world example, and the manuscript as a whole makes extensive and effective use of empirical illustrations. Given this, I would suggest that the introduction focus less on additional examples and instead prioritize clearly framing the problem, introducing key definitions, and establishing the conceptual focus of the paper.
In particular, the scope of the paper would benefit from greater clarification. For example, in line 74 the authors state: “Even when recovery is completed between events, decisions made during the recovery process… can fundamentally shape a society’s resilience to subsequent events.” At this point, it is unclear whether the paper focuses specifically on consecutive disaster recovery, recovery processes more generally, or whether the concept of “consecutive disaster” itself still requires clearer definition. I encourage the authors to explicitly clarify this distinction early in the paper.
Following this, the subsystems proposed by Spaiser et al. (2024) are introduced, and a systemic definition is developed. However, the rationale behind adopting and adapting this framework remains unclear. I recommend explaining in greater detail why this particular framework was selected, how it relates to other existing approaches, and how the authors arrived at the proposed system definition.
Additionally, the introduction refers to “health crises” alongside “disasters.” It would be helpful to clarify whether health crises are conceptualized as disasters within the framework of this paper, or whether they are treated as distinct phenomena.
Regarding Figure 1, since uncertainty in recovery outcomes is a central theme of the paper, it is unclear why uncertainty is not visually represented in the recovery phase following the first event. As currently presented, the figure seems to associate uncertainty only with the recovery from the second event.
Between lines 90 and 100, the authors introduce the concept of societal recovery. It would be beneficial to clearly define this term and explain how it differs from “recovery” more generally. Furthermore, I recommend clarifying the criteria and process used for selecting the recovery events and cases discussed in the paper. Although the authors note that this is not a systematic review, a more transparent methodological explanation would add rigor and clarity.
Finally, briefly outlining the key issues addressed in each section at the end of the introduction with their related section numbers could greatly improve readability and guide the reader through the manuscript.
- Impacts within Human Settlements
In the opening of this section, human settlements and critical systems are defined; however, health is not included among the listed critical systems unlike in section 2.3. This omission raises questions and should be either justified or clarified.
I also suggest shortening the sentence in lines 105–107 for clarity and readability.
In the first pathway, the authors discuss increased exposure and vulnerability through displacement and relocation. However, the connection between this argument and the discussion in lines 114–117 is not entirely clear and could be strengthened.
In lines 139–144, the manuscript refers to the lack of critical services such as WASH infrastructure. Here, it would be useful to clarify whether health-related impacts are considered disasters in their own right, and whether this example is intended to illustrate a consecutive disaster. If WASH is considered part of the health subsystem, human settlement subsystem, or both, this should be explicitly stated.
In Section 2.3, healthcare is identified as a critical system. This highlights the importance of clearly defining what constitutes a “critical system” and how these systems relate to the societal subsystems introduced earlier. Such clarification would also help explain why food systems are treated as critical systems rather than as considering ecosystems as critical system or why the issue of the food is not addressed economy societal subsystem section.
In Section 2.4, it is unclear whether the focus is on resource shortages, unworkable environments, or response capacity. I suggest maintaining a clear focus on one dimension and addressing others in relevant subsections. For example, maybe resource depletion could be discussed more comprehensively within the economic system section.
The statement in lines 184–185 regarding emergency systems remaining functional also requires clarification. If emergency workers are unable to perform their duties, it is difficult to consider the system functional.
Sections 2.5 and 2.6 appear to address similar concepts related to “Build Back Better” and could potentially be merged to reduce redundancy.
In line 198, the discussion refers to situations where recovery is completed, which may fall outside the paper’s stated focus on consecutive disasters. Some examples in Section 2.6 similarly refer to completed recovery and may introduce ambiguity.
- Health-Related Pathways
Section 3 presents four pathways, three of which focus on outbreaks, epidemics, and pandemics, and one on mental health impacts. In my view, these topics could be more effectively integrated into earlier sections. For example, health-related impacts could be framed as disasters shaping a consecutive disaster environment, with health systems treated consistently as critical systems. Changes in working environments due to outbreaks could also be discussed within Section 2.4.
Please consider this as a suggestion. In my view, this section could be structured differently and, rather than being framed around the health system, it could be presented with a stronger focus on human health as shown in figure 1. I recognize that this would require substantial changes to the overall structure of the paper; however, I encourage the authors to reflect on this possibility and to consider alternative conceptual frameworks beyond the current organization around four societal subsystems. There is also an easy solution to partiality address this confusion: just change the title of the section to impact on the human health! and keep it consistent through the section.
- Economic System
Section 4 is, in my opinion, one of the strongest parts of the paper. One aspect that could be further strengthened is a clearer definition of economic recovery. Specifically, what does it mean for an economic system to be considered “recovered”?
Section 4.4 raises important questions about the scale of analysis. I suggest explicitly incorporating scale into the definition of consecutive disasters, clarifying at which spatial and temporal scales events are considered consecutive.
In Section 4.5, the relevance to consecutive disaster recovery is less evident and may require further justification or reframing.
- Socio-Political System
A key issue in this section is whether political instability is conceptualized as a disaster. This point should be clarified explicitly. If political instability is considered a disaster, then it would be useful to apply a similar logic across other societal subsystems (e.g., economic instability as a disaster interacting with other hazards).
I recommend revisiting page 18 with a strong focus on consecutive disaster recovery dynamics. Additionally, the issue mentioned in line 474 would benefit from a concrete example.
Finally, the term “socio-political system” may be too broad to capture the concepts discussed in this section. Alternative framings such as disaster risk governance, institutional capacity, or disaster risk management may be more precise.
Section 5.3, in particular, appears only loosely connected to consecutive disaster recovery dynamics. The ideas presented here may be more effectively integrated into the next sections like in section ragrding the tipping point that leads to increase societal resilience.
- Implications
Section 6 has the potential to be a core contribution of the paper; however, in its current form it reads more as a summary of previous sections. I encourage the authors to deepen the analytical discussion, particularly by strengthening the link between their findings and the framework proposed by Spaiser et al. (2024).
Specifically, clarifying what constitutes tipping points within each societal subsystem and what is meant by “system transformation” would greatly enhance the paper’s theoretical contribution. Examples discussed earlier in the paper, or new illustrative cases, could be used to demonstrate how societal resilience may be improved.
Finally, in line 557 migration is described as a negative feedback loop. I would caution that migration is not necessarily negative and can also be understood as an adaptive strategy, particularly for populations facing adverse or uninhabitable conditions.
Citation: https://doi.org/10.5194/egusphere-2025-3200-RC2 -
AC2: 'Reply on RC2', Sophie L. Buijs, 09 Mar 2026
Response to RC2
- General Comments
Dear Authors,
Thank you for this very interesting and timely manuscript. The paper addresses an important and still underexplored gap in the multi-hazard and multi-risk literature, an area that is rapidly expanding. I believe this work represents a valuable contribution and has the potential to become an influential reference in the field.
My main overarching suggestion is to strengthen and clarify the paper’s focus on consecutive disaster recovery. While I provide a detailed set of comments below, these are intended to be constructive rather than discouraging. On the contrary, I believe the authors have conducted a valuable and insightful study, and that further clarification and refinement would significantly enhance its impact and readability.
We would like to thank the reviewer for the thorough review and constructive feedback. We greatly appreciate the encouraging words and the recognition of the manuscript’s relevance. Based on the detailed comments provided by the reviewer, we have carefully reassessed the manuscript and propose several substantial revisions, as well as smaller changes, that, together, will help to refine the framing, scope, clarity, and conceptual contribution of the paper. Overarching adjustments are outlined below, with individual comments addressed in more detail under the respective points.
I. We agree that the manuscript would benefit from a more explicit and consistent focus on consecutive disasters, avoiding references to more general recurrent disaster contexts. We will therefore revise the manuscript to strengthen the consecutive-disaster focus throughout, removing or reframing examples that assume completed recovery between events and establishing a clear framing from the start. We will take specific care to reevaluate the sections explicitly highlighted by the reviewer.
II. After reassessing the conceptual focus of the paper, based on the reviewers feedback, we recognise that the main value of the paper lies in illustrating the complexity of recovery processes in the context of consecutive disasters, and in showing how recovery can be affected in many different ways within and across societal domains, spatial and temporal scales. The paper therefore is best understood as a perspective, conceptual contribution.
In this context, we chose to frame our manuscript around the tipping point framework of Spaiser et al. (2024), as we recognised the tipping point literature and framing to be relevant in the context of consecutive disasters specifically. The tipping literature was intended to be a conceptual lens to reflect on the possibility that recovery under consecutive disasters may not return to a pre-disaster state and may instead lead to alternative system trajectories. We see how the strong framing around the tipping-point framework has created the suggestion of more quantitative and concrete statements regarding a definition of the identification of tipping thresholds within each subsystem, which tipping points are common, etc. that cannot be provided using the illustrative examples presented in Chapters 2 through 5.
Therefore, more in line with the intended purpose of the empirical examples, we will move away from the Spaiser et al. (2024) framework and present the different societal systems as a cross-section of society that illustrates how consecutive disasters can fundamentally increase the complexity of the recovery process across spatial and temporal scales, with potentially interrelated consequences across societal domains. We will also clarify that these domains do not represent an exhaustive coverage and that additional (societal) domains could be considered.
III. Related to this, we propose to revise and restructure Chapters 6-7 into a single reflection chapter that provides a clear conceptual synthesis of recovery complexity in the context of consecutive disasters. Rather than focusing on recommendations, the revised chapter will explicitly reflect on how recovery, consecutiveness and outcomes are shaped by system definition and analytical choices. This chapter will be structured around three interrelated reflection points that we will outline below.
Reflection point 1: System definition
- This reflection point will address how system definition (i.e., the choice of indicators, societal subsystems, spatial and temporal scales, and the interconnections considered) directly shapes how recovery is defined, how consecutiveness is interpreted, and what recovery outcomes are observed.
- We will reflect on how recovery may appear complete in some subsystems (e.g., physical reconstruction) while remaining incomplete in others (e.g., mental health, governance capacity, or financial buffers), thereby affecting whether subsequent events are classified as consecutive. We will further discuss how incorporating spatial, sectoral, and functional interconnections can reveal forms of consecutiveness that are missed when recovery is assessed for a system or region in isolation. Finally, we will reflect on practical challenges in defining recovery timescales and event boundaries in contexts of prolonged or incomplete recovery, highlighting limitations of existing definitions of consecutiveness and the need for transparent system boundary choices tailored to the research question.
Reflection point 2: Recovery trajectories and “the recovered state”
- This reflection point will integrate the conceptual reflections that were originally embedded in Chapter 6, focusing on the non-linearity in recovery trajectories under consecutive disasters, tipping points, and the definition of “the recovered state”.
- We will discuss how consecutive disasters can increase the likelihood of non-linear responses and the occurrence of threshold effects leading to complex recovery trajectories . Such complex trajectories potentially pushing systems toward alternative, new, or harder to reverse states, a process which may be conceptualized as tipping. Specifically, in this context, recovery does not necessarily imply a return to pre-disaster conditions. We will therefore reflect on how the definition of a “recovered state” is inherently system-dependent, and should encompass non-linear and complex responses, as well as alternative system states. Additionally, we will discuss how what constitutes recovery may differ substantially across systems, such as physical assets, financial systems, governance structures, or social well-being.
- We will also address that while recovery can, in principle, create opportunities for positive transformation, consecutive disasters often inhibit such change by shifting attention and resources from long-term development to short-term response. These reflections highlight why simplified or static recovery assessments are inadequate in consecutive-disaster contexts and motivate the need for more dynamic recovery and risk assessment approaches.
Reflection point 3: Conceptualization of “disasters”
- This reflection point will address conceptual ambiguities surrounding what constitutes a disaster versus an impact or contextual/vulnerability factors, and how this becomes more difficult to separate when adopting an all-hazard, consecutive disaster perspective.
Rather than prescribing a single definition or framing, all of these reflections highlight the need for transparency and reflexivity in analytical framing, recognizing that multiple valid interpretations may exist depending on the research question.
IV. Lastly, we will address the reviewers observations regarding unclarity of some of the terminology used within the paper.
- As described, we will provide more consistent use and framing around the provided definition for consecutive disasters.
- We will add a clear definition of critical systems and services early in the paper, making sure that these definitions are then applied consistently throughout the manuscript.
- At the start of the paper, we will add a section clarifying how economic shocks, health crises and political instability are conceptualised in relation to the concept of “disaster.” Following the all-hazards approach adopted in the UNDRR Hazard Information Profiles (HIPs), phenomena such as political and economic instability can be classified as societal hazards, while infectious diseases are classified as biological hazards. Disasters can then result from the interaction of any hazard type with exposure and vulnerability (H × E × V). In principle, therefore, large-scale health crises, financial shocks or political crises may constitute disasters in their own right. At the same time, we recognize that in much of the disaster risk reduction literature, the term “disaster” is commonly used to refer primarily to events triggered by natural hazards (e.g., meteorological, climatological or geological hazards). To discuss the interaction between natural-hazard-induced disasters in this more traditional sense and health, economic or political crises in a consistent and transparent manner, we will use the term “disaster” in this paper primarily to refer to natural-hazard-induced disasters, while explicitly referring to economic crises, political crises and health crises as distinct phenomena. This while acknowledging that these phenomena can, under an all-hazards perspective, constitute disasters in their own right and may therefore create consecutive-disaster dynamics when their impacts overlap with those of natural-hazard-induced events.
- We will also add a clarifying statement regarding the definition of “societal recovery” in the introduction, explaining that the addition of the word “societal” meant to limit the scope to human systems, which also clarifies the question as to why ecosystem recovery was not included in the review.
- We will add some lines on how the “recovered state” can be defined or conceptualised within each of the different societal domains, yet we do not aim to provide one binding definition for this (as is addressed in reflection point 2).
We believe these changes, made based on the reviewers' elaborate feedback, will result in a more coherent manuscript with a clearer conceptual contribution.
Introduction
The introduction opens with a real-world example, and the manuscript as a whole makes extensive and effective use of empirical illustrations. Given this, I would suggest that the introduction focus less on additional examples and instead prioritize clearly framing the problem, introducing key definitions, and establishing the conceptual focus of the paper.
We thank the reviewer for this suggestion and agree that it would be better to take out the empirical example from the introduction, to instead prioritize a clear framing and scope of the paper, introduction of definitions, and clarification of the conceptual focus as outlined in the overarching response.
In particular, the scope of the paper would benefit from greater clarification. For example, in line 74 the authors state: “Even when recovery is completed between events, decisions made during the recovery process… can fundamentally shape a society’s resilience to subsequent events.” At this point, it is unclear whether the paper focuses specifically on consecutive disaster recovery, recovery processes more generally, or whether the concept of “consecutive disaster” itself still requires clearer definition. I encourage the authors to explicitly clarify this distinction early in the paper.
We agree with the reviewer that the statement in line 74 is not sufficiently specific for consecutive disasters. As outlined in revision I, we will refine the manuscript to limit its focus more strictly to consecutive events, following the original definition provided in the first version of the manuscript.
Following this, the subsystems proposed by Spaiser et al. (2024) are introduced, and a systemic definition is developed. However, the rationale behind adopting and adapting this framework remains unclear. I recommend explaining in greater detail why this particular framework was selected, how it relates to other existing approaches, and how the authors arrived at the proposed system definition.
We thank the reviewer for pointing out this ambiguity. As outlined in revision II, we have reconsidered the role of the Spaiser et al. (2024) framework in the manuscript. While it was initially introduced because we considered tipping-point theory conceptually relevant to recovery under consecutive disasters, we recognize that its inclusion as an organizing structure was not sufficiently justified and may have created expectations that extend beyond the conceptual and illustrative contribution of this work.
We will therefore move away from this framework and instead present the selected societal domains as a cross-section of societally relevant subsystems, in line with the examples’ intended purpose, which was to illustrate how consecutive disasters add complexity to recovery processes across interconnected societal domains.
Additionally, the introduction refers to “health crises” alongside “disasters.” It would be helpful to clarify whether health crises are conceptualized as disasters within the framework of this paper, or whether they are treated as distinct phenomena.
We thank the reviewer for raising this point. As outlined in revision IV, we will add a section to the introduction to address this conceptualisation explicitly. In the HIPS, infectious diseases like cholera and tuberculosis are conceptualised as biological hazards, which means that a large disease outbreak can be conceptualised as a disaster.
At the same time, we emphasize that such classifications are not always clear-cut, particularly in the context of consecutive disasters. Health crises, political instability, and economic shocks may be framed as independent disasters, as impacts resulting from prior disasters, or as contextual conditions or vulnerability factors shaping subsequent recovery trajectories. We will address these blurred boundaries in reflection point 3 of the new reflection chapter.
Regarding Figure 1, since uncertainty in recovery outcomes is a central theme of the paper, it is unclear why uncertainty is not visually represented in the recovery phase following the first event. As currently presented, the figure seems to associate uncertainty only with the recovery from the second event.
We suggest adjusting figure 1 and/or its caption to address the concerns about the additional dimensions of recovery that are currently not visually represented. While we initially chose to limit the visual representation of uncertainty to the second recovery phase, as this is what the paper focuses on specifically, we will clarify either visually or through text that the recovery uncertainty is also relevant to the initial recovery process.
Additionally, we will add a fifth grey box to represent additional societal, or non-human systems, as we show only a cross-section of possible domains that can be included. This will help clarify to the readers that the chosen domains are non-exhaustive.
Between lines 90 and 100, the authors introduce the concept of societal recovery. It would be beneficial to clearly define this term and explain how it differs from “recovery” more generally.
We thank the reviewer for this helpful suggestion. As outlined in revision IV, we will explicitly explain in the introduction that the term “societal” is added before recovery simply to delimit the scope of the paper to various recovery processes within human systems, as similar dynamics are relevant in non-human systems as well but outside the scope of this paper.
Furthermore, I recommend clarifying the criteria and process used for selecting the recovery events and cases discussed in the paper. Although the authors note that this is not a systematic review, a more transparent methodological explanation would add rigor and clarity.
We agree with the reviewer that it would be good to clarify our methodological approach. The initial pathways were identified based on expert input through multiple rounds of discussion among the authors, all of whom have research experience relevant to the topic. These preliminary pathways were subsequently refined through iterative rounds of targeted literature searches and collective discussion. As new insights emerged from the literature, additional iterations were undertaken, leading to the final set of illustrative pathways. We will describe this expert-driven and literature-informed iterative process more explicitly in the revised manuscript to enhance methodological clarity.
Finally, briefly outlining the key issues addressed in each section at the end of the introduction with their related section numbers could greatly improve readability and guide the reader through the manuscript.
We thank the referee for this suggestion and will add a brief description of each section at the of the introduction that will help the reader to navigate the dense content of our manuscript.
- Impacts within Human Settlements
In the opening of this section, human settlements and critical systems are defined; however, health is not included among the listed critical systems unlike in section 2.3. This omission raises questions and should be either justified or clarified.
We thank the reviewer for noting this point. As outlined in revision IV, we will introduce a clearer and more consistent definition of “critical systems” early in the manuscript, including how these systems relate to the concept of “critical services” and to the societal domains analysed in the paper. In this context, healthcare systems (e.g., hospitals) and WASH infrastructure are recognised as critical systems that are relevant both within the human settlements domain and for human health.
While the list of critical systems at the start of Section 2 was not intended to be exhaustive, we agree that the omission of healthcare systems may create unnecessary confusion, particularly as human health is explicitly treated as a societal domain in the paper. We will therefore revise the opening of the human settlements section to explicitly include healthcare systems among the relevant critical systems, thereby clarifying how these concepts are integrated and interconnected across domains.
I also suggest shortening the sentence in lines 105–107 for clarity and readability.
We thank the reviewer for this suggestion and will revise the sentence in lines 105-107.
In the first pathway, the authors discuss increased exposure and vulnerability through displacement and relocation. However, the connection between this argument and the discussion in lines 114–117 is not entirely clear and could be strengthened.
We appreciate this suggestion and will strengthen our argumentation in this section.
In lines 139–144, the manuscript refers to the lack of critical services such as WASH infrastructure. Here, it would be useful to clarify whether health-related impacts are considered disasters in their own right, and whether this example is intended to illustrate a consecutive disaster. If WASH is considered part of the health subsystem, human settlement subsystem, or both, this should be explicitly stated.
We thank the reviewer for this important clarification request. As outlined in revision IV.2, we will introduce a clearer definition of critical systems and their relation to the societal domains discussed in the paper. In this context, WASH infrastructure will be defined as a critical system that is of relevance both within the human settlements and human health domains.
Regarding the conceptualisation of health-related impacts as potential disasters and whether the example intends to illustrate a consecutive disaster, we will revisit the text to ensure consistency with the definitions of disasters as we will introduce as outlined in revision IV.3. Following this conceptualisation and the hazard definitions as provided in the HIPS, we believe that a lack of access to critical services is not to be considered a hazard, but rather a vulnerability factor, increasing the vulnerability to biological hazards such as cholera. An outbreak resulting from this interaction (between the presence of a hazard and heightened vulnerability, for an exposed population) could, within the conceptualisation as defined in IV.3 then be considered a disaster (in the manuscript explicitly referred to as health emergency, to distinguish from natural-hazard caused disasters). We will revise the paragraph accordingly to make explicit that in this case absence of WASH services is conceptualised as a vulnerability driver within the recovery process, while a subsequent disease outbreak may constitute a consecutive disaster. As we discuss also in the reflection chapter we believe that conceptualisations like this are not set in stone and can be done differently depending on your research goal, but we agree that given the purpose, scale, and boundaries of the system of the study it is important that provided definitions are coherent.
In Section 2.3, healthcare is identified as a critical system. This highlights the importance of clearly defining what constitutes a “critical system” and how these systems relate to the societal subsystems introduced earlier. Such clarification would also help explain why food systems are treated as critical systems rather than as considering ecosystems as critical system or why the issue of the food is not addressed economy societal subsystem section.
As outlined in revision IV.2, we will introduce an explicit definition of critical systems early in the paper. In the revised manuscript, critical systems will be defined as functional systems that provide essential services necessary for the maintenance of societal functioning and well-being (critical services such as healthcare, WASH, food, energy, transport). These systems are not conceptualised as separate societal subsystems, but rather as cross-cutting infrastructures that are essential for the functioning of different societal domains.
The critical systems discussed in the manuscript are not an exhaustive list, but rather serve as illustrative examples that help demonstrate how disruptions to essential services can influence multiple societal domains simultaneously. Other systems may equally qualify as critical depending on the specific research context, scale, or hazard type under consideration. While we address the issue of breadbasket failures in the economy social subsystem section, we will make sure to clarify the link to the food system.
With regard to the ecosystems mentioned by the reviewer, we will clarify that the scope of the paper is explicitly focusing on societal recovery within human systems. While ecosystems evidently constitute a network of critical systems in themselves for society, the study of their recovery processes in regards to consecutive events represents an entire distinct field of study. We will explicitly discuss this crucial complexity in the outlook of the paper.
In Section 2.4, it is unclear whether the focus is on resource shortages, unworkable environments, or response capacity. I suggest maintaining a clear focus on one dimension and addressing others in relevant subsections. For example, maybe resource depletion could be discussed more comprehensively within the economic system section.
We thank the reviewer for this suggestion and we will revisit this section to ensure a clear focus and logical structure for the discussed pathways.
The statement in lines 184–185 regarding emergency systems remaining functional also requires clarification. If emergency workers are unable to perform their duties, it is difficult to consider the system functional.
We thank the reviewer for this helpful observation and will revisit the relevant section, adding a more clear explanation of what we mean with system in this regard and ensuring a coherent reasoning in the section.
Sections 2.5 and 2.6 appear to address similar concepts related to “Build Back Better” and could potentially be merged to reduce redundancy.
In line 198, the discussion refers to situations where recovery is completed, which may fall outside the paper’s stated focus on consecutive disasters. Some examples in Section 2.6 similarly refer to completed recovery and may introduce ambiguity.
We thank the reviewer for this helpful observation and for identifying instances where the discussion extends to recovery dynamics more generally, including cases where recovery is completed, rather than focusing explicitly on consecutive-disaster dynamics. In line with revision I, we will carefully reassess Sections 2.5 and 2.6 to determine whether their content aligns with the refined consecutive-disaster scope of the manuscript, reformulating or removing where necessary. Additionally, we will consider merging overlapping components to reduce redundancy.
- Health-Related Pathways
Section 3 presents four pathways, three of which focus on outbreaks, epidemics, and pandemics, and one on mental health impacts. In my view, these topics could be more effectively integrated into earlier sections. For example, health-related impacts could be framed as disasters shaping a consecutive disaster environment, with health systems treated consistently as critical systems. Changes in working environments due to outbreaks could also be discussed within Section 2.4.
Please consider this as a suggestion. In my view, this section could be structured differently and, rather than being framed around the health system, it could be presented with a stronger focus on human health as shown in figure 1. I recognize that this would require substantial changes to the overall structure of the paper; however, I encourage the authors to reflect on this possibility and to consider alternative conceptual frameworks beyond the current organization around four societal subsystems. There is also an easy solution to partiality address this confusion: just change the title of the section to impact on the human health! and keep it consistent through the section.
We thank the reviewer for recognising the possibility of organising the sections in a different manner. We acknowledge some overlap between the pathways discussed in Section 2 and 3, particularly given the cross-cutting nature of health systems as critical systems. However, we argue for maintaining the current structure, as the section on human settlements primarily focuses on the physical and built environment (e.g., housing and critical infrastructure systems, which also includes health-related infrastructure such as hospitals). In contrast, Section 3 discusses human health in a broader sense, including not only disruptions to health-related critical systems, but also other health outcomes such as mental health effects.
We agree with the reviewer that health-related impacts can be conceptualised as disasters that shape a consecutive-disaster environment, and that health systems can be treated consistently as critical systems, as was addressed in earlier comments as well as the overarching response (IV.2 and IV.3).
At the same time, we recognise that the current title of the chapter may create confusion between “health systems” as critical systems and “human health” as a societal domain. To reduce ambiguity, we will revise the title of Section 3 to “Impacts on Human Health”, in line with the reviewers suggestion.
- Economic System
Section 4 is, in my opinion, one of the strongest parts of the paper. One aspect that could be further strengthened is a clearer definition of economic recovery. Specifically, what does it mean for an economic system to be considered “recovered”?
We thank the reviewer for posing this important question. We agree that it’s not inherently clear for each subsystem what it means for that system to be “recovered”, and that this ambiguity deserves an explicit reflection.
Simultaneously, we don’t consider it within the scope of this paper to formalize how recovery should be defined within each societal subsystem. As outlined in the overarching response, we will discuss the “recovered” state, and how this is also indicator and system-dependent, in reflection point 2 of the new reflection chapter.
To highlight this better in the paper we will add a brief remark at the beginning of each subsystem chapter that reflects on how recovery might be operationalised within that specific context. For economic recovery one could consider for instance sales for a store, financial reserves for a government, debt, GDP.
Section 4.4 raises important questions about the scale of analysis. I suggest explicitly incorporating scale into the definition of consecutive disasters, clarifying at which spatial and temporal scales events are considered consecutive.
We thank the reviewer for highlighting the importance of scale in defining consecutiveness. Similarly to the issue of defining the recovered state, we believe that it’s beyond the scope of this paper to prescribe a specific definition. The intended contribution, rather, is to show how decisions in system definition and the analytical choices made directly affect recovery outcomes as well as your definition of consecutiveness. A reflection on scale specifically will be added to reflection point 1, as outlined in the overarching response.
In Section 4.5, the relevance to consecutive disaster recovery is less evident and may require further justification or reframing.
We thank the reviewer for this observation. In line with revision I, in which we sharpen the manuscript’s focus on consecutive disaster recovery specifically, we will revisit Section 4.5 to ensure that its relevance to consecutive-disaster dynamics is made explicit. Where necessary, we will either strengthen the connection to incomplete recovery and subsequent events or reframe (or remove) elements that do not clearly align with the refined consecutive scope of the paper.
- Socio-Political System
A key issue in this section is whether political instability is conceptualized as a disaster. This point should be clarified explicitly. If political instability is considered a disaster, then it would be useful to apply a similar logic across other societal subsystems (e.g., economic instability as a disaster interacting with other hazards).
We thank the reviewer for raising this important point. This concern is closely related to earlier comments regarding the conceptualisation of health crises and economic shocks as potential disasters. We will address this point as outlined in earlier comments as well as the overarching response (explicitly revision IV.3 and reflection point 2). We will ensure that the conceptualisation regarding these concepts is clarified and kept consistently throughout the paper.
I recommend revisiting page 18 with a strong focus on consecutive disaster recovery dynamics.
We thank the reviewer for this comment. This point closely relates to earlier remarks on the scope of the paper (Revision I) and our conceptualisation of disasters (Revision IV.3 and Reflection Point 2).
We will revisit page 18 to ensure a consistent framing, with a clearer focus on consecutive events and a transparent conceptualisation of political instability as outlined in the overarching response.
Additionally, the issue mentioned in line 474 would benefit from a concrete example.
We thank the reviewer for this suggestion and we will revisit this point and complement it with a concrete empirical example, consulting the referenced literature as well as additional relevant sources.
Finally, the term “socio-political system” may be too broad to capture the concepts discussed in this section. Alternative framings such as disaster risk governance, institutional capacity, or disaster risk management may be more precise.
We thank the reviewer for this observation, and we will consider alternative framings to better cover the content of this chapter.
Section 5.3, in particular, appears only loosely connected to consecutive disaster recovery dynamics. The ideas presented here may be more effectively integrated into the next sections like in section regarding the tipping point that leads to increase societal resilience.
We thank the reviewer for this observation. In line with revision I, we will narrow the scope of the main chapters more explicitly to consecutive disasters, excluding examples and pathways, such as the one in Section 5.3, that are only loosely connected to consecutive disasters.
Importantly, and in response both to this comment and to earlier remarks from RC1 regarding the limited discussion of positive examples, we have reassessed how we frame transformative change in the context of consecutive disasters. While disaster recovery may in principle create windows of opportunity for positive transformation, we recognise that the consecutiveness of disasters can itself act as an inhibitor of such change.
Therefore, we will use the examples described in section 5.3 to demonstrate that recovery processes have, in some contexts, created windows of opportunity for positive change. This will be integrated into reflection point 2, as outlined in the overarching response, where we also integrate former Section 6. Here we then critically reflect on how opportunities for positive transformation are affected by consecutive disasters.
Implications
Section 6 has the potential to be a core contribution of the paper; however, in its current form it reads more as a summary of previous sections. I encourage the authors to deepen the analytical discussion, particularly by strengthening the link between their findings and the framework proposed by Spaiser et al. (2024).
Specifically, clarifying what constitutes tipping points within each societal subsystem and what is meant by “system transformation” would greatly enhance the paper’s theoretical contribution. Examples discussed earlier in the paper, or new illustrative cases, could be used to demonstrate how societal resilience may be improved.
We appreciate the suggestions of the reviewer regarding Section 6. Based on the combined feedback from both reviewers, we have reassessed the role of the tipping point discussion as well as the role of the Spaiser et al. (2024) framework, as outlined in revision II and III.
As these revisions explain, we have decided to integrate the former Section 6 into the new reflection chapter (in reflection point 2). This aligns better with the original intention of the tipping-point framing, which was not to introduce a fully operationalised tipping-point framework. Rather, we use tipping dynamics as a conceptual lens, illustrating that recovery processes under consecutive disasters do not necessarily return to pre-disaster conditions but may instead lead to alternative system states.
Moveover, in light of the reviewers’ broader feedback regarding the scope and purpose of the paper and the observation of RC1 that the recommendations in Section 7 were insufficiently connected to Chapters 2 through 5, we have reconsidered the inclusion of policy-oriented recommendations. While we agree that providing suggestions on how to improve societal resilience is valuable, we recognise that particularly the policy-oriented recommendations previously included could not be directly and systematically derived from the empirical and conceptual analysis presented in the manuscript. Additionally, as addressed in the previous reply as well as reflection point 2, we question if with the narrowed consecutive focus positive change within recovery is possible, recognising consecutiveness of disasters as an inhibitor of positive change.
Finally, in line 557 migration is described as a negative feedback loop. I would caution that migration is not necessarily negative and can also be understood as an adaptive strategy, particularly for populations facing adverse or uninhabitable conditions.
We thank the reviewer for this important point and agree that migration should not be framed as inherently negative. We will revise the respective section to clarify that migration can also function as an adaptive strategy.
Peer review completion
Journal article(s) based on this preprint
We explored how disasters that occur before recovery from earlier events is complete can complicate recovery, using examples from science and the real world across different societal domains. We then reflect on why recovery under repeated disasters is more complex than simply returning to normal, and how the way we define recovery and the system we study shapes our understanding. These insights can help us better understand, study, and manage recovery from disasters.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 15,287 | 3,644 | 245 | 19,176 | 305 | 305 |
- HTML: 15,287
- PDF: 3,644
- XML: 245
- Total: 19,176
- BibTeX: 305
- EndNote: 305
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
2 citations as recorded by crossref.
- Multi-hazard risk assessment and management: pathways for the Sendai Framework and beyond T. Tiggeloven et al. https://doi.org/10.5194/gc-9-185-2026
- Reducing risk together: moving towards a more holistic approach to multi-hazard and multi-risk assessment and management P. Ward et al. https://doi.org/10.5194/nhess-26-1325-2026
Chahan M. Kropf
Samuel Juhel
Zélie Stalhandske
Marleen C. De Ruiter
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3143 KB) - Metadata XML
Dear authors,
The paper addresses an interesting topic and is generally well written. I agree with the authors that the recovery process has received comparatively little attention in the literature, especially when it comes to consecutive disasters.
The paper is extremely broad and examines four different societal subsystems: human settlements, human health, the economic system, and the socio-political system. In my view, each of these subsystems is already extremely complex in terms of recovery processes and challenging to address in a single paper, let alone when combining all of them. Moreover, the paper aims to explore the interactions between these systems, considering how they may lead to a deterioration of resilience or to transformative learning. This broad focus inevitably results in a superficial conceptual discussion with rather ad-hoc assembled examples. This issue is also reflected in Figure 1, which essentially indicates that recovery trajectories are diverse, unknown, and involve interactions. Overall, I question whether this approach truly provides many new insights.
In this regard, while the authors accurately describe the examples as empirical, I would consider them anecdotal evidence to support the conceptual points the authors wish to convey. Perhaps presenting the paper as a conceptual piece would be more appropriate, as the title led me to expect something else.
Although the paper largely focuses on the potential negative impacts of consecutive disasters, considerable evidence suggests that disasters can serve as "windows of opportunity" leading to transformative learning (see e.g., Kreibich et al. 2017). This is rather briefly addressed in sections 5.3 and 4.2, but many more examples could be provided for each of the subsystems discussed. For instance, in Section 2, I am sure you can find numerous examples where disasters have led to revised spatial planning policies and/or building codes, significantly reducing the impact of subsequent disasters.
Section 6 also remains vague, offering little insight into what leads to tipping points in terms of deterioration or enhanced learning. The conclusion that these are linked to disaster intensity and frequency seems quite evident based on existing literature and cannot be derived directly from the empirical approach provided in the paper. We gain limited understanding of which discussed pathways are important or common.
Due to the broad focus, most recommendations remain vague, and it is unclear how or why the authors arrived at these specific recommendations. For example, the authors state, “…to support efficient and equitable recovery after disasters, countries need more reliable and proactive financing solutions. These may also include pre-arranged recovery financing mechanisms such as forecast-based financing, where funds are automatically released for humanitarian actions that are agreed upon in advance (IFRC & RCCC, 2020) or parametric insurance, which offers rapid, flexible payouts based on pre-defined parameters such as certain rainfall or wind speed, ensuring rapid payments in post-disaster settings (Ocampo & Moreira, 2024)." (lines 631ff). While I do not disagree with the recommendation itself, I question how this particular suggestion is derived from section 4, and why it was chosen as one of four key recommendations among potentially many others?