Integrating Climate Risk in Cultural Heritage: A Critical Review of Assessment Frameworks

Abstract

Climate change poses an escalating threat to cultural heritage (CH), driven by intensifying climate-related hazards and systemic vulnerabilities. In response, risk assessment frameworks and methodologies (RAFMs) have emerged to evaluate and guide adaptation strategies for safeguarding heritage assets. This study conducts a state-of-the-art (SotA) review of 86 unique RAFMs using a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-guided systematic approach to assess their scope, methodological rigor, alignment with global climate and disaster risk reduction (DRR) frameworks, and consistency in conceptual definitions of hazard, exposure, and vulnerability. Results reveal a growing integration of Intergovernmental Panel on Climate Change (IPCC)-based climate projections and alignment with international policy instruments such as the Sendai Framework and United Nations Sustainable Development Goals (UN SDGs). However, notable gaps persist, including definitional inconsistencies, particularly in the misapplication of vulnerability concepts; fragmented and case-specific methodologies that challenge comparability; and limited integration of intangible heritage. Best practices include participatory stakeholder engagement, scenario-based modeling, and incorporation of multi-scale risk typologies. This review advocates for more standardized, interdisciplinary, and policy-aligned frameworks that enable scalable, culturally sensitive, and action-oriented risk assessments, ultimately strengthening the resilience of cultural heritage in a changing climate.

1. Introduction

Global warming is unequivocally evident in the climate record. Average global surface temperatures in the last decade have already exceeded 1.1 °C above the 1850–1900 pre-industrial baseline levels [1], driving anthropogenic climate change and causing impacts that extend beyond environmental systems. The impact of climate change (CC) on cultural heritage (CH) is one of the aspects gaining more visibility in the international scientific community [2], due to increasing frequencies and intensities of climate hazards and potential risks to tangible and built heritage [3,4]. The exposure and vulnerability of heritage objects, sites, and monuments differ depending on a combination of physical, socio-economic, and environmental factors [5]. Thus, characterizing the potential risk is paramount in order to effectively protect and manage cultural heritage sites from the impacts of climate change.

Climate change-related risk assessment frameworks and methodologies (RAFMs) for cultural heritage are fundamental tools for understanding climate risks and formulating effective adaptation strategies [6]. In this work, RAFMs are defined as structured approaches that aim to identify and evaluate risks posed by climate change-related hazards.

Given the increasingly systemic and cross-cutting nature of climate change impacts, it is essential to adopt a risk analysis framework capable of coherently integrating the various components that shape risk across temporal and spatial scales. In this review, the conceptualization developed by the Intergovernmental Panel on Climate Change (IPCC) is adopted as the principal analytical lens. According to the IPCC’s fifth and sixth assessment reports (AR5 and AR6), climate risk arises from the interaction of three distinct but interrelated components: hazard (the potential occurrence of damaging climate-related events), exposure (the presence of people, assets, or systems in places that could be adversely affected), and vulnerability (the propensity or predisposition to be negatively impacted, shaped by sensitivity and adaptive capacity) [1]. This structure departs from earlier formulations that conflated exposure with vulnerability [7] and offers a clearer, more operational foundation for disaggregating and analyzing climate risk.

The choice of the IPCC framework is not merely conceptual. It provides a standardized and widely endorsed foundation that enhances the comparability of results, supports alignment with international climate and disaster risk reduction (DRR) policies (e.g., the Sendai Framework, Paris Agreement), and facilitates interdisciplinary integration. This is particularly important for fields such as cultural heritage, where existing risk frameworks often remain fragmented or narrowly scoped. While many existing RAFMs still rely on legacy definitions or adopt context-specific interpretations, this study uses the IPCC’s evolving framework to critically assess the degree of conceptual consistency, methodological robustness, and policy coherence of climate risk assessments applied to cultural heritage. In doing so, it seeks to contribute toward the development of more interoperable, scalable, and action-oriented RAFMs that are better suited to the complex challenges posed by climate change.

Each of these RAFMs covers different stages of the adaptation cycle [8], which is a concept that has gone through iterations in literature but fundamentally encompasses the whole cycle of risk: from risk assessment, planning, implementation, and monitoring [9,10]. Moreover, cultural heritage reaches beyond the scope of physical structures such as sites and monuments, as it also encompasses beliefs, traditions, and experiences [11]. For simplicity, the focus on cultural heritage in this work refers specifically to tangible elements, or physical “artefacts, monuments, or group of buildings and sites, and museums that have a diversity of values, including symbolic, historic, artistic, aesthetic, ethnological or anthropological, scientific and social significance” [12].

The universal applicability of existing RAFMs is uncertain, primarily because (1) multiple climate-related hazards or threats are generally quantified differently across various studies [13]; (2) the definitions of climate risk-related terminologies, particularly on vulnerability, are loosely interpreted [11]; and (3) policy frameworks for global disaster risk reduction and cultural heritage have been, from the onset, generally conceptualized separately [14].

These circumstances necessitate a more standardized, interdisciplinary, and policy-aligned approach to climate risk assessment for cultural heritage. Thus, this systematic review primarily aims to appraise, analyze, and synthesize the nature and scope of RAFMs that exist in literature, while particularly focusing on aspects related to hazard characterization, risk-related terminology, and global policy alignment. The research questions addressing these aspects are presented in

Table 1. The three research questions (RQs) presented in this study.

Research Question	Context	Follow-Up Question(s)	Relevance
Are IPCC-based climate scenarios considered in the hazard component of existing RAFMs that apply to cultural heritage?	Many hazard typologies exist, but methods vary widely across studies [13].	What are the more common and less frequently studied CC-related hazards?	Important in identifying methodological gaps and good practices in hazard assessment.
2. How are risk components, particularly exposure and vulnerability, defined by different RAFMs relative to the context of the IPCC?	Definitions of risk have evolved, especially in AR5 and AR6. Some RAFMs still use outdated or inconsistent terms [7].	Has the IPCC’s revised vulnerability concept been understood correctly, adopted, and applied consistently by RAFMs? Are there underlying reasons for these deviations?	Important in promoting consistency in risk terminology and encouraging alignment with current frameworks.
3. Aside from IPCC, to what extent are existing RAFMs aligned with global climate-related and disaster risk reduction policy convention frameworks?	DRR and heritage risk frameworks are often developed separately, with limited integration [14].	Which global DRR framework(s) serve as guidelines? Are there underlying challenges in integrating these frameworks with cultural heritage?	Alignment with DRR frameworks may ensure easier adoption by stakeholders in diverse locations and scales.

, as well as their context and relevance to advancing the CC-CH state-of-the-art (SotA). Shedding light on these questions will help emphasize the best practices in risk assessment that are implemented in various studies, provide good insights on policy recommendations and climate change adaptation, and help develop a more holistic risk framework for cultural heritage sites.

To address this objective, the paper begins by reviewing the main international frameworks and institutional efforts related to climate change and cultural heritage, setting the global policy and conceptual context. It then outlines the systematic methodology used to identify and analyze relevant risk assessment frameworks, including the selection criteria and analytical approach. The main findings are then presented and discussed according to the three guiding research questions, exploring the use of IPCC scenarios, the conceptualization of risk components, the alignment with global DRR policies, and the typological and operational limitations of current RAFMs. This section also identifies emerging best practices and links them with broader adaptation processes. Finally, the key conclusions and recommendations for advancing future research and framework development are synthesized. This paper’s structure seeks to provide a comprehensive and critical appraisal of the current state of climate risk assessment methodologies for cultural heritage, offering both analytical clarity and practical insights.

2. Review of Frameworks

2.1. Global Climate Frameworks

The IPCC is the leading international body that provides updated information on the science of climate change, as well as the associated risks and impacts, in aid of developing climate policies that can be used by governments and institutions. At its core, the sixth and latest IPCC assessment report (AR6) defines risk, which “[arises] from potential impacts of climate change as well as human responses to climate change.” In the context of climate change impacts, risk results from the dynamic interaction among its components: hazard, exposure, and vulnerability, within the affected human or environmental system [15]. Despite having a general inadequacy of focus on cultural heritage [3], IPCC has a widely recognized framework that can serve as a useful foundation for climate risk assessment that can be applied to any field, including cultural heritage.

Apart from the IPCC, this work aims to shed light on the current state of alignment between the RAFMs for cultural heritage and global frameworks aligned with the IPCC to formulate policies related to climate and disaster risk, such as the Sendai Framework for Disaster Risk Reduction 2015–2030, the 2015 United Nations (UN) Climate Change Conference in Paris, France (COP21 or Paris Agreement), and the UN Sustainable Development Goals (SDGs). The Sendai Framework is a global policy blueprint focusing on reducing disaster risk and building resilience through improved risk governance, preparedness, and investment in disaster prevention [16]. The Paris Agreement is a legally binding international treaty under the UN Framework Convention on Climate Change that aims to limit global warming to 1.5 °C and well below 2 °C above pre-industrial levels through mitigation by emissions reduction, climate adaptation, and financial commitments [17]. Meanwhile, the SDGs are a set of 17 global goals by the year 2030 adopted by the UN in 2015 to promote sustainability, social equity, and economic development, with specific targets addressing climate action and cultural heritage protection (SDG 11.4) [18].

2.2. Global Cultural Heritage Frameworks

International bodies with jurisdiction on cultural heritage include the United Nations Educational, Scientific and Cultural Organization (UNESCO), the International Centre for the Study of the Preservation and Restoration of Cultural Property (ICCROM), and the International Council on Monuments and Sites (ICOMOS) [14]. UNESCO is an international agency under the UN that promotes peace and welfare of mankind through educational, scientific, and cultural exchanges [19]. ICCROM is an intergovernmental organization that is based in Rome, Italy, and was created by UNESCO in 1959 to focus specifically on cultural property conservation and restoration through training, research, and technical support but has since become an independent body [20]. Meanwhile, ICOMOS is a non-governmental international organization that serves as an advisory body to UNESCO for advocating the conservation and preservation of cultural monuments and sites around the world through developing policies and guidelines, such as those articulated in the Venice Charter [21].

Global climate and global cultural heritage institutions remain largely decoupled from each other [22]. For instance, the ICCROM guide to risk management of cultural heritage [23] presents a framework that focuses more on a site-specific and industrial approach to risk assessment. Unlike the IPCC’s climate-focused framework, ICCROM’s approach is rooted in heritage conservation practice that emphasizes asset-level risk assessment based on diverse threats, including theft or deterioration. This highlights different risk drivers and scopes, resulting in different intervention strategies. Consequently, the guideline encompasses the assessment of climate and non-climate-related risks through a semi-quantitative approach that substantially differs from IPCC-related frameworks in terms of the conceptual understanding of risk and risk management, as well as the focus and scale of application to assets of heritage. A report [24] revealed that UNESCO-designated world heritage sites lack sufficient risk assessment and management plans, as well as the necessary capacity to implement them, despite their exposure to climate-related hazards. There are studies on disasters affecting world heritage sites and the status of their disaster risk reduction and management strategies [25,26], but these studies highlight the same inadequate institutional coordination and siloed approaches due to conceptual fragmentation and lack of common methodologies.

In recent years, the gap between DRR and CH protection plans has been gradually narrowing, with increasing efforts to integrate archaeology and heritage conservation into climate adaptation and risk management strategies. Systematic literature review papers on cultural heritage in a climate lens, as further discussed in Section 2.3, have been seminal in mapping the state of knowledge and gaps in the field. Sesana et al. [27] demonstrated how cultural heritage experts across Europe are increasingly advocating for proactive climate adaptation measures within heritage management, emphasizing the need to incorporate DRR principles into long-term preservation planning. Kohler and Rockman [28] highlighted this idea by documenting the rising engagement between IPCC and archaeological data along with heritage expertise, advocating for greater use of past conditions to connect with contemporary climate adaptation responses. Complementing these developments, Garcia [29] illustrated how the Culture in City Reconstruction and Recovery (CURE), a framework developed by UNESCO and the World Bank, positions culture as both a tool and asset for resilient urban recovery that promotes policy alignment between DRR efforts and heritage conservation in post-crisis contexts.

2.3. RAFMs and Components of Risk

Several recent systematic reviews, which serve as key references of this work, have touched upon and evaluated several of these frameworks according to their various objectives, scope, methodology, and focus, as summarized in

Table 2. Systematic literature reviews on climate risk assessment for cultural heritage. The rows are arranged based on year of publication.

	Objective	Scope (Years)	Number of Papers	Method	Focus Areas	Key Findings	Gaps Identified	Policy Relevance	Recommendations for Future Research
[3]	Investigates climate change threats to cultural heritage and resources	1900–2015	124	Systematic literature review using Web of Science	Geographical scope, methodologies, barriers	CH research is increasing but remains Eurocentric	Limited focus on adaptation implementation	Suggests policy improvements for CH adaptation	Interdisciplinary research and integration of local community values in conservation planning
[33]	Uses machine learning for disaster risk analysis in CH	1988-–020	565	Machine-human coupled analysis	Topic modeling and classification of disaster risks	Lack of interdisciplinary collaboration in CH risk studies	Low interdisciplinarity, lack of spatially explicit studies	Highlights the importance of spatial risk assessments	Foster interdisciplinary CH risk studies
[35]	Conducts a scientometric analysis of CC impacts on cultural and natural heritage	1999–2020	78	Systematic literature review using Scopus and Web of Science	Trends in scientific production, citation impact, research clusters, co-authorship networks	Found research to be in early stages with limited geographic diversity; lack of standardization in risk assessments	Limited integration of climate change policy frameworks in CH risk assessments	Emphasizes the need for better integration of adaptation strategies into CH risk assessments	Results can be a good basis for future research on the same topic
[30]	Systematically reviews 165 publications on CH and climate change (2016–2020)	2016–2020	165	Systematic review using Web of Science	Physical impacts, adaptation, policy integration	CH research is growing but remains methodologically fragmented	Weak international collaboration and a vague approach to CC processes and timescales	Urges stronger CH inclusion in adaptation policies	Further work to understand CC and its impact on the policy and practice in CH management
[36]	Synthesizes climate change impacts on tangible CH	1999–2020	100	Keyword-based search from ScienceDirect and Google Scholar	Material degradation, disaster events, long-term threats	CH degradation is accelerating due to climate extremes	Need for integrated CH conservation strategies	Links CH degradation to climate policies	Further research in other regions, estimating CC uncertainties and best practices
[11]	Identifies gaps in CH risk assessments	2005–2020	32	Qualitative thematic synthesis using Scopus and NVivo software	Community engagement, vulnerability, resilience	CH risk assessments lack community involvement	Minimal engagement with local communities	Proposes community-driven CH risk assessments	Policy-driver research; include local knowledge in CH risk planning
[31]	Assesses climate risks, vulnerability, and adaptation in CH	2016–2020	167 CH papers (and other non-CH works)	Global assessment and regional analysis	Includes natural heritage and global and regional risk assessments	CH impacts are underestimated in IPCC frameworks	Systematic underrepresentation in climate risk studies	Calls for CH inclusion in IPCC risk frameworks	Integrate heritage into IPCC, develop new modalities to assess CC impacts on heritage
[37]	Reviews frameworks for climate risk management (CRM) in CH	2017–2021	165	Two-phase systematic review using Scopus and Google Scholar	CRM adoption, quantitative vs. qualitative methods	CRM tools are underutilized at the asset level	Limited place-based assessment in CRM frameworks	Recommends multi-sectoral CRM approaches	Strengthen CRM tools for CH preservation
[32]	Strategic review of climate change impacts on CH	2015–2022	688	PRISMA-based review using Scopus, with international (non-English) coverage	Policy integration, research trends	CH literature is expanding but underrepresents intangible heritage	Policy-research gap, weak integration of CH in climate policies	Identifies policy gaps in CH climate change mitigation	Expand CH research in economic and social impacts, including non-Western contexts
[38]	Examines climate change impacts on UNESCO WH cultural properties (2008–2021)	2008–2021	58	PRISMA-based review using Scopus and Google Scholar	Hazard characterization, adaptation, UNESCO policies	CH studies focus on Europe/N. America; adaptation barriers exist	Lack of global south representation, weak policy linkages	Links UNESCO policies with climate adaptation	Enhance global collaboration, integrate CH in CC policy
[34]	Analyzes AI and IoT integration in CH preservation	2015–2024	92	PRISMA guidelines, Scopus database	Artificial Intelligence (AI) and Internet of Things (IoT) applications in CH risk management	IoT enables real-time monitoring, but adoption is uneven	Uneven adoption of AI/IoT in CH risk management	Encourages tech-driven CH conservation policies	Increase adoption of AI/IoT in CH conservation

. Although these reviews have varying scopes and are not exclusively focused on RAFMs, they provide important context for understanding the broader patterns and knowledge gaps that help frame the research questions in this work. These reviews include, among others, Fatorić and Seekamp [3], a pivotal work that was one of the first reviews that attempted to understand the state of knowledge of cultural heritage at risk from climate change threats; Orr et al. [30], a complementary work that covered later publications; also subsequent assessments that included grey literature and gave emphasis to the social perspectives on cultural heritage [31] and expanded the coverage of literature review to non-western contexts [32]; as well as works that incorporated novel technological search methods [33,34].

It is important to look at how different RAFMs incorporate or interpret the different components that make up risk in the context of climate change impacts. On the aspect of hazards, these are generally well covered by most frameworks that are applicable to cultural heritage. However, because these hazards (e.g., flooding, coastal erosion, material desiccation, changes in water salinity and acidity, etc.) are generated by a plethora of climate drivers, such as but not limited to sea level rise and changes in temperature, precipitation, and ocean chemistry [32,36], developing robust and flexible risk frameworks is still a challenging task for climate scientists and heritage practitioners [39]. Many studies resolve this by focusing on specific hazards in case-study assessments [40,41,42] or developing simplified qualitative scales and indices to represent hazard intensity and risk levels [43,44]. While still effective, the notable downsides are inevitable, including limited applicability of these methodologies and introduction of bias in the approach [13]. Additionally, the extent to which the hazard assessments of these RAFMs consider climate change scenarios hinged on global frameworks, such as the IPCC, remains insufficiently examined.

Moreover, definitions of other climate risk-related terminologies, especially vulnerability, are loosely interpreted [11]. This vagueness is evident elsewhere in many disciplines, where vulnerability definitions based on IPCC frameworks can vary depending on the inclusion or exclusion of exposure from sensitivity and adaptive capacity [7], as well as the consideration of cultural, social, and economic aspects in some works [45,46]. In the IPCC third and fourth assessment reports (TAR and AR4, respectively), vulnerability is essentially defined as a function of exposure, sensitivity, and adaptive capacity [47]. In terms of heritage, the operationalization of vulnerability has been introduced by studies such as Daire et al. [48]. Here, vulnerability was defined as a measure of various criteria that affect the coastal archaeological sites in western France, as case studies of threats from sea level rise and erosion. The inclusion of risks, site sensitivity, protection, impacts, and resistance as factors to define vulnerability presents broad parallels with earlier IPCC concepts.

The IPCC fifth and sixth assessment reports have subsequently revised their definition, where vulnerability remains as a function of sensitivity and adaptive capacity, but exposure was entirely separated as another pillar to define the components of risk. It acknowledges that vulnerability can also be shaped by socio-economic factors and hence exists independently of climate hazards [7,49,50]. Fatemi et al. [51] provided examples of these social factors affecting vulnerability to people and places in the context of disaster risk through a systematic literature review. Based on these updated definitions and underlying themes, standardizing and quantifying vulnerability of cultural heritage in relation to multiple climate-related hazards is a relatively new endeavor, with only a handful of studies, such as Giglio et al. [6] and Ravan et al. [52], providing a starting blueprint (see Section 4.4 for more details).

3. Methodology

Building upon the conceptual and methodological inconsistencies in risk assessment, this section outlines the approach and limitations for the synthesis and analysis of published literature. Referring to the typologies outlined by Grant and Booth [53], this study employs a hybrid approach akin to a state-of-the-art review and a systematic literature review of RAFMs to comprehensively examine the current literature trends on climate change and cultural heritage and to ensure alignment with the study’s objectives. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach [54] was used as a guideline to identify, select, and synthesize research on works that present RAFMs, with emphasis on the structure, transparency, and replicability of the search process. As with other systematic literature reviews, the first step is to formulate specific research questions (RQs) [55,56]; in this case, through a root cause analysis on the systemic knowledge gap between CC and CH (Figure 1). Existing RAFMs attempt to fill this gap, but challenges on methodological and theoretical aspects still remain. The methodological aspect pertains to diverse approaches to characterizing the hazards in the risk assessment, as well as inconsistencies in the inclusion of climate change scenarios, while theoretical issues involve varied risk definitions, and interpretations, and ontological differences from global policy frameworks. These considerations form the foundation for the three RQs presented in this review.

3.1. Selection Process

The screening and selection of literature followed a multi-stage process based on structured eligibility and inclusion-exclusion criteria (Figure 2). For the eligibility, the initial search was conducted using the Web of Science (WoS), an online platform owned by Clarivate that provides access to over 34,000 academic journals, conference proceedings, and various dossiers. Boolean search strings and operators were used to initially identify potentially relevant literature on the topic. Boolean search operators were used to construct the search string: “climate change” AND “cultural heritage” AND (“risk assessment” OR “risk analysis” OR “framework”) AND (“adaptation” OR “mitigation” OR “management”).

The search string was applied across all search fields to maximize inclusivity, which returned n = 372 results. The terms “adaptation”, “mitigation”, and “management” were included to capture potential frameworks that also cover climate risk adaptation in their methodologies. The date range of the search was narrowed down to works published online from 1 January 2015 to 31 December 2024 to (1) see potential trends in relation to the amount of research completed and the emergence of common scope and themes on the topic, and (2) see if these studies have adopted the updated concepts of risk, as reflected in and after the publication of IPCC AR5 and the subsequent publication of IPCC AR6, within the last decade.

Next, title and abstract screening was conducted in three rounds of inclusion-exclusion of works, starting with the 372 initial titles. In the first round, titles and abstracts were reviewed based on thematic relevance. A total of 197 titles were excluded in this stage, leaving 175. The purpose of this is to exclude works not directly related to both climate change and cultural heritage. The discarded titles included studies focusing on natural heritage, agriculture, urban or financial risk assessments, works centered on indoor climate monitoring or intangible cultural heritage, and general literature reviews and/or synthesis papers unrelated to a specific RAFM. In parallel, additional works from gray literature and non-WoS articles (22) were included based on expert judgment by the authors, as they were evaluated as works that potentially introduce or discuss a climate change-based RAFM for heritage. Although these referred to reports and documents from non-academic bodies and were often not peer-reviewed, these works were still considered after abstract review, as they met the initial selection criteria despite not being part of the WoS database. This stage resulted in a combined pool of 197 works (175 from WoS + 22 additional).

In the second round, scope and accessibility were used as the criteria for exclusion. Forty-six (46) more titles were excluded from the 197 titles, which narrowed the list to 151. The excluded titles were mostly literature reviews or synthesis papers (36), or the content apart from the abstract was inaccessible for further review (10).

In the third round, ineligible studies were further excluded based on methodological relevance. Twenty-eight (28) studies were excluded for focusing primarily on climate adaptation, management, policy, and socio-economic aspects, rather than directly presenting RAFMs for cultural heritage. Additionally, exclusions covered studies that discuss only tech-related methodologies (i.e., utilization and evaluation of technologies and tools used in data gathering for the risk framework, and not presenting and evaluating a framework per se) (18); studies that explicitly or implicitly discuss only one of the components of risk assessment (hazard, exposure, or vulnerability) (11); and studies that duplicate the same framework, i.e., those that discuss or apply exactly the same RAFMs, despite having been applied in different settings or written by a different set of authors (8). In cases of duplication in the framework, expert judgment by the authors was comparing by comparing the papers to identify the study that best encapsulates the entire framework, which was then retained to represent the framework in this analysis. After this final round, a total of 86 unique RAFMs were identified for inclusion in the analysis and synthesis (see Supplemental table for the list in Supplementary Materials).

3.2. Analysis and Synthesis

Full text reading was completed on the remaining 86 titles that constitute unique RAFMs to gain an in-depth understanding of the topic using a combined quantitative-qualitative analysis. This content synthesis was guided by the three research questions that were presented. Initially, the papers were quantitatively analyzed to address the RQs and derive statistical insights. For example, the analysis examined the number of frameworks incorporating IPCC-based climate change scenarios in hazard assessment (RQ1), as well as grouping and counting the frameworks relative to their interpretation of risk components (RQ2) and analyzing the number of frameworks that reference or implicitly incorporate parts of the global DRR policy framework(s) as a basis for their respective approaches (RQ3). Further qualitative analysis was conducted afterwards to explore the underlying context that these statistical figures may imply. This approach in the synthesis provides a comprehensive coverage of the systematic review, which gives way to the subsequent discussions in Section 4 to answer the RQs and give details on the effectiveness and limitations of current climate-related RAFMs on cultural heritage.

3.3. Limitations

The database used in this study was Web of Science. While previous comparisons with other databases, such as Scopus reveal that while WoS has fewer indexed titles than Scopus, it still has a strong disciplinary focus in the fields of natural sciences, engineering, and biomedical research [57]. Compared to Google Scholar, WoS has a smaller citation list, but its entries generally reflect higher scientific impact, as approximately half of Google Scholar’s unique citations consist of lower-impact sources such as book chapters, unpublished materials, reports, and other grey literature [58]. Although WoS includes many high-impact journals, it may underrepresent humanities-focused research, where publishing in edited books or other means remains common [59]. To address this limitation, relevant non-WoS and grey literature were included based on expert judgment. Also, preceding systematic literature reviews demonstrate the capability of WoS alone in obtaining comprehensive insights in the field of CC and CH, following the lead of previous seminal works [3,30,35]. It is also important to note that both Scopus and WoS are disproportionately biased in the English language. The Boolean search was limited to papers written in English given the constraints of the WoS database and the challenges of translating and interpretating non-English works.

4. Results

4.1. RQ1: Are IPCC-Based Climate Scenarios Considered in the Hazard Component of Existing RAFMs That Apply to Cultural Heritage?

Out of the analyzed RAFMs published from 2015 to 2024, more than three-quarters, or 79% (68 titles), incorporated IPCC-based climate scenarios in their respective hazard assessments, while 21% (18 titles) did not. This suggests that a good majority of risk frameworks have a high level of awareness of the potential impacts of these hazards on cultural heritage and that incorporating climate scenarios is important to understand how risk could potentially change from the present to the future. A growing trend is noticeable in the number of studies integrating IPCC-based climate scenarios, particularly in recent years, with a surge observed in 2020 and 2024 (Figure 3). Only 8% of the studies in the last five years (2020 to 2024) are not IPCC-based compared to 48% in 2015 to 2019. Older studies often omit explicit consideration of climate projections in hazard assessments, while newer papers are more likely to include IPCC scenarios in their frameworks. While there are still works that do not conform to IPCC guidelines, their proportion has decreased in recent years. These findings suggest a growing consensus and converging themes between climate science and cultural heritage.

The substantial number of frameworks that do not explicitly reference IPCC can be roughly grouped according to their intent and reasoning. Some prioritize past and present climatic risks over long-term future projections [60,61], focus on current risk factors and adaptation strategies from these hazards [45,62], or examine hazards such as landslides and wildfires, which are secondary effects or have indirect relationships to climate change [40,63].

Meanwhile, the hazard typology of these frameworks reveals that flooding is the most studied hazard (n = 50). It appears in these categorizations: “flooding” or “flood/s”, which are often referenced without further distinction [64,65,66]; coastal flooding [67]; fluvial flooding [68]; and in many cases, combined with other hazard assessments for frameworks that consider multiple hazards. Also, sea level rise (19), temperature-related (13), coastal erosion (12), precipitation-related (12), landslide/erosion-related (8), and storm surge (5) are the other often-considered hazards in these frameworks. Other works highlight hazards that are seldom studied, such as wildfires [40,69], permafrost thawing [70,71], and salt weathering [42].

The breakdown of hazard assessments reveals that there is an inherent predisposition of studies to carry out risk assessments that deal with the more common hazards, partly due to the convenience brought by the availability of more datasets, more straightforward characterization of hazard variables, and substantial number of preceding assessments as references. This highlights the underrepresentation of the other climate hazards in the literature (e.g., permafrost thawing and wildfires), due to limited precedent datasets and less relevance on a regional or global scale. Knowing how RAFMs present their hazard assessments further reveals that while multiple hazard assessments exist, there is no consensus on which approach works best, as the effectiveness of each method depends on its application on a case-to-case basis. Multi-hazard assessments often fail to be robust because representing different quantitative and qualitative units, spatio-temporal scales, and intensities of hazards is often difficult [13]. Nevertheless, this can be achieved by gaining a deeper understanding of the different types of climate hazards, their interdependencies, and impacts on heritage assets [36]; adopting best practices and effective, scalable methods for hazard assessment; and integrating IPCC-based climate scenarios into future hazard modeling. Likewise, hazard datasets are also often enriched with inputs coming from multiple stakeholders [13], which adds more value in incorporating multidisciplinary knowledge from local communities, specialists, and heritage managers when hazard assessments are performed.

4.2. RQ2: How Are Risk Components, Particularly Exposure and Vulnerability, Defined by Different RAFMs Relative to the Context of IPCC?

It is worth noting that the concept of risk in cultural heritage has been gaining wider adoption. Figure 4 shows that more than three-fourths of the RAFMs (76% or n = 65) align their definitions of risk either with IPCC AR5 or AR6. Only four (4) frameworks base their definitions of vulnerability on the older IPCC TAR or AR4 documents. Meanwhile, the remaining 17 either have risk concepts and definitions that are not clearly defined in their respective methodologies or consistent with any IPCC assessment report. Similar to the growing consensus on the concepts of hazard, this statistic signals better agreement with the updated risk reduction principles.

However, while claiming alignment with the IPCC AR5 or AR6 definitions is one thing, asserting misinterpreted definitions in their respective frameworks is another. Forty-eight percent (48% or n = 31) of the frameworks that explicitly state or imply adherence to IPCC AR5 and/or AR6 either misapply vulnerability definitions (i.e., that exposure is included as a factor that constitutes vulnerability) or fail to clearly justify the basis of their interpretations. In particular, a number of these frameworks still include exposure as a factor within vulnerability instead of being a separate and equal component that helps characterize the risk, which, in essence, contradicts the updated definitions in IPCC AR5 and AR6. The substantial number of frameworks that follow this definition suggests the following key issues. First, it underscores a clear misalignment in understanding the role of exposure within the context of vulnerability and the overall risk to cultural sites and properties. Second, these risk frameworks reveal that they may be relying on outdated disaster risk models, which could result in challenges to effectively operationalize vulnerability within the cultural heritage context. Furthermore, it suggests that the cultural heritage field as a collective body remains slow to adopt updated climate risk principles and encounters substantial barriers to fully integrating with climate science, resulting in the limited application of conceptual standards and tools.

On the subject of exposure, typologies of cultural heritage sites widely vary based on location, age, material, scale, and other factors. These inconsistencies may arise from the diverse sites, materials, and monuments that frameworks prioritize when characterizing exposure, making it difficult to establish a universal understanding of this risk component. For example, some frameworks focus on geospatial exposure [72,73], while others characterize exposure through material degradation [42,71] or through material/site inundation [74,75]. This variation can contribute to inconsistencies in how exposure is defined and quantified, which may lead to overlaps in vulnerability assessments and methodological inconsistencies in risk calculations, and eventually highlights the need for standardizing the approach. Therefore, having consensus on the same basis and definitions could help build on proposed approaches such as Ravan et al. [52], which could then give a better structure to conceptualizing exposure and vulnerability and, ultimately, build on a more robust and standardized framework.

A theoretical implication of this conceptual misalignment is having an overestimated or underestimated perception of risk when quantified. CH sites that are highly exposed to CC-related hazards, for example, may receive disproportionately higher risk assessment scores, which can skew risk prioritization when sites are ranked according to priorities for preservation. Meanwhile, sites that are highly sensitive and have lower adaptive capacity but are in low-exposure areas may appear less vulnerable than they truly are, which can lead to less prioritization for climate change adaptation. This poses practical challenges to CH adaptation because merging exposure and vulnerability may blur the distinction between different intervention strategies that deal with exposure reduction (e.g., physical barriers, relocation of heritage sites), sensitivity reduction (material reinforcement, improved conservation techniques), or improvement of adaptive capacity (community training, increased maintenance funding). Frameworks that vaguely define or inconsistently apply vulnerability and risk concepts could have implications for maladaptation. Adaptation strategies may become ineffective or inadequate when efforts are overly focused on reducing exposure while neglecting measures that address sensitivity and adaptive capacity. Ultimately, this imbalance can leave heritage sites highly sensitive and less adaptable to hazards.

4.3. RQ3: Aside from IPCC, to What Extent Are Existing RAFMs Aligned with Global Climate-Related and Disaster Risk Reduction Policy Convention Frameworks?

Another perspective that this study looks at is how these RAFMs approach congruence with global disaster risk reduction policies. It is notable that the majority of these frameworks recognize IPCC. However, the IPCC is an international body that is only mandated to provide scientific evidence and consensus on the Earth’s climate upon which these studies can build their frameworks and methodologies. It does not prescribe policy commitments nor recommend the working foundations for specific risk management strategies. Looking at how these RAFMs also adhere to DRR policy frameworks is crucial, because these DRR policy frameworks go beyond understanding the climate and knowing the risks. Further, climate change is just one of many central points addressed by these global frameworks. These policy frameworks are complementary to the IPCC and are geared more as blueprints that encourage and bind governments and other institutions, including those working with cultural heritage, to commit to actionable risk management.

Each of the RAFMs covered in this study was reviewed to determine if they either explicitly mention, incorporate parts of, implicitly align with, or follow strategies from these DRR policy frameworks. Of the 86 titles considered, 84% (n = 72) reference one or several global frameworks as the basis for their approaches. The main frameworks that were often referenced are the UN SDGs (32), the Sendai Framework (36), and the COP 21 Paris Agreement (18). The Hyogo Framework for Action was referenced in a few papers (5), along with other related policy frameworks such as the European Union (EU) Flood Risk Directive (3) [76], and the UNESCO Strategy for Risk Reduction in World Heritage properties [77] by one paper. In addition to the global DRR frameworks, region-specific policies such as the EU Flood Risk Directive focus on water-related risks. While not a global framework, its integration suggests that some RAFMs are also in sync with pan-regional policy frameworks. As for the limited reference to the Hyogo Framework, this is likely due to the replacement of the more updated Sendai Framework, which changed from a top-down process to a bottom-up approach in managing disasters [14]. In general, the frameworks that align with or explicitly reference the Sendai or UN SDGs tend to emphasize resilience and sustainability [78,79,80,81,82], while those that align with or reference the Paris Agreement are more common in studies that incorporate climate projections [67,83,84].

The large number of studies referencing DRR frameworks reflects a growing awareness of their importance in cultural heritage, but a closer analysis reveals persistent gaps in this area. A significant number of RAFMs (n = 14) do not reference any global policy framework or show only partial integration, either by only mentioning the DRR framework without incorporating it into their methodology or by focusing solely on cultural heritage perspectives and risk assessment on a localized scale while barely aligning with broader policy goals. This reflects a governance gap: an underlying challenge where local risk-related policies and frameworks exist but remain partially or entirely misaligned with global counterparts. These gaps likely stem from the institutional separation between international bodies concerned with CC and CH, as well as from a siloed approach in some studies, where the primary focus is heritage [85,86], or the framework is developed and focused at a local scale [73,87]. This gap hinders risk managers and policymakers in adopting RAFMs that lack DRR framework alignment, complicates the discovery and sharing of best practices, and ultimately results in missed opportunities for interdisciplinary collaboration.

A closer review of the global DRR policy frameworks generally reveals another challenge: that cultural heritage is a secondary concern in guideline development and rarely reflects on-the-ground focus on the risks that affect heritage [14]. However, these frameworks are gradually evolving to place greater emphasis on the protection of cultural heritage assets. For instance, the transition from the past Hyogo Framework, which made only minimal reference to cultural heritage, to the present UN Sustainable Development Goals (SDGs), where heritage protection is explicitly incorporated into a dedicated development goal, reflects significant progress [14]. Ultimately, ensuring that cultural heritage receives equal focus within global DRR policies is essential for congruence to bridge the institutional and disciplinary divide and, from a top-down approach, facilitate the integration of global policy frameworks into heritage risk assessment methodologies. This disconnect is not unique to cultural heritage. Fragmentation between other fields and climate adaptation/DRR strategies has been noted as well, for example, in public health [56,88]. This reflects a broader systemic issue that cannot be fully resolved in the heritage sector alone. Exploring how other fields have integrated climate and disaster risk in their risk frameworks may offer useful strategies for heritage.

On the other hand, a bottom-up approach is also needed to address the governance gap. Despite this misalignment, local and indigenous governance structures often demonstrate considerable adaptive capacity and practical knowledge, particularly in the monitoring, documentation, and management of heritage sites under climate threat. For example, Carmichael et al. [89] show that local cultural protocols and landscape literacy are important considerations that can better guide assessments of site vulnerability more than metrics from top-down approaches. Carmichael et al. [90] also highlighted indigenous participatory risk analysis tools tailored to specific sites and values, choosing context-appropriate measures such as augmented reality documentation over costly or culturally inappropriate relocation strategies. Such efforts not only build site resilience but also strengthen local adaptive capacity and enhance the legitimacy of adaptation plans by embedding them within community-held values and experiences. Recognizing and supporting these grassroots capacities is essential not only for filling the governance gap but also for ensuring that DRR measures are sustainable, culturally grounded, and based on more inclusive and context-sensitive policies and approaches.

4.4. Typologies and Limitations of Frameworks

Several classifications or ex-ante typologies have been presented and discussed in detail in some previous works to describe the nature, approach, and scope of RAFMs [11,13,30,32,35,36,38]. The common themes of classifications are based on their geographic focus, hazard-specific or multi-hazard orientation, use of qualitative or quantitative methods, and their level of stakeholder engagement.

Beyond these typologies, several framework shortcomings come up. First, common methodological approaches center on modeling common climate hazards and geospatial exposure mapping as explained in Section 4.1 (RQ1), often at the expense of deeper vulnerability and risk analysis. This results in an inconsistent separation between hazard assessment and risk assessment [11]. Second, as discussed in Section 4.1 (RQ1) and Section 4.3 (RQ3), many RAFMs are context-based and were developed to cater to specific case studies, leading to a fragmented list of risk assessment tools with varying degrees of applicability, transferability, and alignment with global climate and policy frameworks. Lastly, these frameworks fall short in capturing the overall risk due to lack of understanding and definitional ambiguities regarding exposure and vulnerability, as discussed in Section 4.2. In relation to answering RQ2, this limitation is especially impactful because it affects not just conceptual clarity but also how risk levels are determined and acted upon considering site prioritization for conservation and adaptation. Many frameworks still misapply or conflate vulnerability and exposure or reduce vulnerability to material sensitivity alone, excluding critical aspects that define the socio-cultural dimensions of vulnerability, such as adaptive capacity and systemic resilience [7,51]. This not only hinders comparative assessments but also leads to skewed prioritization of outcomes, especially in multi-site or multi-hazard contexts where vulnerability may not correlate directly with exposure. Figure 5 is an illustrative example of how updated IPCC definitions can clarify the distinct but interacting roles of exposure and vulnerability in cultural heritage risk assessment. It helps highlight that high exposure does not necessarily imply high vulnerability (and vice versa) and helps shift the focus from hazard-centric models to more distinct and nuanced interpretations of what makes heritage assets truly at risk. It is important to note that these examples are illustrative of generalized present-day levels of exposure and vulnerability, and both components are dynamic, i.e., heritage objects in polar and desert regions may face an increased number of hazards with respect to evolving climate scenarios. This highlights the need for a temporal assessment of hazard evolution and its interaction with exposure and vulnerability in risk frameworks to reflect these changing conditions.

The broader field remains conceptually fragmented and misaligned despite the increasing attention from recent vulnerability-focused studies. For example, the Climate Vulnerability Index (CVI) [46] is a good approach that allows for rapid assessment using participatory methods that evaluate both Outstanding Universal Value (OUV) and community vulnerability. It has methodological strengths by promoting cross-sectoral stakeholder dialogue in the risk assessment process, making it suitable for community-oriented heritage governance and sites with limited technical capacity. However, the CVI framework presents outdated conceptualizations of vulnerability and risk from IPCC AR4, in light of crucial conceptual updates as detailed in AR5 and AR6. Furthermore, its simplification of site-specific hazard modeling limits its applicability where climate projections and scenario-based hazard assessments are necessary for long-term planning. To strengthen its applicability, future versions of the CVI could adopt updated vulnerability concepts, include scenario-based climate projections, and integrate site-specific hazard intensities.

Another structured approach tailored for cultural heritage is presented by ICCROM [23] as a guide for risk management on cultural heritage. This guideline uses the ABC Method for risk management [91] and is hinged on the International Organization for Standardization (ISO) 31000:2009 (International Standards for Risk Management), which is then aligned with several UN SDGs [92]. It offers a clear process for risk identification, analysis, evaluation, and control, as supported by principles such as stakeholder engagement and continuous review. However, it diverges from the IPCC in key conceptual terms. ICCROM defines risk as “the chance of something happening that will have a negative impact on our objectives” [23], while the ABC Method defines risk as “the possibility of a loss of value to the heritage asset” [91]. These definitions, although broad and flexible, can obscure distinctions between climate hazards, material sensitivity, and adaptive capacity. When viewed from a climate risk perspective, this broad framing can lead to failures in clearly defining specific climate-related hazards for assessment, as well as distinguishing spatial scales and temporal dynamics crucial for long-term planning. Also, although ICCROM’s guide is still circulating in its original 2016 edition, the reference ISO 31000:2009 benchmark was already withdrawn and revised to 31000:2018 [92], superseding the 2009 version on which the guide is based. Nonetheless, the ICCROM framework’s emphasis on iterative risk management, organizational responsibility, and integration of values-based decision-making is a notable strength, and it can be conceptually merged with IPCC to achieve procedural robustness and climate-specific analytical clarity. The framework could, for example, be modified into a hybrid framework by embedding climate-related risk components into ICCROM’s risk management cycle.

Later studies present improvements by attempting to refine vulnerability assessments in ways that incorporate IPCC guidance while being operationally relevant to heritage. Ravan et al. [52] proposed a modular vulnerability framework that disaggregates sensitivity and adaptive capacity into site-specific and institutional components, supported by weighted scoring systems. Giglio et al. [6] presented a material-based index system calibrated for heritage structures across Europe, introducing quantifiable metrics that can support comparative assessments. Both frameworks demonstrate that the integration of climate data, stakeholder participation, and technical indicators is not only feasible but also necessary to advance vulnerability assessments.

Lastly, it is important to acknowledge that a crucial limitation persists in the treatment of intangible cultural heritage (ICH). While this study focuses on tangible CH, the stark marginalization of ICH in existing RAFMs is evident and problematic. The British Council’s strategic review [32] affirms that ICH, despite being vital for community resilience and identity, is poorly represented in climate change discourse and adaptation planning. Given its fundamentally distinct nature from tangible CH, being non-physical, experiential, and rooted in evolving practices, this paper argues that it is necessary for ICH to have a separate but complementary framework in climate risk assessment. Attempting to fit ICH into RAFMs designed for material cultural heritage risks oversimplifying its value and diminishing its cultural relevance. Such a framework should be grounded in a combination of climate and social science, incorporate ethnographic methods, and prioritize community-led documentation and safeguarding initiatives.

4.5. Best Practices and Moving Forward

In light of the research questions explored and the underlying challenges discussed, it becomes increasingly evident that the effectiveness of risk assessment frameworks is linked to several key characteristics: integration of both quantitative and qualitative indicators that effectively represent hazard, exposure, and vulnerability; adaptability and flexibility across different scales and types of cultural heritage; incorporation of climate change scenarios; alignment with climate policy frameworks; and incorporation of multi-stakeholder engagement procedures. These criteria are, to varying extents, already reflected in existing frameworks, but what is still needed is to critically identify and evaluate these elements, replicate effective methodological features, and refine or expand upon aspects that remain underdeveloped.

From this, several good practices can be highlighted from existing RAFMs that point the way forward. Recent efforts such as Battisti et al. [93] showcase how multidisciplinary and data-driven approaches can help risk assessments through co-created taxonomies, harmonized conceptual definitions of risk and its components, and utilization of artificial intelligence. Their approach presents quantifiable indices for both natural and anthropic risks, advancing the field toward more scalable, interoperable, and decision-supportive tools.

Table 3. Three-level structure of risk assessment approaches for climate-related hazards in cultural heritage, illustrating differences in scale, methodology, data needs, and expected outputs. Relevant publications are cited for each level to highlight representative frameworks that employ these methodologies in their studies.

Level of Risk Assessment	Approach	Data Prerequisites	Methodologies	Data Output	Examples of Publications That Present RAFMs with Varying Scope and Methodologies
Level 1 Global/ Regional Assessment	More qualitative	Hazard and heritage material typologies Risk indicators	Desktop surveys Risk perception surveys Stakeholder mapping	Thematic maps Stakeholder network diagrams Qualitative risk matrices	Level 1: [94,95]	Levels 1 and 2: [39,46,74,96]		Levels 1, 2, and 3: [70]
Level 2 Regional/ Local Assessment	Hybrid qualitative-quantitative	Risk indicators Stakeholder maps Spatial datasets	Reconnaissance surveys Expert judgment Ethnographic surveys Risk index calculations Spatial analysis Participatory mapping	Hazard, exposure, vulnerability, and risk indices Risk maps Asset prioritization	Level 2: [40,62,68,71,72,73,75,87,97]		Levels 2 and 3: [98,99]
Level 3 Site-specific/Heritage Asset-based Assessment	More quantitative Hybrid qualitative-quantitative	Spatial datasets Atmospheric models Material-specific properties	Fieldwork Ethnographic surveys Numerical modeling Downscaling Experiments	Material-specific vulnerability curves Risk projections Site-specific impact assessments	Level 3: [100,101]

illustrates a three-tier structure that represents different levels of risk assessment: from global/regional and less detailed to site-specific and more detailed scope. Within these levels are common approaches, data requirements, methodologies, and typical output types, while identifying publications that showcase methodological strengths across different contexts. Placing the frameworks in this context serves as a strategic reference point to recognize the current state of practice and the varying degrees to which the key framework characteristics are addressed. This framing underscores where opportunities for methodological improvement can be made. For instance, while site-specific risk assessments tend to excel in contextual accuracy and material-specific analysis, they sometimes break down when scaled up and/or misalign when linked with broader global policy frameworks. Conversely, global or regional frameworks may offer robust indicator sets and broader thematic coverage, but they could fall short in local relevance or implementation feasibility across different heritage sites and objects. The hybrid models at the regional/local level provide a promising middle ground, but they may, for example, still require refinement in balancing expert judgment with participatory mechanisms in their risk calculations.

Aside from noting how current RAFMs are structured and applied across different spatial scales and methodologies, it is also helpful to determine how these frameworks facilitate knowledge transfer and decision-making within their approaches. Accordingly, highlighting the shift from risk diagnostics toward adaptation-oriented strategies is important, as presented in the reviews by Adetunji and MacKee [37] and Santangelo et al. [102]. These systematic reviews discuss in detail several frameworks that not only assess the risk but also transcend into climate action by proposing adaptation measures. Nonetheless, covering the whole adaptation process is not a prerequisite for an RAFM to be an effective approach, as the effectiveness of the framework depends on the execution of the methodology and quality of the results, rather than the extent of coverage.

Moreover, frameworks that transcend risk assessment and also focus on response, adaptation, monitoring, and evaluation are more holistic and have the upper hand when it comes to their potential and applicability when tested across many cultural heritage settings [37]. Ultimately, it is better for RAFMs to bridge the gap between the phases of risk assessment and risk adaptation to be more effective and to gain added value in their respective results. When risk assessments are tied to actionable outcomes, they provide heritage managers with the tools to translate risk knowledge into measures that promote heritage resilience. Several of the frameworks analyzed in this study attempt to do this by recommending strategies in response to risk. While most RAFMs propose adaptation strategies, the extent to which they provide concrete implementation pathways varies. Addressing these challenges is critical for RAFMs to move beyond theoretical recommendations into practical, site-specific adaptation.

Beyond their diagnostic function, RAFMs should be understood as the initial and essential phase of the climate adaptation cycle. As conceptualized in the adaptation literature [8,9], adaptation is not a single event but a process comprising several stages: risk assessment, planning, implementation, monitoring, and iterative learning. Within this cycle, risk assessment plays a foundational role by identifying the specific hazards, exposures, and vulnerabilities that shape the climate-related threats to cultural heritage. This diagnostic phase not only informs prioritization but also determines the types of interventions that are both feasible and effective under given climatic and socio-institutional conditions.

In the context of archaeological sites, many of which are located in coastal, permafrost, or semi-arid regions particularly sensitive to climate change, the need for robust risk assessments is even more acute. Such sites often have limited adaptive capacity due to their material fragility, remoteness, or lack of institutional support. Through detailed risk characterization, including scenario-based modeling and vulnerability profiling, RAFMs can help identify tailored adaptation measures such as physical protection (e.g., drainage systems, flooding protections), site reburial or relocation, community-led monitoring, or policy safeguards. Without this analytical foundation, adaptation risks becoming reactive, fragmented, or misaligned with the actual drivers of risk. Thus, integrating RAFMs into a full-cycle adaptation framework is not optional but critical for ensuring the long-term preservation and resilience of archaeological heritage in a changing climate.

5. Conclusions

This review confirms that meaningful progress has been achieved in the development of climate change risk assessment frameworks and methodologies (RAFMs) for cultural heritage. However, substantial conceptual, methodological, and operational gaps persist, limiting their capacity to inform and support effective adaptation strategies. One of the most significant advances identified is the growing incorporation of IPCC-based climate scenarios into hazard assessments. This reflects an increasing convergence between climate science and heritage risk analysis. Nonetheless, important inconsistencies remain in the way hazards are characterized, particularly regarding less-studied phenomena such as permafrost thaw or wildfires, which are expected to intensify with climate change. The lack of standardized multi-hazard approaches further complicates efforts to ensure methodological robustness and comparability across sites and regions.

A central finding of this review is the persistent conceptual misalignment in how key risk components, particularly vulnerability and exposure, are defined and operationalized. Many RAFMs still adhere to outdated formulations that conflate exposure with vulnerability, contrary to the structure proposed by the IPCC since AR5, where risk is conceptualized as the interaction between hazard, exposure, and vulnerability as distinct but interrelated components. This definitional fragmentation hampers comparability among frameworks, distorts risk estimations, and may lead to maladaptive strategies that prioritize highly exposed sites over those with critical sensitivity or low adaptive capacity.

The degree of alignment with global DRR policy frameworks such as the Sendai Framework, the Paris Agreement, and the UN Sustainable Development Goals remains uneven. While a majority of RAFMs reference these instruments, few incorporate them substantively into their methodological core. This partial integration limits the institutional utility and scalability of many frameworks, particularly in policy or funding contexts that demand coherence with international agendas. At the same time, the disconnect between climate policy and heritage governance continues to create barriers to interdisciplinary collaboration and knowledge transfer. Future work should also investigate how other sectors have bridged the divide between CC and DRR to determine whether transferable solutions exist or if cross-disciplinary frameworks can be co-developed.

Operationally, the dominance of hazard-focused or geospatial exposure-based methodologies often overshadows the social, institutional, and cultural dimensions of vulnerability. Despite this, several best practices are emerging. These include participatory frameworks that engage local actors, modular indices that allow for context-sensitive vulnerability assessments, and hybrid approaches combining quantitative modeling with qualitative, values-based indicators. Such innovations point to promising directions for future development.

Critically, this review highlights that risk assessment must be understood not merely as a diagnostic exercise but as the foundation for climate adaptation. Proper characterization of risk enables strategic prioritization of interventions, more efficient allocation of resources, and the tailoring of adaptation strategies to specific vulnerabilities, whether through reducing exposure, enhancing adaptive capacity, or addressing material sensitivities. The disaggregation of risk into its core components also facilitates the identification of appropriate responses: for instance, engineering solutions may reduce exposure, while community engagement and institutional capacity-building can improve adaptive capacity.

Furthermore, RAFMs that adopt iterative, modular designs are more likely to support dynamic adaptation planning, allowing updates as new data or climate projections emerge. They also provide a platform for linking global climate narratives with place-based cultural values, making adaptation strategies more socially embedded and culturally meaningful. In this sense, risk assessment becomes not just a technical process but a bridge between scientific understanding, policy frameworks, and heritage management practices.

Looking ahead, the development of next-generation RAFMs for cultural heritage must address several priorities. These include standardizing conceptual definitions and indicators; developing tools that are scalable across different types of heritage and geographic contexts; integrating neglected hazards; developing specific tools and complementary frameworks focusing on intangible cultural heritage; and embedding adaptation pathways directly into the structure of the assessment. Interdisciplinary collaboration and co-production of knowledge with local communities will be essential to ensuring that these frameworks are inclusive, actionable, and legitimate.

Ultimately, characterizing risk is not an endpoint but a critical step toward building resilience. In the face of a changing climate, RAFMs must be equipped to translate knowledge into action, enabling the heritage sector to navigate uncertainty, safeguard cultural values, and ensure the continuity of human heritage for future generations.