Research

INTRODUCTION

The concept of resilience has gained increasing prominence in scientific literature and policy discussions (Folke 2006, Walker et al. 2004). Resilience introduced a complex systems perspective to social-ecological systems analyses, emphasizing that system responses to changes can be nonlinear and include regime shifts (Folke 2006, Walker et al. 2004). Numerous definitions and metrics have been developed to operationalize resilience (Baggio et al. 2015, Dakos and Kéfi 2022). However, there are challenges to measuring resilience and being consistent with theoretical work (Allen et al. 2019, Quinlan et al. 2016).

Diversity and agency are recognized as important concepts in the resilience literature. Diversity, broadly defined as the range of components within a system (e.g., species, functions, or institutions, Stirling 2007), has been fundamental to shaping the resilience concept, with evidence from ecology and economics demonstrating how diversity can enhance system stability and adaptability (Biggs et al. 2015, Elmqvist et al. 2003, Walker et al. 2023). Agency, broadly understood as the capability of individuals and institutions to act and make decisions (Duncan 2019), gained importance in resilience theory recently, as it acknowledges the critical roles these actors play in building resilience, shifting the focus toward local-level analyses (Brown and Westaway 2011, Cote and Nightingale 2012, Hahn and Nykvist 2017). Although diversity and agency are recognized as important, incorporating them into resilience metrics can be challenging because metrics often use simpler conceptualizations of resilience (Allen et al. 2019). In this study, we conducted a systematic review to assess how research on social-ecological resilience currently integrates diversity and agency into quantitative metrics and identify future opportunities to enhance their inclusion using insights from other research fields. In this study, we outline our focus on social-ecological resilience and quantitative resilience metrics, discuss the significance of diversity and agency for resilience, and provide an overview of our methodological approach.

Social-ecological resilience is an interdisciplinarity concept

Social-ecological resilience is rooted in ecological resilience (Folke 2006, Walker et al. 2004). Ecological resilience was first introduced in Holling’s (1973) seminal paper, which emphasized ecosystem stability and the existence of systems with multiple stable states. With the recognition of the intertwined relationship of humans and nature (Adger 2000, Berkes et al. 2002), the concept of ecological resilience was expanded to incorporate theories and methods from social sciences (Brown 2014, Carpenter et al. 2001, Walker et al. 2004). Social-ecological resilience theory builds on this long ecological tradition, which has influenced definitions of resilience along multiple research fields (Brand and Jax 2007, Folke 2006, Quinlan et al. 2016). A citation network analysis shows that social-ecological resilience is highly interdisciplinary and more frequently connected to other research fields than alternative conceptualizations of resilience, being strongly linked to ecology, but also with social sciences, psychology, and to a lesser extent, engineering fields (Baggio et al. 2015).

Definitions of resilience vary widely across research fields, although some common ideas are shared. Resilience is understood as the capacity of a system to deal with disturbances across different fields (Baggio et al. 2015). The concept of recovery from disturbance is relevant to most research fields: in engineering, e.g., recovery of water provision after a pipe failure (Pagano et al. 2019); in psychology, e.g., recovery of well-being after a traumatic event (Davydov et al. 2010); and in social-ecological systems, e.g., recovery of grass in rangelands after intense grazing (Anderies et al. 2002). Many fields, including psychology, community, and social-ecological resilience focus on adaptability, transformability, and learning besides recovery capabilities (Brown and Westaway 2011, Davydov et al. 2010, Walker et al. 2004). Social-ecological resilience, for example, focuses on maintaining structure and function against disruptions, which may require transformations rather than simply recovering pre-disturbance configurations (Folke 2006, Walker et al. 2004). Mathematical formalizations of resilience also show metrics from different fields are correlated in specific contexts (Dakos and Kéfi 2022, Meyer 2016, Krakovská et al. 2023); for instance, recovery time is related to probability of regime shift when computed across the entire basin of attraction (Meyer 2016). One relevant difference in how resilience is understood across fields is whether it is considered a descriptive system property, as in social-ecological resilience, or a normative and inherently desirable condition, as in disaster and development resilience (Brand and Jax 2007, Baggio et al. 2015, Hahn and Nykvist 2017). For instance, a system locked in a poverty trap that quickly returns to that state after interventions has high resilience under many social-ecological resilience definitions but low under development resilience (Lade et al. 2020).

We adopted social-ecological resilience as our primary focus while incorporating resilience definitions from other research fields to learn how they address diversity and agency. Social-ecological resilience is inherently interdisciplinary (Baggio et al. 2015, Beichler et al. 2014), valuing insights across fields and recognizing that diverse knowledge systems are essential to addressing complex, real-world environmental challenges (Allen et al. 2019, Beichler et al. 2014, Folke 2006). As discussed in the previous paragraph, different fields define resilience in distinct ways, but they also share a basic definition and have similarities. Therefore, resilience can function as a boundary object, sharing foundational concepts while being adapted for specific fields, thus facilitating communication and breaking down silos between research fields (Baggio et al. 2015). Diversity and agency are also recognized as important elements in resilience applications across multiple fields (Bandura 2006, Fabozzi et al. 2002, Pagano et al. 2019). Accordingly, by analyzing which properties of these concepts are integrated in resilience metrics in different contexts and research fields we can gain insights to improve how future social-ecological resilience applications include diversity and agency.

Measuring resilience is important for decision-making

Social-ecological resilience is a multifaceted concept that is notably challenging to measure but central to decision-making. Questions have been raised about the limitations of understanding such a complex and dynamic concept using only quantitative analysis (Brown and Westaway 2011, Quinlan et al. 2016, Leslie and McCabe 2013). We acknowledge the value of qualitative approaches in providing a deeper understanding of resilience. However, when quantitative metrics are grounded in theory, appropriately contextualized and explicit about their assumptions and limitations, they have great potential to guide policies (Carpenter et al. 2001, Quinlan et al. 2016). Quantitative resilience measures are critical in comparative studies, policy evaluations, and decision-making processes (Béné 2013, Grafton et al. 2019). Moreover, although diversity and agency are critical in both qualitative and quantitative resilience analyses, we suspect they are primarily addressed in qualitative approaches and rarely operationalized in quantitative metrics. Interdisciplinary works with social scientists have progressed in including agency in qualitative resilience assessments (Bohle et al. 2009, Haider and Cleaver 2023), but due to its subjectivity it is challenging to capture the role of agency in quantitative resilience measures, which also usually focus on system and not on agent level dynamics (Brown and Westaway 2011, Otto et al. 2020). Similarly, despite its quantitative nature, diversity is more commonly used as part of resilience frameworks. For these reasons, we focus on quantitative resilience metrics in this review.

Diversity and agency are critical for resilience

Diversity has been a central concept in the development of ecological resilience, with evidence suggesting it enhances resilience across various social-ecological systems. In ecology, debates on the relationship between biodiversity, ecosystem function, and stability have a long history (Elmqvist et al. 2003, Holling et al. 1995, McCann 2000). Besides biodiversity, the importance of diversity to resilience extends to other contexts, such as the diversity of land use, livelihoods, cognitive capacity, knowledge systems, and institutions (Biggs et al. 2015, Freeman et al. 2020, Page 2010). Studies that synthesized lists of properties or principles critical to resilience across domains and systems consistently highlight the importance of diversity, with it included in the so-called resilience principles, attributes and others (Biggs et al. 2015, Carpenter et al. 2012, Rockström et al. 2023, Walker et al. 2006). In general, systems composed of many elements are more resilient than more homogeneous systems (Biggs et al. 2015, Walker et al. 2006), though the impact of diversity can vary depending on the context and its distribution across scales (Biggs et al. 2015, Walker et al. 2023). In some cases, too much diversity might undermine resilience (Biggs et al. 2015, Mori et al. 2013).

The concept of response diversity offers a mechanistic explanation for why diversity can enhance resilience (Elmqvist et al. 2003, Walker et al. 2023). Systems with components that respond differently to disruptions are only partially affected by disruptions, allowing other components to continue functioning and maintaining the system’s resilience (Elmqvist et al. 2003, Walker et al. 2023). This effect has been observed in empirical ecological studies and has also been extended to social and social-ecological systems, as well as local and global scales (Leslie and McCabe 2013, Walker et al. 2023). For example, species diversity is important to seed dispersal in tropical forests. As different species responsible for seed dispersal respond differently to disruptions, events like a cyclone or wildfire only affect some species, preserving the seed dispersal function (Elmqvist et al. 2003, Biggs et al. 2015). Similarly, traditional farmers in some tropical regions choose crops with varying tolerances to water scarcity and flooding, using agricultural diversity to increase their resilience to floods or droughts and maintain some level of crop production (Lin 2011).

The central role of agency in contributing to social-ecological resilience is well-documented in the literature. The integration of social-ecological resilience with social science theories lead to the recognition of the importance of agency for resilience (Bohle et al. 2009, Brown 2014, Brown and Westaway 2011, Cote and Nightingale 2012, Coulthard 2012) and recent resilience frameworks have incorporated agency as a key component (Bohle et al. 2009, Hahn and Nykvist 2017, Haider and Cleaver 2023, Zhang et al. 2025). In multiple contexts, adaptive capacity is built locally, and we need to understand agents' motivations, practices, history and worldviews, besides systemic factors affecting their agency like power and governance structures, to understand their decision-making and how they respond to change (Brown and Westaway 2011, Cote and Nightingale 2012, Haider and Cleaver 2023). Actor motivation and agency can undermine resilience, for example, if a few powerful actors promote their self-interest over collective resilience, but when agency is aligned with resilience-building objectives it can be a powerful mechanism to develop resilience (Zhang et al. 2025). For instance, Bohle et al. (2009) use an agency-based resilience framework to argue that informal food vendors, their social capital and personal compromise are crucial for the resilience of Dhaka’s food system to droughts and resilience-building strategies should focus on enabling their capability and well-being.

Agency is particularly important for building transformative capacities in social-ecological systems. Key actors and institutions are essential for guiding and promoting transformations and social-technical innovation (Brown and Westaway 2011, Westley et al. 2011, Westley et al. 2013). These actors provide leadership, identify leverage points, develop shared visions and coordinate actions (Olsson et al. 2014, Westley et al. 2011). Furthermore, different types of agency and capabilities are important at various stages of the transformation process. Knowledge carriers and innovators are critical in the early phases when new ideas are being developed, while reinforcers and organizers become more relevant in later stages when proposals are scaling (Olsson et al. 2014, Westley et al. 2013).

Diversity and agency are also important for resilience in fields beyond social-ecological systems. In economics, portfolio theory highlights the importance of asset diversification for managing risk (Fabozzi et al. 2002). In engineering and critical infrastructure resilience, researchers recognize the importance of having diverse ways of delivering water in water distribution networks (Pagano et al. 2019). Psychological research investigates how both external and internal factors influence individual agency and the ability to cope with disruptions (Bandura 2006). In business, researchers analyze preparedness and the capability of organizations to act and respond to crises (Rose 2007).

Methodological approach

Here we assess with what frequency and how resilience metrics from multiple fields include diversity and agency with the goals of assessing how research on social-ecological resilience currently integrates these concepts into quantitative metrics and identifying future opportunities to enhance their inclusion using insights from other fields. We choose to analyze these specific concepts and not other concepts relevant to resilience for four reasons. First, there is strong evidence and agreement on their importance, as discussed in subsection - Diversity and agency are critical for resilience. Second, researchers have called for more studies to incorporate social science and agency theories into resilience analysis (Cote and Nightingale 2012, Olsson et al. 2015) and highlight the need to better understand and quantify the relationship between diversity and resilience (Biggs et al. 2015, Mori et al. 2013). Third, they bridge debates from the two main communities of social-ecological systems, the ecology, for diversity, and the social sciences, for agency. Finally, we suspect there is a noticeable gap in operationalizing these concepts in quantitative metrics, as these concepts are mainly used in theoretical and qualitative frameworks, as discussed in subsection - Measuring resilience is important for decision-making. As such, by analyzing how these two concepts are included in resilience metrics across different fields, we aim to identify gaps in the literature and suggest pathways to improve social-ecological resilience metrics.

To assess the inclusion of diversity and agency in quantitative resilience metrics, we first conducted a meta-review. By focusing on review articles, we aimed to capture a broader landscape of resilience metrics across multiple fields without needing to analyze individual case studies. This approach allowed us to understand how different research fields operationalize resilience and engage with agency and diversity, which is critical since resilience is an interdisciplinary concept (subsection: Subsection Social-ecological resilience is an interdisciplinarity concept). Additionally, selecting review articles enabled us to capture diverse quantitative resilience applications, including empirical studies, theoretical frameworks, and simulation models. One outcome of the meta-review was the identification of six categories of resilience metrics. The primary goal of this categorization is to provide a structure for analyzing how resilience metrics include agency and diversity as methodological constraints can make it challenging to include these concepts in certain types of metrics.

We further explore the operationalization of agency and diversity in resilience metrics through their use in simulation models. Analyzing case studies allows us to better understand the context of each problem and how each metric is used. We focus on simulation (or process-based) models — i.e., models which explicitly and often quantitatively generate results showing how systems change over time, often based on theoretical assumptions and mathematical descriptions of systems (Cuddington et al. 2013) — because of their key role in theory development, scenario exploration, and policy design (Cuddington et al. 2013). Moreover, despite their importance in both contexts, we hypothesize that the absence of diversity and agency in resilience metrics is more pronounced in simulation models than in statistical or exclusively data-based analyses. For instance, community resilience metrics from small-scale empirical studies often capture the role of agency (Sharifi et al. 2016). However, simulation models typically operate at broader scales, making it difficult to account for agency, a property defined at the individual or agent level (Macy and Willer 2002).

METHODS

We performed two systematic reviews (Fig. 1): (A) A meta-review of resilience metrics to capture domain-specific reviews and understand how different research fields measure resilience and (B) a review of resilience metrics applied in simulation models. We first describe the aspects of our methods common to both reviews, then review-specific methods. Across both reviews, we utilized the Web of Science and Scopus databases, adopting DOIs as unique identifiers to merge the databases and excluding documents without DOIs.

Given the large number of results, 1825 for reviews and 32,300 for modelling, we used topic modelling in each review component to facilitate the screening process and filter selective reading (Blei et al. 2003). This statistical modelling technique identifies clusters (or topics) among a set of documents to group similar documents. Topic modelling assumes that when authoring a text, an author discusses one or more topics. A topic is a vector of probabilities for each word in the corpus. So, a specific topic can be interpreted as a set of words ranked according to their probability of occurrence, while each document is a mixture of one or more topics. Given a corpus, statistical techniques can infer the topics presented in the corpus and the mixture of topics in each document. Specifically, we used Latent Dirichlet Allocation (LDA; Blei et al. 2003) as our statistical technique for topic modelling due to its stability and widespread adoption. We trained the LDA model with titles, abstracts, and author's keywords as inputs. Then, we computed the topic distribution probability for each document and assigned the topic with the highest probability to that document. To select the number of topics we evaluate the topic coherence - a quantitative measure of the interpretability of a topic - and the topic diversity - measured as the percentage of unique words in the top 20 words of all topics - and select the value that maximizes the product of the two indicators, as done by Dieng et al. (2020). Note that the modelling review and meta-review have different topic models because they use different datasets.

The first three authors read the documents included in this study to collect three key data for each metric in both reviews: the metric category, whether the metric incorporates agency and diversity elements and how these two concepts are included. After the reading process, we validated a random sample of the results, addressing disagreements in discussions with a larger group of co-authors. We included in our final sample any resilience metrics the authors of the analyzed documents defined as resilience, to include how multiple research fields define and operationalize resilience. Because not all papers directly mention diversity and agency, we adopted specific criteria to define when a metric includes these concepts (subsections: Agency criteria, and Diversity criteria). More details on the reading process can be found in Supplementary Methods (Appendix 1).

Meta-review

The search terms utilized for the meta-review are provided in Table 1. We adopted a shorter year range for the meta-review to ensure we selected reviews that provide an up-to-date perspective on each research field, a common practice in systematic reviews (Gough et al. 2017). Older reviews would miss newer approaches to measure resilience, and most of their metrics are already captured in more recent reviews. In addition, the literature on resilience is recent, and we would increase the sample size by less than 10% by widening the year range (Appendix 1: Fig. S1).

We ran the LDA algorithm with seven topics, merging two psychological topics due to their similarity. The algorithm yielded the following topics: urban, community and disaster; health and psychology; supply-chain; environmental science and food; energy and power-grid; ecological, soil, and animal. Then, for each topic, we performed two sortings: by number of citations and by year of publication. The first sorting aimed to identify influential and well-known publications on the topic, while the latter aimed to capture newer publications that might present fresh perspectives not found in the most influential works. Finally, we read the ordered list of documents, stopping when we identified three documents that satisfied the inclusion criteria for each combination of topic and sorting. The process resulted in six documents per topic model - three from the most cited list and three from the most recent list - with 36 documents overall. The inclusion criteria were (1) the document presented multiple quantitative resilience metrics, and (2) it provided sufficient information on most of the metrics, at least sufficient for categorizing the metric and determining if it includes agency and diversity. We considered as a quantitative resilience metric any type of indicator or index, whether composite or single, that assigns numerical values to resilience and enables comparison across different systems, scenarios, or points in time. In the meta-review we lack the context that would allow us to assess inclusion of agency and diversity in individual metrics. Therefore, we classified how frequently applications of metrics in each category take diversity and agency into account, based on our understanding of the literature. We provide more details on how this score was computed in the Supplementary Methods (Appendix 1).

Modelling review

The terms used for the modelling component are provided in Table 1. We followed a similar procedure to the meta-review, using topic modelling to group different documents and select documents from each topic. We ran the LDA algorithm with six topics and merged two psychological topics due to their similarity. The algorithm yielded the following topics: urban, community and disaster; health and psychology; environmental science; energy and power-grid; and ecological. Then, we used the same two sorts from the meta-review, by number of citations and by year of publication. To cover a broader range of applications, we chose a larger sample size than in the meta-review in which each document already covered a wider range of metrics from a research field. We read the ordered list of documents, and for each combination of topic and sorting, we stopped when we identified ten documents that satisfied the inclusion criteria or when fifty documents were read. We added a second stopping criterion because some topics showed a low percentage of valid papers (<20%), mainly due to a low use of simulation models. The process resulted in 72 documents. We had twelve documents for the urban, community and disaster topic and zero for the health and psychology topic. For the remaining topics, we had twenty documents, ten from the most cited list and ten from the most recent list. To cross-validate the results using an alternative methodology, in Supplementary Methods (Appendix 1), we categorized documents from selected journals. We selected at least one journal from each topic, except for the health and psychology topic. For more information, consult Appendix 1.

The inclusion criterion was that the document presents a simulation (or process-based) model that computes a numerical value for resilience at some stage. The exclusion criteria were that the document (1) applies statistical models to data or conceptual models or (2) presents a metric but does not apply it in a simulated model. Example criteria for including a paper include the paper’s ability to present and apply an equation defining resilience in a simulation model, graphically communicate resilience, or provide numeric values of resilience measures. Compared with the meta-review, on the simulation review there is more context available on how the metric is used. Thus, while for the meta-review, we provide a classification of inclusion of diversity and agency by category, for the modelling review we provide a case-by-case classification for each metric.

Agency criteria

To assess whether a resilience metric incorporates agency elements, we applied Anthony Giddens’ structuration theory. Giddens sought to bridge the duality between social structure and agency, arguing that individuals both reproduce and change the structure through their actions, while also being constrained by structural rules and resources (Duncan 2019, Giddens 1984). He identifies knowledgeability and capability as the fundamental characteristics of human agency (Duncan 2019, Whaley 2018). Given the challenges of capturing knowledgeability in our sampled metrics, we focused on capability, which is defined as the capacity of agents to “command relevant skills, access to material and non-material resources and engage in particular organising practices” (Whaley 2018). Based on this formal understanding of agency, capability, and agent, we used the following criterion: does the metric measure agents’ capability to act? Examples of measures of agents’ capability include measures of agents’ capital, options, responses or paths available for the agents, and measures that consider how social structure, norms, power, and agents’ knowledge can limit the agents’ actions and capability. More details on how this was applied can be found in Appendix 1.

We make two notes on how we considered other definitions of agency and interpreted this concept. First, we expanded Giddens’ definition of agency to incorporate groups of individuals. Specifically, we considered an agent to be the entity responsible for the decision-making process, encompassing individuals, households, groups of individuals, and formal institutions. We made this adaptation because, in methodological trials, the sampled metrics commonly focused on composite agents. Furthermore, social psychology and sociology studies recognize the importance of collective actors’ agency (Bandura 2006, Morris et al. 2001, Otto et al. 2020). Second, we acknowledge the significance of non-human agency, which includes the agency of non-human living creatures and materials (Contesse et al. 2021, Duncan 2019). However, we limited our focus to human and groups of humans’ agency due to the challenges associated with defining and identifying non-human agency.

Diversity criteria

Diversity is an attribute of any system composed of elements that can be subdivided into categories (Stirling 2007, Page 2010). These categories may encompass a wide range of entities, such as biological species in an ecosystem, assets in a portfolio, or available actions when facing a problem. Here, we interpret diversity in terms of the three properties identified by Stirling (2007): variety, balance, and disparity. Variety refers to the number of distinct types of elements the system has, for example, the number of species in an ecosystem. Balance relates to the distribution of types in the system or how much of each type of element the system has, for example, the evenness in the distribution of species. Disparity refers to how different from each other are the type of elements the system has, for example, the average genetic distance between all species. We employed a binary classification to evaluate the incorporation of diversity in resilience metrics. A metric incorporates diversity if it explicitly quantifies the variety, balance, or disparity of a specific element within the system.

RESULTS

We present the systematic review results based on its two components. The first two subsections (Identification and categorization of resilience metrics, and Inclusion of diversity and agency in resilience metrics), cover the meta-review results on metric categorization and inclusion of diversity and agency, and the last subsection cover the modelling review. More detailed information on the documents and metrics analyzed can be found in Appendix 1: Table S1.

Identification and categorization of resilience metrics

Based on the trial readings of the meta-review corpus and discussions with the co-authors, we developed six categories of resilience measures, each of which describes a distinct process to measure resilience (Table 2). From the 36 included documents, we collected 312 metrics, averaging 8.6 metrics per document, with each metric classified into one category. Note that the complete set of 312 metrics can contain duplicate metrics, as different documents can present the same metric, and we did not merge them when this happened.

Two metric categories have origins in ecology and are commonly used in research in social-ecological resilience. The shape of potential landscape category holds historical significance as one of the earliest approaches to analyzing resilience in ecology (Holling 1973, Walker et al. 2004). Early-warning signals is a recent model-free approach to analyze the system’s stability based on statistical patterns in the time series of the system state (Scheffer et al. 2015). These two categories were mainly limited to the two topics most associated with social-ecological systems, shown on the left of Figure 2, with early-warning signals accounting for 57% of the metrics in the environmental science and food topic. Surprisingly, they were not the most common categories in two topics closely related to social-ecological systems: ecology, soil, and animal and the urban, community and disaster topics. A possible explanation for this result is the predominance of qualitative approaches in social-ecological resilience research (González-Quintero and Avila-Foucat 2019, Polain de Waroux et al. 2024). In the absence of more operational metrics, researchers often rely on performance under disruption metrics, which are easier to apply.

In the engineering domain, it is possible to distinguish between performance and structure-based resilience metrics (Pagano et al. 2019). The first is represented by the performance under disruption category, which encompasses a wide range of metrics that measure resilience by analyzing the performance of a system over time under disruptive events (Cheng et al. 2022, Steinmann et al. 2024). The second is represented by the system structure category, which assess resilience by measuring the relationships between the system’s elements, encompassing primarily network topology measures (Jufri et al. 2019, Liu et al. 2022). The most mentioned metric category in our sample was the performance under disruption category, accounting for 35% of all sampled metrics (Fig. 2). It was most mentioned in topics closer to engineering, such as energy and power-grid (90%), and urban, community and disaster (61%) topics. Systems structure metrics were mentioned most frequently in the supply-chain topic (30%, Fig. 2).

Lastly, we have compound indicators that encompass metrics that aggregate multiple indicators into a single metric. In compound indicators at the system-level category, metrics typically aggregate socioeconomic, demographic, and infrastructure indicators to derive a comprehensive resilience metric (Sharifi 2016, Laurien et al. 2022). In compound indicators at the individual level measures protective factors associated with coping well with disruptive events or evaluate individuals’ previous experiences in stressful situations (Davydov et al. 2010). These two categories are the only ones that focus on adaptation and transformation capacities (Table 2), emphasizing the need to include multiple research fields to capture different aspects relevant to resilience, as the remaining metrics focus primarily on persistence. Compound indicators at the individual level were largely confined to the health and psychology topic, where they appeared most and dominated (96%). Compound indicators at the system levels appeared most on urban, community and disaster topic (48%, Fig. 2).

Inclusion of diversity and agency in resilience metrics

In the ecology, soil and animal and environmental science and food topics, the topics more aligned with social-ecological systems resilience, we observed a predominance of performance under disruption and early-warning signal categories, which have a low and very low inclusion of diversity and agency, respectively (Table 3). Most metrics closer to social-ecological resilience incorporating diversity and agency were found in compound indicators at the system level, especially if we consider part of the documents in the urban, community and disaster topic. A common element included in compound indicators at the system level that included diversity is economic diversity, such as the diversity of employment sectors, income levels, and livelihoods (Laurien et al. 2022, Sharifi 2016). The City Resilience Profiling Tool considers that a diversified employment base increases the capability to deal with disasters as industries are affected differently, an example of response diversity promoting resilience (UN-Habitat 2018, Walker et al. 2023). The Natural Hazard Resilience Screening Index considers that a diverse pool of skills can support recovery and rebuilding efforts, an example of functional diversity promoting resilience (Summers et al. 2018, Walker et al. 2006). Regarding agency, metrics typically included factors such as population preparedness, training, and access to essential physical resources needed to effectively respond to catastrophic events (Laurien et al. 2022, Sharifi 2016). The Australian Disaster Resilience Index considers that the lack of prior planning, limited access to crucial information, physical resources, and agentic constructs at higher scales, such as community engagement, governance and leadership, can alter the community resilience to disasters, as they affect the agency of individual actors (Parsons et al. 2016, 2020, 2021).

We observed a limited inclusion of diversity in our complete sample, with four out of six metric categories showing low or very low inclusion of diversity (Table 3). Shape of potential landscape and early-warning signal showed the lowest inclusion of diversity, with no positive example found (Table 3). On the other hand, the system structure category, which is highly used in supply-chain resilience, featured the highest inclusion of diversity in resilience metrics (Table 3). In this category, metrics related to the number and density of nodes and links in a network measured resilience by considering the diversity of elements within the system. For instance, in analyzing the resilience of public transportation networks to events like failures at specific train stations, metrics like the average shortest path indicate the impact of removing nodes and whether this leads to the segmentation of the network (Berche et al. 2009, Zhou et al. 2019). Generally, networks with a higher number of nodes performing similar functions depend less on individual components and are more tolerant to failures or targeted attacks on specific nodes or links (Han et al. 2020, Zhou et al. 2019).

We also observed a limited inclusion of agency in our complete sample, with four out of six metric categories showing low or very low inclusion of agency in our sample (Table 3). Shape of potential landscape, early-warning signal and system structure showed the lowest inclusion of agency, with no positive example found (Table 3). On the contrary, agency is most frequently included within the compound indicator at the individual level category, particularly within psychological resilience (Table 3). By placing the individual at the center of the process, psychological resilience typically measures individual resilience by quantifying the presence of protective factors or personality traits associated with a better capability to overcome stressful events (Davydov et al. 2010, Friborg et al. 2003). Measurements encompass aspects ranging from internal factors like emotions and self-esteem to external elements such as relationships within the community and family cohesion, all of which influence an individual’s adeptness at navigating stressful events (Davydov et al. 2010, Pangallo et al. 2015). In these measures, the individual is conceptualized as constantly acting and using available resources to adapt to disruptive or stressful situations, with their agency serving as a core component of the resilience measure.

Resilience metrics in simulation models

In the modelling review, the predominance of the performance under disruption category is even more substantial, with 65% of all sampled metrics categorized as such (Fig. 3). In the ecological and environmental science topics, which are more aligned with social-ecological systems, we observe more traditional ecological and social-ecological resilience metrics such as metrics based on shape of potential landscape (25% of the ecological topic) and early-warning signal (15% of the environmental science topic), though there is a high use of performance under disruption metrics. Performance under disruption metrics were the most used across all topics, likely due to the ease of operationalizing metrics of this category within simulation models. This metric category is especially predominant in the urban, disaster, and community (83%) and energy and power-grid (80%) topics. Compound metrics at the system and individual levels showed the highest overall percentual reduction when compared with the meta-review, 5% and 25% respectively, with the latter being completely absent from this corpus due to the low use of simulation models in psychology. When classifying resilience metrics in five selected journals from different topics, we obtained a similar result, validating our document selection methodology. Performance under disruption is used on 70% of metrics and is predominant in all journals except for Ecology and Society, where the shape of potential landscape category is predominant (54%, Appendix 1: Fig. S5).

In the ecological and environmental science topics, the topics more aligned with social-ecological resilience, we observe a limited inclusion of diversity and agency (15% for diversity and 5% for agency, Fig. 4). Here are a few notable examples of the inclusion of these concepts in social-ecological articles. Bitterman and Bennett (2016) measure the resilience of agricultural systems using a bi-dimensional potential landscape composed of the landscape diversity, which is considered to provide greater functional and response diversity, and the mean net profit, which reflects farmers’ capability to overcome disruptions. Estrada and Bodin (2008) measure the resilience of a landscape in Madagascar to patch loss using the size of the largest network component and clustering coefficient, which capture the diversity and density of links between habitat patches. Lankford et al. (2023) measure the resilience of irrigated agriculture in South Africa to droughts using a performance indicator that captures farmers’ capabilities to respond to droughts.

Few metrics in the complete simulation model sample include diversity measures (14% of all metrics, Fig. 4a). Similar to the meta-review, metrics that include diversity reside predominantly in the system structure category. Metrics in this category predominantly analyze critical infrastructure networks such as power-grid and water distribution networks. For example, systems with energetic redundancy, where water flows can reach nodes from multiple alternative routes, will be more resilient to pipe failure (Pagano et al. 2019). We found a smaller number of metrics including diversity in the compound indicator category — where a metric includes transportation routes diversity during disasters (Hosseini and Barker 2016) as a way to provide response diversity to the analyzed systems — and the performance under disruption category — where a metric includes vegetation diversity as a performance indicator (Gustafson et al. 2023).

Compared to diversity, we observe even fewer metrics including agency in the complete simulation model sample (5% of all sampled metrics, Fig. 4b). Performance under disruption and compound indicators at the system level are the primary source of metrics including agency in the modelling review. Availability of resources during disruptions emerges as a recurrent theme in these metrics, encompassing aspects like the availability of human resources, equipment, and knowledge during disasters (Hosseini and Barker 2016, Pagano et al. 2017). Covering more subjective elements, one metric considered both “real” and “perceived” resilience, with the latter guiding decision-makers’ actions to build disaster recovery plans (Zobel 2011). Multiple articles in the ecological and environmental science topics focused solely on non-human ecological elements, recognizing the human impact but not modelling it. Most articles from other topics had models with agents making decisions. However, they do not consider the agents’ capability to act when measuring resilience.

DISCUSSION

Prominent works have identified the importance of diversity and agency in social-ecological resilience (Biggs et al. 2015, Brown and Westaway 2011). Significant progress has been made in integrating these concepts within qualitative and theoretical contexts (González-Quintero and Avila-Foucat 2019, Hahn and Nykvist 2017). However, studies suggest that there are challenges to operationalizing these concepts in resilience quantitative metrics (Cote and Nightingale 2012, Leslie and McCabe 2013). Our findings confirm the limited use of these concepts: for the sampled documents, we observe that quantitative resilience metrics rarely incorporate diversity and agency (Fig. 5). For instance, in the meta-review, over 60% of metrics fall into categories with low or very low inclusion of either diversity or agency (Fig. 2 and Table 3). Metrics incorporating diversity were predominantly found in the system structure and compound indicators at the system level categories, while agency was primarily included in compound indicators at both the individual and system levels. Simulation models exhibited an even greater disparity, with more than 80% of metrics falling into categories with less than 20% inclusion of either diversity or agency (Fig. 4). The lower inclusion of these concepts in simulation models is largely due to the reduced presence of compound indicators. An interesting next step would be to assess whether the observed lack of diversity and agency in simulation models also applies to empirical application of resilience metrics.

Differences in how research fields conceptualize resilience help explain the inclusion of diversity and agency in Figure 5. Metrics closer to ecological and social-ecological resilience (Fig. 5, left) tend to focus on ecosystem processes and often overlook individual dimensions such as agency. While ecological and social-ecological resilience frequently recognizes the importance of diversity as a property that confers resilience, it typically does not include it explicitly in metrics. Metrics closer to engineering fields (Fig. 5, middle) are highly varied but tend to prioritize recovery of performance indicators and often neglect human dimensions, resulting in a low inclusion of agency. However, surprisingly, system structure metrics, which commonly measure infrastructure and supply-chain network indicators, show the highest inclusion of diversity between all categories (Fig. 5). Compound indicators at the individual level are primarily used in psychology resilience, where humans are central, leading to a high inclusion of agency (Fig. 5). Compound indicators at the system level are predominantly used in community, urban and disaster resilience, where metrics tend to incorporate multiple dimensions such as social, organizational, infrastructure, ecological, and others. The use of multiple indicators and adoption of a multifactorial perspective result in a moderate inclusion of both diversity and agency (Fig. 5), showing a possible pathway to increase the inclusion of these concepts in social-ecological resilience metrics.

Why should social-ecological resilience metrics include diversity and agency?

The failure of many social-ecological resilience metrics to include diversity overlooks a key determinant of resilience and risks poorly designed management. In the subsection: Diversity and agency are critical for resilience, we summarized literature showing the importance of diversity for social-ecological systems. For example, social-ecological systems with greater diversity are often more resilient than homogeneous ones, though not always (Biggs et al. 2015). Our literature reviews subsequently showed that metrics often neglect this important feature of social-ecological resilience. Unintended consequences of omitting diversity include: (1) Metrics focused on detecting proximity or risk of critical threshold can lead to management strategies focused on increasing stability and reducing disturbances, which can reduce variability and stop the system from building response diversity (Carpenter et al. 2015). (2) Management based on these metrics can improve resilience against specific disruptions but can undermine the system’s capacity to handle unknown events and deep uncertainty, while diversity is an emergent property associated with general resilience (Carpenter et al. 2012, Walker et al. 2023). The importance of diversity to resilience depends on the resilience questions: resilience of what and to what (Carpenter et al. 2001). Highly specialized and more homogenous systems might be necessary to handle resilience to specific disruptions, as diversity can increase system complexity and reduce efficiency (Biggs et al. 2015, Carpenter et al. 2012). Generally, including diversity in resilience metrics is especially important when analyzing systems composed of multiple actors or components, as they will likely act differently and respond differently to disturbances. For example, perceptions of reality and cognitive diversity, together with the capability of agents to generate shared goals and diffuse conflict, are a mechanism to solve common-pool problems, promote cooperation and resilience (Freeman et al. 2020, Cronin and Weingart 2019). Overall, diversity is related to resilience in most social-ecological systems at appropriate temporal and spatial scales (Biggs et al. 2015, Walker et al. 2023).

Likewise, the failure of many social-ecological resilience metrics to include agency risks poorly designed management actions. The literature highlights the importance of understanding agents’ motivations for building resilience at local scales (subsection: Diversity and agency are critical for resilience). Neglecting agency in resilience metrics can underestimate actors’ roles in producing change and the factors influencing this capability. For example, (1) Simpler metrics such as performance loss and recovery time often ignore agency and taking management decisions based on them can lead to solutions that ignore how conflicts and power dynamics constraining agency affect the system, favoring the status-quo, a depoliticized description of resilience and missing who benefits from enhancing resilience (Brown 2014, Cote and Nightingale 2012, Semplici et al. 2024, Parsons 2024). Without a deeper understanding of the local level, improving the resilience of fishing communities can promote solutions such as exiting fishing that might reduce performance loss but does not match the community values and undermines their long-term well-being (Coulthard 2012). (2) Beyond these practical implications, neglecting agency can increase the already noticeable gap between practical application and theoretical works (Quinlan et al. 2016, Reyers et al. 2022), as well as making resilience analyses less relevant to social scientists, causing communication problems and making knowledge integration harder (Brown and Westaway 2011, Olsson et al. 2015, Parsons 2024). Analyzing agency is especially relevant when the gap between the agent and system level is small and when differentials of power and influence between agents are low. For example, agency is crucial to resilience in small-scale communities where few agents can each have a considerable influence on their shared natural resources. The impact of each agent in the system is less evident in large-scale and decentralized systems, but even in large-scale systems decision-making and power imbalances make agency relevant (Cote and Nightingale 2012).

Our review shows diversity and agency are underrepresented in quantitative resilience metrics, leading practitioners and researchers to overlook critical aspects of social-ecological resilience. We recognize that other concepts not assessed here, such as feedbacks, governance, and slow and fast variables, may also be important for resilience. However, our findings demonstrate that diversity and agency are important for resilience metrics in multiple social-ecological contexts and are frequently missing from metrics. We, therefore, recommend improvement on the inclusion of diversity and agency into quantitative resilience metrics. Next, we discuss the challenges involved and potential pathways for incorporating diversity and agency into resilience metrics.

Challenges to include diversity and agency in social-ecological resilience metrics

We identify three major challenges to incorporating more diversity and agency in social-ecological resilience metrics. First, metrics need to better represent individual elements. To include diversity or agency, resilience measures must capture individual elements, be it systems components, for diversity, or agent characteristics, for agency. However, metric categories such as shape of potential landscape, early-warning signal, and performance under disruption usually focus on analyzing system-scale dynamics and neglect lower-scales. Indeed, our results show these categories rarely include diversity and agency. Metrics in these categories have valid use cases, but researchers should be aware that choosing a metric can limit what elements it can capture. This challenge is stronger for simulation models, where the choice of models further limits the capacity to include individual elements.

Second, metrics need to account for the complexity of resilience, diversity and agency. Compared to ecological resilience, social-ecological resilience tends to focus on qualitative assessments (González-Quintero and Avila-Foucat 2019, Polain de Waroux et al. 2024). The limited attention to developing better quantitative social-ecological resilience metrics means that simpler and easier-to-apply metrics tend to be used, such as the predominant performance under disruptions metrics observed in this review. However, these metrics rarely include diversity and agency. There are also challenges to better capture the complexity of diversity and agency in metrics that already include them. Our results showed that agency is mainly captured by assessing individual capitals. However, research argues agency analyses should go beyond and include structural elements such as power and relational and dynamic capabilities (Cote and Nightingale 2012, Reyers et al. 2022). These elements are harder to quantify, and there is limited data available, especially at higher scales, which explains why they are rarely included in resilience metrics. Similarly, many studies analyze the relationship between biodiversity and resilience (Biggs et al. 2015, McCann 2000). However, we know less about how this translates to social systems and the diversity of other elements that might be more relevant to promote resilience (Leslie and McCabe 2013, Mori et al. 2013). Improvements are also necessary to use diversity-based analysis to provide concrete management and policy recommendations (Mori et al. 2013).

Third, to include diversity and agency requires integrating knowledge from diverse research fields. Many studies argue for better integrating social theories into social-ecological resilience (Brown 2014, Cote and Nightingale 2012). Resilience can work with different theories of agency, including more established or collective and non-human agency (Contesse et al. 2021, Duncan 2019). More integration between social science knowledge and resilience is necessary to choose relevant agency theories aligned with the purpose of the research (Cote and Nightingale 2012, Semplici et al. 2024). Similarly, to better include diversity in resilience metrics, we need to adapt ecology results to social and social-ecological systems (Leslie and McCabe 2013, Mori et al. 2013).

Pathways to improve inclusion of diversity and agency in social-ecological resilience metrics

We suggest five promising pathways, informed by this article’s results and insights from multiple research fields, for improving the inclusion of diversity and agency in social-ecological resilience metrics and addressing the three previous challenges.

• Using network-based metrics: Networks are a methodology that can capture both local and system-level elements. They can also be a common language to bridge research fields, as they are used in social science, ecology and engineering, and include agency and diversity in resilience metrics (Janssen et al. 2006). The system structure category was the main source of metrics including diversity in this review (Fig. 5). Recent works used networks to estimate the likelihood of reaching critical thresholds (Domínguez-García et al. 2019, Liu et al. 2022, Rocha et al. 2018) and the impact on system performance time series (Meng et al. 2018, Zhou et al. 2019), showing a pathway to expand network-based metrics methodology to include diversity to other categories. Network methodologies have also been used to favor a more relational, less individual-focused understanding of agency (Semplici et al. 2024, West et al. 2024). As such, allowing new insights into how structural properties, such as power and inequalities, influence agency and adaptive and transformative capacities (Barnes et al. 2025, Semplici et al. 2024, West et al. 2024).
• Using response and pathway diversity: These metrics are particularly effective in incorporating both diversity and agency but were rarely found in our review. Response diversity directly incorporates diversity and, in social-ecological systems, frequently involves agent response diversity, thus also including agency. The response diversity concept is originally from ecology but has been extended to include social and economic systems (Leslie and McCabe 2013, Walker et al. 2023). It is not commonly recognized and applied as a metric, though promising efforts are being made to operationalize it (Ross et al. 2023). Pathway diversity builds upon response diversity and previous resilience theories to capture both system and agent components of resilience. The metric defines resilience as higher if more agents’ actions are currently available and can be maintained and enhanced into the future (Lade et al. 2020, Sellberg et al. 2024). It was designed as an operational metric that can work in different social-ecological contexts, and it incorporates diversity and agency. Because it is a new approach, new studies still need to refine it and implement it in different empirical case studies.
• Including diversity and agency in compound indicators: It is relatively easy to include more diversity and agency in compound indicators, and we have already observed a moderate inclusion of these elements in this review (Fig. 5). However, more metrics could include these concepts and their whole complexity. For agency, most approaches include individual capital assessment but neglect structural and subjective elements such as power and dynamic capacities (Cote and Nightingale 2012, Reyers et al. 2022). Multiple psychology resilience metrics capture how external relationships affect individual capability to act and other subjective individual factors. These metrics, or other psychological and sociological assessments, can be adapted for social-ecological problems to guarantee more effective incorporation of agency in social-ecological resilience metrics (Reyers et al. 2022). For diversity, most approaches include biodiversity or economic diversity. However, they are not necessarily the main promoters of response and functional diversity in different systems. More empirical and contextual studies are necessary to identify the key diversity elements causing a system to be resilient before including them on metrics.
• Integrating quantitative methodologies outside resilience theory: Different theories are used to analyze social-ecological problems, and resilience theory can benefit from them. For example, the field of Decision-Making under Deep Uncertainty (DMDU) analyzes how systems can cope with uncertainties in environmental, political, and social variables, dealing with similar problems as resilience (Marchau et al. 2019). In the DMDU literature, methods such as portfolio theory have been used to promote a diversity of strategies and resources (Crowe and Parker 2008), and exploratory modelling has been used to analyze cooperation in regional water management and how power and agency might affect it (Gold et al. 2022).
• Improving the application of resilience in simulation models: Dynamical systems models are heavily used in ecology but usually focus on the system level and consider simplified aggregated agents (Macy and Willer 2002). Agent-based, integrated assessment and network models are more flexible and allow simulation models to focus on the agent perspective (Bianchi and Squazzoni 2015, Liu et al. 2022, Otto et al. 2020). They show a pathway to use behavioral theories and analyze how structural elements influence agent capability to take actions as well as capture agent heterogeneity.

Conclusion

Through this systematic review, we contribute to the debate by showing diversity and agency are central concepts for understanding social-ecological resilience and highlight a critical gap in quantitative resilience metrics: few quantitative resilience metrics sampled include agency and diversity. While resilience definitions and qualitative frameworks incorporate agency and diversity, their translation into operationalized metrics remains rare. This review underscores the need for novel approaches and metrics that explicitly embrace agency and diversity as integral components of resilience measures. The inclusion of these concepts in quantitative methodologies has the potential to promote a more coherent understanding of social-ecological resilience.

LITERATURE CITED

Adger, W. N. 2000. Social and ecological resilience: are they related? Progress in Human Geography 24:347–364.https://doi.org/10.1191/030913200701540465

Allen, C. R., D. G. Angeler, B. C. Chaffin, D. Twidwell, and A. Garmestani. 2019. Resilience reconciled. Nature Sustainability 2(10):898-900. https://doi.org/10.1038/s41893-019-0401-4

Anderies, J. M., M. A. Janssen, and B. H. Walker. 2002. Grazing management, resilience, and the dynamics of a fire-driven rangeland System. Ecosystems 5(1):23-44. https://doi.org/10.1007/s10021-001-0053-9

Baggio, J., K. Brown, and D. Hellebrandt. 2015. Boundary object or bridging concept? A citation network analysis of resilience. Ecology and Society 20(2):2. https://doi.org/10.5751/ES-07484-200202

Bandura, A. 2006. Toward a psychology of human agency. Perspectives on Psychological Science 1(2):164-180 https://doi.org/10.1111/j.1745-6916.2006.00011.x

Barnes, M. L., S. Sutcliffe, I. Muly, N. Muthiga, S. Wanyonyi, P. Matous, M. Murunga. 2025. Agency, social networks, and adaptation to environmental change. Global Environmental Change 92:102983. https://doi.org/10.1016/j.gloenvcha.2025.102983

Beichler, S. A., S. Hasibovic, B. J. Davidse, and S. Deppisch. 2014. The role played by social-ecological resilience as a method of integration in interdisciplinary research. Ecology and Society 19(3):4. https://doi.org/10.5751/ES-06583-190304

Béné, C. 2013. Towards a quantifiable measure of resilience. IDS Working Papers 2013(434):1-27. https://doi.org/10.1111/j.2040-0209.2013.00434.x

Berche, B., C. von Ferber, T. Holovatch, and Yu. Holovatch. 2009. Resilience of public transport networks against attacks. The European Physical Journal B 71(1):125-137. https://doi.org/10.1140/epjb/e2009-00291-3

Berkes, F., J. Colding, and C. Folke. 2002. Navigating Social-Ecological Systems: Building Resilience for Complexity and Change. Cambridge University Press, Cambridge, UK.

Bianchi, F., and F. Squazzoni. 2015. Agent-based models in sociology. WIREs Computational Statistics 7(4):284-306. https://doi.org/10.1002/wics.1356

Biggs, R., M. Schlüter, and M. L. Schoon. 2015. Principles for Building Resilience: Sustaining Ecosystem Services in Social-ecological Systems. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781316014240

Bitterman, P., and D. A. Bennett. 2016. Constructing stability landscapes to identify alternative states in coupled social-ecological agent-based models. Ecology and Society 21(3):21. https://doi.org/10.5751/ES-08677-210321

Blei, D. M., A. Y. Ng, and M. I. Jordan. 2003. Latent Dirichlet Allocation. Journal of Machine Learning Research 3:993-1022.

Bohle, H.-G., B. Etzold, and M. Keck. 2009. Resilience as agency. International Human Dimension Programme Update 2:8-13.

Brand, F., and K. Jax. 2007. Focusing the meaning(s) of resilience: resilience as a descriptive concept and a boundary object. Ecology and Society 12(1):23. https://doi.org/10.5751/ES-02029-120123

Brown, K. 2014. Global environmental change I: A social turn for resilience? Progress in Human Geography 38(1):107-117. https://doi.org/10.1177/0309132513498837

Brown, K., and E. Westaway. 2011. Agency, capacity, and resilience to environmental change: lessons from human development, well-being, and disasters. Annual Review of Environment and Resources 36(1):321-342. https://doi.org/10.1146/annurev-environ-052610-092905

Carpenter, S. R., K. J. Arrow, S. Barrett, R. Biggs, W. A. Brock, A.-S. Crépin, G. Engström, C. Folke, T. P. Hughes, N. Kautsky, C.-Z. Li, G. McCarney, K. Meng, K.-G. Mäler, S. Polasky, M. Scheffer, J. Shogren, T. Sterner, J. R. Vincent, B. Walker, A. Xepapadeas, and A. D. Zeeuw. 2012. General resilience to cope with extreme events. Sustainability 4(12):3248-3259. https://doi.org/10.3390/su4123248

Carpenter, S. R., W. A. Brock, C. Folke, E. H. van Nes, and M. Scheffer. 2015. Allowing variance may enlarge the safe operating space for exploited ecosystems. Proceedings of the National Academy of Sciences 112(46):14384-14389. https://doi.org/10.1073/pnas.1511804112

Carpenter, S., B. Walker, J. M. Anderies, and N. Abel. 2001. From metaphor to measurement: resilience of what to what? Ecosystems 4(8):765-781. https://doi.org/10.1007/s10021-001-0045-9

Cheng, Y., E. A. Elsayed, and Z. Huang. 2022. Systems resilience assessments: a review, framework and metrics. International Journal of Production Research 60(2):595-622. https://doi.org/10.1080/00207543.2021.1971789

Contesse, M., J. Duncan, K. Legun, and L. Klerkx. 2021. Unravelling non-human agency in sustainability transitions. Technological Forecasting and Social Change 166:120634. https://doi.org/10.1016/j.techfore.2021.120634

Cote, M., and A. J. Nightingale. 2012. Resilience thinking meets social theory: Situating social change in socio-ecological systems (SES) research. Progress in Human Geography 36(4):475-489. https://doi.org/10.1177/0309132511425708

Coulthard, S. 2012. Can we be both resilient and well, and what choices do people have? Incorporating agency into the resilience debate from a fisheries perspective. Ecology and Society 17(1):4. https://doi.org/10.5751/ES-04483-170104

Cronin, M. A., and L. R. Weingart. 2019. Conflict across representational gaps: Threats to and opportunities for improved communication. Proceedings of the National Academy of Sciences 116(16):7642-7649. https://doi.org/10.1073/pnas.1805866116

Crowe, K. A., and W. H. Parker. 2008. Using portfolio theory to guide reforestation and restoration under climate change scenarios. Climatic Change 89(3):355-370. https://doi.org/10.1007/s10584-007-9373-x

Cuddington, K., M.-J. Fortin, L. R. Gerber, A. Hastings, A. Liebhold, M. O’Connor, and C. Ray. 2013. Process-based models are required to manage ecological systems in a changing world. Ecosphere 4(2):art20. https://doi.org/10.1890/ES12-00178.1

Dakos, V., and S. Kéfi. 2022. Ecological resilience: what to measure and how. Environmental Research Letters 17(4):043003. https://doi.org/10.1088/1748-9326/ac5767

Davydov, D. M., R. Stewart, K. Ritchie, and I. Chaudieu. 2010. Resilience and mental health. Clinical Psychology Review 30(5):479-495. https://doi.org/10.1016/j.cpr.2010.03.003

Dieng, A. B., F. J. R. Ruiz, and D. M. Blei. 2020. Topic modeling in embedding spaces. Transactions of the Association for Computational Linguistics 8:439-453. https://doi.org/10.1162/tacl_a_00325

Domínguez-García, V., V. Dakos, and S. Kéfi. 2019. Unveiling dimensions of stability in complex ecological networks. Proceedings of the National Academy of Sciences 116(51):25714-25720. https://doi.org/10.1073/pnas.1904470116

Duncan, S. 2019. Agency. Page in P. Atkinson, S. Delamont, A. Cernat, J. W. Sakshaug, and R. A. Williams, editors. SAGE Research Methods Foundations. Sage Publications Ltd.

Elmqvist, T., C. Folke, M. Nyström, G. Peterson, J. Bengtsson, B. Walker, and J. Norberg. 2003. Response diversity, ecosystem change, and resilience. Frontiers in Ecology and the Environment 1(9):488-494. https://doi.org/10.1890/1540-9295(2003)001[0488:RDECAR]2.0.CO;2

Estrada, E., and Ö. Bodin. 2008. Using network centrality measures to manage landscape connectivity. Ecological Applications 18(7):1810-1825. https://doi.org/10.1890/07-1419.1

Fabozzi, F. J., F. Gupta, and H. M. Markowitz. 2002. The Legacy of Modern Portfolio Theory. The Journal of Investing 11(3):7-22. https://doi.org/10.3905/joi.2002.319510

Folke, C. 2006. Resilience: The emergence of a perspective for social-ecological systems analyses. Global Environmental Change 16(3):253-267. https://doi.org/10.1016/j.gloenvcha.2006.04.002

FAO (Food and Agriculture Organization). 2016. Rima-II: resilience index measurement and analysis II. Food and Agriculture Organization, Rome.

Freeman, J., J. A. Baggio, and T. R. Coyle. 2020. Social and general intelligence improves collective action in a common pool resource system. Proceedings of the National Academy of Sciences 117(14):7712-7718. https://doi.org/10.1073/pnas.1915824117

Friborg, O., O. Hjemdal, J. H. Rosenvinge, and M. Martinussen. 2003. A new rating scale for adult resilience: what are the central protective resources behind healthy adjustment? International Journal of Methods in Psychiatric Research 12(2):65-76. https://doi.org/10.1002/mpr.143

Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. University of California Press, USA.

Gold, D. F., P. M. Reed, D. E. Gorelick, and G. W. Characklis. 2022. Power and Pathways: Exploring robustness, cooperative stability, and power relationships in regional infrastructure investment and water supply management portfolio pathways. Earth’s Future 10(2):e2021EF002472. https://doi.org/10.1029/2021EF002472

González-Quintero, C., and V. S. Avila-Foucat. 2019. Operationalization and measurement of social-ecological resilience: a systematic review. Sustainability 11(21):6073. https://doi.org/10.3390/su11216073

Gough, D., S. Oliver, and J. Thomas. 2017. An Introduction to Systematic Reviews. Sage Publications, Inc.

Grafton, R. Q., L. Doyen, C. Béné, E. Borgomeo, K. Brooks, L. Chu, G. S. Cumming, J. Dixon, S. Dovers, D. Garrick, A. Helfgott, Q. Jiang, P. Katic, T. Kompas, L. R. Little, N. Matthews, C. Ringler, D. Squires, S. I. Steinshamn, S. Villasante, S. Wheeler, J. Williams, and P. R. Wyrwoll. 2019. Realizing resilience for decision-making. Nature Sustainability 2(10):907-913. https://doi.org/10.1038/s41893-019-0376-1

Gustafson, E. J., C. C. Kern, and J. M. Kabrick. 2023. Can assisted tree migration today sustain forest ecosystem goods and services for the future? Forest Ecology and Management 529:120723. https://doi.org/10.1016/j.foreco.2022.120723

Hahn, T., and B. Nykvist. 2017. Are adaptations self-organized, autonomous, and harmonious? Assessing the social-ecological resilience literature. Ecology and Society 22(1):12. https://doi.org/10.5751/ES-09026-220112

Haider, L. J., and F. Cleaver. 2023. Capacities for resilience: persisting, adapting and transforming through bricolage. Ecosystems and People 19(1):2240434. https://doi.org/10.1080/26395916.2023.2240434

Holling, C. S., D. W. Schindler, B. W. Walker, and J. Roughgarden. 1995. Biodiversity in the functioning of ecosystems: an ecological synthesis. Pages 44–83 in Biodiversity Loss: Economic and Ecological Issues. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781139174329.005

Han, Y., W. K. Chong, and D. Li. 2020. A systematic literature review of the capabilities and performance metrics of supply chain resilience. International Journal of Production Research 58(15):4541-4566. https://doi.org/10.1080/00207543.2020.1785034

Holling, C. S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systematics 4(1):1-23. https://doi.org/10.1146/annurev.es.04.110173.000245

Hosseini, S., and K. Barker. 2016. A Bayesian network model for resilience-based supplier selection. International Journal of Production Economics 180:68–87. https://doi.org/10.1016/j.ijpe.2016.07.007

Hosseini, S., K. Barker, and J. E. Ramirez-Marquez. 2016. A review of definitions and measures of system resilience. Reliability Engineering & System Safety 145:47-61. https://doi.org/10.1016/j.ress.2015.08.006

Janssen, M. A., Ö. Bodin, J. M. Anderies, T. Elmqvist, H. Ernstson, R. R. J. McAllister, P. Olsson, and P. Ryan. 2006. Toward a network perspective of the study of resilience in social-ecological systems. Ecology and Society 11(1):15. https://doi.org/10.5751/ES-01462-110115

Jufri, F. H., V. Widiputra, and J. Jung. 2019. State-of-the-art review on power grid resilience to extreme weather events: Definitions, frameworks, quantitative assessment methodologies, and enhancement strategies. Applied Energy 239:1049-1065. https://doi.org/10.1016/j.apenergy.2019.02.017

Krakovská, H., C. Kuehn, and I. P. Longo. 2023. Resilience of dynamical systems. European Journal of Applied Mathematics 1-46. https://doi.org/10.1017/S0956792523000141

Lade, S. J., B. H. Walker, and L. J. Haider. 2020. Resilience as pathway diversity: linking systems, individual, and temporal perspectives on resilience. Ecology and Society 25(3):19. https://doi.org/10.5751/ES-11760-250319

Lankford, B., C. Pringle, J. McCosh, M. Shabalala, T. Hess, and J. W. Knox. 2023. Irrigation area, efficiency and water storage mediate the drought resilience of irrigated agriculture in a semi-arid catchment. Science of The Total Environment 859:160263. https://doi.org/10.1016/j.scitotenv.2022.160263

Laurien, F., J. G. C. Martin, and S. Mehryar. 2022. Climate and disaster resilience measurement: Persistent gaps in multiple hazards, methods, and practicability. Climate Risk Management 37:100443. https://doi.org/10.1016/j.crm.2022.100443

Leslie, P., and J. T. McCabe. 2013. Response diversity and resilience in social-ecological systems. Current Anthropology 54(2):114-143. https://doi.org/10.1086/669563

Lin, B. B. 2011. Resilience in agriculture through crop diversification: adaptive management for environmental change. BioScience 61(3):183-193. https://doi.org/10.1525/bio.2011.61.3.4

Liu, X., D. Li, M. Ma, B. K. Szymanski, H. E. Stanley, and J. Gao. 2022. Network resilience. Physics Reports 971:1-108. https://doi.org/10.1016/j.physrep.2022.04.002

Macy, M. W., and R. Willer. 2002. From factors to actors: computational sociology and agent-based modeling. Annual Review of Sociology 28(1):143-166. https://doi.org/10.1146/annurev.soc.28.110601.141117

Marchau, V. A. W. J., W. E. Walker, P. J. T. M. Bloemen, and S. W. Popper, editors. 2019. Decision Making under Deep Uncertainty: From Theory to Practice. Springer Cham. https://doi.org/10.1007/978-3-030-05252-2

McCann, K. S. 2000. The diversity-stability debate. Nature 405(6783):228-233. https://doi.org/10.1038/35012234

Meng, F., G. Fu, R. Farmani, C. Sweetapple, and D. Butler. 2018. Topological attributes of network resilience: A study in water distribution systems. Water Research 143:376-386. https://doi.org/10.1016/j.watres.2018.06.048

Meyer, K. 2016. A mathematical review of resilience in ecology. Natural Resource Modeling 29(3):339-352. https://doi.org/10.1111/nrm.12097

Mori, A. S., T. Furukawa, and T. Sasaki. 2013. Response diversity determines the resilience of ecosystems to environmental change. Biological Reviews 88(2):349-364. https://doi.org/10.1111/brv.12004

Morris, M. W., T. Menon, and D. R. Ames. 2001. Culturally conferred conceptions of agency: a key to social perception of persons, groups, and other actors. Personality and Social Psychology Review. https://doi.org/10.1207/S15327957PSPR0502_7

Olsson, L., A. Jerneck, H. Thoren, J. Persson, and D. O’Byrne. 2015. Why resilience is unappealing to social science: Theoretical and empirical investigations of the scientific use of resilience. Science Advances 1(4):e1400217. https://doi.org/10.1126/sciadv.1400217

Olsson, P., V. Galaz, and W. J. Boonstra. 2014. Sustainability transformations: a resilience perspective. Ecology and Society 19(4):1. https://doi.org/10.5751/ES-06799-190401

Otto, I. M., M. Wiedermann, R. Cremades, J. F. Donges, C. Auer, and W. Lucht. 2020. Human agency in the Anthropocene. Ecological Economics 167:106463. https://doi.org/10.1016/j.ecolecon.2019.106463

Pagano, A., I. Pluchinotta, R. Giordano, and M. Vurro. 2017. Drinking water supply in resilient cities: Notes from L’Aquila earthquake case study. Sustainable cities and society 28:435-449. https://doi.org/10.1016/j.scs.2016.09.005

Pagano, A., C. Sweetapple, R. Farmani, R. Giordano, and D. Butler. 2019. Water distribution networks resilience analysis: a comparison between graph theory-based approaches and global resilience analysis. Water Resources Management 33(8):2925-2940. https://doi.org/10.1007/s11269-019-02276-x

Page, M. J., J. E. McKenzie, P. M. Bossuyt, I. Boutron, T. C. Hoffmann, C. D. Mulrow, L. Shamseer, J. M. Tetzlaff, E. A. Akl, S. E. Brennan, R. Chou, J. Glanville, J. M. Grimshaw, A. Hróbjartsson, M. M. Lalu, T. Li, E. W. Loder, E. Mayo-Wilson, S. McDonald, L. A. McGuinness, L. A. Stewart, J. Thomas, A. C. Tricco, V. A. Welch, P. Whiting, and D. Moher. 2021. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. International Journal of Surgery 88:105906. https://doi.org/10.1016/j.ijsu.2021.105906

Page, S. E. 2010. Diversity and Complexity. Princeton University Press, Princeton, NJ, USA.

Pangallo, A., L. Zibarras, R. Lewis, and P. Flaxman. 2015. Resilience through the lens of interactionism: A systematic review. Psychological Assessment 27:1-20. https://doi.org/10.1037/pas0000024

Parsons, M. 2024. Neoliberalism, normativity and agency: constructive tensions in the application of resilience. Pages 305-318 in Resilience and Riverine Landscapes. Elsevier. https://doi.org/10.1016/B978-0-323-91716-2.00003-0

Parsons, M., S. Glavac, P. Hastings, G. Marshall, J. McGregor, J. McNeill, P. Morley, I. Reeve, and R. Stayner. 2016. Top-down assessment of disaster resilience: A conceptual framework using coping and adaptive capacities. International Journal of Disaster Risk Reduction 19:1-11. https://doi.org/10.1016/j.ijdrr.2016.07.005

Parsons, M., I. Reeve, J. McGregor, P. Hastings, G. R. Marshall, J. McNeill, R. Stayner, and S. Glavac. 2021. Disaster resilience in Australia: A geographic assessment using an index of coping and adaptive capacity. International Journal of Disaster Risk Reduction 62:102422. https://doi.org/10.1016/j.ijdrr.2021.102422

Parsons, M., I. Reeve, J. McGregor, G. Marshall, R. Stayner, J. McNeill, P. Hastings, S. Glavac, and P. Morley. 2020. The Australian Natural Disaster Resilience Index: Volume II - Chapter 2: Indicators. Bushfire and Natural Hazards CRC, Melbourne, Australia.

Polain de Waroux, Y., M.-C. Carignan, O. del Giorgio, L. Díaz, L. Enrico, P. Jaureguiberry, M. L. Lipoma, F. Mazzini, and S. Díaz. 2024. How do we study resilience? A systematic review. People and Nature 6(2):474-489. https://doi.org/10.1002/pan3.10603

Quinlan, A. E., M. Berbés-Blázquez, L. J. Haider, and G. D. Peterson. 2016. Measuring and assessing resilience: broadening understanding through multiple disciplinary perspectives. Journal of Applied Ecology 53(3):677-687. https://doi.org/10.1111/1365-2664.12550

Reyers, B., M.-L. Moore, L. J. Haider, and M. Schlüter. 2022. The contributions of resilience to reshaping sustainable development. Nature Sustainability 5(8):657-664. https://doi.org/10.1038/s41893-022-00889-6

Rocha, J. C., G. Peterson, Ö. Bodin, and S. Levin. 2018. Cascading regime shifts within and across scales. Science 362(6421):1379-1383. https://doi.org/10.1126/science.aat7850

Rockström, J., A. V. Norström, N. Matthews, R. (Oonsie) Biggs, C. Folke, A. Harikishun, S. Huq, N. Krishnan, L. Warszawski, and D. Nel. 2023. Shaping a resilient future in response to COVID-19. Nature Sustainability 6(8):897-907. https://doi.org/10.1038/s41893-023-01105-9

Rose, A. 2007. Economic resilience to natural and man-made disasters: Multidisciplinary origins and contextual dimensions. Environmental Hazards 7(4):383-398. https://doi.org/10.1016/j.envhaz.2007.10.001

Ross, S. R. P.-J., O. L. Petchey, T. Sasaki, and D. W. Armitage. 2023. How to measure response diversity. Methods in Ecology and Evolution 14(5):1150-1167. https://doi.org/10.1111/2041-210X.14087

Satapathy, S., S. Dang, R. Sagar, and S. N. Dwivedi. 2022. Resilience in children and adolescents survived psychologically traumatic life events: a critical review of application of resilience assessment tools for clinical referral and intervention. Trauma, Violence, & Abuse 23(1):288-300. https://doi.org/10.1177/1524838020939126

Scheffer, M., S. R. Carpenter, V. Dakos, and E. H. van Nes. 2015. Generic indicators of ecological resilience: inferring the chance of a critical transition. Annual Review of Ecology, Evolution, and Systematics 46(1):145-167. https://doi.org/10.1146/annurev-ecolsys-112414-054242

Sellberg, M. M., S. J. Lade, J. J. Kuiper, K. Malmborg, T. Plieninger, and E. Andersson. 2024. Operationalizing pathway diversity in a mosaic landscape. Ecology and Society 29(3):26. https://doi.org/10.5751/ES-15255-290326

Semplici, G., L. J. Haider, R. Unks, T. S. Mohamed, G. Simula, P. Tsering (Huadancairang), N. Maru, L. Pappagallo, and M. Taye. 2024. Relational resiliences: reflections from pastoralism across the world. Ecosystems and People 20(1):2396928. https://doi.org/10.1080/26395916.2024.2396928

Sharifi, A. 2016. A critical review of selected tools for assessing community resilience. Ecological Indicators 69:629-647. https://doi.org/10.1016/j.ecolind.2016.05.023

Steinmann, P., H. Tobi, and G. A. K. van Voorn. 2024. Resilience Metrics for Socio-Ecological and Socio-Technical Systems: A Scoping Review. Systems 12(9):357. https://doi.org/10.3390/systems12090357

Stirling, A. 2007. A general framework for analysing diversity in science, technology and society. Journal of The Royal Society Interface 4(15):707-719. https://doi.org/10.1098/rsif.2007.0213

Summers, J. K., L. C. Harwell, L. M. Smith, and K. D. Buck. 2018. Measuring community resilience to natural hazards: The Natural Hazard Resilience Screening Index (NaHRSI) – Development and application to the United States. GeoHealth 2(12):372-394. https://doi.org/10.1029/2018GH000160

UN-Habitat. 2018. Guide to the City Resilience Profiling Tool.

Walker, B., A.-S. Crépin, M. Nyström, J. M. Anderies, E. Andersson, T. Elmqvist, C. Queiroz, S. Barrett, E. Bennett, J. C. Cardenas, S. R. Carpenter, F. S. Chapin, A. de Zeeuw, J. Fischer, C. Folke, S. Levin, K. Nyborg, S. Polasky, K. Segerson, K. C. Seto, M. Scheffer, J. F. Shogren, A. Tavoni, J. van den Bergh, E. U. Weber, and J. R. Vincent. 2023. Response diversity as a sustainability strategy. Nature Sustainability 1-9. https://doi.org/10.1038/s41893-022-01048-7

Walker, B., L. Gunderson, A. Kinzig, C. Folke, S. Carpenter, and L. Schultz. 2006. A handful of heuristics and some propositions for understanding resilience in social-ecological systems. Ecology and Society 11(1):13. https://doi.org/10.5751/ES-01530-110113

Walker, B., C. S. Holling, S. Carpenter, and A. Kinzig. 2004. Resilience, adaptability and transformability in social-ecological systems. Ecology and Society 9(2):5. https://doi.org/10.5751/ES-00650-090205

West, S., L. J. Haider, T. Hertz, M. Mancilla Garcia, and M.-L. Moore. 2024. Relational approaches to sustainability transformations: walking together in a world of many worlds. Ecosystems and People 20(1):2370539. https://doi.org/10.1080/26395916.2024.2370539

Westley, F., P. Olsson, C. Folke, T. Homer-Dixon, H. Vredenburg, D. Loorbach, J. Thompson, M. Nilsson, E. Lambin, J. Sendzimir, B. Banerjee, V. Galaz, and S. van der Leeuw. 2011. Tipping Toward Sustainability: Emerging Pathways of Transformation. Ambio 40(7):762-780. https://doi.org/10.1007/s13280-011-0186-9

Westley, F. R., O. Tjornbo, L. Schultz, P. Olsson, C. Folke, B. Crona, and Ö. Bodin. 2013. A theory of transformative agency in linked social-ecological systems. Ecology and Society 18(3):27. https://doi.org/10.5751/ES-05072-180327

Whaley, L. 2018. The Critical Institutional Analysis and Development (CIAD) Framework. International Journal of the Commons 12(2):137-161. https://doi.org/10.18352/ijc.848

Zhang, J., D. Tyfield, and L. Liu. 2025. How can food system actors influence food system resilience? A literature review via an actor-based lens. Environmental Research: Food Systems 2(2):022001. https://doi.org/10.1088/2976-601X/adce0e

Zhou, Y., J. Wang, and H. Yang. 2019. Resilience of transportation systems: concepts and comprehensive review. IEEE Transactions on Intelligent Transportation Systems 20(12):4262-4276. https://doi.org/10.1109/TITS.2018.2883766

Zobel, C. W. 2011. Representing perceived tradeoffs in defining disaster resilience. Decision Support Systems 50(2):394-403. https://doi.org/10.1016/j.dss.2010.10.001