Promotion of community resilience: does citizens have a role to play?

ABSTRACT

The promotion of community resilience is a multifaceted process that warrants further exploration. This article aims to contribute to this literature stream by assessing the role of citizen science (CS) approaches in the development of community resilience, since CS is considered a promising approach for generating new knowledge by fostering the participation of citizens (non-professional scientists) in research activities. The results show that CS approaches can contribute to community resilience at the institutional level, improving emergency and recovery planning capacity; at the infrastructural level, by contributing to land use planning and urban design; and at the social level, through the creation of community ties and social capital. This contribution is relevant for the development of public policies that are more shaped to local contexts and aligned with communities’ needs and expectations.

KEYWORDS:

• Citizen scientists
• resilient
• citizen science
• community resilience
• participatory methods

Introduction

In recent years, natural and technological disasters with devastating consequences for local communities have become increasingly frequent around the world. As a result, the concept of resilience has been gaining increased relevance in the policy arena, since governments have realised that they cannot prevent all natural, technological or social hazards from occurring, but can instead promote forms of risk adaptation and management to improve resilience and minimise its impact on humans and other systems (Renschler et al. 2010). In this case, resilience is generally defined as an ability or capacity to adapt to change, whether related to an individual or a community (Holling 1973; Magis 2010; Saja et al. 2021). In the case of a community, resilience is perceived as the ability of community members to engage community resources “to thrive in an environment characterized by change, uncertainty, unpredictability, and surprise” (Magis 2010, 402).

Despite the increasing number of publications in the field of community resilience, there is still a paucity of guidance on how to apply these insights in practice (Glass et al. 2022). Thus, “what makes a community resilient?” is the underlying question that many researchers are trying to answer, as finding a straightforward answer is not without its challenges. Firstly, the lack of a widely accepted definition of resilience encourages the emergence of ideas that purport to promote different things. Secondly, a community can suffer the effects of multiple types of natural and technological hazards (floods, earthquakes, chemical contaminations, etc.), so its resilience can change depending on the hazard or if multiple hazards occur simultaneously. For example, a community may be considered resilient to flooding due to the implementation of a certain warning system, but it may be less resilient to a fire or a pandemic. Thus, the factors that contribute to the resilience of a community vary according to the hazard it faces. For this reason, most studies have considered resilience in specific contexts, such as in the case of a specific hazard type (floods, earthquakes, etc.) or location (urban or peripheral location), and cannot be easily extrapolated to other contexts. Yet, Berkes and Ross (2013) highlighted that, in the context of a community, “generalized resilience” is desirable, as it provides flexibility to deal with a wide range of crises. Thirdly, a local community is considered a composite entity, as local organisations and citizens can differ concerning their behaviour, as well as the resources they are able to raise in the face of a disaster. As a consequence, resilience can vary in terms of space and time within a community. For this reason, several researchers have argued that resilience should lead to the long-term incremental evolution of a system (Sharifi 2016), which can only be achieved by considering all dimensions of a community (Sharifi 2016), and where the active participation of citizens cannot be ignored.

This paper explores the role of citizen science (CS) approaches in the development of community resilience. CS refers to the involvement of members of the public in scientific research activities, typically in collaboration with professional scientists or scientific institutions. CS projects often use participatory methods as a means to enroll the general public into the different phases of the research lifecycle along with “professional” scientists (Vadjunec et al. 2022). According to Vadjunec et al. (2022), CS has the potential to change the role of science in the face of the current and imminent challenges to environmental sustainability, and in developing resilience. Compared to other forms of public enrollment in science, CS comprehends the “active engagement” of non-professional scientists in different research activities to different extents (Buytaert et al. 2014). This approach has the potential to strengthen not only local autonomy and adaptive capacity, but also the adoption of strategies and measures that are suited to local contexts and congruent with local worldviews, beliefs, values, and aspirations, and thus produce more effective and sustained outcomes (Kirkby, Williams, and Huq 2018).

The present study aims to contribute to both theory and practice by exploring ways in which community resilience can be developed. In the case of the resilience research field, this article discusses the relevance of community members in the design and implementation of hazard mitigation measures. The literature points out that in the face of major disruptive events, resilience depends first on the actions of people operating at the individual and neighborhood scale (Renschler et al. 2010). Therefore, by demonstrating that citizen engagement in scientific activities can contribute to community capacity to act and recover from disaster events, it would contribute to increasing research in this area and thus developing new insights on how to increase community resilience.

Also, this study can provide potential contributions to practice, since top-down disaster risk measures often fail for vulnerable communities due to low rates of community implementation. The main causes for this to happen are twofold. On the one hand, government authorities often ignore – or underestimate – the local dynamics, culture and activities which are determinant in managing the impacts of hazard events (Cadag and Gaillard 2012). On the other hand, local communities rarely have the technical and scientific knowledge necessary to understand and implement risk prevention and mitigation measures at the local level, which then has an impact on its adoption by citizens (Cadag and Gaillard 2012). Such gap in terms of action and knowledge, is considered a major obstacle for improving disaster-resilience approaches (Cadag and Gaillard 2012). For this reason, there is a general agreement that “a greater understanding of the dynamics of vulnerabilities, hazardous exposure and resilience can only be gained if the knowledge creation process is seated within, and by those affected” (Van Niekerk et al. 2018, 411). Therefore, only by involving the community in risk identification, data collection or data analysis (McCormick 2012), as well as “enhancing the skills, knowledge, and capacities of local communities’”(Van Niekerk and Annandale 2013, 164), it will be possible to develop prevention and mitigation measures that will be more adjusted to local realities, especially in case of state abandonment (Roque et al. 2021).

Community resilience: harnessing citizen contributions

According to Aldrich and Meyer (2015, 255), “community resilience describes the collective ability of a neighbourhood or geographically defined area to deal with stressors and effectively resume the rhythms of daily through cooperation following shocks”. Therefore, resilience is a latent capability of a community that only manifests itself in the face of a disturbance, allowing a community to prepare itself against an identified threat (ability to plan/prepare), to absorb the consequences of a disaster while maintaining a certain degree of system functioning (ability to absorb), recovering at post-shock level (ability to recover) and improving the community's capacity to adapt and recover from future disasters (ability to adapt) (Figure 1). These different resilience abilities can be developed overtime through planning, collective action, innovation, and learning (Franklin, Newton, and Mcentee 2011; Magis 2010).

Figure 1. Resilience abilities. Source: Adapted from National Research Council (2012).

However, communities are not static or well-defined entities that remain constant before, during and after a disaster (Barrios 2014), since those are shaped by the intervention of both external and internal agents before and after disasters. In this case, citizens may play a determinant role in the development of resilience within a specific location.

According to Magis (2010), community resilience is related to the capacity of community members to engage community resources to thrive in the face of unpredictable events. In turn, this ability depends on community’s empowerment, as an empowered community is better able to anticipate and adapt to stresses and changes and transform itself into more desirable development states (Glass et al. 2022; Skerratt and Steiner 2013); as well as community’s social capital, which includes social networks, trust, social resources and community cohesion (Norris et al. 2008; Rasmussen, Armstrong, and Chazdon 2011). Thus, resilient communities are built through active citizenship (Madsen and O’Mullan 2014), such as volunteering activities.

Citizen science (CS) is “a form of research collaboration that involves volunteers in producing authentic scientific research” (Wiggins and Crowston 2015, p. 1). It is considered a participatory science approach in which active citizens and professional scientists work together in the development of research projects (Hecker and Taddicken 2022). There are different levels of citizen involvement in SC, ranging from a minimal level of involvement, where citizens facilitate access to data but do not actively promote their collection, to participation in all phases of a research project, namely the definition of the research question and research approach, as well as in the selection of research methods, conducting pilot study and data collection, data analysis, dissemination of results and implementation of findings (Table 1).

Table 1. Levels of engagement and participation in citizen science projects. Based on the work of Haklay (2013), Azizi et al. (2022) and Gelling (2015).

Type of CS	Crowdsourcing	Distributed intelligence	Participatory science	Extreme citizen science
Level of citizens’ cognitive involvement	Minimal	Simple interpretation activity	Cognitive involvement with support from experts	High cognitive involvement
Information flow	One-way	Bidirectional	Bidirectional	Bidirectional
Stages of the research process*	Data collection Access to the research site and data	Data collection Access to the research site and data	Research question Access to the research site and data Methodology design Data collection Data analysis (with support)	Research question Access to the research site and data Methodology design Data collection Data analysis Dissemination of results
Type of participation	Pooling of resources (e.g. volunteered computing) Data collectors (citizens as sensors)	Citizens as basic interpreters Volunteered thinking	Participation in the problem definition, data collection, and data analysis (with support from experts)	Definition of problems and methods, data gathering and analysis, publication and/or utilisation of results (co-creation)

By implementing a CS approach, it is possible to overcome duration and scope problems in the collection of large amounts of data, which usually require a human observer to be classified at minimal cost (Kullenberg and Kasperowski 2016; Theobald et al. 2015). In addition, CS approaches can contribute to increasing society’s understanding and acceptance of science (Trumbull et al. 2000), as well as bringing new insights and concerns to the academic arena (Shirk and Bonney 2019). As such, “citizen science (…) provides avenues for interrogating topics that have both scientific and social relevance – a prime nexus for informing policy” (Shirk and Bonney 2019, 43). Therefore, CS as the potential to enhance, on the one hand, the participation of community members in defining and implementing disaster mitigation strategies, as well as informing policy makers about the specific needs of a community, as well as existing resources that may impact any design of public policies.

Community resilience: main determinants

Callaghan and Colton (2008) argue that community resilience can be enhanced through investments in all the various forms of community capital, such as environmental capital, human capital, social capital, cultural capital, public structural capital, and commercial capital. In this vein, Vaneeckhaute et al. (2017) suggested that promoting community development can strengthen the resilience of a community. In turn, Norris et al. (2008, 127) argue that “to build collective resilience, communities must reduce risk and resource inequities, engage local people in mitigation, create organisational linkages, boost and protect social supports, and plan for not having a plan”. This could be achieved through the development of a set of primary adaptive capacities, such as economic development, social capital, information and communication, and community competence. In the same vein, Callaghan and Colton (2008) suggest that a community is resilient when the following set of critical elements are in place: plans and strategies to minimise vulnerabilities; communication and crisis response systems; government/private partnerships and independent initiatives that create social support; and strategies that diversify risk across space, time, and institutions.

As community resilience continues to gain relevance, more focus is being placed on developing frameworks and tools to measure it. These measurements can help to identify improvement factors and, thus, aid in the development of public policies and strategies. However, the considerable increase in the number of studies addressing this topic in recent years has led to different definitions of resilience, research approaches, contexts, and focuses (Tariq, Pathirage, and Fernando 2021) that make it difficult to select which tools are better suited to a given context. Nonetheless, considering the review articles published in the last years, there is a general agreement regarding the main dimensions and sub-dimensions of resilience, despite a not-so-consensual selection of indicators/measures (Table 2). For instance, Cutter (2016) proposed dividing the most common elements of community resilience into attributes and assets (economic, social, environmental, and infrastructural) and capacities (social capital, community functions, connectivity, and planning). Also, through a literature review of 36 community resilience tools, Sharifi (2016) identified the most common dimensions for assessing resilience, such as environmental, social, environmental, infrastructural, and institutional. With a similar approach, Tariq, Pathirage, and Fernando (2021) identified six critical dimensions of resilience: physical, health, economic, environmental, social, and governance. Table 2 summarises the main dimensions, sub-dimensions, and indicators of community resilience.

Table 2. Main dimensions, sub-dimensions, and indicators of community resilience. Based on Sharifi (2016), Cutter (2016), and Saja et al. (2021).

Dimensions	Sub-dimensions	Proxy variables used most often
Economic	Structure	Employment; income
	Security and stability	Insurance; social welfare
	Dynamism	Economic diversity
Social	Community bonds	Volunteerism; civic engagement in social networks
	Social capital	Social trust; diverse workforce
	Safety and well-being	Crime prevention and reduction; physical and psychological health; preventive health measures
	Local culture	Past experience with disaster recovery; indigenous knowledge and traditions; cultural and historical preservation
Environmental	Resource management	Resource management (production, consumption, conservation, recycling, etc.)
	Ecosystem protection	Ecosystem monitoring and protection; biodiversity and wildlife protection
	Presence of natural assets	Availability and accessibility of resources (air, energy, water, food, oil, etc.)
Infrastructural	Robustness and redundancy	Redundancy of critical infrastructure, facilities, and stocks; robustness and fortification (of critical infrastructure, vital assets, ecosystems, etc.)
	Land use and urban design	Accessibility of basic needs and services over time (food, water, shelter, energy, health, and education); site selection and avoidance of risk and habitat areas (floodplains, flood-prone areas, exposed coastal zones, etc.)
Institutional	Emergency and recovery planning	Mitigation plans; mitigation activities
	Leadership and participation	Multi-stakeholder planning and decision making; strong leadership; transparency and accountability

Research methodology

This study adopted an evidence-based approach to investigate the contributions of citizen science (CS) initiatives to community resilience. An evidence-based approach involves systematically gathering and evaluating relevant evidence from empirical research, expert opinions, and theoretical frameworks to inform decision-making and practice (Hoon 2013; Rauch, van Doorn, and Hulsink 2014). In this research, we conducted a comprehensive literature review using multiple databases, such as Scopus and Web of Science, to identify relevant articles on CS and resilience building. The search was performed in December 2022, by applying the following search query (resilien* AND (“citizen science” OR “crowd science”)) to the “title, abstract and keywords” field, and specific inclusion criteria were applied to select articles for analysis. These criteria encompassed subject areas such as social sciences, business, management, and economics, document types including articles and review articles, and language (English). No time limitations were imposed to ensure a thorough examination of the existing literature. Following the retrieval of articles, a systematic analysis was conducted to extract key findings, identify common themes, and draw conclusions regarding the role of CS in promoting community resilience.

Figure 2 summarises the number of documents included and excluded from the review. The keyword search retrieved 377 articles from the selected electronic databases, with this number being reduced to 33 articles after applying the selected inclusion criteria, and the removal of duplicates. After the analysis of articles, a final set of 12 articles were considered eligible for analysis.

Figure 2. Summary of the PRISMA method procedure.

After selecting the articles, each author separately classified the articles according to the level of citizen engagement, resilience capacity and impact on resilience dimensions and sub-dimensions. Classification differences were discussed in order to reach a consensus.

Results and discussion

The different case studies reported in the selected articles describe CS initiatives in different geographic locations, such as Nepal, Puerto Rico, Brazil, Italy, USA and Australia, and focus on natural environmental hazards, such as floods, extreme heat and volcanic events (Table 3). Regarding the types of SC engagement, distributed intelligence, in which citizens carry out simple activities of interpretation and data collection, and participatory science, where citizens participate in problem definition, data collection and data analysis (with the support of specialists), were the most used types. Extreme citizen science was not used in either case. Thus, there is a preference for CS approaches that involve less citizen participation in complex activities, as they are possibly easier and faster to implement.

Table 3. Summary of the articles’ main focus.

Reference (author, year)	Study site	Project goal	Type of data collected	Level of citizens’ engagement	Research stages	Focus of resilience ability	Impact on resilience dimensions and sub-dimensions
Parajuli et al. (2020)	Nepal (Bajhang and Bajura)	Co-create geospatial knowledge of Nepal's remotest districts using remote mapping	Geographical data such as the location of community resources (e.g. health facilities, financial institutions, and public assets) using the OpenStreetMap platform	Distributed intelligence	Access to the research site and data; Data collection; Data cleaning and validation (with support)	Plan/prepare	Social / Community bonds Environmental / Resource management; Presence of natural assets Infrastructural / Land use and urban design; Emergency and recovery planning
Pandeya et al. (2021)	Nepal	Explore how low-cost sensor technology can be used with a CS approach to improve local resilience to flood risk	River level data	Participatory science	Methodology design; Access to the research site and data; Data collection; Data analysis (with support)	Plan/prepare	Social / Community bonds; Social capital Environmental / Resource management Infrastructural / Robustness and redundancy Institutional / Emergency and recovery planning
Rossi et al. (2022)	Italy (Perugia)	Assess the state of the urban forest in a neighbourhood of a densely built-up city	Inventory and categorisation of trees	Distributed intelligence	Data collection	Plan/prepare	Social / Social capital Environmental / Ecosystem protection; Presence of natural assets Infrastructural / Land use and urban design
Mahajan et al. (2021)	Taiwan	Monitor air quality	Air quality (PM2.5)	Participatory science	Methodology design; Access to the research site and data; Data collection; Data analysis (with support)	Absorb; Recover	Social / Community bonds; Social capital; Safety and well-being Environmental / Ecosystem protection Infrastructural / Land use and urban design Institutional / Emergency and recovery planning
Zhao et al. (2021)	USA (Phoenix)	Understand how Phoenix residents experience and behave in an extreme heat environment	Location/time, climate, and human activity data on their exposure to heat	Crowdsourcing	Data collection	Absorb	Social / Safety and well-being Institutional / Emergency and recovery planning
Thomas et al. (2016)	Australia	Describe a community program for education, training, and field sampling of acid sulfate soils	Water level and field soil sampling	Distributed intelligence	Access to the research site and data; Data collection	Recover	Social / Community bonds Environmental / Ecosystem protection Institutional / Emergency and recovery planning; Leadership and participation
Alves et al. (2021)	Brazil (Campina Grande)	Evaluate similarities and differences in residents’ risk perception and coping capacity in the context of multiple hazards	Social context in which multiple water-related hazards take place	Distributed intelligence	Access to the research site and data; Data collection	Recover	Social / Local culture Environmental / Presence of natural assets Infrastructural / Land use and urban design Institutional / Emergency and recovery planning
Stablein et al. (2022)	Puerto Rico	Implement disaster response and resilience strategies using a collaborative system between transdisciplinary researchers and local NGOs	Geospatial data	Distributed intelligence	Access to the research site and data; Data collection	Recover	Social / Community bonds; Social capital Environmental / Resource management; Presence of natural assets Infrastructural / Land use and urban design Institutional / Emergency and recovery planning
Stone et al. (2014)	Ecuador (Tungurahua)	Describe a network of volunteers engaged in community-based volcano monitoring to mitigate the risk of volcanic eruptions	Volcano observations	Distributed intelligence	Access to the research site and data; Data collection;	Recover; Adapt	Economic / Structure Social / Community bonds; Social capital Environmental / Ecosystem protection Infrastructural / Land use and urban design Institutional / Emergency and recovery planning; Leadership and participation
Zeng et al. (2020)	China (Wuhan)	Demonstrate the value of public and media observations in understanding the predisposing factors to urban flooding as a contribution to flood hazard assessment	Flooding locations, e.g. streets and roads	Crowdsourcing	Data collection	Adapt	Social / Local culture Infrastructural / Land use and urban design Institutional / Emergency and recovery planning
Hoffman (2020)	USA (Richmond and Virginia)	Assess how extreme heat events are amplified within city landscapes (also known as the urban heat island effect)	Data describing climatological conditions	Participatory science	Methodology design; Data collection; Data analysis (with support)	Adapt	Social / Community bonds; Safety and well-being Environmental / Presence of natural assets Infrastructural / Land use and urban design Institutional / Emergency and recovery planning; Leadership and participation
Vadjunec et al. (2022)	USA (Oklahoma, New Mexico, Colorado, Kansas, and Texas)	Create maps showing areas of drought, flash floods, invasive species, and other environmental hazards	Local knowledge of different types of environmental hazards	Participatory science	Development of the research question; Data collection	Adapt	Social / Community bonds; Social capital; Local culture Environmental / Resource management; Ecosystem protection; Presence of natural assets Infrastructural / Land use and urban design Institutional / Emergency and recovery planning; Leadership and participation

In the case of resilient abilities, three articles reported how CS affects the ability to plan and prepare for hazards, mainly those related to the environment (e.g. droughts, flooding, and invasive species). In these cases, SC initiatives helped to collect data that could be used to better understand local resources, capacities, and vulnerabilities, especially using distributed intelligence approaches. This information was relevant for the development of early-warning systems and hazard prevention plans. Two articles highlighted the ability to absorb the consequences of a shock. Different CS approaches (but mainly participatory science) were used to collect data on people’s perceptions, behaviours, and attitudes during a disaster. In addition, these approaches were useful to understand what actions people took to mitigate hazards, as well as how these mitigation measures impacted local well-being. In turn, four articles focused on the post-shock level, namely how systems recovered their functionality after a major hazard or disaster. These CS projects implemented distributed intelligence approaches and were relevant to understand the communities’ coping capacities and the mitigation strategies adopted to overcome the main consequences of the hazards. This information was important for assessing the communities’ preparedness and their ability to apply measures of protecting against hazards or reducing their impact. Finally, three articles described how CS can improve a system’s capacity to absorb and recover from shocks. The capacity to adapt concerns building the resilience to cope with future hazards and other stressful events by empowering citizens to design and implement preventive measures on their own. These participatory science approaches helped to develop human and social capital, as well as design prevention strategies based on the knowledge and experience of past hazard events.

Regarding the impact of CS on community resilience, all CS initiatives reported an impact on the institutional dimension, especially on improving emergency and recovery planning capacity. Furthermore, most cases (9 cases out of 12) had an impact on the infrastructural dimension by contributing to land use and urban design; and on the social dimension by promoting community ties (8 cases out of 12) and social capital (6 cases out of 12).

Resilience ability: plan/prepare

The ability to plan/prepare is crucial to promote resilience since it allows the effective detection and monitoring of potential hazards, as well as the design of effective measures to mitigate their negative consequences on the community. CS initiatives play a pivotal role in this process by facilitating data collection, analysis, and decision-making processes informed by local knowledge and experiences. For instance, (Pandeya et al. 2021) demonstrated the use of CS in Nepal to collect data on flood-prone areas, leveraging low-cost detection technology. This allowed researchers to gather data on resources, capacities and vulnerabilities of local remote areas and, thereby, overcome the data limitation in a data scarce region, promoting the development of an effective community flood early warning system. In addition, this project allowed educating local stakeholders and empowering them to build resilience to floods.

Similarly, Parajuli et al. (2020) used a CS approach and remote mapping to co-create geospatial knowledge of the remotest districts of Nepal. This process involved people from different backgrounds, such as students, government officials from municipal and ward offices, IT officers from the respective municipalities, representatives from civil society organisations (CSOs), and the Nepal Police and the Nepali Army, who were trained in the preparation and interpretation of maps through practical sessions. Citizen scientists gathered, processed, analyzed, and validated data together with experts. According to Parajuli et al. (2020, 10), as a result of the training and experience gathered in the project, these citizen scientists can “play an important future role in bringing new innovation to shape local-level planning and program implementation”. Furthermore, as mentioned by the mayor of the local municipalities, the information gathered through this CS approach was crucial to build resilience to natural hazards by achieving a better understanding of the local context and training new citizen scientists who are relevant resources for each municipality.

Rossi et al. (2022) described a project related to the urban forest of a neighbourhood in a densely built-up Italian city, Perugia. This project involved citizens in field research to build a tree inventory by collecting data on the localisation, number, and species of trees, percentage of ground, and area covered by trees. Their involvement had two main contributions. On the one hand, it helped to raise awareness of the relevance of urban forests in the urban context. On the other hand, it helped to develop two geo-specific tools: one for supporting tree-planting decisions to enhance ecosystem service provision, and another for supporting economic evaluations differentiated by species in the definition of taxations linked to trees damaged or cut down, which could promote a more resilient city to specific hazards. For example, in urban areas subject to frequent flooding, urban planners should look for trees that help manage rainwater, and, along roads with heavy traffic, they should look for trees that help remove pollution.

These examples underscore the transformative potential of CS in facilitating community-led resilience initiatives and empowering individuals to actively contribute to risk reduction and preparedness efforts. Drawing upon the theoretical frameworks of participatory action research and community engagement, these initiatives highlight the importance of democratizing knowledge production and fostering inclusive decision-making processes for building resilient communities (Fraser et al. 2006).

Resilience ability: absorb

In the face of hazardous events, resilience hinges on the actions of individuals and communities to absorb and adapt to shocks effectively (Renschler et al. 2010). Therefore, it is important to understand people’s behaviour and attitudes in order to assess the social impact of hazards. Nevertheless, observational data during a disaster is often lacking. Moreover, long-term social, economic, and environmental community stresses include chronic issues such as recurring droughts, flooding conditions, and long-term heat stress (Cutter 2021), which can cause similar or even higher levels of damage to human communities over the long run. Despite this, these types of stress can be difficult to track and monitor, because each individual experiences them differently over time (Zhao et al. 2021). CS initiatives contribute to this process by providing real-time data on environmental hazards and facilitating informed decision-making. For instance, Zhao et al. (2021) used a CS project to better understand the risk of heat exposure in the Phoenix area, where summer temperatures can exceed 49°C. In this case, volunteers were recruited to collect data on location/time, climate, human activities and heat exposure during their daily routine. By harnessing citizen-generated data, researchers gained insights into daily behaviour patterns and vulnerability to heat waves, informing targeted interventions to mitigate heat-related risks.

Similarly, Mahajan et al. (2021) implemented a CS initiative in Taiwan to monitor air quality, leveraging participatory sensing systems to collect and disseminate real-time data. This collaborative approach not only raised awareness of air pollution issues but also empowered citizens to make informed choices regarding outdoor activities, thereby enhancing community resilience to environmental hazards.

According to Elwood (2008), participatory sensing and citizen sensing are collaborative processes involved in data collection and analysis, which highlight the role of community engagement and empowerment in generating actionable knowledge and promoting collective action. Similarly, CS initiatives can enhance community resilience by promoting knowledge sharing and adaptive responses to environmental hazards (Goodchild, 2007).

Resilience ability: recover

The aftermath of a disruptive event “can often escalate from something anticipated, prepared for, and seemingly manageable into more complex problems as hazardous events become connected and compound” (Stablein et al. 2022, 1). As each combination of hazards events may impact communities in distinct ways, recovery decisions and plans have to be designed according to the context where the event has occurred. Nevertheless, academic and public policy perspectives on how to handle disaster recovery may be incomplete unless the lived experience of those directly impacted is taken into account (Stablein et al. 2022). CS can provide a deeper understanding of the social environment in which disasters take place, as well as the social vulnerabilities felt by each community. According to Alves et al. (2021), in recent decades, Brazil has been severely affected by droughts and floods. Through the use of several participatory methods, including surveys, informal meetings, workshops, and discussion groups, researchers have been able to map the coping measures in place. In this case, the researchers found that an increased adoption of risk mitigation measures and thus a greater resilience is dependent on three factors: information, incentive, and trust.

Similarly, Thomas et al. (2016) described a project consisting of a community education, training, and field sampling program for measuring acid sulfate soils, which involved citizens collecting samples and categorising them by colour, odour, structure and consistency, and pH. In this case, the development of the project was motivated by the potential increase in acid sulfate soils in lakes due to drought and the resulting decline in water levels. In turn, Mahajan et al. (2021) provided an example of how the results of their project aided the recovery process during a fire incident, highlighting how AirBox visualisation services were used to reduce the damage caused by a pollution episode and allowed local officials to advise citizens to take protective measures, such as wearing dust masks.

Finally, Stablein et al. (2022) presented the implementation of a collaborative system involving transdisciplinary researchers and local NGOs that aimed to implement disaster relief and resilience strategies. This collaboration took place in Puerto Rico, where communities have faced the complex and compounded effects of multiple and diverse disasters. In this case study, the authors presented a virtual geospatial CS program with local stakeholders aimed at co-creating local and external scientific knowledge to strengthen communities’ resilience to disasters.

As mentioned by Pretty, Smith, and Thompson (2003), community-driven approaches and inclusive decision-making processes are relevant to build resilient communities. In this case, CS initiatives can foster community empowerment and resilience through bottom-up approaches to recovery planning and implementation.

Resilience ability: adapt

The design of measures to improve a system’s capacity to absorb and recover from shocks is based on past experience. CS initiatives have the potential to generate knowledge that is useful for the design of disaster management plans and disaster coping strategies. For instance, Vadjunec et al. (2022) used participatory mapping projects to identify places characterised by drought, flash flooding, the encroachment of invasive species, and other environmental hazards within five states in the USA. This initiative had impacts both on social capital and on local awareness regarding the research project. Moreover, it allowed the researchers to gain a better understanding of potentially unforeseen hazards.

In Wuhan (China), Zeng et al. (2020) sought to demonstrate the value of public and media observations of flooded areas, such as streets and roads, after or during storms, in contributing to a better understanding of urban flooding. By helping to identify the locations and time periods that are most sensitive to climate and human influence, these observations contribute to flood hazard assessment and can provide scientific knowledge to aid decision-makers in the planning of flood management and the design of drainage strategies.

Furthermore, Hoffman (2020) described a CS project where community volunteers collected temperature data during certain hours of the day while driving predetermined routes, with the aim of studying the effect of heat in the city of Richmond (USA). The data collected was used to develop a heat vulnerability index map showing areas that may need to be prioritised for action due to excessive vulnerability. Thus, this CS project was relevant not only because it improved citizens’ literacy regarding the effects of heat, but also because the Urban Heat Vulnerability Index can be used as a guide to design urban space in Richmond's hottest and most vulnerable neighbourhoods and develop specific solutions. emergency and recovery measures to increase resilience to extreme heat.

Resilience adaptation is dependent on flexibility, learning, and collaboration in responding to complex challenges (Berkes, Folke, and Colding 2003). CS initiatives can enhance community resilience by promoting adaptive responses that are responsive to local contexts and informed by diverse perspectives.

Citizen science for community resilience enhancement

Community resilience is a multifaceted concept encompassing the ability of communities to prepare for, absorb, recover from, and adapt to various hazards and shocks. Each phase of community resilience benefits uniquely from the participation and integration of citizen science efforts, thereby amplifying the overall effectiveness of resilience-building strategies.

Citizen science plays a pivotal role in the planning and preparation phase by empowering communities to collect and analyse data pertinent to potential hazards. Through active participation in data collection, citizens contribute local knowledge and insights, enhancing hazard assessment and informing the development of early warning systems. By leveraging low-cost technologies and participatory mapping techniques, citizen science initiatives bridge data gaps in resource-limited areas, thereby facilitating more informed decision-making processes and strengthening community preparedness.

During the absorption phase, citizen science initiatives provide real-time data on environmental conditions and community behaviours, enabling communities to respond effectively to immediate hazards. By engaging citizens in monitoring and reporting activities, these initiatives enhance situational awareness and facilitate rapid response efforts. Moreover, citizen science fosters community empowerment and resilience by promoting knowledge sharing and facilitating adaptive responses to changing environmental conditions.

In the aftermath of a hazard event, citizen science contributes to recovery and restoration efforts by assessing damages, mapping coping capacities, and facilitating resource allocation. Through participatory methods and collaborative data collection, citizen science initiatives capture the lived experiences of affected communities, informing post-disaster recovery strategies and guiding equitable distribution of resources. By fostering community-driven approaches to recovery planning and implementation, citizen science enhances social cohesion and promotes sustainable rebuilding efforts.

Finally, in the adapt phase, citizen science initiatives serve as catalysts for long-term resilience building by identifying emerging hazards, informing resilient infrastructure planning, and promoting knowledge sharing and adaptive responses. By engaging citizens in participatory mapping projects and collaborative research efforts, these initiatives enhance community awareness of environmental risks and vulnerabilities, fostering a culture of resilience and empowering communities to proactively adapt to changing conditions. Moreover, citizen science facilitates the co-creation of local and external scientific knowledge, strengthening community capacity to address complex challenges and build adaptive capacity over time.

Thus, by empowering communities to actively engage in scientific research and data collection, citizen science fosters knowledge exchange, promotes adaptive responses, and strengthens community capacity to prepare for, absorb, recover from, and adapt to hazards and shocks enabling continuous cycle of resilience development (Figure 3).

Figure 3. Role of citizen science for community resilience building in diverse environmental context.

Conclusion

The diverse case studies examined in this research shed light on the significant contributions of citizen science (CS) initiatives to building community resilience in various geographic locations and in response to natural environmental hazards. More specifically, SC approaches show great potential for building community resilience, namely: (1) at the institutional level, improving emergency and recovery planning capacity; (2) the infrastructural level, by contributing to land use planning and urban design; and (3) the social level, through the creation of community ties and social capital.

Through distributed intelligence and participatory science approaches, CS initiatives have facilitated data collection (i) from remote locations where data is scarce, allowing a better understanding of community resources, capacities and vulnerabilities; (ii) on people’s perceptions, behaviours and attitudes during a disaster; (iii) on citizens’ actions and activities to mitigate disasters; and, (iv) about potential future dangers, which is crucial for the development and implementation of hazard detection tools and mitigation measures. Also, most cases describe the implementation of training and knowledge-sharing activities to improve citizen’s skills to take protective measures against hazards. Finally, by fostering the cooperation between the community, academia and government, CS approaches can contribute to the development of public policies that reflect local needs and expectations, and thus are more adjusted to local contexts.

In conclusion, this study provides compelling evidence of the effectiveness of CS in promoting community resilience and highlights the importance of adopting evidence-based approaches to inform resilience-building efforts. By leveraging local knowledge and experiences, CS initiatives have the potential to enhance community resilience and contribute to more sustainable and resilient societies.

Limitations

This study used a qualitative and exploratory research method, which has some limitations. On one hand, this methodology precludes the generalisation of findings due to the restricted number of case studies examined. Nevertheless, these cases made enabled the collection of valuable information on various facets of the SC projects, as well as the potential effects on community resilience. On the other hand, the inclusion criteria may have been overly restrictive, potentially resulting in the exclusion of pertinent articles. Additionally, there could be some level of bias in the content analysis stemming from individual researchers’ interpretations of each case study description. However, efforts were made to mitigate this bias through a consensus meeting, during which researchers meticulously analysed the classifications made.

Further research

In order to promote the resilience of communities, there is a need for new, more transdisciplinary and integrative research approaches that allow understanding how to motivate and promote the involvement of citizens in the development of strategies and implementation of actions that promote the resilience of territories. Furthermore, it is important to understand which characteristics and typologies of SC projects best promote community resilience.

Measuring the resilience of territories is still a challenge, so more research in this area will be relevant in order to design tools that allow measuring such a complex phenomenon, since without measuring it is difficult to identify ways of improvement. The same goes for CS projects, given that the number of studies focused on understanding and measuring the impact of these initiatives on territories and communities is still limited.

Finally, quantitative studies on the effects of CS in the development of community resilience should be promoted in order to complement the results obtained in the present article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Portuguese Foundation for Science and Technology (FCT) through the Scientific Employment Stimulus – Institutional Call – reference [grant number: CEECINST/00026/2018]; and the SMART-ER project, funded by the European Union’s Horizon 2020 research and innovation programme under [Grant Agreement: #101016888].