Participatory Flood Risk Management and Environmental Sustainability: The Role of Communication Engagement, Severity Beliefs, Mitigation Barriers, and Social Efficacy

Abstract

Climate change has continued to cause severe coastal flooding, erosion, and storm surge in the northeastern U.S. region. Compounding the coastal storm challenge, this region also experienced multiple 1-in-100-, 1-in-200-, and 1-in-500-year rainfall events in 2024. In recent years, community-based flood risk management has become an important component for generating locally viable mitigation strategies to build environmental sustainability. At the heart of this community engagement paradigm is flood risk communication, which aims to bring together community stakeholders to strengthen their social resilience to collaborate in generating flood risk management solutions. Extant research has rarely examined the direct connection between theory-driven risk communication factors and community-based flood risk management. To better understand the role of risk communication in facilitating participatory flood risk management planning, this study integrated risk communication constructs with the relevant Health Belief Model components to propose and test a conceptual framework. Specifically, this study conducted a survey with 302 residents of a coastal community highly vulnerable to sea level rise, storm surge, and year-round flooding in the coastal region of northeastern U.S. Study results suggested that flood information exposure could drive greater perceived flood risk severity and mitigation barriers, in addition to furthering flood risk information-seeking behavior and affiliated community-engaged flood risk communication. Community-engaged communication was positively linked to perceived social efficacy beliefs in tackling flood risk management, aside from being linked to perceived flood risk mitigation response efficacy. Both perceived social efficacy and response efficacy in flood risk management positively predicted interest in participatory flood risk management planning.

1. Introduction

Climate change has continued to cause severe coastal flooding, erosion, and storm surge in the northeastern region of the U.S. [1]. The environmental hazards created by Hurricane Sandy of 2012 and other subsequent severe weather events have continued to cause visible destructions on properties, local economies, and infrastructures [2]. Compounding the coastal storm challenge, this region also experienced multiple 1-in-100-, 1-in-200-, and 1-500-year rainfall events in 2024; some of these events were triggered by the remnants of Hurricane Beryl and Hurricane Debby, which made their landfall in the Mexican Gulf and South Atlantic [3].

In recent years, the emphasis of flood and erosion hazard management has often involved addressing community-level factors [4] such as physical, political, social, and ecological interactions [5]. Community-based flood risk management has become an important component for generating locally viable mitigation strategies [6]. At the heart of this community engagement paradigm is flood risk communication, which aims to bring together community stakeholders to strengthen their social resilience (via individual and collective efficacy) to collaborate on generating flood risk management solutions [7].

When individuals residing in a flood-prone region are exposed to flood risk information, several likely cognitive responses will help determine whether or how they may engage in risk communication to adapt to or mitigate that risk. These cognitive responses can include whether these individuals (1) perceive the risk as severe enough to pose a real threat to them; (2) consider any recommended preventive or mitigation actions to be achievable; (3) engage in additional risk communication through seeking additional risk information and risk management solutions; and (4) feel confident in their own ability and external resources (including community and government capacity) to help mitigate the risk.

The cognitive responses presented above are similar to the key conceptual components of the Health Belief Model (HBM), which proposes that perceived risk severity and susceptibility, risk prevention benefits and barriers, cues to action, and self-efficacy (not necessarily in a successive order) can lead to an individual taking a protective action [8]. As the conceptual framework of the HBM lacks sequentially ordered theoretical components and comprises a cognition-only focus, researchers have applied the HBM by testing it as a process model [9] through modifying the model components or adding other variables relevant to the health topic under study [10]. The empirical literature has primarily applied the HBM to study the cognitive processes of health risk prevention and intervention [9]. Very limited empirical research has utilized this widely validated conceptual framework to flood risk prevention research.

Against this backdrop, extant research has rarely examined the direct connection between theory-driven risk communication factors and community-based flood risk management. It is worth noting that risk communication is considered a key strategy for natural disaster preparedness, adaptation, and recovery by major U.S. disaster management agencies, such as FEMA [11] and NOAA [12]; these government agencies have developed comprehensive training programs, tools, and materials to support local governments and communities in initiating climate-change-related risk communication.

Past research has shown that when communicating flood risk management, it is necessary to consider individuals’ self-efficacy to effectively prepare for or mitigate a flood risk [13]. Prior work has also suggested that community-engaged flood planning and recovery processes are a promising approach to promoting flood mitigation among participating households [4,14]. Engaging the community to collaborate on flood risk management can also build social resilience [15]. Social resilience in the current study context is conceptualized as a community’s ability to adapt the negative effects of a flood disaster through adapting and/or mitigating such effects for future flood events [16].

Taken together, limited work has explored the cognitive and behavioral factors that may help drive risk communication focusing on flood risk learning and management. To better understand the role of risk communication in facilitating participatory flood risk management planning which involved communities and government risk management entities, this study integrated risk communication constructs with the relevant Health Belief Model components as the basis to propose and test a conceptual framework. This conceptual model aimed to examine how flood risk communication might trigger threat evaluation, mitigation solution assessment, and interest in community-level participatory risk management planning in a coastal community.

Specifically, the current study conducted a survey with residents of a coastal community highly vulnerable to sea level rise, storm surge, and year-round flooding in the coastal region of northeastern U.S. In recent years, this region has experienced alarming severe storm surge [17] and flash floods [18] due to sea level rise and extreme rainfall, which caused costly disasters throughout the year. By focusing on a community that lives with frequent flood event realities and consequences, this study could better provide evidence-based results to benefit scholarly research and practitioners alike.

Following this rationale, only the theoretical components of the HBM directly relevant to exploring and explaining their relationship with risk communication and flood risk management were tested. In particular, residents of a coastal community that suffer from regular flood events are familiar with their vulnerability to flood risks and the benefits of flood mitigation, in addition to flood-related action cues in the media and their immediate environment. These realities have been confirmed by past studies that found perceived susceptibility, as an inconsistent predictor of preparedness action and cues to action, was largely absent in similar empirical research contexts [19,20]. The same is true for the construct of perceived benefits of flood risk management, which was already recognized and understood by at-risk communities [21,22].

2. Background Literature and Research Hypotheses

2.1. Flood Information, Risk Severity, Mitigation Barriers, Information Seeking, and Risk Communication

Flood risk adaptation planning begins with communication that informs individuals of the risk to elicit risk perception [23,24]. The first step of flood risk communication involves exposure to flood risk information from a range of communication sources that formally or informally help disseminate disaster advisories and warnings issued by weather forecast and emergency management units from national, state, and/or local governments [24,25]. These sources can include offline/online traditional new media [26,27,28], social media platforms [29,30,31], interpersonal networks [32,33], and grassroots community groups [24,34].

Past studies have confirmed the significant role of traditional news sources—including newspapers [35], radio [36], and television [29,37]—in facilitating government efforts to inform the public of impending flood and weather disasters. Social media platforms such as Twitter/X and Facebook are also proven venues for the public to receive and share disaster news and information updates [38,39], in addition to enabling them to stay in touch for help seeing [40] and extend social support to others in their virtual network [41]. Interpersonal networks, including friends, family, and neighbors, are important personal risk information sources, as communication between individuals in these networks could help motivate those at risk to take preventive actions [42,43].

As introduced above, the Health Belief Model (HBM) asserts that perceptions of susceptibility and severity associated with a risk—as well as perceived benefits and barriers associated with the recommended risk prevention action in conjunction with self-efficacy—can facilitate the adoption of the promoted protection action [44]. Past research has shown that the measurement scheme operationalized based on the theoretical components of the HBM was validated as general disaster preparedness belief scale [45]. As flood risk related research applying the HBM in the U.S. context remains scarce, the few relevant empirical studies have yielded mixed results due to differences in the socio-cultural backgrounds and demographic characteristics of the samples and the research methodologies employed in these studies.

For instance, an Irish study surveyed 305 participants to examine flood water-caused waterborne infection by comparing the HBM with the Risks–Attitudes–Norms–Abilities–Self-regulation (RANAS) framework [46]; the logistical regression (comparing those with or without direct flood experience) found that only perceived susceptibility and vulnerability were significant predictors of protective health behaviors from both models, aside from the normative and self-regulation factors from the RANAS model. Another flood risk preparedness survey study conducted with 81 participants in Indonesia found that the model components of the HBM, analyzed only as dichotomous variables, yielded no significant chi-square results [47]. It is possible that if these studies had adopted a more robust level of measurement and/or an alternative analytical approach, they might have yielded different results.

By contrast, exposure to flood risk information was found to be related to more HBM-based constructs. For instance, a Malaysian study with 284 survey participants tested an HBM-based educational module addressing flood risk preparedness and reported increased flood disaster knowledge, skills, preparedness, perceived benefit score, perceived barrier score, and cues to action in the intervention group compared to the control group [48]. Another survey study conducted with 285 participants in Turkey established the HBM as a valid conceptual basis for developing a global disaster preparedness belief (GDPB) scale, encompassing self-efficacy, cues to action, perceived susceptibility, perceived barriers, perceived benefits, and perceived severity [45].

Alternatively, according to a systematic literature review (totaling 33 studies), the HBM was found to be more successful in predicting preparedness for influenza than an environmental health challenge of preparedness for heat wave hazards; while attributing such discrepancies in empirical findings to be as a result of population type, sample size, study settings, and method of analysis, the authors recommended the use of a path analysis or structural equation modeling for testing the validity of HBM constructs [49]. As noted in the literature review above, the HBM has rarely been utilized for studying disaster preparedness research in the risk communication context because the model has primarily been tested as a descriptive instead of an explanatory framework [9].

Based on the theoretical discussion and empirical evidence reviewed above, it is logical to assume that flood information exposure will be related to greater perceived risk severity and stronger perceived mitigation barriers.

H1a,b: 

Flood information exposure will be positively related to (a) perceived risk severity and (b) perceived risk mitigation barriers.

As public interaction with different flood information dissemination channels can serve to amplify flood risk communication [50,51], it can also further activate additional flood risk information-seeking behavior [24]. Yet scarce research has addressed the impact of flood information exposure on flood risk information seeking in the context of flood risk management communication. Of the less than a handful of relevant empirical studies, Bird et al. [52] found that Facebook users in two Australian communities engaged in further flood information seeking about the condition in communities in which they and their friends and family resided. Rainear and Lin [53] suggested that public trust in flood information sources was positively related to flood mitigation information seeking. Moreover, Lin’s work [54], for instance, reported that both online news sources and mobile weather apps had a direct effect and offline news sources had an indirect effect on increased flood risk information seeking among residents of a U.S. coastal community.

Once the public has been exposed to the flood information that can impact their community, additional flood risk information seeking is essential to facilitate flood risk management communication at the community level. Community-engaged flood risk communication holds the key for those who face a flood disaster to learn and share useful adaptive preparedness and risk management actions with others in the community. As demonstrated in the relevant literature, individuals engaging and working with community action groups [42] and their interpersonal networks [33] can help drive community-level flood risk management communication processes.

To validate the relationship between flood information exposure and further information seeking, a hypothesis is posited below.

H1c,d: 

Flood information exposure will be positively related to (c) flood risk information seeking and (d) community-engaged flood risk communication.

2.2. Threat Appraisal, Risk Communication, and Efficacy Beliefs

Extant research that directly measured the relationship between flood risk perception and information-seeking behavior among communities that experience frequent flood threats and hazards is scarce, as relevant past studies tended to study the relationships between risk perception and information-seeking needs [54,55,56,57] instead of information-seeking behavior. Of the few directly relevant studies, the empirical findings were both positive and consistent. For instance, a U.S. survey study conducted with 716 participants in the flood-prone south Louisiana reported a significant relationship—between risk perception attitude (measured by perceived flood risk vulnerability and severity) and information-seeking efficacy (assessed by self-efficacy for securing information)—in the context of before, and after a hypothetical future flood event [58].

A Dutch study which tested how low vs. high flood risk level in an experiment suggested that compared to low risk perception, high flood risk perception was a significant predictor of flood risk information-seeking behavior among residents from low-lying communities [59]. Another study conducted in Belgian found that as permanent residency in coastal regions was positively associated with flood risk perception, knowledge, and information needs, flood risk perception was significantly related to flood risk information-seeking behavior [23]. In a related experimental study examining climate change adaptation strategies, a strong belief about human-induced causes of climate change was found to be related to information-seeking intention or behavior instead of perceived climate change consequences among both Dutch and British study participants [60].

Based on the relevant empirical findings explicated above, a hypothesis is proposed to verify the relationship between flood risk severity perception and information-seeking behavior.

H2: 

Perceived flood risk severity will be positively related to flood risk information seeking.

A key objective when communicating flood information, aside from motivating the public to engage in risk information seeking, is to motivate them to engage in flood risk communication in a public venue—such as a local community workshop, town hall, or public meeting led by relevant government agencies, grassroots, and/or citizen action groups—to better understand local government policies and action plans for flood risk preparedness, mitigation, adaptation, and recovery. This observation is similar to the narrative provided by Henstra and McIlroy-Young [61], which states “Public communication is intuitively understood as informing the public about flood risk and management options. Flood risk communication is an important complement to other policy tools for flood risk management that aim to assist communities in preventing, preparing for, and recovering from flood impacts”.

In essence, when the public furthers the flood risk information at hand by seeking to learn about their own flood risk and flood risk management solutions, it can lead them to engage in flood risk management communication at the community level. This is because the adaptation and mitigation solutions associated with flood risk management will require a community-level response supported by the local government policy and implementation strategy [14,62]. Even so, empirical research has yet to explore the relationship between flood risk information seeking and flood risk management communication.

Prior research has suggested community-based flood risk management as an effective approach to both educate and motivate the public to participate in risk mitigation policy planning and implementation [14,62]. Based on Bandura’s theory of efficacy, an individual’s self-efficacy belief can interact with their group efficacy belief through judging their own and their affiliated group’s ability to contribute to accomplishing the common goals [63,64]. Past studies have evidenced that when residents participate in community-based risk communication to engage in flood risk management planning, they will also develop flood risk management efficacy—owing to the strengthening of both their own self-efficacy belief and the community group’s collective efficacy—to work toward the flood risk management planning objectives [54,64].

Even though flood risk information seeking has been validated as an important factor for steering the public to take preventive and adaptive actions to protect themselves [60,65], very little empirical research has investigated the relationship between flood risk information seeking and efficacy beliefs in flood risk management. In particular, a study that surveyed coastal residents in a U.S. coastal community reported a positive relationship between flood risk information seeking (to learn about flood risk management) and collective efficacy beliefs about community-engaged flood risk communication [54]. A second study also supported this relationship, through a survey of a college student sample selected from a large U.S. university located in a state experiencing frequent flood events [53].

To test the role of flood risk information-seeking behavior in relation to community-engaged flood risk management communication and social (individual/group) efficacy beliefs, the following hypotheses are postulated:

H3a,b: 

Flood risk information seeking will be positively related to (a) community-engaged flood risk communication and (b) social efficacy beliefs.

When individuals from a community come together to engage in flood risk management communication, it is common to involve an evaluation of the government’s policies and action plans for countering the flood threat on the community. The potential outcomes for this type of community-based flood risk communication can result in increased cohesion and efficacy on flood risk management, as demonstrated by empirical findings that supported the direct effect from flood risk information seeking to learn about risk management and individual/collective flood preparedness efficacy [54,60,66]. In other words, flood risk management communication can lead to perceived flood risk management efficacy that is achieved by individuals who participate in community-based flood communication activities.

Another likely outcome of community-engaged flood risk management communication may reflect the confidence of community residents in their government policies and strategies for reducing community exposure to and the negative consequences of the flood risk. These policies and strategies, for instance, can include increasing infrastructure resilience such as constructing flood walls or levees, as well as developing living shoreline solutions such as preserving tidal wetlands and salt marshes. As flood risk information seeking to learn about flood risk management was shown to be a significant predictor of flood risk preparedness response-efficacy beliefs [53,54], it is anticipated that flood risk management communication will have the same direct effect on individual and community’s mitigation response efficacy beliefs.

To validate the conceptual links between community-engaged flood risk management communication, social efficacy beliefs, and mitigation response-efficacy beliefs in the current study context, the following hypotheses are proposed:

H4a,b: 

Community-engaged flood risk management communication will be positively related to (a) perceived social efficacy and (b) perceived mitigation response efficacy.

2.3. Mitigation Barriers, Efficacy Beliefs, and Participatory Planning

As discussed above, in communities where flood risk is frequently present, the key factors that can facilitate public awareness of that risk is through exposure to flood risk information sourced from government entities and disseminated by all different communication channels offline and online such as news media outlets, social media platforms, and weak-tie and strong-tie personal networks [24]. In assessing the relationship between flood information exposure and threat appraisal via the HBM, it was also observed that risk information awareness could be related to greater risk severity and mitigation barriers [45,48].

Extending from these observations, it is logical to assume that the greater the belief associated with mitigation barriers, the lower the perceptions of mitigation strategy effectiveness. Limited relevant studies have only addressed how private risk adaptation barriers—such as flood experience frequency and perceived powerlessness—were related to threat appraisal [67,68] or referenced the need for removing private adaptation barriers [69,70]. Hence, empirical research has yet to test the connection between perceived flood mitigation barriers and perceived confidence in mitigation solutions. To verify the assumption of the inverse relationship between these two variables, the following hypothesis is proposed:

H5: 

Perceived flood risk mitigation barriers will be negatively related to perceived mitigation response efficacy.

When considering community-engaged flood mitigation communication, it would be necessary to understand the public perceptions of the formidable undertaking and capital investment associated with executing flood risk mitigation plans. It is through this community-engaged practice, public communication engagement in turn plays a central role in allowing individuals to have their voices heard and to collaboratively engage in forging community-based problem solutions [61]. This scenario is thus consistent with the notion of how public buy-in is an important aspect for achieving successful community-based flood risk management policymaking and policy implementation, as both sets of tasks will affect a community’s economic and social well-being [71].

Widely adopted flood risk mitigation plans are constructed using science-based structural and natural solutions such as seawalls, bulkheads, storm water networks, wetlands, and other natural buffers. Community-engaged risk communication can enable the public to develop their individual and group ability to learn to understand and evaluate these mitigation plans; this type of communal learning experience can then foster individual self-efficacy alongside group efficacy in flood risk management. Empirical research has shown that the more the public believes in their social efficacy, the more they will participate in providing their input to help shape these policies/solutions to best meet the community needs [72]. Past studies have also suggested that communicating flood risk mitigation strategies to facilitate coping appraisal was effective in enhancing individuals’ motivation to adopt protective actions—without elevating the risk itself to heighten threat appraisal [15,68]—and increasing individual commitment to engage in community action [70].

To examine how flood mitigation barriers and mitigation response efficacy may be associated with each other, as well as how social efficacy and mitigation response efficacy may be linked to participatory planning in the current study context, the research hypotheses below will be tested.

H6a,b: 

Perceived (a) social efficacy and (b) mitigation response efficacy will be positively related to participatory flood risk management planning.

3. Proposed Conceptual Model

A conceptual model is shown in Figure 1 below to elucidate the inter-relationships between all the theoretical constructs proposed in the research hypotheses. Specifically, flood information exposure is hypothesized to be linked to perceived risk severity (H1a), perceived mitigation barriers (H1b), risk information seeking (H1c), and community risk communication (H1d). While perceived flood risk severity is connected to risk information seeking (H2), the latter is related to both community risk communication (H3a) and perceived social efficacy (H3b). As community risk communication is associated with perceived social efficacy (H4a) and perceived mitigation response efficacy (H4b), the latter will be negatively predicted by perceived mitigation barriers (H5). Finally, perceived social efficacy (H6a) and perceived mitigation efficacy (H6b) are presumed to be significant predictors of participatory flood risk management planning.

4. Materials and Methods

Data collection was conducted in a coastal community located in northeastern U.S. This community often experiences flooding whenever there is a severe weather event. A cross-sectional online survey was administered with those who reside in areas that are under direct threat of coastal storms. These residents received a study invitation from several community-based organizations that were the partners of a government-led community-engaged partnership program on flood management adaptation and mitigation planning to protect both residential and business properties and critical infrastructure in the community.

4.1. Procedure

Survey respondents were asked to answer questions using their personal knowledge of and experience with past hurricane and severe storm events that created damage to residential homes and infrastructure in their community. The study recruitment and execution took place with prior approval from the Institutional Review Board that evaluated the research protocol to protect the rights of research participants. In terms of measurement, aside from the demographic questions, which indicated participant age, gender, race/ethnicity, education, income, and more, all theoretical constructs were assessed on a 5-point Likert type scale to evaluate the magnitude of cognitive or affective response. As no empirical study has assessed flood risk cognitions utilizing the framework of the HBM in conjunction with evaluating community-based risk communication and flood risk management planning, the measurement items of this study were developed based on those validated by past studies, including Inal et al. [45] and Lin [54]. The definition for each variable is described below. Unless otherwise noted, all constructs were measured on a 5-point Likert type scale.

4.2. Data Analysis

A confirmatory factor analysis (CFA) was conducted using the IBM statistical software package AMOS 29. The CFA procedure evaluated how well the latent variables (the constructs) were measured by their observed indicators (measurement items). After the CFA procedure had established measurement validity for the constructs, a principal component factor analysis (with Varimax rotation) was adopted to ascertain each theoretical construct’s conceptual dimension(s). This procedure then established that the factor structures for all theoretical constructs were deemed statistically sound, as Kaiser–Meyer–Olkin values for all factor analysis results ranged from 0.64 to 0.89 and the affiliated Bartlett’s test of sphericity were statistically significant (p < 0.001). In addition, both the structural equation modeling procedure and Bartlett’s test of sphericity also confirmed a lack of multicollinearity concern.

Based on the factor structure generated by the factor analysis, conceptually similar measurement items were grouped into the same factor dimension. Following that, the measurement items grouped into the same factor dimension were subject to an internal consistency reliability test using Cronbach’s alpha procedure. As all Cronbach’s alpha values were greater than 0.70 and met the required reliability standard, the measurement items that were grouped into the same factor dimension were then merged to create a multi-indicator variable to represent a valid and reliable construct. The Cronbach’s alpha (α) value for each constructed variable was reported below alongside its affiliated mean and standard deviation.

Data analysis began with computing the descriptive statistics for all variables, including demographic characteristics and constructs tested in the conceptual model. The inter-relationships between all variables were calculated with zero-order correlations. A path modeling procedure was used to test the research hypotheses.

4.3. Definitions

Flood information exposure describes how often an individual receives flood information from traditional media, online media, mobile apps, and interpersonal communication sources. These sources include (1) newspapers; (2) radio stations; (3) TV stations; (4) weather channels on cable TV; (5) cable news networks; (6) web portals or search engines; (7) local news outlets on the Internet; (8) major national news outlets on the Internet; (9) Facebook, (10) YouTube, (11) Twitter/X, (12) Instagram, (13) Reddit, (14) Snapchat, (15) Tumblr or WhatsApp; (16) weather-tracking apps, (17) emergency alert apps, (18) emergency alert emails; (19) family members, (20) relatives, (21) friends, and (22) neighbors (α = 0.90; mean =3.43; standard deviation = 0.56). A decision was made to combine all the information sources reported here, since the flood information received from this comprehensive list of information outlets often overlaps with each other.

Flood risk information seeking indicates how often an individual seeks information about the risk associated with (1) their own property; (2) their neighborhood; and (3) their community, as well as (4) what to do to help manage a possible flood (α = 0.73; mean = 3.43; standard deviation = 0.71).

Community-engaged risk communication reflects an individual’s communication activities that address flood risk management at the community level by engaging others through (1) getting involved with a local community group, (2) communicating with the city hall, and (3) participating in public meetings (α = 0.71; mean = 3.41; standard deviation = 0.88).

Perceived flood risk severity describes the perception of how severely an individual is exposed to the following types of risk to their own property and the properties in their neighborhood: (1) flood; (2) flood-induced erosion; and (3) storm surge (α = 0.71; mean = 3.30; standard deviation = 0.51).

Perceived flood risk mitigation barriers are reflected by the degree to which individuals believe in the challenges for implementing flood risk management strategies. These barriers include the following: (1) individuals do not believe a serious flood will occur in the foreseeable future; (2) the flood risk information is unreliable; (3) flood risk management may not be enough to reduce the risk; (4) managing flood risk is challenging to achieve; (5) those who live the flood plan do not care enough about their flood risk; (6) residents do not want taxes to go up to pay for flood risk management; and (7) residents do not want any flood management construction to block the view from their residential property (α = 0.81; mean = 3.39; standard deviation = 0.73).

Perceived social efficacy measures an individual’s perceived self-efficacy and group efficacy in addressing flood risk management at the community level via 12 different items. These items describe the following considerations: (1) come up with a strategy about what to do; (2) construct an actual plan of action; (3) have enough financial resources to implement the action plan (e.g., a flood risk reduction project); (4) have enough support from others (e.g., family and friends) to help implement the action plan; (5) decide how to best handle the financial problem (e.g., taking out a loan, as needed); (6) carry out the action plan regardless of the difficulty; (7) work with others to develop community-based solutions; (8) participate in local activities or events that educate residents about flood risks; (9) contribute money to local events that promote flood risk management; (10) contribute materials/supplies to local events that promote flood risk management; (11) participate in public meetings to offer feedback on flood risk management plans; and (12) attend public workshops to participate in flood risk management planning (α = 0.78; mean = 3.63; standard deviation = 0.48).

Perceived mitigation response efficacy assesses an individual’s perceived confidence in the effectiveness of structural and natural solutions for flood risk management in their community. The measurement items were adopted from Lin’s study [70]. These solutions include the following mitigation strategies: (1) raising levees; (2) deepening river channels; (3) improving the storm water network; (4) building structures such as ripraps, revetments, or bulkheads; (5) moving houses away from high-risk areas; (6) increasing natural buffers like tidal wetlands, salt marshes, and ponds; (7) laying natural barriers such as dunes or intertidal flats; (8) constructing flood walls; (9) modifying buildings (e.g., raising floor levels and utility structures and filling in basements); and (10) restricting new buildings or renovations in areas with high flood risk (α = 0.72; mean = 3.89; standard deviation = 0.45).

Participatory flood risk management planning identifies an individual’s level of interest in participating in the community-engaged flood risk management planning project via (1) visiting the flood risk management project planning website; (2) offering input about flood risk management by contacting the project planning unit; (3) attending the forthcoming public meeting held by the project planning unit; (4) attending the upcoming community workshops held by the project planning unit; and (5) participating in a neighborhood group to address flood risk concerns in one’s neighborhood (α = 0.80; mean = 3.46; standard deviation = 0.70).

5. Results

The self-reported demographic profile of the sample indicated that 97.3% of the survey participants had direct experience with and were affected by flooding and its related hazards. A total of 70.1% of the survey participants were male and 29.9% were female; their ages ranged between 18 and 74 years, with a median age between 35 and 39. The racial and ethnic characteristics were distributed as follows: 76% White (non-Hispanic), 12% Black/African American, 6% Hispanic/Latino, 4.3% Asian, 1.3% Native American, and 0.3% mixed-race. A total of 81.7% of the sample were homeowners and 84.3% had a college degree; the median household income was around USD 80,000. In addition, 85.4% and 10.3% of the sample held full-time and part-time employment, respectively.

Given the fact that flooding can cause a severe financial impact on homeowners, it was not surprising that the sample was over-represented by homeowners (instead of renters). Homeownership tends to be positively correlated with education, income, and employment status, which explained the sample’s high level of college graduates, full-time employment, and household income. It should be noted that the sample’s median household income was lower than the median income of the state in which it is located, as income in this northeastern state is higher than that in most other states in the U.S. As such, the results reported below reflected mostly the responses and experiences of homeowners (81.7%) and just under 20% of non-homeowners.

Table 1. Descriptive statistics and zero-order correlations for theoretical constructs.

	1	2	3	4	5	6	7	8
1. Flood Information Exposure	--
2. Perceived Flood Risk Severity	0.18 *
3. Perceived Barriers for Flood Mitigation	0.45 **	0.36 **
4. Flood Risk Information Seeking	0.54 **	0.38 **	0.34 **
5. Community-Engaged Risk Communication	0.56 **	0.29 **	0.31 **	0.70 **
6. Perceived Social Efficacy	0.42 **	−0.09	−0.03	0.41 **	0.47 **
7. Perceived Flood RiskMitigation Efficacy	0.47 **	−0.04	0.04	0.37 **	0.44 **	0.54 **
8. Participatory Flood Risk Management	0.40 **	0.07	−0.09	0.41 **	0.39 **	0.61 **	0.47 **	--
M	3.43	3.30	3.39	3.43	3.41	3.63	3.89	3.46
SD	0.56	0.51	0.73	0.71	0.88	0.48	0.45	0.70

Note: ** p < 0.001; * p < 0.05.

reports the descriptive statistics and correlations for all scaled variables. While the mean values of these variables ranged from 3.30 (SD = 0.51) to 3.89 (SD = 0.45), significant correlations existed between all scaled variables ranging between 0.35 and 0.70.

As described above, a confirmatory factor analysis (CFA) was performed to examine the validity of the latent variables (constructs) in the measurement model. The CFA results showed an acceptable model fit (χ² = 2839.25; p < 0.001; CMIN/DF = 1.13; CFI = 0.95; TLI = 0.95; IFI = 0.95; RMSEA = 0.021), which ascertained the measurement model’s validity. Based on the CFA results, a path analysis was conducted to test the research hypotheses and verify the hypothesized relationships between those latent variables illustrated in the proposed conceptual model below (see

Table 2. Path analysis results for testing the research hypotheses.

Hypothesis	Independent Variable	Dependent Variable	β	C.R.	p
H1a	Flood Information Exposure	Flood risk Severity Perception	0.19	2.98	0.03
H1b	Flood Information Exposure	Mitigation Barrier Perception	0.48	8.26	<0.001
H1c	Flood Information Exposure	Risk Information Seeking	0.47	8.46	<0.001
H1d	Flood Information Exposure	Community Risk Communication	0.30	5.42	<0.001
H2	Flood Risk Severity Perception	Flood Risk Information Seeking	0.28	5.37	<0.001
H3a	Flood Risk Information Seeking	Community Risk Communication	0.51	9.72	<0.001
H3b	Flood Risk Information Seeking	Social Efficacy Perception	0.48	3.25	0.001
H4a	Community Risk Communication	Social Efficacy Perception	0.26	3.44	<0.001
H4b	Community Risk Communication	Perceived Mitigation Response Efficacy	0.72	6.79	<0.001
H5	Mitigation Barrier Perception	Perceived Mitigation Response Efficacy	−0.16	−2.22	0.026
H6a	Social Efficacy Perception	Participatory Flood Rsk Management	0.54	9.57	<0.001
H6b	Perceived Mitigation Response Efficacy	Participatory Flood Risk Management	0.14	2.05	0.041

). Path modeling results indicated an acceptable model fit (χ² = 22.13; p < 0.005; CMIN/DF = 2.77; CFI = 0.98; TLI = 0.91; IFI = 0.98; RMSEA = 0.077).

Path analysis results for H1a-H1d, exploring the relationship between flood information exposure and a set of four variables (as specified in the conceptual model) showed that flood information exposure was positively related to flood risk severity perception (β = 0.19, p < 0.003), mitigation barrier perception (β = 0.48, p < 0.001), risk information seeking (β = 0.47, p < 0.001), and community communication engagement (β = 0.30, p < 0.001). These results supported H1a–H1d, respectively. Turning to H2, perceived flood risk severity was found to have a positive association with risk information seeking (β = 0.28, p < 0.001), thus validating H2.

H3a and H3b, testing the relationship between risk information seeking and the following two variables, found that it was positively related to both community-engaged risk communication (β = 0.51, p < 0.001) and perceived social efficacy (β = 0.48, p < 0.001). These results sequentially verified H3a and H3b. As expected, the results of analyzing H4a and H4b showed that relationship between community communication engagement and its two dependent variables—perceived social efficacy (β = 0.26, p < 0.001) and perceived mitigation response efficacy (β = 0.72, p < 0.001)—were positive and significant. Thus, both H4a and H4b were confirmed based on these results.

As for H5, perceived flood risk mitigation barriers were found to be a negative predictor of perceived flood risk mitigation response efficacy (β= −0.16, p < 0.026), as anticipated. This result then affirmed H5. Lastly, path analysis results from testing H6a and H6b indicated that perceived social efficacy for flood risk management (β = 0.54, p < 0.001) and perceived flood risk mitigation response efficacy (β = 0.13, p = 0.041) were positively related to interest in participatory community-based flood risk management planning, as speculated. These results substantiated H6a and H6b.

Figure 2 illustrates the inter-relationships between all the theoretical constructs proposed in the conceptual model and their affiliated research hypothesis testing results.

6. Discussion and Implications

This study contributes to the literature by demonstrating how exposure to flood risk information through accessing risk communication channels, both offline and online, may facilitate public engagement—which leads to flood risk communication at the community level—and the interest of community members in partaking in participatory flood risk management planning activities. This study adopts a theory-driven conceptual model to explain risk information seeking and communication behavior in the context of flood risk management that remains under studied. Given that this study was conducted with residents in the area of a coastal community that faces chronic threats from flooding, erosion, storm surge, and sea level rise in the midst of an on-going government-led community-engaged flood risk management planning program, the study results can help inform both scholars and practitioners alike.

Study results establish that flood information exposure could drive greater perceived flood risk severity and mitigation barrier through cognitive threat appraisal, in addition to furthering flood risk information-seeking behavior and affiliated community-engaged flood risk communication activity. The community-engaged communication activity, in turn, was positively linked to perceived social efficacy, which was characterized by the combined individual and collective efficacy beliefs in tackling flood risk management. Similarly, the community-engaged communication activity was also linked to the public’s confidence or perceived response efficacy in flood risk mitigation strategies, which contained both structural and natural solutions. Both perceived social efficacy and response efficacy in flood risk management then positively predicted community residents’ interest in partaking in the on-going participatory flood risk management planning program.

Importantly, the positive relationship between flood information exposure—through a variety of different interpersonal networks, social media platforms, offline/online traditional media outlets, and mobile apps—and additional risk information seeking signifies the necessity of utilizing every type of communication channel to spread news and information on flood risk. Disseminating flood risk information as widely as possible maximizes the opportunity to reach a cross-section of the public—including the hard-to-reach population—regardless of what their capacity for risk communication may be. These findings further confirm past work which found traditional news media outlets online/offline [24,35,36,37,54], social media platforms [38,39,73], mobile apps [54,70], and interpersonal networks [42,43] could all be useful risk communication channels for informing the public about flood risk and to motivate them to seek further information for risk management [52,53,54].

The results suggest a positive link between flood information exposure and perceived risk severity, conceptualized as an HBM construct. The conceptual link revealed here has been widely validated in the literature, often through the lens of protection motivation theory [7,13,15]. Flood information exposure was also positively associated with another HBM construct operationalized for the current study context, perceived flood risk mitigation barriers. This finding is consistent with past research which indicated the need for flood information dissemination to allow the public to educate the public and to allow them to consider their risk adaptation beliefs [15,66], which may or may not yield a resilience mindset. Nonetheless, it is a necessary cognitive process for the public to internalize the flood risk. As coping with flood risk mitigation is a daunting challenge for individual residents and property owners in a community, ideally, exposure to flood risk information will encourage them to seek community-based solutions to help them develop social resilience. It is worth noting that the current study is among the first to test the framework of the HBM via the constructs of perceived risk severity and mitigation barriers and their connection to flood information exposure.

Interestingly, flood information exposure was also a positive predictor of additional flood risk information seeking and community-engaged risk communication about flood risk management. These findings are similar to relevant work that reported how flood information exposure through information dissemination channels can lead to additional flood risk information seeking [52,53,54] and amplified risk communication engagement [50,51]. Community-engaged flood risk communication entails individuals taking the initiative to work with their neighbors and community through a public venue—by joining a grassroots group, townhall, and public meeting—to address flood risk management. These first-time findings thus further evidence how risk communication is a continuous process that should not end with being aware of what a flood risk is and should ideally lead to community-engaged communication about flood risk management.

The current study also validated that flood risk information seeking was positively related to an individual’s perceived social efficacy, as demonstrated by the perceived confidence of their own and their social group when engaging in community-based flood risk management. Perceived social efficacy, as conceptualized in the current study context, is a proxy measure for social resilience, as it reflects a self-affirmation process [74,75] that helps individuals to examine their individual and collective confidence in addressing the challenges of flood risk management. Again, the findings reported here are among the first to explicate the relationship between information seeking and perceived social efficacy (or resilience), as fewer than a handful of relevant studies were available to explain this phenomenon [53,54].

By implication, after residents of a flood-prone community have ascertained their flood risk and the flood risk mitigation solutions, they may turn to working with the community in different public forums to collaboratively consider flood risk management. In doing so, they will evaluate the perceived effectiveness of flood mitigation strategies that include both structural (e.g., raised levees and flood wall construction) and natural (e.g., tidal wetland and salt marsh preservation) solutions. Based on the path modeling results, the more active the community-engaged communication engagement, the stronger the community confidence in the potential effectiveness of these mitigation solutions. In other words, when the public comes together in a group and in a public setting to learn and communicate about flood risk management over time, they may also become more confident in the flood mitigation strategies proposed by the relevant government team (including planners, architects, and engineers). As no existing research has investigated this conceptual connection, this original preliminary finding here then establishes the role of community-engaged flood risk communication in facilitating public discussion about flood risk mitigation solutions.

Turning to perceived flood mitigation barriers, the findings suggested that the greater the perceptions of these barriers, the lower the perceived effectiveness of flood risk management strategies. As flood mitigation barriers were reflective of the doubts toward the need or success for flood risk reduction, potential tax burdens, and personal inconveniences, these barriers then contribute to the lack of confidence in community-based mitigation solutions. The negative association between the perceptions of mitigation barriers and mitigation solution effectiveness seemed to be emblematic of the positive but weak connection between perceived confidence in mitigation solution effectiveness and public interest in participatory flood risk manage planning.

The extant literature has yet to test the conceptual connection between perceived mitigation barriers in flood risk management—a construct derived from the theoretical framework of the HBM—and public confidence in mitigation solutions. The study finding reported here thus implies that when the public is aware of how flood risk mitigation solutions cannot deliver guaranteed success, a certain degree of doubt about the effectiveness of mitigation solutions will exist among some who reside in an area that experiences frequent flood threats.

Another factor that may signify the public’s confidence in flood mitigation solutions could be the degree of interest they demonstrate in supporting the flood risk management planning activities that are typically led and organized by federal, state, and/or local agencies. The current study tested this relationship and found that perceived confidence in mitigation solutions was a significant but weak predictor of interest in participatory risk management planning. The weak connection between these two variables may stem from the fact that the time and effort needed to contribute to a joint community–government risk management planning group may not be feasible to most people in the community.

Contrastingly, perceived social efficacy in flood risk management—partly attributed to the public’s risk information-seeking behavior and community-based risk communication activities—was a strong predictor in explaining the public’s interest in participatory flood risk management planning. This finding confirms the result from prior research that revealed the same positive relationship between collective group efficacy and collective community action [54]. By implication, the confidence and resilience beliefs in individuals and their community group toward communication activities associated with flood risk management can help pave the way for supporting government-led community-engaged flood risk management planning programs. The evidence generated from the current study thus affirms the value of participatory community-engaged planning action, similar to the community co-design concept, as a potentially effective model for gaining positive public policy outcomes [70,72].

Taken together, by drawing from the discussion of the research findings above, the following practical implications can be advanced: Firstly, as flood information exposure can help increase flood risk and mitigation challenge perceptions, it is essential that the flood information transmitted by the relevant national (e.g., National Weather Service), state (e.g., public health department), and municipal (e.g., emergency management department) government units also emphasize practical risk management tips and adaptation solutions. These institutions should also utilize every available offline and online communication channel and network at their disposal to transmit the risk and solution messaging to reach the target population as widely as possible.

Secondly, since flood information exposure can potentially trigger additional risk information seeking and bottom-up grassroots community-driven efforts to initiate public engagement in flood risk management communication, this suggests that practitioners (e.g., emergency management personnel) should also develop a community-engaged risk communication strategy. An effective risk communication strategy that aims to emulate a respectful and reciprocal partnership to engage community members will be valuable in boosting their social efficacy and response efficacy to collectively address flood risk management challenges.

Thirdly, as social efficacy and response efficacy can facilitate public participation in government-led flood risk management planning programs, it is important for scientists, engineers, policy experts, resource economists, planning agencies, and other practitioners to conduct risk communication with community members by potentially considering a “co-design” approach. Using this “co-design” modality, a participatory design process involving both community members and government planning agencies can be supported by expert-guided—as well as science- and evidence-based—practices to develop a more comprehensive flood risk management policy and an economically viable implementation strategy.

7. Conclusions

The current study validated the role of risk communication in facilitating the appraisal of flood risk severity and mitigation barriers—as well as initiating flood risk information seeking and communication to engage others for learning, evaluating, and shaping risk management solutions—leading to participatory risk management planning at the community level. One of the new findings reported by this study revealed a positive association between flood information exposure and perceived mitigation barriers. Past studies have observed that hurricane risk information dissemination—including misinformation such as rumor [74]—rarely motivates the public to take protective action, even though government agencies such as FEMA have widely promoted risk communication strategies [70,73]. To avoid triggering negative beliefs toward risk mitigation, it would be important for those who disseminate flood information to engage in risk communication that provides well-reasoned, science-based solutions to boost their confidence in supporting flood risk management plans instead of writing them off.

Other first-time findings generated by the current study illustrate how risk information seeking can motivate the public to engage other community members for flood risk management communication and develop greater perceived social efficacy and perceived confidence in mitigation solutions. These results have strong theoretical and practical implications, as both information-seeking and risk communication behaviors can influence social resilience and confidence in mitigation solutions that are necessary for promoting a community-based flood risk management program [61].

Several study limitations should be noted here. First, this study was a retrospective study that asked coastal residents to recall their own hurricane experience as the basis for responding to the survey questions. To establish the empirical effects of risk communication, an experimental study can be conducted to compare residents who are instructed to engage in flood risk management communication vs. those who are not assigned to the same risk communication exercise. Second, the current study presented data that captured a snapshot in time. This suggests that a longitudinal survey study will be able to further validate both the validity and reliability of the study measures and the proposed conceptual model. Lastly, the current study was conducted in a northeastern coastal community where the majority of its residents were college-educated and likely more civic-minded. To better understand whether a participatory flood risk management model is feasible in other communities facing the same flood risk, it would be necessary to replicate the same survey elsewhere.

In conclusion, the current study has evidenced the essential role of information and communication behaviors that can help awaken risk perceptions and mitigation barriers, as well as cultivate community resilience, develop mitigation solutions, and participate in flood risk management planning. Future research will benefit from verifying the applicability of the proposed theory-based conceptual framework to further our understanding of how risk information and communication factors can help enhance the success of community-engaged flood risk management.