Climate change, coupled with urban development and aging infrastructure, is driving more pluvial flooding in cities (Berndtsson et al., 2019; National Academies of Sciences Engineering and Medicine, 2019; O'donnell and Thorne, 2020). The management of urban stormwater and flooding risks increasingly aims for resilience, the capacity to absorb, recover from, and adapt to extreme storm events and their uncertain impacts (Liao, 2012; Disse et al., 2020; Mcclymont et al., 2020). Such a shift calls for changes in urban landscapes to make space for water, and many government agencies and organizations are developing nature-based solutions (NBS) as a promising approach (Hobbie and Grimm, 2020; Axelsson et al., 2021). NBS include various practices (e.g., retention ponds, detention swales and basins, constructed wetlands) that seek to use natural processes to manage stormwater and mitigate flooding while offering other societal benefits (Lennon et al., 2014; Hobbie and Grimm, 2020; O'donnell et al., 2020).

One increasingly discussed NBS innovation is floodable sites, urban spaces designed to accommodate different dry and wet weather functions (Palazzo, 2019; Ashley et al., 2020; Kuang and Liao, 2020; La Loggia et al., 2020). In dry weather, they are used for everyday activities (e.g., recreation, parking, light traffic); in wet weather, they are temporarily inundated to manage excessive storm runoff and mitigate flooding (Mariano and Marino, 2018; Silva and Costa, 2018; Lund et al., 2019; Rogers et al., 2020). Floodable sites can include diverse types of urban spaces for stormwater and flood risks management. Some scholars have also argued that introducing such practices may help urban residents observe and learn about stormwater, thus encouraging a shift in social-cultural norms to “live with water” (Lennon et al., 2014; Silva and Costa, 2018; Kuang and Liao, 2020; Mcclymont et al., 2020).

However, scant research exists on how community members perceive novel NBS practices like floodable sites. There are anecdotes of residents calling a water plaza designed as a floodable site a “drowning plaza” (Silva and Costa, 2018). In cities, drainage systems have long been adopted to discourage standing water, treating stormwater as a nuisance that can disturb daily activities (Tempels and Hartmann, 2014; Ashley et al., 2020; Kuang and Liao, 2020). The visible ponding and puddling in typically dry floodable sites might feel unpleasant and unsafe to urban residents. This can undermine public acceptance of novel NBS practices despite their environmental benefits (Derkzen et al., 2017; Frantzeskaki, 2019; Han and Kuhlicke, 2019; Anderson and Renaud, 2021; Li and Nassauer, 2021).

To offer new insights for managing urban flooding through NBS, this study uses “risk as feelings” as an organizing framework to investigate community members' risk perceptions. This concept is embedded in a broader theory in cognitive psychology and neuroscience which asserts that humans use both affective and cognitive processing—a “dual process” model—to comprehend the world (Loewenstein et al., 2001; Slovic et al., 2004). Affective processing, which occurs when people perceive something as good or bad, is often labeled experiential, intuitive, and automatic; cognitive processing, which occurs when people use explicit reasoning, is often labeled analytical and deliberative (Epstein, 1994; Kahneman, 2011). Both processes are now recognized as critical to decision-making. When faced with risk or uncertainty, though, people are prone to make judgements driven by how they feel about an outcome rather than by deliberation about its probability and consequences (Slovic, 1987; Slovic et al., 1998; Disse et al., 2020). For example, if people have unfavorable feelings toward using pesticides, they tend to perceive high risk and low benefit of pesticides; if they have favorable feelings, they tend to perceive the opposite (Alhakami and Slovic, 1994). The risk perception of a phenomenon can be strongly affected by past experiences of the same phenomenon or one with similar perceptible characteristics (Loewenstein et al., 2001). Through lived experience and learning, certain sights, sounds, smells, ideas, and words can become associated with positive and negative feelings, forming “affective images” that guide responses in future situations (Slovic et al., 1998).

“Risk as feelings” has informed many studies on flood risk perception—the perceived likelihood and potential damage of flooding (Botzen et al., 2009; Kellens et al., 2013; Wachinger et al., 2013; Birkholz et al., 2014; O'neill et al., 2016; Lechowska, 2018). Flood risk perception often differs from expert technical assessments of flood risks and is consistently reported to relate to past experiences of flooding (Botzen et al., 2009; Kellens et al., 2013; Wachinger et al., 2013; Birkholz et al., 2014; O'neill et al., 2016; Lechowska, 2018). For example, Botzen et al. (2009) found that individuals who had experienced or been evacuated from a historic flood event in the Netherlands reported a higher perceived probability of flooding. Besides flood experience, socio-demographic characteristics (e.g., age, gender, income, education, having young children) may also influence flood risk perception, though their reported effects are not consistent (Kellens et al., 2013; Wachinger et al., 2013; Lechowska, 2018). For example, some studies found that people with higher education perceived lower flood risks (Botzen et al., 2009; Bradford et al., 2012), whereas other studies reported no association betweeen eduction and risk perception (Kellens et al., 2011; O'neill et al., 2016).

While the growing literature on flood risk perception provides important insights for flood risk communication and management, most studies have focused on major riverine and coastal flooding. Urban pluvial flooding, which results from overwhelmed drainage systems and occurs more frequently, calls for more investigation (Netzel et al., 2021). Further, risk perception related to novel NBS that are designed to manage urban pluvial flooding has not yet been examined, to our knowledge.

This study investigates the risk perception of localized flooding and the perceived safety of novel and familiar NBS practices. We examine how people with different past experiences with localized flooding perceive the risk of urban pluvial flooding as well as the perceived safety of novel and familiar NBS under storm conditions. We compare perceptions of floodable sites, a novel NBS practice that temporarily inundates urban landscapes under the storm condition, with perceptions of stormwater retention ponds, a familiar NBS practice that can hold excess water under the storm condition. We also explore whether other contextual (e.g., residence location relative to flood zone, environmental knowledge) and socio-demographic factors are associated with perceived flood risk and perceived safety of NBS practices. Specifically, we address four questions:

This study aims to inform resilient approaches to urban flood risks and stormwater management that can better respond to growing extreme weather events under climate change. We investigated community members' risk perception of urban pluvial flooding and perceived safety of NBS practices, using “risk as feelings” to frame our research questions. To shed light on how perceived safety may vary by different design solutions for storing stormwater, we compared two types of practices: floodable sites, a novel NBS practice that temporarily inundates urban landscapes, and stormwater retention ponds, a familiar NBS practice that always has water. We also examined how flood risk perceptions and perceived safety of NBS practices are associated with experiences of localized flooding, as well as other contextual and socio-demographic factors.

The majority of our study respondents indicated some degree of worry about potential damage to property when noticing standing water near home. This result is somewhat expected. Recent studies have reported that community members are generally aware of urban flooding problems (Derkzen et al., 2017; Meerow et al., 2021). Regarding the question of perceived safety of NBS practices, this study indicates that community members may perceive NBS practices that visibly change stormwater levels in surrounding landscapes as unsafe. Both floodable sites and retention ponds were perceived as less safe under storm conditions than under non-storm conditions. Furthermore, ponds were perceived as safer than floodable sites under storm conditions. Reflecting on these results through the “risk as feelings” framework, we speculate that people might intuitively perceive inundated floodable sites that look like flooding events as less safe, whereas water level fluctuations in retention ponds are more expected because it is a natural-looking practice where water is typically present. While we did not directly measure feelings (e.g., through a psychometric paradigm or physiological change), our study provides a basis for future research to examine affective reactions to novel and familiar NBS practices that have noticeable water level changes. We also call for more research on why people view water level changes as unsafe and what influences such perceptions might have on well-being or support for the adoption of NBS practices.

Prior studies have shown associations between personal experiences of major riverine and coastal flooding and greater perceived risks and discussed how witnessing disastrous events may help people envision low-probability events and their consequences (Botzen et al., 2009; Kellens et al., 2011; O'neill et al., 2016). Importantly, we found that the experience of less intense localized events—seeing standing water and flooding near home—are also associated with greater flood risk perception. As “risk as feelings” implies, intuitive perceptions of risks involve affective processing and do not always reflect the actual magnitude of damage. Further, experience of localized flooding may also undermine perceived safety of floodable sites under storm conditions. However, we found no associations between such experience and perceived safety of retention ponds under storm conditions. This has important implications for stormwater management interventions that are more likely to gain support in communities having experienced localized flooding.

We also found notable effects of other contextual and socio-demographic factors on the risk perception of urban pluvial flooding and the perceived safety of NBS practices. Consistent with research focused on riverine and coastal flooding (Kellens et al., 2013; Lechowska, 2021), we also observed strong associations between gender (female) as well as past property damage and greater perceived risks of localized flooding. Different from previous studies (Botzen et al., 2009; Kellens et al., 2013), we did not find associations between higher education level and lower flood risk perception. However, our results suggest more nuanced effects of knowledge and behavior specifically related to stormwater. In this study, respondents who indicated more knowledge about local water quality or had participated in activities addressing stormwater management issues (i.e., wrote a letter or made a phone call, attended or arranged a public meeting, or talked to a manager face to face) in the past 2 years were more worried about potential stormwater-related damage. In contrast, respondents who had participated in activities promoting general environmental sustainability (i.e., made donations, volunteered or served in a leadership positions for an organization or advocacy group, voted for a candidate for public office) in the past 2 years were less worried about potential stormwater-related damage. Perhaps people who participate in activities to promote general sustainability are not necessarily interested in or well-informed about stormwater management and focus more on other environmental issues (e.g., greenhouse gas reduction, wildlife habitat). Research on flood risk perception should continue to investigate knowledge and behaviors related more specifically to stormwater and water systems to further understand their influences.

Perceived safety of floodable sites was associated with only one socio-demographic factor—gender, whereas perceived safety of retention ponds was associated with several socio-demographic factors. The lower perceived safety of both practices among female respondents might be attributed to social norms for males to not express worry (Sutton and Farrall, 2005). It might also relate to females' primary role in taking care of children and home. Future research may conduct more qualitative analyses of the potential gender differences in perceived safety of NBS practices for stormwater management. For retention ponds, respondents with children under the age of 12 in their households perceived retention ponds under storm conditions as less safe. This reflects concerns about drowning hazards and personal safety, especially for children, that have been observed in previous studies (Bastien et al., 2012; Jarvie et al., 2017; Williams et al., 2019). Moreover, non-white respondents perceived retention ponds as less safe, while older respondents perceived retention ponds as safer. People of color may have less experience with well-maintained retention ponds given their often more limited access to high-quality, large greenspace (Rigolon, 2016). Older people may have more contact with nature and thus are more familiar with the fluctuation between typical and high water levels. These results point to potentially different causes that shape perceived safety of floodable sites and stormwater ponds under storm conditions.

Further, the adjusted R-squared for the perceived safety model of floodable sites (12.0%) is smaller than that for the model of retention ponds (39.0%), indicating a lower explanatory power. Therefore, factors besides the contextual and socio-demographic variables examined in this study, such as landscape design choices and environmental values, may impact the perceived safety of floodable sites, and more research is needed.

We assert that urban flood risk management should engage with social science theories such as “risk as feelings” to account for peoples' affective responses and intuitive perceptions. Drawing on the concept of “risk as feelings,” this paper deepens the understanding of how community members perceive urban pluvial flooding by highlighting the effects of experience of localized flooding and other contextual and socio-demographic factors. It also provides new insights into developing nature-based solutions (NBS) for managing urban stormwater by elucidating how noticeable stormwater level changes in novel and familiar NBS practices can elicit safety concerns. Notably, we found that, under storm conditions, floodable sites, a novel NBS practice that temporarily inundates urban spaces to manage stormwater, were perceived as less safe than retention ponds, a familiar NBS practice that always has water. Further, community members with more experiences of localized flooding perceived floodable sites as less safe, while those who were younger, non-white, or had children in their households perceived retention ponds as less safe. The difference in perceived safety of these two types of NBS practices have implications for public support for their adoption. We call for inter and trans-disciplinary collaborations in designing new landscape interventions to address extreme weather events and the increasing urban flood risks. Pervasive adoption of NBS must consider potential impacts on people's everyday experiences in their neighborhoods and communities, in addition to stormwater management objectives, to gain broad societal support.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Institutional Review Boards, University of Michigan. The patients/participants provided their written informed consent to participate in this study.

JL: conceptualization, methodology, formal analysis, data curation, and writing—original draft preparation. JN: conceptualization, methodology, funding acquisition, supervision, and writing—review and editing. NW: methodology, investigation, funding acquisition, and writing—review and editing. SP: conceptualization, data interpretation, and writing—review and editing. LM: investigation, funding acquisition, and writing—review and editing. All authors contributed to the article and approved the submitted version.