Mapping Cross-Sector Communication Networks for Natural Hazards: A Qualitative Exploration of Formal and Informal Flood Warning Communication Pathways for Enhancing Community Resilience

Resilience is the ability of people, households, communities, and systems to mitigate, adapt to, and recover from shocks and stresses in a manner that reduces chronic vulnerability and facilitates inclusive growth. The ways in which communities have historically responded to stressors like extreme weather events may not be sufficient for the increasing frequency and intensity of climate-related shocks. We have more accurate forecasts than at any other point in history, but the benefits of those complex weather forecasts are only realized when the information is timely and successfully integrated into decision-making. For emergency managers, local organizations, and the public to use forecast products effectively and build community resilience, dissemination and presentation must account for the range of technical understanding and the complex information landscapes and systems in which professionals and the public operate. In our study, we aim to capture such complex and dynamic networks community organizations use to receive and share advanced flood information so we can enhance future risk communication.

How forecast data is communicated directly affects its interpretation. For example, certain types of information (e.g., probabilistic) can be difficult to understand for those without training, while visual characteristics (e.g., text format, color contrast) can impact attention and risk perception. Factors like emotions, presence or absence of communication networks, and trust in the information source also affect engagement with environmental warnings and whether someone takes action or seeks additional information. At the individual level, one’s social interactions (e.g., with their neighbors) can impact planning for and responses to emergencies (e.g., home weatherization behavior). As emergency weather events become more prevalent, it’s crucial to understand the factors that impact receipt and interpretation of weather-related risk communication, including for flood forecasts.

One key factor which may impact the receipt, interpretation, and dissemination of weather-related risk communication are the communication pathways in the information network of a local community. Qualitative social network analysis has become an increasingly visible methodology in disaster research (e.g., Benedict et al., 2024) as a means of providing rich data about complex systems (Heath et al., 2009; Schepis, 2011). Mapping out community networks through qualitative methods offers a comprehensive picture of the social structures and relationship dynamics (Emmel & Clark, 2009) which may affect how flood information is received, understood, and shared.

Even though the launch of innovative flood inundation map (FIM) services will make street-level flood forecasts publicly available for the first time, there remain questions regarding the perception and application of these services by end-users. To answer some of these questions, we are conducting a multi-year, mixed-methods project to understand flood forecast receipt and use by local organizations, including conducting a qualitative social network analysis mapping to more deeply understand local community networks and how those can be effectively leveraged. Our primary goal for this phase is to understand the current process by which local organizations (i.e., local authorities and decision makers, health and human service organizations, and community interest organizations) receive and act on flood information. We set out to understand what information they receive, how they receive it, their key sources of information, who they share information with, how they share information, and if that information changes in any way between receipt and sharing.

The insights we detail here are informed by an ongoing study funded by NOAA through the Cooperative Institute for Research to Operations in Hydrology. Starting in June 2023, our team of researchers from RTI international and the University of Vermont launched an exploratory study to understand how communications about upcoming flooding filter through communities to reach their intended audiences. Specifically, we are investigating how flood information is received and shared by organizations within six communities. The initial phase of this research included 15 focus groups and 16 interviews, which were followed by a national survey to expand on our understanding of emergency information flow.

Using both qualitative and quantitative data, we will present network maps based on three audience segments:

• Local Authorities/Decision Makers: Leaders within the community who make decisions about spending and implementing policies (e.g., city communications department, department of transportation, public works, utilities, public safety, news media). Information sharing policies and decisions are focused on taking community-level resilience actions.
• Health and Human Service Organizations: Decision-makers and staff who work in the community to provide services for general and/or vulnerable populations (e.g., organizations serving unhoused, universities/school districts, hospitals/clinics, public health departments). Information sharing and policies are aimed at preparing for resilience service needs.
• Community Interest Organizations: People who are embedded in the day-to-day activities of a community and may have an influence on the actions that both residents and decision makers take (e.g., faith-based organizations, cultural groups, large scale employers, watershed groups). Information sharing policies and decisions are centered on enhancing resilience by keeping community members informed, safe, and connected to services.

The network maps reveal interesting differences between community segments as well as between community locations, including more formalized and routine pathways and touchpoints for local authorities and decision makers with information flowing both from and back to the NWS in advance of flood events. This more systematic receipt of information about forthcoming flooding aligns with their role as decision-makers, though there were more inconsistencies across participants in how they share information to others in the community. Community interest organizations generally had less formalized methods for both receiving and sharing information and relied on standard announcements, alerts, and apps. The social network maps identify the complex nature of these interactions, reveal how flood communication flows through them, and highlight barriers in the communication process.

We will share recommendations based on our network maps for how the design and content of flood warning information messages can operationally account for variability in local audience engagement including potential new strategies for NWS partner engagement. In doing so, we aim to provide practical ways for emergency managers and forecasting offices to enhance flood communication, and how this is one potential step to build community resilience to extreme weather events.