Gaps and Opportunities for Collaborative Flood Governance: Network‐Analytic Insights From the US Gulf Coast

Abstract

Flood governance often operates as a polycentric network where authority, resources, and responsibilities are distributed across multiple organizations, yet the structural and cognitive factors that shape collaboration remain underexplored in flood contexts. We examine inter‐organizational ties in the Beaumont–Port Arthur region, Southeast Texas, using an organizational survey and self‐bounded network. We build three directed, unweighted collaboration networks: an All‐Issues network and two phase‐specific networks for preparation/mitigation and response/recovery. We decompose the survey data into nine driver‐ and impact‐specific layers to examine how collaboration varies across flood‐governance concerns. We report whole‐network statistics and estimate Exponential Random Graph Models (ERGMs) to test cognitive homophily, shared risk framing, and perceived competence. All‐Issues collaboration is denser, more reciprocal, and more clustered than phase‐specific networks, with response/recovery the most decentralized. Subnetworks focused on infrastructure and geophysical drivers are better connected than socio‐economic‐driver or social‐impact subnetworks. ERGMs show that collaboration does not favor same‐type partners. Instead, shared risk framing predicts tie formation in phase‐relevant ways: higher precipitation importance increases response ties, higher infrastructure‐driver importance increases preparation ties, and similarity in precipitation assessments increases ties in the full network. Perceived competence alignment also matters: smaller differences in operational‐confidence ratings increase the likelihood of collaboration. Across models, organizations with shared collaborators are more likely to be connected, suggesting that flood‐governance collaboration is organized through locally reinforcing clusters. These findings point to two actionable strategies: foster shared framing around less connected social and economic issues and empower brokerage among less connected actors while maintaining robust technical coordination on infrastructure.