Streets, storm surge, and the frailty of urban transport systems: A grid-based approach for identifying informal street network connections to facilitate mobility

A variety of real-world infrastructure systems, including power grids, telecommunications, roads and social networks can be conceptualized and modeled using basic concepts from graph theory and network science. The abstraction of real-world systems into an edge-node topology enables analysts to quantify and characterize a range of network properties, including structure, connectivity, and modularity, as well as the attributes of individual nodes and edges such as degree, criticality, and betweenness. One major problem with the abstraction process associated with traditional network analysis is that edges and nodes are stripped of important proximal, contextual information that is critical for understanding how network elements operate in situ. For example, the geographic context of an elevated causeway is significantly different than a landlocked residential street, yet they are both simply represented as edges in traditional network modeling. The purpose of this paper is to explore the implications associated with network abstractions, including the loss of geographic fidelity, and why it matters for vulnerability analysis and emergency response during extreme events. Specifically, this paper introduces an alternative network conceptualization that preserves important on-network information, but also provides in situ context for local network elements, helping to deepen our understanding of local operating conditions. A case study for hurricane storm surge flooding in Volusia County, Florida is utilized to highlight how the developed approach better captures the realities of how evacuees and emergency response teams can leverage information for both on- and off-network space during a major disaster.