Evaluation of Road Network Resilience to Natural Hazards using Network Analysis

Roadway vulnerability assessments are often used to predict which routes are currently, or may in the future, be subject to natural hazards. However, these assessments are often conducted for individual roadways and therefore do not assess to what degree road closures affect the connectivity of road networks – i.e., the ability for a user to access other roads in the network. A consequence of this is that future roadway retrofits, such as raising the elevation of roadways, could alter network connectivity in a way that has cascading impacts on community accessibility during extreme events. The principal goal of this project is to improve predictions of roadway vulnerability by using network science and network analysis to understand the connectivity of road networks during extreme events. By treating road intersections as ‘nodes’ and road segments as ‘edges’, the research team can successively remove nodes based on some criteria (such as increasing elevations, akin to flooding or another extreme event) to identify the threshold where the entire network starts to break apart. The network analysis proposed in this project is focused on coastal settings, and specifically flood hazards, but the methodology is broadly applicable to other regions of North Carolina and additional natural hazards (e.g., landslides). To achieve the project goal, the team proposes the following objectives: (1) Evaluate network connectivity on NC barrier islands for a range of node removal proxies. The team will adapt an existing network model, developed by the PI for node removal by elevation, to evaluate vulnerability metrics currently in use by the North Carolina Department of Transportation (NCDOT) (e.g., distance to shorelines). (2) Compare mathematical models of network failure to real-world examples. The team will use NCDOT data of past roadway incidents and hydrodynamic model outputs to assess the efficacy of different node removal proxies. This data will be supplemented with images from new cameras deployed at critical intersections identified by the models. (3) Explore roadway retrofit scenarios for climate change adaptation using the network models. The team will work with NCDOT to evaluate how potential structural changes to existing roadway networks in response to climate impacts might affect future roadway connectivity. (4) Extend the network model and analysis to an inland location (e.g., Kinston, NC). This will allow the research team to test the viability of the method in different (non-barrier island) settings with compound flood drivers (i.e., riverine flooding, rainfall runoff, and storm surge). The associated research products will include a map of the critical nodes – the intersections that break network connectivity when not functioning – for different node removal proxies and retrofit scenarios, which could be computed and identified in FIMAN-T alongside storm or shoreline change scenarios. The map could also be used to inform locations for longer-term monitoring by NCDOT during disruptions to ensure public safety. More broadly, this project will lead to a more holistic framework for identifying roadway and network vulnerability to a range of hazards and inform resilient management of roadway networks in a changing climate.