Beyond 'flood hotspots': Modelling emergency service accessibility during flooding in York, UK

This paper describes the development of a method that couples flood modelling with network analysis to evaluate the accessibility of city districts by emergency responders during flood events. We integrate numerical modelling of flood inundation with geographical analysis of service areas for the Ambulance Service and the Fire & Rescue Service. The method was demonstrated for two flood events in the City of York, UK to assess the vulnerability of care homes and sheltered accommodation. We determine the feasibility of emergency services gaining access within the statutory 8- and 10-min targets for high priority, life-threatening incidents 75% of the time, during flood episodes. A hydrodynamic flood inundation model (FloodMap) simulates the 2014 pluvial and 2015 fluvial flood events. Predicted floods (with depth >25 cm and areas >100m(2)) were overlain on the road network to identify sites with potentially restricted access. Accessibility of the city to emergency responders during flooding was quantified and mapped using; (i) spatial coverage from individual emergency nodes within the legislated timeframes, and; (ii) response times from individual emergency service nodes to vulnerable care homes and sheltered accommodation under flood and non-flood conditions. Results show that, during the 2015 fluvial flood, the area covered by two of the three Fire & Rescue Service stations reduced by 14% and 39% respectively, while the remaining station needed to increase its coverage by 39%. This amounts to an overall reduction of 6% and 20% for modelled and observed floods respectively. During the 2014 surface water flood, 7 out of 22 care homes (32%) and 15 out of 43 sheltered accommodation nodes (35%) had modelled response times above the 8-min threshold from any Ambulance station. Overall, modelled surface water flooding has a larger spatial footprint than fluvial flood events. Hence, accessibility of emergency services may be impacted differently depending on flood mechanism. Moreover, we expect emergency services to face greater challenges under a changing climate with a growing, more vulnerable population. The methodology developed in this study could be applied to other cities, as well as for scenario-based evaluation of emergency preparedness to support strategic decision making, and in real-time forecasting to guide operational decisions where heavy rainfall lead-time and spatial resolution are sufficient. (C) 2017 The Author(s). Published by Elsevier B.V.