Fine-resolution evaluation of urban fire service accessibility under the impact of a 100-year pluvial flood

Taking the central urban area of Shanghai Municipality as the study area and using the enhanced two-step floating catchment area method (E2SFCA), this study evaluated the accessibility of urban fire service (UFS) under the impact of a 100-year pluvial flood, which can cause extensive road closures. A high-resolution hydrodynamic model (FloodMap-HydroInundation2D) was used to produce flood scenarios at very fine spatiotemporal scales. The results show that: 1) Under the flood scenario, the most serious flood extent (water depth > 50 cm) totals about 1.5 km² and may cause 471 road closures (accounting for 5.11% of the entire road length). Flooding on the west side of the Huangpu River is slightly more severe than the east. 2) The spatial distributions of UFS accessibility are not the same at different time periods of the day, but they show a very similar pattern that the accessibility decreases from both sides of the Huangpu River to the northwest and the southeast peripheral areas. 3) Compared with normal weather situation, the number of inaccessible spatial units (250 m × 250 m) at the night, morning peak, nonpeak daytime, and evening peak hours increases by 36.32%, 35.89%, 39.07% and 32.01% respectively under the flooding scenario. During the entire process of the pluvial flood event (120 min), both the maximum number of inaccessible units and the most significant spatial variance emerge at the second phase of the rain peak (30-45 min). 4) Urban fire service accessibility displays significant spatial agglomeration characteristics under both the normal and the flooding scenarios. The High-High clusters are mainly located along the two sides of the Huangpu River and the western peripheral areas, while the Low-Low clusters are primarily distributed in the northwest and southwest. Neither the High-Low clusters nor the Low-High clusters are significant. 5) The spatial mismatch of UFS demand and accessibility (low demand-high accessibility or high demand-low accessibility) is evident even under the normal weather situation, and can be further aggravated if a 100-year pluvial flood occurs. These results can provide a scientific support for improving the smart management of pluvial flood disaster and emergency response service in urban areas.