Evaluating the importance of catchment hydrological parameters for urban surface water flood modelling using a simple hydro-inundation model

The influence of catchment hydrological processes on urban flooding is often considered through river discharges at a source catchment outlet, negating the role of other upstream areas that may add to the flooding. Therefore, where multiple entry points exist at the urban upstream boundary, e.g. during extreme rainfall events when surface runoff dominates in the catchment, a hydro-inundation model becomes advantageous as it can integrate the hydrological processes with surface flow routing on the urban floodplain. This paper uses a hydro-inundation model (FloodMap-HydroInundation2D) to investigate the role of catchment hydrological parameters in urban surface water flooding. A scenario-based approach was undertaken and the June 2007 event occurred in Kingston upon Hull, UK was used as a baseline simulation, for which a good range of data is available. After model sensitivity analysis and calibration, simulations were designed, considering the improvement of both the urban and rural land drainage and storage capacities. Results suggest the model is sensitive to the key hydrological parameter soil hydraulic conductivity. Sensitivity to mesh resolution and roughness parameterisation also agrees with previous studies on fluvial flood modelling. Furthermore, the improvement of drainage and storage capacity in the upstream rural area is able to alleviate the extent and magnitude of flooding in the downstream urban area. Similarly urban drainage and storage upgrade may also reduce the risks of flooding on site, albeit to a less extent compared to rural improvements. However, none of the improvement scenarios could remove the flow propagation completely. This study highlights that in some settings, urban surface water flood modelling is just as strongly controlled by rural factors (e.g. infiltration rate and water storage) as internal model parameters such as roughness and mesh resolution. It serves as an important reminder to researchers simulating urban flooding that it is not just the internal parameterisation that is important, but also the use of correct inputs from outside the area of study, especially for catchments with a mixture of urban and rural areas.