Progress towards regional flash flood modelling

Abstract:

Flash flooding causes widespread disruption and damage across the UK, with recent research indicating that the occurrence and severity of intense rainfall is likely to increase in the future. To date, our ability to model such events at anything other than local scales has been hindered both by a lack of data at adequate spatial and temporal resolutions and a limited understanding of the processes involved in flooding from short duration, high intensity rainfall events. To enable effective flood risk management and decisions, it is essential that we improve our understanding of the variability in risk from such events across the UK and this requires an ability to undertake flash flood modelling at regional scales.

In this study, we have implemented a coupled hydrological – hydrodynamic model for the representation of flash flooding at regional scales over long durations. To provide inputs to these models, we have developed a gridded sub-daily (hourly) rainfall record of the UK from 1993 to 2011. This enables us to more accurately represent flooding resulting from short duration rainfall events that are poorly represented by commonly utilised daily rainfall data. For a given region of interest we cascade rainfall estimates from our sub-daily dataset into a semi-distributed hydrological model (Dynamic Topmodel) in order to generate river discharge estimates which are then used to force a widely utilised inundation model (LISFLOOD-FP).

Here we present the results from a test case in order to demonstrate the capabilities of the modelling framework over a variety of events with differing characteristics. The results will provide an insight into our capabilities of representing flash flooding and highlight key areas for future model development and enhanced process understanding.