Recommended Design Storm Duration
The method for estimating the duration of a design storm in the ReFH method was adopted from the FSR/FEH method, where the design storm duration (D) is based on a formula, which approximates the duration giving the largest flood magnitude (Houghton Carr, 1999). The design storm duration for a particular catchment depends on the response time of the catchment (time to peak, Tp, as estimated from catchment descriptors) and the general wetness of the catchment (as measured by the standard average annual rainfall, SAAR):
Sensitivity analysis shows that curves of flood magnitude against storm duration are generally flat, indicating that peak flow estimated generated using the ReFH method are not overly sensitive to the choice of storm duration. However, the duration will have an impact on the volume of the generated design flood event. The longer the design storm, the larger the volume of the resulting design flood.
The duration estimated in this way is commonly referred to as the Recommended Duration. In practice the duration that gives the largest peak flow for a given design package of parameters and initial conditions, the Critical Duration, may depart from this. This departure will be different for the two FEH13 rainfall base design packages and for the legacy FEH99 package. Caution should be used when adopting the Critical Duration as the design packages are calibrated around the assumption of the Recommended Duration.
When applying very long duration storms to the FEH13 non water balance and FEH99 design packages, ReFH may significantly over-estimate the baseflow as the model is being applied outside of it’s calibrated range. This is addressed within the FEH13 water balance option as described in Water balance closure in the Rural model.
The ReFH2 software ensures that the default estimate of design storm duration is an odd integer multiple of the chosen model time step to enable the design storm hyetograph to be defined with a central peak.