Use of radar rainfall data to improve calibration of rainfall-runoff routing model parameters

Meteorological radar has advantages for rainfall estimation in hydrology, providing detailed spatial and temporal resolution over an extensive spatial domain. This paper describes a new application of radar rainfall data to calibration of a RORB rainfall-runoff routing model, for the purpose of design flood estimation. A large flood event occurred in the Loddon River upstream of Cairn Curran Dam in October 2000. Observations at a streamflow gauge upstream of the dam indicate that this event had an annual exceedance probability between 1 in 100 and 1 in 200. The radar rainfall approach was able to achieve a much better fit to the observed hydrograph than the conventional raingauge approach. Observations of rainfall accumulations from the radar during the event revealed that all of the five pluviographs missed the most intense rainfall accumulations that fell over the upper parts of the catchment. The large improvement in the quality of the fit between the observed and modelled hydrograph from use of radar data allowed a large reduction in the potential bias associated with estimation of extreme floods using the RORB model.