Author(s): James E Ball
Linked Author(s): James Ball
Keywords: Floods; Data; Gauging; Uncertainty
Abstract: Design flood estimation remains a problem for many professionals involved in the management of rural and urban catchments. An essential aspect of this problem is prediction of the magnitude of the hazard and the exceedance probability of the hazard from suitable data. This suitable data can be obtained from either monitoring of the floods in the catchment or from catchment modelling aimed at replicating floods that would have been monitored were monitoring systems in place. If the suitable data is obtained from monitored data, then the predictions are obtained from a frequency analysis of the monitored data. If the suitable data is generated by catchment modelling, then monitored data is needed to ensure application of the catchment modelling system results in flood predictions that are suitable replications of the data that could have been recorded at the site of interest. In both cases, there is a need to ensure the reliability of the gauged data in terms of flood flows. A typical gauging station, however, monitors river stages and requires a stage-discharge relationship to convert the recorded levels to an equivalent flow. For design flood systems, extrapolation of the stage-discharge usually is required. There is a need to consider the impact of this extrapolation of the stage-discharge on the predicted flood. Presented herein will be the results of an investigation into stage-discharge relationships and design flood estimation.