The influence of rainfall time distribution on the design discharge

The peak discharge in sewer systems is usually estimated by design models based on the uniform hyetograph; in this hypothesis, for a fixed duration the rain storm has a constant intensity, which value can be calculated by the depth-duration-frequency (DDF) curve, referred to a selected return period. On a theoretical point of view, this approximation leads to a discharge value which is smaller than the expected value generated by a real rainfall, having the same cumulated rain depth. In this paper the historical series recorded at the rain gages of Naples and Milan (Italy) have been used as rain input for a continuous simulation based on the linear reservoir model. Consequently, it was possible to evaluate the effect of the two fundamental hypotheses of the linear reservoir method: · the hyetograph is constant; · the reservoir is empty when the rainfall starts. From the data analysis some interesting conclusions arise. In general, the two hypotheses, before mentioned, lead to an unsafe estimation of the peak discharge. Selected design hyetographs (Chicago, Sifalda, and Desbordes) have been considered in order to take in account the effect of the uniform hyetograph assumption. Finally it is shown that, despite of a larger computational cost, the design hyetographs do not lead to more accurate results in terms of estimation of the peak discharge.