Author(s)

G. Becciu & A. Raimondi

Abstract

Water Storage tanks have proved to be effective for runoff control in drainage systems. Standard design procedures follow an event-based approach: a single flood at a time is considered and tank is assumed completely empty at the beginning of its filling. The possibility of pre-filling from previous events is then neglected and underestimation of storage volume can occur. In this paper an analytical probabilistic approach to estimate the probability of pre-filling is presented and effects due to outflow rate and storage volume are investigated. Derived formulas are validated by their application to a case study. Keywords: water storage tanks, analytical probabilistic approach, pre-filling probability. 1 Introduction In last decades the significant and rapid increase of impervious surfaces has made more frequent the overload of drainage systems, with uncontrolled polluted spills into receivers. Runoff control for reduction of combined and separated sewer overflow spills is typically accomplished through the implementation of storage tanks. Although the proper modeling of these facilities should be based on the analysis of stochastic process of flood events, often an event-based approach is adopted for their design. A design (critical) rainfall, extracted from a recorded series or defined by a standard pattern, is used as input of a rainfall – runoff model to reproduce a flood hydrograph from which the needed minimum tank capacity is estimated. This flood event, although associated with a return period, is considered isolated from the whole stochastic process and the tank is assumed always empty at its beginning. Pre-filling from previous events is then neglected and an underestimation of the minimum storage size may occur.

Keywords

water storage tanks, analytical probabilistic approach, pre-filling probability