Performance measures relating to combined sewer overflow (CSO) operation are widely known to have limitations. In the UK they have evolved from simplistic settings based on six times dry weather flow, to Formula A (Government, 1970) and eventually to more complex consent conditions of spill volumes, spill frequency and water quality. Performance measures related to the quality of the receiving waters have also been developed in an attempt to better represent the impact of CSO spills. However, difficulties associated with measuring CSO spills and associated river flows (including the forecasting of events, taking a representative number of water quality samples, and the general costs of this type of activity) make it very difficult to accurately quantify CSO performances (Blanksby, 2002).
From a network modelling point of view, consent conditions are still a valid first step for assessing CSO performance where a computational model has been built and adequately calibrated. This initial assessment can help to identify the critical CSOs, allowing the prioritisation of resources in these areas. This is particularly useful outside of the UK, where sometimes there is less of a tradition for computational modelling (in particular calibration and verification against survey data) and there are less economical resources available.
In these circumstances, ideally three performance measures should be determined from the model: spill volumes, spill frequency and degree of dilution of the spill. The use of long time series rainfall (based on historical or synthetic data) is common practice these days for defining spill volumes and spill frequency. However, methods for determining the degree of dilution tend to be based on a large number of assumptions and, therefore, this performance measure is often neglected. It is widely accepted that overflow spill volume and spill frequency, used with care, can be considered as receiving water quality impact indicators (Lau et al., 2002). However, two different CSOs with the same spill frequency and spill volume could have very different pollution loads. An estimation of the degree of dilution helps to quantify this and allows a comparison between the performance of different CSOs for a same catchment area.
A new and simple methodology to determine the degree of dilution is presented in this paper. This methodology is based on using the Water Quality model available in InfoWorksCS as a ‘tracer’ for foul flows. This methodology is illustrated with a real-life example: the Fano sewer Master Plan (Italy), where HR Wallingford assessed CSO performance for the local drainage company. The strengths and limitations of this methodology are discussed in the conclusions of this paper.