Elsevier

Urban Water

Volume 4, Issue 2, June 2002, Pages 181-189
Urban Water

Is combined sewer overflow spill frequency/volume a good indicator of receiving water quality impact?

https://doi.org/10.1016/S1462-0758(02)00013-4Get rights and content

Abstract

It is often assumed that the frequency or volume of combined sewer overflow (CSO) spill is a good indicator of receiving water pollution impact. Whilst this assumption would appear to be true, recently there have been challenges to its veracity. To test this basic premise, an integrated model (SYNOPSIS) has been applied to the urban wastewater system of a semi-hypothetical catchment. By increasing the storage volume at a single downstream tank in the drainage system, the CSO spill frequency and volume was reduced. River water quality criteria, based on UPM standards, were calculated and related to spill frequency and volume over a series of long-term simulation runs. It was found that, up to certain storage volume levels, decreasing overflow frequency improved river DO and BOD and total ammonia. Beyond these volumes, however, there was no further improvement in DO/BOD and an increase in total ammonia. It is concluded that overflow frequency/volume can be used as a performance indicator for receiving water quality, provided its significant limitations are understood.

Introduction

Combined sewer overflows (CSOs) are used to divert excess stormwater, above the downstream limit, into a nearby receiving water. The original aim of this was to prevent flooding in urban areas and to limit the amount of flow channeled to the wastewater treatment plant (WWTP). This negated the need to provide large downstream sewers and considerable amounts of infrequently used treatment capacity at the WWTP. However, with the widespread improvements in the quality of WWTP discharges it is now fully recognised that CSOs are themselves significant sources of pollution of receiving water systems (Butler & Davies, 2000).

In the UK and throughout Europe, it is widely assumed that the frequency or total volume of overflow discharges is a good indicator of the pollution impact on receiving waters. The less frequently an overflow spills, the lower will be its negative impact on receiving water quality. At the very least, the reduction of overflow frequency leads to the reduction in aesthetic pollution. Regulatory bodies in recent years have focused on identification and improvement of unsatisfactory intermittent overflow discharges under the aegis of the EU Urban Waste Water Treatment Directive (CEC, 1991). In the UK, the third Periodic Review (AMP3: 2000–2005) specifies a timetable for water companies to reduce the number of unsatisfactory intermittent discharges caused by CSOs (Morris, 1999). The focus is particularly on amenity standards where reducing CSO spills should gain improvement, but also aquatic life standards and bathing standards where appropriate.

A common approach to reduce aquatic life pollution caused by overflows is to introduce a storage tank to temporarily store wastewater flows. The stored volume of wastewater will then be diverted back to the WWTP for treatment when there is available capacity. However, prolonged flow to the WWTP may also lead to further negative impacts on the receiving water. This is manifested not only in terms of the increase in total pollutant load discharged but also the possibility of breakdown in plant capacity (e.g. loss of secondary clarifier sludge blanket) due to shock loading (Rauch & Harremoës, 1995). Ideally then, interactions between the urban drainage system, treatment plant and receiving water have to be more fully considered. Therefore, the appropriateness of specifying overflow spill frequency/volume as a direct indicator of receiving water quality is debatable.

This paper presents results of an investigation into the complex relationship between overflow spill frequency and volume, and receiving water quality. Due to the complex nature of processes and interactions that are involved, the Urban Water Research Group's integrated modelling tool, SYNOPSIS (Schütze, 1998), has been used to model the whole urban wastewater system (UWWS). The aim is to demonstrate whether or not there is a strong correlation between overflow spill frequency/volume and receiving water quality.

Section snippets

Synopsis: integrated modeling tool

SYNOPSIS was developed to assist studies of the urban wastewater system following the need for an integrated perspective. The simulation package consists of three main simulation sub-programs for modelling water flow and quality processes in the urban drainage system, WWTP and river system. A number of auxiliary programs are also used for control, optimisation and file management. The following is an outline description of each of the sub-models and a number of important details. A more

Overview of integrated catchment

The catchment used for the study is illustrated in Fig. 2 and is semi-hypothetical in origin (Schütze et al., 1999). The urban drainage system consists of a network of seven sub-catchments, based on an example in the German ATV 128 document (1992) and rescaled to match the capacity of the WWTP. The total impervious area amounts to 725.8 ha. As a simplification, a single on-line storage tank is modelled at the downstream end of the urban drainage system. The flow setting, limiting the maximum

Definition of overflow spill frequency/volume and receiving water quality

The assessment of overflow spill volume is accomplished by totalling the discharge volume of all overflow events. This is easily incorporated into existing auxiliary routines in SYNOPSIS. However, the assessment of overflow frequency is more ambiguous.

As overflow spills are not identical events of a set duration, this leads to difficulty in assessment of overflow frequency in modelling. It is usual for simulation models to be conducted according to a convenient simulation time step. For

Results of simulations

The simulation period chosen consists of a six-month rainfall precipitation record (see Fig. 8(a)). This data is based on the rainfall record in Fuhrberg, Germany for the year 1977. This length of simulation provides adequate consideration of antecedent conditions that are neglected when design storms, single historical events or even annual time-series are modelled.

A total of 120 simulation runs were performed for the six months period. The storage tank volume was increased from a nominal

Discussion of results

Fig. 3 indicates the impact of increasing the volume of drainage system storage up to 68 m3/ha (or 50,000 m3) (i.e. decreasing overflow spill volume and decreasing overflow frequency) on the chosen river DO criterion (i.e. duration of time DO concentration is below the 4 mg/l threshold). Fig. 3 indicates that as storage volume increases (up to 6 m3/ha), DO–Du decreases. However, a level is reached (approximately 4.7%) at which further increase in storage volume has no significant added benefit.

Conclusions

Overflow spill frequency or volume can be used as an indicator of receiving water quality impact, but must be used with considerable care. This is because of the subtle relationships between the individual sub-systems of the UWWS. If additional flow is channelled to the WWTP due to decreased overflow spills, the complex interactions between CSO tank outflow, storm tank overflow and treatment plant effluent can lead to unexpected results, as demonstrated in this work. Therefore, it is

References (17)

There are more references available in the full text version of this article.

Cited by (0)

View full text