Stofkoncentrationer i regnbetingede udledninger fra fællessystemer

Summary and conclusions

Analysis of Danish measurement data
This project is concerned with collection, quality control and detailed analysis of flow and runoff concentrations during wet weather in combined sewer systems in Denmark. The data originates from the national Danish surveillance program (in Danish: "Vandmiljøplanens intensive overvågningsprogram") and from earlier investigations that formed the basis for recommended pollution concentrations previously used for calculating annual loads from combined sewer overflows.

International literature review
The data analysis is put in perspective through an international literature review summarising the available knowledge in the area. New recommended concentrations are put forward on this basis, and variations of discharge volumes and pollutant loads are calculated for typical catchments in Denmark, illustrating the implications of the regional precipitation pattern that has been identified recently.

Present calculation practice
The present calculation practice is defined by the Danish Environmental protection Agency (Miljøstyrelsen, 1990b) and used by the Danish counties when reporting annual pollution discharges from combined sewer overflows. Three standard methods of increasing complexity are defined, the most simple being use of areal unit pollution loads and the most advanced being computer simulations with runoff models taking historical rain series as input. Typically, the SAMBA model is used in Denmark for these calculations. The intermediate level based on standard curves and hand calculations is considered obsolete.

Pollution concentrations in stormwater*
The primary focus is estimating concentrations in stormwater* (in Danish: "overvand"). Stormwater* is defined as the increased water and pollution runoff in combined sewers during rain. While the increased flow results solely from rain, the increased pollution runoff results from matter contained in precipitation and accumulated on urban surfaces, and resuspended matter accumulated in the sewer system during dry weather.

This definition of stormwater* is not widespread internationally, and concentrations reported in the international literature typically refer to total concentrations in CSO water. Stormwater* concentrations and total concentrations in CSO water are often comparable in Denmark since stormwater* accounts for a dominating part of the total runoff during rain. Sewer networks are often designed with less capacity in other countries and thus, overflow occurs more frequently and with a larger fraction of wastewater than in Denmark. Comparison of Danish and international concentration levels should therefore be made with caution. However, larger uncertainties probably result from a generally poor documentation of sampling procedures, whether there are upstream overflow structures, how events are defined and whether samples are taken in combined sewer systems or separate storm sewers with a considerable amount of illegal wastewater connections.

Variations of pollution concentrations
International investigations report on substantial variations of pollution concentrations both during rain, between rain events, between sewer catchments and between countries. Different design practices and variations of building materials, waste handling and street cleaning may explain variations between countries. Differences in precipitation, size and type of catchments, the physical conditions of the sewer network and the population density may describe variations between sewer catchments. Finally the properties of rainfall events, the antecedent dry weather period, street cleaning practices and the physical conditions of the sewer system may describe variations between and during rain events. In some countries, design practices account for significant variations of pollution concentrations during runoff events in order to retain the most polluted water in structures. However, only part of the observed variation can be explained.

Three new monitoring sites
The new Danish data is measured at three different sites, one of which is characterised by badly maintained pipes and well-known sedimentation problems. The concentrations at this site are significantly higher than at the two other sites and the sites previously investigated. Therefore, data from this site is only used qualitatively; it is not included when calculation recommended concentrations for use in the annual reporting of discharges from combined sewer systems.

Part of the variation can be explained
By analysing data from the Danish monitoring sites it is concluded that properties of the rainfall, the runoff and the antecedent dry weather period can significantly reduce the observed variation of pollution concentrations. There is, however, still a significant residual variation that cannot be explained by the simple models employed in this investigation. It is concluded that the following stormwater* concentrations should be used for estimation of combined sewer overflows in Denmark in the future:

New Danish recommended stormwater* concentrations
The recommended concentrations of COD and SS are adjusted slightly compared with previous values. Furthermore, lower concentrations are recommended when large off-line storage volumes are present in the sewer system to compensate for the generally decreasing runoff concentrations after 3-5 mm of runoff volume. The lower recommended COD concentrations for extreme events also reflect this effect. Comparison with similar international investigations did not justify changing the recommended concentrations.

New Danish areal unit loads
Areal unit pollution loads are computed based on the new recommended concentrations and the latest knowledge on regional variation of rainfall in Denmark. The unit loads are significantly higher than previous unit loads, particularly in Copenhagen and Western Jutland.

Perspectives
The analysis points at a need for more homogenous calculation of CSO loads that accounts for the latest knowledge. Based on the present data and knowledge, it is possible to develop a simple spreadsheet-based model that calculates discharge volumes and pollution loads based on input detailing the catchment characteristics, pollution concentrations and mean annual rainfall. In combination with future goal-directed monitoring programs it will be relevant to develop more detailed pollution transport models. Such models will be of use when evaluation the potential pollution reduction of e.g. detention ponds, but they will be of limited use in combination with the national Danish surveillance program.