The Agency has been conducting research into the endocrine disrupting impact of sewage effluents on fish populations for many years. It has shown that steroid oestrogens - natural and artificial - are the most important compounds and that the effects can cause significant declines in the fertility of male fish.
The Agency has been urging that it is time to investigate practical measures to demonstrate how the impact of oestrogenic sewage discharges on fish populations might be reduced. However, the water industry and its economic regulator Ofwat have been sceptical that large amounts of money should be invested in additional sewage treatment (ENDS Report 352, pp 5-6 ).
An ambitious programme to test two different treatment technologies at several sites, costing around £100 million, was put forward by the Agency as part of the water industry's investment plan for 2005-2010. This has now been slimmed down, but draft price limits announced this month by Ofwat included provision for research at two sites costing £25 million (see pp 11-12 ).
A continuing research effort by the Agency has underpinned this success. The latest contribution passed to ENDS this month is a review paper including the latest results of field studies and chemical analyses which present a strong body of evidence in support of the Agency's position.1The review includes a package of new studies, partly conducted in-house by the Agency but also in collaboration with experts from Brunel and Exeter Universities and the Centre for Ecology and Hydrology.
A key part of the review is the largest survey yet of endocrine disrupting effects in wild fish in the UK. The survey aimed to look at the prevalence of endocrine disruption in wild fish rather than concentrating on known hot spots, and a further aim was to develop a model to predict the impact of sewage effluent discharge on fish populations.
The researchers selected a total of 466 sewage treatment works with population-equivalents of over 10,000 people and classified them into high, medium and low-risk categories. The assessment was based on the likely load of steroid oestrogens entering a sewage works, the removal efficiency during treatment and the dilution present in the receiving water.
Some 142 of the works were classified as high risk (30%), 192 as medium risk (41%) and 132 (28%) as low risk.
Wild roach populations were sampled from rivers downstream of 57 of the sewage works and assessed for the effects of endocrine disruption. A total of 1,615 roach were examined, 21 of which were classed as high risk, 30 as medium risk and six as low risk.
The fish were examined for the prevalence of oestrogen-induced deformities, such as ovotestis - the formation of eggs within the testes of male fish - and the presence of female reproductive ducts in male fish. The fish were scored on a feminisation index on the basis of the number of eggs in the testes and the presence or absence of abnormal reproductive ducts.
Previous research has shown that the ovotestis can be caused by exposure to oestrogens but the results of the study strongly indicated that its extent is dependent on the level of oestrogens present. High-risk sites had a much higher incidence of intersex than medium-risk sites, which in turn had higher incidence than low-risk sites (see figure).
The survey also showed that intersex became more pronounced in older fish, but female oviducts appeared to be formed only in juvenile fish which developed under the influence of oestrogens.
In all, some 218 fish were intersex - a third of all males sampled - and some 117 of these also had abnormal reproductive ducts. Intersex was found at almost all of the sites, downstream of 44 of the 57 discharges studied.
The research continued by examining the chemistry of discharges and comparing this with predictions about the oestrogenic potential of the discharges. Effluent samples were collected from a total of 41 sewage works on each of two occasions. They were analysed for the natural oestrogens 17ß oestradiol and oestrone, the synthetic oestrogen used in the birth control pill ethinyl oestradiol, and the industrial chemical nonyl phenol.
The effluents were also tested for total oestrogenic activity using a yeast-based assay, the YES screen, and for anti-androgenic activity using a similar yeast-based anti-androgen screen known as YAS.
All of the effluents tested showed oestrogenic activity. Most contained detectable levels of steroid oestrogens likely to cause oestrogenic effects on fish, even after dilution in the watercourse. However, the levels of nonyl phenol measured - up to 7.7µg/l - were judged unlikely to be responsible alone for oestrogenic effects on fish, although they might be a significant contributor via a cocktail effect (ENDS Report 336, pp 20-23 ).
One surprise was that most effluents showed significant anti-androgenic activity due to unknown chemicals. Up to 764µg/l of flutamide equivalents were measured - expressed as the amount of a strong synthetic anti-androgen required to cause similar impacts. The study concluded the levels were "sufficient to induce biological responses in fish".
The results suggest fertile ground for further research on endocrine disruptors in sewage. At present it remains pure speculation whether these anti-androgens are the result of natural or industrial inputs to the sewer.
Charles Tyler of the University of Exeter said that a toxicity identification evaluation of the kind that tracked down the oestrogenic activity in effluents would be likely to answer the question and that that research was now under way.
Finally, the research compared the actual levels of oestrogenic compounds measured in effluents with the amounts predicted using model assumptions.
The results were encouraging (see table). The majority of results agreed well - suggesting that the model was broadly accurate. Where discrepancies occur these could well be due to the fact that only two effluent samples were measured from each treatment works.
The review concludes: "There is sufficient weight of evidence to show that sewage effluent discharges are leading to the feminisation of male fish in English rivers....The implication of these effects is potential impact on the sustainability of some fish populations. However, the effects appear to be permanent, progressive and can be concluded as harm to individual male fish."
The effect on reproductive health has been shown in moderate to severe levels of effects and the progressive nature of the condition indicates that males' functionality decline as they get older."
The Agency's director of environmental protection, Andrew Skinner, added: "The latest survey reinforces the need to look at cost-effective ways of minimising endocrine-disrupting substances. We are responsible for the protection of the environment through the control of pollution and we have specific duties for the protection of fish. There is sufficient evidence of harm...that action needs to be taken now to find out how to control this."
Catherine Wright, the Agency's policy officer on water industry investment, explained that the current plans were to test granular activated charcoal treatment of effluents at two sewage works: Thames Water's Swindon works and Severn Trent's Hallam Fields works at Ilkeston.
Plans to test ozone treatment have been dropped, although there is a possibility that existing ozone facilities at Leek sewage works, developed to deal with colour effluents from textile dyeing, might be a suitable surrogate.
In addition, some 17 other sites would be studied intensively to examine how different treatment methods perform in degrading steroid oestrogen inputs.