Technical Report
20.02.2005

TR 094 – Whole Effluent Assessment

TR 094 : Whole Effluent Assessment | February 2005

There is increasing recognition by regulators that there are limitations to the substance-specific approach for assessing and controlling the environmental fate and effects of effluents. Consequently, many regulators are seeking more holistic techniques such as whole effluent assessment (WEA) to supplement existing approaches. Even in countries where whole effluent toxicity (WET) is already assessed there is growing desire to address other issues including persistence and bioaccumulation of effluent components. It is inevitable that new WEA approaches will reveal different issues from those raised by existing substance controls. However, to ensure that these approaches are capable of indicating potential environmental effects, it is important that developing WEA protocols are scientifically robust, sustainable and, ultimately, fit for purpose. This report provides an overview of WEA approaches in terms of their applicability to existing regulation, the types of tests being considered and how WEA could be tailored to meet specific objectives. In addition, case studies are provided with recommendations made regarding both the applicability of a number of WEA approaches and when these should be considered and applied to improve the environmental hazard/risk assessments.

The information reviewed suggests that WEA approaches will increasingly be incorporated into effluent assessment and control schemes. In many of these schemes WEA approaches are seen as new (developing) tools for assessing effluent quality that should be applied in combination with (and not instead of) the substance-oriented approach. Within Europe, WEA-type schemes are generally seen as supporting the hazardous substance strategies of OSPAR and the Water Framework Directive (WFD). As with any initiative there are advantages and disadvantages of WEA approaches. One of the principal advantages of WEA approaches is that these can improve the information relating to environmental hazard of poorly characterised and complex effluents (i.e. those containing unknown mixtures of chemicals) and hence help their risk assessment. Disadvantages could potentially occur if the tests are inappropriate and/or incorrectly applied and interpreted, leading to demands for unjustified risk reduction measures.

The most widely applied WEA schemes assess toxicity to aquatic organisms. These have relevance for the protection of ecosystems although the relevance and interpretation of results ultimately depends on the tests used. For example, experience in the US (Diamond et al, 1999) reveals that acute toxicity observed in WET tests may be traced by impact assessment in the environment. However, more refined toxic endpoints (e.g. those used in chronic toxicity) are not so easily traced to the environment as these can be of limited significance compared to other stressors (effects of shipping, diffuse inputs, etc.). The ephemeral and often intermittent nature of the chronic toxicity for many effluents can make it extremely difficult to pinpoint simple solutions (Diamond et al, 1999; WERF, 2000). WEA can also be extended to the receiving environment to provide additional data to complement existing analytical and biological diversity studies and thereby improve the assessment of both sediment and water quality. In many countries methods are being developed to assess persistence (P) and bioaccumulation (B) of effluent components. Such tests can potentially improve the risk assessment process for discharges but it is important that their limitations are recognised and put into context.

The results of this review indicate that there is considerable practical experience with WET (i.e. whole effluent toxicity) testing and many of the pitfalls and practical problems have been identified. There is reasonable confidence that an appropriate set of tests (at least for acute and, to some extent, chronic toxicity assessments) exist which have guidance on limit conditions for testing and interpreting results. However, this is not the case for methods for assessing the persistence and bioaccumulative characteristics of effluent components. These tests are in a much earlier stage of development and will require more practical experience, standardisation and verification in their application to effluents to demonstrate their usefulness and feasibility. To help facilitate this process an overview of P and B tests and their suitability for incorporation into effluent control schemes has been provided in this report.

The procedures used should ensure that the test results reflect the properties of the sample rather than circumstantial conditions or confounding factors. Thus when measuring toxicity, critical parameters that should be within restricted limits include pH, temperature, dissolved oxygen, hardness, salinity, suspended solids and colour. These parameters may require different limits for different organisms and practical experience suggests that certain substances are often the cause of the toxicity in a sample (e.g. ammonia is relatively toxic and a common component of many effluents). The presence of such substances may mask or interfere with other effects of importance.

One of the key factors that must be considered for any WEA test method is its relevance to the environment to be protected. In a move to increase the sensitivity of biological monitoring and toxicity assessments over and above that seen in traditional bioassays a range of immunological and biochemical tests have been developed. These approaches are referred to as biomarkers because they measure biochemical, cellular or molecular responses (but not adverse effects) induced by exposure to certain stressors. However, it is important to balance sensitivity with environmental relevance and to recognise that not all responses of biomarkers represent adverse effects. For example, it is not surprising that biomarkers are amongst the most sensitive of assays because induction of stress proteins and detoxification systems is the natural response for an organism subjected to a toxicant. However a number of other non-toxicant factors may influence these biomarker responses. If the biomarker is intended for use in WEA, it is important to be able to differentiate whether or not the response can be related to real toxicant environmental effects. For example, in effluent assessments the quality of the water may be impacted by a number of factors in addition to contaminants (e.g. hardness, ionic composition, salinity, pH). These may induce biomarker stress responses that are not contaminant related.  Furthermore, there is no scientific basis to apply screening tests for endpoints that have no relevance to an ecosystem functioning under real world conditions. For example, while several in vitro genotoxicity screening tests, originally developed for human health purposes, have been applied to effluents and environmental samples, these methods are not suitable for use in WEA. Although not yet sufficiently validated for use in WEA, there are several published in vivo methods in aquatic organisms that could be used to assess genotoxic hazard (especially for developmental or reproductive impacts) in a WEA context. Furthermore, there is a continued lack of understanding of the implications of naturally occurring endocrine materials in the wider environment. At this moment therefore it is believed to be inappropriate to recommend the widespread use of in vivo endocrine disruptor testing in a whole effluent assessment programme. However, there may be specific circumstances where such tests should be considered (e.g. production of known endocrine disrupting chemicals).

While the environmental relevance of fish toxicity tests is clear, there are concerns over the relative sensitivity of fish as well as the ethics of their use in WEA. A number of studies suggest (Walker et al 1991; Fentem and Balls, 1993; Weyers et al, 2000) that fish are rarely the most sensitive species in effluent assessments. However, experience from the USA indicates that fish are the more reliable test species. Dyer and Wang (2002) and related studies showed that fish and macroinvertebrates may exhibit different levels of discrimination with fish indicating change more often than macroinvertebrates. It may be that fish and invertebrates pick up on different types of stressor and that more than sensitivity should be considered. In certain circumstances (e.g. need to protect fish spawning grounds or as potential indicators of endocrine disruption) their use will currently be unavoidable. It is therefore unclear at the moment whether or not these should be incorporated into routine WEA programmes. It is important that the range of tests selected from the battery available gives maximum protection to wild fish populations while at the same time minimises the number of fish used in effluent testing.

Site-specific considerations make it impractical to recommend a single standardised WEA testing programme for all effluents. Ultimately these will be tailor-made and influenced by the objectives (e.g. is it for local compliance, environmental impact, tracing source or nature of toxicity components), the nature of the effluent(s) being assessed (i.e. is it a single discharge point or combined discharge from different processes, batch or continuous processes, etc) and finally, the local environmental situation (is the receiving water salt or freshwater, protected ecosystem, etc.). General aspects to be considered when developing a testing strategy are discussed in Chapter 5.

Biological and chemical monitoring is part of the WEA methodology and can be applied for two different purposes. One is to monitor the receiving water to assess whether or not reduction measures have been successful. The second use is in monitoring the receiving water during in the development phase of WEA tests and programmes in order to assess whether the results of such tests are capable of predicting environmental impact. Again the relevance of the tests can be affected by many factors and there is no 'one size fits all' approach for monitoring. Nevertheless there are good examples of a tailored approach yielding good data on discharge impacts. The UK DTA programme on the river Tees used acute toxicity in the receiving water to identify zones of impact from discharges. The oil industry in the North Sea has also utilised similar approaches to assess the very localised impact of produced water discharges.

Since a large number of WEA methods are not fully understood in terms of factors which influence their variability and reliability, the TF recommends that these need to be developed by applying them in practice. In this respect, fact-finding projects carried out jointly by industry and authorities to identify specific areas for improvement appear to produce more meaningful results and are preferable to the introduction of a strict legislative or penalty-based system. Experience with many forms of hazard and risk assessment has shown that ultimately a flexible stepwise approach is advisable when new procedures and methods are being validated.