In vitro bioassays are sensitive screening tools to detect complex mixtures of micropollutants in water samples, but micropollutants are often present at low concentrations, particularly in drinking water and clean source water. Therefore, there is a need to enrich samples prior to bioanalysis. This report summarizes commonly applied sampling strategies and sample preparation options and proposes a decision-making tool to assist users to select appropriate sampling methods.

The sampling strategy will depend on the purpose and objective of the sampling campaign, as well as the sample context. For example, if the purpose of a sampling campaign is to assess the product quality of a drinking water treatment plant, with the objective of comparing the effect in the treated product water to an effect-based trigger value, then only the product water needs to be collected. If we want to understand critical processes, it might b e required to sample at different stages of a treatment chain. The type and volume required will depend on the sample, with composite sampling and smaller volumes recommended for wastewater influent and effluent. In contrast, grab sampling and larger sample volumes are more suitable for drinking water and recycled water.

Once a water sample has been collected, a number of decisions regarding sample pre-treatment need to be made. For example, samples should be processed within 48 hours of collection, while samples containing chlorine should be quenched immediately after collection. While many studies filter samples prior to enrichment, the type of filter and the filter pore size used varies widely in the literature. We recommend filtering water samples with a turbidity of 5 nephelometric turbidity units (NTU) or greater using glass fibre filters with a pore size between 0.7 to 1.5 μm.

Common sample extraction methods include solid-phase extraction (SPE), passive sampling and liquid-liquid extraction, with SPE the most frequently used method to enrich micropollutants prior to bioanalysis. Testing native or unenriched water samples is not recommended as it is not possible to differentiate the effect of organic micropollutants from other components in water. Different SPE sorbents have been used in the literature, with multilayer SPE containing different sorbents to capture a wider range of micropollutants increasingly being applied. However, blank effects have previously been detected, so it is important to select an extraction method with no blank effects as bioassays cannot differentiate between effects from a water sample and effects from impurities from sample processing. Few studies have evaluated effect recovery by SPE, with one study finding effect recovery to be within a factor of two of the optimal 100% recovery for most bioassays.

As there are a number of decisions to be made regarding sample collection, pre-treatment and enrichment a decision-making flow chart was developed to guide users through the key steps. Once the final sample pre-treatment and processing methods have been selected, it is important to use the same approach for all samples that you want to compare. It is important to note that the information used to support the sample processing decisions is often based on user experience and chemical analysis protocols, with few studies investigating the impact of different sample processing options on the biological effect. One of the least standardised, but very important, pre-treatment steps is sample filtration, with further experimental work required to validate the approach proposed in this report.