The adverse effects of contaminants on predator-prey interactions : implications for ecological risk assessment.

Predator-prey interactions play an important role in maintaining community structure and thus ecosystem function. However, these interactions can be disrupted by exposure of prey to contaminants released into the environment. Depending on the exposure concentration, prey can experience lethal or sublethal toxic effects. They may also take up contaminants into their tissues. This may affect prey availability or the transfer of contaminants to, and potentially toxic effects in, predators. The aim of this thesis was to examine whether changes in predator-prey interactions could exacerbate the negative impacts of contaminants via such effects. An experimental approach was taken, using a macroinvertebrate food web consisting of two predator (Notonecta glauca and Ischnura elegans) and three prey species (Asellus aquaticus, Chironomus riparius and Cloëon dipterum), and two contaminants, a non-essential metal (cadmium) and a hydrophobic organic (benzophenone). Three main conclusions were drawn from this work. Firstly, the extent of trophic transfer depends on prey exposure pathways and predator feeding behaviours, with subsequent secondary poisoning depending on predator critical body residues, contaminant modes of action, and species traits. Secondly, contaminants with low acute to chronic toxicity ratios can act by reducing prey survival, resulting in increased predation or altered prey choice by predators, thereby reducing prey populations that are unable to recover. Thirdly, contaminants with high acute to chronic ratios can have sublethal effects on prey, affecting prey susceptibility to predation or reducing prey populations. The impact of the resulting increased predation depends on the type of sublethal effect, predator traits, and the potential for recovery. This thesis indicates that the current risk assessment approach of focusing on direct exposures of single species to contaminants may be under protective. Instead, a scenario-based approach may be more appropriate, taking into account the interactions between contaminant properties, species traits, landscape profiles and exposure durations to identify situations where changes in predator-prey interactions may be particularly detrimental.