Toxicity characterisation of organic contaminants in stormwaters from an agricultural headwater stream in South East England

Abstract

The transient movement of pesticides at biologically active concentrations during storm events is considered to be a cause of biological impoverishment in some headwater streams. The programme of work described sought to identify compounds that are the cause of toxic effects during such events. Along with targeted pesticide analysis, toxicity identification evaluation (TIE) procedures were used to identify compounds with a demonstrated toxic effect. These procedures were specifically directed towards isolating and attributing toxicity to classes of organic contaminants in samples collected from an English headwater stream during a storm event. The organic load was isolated by means of solid-phase extraction (SPE). Bioassay of the SPE extract at ×100 whole water concentrations confirmed that the samples contained substances toxic to Daphnia magna, although the raw samples were not toxic. Targeted pesticide analysis identified simazine and diuron as the major pesticides present and, using a toxicity unit (TU) approach, were shown to be responsible for a significant amount of the observed concentrate toxicity during a runoff event. However, they were not present in sufficient quantities to be totally responsible for a more toxic later event. By simplification of the SPE isolate using reverse-phase HPLC, fractions from which were tested for toxicity, the cause of concentrate toxicity in the later event was isolated to two discrete fractions. GC–MS analysis of these fractions identified nonylphenol, endosulfan sulphate and pendimethalin as present, with the majority of toxicity attributed to nonylphenol (NP). The main advantage of the TIE approach is that it allows biological active compounds with a demonstrated effect to be identified that may not be selected by more traditional techniques.