Ptaquiloside from bracken in stream water at base flow and during storm events.

The bracken fern (Pteridium spp.) densely populates both open and woodland vegetation types around the globe. Bracken is toxic to livestock when consumed, and a group of potent illudane-type carcinogens have been identified, of which the compound ptaquiloside (PTA) is the most abundant. The highly water soluble PTA has been shown to be leachable from bracken fronds, and present in the soil and water below bracken stands. This has raised concerns over whether the compound might pose a risk to drinking water sources. We investigated PTA concentrations in a small stream draining a bracken-infested catchment at base flow and in response to storm events during a growth season, and included sampling of the bracken canopy throughfall. Streams in other bracken-dominated areas were also sampled at base flow for comparison, and a controlled pulse experiment was conducted in the field to study the in-stream dynamics of PTA. Ptaquiloside concentrations in the stream never exceeded 61 ng L(-1) in the base flow samples, but peaked at 2.2 μg L(-1) during the studied storm events. The mass of PTA in the stream, per storm event, was 7.5-93 mg from this catchment. A clear temporal connection was observed between rainfall and PTA concentration in the stream, with a reproducible time lag of approx. 1 h from onset of rain to elevated concentrations, and returning rather quickly (about 2 h) to base flow concentration levels. The concentration of PTA behaved similar to an inert tracer (Cl(-)) in the pulse experiment over a relative short time scale (minutes-hours) reflecting no PTA sorption, and dispersion and dilution considerably lowered the observed PTA concentrations downstream. Bracken throughfall revealed a potent and lasting source of PTA during rainfall, with concentrations up to 169 μg L(-1), that did not decrease over the course of the event. In the stream, the throughfall contribution to PTA cannot be separated from a possible below-ground input from litter, rhizomes and soil. Catchment-specific factors such as the soil pH, topography, hydrology, and bracken coverage will evidently affect the level of PTA observed in the receiving stream, as well as the distance from bracken, but time since precipitation seems most important. Studying PTA loads and transport in surface streams fed by bracken-infested catchments, simply taking occasional grab samples will not capture the precipitation-linked pulses. The place and time of sampling governs the findings, and including event-based sampling is essential to provide a more complete picture of PTA loads to surface water.