Identifying the risk to the aquatic environment of endocrine disrupters derived from agriculture

Matthiessen, Peter; Arnold, David; Johnson, Andrew; Pepper, Tim; Pottinger, Tom G.; Pulman, Kim G. T.; Williams, Richard. 2005 Identifying the risk to the aquatic environment of endocrine disrupters derived from agriculture. Defra, 67pp. (CEH Project no: C02581) (Unpublished)

Before downloading, please read NORA policies.
Endocrine disrupters derived from agriculture (Defra).pdf - Accepted Version

Download (1MB) | Preview


EXECUTIVE SUMMARY 1. The purpose of this project was to review possible inputs to UK headwater streams of steroid hormones originating from livestock, to investigate hormone contamination in some streams in which concentrations were expected to be maximal, and to draw conclusions about possible risks that these hormones may pose to aquatic organisms. 2. The review concluded that although livestock in the UK excretes more steroid sex hormones (oestradiol and testosterone) than the human population, almost all of the material deposited on soil by livestock and by manure/slurry spreading is likely to be adsorbed and/or degraded in soil before reaching surface waters. Concentrations of oestrogens in field drains are unlikely to exceed 1 ng/l (expressed as 17β-oestradiol equivalents), a concentration that is probably harmless. However, it is possible that direct excretion by livestock into unfenced streams, and direct run-off to surface waters from slurry stores and hard-standing in livestock farms, may contribute higher concentrations. In other words, poor farming practice may lead to significant steroid hormone pollution. 3. The review also concluded that surface waters in some other countries are contaminated with oestrogens at potentially active concentrations, so it was considered that a survey of UK headwater streams for hormonal activity was justified. The literature search clearly showed that pregnant cattle are the single most important source of natural oestrogens on livestock farms. 4. The chosen sampling strategy was to focus on a limited number of predominantly dairy farms that were considered to represent worst-case conditions for hormone translocation to small headwater streams. Criteria that contributed towards the choice of field sites included stocking type and density, soil type and slope, access of livestock to the stream, application of manure or slurry to the land, possible direct drainage to the stream of waste from leaking slurry stores and hard-standing areas used by livestock, and access permission from the land-owner. Confounding factors such as upstream inputs of hormonally active material from sewage treatment works, septic tank soak-aways, and industrial discharges, were excluded as far as possible from the study. 5. In order to obtain semi-quantitative, time-integrated samples of hormones in water, locations up- and downstream of livestock activity were sampled on 10 farms using a passive, solid-phase device known as a Polar Organic Chemical Integrative Sampler (POCIS). These were deployed between November 2004 and January 2005 for 3 to 10 weeks (mean = 39 days). At an eleventh site, a field drain issuing from an experimental plot of cracking clay soil treated solely with dairy cow slurry was also sampled with POCIS. At one site, an automatic flow-driven water sampler was deployed alongside the POCIS to capture water soon after heavy rainfall. 6. POCIS and water extracts were assayed for estrogenic and androgenic activity using the in vitro yeast estrogen screen (YES) and yeast androgen screen (YAS), respectively. As part of a separate project, POCIS extracts were also analysed chemically for oestrone (E1), 17β-oestradiol (E2) and 17α-ethinylestradiol (EE2) by the Environment Agency. 7. The flow from only one rainfall event was captured in its entirety by the autosampler, but this revealed a background concentration (E2 equivalents) of 0-0.3 ng/l, rising to a transient peak of 9.4 ng/l. Average E2 activity at this site as determined from the POCIS samplers was 1.8-2.7 ng E2 equiv./litre, which provides confidence that the POCIS results are reliable. 8. Estimated oestrogenic activity across all sites (with one exception) lay in the range zero-26.5 ng E2 equiv./litre (mean = 2.0 ng/l; standard deviation = 5.1), based on the POCIS samples. The outlier was 292 ng/l, but this could not be specifically linked with intensive livestock rearing. 92% of monitoring stations (at least one on each farm) contained some oestrogenic activity. 9. In 5 of 9 livestock farms where upstream/downstream comparisons were possible, the downstream oestrogenic activity was higher than upstream, implying inputs from the farms under study. There was one case (Farm 3) where there were no known confounding factors whatever, with very little upstream contamination, and the farm increased activity by a factor of 7. 10. However, upstream activity was sometimes higher than downstream, suggesting possible inputs from phyto-oestrogens and scattered septic tank overflows, and in-stream adsorption and/or degradation. There was a low background level of oestrogenic activity in all but two locations. 11. The data did not generally permit discrimination between different potential sources on the farms, but it seems likely that the observed oestrogenic activity was mainly caused by a combination of slurry spreading and farmyard runoff, with direct excretion to pasture by livestock probably contributing less. In one case (Farm 7 slurry application experiment), activity in the field drain was directly attributable to dairy slurry alone. 12. On 8 of the 11 surveyed farms, oestrogenic activity in the stream (or field drain in one case) exceeded the Predicted-No-Effect-Concentration for E2 of 1 ng/l. In two cases, activity was probably high enough to damage reproduction in fish, although in neither case was livestock itself likely to have been the primary cause. 13. Although no EE2 was detected analytically in any stream, E1 and E2 were ubiquitous, with E2 equivalents ranging from 0.04 to 3.62 ng/l across all but two sites. Furthermore, concentrations downstream of livestock were generally higher than upstream, more markedly so than for the YES data. The oestrogen concentrations agree well with the YES data and these observations suggest that most of the detected activity was attributable to E1 and E2 derived from livestock. However, the low levels of oestrogenic activity detected by the YES upstream at several stations, and the much higher upstream levels at Farms 11 and 13, could not be explained by E1 or E2, and it is postulated that phytooestrogens may have contributed to this signal. 14. Although all streams were assayed with the YAS for androgenic activity, this was only detectable in two cases, and at levels which are unlikely to pose a threat to fish. However, it should be noted that uptake of testosterone by the POCIS has not yet been calibrated. 15. On the basis of this survey, the possibility that natural oestrogens (from both livestock and other sources) in headwater streams are causing adverse effects in fish cannot be excluded. 16. Recommendations are made for further research to discriminate between sources, and to evaluate the risks to fish.

Item Type: Publication - Report
UKCEH and CEH Sections/Science Areas: _ Environmental Chemistry & Pollution
_ Water Quality
Funders/Sponsors: Department for Environment, Food and Rural Affairs (Defra)
Additional Keywords: oestrogens, farm animal excretion, manure, slurry, headwater streams, endocrine disruption, run-off
NORA Subject Terms: Ecology and Environment
Agriculture and Soil Science
Biology and Microbiology
Date made live: 24 Sep 2015 14:04 +0 (UTC)

Actions (login required)

View Item View Item

Document Downloads

Downloads for past 30 days

Downloads per month over past year

More statistics for this item...