U.S. Geological Survey, California Water Science Center, Placer Hall, 6000 J Street, Sacramento, CA 95819, USA.
U.S. Geological Survey, Washington Water Science Center, 934 Broadway, Suite 300, Tacoma, WA 98402, USA.
Sci Total Environ. 2018 Feb 1;613-614:1469-1488. doi: 10.1016/j.scitotenv.2017.06.156. Epub 2017 Aug 9.
Aquatic organisms in streams are exposed to pesticide mixtures that vary in composition over time in response to changes in flow conditions, pesticide inputs to the stream, and pesticide fate and degradation within the stream. To characterize mixtures of dissolved-phase pesticides and degradates in Midwestern streams, a synoptic study was conducted at 100 streams during May-August 2013. In weekly water samples, 94 pesticides and 89 degradates were detected, with a median of 25 compounds detected per sample and 54 detected per site. In a screening-level assessment using aquatic-life benchmarks and the Pesticide Toxicity Index (PTI), potential effects on fish were unlikely in most streams. For invertebrates, potential chronic toxicity was predicted in 53% of streams, punctuated in 12% of streams by acutely toxic exposures. For aquatic plants, acute but likely reversible effects on biomass were predicted in 75% of streams, with potential longer-term effects on plant communities in 9% of streams. Relatively few pesticides in water-atrazine, acetochlor, metolachlor, imidacloprid, fipronil, organophosphate insecticides, and carbendazim-were predicted to be major contributors to potential toxicity. Agricultural streams had the highest potential for effects on plants, especially in May-June, corresponding to high spring-flush herbicide concentrations. Urban streams had higher detection frequencies and concentrations of insecticides and most fungicides than in agricultural streams, and higher potential for invertebrate toxicity, which peaked during July-August. Toxicity-screening predictions for invertebrates were supported by quantile regressions showing significant associations for the Benthic Invertebrate-PTI and imidacloprid concentrations with invertebrate community metrics for MSQA streams, and by mesocosm toxicity testing with imidacloprid showing effects on invertebrate communities at environmentally relevant concentrations. This study documents the most complex pesticide mixtures yet reported in discrete water samples in the U.S. and, using multiple lines of evidence, predicts that pesticides were potentially toxic to nontarget aquatic life in about half of the sampled streams.
溪流中的水生生物会接触到随时间变化而组成不同的农药混合物,这些变化是由水流条件变化、农药输入溪流以及农药在溪流中的归宿和降解引起的。为了描述中西部溪流中溶解相农药和降解产物的混合物,2013 年 5 月至 8 月在 100 条溪流中进行了一项同步研究。在每周的水样中,检测到 94 种农药和 89 种降解产物,中位数为每个样品检测到 25 种化合物,每个站点检测到 54 种。在使用水生生物基准和农药毒性指数(PTI)进行的筛选水平评估中,大多数溪流中鱼类受到影响的可能性不大。对于无脊椎动物,预测 53%的溪流存在慢性毒性,12%的溪流存在急性毒性暴露。对于水生植物,预测 75%的溪流中会出现急性但可能可逆的生物量效应,9%的溪流中可能对植物群落产生长期影响。在水中,相对较少的农药(莠去津、乙草胺、甲草胺、氯吡虫啉、氟虫腈、有机磷杀虫剂和多菌灵)被预测会成为潜在毒性的主要贡献者。农业溪流对植物的潜在影响最大,尤其是在 5 月至 6 月,此时春汛除草剂浓度较高。与农业溪流相比,城市溪流中杀虫剂和大多数杀菌剂的检出频率和浓度更高,无脊椎动物毒性更高,在 7 月至 8 月达到峰值。基于分位数回归的定量回归表明,在 MSQA 溪流中,Benthic Invertebrate-PTI 和氯吡虫啉浓度与无脊椎动物群落指标之间存在显著关联,以及用氯吡虫啉进行的中尺度毒性测试表明,在环境相关浓度下,氯吡虫啉对无脊椎动物群落有影响,这些结果支持了无脊椎动物毒性筛选预测。本研究记录了美国离散水样中迄今为止最复杂的农药混合物,并使用多种证据表明,约有一半采样溪流中的农药可能对非目标水生生物有毒。