U.S. Geological Survey (USGS), 821 E Interstate Avenue, Bismarck, ND 58503, USA.
USGS, 6000 J Street, Placer Hall, Sacramento, CA 95819, USA.
Sci Total Environ. 2015 Dec 15;538:431-44. doi: 10.1016/j.scitotenv.2015.06.095. Epub 2015 Aug 25.
Trends in pesticide concentrations in 38 major rivers of the United States were evaluated in relation to use trends for 11 commonly occurring pesticide compounds. Pesticides monitored in water were analyzed for trends in concentration in three overlapping periods, 1992-2001, 1997-2006, and 2001-2010 to facilitate comparisons among sites with variable sample distributions over time and among pesticides with changes in use during different periods and durations. Concentration trends were analyzed using the SEAWAVE-Q model, which incorporates intra-annual variability in concentration and measures of long-term, mid-term, and short-term streamflow variability. Trends in agricultural use within each of the river basins were determined using interval-censored regression with high and low estimates of use. Pesticides strongly dominated by agricultural use (cyanazine, alachlor, atrazine and its degradate deethylatrazine, metolachlor, and carbofuran) had widespread agreement between concentration trends and use trends. Pesticides with substantial use in both agricultural and nonagricultural applications (simazine, chlorpyrifos, malathion, diazinon, and carbaryl) had concentration trends that were mostly explained by a combination of agricultural-use trends, regulatory changes, and urban use changes inferred from concentration trends in urban streams. When there were differences, concentration trends usually were greater than use trends (increased more or decreased less). These differences may occur because of such factors as unaccounted pesticide uses, delayed transport to the river through groundwater, greater uncertainty in the use data, or unquantified land use and management practice changes.
评估了美国 38 条主要河流中 11 种常见农药化合物的使用趋势与农药浓度变化趋势之间的关系。对水中监测到的农药进行了分析，以评估三个重叠时期（1992-2001 年、1997-2006 年和 2001-2010 年）的浓度变化趋势，以便在具有不同时间样本分布的地点之间以及在不同时期和持续时间内使用变化的农药之间进行比较。使用 SEAWAVE-Q 模型分析浓度趋势，该模型结合了年内浓度变化和长期、中期和短期径流量变化的测量。使用间隔censored 回归确定每个流域内的农业使用趋势，该回归使用了使用的高估计值和低估计值。受农业使用强烈支配的农药（氰草津、甲草胺、莠去津及其降解产物去乙基莠去津、甲草氯和克百威）的浓度趋势与使用趋势之间存在广泛的一致性。在农业和非农应用中大量使用的农药（西玛津、毒死蜱、马拉硫磷、二嗪农和克百威）的浓度趋势主要由农业使用趋势、监管变化以及从城市溪流中浓度趋势推断出的城市使用变化共同解释。当存在差异时，浓度趋势通常大于使用趋势（增加更多或减少更少）。这些差异可能是由于未考虑到的农药使用、通过地下水向河流的延迟运输、使用数据的更大不确定性或未量化的土地利用和管理实践变化等因素造成的。
