Results of two studies are presented in this report. The first part of the report presents results of pesticide data collected at surface-water sites in the agricultural areas along the South Platte River from Henderson, Colorado, to North Platte, Nebraska, during the 1994 growing season. Samples were collected at 16 to 20 sites on three occasions from May through August 1994 (synoptic samples) and at 2 sites approximately biweekly during the same period (anchor-site samples). The second part of the report presents results of pesticide data collected at two surface-water sites in the Denver metropolitan area. One site was located on the main-stem South Platte River in Denver, whereas the second was located on a tributary (Cherry Creek) at its confluence with the South Platte River, 0.4 miles upstream from the main-stem site. Eighteen samples were collected at each site from December 1993 through November 1994.
Thirty-nine pesticides were detected at least once during the agricultural study. Ten pesticides accounted for about 80 percent of the synoptic sample detections. Of the 10 pesticides, 6 commonly are used for irrigated agriculture in the study area (atrazine, metolachlor, DCPA, cyanazine, EPTC, and carbofuran); one (desethylatrazine) is a metabolite of atrazine; two (prometon, simazine) primarily are used for long-term weed control in noncropped areas; and one (diazinon) is an insecticide used around households and commercially on some vegetable crops.
Of the 39 pesticides detected during the agricultural study, only 8 had median concentrations greater than their respective method detection limits (the 8 median concentrations ranged from 0.004 to 0.15 microgram per liter). Collectively, there were 22 detections of 10 pesticides that were measured at concentrations greater than 1 microgram per liter. The herbicide DCPA was measured at the highest concentration of 30 micrograms per liter. Pesticide concentrations generally were less than standards or guidelines for drinking water and aquatic life.
Atrazine, desethylatrazine, and prometon were detected throughout the agricultural study area and each persisted throughout the growing season at consistent concentrations. Groundwater discharge may transport atrazine, desethylatrazine, and prometon to surface waters in the study area as these three compounds also have been detected at equivalent concentrations in alluvial groundwater in the study area. Other commonly detected pesticides occurred less frequently and had geographic or temporal fluctuations in their occurrence, indicating that pesticide use varies throughout the study area. For example, occurrences of metolachlor, which commonly is applied to corn, primarily were limited to sites located in corn production areas. Occurrences of diazinon, which is used in urban areas and on vegetables, primarily were limited to sites in the upper one-half of the study area where major urban centers are located and where most vegetable production occurs. The timing of pesticide applications resulted in temporal variations of concentrations. For example, the concentrations of the herbicide DCPA were highest in May, following early-season application, and steadily declined throughout the growing season. The insecticide propargite was detected for only several weeks following its application to corn in July, and the insecticide carbofuran was only detected in June following its application to alfalfa and corn.
Twenty-eight pesticides were detected at the two sampling sites during the urban study. Nine pesticides accounted for about 80 percent of the detections at both sites, indicating that pesticide use is similar in each basin. Of the nine most commonly detected pesticides, five (carbaryl, chlorpyrifos, DCPA, diazinon, and malathion) commonly are used by homeowners or commercial applicators in urban areas of the South Platte River Basin. Three (prometon, simazine, and tebuthiuron) are herbicides generally associated with nonagricultural uses in Colorado and are used for long-term, nonselective weed control; and one (atrazine) has had nonagricultural uses in Colorado limited to roadside and turf application since 1992.
Pesticide concentrations measured in the urban samples were small. Seventy-eight percent of the pesticide detections were less than 0.1 microgram per liter, and about 94 percent were less than 0.5 microgram per liter. Carbaryl and 2,4-D were the only pesticides detected at concentrations greater than 1 microgram per liter, and carbaryl was measured at a maximum concentration of 5.5 micrograms per liter. Pesticide concentrations generally were less than standards and guidelines for drinking water and aquatic life, although the U.S. Environmental Protection Agency health advisory for diazinon was exceeded in one sample, and the aquatic-life guidelines for carbaryl, diazinon, and chlorpyrifos were exceeded in several samples. Multiple pesticides were detected in all of the samples.
Individual pesticide concentrations were higher in storm-runoff samples than in nonstorm-runoff samples. The insecticide carbaryl was the dominant pesticide in most of the storm-runoff samples. Carbaryl concentrations in storm-runoff samples generally were 1 to 2 orders of magnitude higher than other pesticide concentrations. DCPA, chlorpyrifos, and malathion generally were detected only during storm-runoff events. Several detections of atrazine, carbaryl, diazinon, prometon, and simazine during dry weather indicate that mechanisms other than storm runoff transport pesticides to streams in the urban environment. Prometon and simazine most likely are being transported by groundwater in the Cherry Creek Basin. Prometon and simazine were consistently detected in dry-weather samples when groundwater was the predominant source of streamflow in Cherry Creek. Comparisons of drainage area, urban area, lawn area, and mean streamflow indicate that the small urban basin (Cherry Creek) is representative of the larger urban basin (South Platte River); however, the percentage of carbaryl load in the South Platte River at Denver attributed to Cherry Creek was not consistent and correlated poorly with the contribution of streamflow from Cherry Creek. Overall, the median percentage of pesticide load in the South Platte River at Denver that was attributed to Cherry Creek (equal to 21 percent) was greater than the median contribution of streamflow from Cherry Creek (12.5 percent).
1 U.S. Geological Survey, Box 25046, MS 415, Denver Federal Center, Lakewood, CO 80225