Water Quality in the Willamette Basin, Oregon, 1991-95

Table of contents || Previous section || Next section

Pesticides and Volatile Organic Compounds

About 4.5 million pounds of pesticides are used each year in the Willamette Basin to control weeds, insects, and other pests in agricultural and urban settings (Rinehold and Witt, 1992). Many pesticides are relatively soluble in water, whereas others attach strongly to soil particles. Pesticides are transported from the land surface to streams through a combination of subsurface drainage, surface runoff, and soil erosion. Infiltration of rain and irrigation water facilitate transport of pesticides to ground water.

Volatile organic compounds (VOCs) are synthetic organic compounds that include two main categories: (1) non-halogenated, fuel-related components, such as benzene, toluene, ethylbenzene, and xylenes (BTEX), and (2) chlorinated solvents, such as chloroform, trichloroethene (TCE), and tetrachloroethene (PCE). Because of their uses, VOCs are typically associated with urban environments. Examples of VOC sources include leaky underground storage tanks and emissions from automobiles. In drinking water, they may be carcinogenic or otherwise harmful to human health.

Photo (22,238 bytes)

Herbicides, such as atrazine, are applied primarily in spring and early summer.

Pesticides in Streams

Of 86 dissolved pesticides (herbicides, insecticides, fungicides) and pesticide degradation products analyzed in streams as part of this study, 50 were detected at concentrations greater than or equal to about 0.001 µg/L (micrograms per liter). Of the most frequently detected pesticides, the majority were herbicides.

Atrazine, simazine, metolachlor, deethylatrazine, diuron, and diazinon were the most commonly detected pesticides in stream water. All were found in more than 50 percent of the samples analyzed. Other pesticides were detected in less than 50 percent of all samples.

Maximum recorded pesticide concentrations in stream water were: diuron (14 µg/L), carbofuran (9.0 µg/L), simazine (5.8 µg/L), atrazine (4.5 µg/L), and metolachlor (3.3 µg/L). Eight pesticides were detected at 1-2 µg/L.

Bar chart:Detection Frequency, in percent (25,283 bytes)

Most of the pesticides detected in streams and ground water were herbicides.

Criteria for the protection of freshwater aquatic life from chronic toxicity were exceeded for 10 pesticides in streams.

Criteria for the protection of freshwater aquatic life from chronic toxicity (as summarized by Gilliom and others, in press) were exceeded for 10 pesticides, including atrazine (4 of 183 detections), azinphos-methyl (3 of 3 detections), carbaryl (17 of 46 detections), carbofuran (3 of 51 detections), chlorpyrifos (4 of 65 detections), diazinon (66 of 105 detections), p,p'-DDE (6 of 8 detections), diuron (24 of 83 detections), lindane (1 of 13 detections), and malathion (1 of 12 detections).

In one sample each, atrazine and simazine exceeded their maximum contaminant levels (MCLs) of 3 and 4 µg/L, respectively, as established by the USEPA (U.S. Environmental Protection Agency, 1996) for protection of drinking water. Although the stream where the exceedances occurred (Zollner Creek, northeast of Salem) was not used as a drinking water source, the MCLs are widely accepted criteria for comparing concentrations.

All aquatic life criteria and MCL exceedances occurred at sites draining mostly (greater than 50 percent, by area) agricultural or urban land; the highest pesticide concentrations occurred in agricultural basins.

Atrazine concentrations in streams, like those for nitrate, increased with percent of drainage area in agricultural land, as seen for 20 watersheds less than 490 square miles in the Pudding and Molalla Basins during April 26-29, 1993. Simazine and metolachlor concentrations at these sites behaved similarly to atrazine.

Forty-nine pesticides were detected in streams draining predominantly agricultural land. Forty-three of these pesticides were found in Zollner Creek, which has a drainage basin that is 99 percent agricultural. In contrast, 25 pesticides were detected in urban streams, and 23 of these were found in Fanno Creek, a basin in the Portland metropolitan area that is 92 percent urbanized. Herbicides were detected at twice the rate of insecticides in both agricultural and urban streams. Only atrazine and deethylatrazine were detected in streams draining forested basins (greater than 90 percent forest, by area), and these compounds were present at extremely low concentrations (0.002 to 0.004 µg/L).

Of the six most frequently detected pesticides, all but diazinon were found at highest concentrations in predominantly agricultural streams. Diazinon concentrations were similar in agricultural and urban streams. Three herbicides--prometon, tebuthiuron, and dichlobenil--were detected most frequently in urban streams.

Concentrations of the six most frequently detected pesticides varied seasonally. For example, the graph for Zollner Creek shows that the highest atrazine concentrations generally coincided with spring runoff following application; however, relatively high concentrations also were found during late fall/early winter when increasing rainfall flushed land that had been dry most of the summer.

Atrazine Concentration (22,619 bytes)

Atrazine concentrations increased with percent of drainage area in agriculture for Pudding and Molalla Basin streams, April 26-29, 1993.

Chart:Atrazine concentrations following pesticide application (34,237 bytes)

Atrazine concentrations were highest during spring following pesticide application, but relatively high concentrations also occurred during fall runoff.

Pesticides and Volatile Organic Compounds in Ground Water

One to 5 pesticides were detected in water from each of 23 domestic wells (about one-third of the wells sampled); 13 different pesticides were detected at these sites. Concentrations ranged from less than 0.001 to 0.89 µg/L, with one exception: dinoseb (an herbicide) was detected in water from one well at 7.9 µg/L, which is greater than the MCL of 7 µg/L. This was the only measured exceedance of an MCL for pesticides in ground water.

Ground water with detected pesticides generally came from domestic wells penetrating smaller thicknesses of overlying clay than did ground water containing no detected pesticides. This is partly because pesticides tend to be retained by clay particles. In addition, ground water flows more slowly through clay than it does through coarser grained materials, and the increased contact time with the clay allows greater pesticide degradation.

Only one pesticide and one VOC were detected in ground water from domestic wells at concentrations greater than the USEPA MCL for drinking water.

Pesticides were present at lower concentrations and detection rates in water from urban monitoring wells than from the mostly agricultural domestic wells, possibly reflecting smaller pesticide application rates in urban areas compared to agricultural areas. Trace concentrations (less than 0.01 µg/L) of one or two pesticides were detected in water from three urban monitoring wells, but no USEPA MCLs were exceeded.

One to five VOCs were detected in water from each of seven domestic wells. Six different VOCs were detected, but only tetrachloroethene (one occurrence) exceeded the 5 µg/L MCL.

VOCs were detected in water from a greater percentage of urban monitoring wells than from the mostly agricultural domestic wells, probably reflecting a larger number of point and nonpoint sources of VOCs associated with urban land use practices. From one to four VOCs were detected in water from eight monitoring wells. Five different VOCs were detected; tetrachloroethene exceeded the MCL in one well.

Chart:VOCs in ground water (15,461 bytes)

VOCs in ground water were detected mostly in urban areas.

Additional Information

During 1993-95, 195 samples were collected from 43 stream sites in the Willamette Basin for analysis of dissolved pesticides and pesticide degradation products (p. 25). At four sites, pesticides were analyzed monthly and during extremes in streamflow. Pesticides were analyzed at 39 additional sites during periods of one week or less to examine spatial distributions during high and low streamflow. Site locations are shown on the land use map on page 24.

During 1993-95, 69 of 70 shallow domestic wells and 10 shallow USGS monitoring wells were sampled for pesticides, whereas 65 of 70 shallow domestic wells and the 10 USGS monitoring wells were sampled for VOCs. Site locations are shown on the hydrogeology map on page 24. Because more than 80 percent of ground water used in the Willamette Basin is pumped from alluvial aquifers, only wells completed in alluvium were sampled. Most of the domestic wells were in agricultural settings. USGS monitoring wells were in urban areas comprising primarily residential land, with small amounts of interspersed commercial property.

Laboratory procedures for VOCs and pesticides are given by Rose and Schroeder (1995), Werner and others (1996), and Zaugg and others (1995). Detailed analyses of the Willamette Basin data are presented by Rinella and Janet (1998) for surface water and by Hinkle (1997) for ground water.

U.S. Geological Survey Circular 1161

Table of contents || Previous section || Next section

Suggested citation:
Wentz, D.A., Bonn, B.A., Carpenter, K.D., Hinkle, S.R., Janet, M.L., Rinella, F.A., Uhrich, M.A., Waite, I.R., Laenen, A., and Bencala, K.E., 1998, Water Quality in the Willamette Basin, Oregon, 1991-95: U.S. Geological Survey Circular 1161, on line at <URL:>, updated June 25, 1998 .

This page is a subpage of <URL:>
Email questions and comments to
Last modified: Mon Aug 17 14:17:02 1998