HRGSA BMP Project / Areawide II Handbook

Pesticide Monitoring in Hood River Area Streams

Since the Oregon Department of Environmental Quality (DEQ) began sampling in 1999, there has been an ongoing monitoring program for pesticides in Hood River area streams. The program has been conducted as a joint project between Oregon State University (OSU) and DEQ scientists. This has been a collaborative effort with broad support from local and state entities including the Hood River Grower-Shipper Association, the Hood River Soil and Water Conservation District, the Oregon Department of Agriculture, OSU-MCAREC, and the Confederated Tribes of Warm Springs.

Water samples were collected during two different periods of each growing season from several locations in the Hood River Valley. Sampling sites were instrumented to collect local weather and stream height data. Sampling focused on the organophosphate (OP) insecticides chlorpyrifos (Lorsban) during the pre-bloom period (March to early May), and azinphos-methyl (Guthion) during early summer (June to mid-July.) Additionally, DEQ has analyzed samples for a wide array of other pesticides including phosmet (Imidan), malathion, and simazine, and conducted instream studies with caged fish, laboratory bioassays with aquatic invertebrates and juvenile steelhead, and instream aquatic invertebrate surveys.

As part of an ongoing monitoring program, reported here are the chlorpyrifos and azinphos-methyl monitoring results for the years 2000 to 2003. Results to date are presented to show overall progress towards reducing the occurrence of these commonly used OP pesticides in Hood River area streams.

Monitoring for OPs and other pesticides in water at low parts per trillion levels (1 ppt = 0.001 micrograms per liter or ug/l) is analytically challenging. In addition, the use of rapid and cost effective methods for sampling and analysis often requires that analytical instrumentation be pushed to its limits. However, the ability to detect pesticides at such low levels is necessary in order to assess stream loading relative to the Clean Water Act (CWA) criteria for the protection of aquatic life. The CWA standards for chlorpyrifos are 0.042 micrograms per liter for chronic exposure and 0.081 micrograms per liter for acute exposure. Currently, azinphos-methyl has only a chronic water quality standard of 0.010 micrograms per liter.

Different strategies were employed by DEQ and OSU to address the analysis challenge. DEQ used multi-residue methods to analyze for total residues (bound to sediment and dissolved), while OSU used single residue methods to analyze for dissolved residues only. DEQ reports values above the method detection limit; generally 0.02 micrograms per liter, whereas OSU reports values above the instrument detection limit of 0.001 micrograms per liter. In addition, OSU sampled Neal, Evans, and Indian creeks, collecting triplicate samples three times weekly at about mid-day, whereas DEQ sampled Neal, Evans, and Lenz creeks, collecting single samples less frequently. These Hood River tributaries were selected to represent areas of the lower valley where land use is primarily agriculture (Neal and Lenz creeks), as well as mixed urban and agricultural use (Indian Creek). Evans Creek is located in the upper Hood River valley where land use is primarily agricultural. To better understand the fluctuation in concentrations, OSU also collected hourly samples continuously for 48 hours (2000 and 2002 seasons) and 96 hours (2003 season) from Neal Creek on two separate occasions during both chlorpyrifos and azinphos-methyl use periods. Results of continuous sampling indicate that concentrations can fluctuate significantly over a period of a few hours, suggesting that OP concentrations in grab samples collected at a different time of day, or on subsequent days, may vary considerably.

In addition, in successive years sampling plans have been modified slightly. For example, between 2001 and 2002 the sampling site on Indian Creek was moved downstream approximately 0.5 mile, and the azinphos-methyl sampling period was extended from the end of June to mid-July.

Differences in sampling times, analysis methods, and detection limits must be considered when attempting to make comparisons between years and between DEQ and OSU monitoring results. However, based on sampling and analysis methods described above, several general observations can be made:

s DEQ and OSU data from common streams showed similar trends for samples with detectable concentrations of chlorpyrifos and azinphos-methyl, and samples exceeding water quality standards, although maximum concentrations reported by DEQ were generally higher.

s Concentrations which exceeded the acute water quality standard (WQS) for chlorpyrifos were not frequent and generally decreased from 2000 to 2003, except at Lenz Creek.

s Although more frequent, concentrations which exceeded the chronic WQS for chlorpyrifos generally decreased from 2000 to 2003. OSU results show that concentrations exceeding the chronic WQS were transient; often for only one sampling date, and no more than two successive sampling dates.

s Concentrations which exceeded the chronic WQS for azinphos-methyl were relatively common and have become more frequent at some sample collection locations.

s Both grab samples and continuous sampling periods show no relationship between rainfall patterns and OP concentrations in water.

Results of the OSU monitoring program from 2000 to 2003 (beginning with the azinphos-methyl use period in 2000) are shown in Figures B-1 to B-7.

Both DEQ and OSU sampling found very few detections of phosmet. As phosmet has a short half-life in solutions at pH greater than 6.5, rapid degradation in Hood River streams may be responsible, in part, for the lack of detections.

Although the results for chlopyrifos and phosmet are encouraging, those for azinphos-methyl are cause for concern. Regulatory action may result if voluntary efforts to correct the situation are not successful.

Lack of a discernable relationship between rainfall and OP concentrations in water is suggestive that runoff may not be a significant factor in stream loading. Further research is necessary to investigate this preliminary finding.

Assuming responsible mixing/loading and storage practices, drift associated with pesticide application practices continues to be implicated as the primary contributor to stream loading. A major focus of future research should be improvement of drift mitigation measures which prevent the direct input of spray drift into streams and minimize deposition in riparian areas surrounding these aquatic resources.

For more information on the OSU study Contact Jeffrey Jenkins, Department of Environmental and Molecular Toxicology, Oregon State University, jenkinsj@ace.orst.edu.