Methods

Concentrations of pesticides and trace elements used for this report were obtained from the ODEQ’s Laboratory Analytical Storage and Retrieval (LASAR) database (Oregon Department of Environmental Quality, 2008). Only primary (not quality assurance) samples were analyzed. The data were retrieved on August 9, 2010. The pesticide data were organized into two datasets. The trace element data were organized into a third dataset. Each is briefly described below. Common chemical names for pesticide active ingredients are used in this report and differ in some cases from the names used in the ODEQ’s database. The ODEQ’s database includes some pesticide product names and alternate spellings of chemical names, Baygon for propoxur, Imidan for phosmet, Guthion for azinphos-methyl, oxygen analog for oxon, and chlorpyriphos or Dursban for chlorpyrifos, for example.

Pesticide Concentration Data, Ambient Stream Water, 1999–2009

Data were collected at 16 sites in the Hood River basin from 1999 through 2009. Catchment area and land use information for those sites are provided in appendix A. Sample counts for all sites are shown in table 1. Nine hundred fifty-three (953) surface-water pesticide samples were collected between 1999 and 2009. Most samples were collected during the spring and summer, to coincide with orchard pesticide application in the basin (78 percent of samples were collected March–June). Data from 1999 through 2002 came from a Portland State University and ODEQ study on the effects of instream exposure to pesticides on threatened steelhead in the Hood River basin (Eugene Foster, Portland State University, written commun., 2003; Oregon Department of Environmental Quality, 2008). That study examined nine pesticides at eight sites. The remaining water-quality data from routinely monitored sites came from the Hood River Pesticide Stewardship Partnership (HRPSP) project, which included as many as 10 organophosphate insecticides, 4 organophosphate degradation products, and 2 herbicides at a subset of the 16 sites through 2008. In 2009, an expanded list of 100 pesticides was analyzed for 8 of those sites to account for changes in pesticides used in the basin. Appendix B shows the number of samples collected at each site during each month and year.

Pesticide Concentration Data, Special Study on Effluent from Fruit Packers, 2004–2005

Fifty pesticide samples from 37 surface-water sites in the Hood River basin were collected in 2004 (n = 12) and 2005 (n = 38) for the National Pollutant Discharge Elimination System permit monitoring project for fruit packing facilities. Ten pesticides were analyzed in that study. Samples from 13 sites are from effluent from packing plants, and the remaining samples are from stream water. Sample counts by site are shown in table 1, with the receiving stream for effluent sites listed. Surface-water samples were collected by ODEQ staff at, upstream, and downstream of fruit-packing-plant discharge sites.

Trace Element Concentration Data, 1999–2009

Two hundred fifty-five (255) surface-water samples collected from 53 sites in the Hood River basin during 1999–2002 and 2009 were analyzed for trace element concentrations. These data came from the ODEQ’s LASAR database, although they were not all collected as part of the HRPSP project. Twenty-seven trace elements were analyzed. Most sites were sampled once; some had as many as 40 samples.

Reporting Limits and Data Screening

The reporting limit (RL) is the value at which a laboratory reports a concentration as undetected. The value is said to be censored at that RL. No determination can be made about the magnitude of a concentration less than the RL; it might be slightly less than the RL or it might be zero. The RL is based on the laboratory’s analysis of several types of quality control samples, including blanks, replicates, matrix spikes, and surrogate spikes. A laboratory’s ability to quantify a concentration often changes over time as a result of changing laboratory techniques, equipment, and analysts, and also because of sample-to-sample differences in water chemistry. As a result, RLs change over time.

From 1999 through 2009, RLs for all pesticides varied among and within years. Figures 2 and 3 show reporting limits by year for azinphos-methyl and simazine, respectively, to exemplify the variability in reporting limits in this dataset. Comparing samples with different reporting limits can misrepresent the frequency and distribution of occurrence. Pesticides may seem to be more widely distributed or to be detected more frequently during periods of time with lower RLs compared to periods with higher RLs.

To address the issue of multiple RLs, the pesticide data from 1999 through 2009 were screened for some analyses in this report. Often, the screening level is set equal to the highest RL in a dataset. This technique would have resulted in the loss of a large amount of information in this dataset. Instead, the screening level was set to minimize the loss of data. The screening level for each pesticide is shown in table 2. Positive detections less than the screening level were censored at the screening level. Censored values less than the screening level were recensored at the screening level. Data censored at a RL greater than the screening level were removed from the dataset. Similarly, positive detections collected during periods when the RL was greater than the screening level also were removed to avoid biasing the data. Four hundred four (404) data points were removed from the screened dataset. Of those, only 2 were detected concentrations.

Screened and unscreened data were used in different sections of this report. Each section or analysis includes a description indicating which data were used.

Pesticide Use

Watershed-level pesticide-use information is difficult to obtain. The mid-Columbia basin, which includes the Hood River basin, had the highest reported amount (by weight) of pesticides applied across Oregon in 2007 and 2008 (55 and 38 percent of the total amount reported statewide, respectively) (Oregon Department of Agriculture, 2008, 2009). The high amount of pesticides used in the mid-Columbia basin was driven by the heavy use of metam sodium, a soil fumigant (Oregon Department of Agriculture, 2008, 2009). Soil fumigants often account for a large proportion of pesticide use when reported by weight due to their high application rates relative to other types of pesticides and are more heavily used on crops grown elsewhere in the mid-Columbia basin than on orchards (U.S. Environmental Protection Agency, 2005d). Total acreage treated with chemicals to control insects, weeds, grass, brush, or diseases in crops and orchards, and to control growth, thin fruit, ripen, or defoliate crops in Hood River County increased from 2002 to 2007 (U.S. Department of Agriculture, 2007). Information on pesticides registered and deemed appropriate for use by crop are available from Oregon State University Extension Service’s 2009 Integrated Pest Management Handbooks for Weeds, Insects, and Diseases (Hollingsworth, 2009; Peachey, 2009; Pscheidt and Ocamb, 2009) and 2010 Pest Management Guide for Tree Fruits in the mid-Columbia Area (Oregon State University Extension Service, 2010). Appendix C lists pesticides considered suitable for the major land uses of the Hood River basin.

Agriculture

Table 3 lists pesticides known to be commonly used on agricultural crops in the Hood River basin at the time of the writing of this report (Steve Castagnoli, Oregon State University Extension Service, oral commun., 2010). The list includes herbicides for weed control, insecticides (including miticides and acaracides) for arthropod control, and fungicides to control various blights, rusts, and molds. Numerous others are registered for use on crops grown in the Hood River basin (appendix C). Most of the pesticides in table 3 and appendix C were not analyzed for this project, which focused on organophosphates through 2008 due to the existence of State water-quality standards for those pesticides. Agricultural pesticide use in the basin varies across years in response to changes in pest occurrence and to reduce the potential for pesticide resistance.

Forestry

Herbicides are the most common class of pesticides used in forests. Overall, herbicide use fluctuates across sites and years in order to meet localized needs (Doug Thiesies, Oregon Department of Forestry, oral commun., 2010). Forestry herbicides commonly used in the Hood River basin are listed in table 4. Sulfometuron methyl, glyphosate, and 2,4-D are commonly used in the fall for site preparation (Doug Thiesies, Oregon Department of Forestry, oral commun., 2010); however, only 2,4-D was analyzed in this project. In the Pacific Northwest, 2,4-D, glyphosate, imazapyr, picloram, or triclopyr are used for nearly all brush and weed tree control, although other pesticides are registered for this purpose (Peachey, 2009). Hexazinone is used in the spring at the time of planting (Doug Thiesies, Oregon Department of Forestry, oral commun., 2011) and was analyzed in this project in 2009. Insecticide use is rare, although widespread applications may occur in response to the outbreak of a specific pest (Doug Thiesies, Oregon Department of Forestry, oral commun., 2010).

Rights-of-Way

Bromacil, 2,4-D ester, diuron, glyphosate, sulfometuron methyl, and triclopyr are known to have been used for weed and brush control adjacent to roads or irrigation canals since 2009 (Brian Walker, Oregon Department of Transportation, oral commun., 2010; John Buckley, East Fork Irrigation District, oral commun., 2010; Nate Lain, Hood River County Weed and Pest Division, oral commun., 2010). 2,4-D, bromacil, diuron, and triclopyr were analyzed for this project in 2009.

Household Use

Household use of pesticides is difficult to assess. The Oregon Department of Agriculture’s Pesticide Use Reporting System (PURS) includes an annual survey of about 1,500 Oregon households regarding pesticide use. However, PURS has been suspended since 2009 due to budget constraints. Combined data for Hood River and Wasco counties indicate that herbicides were the primary class of pesticides used in households in 2007 and 2008 (99.4 and 81.3 percent, respectively) (Oregon Department of Agriculture, 2008, 2009). Statewide, glyphosate and 2,4-D accounted for approximately 78 percent of reported household herbicide use in 2007 and 2008. Fipronil and S-methoprene (2007) and malathion (2008) were the insecticides with the highest reported use statewide. Both years, DEET (N,N-Diethyl-meta-toluamide) was by far the most common form of insect repellent used. Calcium polysulfide and Captan accounted for 62 to 74 percent of reported fungicide use in households.

First posted June 17, 2011