Results

Pesticide Concentration Data from Ambient Stream Water, 1999–2009

As part of the following discussions of each detected pesticide, the range of detected concentrations is compared to established water-quality standards and mortality and sublethal effect values. Sublethal effects are physiological or behavioral changes that occur to an organism after exposure to a contaminant at a less-than-fatal concentration. In figures 4–22, each sublethal effect value from the literature is represented by a single black square and each no observed effect concentration (NOEC) value is shown by an “X”. Multiple values are shown for a given sublethal endpoint when multiple values were available from the literature and may reflect differences in test species, product formulation, exposure duration, or other variations in experimental design. The established water-quality standards are freshwater values from the ODEQ 2004 water-quality criteria, U.S. Environmental Protection Agency (USEPA) national recommended water-quality criteria, and USEPA Office of Pesticide Programs’ aquatic life benchmarks (appendix D) (Oregon Department of Environmental Quality, 2004; U.S. Environmental Protection Agency, 2005a, 2009b). Toxicity and sublethal endpoint data were obtained from USEPA Reregistration Eligibility Decisions (U.S. Environmental Protection Agency, 1994, 1996, 1997, 1998a-b, 1999, 2003, 2005e, 2006a-g, 2009d), USEPA pesticide fact sheets and ecological risk assessments (U.S. Environmental Protection Agency, 2001, 2005c, 2008, 2009a), National Marine Fisheries Service biological opinions (National Marine Fisheries Service, 2008, 2009), USEPA ECOTOX database (U.S. Environmental Protection Agency, 2007), U.S. Office of Pesticide Programs Pesticide Toxicity database (U.S. Environmental Protection Agency, 2005b), U.S. Geological Survey Columbia Environmental Research Center Acute Toxicity database (U.S. Geological Survey, 2004), pesticide product labels (Gowan Company, 2004; BASF Corporation, 2010), a National Pesticide Information Center fact sheet (National Pesticide Information Center, 2010), and selected peer-reviewed literature (Julin and Sanders, 1977; Spehar and others, 1981; Mayer and Ellersieck, 1988; Sheedy and others, 1991; Beketov and Liess, 2008; Tierney and others, 2010). Fish toxicity and sublethal effect data are for salmonids (Oncorhynchus spp. and Salmo spp.), except for a few specified values for fathead minnows (Pimephales promelas), which were used where no salmonid data exist. Fish toxicity values are for 96-hour tests. Invertebrate endpoints were selected for common toxicity test species that are likely to occur in the Pacific Northwest; for example, various species of stonefly, mosquito, scud, and zooplankton. Invertebrate toxicity test procedures are more variable than those for fish. Exposure durations for invertebrate toxicity tests were generally 24, 48, or 96 hours. This accounts for some of the variability in the salmonid prey toxicity values provided for a given pesticide.

A summary of the use and environmental fate of all detected pesticides is provided in appendix E. A complete list of all pesticides analyzed is provided in appendix F.

Pesticide Occurrence for Pesticides Analyzed 1999–2009

From 1999 through 2009, two herbicides and five insecticides were detected in the basin (table 5). At least one pesticide was detected at 13 of 16 sampling sites. The only sites without a pesticide detection were West Fork Hood River at Moving Falls (RM 2.5), West Fork Hood River at Lost Lake Road (River Mile 4.7), and Dog River below Puppy Creek confluence.

Concentrations at eight sites exceeded the USEPA or Oregon water-quality standards (table 6) for one or more pesticides at least once during the project period. However, because the standards are based on specific exposure durations (acute = 24 hours, chronic = 96 hours), point-in-time samples are not directly comparable to the standards.

Occurrence, Detection Frequency, and Potential Impacts on Aquatic Life for Pesticides Analyzed 1999–2009

Additional information on the uses and environmental fate of each pesticide is provided in appendix C and appendix E.

Atrazine (herbicide)

Atrazine was detected in March 2003 and May 2004 at Lenz Creek at mouth. The detections were more than an order of magnitude (10 times) lower than concentrations known to cause changes in olfactory-mediated behavior and over three orders of magnitude (1,000 times) lower than the lowest aquatic life benchmark (fig. 4).

Azinphos-methyl (insecticide)

Azinphos-methyl was the second most frequently detected pesticide since sampling began in 1999 (n = 76 detections). Most detections occurred at Neal Creek at mouth (47 percent of detections) and Lenz Creek at mouth (43 percent). Most detections occurred in the summer and fall; more than 20 percent of samples collected during August through October had azinphos-methyl detected. During August to October, 53 percent of azinphos-methyl detections were in Lenz Creek; 47 percent were in Neal Creek (Middle or at mouth). Only six detections occurred in March–May, even though more than half of all samples were collected during those months. Twenty-one samples had detectable concentrations of azinphos-methyl oxon, a degradation product of azinphos-methyl. In 18 of those samples, azinphos-methyl and its degradation product were both detected. Detections of the degradation product were generally more common with higher measured concentrations of azinphos-methyl and were most common at Lenz Creek at mouth and Neal Creek at mouth. All detected concentrations of azinphos-methyl exceeded the Oregon chronic criterion (0.01 µg/L) and 37 percent of the detected concentrations exceeded the USEPA benchmark for acute exposures for invertebrates (0.08 µg/L). The highest detected concentrations were less than one order of magnitude lower than concentrations associated with sublethal changes to salmonids and mortality to their prey (fig. 5).

Chlorpyrifos (insecticide)

Chlorpyrifos was detected 66 times since 1999, last in 2008. The reporting limit in 2009 was higher than concentrations detected in 2008, which may account for the absence of detections in 2009. However, in 2009, chlorpyrifos was not detected at a concentration equal to or greater than 0.041 µg/L, the concentration of the lowest water-quality criterion, in 105 samples with reporting limits of 0.041 µg/L or less. The majority of detections were at Neal Creek at mouth (47 percent) and Lenz Creek at mouth (18 percent). Forty-seven (47) of 66 detections occurred in March; the rest were in April. Chlorpyrifos-oxon, a degradation product of chlorpyrifos, was not detected in the 503 samples in which it was analyzed through 2008. Detections of chlorpyrifos were at concentrations exceeding the most stringent USEPA aquatic life benchmarks and at concentrations that can cause harmful effects to salmonids and their prey (fig. 6).

Diazinon (insecticide)

Diazinon was detected once at Neal Creek at mouth in June 2000 and March 2003 and twice in Baldwin Creek in June 2005. The detected concentrations exceeded the lowest USEPA benchmarks for acute and chronic exposures (for invertebrates) and were within a range known to cause invertebrate mortality and changes in salmonid olfaction (fig. 7).

Malathion (insecticide)

Malathion was detected seven times from 2000 through 2004 at concentrations that exceeded the minimum USEPA benchmarks for invertebrates and were within a range known to cause sublethal effects to invertebrates (fig. 8). Detections occurred in June and early July at Lenz Creek at mouth (n = 1) and Neal Creek at mouth (n = 6). Malathion-oxon, a degradation product, was detected in 3 of 460 samples through 2008, twice at Neal Creek at mouth and once at Lenz Creek at mouth. Of those, malathion was detected only in the sample with the highest malathion-oxon concentration.

Phosmet (insecticide)

Phosmet was detected in 14 samples from 2001 through 2008, mostly in Lenz Creek. Most detections (57 percent) occurred in the fall, even though most samples were collected during the spring and summer. The remaining 43 percent of detections were in April–June. Phosmet-oxon, a phosmet degradation product, was not detected in 459 samples in which it was analyzed. Phosmet detections were generally one order of magnitude lower than the minimum USEPA benchmarks, but overlapped the lower range of invertebrate toxicity values (fig. 9).

Simazine (herbicide)

With 157 detections from 1999 through 2009, simazine was the most commonly detected pesticide. Most detections occurred at Lenz Creek at mouth (44 percent of detections) and Neal Creek at mouth (43 percent). The highest concentrations and detection counts were in June, when 43 percent of simazine detections occurred. The maximum concentration was slightly less than a concentration known to cause olfactory changes in Atlantic salmon (fig. 10).

Pesticide Occurrence for Pesticides Analyzed Only in 2009

In 2009, the number of pesticides analyzed increased to 100, compared with 10 in 2008. Because of the increase in the number of analytes, 12 pesticides were detected in 2009 that had not previously been detected in the Hood River basin: 2 fungicides, 4 herbicides, and 6 insecticides. Pesticides were detected at seven of eight sites. No pesticides were detected at West Fork Hood River at Moving Falls (RM 2.5); however, only one sample was collected at this site. All other sites were sampled 15 or 16 times in 2009. Table 7 provides sample counts, detection counts, and maximum detected concentrations for each pesticide detected in 2009 only.

Of the new pesticides detected in 2009, only one, endrin, was detected at a concentration exceeding the USEPA or Oregon water-quality standards (table 8). However, the standards are based on specific exposure durations (acute = 24 hours, chronic = 96 hours), so point-in-time samples are not directly comparable to the standards.

Occurrence, Detection Frequency, and Potential Impacts on Aquatic Life for Each Detected Pesticide

Additional information on the uses and environmental fate of each pesticide is provided in appendix C and appendix E.

Carbaryl (insecticide)

Carbaryl was detected a total of nine times in 2009 in Middle Neal Creek and at the mouths of Neal Creek, Lenz Creek, and Hood River. Eight of the detections occurred in May or June; the other was in October. The highest concentration of carbaryl was more than an order of magnitude less than the lowest USEPA benchmark (0.5 µg/L) and the lowest available ecotoxicological value (1.5 µg/L, for invertebrate reproduction) (fig. 11).

DEET (insecticide)

DEET was detected twice in 2009, once each at West Fork Hood River at mouth in March and Lenz Creek at mouth in April. The detections were nearly seven orders of magnitude less than the acute toxicity values for salmonids (fig. 12). Water-quality standards do not exist for DEET.

Diuron (herbicide)

Diuron was the most frequently detected pesticide in 2009, found in 41 of 104 samples. All detected concentrations were less than the lowest aquatic life benchmarks set by the USEPA (fig. 13). The highest concentrations of diuron were found in Lenz Creek at mouth, followed by Neal Creek at mouth. Concentrations peaked in May at both sites.

Endrin (insecticide)

Endrin was detected once in Rogers Spring Creek in April 2009 at a concentration approximately 1.5 orders of magnitude higher than the ODEQ chronic water-quality criterion and within an order of magnitude less than the USEPA acute water-quality criterion and salmonid toxicity values (fig. 14). Endrin has not been registered for use since 1991, but it is very persistent in soils.

Fluometuron (herbicide)

Fluometuron was detected once in Upper Neal Creek in April 2009 at a concentration four orders of magnitude less than the lowest USEPA benchmark (fig. 15). However, the detection was unexpected given that fluometuron is registered for use only on cotton, a crop not grown in Oregon.

Hexazinone (herbicide)

Hexazinone was detected in 34 of 111 samples in March through June 2009, with the highest concentrations occurring in April and May. All detections except one were from the three Neal Creek sites (Upper below agricultural diversion, Middle, and at mouth). The other detection was in Lenz Creek at mouth. The highest concentration of hexazinone found was more than five orders of magnitude lower than the lowest USEPA benchmark (fig. 16). Hexazinone is a broad-range triazine herbicide, predominantly used in forestry, the dominant upstream land use in the Neal Creek basin. While it is not highly toxic to salmonids, it can harm salmonid habitat by affecting vegetation as far as 100 meters from the application site because of its persistence and mobility in soil and surface waters (Wan and others, 1988).

Imidacloprid (insecticide)

Imidacloprid was detected once at Lenz Creek at mouth in October 2009. Its concentration was more than an order of magnitude lower than the lowest USEPA benchmark (fig. 17).

Methomyl (insecticide)

Methomyl was detected at four sites on April 29, 2009: at Upper Neal and Rogers Spring Creeks, and at more dilute concentrations at Middle Neal Creek and at Lenz Creek at mouth. All detections were several orders of magnitude lower than USEPA benchmarks or concentrations known to induce sublethal responses in salmonids or their prey (fig. 18).

Norflurazon (herbicide)

Norflurazon was detected once in March 2009 in Lenz Creek at mouth at a concentration several orders of magnitude less than salmonid toxicity endpoints and USEPA benchmarks (fig. 19).

Propiconazole (fungicide)

Propiconazole was detected six times in 2009; all concentrations were much less than the lowest water-quality benchmarks (95 and 425 µg/L) (fig. 20). It was detected at the three Neal Creek sites (Upper below agricultural diversion, Middle, and at mouth) in March, May, and October, with generally higher concentrations detected at the upstream sites.

Propoxur (insecticide)

Propoxur was detected in Middle Neal, Upper Neal, Lenz, and Rogers Spring Creeks on April 29, 2009. Detected concentrations were several orders of magnitude less than those known to harm salmonids or their prey (fig. 21).

Pyraclostrobin (fungicide)

Pyraclostrobin was detected a total of seven times in 2009: at the mouths of Neal Creek, Lenz Creek, Hood River, and West Fork Hood River and in Middle Neal Creek. Detections occurred in April, except for one in May at Lenz Creek at mouth. All detected concentrations were less than acute toxicity values for salmonids and invertebrates (fig. 22). No State or USEPA benchmarks exist for pyraclostrobin.

Trends in Pesticide Detections and Concentrations

Subsets of the 1999–2009 dataset from the mouths of Neal Creek, Lenz Creek, and Hood River were used to examine long-term trends throughout the basin. Only one site, Neal Creek at mouth, was monitored every year from 1999 through 2009 (162 samples). Lenz Creek at mouth was monitored from 2001 through 2006 and 2008 through 2009 (114 samples). The mouth of Hood River was monitored during 1999–2002 and 2005–2009 (90 samples). Data were screened in order to draw comparisons across years with different reporting limits (refer to the Methods section for more information on data screening). Appendix G contains screened sample counts and detections by site for all pesticides.

Neal Creek at Mouth

Azinphos-methyl, chlorpyrifos, and simazine were the most frequently detected pesticides at Neal Creek at mouth. Azinphos-methyl was detected mostly in the summer and fall and has not been detected since 2007 (fig. 23). About half of azinphos-methyl samples from this site in 1999, 2008, and 2009 were removed during the data screening process due to high reporting limits. Chlorpyrifos was detected during March, except one detection in April, with the last detections in 2005 (fig. 24). However, fewer samples collected in March and April were analyzed for chlorpyrifos after 2005 compared to previous years. All chlorpyrifos data at this site for 2009 were removed from the screened dataset due to the high reporting limit. However, chlorpyrifos was not detected in 14 samples in 2009 at or exceeding the lowest water-quality criterion (0.041 µg/L). The highest detection counts and concentrations of simazine occurred during the summer months, but simazine was also detected in the spring. It was detected every year except 2008, with general downward trends in the percentage of samples with detections and in annual maximum concentration (fig. 25). From 1999 through 2006, simazine was detected in 43 percent of samples at Neal Creek at mouth. From 2007 through 2009, it was detected in 14 percent of samples.

Lenz Creek at Mouth

Azinphos-methyl, chlorpyrifos, phosmet, and simazine were the most frequently detected pesticides at Lenz Creek at mouth since 2001. Azinphos-methyl detections were nearly evenly split between summer and fall. Azinphos-methyl was found in 0 to 70 percent of samples from 2001 through 2005 (fig. 26), when the annual sample counts ranged from 16 to 20. Only three samples were collected in 2006. On average, it was present in 32 percent of samples from 2001 through 2006 and 0 percent of samples since 2008 (there are no samples from 2007). However, many azinphos-methyl samples from 2008–09 were screened out.

Chlorpyrifos was detected one to four times in March during each year from 2001 through 2005 and has not been detected since (fig. 27). However, no chlorpyrifos samples were analyzed in March in 2006–2008. Moreover, there were 87 samples from 2001 to 2005 and only 10 since 2006 due to a decrease in total sample counts in 2006 and 2008, the complete absence of samples in 2007, and the censoring of 2009 data due to the high reporting limit.

Phosmet was detected one to three times per year from 2002 through 2005 (fig. 28), representing 17 percent or less of yearly samples. Overall, it was detected in 7.7 percent of samples through 2006 and in 4.3 percent of samples since 2007. It was detected mostly during the fall. The absence of samples collected during fall after 2005 may have caused the nearly complete absence of detections since then.

Simazine was most frequently detected and was found at the highest concentrations in Lenz Creek during the summer. It was detected in at least 28 percent of screened samples each year in 2001–06 and 2008–09 (fig. 29). Three to 20 samples were collected per year. Prior to 2007, simazine was detected in 61 percent of screened samples from Lenz Creek. Since 2008, it was detected in 53 percent of samples. Many simazine samples from 2008 to 2009 were removed because they had a higher reporting limit than the censoring level (0.027 µg/L).

Hood River at Mouth

The mouth of Hood River was monitored from 1999 through 2002 and 2005 through 2009. Samples were collected from two sites at the mouth of Hood River: Hood River downstream of Ppl Powerdale Powerhouse (1999-2001) and Hood River at footbridge downstream of I-84 (2002 and 2005–2009). Pesticides were detected in fewer samples at the mouth of Hood River than at Lenz Creek at mouth or Neal Creek at mouth. Chlorpyrifos was detected once each in 1999 and 2001 (fig. 30). Approximately twice as many samples were analyzed for chlorpyrifos each year after 2001 compared to before 2001, so the absence of detections was not caused by the number of samples collected. It was found in 4 percent of 45 samples through 2006 and 0 percent of 23 samples in 2007–08. Chlorpyrifos data from 2009 were screened out due to the higher reporting limit. Simazine was detected in 2 of 4 samples in 2000, 2 of 15 samples in 2005, and 2 of 9 samples in 2009 (fig. 31). Four of the six samples with detections were collected in May or June. Simazine was found in 4 percent of 47 samples through 2006 and 6 percent of 32 samples since 2007.

Pesticide Mixtures

Of the 953 pesticide unscreened samples from 1999 through 2009, 116 (12 percent) had multiple pesticides detected in the same sample (table 9). Simazine was the pesticide most commonly detected with other pesticides in the same sample (79 percent of mixture samples). Azinphos-methyl was found in 43 percent of mixture samples, followed by diuron (31 percent), chlorpyrifos (20 percent), and hexazinone (13 percent). The most commonly detected pesticide pair was azinphos-methyl and simazine (n = 47). Most pesticide mixture samples were collected from Neal Creek at mouth and Lenz Creek at mouth (45 and 38 percent, respectively). Most mixtures were detected in 2009, when the list of pesticides analyzed was much larger than in previous years. The second-highest number of pesticide mixtures was detected in 2003, the year with the highest sample count from Lenz Creek at mouth and Neal Creek at mouth. The highest number of mixture samples was collected in June (n = 36). October had the highest proportion of samples collected with mixtures present, possibly driven by the low number of samples collected in October.

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

Four of 10 pesticides were detected from the 2004–05 special study on effluent from fruit packers. The number of detections and minimum, median, and maximum detected concentrations of the four detected pesticides at each site are shown in table 10. Simazine was the only pesticide that was detected at concentrations less than the lowest water-quality standard (table 11). Azinphos-methyl was the most commonly detected pesticide in these samples and it occurred in the highest concentrations, with more than half of samples exceeding the most stringent USEPA or ODEQ standard. Its degradation product, azinphos-methyl oxon, was detected in 21 of 50 samples at concentrations of 0.05–10 µg/L, with higher concentrations in fruit packing effluent and surface waters downstream of effluent discharge points. Malathion was detected once in Lenz Creek in September 2005 at a concentration exceeding the lowest acute and chronic USEPA standards, which are set to protect invertebrates (table 11). Two samples exceeded the minimum water-quality criterion for phosmet. Malathion- and phosmet-oxons were not detected.

Trace Element Concentration Data, 1999–2009

A summary of available data for priority trace elements is shown in table 12, along with USEPA and Oregon water-quality standards and values from published studies on salmonid exposure to trace elements. Table 13 shows the same information for nonpriority trace elements.

Based on total recoverable trace element concentration data from 2000 through 2002 at the nine most frequently sampled sites (n = 13–39 per site), aluminum, copper, zinc, and nickel were the most commonly detected priority trace elements. The percentage of samples with those trace elements detected was highest at the mouths of Lenz Creek, Neal Creek, and Hood River and at Evans Creek at bridge. Middle Fork Hood River, West Fork Hood River, and Dog Creek had the lowest incidence of priority trace element detections from 2000 through 2002. Aluminum was detected in 100 percent of samples at all frequently sampled sites except Middle Fork Hood River and West Fork Hood River. Total recoverable copper and nickel were frequently detected at Evans Creek at bridge and the mouths of Hood River, Lenz Creek, and Neal Creek. Copper was also commonly detected at East Fork Hood River, and nickel was frequently detected at Upper Neal Creek. Zinc was detected with the same spatial pattern as nickel, but generally with lower detection frequencies. The sources of these elements in the environment could not be assessed from the available data.

First posted June 17, 2011