SEARCH

SEARCH BY CITATION

Keywords:

  • Inhibition of acetylcholinesterase activity;
  • Carbamate;
  • Organophosphate;
  • Pacific salmon;
  • Pesticide mixtures

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

Salmon populations in the Pacific Northwest are being affected by a variety of environmental stressors including intense fishing pressure, parasites and disease, climatic variability and change, land development, hatchery production, hydropower operations, stormwater runoff, and exposure to toxic contaminants. In recent years, there has been much concern that mixtures of pesticides are causing toxic effects to Pacific salmon. In this study, we compared measured stream water concentrations from 2 monitoring studies conducted in the Pacific Northwest with concentration-response curves derived for inhibition of brain acetylcholinesterase activity in juvenile coho salmon (Oncorhynchus kisutch) for mixtures of organophosphate (OPs) and carbamate (CBs) pesticides. In the first monitoring study, samples were collected from 2003 to 2007 in salmonid-bearing waters of 5 urban or agricultural watersheds in Washington State. This study was targeted to areas of high pesticide use and generally involved weekly sampling during the pesticide use season. The second monitoring study was the United States Geological Survey National Water Quality Assessment that included samples taken from 2003 to 2010 in California, Idaho, Oregon, and Washington. OPs and CBs were frequently detected in both studies. The available monitoring data collected since 2003, however, demonstrates that mixtures of OPs and CBs in surface waters rarely occur at levels capable of producing significant physiological and behavioral effects in Pacific salmon. The observed mixtures never reached concentrations capable of causing mortality. We conclude that mixtures of organophosphates and carbamates do not pose a significant direct risk to Pacific salmon. Integr Environ Assess Manag 2013; 9: 70–78. © 2012 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

More than 450 million kilograms of conventional pesticides are applied annually in the United States, which has led to the frequent occurrence of mixtures of pesticides in the environment (Gilliom et al. 2006). In the United States, the US Environmental Protection Agency (USEPA) registers pesticides under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and the Federal Food, Drug, and Cosmetic Act. The registration process includes human health and ecological risk assessments. The FIFRA process does not require ecological risk assessments of pesticide mixtures or any combination of multiple stressors. Similarly, under the Endangered Species Act (ESA), pesticides are evaluated individually. Given the widespread occurrence of pesticide mixtures in the environment, there is concern that the FIFRA and ESA processes are overlooking the potential for significant cumulative toxicological impacts of pesticide mixtures on aquatic life and wildlife (Belden et al. 2007; Laetz et al. 2009; Relyea 2009).

In this article, we present a case study on the effects of pesticide mixtures to salmon in the Pacific Northwest. Salmon populations are faced with many environmental stressors including intense fishing pressure, parasites and disease, climatic variability and change, land development, hatchery production, hydropower operations, stormwater runoff, and others (NMFS 2010; Spromberg and Scholz 2011). The potential for toxic effects to Pacific salmon from exposure to pesticide mixtures has been the subject of serious debate in recent years (e.g., Scholz et al. 2006; Laetz et al. 2009; NMFS 2010; Lisker et al. 2011).

The pesticides of greatest concern for direct effects to Pacific salmon are carbamates and organophosphates. Both classes of pesticides are highly toxic to salmon (Laetz et al. 2009; NOAA 2009) and are frequently detected in salmonid-bearing streams in the Pacific Northwest (Anderson et al. 2004, 2007; Burke et al. 2005, 2006; Anderson and Dugger 2008; Lisker et al. 2011). They share the same mode of action, i.e., inhibition of brain and plasma acetylcholinesterase (AChE) activity, and thus would be expected to cause additive toxicity to fish (Deneer 2000; Kudsk et al. 2005; Belden et al. 2007). In this article, the measured stream water concentrations from monitoring studies conducted in the Pacific Northwest are compared to concentration–response curves derived for inhibition of brain acetylcholinesterase activity in juvenile coho salmon (O. kisutch) for mixtures of organophosphate and carbamate pesticides (Laetz et al. 2009; NOAA 2009). The pesticides for which Laetz et al. (2009) and NOAA (2009) derived concentration–response curves include a range of carbamate (e.g., carbaryl, carbofuran) and organophosphate (e.g., azinphos-methyl, chlorpyrifos, diazinon, dimethoate, ethoprop, malathion, methidathion, methyl parathion, naled, phorate, and phosmet) pesticides. Two monitoring studies are used to estimate exposure of salmon to mixtures of carbamates and organophosphate pesticides: the Washington State monitoring study and the United States Geological Survey (USGS) National Water Quality Assessment monitoring study for the Pacific Northwest.

THE WASHINGTON STATE MONITORING STUDY

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

The Washington State Department of Agriculture (WSDA) and the Washington State Department of Ecology have been conducting a multiyear monitoring study to characterize pesticide concentrations in selected salmonid-bearing streams in Washington State (Anderson et al. 2004, 2007; Burke et al. 2005, 2006; Anderson and Dugger 2008). Samples have been typically collected weekly during the pesticide-use season (February to September) and 5 years of data (2003–2007) were available when the analysis for this article was conducted. Five water resource inventory areas (WRIA) have been included in the monitoring study to date (Figure 1). Each of the WRIAs supports salmon populations and has a high proportion of agricultural and/or urban land area.

thumbnail image

Figure 1. Water Resource Inventory Areas (WRIAs) and sample locations in the Washington State monitoring program (2003–2007).

Download figure to PowerPoint

Thornton Creek in the Cedar-Sammamish WRIA was selected as the urban water body. The remaining WRIAs represent the range of agricultural practices in Washington State: Lower Skagit-Samish (western Washington land-use practices), Lower Yakima (eastern Washington irrigated land-use practices), and Wenatchee and Entiat (both represent central Washington tree-fruit practices). Because the Washington State monitoring study is comprehensive, long-term, and focused on salmonid-bearing streams in areas of high pesticide usage, we chose to use this data set as 1 of the 2 data sets to estimate exposure of Pacific salmonids to mixtures of pesticides.

The Washington State monitoring study grouped pesticides into 8 categories: herbicides, organochlorines, carbamates, organophosphates, pyrethroids, sulfite esters, wood preservatives, and degradates. Of these, the sulfite esters, pyrethroids, and organochlorines were rarely detected from 2003 to 2007 (≤1% samples). The only wood preservative detected was pentachlorophenol (6% of samples) and only at levels that were orders of magnitude below the EPA ambient water quality criterion (AWQC) for chronic exposure (AWQC = 2.7 to 111 µg/L for the pH range of 6.1 to 9.8 observed in Washington State WRIAs in 2007; Anderson and Dugger 2008). Herbicides such as 2,4-D, dichlobenil and atrazine were commonly detected from 2003 to 2007 in most sampling locations. Although herbicides could indirectly affect salmonids via effects to algae and plants, direct effects to fish from these pesticides are much less likely. In contrast to the other pesticide groups, carbamates and organophosphate pesticides are highly toxic to salmonids (Beauvais et al. 2000, 2001; Sandahl and Jenkins 2002; Sandahl et al. 2005; Laetz et al. 2009; NOAA 2009) and were frequently detected in the Washington State WRIAs from 2003 to 2007. Thus, carbamate and organophosphate pesticides are the pesticides of greatest concern for direct effects to Pacific salmonids. Examples of carbamates that were detected in the Washington State watersheds include carbaryl, carbofuran, oxamyl, and methomyl (Anderson et al. 2004, 2007; Burke et al. 2005, 2006; Anderson and Dugger 2008). Chlorpyrifos, diazinon, azinphos methyl, and malathion are examples of organophosphates detected in the monitoring study (Anderson et al. 2004, 2007; Burke et al. 2005, 2006; Anderson and Dugger 2008).

The Washington State, Department of Ecology, Environmental Information Management (EIM) system (WSDE 2009) was queried to isolate data from the Washington State monitoring study. The EIM contains records on physical, chemical, and biological analyses and measurements. Supplemental information about the data (metadata) is also available, including information about study protocols, monitoring locations, and data quality.

Data were first identified by querying the EIM for all available data from each of the monitoring years. Unique samples were identified by location and sample date. The data files for the first 5 years of the monitoring program were downloaded from the EIM (WSDE 2009). Each download consisted of 3 files: 1) study identification information, 2) monitoring station location information, and 3) monitoring results for that year. A total of 258 unique pesticides and degradates were identified in the 5 study years. Each of these unique products was subsequently identified through literature searches or other means as either carbamates (C) or organophosphates (OP) or other.

NAWQA DATA SET FOR THE PACIFIC NORTHWEST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

The USGS National Water Quality Assessment (NAWQA) program is a long-term national monitoring study on flowing water and groundwater. The program began in 1991 and uses a nationally consistent study design and methods where possible to facilitate comparison of results. For the analyses conducted herein, data for 2003 to 2010 were isolated as of June 30, 2010. Both location and date of the sample were used to isolate data. Only NAWQA data for the states containing evolutionarily significant units (ESU) and distinct population segments (DPS) for Pacific salmon were queried. These included the following NAWQA study units: Sacramento and San Joaquin-Tulare River Basins, California; Great Salt Lake Basins and Upper Snake River Basin, Idaho; Willamette Basin, Oregon; and Central Columbia Plateau and Yakima River Basin, Washington. A total of 17, 5, 26, and 30 unique sampling stations in California, Idaho, Oregon, and Washington State, respectively, were found.

NAWQA data are accessible through the web interface at the USGS NAWQA data warehouse using the following link: http://infotrek.er.usgs.gov/traverse/f?p=NAWQA:HOME:0. Data for 2003 to 2010 were isolated using the web-based interface. Data from the surface water and bed sediments category were selected. The surface water and bed sediments category provides water quality results from NAWQA surface water sites (USGS 2010). All results used were from filtered water samples.

As of June 30, 2010, a total of 16 467 records were isolated using these queries. The raw NAWQA data were imported into MS-Access© 2007 to facilitate organization and subsequent queries. An analysis of the data indicated that there were approximately 161 unique pesticides, and others that were volatile organic compounds (VOCs). Raw data were filtered using the USGS parameters code definition file USGS (2009) to isolate data for the carbamates and organophosphates. Carbamates and organophosphates were frequently detected in surface water samples though less commonly than in the more targeted Washington State monitoring study.

TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

The primary mode of toxicity for carbamates and organophosphates is inhibition of cholinesterase (ChE) activity. Inhibition of the enzyme acetylcholinesterase (AChE) activity results in the buildup of acetylcholine (ACh) at cholinergic nerve endings causing continual nerve stimulation (Scholz et al. 2006). Inhibition of AChE activity occurs in the central nervous system (CNS) and in the plasma and muscle of fish (Perkins and Schlenk 2000). Studies with a number of carbamate and organophosphate pesticides (e.g., azinphos-methyl, chlorpyrifos, malathion, diazinon, carbaryl) have demonstrated an association between inhibition of brain AChE activity and behavioral endpoints such as reduction in spontaneous swimming and feeding rates, swimming stamina, predator detection, and homing behavior (Post and Leasure 1974; Moore and Waring 1996; Van Dolah et al. 1997; Beauvais et al. 2000, 2001; Scholz et al. 2000; Sandahl et al. 2005; Labenia et al. 2007).

The most sensitive effect reported to date was reduced swimming stamina for coho salmon when brain AChE activity was inhibited by 22.8% after exposure to 0.6 µg/L of chlorpyrifos (Sandahl et al. 2005). Adverse effects on physiology and behavior can result in reductions of somatic growth and, as a result, size-dependent mortality in juvenile Chinook salmon (Oncorhynchus tshawytscha) (Baldwin et al. 2009). Studies with estuarine fish have suggested that inhibition of brain AChE activity of greater than 50% may be associated with mortality, though some species of fish appear capable of tolerating greater than 90% inhibition of brain AChE activity (Fulton and Key 2001; Sandahl et al. 2005).

Although carbamates and organophosphates share a common mode of toxic action, the latter bind the AChE receptor irreversibly whereas the former do not. Thus, recovery from exposure to organophosphates requires the synthesis of new enzyme and can take up to several weeks (Ferrari et al. 2004). Recovery from exposure to carbamates is rapid and occurs on a timescale of hours (Ferrari et al. 2004; Labenia et al. 2007). As a result of these and other differences in the pharmaco- and toxicokinetics of carbamates and organophosphates, the USEPA conducted separate cumulative risk assessments for these groups of compounds for human health (USEPA 2006, 2007). Thus, the assumption of concentration additivity used in this assessment of risks of carbamates and organophosphates to salmon (see below) may be somewhat conservative.

The time course for recovery from inhibition of AChE activity may be longer than the time course for recovery from physiological or behavioral effects. For example, signs of toxicity of mosquito fish ceased even when brain AChE activity was inhibited by more than 70% (Boone and Chambers 1996).

ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

Several studies have shown that the combined effects of pesticides sharing a common mode of toxic action are best predicted by a concentration addition (CA) model (Faust et al. 2000; Deneer 2000; Kudsk et al. 2005; Belden et al. 2007). According to the CA model, pesticide concentrations are first normalized for potency and then the total concentration of the mixture is calculated and used to predict mixture toxicity (Faust et al. 2000). In this article, we used the CA model to estimate the combined toxicity of mixtures of carbamate and organophosphate pesticides occurring in salmonid-bearing waters in Washington State, and other states in the Pacific Northwest.

Laetz et al. (2009) conducted a study to determine the effects of binary mixtures of organophosphate and carbamate pesticides on inhibition of brain AChE activity in juvenile coho salmon. Their study included all individual and pairwise combinations of the organophosphate pesticides diazinon, malathion and chlorpyrifos and the carbamate pesticides carbaryl and carbofuran. Using methods similar to those of Laetz et al. (2009), NOAA (2009) derived concentration-response curves for 9 additional organophosphate pesticides for inhibition of brain AChE activity in juvenile coho salmon: azinphos-methyl, dimethoate, disulfoton, ethoprop, methidathion, methyl parathion, naled, phorate, and phosmet. NOAA (2009) tested 3 other organophosphate pesticides (bensulide, fenamiphos, methamidophos) but these pesticides did not have concentration–response relationships in the range of concentrations tested.

The 14 pesticides in the Laetz et al. (2009) and the NOAA (2009) studies accounted for 94.4% of the carbamate (CB) and organophosphate (OP) pesticides detected in the Washington State monitoring study from 2003 to 2007 (Table 1). Similarly, the 14 pesticides of interest accounted for 99.5% of the CB and OP pesticides detected in the NAWQA monitoring study in the Pacific Northwest (Table 2). Data on the inhibition of brain AChE activity of other CB and OP pesticides detected in the Washington State and NAWQA monitoring studies (e.g., aldicarb, methomyl, azinphos-ethyl) in salmonids are lacking. Given that the pesticides used in the Laetz et al. (2009) and NOAA (2009) studies accounted for the vast majority of detected CBs and OPs in the Washington State and NAWQA monitoring studies, we believe that mixture toxicity is not being significantly underestimated due to lack of toxicity data for some CBs and OPs.

Table 1. Percent of total carbamate and organophosphate pesticide detections in the Washington State monitoring study for each of the 14 pesticides of interest
PesticidePercent of total carbamate and organophosphate detections
20032004200520062007Average
  1. OP = organophosphate.

Azinphos-methyl11.611.021.411.35.112.1
Carbaryl2.1005.023.96.2
Carbofuran00001.40.3
Chlorpyrifos32.637.035.753.834.838.8
Diazinon29.59.612.512.55.814.0
Dimethoate12.66.801.31.44.4
Disulfoton02.70000.5
Ethoprop1.19.67.12.51.44.3
Malathion7.423.323.27.56.513.6
Methidathion000000
Methyl parathion000000
Naled000000
Phorate000000
Phosmet1.100000.2
Total percent97.910010093.880.394.4
Nr detections for all carbamates and OPs1261027578138104
Table 2. Percent of total carbamate and organophosphate pesticide detections in NAWQA database for each of the 14 pesticides of interest in California, Idaho, Oregon, and Washington
PesticidePercent of total carbamate and organophosphate detections
200220032004200520062007200820092010Average
  1. OP = organophosphate.

Azinphos-methyl3.39.29.02.8000.6002.8
Carbaryl24.627.024.929.226.531.324.041.130.928.8
Carbofuran8.24.82.83.93.12.00.6002.8
Chlorpyrifos24.620.123.625.334.729.333.037.032.728.9
Diazinon24.620.422.118.516.319.019.66.820.018.6
Dimethoate3.33.00.93.95.13.45.011.05.54.6
Disulfoton00.500000.6000.1
Ethoprop4.96.24.27.910.210.27.82.77.36.8
Malathion5.75.53.96.22.04.15.603.64.1
Methidathion0.83.03.31.11.003.41.401.6
Methyl parathion00.50.41.11.00.70000.4
Naled0000000000
Phorate0000000000
Phosmet0000000000
Total percent10010095.210010010010010010099.5
Nr detections for all carbamates and OPs1224544561781041521798288171

For each sample taken in the Washington State and NAWQA monitoring studies, we estimated inhibition of brain AChE activity for juvenile coho salmon exposed to the organophosphate and carbamate pesticides of interest via the following steps.

For each detected pesticide, percent inhibition of brain AChE activity was estimated using the following general logistic equation:

  • equation image(1)

where Yi is percent inhibition of brain AChE activity caused by pesticide i, Xi is sample concentration (µg/L) for pesticide i, and the EC50i and HillSlopei are fitted parameters from the studies by Laetz et al. (2009) and NOAA (2009) on juvenile coho salmon for pesticide i. The variable HillSlope describes the steepness of the concentration–response relationship. Separate fitted equations were developed by Laetz et al. (2009) for carbaryl, carbofuran, chlorpyrifos, diazinon, and malathion and by NOAA (2009) for azinphos-methyl, dimethoate, disulfoton, ethoprop, methidathion, methyl parathion, naled, phorate, and phosmet (see Table 3 for parameter values).

Table 3. Parameters of the logistic concentration-response models for juvenile coho salmon (O. kisutch) exposed to individual carbamates and organophosphatesa
PesticideEC50 (µg/L)HillSlope
  • a

    Taken from Laetz et al. 2009 and NOAA 2009.

Azinphos-methyl0.16−1.9
Carbaryl146−0.81
Carbofuran58.4−0.82
Chlorpyrifos2.0−1.5
Diazinon145−0.79
Dimethoate273−0.86
Disulfoton488−0.32
Ethoprop90.6−1.3
Malathion74.5−1.32
Methidathion1.1−0.92
Methyl parathion28.8−0.7
Naled7.8−1.3
Phorate0.57−1.6
Phosmet3.3−1.0

When a sample had 2 or more detections of the pesticides of interest, joint toxicity to inhibition of brain AChE activity was estimated by: 1) normalizing the concentration of each detected pesticide to its respective EC50, and 2) summing the EC50-normalized concentrations for the detected pesticides using the fitted equation from Laetz et al. (2009):

  • equation image(2)

where Ymixture is the joint toxicity of the detected pesticides, Xi is the concentration of pesticide i in the sample, EC50i is the concentration causing a 50% decrease in brain AChE activity relative to carrier controls, and n is the number of detected pesticides for the OPs and CBs of interest. Equation 2 assumes concentration additivity.

For duplicate and triplicate samples, concentrations were averaged for the above calculations except for cases where the concentration in 1 or more of the samples was below the detection limit. In the latter cases, only the detected values were used to estimate inhibition of brain AChE activity in juvenile coho salmon.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

The monitoring data from Washington State indicate that brain AChE activity in Pacific salmon could potentially be inhibited by more than 20% in only 9 of the 1971 samples collected from 2003 to 2007 (0.46%) (Figure 2). By far, the major contributor to the estimated effects of the OP and CB mixtures on brain AChE activity in salmon in Washington State was azinphos-methyl. When azinphos-methyl was removed from the mixtures analysis, none of the 1971 samples would be expected to cause greater than 20% inhibition of brain AChE activity (Figure 2). The most sensitive sublethal endpoint in salmon, swimming stamina in coho salmon, occurs at 22.8% inhibition of brain AChE activity (Sandahl et al. 2005).

thumbnail image

Figure 2. Reverse cumulative frequency distribution for estimated inhibition of brain acetylcholinesterase activity in Pacific salmon exposed to mixtures of organophosphates and carbamates in five WRIAs in Washington State (2003–2007).

Download figure to PowerPoint

The results from the monitoring data in the NAWQA database for California, Idaho, Oregon, and Washington are similar to those observed in Washington State. Of the 1580 samples taken from January 2003 to July 30, 2010, 11 (0.7%) had the potential to cause greater than 20% inhibition of brain AChE activity in Pacific salmonids (Figure 3). When azinphos-methyl was removed from the analysis, only 1 of the 1580 samples (0.06%) had the potential to cause greater than 20% inhibition of brain AChE activity (Figure 3).

thumbnail image

Figure 3. Reverse cumulative frequency distribution for estimated inhibition of brain acetylcholinesterase activity in Pacific salmon exposed to mixtures of organophosphates and carbamates in California, Idaho, Oregon and Washington (2002–2010).

Download figure to PowerPoint

Three sampling locations in the Washington State monitoring program were selected for additional analysis: Marion Drain 2, Sulphur Creek Wasteway 1, and Thornton Creek 3. Each of these sampling locations has a complete data set from 2003 to 2007. The Marion Drain and Sulphur Creek Wasteway are in the Lower Yakima basin and were selected because they have the highest percentage of land with crops and a diversity of agriculture within the drainage (Burke et al. 2006). The Marion Drain is a 19 mile long drainage ditch and 59% of the watershed is in agricultural crops (e.g., apple, hops, corn) (Burke et al. 2006). The Marion Drain provides spawning habitat for fall chinook salmon and summer steelhead (Oncorhynchus mykiss), and coho salmon have been observed in the drain (Burke et al. 2005). Of the 239 samples taken at the Marion Drain 2 sampling location, only 1 approached, but did not exceed, 22.8% estimated inhibition of brain AChE activity. Sampling conducted 4 days before and 8 days after the highest observed effect indicated less than 1% inhibition of brain AChE activity (Figure 4).

thumbnail image

Figure 4. Estimated inhibition of brain acetylcholinesterase activity in Pacific salmon in the Marion Drain 2 sampling location in Washington.

Download figure to PowerPoint

The Sulphur Creek Wasteway is a highly channelized agricultural conveyance located in a watershed in which 35% of the land area is in agricultural crops (e.g., apple, grapes, and corn) (Burke et al. 2006). Salmon are attracted to the wasteway by the high volume of irrigation return flows. As a result, summer steelhead, fall chinook salmon, and spring chinook salmon have been observed in the wasteway (Burke et al. 2005). From June 25 to July 6, 2005, 4 consecutive samples had the potential to cause greater than 14% inhibition of brain AChE activity (Figure 5), with 1 of those samples resulting in an estimated 40.3% inhibition (not shown in the figure). Nearly all of the observed effects could be attributed to high concentrations of azinphos-methyl. Overall, 93.6% of samples at this location were associated with less than 5% estimated inhibition of brain AChE activity (Figure 5).

thumbnail image

Figure 5. Estimated inhibition of brain acetylcholinesterase activity in Pacific salmon in the Sulphur Creek Wasteway 1 sampling location in Washington.

Download figure to PowerPoint

Thornton Creek is located in the Cedar-Sammamish WRIA and was selected to represent pesticide exposure in an urban basin. The creek drains a 12-square mile watershed, has 75 000 to 100 000 residents, and impervious surfaces cover approximately 50% of the watershed (Burke et al. 2006). Thornton Creek is located within the Puget Sound Evolutionarily Significant Unit for chinook salmon and the Puget Sound Distinct Population Segment for bull trout (Salvelinus confluentus) and Puget Sound steelhead. Puget Sound coho salmon are a species of concern (Burke et al. 2005). Despite being located in a highly urbanized WRIA, mixtures of organophosphates and carbamates were never sufficient to cause estimated inhibition of brain AChE activity greater than 0.5% in the period 2003 to 2007 (Figure 6).

thumbnail image

Figure 6. Estimated inhibition of brain acetylcholinesterase activity in Pacific salmon in the Thornton Creek 3 sampling location in Washington.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

Our analyses indicate that it is relatively rare for organophosphates and carbamates to cause even brief, minor inhibition of brain AChE activity in Pacific salmon in salmonid-bearing waters in Washington State (Figure 2) and the Pacific Northwest (Figure 3). The observed effects of organophosphates and carbamates on brain AChE activity in Pacific salmon were primarily due to exposure to azinphos-methyl (Figures 2 and 3). The USEPA published a product cancellation order on February 20, 2008 (Federal Register Notice, 73 FR 9328) specifying that all registrations of azinphos-methyl will be canceled effective September 30, 2012. This cancellation effectively eliminates the risks posed by azinphos-methyl to salmon in the Pacific Northwest.

There were several important assumptions in our analyses: 1) the protocols used by Washington State and the NAWQA programs were able to capture peak exposures of salmon to organophosphates and carbamates, and 2) mixture effects can be estimated by normalizing responses to pesticide EC50s and assuming concentration additivity. We explore each of these assumptions in turn.

Peak exposures

Capturing peak exposures may be critical because the mode of action of OP and CB pesticides, inhibition of acetylcholinesterase activity, occurs within hours of initial exposure (Boone and Chambers 1996). However, the toxicity data to which exposure concentrations were compared (i.e., Laetz et al. 2009; NOAA 2009) were based on exposures of 96 h. In most flowing waters, concentrations of OPs and CBs will decline rapidly after the initial peaks arising from spray drift, and to lesser extent, runoff. Temporally averaged 96-h concentrations in salmonid-bearing waters would be far lower than peak instantaneous concentrations. Thus, it is highly conservative to compare peak instantaneous concentrations with 96-h effects metrics.

The above argues that the absence of peak instantaneous concentrations in the monitoring data sets is irrelevant. Even so, it is hard to accept the argument that peak exposure concentrations were missed by the monitoring studies conducted to date in California, Idaho, Oregon and Washington. In the Washington State monitoring program, 1971 samples were targeted to areas of intense pesticide use during the growing season (Anderson et al. 2004, 2007; Burke et al. 2005, 2006; Anderson and Dugger 2008). Although the NAWQA data set of 1580 samples was not as targeted to the timing and location of pesticide applications, many stations were located in agricultural and urban areas where pesticide use has been high (e.g., Central Valley of California). Thus, it is highly unlikely that exposure peaks meaningful to populations of salmon over appropriate exposure durations (96 h) would have been missed in both of these data sets.

Method to estimate joint toxicity

In our analyses, we estimated the joint toxicity of organophosphates and carbamates to brain AChE activity by normalizing the concentration of each detected pesticide to its respective EC50 and summing the EC50-normalized data for the detected pesticides. The summing was done according to a fitted equation from Laetz et al. (2009) that assumed concentration additivity. Normalizing the pesticide concentrations to their respective EC50s assumes that the slopes of the concentration-response curves for the tested OPs and CBs in Laetz et al. (2009) and NOAA (2009) do not differ significantly. It is doubtful whether this is the case. Laetz et al. (2009) state that “the slopes of the concentration–response curves were not significantly different (average = 0.96, F test, p = 0.1).” However, the statistical analysis involved a one-way analysis of variance (1-way ANOVA) followed by Dunnett's post hoc test to establish differences between groups. The post hoc procedure would unnecessarily dilute power to detect a significant difference in slopes.

An F test from a 1-way ANOVA using the slopes and standard errors listed in Table 4 of Laetz et al. (2009) showed a significant difference in the 5 slopes listed (p = 0.024). The most powerful and appropriate test of the null hypothesis of equal slopes from a regression would be an F test from analysis of covariance for linear regression. From the slopes shown in Table 4 of Laetz et al. (2009) and the slopes shown in Table 4 of NOAA (2009), using an ANCOVA approach would very likely yield a significant p value for slope differences. For example, the slopes of −1.9 and −1.5 for azinphos-methyl and chlorpyrifos, respectively, appear to be quite different from the slopes of −0.32 and −0.79 for disulfoton and diazinon, respectively. However, to do the test would require the raw data from Laetz et al. (2009) and NOAA (2009), which we do not have.

Although the slopes of the tested organophosphates and carbamates are likely to be significantly different, the approach used by Laetz et al. (2009) to fit an equation to the combined EC50-normalized data is an unbiased one. Thus, although individual sample estimates of mixture toxicity to brain AChE in salmon may be somewhat uncertain, the overall conclusion that organophosphates and carbamates have not caused significant effects to Pacific salmon in recent years is unaffected.

The assumption of concentration additivity to estimate joint toxicity of organophosphates and carbamates appears to be reasonable. As noted earlier, recent studies and reviews have shown that the combined effects of pesticides sharing a common mode of toxic action are best predicted by a concentration addition (CA) model (Faust et al. 2000; Deneer 2000; Kudsk et al. 2005; Belden et al. 2007). Laetz et al. (2009) observed both concentration additivity and synergism in the binary combinations of carbamates and organophosphates tested, with a greater degree of synergism at higher exposure concentrations. However, in the 1580 samples in the NAWQA data set and 1971 samples in the Washington State data set, there was not a single instance where concentrations of 2 or more pesticides reached the lowest concentrations of any of the 10 binary mixture combinations tested by Laetz et al. (2009). The high concentrations used in the Laetz et al. (2009) mixture studies may have affected secondary, nontarget sites of action, which would not have been affected at environmentally relevant concentrations. Past studies that were designed to determine mixture toxicity at environmentally relevant concentrations (e.g., George et al. 2003; Teather et al. 2005; Brander et al. 2009) did not produce effects beyond those predicted by assuming concentration addition.

Other lines of evidence

Migratory organisms such as Pacific salmon present a challenge when predicting exposure to pesticides given their uneven distribution in time and space. Our analysis assumed that salmon were present at the sampling locations at the times when organophosphates and carbamates were being applied and transported to the water bodies. Although there is some overlap between the presence of salmon and the pesticides of interest, there are arguments indicating that the degree of overlap may not be high. For example, Waite and Carpenter (2000) sampled fish assemblages and habitat features in 24 Willamette River valley streams from 1993 to 1995. They found that the 3 forested coldwater sites had a high percentage of salmonids (24.1%) whereas the 21 valley-bottom streams had few salmonids (0.2% to 1.0%) (Teply et al. 2012). Furthermore, Hughes and Gammon (1997) found that salmonids tended to be associated with “fast water, rubble or gravel bottoms, cold water, and high dissolved-oxygen concentrations.” The forested coldwater sites with fast water and rubble or gravel bottoms would likely have much lower exposure to pesticides than would valley-bottom streams because the former tend to occur at higher elevations with limited surrounding agriculture and urban development. As the Washington State monitoring program was biased to urban areas and areas with intense agriculture, i.e., areas with generally lower numbers of salmon, there is a distinct possibility that we overestimated the risks of organophosphates and carbamates to salmon in the Pacific Northwest.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

Collectively, the evidence from the available monitoring data collected since 2003 demonstrates that mixtures of organophosphates and carbamates in surface waters inhabited by Pacific salmon rarely occur at levels capable of producing significant physiological and behavioral effects. Furthermore, these mixtures never reached concentrations capable of causing mortality. The Washington State and NAWQA data sets had a large number of stations in areas of intense agriculture. Furthermore, sampling intensity and timing were sufficient to have detected peak concentrations of organophosphates and carbamates. Given the phase-out of azinphos-methyl in 2012, mixtures of organophosphates and carbamates are highly unlikely to pose a direct risk to Pacific salmon in the future.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES

This project was funded and guided by the Ecological Risk Assessment Committee of CropLife America. The authors are grateful for their support and contributions to this manuscript. In particular, we would like to acknowledge the helpful comments provided by Lisa Ortego and Nick Poletika on an earlier draft of this manuscript. The authors would also like to thank the anonymous reviewers of the original draft manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. THE WASHINGTON STATE MONITORING STUDY
  5. NAWQA DATA SET FOR THE PACIFIC NORTHWEST
  6. TOXICITY OF CARBAMATES AND ORGANOPHOSPHATES
  7. ESTIMATING THE EFFECTS OF MIXTURES OF CARBAMATE AND ORGANOPHOSPHATE PESTICIDES
  8. RESULTS
  9. DISCUSSION
  10. CONCLUSIONS
  11. Acknowledgements
  12. REFERENCES
  • Anderson P, Dugger D, Burke C. 2007. Surface water monitoring program for pesticides in salmonid-bearing streams, 2006 data summary. Olympia (WA): Washington State Departments of Agriculture and Ecology. Ecology Publication Number 07–03–016. [cited 2010 January 5]. Available from: http://www.ecy.wa.gov/biblio/0703016.html
  • Anderson P, Jack R, Burke C, Cowles J, Moran B. 2004. Surface water monitoring program for pesticides in salmonid-bearing streams, April to December, 2003. Olympia (WA): Washington State Department of Ecology. Ecology Publication Number 04–03–048. [cited 2010 January 5]. Available from: http://www.ecy.wa.gov/biblio/0403048.html
  • Anderson PD, Dugger D. 2008. Surface water monitoring program for pesticides in salmonid-bearing streams, 2007 data summary. Olympia (WA): Washington State Departments of Agriculture and Ecology. Ecology Publication Number 08–03–009. [cited 2009 November 1]. Available from: http://www.ecy.wa.gov/biblio/0803009.html
  • Baldwin DH, Spromberg JA, Collier TK, Scholz NL. 2009. A fish of many scales: Extrapolating sublethal pesticide exposures to the productivity of wild salmon populations. Ecol Appl 19:20042015.
  • Beauvais SL, Jones SB, Brewer SK, Little EE. 2000. Physiological measures of neurotoxicity of diazinon and malathion to larval rainbow trout (Oncorhynchus mykiss) and their correlation with behavioral measures. Environ Toxicol Chem 19:18751880.
  • Beauvais SL, Jones SB, Parris JT, Brewer SK, Little EE. 2001. Cholinergic and behavioral neurotoxicity of carbaryl and cadmium to larval rainbow trout (Oncorhynchus mykiss). Ecotoxicol Environ Saf 49:8490.
  • Belden JB, Gilliom RJ, Lydy MJ. 2007. How well can we predict the toxicity of pesticide mixtures to aquatic life. Integr Environ Assess Manag 3:364372.
  • Boone JS, Chambers JE. 1996. Time course of inhibition of cholinesterase and aliesterase activities, and nonprotein sulfhydryl levels following exposure to organophosphorus insecticides in mosquitofish (Gambusia affinis). Fund Appl Toxicol 29:202207.
  • Brander SM, Werner I, White J, Deanovic LA. 2009. Toxicity of a dissolved pyrethroids mixture to Hyalella azteca at environmentally relevant concentrations. Environ Toxicol Chem 28:14931499.
  • Burke C, Anderson P, Cowles J, Moran B. 2005. Surface water monitoring program for pesticides in salmonid-bearing streams, April through October, 2004. Olympia (WA): Washington State Departments of Agriculture and Ecology. Ecology Publication Number 05–03–025. [cited 2010 January 5]. Available from: http://www.ecy.wa.gov/biblio/0503025.html
  • Burke C, Anderson P, Dugger D, Cowles J. 2006. Surface water monitoring program for pesticides in salmonid-bearing streams, 2003–2005. Olympia (WA): Washington State Departments of Agriculture and Ecology. Ecology Publication Number 06–03–036. [cited 2010 January 5]. Available from: http://www.ecy.wa.gov/biblio/0603036.html
  • Deneer JW. 2000. Toxicity of mixtures of pesticides in aquatic systems. Pest Manag Sci 56:516520.
  • Faust M, Altenburger R, Backhaus T, Boedeker W, Scholze M, Grimme LH. 2000. Predictive assessment of the aquatic toxicity of multiple chemical mixtures. J Environ Qual 29:10631068.
  • Ferrari A, Venturino A, Pechen D'Angelo AM. 2004. Time course of brain cholinesterase inhibition and recovery following acute and subacute azinphosmethyl, parathion, and carbaryl exposure in the goldfish (Carassius auratus). Ecotoxicol Environ Saf 57:420425.
  • Fulton MH, Key PB. 2001. Acetylcholinesterase inhibition in estuarine fish and invertebrates as an indicator of organophosphorus insecticide exposure and effects. Environ Toxicol Chem 20:3745.
  • George TK, Liber K, Solomon KR, Sibley PK. 2003. Assessment of the probabilistic ecological risk assessment-toxic equivalent combination approach for evaluating pesticide mixture toxicity to zooplankton in outdoor microcosms. Arch Environ Contam Toxicol 45:453461.
  • Gilliom RJ, Barbash EJ, Crawford CG, Hamilton PA, Martin JD, Nakagaki N, Nowell LH, Scott JC, Stackelberg PE, Thelin GP., et al. 2006. The quality of our nation's waters—Pesticides in the nation's streams and ground water, 1992–2001. US Geological Survey Circular 1291.
  • Hughes RM, Gammon JR. 1997. Longitudinal changes in fish assemblages and water quality in the Willamette River, Oregon. Trans Am Fish Soc 116:196209.
  • Kudsk P, Andersen HR, Cedergreen N, Mathiassen SK, Mohlenberg F, Streibig JC, Vinggaard AM. 2005. Combined effects of pesticides. Danish Environmental Protection Agency. No. 98 2005.
  • Labenia JS, Baldwin DH, French BL, Davis JW, Scholz NL. 2007. Behavioral impairment and increased predation mortality in cutthroat trout exposed to carbaryl. Mar Ecol Prog Ser 329:111.
  • Laetz CA, Baldwin DH, Collier TK, Hebert V, Stark JD, Scholz NL. 2009. The synergistic toxicity of pesticide mixture: Implications for risk assessment and the conservation of endangered Pacific salmon. Environ Health Perspect 117:348353.
  • Lisker EB, Ensminger MP, Gill SL, Goh KS. 2011. Detections of eleven organophosphorus insecticides and one herbicide threatening Pacific salmonids, Oncorhynchus spp., in California, 1991–2010. Bull Environ Contam Toxicol 87:355360.
  • Moore A, Waring CP. 1996. Sublethal effects of the pesticide diazinon on olfactory function in mature male Atlantic salmon parr. J Fish Biol 48:758775.
  • [NMFS] National Marine Fisheries Service. 2010. Endangered Species Act Section 7 consultation. Draft biological opinion. Environmental Protection Agency registration of pesticides containing azinphos methyl, bensulide, dimethoate, disulfoton, ethoprop, fenamiphos, naled, methamidophos, methidathion, methyl parathion, phorate, and phosmet. Endangered Species Division, Office of Protected Resources.
  • [NOAA] National Oceanic Atmospheric Administration. 2009. Toxicity of twelve organophosphate pesticides to juvenile coho salmon. NOAA Northwest Fisheries Science Center, Seattle, WA.
  • Perkins EJ Jr, Schlenk D. 2000. In vivo acetylcholinesterase inhibition, metabolism, and toxicokinetics of aldicarb in channel catfish: Role of biotransformation in acute toxicity. Toxicol Sci 53:308315.
  • Post G, Leasure R. 1974. Sublethal effect of malathion to three salmonid species. Bull Environ Toxicol Chem 12:312319.
  • Relyea RA. 2009. A cocktail of contaminants: How mixtures of pesticides at low concentrations affect aquatic communities. Oecologia 159:363376.
  • Sandahl JF, Baldwin DH, Jenkins JJ, Scholz NL. 2005. Comparative thresholds for acetylcholinesterase inhibition and behavioral impairment in coho salmon exposed to chlorpyrifos. Environ Toxicol Chem 24:136145.
  • Sandahl JF, Jenkins JJ. 2002. Pacific steelhead (Oncorhynchus mykiss) exposed to chlorpyrifos: Benchmark concentration estimates for acetylcholinesterase inhibition. Environ Toxicol Chem 21:24522458.
  • Scholz NL, Truelove NK, French BL, Berejikian BA, Quinn TP, Casillas E, Collier TK. 2000. Diazinon disrupts antipredator and homing behaviours in chinook salmon (Oncorhynchus tshawytscha). Can J Fish Aquat Sci 57:19111918.
  • Scholz NL, Truelove NK, Labenia JS, Baldwin DH, Collier TK. 2006. Dose-additive inhibition of Chinook salmon acetylcholinesterase activity by mixtures of organophosphate and carbamate insecticides. Environ Toxicol Chem 25:12001207.
  • Spromberg JA, Scholz NL. 2011. Estimating the future decline of wild coho salmon populations resulting from early spawner die-offs in urbanizing watersheds of the Pacific Northwest, USA. Integr Environ Assess Manag 7:648656.
  • Teather K, Jardine C, Gormley K. 2005. Behavioral and sex ratio modification of Japanese Medaka (Oryzias latipes) in response to environmentally relevant mixtures of three pesticides. Environ Toxicol 20:110117.
  • Teply M, Cramer SP, Poletika NN. Forthcoming. 2012. A spatially and temporally explicit model for determining the exposure of juvenile salmon to agricultural pesticides in freshwater. Integr Environ Assess Manag 8:271284.
  • [USEPA] United States Environmental Protection Agency. 2006. Organophosphorus cumulative risk assessment—2006 update. Office of Pesticide Programs, Washington DC: USEPA. EPA-HQ-OPP-2006-0618.
  • [USEPA] United States Environmental Protection Agency. 2007. Revised n-methyl carbamate cumulative risk assessment. Office of Pesticide Programs, Washington DC: USEPA. EPA-HQ-OPP-2007-0935.
  • [USGS] United States Geological Survey. 2010. USGS National Water Quality Assessment Data Warehouse. [cited 2010 July]. Available from: http://infotrek.er.usgs.gov/traverse/f?p=NAWQA:HOME:0
  • [USGS] United Stated Geological Survey. 2009. National Water Systems Information System: Web interface. Parameter code definitions. [cited 2009 October]. Available from: http://nwis.waterdata.usgs.gov/usa/nwis/pmcodes
  • Van Dolah RF, Maier PP, Fulton MH, Scott GI. 1997. Comparison of azinphosmethyl toxicity to juvenile red drum (Sciaenops ocellatus) and the mummichog (Fundulus heteroclitus). Environ Toxicol Chem 16:14881493.
  • Waite IR, Carpenter KD. 2000. Associations among fish assemblage structure and environmental variables in Willamette Basin streams, Oregon. Trans Am Fish Soc 129:754770.
  • [WSDE] Washington State Department of Ecology. 2009. Environmental Information Management System. [cited 2009 October]. Available from: http://apps.ecy.wa.gov/eimreporting/Search.asp?Type=StudyName