		








Drinking Water Exposure Assessment for Proposed New Uses of Abamectin on Chives, Dried Beans, and Bulb Onions (Crop Subgroup 3-07A)
                                       
                                       


                                   Abamectin
                        CAS# 71751-41-2	PC Code 122804
                                       


Prepared by:
Greg Orrick, Environmental Scientist
U.S. Environmental Protection Agency
Office of Pesticide Programs
Environmental Fate and Effects Division
Environmental Risk Branch II
1200 Pennsylvania Ave., NW
Mail Code 7507P
Washington, DC 20460
Reviewed by:
Nelson Thurman, Senior Science Advisor
Brian Anderson, Branch Chief

      

                               Table of Contents
                                       
                                       
1.	EXECUTIVE SUMMARY	3
2.	PROBLEM FORMULATION	4
2.1.	Use Characterization	4
2.1.1.	Usage Summary	7
3.	ANALYSIS	8
3.1.	Fate and Transport Characterization	8
3.1.1.	Residues of Concern	11
3.2.	Drinking Water Exposure Modeling	11
3.2.1.	Exposure Modeling Approach	11
3.2.2.	Input Parameters	12
3.2.3.	Modeling Results	14
3.2.4.	Monitoring Data	16
3.2.5.	Drinking Water Treatment	16
4.	CONCLUSIONS	16
5.	REFERENCES	17
Appendix A.  SCI-GROW Input/Output Data.	20
Appendix B.  PE Input Files.	21
                                       
1. 
2. EXECUTIVE SUMMARY

      This assessment provides Estimated Drinking Water Concentrations (EDWC) in surface water and in ground water from the use patterns of maximum abamectin exposure in support of the Health Effects Division's (HED) human health risk assessment (Table 1.1).  EDWCs of abamectin were modeled for the proposed uses on dried beans, fresh and dried chives, and the bulb onions crop subgroup (3-07A) and current uses on cotton and tomatoes.  The proposed use on dry beans is the use pattern of maximum exposure used to represent potential drinking water exposure for the dietary assessment of the proposed uses.  Exposure estimates reflect all residues of abamectin because the compound is stable to abiotic hydrolysis and acceptable data are not available to estimate the formation and decline of residues of concern via other degradation routes.  For characterization, exposure estimates are also provided that reflect the degradation rates used in the previous assessment that are based on studies currently determined to be unacceptable.  Surface water exposure estimates are adjusted by the national default percent cropped area (PCA) adjustment factor (87%).  This default PCA does not necessarily mean that dry beans will occupy 87% of the watershed, but reflects the likelihood that abamectin may be used on multiple crops in the same watershed.  Water resource monitoring data for abamectin are not available.  Therefore, exposure estimates cannot be evaluated with monitoring data.
      
Table 1.1.  Abamectin Total Residue EDWCs in Surface Water and Ground Water for Use in Dietary Risk Assessment Based on the Use Pattern of Maximum Exposure, Dry Beans. A
                              Use (maximum rate)
                                    Source
                                    PCA (%)
                       1-in-10-year Peak Exposure (ppb)
                    1-in-10-year Annual Mean Exposure (ppb)
                          30-year Mean Exposure (ppb)
                                   Dry Beans
                            (0.056 lbs a.i./A/year)
                                 Surface water
                                      87
                                  2.3 [1.1] A
                                 1.3 [0.41] A
                                 0.94 [0.27] A
                              Tomatoes, cucurbits
                            (0.18 lbs a.i./A/year)
                                 Ground water
                                      N/A
                                 1.6 x 10[-3]
[A]  EDWCs assume persistence in degradation processes.  Values based on (currently unacceptable) degradation inputs used in previous assessments are in brackets for characterization purposes.

      The surface water EDWCs in Table 1.1 are 5x to 6x higher than those reported in the previous assessment (DP barcode 335118; USEPA, 2007).  This increase is due to modeling different model scenarios and application dates in this assessment as well as the assumption in this assessment that the abamectin is stable to degradation in the absence of acceptable data.
      
      The environmental fate of abamectin is uncertain because submitted environmental fate data are not acceptable other than submitted hydrolysis, batch equilibrium, and terrestrial field dissipation studies.  The compound mixture appears to readily photolyze, biodegrade moderately slowly in aerobic soil, and persist in anaerobic soil.  However, the studies supporting that characterization are considered unacceptable and new data need to be submitted.  Abamectin does not hydrolyze and data have not been submitted on its biodegradation in water.  The residues of concern of abamectin are unclear due to uncertainty in the environmental fate data.  Abamectin is nonvolatile, has a wide range of low mobility in soil, and is expected to be transported off of agricultural fields mainly via spray drift and runoff of residues sorbed to suspended solids.  Field studies indicate that abamectin moderately dissipates with a field dissipation half-life of 31 days.

      Exposure estimates for the proposed uses of abamectin likely exceed actual values that occur in drinking water reservoirs and ground water due to the screening design of the exposure models and variability and uncertainty in the submitted environmental fate data.
      
      
3. PROBLEM FORMULATION
      
      Abamectin was first registered in 1985 for fire ant control on turf, lawns, and non-agricultural areas, followed by registrations in 1987 for use on ornamentals and in 1989 for use on cotton and citrus, the first agricultural uses.  A variety of uses have been registered since.  The Interregional Research Project No. 4 (IR-4) is now proposing use on dry beans, fresh and dried chives, and bulb onions (i.e., crop subgroup 3-07A).  The bulb onions include "daylily, bulb; fritillaria, bulb; garlic, bulb; garlic, great-headed, bulb; garlic, serpent, bulb; lily, bulb; onion, bulb; onion, Chinese, bulb; onion, pearl; onion, potato, bulb; shallot, bulb; cultivars, varieties, and/or hybrids of these" (40 CFR Part 180.41; Jul. 1, 2010).
      
      This drinking water assessment uses modeling to provide estimates of surface water and ground water concentrations of residues in drinking water source water (pre-treatment) resulting from abamectin use on vulnerable sites according to current and proposed labels.  EDWCs reflect drinking water exposure to residues of concern for abamectin, which the Health Effects Division's 2008 dietary risk assessment reported is the parent compound and a mixture of 8-α-hydroxy and ring-opened aldehyde derivatives ("Compound 7"; DP 359583; USEPA, 2008).  Primary routes of transport to source water include runoff, including erosion, and spray drift to surface water.  Leaching of the residues of concern to ground water is not expected due to the compounds' affinity for soil particles.
      
      The models PRZM and EXAMS, coupled with the PE interface, were used to assess exposure in surface water due to runoff, erosion, and spray drift from the proposed and current abamectin uses.  Exposure in ground water due to leaching was assessed with the screening model SCI-GROW.

3.1.    Use Characterization

	Abamectin is a mixture of avermectins containing at least 80% avermectin B1a (5-0-demethyl avermectin A1a) and up to 20% avermectin B1b (5-0-demethyl-25-de(1-methylpropyl)-25-(1-methylethyl) avermectin A1a).  The avermectins are natural fermentation products of the soil bacterium Streptomyces avermitilis.  The active ingredient, abamectin, is used as an insecticide (and miticide) and as a seed protectant nematocide.  The compound acts as a chlorine channel agonist in invertebrates (Fritz, et al., 1979, Mellin et al., 1983 and Arena et al., 1991 in Sherma and Cairns, 1993), and may function as a gamma-aminobutyric acid (GABAergic) agonist (Kass et al., 1980, 1984 in Sherma and Cairns, 1993).  It acts by stimulating the release of gamma-aminobutyric acid, an inhibitory neurotransmitter, thus causing paralysis (Tomlin, 1994).  The difference in toxicity between invertebrates and mammals may be partially due to different distribution of the GABAergic neurons (Turner and Schaeffer, 1989 in Sherma and Cairns, 1993).

      The compound is currently registered on dozens of Section 3 end-use labels with uses on numerous fruit, vegetable, and other agricultural crops, including seed treatment uses on cotton, corn, cucurbit, and tomato seeds, cattle ear-tag uses, and non-agricultural uses including ornamental uses, bait uses, and use in residential, commercial, industrial and other non-crop areas for fire ant- and related ant-control and for crack and crevice treatments.  Proposed uses are for control of leafminers and thrips in bulb onions and control of leafminers and spider mites in dry beans and fresh and dried chives.
	
	This assessment analyzes the proposed uses as well as currently labeled uses of abamectin.  Table 2.1 displays the maximum labeled use patterns for each proposed use (dry beans, bulb onions, and chives) and the current uses of maximum exposure (cotton, tomatoes, and cucurbits).  Exposure estimates generated from these maximum use patterns were used to characterize drinking water exposure from abamectin use, both in surface water and in ground water.

Table 2.1.  Maximum Use Patterns for Proposed and Selected Current Abamectin Uses. [A]
                                  Use Pattern
                                   Current/
                                   Proposed
                         Single App. Rate (lbs a.i./A)
                             Max. No. of App. [B]
                      Seasonal App. Rate (lbs a.i./A) [C]
                           App. Interval (days) [B]
                               Application Type
                                   Dry beans
                                   Proposed
                                     0.019
                             Sets of 2 in sequence
                                     0.056
                                       6
                                 Aerial/ground
                                  Bulb onions
                                   Proposed
                                     0.019
                             Sets of 2 in sequence
                                     0.056
                                       7
                                 Aerial/ground
                                    Chives
                                   Proposed
                                     0.019
                                 2 per cutting
                                     0.056
                                       7
                                    Ground
                                    Cotton
                                    Current
                                     0.019
                                       2
                                     0.038
                                      21
                                 Aerial/ground
                              Tomatoes, cucurbits
                                    Current
                                     0.019
                             Sets of 2 in sequence
                                     0.056
                                       7
                                 Aerial/ground
                                       
                                       
                                     0.12
                                       1
                                     0.12
                                      N/A
                                Seed treatment
[A] Current uses selected for this table are those that result in maximum exposure.
[B] Applications to chives are limited to two per cutting or harvest (which occur approximately six weeks apart, five to nine times per year; USEPA, 2010a).  Applications to dry beans and bulb onions are limited to two sequential applications, with additional sets of two sequential applications allowed if a different insecticide is applied (twice in the case of bulb onions) in between sets of abamectin applications.
[C] One crop season occurs per year for the proposed and current crops.
      
      Seasonal application rates are equivalent to annual application rates for the proposed and current uses because only one crop season occurs per year.  The current uses on tomatoes and cucurbits, which were the uses of maximum exposure in previous drinking water exposure modeling (USEPA, 2007), have seed treatment uses as well as foliar uses.  When the uses are evaluated collectively, this makes them the use patterns of highest ground water exposure.  The current use on cotton has a seed treatment use as well.  However, the application rate per seed (0.15 mg) is 25% less than that for tomatoes and cucurbits (0.60 mg), resulting in an insubstantial increase in exposure.  The proposed use on dry beans (with no seed treatments) is the use pattern of highest surface water exposure.  The maximum application rate for dry beans is 0.019 pounds of active ingredient per acre (lbs a.i./A), limited to sets of two sequential applications six days apart that are separated by application of a different insecticide (interval not provided) up to a total seasonal (or annual) application rate limit of 0.056 lbs a.i./A.
      
      The proposed uses may be applied by aerial or ground application equipment, with the exceptions that 1) use on chives is limited to ground equipment and 2) aerial application is prohibited in New York State for the other proposed uses on dry beans and bulb onions.
      
      Figures 2.1 and 2.2 display Census of Agriculture maps of harvested acres of dry beans and dry onions, respectively, in 2007 (USDA, 2010).  Maps of harvested acres for chives or other more specific proposed uses are not available from the Census.
      

Figure 2.1.  Harvested Acres of Dry Edible Beans Other Than Limas in 2007 (USDA, 2010).


Figure 2.2.  Harvested Acres of Dry Onions in 2007 (USDA, 2010).

3.1.1.       Usage Summary

	Figure 2.3 presents the national agricultural usage pattern of abamectin in 2002 (USGS, 2010).  At that time, 21% of the national agricultural usage was on citrus fruits, followed by 19% on cotton, 13% on almonds, 9% on pears, and <6% on each of the remaining crops, grapes, tomatoes, strawberries, lettuce, avocados, and apples.  These data are relatively consistent with BEAD's Screening Level Usage Analysis (SLUA) of abamectin (dated October 6, 2010) based on source data from 2001 to 2008 (USEPA, 2010).  However, usage on most crops is reportedly higher than USGS estimates, with amounts used on cotton and almonds surpassing that on citrus.  The SLUA reports that cotton (10,000 lbs a.i.), almonds (6,000 lbs a.i.), and citrus (grapefruit, lemons, oranges, tangelos, and tangerines; collectively 3,000-4,500 lbs a.i.) account for the greatest amount of usage, followed by avocados, grapes, pears, strawberries, and tomatoes (accounting for 1,000 lbs a.i. of usage each).
      

Figure 2.3.  National Agricultural Usage of Abamectin in 2002 (USGS, 2010).


4. ANALYSIS

4.1.    Fate and Transport Characterization

      Abamectin (CAS No. 71751-41-2; PC code 122804; chemical structure in Figure 3.1) is a mixture of avermectins containing at least 80% avermectin B1a (5-0-demethyl avermectin A1a; CAS No. 65195-55-3) and up to 20% avermectin B1b (5-0-demethyl-25-de(1-methylpropyl)-25-(1-methylethyl) avermectin A1a; CAS No. 65195-56-4).  The environmental fate of this compound mixture is uncertain because submitted environmental fate data are not acceptable other than the submitted hydrolysis, batch equilibrium, and terrestrial field dissipation studies.  Based on submitted data, including the currently unacceptable data used in recent assessments, abamectin moderately degrades in aerobic soil (with half-lives of 15 to 47 days) and persists in anaerobic soil (half-life of 267 days).  Data have not been submitted on the biodegradation of abamectin in water; however, its close derivative ivermectin has an aerobic aquatic DT50 of 15 days according to Loffler et al. (2005).  The compound mixture is stable to hydrolysis.  Illuminated, abamectin photolyzes readily (half-life of <1 day) on soil surfaces and in clear, shallow waters.  However, there is uncertainty in the photolysis data because the studies are unacceptable.  Photolysis is not likely to be substantial where abamectin is incorporated or under canopy.  Because the submitted photolysis and biodegradation studies of abamectin are not acceptable, the compound mixture will be conservatively assumed stable to these routes of degradation, which includes the assumption that any observed degradation results in structurally similar degradates of similar toxicity and environmental fate to the parent compound mixture, until acceptable data are submitted to refine this assumption.
                                       
Figure 3.1.  Chemical structure of abamectin (mixture of avermectins B1a and B1b; MRID 46877502).
      
      Abamectin is soluble in water only up to 0.010 mg/L (at 21°C) and is non-volatile under field conditions (vapor pressure of 1.5 x 10[-9] torr at 25°C).  In soil, abamectin displays a wide range of low mobility (Kd range of 18 to 132 L/kg), with no correlation between absorption to soils and their organic carbon content.  Therefore, the compound mixture is not expected to be a ground water concern.  Important means of transport off of agricultural fields following application will include spray drift and runoff of residues sorbed to suspended solids.  Field dissipation data appear to be consistent with the laboratory data, indicating that abamectin dissipation on the field is moderate (dissipation half-life of 31 days).  However, leaching of abamectin and its delta-8,9-isomers to the 12-18-inch depth was observed by 28 days after application, indicating that abamectin residues may have increased mobility in some soils.  A summary of the environmental fate characteristics of abamectin from registrant-submitted studies is presented in Table 3.1.
      
Table 3.1. General Physicochemical Properties and Environmental Fate Parameters of Abamectin.
Parameter
Value
Reference
                         Physical/Chemical Parameters
Molecular mass
avermectin B1a: 873.1 g/mol
avermectin B1b: 859.1 g/mol
Calculated
Vapor pressure (25°C)
1.5 x 10[-9]  torr
MRID 47436204
Henry's Law constant (25°C)
1.7 x 10[-][7] atm·m[3]/mol
Calculated
Water solubility (21°C)
0.010 mg/L
MRID 47436204
Octanol-water partition coefficient (KOW); (log KOW) (25°C)
avermectin B1a: 300,000; (5.5)
avermectin B1b: 100,000; (5.0)
MRID 46877502
Dissociation constant (pKa)
7.22
MRID 46877502
                             Persistence in Water
Hydrolysis half-life (pH 5, 7, and 9)
No evidence of degradation 
MRID 131083
Aqueous photolysis DT50
[3.5, 12 hrs (0.29, 1.0 solar day)] [A]
[MRID 159163] A
Aerobic and anaerobic aquatic metabolism half-lives
(No submitted data)
--
                              Persistence in Soil
Soil photolysis DT50
[21 hrs (1.7 solar days)] A
[MRID 159163] A
Aerobic soil metabolism half-lives
[15, 20, 20, 28, 40, 47 d] A
[MRID 159163] A
Anaerobic soil metabolism half-life
[267 d] A
[MRID 159163] A
                                   Mobility
Soil-water partitioning coefficient (Kd)
77 +- 58 L/kg (range: 9.7-161 L/kg; 3 soils)
MRID  40856301
                               Field Dissipation
Field dissipation half-life
31+-6 d (leaching to 12-18" depth at 28 days)
MRID 42627601; 41191501
[A] Data in brackets are unacceptable and provided for characterization.
      
      The major abamectin degradates in soil included a mixture of 8-α-hydroxy and ring-opened aldehyde derivatives of the parent compound mixture ("Compound 7"; maximum of 20% of the applied 56 days after treatment) and two other unidentified compounds or mixtures ("Compound 4", maximum of 12% of the applied at 84 days, and "Compound 12", maximum of 12% of the applied at 28, 56, 112 days).  These major degradates appeared to be more persistent in soil than the parent compound mixture.  Their mobility in soil is not determined.
      
      The major abamectin photodegradates were not identified.  However, three chromatographic fractions representing degradates were described, including a major non-polar fraction consistent with the delta-8,9-isomers of abamectin, a major "moderately polar" fraction, and a major polar fraction that consisted of multiple components.

4.1.1.       Residues of Concern

      The Health Effects Division's 2008 dietary risk assessment included abamectin and its soil degradate mixture "Compound 7" as residues of concern in drinking water (DP barcode 359583; USEPA, 2008).  Compound 7 was considered as toxic as the parent compound due to structural similarity and because its toxicity is unknown.  Other major environmental degradates may meet these criteria as well pending their identification.  Because submitted photolysis and biodegradation studies are not acceptable due to failure to identify major degradates among other deficiencies, abamectin is conservatively assumed stable to these routes of degradation, as stated above, which includes the assumption that any observed degradation results in structurally similar degradates of similar toxicity and environmental fate to the parent compound mixture.  Due to this assumption of no degradation, only abamectin is modeled in this assessment as the residue of concern.
      
      Once acceptable environmental fate data for abamectin are submitted, a total residues approach will be used to estimate exposure to the abamectin residues of concern.  This approach involves the use of total-residues degradation half-lives based on the sum of residues of concern at each sampling interval in submitted laboratory degradation studies.  Also, the chemical properties and soil mobility of the residue of concern most vulnerable to environmental transport are typically used to represent the chemical properties and soil mobility of all residues.

4.2.    Drinking Water Exposure Modeling

4.2.1.       Exposure Modeling Approach

      The Tier II models, Pesticide Root Zone Model (PRZM v3.12.2, May 12, 2005) linked with EXposure Analysis Modeling System (EXAMS v2.98.04.06, Apr. 25, 2005) via the PE model shell (PE v5.0, Nov. 15, 2006, i.e., PRZM/EXAMS) were run to estimate screening-level exposure of surface water drinking water sources to abamectin following the proposed and current uses.  Ground water drinking water sources were assessed with the Tier I model, Screening Concentration in Ground Water (SCI-GROW v2.3, Jul. 29, 2003).

      The PE model generates estimated drinking water concentrations (EDWC) of abamectin residues that may occur in surface water used as drinking water down gradient from applications on terrestrial crops.  The PRZM model simulates pesticide movement and transformation from crop application through soil residue processes.  The EXAMS model simulates pesticide loading via runoff, erosion, and spray drift assuming a standard watershed of 172.8 ha that drains into an adjacent standard drinking water index reservoir of 5.26 ha and an average depth of 2.74 m.  A more detailed description of the index reservoir watershed can be found in Jones et al., 2010.

      SCI-GROW estimates a single screening-level ground water concentration that might be expected in shallow unconfined aquifers under highly permeable soils, for use as both an acute and chronic drinking water endpoint.  SCI-GROW was developed by fitting a linear model to ground water concentrations with the Relative Index of Leaching Potential (RILP) as the independent variable.  Ground water concentrations were taken from 90-day average high concentrations from Prospective Ground Water studies.  The RILP is a function of aerobic soil metabolism and the soil-water partition coefficient.  The output of SCI-GROW represents the concentrations of abamectin residues that might be expected in shallow unconfined aquifers under sandy soils, which is representative of the ground water most vulnerable to pesticide contamination and likely to serve as a drinking water source.

      The coupled PRZM/EXAMS (PE) model, SCI-GROW model, and associated users manuals may be downloaded from the U.S. Environmental Protection Agency (EPA) Water Models web-page (USEPA, 2010a).

4.2.2.       Input Parameters

	Ground Water Exposure

      Input parameters for the SCI-GROW model appear in Table 3.2.  The selected application inputs represent the maximum proposed or current use pattern, which is for the current uses on tomatoes and cucurbits (one seed treatment per year at 0.12 lbs a.i./A and three foliar applications per year at 0.019 lbs a.i./A).  The seed treatment application rate of 0.12 lbs a.i./A was multiplied by 6%, producing 0.0072 lbs a.i./A, to account for the availability of abamectin on treated seeds (MRID 47000503).  Because SCI-GROW cannot accept multiple applications at different rates (and abamectin is persistent below the soil surface), all applications were summed and input as one total application of 0.064 lbs a.i./A.

Table 3.2.  SCI-GROW Input Parameters for Abamectin Residues. [A]
Input Parameter
Value
Source
Application Rate
0.064 lbs a.i./A [B]
Current labels
Applications per Year
1 B
Current labels
Organic Carbon Partition Coefficient (KOC)
4700 L/kgOC
MRID 40856301
Aerobic Soil Metabolism
Half-life (from previously assessments) [C]
10,000 d (101 d)
No acceptable data available (USEPA, 2007)
[A]  Source data are in Tables 2.1 and 3.1.
[B]  Input is annual sum of applications [(0.019 lbs a.i./A x 3) + 0.0072 lbs a.i./A = 0.064 lbs a.i./A] because SCI-GROW does not accept different application rates per year.
[C]  For characterization (in parentheses), the total residue half-life used in previous assessments (USEPA, 2007) was also modeled.
      
      The lowest measured mean organic carbon partition coefficient (4700 L/kgOC) from the submitted batch equilibrium study of abamectin (MRID 40856301) was input into SCI-GROW, in accordance with current input parameter guidance (USEPA, 2009), because of the greater than three-fold variation in measured values.  Because acceptable aerobic soil metabolism data have not been submitted, the compound was assumed stable for the aerobic soil metabolism half-life input (represented as 10,000 days).  For characterization, the total residues half-life value (101 days) used in previous assessments (USEPA, 2007) was modeled as well, although the data represented by the half-life (MRID 159163) have since been reclassified as unacceptable.  Input and output data for SCI-GROW are attached in Appendix A.

	Surface Water Exposure
      
      The model input parameters used in PE to simulate the scenarios and application practices for the proposed and current uses of abamectin are provided in Table 3.3.  Input files used for modeling results are listed in Appendix B.
      
Table 3.3.  PRZM Scenarios and Input Parameters Describing Proposed and Current Maximum Use Patterns of Abamectin.
                                      Use
                                   Scenario
                             Date of Initial App.
                            App. Rate in lbs a.i./A
                                 (kg a.i./ha)
                                 App. per Year
                             App. Interval (days)
                                   CAM Input
                                 IPSCND Input
                      Application Efficiency/ Spray Drift
                                   Dry beans
                                   MI beans
                                    Aug. 1
                                 0.019 (0.021)
                                       3
                                    6, 12 A
                                       2
                                       1
                                   0.95/0.16
                                 Bulb onions 
                                   GA onion
                                    Apr. 24
                                 0.019 (0.021)
                                       3
                                   7, 21 [B]
                                       2
                                       1
                                   0.95/0.16
                                    Chives
                                    OR mint
                                    Jul. 1
                                 0.019 (0.021)
                                       3
                                       7
                                       2
                                       1
                                  0.99/0.064
                                    Cotton
                                   MS cotton
                                    Jun. 15
                                 0.019 (0.021)
                                       2
                                      21
                                       2
                                       1
                                   0.95/0.16
                                   Tomatoes
                                   FL tomato
                                    Mar. 5
                                 0.019 (0.021)
                                       3
                                    14, 7 C
                                       2
                                       1
                                   0.95/0.16
                                       
                                       
                                    Jan. 18
                              0.0072 (0.0081) [D]
                                       1
                                      --
                                     8 [E]
                                       1
                                    1.0/0.0
[A]  The second application occurs after 6 days.  The third application was assumed to occur after 12 days, allowing for an application of an alternative insecticide after the second application.
[B]  The second application occurs after 7 days.  The third application was assumed to occur after 21 days, allowing for two applications of an alternative insecticide after the second application.
[C]  The second application was assumed to occur after 14 days, following an application of an alternative insecticide.  The third application occurs 7 days after the second application.
[D]  The application rate of 0.12 lbs a.i./A was multiplied by 6%, producing 0.0072 lbs a.i./A, to account for the availability of abamectin on treated seeds (MRID 47000503).
[E]  Treated seeds are planted at a 0.5-inch (1.3-cm) depth.  However, a 2.0-cm incorporation depth was modeled due to limitations with PRZM.
      
      Applicable scenarios were modeled for each proposed use.  The Oregon mint scenario was used as a surrogate for chives in the absence of a scenario specific to chives and because mint is an herb as well.  Scenarios listed in Table 3.3 are those that are most vulnerable to exposure for each use.  Dates of initial application were selected that were within the application window and characterized by vulnerability to runoff.  Runoff vulnerability was explored by modeling a range of initial application dates approximately 14 days apart within the labeled application timing and scenario crop season, and selecting the date for which exposure estimates were highest.  This was expected to produce high-end exposure estimates that are conservative but that may not represent the highest possible exposure.  Use on cucurbits was not modeled, as was done in previous assessments, because use on tomatoes produces higher exposure.  All foliar uses were modeled as aerially applied, with the exception that use on chives was modeled as a ground application per labeled directions.
      
      For seed treatment of tomatoes, the application rate of 0.12 lbs a.i./A was multiplied by 6% (producing 0.0072 lbs a.i./A) to account for the availability of abamectin on treated seeds (MRID 47000503).  The entire application (100% efficiency) was modeled with a 2.0-cm incorporation depth (chemical application method 8) due to limitations with PRZM, although treated seed may be planted at a 0.5-inch (1.3-cm) depth (DP 335118; USEPA, 2007).

      The chemical and environmental fate data for abamectin listed in Table 3.1 were used for generating model input parameters for PE (listed in Table 3.4).  The chemical inputs were based on abamectin parent and reflect stability to degradation due to a lack of acceptable degradation data, with the exception that hydrolysis data indicate that abamectin is stable to abiotic hydrolysis.  This results in exposure estimates that are conservative unless abamectin degrades into persistent residues of similar or greater toxicity and mobility.  The soil-water partition coefficient (Kd) input (77 L/kg) is the mean of three mean values for abamectin.
      
Table 3.4.  PE Chemical Input Parameters for Abamectin. [A]
                                Input Parameter
                                     Value
                                 Justification
                                    Source
Molecular Mass (g/mol)
                                     873.1
                   MW of >=80% of avermectins in abamectin
                                 (Calculated)
Henry's Law Constant (atm m[3]/mol)
                                1.7 x 10[-][7]
                  Ratio of vapor pressure to water solubility
                        (Calculated from MRID 47436204)
Solubility in Water (mg/L)
                                     0.010
                                  Study value
                                 MRID 47436204
Soil-water Partition Coefficient (Kd) (L/kg)
                                      77
                           Mean of three mean values
                                 MRID 40856301
Aerobic Soil Metabolism Half-life (days)
                                    0 [150]
Assumed stable in absence of acceptable data [value from previous assessments]
                       No acceptable data [USEPA, 2007]
Aerobic Aquatic Metabolism Half-life (days)
                                    0 [300]
Assumed stable in absence of acceptable data [value from previous assessments]
                       No acceptable data [USEPA, 2007]
Anaerobic Aquatic Metabolism Half-life (days)
                                       0
                 Assumed stable in absence of acceptable data
                              No acceptable data
Hydrolysis Half-life at pH 7 (days)
                                       0
                                  Study value
                                  MRID 131083
Aqueous Photolysis
Half-life (days)
                                    0 [0.5]
Assumed stable in absence of acceptable data [value from previous assessments]
                       No acceptable data [USEPA, 2007]
[A]  For characterization, values in brackets represent total residue half-lives used in previous assessments (USEPA, 2007).

4.2.3.       Modeling Results

	Ground Water Results

      The single exposure estimate from the SCI-GROW model is used as a conservative peak and annual mean value in ground water leachate.  The exposure estimate for the maximum use patterns (tomatoes and cucurbits) is 1.6 x 10[-3] ug/L (1.6 ng/L) assuming residues do not degrade.  Using the mean total residue aerobic soil degradation half-life value used in previous assessments, for characterization, the maximum exposure estimate is 9.1 x 10[-][4] ug/L (0.91 ng/L); however, the degradation data used to calculate this value have been downgraded to unacceptable.  These values support the expectation that abamectin residues will not produce appreciable exposure in ground water due to low mobility in soil and low application rates.

	Surface Water Results
      
      Proposed and current use patterns were modeled to estimate exposure in surface water, as described above.  Maximum abamectin EDWCs in surface water per use are listed in Table 3.5.  Exposure estimates were adjusted by the default national percent cropped area (PCA) value (87%) that accounts for the maximum area within any watershed that may be in agriculture (Jones et al., 2010a).  Modeled daily EDWCs for foliar and seed-treatment uses on tomatoes were combined in an Excel spreadsheet prior to adjusting for the PCA and calculating the EDWCs reported in Table 3.5.
      
Table 3.5.  Abamectin EDWCs in Surface Water from Proposed and Current Uses. [A]
                                      Use
                                 PRZM Scenario
                       1-in-10-year Peak Exposure (ppb)
                    1-in-10-year Annual Mean Exposure (ppb)
                          30-year Mean Exposure (ppb)
                                 Proposed Uses
                                   Dry beans
                                   MI beans
                                   2.3 [1.1]
                                  1.3 [0.41]
                                  0.94 [0.27]
                                 Bulb onions 
                                   GA onion
                                  1.4 [0.81]
                                  0.63 [0.26]
                                  0.49 [0.19]
                                    Chives
                                    OR mint
                                  0.91 [0.27]
                                  0.67 [0.15]
                                  0.54 [0.12]
                                 Current Uses
                                    Cotton
                                   MS cotton
                                  1.2 [0.74]
                                  0.58 [0.22]
                                  0.42 [0.15]
                                   Tomatoes
                                   FL tomato
                                  1.5 [0.67]
                                  0.63 [0.22]
                                  0.53 [0.18]
[A]  Maximum values are in bold.  Values based on degradation inputs used in previous assessments are in brackets.

      Exposure estimates from the proposed maximum use pattern for dry beans are greater in this national screen than those from other proposed and currently labeled uses.  This use pattern of maximum exposure results in a 1-in-10-year peak EDWC (2.3 ug/L) that is 5x higher than that reported in the previous assessments for use on cucurbits (0.46 ug/L; USEPA, 2007).  The 1-in-10-year annual mean EDWC (1.3 ug/L) for use on dry beans is 6x higher than that reported in the previous assessments for use on tomatoes (0.21 ug/L).  However, these 1-in-10-year peak and annual mean exposure estimates for use on dry beans are only 2x and 3x higher, respectively, than the exposure estimates based on the degradation inputs used in the previous assessments for characterization (listed in brackets in Table 3.5).  Addition of seed treatment to the foliar use on tomatoes did not change 1-in-10-year peak exposure estimates for foliar use and increased 30-year mean exposure estimates by only 15%.
      
      Values provided for characterization should not be used quantitatively in exposure assessment because they are based on unacceptable data.  They indicate that the current version of PE, with most of the same inputs (dates of initial application used in previous modeling are unknown), results in 1-in-10-year peak and annual mean values for use of abamectin on tomatoes that are similar (1.0-1.5x) to those resulting from the previous version of PE.
      
      There are a number of factors inherent in the modeling that can affect the accuracy of this analysis; these include the availability and quality of the model input data and the use of maximum labeled practices and environmental conditions that may differ from actual practices and conditions.  The exposure values listed above are conservative for a number of reasons.  The actual application area of these uses in drinking water basins is likely to be less than the total cropped area of the watershed.  Also, degradation rates for abamectin residues of concern were assumed stable in the absence of acceptable photolysis and biodegradation data.
      
4.2.4.       Monitoring Data

      This assessment contains a cursory review of available surface water and ground water monitoring data sources for data on abamectin.  Abamectin is not a regulated chemical under the Safe Water Drinking Act (SWDA) or related statutes.  The compound is not an analyte listed in the USEPA STORET database (USEPA, 2010b), the California Department of Pesticide Regulation (DPR) Surface Water Database (CDPR, 2010), or the USGS NAWQA program (USGS, 2010a).  Therefore, water monitoring data on abamectin have not been located.

4.2.5.       Drinking Water Treatment

      Because direct data regarding the effects of drinking water treatment on abamectin residues of concern have not been provided to the Agency and could not be located in the public literature, drinking water treatment effects on abamectin have been considered only qualitatively based on the physicochemical properties of the compound.  Carbon filtering is expected to reduce abamectin residue concentrations due to the compound's affinity to organic carbon.
      
      Flocculation and sedimentation removal may be effective at reducing abamectin residue concentrations, especially if residues are bound to suspended sediment.  Lime softening is not expected to reduce concentrations of abamectin residues because the compound mixture does not indicate susceptibility to abiotic hydrolysis.


5. CONCLUSIONS

      Drinking water exposure estimates for the proposed uses of abamectin on dry beans, chives, and bulb onions are represented by the use pattern of maximum abamectin exposure, dry beans (Tables 1.1 and 3.5).  Maximum acute, chronic, and cancer EDWCs in surface water are 2.3, 1.3, and 0.94 ug/L, respectively.  The maximum exposure estimate for ground water (from use on tomatoes and/or cucurbits) is only 1.6 x 10[-3] ug/L.  Monitoring and drinking water treatment data on abamectin are not available.
      
      Abamectin was modeled in this assessment as stable to degradation because it is stable to hydrolysis and submitted photolysis and biodegradation data are not acceptable.  It could not be concluded, based on the unacceptable data, that any observed degradation of abamectin resulted in degradates of much less toxicity.
      
      Assessment modeling relied on maximum labeled use patterns and a default national PCA value.  Where actual use patterns are less than the labeled maximums and location-specific PCAs are less than assumed in this assessment, actual environmental exposures may be lower.


6. REFERENCES

California Department of Pesticide Regulation (CDPR).  2010.  Surface Water Database.  California Environmental Protection Agency, Department of Pesticide Regulation, Surface Water Protection Program.  Data available through June, 2008.  Online at: http://www.cdpr.ca.gov/docs/sw/surfdata.htm

Jones, R. D., K. Costello, J. Hetrick, J. Lin, R. Parker, N. Thurman, C. Peck.  2010.  Development and Use of the Index Reservoir in Drinking Water Exposure Assessments.  U.S. Environmental Protection Agency, Office of Pesticide Programs.  April 15, 2010.  Online at: http://www.epa.gov/oppefed1/models/water/index_reservoir_dwa.html

Jones, R. D., K. Costello, J. Hetrick, J. Lin, R. Parker, N. Thurman, C. Peck, G. Orrick.  2010a.  Development and Use of Percent Cropped Area Adjustment Factors in Drinking Water Exposure Assessments.  U.S. Environmental Protection Agency, Office of Pesticide Programs.  September 9, 2010.  Online at: http://www.epa.gov/oppefed1/models/water/pca_adjustment_dwa.html

Loffler D., J. Rombke, M. Meller, and T. Ternes.  2005.  Environmental Fate of Pharmaceuticals in Water/Sediment Systems.  Environ. Sci. Technol., 39, 5209-5218.

Sherma, J. and Cairns, T.  Comprehensive analytical profiles of important pesticides.  1993.  CRC Press, Inc. Boca Raton, Fl. p 75.

Tomlin, C.D.S. (ed.). 1994.  The Pesticide Manual - World Compendium. 10th ed. Surrey, UK. The British Crop Protection Council, p. 4.

U.S. Department of Agriculture (USDA).  2010.  2007 Census of Agriculture: United States: Summary and State Data.  U.S. Department of Agriculture, National Agricultural Statistics Service.  Issued February, 2009; updated December, 2009.  Online at: http://www.agcensus.usda.gov/Publications/2007/index.asp

USEPA.  2007.  Abdel-Saheb, I.  Revised Drinking Water Assessment for Abamectin as a Seed Treatment and as a Foliar Application on Cucurbits and Tomato Based on New Seed Leaching Study Submitted to EFED.  DP Barcode 335118.  U.S. Environmental Protection Agency, Office of Prevention, Pesticides, and Toxic Substances, Environmental Fate and Effects Division.  Memorandum to the Registration Division.  Mar. 20, 2007.

USEPA.  2008.  Piper, S.  Abamectin: (Acute and Chronic Aggregate Dietary Exposure and Risk Assessments for the Section 3 on Stone Fruit Crop Group 12, Tree Nut Crop Group 14, Pistachio, Tuberous and Corm Vegetables Subgroup 1C and Cattle Ear Tag Use.  DP Barcode 359583.  U.S. Environmental Protection Agency, Office of Prevention, Pesticides, and Toxic Substances, Health Effects Division.  Memorandum to the Health Effects Division and Registration Division.  Dec. 13, 2008.

USEPA.  2009.  Guidance for Selecting Input Parameters in Modeling the Environmental Fate and Transport of Pesticides.  Version 2.1.  U.S. Environmental Protection Agency, Office of Pesticide Programs, Environmental Fate and Effects Division, Oct. 22, 2009.  Online at: http://www.epa.gov/oppefed1/models/water/input_parameter_guidance.htm

USEPA.  2010.  Screening Level Usage Analysis (SLUA) of Abamectin (122804).  U.S. Environmental Protection Agency, Office of Chemical Safety and Pollution Prevention, Biological and Economic Analysis Division.  Oct. 6, 2010.

USEPA.  2010a.  Water Models.  U.S. Environmental Protection Agency, Pesticides: Science and Policy.  Last updated: Dec. 8, 2010.  Online at: http://www.epa.gov/oppefed1/models/water/

USEPA.  2010b.  STORET.  U.S. Environmental Protection Agency, Office of Water.  Online at: http://www.epa.gov/STORET/

United States Geological Survey (USGS).  2010.  2002 Pesticide Use Maps.  United States Geological Survey, National Water-Quality Assessment (NAWQA) Program, Pesticide National Synthesis Project.  Online at: http://water.usgs.gov/nawqa/pnsp/usage/maps/

USGS.  2010a.  National Water-Quality Assessment (NAWQA) Program.  United States Geological Survey.  Data available through Sep. 30, 2007.  Last updated: Dec. 14, 2010.  Online at: http://infotrek.er.usgs.gov/apex/f?p=NAWQA:HOME:3966441777962223 

Submitted Environmental Fate and Product Chemistry Data

MRID 131083. Maynard, M.; Ku, C.; Gruber, V.; et al. (1982) Hydrolysis of Avermectin B1a (MK-0936). (Unpublished study received Jan 3, 1983 under 618-EX-10; submitted by Merck & Co., Inc., Rahway, NJ; CDL:249152-F)

MRID 159163. MSD AGVET, Inc. (19??) Chemical and Physical Properties of Technical MK-936 (Avermectin B1) and Avid 0.15 EC Miticide/Insecticide for Experimental Use on Citrus. (Compilation; unpublished study received Dec 29, 1983 under 50658-EX-1; CDL:252113-A)

MRID  40856301. Wislocki, P. (1988) Position Paper: Abamectin--Soil Dissipation and Crop Rotation: Supplemental: Project ID: 618-936-SD-CR. Unpublished study prepared by Merck Sharp & Dohme Research Labora- tories. 47 p.

MRID 41191501. Wehner, T. (1989) Additional Soil Residue Data in Support of Applications for Registration of Abamectin Soil Leaching and Dissipation Study: Project ID: Merck No. 001-87-6045R; ADC No. 992. Unpublished study prepared by Merck Sharp & Dohme Research Laboratories and Analytical Development Corp. 18 p.

MRID 42627601. Wertz, P. (1993) Soil Residue Data in Support of an Application for Registrations of Abamectin: Soil Leaching and Dissipation Study 001-87-6045R: A Supplement: Lab Project Number: 618-936-6045. Unpublished study prepared by Analytical Development Corp. 603 p.

MRID 46877502. Friis, M. (2006) Product Chemistry - Abamectin Technical (CHA 204) - EPA Reg. No. 4787 - (none yet) Group B - Physical /Chemical Properties. Project Number: MVF/060500/ABA/204/B/USA, PYC/170, PYC/171. Unpublished study prepared by Cheminova Agro A/S. 399 p.

MRID 47436204. Wagner, J. (2008) Abamectin Technical Herbicide: Stability to Normal and Elevated Temperatures, Metals, and Metal Ions: Oxidation/Reduction: Flammability: Explodability: Storage Stability: Miscibility: Corrosion Characteristics: Dielectric Breakdown Voltage: Viscosity: Boiling Point: Dissociation Constants in Water: Particle Size, Fiber Length, and Diameter Distribution: Partition Coefficient: Water Solubility: Vapor Pressure. Unpublished study prepared by Wagner Regulatory Associates, Inc. 6 p.


MRID 47000503. Clark, A. (2005) Abamectin FS 400 Pre-Treated Tomato Seeds and Abamectin FS 500 Pre-Treated Corn Seeds: Desorption of (Carbon 14) Abamectin Following Applications of Pre-Treated Corn and Tomato Seeds to Various Soils: Final Report. Project Number: T005067/05. Unpublished study prepared by Syngenta Crop Protection. 63 p.


 Appendix A.  SCI-GROW Input/Output Data.


SCI-GROW Input/Output File.

 SciGrow version 2.3
 chemical:Abamectin
 time is  1/10/2011  13:47:22
 ------------------------------------------------------------------------
  Application      Number of       Total Use    Koc      Soil Aerobic
  rate (lb/acre)  applications   (lb/acre/yr)  (ml/g)   metabolism (days)
 ------------------------------------------------------------------------
      0.064           1.0           0.064      4.70E+03    10000.0
 ------------------------------------------------------------------------
 groundwater screening cond (ppb) =   1.58E-03 
 ************************************************************************
 
 SciGrow version 2.3
 chemical:Abamectin
 time is  1/10/2011  15:22:12
 ------------------------------------------------------------------------
  Application      Number of       Total Use    Koc      Soil Aerobic
  rate (lb/acre)  applications   (lb/acre/yr)  (ml/g)   metabolism (days)
 ------------------------------------------------------------------------
      0.064           1.0           0.064      4.70E+03      101.0
 ------------------------------------------------------------------------
 groundwater screening cond (ppb) =   9.14E-04 
 ************************************************************************


 Appendix B.  PE Input Files.

Table B.1.  Summary of PE Input File Names.
                                      Use
                                  File names
                                     Date
                                   Dry beans
Aba_bean.out	Aba_bean_char.out
                                 Jan. 11, 2010
                                  Bulb onions
Aba_onio.out	Aba_onio_char.out
                                       
                                    Chives
Aba_mint.out	Aba_mint_char.out
                                       
                                    Cotton
Aba_cott.out	Aba_cott_char.out
                                       
                                   Tomatoes
Aba_toma.out	Aba_toma_char.out
                                       
                                       
Aba_toma_seed.out	Aba_toma_seed_char.out
                                       


Example Input File (Use on Dry Beans)

Output File: Aba_bean
Metfile:	w14826.dvf
PRZM scenario:	MIbeansSTD.txt
EXAMS environment file:	ir298.exv
Chemical Name:	Abamectin
Description	Variable Name	Value	Units	Comments
Molecular weight	mwt	873.1	g/mol
Henry's Law Const.	henry	1.7e-7	atm-m^3/mol
Vapor Pressure	vapr		torr
Solubility	sol	0.010	mg/L
Kd	Kd	77	mg/L
Koc	Koc		mg/L
Photolysis half-life	kdp	0	days	Half-life
Aerobic Aquatic Metabolism	kbacw	0	days	Halfife
Anaerobic Aquatic Metabolism	kbacs	0	days	Halfife
Aerobic Soil Metabolism	asm	0	days	Halfife
Hydrolysis:	pH 7	0	days	Half-life
Method:	CAM	2	integer	See PRZM manual
Incorporation Depth:	DEPI		cm
Application Rate:	TAPP	0.021	kg/ha
Application Efficiency:	APPEFF	0.95	fraction
Spray Drift	DRFT	0.16	fraction of application rate applied to pond
Application Date	Date	01-08	dd/mm or dd/mmm or dd-mm or dd-mmm
Interval 1	interval	6	days	Set to 0 or delete line for single app.
app. rate 1	apprate		kg/ha
Interval 2	interval	12	days	Set to 0 or delete line for single app.
app. rate 2	apprate		kg/ha
Record 17:	FILTRA	
	IPSCND	1
	UPTKF	
Record 18:	PLVKRT	
	PLDKRT	
	FEXTRC	0.5
Flag for Index Res. Run	IR	Reservoir
Flag for runoff calc.	RUNOFF	total	none, monthly or total(average of entire run)
