  SEQ CHAPTER \h \r 1 

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

WASHINGTON, D. C.  20460

		

	OFFICE OF

PREVENTION, PESTICIDES

AND TOXIC SUBSTANCES



October 23, 2006

									PC Code: 	098301

									DP Barcode:	333309

MEMORANDUM

SUBJECT: 	Drinking Water Exposure Assessment for Total Aldicarb Residues
(Parent, Aldicarb Sulfoxide, and Aldicarb Sulfone) Based on the N-Methyl
Carbamate Cumulative Risk Assessment

FROM: 	Nelson Thurman, Senior Environmental Scientist

		Jonathan Angier, Ph.D.

Environmental Risk Branch 2

		Environmental Fate and Effects Division (7507P)	

		

TO:		Sherrie Kinard, PM

Robert McNally, Branch Chief

		Special Review and Reregistration Division (7508P)

		

		Felicia Fort

		Michael Metzger, Branch Chief

		Health Effects Division 

THROUGH:	Tom A. Bailey, Ph.D.		

		Chief, Environmental Risk Branch 2

		Environmental Fate and Effects Division (7507C)	

This updated assessment, based on the work for the NMC cumulative
assessment, was requested by SRRD and HED. It reflects drinking water
exposure estimates used for the N-Methyl Carbamate (NMC) cumulative risk
assessment (CRA) and is based on typical application rates. The attached
document provided estimated exposures for both ground- and surface-water
sources of drinking water, characterizes the variability in the exposure
estimates, compares estimated concentrations to available monitoring
data, and describes the potential extent of high aldicarb exposure
areas.

Private well monitoring data submitted by Bayer CropScience have not yet
been fully analyzed. Such information will help in the characterization
of the potential extent of aldicarb exposure in private wells.

Drinking Water Exposure Assessment for Total Aldicarb Residues (Parent,
Aldicarb Sulfoxide, and Aldicarb Sulfone) Based on the N-Methyl
Carbamate Cumulative Risk Assessment

  TOC \o "2-3" \h \z \t "Heading 1,1"    HYPERLINK \l "_Toc149464273" 
Summary	  PAGEREF _Toc149464273 \h  3  

  HYPERLINK \l "_Toc149464274"  Groundwater	  PAGEREF _Toc149464274 \h 
3  

  HYPERLINK \l "_Toc149464275"  Surface Water	  PAGEREF _Toc149464275 \h
 5  

  HYPERLINK \l "_Toc149464276"  Ground Water Exposure Assessment	 
PAGEREF _Toc149464276 \h  6  

  HYPERLINK \l "_Toc149464277"  Ground water modeling approach	  PAGEREF
_Toc149464277 \h  6  

  HYPERLINK \l "_Toc149464278"  Estimated total aldicarb residues in
ground water	  PAGEREF _Toc149464278 \h  9  

  HYPERLINK \l "_Toc149464279"  Variability due to well setback
distances	  PAGEREF _Toc149464279 \h  9  

  HYPERLINK \l "_Toc149464280"  Variability due to well depth	  PAGEREF
_Toc149464280 \h  10  

  HYPERLINK \l "_Toc149464281"  Variability due to application rates	 
PAGEREF _Toc149464281 \h  12  

  HYPERLINK \l "_Toc149464282"  Comparisons with monitoring	  PAGEREF
_Toc149464282 \h  12  

  HYPERLINK \l "_Toc149464283"  Lake Wales Ridge, FL, ambient
groundwater monitoring	  PAGEREF _Toc149464283 \h  13  

  HYPERLINK \l "_Toc149464284"  Private drinking water well monitoring
in FL	  PAGEREF _Toc149464284 \h  15  

  HYPERLINK \l "_Toc149464285"  Private drinking water well monitoring
by Bayer CropScience	  PAGEREF _Toc149464285 \h  15  

  HYPERLINK \l "_Toc149464286"  Characterization of the spatial extent
of high potential exposure	  PAGEREF _Toc149464286 \h  15  

  HYPERLINK \l "_Toc149464287"  Comparison of high leaching potential
areas with current aldicarb label setbacks	  PAGEREF _Toc149464287 \h 
18  

  HYPERLINK \l "_Toc149464288"  Surface Water Exposure Assessment	 
PAGEREF _Toc149464288 \h  19  

  HYPERLINK \l "_Toc149464289"  Surface water modeling approach	 
PAGEREF _Toc149464289 \h  19  

  HYPERLINK \l "_Toc149464290"  Estimated total aldicarb residues in
surface water	  PAGEREF _Toc149464290 \h  21  

  HYPERLINK \l "_Toc149464291"  Uncertainty due to application method	 
PAGEREF _Toc149464291 \h  21  

  HYPERLINK \l "_Toc149464292"  Uncertainty due to application rate	 
PAGEREF _Toc149464292 \h  22  

  HYPERLINK \l "_Toc149464293"  Comparisons with monitoring	  PAGEREF
_Toc149464293 \h  23  

 

Summary

This updated drinking water assessment for aldicarb and its major
degradates (aldicarb sulfoxide and aldicarb sulfone) reflects total
aldicarb residue exposures used in the N-methyl carbamate (NMC)
cumulative exposure assessment. While that assessment includes multiple
NMC pesticides, the ground water exposure assessment focused on high
aldicarb use areas and reflects the high potential exposure areas for
aldicarb based on current use. The surface water assessment represents
areas of highest combined NMC use and may not reflect the highest
potential exposure sites for aldicarb alone. Both of the assessments
represent exposures based on typical application rates and do not
reflect potential exposures that can occur when aldicarb is used at
maximum allowable label rates to address pest pressures. The
characterization section addresses variability in exposure estimates
resulting from differing application rates, depths to ground water, and
setback distances between the well and the application area.

At this level of refinement, EFED focused on those areas that have the
greatest potential for exposure to aldicarb. As noted in the preliminary
NMC cumulative risk assessment, total aldicarb residues are not expected
to occur at levels that will contribute to dietary exposures for most of
the country (available at   HYPERLINK
"http://www.epa.gov/scipoly/sap/meetings/2005/august/preliminarynmc.pdf"
 http://www.epa.gov/scipoly/sap/meetings/2005/august/preliminarynmc.pdf 
). However, total aldicarb residues estimated for vulnerable private
wells in some areas of Florida (primarily along the central ridge) and
high leaching potential areas in the southeastern coastal plain approach
levels that may be a concern for total dietary exposures. These areas
represent a relatively small area of the country where the estimated
ground water residues are reasonable estimates of drinking water
exposure for residents who get their drinking water from shallow private
wells.

Groundwater

The estimated distributions of total aldicarb residues (parent plus the
sulfoxide and sulfone transformation products) in ground water reflect:

Shallow (30-ft) private wells. Concentrations will vary with varying
depths to ground water and well depths. Higher concentrations would be
expected in more shallow wells while lower concentrations would be
likely in deeper wells.

High leaching potential soils (as classified by the USDA Natural
Resources Conservation Service), with similarly permeable conditions
extending through the vadose zone to ground water. Soil and vadose zone
permeability will vary across the landscape and with depth. Where the
vadose zone includes low-permeability layers, concentrations will be
lower.

Aldicarb applications to fields at label setback distances between the
field of application and the well, as specified on the current aldicarb
label. Since travel time varies with well depth, a longer setback
distance would be needed for shallower wells while a shorter setback
distance might be sufficient for deeper wells. 

A high-end typical lateral flow velocity to estimate the travel time
from the field of application to the well based on well setback
distance. This assumes that the ground water is flowing from the field
of application toward the well.

Typical application rates for aldicarb, provided by the Biological and
Economic Analysis Division (BEAD). Concentrations would be greater if
maximum label rates are used.

Acidic soil and ground water, which favor the persistence of the
sulfoxide and sulfone transformation products (both degrade rapidly
under alkaline conditions; the parent aldicarb is less susceptible to
alkaline hydrolysis). 

Table 1 summarizes the estimated distributions of total aldicarb
residues for various crop/site scenarios and label-specified well
setback distances based on typical application rates. While the soils
along the central ridge of Florida are typically more vulnerable to
leaching in comparison to other regions, estimated total aldicarb
residues used for dietary exposures are less than for the Georgia
coastal plain scenario because of a greater setback distance (1000 ft
for citrus in FL compared to 300 ft in GA).

Table 1: Estimated concentrations of total aldicarb residues in private,
shallow (30-ft) wells in high leaching potential soils in FL and the
southeastern Coastal Plain based on typical rates.

Scenario	Well setback distance	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile	50th %ile

FL central ridge citrus	1000 ft	3.0	2.8	2.6	2.4	2.1	2.0	1.7

GA Coastal Plain Peanuts/ cotton	300 ft	6.5	6.0	5.1	4.8	4.3	4.1	3.1

NC Coastal Plain Peanuts/ cotton	300 ft	1.3	1.2	1.1	1.0	0.9	0.8	0.6



The estimated ground water concentrations are not national numbers but
are reasonable for people living in those vulnerable areas who get their
drinking water from shallow private wells. The exposure estimates
represented by these scenarios are limited to these crops and
conditions:

The Florida central ridge citrus scenario represents those high leaching
potential soils that occur along the central ridge of Florida and are in
citrus production. Outside of the central ridge, the soils are typically
less permeable and/or the ground water is alkaline (pH>7) and aldicarb
residues are not expected to persist.

The Georgia coastal plain peanuts/cotton scenario represents high
leaching potential soils in the southern coastal plain (GA, AL, SC)
underlying cotton and peanuts. Although aldicarb use on pecans was not
included in this assessment, it is likely that, under the same high
leaching potential soils, total aldicarb residues in private wells from
pecan use would be greater than estimated for peanuts because of a
higher typical application rate (3.22 kg/ha for pecans compared to 1.10
kg/ha for peanuts). Pecans were not modeled because, according to BEAD
estimates, only a small portion of pecan acres (1%) are treated with
aldicarb.

The North Carolina coastal plain peanuts/cotton represents high leaching
potential soils in the northern coastal plain (NC, VA, Delmarva
pensinsula) underlying peanuts or cotton. 

The ground water exposure represents private drinking water wells. The
Agency assumed in this assessment that, in general, public water
supplies supplied by ground water will typically draw from deeper
aquifers and/or aquifers that have a relatively impermeable layer
between the surface and the water supply. Such supplies are expected to
be much less vulnerable to pesticide contamination. Public water
supplies have a higher probability of being treated, although
conventional treatments processes are likely to result in little or no
reduction of aldicarb residues in water. However, where lime softening,
which will accelerate pH-dependent hydrolysis for aldicarb sulfoxide and
sulfone, or activated carbon filtration is used, some reduction in
aldicarb residues between untreated and treated water may occur.

Surface Water

The estimated surface water exposures reflect areas of high combined NMC
use, but not necessarily the highest aldicarb use areas. An evaluation
of the scenarios used for the NMC CRA assessment indicate that three
scenarios – FL central ridge (citrus), NC coastal plain (peanuts,
cotton), and LA/MS Mid-south (cotton) – coincide with high aldicarb
use areas where drinking water intakes are likely to occur.

For the NMC CRA, OPP used typical application rates and acres treated
because of a low likelihood that all of the NMC pesticides will be used
at maximum rates on all of the crop acreage in a watershed at the same
time. This assumption does not hold up for the individual pesticides
because high exposures are likely where the pesticide is applied in
response to pest pressures, often treating the entire crop on a local
area. The typical application rates and percent acres treated are
derived from state-level survey data and assume uniform use practices
across the state. Indeed, an uneven distribution of application rates
and percent acres treated is expected in response to differing pest
pressures. This assumption will underestimate areas where pest pressures
may dictate a higher percentage of acres treated in a given year. Thus,
for this aldicarb assessment, OPP assumed that all of the crop acres in
the watershed were treated. While reported acres treated ranged from 14%
for citrus in FL to 63% for peanuts in NC, the use data do not allow for
a reasonable upper bound estimate on percent treated in localized areas
across the state.

The estimated surface water concentrations (Table 2) represent total
aldicarb residues reservoirs located in high use areas which coincide
with relatively high runoff conditions. They reflect typical application
rates, but assume that the entire crop in the watershed is treated.

Table 2: Estimated concentrations of total aldicarb residues in surface
water sources of drinking water in high runoff potential areas based on
typical rates.

Scenario Location	Crops	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile

FL central ridge citrus	Orange, grapefruit	10.2	1.6	0.3	0.1	0.02	0.01

NC Coastal Plain 	Peanuts, cotton	4.6	1.0	0.2	0.1	0.01	0.005

LA/MS Mid-south	Cotton	0.8	0.2	0.02	0.004	<0.001	<0.001



The variability in concentrations from year to year in the surface water
distributions reflect the variability due to varying weather patterns
over 30 years of actual weather data. Because these scenarios target
high aldicarb use areas associated with drinking water intakes in high
runoff areas, they represent high end exposure sites. Total aldicarb
residues in surface water sources in the rest of the country are
expected to be lower.

The sections that follow include a brief write-up of the ground water
modeling approach used to estimate total aldicarb residues in ground
water, a summary of the results and a comparison against recent
monitoring, and a characterization of the spatial extent of the
potential high exposure areas. More detail on the ground water
conceptual model and the model approach can be found in the NMC
cumulative risk assessment (FIFRA SAP, 2005). The surface water modeling
approach is also described. 

Ground Water Exposure Assessment

Ground water modeling approach

Although previous drinking water exposure assessments for aldicarb
relied on a summary of available monitoring data, the vast majority of
the monitoring represents unknown conditions (in particular, no
information on aldicarb rates, distances between fields and wells,
ground water depth, type of well, soil or hydrogeologic conditions, or
ground water pH) and represented monitoring prior to label changes.
Bayer CropScience has recently submitted a compilation of recent
monitoring of private wells in selected areas of the US. Although EPA
has not yet had time to fully evaluate the monitoring (in particular,
the correlation of aldicarb detects with high leaching potential soils,
distance between field and well, depth to ground water, and nature of
well), a brief review of study results indicates that the estimated
exposures reported below are on the same order as reported detections. 

EPA used the Pesticide Root Zone Model (PRZM) to simulate transport
processes through high leaching potential soils to a shallow unconfined
aquifer with a water table at 30 feet (approximately 9 m) below the
surface. The well screen extended an additional 1 m below the water
table. While information on typical depths of private wells is not
readily available, USGS NAWQA and ground water atlases suggest that 30
feet is a reasonable depth for shallow ground water supplying private
wells in the southeastern coastal plain and Florida.  

The well concentration is the average pore water concentration across
the length of the screen. PRZM was set up to deliver the average pore
water concentration in the ‘saturated’ soil profile in the upper
meter of the ground water zone. 

The modes and rates of degradation for aldicarb residues changed through
the soil profile. EPA used the pesticide aerobic soil metabolism rate
for the top 25 cm, linearly decreasing the rate with depth to 1 m. Below
that, EPA used hydrolysis. Table 3 summarizes the pertinent aldicarb
properties used for this assessment. These properties came from an
evaluation of registrant-submitted studies. Other chemical properties
are required as inputs, but have negligible effect on model output;
these properties can be found in the model input files as well as in the
RED. Properties for aldicarb represent total residue (parent aldicarb,
plus the degradates aldicarb sulfoxide and aldicarb sulfone) properties.

Table 3: Summary of aldicarb fate and transport properties for leaching.


Input Parameter	Value	Reference/Comment

Kd	Kd = 0.12 mL/g 

(Koc = 10 mL/g)	Value for aldicarb sulfone (MRID 43560302)

Hydrolysis	Parent degrades slowly at pH9

Sulfoxide 2-3 days @ pH9

Suflone 60-63 days @ pH7, 6 days @pH8, 1 day @ pH9	Parent hydrolyzed
only at pH 9 (MRID 00102065) – degradates hydrolyze more rapidly at
neutral-to-high pH

Soil Half-life	55 days for total aldicarb residues for the top 25 cm;
decreased linearly from 25 to 100 cm	Revised from 2001 Aldicarb RED;
Upper 90th pct bound on mean for combined parent+sulfoxide+sulfone
half-life from 19 soils

Additional Notes	Modeled total aldicarb residues	Half-life values used
in inputs based on combined aldicarb + sulfone + sulfoxide residues;
lowest Kd of the 3 chemicals used for mobility. Assumes equal toxicity
of parent, degradates



Aldicarb was modeled using a unit rate of 1 kg ai/ha for each scenario.
Because the exposure concentrations are linearly related to the
application rate, the resulting rates were multiplied by the typical
application rates (Table 4). The maximum label application rates are
also provided in contrast. If these rates are used, the concentrations
would increase proportionally.

Table 4: Summary of aldicarb seasonal application rates used for ground
water exposure modeling.

Location	Crop	PRZM scenario	App. Rate (kg/ha)1	Max. Label Rate (kg/ha)1
App. Date	Well setback (ft)

Southeast coastal plain/ NC	Peanut, Cotton	NC Cotton	1.10 (peanut)	5.55
(cotton); 3.66 (peanut)	10-Apr	300

Southeast coastal plain/ GA	Peanut, Cotton	GA Peanut	1.00 (peanut)	5.55
(cotton); 3.67 (peanut)	10-Apr	300

FL Central Ridge	Oranges / grapefruit	FL Citrus	4.27	5.49	1-Apr	1000

Northeast FL	Potato	FL potato	0.80	3.66 	20-Jun	300

Central WA      	Potato	WA potato	3.21 	3.66	15-May	300

1 Application rates, typically reported in pounds per acre (lb/A), has
been converted to kg/ha for modeling purposes. 

Because aldicarb labels specify well setback distances (varying with
soil type and among states), EPA used a plug flow model to simulate the
additional travel time for a pesticide to reach a drinking water well
from point of application. This is explained in detail in the
preliminary NMC cumulative assessment (USEPA/FIFRA SAP, 2005). 

Well setback distances result in additional travel time for the chemical
to move laterally to the well. This results in additional degradation.
Reductions in concentration are calculated in these assessments by a
plug flow approximation:

 

where 	C = concentration at well [mass/volume]

		C0= concentration at point of application [mass/vol]

		L = well setback distance [length]

		v = lateral groundwater velocity [length/time]

		k = degradation rate in aquifer [time-1]

Table 5 provides estimated travel times and concentration reduction
factors for varying well setback distances. For the ground water
exposure assessment, EPA used the reduction factor associated with the
corresponding well-setback distance on the aldicarb label (1000 ft for
citrus in FL; 300 ft for other crops in FL and in the southeastern
coastal plain). The reduction factor is based on a typical high-end
lateral groundwater velocity of 0.15 m/da (0.49 ft/da) for the FL
central ridge as reported by Russell et al, 1987.  Other studies
reported velocities ranging from 0.09 to 0.27 m/da (Paramasivam et al,
1999). Based on that range in groundwater velocities, resulting
concentration reduction factors for a 300-foot setback distance ranged
from 43% lower (0.244) to 46% higher (0.625) than what was calculated
with the 0.15 m/da velocity. 

Table 5. Travel time and concentration reduction factors for varying
well setback distances for aldicarb (based on 500-day half-life for
hydrolysis @ pH5).

Setback distance (ft)	Travel time (da)	Reduction factor

100	203	0.754

300	610	0.429

500	1016	0.244

1000	2033	0.060



Estimated total aldicarb residues in ground water

Table 6 summarizes the distributions of total aldicarb residues for
various scenarios and varying well setback distances. The distributions
represent 25 years of simulations. The distributions highlighted in bold
in the table represent estimated residues for the labeled setback
distance between the treated field and the well. The 0-foot
“setback” estimates in-field concentrations which were used to
compare model estimates with in-field ground water monitoring data. 

Table 6: Estimated concentrations of total aldicarb residues in private,
shallow (30-ft) wells. Concentrations represent typical application
rates in high leaching potential soils.

Scenario	Well setback	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile	50th %ile

FL Central Ridge/ Citrus	0 ft	58.5	55.5	50.9	48.5	41.8	40.3	33.8

	300 ft	24.9	23.6	21.6	20.6	17.8	17.2	14.4

	1000 ft	3.0	2.8	2.6	2.4	2.1	2.0	1.7

FL Potatoes (alkaline GW)	0 ft	3.9e-05	3.0e-05	1.9e-05	1.3e-05	8.1e-06
6.2e-06	2.3e-06

	300 ft	1.7e-05	1.3e-05	8.0e-06	5.7e-06	3.5e-06	27.e-06	9.9e-07

GA Coastal Plain Peanuts/ cotton	0 ft	15.2	14.1	12.0	11.2	10.1	 9.6	 7.2

	300 ft	 6.5	 6.0	 5.1	 4.8	 4.3	 4.1	 3.1

	500 ft	3.7	3.4	2.9	2.7	2.5	2.4	1.8

	1000 ft	0.9	0.8	0.7	0.7	0.6	0.6	0.4

NC Coastal Plain Peanuts/ cotton	0 ft	 3.1	 2.9	 2.5	 2.3	 2.0	 2.0	 1.5

	300 ft	 1.3	 1.2	 1.1	 1.0	 0.9	 0.8	 0.6

	500 ft	0.8	0.7	0.6	0.6	0.5	0.5	0.4

	1000 ft	0.2	0.2	0.2	0.1	0.1	0.1	0.1

WA Potato (alkaline soil, GW)	300 ft, 15-ft depth	0.001	0.001	<0.001
<0.001	<0.001	<0.001	<0.001



Variability due to well setback distances

The exposure estimates for total aldicarb residues assume that aldicarb
is applied at the appropriate setback distances specified on the label
– 300 feet between the field of application and the well for most
crops; 1000 feet for citrus in Florida. Figure 1 illustrates the
estimated effect of varying well setback distances on total aldicarb
residues in the southern coastal plain of Georgia. Similar responses can
be seen in the other regional scenarios (Table 6). The graph plots
estimated concentrations of total aldicarb residues in a 30-foot well
located near treated peanut fields in high leaching potential soils in
the coastal plain. 

Figure   SEQ Figure \* ARABIC  1 : Estimated concentrations of total
aldicarb residues in groundwater at 30 feet under peanuts in the Georgia
Coastal Plain with varying setback distances.

The Agency is not aware of studies that would quantify the effect of
varying setback distances on concentrations of total aldicarb residues
in ground water. Thus the relative differences in concentration
magnitudes based on varying setback distances shown in the figure are
more illustrative than quantitative. The net effect of the setback
distances will vary depending on predominant direction and velocity of
ground water flow. Other factors, such as the influence of irrigation
wells on shallow ground water flow, are not easily quantifiable and are
not accounted for in this assessment.

Variability due to well depth

The estimated drinking water exposures for shallow private wells are
based on a ground water depth of 30 feet. For comparisons, OPP used PRZM
to estimate concentrations from wells drawing from 15, 30, and 50 feet
(Table 7). The 30-50 foot depths are more representative of shallow
private wells, while the 15-foot depth may represent shallow dug wells.
With the increased travel time allowing for more degradation, estimated
aldicarb residues decreased by nearly an order of magnitude between 15
and 30 feet (Table 7). Estimated concentrations at 50 feet were
approximately 3 times lower than those at 30 feet.

Table 7: Estimated concentrations of total aldicarb residues in private,
shallow wells of varying depths. Concentrations represent typical
application rates in high leaching potential soils.

Scenario	Well Depth, ft	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile	50th %ile

FL Central Ridge/ Citrus (1000’ setback)	15	32.6	30.3	26.2	23.7	21.7
20.7	16.6

	30	3.0	2.8	2.6	2.4	2.1	2.0	1.7

	50	0.9	0.9	0.8	0.7	0.6	0.6	0.5

GA Coastal Plain Peanuts/ cotton (300’ setback)	15	36.8	32.7	28.6	25.9
23.1	21.6	14.0

	30	 6.5	 6.0	 5.1	 4.8	 4.3	 4.1	 3.1

	50	2.0	1.9	1.8	1.7	1.5	1.4	1.1

NC Coastal Plain Peanuts/ cotton (300’ setback)	15	17.2	16.5	14.0	12.5
9.5	8.7	7.4

	30	 1.3	 1.2	 1.1	 1.0	 0.9	 0.8	 0.6

	50	0.2	0.2	0.2	0.2	0.2	0.2	0.1



Estimated concentrations at the 15-foot depth showed more variation in
concentrations over time (Figure 2), reflecting the shorter time frame
for movement. Seasonal and yearly variations due to weather are less
pronounced with depth. This is consistent with monitoring data which
show that mobile chemicals applied to the surface of highly permeable
soils can reach shallow groundwater in the same season or year.

Figure   SEQ Figure \* ARABIC  2 . Estimated concentrations of total
aldicarb residues at different depths in groundwater near citrus in the
Central Ridge of Florida (1000 foot setback from well).

Variability due to application rates

For the estimated concentrations for total aldicarb residues in private
wells, based on the NMC CRA, typical application rates specific to the
regional scenarios (reported at the state level) were used. These
typical distributions have been provided for the aldicarb aggregate
exposure assessment. For ground water 30 to 50 feet or more below the
surface, typical transport time for aldicarb residues is likely to span
multiple seasons. While the typical rates used in the exposure
assessment may reflect the integrated application of aldicarb over
multiple seasons, it will underestimate the amount of aldicarb residue
available for transport in high application seasons. 

Table 8 contrasts estimated concentrations of total aldicarb residues
for the three high exposure scenarios based on both typical and maximum
label application rates. The range in concentrations between typical and
maximum application rates reflects the potential variability in high-end
exposures based on application. Obviously, the total aldicarb load
available for transport to ground water would be lower in those years
when little or no aldicarb is applied.

Table 8: Estimated concentrations of total aldicarb residues in private,
shallow wells (30 ft) in high leaching potential soils contrasting
typical and maximum application rates.

Scenario	App. Rate, kg/ha	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile	50th %ile

FL Central Ridge/ Citrus (30’ depth; 1000’ setback)	5.5 (max)	4.5
4.3	3.9	3.7	3.4	3.1	2.6

	4.3 (typ)	3.0	2.8	2.6	2.4	2.1	2.0	1.7

GA Coastal Plain Peanuts/ cotton (30’ depth; 300’ setback)	3.3 (max)
19.8	18.4	15.7	14.5	13.2	12.5	9.4

	1.0 (typ)	 6.5	 6.0	 5.1	 4.8	 4.3	 4.1	 3.1

NC Coastal Plain Peanuts/ cotton (30’ depth; 300’ setback)	3.3 (max
– peanut)	4.2	3.9	3.4	3.0	2.8	2.6	2.0

	1.1   (typ - peanut)	 1.3	 1.2	 1.1	 1.0	 0.9	 0.8	 0.6

	5.5 (max – cotton)	6.9	6.4	5.6	5.0	4.7	4.3	3.3

	0.7   (typ – cotton)	0.9	0.8	0.7	0.7	0.6	0.6	0.4



Comparisons with monitoring

OPP compared estimated aldicarb concentrations from PRZM modeling to two
recent groundwater monitoring datasets from Florida. The first study,
conducted by the USGS and the Florida Department of Agriculture,
measured aldicarb concentrations in monitoring wells located in citrus
groves along the Central Ridge of Florida (Lake Wales Ridge). These
monitoring wells are not drinking water wells, but reflect ambient
pesticide concentrations in ground water beneath the citrus groves.
Since these wells are located within the treated fields, OPP used PRZM
concentrations with no well setback adjustments (0-ft well setback) for
comparisons to the monitoring data. 

The second monitoring dataset consists of private well monitoring data
collected by the FL Department of Environmental Protection across the
state of Florida. While the data represent potable drinking water wells,
no information is available on well depth, aldicarb use in the vicinity,
or distance between the well and the treated field. A third monitoring
set recently submitted by Bayer CropScience provides recent (2005)
monitoring of aldicarb residues in private drinking wells in other parts
of the US. These data are currently being analyzed.

Lake Wales Ridge, FL, ambient groundwater monitoring 

In-field concentrations (0-ft well setback) of estimated total aldicarb
residues from the FL Central Ridge Citrus scenario were compared to an
on-going groundwater monitoring study on the Florida Central Ridge ( 
HYPERLINK "http://fisc.er.usgs.gov/Lake_Wales_Ridge/" 
http://fisc.er.usgs.gov/Lake_Wales_Ridge/  ). The USGS and the Florida
Department of Agriculture is monitoring 31 wells within and around
citrus groves on the Ridge (the area of the OPP scenario). Well depths
range from 4 feet to 110 feet deep (two thirds in the 20 to 60 foot
range), and pH ranged from 3.9 to 6.9 (median about 5). Concentrations
as high as 23 ppb have been recorded in one 26-ft well, while a 4-ft
well had reported concentrations as high as 21 ppb. This study is not
targeted for any specific pesticide, but rather is designed as a survey
mechanism—that is, it is not known how much aldicarb was used nor is
it known how far aldicarb was used from the wells. 

Figure 3 compares the monitoring results from the Lake Wales Ridge study
with PRZM-modeled estimated aldicarb residues at a 50-foot well depth.
The PRZM estimates are shown in pale yellow and represent roughly 25
years of simulations. When compared to wells of similar depth (TURKEY,
ARBUCKL, MTNLKN, NLKPATK), the estimated exposures are in the same
concentration range. While the median estimated concentrations for total
aldicarb residues at a 30-foot well depth were typically greater than
those found in the wells at similar depths (GLENNST, JACKS2), the
measured detections were still within the range of estimated
concentrations. 

Figure   SEQ Figure \* ARABIC  3 : Comparison of estimated
concentrations of total aldicarb residues in groundwater at 50 feet
(top, yellow) and 30 feet (bottom, blue) with in-field monitoring from
the USGS/FL Dept. of Ag. Lake Wales Ridge monitoring study.

Private drinking water well monitoring in FL

The Florida Department of Environmental Protection (FDEP) monitors
private drinking water wells in rural areas. The monitoring is not
comprehensive, but instead is instituted when there has been an
indication of a problem (personal communication, FDEP). Total aldicarb
residues (parent, sulfoxide and sulfone degradates) as high as 47 ppb
were reported in private drinking water wells in the early 1990s in the
FDEP study. The concentrations dropped off in subsequent years. The
reduction in concentrations of aldicarb may have resulted from label
changes which reduced application rates and applied well setback
requirements. Specific reductions at home sites also were also likely
the result of a Florida State program to install carbon filters or to
pipe water in from treatment facilities when contamination was found.
Other reasons for the decline include the possibility of discontinued
use in the vicinity of the contaminated areas (personal communication
FDEP) or increased method detection limit.  

Method detection limits (MDL) for aldicarb residues vary over time in
this monitoring study. In 1999 and earlier, the MDL for aldicarb sulfone
and aldicarb sulfoxide ranged from 0.077 to 0.73 ug/L. Between 2000 and
2004, the MDL ranged from 2.1 to 3.3 ug/L for aldicarb sulfone and from
2.4 to 4.0 ug/L for aldicarb sulfoxide. Estimated concentrations for
total aldicarb residues are below the high MDL for the individual
degradates. This further complicates interpretations regarding the
effectiveness of label changes in reducing aldicarb residues in private
wells.

Private drinking water well monitoring by Bayer CropScience

Bayer CropScience recently submitted retrospective groundwater
monitoring studies for aldicarb and its metabolites for the southeastern
US, Mississippi Delta, Pacific Northwest, California, and Texas. The
studies included 1,673 drinking water wells and collected information on
groundwater depth, well depth, casing depth, well type and age, soil
types, recent aldicarb use history, crops, and distance of the well from
the treated field. The Agency has not completed its review of these
studies. However, a preliminary evaluation of the results of the
monitoring for the southeastern US indicates that roughly 16% of sampled
wells had aldicarb residues above the method detection limit, with a
maximum reported concentration of 2.9 ug/L. The precursory review
suggests that the estimated aldicarb exposures from the PRZM simulations
are in line with the reported monitoring results.

Characterization of the spatial extent of high potential exposure

Three of the scenarios – FL citrus on the Central Ridge, NC
peanuts/cotton/tobacco on the coastal plain, and GA peanuts/cotton on
the coastal plain – represents areas with the potential for high total
aldicarb residue concentrations as a result of use, high leaching
potential, and acidic groundwater. Aldicarb residues are not expected to
persist in sites represented by WA potatoes and FL potatoes because
alkaline soil and/or aquifer conditions favor more rapid hydrolysis of
the aldicarb residues. 

In other regions of the country, anticipated exposure to aldicarb
residues in groundwater is expected to be lower than estimated in these
scenarios or in surface water scenarios representing those regions. In
the north and north-central, groundwater exposures are expected to be
lower because of low use or because aldicarb is no longer labeled for
use, particularly in northeastern states where acidic conditions and
high leaching potential conditions exist. In the mid-south, drinking
water is drawn predominantly from public ground water supply from deep,
protected aquifers and aldicarb contamination not expected. In the Great
Plains and lower Midwest, anticipated exposure is expected to be lower
because of low rainfall and deeper aquifers than in the southeast and
Florida. In many parts of the west, alkaline soils and/or ground water
will also facilitate the rapid breakdown of aldicarb residues into
non-toxic degradation products.

Figure 3 illustrates the spatial extent of high leaching potential soils
(shown in red) in the aldicarb use areas (shown in green) in the
southeastern US. Although county-level soil information is not available
for the entire region, such soils can be identified and included on the
aldicarb label. While the map includes acidic, high leaching potential
soils, it does not reflect depth or pH of the ground water, or the
relative permeability of the underlying vadose zone and aquifer. Neither
is information on the location or depth of private drinking water wells
available.

The scenarios represent potential high exposure areas for private wells:

High leaching potential soils (as classified by the USDA NRCS)

Shallow depth to ground water (roughly 30 feet, though the depth may
vary, depending on the permeability of the overlying soil and vadose
zone)

Acidic soils and ground water

Figure   SEQ Figure \* ARABIC  4 : Areas of high leaching potential
(red) soils in aldicarb use areas (green) in the southeast US.

Comparison of high leaching potential areas with current aldicarb label
setbacks

The current label for aldicarb specifies setback distances between the
field of application and drinking water wells based on soil properties,
water table depth, and well condition. Depending on the state, soils for
which restrictions apply are identified as having either

Loamy sand or sand surface soils and subsoils with an average organic
matter in the upper 12 inches of less than 2% by weight (primarily in
the south and southeast), or 

Sandy loam, loamy sand, or sand surface soils, and loamy sand or sand
subsoils, with an average organic matter in the upper 12 inches of less
than 2% by weight (primarily in the Midwest).

Soil texture and organic matter content act as surrogates for
identifying those soils that have a high potential for leaching and a
low capacity for retaining the pesticide. While these properties are
relatively simple to describe and can be identified with a knowledge of
the soil series mapped for the treated fields, they do not fully
integrate those soil properties that affect leaching potential and are
imperfect indicators of the potential for aldicarb residues to move to
groundwater. 

For this assessment, OPP used the USDA NRCS classification for soil
leaching potential for pesticides (NRCS. 2003. Florida NRCS Field Office
Technical Guide. Section II. Water Quantity and Quality. Available from
the electronic FOTG site at   HYPERLINK
"http://efotg.nrcs.usda.gov/treemenuFS.aspx" 
http://efotg.nrcs.usda.gov/treemenuFS.aspx  ).  Rating criteria are
provided in Table 7. The predicted exposure estimates for total aldicarb
residues represent properties of soils identified as having a high soil
leaching potential for pesticides. The soil leaching potential rating
has been derived for all soils in the SSURGO county surveys for Florida
(available for download from the USDA NRCS Soil Data Mart at   HYPERLINK
"http://soildatamart.nrcs.usda.gov/"  http://soildatamart.nrcs.usda.gov/
). OPP used the criteria in Table 9 to estimate the leaching potential
for soils in counties in other states (for the NMC cumulative, OPP
focused on coastal plain soils in VA, NC, SC, GA, and AL). The criteria
could be used to identify soils that are vulnerable to pesticide
leaching throughout the country. This list of soils should provide a
more definitive list of vulnerable soils than the current list of soils
based solely on soil texture and organic matter content.

Table 9: USDA NRCS Criteria Used for Soil Leaching Potential for
Pesticides (NRCS, 2003).

Rating	Criteria

High	Hydrologic Group = A and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon <= 30 or

Hydrologic Group = B and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon <= 9 and the K Factor is <= 0.48 or

Hydrologic Group = B and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon <= 15 and the K Factor is <= 0.26

Low	Hydrologic Group = B and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon >= 35 and the K Factor is >= 0.40 or

Hydrologic Group = B and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon >= 45 and the K Factor is >= 0.20 or

Hydrologic Group = C and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon <= 10 and the K Factor is >= 0.28 or

Hydrologic Group = C and % Surface Horizon Organic Matter Content X
Depth of the First Soil Horizon >= 10

Very Low	Hydrologic Group = D

Intermediate	All other conditions



The exposure estimates for total aldicarb residues in private wells were
modeled at a 30-foot depth. Current label restrictions for aldicarb
apply if vulnerable soils are present and the water table is less than
25 feet below the ground surface. Exposure in private wells is a
function of depth to ground water/ well screen, permeability of the
overlying soil and vadose zone, the amount of precipitation in excess of
evapotranspiration (to leach the chemical through the soil and vadose
zone) distance between the field of application and the well, and the
direction and velocity of lateral ground water flow. No single ground
water depth provides a bright line between vulnerable and not
vulnerable. The current label restrictions do not reflect the true range
in vulnerability with depth.

  SEQ CHAPTER \h \r 1 Surface Water Exposure Assessment

The revised surface water exposure assessment focused on three high
aldicarb use/ exposure scenarios: Florida citrus (central FL),
Mississippi cotton, and North Carolina peanuts/cotton.  While these
scenarios were selected based on combined NMC uses in the vicinity of
drinking water intakes in relatively high runoff potential areas, they
represent areas of relatively high aldicarb use. Thus, the scenarios
represent drinking water intakes with relatively high potential for
aldicarb exposure.

The conceptual model for surface water exposure, scenario selection, and
modeling approach is well-documented in the preliminary NMC Cumulative
Risk Assessment (available at   HYPERLINK
"http://www.epa.gov/scipoly/sap/meetings/2005/august/preliminarynmc.pdf"
 http://www.epa.gov/scipoly/sap/meetings/2005/august/preliminarynmc.pdf 
) and will not be duplicated here. Only those aspects of the aldicarb
aggregate exposure assessment that differ from the NMC CRA are
documented.

Surface water modeling approach

This revised surface water exposure assessment used the same chemical
input parameters that have been previously documented in the revised RED
for aldicarb and in previous drinking water exposure assessments. These
values, which represent the combined residues of parent aldicarb and
sulfoxide and sulfone degradates, are documented in Table 10.

  SEQ CHAPTER \h \r 1 Table 10: Aldicarb-specific PRZM-EXAMS Input
Parameters. Input Parameter	Value	Reference/Comment

Molecular Weight 	190.2 g/mol	MRID 00152095

Henry’s Law Constant	1.7 E-10 atm-m^3/mol	Acc 255979

Vapor Pressure	1 E-6 @ 25°C	MRID 00152095

Solubility	6,000 mg/L	Acc 255979

Kd	0.12	Minimum non-sand value for aldicarb sulfone (MRID 43560302)

Hydrolysis	pH 5, stable (0)

pH 7, stable (0)

pH 9, >30 days	Parent hydrolyzed only at pH 9 (MRID 00102065) –
degradates may hydrolyze more rapidly at neutral-to-high pH

Aqueous Photolysis Half-life	4 days	MRID 42498201

Water Half-life	12 days 	MRID 44592107. Single acceptable guideline
study for parent / sulfoxide / sulfone (4days) x 3; corresponds w/ DT90

Benthic Half-life	24 days	No data; use 2X aerobic aquatic half-life

Soil Half-life	55 days	Revised from 2001 Aldicarb RED; Upper 90th pct
bound on mean for combined parent+sulfoxide+sulfone half-life from 19
soils

FILTRA, UPTKF, PLVKRT, PLDKRT	0	Default values

FEXTRC	0.5	Default value

Additional Notes	Modeled total aldicarb residues	Half-life values used
in inputs based on combined aldicarb + sulfone + sulfoxide residues;
lowest Kd of the 3 chemicals used for mobility. Assumes equal toxicity
of parent, degradates



The NMC CRA used region-specific typical application rates (see 2005 NMC
CRA for documentation of the use information). These rates, along with
the number of applications are less than the maximum label rates. While
typical rates, representing an “average” of high and low pest
pressures over time, might be reflective of ground water exposures, in
which the length of time for transport from the surface to groundwater
tends to lessen the variability in concentrations over time, they are
more likely to underestimate surface water concentrations in the case of
maximum use in response to high pest pressures. Likewise, they will
likely overestimate surface water concentrations in the case of low pest
pressures.  Table 11 documents the application-related parameters used
as inputs for PRZM-EXAMS in this revised surface water assessment.

  SEQ CHAPTER \h \r 1 Table 11: Application-specific PRZM-EXAMS Input
Parameters. 

Scenario Location	NC Coastal Plain	Central Florida	MS/LA

Crop/Use	Cotton	Peanut	Oranges	Grapefruit	Cotton

Counties in watershed area (used to collect use, crop data)	Edgecombe,
Halifax, Northhampton, NC	Polk, Hillsborough, Manatee, FL	Franklin,
Madison, Tensas, LA

Typical App. Rate, lb ai/A	0.73	1.08	3.85	3.89	0.53

Typical App. Rate, kg/ha 1	0.81 (0.12)	1.20 (0.18)	4.27 (0.64)	4.32
(0.65)	0.59 (0.09)

Max. App. Rate, kg/ha	4.5 (0.82)	3.7 (0.5)



	Most Active Range	May 1-15	Apr10-20	Apr-Nov	Apr-Nov	May1-15

App. Date	1-May	10-Apr	1-Apr	1-Apr	1-May

No. Apps.	1.0	1.0	1.5	1.5	1.0

Interval between apps. (da)	na	na	121	121	na

PCA for Region	0.61	0.61	0.19	0.19	0.20

Crop ratio 2	0.82	0.18	0.94	0.06	1.00

1 – Application rate was multiplied by 0.15 (shown in parentheses) to
reflect 15% of granules on the surface for CAM 1 application. 

2 – Crop ratio is based on the relative ratio of crops in the scenario
watershed, based on USDA 2002 AgCensus data for the counties in the
scenario area. Each crop-use is run with PRZM/EXAMS. The resulting
concentration distribution is multiplied by the crop ratio before
summing the daily distributions to represent total aldicarb residues
from all contributing uses in the watershed.



Estimated total aldicarb residues in surface water

Table 12 summarizes the distributions of total aldicarb residues for
various scenarios, reflecting the relative contributions of aldicarb
from multiple crop uses in the watershed. 

Table 12: Estimated concentrations of total aldicarb residues in surface
water sources of drinking water in high runoff potential areas based on
typical rates.

Scenario Location	Crops	Concentrations, ug/l



Max-imum	99th %ile	95th %ile	90th %ile	80th %ile	75th %ile

FL central ridge citrus	Oranges	9.6	1.5	0.24	0.08	0.014	0.007

	Grapefruit	0.6	0.1	0.02	0.005	0.001	0.0005

	Aggregate	10.2	1.6	0.26	0.85	0.015	0.007

NC Coastal Plain 	Cotton	4.5	1.0	0.14	0.04	0.004	0.001

	Peanuts	0.8	0.1	0.03	0.01	0.001	<0.001

	Aggregate	4.6	1.0	0.19	0.08	0.01	0.005

LA/MS Mid-south	Cotton	0.8	0.2	0.02	0.004	<0.001	<0.001



These estimates are based on the reported typical application rates for
aldicarb, with an assumption that approximately 15% of the applied
pesticide is available at or near the surface for runoff. The
uncertainties in these assumptions are addressed in the sections that
follow.

Uncertainty due to application method

For these scenarios, OPP modeled CAM 1 in PRZM (broadcast application on
the surface), using 15% of the application rate to reflect the
assumption that 15% of the granules remain on the surface after
application. This assumes that none of the remaining application is
available for runoff. To characterize uncertainties related to the type
of application method used, OPP also modeled CAM 7 in PRZM, equivalent
to a T-band application, assuming that 15% of the application was
incorporated in the top 2 cm while the remaining 85% was incorporated in
the lower 10 cm. Peak estimates were greater with the CAM 7 simulation,
which more closely reflects banded applications (Table 13). However, the
peaks were generally less than 20% greater than that estimated with CAM
1, which more closely reflects broadcast applications. 

  SEQ CHAPTER \h \r 1 Table 13: Comparison of estimated total aldicarb
residues using different application methods and assumptions on the
fraction of applied aldicarb available for runoff.

Region	 App Method / Rate	15% at surface	1% at surface



Max	99th	95th	90th	Max	99th	95th	90th

NC/ Coastal	CAM 1 / typical rate	4.6	1.0	0.19	0.08	0.30	0.07	0.01	0.005

	Cam 7 / typical rate	5.3	1.1	0.21	0.08	0.35	0.08	0.01	0.005

Central FL Citrus	CAM 1 / typical rate	10.2	1.6	0.26	0.08	0.68	0.12	0.02
0.007

	Cam 7 / typical rate	11.8	1.8	0.30	0.09	0.79	0.14	0.02	0.008

LA/ Midsouth	CAM 1 / typical rate	0.84	0.17	0.02	0.004	0.06	0.01	0.001
<0.001

	Cam 7 / typical rate	0.99	0.25	0.03	0.007	0.07	0.02	0.002	<0.001



The registrant contends that only 1% of the granules remain
unincorporated. Such an assumption would result in estimated exposures
that are no more than 1/15th of the estimated concentrations reported in
Table 2 (Table 13). 

Uncertainty due to application rate

The estimates used in the NMC cumulative drinking water exposure
assessment and provided to HED for surface water exposure are based on
typical application rates. While “typical” application rates may be
reflective of average conditions over an extended period of time, they
fail to capture the variability in application rates that may occur from
one year to the next as a result of varying pest pressures. Such
variations are more likely to be reflected in surface water exposures,
which are subject to short-term variations in factors such as
application rates, acres treated, and the timing and intensity of
rainfall after application. Table 14 summarizes the estimated
distributions of total aldicarb residues based on both maximum and
typical application rates. It also characterizes the variability in
estimated concentrations depending on the amount of applied pesticide
remaining at or near the surface (and, thus, vulnerable to runoff).
While the assumption of 15% granules left on the surface after
incorporation may be a high-end estimate, the assumption of only 1% of
the granules left on the surface is a “best case” condition. The
actual amount left on the surface is likely somewhere in between.

  SEQ CHAPTER \h \r 1 Table 14: Comparison of estimated total aldicarb
residues using different application rates and assumptions on the
fraction of applied aldicarb available for runoff.

Region	App Rate	15% at surface	1% at surface



Max	99th	95th	90th	Max	99th	95th	90th

NC/ Coastal	Typical 	4.6	1.0	0.19	0.08	0.30	0.07	0.01	0.005

	Maximum	30.7	6.6	1.1	0.39	2.0	0.44	0.07	0.03

Central FL Citrus	Typical 	10.2	1.6	0.26	0.08	0.68	0.12	0.02	0.007

	Maximum	13.2	2.3	0.39	0.13	0.88	0.15	0.03	0.009

LA/ Midsouth	Typical	0.84	0.17	0.02	0.004	0.06	0.01	0.001	<0.001

	Maximum	6.4	1.3	0.14	0.03	0.43	0.08	0.01	0.002



Comparisons with monitoring

Aldicarb residues have not been detected frequently or in high amounts
in surface water in the USGS NAWQA monitoring studies – 0.2%
detections with a maximum concentration of 0.5 ug/L based on the 2001
national summary by Martin et al (2003; see   HYPERLINK
"http://ca.water.usgs.gov/pnsp/pestsw/Pest-SW_2001_Text.html" 
http://ca.water.usgs.gov/pnsp/pestsw/Pest-SW_2001_Text.html  ). While
the NAWQA monitoring sites are not targeted to aldicarb use areas and
the frequency of sampling is not designed to capture peak concentrations
in surface water, the results suggest that actual concentrations of
aldicarb residues in surface water are likely to be closer to the single
or sub-parts per billion range than to 10-30 ppb.

 PAGE   

 PAGE   2 

