Superfund's
Standard Default Exposure Factors
for the
Central Tendency
and
Reasonable Maximum Exposure
-------
CONTENTS A
%-j.
1.0 INTRODUCTION 1 /
1.1 Central Tendency 3
1.2 Reasonable Maximum Exposure (RME) 3
1.3 Residential Exposure 4
1.4 Occupational Exposure 4
2.0 CONCENTRATION 5
3.0 EXPOSURE FREQUENCY 5
3.1 Central Tendency 5
3.1.1 Residential 5
3.1.2 Occupational 6
3.2 Reasonable Maximum Exposure 6
3.2.1 Residential . 6
3.2.2 Occupational 6
4.0 EXPOSURE DURATION 6
4.1 Central Tendency 6
4.1.1 Residential 6
4.1.2 Occupational . ^ . 6
4.2 Reasonable Maximum Exposure 6
4.2.1 Residential 6
4.2.2 Occupational 7
5.0 BODY WEIGHT . 7
5.1 Chiild 7
5.2 Adult 7
6.0 INGESTION OF POTABLE WATER 7
6.1 Central Tendency 7
6.1.1 Residential Ingestion Rate 7
6.1.2 Occupational Ingestion Rate 7
6.2 Reasonable Maximum Exposure 8
6.2.1 Residential Ingestion Rate 8
6.2.2 Occupational Ingestion Rate . , 8
-------
7.0 INGESTION OF SOIL AND DUSTS ^"£0
7.1 Central Tendency Exposure
7.1.1 Child's Ingestion Rate • ,8
7.1.2 Adult Ingestion Rate; Non-Contact Intensive 9
7.1.3 Adult Ingestion Rate: Contact Intensive 9
7.1.4 Residential Exposure Scenario; Child + Adult 9
7.1..5 Exposure Frequency and Duration: Central Tendency , . 9
72 Reasonable Maximum Exposure 10
7.2.1 Child's Ingestion Rate • 10
7.2.2 Adult Ingestion Rate: Non-Contact Intensive 10
7.2.3 Adult Ingestion Rate: Contact Intensive 10
7.2.4 Residential Exposure Scenario; Child 4- Adult 10
7.2.5 Exposure Frequency aad Duration: RME 11
7.3 General Exposure Frequency and Duration Considerations . 11
7.4 Fraction Ingested fronHhe Contaminated Source 12
7.5 Matrix Effect 12
8.0 INHALATION OF CONTAMINANTS 12
9.0 INGESTION OF LOCALLY CAUGHT FISH 13
10.0 INGESTION OF PRODUCE 14
10.1 Total Produce Consumption Rates 15
10.2 Crop Specific Ingestion Rates 15
SUMMARY TABLES OF DEFAULT EXPOSURE FACTORS -17
FRUIT AND VEGETABLE CONSUMPTION QUANTITIES 18
REFERENCES
19
-------
1.0 INTRODUCTION
Last spring, EPA's Risk Assessment Council released a memorandum entitled "Guidance
on Risk Characterization for Risk Managers and Risk Assessors" (U.S. EPA 1992) in
which the council advocated greater interface between risk assessment and risk
management, greater discussion of confidence and uncertainty in the risk assessment/
and presentation of the range of possible exposures including the use of multiple risk
descriptors. Focusing specifically on this last point regarding the exposure assessment,
the Risk Assessment Council (RAC) clearly Indicated that it expects all risk assessments
"to address or provide descriptions of (1) individual risk to include the central tendency
and high end portions of the risk distribution, (2) important subgroups of the population
such as highly exposed or highly susceptible groups or individuals, if known, and (3)
population risk."
For several years now, the Super fund program has considered exposure to sensitive
subgroups or populations as applicable and has been estimating individual risk
corresponding to the reasonable maximum exposure (RME). The Risk Assessment
Guidance for Superfund: Human Health Evaluation Manual (Part A) (U.S. EPA 1989)
also known as RAGS, defines the RME as the highest exposure that, is reasonably
expected to occur at a site and in practice is estimated by combining 90 - 95th percentile
values for some but not all exposure parameters. Shortly after RAGS was released, the
Superfund Program developed the "Standard Default Exposure Factors" Supplemental
Guidance (U.S. EPA 1991) to promote consistency in the evaluation of the RME exposure
in baseline risk assessments when site-specific data were lacking. It is the position of
the Superfund Program that RAGS and the standard default values for the RME are
consistent with the Risk Assessment Council's expectation to provide a description of the
high-end portion of the risk distribution..
Until the guidance contained herein was developed, existingSuperfund guidance did not
provide a framework in which to estimate risk corresponding to the central tendency
portion of the risk distribution as called for by the Risk Assessment Council. Perceiving
a need to fill this void, a workgroup was organized by the Superfund Program in
October of 1992, comprised mainly of EPA Regional Superfund risk assessors, with the
purpose of defining the central tendency for use in Superfund baseline risk assessments.
Over the course of the following six months, the workgroup convened periodically to
discuss an approach and identify standard default exposure factors for the centra]
tendency. In doing so, the workgroup also felt it beneficial to review the current default
exposure factors for the RME and identify whether any changes were warranted at this
time. Consequently, this guidance builds on the concepts identified in RAGS Part A and
the Risk Assessment Council's recommendations regarding risk descriptors for the
-------
central tendency, It supersedes the standard default exposure factors for the
contained in the guidance of the same name (U.S. EPA 1991). j
Reliance on the point estimate approach versus the use of monte carlo techniques to
characterize the range of possible exposure estimates was initially discussed by the
workgroup as both approaches have merit in addressing the Risk Assessment Council's
call to present the range of possible exposures and risk outcomes. In the end, the
workgroup concluded that too many issues regarding the practical application of monte
carlo techniques remained unresolved and a significant investment of time and resources
would be required to address the issues. Thus, the traditional point estimate approach
to exposure assessments was favored by the workgroup at that time. Additionally, the
point estimate approach to exposure was deemed fully consistent with the intent of the
Risk Assessment Coundl in their memo.
As there is presently an agency-wide effort underway to address all of the Risk
Assessment Council's recommendations (including the use of monte 'carlo techniques
and revisions to EPA's Exposure Factors Handbook incorporating distributions for the
various exposure parameters), the guidance contained herein for the Superfund Program
is subject to change and consequently should be viewed as interim in status. When such
agency-wide guidance is available, it is expected that it may supersede this guidance.
The guidance contained herein has been developed to encourage a consistent approach
to assessing exposures when there is a lack of site-specific data or consensus on which
parameter value to choose, given a range of possibilities. -Accordingly, the exposure
factors presented in the Summary Tables on page 17 of this document are generally
considered most appropriate and should be used in baseiine risk assessments unless
alternate or site-specific values can be clearly justified by supporting data.
Supporting data for many of the exposure factors presented in this guidance can be
found in the Exposure Factors Handbook (EFH; U.S. EPA, 1990). Additionally, in some
instances, peer reviewed studies were utilized to identify suitable default values as well
as group consensus techniques when a faced with a great deal of uncertainty. In these
instances, either the study or a clearly documented logical approach used to identify
default factors is referenced.
2
-------
^6
The general exposure equation into which these standard factors are to be utilizf£|is'-£s
C - Concentration of the contaminant in a given medium
IR - Intake/Contact Rate; the amount of contaminated medium contacted
per unit time or event
EF = Exposure Frequency
ED = Exposure Duration
BW = Body Weight
AT « Averaging Time (equal to exposure duration for non-carcinogens and
70 years for carcinogens)
1.1 Central Tendency (CT)
The Risk Assessment Council defined the central tendency risk descriptor as
either the arithmetic mean risk or the median risk and continues to say'that the
arithmetic mean risk can be derived by using average values for all the exposure
factors though cautions that when dealing with skewed data, the median or 50th
percentile may better approximate the midpoint of a distribution (U.S. EPA 1992).
As a result, any approach to the identification of default factors for the central
tendency should seek to identify average or 50th percentile values whenever
possible. In keeping with this approach, default exposure factors approximating
the average or 50th percentile value have been identified whenever possible for
use in central tendency exposure evaluations.
1.2 Reasonable Maximum Exposure (RME)
The Risk Assessment Council denned a high end risk descriptor as one which
characterizes risk to an individual at the upper end of the risk distribution.
Conceptually, it describes exposures above the 90th percentile of the population
distribution (U.S. EPA 1992). As previously indicated, the reasonable maximum
exposure (RME) terminology used by the Superfund Program is believed
consistent with this description. The Risk Assessment Guidance for Superfund:
Human Health Evaluation Manual (Part A) (RAGS) defines the RME as the
highest exposure that is reasonably expected to occuj at a site and in practice is
estimated by combining upper bound (90th to 95th percentile) values for some
but not all exposure parameters. Consequently, the Superfund Program will
continue to use the current terminology of reasonable maximum exposure (RME)
follows:
Intake ¦= C x IR x EF x ED where
BW x AT
3
-------
>r
* j
in fulfilling the Risk Assessment Council's mandate to evaluate a high end rislT^VV
descriptor.
In keeping with the previous default factor guidance (U.S. EPA 1991), 90 to 95th
percentile values were targeted in this guidance document when identifying
default factors for intake/contact rate, exposure frequency, and exposure
duration. - An average value or conservative estimate of the media average
contacted over the exposure period was targeted for identification of default
values for body weight and exposure concentration respectively.
Within the context of this guidance, standard defaultexposure factors have been
identified for various exposure pathways aad receptor populations owing to the
different nature and magnitude of the assumed exposure. Generally speaking,
default values for residential and occupational receptors have been identified and
serve as the genera] basis for organization of this guidance.
1.3 Residential Exposure
Residential default exposure factors are generally relevant whenever there are or
reasonably may be expected to be residences on or adjacent to the site. The.
contamination may be on the site itself or may have migrated from it. With the
exception of exposure to contaminated soils, distinctions are not usually made in
the default parameters for exposures to different aged receptors. Because of the
higher intake to body weight ratio presumed to occur during the early years (ages
1-6) for this exposure pathway, special attention should be given to evaluating
exposure for this pathway as discussed in sections 7.4. and 7.5.
1.4 Occupational Exposure
Occupational default exposure factors are generally relevant whenever the site
serves or may reasonably be expected to serve as a place of temporary or
permanent employment. -Examples of employment in which one may be
presumed to come in contact with contaminated media'might include employ-
ment at the facility itself or nearby facilities (commercial/industrial), servicing of
the facility (grounds keeper/utility maintenance), or construction of new facilities
or the demolition of old facilities on or adjacent to the site.
4
-------
2.0 CONCENTRATION
Centra] Tendency ana RME
The concentration term in the intake equation is the arithmetic average of the
concentration that is contacted over the exposure period. Because of the uncertainty
associated with any estimate of exposure concentration, the 95% percent upper
confidence limit on the arithmetic average concentration will be used for this variable
in both the central tendency and reasonable maximum exposure estimates.
Consideration should De given to the data set upon which the 95% upper confidence
limit of the mean value is generated so as to represent as closely as possible the nature
(acute vs. chronic) of potential exposures. In some instances, there may be great
viariability iri measured or modeled concentration values such as when too few samples
are taken or'when model inputs are uncertain. In these cases, the upper confidence limit
on the average concentration may even' exceed the maximum value observed or
predicted,- Should this scenario arise, the simple arithmetic mean and maximum
concentrations should, in general, be used for the central tendency and reasonable
maximum exposure concentrations respectively. Although, in general, the mean and
maximum concentrations will he used in this situation, thisxuill not always be the case
and is a site-specific decision.
3.0 EXPOSURE FREQUENCY
The following default exposure frequencies may be utilized unless otherwise indicated
or site-specific data is available.
3.1 Central Tendency
3.1.1 Residential
The central tendency residential default exposure frequency of 234 days/year
corresponds to the fraction of time estimated that is actually spent at home (64
percent) for both men and women, based on a study of time use patterns
summarized in the EFH (U.S. EPA 1990). Because the study included both
personal and work-related travel, a 365 day year was used from which to
compute the 64 percent.
3.2.2 Occupational
The cctitral tendency occupational default exposure frequency of 219 dayslyear
corresponds to the average time spent at work by both full-time and part-time
workers, based on 1991 data from the Bureau of Labor Statistics.
5
-------
3.2 Reasonable Maximum Exposure
3.2.1 Residential ^
The RME residential default exposure frequency of 350 days/year is based on the
previously identified default value which assumes a two week vacation each year.
This is viewed as a reasonably conservative estimate of exposure frequency absent
site-specific data..
3.2.2 Occupational
The RME occupational default exposure frequency of 250 days/year is consistent
with the previously identified default value and is based on a 5 day work week
with two weeks of vacation each year, This is viewed as a reasonably conserva-
tive estimate of exposure frequency absent site-specific data.
4.0 EXPOSURE DURATION
The following default exposure durations may be utilized unless otherwise indicated or
site-specific data is available.
4.1 Central Tendency
4.1.1 Residential
The residential central tendency default exposure duration of 9 yeajs is based on
data summarized in the EFH (U.S. EPA 1990) in which the average length of
residence in the same house of people who own their own home was estimated
to be 9 years.
4.1,2 Occupational
The occupational central tendency default exposure duration of 5 years is based
on . . ,
4.2 Reasonable Maximum Exposure
4.2.1 Residential
The RME residential default exposure duration of 30 years is based on data
summarized in the EFH (U.5, EPA 1990) in which the 90th percentile for the
length of residence in the same house of people who own their own home was
estimated to be 30 years.
4.2.2 Occupational
The RME occupational default exposure duration of 25 years is based upon the
6
-------
95th percentile for the number of years worked at the same location as reporteaV^
by the U.S. Bureau of Labor Statistics, 1990.
5.0 BODY WEIGHT
The average body weight is to be utilized for both the central tendency and RME
exposure evaluations in keeping with the respective definitions.
5.1 Child
The approximate average body weight of young children (boys and girls combined)
under the age of 6 years is approximately 15 kg (U.S. EPA 1990). Distributions of
body weights and average body weights and for other age gToups can be found in
the EFH (U.S. EPA 1990).
b.'i Adult
The average body weight, ot 70 kg corresponds to the average weight o: men and
women age 18-75 as reported in EFH (U.S. EPA 1990). Distributions of body weights
and average body weights for other age groups can be found in the EFH (U.S. EPA
1990).
6.0 INGESTION OF POTABLE WATER (Entries la and lb on Summary Tables)
6.1 Central Tendency
6.1.1, Residential Ingestion Kate
The central tendency potable water ingestion rate for an adult of 1.4 L/day is
based on the average intake observed from five studies as summarized in the
EFH (U.S. EPA 1990). The observed range reported across the five studies was
from 0.26 to 2.8 L/day.
6.1.2 Occupational Ingestion Rate
The central tendency potable water ingestion rate fof an adult of 1,4 Uday is
based on the average intake observed from five studies as summarized in the EFH
(U.S. EPA 1990). The observed range reported across the five studies was from
0.26 to 2.8 L/day.
6.2 Reasonable Maximum Exposure
6.2.1 Residential Ingestion Rate
The RME potable water ingestion rate of 2 L/day is dose to the 90th percentile
of values measured and estimated by researchers as summarized in EFH (U.S.
7
-------
EPA 1990). It is also the value currently used by EPA's Office of Water inf.
establishing drinking water standards.
6.2.2. Occupational Ingestion Rate
No data upon which to base a default value.
7.0 INGESTION OF SOIL AND DUST (Entries 2a through 2d on Summary Tables)
Due to the importance of the receptors age arid behavioral characteristics, default
ingestion rates for this exposure pathway have been established based on the
characteristics of the receptor rather than on the location of the exposure (residential vs.
occupational). Default ingestion rates for this pathway are as described below in
Sections 7.1 and 7.2.
7.1 Central Tendency
7.1.1. Child's Residential Ingestion Rate
Numerous studies have documented that the propensity to ingest non-food items
is greatest in the early years of development. As a result, children between the
ages ox 1 and 6 years are of greatest concern as they are expected to have the
greatest exposure to contaminated soils and dusts via ingestion. .Numerous
studies (tracer studies ai\d estimates of deposition/exposed surface area) have
resulted in wide ranging estimates of the amount of soil and dusts ingested by
young children making it difficult to identify a single value for use as the central
tendency. Additionally, owing to the nature of the experimental studies, it is
extremely difficult to separate the contribution to exposure resulting from exterior
soils vs. interior dusts. As a result the ingestion rate is reported as the combined
xate for soils and dusts.
It was believed by a consensus of workgroup members that the ingestion rate of
100 mg/day as a central tendency Ingestion rate for a child between the ages of
1-6 years was within reason based on results using tracer elements (Davis el al.
1990 and Calabrese 1989). Furthermore, 100 mg/day is nearly identical to the
ingestion rate for this age group based on age-specific values utilized in support
of the NAAQS for lead (U.S. EPA 1989b) and the lead biokinetic uptake model.
7.1.2 Adult's Residential and Occupational Ingestion Rate: Non-Contact Intensive
For the adult who does not engage in soil- or dust-contact-intensive activities on
a regular basis (apartment dweller, typical homeowner, office, worker, teacher,
professional, etc.) the soil and dust default ingestion rate for the central tendency
8
-------
rhA i\u. ouodboidou r• 10
-ch nun 10. oo onwn an . . . ...
of 50 "mg/day based on. a stvidy of Calabrese 1990 (with supporting estimate,
from Hawley 19S5).
7.1.3 Adult's Ingestion -Rate.' Contact Intensive
For adults who routinely engage in heavy contact with soils and dusts on a
regular basis (including seasonal work), the workgroup was unable to identify a
default soil ingestion rate corresponding tp the central tendency given the data
available. It is suggested that an evaluation of the RMS scenario for this receptor
be conducted.
7.1.4. Residential: Child -f Adult Combined
In evaluating a residential exposure scenario for this pathway, a weighted average
of the child's and adult's exposure is to be utilized. The duration of exposure for
the central tendency has been defined.as consisting of 9 years (average number
for years at the same dwelling).. It is the default position to assume that for 2 of
the 9 years, intake will be at the child's rate and for the remaining 7 years, intake
will be at the adult rate. This is consistent with the proportion of time one is
assumed to be a young child that is- utili2ed for RME residential calculations.
Thus residential exposure for the central tendency should generally be evaluated
as follows:
2 years x 100 me/dav + 7 years x 50 mg/dav
15 kg 70 kg
7.1.5 Exposure Frequency and Duration
The default value for the duration of exposure for the central tendency scenario
is 9 years for a residential exposure based on the average length of stay in a home
as reported in the EFH (U.S.EPA 1990). It should be noted that generally the
intake over the 9 year exposure period is to be computed as described in section
7.1.4, The default exposure frequency for the central tendency is 350 days/year
due to the nature in which the soil ingestion rates have been computed (average
daily exposure). Although the default exposure frequency in most regions is 350
days per year, in some regions a consideration of climate conditions will result
in an exposure frequency less than 350 days per year. Regional offices should be
consulted for guidance.
A default exposure frequency and duration has not been specified for the central
tendency contact-intensive occupational scenario at this time as it has not been
discussed -by the workgroup.
9
-------
7,2 Reasonable Maximum Exposure
7.2.1 Child's Residential Ingestion Rate
The default RME ingestion rate for a young child age 1-6 years of age of 200
xng/day represents.the consensus opinion, oi'-the workgroup based on review of
available data and is believed to correspond to a conservative estimate of an
average ingestion rate for this age group over a chronic period of exposure.
Unfortunately, the available data did not support identification of the 90th or 95th
percentile value. It was the' consensus among workgroup participants that over
the 6-year period of concern for this receptor category, the value of 200 mg/day
was reasonable to assume. It should be noted that this value was not necessarily
deemed relevant for acute exposures when a child may engage in intensive
contact with soils and dusts for a brief period of time. In these situations,
ingestion rates greater than this value may be warranted,
7.2.2 Adult's Residential and Occupational Ingestion Rate: Non-Contact Intensive
The RME default soil and dust ingestion rate of 100 mg/day is based a study of
Sedman (1989). This value is presumed suitable for non-contact intensive
scenarios (apartment dweller, typical homeowner, office worker, teacher, profes-
sional, etc.).
7.23 Adulfs Residential ar.d Occupational Ingestion Rate-. Contact intensive
The RME default soil and dust ingestion rate of 480 mg/day is deemed
appropriate for acute exposures (those less than a year in duration such zs
construction or gardening). . This value is based on estimates made by- Hawley
(1985) in which he estimated deposition rates, exposed surface areas of the hands,'
and the fraction inadvertently consumed.
7.2.4 Residential: Child + Adult
In evaluating a residential RME exposure scenario, the exposure duration for the
RME has been denned as consisting of 30 years (90th percentile for years at the
same dwelling, U.S. EPA 1990). It shall generally assumed when evaluating the
RME residential exposure for the ingestion or soil and dusts tiw„ for 6 of the 30
years, intake will be at the child's rate and for the remaining 24 years, intake will
be at the adult rate. Thus residential RME exposure for this pathway should
generally be evaluated as follows:
6 years x 200 mg/dav + 24 years x 100 mg/dav
15 kg 70 kg
10
-------
•f
7.2.5 Exposure Frequency and Duranon.
The default value for the duration of exposure for the RiYfE scenario is 30 years
for a residential exposure based on the 90th percentile for the length of stay Ln a
home as reported in the EFH (U.S.EPA 1990), It should be noted that generally
the intake over the 30 year exposure period is to be computed as described iri
section 7.2.4.
The default exposure frequency for the RME is 350 days/year due to the nature
in which the soil ingestion rates have been computed (average daily exposure)
and assuming a two week period away from home each year.
The default value for the duration of exposure for the RME occupational scenario
is 25 years based on the 95th percentile for the number of years worked at the
same location (Bureau of Labor Statistics 1990). The exposure frequency of 250
days/year corresponds to a five day work week. In some regions, a consideration
of climate conditions will result in a downward adjustment of exposure
frequency. Regional offices should be consulted fcr guidance.
7.3 General Exposure Frequency and Duration Considerations
Owing -to the strong age- and behavior-dependent nature of this exposure,
exposure durations and frequencies other than the default values may be
warranted for this exposure pathway. For example, a situation may arise in
which a child-care facility is of concern and the residential default values for
exposure frequency and duration may not be appropriate. Similarly, certain
occupations may lead to intensive exposure but for brief periods of time (u.
construction workers/ field laborers, seasonal workers, etc.) rendering use of the
occupational default values for exposure frequency and duration inappropriate.
Additionally, there may be situations in which a Region believes it necessary to
adjust the exposure frequency to account for meteorological conditions which may
be presumed to drastically re'duce or eliminate exposure to potential contaminants
via soil ingestion, In these situations, any adjustments to the exposure frequency
to reflect local weather patterns should first be approved by the Regional Office.
For these reasons, the default exposure durations and exposure frequencies may
not always be relevant for the exposure at hand. Extra care snould be taken
when identifying suitable exposure frequencies and durations for this exposure
pathway.
11
-------
7.4 Fraction Ingested From the Contaminated Source ^
The fraction ingested from the contaminated source is a variable that often gets
overlooked -when evaluating scenarios that are largely dependent on the receptor
coming to the source of contamination rather than the contamination migrating
to the receptor. Due to variations in the proximity of the receptor to the
contaminated source, size of the contaminated source, receptors of concern,
mobility of receptors, and the nature of exposure, default values for the fraction
ingested from the contaminated source are not possible. However, it is advocated
that this factor be givzn consideration when evaluating this exposure pathway.
7.5 Matrix Effect
A parameter unique to all combinations of compounds and soil types — the
matrix effect -— accounts for the tendency of a compound to bind to soils. The
more "soil loving" a compound is, the less likely it is to desorb and. become
bioavailable in the gastrointestinal tract once ingested. 'Chemical and physical'
properties of contaminants and the soil can thus have a profound effect on the
bioavailability of a compound. Unfortunately the data do not exist to support
default desorption values for all compounds at this time though work is currently
underway to develop some guidance in this area. At present, any adjustments
for this phenomenon are left open to the discretion of the Regional Office.
8.0 INHALATION OF CONTAMINANTS
It is anticipated that at some time in the future, inhalation exposures will be evaluated
vising inhalation reference concentrations. However, at this time, the methodology is not
available and, consequently/ inhalation rates and-resulting dose (mg/kg/day) are the
approach that is advocated for this exposure pathway. Inhalation rates are dependent
on age, sex, and activity level to name just a few factors and can be found in the
Exposure Factors Handbook (U.S. EPA 1990).
The same default inhalation rate has been identified for both the central tendency and
the RME exposure scenarios. This is in keeping with the assumption regarding
inhalation rate used in the derivation of cancer potency estimates and inhalation
reference concentrations. The default value of 20 mVday corresponds to a reference
man's inhalation rate who is at rest 8 hours/day and at a light activity level (i.e.
domestic work, personal care, hobbies, minor indoor home improvements) for the
remaining 16 hours/day.
12
-------
9.0 INGESTION OF LOCALLY CAUS KT PISH
The evaluation of this exposure pathway Will not always be relevant to every site. The ?
receptor of concern for this pathway is ap^ to include both the recreational fisherman
and a subsistence fisherman and their famllv. The preferred approach to the evaluation
of this exposure pathway is to obtain site-specific data regarding-consumption rales and
fishing habits. This is due to the strong influence of local habits, populations, and
conditions on the resulting exposure.
When site-specific data are not feasible to obtain, the default approach suggested for this
exposure pathway is based on an estimate of the averse size of a fish meal and merely
varies the exposure.frequency, duration, and fraction ingested from the contaminated
source between the central tendency' and the RME estimates. With this approach,
recreational and subsistence fishermen can "be assumed to consume the same amount of
fish per eating occasion yet differ in the frequency or number of fish meals actually
consumed and the fraction of fish meals consumed that originated from the contaminat-
ed source. This change in approac:,*' Was adopted qecause it was believed to better
characterize exposure resulting from an interment and often infrequent exposure
pathway than the default approach previously advocated which relied on an intake rate
averaged over a year of exposure.
The average amount of fish consumed per eating occasion was observed to be 145
g/meal or about 5 ounces as reported in the study oi Pao et ai: (1982). The range
reported for the size of the fish meal was from 43 g/meal (5th percentile) to 565 g/meal
(99th percentile). The study was based on the results of a self-administered USDA
nationwide consumption survey from 1977-78 of individuals in 48 states. The amount
of fish corresponds to consumption ^abits for fin-fish as reported on a wet weight basis.
It does not include shellfish. Although fi;h consumption habits have likely increased
oyer the past 15 years, the Pao study was believed to be the best study available upon
which to base-a default value.
Owing to the very site-specific nature of the frequency of this exposure, no defaults are
given at this time for exposure frequency (fish meals/year). However, estimates of the
average and 90th to 95th percentile wr the frequency of exposure should be used for the
central tendency and RME respectively. Default values for exposure duration are those
which are consistent with residential default values previously identified of 9 years for
the central tendency and 30 years for the RME. Additionally, it was believed1 that a site-
specific value for the fraction of fish consumed from the contaminated source was
appropriate rather than establishing a default value for this factor. The average and the
13
-------
uon lO'Jo ofirtiron _ '*9rt IVJ' Juocooi^ou F,c-L
«0
N<&*
n* -
*<*&
90th to 95th percentile values are suggested for the central tendency and RME for this^ A
parameter respectively.
10.0 INGESTION'OF PRODUCE
The following approach has been suggested for this exposure pathway provided it is
relevant to the risk evaluation:
a. Strongly consider evaluating consumption of homegrown produce if it constitutes
a current exposure pathway and if produce is available for analysis. If produce
is not available for analysis, evaluation of this exposure pathway is open to the
discretion of the Regional Office (recognizing that this decision is apt to depend
on the level of confidence in available plant uptake models).
b. If the decision is made to employ an uptake model, the Region is strongly-
encouraged to seek the assistance and/or review of the proposed approach by
ECAO-Cincinnafci.
c. When evaluating this exposure pathway, preference should be given for site-
specific consumption rates (obtainable via door-to-door surveys) if feasible. When
site-specific consumption rates are not feasible, either generic defaults regarding
total consumption rates for all fruits combined or all vegetables combined (USDA
1980) or defaults based on the average amount of a fruit or vegetable consumed
on a given eating occasion (Pao et al. 1982) together with site-specific exposure
frequencies is suggested.
d. The fraction ingested assumed to originate from a contaminated source will
always be a site specific determination.
The choice of which of the approaches described below should be utilized for the
identification of default ingestion rate values is left up to the risk assessor based on their
understanding of the site. The USDA (1980) results are based on the average
consumption rate as self-reported over a three-day period and included non-consumers
as well a$ consumers in the calculation. In contrast, the data of interest from Pao et al
(1982) focused on the amount consumed of various food crops for a given eating
occasion. If and when default values are used, the same ingestion rate utilized for the
central tendency is advocated for use in evaluating the RME scenario, It is suggested
that in these instances, merely the exposure frequency, duration, and the fraction
14
-------
rnA ihu, ouo^aoi^ou
i,un lo• oy orihu on _ _ .
' ¦' a
' *
ingested from the contaminated source vary between the centra] tendency and the
evaluations.
10.1 Total Produce Consumption Rates (USDA 1980, U.S. EPA 1990)
As summarized in the EFH (U.S. HPA1990), the USDA estimated the average intake
on any one day of all fruits combined ?.?.142 g/day per person and approximately
1/5 of this (28 g/day) could be assumed to be homegrown on average or as much
as 3/10 of this (42 g/day) could be assumed to be homegrown as a reasonable
maximum exposure case.
The average intake on any one day for all vegetables combined was estimated as 201
g/day. Furthermore, approximately 1/4 (50 g/day) of this amount could be assumed
to be homegrown on average'and as 'much as 2/5 (80 g/day) could be assumed to
be homegrown as a reasonable maximum exposure case.
Due to the nature of the study, (a daily average intake over a three-day exposure
period), it can be assumed that the contact rates do not represent a chronic value. If
this approach is selected, .then the exposure frequency for the central tendency and
RME should be 350 days/vear. The default exposure duration reflects the residential
central tendency value of 9 years or 30 years for the RME scenario. Assumptions
regarding the fraction ingested from the contaminated' source are not specified,
though national averages for the fraction that can be assumed to be homegrown have
been suggested as a described above.
10.2 Crop-Specific Consumption Habits (Pao et ah 19S2)
As summarized in the attached table, average values for the amount of a particular
fruit or vegetable consumed on a given eating occasion can be identified based on the
results of a nationwide swey conducted by the USDA as summarized in Pao et al
(1982), Additionally, the authors reported the distribution of consumption values
observed for each fruit or vegetable included in .the survey. The Pao et al data was
based on the USDA nationwide food consumption survey conducted in 1977-78.
Default values for the frequency of exposure have not been identified and are subject
to site-specific determinations reflecting local consumption habits.
The default exposure duration reflects the residential central tendency value of 9
years or 30 years for the RME scenario. The fraction ingested originating from the
contaminated source has not been specified but is open to consideration of site-
specific factors.
V"
15
-------
in»\"£ii -o-i nun io-oo on*n on
rn<\
tvj. ouocooic:ou
r. co
CENTRAL TENDENCY
Exposure Pathway
Contact Rate
frequency
Duration
Eody We-jjnt
1. ingestion of Drinking Water
la, Residential
1.4 L/day
234 davs/vr
9 years
'"V
70 kg K-
lb. Occupational
1.4 L/day
219 davs/vr
S vears
70 kg
2. Ingestion of Soils ar.d Dusts2
2a. Child - residential
100 mg/day-
350 days/yr'
2 years
15 kg
2b. Adult - Non-contact residential
SO mg/day
3S0 days/yr*
7 years
70 kg
2c. Adult - Non-contact occupational
50 mg/day
234 davs/vr'
70 kg
2d. Adult - Contact Intensive
data insufficient
70 kg
3. Inhalation
3a. Residential
20 m'/day
234 days/yr
P years
70 kg
3b. Occupational
?
219 davs/vr
5 vears
?0 kg
4. Ftsh Ingestion1
145 g/meal
Site specific
average
9 years
70 kg
5. Ingestion of Produce1
142 g/day (fruits).
20\ g/iay (ves.) or
produce specific for
amount per meal (see
attachment)
350 days/yr for
values indicated or
site-specific average
if usa ami/meal
9 years
70 kg
'
REASONABLE MAXIMUM EXPOSURE
Exposure Pathway
Contact rlate
Frequency
Duration
Body Weight
1. ingestion of Drinking Water
la. Residential
2 L/day
350 days/yr
30 years
70 kg
lb. Occupational
?
250 davs/vr
25 years
70 ke
2. Ingestion of Soils and Dusts2
2a. Child - residential
200 mg/day
350 days/yr'
6 years
IS kg
2b. Adult - Nor.-contatct residential
ICO mg/day
350 days/yr'
24 years
70 kg
2c. Adult - N'pn-contact occupational
100 mg/day
250 dsys/yr'
25 years
70 kg
2d. Adult - Contact Intensive
4S0 mg/day
s'.te specific
sits
specific
70 kg
3, Inhalation
2a. Residential
20 m'/day
350 days/yr1
30 years
70 k3
3b. Occupational
7
250 days/yr'
25 years
70 kg
4. Fish Ingestion'
145 g/meal
site specific
S0-9£th *
30 years
70 kg
S. Ingestion of Produce1
142 g/day (fruits) 350 days/yi for
201 g/day (veg.) or produce values Indicated
specific value for amount or site-specific
per mead (see attachment) S0-95th % If use
azU./meal
30 years
70 kg
Adjustment* basa<< on behavioral or mcteorlogical conditions (such as a downward adjustment for exspojurc frequency for days
that the ground b snow-convcnsd) my be warranted biied on site-specific conditions and Regional policies.
Though not specified, exposure pathway should include a sitc-specfic value for the fraction ingested originating from the
con:amina:cd source.
16
-------
a tTaxiaus I'taits and Vegetables (grams)
Pao et nl. 1982
"Cooked vegetables includes conned.
l'ciccntile
Std.
Food category
Avg.
Avg.
5lh
25th
50 Ih
75lli
90£h
95Ub
99th
Max.
fresh grapefruit
159
58
•106
13-1
m
165
268
268
330
660
fresh oranges
146
57
73
145
145
145
180
228
360
1160
raw apples
141
49
69
133
133
138
212
212
276
636
bananas
106
37
50
95
119
119
136
136
238
476
Fruit
cantaloupe
171
91
61
136
136
272
272
272
529
896
raw pears
163
69
82
164
164
164
164
328
328
2132
raw peaches
160
75
76
152
152
152
3M
301
456
760
raisins
33
28
3
14
28
43
73
73
145
290
raw strawberries
100
58
37
75
75
149
149
180
298
447
white potatoes
125
90
29
63
105
170
235
280
426
1260
cabbage/colc slaw
68
45
15
40
60
90
120
120
240
1020
raw carrots
43
•10
4
13
31
55
100
122
183
500
Raw
raw celery
33
24
8
17
28
40
60
80
120
20i
Vegetables
raw cucumbers
K0
76
8
24
70
110
15S
220
316
8'10
lettuce/tossed salads
65
59
10
20
55
93
140
IB6
270
1030
raw onions
31
33
3
17
18
36
57
72
180
350
raw tomatoes
SI
55
30
45
62
113
123
182
246
728
cooked broccoli
112
68
30
78
90
155
185
190
350
6S0
cooked cabbage
128
83
28
75
145
150
225
300
450
610
cooked carrots
79
50
19
46
75
92
150
155
276
736
com on/off coli
95
56
21
65
83
123
170
170
330
850
limn beans
110
75
21
67
88
170
175
219
350
875
Cooked
cow peas, field peas and
blnckeye peas
131
88
22
SS
83
175
196
350
350
700
cooked green peas
90
57
20
43
65
85
170
170
330
680
cooked spinach
121
70
2-1
78
103
185
205
205
380
454
string beans
86
54
18
67
70
135
140
140
280
810
cooked summer squash
145
98
27
105
103
215
215
352
430
860
cooked sweel potatoes
136
87
30
86
114
185
225
238
450
1020
cucumber/pickles
45
45
7
16
30
65
90
130
222
45s{
& 'y
V
17
-------
ikiiV iJVi i
tiui 10.41 0,1^1 J.. • REPEJ^ENCES'
•O
'<¦• A
Calabrese, E.J., Barnes, Rv Stanek, E.J., Pastides, H., Gilbert, C.E., Veneman, P., Wang, X., Lasztity, A.,
and P.T. Kosteck. 1989. How Much Soil Do Young Children Ingest; an Epidemiologic StudyyRe^.
Tox, arid Pharmac. 10:123-137. ^ \
&
Davis, Sv Waller, PI, Buschbom, R., Ballou, J. and P. White. 1990. Quantitative Estimates of Soil
Ingestion in Normal Children.between the Ages of 2 and 7 Years; Population-based Estimates Using
Aluminum, Silicon, and Titanium as Soil Tracer Elements. Axe. Environ. Health. 45(2):112-122.
Hawley, J.K 1985. Assessment of health risk from exposure to contaminated soil. Risk Analysis 5(4):
289-302.
Pao, E.M, Fleming, K.H., Guenther, P.M. ei ah 1982, Poods commonly eaten by individuals: amount
per day and per eating occasion. U.S. Department of Agriculture. Home Economics Report No. 44.
Sedman, R. 1989. Development of Applied Action Levels for Soil Contact: A Scenario for the
Exposure of Humans to Soils in a Residential Setting. Environmental Health Perspectives. Vol 79,
pg 291-313.
U.S. Bureau of Labor Statistics. 1990. Statistical summary: tenure with current employer as of January
1987. (transmitted via facsimile, Sept. 7, 1990).
t/SDA., 1980. U.S. Department of Agriculture. Food and nutrient intakes of individuals in one day
in the United States, Spring 1977. Nationwide Food Consumption Survey 1977-1978. Preliminary
Report No.2.
i.S. EPA 1989. Risk Assessment Guidance for Superfund: Human Health Evaluation Manual (Part
4). EPA/540/1-89-002. December 1989.
Jj.S. EPA 1989b. Review of the National Ambient Air Quality Standards for Lead: Exposure Analysis
nethodology and Validation. USEPA Office of Air Quality Planning and Standards. EPA 450/2-
'9/011.
I.S. EPA 1990. Exposure Factors Handbook. EPA/600/8-89-043. March 1990.
j.S. EPA 1991. "Human Health'Evaluation Manual, Supplemental Guidance: Standard Default
Wposure Factors". OSWER Directive 9285.6-03. March 25,1991.
'•S. EPA 1992. "Guidance on Risk Characterization for Risk Managers and Risk Assessors.
Memorandum from Henry Habicht to Assistant Administrators. Feb. 26, 1992
18
------- |