Review Draft— Do Not Cite or Quote—December 1994
                                                                                                     w-Ott. OSUjg-A^
                                United States
                                Environmental Protection
                                Agency
                                                            Office of
                                                            Solid Waste and
                                                            Emergency Response
             9355.-4-14FS
            EPA/540/R-94/101
            PB95-963529
            December 1994
                                Soil  Screening  Guidance
    Office of Emergency and Remedial Response
    Hazardous Site Control Division
                                                                                   Quick Reference Fact Sheet
    NOTICE: This document is draft for review only and should not be used until the guidance is finalized following public comment and peer review.
BACKGROUND

On June 19,1991, the U.S. Environmental Protection Agency's
(EPA's) Administrator charged the Office of Solid Waste and
Emergency Response  (OSWER) with conducting a 30-day
study to outline options for accelerating the rate of cleanups at
National Priorities List  (NPL) sites.   One of the  specific
proposals of the study was for OSWER to "examine the means
to develop standards or guidelines for contaminated soils."

On June 23, 1993, EPA announced the development of "Soil.
Trigger Levels" as one of the Administrative Improvements to
   ISuperfund program.  On September 30, 1993, a draft fact
  eet was released that presented generic Soil Screening Levels
(SSLs) for 30 chemicals.  The fact sheet presented standard-
ized  equations to model exposures to soil contaminants via
ingestion, inhalation, and migration to ground water. The fact
sheet provided generic defaults for each parameter in the equa-
tions and a sampling methodology to measure soil contaminant
levels.  The SSL initiative underwent widespread review both
within  and outside the Agency.  Suggestions were made on
how  to improve the methodology and increase the usefulness
of screening levels by finding simple ways to modify them
using site-specific data.

Based  on  that review, EPA modified the  SSLs into a Soil
Screening framework that emphasizes the  application  of
standardized equations for the site-specific  evaluation of soil
contaminants.   This framework provides an overall approach
for developing SSLs for specific contaminants and exposure
pathways at a site under a residential land use scenario. Areas
with soil contaminant concentrations below SSLs generally
would  not warrant  further  study  or action under  the
Comprehensive Environmental Response, Compensation, and
Liability Act (CERCLA).
^jne
The Soil Screening framework's point of departure is a simple
niethodology for calculating  site-specific SSLs using easily
  itained site data with standardized equations.  An option for
'conducting  a  more  detailed site-specific analysis is  also
included in the framework. In addition, default parameters are
                                                              used in  the standardized equations to produce a table of
                                                              generic Soil Screening Levels for 107 chemicals that update
                                                              those presented in the September 30,  1993, draft SSL fact
                                                              sheet. These generic SSLs are included in the framework as
                                                              a default option for use  when site-specific values  are not
                                                              available.

                                                              PURPOSE OF SOIL SCREENING
                                                              FRAMEWORK

                                                              The Soil Screening framework represents the first of several
                                                              tools EPA plans to develop to standardize the evaluation and
                                                              cleanup of contaminated soils.  SSLs streamline the remedial
                                                              investigation/feasibility study (RI/FS) process by accelerating
                                                              and  increasing consistency in  decisions  concerning soil
                                                              contamination.  As a future companion to the Soil Screening
                                                              framework, EPA also  intends to develop a methodology to
                                                              identify levels of contamination that clearly warrant a response
                                                              action or, possibly, concentrations for which treatment would
                                                              be required.  The screening levels at the  low  end  and  the
                                                              higher concentration values that warrant response can be used
                                                              to identify the bounds of a risk management continuum (Figure
                                                              1).  Generally, within this continuum lies a range of possible
                                                              cleanup levels that will continue to be determined on a site-
                                                              specific basis.

                                                              EPA anticipates the use of the Soil Screening framework as a
                                                              tool to facilitate prompt identification of the contaminants and
                                                              exposure areas  of concern during both remedial actions and
                                                              some removal actions under CERCLA.  SSLs do not trigger
                                                                     No further study
                                                                     warranted under
                                                                       CERCLA
                                                                                 Site-specific
                                                                                   cleanup
                                                                                  goal/level
                     Response
                    action clearly
                     warranted
                                                                 "Zero"
                                                               concentration
Screening
  level
                                                                                         Response
                                                                                           level
 Very high
concentration
                                                                    Figure 1.  Risk management spectrum for
                                                                               contaminated soil.

-------

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
 the need for response actions or define "unacceptable" levels
 of contaminants in soil.  SSLs may serve as Preliminary
^Remediation Goals  (PRGs) under  certain conditions  (see
 section  on  Use  of  SSLs  as  Preliminary  Remediation
      /Cleanup  Levels).  In the future, EPA will consider
          the guidance to address the Resource Conservation
 and Recovery Apt (RCRA) Corrective Action program.

 The SSLs are, as noted above, intended for use as a tool; their
 use is not mandatory at sites being addressed under CERCLA.
 The framework leaves a broad range of discretion to the site
 manager, both on whether the SSL approach is appropriate for
 a site and,  if it is  used, on the appropriate method.  This
 guidance anticipates three optional approaches—simple site-
 specific, detailed site-specific, and generic.  In  the first two,
 some or all default values would be replaced as appropriate
 with site-specific data. Furthermore, the models themselves
 are not codified as rules and can be modified if appropriate,
 although  some  explanation should be provided if such
 modification is made.

 SOIL SCREENING  FRAMEWORK

 A Soil Screening Level is a chemical  concentration in soil that
 represents a level of contamination below which there is no
 concern under CERCLA, provided conditions associated with
 the SSLs are met. Generally, if contaminant concentrations in
 soil fall below the SSL, and there are  no significant ecological
 receptors  of concern, then  no further study  or  action is
      nted for residential use of that area. (Some States have
  Iveloped screening numbers that are more stringent than the
 generic SSLs presented in this fact sheet; therefore further
 study may be warranted under State programs.) Concentra-
 tions in soil above either the generic or site-specific screening
 level  would not automatically  designate a  site  as  "dirty" or
 trigger a response action.  However, exceeding a  screening
 level  suggests that a further evaluation of the potential risks
 that may be posed by site  contaminants  is appropriate to
 determine the need for a response action.

 The Soil Screening framework presents  three approaches for
 establishing  screening  levels. The option emphasized in this
 Fact Sheet is a simple method that incorporates readily obtain-
 able,  site-specific data into  standardized equations  to derive
 site-specific screening levels for selected contaminants. When
 questions still exist at a site regarding whether or not contam-
 inant  levels are of concern, as a  second approach, more
 tailored screening levels can be derived for most contaminants
 by incorporating additional site data  into more complex fate
 and transport models.  The  third approach is  to apply the
 generic SSLs presented in Appendix A.  Although the default
 parameters used to derive the generic  SSLs are not necessarily
 "worst case," they are  conservative.

 The progression from generic  to simple  site-specific and
  etailed (full-scale) site-specific SSLs usually will involve an
 •crease in investigation costs and a decrease in conservatism
 (Figure 2).  Generally, the decision of which method to use
    More
                 Conservatism
                                 •>•  Less
Generic
    SSL
                     Simple
                  Site-Specific
                    Method
                                       Detailed Srte-
                                   >•  Specific Method
               Investigation Costs
    Less ^	>•  More
    Figure 2. Components of the Soil Screening
                      framework.
involves balancing the increased investigation costs with the
potential savings associated with higher (but protective) SSLs.
Therefore, the framework promotes the option of using  site-
specific data  to derive screening  levels.   More guidance
regarding which option to use is presented later in this fact
sheet.

Site-Specific SSLs:   Simple Method

The simple method for developing site-specific SSLs requires
the collection of a  small  number of easily obtained site
parameters (e.g., fraction organic carbon, percent soil moisture,
and dry bulk; density) for use in the standardized equations so
that the calculated screening levels can be appropriately con-
servative for the site but not as conservative as the generic
values.  Once derived, the user then compares measured site or
area contaminant concentrations to the site-specific screening
levels.  If concentrations do not exceed the SSLs for each
pathway of concern,  it would generally be appropriate to
exclude the area from further investigation.  If the levels are
exceeded,   the site  manager  may  decide  that  a more
comprehensive evaluation  is  needed  to determine the  risk
posed  via  a  particular exposure pathway (see Technical
Background section).

Site-Specific SSLs:   Detailed  Approach

A more detailed method for developing site-specific SSLs is a
full-scale model evaluation requiring  the collection of addi-
tional site data. Full-scale modeling allows the application of
complex transport and fate models and allows for consideration
of a finite contaminant source.  Applying these models  will
further  define the risk associated with  exposure via the
inhalation or migration to ground water pathway. The model
application may show that there is no concern over exposure
from the pathway, thereby eliminating it from further concern.
This potential outcome provides the incentive for incurring the
cost and time to conduct a comprehensive site evaluation.

-------
                                 Review Draft—Do Not Cite or Quote—December 1994
Generic SSLs

Generic SSLs can be used in place of site-specific screening
levels. The decision to use generic SSLs will likely be driven
by time and cost.  The site manager must weigh the cost of
conducting a more site-specific investigation with the potential
for deriving a higher SSL that provides for an appropriate level
of protection.   The Technical Background section of this
guidance presents a more detailed discussion of the level of
effort required to conduct further study of site conditions and
risks. Appendix A provides generic SSLs for 107 chemicals.

SCOPE OF SOIL SCREENING FRAMEWORK

The Soil Screening framework has been developed for 107
chemicals using assumptions for residential land use activities
for three pathways of exposure (see Figure 3):

•   Ingestion of soil

•   Inhalation of volatiles  and fugitive dusts

•   Ingestion of contaminated ground water caused by migra-
    tion of chemicals through soil to an underlying potable
    aquifer.

Reviews of risk assessments at hazardous waste sites indicate
that these pathways are the most common routes of human
exposure to contaminants  in the residential setting. These are
also the pathways for which generally accepted  methods,
models,  and assumptions have  been developed that lend
themselves to a  standardized approach.   Data on  dermal
exposures have also been  considered, and the generic SSL for
                   Direct Ingestion
                     of Ground
                   Water and Soil
Inhalation
                                     Blowing
                                     Dust and'
                                     Volatilization
                      Highlight 1: Key Attributes of the SSL Framework

                      •  Standardized equations are presented to address
                         three individual human exposure pathways.

                      •  Parameters are identified for which site-specific
                         information is needed to develop site-specific SSLs.

                      •  Default values are provided and used to calculate
                         generic SSLs that are consistent with Superfund's
                         concept of "Reasonable Maximum Exposure" (RME).

                      •  SSLs are generally based on a 10"6 risk for
                         carcinogens, or a hazard quotient of 1 for noncar-
                         cinogens.  SSLs for migration to  ground water are
                         based on nonzero maximum contaminant level goals
                         (MCLGs), or, when not available, maximum contami-
                         nant levels (MCLs). Where neither of these are
                         available, the aforementioned risk-based targets  are
                         used.
   Figure 3.  Exposure pathways addressed by the
              Soil Screening framework.
pentachlorophenol has been modified accordingly. The scope
of the SSL framework is limited to human exposure via the
pathways listed above; therefore, sites with other significant
exposure  pathways, nonresidential land  uses,  possible
ecological concerns, or  unusual site conditions should
consider their associated risks on a site-specific basis apart
from the SSL  framework.   Key  attributes  of  the  Soil
Screening framework are given in Highlight 1.

Soil Ingestion Pathway

For the direct soil ingestion pathway,  only generic SSLs were
developed. Simple and full-scale site-specific methods were
not developed because cost and complexity make developing
site-specific data for this pathway, such as soil ingestion rates
or chemical-specific bioavailability, generally impracticable.
However, EPA is evaluating  the  data available to support
adjustment of the exposure frequency term based on regional
climatic conditions.

Inhalation  Pathway

For  inhalation of volatiles and fugitive dust,  both generic
values and a method for incorporating site-specific data into
the standardized equations have been  developed. To estimate
the site-specific potential for volatilization of contaminants, soil
conditions such  as  fraction organic carbon, soil moisture
content, and dry bulk density must  be evaluated. To estimate
the site-specific potential for generation of fugitive dusts, other
parameters must be evaluated, such as mean annual windspeed,
threshold friction velocity, and the mode soil aggregate size to
further tailor the SSLs to the site.  For both the inhalation of
volatiles  and  fugitive  dust  pathways,  a  site-specific
determination of the area of contamination and  meteorologic
inputs can be incorporated into dispersion calculations.

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
 Migration to Ground Water

*The  simple  site-specific  method for addressing  potential
      :ninant migration to ground water uses the same soil
      eters required to address volatilization, along with easily
      able hydrogeologic parameters.  The simple site-specific
 method for this exposure pathway also requires a determination
 of the area of contamination.

 Other Pathways

 Additional exposure pathways to contaminants in soil—dermal
 absorption,  plant  uptake,  and migration of volatiles  into
 basements—may contribute significantly to the risk to human
 health in a residential setting.  The  Superfund program has
 evaluated the data and methods  available  to address these
 potential exposures and has incorporated as much information
 as possible into the SSL framework.

 Based  on limited empirical  data,  the  ingestion  SSL  for
 pentachlorophenol has been adjusted to account for potential
 dermal exposure.  Additionally, empirical data  indicate that
 plant uptake may  be important for some chemicals (Le., As,
 Cd, Hg, Ni, Se, Zn).  The fact that these chemicals' potential
 for plant uptake and  dermal  absorption has been noted  in
 Appendix A should not be misinterpreted to mean that other
 chemicals are not of potential concern for dermal exposure or
 plant uptake.  As additional information becomes available,
 other chemicals may be addressed as well.

  It this time, Superfund does not believe that the potential for
 migration of contaminants  into basements can be reasonably
 incorporated into the SSL framework.  The parameters required
 for the  models (e.g., the  number  and size of cracks  in
 basement walls) do not lend themselves to standardization or
 to evaluation of potential future exposure, and the models have
 not been adequately validated.  The Technical Background
 Document (U.S. EPA, 1994e)  provides a detailed analysis of
 available modeling of this pathway.

 Other Land Uses

 Longer-term efforts will be required to develop standardized
 tools to address exposures relevant to other land uses such as
 industrial land use.   The results of these efforts  may  be
 included in future revisions of this guidance.

 Ecological Receptors

 As part of the baseline risk assessment, an ecological assess-
 ment should be conducted at every Superfund site.  The SSL
 framework does not attempt to define significant ecological
 receptors or quantify ecological risks. However, a comparable
 list of screening level benchmarks, called Ecotox Thresholds,
 is being developed by Office of Emergency and Remedial
  Response (OERR) for application during the ecological risk
 assessment addressed in OSWER Directive No. 9285.7-17
 (U.S. EPA,  1994d).   These  values are defined as media-
specific chemical concentrations above which there is sufficient
concern regarding adverse effects to ecological receptors to
warrant further site  investigation.    OERR  is  developing
guidance  on  designing  and  conducting  ecological  risk
assessments, that will describe the use of such screening values
in the Superfund Remedial Investigation process.

HOW TO USE THE SOIL SCREENING
FRAMEWORK

The decision to use the Soil Screening framework at a site will
be driven by the potential benefits of eliminating areas,
exposure pathways, or contaminants from further investigation.
By identifying areas where concentrations of contaminated soil
are below levels of concern under CERCLA,  the framework
provides a means to focus resources  on exposure areas,
contaminants, and exposure pathways of concern.

Highlight 2 outlines flie process of applying the Soil Screening
framework at a site.   To enable early comparison with site
background  concentrations  and  to provide  information
necessiary  for  determining  an  adequate sample  size,  site-
specific SSLs should be developed as early in the process as
possible.   They  can be  adjusted  during the  process  to
accommodate additional site information and the resulting
changes to the conceptual site model.

Developing a  Conceptual Site Model

The primary condition for use of SSLs is that exposure path-
ways of concern and conditions at the site match those taken
into account by the Soil Screening  framework.  Thus, at  all
sites it will be necessary to develop a conceptual site model to
identify likely contaminant source areas, exposure pathways,
and potential receptors.   This information can be used to
   Highlight 2:  Using the Soil Screening Framework

   •  Develop site conceptual model and compare with
     SSL conceptual model to determine applicability of
     framework.

   •  Determine if background contaminant concentrations
     are above generic SSLs.

   •  Select approach (simple or detailed she-specific,
     generic) and develop SSLs.

   •  Measure average soil contaminant concentrations in
     exposure areas (EAs) of concern.

   •  Compare average soil concentrations with SSLs and
     eliminate site or area of site where EA mean
     concentration is less than SSL.

   •  Consider further study or use of SSLs as PRGs for
     sites or site areas with contaminant concentrations
     greater than SSLs.

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
determine the applicability of the framework at the site and the
need for additional information.

A conceptual  site model is developed  from available site
sampling data, historical records, aerial photographs, and
hydrogeologic information. The model establishes a hypothesis
about possible contaminant sources, contaminant  fate and
transport, exposure pathways,  and potential receptors.  The
DQO Guidance for Superfund (U.S. EPA, 1993a) provides an
excellent discussion on the development of a conceptual site
mode].  The rationale for including the contaminant migration
to ground water exposure pathway should be consistent with
EPA ground water policy (U.S. EPA,  1988,  1990b,  1992a,
1992b, 1993b).

The conceptual model upon which the generic SSLs are based
is a 30-acre property that has been divided up for residential
use.  Thus, the generic  SSLs have been developed to be
protective for source areas up to 30 acres. The contamination
is assumed  to be evenly distributed across the area of concern
and extends from the ground surface to the top of the aquifer.
The soil type is  assumed to  be loam that  has 50 percent
vegetative  cover.   Loam is soil with  approximately equal
proportions of sand and silt  Exposure to contaminants can
occtir via ingestion of soils, inhalation of volatiles and fugitive
dusts, or migration to ground water.

For the migration to  ground  water pathway, the point  of
compliance is assumed to be at the edge of the site, which is
assumed to be homogeneously contaminated. No attenuation
is considered  in the unsaturated  zone; however, dilution is
assumed within the aquifer to the point of compliance.  For the
generic conceptual site model, the source is assumed to extend
acrpss  the  entire  site.  See Figures 3  and 4 for a  graphic
representation of aspects of the  conceptual model applicable to
the Soil Screening framework.

Partitioning of contaminant   mass  between  media  is not
addressed in the SSL framework because the fate and transport
models used to derive the generic SSLs are based on the
assumption of an infinite  source.  Although the assumption is
highly conservative, a  finite source  model cannot be  applied
unless there  are  accurate data regarding  source  size and
volume. Obviously, in  the case of the generic SSLs, such data
are not available.  It is also unlikely that such data  will  be
available from the limited subsurface sampling that is  done to
apply the simple site-specific method. Thus, it is most likely
that a finite source model would  be  applied  as part of a
detailed site-specific investigation. EPA will continue to seek
consensus  on the  appropriate methods  to  incorporate
contaminant partitioning  and a finite source into the simple
site-specific method.   The results  of these efforts may  be
included in future updates to this guidance.

The Technical Background Document (U.S. EPA,   1994e)
presents information  on equations and models  that can
accommodate finite sources and predict the subsequent impact
on either ambient air or ground water.  However, when using
                   SECTION VIEW
                 Receptor
                   Well
                                          Land Surface
                              Unsaturated
                                 Zone     Water Table
    Ground Water
     -\_
      Flow
                                        Saturated Zone
         Default assumptions:
         • Infinite source
         • Source extends to water table
         • Well at downgradient edge of source
         • 30-acre source size
   Figure 4. Migration to ground water pathway-
               SSL conceptual model.
a finite source model, the site manager should recognize the
uncertainties inherent in site-specific estimates of subsurface
contaminant distributions and use conservative estimates of
source size and concentrations to allow for such uncertainties.

The following questions should always be considered in the
development of the conceptual site model before applying the
Soil Screening framework:

•  Is the site adjacent to surface waterbodies where the
   potential for contamination of surface water by overland
   flow or release of contaminated ground water should be
   considered?

•  Are   there potential  terrestrial or aquatic  ecological
   concerns?

•  Is there potential for land use other than residential?

•  Are  there other likely human exposure  pathways  that
   were not considered in development of the SSLs (e.g., local
   fish   consumption; raising  of beef,   dairy,   or  other
   livestock)?

•  Are there unusual site conditions (e.g., area of contamina-
    tion  greater than 30 acres,  unusually high fugitive dust
    levels due to soil being tilled for agricultural use, or heavy
    traffic on unpaved roads)?

-------
Review Draft—do Not Cite or Quote—December 1994
  If ftie conceptual site model indicates that residential assump-
. tions are appropriate for your site and no pathways of concern
  other than those covered by the Soil Screening framework are
     ent, then the framework may be applied directly to (he site.
   i the conceptual site model indicates that the site is more
  . omplex than the scenario outlined in this guidance, the frame-
  work above will not be sufficient. Additional pathways, recep-
  tors, or chemicals must  be evaluated on a site-specific basis.

  Considering Background Contamination

  A necessary step  in  determing  the  usefulness of the SSL
  framework  is the  consideration of background contaminant
  concentrations,  since  the framework  will have little  utility
  where background concentrations exceed the SSLs.

  EPA may be concerned with two types of background at sites:
  naturally occurring and anthropogenic. Natural background is
  usually limited to metals whereas anthropogenic (i.e., human-
  made) background includes both organic and inorganic contam-
  inants.

  Generally, EPA does not clean up below natural background;
  however, where anthropogenic background levels exceed SSLs
  and EPA has determined that a response action is necessary
  and feasible, EPA's goal will be to develop a comprehensive
  response to the widespread contamination.  This will often
  require  coordination  with different  authorities  that  have
  jurisdiction  over other sources of contamination in the area
   :uch as  a regional air board or RCRA program).  This will
    :lp  avoid response actions that create "clean islands" amid
  widespread  contamination. The background information and
  understanding of the site developed as part of the conceptual
  model can help determine background concentration.

  When considering background, one should also consider the
  bioavailability and mobility of compounds. Some compounds
  may form complexes that are immobile and unlikely to cause
  significant risk. This situation is more likely to occur with
  naturally occurring compounds.  Therefore, background con-
  centrations  of compounds exceeding the SSLs do not neces-
  sarily  pose a threat.   Alternately,  activities at a site can
  adversely affect the natural soil geochemistry, resulting in the
  mobilization of compounds.  Consequently, background con-
  tamination should be considered carefully.  Regardless, where
  background concentrations are higher than the SSLs, the SSLs
  generally will not be the best tool for site decisionmaking.

  Sampling Exposure Area

  After the conceptual site model has been developed, and the
  applicability of the Soil Screening framework is determined,
  the next step is to  collect a representative sample set for each
  exposure area. An exposure area is defined as that geographic
  ^area in which an individual may be exposed to contamination
  Dver time.   Because SSLs are developed  for a residential
  Scenario, EPA assumes the exposure area is a 0.5-acre
  residential lot.
                             In those situations where little or no sampling has been done,
                             it will be beneficial to collect the site data required for the
                             simple site-specific methodology in tandem with the collection
                             of samples to identify contaminant concentrations.  The site
                             manager should work to limit the total number of trips  to the
                             site by maximizing the usefulness of the samples collected.
                             (See section on Measuring Contaminant Concentrations in Soil
                             for additional guidance.)

                             Comparing Exposure Area Concentration
                             to SSLs

                             The fourth step is to compare onsite soil contaminant concen-
                             trations with site-specific SSLs or the generic SSLs listed in
                             Appendix A.  At this point, it is reasonable to review the
                             conceptual site model with  the actual site data in hand to
                             reconfirm the accuracy of the conceptual site model and the
                             applicability of the Soil Screening framework.   Once this is
                             confirmed, site contaminant levels may be compared with the
                             SSLs.

                             In Appendix A, the first column to the right of the chemical
                             name presents levels based on direct ingestion of soil. The
                             second column presents the levels based on inhalation of vola-
                             tiles or soil particulates. The third column presents SSL values
                             for the migration to ground water pathway multiplied by a
                             dilution and attenuation factor (DAF) of 10 to account for
                             natural processes that reduce contaminant concentrations in the
                             subsurface.  The fourth  column contains the SSL multiplied
                             by a DAF of 1, which may be appropriate to use in instances
                             where there are high water tables, karst topography, fractured
                             bedrock, or source size greater than 30 acres. The lowest SSL
                             of the three pathways (ingestion, inhalation, and ground water
                             with DAF of 10) is highlighted in bold for each contaminant.

                             Generally, the comparison of SSLs to site contaminant levels
                             will result in one of three outcomes:

                             1. Site-measured values indicate that an area falls below all of
                                the SSLs. Soils from these areas of the site generally can
                                be eliminated from further evaluation under CERCLA.

                             2. Site-measured data indicate that one or more SSLs have
                                clearly been exceeded. In this case, the SSLs have helped
                                to identify site areas, contaminants, and exposure pathways
                                of potential concern on which  to focus further .analysis or
                                data-gathering efforts.

                             3. A site-measured value exceeds one pathway-specific value
                                but not others.   In  this case, it  is  reasonable to focus
                                additional site-specific data collection efforts only on data
                                that will help determine whether there is truly a risk posed
                                via that pathway or by a limited set of chemicals  at the
                                site.   When an exceedance is marginally significant, a
                                closer look at site-specific conditions and exposures may
                                result in the area being eliminated from further study.

-------
                                 Review Draft—Do Not Cite or Quote—December 1994
Use of SSLs as Preliminary Remediation
Goals/Cleanup Levels

SSLs are not nationwide cleanup levels or standards.  Where
the basis for response action exists and all exposure pathways
of concern are addressed by the SSLs, the SSLs may serve as
PROs as defined in HHEM, Part B (U.S. EPA,  1991d).  A
PRO is a strictly risk-based value that serves as the point of
departure for the establishment of site-specific cleanup levels.
PRGs are modified to become final cleanup levels based on
a consideration of the nine-criteria analysis described in the
National Contingency  Plan (NCP; Section 300.430  (3)(2)
(i)(A)),  including  cost,  long-term  effectiveness,  and  imple-
mentability.  See  Role  of the  Baseline Risk Assessment in
SuperfundRemedy Selection Decisions (U.S. EPA, 1991e) for
guidance on how to modify PRGs to generate cleanup levels.

The SSLs should only be used as site-specific cleanup levels
when a  nine-criteria evaluation using the SSLs as PRGs for
soils indicates that a selected remedy achieving the SSLs is
protective, ARAR-compliant, and appropriately balances the
other criteria, including cost. An example is a small site or
exposure area where the cost of additional study would exceed
the cost of remediating to the generic SSLs.

Addressing Exposure to Multiple Chemicals

The SSLs generally correspond to a 10"6 risk level for carcino-
gens and a hazard quotient (HQ) of 1 for noncarcinogens.
This "target" hazard quotient  is  used to calculate  a  soil
concentration below which it is unlikely  for even sensitive
populations to experience adverse health effects. The potential
for additive effects has not been "built in" to the SSLs through
apportionment. For carcinogens, EPA believes that setting a
lfJ6risk level for individual chemicals and pathways generally
will lead to cumulative risks within the 10~4 to 10"6 risk range
fo^ the combinations of chemicals typically found at Superfund
sites.

For noncarcinogens, there  is no widely accepted risk range.
Thus, for  developing national  screening levels,  options are
either (1) to set the risk level for individual contaminants at the
Rfb or RfC (i.e., a hazard quotient of 1), or (2)  to  set
chemical-specific concentrations by apportioning risk based on
some arbitrarily chosen fraction of the  acceptable risk level
(e.g,, one-fifth or one-tenth the RfD or RfC).  The Agency
believes, and EPA's Science Advisory Board agrees (U.S.
EPA, 1993d), that noncancer risks should be added only for
those chemicals with the same toxic endpoint or mechanism of
action.

Highlight 3 lists the chemicals from Appendix A that have
SSLs based on noncarcinogenic toxicity and affect the same
target organ. If more  than one chemical detected at a site
affects the same target organ (i.e., has the same critical effect
as defined by the RfD methodology), site-specific SSLs for
each chemical in the group should be divided by the  number
of chemicals present The concentration of contaminants at the
  Highlight 3: SSL Chemicals with Noncarcinogenic
              Toxic Effects on Specific Target Organs
  Kidney
    Acetone
    1,1-Dichloroethane
    Dimethyl phthalate
    2,6-Dinftrotoluene
    Di-n-octyl phthalate
    Nitrobenzene
    2,4,5-Trichlorophenol
    Vinyl acetate

  Liver
    Acetone
    Chlorobenzene
    Di-n-octyl phthalate
    Nitrobenzene
    2,4,5-Trichlorophenol

  Central Nervous System
    Butanol
    2,4-Dichlorophenol
    2,4-Din'rtrotoluene
    2,6-Dinitrotoluene
    2-Methylphenol
Circulatory System
 Antimony
 Barium
 p-Chloroaniline
 c/s-1,2-Dichloroethylene
 Nitrobenzene
 Zinc

Reproductive System
 Carbon disulfide
 2-Chlorophenol
 1,2,4-Trichlorobenzene

Gross Pathology
 Diethyl phthalate
 2-Methylphenol
 Naphthalene
 Nickel
 Vinyl acetate
site should then be compared  to the SSLs that have been
modified to account for this potential additivity.

Because the combination of contaminants will vary from site |
to site,  apportioning  risk to account for potential additive
effects could not be considered in the development of generic
SSLs. Furthermore, for certain noncarcinogenic organics (e.g.,
ethylbenzene, toluene), the; generic SSLs are not based on
toxicity  but are determined instead by a "ceiling limit"
concentration (C^ at which these chemicals may occur as
nonaqueous  phase  liquids (NAPLs) in soil (see Technical
Background section).   For these reasons,  the potential for
additive effects and the need to  apportion risk must be a site-
specific  determination.

TECHNICAL BACKGROUND

The models and assumptions supporting the Soil Screening
framework were developed to be consistent with Superfund's
concept of "reasonable maximum exposure" (RME) in the
residential  setting.    The Risk Assessment Guidance  for
Superfund, Volume 1  (U.S. EPA,  1989b) and the Standard
Default Exposure Factors guidance (U.S. EPA, 1991b) outlined
the Superfund program's approach to calculating an RME.
Since that time, the Agency (U.S. EPA, 1991a) has coined a
new term that the Superfund program believes corresponds to
the definition of RME: "high-end individual exposure."

The Superfund  program's method to estimate the RME for
chronic  exposures  on a site-specific basis is to combine an
average  exposure  point concentration  with  reasonably

-------
                                    Review Draft—Do Not Cite or Quote—December 1994
   conservative values for intake and duration in the exposure
 ^ calculations. The default intake and duration assumptions are
  * presented in the Standard Default Exposure Factors guidance
   (U.S. EPA,  1991b). The duration assumptions were chosen to
^kpresent individuals  living  in  a  small town  or  other
^Kmtransient community.  (Exposure to members  of a more
   transient  community  is assumed  to be  shorter and  thus
   associated with lower risk.) Exposure point concentrations are
   either measured at the site (e.g., ground water concentrations
   at a receptor well) or estimated using exposure models with
   site-specific model inputs.  An average concentration term is
   used in most  assessments where the focus is on  estimating
   long-term, chronic exposures.  Where the potential for acute
   toxicity is of concern, exposure estimates based on maximum
   concentrations may be more appropriate.

   The resulting site-specific estimate of RME is then compared
   with chemical-specific toxicity criteria such as RfDs or RfCs.
   EPA recommends using criteria from the Integrated  Risk
   Information System  (IRIS)  (U.S. EPA, 1994c) and  Health
   Effects Assessment Summary Tables (HEAST) (U.S. EPA,
   1993c), although values from other sources may be used in
   appropriate  cases.

   The Soil Screening  framework differs from  a site-specific
   estimate of risk in that the exposure equations and models are
   run  in reverse to backcalculate to an "acceptable level"  of
   contaminant in soil.  Toxicity criteria are used to define the
   acceptable level:  a level corresponding to a 10~6 risk for
  _carcinogens and a hazard quotient of 1  for noncarcinogens.
    tie concept of backcalculating to an acceptable level in soil
   Pas presented in RAGS Part B (U.S. EPA, 199Id), and the
   Soil  Screening  framework serves  to update  Part  B for
   addressing residential soils.  Site-specific SSLs are consistent
   with the Superfund approach to estimating RME on a site-
   specific basis.  Standard default factors are used for the intake
   and duration assumptions, site-specific inputs are used in the
   exposure  models,   and  chemical-specific   concentrations
   averaged over the exposure area are used for comparison to the
   SSLs.

   Consistent with the site-specific SSLs, the generic SSLs use
   the same intake and duration assumptions and are compared to
   area average concentrations. However, the generic SSLs are
   based on a hypothetical site  model.  In developing the
   parameters for the hypothetical site, the Superfund program
   considered the conservatism inherent in the exposure models
   (e.g., assumption of an infinite source)  and then combined
   high-end and central tendency parameters  for size, location,
   and soil characteristics.  The resulting generic SSLs should be
   protective for most site conditions across the Nation.
   OERR performed a sensitivity analysis to  determine which
   parameters most influenced the output of the volatilization and
   fugitive dust models used to calculate SSLs  for the inhalation
    ithway.  For fugitive dusts, the particulate emission factor
        was most sensitive to threshold friction velocity,, which
   was set at a  "high-end"  value.   For  calculation of the
«thw
EF)
volatilization factor (VF), soil moisture content was set at a
conservative value because it drives the air-filled soil porosity
that in turn provides the pathway for chemicals to volatilize
from soils. Climatic conditions have a significant impact on
dispersion of both volatile and particulate emissions and were
set at high-end values to be protective for conditions at most
sites.    Different high-end  meteorological data  sets  were
selected to calculate 90th percentile dispersion coefficients for
the VI1 and for the PEF.

For the migration of contaminants from soils to ground water,
only average soil conditions are used to calculate generic SSLs
because of the conservatism inherent in the partition equation.
The generic DAF for this pathway was  developed using  a
weight of evidence approach to be protective under  most
hydrogeologic conditions across the country as described in the
following section on the migration to ground water.

Characteristics of the generic, hypothetical site used to develop
generic  SSLs were  described  previously in the  section
discussing  the  conceptual  site  model.   The  Technical
Background Document (U.S. EPA, 1994e) accompanying this
guidance   describes    the   pathway-specific   equations,
assumptions, and methodology that form the basis for both the
simple site-specific approach  and the generic SSLs.  The
Technical Background Document also describes development
of the specific default input  values used to calculate  generic
SSLs  for the inhalation and  migration  to  ground water
pathways.

The generic SSLs are based on default assumptions.  EPA
recognizes that site-specific conditions may differ significantly
from  these  default  assumptions.    The  Soil  Screening
framework emphasizes the substitution of some of the generic
fate and transport assumptions with site-specific data to derive
site-specific SSLs.  However, one purpose of the SSLs is to
define a level  in soil below which no further study or action
would be required.  Therefore, alternative levels that are set
using site-specific data should generally be calculated assuming
the RME/"high-end" individual exposure.

The following sections present the standardized equations and
default assumptions that form the basis for the simple site-
specific methodology and the generic SSLs. The soil ingestion
discussion is  limited  to default assumptions  because  only
generic SSLs have been  developed for this pathway.

Direct Ingestion

Agency toxicity criteria for noncarcinogens establish a level of
daily exposure that is not expected to cause deleterious effects
over a lifetime (i.e., 70 years). Depending on the contaminant,
however, exceeding the RfD (i.e., the "acceptable" daily level)
for a  short period of time may  be  cause for concern.   For
example, if there is reason to believe that exposure to soil may
be higher at a particular  stage of an individual's lifetime, one
would need to protect for that shorter period of high exposure.
Because  a number of studies have shown that inadvertent

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
ingestion of soil is common among children age 6 and younger
(Calabrese et al., 1989; Davis et al., 1990; Van Wijnen et al.,
1990), the SSLs in the default option are set at concentrations
that are protective of this increased exposure during childhood
by ensuring that the chronic reference dose (or RfD) is not
exceeded during this shorter (6-year) time period (Equation 1).
If there is reason to believe that  exposures at a site may be
significant over a short period of time (e.g., extensive soil
excavation work in a dry region), depending on the contami-
nanj, the site manager should consider the potential  for acute
health effects as well.
Equation 1 : Screening Level Equation for
Ingestion of Noncarcinogenic
Contaminants in Residential Soil
omnni , ,.,„, (m3^j . THQ x BW x AT x 365 d/yr
' " " 1/RfD0 x 10"6 kg/mg x EF x ED x IR
Parameter/Definition (units)
THQrtarget hazard quotient (unitless)
BW/body weight (kg)
AT/averaging time (yr)
Rf D0 /oral reference dose (mg/kg-d)
EF/exposure frequency (d/yr)
ED/exposure duration (yr)
IR/soiJ ingestkm rate (mg/d)
Default
1
15
6s
chemical-specific
350
6
200
* For noncardnogens, averaging time is equal to exposure duration.
In some cases, children may ingest large amounts of soil (i.e.,
3 to 5 grams) in a single event. This behavior, known as pica,
may  result  in  relatively  high  short-term  exposures  to
contaminants in soils. Such exposures may be of concern for
contaminants  that primarily  exhibit acute health effects.
Review of clinical reports on contaminants  addressed in this
guidance suggests that acute effects of cyanide and phenol may
be of concern in children exhibiting  pica behavior.  If soils
containing cyanide  and  phenol  are of concern and pica
behavior  is expected  at a site,  the protectiveness  of the
ingestion SSLs for these chemicals should be reconsidered.

For carcinogens, both the magnitude and  duration of exposure
are important Duration is critical because the toxicity criteria
arc based on "lifetime average daily dose." Therefore, the total
dose received, whether it be over 5 years or  50 years, is
averaged  over a lifetime of 70 years.  To be protective of
exposures  to  carcinogens in  the  residential setting, OERR
foojses on exposures to individuals who may live in the same
residence for a "high-end" period of time (i.e., 30 years). As
mentioned previously, exposure  to   soil  is  higher during
childhood and decreases  with age. Thus, Equation 2 uses a
time-weighted average soil ingestion rate  for  children  and
adults.   The derivation of  this  time-weighted average is
presented in U.S. EPA (1991d).
Equation 2: Screening Level Equation for
Ingestion of Carcinogenic
Contaminants in Residential Soil
Screening Level _ TR x AT x 365 d/yr
(mg/kg) SF0 x 10-* kg/mg x EF x IFsoH/adj
Parameter/Definition (units:)
TR/target cancer risk (unrtlesss)
AT/averaging time (yr)
SF0 /oral slope factor (mg/kg-d)"1
EF/exposure frequency (d/yr)
I Fsoii/ajj /age-adjusted soil ingestion
factor (mg-yr/kg-d)
Default
10"6
70
chemical-specific
350
114
Inhalation of Volatiles and Fugitive Dusts

Agency toxicity data indicate that risks from exposure to some
chemicals via inhalation far outweigh the risks via ingestion.
The models and assumptions  used to  calculate SSLs for
inhalation of volatiles and fugitive dusts are updates of the
equations presented in U.S. EPA's HHEM Part B guidance
(U.S. EPA, 1991d).  The volatilization factor (VF), soil
saturation limit (Csat), particulate emission factor (PEF),
and dispersion model have all been revised.

Another change from the Part B methodology is the separation
of the ingestion and inhalation pathways.  Toxicity criteria for
oral exposures are presented as administered doses in units of
milligrams per kilograms per day (mg/kg-d);  whereas,  the
inhalation criteria are presented as concentrations in air (pg/m3
or mg/m3) that require conversion to an estimate of internal
dose to be comparable  to the oral route.  EPA's Office of
Research  and Development (ORD)  now  believes that  the
conversion from concentration in air to  internal  dose is  not
always appropriate and suggests evaluating these  exposure
routes separately.

As explained  in HHEM Part  B, the basic principle of the
volatilization model is applicable only if the soil concentration
is at or below soil saturation (C^.  Above this  level  the
model cannot predict an accurate VF. C^ is the concentration
at which soil air, pore water, and sorption sites are saturated
and above which free-phase contaminants may be present For
compounds that  are liquid at ambient soil temperatures, C^
indicates  a concentration  above which  NAPLs  may be
suspected  in  site  soils and further investigation may be
necessary. Thus,  for liquid compounds for which the SSL
exceeds C^, the SSL is set: at C^. For compounds that are
solid at soil  temperatures for which the SSL exceeds C^
volatile emissions can be assumed to be of no concern and the
SSL is calculated considering particulate emissions only (i.e.,
the 1/VF  term in Equation 3 or 4 is set to zero).

-------
                                     Review Draft—Do Not Cite or Quote—December 1994
    Equation 3: Screening Level Equation for
                 Inhalation of Carcinogenic
                 Contaminants in Residential Soil
       lening Level
                        TR x AT x 365 d/yr
                   URF x 1000 ng/mg x EF x ED x   1  +   1
    Parameter/Definition (units)

    TR/target cancer risk (unitless)
    AT/averaging time (yr)
    URF/inhalation unit risk factor
      (ug/m3)"1
    EF/exposure frequency (d/yr)
    ED/exposure duration (yr)
    VF/soil-to-air volatilization factor
      (m3/kg)
    PEF/particulate emission factor
      (m3/kg)
                                  Default

                                  1.0'6
                                  70
                                  chemical-specific

                                  350
                                  30
                                  chemical-specific

                                  6.79 x108
    Equation 4: Screening Level Equation for
                 Inhalation of Noncarcinogenic
                 Contaminants in Residential Soil
         Screening Level
             (mg/kg)
                        THQ x AT x 365 d/yr
                   EF x ED x
f
    Tl-l
                                   [RTC x (w * PEF J
rameter/Definition (units)
    THQ/target hazard quotient (unitless)
    AT/averaging time (yr)
    EF/exposure frequency (d/yr)
    ED/exposure duration (yr)
    RfC/inhalation reference concentration
       (mg/m3)
    VF/soil-to-air volatilization factor
       (m3/kg)
    PEF/particulate emission factor
       (m3/kg)
Default

1
30
350
30
chemical-specific

chemical-specific

6.79x10s
   Equations 3 through 7 form the basis for deriving both simple
   site-specific and generic SSLs for the inhalation pathway. The
   following parameters in  the standardized  equations can  be
   replaced with specific site data to develop a more site-specific
   SSL:

   •  VF and C^
      — Average soil moisture content
      — Average fraction organic carbon content
      — Dry soil bulk density
                                                              Equation 5: Derivation of the Volatilization Factor
                                VF (m3/kg) = Q/C x   Q.14 x a. x T)1/g  x 10-4m2/cm2
                                                  (2 x Dai x  6a x KJ
                                where
                                                       Pel X 6a
                                               " ea + (Ps) (1
Parameter/Definition (units)

VF/volatilization factor (m3/kg)
Q/C/inverse of the mean cone, at the
   center of a 30-acre-square source
   (g/m2-s per kg/m3)
T/exposure interval (s)
Dei /effective diffusivity (cm2/s)
6a/air-filled soil porosit
Dj /d'rffusivity in air (cm /s)
ft/total soil porosity (Lp^/L^,,)
w/average soil moisture content
   (9v,ate/9soil or cn^wate/gsoij)
pb/dry soil bulk density (g/cnT3)
ps/soil particle density (g/cm3)
Kas/soil-air partition coefficient
   (g-soil/cm3-air)
H/Henry's law constant (atm-m3/mol)
Kd /soil-water partition coefficient
   (cm3/g)
Koc/organic carbon partition coefficient
   (crrfrg)
foc/organic carbon content of soil (g/g)
                                                                 Default
                                                                 35.10
                                                                 9.5x108s
                                                                 D,(eBa8S/h2)
                                                                 0.28 or n-wpb
                                                                 chemical-specific
                                                                 0.43 (loam)
                                                                 0.1 (10%)

                                                                 1.5 or (1  - n) ps
                                                                 2.65
                                                                 (H/Kd) x 41 (41 is a
                                                                 conversion factor)
                                                                 chemical-specific
                                                                 Koc x foe

                                                                 chemical-specific

                                                                 0.006 (0.6%)
                                                               Equation 6: Derivation of the Soil Saturation Limit
                                                                                   _£
                                                                                   Tb
                                                               Parameter/Definition (units)
                                                                      saturation concentration
                                                                  (mg/kg)
                                                               S/solubility in water (mg/L-water)
                                                               pb/dry soil bulk density (kg/L)
                                                               n/total soil porosity (L^Ls,,,,)
                                                               ps /soil particle density (kg/L)
                                                               Kj/soil-water partition coefficient (L/kg)
                                                               Koc/soil organic carbon/water partition
                                                                  coefficient (L/kg)
                                                               foc/fraction organic carbon of soil (g/g)
                                                               6^/water-filled soil porosity (L^ef/Lgoj,)
                                                               6a /ait-filled soil porosity (Lai/Lsoi|)
                                                               w/average soil moisture content
                                                                  (kUwate/kSsoil or Lwate/kSsoil)
                                                               H'/Henry's law constant (unitless)

                                                               H/Henry's law constant (atm-m3/mol)
                                     Default
                                                                 chemical-specific
                                                                 1.5 or (1 - n) ps
                                                                 0.43 (loam)
                                                                 2.65
                                                                 Koc x foc (organics)
                                                                 che m ical-specif ic

                                                                 0.006 (0.6%)
                                                                 wpb or 0.15
                                                                 n - wpb or 0.28
                                                                 0.1 (10%)

                                                                 H x41, where 41 is
                                                                 a conversion factor
                                                                 chemical-specific
                                                               10

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
Equation 7: Derivation of the Paniculate Emission
Factor
occ/~3n,~\ _ r>tf> s, 3600s/h

0.036 X (1-V) X (Um/Ut)3 X F(X)
Parameter/Definition (units)
PEF/particuIate emission factor
(m3/kg)
Q/C/inverse of the mean cone, at the
canter of a 30-acre-square source
(g/m2-s per kg/m3)
V/fractfon of vegetative cover
(unitless)
Um /mean annual windspeed (m/s)
U, /equivalent threshold value of wind-
speed at 7 m (m/s)
F(x)/functton dependent on Um/Ut
derived using Cowherd (1 985)
(unitless)
Default
6.79 x108

46.84


0.5 (50%)

4.69
11.32

0.194


•  PEF
   — Mean annual windspeed
   — Threshold friction velocity (as determined by):
      -  mode of the surface soil aggregate size
      -  roughness height
      -  correction for nonerodible particles
      -  f(x)
   — Equivalent threshold windspeed at a 7-m anemometer
      height

Site  location (to  some  extent)  and site  size (i.e., "area of
contamination") can be factored into the  simple site-specific
methodology for  the  inhalation pathways.  The dispersion
factor  (Q/C)  for both volatiles  and fugitive dusts was
calculated using  a  90th percentile meteorological data set
selected from 29 data  sets across the  United States (see
Technical Background Document [U.S. EPA, 1994e]). Los
Angeles was  selected as  the 90th percentile  data  set for
volatiles and Minneapolis was selected as the 90th percentile
data set for fugitive dusts.  Replacing the default city and site
she  of 30 acres will affect the Q/C values in both the VF and
PEF  equations   (Equations  5  and 7).    The  Technical
Background Document supporting this  guidance (U.S. EPA,
1994e) provides a table of Q/C values for 29 cities across the
country over a range of contaminant source areas for use in the
simple site-specific method.

The  particulate emission factor derived by using the default
values in Equation 7 results in an ambient air concentration of
approximately 1.5 pg/m3.  This represents an annual average
emission  rate that is based on  wind erosion  and is  not
appropriate  for  evaluating  the  potential  for more acute
exposures.
Migration to Ground Water

The methodology for addressing migration of contaminants
from soil to ground water reflects the complex nature of
contaminant fate  and transport in the subsurface.   In  this
methodology, a concentration in soil is backcalculated from an
acceptable  ground water concentration.   The  generic SSLs
presented in  Appendix  A  for  this  pathway represent  a
conservative estimation of the concentration of a contaminant
in soil that would not result in exceedances of the acceptable
concentration of a contaminant in ground water. Flexibility to
consider site-specific conditions is addressed in the simple and
detailed site-specific methodologies.

The first step in applying the SSL framework is a comparison
of the SSL conceptual model presented earlier in this document
with the conceptual model developed for the site. This forms
the basis for determining the appropriateness of conducting a
more detailed investigation and the applicability of the SSL
guidance for the migration to ground water pathway. Some of
the assumptions used to develop the SSL conceptual model
have implications for the ground water pathway.  Highlight 4
lists assumptions implicit in the conceptual model that should
be understood before applying the SSL ground water frame-
work.

Both the simple site-specific and generic methods are based on
the commonly used equilibrium soil/water partition equation
(Equation 8) that describes the ability of contaminants to sorb
 Equation 8: Soil Screening Level Partitioning
              Equation for Migration to Ground
              Water
           Screening Level
            in Soil (mg/kg)
= C IK  + (9"*e*H>)
   w[ "  "  Pb
 Parameter/Definition (units)

 C^/target soil leachate
    concentration (mg/L)
 Kysoil-water partition coefficient
 Koc/soil organic carbon/water
    partition coefficient (L/kg)
 foc /fraction organic carbon in soil
    (g/g)
 e^/water-filled soil porosity
    C-wate/'-soil)
 w/average soil moisture content
    (kgwate/kSsoil or l-wate/kgsoil)
 pb/dry soil bulk density (kg/L)
 n/soil porosity (Lp^/L^,)
 Ps/soil particle density (kg/L)
 6a/air-filled soil porosity (
 H'/Henry's law constant (unitless)
 H/Henry's law constant
    (atm-m3/mol)
         Default

         nonzero MCLG, MCL,
         or HBL x 10 DAF
         chemical-specific, «„,,
         xfoc(organics)
         chemical-specific

         0.002 (0.2%)

         0.3 or wpb

         0.2 (20%)

         1.5 or (1  - n) ps
         0.43 (loam)
         2.65
         0.13 or (n - 6W)
         Hx41
         chemical-specific
         (assume to be zero
         for inorganic con-
         taminants except
         mercury)
                                                           11

-------
Review Draft—Do Not Cite or Quote—December 1994
   Highlight 4: Simplifying Assumptions of the Default
               Conceptual Model for Ground Water

   1.  The source of contamination is defined as an evenly
      contaminated 30-acre site.  Source size has signifi-
      cant implications for the development of the dilution/
      attenuation factor. Large sources generally tend to
      result in low DAFs, while smaller sources generally
      justify higher DAFs. Where actual source size differs
      significantly from the default 30-acre assumption, the
      user should consider a site-specific evaluation to
      develop a more site-specific DAF.

   2.  The soil contamination extends from the surface to
      the top of the aquifer.  This is a reasonable assump-
      tion for sites where the water table is fairly shallow
      (e.g., 5 to 10 feet below surface). However, in areas
      where the water table is very deep, this assumption
      may not be valid and should be considered in the
      decision to apply a detailed site-specific evaluation.

   3.  No attenuation is considered in the unsaturated
      zone.  This assumption also has implications for the
      DAF. As discussed above, a detailed site-specific
      evaluation  should be considered at sites that have a
      very thick uncontaminated unsaturated zone because
      a higher DAF may be  justified.

   4.  The point of compliance is at the edge of the site,
      which is assumed to be  uniformly contaminated.
      This conservative assumption also has implications
      for the calculation of the DAF.  The user should
      consider whether this  assumption is valid for the site
      in question and whether further  evaluation would be
      appropriate.

   5.  The simple site-specific or generic DAF assumes;
      that an unconfined, unconsolidated aquifer with
      homogeneous and isotropic hydrologic properties
      underlies the site.  A DAF greater than 1 may not be
      appropriate for soils underlain by karst or fractured
      rock aquifers.

   6.  NAPLs are not present.  If NAPLs are present in
      soils, the SSLs do not apply (i.e., further investiga-
      tion is  necessary).
to organic carbon in soil (Dragun, 1988).  An adjustment to
relate sorbed concentration in soil to the analytically measured
total soil concentration has been added to the equation.

The partition  equation  contains  parameters  for chemical-
specific (Henry's law constant; Kd or K^ and  subsurface
characteristic variables (dry bulk  density, porosity, air-filled
and water-filled pore space).  In the default method, national
default values for the parameters in the partition equation were
used to calculate the generic SSLs in Appendix A. Nonzero
ground water  maximum  contaminant level goals (MCLGs)
 •ere  used as  the  acceptable ground water limits for each
  ntaminant in the partitioning equation.  If nonzero MCLGs
were not available, maximum contaminant levels (MCLs) were
                              used.  If MCLs were not available, concentrations associated
                              with a target cancer risk of 10"6 and/or a noncancer HQ of 1
                              were derived using Agency toxicity criteria.  The acceptable
                              ground water limit is multiplied by the DAF of 10 to obtain a
                              target soil leachate concentration for calculating generic SSLs.

                              In the simple site-specific method, site-measured data would
                              replace the default values for the subsurface characteristic and
                              soil variables (i.e., fraction organic carbon, dry bulk density,
                              average soil moisture content). These variables would then be
                              used to calculate a more site-specific screening value.   Even
                              this screening number is fairly conservative because of the
                              underlying assumptions regarding the absence of attenuation
                              and placement of the well adjacent to  the source.

                              As described above, the C^ ceiling limit defines (for organic
                              chemicals that are liquid at soil temperatures) a concentration
                              above which chemicals may occur as NAPLs in soil.  For
                              liquid chemicals present at concentrations greater than  C^,
                              NAPL presence may be suspected and the Soil Screening
                              framework would not be applicable (i.e., further investigation
                              is necessary).  See U.S. EPA (1992b) for guidance on deter-
                              mining the likelihood of NAPL occurrence in the subsurface
                              and on conducting the additional investigations  necessary if
                              NAPL, occurrence is suspected at a site.

                              Partitioning of inorganic constituents in the subsurface is more
                              complex than for organics.  A variety of soil conditions affect
                              the derivation of the partitioning coefficient for inorganics,
                              while organic carbon is the parameter that most affects organic
                              partitioning. For this reason, the EPA MINTEQ2 equilibrium
                              geochemical speciation model was used to calculate Kd values
                              for the metals, which were then used in Equation 8. Kd values
                              for metals are  most significantly affected by pH; therefore,
                              metal Kd values were calculated over a range of subsurface pH
                              conditions (4.9 to 8.0).  Kd values  corresponding to this pH
                              range  are  presented in the revised  Technical  Background
                              Document (U.S.  EPA,  1994e) for use in the  simple  site-
                              specific method.  Based on the pH at the site, the appropriate
                              Kd should be selected and used in the calculation. Also note
                              that all metals except mercury are essentially nonvolatile and
                              their Henry's law constant (HO in Equation 8  should be set at
                              zero.

                              Generic SSLs for inorganics corresponding to a pH of 6.8 are
                              presented in; Appendix A for the default method.  Table 1 lists
                              inorganic SSLs corresponding to pH values of 4.9 and 8.0 and
                              a DAI7 of  10.  If pH conditions at  a site are not known, the
                              generic SSL corresponding to a pH of 6.8 should be used in
                              the default method.  Table  1 also includes SSLs for ionizing
                              organics, whose partitioning behavior is also pH dependent.
                              Readers are referred to the Technical  Background Document
                              (U.S. EPA, 1994e) for a  more detailed discussion  of the
                              derivation  of Kd values for  inorganics and  K,,,. values for
                              ionizing and nonionizing organics.

                              The framework also includes the option of using a leach test
                              instead of the partitioning equation.  In some instances a leach
                         12

-------
                                 Review Draft—Do Not Cite or Quote—December 1994
        Table 1. pH-Speclfic SSLs for Metals
     and Ionizing Organlcs (mg/kg) (DAF = 10)
Chemical
Arsenic
Barium
Beryllium
Cadmium
Chromium (+6)
Mercury
Nickel
Selenium
Thallium
Zinc
Benzoic acid
2,4-Dichlorophenol
Pentachtorophenol
2,4,5-Trfehlorophenol
2,4,6-Trfchlorophenol
pH4.9
13
16
0.1
0.06
31
0.006
1
9
0.2
180
300
0.5
0.2
200
0.07
pH8
16
340
19,000
230
14
4
140
1
0.5
1.6E+6
280
0.3
0.01
26
0.01
test may be more useful than the partitioning method, depend-
ing on the constituents of concern and the possible presence of
RCRA wastes.   This  guidance suggests using  the EPA
Synthetic Precipitation Leaching Procedure (SPLP, EPA SW-
846 Method 1312, see the Technical Background  Dcoument
[U.S. EPA, 1994e]).  The SPLP was developed to model an
acid rain leaching environment and is generally appropriate for
a contaminated soil scenario.  Like most leach tests, the SPLP
may  not  be  appropriate  for all situations (e.g., soils
contaminated with oily constituents may  not  yield suitable
results).  Therefore, discretion is advised when applying the
SPLP.

The Agency is aware that there are many leach tests available
for application at hazardous waste sites, some of which may be
appropriate in specific situations (e.g., the Toxicity Charac-
teristic Leaching Procedure, known as the TCLP, models
leaching in a municipal landfill environment). It is beyond the
scope of this  document to discuss in detail other leaching
procedures and the appropriateness of their use. Stabilization/
Solidification  of CERCLA and RCRA Wastes (U.S. EPA,
1989c) and  the SAB's  review of leaching tests (U.S. EPA,
199 lc) contain information on the application of various leach
tesls  to  various  waste disposal  scenarios.   The user is
encouraged to consult these doucments for further information.

DETERMINING  THE DILUTION/
ATTENUATION FACTOR

As contaminants move through soil and ground water, they are
subjected to a number  of physical, chemical,  and biological
processes that generally reduce  the  eventual contaminant
concentration  level at  receptor points.   The reduction in
concentration can be expressed succinctly by the DAF, defined
as the ratio of the soil leachate concentration to the receptor
point concentration. The lowest possible value of DAF is 1,
corresponding  to the  situation where there is no dilution or
attenuation of a contaminant; i.e.,  the concentration at the
receptor point  is the same as that in the soil leachate. High
DAF values, on the other hand, correspond to a high degree of
dilution and attenuation of tlie contaminant from the leachate
to the receptor point.

The soil/water partition equation relates concentrations of
contaminants adsorbed to soil organic carbon to soil leachate
concentrations in the unsaturated zone.  Contaminant migration
through the unsaturated zone  to the water  table generally
reduces the soil leachate concentration by attenuation processes
such as adsorption and degradation.  Ground water transport in
the saturated  zone further reduces concentrations through
adsorption, degradation, and dilution. Generally, to account for
this reduction in concentration, acceptable ground water limits
are multiplied  by  a  DAF to obtain  a target soil leachate
concentration for the  partition equation.
               !           !                          I
A default DAF of 10 is applied to calculate the generic SSLs.
A weight of evidence method  was used  to determine this
default DAF.   In  the  weight-of-evidence  approach, OERR
evaluated  a number of  methods  for  calculating DAFs.
Included  in this  approach  was  an  evaluation  of  DAFs
calculated  by  the EPACMTP model,  using a  range of
assumptions including those ;associated with the conceptual site
model for the generic SSLs.  The comparison also included
DAFs calculated from a mores simplified mixing-zone equation,
as well as  acceptable DAFs used in existing State programs.
The comparison indicated that, for the default scenario, a DAF
of 10 is conservatively protective of the majority of site
conditions,  including the  site  scenario developed  for the
generic SSLs.  The  Technical Background Document (U.S.
EPA, 1994e) supporting this guidance contains additional detail
on the development of the generic DAF.

The simple site-specific  method  relies  on  a fairly simple
mixing zone equation (Equation 9)  to calculate a site-specific
dilution factor to be used instead of the default DAF. In this
method,  site-measured values for hydraulic gradient, hydraulic
 Equation 9:   Derivation of Dilution Factor
                  dilution factor = 1 + _1_
  Parameter/Definition (units)

  dilution factor (unitless)
  K/aquifer hydraulic conductivity (m/yr)
  i/hydraulic gradient (m/m)
  d/mixing zone depth (m)
  I/infiltration rate (m/yr)
  L/source length parallel to ground water flow (m)
                                                          13

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
conductivity, and estimates of infiltration, contaminant source
length, and mixing-zone depth are used to calculate the dilution
factor. The mixing-zone depth is estimated from an equation
relating it to aquifer thickness, infiltration rate, ground water
  ilocity, and source length parallel to flow (Equation 10).
 Equation 10: Estimation of Mixing Zone Depth
        d = (0.0112 L2)0'5 + da {1 - exp[(-LI)/(Kid,)]}
 Parameter/Definition (units)

 d/mixing zone depth (m)
 L/source length parallel to ground water flow (m)
 1/infiftration rate (m/yr)
 K/aquifer hydraulic conductivity (m/yr)
 da/aquifer thickness (m)
Detailed Site-Specific Method

In this investigation, site-specific data are collected and used
in a fate and transport model to determine whether a threat to
ground water exists and, if so, to further determine site-specific
cleanup goals  as would typically be done for the remedial
investigation/feasibility   study  (RI/FS).    Consequently  it
represents the highest level of site-specificity in evaluating the
migration to ground water pathway. A DAF is not used in this
   :1iod because the model would account for fate and transport
    nanisms in the subsurface.  The advantage of this approach
    mt it accounts  for site hydrogeologic,  climatologlc, and
contaminant source characteristics and may result in fully
protective but less stringent remediation goals. However, the
additional cost of collecting the data required to apply the
model should be factored into  the  decision to conduct a
detailed site-specific investigation.

Choosing a model for site-specific application is integral to an
accurate evaluation of potential concern. However, the data
used in the application and interpretion of  the results are
equally important.  In an effort to provide useful information
for a model application, EPA's  ORD Laboratories in Ada,
Oklahoma, and Athens, Georgia, conducted an evaluation  of
nine unsaturated zone fate and transport  models.  The infor-
mation  in this report  is summarized in  the  Technical
Background Document (U.S. EPA,  1994e)  supporting this
guidance. These nine models are only a  subset of the poten-
tially appropriate models available to the public and are not
meant to be construed as having received EPA approval. EPA
also has developed guidance for the selection and application
of ground water transport and fate models and for interpreta-
tion of model  applications. The user is referred to Ground
Water Modeling Compendium (U.S. EPA, 1994b) and Frame-
work for Assessing Ground Water Model Applications (U.S.
   'A, 1994a) for further information.
MEASURING CONTAMINANT
CONCENTRATIONS IN SOIL

In order to compare site soil concentrations with the SSLs, it
is important to develop a sampling strategy that will result in
an accurate representation of site contamination.  This Soil
Screening Guidance recommends  that site managers use the
Data Quality Objectives (DQO) process (Figure 5) to develop
a  sampling strategy that will satisfy  Superfund  program
objectives.  The site manager can use the DQO process to
conveniently organize  and  document many  site-specific
features and assumptions underlying the sampling plan.  In the
last step  of the DQO  process, "Optimize  the Design for
Obtaining Data," the site manager can choose between two
alternative approaches to measuring surface soil contaminant
concentrations.  The first is  a site-specific strategy that uses
site-specific estimates of contaminant variability to determine
how many  .samples are needed  to support the screening
decision. The second is a fairly prescriptive approach that can
be used in lieu of the site-specific strategy. Recommendations
for subsurface sampling that can be modified to accommodate
site-specific conditions are also included in the guidance.

Exposure to site contaminants over a long (chronic) period of
time is best represented by an arithmetic average concentration
for an exposure area (U.S. EPA, 1992d). Therefore, measure-
ment of site concentrations for comparison to the SSLs should
                     State the Problem
                    Identify the Decision
                Identify Inputs to the Decision
                 Define the Study Boundaries
                  Develop a Decision Rule
               Specify Limits on Decision Errors
            Optimize the Design for Obtaining Data
                                                                  Figure 5.  The Data Quality Objectives process.
                                                          14

-------
                                   Review Draft—Do Not Cite or Quote—December 1994
be based on the arithmetic mean concentration as well. For the
purposes of this guidance, the Agency has assumed that the
size of a typical residential lot (0.5 acre)  is an appropriate
averaging area for residential land use. For large sites that
could be divided into multiple residential lots, the site should
be sectioned into appropriate 0.5-acre parcels.

For measurement of surface soil samples for the inhalation and
ingcstkm pathways, samples should be collected over a depth
of 6 inches because it is the top 6 inches of soil that is most
likely to be ingested or inhaled as fugitive dusts. Additional
sanipling beyond 6 inches may be appropriate, depending on
the contaminant's mobility. If soils at the site are of concern
for the migration  to ground  water pathway  as  well as the
ingesdon and/or inhalation pathways, then surface soils should
be  sampled  first since the results of the composite samples
may indicate source areas to target for subsurface sampling.

As discussed previously, the initial steps for implementing the
Soil Screening framework are to  (1) develop the conceptual
site model and determine the applicability of the framework;
(2)  determine if background  concentrations  exceed the
(generic) SSLs; and (3) select the method (simple site-specific,
detailed site-specific, or generic) to determine the SSLs.  Once
these steps have been completed,  it will then be necessary  to
choose either a site-specific or a generic, prescriptive sampling
strategy for surface soils.

Surface Soils—Site-Specific Strategy

The site-specific sampling strategy utilizes a sampling design
approach that allows statistically valid conclusions to be drawn
about contaminant concentrations  at a site based on relatively
limited sampling.  EPA recommends that site managers use
this strategy to determine the number  of samples needed  to
compare average  contaminant concentrations within  each
exposure area against the SSLs.   The site-specific  strategy
provides procedures for ensuring that screening decisions can
be made  with  acceptable  levels  of confidence  despite
variability  in  soil  contaminant concentrations  that can
sometimes mask true conditions  at the site.   This approach
provides flexibility to incorporate site-specific information
about  likely contamination patterns so that sampling can  be
concentrated in areas where uncertainty about the risk posed  by
soil contaminants is greatest.

The sampling design developed for the site should be based  on
the conceptual site model and should reflect conditions at the
site. It is flexible in that the information used to develop the
conceptual site model (historical records, aerial photographs,
existing sampling  data, etc.) can  also be used to develop  an
appropriate sampling  strategy.  Such a strategy  may include
stratification of the site, if appropriate, into areas where soil
contaminant concentrations are expected to clearly exceed the
SSLs, areas where soil contaminant concentrations are expected
to fall well below the SSLs, and areas of the  site where there
is  greater  uncertainty  as  to   whether  soil  contaminant
concentrations exceed the SSLs.
This classification of areas of the site can help in designing an
efficient sampling plan, since the number of samples required
to support good decision making depends on the contaminant
variability  likely  to  be  encountered and  how  greatly
contaminant concentrations differ from the SSLs. By grouping
similar areas together, each area can be sampled in accordance
with the level of uncertainty or variability associated with that
area. For example, EPA expects that a relatively small number
of samples will be needed to make the screening decision
where average contaminant concentrations clearly exceed or
are well below the SSLs. More intensive sampling is expected
for those areas where relatively high contaminant variability or
concentrations close to the SSLs make it more difficult to
determine with confidence whether the average contaminant
concentration exceeds the screening level.

Inherent in the statistically based site-specific sampling strategy
is the  specification of limits  on decision  errors, which is
performed in the sixth  step of the DQO process.  Limits on
decision errors are quantitative performance requirements for
the quality and quantity of data that will support the screening
decision.  These performance requirements are specified in
terms of the probability of making a decision error, which can
occur in two ways:

•  Type  I:    The data mislead the  site manager into
   deciding that the exposure area concentration is below
   the  SSLs  when  the  true   average   contaminant
   concentration exceeds the screening level; or

•  Type  IT:   The  data  mislead  the site  manager into
   deciding that the exposure area  concentration is above
   the SSL and further investigation is required when in
   fact the true average contaminant concentration is less
   than the SSL.

To ensure consistency  in applying the framework, EPA has
specified  tolerable limits on decision errors at the program
level.   The Technical Background  Document (U.S. EPA,
1994e)  provides a full  discussion  of the Soil  Screening
framework's limits on decision errors and of the site-specific
strategy in general. EPA encourages the project manager to
seek the assistance of a  statistician or the Regional quality
assurance staff for the development of the sampling strategy.
For  more detailed  guidance on the DQO  process the  user
should refer to the Technical Background Document and Data
Quality Objectives for Superfund (Interim Final) (U.S. EPA,
1993a).

Surface Soils—Prescriptive Approach

The guidance provides a second sampling methodology—a
"prescriptive approach"—that can be used as an alternative to
the  site-specific  approach.    A sampling  design effort is
required for the site-specific strategy, whereas the prescriptive
approach  provides a simple, standard sampling approach that,
will be most  useful for  small sites that do not  warrant an'
extensive design effort. It emphasizes composite sampling for
                                                            15

-------
                                  Review Draft—Do Not Cite or Quote—December 1994
 nonvolatile contaminants and specifies the number of samples
 to be collected for analysis of volatile contaminants.  It differs
 from the site-specific approach in that  the same sampling
 strategy  must be applied to each 0.5-acre exposure  area.
          it does  not  explicitly  control decision errors,
            simulations suggest that it does not underestimate
 mean concentrations for commonly occurring patterns of soil
 contamination.     Additional  simulations  comparing  the
 performance of the prescriptive approach to the site-specific
 strategy will be a subject of peer review.

 Studies by ORD indicate that at least 20 samples per exposure
 area are needed to closely estimate the true mean.  To balance
 the need for statistical confidence in determining a meaningful
 arithmetic mean contaminant concentration with the costs of
 analyzing multiple samples for each exposure  area,  EPA
 recognizes the benefits of composite  samples and advocates
 compositing, where appropriate.   Compositing  may mask
 contaminant levels that are slightly higher than the SSL, but
 areas of high contamination will still be detected. Compositing
 is a reasonable approach and an efficient use of resources since
 the Superfund program is interested in the average exposure
 over time.  (See  the Technical Background Document [U.S.
 EPA, 1994e] for a more detailed discussion of compositing and
 its limitations.)

 Using the prescriptive approach, 20 discrete samples can be
 reduced to four composite samples.  (The exposure area can be
 divided  into quadrants and  five random samples  can  be
 Collected and composited within each quadrant.) The contam-
  tiant concentrations from the four composite samples should
"be compared directly with their respective SSLs. If any one of
 the composites equals or exceeds the SSL, then that portion of
 the exposure area should be studied further.

 Compositing is not appropriate for volatile organic compounds
 (VOCs)  since much of the contaminant will be lost during
 homogenization of the soil (U.S. EPA, 1989a, 1992c).  For
 VOCs, 10 discrete samples can be taken per exposure area and
 any  sample above  the  SSL would  trigger  the need  for
 additional study in that exposure area. Additionally, it is not
 appropriate to average the contaminant levels in each exposure
 area and evaluate the mean concentration against the SSLs
 because  10 discrete samples may underestimate the tree mean.

 Subsurface Sampling

 For the migration to ground water pathway, subsurface soils
 that have constituents that might contribute to ground water
 contamination are of primary concern.  Therefore, it is  the
 source areas that are of interest and not necessarily a 0.5-acre
 exposure area as specified for the ingestion  and inhalation
 pathways. To determine whether contaminants in the subsur-
 face soils (defined as below 6 inches for the purposes of
 implementing SSLs) potentially pose a risk to ground water,
     guidance suggests sampling at least two boreholes using
   lit spoon or Shelby  tube  samples in each  source area.
 Samples should begin at 6 inches below ground surface and
continue at  2-foot  intervals until  no  contamination is
encountered.  If the  average  concentration in any borehole
exceeds the SSL, then further site-specific study is warranted.

Subsurface sampling depths and intervals can be adjusted at a
site to accommodate site-specific information on subsurface
contaminant  distributions  and  geological conditions.   In
addition, soil investigation for the migration to ground water
pathway should not be conducted independent of ground water
investigation. Ground water should be sampled to determine
whether there is concern for existing ground water contam-
ination, and the results should be considered in the holistic
application of the Soil Screening framework.

Geostatistics

If the SSLs are to be compared with the data resulting from
the  initial  sample  collection  efforts  of  the remedial
investigation,  the site manager may want to consider using
geostatistics to estimate contaminant concentrations across the
site. Geostatistics is probably  most appropriate to use in the
detailed site-specific approach. Geostatistics is a field of study
in which statistical analyses of geologic or environmental data
are conducted. It differs from single-sample classical statistics
in that it assumes that variability and independence between
samples is not random, but that there is some spatial continuity
between samples.  Geostatistics can  be used to  estimate
contaminant concentrations at unsampled points and estimate
average contaminant concentrations across the site.

Software  packages  have been  developed  to  facilitate
geostatistical analyses. One package is GEO-EAS, developed
by EPA's Environmental Monitoring Systems Laboratory in
Las Vegas, Nevada.  Assistance and consultation with skilled
geostatisticians is recommended prior  to   initiating  any
sampling plan to ensure that the sampling strategy will capture
the critical data necessary for the geostatistical analyses.

WHERE TO GO FOR  FURTHER
INFORMATION

More detailed discussions of the  technical background and
assumptions supporting the development of the Soil Screening
framework  are  presented in  the  Technical  Background
Document, for Soil Screening Guidance (U.S. EPA, 1994e).
For additional copies of this Fact Sheet and/or the Technical
Background Document, call the National Technical Information
Service (NTIS) at (703)  487-4650.
                                                           16

-------
                                Review Draft—Do Not Cite or Quote—December 1994
  NOTICE: This guidance is based on policies in the Final Rule of the National Oil and Hazardous Substances Pollution
  Contingency Plan (NCP), which was published on March 8, 1990 (55 Federal Register 8666).  The NCP should be considered
  the authoritative source.

  The policies set out in this document are intended solely as guidance to the U.S. Environmental Protection Agency (EPA)
  personnel; they are not final EPA actions and do not constitute rulemaking. These policies are not intended, nor can they be
  rel|ed upon, to create any rights enforceable by any party in litigation with the United States.  EPA officials may decide to follow
  the guidance provided in this document, or to act at variance with the guidance, based on an analysis of specific site
  circumstances.  EPA also reserves the right to change the guidance at any time without public notice.
REFERENCES

Calabrese, EJ., H. Pastides, R. Barnes, et al.  1989.  How
   Much Soil Do Young Children Ingest: An Epidemiologic
   Study. In: Petroleum Contaminated Soils, Vol. 2.  EJ.
   Calabrese and P.T. Kostecki, eds. pp. 363-417.  Chelsea,
   MI, Lewis Publishers.

Cowherd, C., G. Muleski, P. Engelhart, and D. Gillette. 1985.
   Rapid Assessment of Exposure to Paniculate Emissions
   from Surf ace Contamination. EPA/600/8-85/002.  Prepared
   for Office  of Health and Environmental Assessment, U.S.
   Environmental Protection Agency, Washington, DC. NHS
   PB85-192219 7AS.

Davis, S., P. Waller, R. Buscnom, J. Ballou, and P. White.
   1990.  Quantitative estimates of soil ingestion in normal
   children between the ages of 2 and 7 years: population-
   based estimates using Al, Si, and Ti as soil tracer elements.
   Archives of Environmental Health, 45:112-122.

Dragun, J. 1988. The Soil Chemistry of Hazardous Materials.
   Hazardous Materials Control Research Institute, Silver
   Spring, MD.

U.S. EPA.   1988.   Guidance on Remedial Actions for
   Contaminated Ground Water at Superfund Sites.  Directive
   9283.1-2.  EPA/540/G-88/003. Office of Emergency and
   Remedial  Response, Washington, DC.    NTIS PB89-
   184618/CCE.

U.S. EPA.  1989a. Methods for Evaluating the Attainment of
   Soil  Cleanup Standards.   Volume 1:  Soils and Solid
   Media.  EPA 230/02-89-042.  Statistical Policy Branch,
   Office of Policy, Planning, and Evaluation, Washington,
   DC.

U.S. EPA.  1989b. Risk Assessment Guidance for Superfund:
   Volume 1: Human Health Evaluation Manual, Part  A,
   Interim Final.  EPA/540/1-89/002.  Office of Emergency
   and Remedial Response, Washington, DC.  NTIS PB90-
   155581/CCE.

U.S. EPA. 1989c. Stabilization/Solidification ofCERCLA and
   RCRA Wastes. EPA/625/6-89/022.
U.S. EPA.   1990a.  Guidance  on Remedial Actions for
   Superfund Sites with PCB Contamination.  EPA 540G-
   90/007.   Office of Emergency and Remedial Response,
   Washington, DC. NTIS PB91-921206/CCE.

U.S. EPA.   1990b.  Suggested ROD Language for Various
   Ground Water Remediation Options. Directive 9283.1-03.
   Office   of  Emergency   and  Remedial   Response,
   Washington, DC. NTIS PB91-921325/CCE.

U.S. EPA.   199 la.  Guidance for Risk Assessment.   Risk
   Assessment  Council,  Office  of  the  Administrator,
   Washington, DC.

U.S. EPA.   199 Ib.  Hwnan Health Evaluation  Manual,
   Supplemental  Guidance:   Standard Default Exposure
   Factors.  Publication 9285.6-03.  Office of Emergency and
   Remedial Response, Washington, DC. NTIS PB91-921314.

U.S. EPA.  199 Ic.  Leachability Phenomena.  Recommenda-
   tions and Rationale for Analysis of Contaminant Release by
   the Environmental Engineering Committee.   EPA-SAB-
   EEC-92-003.  Science Advisory Board, Washington, DC.

U.S. EPA.  1991d.  Risk Assessment Guidance for Superfund,
   Volume  1:  Human Health Evaluation Manual (Part B,
   Development  of Risk-Based Preliminary  Remediation
   Goals). Publication 9285.7-01B. Office of Emergency and
.   Remedial Response, Washington, DC. NTIS PB92-963333.

U.S. EPA.  1991e. Role of the Baseline Risk Assessment in
   Superfund  Remedy  Selection Decisions.    Publication
   9355.0-30.  Office of Emergency and Remedial Response,
   Washington, DC. NTIS PB91-921359/CCE.

U.S.  EPA.    1992a.    Considerations in  Ground-Water
   Remediation at Superfund  Sites and RCRA Facilities—
    Update.  Directive 9283.1-06.  Office of Emergency and
   Remedial Response,  Washington, DC.    NTIS  PB91-
   238584/CCE.

U.S. EPA.   1992b.  Estimating Potential for Occurrence of
   DNAPL  at Superfund Sites.   Publication  9355.4-07FS.
   Office of Emergency and Remedial Response, Washington,|
   DC. NTIS PB92-963338.
                                                        17

-------
                                Review Draft—Do Mot Cite or Quote—December 1994
U.S. EPA.   1992c.  Guidance for Data Usability  in Risk
   Assessment (Part A).  Office of Emergency and Remedial
 1  Response, Washington, DC. NTIS PB92-963356.

  _ EPA.   1992d.   Supplemental Guidance to  RAGS:
  ^Calculating the Concentration Term, Volume 1, Number 1.
   Publication  9285.7-011.   Office of  Emergency  and
   Remedial Response, Washington, DC. NTISPB92-963373.

 U.S. EPA.  1993a.  Data Quality Objectives for Superfund:
   Interim Final Guidance.  EPA 540-R-93-071.  Publication
   9255.9-01. Office of Emergency and Remedial Response,
   Washington, DC. NTIS  PB94-963203.

 U.S. EPA.   1993b.   Guidance for Evaluating Technical
   Impracticability of Ground-Water Restoration. Directive
   9234.2-25. EPA/540-R-93-080. Office of Emergency and
   Remedial Response, Washington, DC.

 U.S. EPA.  1993c.   Health  Effects Assessment Summary
   Tables (HEAST): Annual Update, FY1993. Environmen-
    tal  Criteria and Assessment Office, Office of Health and
   Environmental Assessment,  Office of Research and
    Development, Cincinnati, OH.

 U.S. EPA.  1993d.  Science Advisory Board Review of the
    Office of Solid Waste and Emergency Response draft Risk
    Assessment  Guidance for Superfund  (RAGS),  Human
    Health Evaluation Manual (HHEM).  EPA-SAB-EHC-93-
    007.  Science Advisory  Board, Washington, DC.
U.S. EPA, 1994a. Framework for Assessing Ground Water
   Modeling Applications.  EPA-500-B-94-004.  Resource
   Management and Information Staff. Office of Solid Waste
   and Emergency Response, Washington, DC.

U.S. EPA.  1994b. Ground Water Modeling Compendium,
   Second  Edition.     EPA-500-B-94-003.    Resource
   Management and Information Staff. Office of Solid Waste
   and Emergency Response, Washington, DC.

U.S. EPA. 1994c. Integrated Risk Information System (IRIS).
   Dulutli, MN.

U.S. EPA. 1994d. Role of the Ecological Risk Assessment in
   the Baseline Risk Assessment.   OSWER Directive No.
   9285.7-17.   Office of  Solid Waste and  Emergency
   Response, Washington, DC. August 12.

U.S. EPA.  1994e. Technical Background Document for Soil
   Screening Guidance.   EPA/540/R-94/102.   Office of
   Emergency and Remedial Response, Washington, DC.
   PB95-963530.

Van Wijnen, JJEL, P. Clausing, and B. Bmnekreef.  1990.
    Estimated Soil Ihgestion  by Children.  Environmental
    Research, 51:147-162.
                                                         18

-------

-------
        Review Draft—Do Not Cite or Quote—December 1994
Appendix A. Generic Soil Screening Levels for Superfund*
NOTICE: These
specific exposure
CAS No.
83-32-9
67-64-1
309-00-2
120-12-7
71-43-2
56-55-3
205-99-2
207-08-9
50-32-8
111-44-4
117-81-7
75-27-4
75-25-2
71-36-3
85-68-7
86-74-8
75-15-0
56-23-5
57-74-9
108-90-7
124-48-1
67-66-3
218-01-9
72-54-8
72-55-9
50-29-3
53-70-3
84-74-2
95-50-1
106-46-7
91-94-1
75-34-3
107-06-2
75-35-4
156-59-2
156-60-5
78-87-5
542-75^6
60-57-1
84-66-2
131-11-3
121-14-2
606-20-2
values were developed for use in application of the Soil Screening Guidance
pathways constituting a residential scenarb and should only be used in that
, 	 \
JB|0 K
Chemical
Acenaphthene
Acetone
Aldrin
Anthracene
Benzene
Benzo(a)anthracene
Benzo(6)fluoranthene
Benzo(/c)fluoranthene
Benzo(a)pyrene
Bis(2-chlorethyl)ether
Bis(2-ethy!hexy!)phthalate
Bromodichloromethane
Bromoform
Butanol
Butyl benzyl phthalate
Carbazole
Carbon disulfide
Carbon tetrachloride
Chlordane
Chtarobenzene
Chlorodibromomethane
Chtaroform
Chrysene
ODD
DDE
DDT
Dibenzo(a,/i)anthracene
Di-/7-butyl phthalate
1 ,2-Dichlorobenzene (o)
1 ,4-Dichbrobenzene (p)
3,3-Dichlorobenzidine
1,1-Dichbroethane
1 ,2-Dichtaroethane
1 , 1 -Dfchbroethylene
cfe-1 ,2-Dichloroethylene
frans-1 ,2-Dichloroethylene
1 ,2-Dichbropropane
1 ,3-Dichloropropene
Dieldrin
Diethyl phthalate
Dimethyl phthalate
2,4-Dinrtrotoluene
2,6-Dinitrotoluene
Pathway-specific values for
surface soils
(mg/kg)
Ingestion Inhalation
4,700 b —c
7,800 b 62,000 d
0.04 e 0.5 e
23,000 b — c
22 e 0.5°
0.9 8 — c
0.9 e - c
g e 	 e
0.09 e-f -c
0.6 e 0.3 e'f
46° 210 d
5e 1,800 d
81 a 46 e
7,800 b : 9,700 d
1 6,000 b 530 d
32 e ~- c
7,800 b 11 b
5el 0.2 e
0.5 e 10 e
1,600b 94 b
8e 1,900 d
110 e 0.2°
88 e — c
38 	 C
2 e 	 <=
2e 80 e
0.09 e-f . •— c
7,800 b 100d
7.IDOO b 300 d
27 e 7,700 b
-j a 	 c
7,800 b 980 b
7 e 0.3 e
1 e 0.04 e
780 b 1,500d
1,600b 3,600 d
9e 11 b
4e 0.1 e
0.04 e 2 e
63,000 b 520 d
7.8E+5b 1,600d
160 b — c
78 b' — °
only. They were devebped for
context.
Migration to ground water
pathway levels (mg/kg)
With 10
DAF
200 b
8b
0.005°
4,300 b
0.02
0.7
4
4
4
3E-4 e-f
11
0.3
0.5
8b
68
0.2 e-f
14 b
0.03
2
0.6
0.2
0.3
1
0.7*
0.5 e
18
11
120 b
6
1
0.01 "•'
11 b
0.01 f
0.03
0.2
0.3
0.02
0.001 e-f
0.001 e'f
110 b
1,200 b
0.2 b'f
0.1 b-f
Wtthl
DAF
20 b
0.8 b
5E-4e'f
430 b
0.002 f
0.07 f
0.4
0.4
0.4
3E-5 "•'
1
0.03
0.05
0.8 b
7
0.02 "•'
1b
0.003 f
0.2
0.06
0.02
0.03
0.1 f
0.07 e
0.05 e
0.1 e
1
12 b
0.6
0.1'
0.001 °'f
1b
0.001 f
0.003 '
0.02
0.03
0.002 f
1E-4e'f
1E-4e'f
11 b
120 b
0.02 b>f
0.01 b'f
                            19
(continued)

-------

-------
Review Draft—Do Not Cite or Quote—December 1994
          Appendix A (continued)
»
CAS No.
117-84-0
115-29-7
72-20-8
100-41-4
206-44-0
86-73-7
76-44-8
1024-57-3
118-74-1
87-68-3
319-84-6
319-85-7
58-89-9
77-47-4
67-72-1
193-39-5
78-59-1
72-43-5
74-83-9
|75-09-2
*91-20-3
98-95-3
1336-36-3
129-00-0
100-42-5
79-34-5
127-18-4
108-88-3
8001-35-2
120-82-1
71-55-6
79-00-5
79-01-6
108-05-4
75-01-4
1330-20-7

65-85-0
106-47-8
95-57-8
120-83-2
^105-67-9
F51-28-5
95-48-7
/ 	 \
010
Chemical
Di-n-octyl phthalate
Endosulfan
Endrin
Ethylbenzene
Fluoranthene
Fluorene
Heptachlor
Heptachlor epoxide
Hexachlorobenzene
Hexachloro-1 ,3-butadiene
a-HCH (a-BHC)
p-HCH (p-BHC)
Y-HCH (Lindane)
Hexachlorocyclopentadiene
Hexachloroethane
lndeno(1 ,2,3-c,d)pyrene
Isophorone
Methoxychlor
Methyl bromide
Methylene chloride
Naphthalene
Nitrobenzene
Polychlorinated biphenyls (PCBs)
Pyrene
Stryene
1 ,1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
Toxaphene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 ,1 ,2-Trichloroethane
Trichloroethylene
Vinyl acetate
Vinyl chloride
Xylenes (total)
lonizable Organics
Benzoic acid
p-Chloroaniline
2-Chlorophenol
2,4-Dichiorophenol
2,4-Dimethylphenol
2,4-Dinitrophenol
2-Methylphenol
Pathway-specific values for
surface soils
(mg/kg)
Ingestion Inhalation
1,600b
470 b
23 b
7,800 b
3,100 b
3,100 b
0.1 •
0.07 e
0.4 •
8e
0.1 •
0.4 e
0.5 e
550 b
46 e
0.9 e
670 e
390 fcl
110b
85 e
3,100 b
39 b
1h
2,300 b
1 6,000 bl
3e
12 e
1 6,000 b
0.6 *
780 b
	 c
11 e
58*
78,000 b
0.3 e
1.6E+5b

3.1E+5b
310 b
390 b
240 b
1,600b
160 b
3,900 b
	 c
	 c
	 c
260 d
... c
	 c
0.3 e
1 e
1 e
1 e
0.9 e
16e
... c
2b
49 e
	 c
3,400 d
	 c
2b
7e
	 c
110b
	 c,h
	 c
1,400d
0.4 e
1 1 e
520 d
5d
240 b
980 d
0.8 e
36
370 b
0.002 e'f
320 d

	 c
	 c
53,000 d
-- . c
	 c
._ c
	 c
Migration to ground water
pathway levels (mg/kg)
With 10
DAF
	 g
4b
0.4
5
980 b
160 b
0.06
0.03
0.8
0.1 f
4E-4 e'f
0.002 e
0.006
10
0.2 e'f
35
0.2 e-f
62
0.1 b
0.01 f
30 b
0.09 b'f
	 h
1,400b
2
0.001 e'f
0.04
5
0.04 f
2
Q.9
0.01 f
0.02
84 b
0.01 f
74

280 b'!
0.3 b-f-i
2W
0.5 "•'
3 bli
0.1 b'f'i
6b.i
Withl
DAF
	 g
0.4 b
0.04
0.5
98 b
16b
0.006
0.003
0.08 f
0.01 f
4E-5e-f
2E-4 e'f
6E-4f
1
0.02 e'f
3
0.02 e>f
6
0.01 blf
0.001 '
3b
0.009 b'f
	 h
140 b
0.2
1E-4e'f
0.004 f
0.5
0.004 f
0.2 f
0.09
0.001 f
0.002 f
8b
0.001 f
7

28 w
0.03 Wi
0.2 b>fli
0.05 b-f'j
Oo b.f.i
.0
0.01 blf-i
0.6 bli
                    20
(continued)

-------
                                 Review Draft—Do Not Cite or Quote—December 1994

                                            Appendix A (continued)
                                                       Pathway-specific values for
                                                              surface soils
                                                       Migration to ground water
CAS No.
86-30-6
621-64-7
87-86-5
108-95-2
95-95-4
88-06-2

7440-36-0
7440-38-2
7440-39-3
7440-41-7
7440-43-9
7440-47-3
7439-92-1
74J39-97-6
7440-02-0
7782-49-2
74^0-22-4
7440-28-0
7440-62-2
7440-66-6
57-12-5
BHB
Chemical
W-N'rtrosodiphenylamine
W-Nitrosodi-n-propylamine
Pentachlorophenol
Phenol
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
Inorganics
Antimony
Arsenic &
Barium
Beryllium
Cadmium &
1
Chromium (6+)
Lead
Mercury &
Nickel 4?
1
Selenium &
Silver
Thallium
Vanadium
Zinc ^
Cyanide
(mg/kg)

Ingestion Inhalation
130 •
0.09 e-f
3e,j
47,000 b
7,800 b
58 e

31 b
0.4 e
5,500 b
0.1 e
39 b
390 b
400 '
23 b
1,600b
390 b
390 b
... c
550 b
23,000 b
1,600b

__ c
	 c
	 c
	 c
210 e

__°
380 e
3.5E+5 b
690 e
920 e
140 e
—
7b.i
6,900 e
	 c
	 c
	 c
... c
	 c
	 c
pathway leve
With 10
DAF
0.2 e'f>l
2E-5 ••«
0.01 w
49 w
120 w
0.06 «•*'

	 k
15s
32'
180 '
6!
19 '
—
3j
21 '
3'
	 k
0.4 '
	 k
42,000 b,i
	 k
is (mg/Kg)
Withl
DAF
0.02 e'f>l
2E-6 e-fli
0.001 fli
5b,i
12 w
0.006 e'f'i

	 k
1 '
3j
181
0.61
2!
—
0.3 '
21
0.31
-k
0.04 !
	 k
4,200 b,i
	 k
   DAF m Dilution and attenuation factor.
   Screening levels based on human health criteria only.
   Calculated values correspond to a noncancer hazard quotient of 1.
   No toxte'rty criteria available for that route of exposure.
   Soil saturation concentration  (C^).
   Calculated values correspond to a cancer risk level of 1 in 1,000,000.
   Level is at or below Contract Laboratory Program required quantitation limit for Regular Analytical Services (HAS).
   Chemical-specific properties  are such that this pathway is not of concern at any soil contaminant concentration.
   A preliminary remediation goal of 1 ppm has been set for PCBs based on Guidance on Remedial Actions for Superfund Sites with
   PCB Contamination, EPA/540G-90/007, Office of  Emergency and Remedial Response, U.S. Environmental Protection Agency,
   Washington, DC, 1990, and on Agency-wide efforts to manage PCB contamination.
1   SSL for pH of 6.8.
1   Ingestion SSL adjusted by a factor of 0.5 to account for dermal exposure.
k  Soil/Water partition coefficients not available at this time.                                            _         ^CD^, „ 0-»
1   A preliminary remediation goal of 400 mg/kg has been set for lead based on Revised Interim Soil Lead Guidance for CERCLA Sites
   and RCRA  Corrective Action Facilities, OSWER Directive #9355.4-12, Office of Solid Waste and Emergency Response,  U.S.
   Environmental Protection Agency, Washington, DC, July 14, 1994.

 &  Indicates potential for soil-plant-human exposure.

 Levels developed for residential use only:
             Residential
Industrial
Agricultural


   21

-------