United States Science Advisory Board EPA-SAB-EEC-95-010
Environmental AUGUST 1995
Protection Agency Washington, DC
AN SAB REPORT: REVIEW
SEPA
EpA,s COMPOSITE
MODEL FOR LEACHATE
MIGRATION WITH
TRANSFORMATION
PRODUCTS - EPACMTP
PREPARED BY THE OSWER
EXPOSURE MODEL
SUBCOMMITTEE OF THE
ENVIRONMENTAL ENGINEERING
COMMITTEE
-------
August 9, 1995
EPA-SAB-EEC-95-010
Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
RE: SAB Review of EPA's Composite Model for Leachate Migration with
Transformation Products - EPACMTP
Dear Ms. Browner:
The Science Advisory Board (SAB) has completed its review of the Office of Solid Waste
and Emergency Response (OSWER) Composite Model for Leachate Migration with
Transformation Products (hereafter the model is referred to as EPACMTP). This review was part
of a continuing effort by the Environmental Engineering Committee to promote improvements in
the development and external peer review of environmental regulatory models. The OSWER
Exposure Model Subcommittee (OEMS) met on March 8, 1995 in Washington, DC to conduct
this review.
EPACMTP is designed to predict human exposure to groundwater pollutants in a
domestic drinking water receptor well that is impacted by releases from land disposal sites. This
model is applied nationally in support of development of regulations for management and control
of hazardous wastes, and it is not intended for site-specific applications. The Subcommittee
commends the Agency for making enhancements to earlier models, responding, in part, to SAB
suggestions and recommendations. The EPACMTP modeling approach incorporates greater
flexibility and versatility in the simulation capability than its predecessor (EPA's Composite Model
for Landfills). EPACMTP model explicitly considers: a) chain transformation reactions and
transport of daughter products, b) effects of water-table mounding on groundwater flow and
-------
contaminant migration, c) finite source, as well as continuous source, scenarios, and d) metals
transport by linking EPACMTP with the MINTEQ metals speciation model outputs. The
Subcommittee also identified a few important deficiencies in EPACMTP that should be addressed
by the Agency in completing the model development and verification before it is widely used.
The charge for this review consisted of the following four questions:
a) Is the mathematical formulation in the EPACMTP of the subsurface fate and
transport of daughter products from degrading organic chemical constituents
appropriate for EPA to use in establishing nation-wide exit levels for hazardous
waste in future regulations?
b) Is the regional approach, using hydrogeologic data from specific sites within
regions, better or should OSW continue to use inputs based on national
distributions?
c) Is the finite-source approach adequate for regulatory purposes?
d) Should the metal speciation model (MINTEQ) be linked to the EPACMTP model
to assess the subsurface fate and transport of metals as part of EPA's national
rulemaking efforts?
The incorporation of the daughter products into the model makes it more complete and
the mathematical formulation appears to be correct. However, EPA should verify that the
software works properly by further testing and documenting how these daughter products are
used in the simulation analysis. In addition, the Agency should document known instances where
very toxic daughter products are formed from multiple parent chemicals (i.e., exposure and risks
may be underestimated) as well as instances where biodegradation or inhibiting factors may affect
the transformation rate (i.e., exposure may be overestimated by the model).
The regional approach, using a stratified sample, which allows the incorporation of
inherent correlations and trends, is scientifically superior to the previous nation-wide approach,
and it responds directly to an earlier SAB recommendation (SAB, 1990). The Subcommittee
recommends that EPA compare the differences in results obtained by using the two approaches,
document the sensitivity of the key variables that affect the time to achieve a peak concentration,
and derive the additional insights necessary to ensure that site- and region-specific values are
available for the most sensitive parameters.
The finite-source approach is very appropriate for EPA to use. However, clear and
precise definition of the source terms must be developed to insure that the approach is used
properly. For example the definition for Q,, the waste concentration, is imprecise. In addition to
-------
the definitions, it is important to explain how the source terms are related to mass transfer, time
dependence, and the availability of the source. The model should also be run under contrasting
scenarios: liners versus no liners and under different closure alternatives that are specified in
existing regulations. It is important that all options offered by EPACMTP are tested to verify that
they perform properly before the model is released.
The MINTEQ model used for metal speciation was not evaluated as part of this review.
While the addition of metal speciation to the EPACMTP model is valuable, the accuracy of the
model estimates must be verified and the documentation of this use needs to be clarified. For
example, the pH difference between the soil and the leachate could cause significant errors. These
possible errors need to be evaluated to be sure that the MINTEQ-generated curves are being
properly generated and appropriately used in the EPACMTP model. Geochemical data on
chromium (VI), selenium, and cadmium are available which should be incorporated in the
MINTEQ code for use with EPACMTP.
In the course of this review, the Subcommittee has referred to the Agency's "Guidance for
Conducting External Peer Review of Environmental Regulatory Models" which was itself
reviewed by the SAB (SAB, 1993d).
We appreciate the opportunity to review the improvements that EPA has made to develop
the EPACMTP model for analyzing the transport and fate of chemical releases to groundwater
from land disposal facilities. Again, we commend the Agency for its progress, and we appreciate
EPA's efforts to respond directly to our past recommendations. We look forward to your
response to our recommendations on the EPACMTP.
Sincerely,
Dr. Genevieve M. Matanoski, Chair
Executive Committee
kJ,
Dr. Ishwar P. Murarka, Chair (Dr. James W. Mercer, Chair
Environmental Engineering OSWER Exposure Model
Committee Subcommittee
-------
NOTICE
This report has been written as part of the activities of the Science Advisory Board, a
public advisory group providing extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency. The Board is structured to provide
balanced, expert assessment of scientific matters related to problems facing the Agency. This
report has not been reviewed for approval by the Agency and, hence, the contents of this report
do not necessarily represent the views and policies of the Environmental Protection Agency, nor
of other agencies in the Executive Branch of the Federal government, nor does mention of trade
names or commercial products constitute a recommendation for use.
-------
ABSTRACT
The OSWER Exposure Modeling Subcommittee (OEMS) of the Environmental
Engineering Committee of the Environmental Protection Agency's Science Advisory Board (SAB)
reviewed the Agency's Office of Solid Waste and Emergency Response (OSWER) Composite
Model for Leachate Migration with Transformation Products (EPACMTP) on March 8, 1995.
The review examined the mathematical formulation, a site-based approach using hydrogeologic
regions, and the metal speciation model (MINTEQ) linked to the EPACMTP model to assess the
subsurface fate and transport of metals.
The Subcommittee concluded that the mathematical formulation incorporating daughter
products into the model appeared to be correct and quite useful. The regional site-based
approach is better than the national distribution approach used in the previous model. In addition,
the finite-source approach will be appropriate for regulatory analysis once the definition of the
source term has been clarified and validated. The Subcommittee encourages the Agency to
improve its documentation of the performance of its modules and to document how they had
responded to past peer reviewer comments. The MINTEQ model for metal speciation is a
valuable addition, but it needs to have further testing and other databases need to be consulted.
Overall, the Subcommittee commends OSWER for its improvements to the EPACMTP model
and its responsiveness to previous SAB suggestions.
KEYWORDS: Groundwater Modeling, Transport and Fate, RCRA
-------
U.S. Environmental Protection Agency
Science Advisory Board
Environmental Engineering Committee
OSWER MODEL EXPOSURE SUBCOMMITTEE
Chair
DR. JAMES W. MERCER, President, GeoTrans, Inc., Sterling, Virginia
Members and Consultants
DR. LINDA M. ABRIOLA, Associate Professor, Department of Civil Engineering, University of
Michigan, Ann Arbor, Michigan
DR. CALVIN C. CHIEN, Environmental Fellow, Corporate Remediation Group, E.I. dupont de
Nemours & Company, Wilmington, Delaware
DR. ISHWARP. MURARKA, Target Manager, Health Sciences Business Unit, Electric Power
Research Institute, Palo Alto, California
Ms LYNNE PRESLO, Senior Vice President, Earth Tech, Inc., Berkeley, California
DR. MITCHELL J. SMALL, Professor, Department of Civil Engineering, Carnegie-Mellon
University, Pittsburgh, Pennsylvania
Science Advisory Board Staff
DR. EDWARD S. BENDER, Designated Federal Official, Science Advisory Board, U.S. EPA,
401 M Street, SW (1400F), Washington, DC 20460,
MRS. DOROTHY M. CLARK, Staff Secretary, Science Advisory Board, U.S. EPA, 401 M Street,
SW (1400F), Washington, DC 20460
-------
TABLE OF CONTENTS
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 2
2.1 Background and Prior SAB Reviews of Related Documents 3
2.2 Charge 4
3. RESPONSE TO THE CHARGE QUESTIONS 5
3.1 Daughter Products 5
3.2 Regional Approach 6
3.3 Finite-Source Approach 7
3.4 Metal SpeciationModel 10
4. ADDITIONAL COMMENTS 11
REFERENCES R-l
APPENDIX A. GLOSSARY OF SELECTED TERMS A-l
-------
1. EXECUTIVE SUMMARY
The OSWER Exposure Model Subcommittee (OEMS)of the Environmental
Engineering Committee of the Science Advisory Board of EPA has reviewed the EPA's
Composite Model for Leachate Migration with Transformation Products (EPACMTP). The
OEMS began this review process through a consultation with OSWER staff and consultants
on March 2-3, 1994 on the concepts for the determination of soil screening levels. Following
further development by OSWER, OEMS reviewed changes that OSWER made to the model
on March 8, 1995. The charge for this review was based on four questions:
a) Is the mathematical formulation in the EPACMTP of the subsurface fate and
transport of daughter products appropriate for EPA to use in establishing nation-wide
exit levels for hazardous waste in future regulations?
b) Is the regional approach, using hydrogeologic from specific sites within regions,
better or should OSW continue to use inputs based on national distributions?
c) Is the finite source approach adequate for regulatory purposes?
d) Should the metal speciation model (MINTEQ) be linked to the EPACMTP model
to assess the subsurface fate and transport of metals as part of EPA's national
rulemaking efforts?
The Subcommittee concluded that the mathematical formulation incorporating
daughter products into the model appeared to be correct and useful. The regional approach
is better than the national distribution approach used in the previous model. However,
variability and sensitivity of the input variables need to be clarified. The finite-source
approach is appropriate, but the definitions of the source terms must be clarified to assure
proper use of the approach. The MINTEQ model is a valuable addition, but it needs to have
further testing to verify its accuracy in this context. Overall, the Subcommittee commends
OSWER for its improvements to the EPACMTP model and its responsiveness to previous
SAB suggestions.
-------
2. INTRODUCTION
EPA's Composite Model for leachate Migration with Transformation Products
(EPACMTP) computer code is a simulation model for subsurface fate and transport of
contaminants released from land disposal sites. EPACMTP (EPA, 1994a) is designed to
predict human exposure to groundwater pollutants in a domestic drinking water receptor well
impacted by such releases. The model is applied to support development of regulations for
management and disposal of hazardous wastes. Simulations are performed using probabilistic
input specifications, where the model is designed to be used for generic, nationwide
assessments using Monte Carlo simulation techniques. It is not intended for site-specific
applications. EPACMTP extends and enhances the modeling approach adopted for the 1990
Toxicity Characteristic (TC) Rule promulgated by the U.S. Environmental Protection Agency
in March 1990. For that rule, the Agency used EPACML (EPA's Composite Model for
Landfills) (EPA, 1990) to estimate the potential human exposure to hazardous chemicals
leaching from land disposal facilities. EPACML accounts for first-order decay and linear
equilibrium sorption of chemicals, but disregards the formation and transport of
transformation products (also known as daughter products). The analytical groundwater
transport solution technique employed in EPACML further imposes certain restrictive
assumptions; specifically, the solution can handle only uniform, unidirectional groundwater
flow and thereby ignores the effects of groundwater mounding on contaminant migration and
groundwater flow. To address the limitations of EPACML, the modeling approach has been
enhanced and implemented in EPACMTP. The EPACMTP modeling approach incorporates
greater flexibility and versatility in the simulation capability; i.e., the model explicitly can take
into consideration:
a) chain transformation reactions and transport of daughter products,
b) effects of water-table mounding on groundwater flow and contaminant
migration,
c) finite source, as well as continuous source, scenarios, and
d) metals transport by linking EPACMTP with outputs from the MINTEQ metals
speciation model.
To facilitate the use of the model, interactive pre- and post-processors have been developed to
assist in problem set-up and analysis.
EPACMTP contains a vadose zone module called Finite Element Contaminant
-------
Transport in the Unsaturated Zone (FECTUZ), a saturated zone module called Combined
Analytical-Numerical SAturated Zone in 3-Dimensions (CANSAZ-3D), and a Monte Carlo
module for nationwide uncertainty analysis. The FECTUZ module is designed to simulate
vertically downward steady-state flow and contaminant transport through the unsaturated
zone above an unconfmed aquifer. FECTUZ is based on the EPA's numerical unsaturated
zone simulator, VADOFT, but with extensions and enhancements to optimize the
computational efficiency for Monte Carlo analyses (McGrath and Irving, 1973), and to handle
multi-species decay chains. The transport simulator of FECTUZ can accommodate advection,
longitudinal dispersion (in the vertical direction), first-order degradation with daughter
product formation, and linear or nonlinear Freundlich equilibrium sorption. In cases where the
transformation products are of concern, FECTUZ can handle either straight or branched
decay chains with up to seven different chemical species, i.e., a parent and up to six daughter
products. FECTUZ predicts concentrations at the water table, which provides the input for
CANSAZ-3D. The CANSAZ-3D module simulates 3-D steady-state groundwater flow and
transient or steady-state contaminant transport. The flow simulator of CANSAZ-3D accounts
not only for ambient groundwater flow, but also for leakage from a land disposal unit and
regional recharge. The transport simulator of CANSAZ-3D accounts for 3-D advection and
dispersion, first-order decay with daughter product formation, and linear or non-linear
Freundlich equilibrium sorption. In cases where daughter product formation is of interest,
CANSAZ-3D can accommodate up to seven different species, i.e., a parent with up to six
daughter products.
2.1 Background and Prior SAB Reviews of Related Documents
The U.S. Environmental Protection Agency (EPA), Office of Solid Waste and
Emergency Response (OSWER) has been using and improving mathematical models since the
early 1980s when the Vertical Horizontal Spread (VHS) model (Domenico and Palciauskas,
1982) was used. This model was replaced by the Combined Analytical-Numerical SAturated
Zone (CANSAZ) flow and transport model used in the EPA Composite Model for Surface
Impoundments (EPACMS). The CANSAZ model was reviewed by the Science Advisory
Board (SAB) in 1990 (SAB, 1990). Earlier in 1988, the SAB had reviewed the Unsaturated
Zone Code (FECTUZ) (SAB, 1988). In 1989, the SAB issued a resolution on use of
mathematical models by EPA for regulatory assessment and decision-making (SAB, 1989),
which was directed, in part, at OSWER. In 1991, the SAB provided a review of OSWERs
draft report on the usage of computer models in the hazardous waste/Superfund programs
(SAB, 1991b). As recently as March 2-3, 1994, the SAB provided a consultation on
EPACMTP (SAB, 1994). OSWER has been receptive to previous SAB review comments
-------
and consultation. Thus, this SAB review of EPACMTP represents a peer review of
groundwater models developed and significantly improved by OSWER for use in
RCRA/Superfund regulations. This current review followed EPA guidance (EPA, 1994b) that
was reviewed by the SAB (SAB, 1993c). There are also four additional SAB publications on
groundwater modeling by the EEC that are listed in section 5 of this report.
2.2 Charge to the Subcommittee
The areas listed below were identified for SAB review in a request from the Office of
Solid Waste and Emergency Response.
a) EPACMTP is the latest and most advanced of the OSW subsurface fate and
transport models designed to be computationally efficient for usage in Monte
Carlo analysis for national rule-making. Is the mathematical formulation in
EPACMTP for the subsurface fate and transport of daughter products from
degrading organic chemical constituents sound, and is it appropriate for EPA
to use this approach to establish nationwide exit levels for hazardous waste in
future regulations?
b) The OSW has been using a national Monte Carlo procedure in which national
distributions of parameters are used as input to the model. An alternative
approach has been developed using a regional site information in which
hydrogeologic model input parameters are selected from specific sites within
hydrogeologic regions and, in general, have cross-correlations. Is this regional
approach better, or should EPA continue to use the approach based on national
distributions of input parameters?
c) The OSW's most recently-used approach is based on an infinite source steady-
state model. EPA has developed a finite-source approach for use with
EPACMTP. Is this approach adequate for regulatory purposes?
d) MINTEQ (metal speciation model) was developed by EPA. EPA has recently
developed the linkage of the output of the model with EPACMTP to assess the
subsurface fate and transport of metals. Is this linkage for metals appropriate
for national rule-making efforts?
-------
3. RESPONSE TO THE CHARGE QUESTIONS
OSWER has been responsive to SAB comments on prior modeling efforts and has
developed an improved model and approach to simulate transport and fate of contaminants
released to the subsurface. Because of the nature of the review and Subcommittee charge, the
focus is on recommendations to improve the model further. Comments that follow address
ways that the model and/or its documentation and use can be enhanced. Generally the
comments include a brief discussion followed by recommendations. Each question from the
charge is addressed in separate subsections below.
3.1 Daughter Products
Incorporation of daughter products into EPACMTP makes the model more complete.
The treatment of degradation with a first-order decay rate is consistent with available scientific
information. The mathematical formulation appears to be correct. Degradation is an
important process that should be considered and, therefore, its incorporation into EPACMTP
is conceptually appropriate. The SAB cannot comment on information/data input on
degradation because these were not reviewed. These data are an important aspect of proper
implementation of modeling degradation. Although parameters and data associated with
hydrolysis tend to be fairly well defined (given pH and temperature conditions), it should be
recognized that parameters and data associated with biodegradation are more imprecise, vary
over a wide range, and are site specific. Because of this uncertainty associated with the
biodegradation of certain chemicals, the three following recommendations are made.
Recommendation 1 - EPA should carefully consider whether to use biodegradation
and associated data when applying EPACMTP on a national basis to certain chemicals that are
known to only be moderately (or possibly) biodegradable (see, e.g., Wilson and McNabb,
1983).
Although the omission of biotransformation term from the model may yield
conservative predictions for some chemicals (e.g., BTEX ), this omission will not yield
conservative predictions for other chemicals that have more toxic daughter products (e.g.,
trichloroethene degrading anaerobically to vinyl chloride). In addition, EPACMTP does not
consistently account for multiple parent compounds that produce the same daughter
compound. From the mathematical formulation (see Equation 2.16 in EPA, 1994a, p. 1-14),
it appears that the code has the capacity to deal with multiple source compounds. However,
-------
in the presentation to the Subcommittee, OSWER consultants suggested that all daughter
products are treated "independently" in the Monte Carlo analysis. Such an approach could
significantly underestimate the amount of a daughter compound that is formed. There is not
enough documentation on how the model is actually implemented to determine whether this
aspect of daughter formation is properly addressed, and more clarification is required.
Furthermore, there appears to be inadequate numerical verification of the daughter products
modules (particularly those which are coupled with nonlinear sorption).
Recommendation 2 - EPA should perform further verification (to ensure proper
coding) and document how the daughter products modules of EPACMTP are implemented.
Recommendation 3 - EPA should document known instances in which more toxic
daughter products are formed from multiple parent chemicals and, in the EPACMTP modeling
process, allow for special postprocessing of those chemicals to ensure that every degradation
pathway is counted, with consideration of rate limiting steps.
3.2 Regional Approach
The regional approach for describing national variability in site conditions represents
improvement over the previous approach in which groundwater fate and transport parameters
were independently sampled from national distributions. That approach fails to account for
the correlation which occurs between parameters at a site due to physical relationships among
soil properties and regional trends in climate and geohydrology. The sampling of parameter
sets from actual sites, based on a regionally stratified sample, allows these inherent
correlations and trends to be properly incorporated by using a more localized data set. This
modification in approach responds to a specific recommendation that was made by the SAB in
reviewing CANSAZ (SAB, 1990), and it should be implemented instead of the national
approach in the EPACMTP. The EPA has provided good documentation on the national
distribution of hydrologic parameters that result. It should be noted, however, that some
parameters are still sampled independently based on a national estimate, because the data for
these parameters are not yet generally available for the sites included in the regional site data
base. EPA has recently published guidance on documenting the costs and performance of
remediation projects (EPA, 1995), which includes site-specific data that might be useful for
this purpose.
The following recommendations concern data and the sensitivity of the calculation to
that data.
-------
Recommendation 4 - EPA should compare the variability distributions of the dilution
and attenuation factors (DAFs) that result from the regional (new) and independent national
sampling (old) approaches, and document the differences and the reasons for those
differences.
Recommendation 5 - EPA should determine and document the sensitivity of
computed DAFs to key hydrologic and chemical variables. This analysis should include
varying distributions and documenting how DAFs are impacted. Discussions should be
provided to help clarify how much of the variation in DAFs is associated with parameter
variability from site to site, and how much is associated with uncertainty in the parameter
values. An example of such a study is provided in Chiang et al. (1995).
Recommendation 6 - EPA should perform the necessary additional data collection of
uncertain parameters and parameters that are currently unavailable on a site-specific basis,
especially parameters that are found to have major effects on DAFs. If data cannot be
obtained, then EPA should attempt to identify correlations that can be used on a regional basis
to estimate these parameters in a manner that can account for the physical conditions at sites.
3.3 Finite-Source Approach
EPACMTP model predictions of concentrations at a receptor well are closely tied to
source term concentration inputs to the code. The previous version of the model, EPACML,
was designed to handle only a constant and continuous source concentration. Such an
approach implies an infinite contaminant mass within the waste, and may lead to unrealistic
and overly conservative model predictions. In response to this recognized limitation of
EPACML, the new EPACMTP model has a number of refinements that are directed towards
the improvement of the representation of the source term. The model can now handle a
source of limited mass and finite duration. Options also have been incorporated to treat a
time-varying source concentration. This variation in time may be the consequence of first
order decay, or production or desorption from the waste solids, or continuous loading.
Although these model refinements in the source term treatment represent a substantial
conceptual improvement for the model, the Subcommittee has identified a number of concerns
with the source term documentation and mathematical development which should be
addressed.
Insufficient information is given in the background documents and user guide for
EPACMTP to identify underlying assumptions in source term quantification, to precisely
-------
define important source term parameters, or to understand how the parameters will be
evaluated for code implementation in the Monte Carlo analysis. In general, more attention
needs to be directed to the definition and evaluation of source term parameters and the
validation of the appropriateness of the underlying model assumptions for the description of
contaminant leaching. Some additional Subcommittee concerns relating to the above issues
are summarized below.
a) The EPACMTP background documents fail to precisely define Cw, the "waste
concentration." An analysis of the source term governing equations suggests
that the units of this parameter are mass of a particular waste constituent per
total mass of waste (wet waste mass). It is unclear whether the term is meant
to account for the total mass of the contaminant in the waste or only that
portion which may enter the aqueous phase and be transported into the
unsaturated zone (the teachable portion). The latter definition would appear to
be more consistent with the intended model use. Clarification is needed on this
point. Guidelines and references also need to be supplied in the documentation
as to how this waste concentration can be precisely measured for a specific
waste.
b) New model options in EPACMTP permit computation of a varying CL at the
source. Here CL represents the mass of a waste constituent per volume of
leachate solution. Two alternative models for the functional time dependence
of CL are incorporated in the simulator. A number of fundamental assumptions
appear to be critical to the mathematical representation of these source models,
but they are not explicitly stated nor justified in the documentation. One
mathematical model describes a waste in which a portion of the contaminant
mass is associated with the waste solids (EPA, 1994a, p. 20-21). An
assumption of equilibrium partitioning between the solid and aqueous phases
results in an exponentially decaying source term. Here an implicit but unstated
assumption is that the sorption is linear.
The second model is designed to model a source constituent, which is subject
to first-order reaction (decay and/or production). The source concentration
expression, as given by Equations 18a and 18b (EPA, 1994a), however, does
not appear to represent a correct mass balance expression. No accounting is
made of the waste that may be associated with the solids, nor is there a term to
-------
account for the mass that leaches out of the system. The intended assumptions
should be explicitly stated and justified. The mass associated with the waste
solids should also be considered, as well as its potential for reaction. For
microbial reactions, such bound mass is typically unavailable to the organisms.
The expression for source term duration given by Equation 19 (EPA, 1994a)
would appear to represent a minimum duration. This expression should be
justified.
c) The modeling approach also implicitly assumes that the transformation rate is
constant, unaffected by inhibition or other limiting factors. While this may be
reasonable for an abiotic chemical degradation process, it may not be so for a
biologically-mediated process.
As a result of this discussion, the following recommendations are made.
Recommendation 7 - EPA should carefully define Cw and CL and indicate how decay,
including a decaying source, impacts these parameters and tp.
Recommendation 8 - A validation/verification of calculated DAFs from a "fresh" finite
source would be helpful. Most validation studies presented are ones that have been used
historically, but these do not test new code features.
The assumptions that underlie the source terms included in the EPACMTP model are
almost exclusively relevant to situations in which the contaminant source is uncontained.
Currently, containment systems such as liners, covers, and slurry walls have been implemented
at contaminated sites on a national basis. The design (configuration, material properties, and
dimensions) of a containment system influences the source term for contaminant transport.
Thus the application of the EPACMTP model to contaminant transport from contained
wastes, without modification of the source term can result in overestimation of the rate of
contaminant migration from a source. More work needs to be performed to relate source
term decay or growth (with time) to the design and performance of barriers.
Recommendation 9 - A scenario analysis of the source term is appropriate. This
analysis should include the remediation options of liners versus no liners and cap/closure
versus no closure that are specified in existing regulations. An example of such a study is
provided in Chiang et al. (1995).
-------
3.4 Metal Speciation Model
For metal speciation, EPA is using MINTEQ. In MINTEQ, precipitation and sorption
reactions are treated using a two-step process; they are not considered simultaneously. As a
result of the two-step process, an initial leachate concentration and distribution coefficient
isotherms (as a function of concentration) are generated. The pH difference between the
leachate and the environment could cause the initial concentration to exceed the chemical
solubility, which would impact solute transport and DAF calculations. This issue is not
considered in the present EPACMTP model.
Recommendation 10 - EPA should ensure that initial concentrations do not exceed the
chemical solubility. EPA should make sure the MINTEQ approach is consistent with the
teachability phenomena discussed in the SAB report EPA-SAB-EEC-92-003.
The MINTEQ code is composed largely of chemical data bases. The SAB did not
review MINTEQ and thus cannot comment on the inputs to EPACMTP for metals. The
version that EPA is using does not contain the latest data available for cadmium, chromium
(VI), and selenium, which will impact calculations for these chemicals. Additional useful
information on metals availability may be available from the Office of Water through its
Contaminated Sediment Research program.
Recommendation 11 - EPA should update MINTEQ using the latest chemistry data
available. This is especially true for chromium (VI), and may be true for other metals such as
selenium and cadmium.
For sorption processes, more discussion is necessary to justify the approach. Sorption
in the vadose zone is treated as nonlinear, whereas that in the saturated zone is assumed to be
linear. Data need to be provided supporting the different approaches used above and below
the water table.
Recommendation 12 - EPA should better document the choice of vadose versus
saturated zone distribution coefficient isotherms, and explain differences where they occur.
For some chemicals, especially some metals, transport times to reach peak
concentrations may be on the order of centuries or longer. It is unclear if and how EPA
factors this time frame to reach peak concentrations into the DAF calculation.
10
-------
4. ADDITIONAL COMMENTS
During the review, discussions indicated that certain portions of EPACMTP are used
more frequently than other portions, and that certain code options have had only limited
testing and some have none at all. For example, in the metals background document (p. 5-4),
the statement is made that, "The numerical nonlinear isotherm scheme works only for mildly
nonlinear cases." It is important, before releasing EPACMTP, that all options offered should
be tested to ensure that they work as designed.
Having independent reviewers review EPACMTP was good, and consistent with EPA
external peer review guidance. These reviewers raised a number of important questions
concerning the model assumptions and formulations. As part of the ensuing guidance, a
response to reviewer comments should be adequately documented. Such documentation was
missing or inadequate in this case. To provide a complete record and be consistent with peer
review guidance, better documentation needs to be provided on all responses to reviewer
comments.
There is a large family of EPA codes, many of which have similar origins. That is,
many of the codes have subcodes that are the same. Examples include EPACMS, EPACML,
EPACMTP, MULTEVIED, and MMSOILS. It would be helpful for EPA to discuss and
document the relationship among these codes. This would represent the first step in an
ongoing communication that should occur within EPA. These various codes are undergoing
continual testing and errors are noted and corrected in some of these codes. It appears that
documentation of corrected errors does not occur and is not distributed to other portions of
EPA where similar codes with the same subcode are being used.
Recommendation 13 - EPA should establish a central location where this debugging
(code correction) information is contained and distributed. This function should be
coordinated with the Agency working group on regulatory models and peer review to
promote consistency and advise the users about enhancements.
The validation studies provided by EPA are good and are designed to test standard
flow and transport conditions. Interestingly, the three validation studies are site-specific
applications, for which EPACMTP is not designed as stated by the Agency. Unfortunately,
the studies provided by OSWER do not test the new features in EPACMTP. EPA needs to
perform confirmation testing of a) daughter products, b) metals speciation, and c) finite-
source. EPA should also explore ways to confirm the regional distribution approach. One
11
-------
approach is to present various scenarios and examples, and check for reasonableness. This
effort should include a discussion on the sensitivity of DAFs to different parameters and
scenarios.
12
-------
REFERENCES
Background documents and literature
Chiang, C.Y., P.D. Petkovsky, and P.M. McAllister. 1995. A risk-based approach for
managing hazardous waste, Ground Water Monitoring & Remediation, Winter, pp.
79-89.
Domenico, P. A. and V.V. Palciauskas. 1982. Alternative boundaries in solid waste
management, Ground Water, 20:301-311.
EPA. 1990. Background document for EPA's composite model for landfills (EPACML),
Office of Solid Waste, USEPA, Washington, DC.
EPA. 1994a. Background documents for EPACMTP: Finite source methodology for non-
degrading and degrading chemicals with transformation products, Office of Solid
Waste, USEPA, Washington, DC.
EPA. 1994b. Guidance for Conducting External Peer Review of Environmental Regulatory
Models. Agency Task Force on Environmental Regulatory Modeling. USEPA,
Washington, DC. EPA-100-B-94-001. July 1994.
EPA. 1995. Guide to Documenting Cost and Performance for Remediation Projects.
Prepared by Member Agencies of the Federal Remediation Technologies Roundtable.
USEPA, Washington, DC EPA-542-B-95-002.
McGrath, EJ. and D.C. Irving. 1973. Techniques for efficient Monte Carlo simulation,
Technical report prepared for Department of the Navy, Office of Naval Research,
Arlington, VA.
Wilson, J.T. and J.F. McNabb. 1983. Biological transformation of organic pollutants in
groundwater, EOS, 64(33).
R-l
-------
Previous SAB Reviews of Groundwater Modeling Approaches
Science Advisory Board. 1988. Review of the Unsaturated Zone Code for the OSW Fate and
Transport Model. Report of the Unsaturated Zone Code Subcommittee. SAB-EEC-
88-030. June, 1988.
Science Advisory Board. 1989. Resolution on Use of Mathematical Models by EPA for
Regulatory Assessment and Decision-Making. Report of the Environmental
Engineering Committee. EPA-SAB-EEC-89-012. January 1989.
Science Advisory Board. 1990. Review of the CANSAZ Flow and Transport Model for Use
in EPACMS. Report of the Saturated Zone Model Subcommittee. EPA-SAB-EEC-
90-009. March, 1990.
Science Advisory Board. 1991a. Consultation on Review of Selection Criteria for
Participation of EPA staff on the Proposed Agency Task Force on Modeling. EPA-
SAB-CON-EEC-91-002. April 29, 1991.
Science Advisory Board. 1991b. Usage of Computer Models in the hazardous Waste and
Superfund Programs. A review of OSWER's draft report and pilot study. EPA-SAB-
EEC-91-016. September, 1991
Science Advisory Board. 1991c. Leachability Phenomena: Recommendations and Rationale
for Analysis of Contaminant Release by the Environmental Engineering Committee.
EPA-SAB-EEC-92-003. October, 1991.
Science Advisory Board. 1993a. Consultation on Groundwater Modeling Pathways for
Radioactive Wastes. EPA-SAB-EEC-CON-93-004. April, 1993
Science Advisory Board. 1993b. Review of the Assessment Framework for Ground-Water
Model Applications. Review of the OSWER response assessment framework for
ground-water model applications. EPA-SAB-EEC-93-013. June, 1993.
Science Advisory Board. 1993c. Review of Draft Agency Guidance for Conducting External
Peer Review of Environmental Regulatory Modeling. EPA-SAB-EEC-LTR-93-008.
July, 1993.
R-2
-------
Science Advisory Board. 1993d. Review of MMSOILS component of the Proposed RIA for
the RCRA Corrective Action Rule. Review of OSWER and ORD draft
documentation and user's manual and RIA of the MMSOILS multimedia contaminant,
fate, transport, and exposure model. EPA-SAB-EEC-94-002. November, 1993.
R-3
-------
APPENDIX A. GLOSSARY OF SELECTED TERMS
1. CANS AZ-Combined Analytical-Numerical Saturated Zone (CANS AZ) a two-dimensional
flow and transport model.
2. CANSAZ-3D-Combined Analytical-Numerical SAturated Zone in 3-Dimensions
(CANSAZ-3D), a module in EPACMTP for the saturated zone of groundwater.
3. DAF-Dilution and attenuation factors.
4. EPACML-EPA's Composite Model for Landfills (EPA, 1990), an earlier OSWER
computer model used to estimate the potential human exposure to chemicals leaching
from land disposal facilities.
5. EPACMS-EPA Composite Model for Surface Impoundments, the precursor to EPACML.
6. EPACMTP-EPA's Composite Model for leachate Migration with Transformation
Products. This computer simulation model is intended to predict the subsurface fate
and transport of contaminants released from land disposal sites.
7. FECTUZ-Finite Element Contaminant Transport in the Unsaturated Zone, a module in
EPACMTP for the vadose or unsaturated zone of groundwater.
8. OSWER-the Office of Solid Waste and Emergency Response
9. MINTEQ (metal speciation model) was developed by EPA whose output is used by
EPACMTP.
10. OSW-Office of Solid Waste
11. TC- Toxicity Characteristic, this attribute, defined by a 1990 EPA rule is one criterion
used to classify wastes as hazardous or non-hazardous.
12. VMS-Vertical and Horizontal Spread (VHS) model (Domenico and Palciauskas, 1982),
the first EPA groundwater contaminant model reviewed by SAB.
A-l
-------
DISTRIBUTION LIST
Deputy Administrator
Assist Administrators
EPA Regional Administrators
EPA Laboratory Directors
Deputy Assist Administrator for Office of Solid Waste and Emergency Response (OSWER)
Director, Office of Solid Waste (OSW)
Deputy Director, OSW
Director, Office of Emergency and Remedial Response (OERR)
Deputy Director, OERR
Deputy Assistant Administrator for Office of Prevention, Pesticides and Toxic Substances
(OPPTS)
Deputy Assistant Administrator for Office of Research and Development (ORD)
Director, Office of Environmental Engineering and Technology
Demonstration (OEETD)
Deputy Director, OEETD
Director, Office of Monitoring, Modeling and Quality Assurance (OMMSQA)
Director, Center for Environmental Research Information (CERI)
EPA Headquarters Library
EPA Regional Libraries
EPA Laboratory Libraries
Congressional Research Service
Library of Congress
------- |