i 6jo-uiise-MMM//:dnM . 8996-ZE8-0 L9 =XVJ 9896-ZE8-OL9
6S6Je-8Zfr6L Vd 'ua>poi|oqsuoD jsaft/j SAIJQ joqjBH JJcg 001 JAU.SV
, ,-, -., ,-,--^. .-,-,:,-,, ,- buijsei JO} A;3pos UBopaiuvam o; 1-0-80Z9S8 X#
juaoiaajBe epuBjsissB japun AouaBy uojpsiojd EtuauiuojiXug S9)e;s pajiun 341 ^q AHOLJM papunj ussq seq jusmnoop smx
aoNvain Nonoanas
PUB lAiniQN3dlAIOO
:Sl3aOIAI lubdSNVdl PUB
spjepuejs IAIJJSV jo uo!JBD||ddv pue
'asn 'aDueiujo^iBd aqi uj Bumiejj.
WN
-------
-------
ADDITIONAL INFORMATION
Do you want additional copies of this Document?
You can write, fax, or phone your request to:
ASTM
Attn: Technical & Professional Training Dept.
100 Barr Harbor Drive
West Conshohocken, PA 19428
Tel: 610-832-9685
Fax: 610-832-9668
An electronic copy of this document can be downloaded at the web site:
http://www.epa.gov/oust/rbdni
Do you want more information about ASTM or its voluntary consensus
standards on Risk-Based Corrective Action?
You can contact us via our web site at:
http://www.astm.org
Do you want more information on the U.S. EPA's Underground Storage Tank
(UST) Program or Risk-Based Decision Making?
Visit the U.S. Environmental Protection Agency's UST program website at:
http://www.epa.gov/oust
Or you can call EPA's RCRA/Superfund Hotline, Monday through Friday, 8:30 a.m.
to 7:30 p.m. EST. The toll-free number is 800-424-9346
This document is not a standard and has not been approved by the ASTM
consensus system.
Copyright ฉ 1999 American Society for Testing and Materials, West
Conshohocken, PA. All rights reserved. This document may not be reproduced
or copied, in whole or in part, by any means without the express written
approval of the President, ASTM.
-------
-------
ACKNOWLEDGMENTS
The American Society for Testing and Materials (ASTM) would like to acknowledge the
many individuals who contributed to this document.
Mr. Richard Mattick was the USEPA OUST project officer and coordinated the technical
review of the document.
Mr. Scott Murphy was the ASTM project manger.
The guidance was developed by Foster Wheeler Environmental Corporation by Dr. James
Kennedy, principal author, with contributions from Dr. Ronald Marnicio, Ms. Monica Caravati
and Dr. Emily Kennedy.
This document received extensive peer review from State programs, USEPA and the
National Partnership in RBCA Implementation (PIRI). ASTM and USEPA would like to thank
those who commented and participated in the review process. These comments helped to
significantly shape this Guidance into a product that has targeted "real world" modeling issues of
State corrective action programs for leaking underground storage tanks. These reviewers include
Gilberto Alvarez (USEPA), Michael R. Anderson (Oregon Department of Environmental
Quality), David Ariail (USEPA), Steven Bainbridge (Alaska Department of Environmental
Quality), Phil Bartholomae (BP Oil Company), Paul Bauer (New Jersey Department of
Environmental Protection), Dave Brailey (OilRisk Consultants), Chet Clarke (Texas Natural
Resource Conservation Commission), Tom Conrardy (Florida Department of Environmental
Protection), Scott Ellinger (USEPA), Geoff Gilman (Amoco Corporation), Annette Guiseppi-Elie
(Exxon Biomedical Sciences Inc.), John Gustafson (Equilon Enterprises LLC), Merlyn Hough
(Oregon Department of Environmental Quality), Steve Howe (Unocal Corp.), Walter Huff
(Mississippi Department of Environmental Quality), Jack Hwang (USEPA), Robin Jenkins (Utah
Department of Environmental Quality), Karen Lyons (Equilon Enterprises LLC), Mark Malander
(Mobil Oil Corp.), Donna Miller (Chevron), Norm Novick (Mobil Oil Corp.), Richard Oppel
(Oklahoma Corp. Comm.), Roger Przybysz (Michigan Department of Environmental Quality),
Jim Rocco (BP Exploration and Oil Co.), Matthew Small (USEPA), Ken Springer (Shell Oil
Co.), Sandra Stavnes (USEPA), John Stephenson (Pennsylvania Department of Environmental
Protection), Karen Synoweic (Chevron Research and Technology Co.), Michael Trombetta
(Montana Department of Environmental Quality), James Weaver (USEPA), and Joe Williams
(USEPA).
-------
-------
TABLE OF CONTENTS
1.0 INTRODUCTION 1
1.1 PURPOSE..: 1
1.2 METHODS 1
1.3 ORGANIZATION 2
2.0 DESCRIPTIVE MODEL INFORMATION 4
2.1 FATE AND TRANSPORT PROCESSES 4
2.1.1 Advection 5
2.1.2 Dispersion 5
2.1.3 Diffusion : 6
2.1.4 Equilibrium Partitioning 6
2.1.5 Biodegradation/Transformation 7
2.1.6 Separate Phase Flow 7
2.2 TYPES OF FATE AND TRANSPORT MODELS ; 8
2.2.1 Analytical Models 9
2.2.2 Numerical Models 10
2.3 SPECIFIC MODEL INFORMATION 11
3.0 INFORMATION REQUIRED FOR SELECTION OF MODELS 12
3.1 SITE CONDITIONS FOR MODEL APPLICATION 12
3.2 INPUT PARAMETERS 12
3.2.1 Sources of Input Parameter Values 12
3.2.2 Techniques for Measuring Input Parameters 13
3.2.3 Sensitivity of Model Output to Input Parameters 15
4.0 MODEL SELECTION....... 17
4.1 MODEL SELECTION CRITERIA 17
4.2 SELECTION OF MODELS FORTIER2 AND TIER 3 RBCA EVALUATIONS 17
4.2.1 Tier 2 Usage 17
4.2.2 Tier 3 Usage 19
4.3 MODEL PACKAGES 20
4.4 MODEL CALIBRATION AND VALIDATION 21
4.4.1 Calibration 21
4.4.2 Validation 22
4.4.3 Modeling versus Field Data 23
5.0 DEFINITION OF TERMS 24
6.0 BIBLIOGRAPHY 26
MATRICES
Matrix 1: Key Model Information
Matrix 2: Generic Site Conditions for Model Application
Matrix 3: Key Input Parameters
FIGURES
Figure 1:
Figure 2:
Figure 3:
Decision Diagram
Analytical Model Selection Process Diagram
Numerical Model Selection Process Diagram
APPENDIX A Model Summaries
-------
Forward
This Guidance document catalogs and describes non-proprietary fate and transport models
that are readily available and in common use for risk-based corrective action (RBCA) at the time
of publication. It is meant to function as a compendium and resource guide, assisting the user in
the model selection process. It is not intended to be a comprehensive review of every available
fate and transport model nor a comprehensive guidance on the use of any single model. The
Guidance does not endorse models listed nor attempts to rank them or evaluate their performance
or accuracy. Models other than those included in this Guidance may be appropriate choices for
fate and transport modeling at any site. It is the responsibility of the experienced fate and
transport modeler to select the appropriate model. The Guidance does not, at this time, include
complex multi-phase, multi-component models for simulating movement of nonaqueous phase
liquid; models for constituent movement through fractured media; nor does it include proprietary
models.
Regulatory agencies may have certain technical preferences or requirements regarding the
selection or use of fate and transport models. For example, certain agencies may require the use
of models that are peer reviewed and within the public domain (i.e., readily available, widely
distributed, and generally accepted). These preferences or requirements should be considered
when selecting a fate and transport model. Determination of the degree of model calibration (or
the determination on whether or not a model can be calibrated) should also involve consultation
with the appropriate regulatory agencies.
Fate and transport modeling is only one of the many tools needed to successfully implement
the Risk-Based Corrective Action (RBCA) process. The purpose of this Guidance is to assist in
selection of models that can be used to implement the RBCA process, and not to be a substitute
for sound professional judgment. The Guidance does not advocate modeling over the collection
and interpretation of quality media-specific site data.
u
-------
1.0 INTRODUCTION
1.1 Purpose
The purpose of this Guidance Document on Fate and Transport Modeling (Guidance) is to
provide a compendium of commonly used fate and transport models and pertinent information to
aid users in the selection of an appropriate model to be used in the Risk-Based Corrective Action
(RBCA) process. Various formulations of fate and transport models have been used for more
than twenty years to assess and predict movement and behavior of chemicals in the environment.
Over time, more sophisticated fate and transport models have been developed to take advantage
of advances in computer hardware and software technologies, and of improved understanding of
fate and transport processes. There are now many models ranging from very simple to very
complex.
Fate and transport models may utilize simple equations that require minimal data input, or
complex equations that require detailed site-specific information. The RBCA process advocates
a gradual process of using models, starting from simple approaches that will produce
conservative results (i.e., over-prediction of likely constituent concentrations) and moving, as
needed, to complex approaches requiring more data and time. Objectives of modeling should be
defined before model selection begins for it is possible that a simple model will be adequate to
provide the desired information. The complexity of selected models should balance the quantity
and quality of available input data (or of data which can be obtained easily) with the desired
model output.
Fate and transport models are most often used to simulate or predict the distribution of
constituent concentrations in environmental media. In some situations, the collection and
interpretation of good quality data on constituent concentrations in soil and groundwater can
defer the need for modeling. Also, situations may arise where fate and transport models cannot
be adequately calibrated or validated, in which case it may be best to use field data rather than
modeling results in the RBCA process. An application of the RBCA process should consider
both data collection and modeling options for meeting information needs.
This Guidance is presented in a way that information can be used by audiences with varying
levels of experience in fate and transport modeling. It addresses a multitude of chemical fate and
transport pathways, including vapor migration, soil leaching, and groundwater transport
pathways. The Guidance contains information on specific types of models, describes governing
equations and model applicability, lists key input parameters for each model, describes model
output formats and limitations, and presents procedures for sensitivity testing of input parameters
and for validating individual model simulations and predictions.
* 1.2 Methods
The sources of information used to describe the models included in this document are listed
in the Bibliography section of the document. The survey of publications focused on those
aspects of models noted in the Introduction. The survey did not focus on the history of
development of each model, or on literature critiques of the use of a model, except where such
critiques provide insight on the applicability or limitations of a model. Models in the Guidance
are applicable to movement of constituents in porous media; none of the models specifically
-------
address movement in fractured media. This Guidance addresses models which are, for the most
part, referenced in the American Society for Testing and Materials (ASTM) Standard Guide for
Risk-Based Corrective Action Applied at Petroleum Release Sites (E 1739-95), or in documents
cited by the Guide.
This Guidance describes readily-available and published models that were in common use at
the time of writing. Models include those in the public and private domain. For the purpose of
this Guidance, public domain models are considered to be those which can be obtained without
cost from government agencies, such as the U.S. EPA Center for Subsurface Modeling Support
at the Robert S. Kerr Environmental Research Center (http://www.epa.gov/ada/models/html) and
the U.S. Geological Survey (http://water.usgs.gov/software/ground_water.html), where models
can be downloaded from the Internet. Private domain models are considered to be those that can
be obtained at cost from trade associations, university research associations, and commercial
vendors. Specific sources of models, including URL addresses, are included on the model
summaries in an appendix to this Guidance. The models listed in the Guidance have been
through various degrees of peer review. The user should be aware of peer review or other model
use or selection policy requirements of a specific RBCA program and the implementing
regulatory agency.
1.3 Organization
This Guidance is presented as five components:
Text
Bibliography
Matrices
Figures
Model Summaries-Appendix A
Information in each of the matrices is grouped by fate and transport pathway. Matrix 1
presents a summary of key information for various models, including:
Model/algorithm name;
Description of model processes and simulations;
Type of model code/algorithm;
Model outputs;
Model features, characteristics, use conditions, and limitations;
Computer needs; and
Sources of additional information.
The matrix provides a snap-shot of commonly-used models allowing a user of the Guidance
to, for example, quickly identify which models:
Are applicable to which fate and transport pathway;
Use analytical methods, and may be relatively simple:
Are more complex, using numerical methods; and
Can be run using standard spreadsheet applications.
Matrix 2 correlates specific models with generic site conditions. The matrix allows a user of
the Guidance to, for example, distinguish those soil-to-ambient-air models applicable to infinite
-------
source depth from those applicable to finite sources. Distinctions are also made on the basis of
soil or aquifer homogeneity and isotropy, steady-state versus transient conditions, and
incorporation of biodegradation and transformation, among other site conditions. Matrix 3
identifies key input parameters for models and comments on sensitivity of model output to the
input parameter. Input parameters are those commonly needed for fate and transport modeling,
grouped by fate and transport pathway. Sensitive input parameters are highlighted in Matrix 3.
'Figures 1, 2, and 3 illustrate the process of selecting a fate and transport model. Figure 1
addresses the decision process for selecting analytical versus numerical models. The figure is in
the form of a decision diagram considering questions on regulatory requirements for modeling,
model calibration, site complexity, and availability of input parameter values. Figure 2
illustrates the process for selecting analytical fate and transport models for the pathways:
Soil-to-ambient air;
Soil-to-indoor-air;
Soil-to-groundwater;
Groundwater-to-ambient-air;
Groundwater-to-indoor-air; and
Groundwater-transport.
Figure 3 illustrates the process for selecting numerical fate and transport models for the
pathways:
Soil-to-groundwater; and
Groundwater-transport.
Both Figures 1 and 2 present information on input data requirements and model output that
correlate with information in the matrices.
Each of the fate and transport models included in the matrices and figures are summarized in
the appendix to this Guidance. The summaries include information on model operation, key and
sensitive input parameters, applicability of the model, and sources of additional information on
the model. The distinction is made between models for which computer programs are available
from common sources, and models that are in the form of equations typically executed in a
spreadsheet environment. Where available, URL locations of model information are included in
the summaries. The summaries are intended to allow further screening of fate and transport
models selected using information in the matrices and figures.
The text of the Guidance intentionally does not refer to specific fate and transport models so
that selection of a model can be made using information in the matrices, figures, and appendix.
Instead, the text provides general information on fate and transport process and types of fate and
transport models. It describes site conditions for model application, provides information on
model input parameters, and describes model selection criteria relative to RBCA-process tier
levels. The text describes packages incorporating models for a variety of fate and transport
pathways, and describes the process of model calibration and validation. The Guidance includes
a Bibliography with references on fate and transport processes, specific fate and transport
models, measurement of model input parameters, and model packages.
-------
2.0 DESCRIPTIVE MODEL INFORMATION
2.1 Fate and Transport Processes
A principal purpose of fate and transport modeling is to predict and quantify migration of
constituents in the environment that are subject to one or more transport mechanisms. For
example, within ASTM and state RBCA programs, fate and transport modeling is one of the
tools used to establish exposure point concentrations and their corresponding risk-based
screening and cleanup levels.
Fate and transport models are used to predict the migration of chemical constituents through
soil, groundwater and air (or a combination thereof) over time, with most models focusing on
specific fate and transport processes. Fate (i.e., chemical) processes address persistence of a
constituent along the migration pathway while transport (i.e., physical) processes address
mobility of the constituent along the migration pathway. The processes incorporated into fate
and transport models include:
Advection, the movement of dissolved constituents caused by the bulk movement of
fluid (liquids and gasses);
Dispersion, the three-dimensional spreading of dissolved constituents as fluid
migrates through environmental media;
Diffusion, the spreading of a mass of constituents as a result of concentration
gradients;
Equilibrium partitioning of constituent mass between solid and fluid (i.e., liquid and
gas) portions of the environmental medium as a result of sorption, solubility, and
equilibrium chemical reactions;
Biodegradation of constituents by indigenous microorganisms along the migration
pathway; and
Phase separation of immiscible liquids.
Fate and transport models developed for constituent migration analyses have been cited in
numerous guidance documents. Models incorporate, to varying degrees, one or more of the fate
and transport processes highlighted above. For example, a model of vapor migration from soil to
ambient air may incorporate the processes of diffusion and advection for vapor movement to the
ground surface, and atmospheric dispersion of vapors emanating from the ground surface.
Information in this Guidance is grouped into the following fate and transport pathways:
Vapor migration from soil with dispersion in ambient air;
Vapor migration from soil to enclosed spaces and indoor air;
Vapor migration from groundwater to ambient air;
Vapor migration from groundwater to indoor air;
Transfer of constituents from soil to groundwater;
Groundwater transport of dissolved constituents.
Following are brief descriptions of the principal processes incorporated into most fate and
transport models or modeling approaches.
-------
2.1.1 Advection
Advective transport processes are modeled to quantify movement of fluids. Advection is the
dominant mass transport process in groundwater flow systems (Domenico and Schwartz, 1990).
Within a groundwater flow system, for example, advective movement of water occurs through
pores and fractures within soil or rock (often referred to as the "water bearing medium" or
"aquifer"). Equations for advective movement of groundwater therefore require information on
material properties of the soil or rock (e.g., hydraulic conductivity, effective porosity) and a
quantitation of the potential gradient driving groundwater movement (hydraulic gradient).
Conservative constituents do not partition to the environmental media and therefore move at
the same velocity as groundwater. Other constituents move at a velocity less than that of the
bulk groundwater movement due to partitioning between solid and fluid portions of the water-
bearing medium. The retardation equation generates a ratio of the groundwater and dissolved
constituent movement velocities called the retardation factor.
Calculation of the retardation factor for organic constituents requires information on soil
bulk density and effective porosity, fraction of organic carbon in the water-bearing medium, and
the organic carbon partitioning coefficient of the constituent. For inorganic constituents, the
fraction of organic carbon and organic carbon partitioning coefficient are replaced with
analogous coefficients and parameters such as the selectivity coefficient, cation exchange
capacity, and total competing cation concentration in solution (Domenico and Schwartz, 1990),
and information may be needed on geochemical properties such as pH or Eh. It must be noted
that the retardation equation incorporates assumptions on equilibrium partitioning (discussed in a
following paragraph) and may not be representative of all situations.
Advective transport is an important process for vapor movement in the vadose zone.
Advective movement of vapors can be caused by both temperature and pressure gradients.
Temperature gradients can be caused by seasonal or diurnal heating of shallow soil, and pressure
gradients can be caused by wetting fronts of groundwater recharge that trap and compress soil
vapors. Pressure differentials can also be caused by building ventilation systems, or by winds
blowing over a structure, which can result in advective movement of vapors from soil to interior
spaces. Impermeable geologic strata and man-made structures such as pavements can redirect
advective movement of vapors and must be considered in fate and transport modeling.
2.1.2 Dispersion
Dispersion is characterized by the tortuous movement of fluid through an environmental
medium and results in spreading of constituent mass beyond the region that would be occupied
due solely to advective movement of fluid (Domenico and Schwartz, 1990). In the modeling of
groundwater flow systems, coefficients of hydrodynamic dispersion are calculated using a
characteristic of the solid medium referred to as dispersivity and the advective velocity of
groundwater movement. Dispersivity, which is a quantitation of the mechanical mixing that
occurs as a consequence of local variations in flow velocity around the mean velocity, can be
measured or estimated statistically. Dispersivity is often calculated in a fate and transport model
as a scale- and direction-dependent coefficient of the downgradient distance of groundwater
movement. Dispersivity is multiplied by the advective velocity to yield the dispersion
coefficient. The dependence of dispersion on advection is captured in the advection-dispersion
-------
equation, which is the. principal differential equation describing mass transport of dissolved
constituents in groundwater flow systems.
Subsurface vapors emanating to ambient air are dispersed by wind and other atmospheric
phenomena. Atmospheric dispersion is the process of growth of the volume of ambient air in
which a given amount of emanated vapor is spread or mixed. The growth of the imaginary
"balloon" containing the emanated vapor arises from a combination of distortion, stretching and
convolution whereby a compact "blob" or "puff of released vapor is distributed in an irregular
way over a volume which is larger owing to the effective capturing and enclosure of "clean" air
(Pasquill, 1974). Unlike dispersion in groundwater flow systems, atmospheric dispersion
incorporates turbulent movement of the fluid medium. Equations for calculation of atmospheric
dispersion require information on emission rates or fluxes of vapors or surface particles, wind
speed and direction, lateral and vertical dispersion factors, ground-surface characteristics, and
mixing heights.
2.1.3 Diffusion
The process of diffusion occurs as a result of concentration gradients. Constituent molecules
in an environmental medium will move toward media characterized by lower constituent
concentrations. Unlike dispersion, diffusion can occur both in the absence or presence of
advective flow. The diffusive flux of vapors is characterized by an effective vapor phase
diffusion coefficient which is affected by the porosity and moisture content of the environmental
medium, and by the size and structure of constituent molecules.
In groundwater flow systems, the process of diffusion is quantified using the diffusion
coefficient of the constituent and the concentration gradient of the constituent in groundwater. In
the advection-dispersion equation, a coefficient of molecular diffusion can be included in the
coefficient of hydrodynamic dispersion. The coefficient of molecular diffusion is often
negligible compared to the dispersivity term and is typically ignored, except when groundwater
is not moving or the velocity of movement is very small.
Diffusion of soil vapors also occurs as a result of concentration gradients. Depending on the
soil porosity, diffusion may be the major component of vapor movement. However, as pore
spaces decrease in size or become filled with liquids, vapor diffusion decreases. Soil moisture
content and water-filled porosity are therefore important considerations in modeling of fate and
transport of soil vapors.
2.1.4 Equilibrium Partitioning
When groundwater containing constituent contamination is mixed with a solid medium, the
constituent mass begins to partition between the solution, the solid, and any gas present in the
medium (Domenico and Schwartz, 1990). A partitioning coefficient is used to relate the
constituent concentrations in the liquid and solid portions of the medium. The sorption process
is very complex and influenced by physical and mineralogical properties of the solid media,
chemistry of the groundwater, temperature, and pressure. The retardation equation cited in the
preceding description of the advective transport process is a means of quantifying the sorption
process.
-------
Equilibrium partitioning of constituents in environmental media dictates that the total mass
of constituent is equal to the sum of the masses of constituent in the dissolved and vapor phases,
and the mass of constituent sorbed to solid media. When free-phase of the constituent is present,
the total mass of constituent is equal to the sum of the masses in the dissolved, vapor, sorbed,
and free phases. The presence of free phase must be considered so that contaminant mass is not
inappropriately allocated to the other three phases.
The amount of constituent in the vapor and sorbed phases is related to the amount in the
dissolved phase by equations involving Henry's Law constant for vapor phases and partition
coefficients for sorbed phases. Estimating constituent concentration under equilibrium
partitioning conditions requires information on dissolved constituent concentration, water
content and bulk density of the solid medium, distribution coefficient between dissolved and
sorbed phases, Henry's Law constant, and vapor content of the medium.
2.1.5 Biodegradation/Transformation
Biodegradation and transformation are processes that reduce constituent concentrations by
changing the form in which the individual chemical components exist. The most significant
rates of biodegradation/transformation occur by means of aerobic reactions where constituents
act as an electron donor, energy source, arid source of carbon for growth of microorganisms (e.g.,
biodegradation of petroleum hydrocarbon constituents). Oxygen acts as the electron receptor for
aerobic processes and is reduced to water, causing a decrease in dissolved oxygen concentrations
(Wiedemeier, et al, 1995). Availability of oxygen and the rate of oxygen transport are the factors
that most significantly control aerobic processes in subsurface environments. Nitrate, sulfate,
ferric iron, and carbon dioxide can be electron receptors in anaerobic processes, which tend to
have slower reaction rates than aerobic processes.
In some of the less-complex fate and transport models, biodegradation and transformation
reactions can be incorporated as first-order reactions where the decay rate is proportional to the
constituent concentration. Reductions in constituent concentrations (or mass) are calculated
using rate constants and incorporate the concept of half-life, defined as the time it takes for
constituent concentration to be decreased by one-half due to biological degradation or
transformation processes. More complex models can utilize more fundamental approaches for
incorporating the processes. If rates of biodegradation or transformation are unknown, or not
considered appropriate by regulatory agencies or others (e.g., if a conservative over-estimation of
constituent concentrations is desired), the effect of these processes can be eliminated from most
fate and transport models.
2.1.6 Separate Phase Flow
Movement of immiscible liquids can result in migration of liquids under gravitational forces.
Within a groundwater system, light nonaqueous phase liquids (LNAPL) such as petroleum
hydrocarbons that are released at or near the ground surface will move vertically downward to
the water table. The buoyant volume of immiscible liquid will then move horizontally to flatten
out. The LNAPL layer may concurrently move hydraulically downgradient with groundwater.
Dense nonaqueous phase liquids (DNAPL) will move vertically downward, penetrate the water
table, and continue to move vertically downward until gravitational movement is restrained by
physical barriers (e.g., an impermeable geologic stratum) or until the DNAPL volume has been
depleted by residual containment in the zone through which the DNAPL is descending
-------
(Domenico and Schwartz, 1990). Both LNAPLs and DNAPLs are identified as secondary
sources and transport mechanisms in the ASTM Standard Guide for Risk-Based Corrective
Action Applied at Petroleum Release Sites (E 1739-95).
2.2 Types of Fate and Transport Models
A model is any device or construct used to represent or approximate a field situation
(Anderson and Woessner, 1992). They are an assembly of concepts in the form of mathematical
equations that represent some understanding of natural phenomena. Models can be conceptual
representations, physical representations, or mathematical representations (i.e., an equation or
series of equations representing the governing physical processes and boundary conditions).
Modeling is an iterative series of questions and decisions, the first question being the
purpose of the model. Once the purpose is established, a conceptual model is developed. This is
often a pictorial representation of the site to be studied that distills the available field data and
descriptive site information into a simplified representation of the study area. This simplified
representation of natural processes and settings can be more easily represented by a
mathematical model. Typically, simplifying assumptions are made to allow the fate and
transport processes to be represented in mathematical terms.
An equation or computer code is then selected that can both satisfy the modeling purpose(s)
and represent the conceptual model. The model is constructed using field, laboratory, and
literature data, and can be calibrated to observed conditions. Following the completion of the
model run, output data are checked against the simplifying assumptions to confirm that none of
the assumptions were violated, and if so, to what degree.
Fate and transport models can be applied in a forward-calculation mode where constituent
concentrations are predicted based on source area concentrations. Some of the less complex
(typically analytical) models can also be applied in a back-calculation mode where one or more
models are combined to determine the source-area constituent concentration corresponding to an
acceptable concentration at the point of interest (ASTM, 1995). Calculations in either mode
require information on the physical and chemical properties of the constituent; mechanism of
releases of constituents to environmental media; physical, chemical and biological properties of
the media through which migration occurs; and interactions between the constituent and medium
along the migration pathway. Models focusing on specific processes vary in complexity and
information requirements depending on assumptions made during model development and use.
Models are categorized as analytical, numerical, or a hybrid of the two. Some models are
analytical, in which the governing equation is solved directly or by means of a simplified
solution to the governing equation. Numerical models use techniques such as finite difference or
finite element methods to solve the governing equation. Different types of models may be used
in different phases of the RBCA process. Analytical models are typically used in simplistic
screening-level fate and transport analyses while more complex numerical models may be used
for:
Analyses for which more detailed output are needed or desired;
Analyses where analytical models do not or cannot yield acceptable output due to
conditions such as heterogeneity of environmental media; or
Analyses for which applicable analytical approaches are not available.
-------
Limits on available data and the resulting need for simplifying assumptions can result in
complex models reducing to the more simplistic models. Unless superseded by one of the above
or other considerations, analytical models are typically used in RBCA Tier 1 and Tier 2 analyses
while numerical models, if used at all, are limited to Tier 3 analyses.
Models can be described further as either steady-state or transient. Steady-state models do
not include a time domain and do not project variations over time. An example of a time-
independent input value is constant source-area concentrations of constituents. Transient models
incorporate a time domain, and model input and output values can vary over time. Transient
models can incorporate time-dependent input such as varying source-area concentrations and
groundwater recharge rates. Using the specific example of source-area concentrations, a steady-
state model incorporating constant concentrations may over-estimate constituent concentrations
at some times or locations in the model domain when compared to output from an analogous
transient model incorporating source-area concentrations that are decreasing due to migration of
the constituent mass or biodegradation/transformation.
2.2.1 Analytical Models
Analytical models use mathematical solutions to governing equations that are continuous in
space and time and applicable to the mass flow and constituent transport processes. They are
generally based on assumptions of uniform properties and regular geometry. Most analytical
models have a simple mathematical form and are based on multiple limiting assumptions rather
than on actual phenomena. A major advantage of analytical models is that such models are
relatively quick to setup and use (ASTM, 1995). Other advantages include:
Analytical models are easy to apply;
Analytical models can be solved for a set of input parameters and used to validate
other numerical codes;
Analytical models can accommodate some anisotropic medium properties;
Analytical models are numerically stable; and
Analytical models can be used as quick, conservative screening tools before using
more complex models.
Analytical models also can be used to quickly develop insight on how model output is
affected by ranges of values for input parameters. A limitation of analytical models is that, in
many cases, such models are so simplistic that important aspects of the environmental system
may be neglected (ASTM, 1995). Other limitations include:
Analytical models cannot accommodate heterogeneous medium properties (i.e.,
medium properties must be constant or uniform in space or time);
Analytical models may not be able to accommodate multiple sources contributing to
a single plume; and
Analytical models may not be able to accommodate irregular site boundaries.
In the matrices presented later in this Guidance, analytical model dimensions are described
as one-dimensional (ID), two-dimensional (2D), or three dimensional (3D) depending on the
number of directions in which model parameters can vary and for which output can be
generated. Forms of the governing equation are described as linear (7= A + B x^Q,
geometric (7= A + B x X*), exponential (Y=A+BxexorY=A+BnlnX)ora
transformation (e.g., Y= -A + B x erf Xwhere "erf is the error function transformation,
-------
which is a mathematical technique for linearizing the governing equation describing a free-
surface boundary condition such as a water table).
2.2.2 Numerical Models
Compared to analytical models, numerical models can accommodate more complex
heterogeneous systems with distributed, non-uniform properties and irregular geometry.
Advantages of numerical models include the ability to:
Simulate more complex physical systems;
Simulate multi-dimensional systems;
Incorporate complex boundary conditions;
Accommodate spatial variability of input parameters;
Accommodate both steady-state and transient conditions; and
Simulate both spatial and temporal distributions of model output.
Numerical models are, in comparison to analytical models, better suited to simulating
multiple combinations of spatially variable input parameters and boundary conditions for the
purpose of calibrating model output to measured site conditions.
Common limitations of numerical models include the:
Requirement of more development time compared to an analytical model of the
same transport process;
Requirement of greater amounts of input information; and
Possibility of numerical instability, which may cause the numerical model to become
difficult to implement without major modifications to the geometric layout of the
model domain.
Numerical models of constituent transport processes are solved using either finite difference
or finite element methods. In each method, the area to be modeled (the model domain) is
divided into sub-areas (i.e., discretized) and the governing differential equation is replaced by a
difference equation (Freeze and Cherry, 1979). In finite difference models, the model domain is
discretized into a finite number of blocks using an orthogonal grid and each block is assigned its
own properties. In the finite element method, the model domain is discretized using an irregular
triangular or quadrilateral mesh. This can result in a smaller nodal grid to model the area of
interest while accommodating irregular boundaries.
The properties can be different within each block (within limits) which allows for numerical
models to accommodate heterogeneous conditions. The difference equation is formulated with
increments of Ax, Ay, and Az for the spatial coordinates, and At for time. A solution is obtained
by solving the sets of difference equations for nodes along the rows or columns of the grid. A
model domain may comprise several hundred or thousands of nodes so that a large number of
equations must be solved simultaneously to obtain the output value at each block center (Fetter,
1980). Model output is calculated for the center of each block. Finite difference models can be
limited by their low accuracy for solving some fate and transport problems, and by the
requirement for a regular gridding of the model domain.
10
-------
2.3 Specific Model Information
This Guidance is a compendium of available, published fate and transport models that
address multiple pathways. Matrix 1 presents information regarding various fate and transport
models so key algorithms and parameters can be readily identified and directly compared.
Matrix 1 includes the following information:
Fate and Transport Pathway;
Name of Model/Algorithm;
Model Description/Process Simulation;
Type of Code/Algorithm;
Model Outputs;
Features/Characteristics;
Computer Needs;
Use Conditions/Technical Support;
References to Model Use; and.
Sources to obtain the Model/Algorithm.
Additional information on operation, input parameters, applicability, and sources of
additional information for the models are presented in an appendix to this Guidance.
11
-------
3.0 INFORMATION REQUIRED FOR SELECTION OF MODELS
ป 3.1 Site Conditions for Model Application
Different fate and transport models are applicable under different conditions relating to:
Properties of environmental media;
Sources and distributions of constituents in environmental media;
Physical pathways available for constituent migration;
Geometric constraints on constituent migration;
Temporal variance of fluid movement (i.e., steady-state or transient flow
conditions); and
Attenuation of constituents, or lack thereof, during transport.
Matrix 2 summarizes generic site conditions for application of various fate and transport
models. For each fate and transport pathway, candidate models are identified for specific site
conditions.
* 3.2 Input Parameters
Input parameters commonly needed for fate and transport modeling are summarized in
Matrix 3. The matrix indicates the typical parameter symbol and units, and comments on the
sensitivity of model output to the input parameter. Model output does not have the same degree
of sensitivity to each input parameter. Variation in certain input parameters will have a greater
affect on model output than other input parameters, depending on the fate and transport process
being modeled, assumptions incorporated into the conceptual development of the model, and the
equation or computer code used to implement the model. Input parameters are grouped by fate
and transport pathway in Matrix 3 and the generally more sensitive input parameters are
highlighted. Sensitivity of specific models to input parameters is indicated on the model
summaries in the appendix to this Guidance. The purpose of the matrix is to highlight sensitive
input parameters and not to provide a comprehensive compilation of every required input
parameter for the fate and transport model under consideration.
)
3.2.1 Sources of Input Parameter Values
Values for input parameters may be measured or obtained from published literature.
Published parameter values are generally based on direct measurements or on calculations made
using direct measurements. Repeating measurements for site or chemical-specific parameters is
often beyond the scope of the effort with which the modeling is associated, or is unjustified
given the defined modeling objectives. This can be the situation for chemical and physical
properties of constituents, and for some properties of the environmental media. Published values
of such input parameters can be evaluated for use at a particular site in lieu of data generated
from site-specific measurements (evaluation of the sensitivity of model output to values of input
parameters is discussed in a later section of this Guidance).
Data input requirements may be fulfilled using default or site-specific values that can be
obtained from published literature or established through measurement. Default values may be
selected from the model itself, the governing regulatory agency, or literature values. Literature
values for many input parameters are often presented as broad ranges, which can confound the
12
-------
selection of a specific value (e.g., values of hydraulic conductivity are generally given in order-
of-magnitude ranges). The candidate fate and transport model may be sensitive to the value
given the input parameter in which case data from direct measurements should be considered for
use with the model. Use of a complex model to simulate site-specific conditions can increase the
need for direct measurement of input parameter values. Often, numerical fate and transport
models cannot be adequately calibrated for their intended use without data from direct
measurements.
3.2.2 Techniques for Measuring Input Parameters
When the need for fate and transport modeling is anticipated, consideration should be given
to the techniques and methods for measuring the physical and chemical properties of
environmental media that may be required as model inputs. Values for input parameters can be
obtained from laboratory measurements made on samples collected from the site, or from direct
measurements made at the site. The input parameters listed in Matrix 3 include those that can be
measured in either the field or the laboratory. By identifying required, sensitive, or influential
input parameters, and planning for their measurement during the assessment of the nature and
extent of constituents, the efficiency of site-specific data collection efforts can be increased and
costs associated with multiple data-collection efforts can be minimized or eliminated.
Methods typically used for collection of soil samples for chemical analyses are generally not
adequate for obtaining samples for geotechnical analyses. The former samples are usually
disturbed during collection while geotechnical samples should be undisturbed to produce
representative values of parameters such as bulk density, total porosity, and natural moisture
content. Undisturbed samples can be collected using thin-walled sampling devices (i.e., Shelby
tubes) advanced using standard subsurface drilling and soil sampling equipment, or from bulk
undisturbed samples collected from excavations. Undisturbed samples should be preserved in
the field to retain their structural integrity and moisture content (e.g., by sealing the sample in
wax) and later submitted to a geotechnical laboratory for analyses.
Grain size distribution can be measured using either undisturbed or disturbed samples (e.g.,
split-spoon samples). Sieve analyses of samples will define the distribution of gravels and sands,
and will indicate the total percentage of silts and clays (i.e., percent passing the #200 sieve), and
hydrometer analyses of samples can be used to determine the distribution of silts and clays.
Grain size distribution curves can be used as an indicator for many other input parameters, which
cannot otherwise be, measured easily (e.g., intrinsic permeability, and thickness of capillary
fringe).
The fraction of organic carbon (foc) in soil is an important input parameter for fate and
transport modeling organic constituents, as it is needed to calculate the soil sorption coefficient.
Fraction of organic carbon can be measured on samples collected specifically for this purpose, or
on samples collected for analysis of constituent concentrations or geotechnical properties.
Measurements can be made on samples collected from contaminated.areas of a site or from areas
where constituents are absent. Where possible, it is best to make measurements on samples
collected from the lithologic zone(s) incorporated in the model. There are many procedures
available for measurement of f^in soil. Users of fate and transport models should assure that foc
measurements are expressed in the form (i.e., units) required for the model being used.
13
-------
It is best to obtain site-specific data for some of the input parameters in Matrix 3. These
parameters may include:
Soil properties such as grain size distribution, bulk density, total porosity, and
natural moisture content (for calculation of volumetric water- and air-content);
Infiltration rate for the soil-to-groundwater pathway, which can be measured using
lysimeters or double-ring infiltrometers;
Saturated hydraulic conductivity for the soil-to-groundwater and groundwater
transport pathways, which can be measured using single-well slug tests, pumping
tests of single wells, or aquifer tests incorporating pumping and observation wells;
and
Hydraulic gradient for the soil-to-groundwater and groundwater transport pathways,
which can be measured from contours of groundwater elevations (i.e., potentiometric
surface contours) generated from concurrent water level measurements in a
distributed set of wells and piezometers.
Chemical-specific properties such as carbon-water sorption coefficient (K^, also called the
organic carbon partition coefficient) and biodegradation rates can be determined from laboratory
experiments conducted on site-specific samples. However, values for these input parameters are
often obtained from literature. The modeling objectives and sensitivity of model output may,
however, justify the cost of such laboratory measurements. Similarly, dispersivity values can be
obtained from in-field tracer testing of water-bearing units, but such testing is also often beyond
the scope of the modeling effort. Values of diffusivity used in modeling of vapor migration are
typically default values based on soil type.
Care must be taken when adopting literature values for use alone, or in combination with
site-specific measured values, as model input parameters. The usefulness of many input
parameters may depend on site characteristics not well documented in the literature, which can
make it difficult to evaluate the appropriateness of the parameter value for use in the chosen fate
and transport model. Measurement of certain indicator parameters (e.g., grain size distribution)
can be performed to provide a basis for selection of appropriate literature values for input
parameters that would be impractical or expensive to measure directly.
Many input parameters to fate and transport models are related to spatial and geometric
factors such as source width, area of enclosed building, area of floor cracks, thickness of affected
soil zone, thickness of vadose zone, saturated thickness of water-bearing unit, and distance along
a flow-path from the downgradient edge of a plume. Values for these case-specific geometric
input parameters can be estimated based on local or regional maps and cross-sections available
prior to collection of site-specific data, from measurements made by on-site personnel, or from
maps and cross-sections generated as part of data collection efforts.
Data quality and quantity requirements should be linked to modeling objectives, the
complexity of the selected model, and the RBCA tier-level requirements. In Tier 1 and Tier 2
analyses, for example, conservative default values can be used to characterize a range of
potential site conditions. As conservative default values are replaced by measured values in
higher tier analyses, more site-specific data may be required to produce the desired quality of
model output, particularly if model output is sensitive to input parameter values. Design of
sampling programs to collect site-specific data should balance modeling objectives and model
output sensitivity to the cost of data collection.
14
-------
This Guidance is not intended to provide detailed information on measurement of input
parameters. Such information is available in the broad-based published literature. However, key
references from this literature on the measurement of input parameters are cited together in a
separate section of the Bibliography of the Guidance.
3.2.3 Sensitivity of Model Output to Input Parameters
Sensitivity testing is the process of determining the degree to which output of a fate and
transport model changes as values of input parameters are changed. Sensitivity testing can:
Identify the fate and transport process(es) with the greatest influence on model
output;
Quantify change in the model output caused by uncertainty and variability in the
values of input parameters; and
Identify the input parameters that have the most influence on model output and
overall model behavior (ASTM, 1995).
A model is considered to be sensitive to an input parameter if model output changes notably
when the value of the input parameter is changed only slightly. Sensitivity of a fate and
transport model to input parameter values depends on the governing equation of the model, the
form of the solution to the governing equation and simplifications made in the model to allow
solution of the governing equation.
Many input parameters used in fate and transport models are best characterized as ranges of
reasonable values. Published values of input parameters are often given as ranges, and field
measurements often produce a range of reasonable values. A procedure for using sensitivity
analyses to determine how model output varies as the range of parameter values are used is:
Identify input parameters for which a range of reasonable values exists.
Conduct model runs varying the value of the target input parameter while holding
other values of other input parameters constant.
The number of model runs needed to determine sensitivity of an input parameter will
depend on how the parameter is incorporated into the solution of the governing
equation. Fewer model runs are needed if the input parameter is used in a linear
form than if it is used as an exponent, raised to a power, used as a logarithm, or
incorporated into a functional transformation.
Compare model runs incorporating uncertainty and variability of the various input
parameters and identify the most and least sensitive input parameters for the model
algorithm.
If model output is not or only slightly sensitive to the range of reasonable values used for an
input parameter, there is generally little or no need for additional effort to better define the value.
On the other hand, if model output is highly sensitive to an input parameter for which an
assumed or default value has been used, consideration should be given to:
Using a model which is less sensitive to the input parameter;
Using a model that has greater flexibility (and therefore is probably more complex)
and thereby allows manipulation of boundary conditions or other input parameters to
compensate for sensitivity to the input parameter;
Obtaining more relevant values of the input parameter from literature; or
15
-------
Making field or laboratory measurements of the input parameter.
Analyses of model sensitivity to values of input parameters can sometimes be used to select
parameter values. This process is sometimes referred to as parametric analysis. A determination
of the sensitivity of model output to a reasonable range of input parameter values is derived. If
model output is not sensitive to the input parameter value (or if model output falls within a
reasonably expected range), a value for the input parameter can be selected from the range of
values used in the sensitivity analysis. For example, if constituent concentration at a
downgradient location is not sensitive to a reasonable range of decay constants, but is sensitive
to a reasonable range of aquifer hydraulic conductivities, a value for decay constant can be
selected from the tested range while additional measurements or analyses may be needed to
select an appropriate hydraulic conductivity value. Sensitivity analysis operates on the
assumption that input parameters are mutually independent. However, some parameter are
correlated to some degree (e.g., effective porosity and hydraulic conductivity) Therefore care
must be taken when conducting parametric analyses to assure that the model has been calibrated
and validated by means of comparisons to input parameters other than the one(s) for which
parametric analyses are being conducted.
16
-------
4,0 MODEL SELECTION
4.1 Model Selection Criteria
Criteria for selection of an appropriate fate and transport model include:
Type of information required from the model (e.g., screening versus detailed
evaluation);
The fate and transport pathway to be modeled;
Complexity of available models;
Required input parameters;
Availability of data on input parameter values;
Model output requirements;
Limitations on model use and output; and
The user's and target audiences' familiarity and comfort with the model.
Criteria for selecting a fate and transport model are illustrated in the process diagrams
presented as Figures 1, 2, and 3. The issue of model complexity is addressed in Figure 1 where
the selection of analytical versus numerical models is illustrated. Figure 2 illustrates the criteria
for selecting an analytical model for a particular fate and transport pathway, given input data and
model output requirements. In a similar manner, Figure 3 illustrates criteria for selecting a
numerical model. Information on the principal limitations of each model is presented in Matrix
1 and in the model summaries in the appendix to this Guidance. Regulatory agencies often
prefer particular models based on familiarity, output formats, and ease of use. These preferences
should be considered when selecting a fate and transport model.
Selection of an appropriate model can be an iterative process, involving use of more than one
model to achieve the desired results. For example, previous modeling results may support
switching to another model to satisfy needs for more detailed output or output which is less
sensitive to input parameters. In some cases, site-specific values for key input parameters may
not be available, forcing the user to rely on default values for the input parameters. The default
values for a particular model may not be a good match for the site or constituents, which may
cause modeling results to be less representative than desired for making necessary decisions.
4.2 Selection of Models for Tier 2 and Tier 3 RBCA Evaluations
4.2.1 Tier 2 Usage
Migration and/or transformation of constituents in Tier 2 usage of the RBCA process is
typically predicted using one or a combination of relatively simple analytical fate and transport
models. Use of analytical models requires the acceptance of .simplifying assumptions regarding
material properties and migration processes. The models attempt to capture the operative
physical and chemical phenomena relevant to the fate and transport process. Unlike the Look-
Up Tables generated for Tier 1 usage in the RBCA process, analytical models used in Tier 2 can
be tailored to reflect site-specific conditions. The ability to simulate fate and transport processes
in a cost-effective manner makes analytical modeling a good middle-ground between the Tier 1
Look-Up Tables and the complex numerical modeling typically conducted for Tier 3 usage.
17
-------
The decision for tier upgrade, or for the use of complex rather than simple models, can be
predicated on several factors, including:
How well the site conditions are accommodated by the conceptual basis of the
selected model;
The potential differences between the current-tier cleanup targets and the cleanup
targets likely to be associated with the higher-tier analyses;
The cost for collection of additional site-specific data; and
The acceptability and reasonableness of corrective action alternatives suggested by
lower-tier analyses.
Use of analytical models can result in predicted constituent concentrations that are greater
than those that will actually occur. This over-estimation of constituent concentrations (i.e.,
conservative predictions of constituent migration) is an important consideration in the selection
of fate and transport models in the RBCA process (ASTM, 1995). Evaluations based on
conservative predictions can preclude the need to collect additional site-specific data in
situations where conservatively predicted constituent concentrations do not exceed acceptable
levels. This may not be true, however, for all situations and model applications, and the model
selection process should consider this possibility.
Data collection for fate and transport models in Tier 2 usage is typically limited to
economically or easily obtained site-specific data, or to easily estimated quantities. Most of the
data collected for Tier 2 usage are related to geometric descriptions of the model area, physical
properties of the environmental media through which migration is occurring, potential gradients
causing advective movement of fluids, and constituent concentrations in source areas. When
selecting a fate and transport model for Tier 2 usage, availability of values for key and sensitive
input parameters should be considered. In general, the fewer the measured data available for
input parameters, the simpler should be the fate and transport model selected for Tier 2 usage.
By the same token, if the scope of the effort associated with the fate and transport modeling is
limited to collection of only limited data for input parameters, simpler models should be selected
for Tier 2 usage.
Input parameters for which measurement data have not been generated are given assumed or
default values in Tier 2 usage of fate and transport models. Default values are typically used for
chemical and physical properties of constituents and some properties of the environmental
medium. Assumed values can usually be based on reasonable application of published data, or
can be obtained from regulatory agencies. Fate and transport models selected for Tier 2 usage
should incorporate assumed and default values which are reasonably appropriate to the site to be
modeled, and which are consistent with regulatory requirements for modeling (if any). Default
values determined to be unrepresentative can be measured.
Uncertainties associated with Tier 2 usage of fate and transport models result from:
Simplification of site geometry;
Simplification of physical properties of environmental media through which
migration occurs (e.g., homogeneity);
Inaccurate definition of site geology and hydrogeology;
Simplification of potential gradients causing fluid movement;
Inability to incorporate time-dependent values of input parameters such as source-
area constituent concentrations;
18
-------
Potential inability to predict time-dependent constituent concentrations;
Use of assumed or default values for many input parameters; and
Use of simplified representations of some of the fate and transport mechanisms
incorporated in the model.
The conservative nature of many fate and transport models associated with Tier 2 usage
compensates to varying degrees for uncertainties in the modeling process. However, care must
be taken to select fate and transport models that will, in fact, result in conservative predictions of
constituent concentrations given the availability of data on input parameters.
4.2.2 Tier 3 Usage
Fate and transport modeling in Tier 3 usage may involve use of numerical models which can
accommodate time-dependent constituent migration under conditions of spatially-varying
properties of the environmental media through which migration is occurring. Tier 3 usage does
not always involve use of numerical models. To meet modeling objectives, a higher-tier analysis
may only require use of more sophisticated analytical models or use of the lower-tier models
with additional site-specific values for input parameters. However, numerical models are not
commonly used for Tier 1 or Tier 2 analysis.
Tier 3 evaluations commonly involve collection of additional site information and
completion of more extensive fate and transport model development and verification than for
Tier 2 usage. In certain situations, successful use of complex fate and transport models in Tier 3
usage may require field and laboratory measurement of many of the default input parameters, or
of input parameters for which values are assumed in the simpler Tier 2 analytical models.
Data collection objectives for numerical fate and transport models in Tier 3 usage include the
data required for Tier 2 usage of analytical models plus additional information on boundary and
initial conditions. Data collected for Tier 3 usage include geometric descriptions of the model
domain and physical properties of the environmental media through which constituent migration
is occurring. The models will generate potential gradients driving advective movement of fluids.
Data objectives for Tier 3 solute transport models include source-area concentrations of
constituents, the initial distribution of dissolved constituents throughout the model domain, and
constituent loading to environmental media in the source area. Data objectives for Tier 3 usage
of fate and transport models should include measurement of constituent concentrations for use in
model calibration.
Fate and transport models for Tier 3 usage can incorporate the same assumptions and
defaults used in Tier 2 usage. However, the value and usefulness of simulations generated using
the complex numerical models typical of Tier 3 usage can be eroded if many assumed and
default values are used as input parameters. However, as with Tier 2 usage, assumed and default
values are still typically used for input parameters associated with chemical and physical
properties of constituents.
Uncertainties associated with Tier 3 usage of fate and transport models can be the same as
those associated with Tier 2 usage of models. The degree of uncertainty depends on the
complexity of the numerical model grid and the assumptions and default values used for input
parameters. The complex methods used to solve governing differential equations in Tier 3
usage, and the ability to adjust boundary and initial conditions, provides a greater ability to
19
-------
calibrate models to measured site conditions than models typical of Tier 2 usage, thus reducing
some of the uncertainty associated with model output.
4.3 Model Packages
Packages of fate and transport models have been developed to incorporate models for a
variety of different pathways and to link model outputs and inputs. References to specific model
packages are cited in a separate portion of the Bibliography section of this Guidance. Use of a
modeling package can decrease the time and cost of performing a model evaluation, assure a
uniform approach to modeling fate and transport processes at a variety of sites, and standardize
data input and model output formats to simplify training on model usage and review of model
output.
Important technical considerations in selection of a model package(s) are:
The algorithm(s) used to model each fate and transport pathway, and the inherent
limitations on applicability of each model;
Degree of documentation, validation, and general acceptance of algorithms
incorporated in the package;
Ability to access and modify data fields for input parameters (i.e., are input values
"hard-wired" from databases of default values or can individual input parameters be
tailored to site-specific conditions);
How the model results or output from individual fate and transport models are
reported and linked to other model components; and
Familiarity of the user with various risk assessment components (i.e., model
packages are not intended to be expert systems for use by those with little or no risk
assessment expertise).
Each model package will have some level of documentation describing fate and transport
algorithms, required formats for data input, model output options, hardware and supporting
software requirements (e.g., spreadsheet software external to the model package), installation
instructions, and troubleshooting aids. Model packages can embed fate and transport models to
estimate cross-media transfer or migration of constituents (i.e., transport of constituents from one
environmental medium to another, such as from soil-to-ambient air or soil-to-groundwater) and
to calculate target cleanup levels for the various media.
Packages may allow both "forward" calculations (i.e., calculations to assess potential
adverse impacts associated with user-specified constituent concentrations) and "backward"
calculations (i.e., calculations of cleanup levels corresponding to acceptable risk targets for
limiting potential adverse impacts), incorporate Monte Carlo simulation capabilities to quantify
uncertainties in input parameters, a chemical database, tools for statistical analyses of site data,
and an option to consider additive risk due to multiple pathways and constituents.
Model packages can include relatively simple analytical fate and transport models for
predicting constituent concentrations incorporated into a spreadsheet workbook. Spreadsheet
frameworks may consist of a group of spreadsheets integrated by a macro interface. The
spreadsheets can be used to calculate baseline risks and soil and groundwater cleanup standards
(i.e., "forward" and "backward" calculations, respectively) for each constituent of concern.
Input parameters and calculated results generated by the package can be contained within linked
worksheets that can be saved, viewed on-screen, or selectively printed.
20
-------
Model packages can generate pathway-specific attenuation factors corresponding to either
cross-media (migration of constituents from one environmental medium to another) or lateral
migration of constituents. Examples of cross-media attenuation factors are:
Surface Volatilization Factor
Particulate Emission Factor
Subsurface Volatilization Factor
Soil-to-Enclosed Space Volatilization Factor
Groundwater Volatilization Factor
Groundwater-to-Enclosed Space Volatilization Factor
Soil-to-Leachate Partition Factor
Leachate-Groundwater Dilution Factor
Lateral transport factors apply to constituent migration within air or groundwater where
concentrations are diminished due to mixing and attenuation effects. Examples of such
attenuation factors are:
Lateral Air Dispersion Factor
Lateral Groundwater Dilution-Attenuation Factor
Model packages can include modules linked in an integrated exposure/risk assessment
framework. The modules can include:
Development of a conceptual model of the site;
Fate and transport models to simulate movement of constituents from sources to
receptors;
A module which uses internally-calculated exposure-point concentrations or user-
entered concentrations to estimate chemical intake; and
Presentation of estimated chemical intake, carcinogenic risk, and hazard indices in
tabular and graphical formats.
4.4 Model Calibration and Validation
4.4.1 Calibration
Model calibration is the process of adjusting the model geometry or input parameter values
so that the model output matches observed conditions at a site. In developing a strategy for
model calibration, decisions are needed on whether calibration is to be steady-state, transient, or
both; what data are to be matched to achieve calibration; and what input parameter value(s) or
boundary condition(s) are be adjusted to achieve calibration. Examples of model calibration
include:
Adjustment of source area constituent concentrations or average linear velocity of
groundwater movement so that predicted concentrations at locations downgradient
of the source area better match measured concentrations.
Adjustment of volumetric water and air contents in vadose zone soil so that
predicted migration of vapors from subsurface soil to ambient air better matches
measurements of constituent concentrations in air at the ground surface above the
source area.
21
-------
Adjustment of hydraulic head or flow at boundaries of a numerical groundwater flow
model so that the hydraulic heads simulated by the model better match
potentiometric surface contours generated from groundwater elevations calculated
from well measurements.
Model calibration is typically accomplished through trial-and-error adjustment of the input
parameter values. Calibration of a model is most often evaluated through analysis of residuals,
which are the differences between the predictive model output and measurements of actual
conditions (ASTM, 1995). Knowledge of the model algorithm used to solve the governing
equation and knowledge of model sensitivity to various input parameters can reduce the amount
of trial-and-error adjustments needed to calibrate a model. The calibration process should
continue until the degree of correspondence between model output and actual conditions is
consistent with objectives of the modeling effort (ASTM, 1995).
The degree of model calibration required can depend on how model output will be used in
the overall RBCA process. If, for example, fate and transport modeling is being used to predict
constituent concentrations at a critical water supply or in indoor air of an occupied building, a
greater degree of calibration may be needed than if the model is used to predict downgradient
movement of dissolved constituents in groundwater not used for potable supplies or to predict
vapor migration to ambient air at an unoccupied site. However, even conservative models may
require some type of calibration for certain applications. Determination of the degree of model
calibration should consider stakeholder concerns and should involve consultation with over-
seeing regulatory agencies. The degree of model calibration can be determined during
development of the conceptual site model.
Calibration of a model to a single set of field measurements does not guarantee a unique
solution of the model algorithm (ASTM, 1995). Uniqueness of model solutions can be tested by
running the model using different input parameter values or boundary conditions than those used
to generate the desired output and comparing the model output to a separate set of independent
calculations or field measurements. If the initial model runs can be calibrated, but output from
subsequent model runs does not adequately match the corresponding calculations or
measurements, additional model calibration or definition of input parameter values may be
warranted.
ซ 4.4.2 Validation
Validation is the process of determining how well the fate and transport model describes
actual system behavior (ASTM, 1995). Validation of the model can be achieved by matching
model output to measurements (Wang and Anderson, 1982). It involves the process of using a
set of input parameter values and boundary conditions for a calibrated model to approximate,
within an acceptable range, an independent set of measurements made under conditions similar
to the model conditions (ASTM, 1995). A calibrated but unverified model may be used to model
fate and transport of constituents if sensitivity analyses indicate that model output is not sensitive
to variability in the portions of the model which cannot be verified (ASTM, 1995).
An analytical model run using a computer spreadsheet can be validated by comparing model
output to independent calculations (e.g., calculations generated using a different "reference"
model or by "pencil-and-paper" calculations) of the output values. Numerical models used to
predict spatial and temporal changes in dissolved constituent concentrations can be validated by
22
-------
determining concentrations of dissolved constituents at locations where initial concentrations are
not known, and by time-series sampling at locations where initial conditions are known.
Care must be taken to ensure that the number of independent calculations or field
measurements is sufficient to effectively validate the model. The number and extent of
calculations and measurements needed to validate a model can increase as the complexity of the
model algorithm increases. If an analytical model is composed of a combination of independent
equations, several independent calculations may be needed to validate a single model output.
The Domenico (1987) model incorporates an average linear velocity of groundwater movement
calculated from hydraulic conductivity, hydraulic gradient and effective porosity. Validation of
output from a Domenico model can therefore require independent calculation of both the
groundwater velocity, using the appropriate linear equation, and calculation of the downgradient
constituent concentration using the error function transformation of the advection-dispersion
equation.
4.4.3 Modeling versus Field Data
There is always the possibility that a model cannot be calibrated to field measurements. For
example, assigning source area concentrations that match present conditions, but do not match
previous conditions, may result in the inability to calibrate a modeled groundwater plume of
dissolved constituents that formed under past constituent loading conditions. This may occur in
a model that does not allow for time-variation of source area concentrations, and may limit the
predictive capabilities of the model. If a model process and algorithm are not representative of
site conditions, it may not be possible to calibrate the model even when measured values for
input parameters are use. This could occur when a steady-state model is used to simulate
transient fate and transport processes, or when a model used to simulate fate and transport of
degradable constituents does not incorporate biodegradation or transformation.
When a selected model can not be calibrated sufficiently to meet modeling objectives,
consideration should be given to using field data in lieu of modeling. Overseeing regulatory
agencies often prefer field data to simulations generated using models that cannot be adequately
calibrated. Collecting field data on constituent concentrations may, in fact, be less expensive
than collecting the data on sensitive input parameters needed to calibrate a model, or than using a
more complex model requiring greater user skill and operation time. Where field information is
adequate, such as where spatial measurements define the full extent of contamination and time-
series measurements indicate decreasing constituent concentrations, fate and transport modeling
of any sort, whether or not it can be calibrated, may not be necessary to implement the RBCA
process.
23
-------
. 5.0 DEFINITION OF TERMS
Anisotropic Conditions: Exhibiting properties with different values when measured along
axes in all directions; opposite of isotropic.
Boundary Conditions: The physical or chemical conditions at the boundary of the area to
be modeled. Boundary conditions must be defined, but are often assumed, to allow for
mathematical solution of governing differential equations.
Capillary Zone: Region in a solid environmental medium in which water is held by
capillary tension at pressure heads less than one atmosphere. The zone may be saturated and
referred to as the tension-saturated zone.
Computer or Source Code: The computer program or software used to run a fate and
transport model.
Deterministic Risk Characterization: The process of determining risk by use of
established, single-valued exposure parameters and direct calculation of constituent
concentrations at point(s) of exposure to human or environmental receptors.
Dispersivity: Characteristic property of a porous environmental medium quantifying the
process of dispersion.
Effective Porosity: The porosity of the environmental medium through which groundwater
movement occurs (i.e., does not include porosity containing water which does not move with
groundwater flow).
Environmental Media: Soil, soil vapor, soil pore water, groundwater, leachate, surface
water, indoor air, or the ambient atmosphere which may be a source of constituents, or which
may be a pathway(s) for migration of constituents from the source to the point of exposure to
human or environmental receptors.
Evapotranspiration: A combination of evaporation from open bodies of water, evaporation
from soil surfaces, and transpiration from the soil by plants.
Heterogeneous Conditions: Properties are not the same at each location in an
environmental medium; opposite of homogenous.
Homogenous Conditions: Properties are the same at each location in an environmental
medium; opposite of heterogeneous.
Hydraulic Conductivity: A physical property measuring the ability of groundwater to
move through an environmental medium under a unit hydraulic head.
Hydraulic Gradient: The maximum slope of the water table or potentiometric surface.
Immiscible Liquids: Liquids which to not readily mix at standard temperature and
pressure.
Isoconcentration Contours: Contours of equal concentrations of constituents in
environmental media (analogous to topographic elevation contours).
Isotropic Conditions: Exhibiting properties with the same values when measured along
axes in all directions; opposite of anisotropic.
24
-------
Leaching: The process whereby constituents in soil are transferred to water infiltrating
through the vadose zone.
Model Algorithm: A procedure for solving a mathematical problem (e.g., an equation) in a
fate and transport model.
Model Domain: The area to be modeled.
Natural Attenuation: The combination of naturally occurring physical and chemical
processes causing concentrations of constituents in environmental media to decrease over
time.
Organic Carbon Partitioning Coefficient: A chemical-specific property related to the
distribution of constituents between solid and liquid environmental media under equilibrium
conditions.
Orthogonal Model Coordinates: Coordinate axes each of which are perpendicular to the
other axes (e.g., X, Y, and Z axes of Cartesian coordinates).
Probabilistic Risk Characterization: The process of characterizing risk by statistical
evaluation, using Monte Carlo or similar analyses, of exposure parameters and constituent
concentrations at the points of exposure to human and environmental receptors.
Risk Assessment: Risk assessment is the systematic, scientific characterization of potential
adverse effects of exposure of human or environmental receptors to hazardous agents or
activities.
Risk-Based Corrective Action: Risk-based corrective action (RBCA) is incorporation of
risk-based decision making into the underground storage tank corrective action process. It is
typically a tiered decision-making process for the assessment and response to a release of
constituents, based on the protection of human health and the environment.
Risk-Based Decision Making: A process that utilizes risk and exposure methodology to
help implementing agencies make determinations about the extent and urgency of corrective
action and about the scope and intensity of their oversight of corrective action by UST
owner/operators. The process is flexible to allow for varying implementation concerns of the
implementing program.
Steady-State Conditions: Conditions when the magnitude and direction of groundwater
movement at any point in a flow field are constant with time.
Transient Conditions: Conditions when the magnitude and direction of groundwater
movement at any point in a flow field change with time.
Vadose Zone: Thezoneof unsaturated soil above the water table.
Water Table: The level to which groundwater will rise in a well open to the atmosphere.
25
-------
6.0 BIBLIOGRAPHY
1. American Petroleum Institute (API), 1994: Transport and Fate of non-BTEXPetroleum
Chemicals in Soil and Groundwater, API Publication No. 4593, Washington DC.
2. American Petroleum Institute (API), 1996: Estimation of Infiltration and Recharge for
Environmental Site Assessments, API Publication No. 4643, Washington DC.
3. American Society of Civil Engineers (ASCE), 1995: Groundwater Quality: Guideline for
Selection of Commonly Used Groundwater Models, New York, NY.
4. American Society for Testing and Materials (ASTM), 1993: Guide for Application of
Groundwater Flow Models to a Site Specific Problem, West Conshohocken PA.
5. American Society for Testing and Materials (ASTM), 1995: Standard Guide for Risk-Based
Corrective Action Applied at Petroleum Release Sites, E 1739-95, West Conshohocken PA.
6. Anderson, M.P., and W.W. Woessner, 1992: Applied Groundwater Modeling, Academic
Press, Inc., San Diego CA.
7. Association of American Railroads, 1996: Risk-Based Management of Diesel-Contaminated
Soil, Publication No. R-897, Washington DC.
8. Bauer, P., 1998: Disperse: Advection/Dispersion Model for MTBE and TBA, New Jersey
Department of Environmental Protection.
9. Bear, J., 1972: Dynamics of Fluids in Porous Media, Dover Publications, Inc., Mineola NY.
10. Bedient, P.B., et al., 1995: Groundwater Contamination Transport and Remediation,
Prentice-Hall, Inc., Englewood Cliffs NJ.
11. Bonazountas, M., and J.M. Wagner, 1984: SESOIL: A Seasonal Soil Compartment Model;
prepared for the US EPA. Arthur D. Little, Inc., Cambridge MA.
12. Campbell, G.S., 1974: "A Simple Method for Determining Unsaturated Hydraulic
Conductivity from Moisture Retention Data", Soil Science, Vol. 117, pp. 311-314.
13. Clapp, R.B., and G.M. Hornbeyer, 1978: "Empirical Equations for Some Hydraulic
Properties", Water Resources Research, Vol. 14, pp. 601-604.
14. Cline, P.V., et al., 1991: "Partitioning of Aromatic Constituents Into Water from Gasoline
and other Complex Solvent and Mixtures", Environmental Science and Technology, Vol. 25,
pp. 914-917.
15. Connor, J.A., et al., 1996: Soil Attenuation Model for Derivation of Risk-Based Soil
Remediation Standards, Groundwater Services, Inc., Houston TX.
16. Crum, J.A., 1997: Generic Groundwater and Soil Volatilization to Indoor Air Inhalation
Criteria: Technical Support Document, Michigan Department of Environmental Quality,
Environmental Response Division.
17. Domenico, P.A., 1972: Concepts and Models in Groundwater Hydrology, McGraw-Hill,
New York NY.
18. Domenico, P.A., 1987: "An Analytical Model for Multidimensional Transport of a Decaying
Contaminant Species", Journal of Hydrology, Vol. 9, pp. 49-58.
26
-------
19. Domenico, P.A. and F.W. Schwartz, 1990: Physical and Chemical Hydrogeology, John
Wiley & Sons, New York NY.
20. Eagleson, P.S., 1978: "Climate, Soil and Vegetation", Water Resources Research, Vol. 14,
pp. 705-776.
21. Electric Power Research Institute (EPRI), 1995: Mineral Insulating Oil Characterization
Report, prepared by CH2M Hill and META Environmental.
22. EPA, 1991: MOFAT: A Two-Dimensional Finite Element Program for Multiphase Flow and
Multicomponent Transport, U.S. EPA Robert S. Kerr Environmental Research Laboratory,
Ada OK, EPA/600/2-91/020, PB91-191692.
23. EPA, 1993: MUL TIMED: The Multimedia Exposure Assessment Model for Evaluating the
Land Disposal of WastesModel Theory: Project Summary, U.S. EPA Environmental
Research Laboratory, Athens GA, EP1.89/2:600/SR-93/081, PB93-1048-M.
24. EPA, 1989: Risk Assessment Guidance for Superfund, Vol. 1: Human Health Evaluation
Manual, Part A, EPA/540/1-89/002.
25. EPA, 1994: Assessment Frameworkfor Ground-Water Model Applications, Office of
Emergency and Remedial Response, Washington DC, EPA 500-B-94-003.
26. EPA, 1994: Ground-Water Modeling Compendium-Second Edition, Office of Emergency
and Remedial Response, Washington DC, EPA 500-B-94-004.
27. EPA, 1996: Soil Screening Guidance: Technical Background Document, Office of
Emergency and Remedial Response, Washington DC, EPA/540/R-95/128, PB96-963502.
28. EPA, 1996: Soil Screening Guidance: Users Guide, Office of Emergency and Remedial
Response, Washington DC, EPA/540/R-96/018, PB96-963505.
29. EPA, 1995: SCREEN3 Model User's Guide, Office of Air Quality Planning and Standards,
Emissions, Monitoring, and Analysis Division, Research Triangle ParkNC, EP4.8: SCR2/2,
PB97-0383-M.
30. EPA, 1998: Superfund Exposure Assessment Manual, Office of Emergency and Remedial
Response, Washington DC, OSWER Directive 9285.5-1.
31. EPA, 1992: User's Guide for the Industrial Source Complex (ISC2) Dispersion Models,
Office of Air Quality Planning and Standards, Technical Support Division, Research
Triangle ParkNC, EP4.8: IN2/V.2, PB93-0858-M.
32. Farmer, W.J., et al., 1980: "Hexachlorobenzene: Its Vapor Pressure and Vapor Phase
Diffusion in Soil", Soil Science Society of America Journal, Vol. 44, pp. 445-450.
33. Fetter, C.W., 1988: Applied Hydrogeology, 2nd Edition, Merrill Publishing, Columbus, OH.
34. Freeze, A. R. and J.A. Cherry., 1979: Groundwater, Prentice-Hall, Inc.-, Englewood Cliffs NJ.
35. Groundwater Services, Inc., Tier 2 RBCA Guidance Manual for Risk-Based Corrective
Action, Houston TX.
36. Gustafson, J., et al., 1996: "Selection of Representative TPH Fractions Based on Fate and
Transport Considerations", Total Petroleum Hydrocarbon Criteria Working Group, Volume
3.
37. Hayden, A.J., et al., 1992: "Prediction of Leachate Concentrations in Petroleum-
Contaminated Soils", Journal of Soil Contamination, Vol. 1, pp. 81-93.
27
-------
38. Hemond, H.F., and EJ. Fechner, 1994: Chemical Fate and Transport in the Environment,
Academic Press, New York NY.
39. Howard, P.H., et al, 1991: Handbook of Environmental Degradation Rates, Lewis Publishers
Inc., Chelsea MI.
40. Johnson, P. C. and R.A. Ettinger, 1991: "Heuristic Model for Predicting the Intrusion Rate of
Contaminant Vapors into Buildings", Environmental Science and Technology, Vol. 25, No.8,
pp. 1445-1452.
41. Jury, W.A., et al., 1990: "Evaluation of Volatilization by Organic Chemicals Residing Below
the Soil Surface", Water Resource Research, Vol. 26, pp. 13-20.
42. Konikow, L.F., et al., 1994: User's Guide to RevisedMethod-of-Characteristics Solute-
Transport Model (MOC-Version 3.1), U.S. Geological Survey Earth Science Information
Center, Denver CO.
43. Liptak, J.F. andG. Lombard, 1996: "The Development of Chemical-Specific, Risk-Based
Soil Cleanup Guidelines Results in Timely and Cost-Effective Remediation", Journal of
Soil Contamination, Vol. 5, pp. 83-94.
44. Lyman, W.J., et al., 1982: Handbook of Chemical Property Estimation Methods, McGraw
Hill, New York NY.
45. Lyman, WJ. et al., "Organic Contaminants in Subsurface Environments ", Mobility and
Degradation of Organic Contaminants in Subsurface Environments, Camp Dresser and
McKee, Inc., Chelsea MI.
46. Mercer, J.W. and R.M. Cohen, 1990: "A Review of Immiscible Fluids in the Subsurface:
Properties, Models, Characterization and Remediation," Journal of Contaminant
Hydrogeology, Vol. 6, 1990, pp. 107-163.
47. Nevin, J.P., 1997: "FATES: A Natural Attenuation Calibration Tool for Groundwater Fate
and Transport Modeling", 1996 Petroleum Hydrocarbons and Organic Chemicals in Ground
Water: Prevention, Detection, and Remediation Conference, National Ground Water
Association Catalog #T539.
48. Odencrantz, J.E., et al, 1990: "Transport Model Sensitivity for Soil Cleanup Level
Determination Using SESOIL and AT 123D in the Context of the California Leaking
Underground Fuel Tank Field Manual", Proceedings Fifth Annual Conference on
Hydrocarbon Contaminated Soils, AmherstMA.
49. Odencrantz, J.E., et al., 1991: "A Better Approach to Soil Cleanup Level Determination",
Sixth Annual Conference on Hydrocarbon Contaminated Soils, Amherst MA.
50. Pasquill, R., 1974: Atmospheric Diffusion, John Wiley & Sons, New York NY.
51. Penman, H.L., 1963: "Vegetation and Hydrogeology", Technical Comment No 53,
Commonwealth Bureau of Soils, Harpenden, England.
52. Pollock, D., 1989:^4 Graphical Kernel System (gks) Version of Computer Program
Modpath-Pilot for Displaying Pathlines Generated from the U.S. Geological Survey
Modular Three-dimensional Ground-water Flow Model, U.S. Geological Survey.
53. Prickett, T.A. and C. G. Lonquist, 1971: Selected Digital Computer techniques for
Groundwater Resource Evaluation, Illinois State Water Survey, Bulletin 55.
28
-------
54. Ravi, V., and J.A. Johnson, 1997: VLEACH: A One-Dimensional Finite Difference Vadose
Zone Leaching Model, U.S. EPA Robert S. Kerr Environmental Research Laboratory, Ada
OK.
55. Rawls, W.J., et al., 1985: "Prediction of Soil Water Properties for Hydrologic Modeling
Proceedings", Watershed Management in the Eighties, American Society of Civil Engineers,
New York NY.
56. Reynolds, B., 1984: "A Simple Method for the Extraction of Soil Solution by High Speed
Centrifugation" Plant and Soil, Vol. 78, pp. 437-440.
57. Sanders, P.F., and A.H. Stern, 1994: "Calculation of Soil Cleanup Criteria for Carcinogenic
Volatile Organic Compounds as Controlled by the Soil-to-Indoor Air Exposure Pathway",
Environmental Toxicology and Chemistry, Vol. 13, pp. 1367-1374.
58. Schroeder, P.R., et al., 1994: The Hydrologic Evaluation of Landfill Performance (HELP)
Model, Version 3. Risk Reduction Engineering Laboratory, Cincinnati OH, EPA/GOV/R-
94/168b.
59. Shiozawa, S. and G.S. Campbell, 1991: "On the Calculation for Mean Particle Diameter and
Standard Deviation for Sand, Silt and Clay Fractions", Soil Science, Vol. 152, pp. 427-431.
60. Schulz, E.F., 1976: Problems in Applied Hydrology., Water Resources Publications, Ft.
Collins CO.
61. Thibodeaux, LJ. and S.T. Hwang, 1982: "Landfarming of Petroleum Wastes - Modeling the
air emissions Problem", Environmental Progress, Vol. 1, No. 1, pp. 44-46.
62. Tomich-Kent, L.J., 1998: Guidance for Fate and Transport Modeling, Alaska Department of
Environmental Conservation, Division of Spill Prevention and Response, Contaminated Sites
Remediation Program, Guidance No. CSRP-98-001.
63. Ungs, J.U., 1997: "IBM-PC Applications in Risk Assessment, Remediation and Modeling",
National Ground-water Education Foundations Short Course, Orlando FL, January 8-12.
64. Voss, C.I., 1984: A Finite-Element Simulation Model for Saturated-Unsaturated, Fluid
Density-Dependent Ground. Water Flow with Energy Transport or Chemically-Reactive
Single-Species Transport", U.S. Geological Survey Report No. 84-4369.
65. Walton, W.C., 1985: Practical Aspects of Groundwater Modeling, Second Edition, national
Water Well Association.
66. Wang, H.F. and M.P. Anderson, 1982: Introduction to Groundwater Modeling, W.H.
Freeman and Company, San Francisco CA.
67. Weidemeier, T.H., etal., 1995: Technical Protocol for Implementing Intrinsic Remediation
with Long-Term Monitoring for Natural Attenuation of Fuel Contamination Dissolved in
Groundwater, Air Force Center for Environmental Excellence.
68. Yeh, G.T., 1981; AT123D: Analytical Transient One-, Two-, and Three-Dimensional
Simulation of Waste Transport in the Aquifer System, Oak Ridge National Laboratory
Publication No. 1439.
69. Zheng, C., 1990: A Modular Three-Dimensional Transport Model for Simulation of
Advection, Dispersion and Chemical Reactions of Contaminants in Groundwater Systems,
S.S. Papadopulos & Associates, Inc.
29
-------
70. Zheng C., and G.D. Bennett, 1995: Applied Contaminant Transport Modeling, Van Nostrand
Reinhold, New York NY.
Bibliography on Measurement of Input Parameters:
71. Clapp, R.B. and G.M. Hornberger, 1978: "Empirical Equations for Some Soil Hydraulic
Properties", Water Resources Research, Vol. 14, No. 4.
72. Connor, J.A., et al., 1997: "Parameter estimation Guidelines for Risk-Based Corrective
Action (RBCA) Modeling", 1996 Petroleum Hydrocarbons and Organic Chemicals in
Ground Water: Prevention, Detection, and Remediation Conference, National Ground Water
Association Catalog #T539.
73. EPA, 1991: Characterizing Soils for Hazardous waste Site Assessments, Office of Research
and Development, EPA/540/4-91/003.
74. EPA, 1991: Compendium of ERT Soil Sampling and Surface Geophysics Procedures,
OSWERPublication 9360.4-02, EPA/540/P-91/006, PB91-921273.
75. EPA, 1991: Compendium of ERT Surface Water and Sediment Sampling Procedures,
OSWER Publication 9360.4-03, EPA/540/P-91/005, PB91-921274.
76. EPA, 1992: Compen dium of ERT Air Sampling Procedures, OSWER Publication 9360.4-05,
PB92-963406.
77. EPA, 1991: Compendium of ERT Ground Water Sampling Procedures, OSWER Publication
9360.4-06, EPA/540/P-91/007, PB91-921273.
78. EPA, 1998: Estimation of Infiltration Rate in the Vadose Zone: Application of Selected
Mathematical Models, Office of Research and Development, EPA/600/R-97/128a and 128b.
79. EPA, 1993: Subsurface Characterization and Monitoring Techniques: A Desk Reference
Guide, Center for Environmental Research Information, Environmental Monitoring Systems
Laboratory, EPA/625/R-93/003a.
80. Rawls, W.J., et al., 1983: "Green-Ampt Infiltration Parameters from Soils Data", Journal of
Hydraulic Engineering, Vol. 109, No. 1.
81. Rawls, W.J. and D.L. Brakensick, 1989: Estimation of Soil water Retention and Hydraulic
Properties; Unsaturated Flow in Hydrologic Modeling, Kluwer Academic Publishers, 1989.
Bibliography on Fate and Transport Model Packages:
82. Spence, L.R. and T. Walden, 1997: Risk-Integrated Software for Clean-Ups: User's Manual
Version 3.0, BP Oil.
83. Connor, J.A., etal., 1995: Tier 2 Guidance Manual for Risk-Based Corrective Action,
Groundwater Services, Inc., Houston TX.
84. American Petroleum Institute (API), Exposure and Risk Assessment Decision Support
System (DSS) Software: Version 1 (find at http:///www.api.org/ehs/software.htm).
30
-------
MATRICES
-------
s s i :ป! ;;i;: fflWS*^ ii j ** l -! ฃ ii W
'< n ' '
jiijii': |i|iiiiiii|iiiii'iii'i ,iii!|,|,ii,i,,D|i|i' i ii, i'!ii'''ii|||,:, !''i ! :'',i ii1' !'|>i| i S';;: ',4 ,|,i;p 'iiJilli::,,:' | m\tS\/ 'ilHillii:: illllln: p|illilll|iiiilll|||||pii,'li' i,ipill|lllllll|iiilllli, ' ^ivpv' ^ * ww?. \ w-' 711 -1 "ii!"' i > iw? ft' " ^ ^'':'': ^ iw:' it* ? ^,1 siiii,,,1 i iii wvt - tiliiiliPiijiiliiiii in in
IT:: nnilllpilpiil"!!!!' (iniiffi, ill liiii'iijilliillllimil 'liilllliii'iniil'lulillll Itii ;nilllilliil|,jiin.il JLlvilil'i; Hi iili1:^:.^!*.]'!!!1 '''Iliiinillili'''^!^!!''!!!'!!!! i'' '*; lii*1!1*' iili'iil'ilirin'gJ'Pll.i'.iill, "iijU'ii '"i i'11',1!!,!''/-'!'!1''!'!'!!!'; V??', i" f'i livJii Ui'llinniHi,,!,!!'! ''fiiiiriliii!"! ""ll'i'iHillHIi'iiilllii' I, 'ifH!111!:"1,, '> 'i""!1,!'1' :l':,pi1;1 liimjn,1, ""illi,"'!* i,,1,1::1,,,:,1 ^ti jIJi1, i"p,r :: |i:,ซ,;, 'iiJJviniilll', i'iiiillllilTlliliLiiilii'nili: P|!iXnilllllllllll|jlllll|il!|i', nillllliiltji'iini iriiiSII I
l|||||'i':ii'"H'i i II '
< ,i' 'illii Inl'l1 1 1 , ' i :: Illlill' 1", "1 1 1'lllli iilili* ll'llnnll'liPII''!!!,:.; ! IJiHlllWh! " V i-fillllllillilli J I ' ..... IP,!" ..... r W*i' SBhi1. UlP!l' "ii 'i i iiiiii :i i i'ili! i' ii,ป: iii", I iln'i'ii ifiiii! iiii iiiii",'' v'i:,, :''' iiii 'i i iiii iiiiii iiin 'i' ' ,' iiiiiiii, <' iTiiiiiiiiiiiiiiiilij ijijiivji iiiiii!ii!iiii|iiiiiiii, liiiliiiiiiiiiiiii ii" iniiiiiiiiiiiiiiniii |
"" ! ! ! ! I! i !! ""'' ,! !!'' !'"' i "' ; ! ' i " '""' "! ; '"!''! !
" ""nl , :> 1IM!!1 ..... ! !! I!'1 ^ j'li ! "'"liEnniliiiliil "lifC ;:',iffi!iii!':i!" "" F! >:::!n!li||j|! ^'
V T , ' !> IIP! ! 'i i'ป :! Ii' : ,"i: :"KS 'SM Ji '!', :' .,'"' ll'l ':'!:' i, tf "!! il!i ': I, t1 li'iHr 'i i" ;!' ^ ' Vs'H ' :' ; Ii ..... "i ' L Mi ii'Hi < "ll, i:il '"I ili|!lii!l _
^ i ป
i' I'i'i"'"'
,,'":*!",,,:i!i":T:!T|!!!!f: : i,i:!:"":!:::'"::i' ',:, "S, I
,|in ,| i i i f
, ,,,,"!l1 i,' ,' !, '
!',""!!ฃ", '!"in' Ii'lii Illllilj!! \ MJ!!:!!"!!!': "I1!,1, lilltu" j
^^^^
i:;! TV ii "
::::; I:!;:
"^ 'i ;t'::4 il'::;,:::!
liin: i , i i ,:,,
!! : i,,
; ;; ;,; j,
: : ! : ;:; s1;11! : : j : :" ": , ::: I!1: : , i: : ;! i11!!1:
'':,>{!!! ": i< iiiina^ ii:,,i,.s iiiiUii'i ii :i,iiiiii|i,^iซ^^^^^^ :M)VH^^ ,>;::i,>>,-,11: K:;:! rwistปf;IHIB*H-':'M > <>: ''tin1! ป>:,:> "ii"' ,F<ซ:;i!!!,<4'i f^'u "vKiiiiii::,!!,:: ::i":'< -it' ff, ii r'!! "m- > t ' >: n:, iiifil liiiiiiii111 M "t
'riP: Fin! I
',; i '* ,<:<
iii wi" i wiiiiiff i lipiii! !'',: i, ,i i,,' 'mi w i fi i* i: ,ii:,! i;, iwiiiai,:ii jf:11:,, agw >mA r "iniiiiilf, l la1*; ['it ii i|>: v Bp1 iiiiMa ill"!' wi iiii l K^^ :', i', ft, ii fi I'HiMi'p";, Viioii "^/i: ;,!iii^ rav, wfi :i i"1!1 i i:11,'" itfTL1,: i ''v "ii,* ""fti: i1!", : ,r l, fl' f l' pi ii iip11;! fBii ll ,,if iiilil'ii 'i, i1 jfeiiir! lhซ
iw,, iiiiiiiiiiipirtiiiis1 iiiiii iiii ii' ,<; i iiiiiiia^^^^ ii1,1'!!1!;,;1!!, i inniufi11,!!1,1 -THB it i A siiiiiin^ i mi 'i!,!iiiiiiiiiiiaii!!im^^ , 'iitii;,,,,!!!,,,1!,',!!,' ,vinr n ', p IP,,, iiihAiiiipi IB v i iii: ,n 1^1,4 nil 'U, j,, JUTI, !>l ซ, iui:, "I!I:,PPII "H,!,i luniiiPiiiiiii ii t! , , 'r," - i pjiuM ipi'1" :, tj p, .irn,'! ' "j1 p :\ : 4 ,11:', ppnpninpiMipp'ipippiiiiniiiii pull ;pipi! ppppH 'iiii
I:.nil:,'!, PIP1!1?!!,!!!!1:'1':'"1!11! L!!!!
i-i ' I"',' '' ''Ll ป ' !'" '"'*,' '':i ;; : p-'Mi'l.!"11,'I,1 .ij. ..iii i I illli P! ".'l.il.. ,. '.I"..' n1.,. . . Pi ' I.. 'I. I. 1 .! . ' .1'. "* plปl...|l lll.l. .'f slllplllpll.1
, !*!! ;!;' | > ป!; I
-------
MATRIX 1
Key Model Information
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References;
Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Soil to Ambient Air
Jury - Infinite Source
Jury - Finite Source
Farmer
Thibodeaux-Hwang
Box
Vapor Migration from the
surficial soils to ambient air.
Vapor Migration from the
surficial soil to ambient air.
Vapor Migration from
subsurface soils to ambient air.
Vapor Migration from
subsurface soils to ambient air.
Dispersion of Vapors in
Ambient Air, no
biodegradation
ID Analytical
Geometric
ID Analytical
Geometric-Exponential
ID Analytical - Linear
ID Analytical -
Geometric
ID Analytical - Linear
Average flux at
surface
Flux to ambient air
over time
Instantaneous flux
at surface
Average flux at
surface
Breathing zone
concentration
Assumes soils are impacted from the
surface to an infinite depth, no leaching or
evaporation, no soil-air boundary layers,
and soil concentration is in the dissolved
phase only (no residuals). Appropriate for
thick zones of impacted soil or short
exposure time. Assumes the effective
diffusion coefficient is constant in
isotropic/homogeneously mixed soil
Assumes characteristics of the infinite
model except soils are impacted from the
surface to a finite depth. Appropriate for
defined zones of impacted soil.
Assumes the location and source
concentration remain constant and that
there is a discrete layer of unimpacted soil
between the atmosphere and the impacted
zone. Simplest model, since the
concentration remains constant, the
surface flux term does not change with
time.
Assumes that concentrations near the
surface and surface flux decrease with
time. Developed for land-farming
processes. Biodegradation is not easily
incorporated into the model. Most
representative for low biodegradable
petroleum compounds.
Assumes complete and total mixing,
constant wind velocity, no degradation.
The mixing zone is rectangular with one
side parallel to the wind direction.
Assumes simple vapor dispersion from
constant soil emissions. In common use
and readily available.
Standard
spreadsheet
application
Standard
spreadsheet
application
Standard
spreadsheet
application
Standard
spreadsheet
application
Standard
spreadsheet
application
Jury etal., 1983;
ASTM Risk-Based
Corrective Action
(RBCA) Guidance,
Soil Screening
Guidance (SSG)
Jury etal., 1990;
SSG, EMSOFT
Farmer etal., 1980;
ASTM RBCA,
SSG
Thibodeaux and
Hwang, 1982;
ASTM RBCA,
SSG
SEAM, 1988;
ASTM RBCA,
SSG
Pagel
-------
-------
. MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References/
"Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Soil to Ambient Air
(continued)
Soil to Indoor Air
SCREENS
ISCST 3
Farmer
Farmer
(modified)
Dispersion of vapors in
ambient air. Can be configured
to model worst-case
atmospheric conditions and
multiple sources
Dispersion of Vapors in
Ambient Air - Can adequately
model complex geometrical
configurations of the source(s)
and receptors. Revised to
perform a double integration of
the Gaussian plume kernel for
area sources
Vapor diffusion from soil
through floor or foundation
Vapor diffusion from soil
through floor or foundation ,
considers advection.
ID Analytical -
Exponential
ID Gaussian plume
model
ID Analytical - Linear
ID Analytical - Linear
1 hour average
concentration
above the ground
"N"-day average
concentration or
total deposition
calculated at each
receptor for any
desired source
combinations
Instantaneous flux
at surface
Instantaneous flux
at surface
Allows input of mixing zone and down-
wind distance to exposure point. Does not
incorporate the effects of terrain.
Appropriate for area, volume and point
(stack) sources. Also appropriate for one
rectangular source and a limited number
of receptors. Requires dimensions of
source, emission rate, and downwind
receptor distance. Does not consider
particle settling, deposition, or wind
direction. Commonly used, easy model
with extensive testing.
Appropriate for multiple sources,
numerous receptors, and where the source
and receptor are separated by some
distance. Will predict deposition rates.
Considers terrain and hourly
meteorological data. Chemical half-life
transformations possible. Requires
dimensions and emissions rate for each
source, hourly meteorological data, and
receptor locations. Can consider particle
settling, depositions rates, and
rudimentary chemical reactions.
Commonly used model with extensive
testing.
Assumes that the floor provides resistance
to diffusion. Models indoor air mixing
based on a box model with air exchange
rate and dimensions of the enclosed space
as input.
Assumes that the floor provides resistance
to diffusion. Considers advection and the
permeability of site soils. Not a
conservative model when sites have
highly permeable soils.
Intel 80286, DOS
3.0 or higher, 640
Kb RAM, 500 Kb
free disk space,
math coprocessor
486/Pentium with 8
MB RAM running
Windowsฎ 3.1,
Windowsฎ 95 or
Windowsฎ NT
Standard
spreadsheet
application
Standard
spreadsheet
application
SCREENS User's
Guide, EPA, 1995;
SSG
Superfund
Exposure
Assessment
Manual (SEAM),
1988; EPA, 1992;
Scientific Software
Group, National
Technical
Information
Service (NTIS)
Jury, Farmer, 1983;
SSG
Jury, Farmer, 1983;
SSG
Page 2
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References./
Pathway" Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Soil to Indoor Air
(continued)
Soil to Groundwater
Johnson and Ettinger
LEACH
SAM
VADSAT
Vapor migration from
subsurface soil through a
cracked foundation. Includes
diffusion and advection
processes but no
biodegradation.
Calculates soil leaching
partitioning factor and an
attenuation factor for mixing
with groundwater specifically
developed for use with
hydrocarbon fractions. Has
linear equilibrium partitioning,
no biodegradation and well-
mixed dispersion in
groundwater.
A modification of the LEACH
model to provide a more
rigorous characterization of
soil to groundwater process
with dilution,
evapotranspiration, sorption,
biodegradation time average
factor.
Contaminant transport through
unsaturated soil using
compartmental approach with
different models to describe
source zone, vadose zone
above the source, and vadose
zone between source and
groundwater.
ID Analytical -
Exponential
ID Analytical - Linear
ID Analytical -
Exponential
ID Analytical -
Exponential
Average flux at
surface and indoor
air concentration
Leaching factor
Leaching factor
with
biodegradation/
time-average factor
Contaminant
transfer to
groundwater,
volatilization
losses
Similar to Farmer model but adds set of
terms to account for flow resistance due to
a floor or foundation. Assumes constant
soil concentration, no biodegradation, no
leaching, and all soil vapors will enter
building, primarily through cracks and
openings in the basement wall or
foundation. Assumes advective air flow
from the soil into the enclosed space.
Assumes all chemical vapors below the
basement will enter and will have a well-
mixed dispersion in air once in the
building.
Assumes constant concentration in
subsurface soils, linear equilibrium
partitioning, steady-state leaching from
the soil to groundwater, no
biodegradation, and well-mixed dispersion
of leachate in groundwater. Relatively
simple and very conservative. Commonly
used for Tier 1 .
Augments the LEACH model to
characterize critical input parameters and
more accurately simulate rainfall
infiltration and leachate migration.
Applicable to analysis of porous media
soils impacted by either organic and
inorganic constituents in the absence of
NAPLs. Can predict groundwater
concentration given affected soil value or
calculate a SSTL given a groundwater
exposure limit
Homogenous/uniform soil conditions
below source, hydraulic conductivity
calculated as a function of constant
moisture content, assumes source has
uniform concentration, does not consider
water table fluctuations. Considers finite-
mass source zone, pseudo steady-state
volatilization , diffusive vapor transport
from source to ground surface, leaching
from source zone
Standard
spreadsheet
application
386/486 with math
coprocessor, 4 MB
RAM, 2.5 MB free
disk space, and
DOS 3.0 or higher
386/486 with math
coprocessor, 4 MB
RAM, 2.5 MB free
disk space, and
DOS 3.0 or higher
IBM 486 or
compatible, 10MB
RAM, 8 MB free
disk space,
Windowsฎ 3.1
Johnson and
Ettinger, 1991;
ASTM RBCA,
SSG
ASTM.1995;
ASTM RBCA,
SSG
J. A. Connor et al,
1996;TNRCC
Scientific Software
Group
Page 3
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References/
Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Soil to Groundwater
(continued)
Jury-Unsaturated
SESOIL
HELP
Designed to simulate chemical
flux in vadose zone. Can
predict concentration in the
aqueous phase and estimate
mass loading to groundwater
over time.
Flow and Transport. Describes
chemical fate and transport in
the vadose zone with
dissolution, diffusion,
absorption, dispersion,
biodegradation, and
volatilization.
Simulates the water balance in
unsaturated and variably-
saturated soils. Developed for
landfills and solid waste
containment facilities as a tool
to evaluate impacts of design
alternatives.
ID Analytical
ID - Hybrid analytical -
numerical
Quasi 2D Deterministic
Concentration with
depth, flux to
ambient air, flu to
groundwater
Concentration with
depth, flux to
ambient air, flux to
groundwater
Infiltration rate
Accounts for capillarity, advection,
diffusion, infiltration, recharge,
absorption, degradation. Uses a
multiphase partitioning equation to relate
concentration between media. Assumes
uniform and steady infiltration. Most
appropriate for time-varying volatile flux
simulations. Assumes homogeneous soils
with uniform chemical distribution within
the source layer. The hydrology model is
very simple. Commonly used for Tiers 2
and 3.
Assumes a finite source. The most
sensitive parameters are biodegradation
rate, soil organic carbon content, annual
precipitation, and depth to groundwater.
Combines 3 modules: a hydrologic
module simulating the water balance, a
pollutant transport module simulating
chemical fate and transport, and a
sediment erosion module. Does not
address contaminant movement in
saturated zone. Widely used, readily
available, and commonly used for Tiers 2
and 3.
Considers effects of vegetation,
topography, engineered covers and liners,
and differential soil layers on runoff and
interception of precipitation. Includes a
large database for weather data for
different cities. Can calculate unsaturated
hydraulic conductivity and soil particle
size distribution from input data. Does not
address transport processes. User-friendly
and commonly used over several tiers.
Intel 80i86, DOS
3.0 or higher, 640
Kb RAM, 3MB
free disk space, and
math coprocessor
Intel 80i86, DOS
5.0 or higher, 2MB
RAM, 2 MB free
disk space, and
math coprocessor
Written in Basic
Language for use
under DOS 3.1 or
higher in IBM-PC
or compatible
computers with 3
MB free disk space
W. A. Jury, D.
Russo, G. Streile,
H. El Abd, 1990;
SSG
Bonazountas and
Wagner, 1984;
Scientific Software
Group,
International
Ground Water
Modeling Center
(IGWMC)
Payton, R. and P.
Schroeder, 1994;
IGWMC
Page 4
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References)
Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Soil to Groundwater
(continued)
Groundwater to
Ambient Air
VLEACH
SUTRA
MOFAT
VS2DT
Farmer
Describes movement of
organic constituents within and
between three phases: solute
dissolved in groundwater, gas
in the vapor phase, adsorbed
compound in the solid phase.
Leaching is simulated in a
number of distinct, user-
defined polygons vertically
divided into a series of user-
defined cells.
Steady-state or transient flow,
saturated and unsaturated
conditions, simulates flow
under variable density
conditions with transport of
energy or dissolved
substances.
Flow and transport of three
fluid phases. Includes
advection, dispersion,
diffusion, sorption, decay, and
mass transfer. Handles cases
in which gas and/or NAPL
phases are absent in part or all
of the domain.
Simulates contaminant
transport in the vadose zone,
simulating variably saturated
soils.
Simulates vapor diffusion from
groundwater through soil and
vapor dispersion in air
assuming an infinite source.
ID Numerical Finite
Difference
2D Numerical Hybrid
Finite-difference and
Finite-element
2D Numerical Finite
Element
2D Numerical Finite
Difference
ID Analytical - Linear
Equilibrium
distribution of
constituent mass
between liquid,
gas, and sorbed
phases. Area-
weighted
groundwater
impact for modeled
area.
Pressure heads,
concentration
distribution over
time
Distribution of
constituent
concentration
Time history,
spatial profiles of
pressure and total
head, volumetric
moisture content,
saturation,
velocities, solute
concentration
Contaminant flux
at surface
Assumes vadose zone is in a steady-state
condition with respect to water movement.
Assumes moisture profile within vadose
zone is constant. Assumes homogenous
soil conditions within polygon. Does not
incorporate biodegradation. Does not
account for nonaqueous phase liquids.
Accounts for capillarity, convection,
dispersion, diffusion, absorption. Allows
sources, sinks, and boundary conditions to
be time-dependent.. Links both
unsaturated leaching and saturated
groundwater flow. Relatively complex
site-specific model commonly used for
Tier 3. Requires experienced user and
reviewer.
Accounts for advection, dispersion,
diffusion, absorption, decay, mass
transfer. Can represent the transport of up
to 5 chemicals in four phases (water, air,
soil, and oil) while allowing up to 10
layers of differing soil layers. Difficult to
use and does not have the same regulatory
acceptance as SESOIL. Commonly used
for Tier 3.
Accounts for evaporation, infiltration,
plant uptake. Considers non-linear storage,
conductance, and sink terms and boundary
conditions. It is widely used, has a high
degree of credibility and peer review, and
is highly sophisticated. Most commonly
used for higher tier analyses.
Assumes the flux term is constant, the
water in the capillary fringe is clean, has
high moisture content, and has low air-
filled porosity. The thickness of the
capillary zone affects the resistance to
diffusion. A thin fringe can reduce the
rate of vapor diffusion
Intel 8086, 80286,
80386, 80486,
256Kb RAM, DOS
2.0 or higher, CGA
board, math
coprocessor
Intel 80i86, DOS
3.0 or higher, 640
Kb RAM, 3MB
free disk space, and
math coprocessor
3 86/486 with math
coprocessor, 4 MB
RAM, 2.5 MB free
disk space, and
DOS 3.0 or higher
386/486 with math
coprocessor, 4 MB
RAM, 2.5 MB free
disk space, and
DOS 3.0 or higher
Standard
spreadsheet
application
Ravi, V. and J.A.
Johnson, 1997;
Center for
Subsurface
Modeling Support
(CSMoS);
Scientific Software
Group
CI.Voss, 1984;
IGWMC, Scientific
Software Group,
U.S. Geological
Survey (USGS)
ESTI, 1991; EPA
1991;CSMoS,
Scientific Software
Group
Healy, R.
1988, IGWMC,
Scientific Software
Group, USGS.
Farmer, 1980,
ASTM RBCA
PageS
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References./
Pathway 'Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Groundwater to Indoor
Air
Oroundwater Transport
Farmer
Johnson and Ettinger
(modified)
Disperse
SOLUTE
AT123D
Simulates vapor diffusion from
groundwater through soil and
vapor dispersion in air.
Vapor migration from
groundwater through a cracked
foundation. Includes diffusion
and advection processes but no
biodegradation.
Calculates conservative
estimates for the size and
duration of a MTBE or TBA
plume using finite mass
advection/dispersion equation.
A set of five programs based
on analytical solutions of the
advection-dispersion equation
for a non-conservative tracer
solute.
Mass Transport, uniform
stationary regional flow, 3D
dispersion, first order decay,
retardation
ID Analytical - Linear
ID Analytical -
Exponential
2D Analytical
ID, 2D, 3D, and
Radialsymetric
Analytical
3D Hybrid analytical -
numerical
Contaminant flux
at surface
Average flux at
surface and indoor
air concentration
Distribution of
constituent
concentration
Distribution of
constituent
concentration
Distribution of
constituent
concentration
Can calculate flux with or without
advection through a modified equation.
The effects of a capillary fringe are
included through a modified diffusion
coefficient
Modification of the Johnson and Ettinger
(1991) model. Assumes constant soil
concentration, no biodegradation, no
leaching, and all soil vapors will enter
building, primarily through cracks and
openings in the basement wall or
foundation. Assumes advective air flow
from the soil into the enclosed space.
Assumes all chemical vapors below the
basement will enter and will have a well-
mixed dispersion in air once in the
building.
Assumes horizontal, homogenous aquifer;
constant velocity; constant dispersion
coefficient proportional to velocity. To be
used for slug release of constituents.
ID and radialsy metric models simulate
effects of a single source; 2D and 3D
models support multiple point sources
using superposition to calculate
accumulated effects or to represent line or
area! sources.
Assumes stationary flow field parallel to
the source. Source release may be
instantaneous, continuous, or finite step-
wise duration and is equally distributed
over the source area or volume. Water
table does not fluctuate, flow direction is
uniform and ID. Simulates mass transport
of dissolved phase, radionuclides, or heat.
Standard
spreadsheet
application
Standard
spreadsheet
application
Standard
spreadsheet
application
Intel 80i86, DOS
3. lor higher, 640
Kb RAM, VGA
graphics, math
coprocessor
DOS 2. lor higher,
640 Kb RAM, 1
MB free disk space
and a math
coprocessor
Farmer, 1980;
ASTM RBCA
Crum, J.A., 1997
Bauer,?., 1998
IGWMC
Yeh,G.T., 1981;
IGWMC, Scientific
Software Group
Page 6
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport - Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References^/
' *" Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Groundwater Transport
(continued)
,
Domenico
FATES
MULTIMED
Summers
BIOSCREEN
Dispersion in three dimensions
over time.
Determine site-specific natural
attenuation rates for organic
constituents dissolved in
groundwater (enhancement to
Domenico analytical model)
ID unsaturated dispersion with
volatilization, biodegradation,
and decay. Saturated transport
with 3D dispersion, linear
absorption, 1st order decay,
steady state or transient flow,
single aquifer and dilution due
to recharge.
Simulates non-dispersive mass
transport in a single layer of
soil from an infinite source.
Steady-state flow conditions
and equilibrium between
absorbed and dissolved phase.
Dispersion in two dimensions,
retardation, and biodegradation
3D Analytical -
Exponential, Error
Function
Transformation
(ID flow, 3D transport)
3D Analytical -
Exponential, Error
Function
Transformation
(ID flow, 3D transport)
3D Semi-Analytical -
Linear
ID Analytical - Linear
(mixing equation)
2D Analytical -
Exponential, Error
Function
Transformation
(ID flow, 2D transport)
Normalized
concentration at
specified location
Normalized
concentration at
specified location
Leachate flux
Constituent
concentration in
groundwater
downgradient of
source
Constituent
concentration in
groundwater
downgradient of
source
Transport is ID along the centerline,
between the source and receptor, the
transport is 3D due to dispersion, and
accounts for transport across the site over
time. Requires input on advective flow
velocity, dispersivity, source
concentration and geometry. Can
accommodate biodegradation. Commonly
used to conduct a Tier 2 evaluation.
Same as Domenico. Includes optimization
routine to match model results to
measured site concentrations, database of
chemical property data, calculation of time
needed for a plume to reach steady-state
conditions.
Assumes constant source concentration,
homogeneous and isotropic environment.
Developed for landfills. Simulates
precipitation, runoff, infiltration,
evapotranspiration, barrier layers, and
lateral drainage. Uses a finite thickness
saturated zone and finite infiltration rate.
Must specify vertical dispersivity and
disposal facility parallel to flow. Not
actively updated, functionally duplicated
by other current software.
Assumes complete mixing of the water-
bearing zone. Developed as screening
model to conservatively estimate
concentrations in groundwater directly
beneath vadose-zone source. Does not
consider biodegradation, first-order decay
or volatilization. Very conservative and
appropriate for screening level.
Can run in a deterministic mode to
compute concentration versus time at a
given location or in the Monte Carlo mode
to compute probability for occurrence of a
concentration. Includes databases for soil
and chemical properties and their
variability. Requires planar groundwater
flow field.
Standard
spreadsheet
application
Standard
spreadsheet
application
DOS-based, 640
Kb RAM with
math coprocessor
Standard
spreadsheet
application
Intel 80486, DOS
3.1 or higher, 2MB
RAM, graphics
adapter
Domenico, 1987;
ASTM RBCA,
SSG
Nevin, J.P., 1997;
Oroundwater
Services, Inc.
Salhotra, 1990;,
SSG, Scientific
Software Group
Summers, 1982;
IGWMC
CSMoS; American
Petroleum Institute
(API)
Page?
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References;
Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Qroundwater Transport
(continued)
VADSAT
MODFLOW
PLASM
MOC
BIOPLUME
Random Walk
Chemical movement from a
source in the unsaturated zone
or below the water table,
considering evaporation of
VOCs, leaching of
constituents, planar
groundwater flow field,
dispersion, adsorption, first-
order decay.
Saturated, steady-state or
transient flow for single or
multiple aquifers, commonly
used for Tiers 2 or 3.
Saturated, steady-state or
transient flow for single or
multiple aquifers.
Groundwater flow and mass
transport model, steady state
or transient flow for a single
aquifer. Considers advection,
dispersion, and diffusion.
Contaminant transport under
influence of oxygen limited
biodegradation; Version III
incorporates influence of
oxygen, nitrate, iron, sulfate,
and methanogenic
biodegradation.
Qroundwater flow and mass
transport model, steady state
or transient flow
heterogeneous aquifers.
Considers convection, ,
dispersion, first-order decay,
and retardation.
3-D Analytical
2D or 3D Numerical
Finite Difference
2D or 3D Numerical
Finite Difference
2D Numerical - Finite
Difference
2D Numerical - Finite
Difference
(based on MOC)
2D Numerical - Finite
Difference
Peak constituent
concentration in
groundwater at
receptor, time to
reach peak
concentration, time
for source
depletion
Hydraulic head
Hydraulic head
Distribution of
constituent
concentration
Distribution of
constituent
concentration,
velocity vectors,
time history plots
at user-defined
observation points
Hydraulic head,
distribution of
constituent
concentration
Ability to simulate advection, dispersion,
adsorption, aerobic and anaerobic decay.
Do not apply where pumping systems
create a complicated flow system.
Assumes unidirectional groundwater
movement, constant flow rate. Easy
screening tool.
Assumes saturated zone can be
heterogeneous and anisotropic, confined
or unconfined aquifer system. Limited to
groundwater flow. Commonly used for
Tiers 2 or 3.
Assumes saturated zone can be
heterogeneous and anisotropic, confined
or unconfined aquifer system. Limited to
groundwater flow. Does not consider
advection, diffusion, or dispersion.
Commonly used for Tiers 2 or 3.
Assumes saturated zone can be
heterogeneous and anisotropic, confined
aquifer system. Commonly used for Tiers
2 or 3.
Simulates processes of advection,
dispersion, sorption, aerobic and anaerobic
biodegradation, and reaeration. Version III
includes biodegradation through
instantaneous, first, or zero order decay; or
Monod kinetics. Hydrocarbon source and
each active electron acceptor are
simulated as separate plumes.
Assumes saturated zone can be
heterogeneous, isotropic or anisotropic,
confined or unconfined aquifer system.
Commonly used for Tiers 2 or 3.
Intel 80286, DOS
3.0 or higher, 640
Kb RAM, 500 Kb
free disk space,
math coprocessor
Intel 80286, DOS
3.0 or higher, 640
Kb RAM, 500 Kb
free disk space,
math coprocessor
Intel 80i86, DOS
2. lor higher, 640
Kb RAM, 1.5MB
free disk space,
math coprocessor
386/486 processor
with math
coprocessor, 4 MB
RAM, DOS 5.0 or
higher, at least 2
MB free disk space
386/486 processor
with math
coprocessor, 4 MB
RAM, DOS 5.0 or
higher, at least 2
MB free disk
space; Windows
95ฎ for Version III
Intel 80i86, DOS
3.0 higher, 640 Kb
RAM, 2.0 MB free
disk space, math
coprocessor
CSMoS; Scientific
Software Group
McDonald, M. and
Harbaugh, A.,
1988; IGWMC,
USGS
Prickett, T. and
Lonnquist, C,
1971; IGWMC
Konikow, L. and
Bredehoeft, J.,
1994; IGWMC,
USGS
CSMoS; Scientific
Software Group
Prickett, T.;
Naymik, T.;
Lonnquist, C.,
1981; IGWMC,
Scientific Software
Group
PageS
-------
-------
MATRIX 1
Key Model Information
(continued)
Fate & Transport Name of Model Description/ Type of Code/ Model Features/Characteristics/ Computer References;
1 Pathway Model/Algorithm Process Simulations Algorithm Outputs Use Conditions/Limitations Needs Sources
Groundwater Transport
(continued)
MT3D
MODPATH
Mass Transport in the
saturated zone, steady-state or
transient flow for single or
multiple aquifers.
Semi-analytical Particle
Tracking Scheme for steady-
state flow, single or multiple
aquifers
3D Numerical - Finite
Difference
3D Numerical Finite
Difference
Simulates changes
in concentration
Computes 3D path
lines
Assumes saturated zone can be
heterogeneous and anisotropic, confined
or unconfmed aquifer system Handles a
variety of discretization schemes and
boundary conditions. Commonly used for
Tiers 2 or 3.
Assumes saturated zone can be
heterogeneous and anisotropic confined or
unconfined aquifer system. Can handle
multiple release times for particles and can
draw true cross-section grids displaying
spatial data. Superimposes particle tracks
on flow field typically generated using
another model.
386/486 with math
coprocessor, 2 MB
RAM, DOS 3.0 or
higher
Requires 386/486
with math
coprocessor, 4MB
RAM 5MB free
disk space, DOS
3.0 or higher
Zheng, C., 1990;
IOWMC, Scientific
Software Group
Pollock, D. W.
1989;IGWMC,
Scientific Software
Group, USGS
Page 9
-------
-------
MATRIX 2
Generic Site Conditions for Model Application
Site Condition for
Model Application ,
, t 'Soilto '!'.CN,
^ Ambient Air
Homogenous/isotropic soil
Infinite source depth
Finite source depth
Constant source
Dissolved-phase constituents
Depth to source increases
Unimpacted soil above source
Constant diffusion coefficient
Dispersion w/complete mixing
Constant wind speed
Downwind receptor
Dispersion w/ multiple sources
Dispersion considers terrain
Particle settling
Biodegradation/transformatiora
- ', Soil to (
' sl Indoor Air > > '^
Floor provides resistance
Mixing of indoor air
Considers advection
Considers soil permeability
Constant soil concentrations
All soil vapors enter building
So!) to
x Groundwater ,
Homogenous soil conditions
Layered soil conditions
Finite source
Constant source concentration
Constant moisture content
Linear equilibrium partitioning
Steady-state vadose zone cond.
Transient vadose zone cond.
Biodegradation/transformation
Well-mixed leachate dispersion
Considers vegetation/topo.
Rainfall infiltration
Analytical model
- ' -
Jury
Infinite
Source1
Farmer
LEACH
-
\
Jury Finite
* Source
Farmer
(modified)
SAM
x,
\"*
Farmer
ป
Johnson
Etlinger
VADSAT
Fhibodcau
x/Ilwang
ป
-
SESOIL
*
r |
Box
-
HELP
SCREENS
ซ
-
VLEACH
ICST3
SWRA
Candidate
- Models
'
'JuryUri-;
saturated
;MbFAT-
VS2DT
1 ^
^, ', ' V
IS
-------
-------
MATRIX 2
Generic Site Conditions for Model Application
(continued)
Site Condition for
Model Application
Numerical model
Engineered covers/liners
Accounts for capillarity
Uniform, steady infiltration
Includes evaporation
Considers multiple sources
Handles non-aqueous phase
Considers sinks
, ' Groundwater to ,
Ambient Air
Constant flux term
Clean capillary water in fringe
High soil moisture content
Low air-filled porosity
Groundwater to i
" , t Indoor Air
Constant flux term
Clean capillary water in fringe
High soil moisture content
Low air-filled porosity
Groundwater
.} , ป Transport * ' "*"
One dimensional
Multi-dimensional
Steady-state conditions
Transient conditions
Finite difference form
Analytical model
Hybrid analytical/numerical
Unconfined aquifers
Confined aquifers
Homogenous/isotropic aquifer
Horizontal water-bearing units
Heterogeneous aquifer
Constant groundwater velocity
Calculates velocity
Calculates constituent cone.
Calculates hydraulic head
Oroundwater flow paths
Considers dispersion
j , '\. Candidate
v ' " n Models : -. ,,.."
Farmer
farmer
0
Disperse
*
*
i . i
^ ' r- . \ x
i \
Johnson
jEttinger.
SOLUTE
*
* ! '
AT1230
I)omemeG
*
*
*
*
*
FATfcS
ป
-
i?MLULTf-
^ME0
ป
Summers
*
B1O-
SfcREEN
VADSAT
*
MOD-
FLOW
*
PLASM
*
;.r-MOG ''-.
*
m^-
PLUME
Random
5^alk
i
MT3D
*
MOO-
PATH
* ?
t
*
Page 2
-------
-------
MATRIX!
Generic Site Conditions for Model Application
(continued)
Site Condition for
Model Application
Adsorption/retardation
Continuous source
Instantaneous/finite source
Variable source concentrations
Uniform flow direction
Biodegradation/transformation
Mass transport
Mixing of water-bearing zone
Run in probabilistic mode
Chemical property database
8
,
f
, (
Candidate
Modets
t
*
>
-
ป
Page3
-------
-------
MATRIX 3
Key Input Parameters
. Fate and Transport s
Pathway
Soil to Ambient Air
Soil to Indoor Air
(in addition to input
parameters for soil
to ambient air)
Soil to Groundwater
Input
* ' 'Parameter ^ i t
Source area concentration
Volumetric air content in vadose zone soil
Volumetric water content in vadose zone soil
Total soil porosity
Depth to soil contamination
Thickness of soil contamination
Diffusion coefficient in air
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Soil bulk density
Wind speed above ground surface
Ambient air mixing zone height
Source width parallel to wind
Enclosed-space volume/infiltration area ratio
Enclosed space air exchange rate
Thickness of foundation/floor
Areal fraction of cracks in foundation/walls
Volumetric water content in cracks
Volumetric air content in cracks
Effective diffusion coefficient through crack
Floor/wall seam perimeter
Depth of crack below ground surface
Effective radius of crack
Source area concentration
Total soil porosity
Fraction of organic carbon
Carbon-water sorption coefficient
Soil-water sorption coefficient
Width of source area parallel to groundwater flow
Soil bulk density
Volumetric air content in vadose zone soil
Parameter .:
Symbol (typ.)
Cs
ฉas
ฉws
ฉT
Ls
L
Dair
*oc
H
KM
K.
Ps
uair
5air
W
LB
ER
^crack
T|
ฉwcrack
ฉacrack
Dcrack
Xaack
ZCrack
rcrack
cs
0T
foe
KOC
KS
W
ps
ฉas
Psifameter
Units (typ.)
mg/Kg
cm3/cm3
cnrVcm3
cm3/cm3
cm, ft.
cm, ft.
cnrVsec.
g-C/g-Soil
cm3-H20/cm3-air
cm3-H2O/g-C
crn3-H2O/g-soil
g/cm3
cm/sec., mi./hr.
cm
cm
cm
L/sec., L/hr.
cm, in.
cm2-cracks/cm2
cm3-H2O/cm3
cm3-ak/cm3
cm2/sec.
cm, in.
cm, in.
cm, in.
mg/Kg
cmVcm3
g-C/g-Soil
cm3-H2O/g-C
cm3-H2O/g-soil
cm
g/cm3
cm3/cm3
Comment on Sensitivity to
Input Parameter
Site-specific; sensitive parameter
Variation effects water content; sensitive parameter
Variation effects air content; sensitive parameter
Correlated with volumetric air/water contents; sensitive parameter
Highly variable, site-specific; sensitive parameter
Highly variable, site-specific; sensitive parameter
Chemical-specific; limited sensitivity
Not a sensitive parameter for this pathway
Chemical-specific; limited sensitivity
Chemical specific; moderate sensitivity
f^x Koc; moderate sensitivity
Varies little for common soil types; limited sensitivity
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Highly variable, site-specific; moderate sensitivity
Relates to volume of air in enclosed space; sensitive parameter
Causes advective flow of vapors to building; sensitive parameter
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Chemical-specific; limited sensitivity
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Site-specific; sensitive parameter
Correlated with volumetric ak/water contents; sensitive parameter
Highly variable, site-specific; sensitive parameter
Chemical specific; sensitive parameter
focx K,,,.; sensitive parameter
Highly variable, site-specific; moderate sensitivity
Varies little for common soil types; limited sensitivity
Not a sensitive parameter for this pathway
Pagel
-------
-------
MATRIX 3
Key Input Parameters
(continued)
Fate and Transport
Pathway
Groundwater to Ambient
Air
Groundwater to Indoor
Air (in addition to input
parameters for ground-
water to ambient air)
Input
- . sl Parameter
Volumetric water content in vadose zone soil
Infiltration rate of water through soil
Groundwater mixing zone thickness
Groundwater Darcy velocity
Degradation rate in vadose zone
Depth to subsurface soil sources
Thickness of vadose zone
Pure constituent solubility in water
Source area concentration
Thickness of capillary fringe
Volumetric air content in vadose zone soil
Volumetric water content in vadose zone soil
Total soil porosity
Depth to Groundwater
Diffusion coefficient in air
Diffusion coefficient in water
Volumetric water content in capillary fringe
Volumetric air content in capillary fringe
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Soil bulk density
Wind speed above ground surface
Ambient air mixing zone height
Source width parallel to wind
Enclosed-space volume/infiltration area ratio
Enclosed space air exchange rate
Thickness of foundation/floor
Areal fraction of cracks in foundation/walls
Volumetric water content in cracks
Volumetric air content in cracks
Parameter
Symbol (typ.)
ฉWS
I
Sew
UEw
A,
Ls
hv
S
cw
iu
ฉas
ฉWS
0T
LOW
Dair
^water
^wcap
ฎacap
foe
H
Koc
Ks
ps
uair
6air
W
, LB
ER
Lcrack
n
^wcrack
ฎacrack
Parameter
Units (typ.)
cm3/cm3
cm/yr., in./yr.
cm
cm/yr., ft/day
yr.'1
cm, ft.
cm, ft.
mg/L
ug/L
cm, m.
cmVcm3
cm3/cm3
cm3/cm3
cm, ft.
cm2/sec.
cmVsec.
cm3-H2O/cm3-soil
cm3-air/cm3-soil
g-C/g-Soil
cm3-H2O/cm3-air
cm3-H20/g-C
cm3-H20/g-soil
g/cm3
cm/sec., mi./hr.
cm
cm
cm
L/sec., L/hr.
cm, in.
cm2-cracks/cm2
cm3-H2O/cm3
cm3-air/cm3
Comment on Sensitivity to v,
input Parameter
Not a sensitive parameter for this pathway
Highly variable, site-specific; moderate sensitivity
Depends on soil type and does not very greatly; limited sensitivity
Volume flux, Ugw/area = K,. x i ; moderate sensitivity
Chemical specific, affected by site conditions; moderate sensitivity
Highly variable, site-specific; moderate sensitivity
Highly variable, site-specific; moderate sensitivity
Chemical specific; moderate sensitivity
Site-specific; sensitive parameter
Serves as barrier to vapor transport; sensitive parameter
Variation effects water content; sensitive parameter
Variation effects air content; sensitive parameter
Correlated with volumetric air/water contents; sensitive parameter
Highly variable, site-specific; sensitive parameter
Chemical-specific; limited sensitivity
Chemical-specific; limited sensitivity
Correlated with thickness of capillary fringe; moderate sensitivity
Correlated with thickness of capillary fringe; moderate sensitivity
Not a sensitive parameter for this pathway
Chemical-specific; limited sensitivity
Chemical specific; moderate sensitivity
4.x ]ฃ.; moderate sensitivity
Varies little for common soil types; limited sensitivity
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Highly variable, site-specific; moderate sensitivity
Relates to volume of air in enclosed space; sensitive parameter
Causes advective flow of vapors to building; sensitive parameter
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Page 2
-------
-------
MATRIX 3
Key Input Parameters
(continued)
Fate and Transport
, Pathway x ','
Groundwater Transport
Input ' s ^ '
Parameter
Effective diffusion coefficient through crack
Floor/wall seam perimeter
Depth of crack below ground surface
Effective radius of crack
Source area concentration
Fraction of organic carbon
Carbon-water sorption coefficient
Soil-water sorption coefficient
Downgradient distance to nearest receptor
Saturated hydraulic conductivity
Hydraulic gradient
Average linear velocity
Width of source area parallel to groundwater flow
Total soil porosity
Soil bulk density
Saturated thickness
Storativity (storage coefficient)
Infiltration rate of water through soil (recharge)
Longitudinal dispersivity
Transverse dispersivity
Vertical dispersivity
Degradation rate
Time since release
Parameter
Symbol ,(typ.)
Doack
XCrack
ZCrack
rcrack
Cs
4
Koc
K,
X
Ks
1
V
W
ฎT
Ps
b
S
I
ax
^
az
K
t
Parameter
Units (iyp.)
cnrVsec.
cm, in.
cm, in.
cm, in.
ug/L
g-C/g-Soil
cm3-H20/g-C
cm3-H2O/g-soil
cm, ft.
cm/sec., ft./min.
ft/ft.
ft./day, ft./yr.
cm
cm3/cm3
g/cm3
cm/ ft.
unitless
cm/yr., in./yr.
cm
cm
cm
yr.-1
days, yr.
Comment on Sensitivity to
Input Parameter
Chemical-specific; limited sensitivity
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Not a sensitive parameter for this pathway
Site-specific; sensitive parameter
Highly variable, site-specific; sensitive parameter
Chemical specific; sensitive parameter
fM x KO,.; sensitive parameter
Highly variable, site-specific; sensitive parameter
Highly variable, site-specific; sensitive parameter
Highly variable, site-specific; sensitive parameter
v = Ks x i / 0T, site-specific; sensitive parameter
Highly variable, site-specific; moderate sensitivity
Affects velocity and retardation factor; moderate sensitivity
Varies little for common soil types; limited sensitivity
Site-specific; moderate sensitivity in numerical models
Depends on confined/ unconfined aquifer; limited sensitivity
Highly variable, site-specific; limited sensitivity
Varies little for common soil types; limited sensitivity
Varies little for common soil types; limited sensitivity
Varies little for common soil types; limited sensitivity
Chemical specific, affected by site conditions; moderate sensitivity
Highly variable, site-specific; moderate sensitivity
Note: The purpose of Matrix 3 is to highlight sensitive input parameters and not to provide a comprehensive compilation of all input parameters for every possible fate
and transport model. Sensitive input parameters are highlighted in bold italics. Input parameters are those commonly needed for fate and transport modeling, grouped by
fate and transport pathway. Sensitivity of specific models to input parameters is indicated in the model summaries in Appendix A.
Page3
-------
-------
FIGURES
-------
ll'HIII'ii'llll'lll'lljljI'il
"I : I !"; I !l""l
. 1111!!1,!,1!!.; '!!, L I"1 tfliflK I1 iliiilll'il i'.''!' i i M illiiii!i!':fflflfl !!:!i;i!: Httitiirtitt !!i'! H J^UKwiOK. IHrtftl iiiiiilliiliili iiii iii'li V
f i*ij!tii t '.wifiwt'^wfl 'Wflf** ipi^iiiiiiw'?!^!:'^^'^^
I iiiiiiinBilihiiiiii!':, lijiiiiiiiiiiiiiiii'ii i Ijiiiiiiii;iih!iil<"!ii! iiiiiiiiiiniiiriliiii'ii "i i> ซซ':.!, i ,,3! :' i BUI! 'iinii,,! iiii i-' >,''": !!> iii", li*iiiiii,:,iii iiiiiisiSB1'' IBIB: "f lii: liiiiiii tiiuiiiii! iiป ,'!'i; iliif" ' 'iiiiiiiiiii': pii'lrli "i.,'' '!H!!' i 'illliiyiiiiijiiili!!!'!1!^.!!!1 ill 'jfi\v:v mat' LIP, < 'inniiii!!:,11 ill' iiiiiiiiniiiipipiiiii iLi'ii'iiiiiPiiii'iiiii >, ji
l|ii, i,
jiilil, r !|i|il|ll|llllllllliliilllll p, i ri'r iiliii''"iiiiiiiiiiiiiiiiii i l'"i''i niiiii' ii'pi <'i:jyiniii ip i i!", T ;'"
ii^ i^M
1
:llllliliii|i!ii|!!p Hi1" ii:|,|lii'ii|||||l|i,i'ii'' 'E||iill||||iilili"!li|.!,ii!iliii ii!ilDiiilll;i' |iiil'r"ii'iii p" I1 "'UlBTI':,!, H :t.h:!MซA ,1 1 "f Ilil: <:!' :li I'liiii'lti" 'iliiiiliidiiiillimilUiiiii'liilvi''!.!;;;, ,i!,ill i "III Jllili! BiiiilH i,' 'II" "lllllllEllllP' Li' ,i, liliiliiiRii'liJ' i|ii:!l,i'l'"ill'll!i;il n, p'!:!lii'!!ll|il!lปl! IIIIIIH 'ilnlli BI|BB IL |: J IIEui ,i 'III 'IBEiBiipllTiiEl IXEillUEi Hiiili il'lPII IE riBBBllllli " I1,, IB i L T >': < i IIIIEIi i,liil|!:||i:E;illlllii,i"llllllll!il!IIR< "III
< |l ill ll||l,lllll'll,,< ' ',1 '!',!< 'n n , h< ' 'I -', ,' "I I' ',1 I <, I ,1 "I ' ,1, - "|i , r 'I' iU| ," < | 1 n" ", ', ', ,,',, ,l< , i',l|| ' I ll"|,<", , | "I < ", llplpllil , I , < llll ,, II, < i <
11 IP p;pj[piซipi|i|pi I'lilBiiHii in!'!'1''1'!1 'iii|i|iii||p' "i hiif i'1'iipi PI i'li|i # vf v\ iilhi.1' '.' iiiiiii'iiiiiilii' 'li'ii'iiiiii'iil'MPi'iiiiiiiiiiiiPippniipiiiii: 'iniii, ;Bi"iiii'' ซi'i|pipiipiiiiiiiii,iippip'ii!!'" 'I1' l|ll||ll IIPIIPI''"! 111'1!1 ill PIIIIIPBBIITi'Pi' IIP" I'P'II'TI'lllllR1'!1, 'll'iil !'ปI li'ilJIIIIIIll! i n' TI'il'IPip' ll'
-------
FATE & TRANSPORT
PATHWAY
ANALYTICAL & NUMERICAL METHODS
ANALYTICAL METHODS ONLY
SOIL TO GROUNDWATER
GROUNDWATER TRANSPORT
SOIL TO AMBIENT AIR
SOIL TO INDOOR AIR
GROUNDWATER TO AMBIENT AIR
GROUNDWATER TO INDOOR AIR
CAN ANALYTICAL
MODELS BE CALIBRATED
AND MEET MODELING
OBJECTIVES?
DO REGULATIONS
REQUIRE ANALYTICAL
MODELING?
DO REGULATIONS
REQUIRE NUMERICAL
MODELING?
REFORMULATE CONCEPTUAL
MODEL/INPUT PARAMETERS/
ASSUMPTIONS
ARE SITE SPECIFIC
DATA AVAILABLE OR
EASILY OBTAINED?
CONSIDER USE OF
NUMERICAL MODEL
COLLECT SITE-SPECIFIC DATA
ARE SITE
CONDITIONS
COMPLEX?
Figure 1
DECISION DIAGRAM
FOSTER WHEELER ENVIRONMENTAL CORPORATION
-------
-------
Fate & Transport
Pathway
Input Data
Requirements ^
Model
Output 2
Applicable
Model
SOIL TO
AMBIENT AIR
CHEMICAL-SPECIFIC PROPERTIES
SOURCE GEOMETRY
SOURCE CONCENTRATION
SOIL PHYSICAL PROPERTIES
ADDITIONAL DATA
METEOROLOGICAL CONDITIONS
STEADY-STATE EMISSIONS
TRANSIENT EMISSIONS
EXPOSURE POINT CONCENTRATIONS
FARMER
JURY-FINITE
JURY-INFINITE
THIBODEAUX-
BOX
SCREEN 3
ISCST 3
SOIL TO
INDOOR AIR
AREA OF ENCLOSED SPACE
CHEMICAL-SPECIFIC PROPERTIES
SOURCE GEOMETRY
SOURCE CONCENTRATION
SOIL PHYSICAL PROPERTIES
ADDITIONAL DATA
BUILDING VENTILATION RATE
BUILDING FOUNDATION PROPERTIES
PRESSURE DIFFERENTIAL
NOTES
1) CAN BE DEFAULT OR SITE-SPECIFIC VALUES
2) REFER TO MATRICES
FOSTER WHEELER ENVIRONMENTAL CORPORATION
DIFFUSIVE FLUX
DIFFUSIVE AND ADVECTIVE FLUX
FARMER
JOHNSON & ETTINGER
Figure 2
ANALYTICAL MODEL
SELECTION PROCESS DIAGRAM
SHEET 1 OF 3
-------
-------
Fate & Transport
Pathway
SOIL TO
GROUNDWATER
Input Data
Requirements 1
PRECIPITATION CONDITIONS
CHEMICAL-SPECIFIC PROPERTIES
SOURCE GEOMETRY
SOURCE CONCENTRATION
SOIL PHYSICAL PROPERTIES
ADDITIONAL DATA
Model
Output 2
LEACHING FACTOR
CONCENTRATION WITH DEPTH
Applicable
Model
LEACH
SAM
VADSAT
JURY UNSATURATED
SESOIL
TOPOGRAPHY
DIFFERENTIAL SOIL LAYERS
PRESSURE DIFFERENTIAL
INFILTRATION RATE
HELP
GROUNDWATER TO
AMBIENT AIR
GROUNDWATER TO
INDOOR AIR
WIND SPEED
CHEMICAL-SPECIFIC PROPERTIES
SOURCE GEOMETRY
SOURCE CONCENTRATION
SOIL PHYSICAL PROPERTIES
ADDITIONAL DATA
ENCLOSED SPACE GEOMETRY
VENTILATION RATE
NOTES
1) CAN BE DEFAULT OR SITE-SPECIFIC VALUES
2) REFER TO MATRICES
FOSTER WHEELER ENVIRONMENTAL CORPORATION
CONTAMINANT FLUX
CONTAMINANT FLUX
FARMER
FARMER
JOHNSON & ETTINGER
Figure 2
ANALYTICAL MODEL
SELECTION PROCESS DIAGRAM
SHEET 2 OF 3
-------
-------
Fate & Transport
Pathway
Input Data
Requirements i
Model
Output 2
Applicable
Model
GROUNDWATER
TRANSPORT
TARGET GROUNDWATER CONCENTRATION
AQUIFER GEOMETRY
SOIL PHYSICAL PROPERTIES
SOURCE GEOMETRY
ADDITIONAL DATA
DISPERSIVITY
RECEPTOR DISTANCE
CHEMICAL PROPERTIES
EXPOSURE POINT CONCENTRATION
SUMMERS
DISPERSE
SOLUTE
EXPOSURE POINT CONCENTRATION
DISPERSION FACTOR(S)
LEACHATE FLUX
CONTAMINANT TRANSPORT
STEADY-STATE/TRANSIENT
BIOSCREEN
VADSAT
DOMENICO
FATE 5
MULTIMED
AT123D
NOTES
1) CAN BE DEFAULT OR SITE-SPECIFIC VALUES
2) REFER TO MATRICES
FOSTER WHEELER ENVIRONMENTAL CORPORATION
Figure 2
ANALYTICAL MODEL
SELECTION PROCESS DIAGRAM
SHEET 3 Of 3
-------
-------
Fate & Transport
Pathway
SOIL TO
GROUNDWATER
GROUNDWATER
TRANSPORT
NOTES
Input Data
Requirements i
GROUNDWATER FLOW
INFILTRATION RATE
FRACTION ORGANIC CARBON
SOURCE GEOMETRY
SOURCE CONCENTRATION
SOIL PHYSICAL PROPERTIES
CHEMICAL-SPECIFIC PROPERTIES
FLUID PROPERTIES
AQUIFER HYDRAULIC PROPERTIES
HYDRAULIC GRADIENT
RECHARGE
FLOW SYSTEM GEOMETRY/BOUNDARIES
ADDITIONAL DATA
SOURCE AREA CONCENTRATION
SOURCE GEOMETRY
AQUIFER DISPERSIVITY
SOIL PHYSICAL PROPERTIES
TIME SINCE RELEASE
1) CAN BE DEFAULT OR SITE-SPECIFIC VALUES
2) REFER TO MATRICES
FOSTER WHEELER ENVIRONMENTAL CORPORATION
Model
Output 2
Applicable
Model
CHEMICAL CONCENTRATION
IN SOIL/FLUID
MOFAT
CHEMICAL CONCENTRATION IN SOIL
SESOIL
MOFAT
CHEMICAL FLUX
VS2DT
EQUILIBRIUM DISTRIBUTION
OF CHEMICAL MASS
CHEMICAL FLUX
PORE FLUID PRESSURES
HYDRAULIC HEAD
HYDRAULIC GRADIENT
STEADY STATE/TRANSIENT
WATER FLUX
VLEACH
SUTRA
MODFLOW
PLASM
MODPATH
CONTAMINANT TRANSPORT
STEADY-STATE/TRANSIENT
MOC
RANDOM WALK
MT3D
BIOPLUME
Figure 3
NUMERICAL MODEL
SELECTION PROCESS DIAGRAM
-------
-------
APPENDIX A
-------
i S i .; ,,|i|
iizi.:s?n!.
i IE EiiiiiiBEillEIEEi'lE
iiiijiiiiii: ii, ii,!;ii!,,,i,'iis' i,"i,,i iii; ii", moil iiiii:iii,N!ii,,,:ini!'i'1 npi.it.1 in: lii'ijiapgpphppip :h'ipiii|,i Pi Jiiiiin '.ii; ipii|piipi"i!!ii"npi iiiiiiiiiiilliiiiiiiHiiiiii'iiiii!,1: ipiiiiiiiiiLiiiinnini PiiiiimiPiPiiiPilPiiip1 pppiPipppiipsiii IP ' inn,;11!!!!,!!;!!1 inini; iJiPii'iiiipipiiiiiiiiili! ii'ii'i'11 niiiinbiin iiiiiiiiihiiiiippiifiiLiiiiiiipiiiiii" wiihi
Eii'lK^^^^^^
r,, '''iilliiiiiiiiiii'iiiif.
;!', ', v1'1!'1 i'1'1:1" - :' Ii1 :;'" '"" ', ' ' ; ' , , !'!- ,'!,: . ': : "" , :' '"!:[!'!?''' , ,' , " ',,": ;"' :
j ,' .,,/'", ,,,i .ij,,, :,' ;i , ,, .,' " ' ..,,.- M, , , ^.j,,^ ,,., , ,, , , in, ,:, ', ',, ,;,, , , ' ,;;;,'!,,. , . , ,, in;, , , ป- , ',j' :,,;,;;!, , ' ,;, | ', ;,
1,'ipl ir 'iiii'iiiiiliPiiP" iiiippiiiiii;11 n iiiiiii iiiiiiliiipppiiijlliiiiini i iiiiitiii i hiniiiiini'iiiiiipi piiiiiiiiiiiiiiin I,IIPIII,I,IIIIII piiiiiniiiiPiipii" ,1,11111 < ii;iiiitiiiiiii,iiiin< i iiii"ii,iii|iii,iiiiir'' in1" ni1 ii'iPiiiiiiiiniinipinniiiiiiiii inn < iiiiiiiiiiiniiiiiiniiiipi'ii1 P i iiiiiiii"iiiiiiii;ni
i I,,,,.
i^;:!:!!;!;'!,!!!!^ ..... '
.. ...... !";i'lliii!:!JIVปliP!! ...... HJililihiiHl ...... Mi'it" ..... HllltK1 ...... II ..... llliiilllililiiKiliii^iiniiliiiiilll ....... !iiiiii9i4il!!liiii!iii
npi!!i!|!P!li liiiiiiiiiiiilliipii1 ilpnnipijjijIjppiinnnF'piniP1 ' f nliii'iiii iil'i'piin ,m : "W't^ " I'l ' H r, m, miji \ ms f ...... ii i",i< i jiiliiiii'SnlPliii'ii >i|i,np|:iปniil|p<>ii iป|i i'J'P'i'llJli'jPIIP'lil! pp Iliil''' Hh 11 illlllilPiliiiiinill ii'li; illlillliililllliili'lliiii'l'iiiP'i J'nPllllll'i i,, 'illjlliiiiniiiilli" iii'iSli iiUllllNlillllllllIlp1, ' ill|Flii!ili!PP|liillillllllll|lili, lll i n;;"' jfl iiiri'ia1:!!!:!!!,;, j ^iiii'iiiiiliiiiiii nii:!i: I'l'iiuiHft z,\ p^asix, iseiiW'? is"' "leajiliiiii!;:1:!;!]!! iSiifiiiiiiBiBii; iiciiiflj!:1:!'1*!!]!*!!!!!!, rs! l:i;;:i!!IJ!iiilliii a, :ii ,L ci iPiiipai^iiEiiiisiipiii!'1;!!1? at m pi1!,
-------
SOIL TO AMBIENT AIR
-------
-------
JURY INFINITE SOURCE
MODEL OPERATION
This model assumes an infinite source for migration of volatiles from soils to ambient air and
enclosed spaces. The model assumes that the soils are initially contaminated from the ground
surface to an infinite depth. As the petroleum constituents diffuse to the ground surface, the
concentration in the shallow soil decreases. The flux or rate of vapor migration to the ground
surface decreases with time as the shallow soils become less contaminated. Because the flux
changes with time, an average flux is used in the volatilization factor. Model assumes:
No biological degradation
One-dimensional flow field (no horizontal dispersion)
Contaminated soil extends from the surface to an infinite depth
Diffusion in both the liquid and vapor phases
Equilibrium partitioning between sorbed, dissolved, and vapor phases
Reversible mass transfer between sorbed, dissolved, and vapor phases
SENSITIVE INPUT PARAMETERS
Source area concentration
Depth to soil contamination
Volumetric air content in vadose zone soil
Total soil porosity
Volumetric water content in vadose zone soil
KEY INPUT PARAMETERS
Soil bulk density
Diffusion coefficient in air
Diffusion coefficient in water
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Averaging time for fluxes
Wind speed above ground surface
Soil intrinsic permeability
Source width parallel to wind
Ambient air mixing zone height
Note : The parameter DA combines variables that relate to soil porosity and moisture content,
diffusion coefficients in the vapor and aqueous phases, and partitioning coefficients that describe
relationships between concentrations in the solid, aqueous, and vapor phases. These variables
are combined together in the DA parameter to make the equations more concise and readable.
APPLICABILITY
Focus of multiple studies, the model is highly used and tested.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
-------
THE JURY FINITE SOURCE
MODEL OPERATION
The Jury Finite Source Model is an alternative for the infinite source model for migration
from surficial soils to ambient air and enclosed spaces. This model assumes that the
contaminated soil has a finite depth. The equation used for the finite source model requires that
values be averaged over a short period of time. Model assumes:
Biological degradation can be included
One-dimensional vertical transport model, dispersion considered in vertical direction only
Contaminated soil has a finite depth
Diffusion in both the liquid and vapor phases
Equilibrium partitioning between sorbed, dissolved, and vapor phases
Reversible mass transfer between sorbed, dissolved, and vapor phases
KEY INPUT PARAMETERS
Soil bulk density
Diffusion coefficient in air
Diffusion coefficient in water
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Averaging time for fluxes
Wind speed above ground surface
Ambient air mixing zone height
Source width parallel to wind
APPLICABILITY
Very simple and easy to use.
ADDITIONAL INFORMATION
ASTM 1739-95 Risk-based Corrective Action Guidance
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
API DSS manual
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
SENSITIVE INPUT PARAMETERS
Source area concentration
Volumetric air content in vadose zone soil
Total soil porosity
Depth to soil contamination
Thickness of soil contamination
Volumetric water content in vadose zone soil
-------
FAELMER MODEL
MODEL OPERATION
The Farmer model estimates the migration of vapors from soil to ambient air. The model
assumes that the concentration in the contaminated soils and the depth to the contaminated soils
do not change with time. This is equivalent to assuming that the soils represent an infinite source
for contamination. Farmer is a soil emission model, air dispersion is modeled separately. The
Model assumes:
No biological degradation
One-dimensional flow field (no horizontal dispersion)
Constant source composition and concentrations
Diffusion in both the liquid and vapor phases
Equilibrium partitioning between sorbed, dissolved, and vapor phases
Reversible mass transfer between sorbed, dissolved, and vapor phases
SENSITIVE INPUT PARAMETERS
Source area concentration
Total soil porosity
Volumetric air content in vadose zone soil
Depth to soil contamination
Thickness of soil contamination
Volumetric water content in vadose zone soil
KEY INPUT PARAMETERS
Diffusion coefficient in air
Diffusion coefficient in water
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Soil bulk density
Averaging time for fluxes
Soil intrinsic permeability
APPLICABILITY
Simplest of the soil to ambient air models and highly used.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources
-------
THIBODEAUX-HWANG MODEL
MODEL OPERATION
The Thibodeaux-Hwang model assumes that the concentration in the soil remains constant
however the distance to the top of the contaminated layer increases as contaminants are
volatilized. The model is only slightly more complicated to use than the Farmer model, yet
provides significantly more realism. For petroleum compounds not readily biodegradable, the
Thibodeaux-Hwang model should be used.
The Thibodeaux-Hwang model is an alternative for the Farmer model in that it assumes the
near-surface soil concentrations decrease with time. The Thibodeaux-Hwang equation provides
an estimate of the average flux over the time period, which produces a more realistic long-term
estimate of vapor flux than the instantaneous flux model. The effects of biological degradation
can be incorporated into the soil to ambient air models if the assumption is made that biological
degradation follows a first-order decay equation.
SENSITIVE INPUT PARAMETERS
Source area concentration
Total soil porosity
Depth to soil contamination
Thickness of soil contamination
Volumetric air content in vadose zone soil
Volumetric water content in vadose zone soil
KEY INPUT PARAMETERS
Diffusion coefficient in air
Diffusion coefficient in water
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Soil bulk density
Averaging time for fluxes
Soil intrinsic permeability
APPLICABILITY
Highly tested and used when there is a finite source.
ADDITIONAL INFORMATION
ASTM 1739-95 Risk-based Corrective Action Guidance
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
-------
BOX MODEL
MODEL OPERATION
ASTM uses a simple box model approach. A "box" model assumes the contaminant vapors
from the soil are mixed with clean air within some box-shaped breathing zone near the ground
surface. This breathing zone, which is assumed to be located immediately above the
contaminated soil, is dependent upon the width of the contaminated soil parallel to the wind and
a mixing height that is generally assumed to be 2 meters. The amount of mixing that occurs
within this breathing zone is determined by the average wind speed in the breathing zone. The
assumptions used to develop fixed-box models are: the mixing zone is a rectangle with one side
parallel to the wind direction; atmospheric turbulence produces complete and total mixing of the
contaminants up to some mixing height, H, and no mixing above this height; the turbulence is
strong enough in the upwind direction that the contaminant concentration is uniform throughout
the mixing zone and not higher at the downwind side than the upwind side; the velocity of the
wind is independent of time, location, or elevation above the ground surface; the concentration of
the contaminant in the air entering the mixing zone is zero; the contaminant emission rate from
the soil is constant and uniform over the base of the mixing zone; no contaminant enters or
leaves through the top of the mixing zone nor through the sides that are parallel to the wind
direction; and the contaminant does not degrade in the atmosphere.
KEY INPUT PARAMETERS
Length of the mixing zone in the direction of the wind
Wind speed above the ground surface
The ambient air mixing zone height
The width of the source parallel to wind
Contaminant flux into the box (soil emissions rate)
APPLICABILITY
Useful model for screening purposes due to its conservative assumptions.
ADDITIONAL INFORMATION
ASTM 1739-95 Risk-based Corrective Action Guidance
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
-------
SCREEN 3
MODEL OPERATION
SCREEN 3 uses a Gaussian plume model that incorporates source-related factors and
meteorological factors to estimate pollutant concentration from continuous sources. It is assumed
that the pollutant does not undergo any chemical reactions and that no other removal processes,
such as wet or dry deposition, act on the plume during its transport from the source. It models
the plume impacts from point sources, flare release, and volume releases in SCREEN. The
SCREEN model uses a numerical integration algorithm for modeling impacts from area sources.
The area source is assumed to be a rectangular shape, and the model can be used to estimate
concentrations within the area.
KEY INPUT PARAMETERS
Background air concentration
Stack height wind speed
Vertical dispersion parameter
Plume centerline height
Emission rate
Lateral dispersion parameter
Receptor height above ground
Mixing height
APPLICABILITY
Commonly used, easy model with extensive testing.
ADDITIONAL INFORMATION
The Gaussian model equations and the interactions of the source-related and meteorological
factors are described in Volume II of the ISC User's Guide (EPA, 1995b), and in the Workbook
of Atmospheric Dispersion Estimates (Turner, 1970).
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
-------
ISCST3
MODEL OPERATION
The ISCST3 model may be used to model primary pollutants and continuous releases of toxic
and hazardous waste pollutants. It can handle multiple sources including point, volume, area, and
open pit source types. Line sources may also be modeled as a string of volume sources or as
elongated area sources. Source emission rates can be treated as constant or may be varied by
month, season, hour-of-day, or other optional periods of variation. These variable emission rate
factors may be specified for a single source or for a group of sources. The model can account for
the effects of aerodynamic down-wash due to nearby buildings on point source emissions. The
model contains algorithms for modeling the effects of settling and removal (through dry
deposition) or large particulates and for modeling the effects of precipitation scavenging for
gases or particulates. Receptors locations can be specified as gridded and/or discrete receptors in
a Cartesian or polar coordinates. The model uses real-time meteorological data to account for the
atmospheric conditions that affect the distribution of air pollution impacts on the modeling area.
KEY INPUT PARAMETERS
Location of the source
Physical stack height
Source elevation
Building dimensions
Stack gas exit velocity
Emission rate
Variable emission rates
Particle size distributions
APPLICABILITY
Commonly used model and widely tested.
ADDITIONAL INFORMATION
Scientific Software Group
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
-------
-------
SOIL TO INDOOR AIR
-------
-------
FARMER MODEL
MODEL OPERATION
The same model used to estimate emissions to ambient air can be adapted to model emissions
to enclosed spaces or indoor air. The Farmer model assumes that the concentration in the
contaminated soils and the depth to the contaminated soils do not change with time. This is
equivalent to assuming that the soils represent an infinite source for contamination. For
petroleum fractions that are biodegradable., a modified Farmer model can be used. Model
assumes:
No biological degradation
One-dimensional flow field (no horizontal dispersion)
Constant source composition and concentrations
Diffusion in both the liquid and vapor phases
Equilibrium partitioning between sorbed, dissolved, and vapor phases
Reversible mass transfer between sorbed, dissolved, and vapor phases
KEY INPUT PARAMETERS
Source area concentration
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Volumetric air content in vadose zone soil
Total soil porosity
Soil bulk density
Area of cracks through which vapor enter the enclosed
space or building
Thickness of the foundation or floor of the enclosed
space or building
Effective diffusion coefficient through the crack
Effective radius of crack
Depth to soil contamination
Thickness of soil contamination
APPLICABILITY
Simplest of the soil to ambient air models and highly used.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
Averaging time for fluxes
Soil intrinsic permeability
Building under pressure
Diffusion coefficient in air
Diffusion coefficient in water
Floor/wall seam perimeter
Viscosity of gas
Depth of crack below ground surface
SENSITIVE INPUT PARAMETERS
Enclosed space volume/infiltration area ratio
Ventilation rate for the enclosed space or building
-------
JOHNSON AND ETTINGER MODEL
MODEL OPERATION
The Johnson/Ettinger Model includes advective flux and is recommended for high
permeability sites. For low permeability sites, these effects are less important. The effects of
advective flow may be important for higher permeability sites. Neglecting advection may result
in non-conservative cleanup levels.
The flux term for the Johnson and Ettinger model is based on the same model used to
simulate migration from subsurface soils to ambient air (i.e., the Farmer model). An additional
set of terms has been added to the contaminant flux term to account for the resistance to flow
that is provided by the floor or foundation of the enclosed space. This resistance is quantified
using parameters that describe the number and widths of cracks in the foundation floor. The
importance of advection from the soil into enclosed spaces will depend upon the magnitude of
the sub-atmospheric pressures in the enclosed space, on the number and size of cracks in the
floor or basement of the enclosed space, and on the permeability of the soil. The effects of soil
permeability are especially significant. The effects of biological degradation can be incorporated
into the soil to enclosed space models if the assumption is made that biological degradation
follows a first-order decay equation.
SENSITIVE INPUT PARAMETERS
Enclosed space volume/infiltration area ratio
Ventilation rate for the enclosed space or building
KEY INPUT PARAMETERS
Effective diffusion coefficient through the crack
Building under pressure
Soil permeability
Floor/wall seam perimeter
Viscosity of gas
Depth of crack below ground surface
Effective radius of crack
Area of cracks through which vapor enter the enclosed
space or building
Thickness of the foundation or floor of the enclosed
space or building
APPLICABILITY
This model is widely tested and used especially for screening purposes due to its conservative
assumptions.
ADDITIONAL INFORMATION
ASTM 1739-95 Risk-based Corrective Action Guidance
BP Oil RISC model
SOURCES
Groundwater Services, Inc.
2211 Norfolk, Suite 1000
Houston, Texas 77098-4044
Phone:(713)522-6300
Fax: (713) 522-8010
-------
SOIL TO GROUNDWATER
-------
-------
LEACH
MODEL OPERATION
The model, developed for ASTM (1995), calculates a soil leaching partitioning factor and an
attenuation factor for mixing with groundwater. Dissolution of contaminants into infiltrating
precipitation is estimated using equilibrium partitioning (which can be capped at the effective
solubility), and dilution into groundwater is estimated using a relatively simple box model.
Calculation of the leaching factor is based on the following assumptions: A constant
chemical concentration in subsurface soils; linear equilibrium partitioning within the soil matrix
between sorbed, dissolved, and vapor phases, where the partitioning is a function of constant
chemical- and soil-specific parameters; steady-state leaching from the vadose zone to
groundwater resulting from the constant leaching rate I [cm/s]; no loss of chemical as it leaches
toward groundwater (that is, no biodegradation); and steady well-mixed dispersion of the
leachate within a groundwater mixing zone.
LEACH assumes that no attenuation of the compounds or fractions occurs from the source
area to the groundwater. Thus, the concentrations entering the groundwater are identical to those
in the pore water leaving the impacted source area.
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
KEY INPUT PARAMETERS
Thickness of affected soil zone
Bulk density
Volumetric water content
Soil-water sorption coefficient
Henry's Law Constant
Volumetric air content
Dilution factor
Darcy groundwater velocity
Mixing zone depth
Infiltration rate
Source width parallel to the groundwater flow
APPLICABILITY
Relatively simple and very conservative.
ADDITIONAL REFERENCES
ASTM 1739-95 Risk-based Corrective Action Guidance
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
10
-------
SAM
MODEL OPERATION
A modification of LEACH is known as the Soil Attenuation Model or SAM. The soil-to-
groundwater leachate process is characterized as a three-step procedure, beginning with 1)
equilibrium partitioning of soil contaminants from a finite source mass to infiltrating rainwater,
followed by 2) sorptive redistribution of contaminants from the leachate onto underlying clean
soils, and 3) subsequent leachate dilution within the receiving groundwater flow system.
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
KEY INPUT PARAMETERS
Thickness of affected soil zone
Biodecay rate of COC in vadose zone
Bulk water partitioning coefficient
Time averaging factor
Net infiltration
Distance from top of affected soil zone to top of water-
bearing unit
Distance from top of affected soil zone to top of water-
bearing unit
APPLICABILITY
The SAM model has undergone peer review and has recently been adopted by the state of
Texas for use in deriving risk-based screening levels.
ADDITIONAL INFORMATION
Texas Natural Resource Conservation Commission
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
11
-------
SESOIL
MODEL OPERATION
SESOIL (the Seasonal SOIL Component Model) is a one-dimensional model developed by
Bonazountas and Wagner (1984) to describe pollutant fate and transport in the unsaturated zone.
Transformations through biodegradation, hydrolysis and cation exchange can also be simulated.
The model allows input of up to four soil layers, the hydrology calculations use only a depth-
weighted average value. This component of the model limits its applicability to site-specific
assessments.
The model uses a mass balance approach, continuously calculating the mass input and
removal from each layer or sublayer and the masses in each of three phases: solid, liquid (non-
aqueous phase), dissolved liquid (soil moisture), and soil gas. Communication between layers is
through advection and diffusion. Importantly, SESOIL assumes all phases are in equilibrium at
all times, using partitioning equations such as Henry's law, and Freundlich adsorption isotherms
to calculate concentrations in different phases. The model does not include surface ponding, or
plant uptake (unless the user specifically inputs an evapotranspiration rate to account for this
mechanism). SESOIL can be used to calculate time until a plume reaches groundwater, as well as
the peak concentrations reaching groundwater.
KEY INPUT PARAMETERS
First-order decay, biodegradation rate
Hydrolysis rate
Soil disconnectedness index
Cation exchange
Depth to groundwater
Precipitation by month
Albedo
Relative humidity
Number of storms per month
Average storm duration
Temperature
APPLICABILITY
Evapotranspiration
Effective solubility
Intrinsic permeability
Diffusion coefficients
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
Has been widely adopted for its ease and scientific credibility.
ADDITIONAL INFORMATION
American Petroleum Institute's Decision Support System
EPA's Graphical Exposure Modeling System
California Leaking Underground Fuel Tank Program
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc/
12
-------
HELP
MODEL OPERATION
The HELP (Hydrogeologic Evaluation of Landfill Performance) model (Schroeder et al.,
1994) is a quasi-two-dimensional, deterministic water-routing model for evaluating the water
balance at sites. It was developed for landfills and solid waste containment facilities, as a tool
for evaluating the impacts of various design alternatives. It is therefore very applicable to
evaluating hydrocarbon leaching at contaminated sites, including the assessment of the impacts
of different remedial design alternatives on leaching potential.
HELP is very user-friendly, and is written in the Basic language for use under DOS in IBM-
PC or compatible computers. The program includes a large database for weather data for
different cities, or more site-specific weather data can be input. It also includes default values for
the hydrogeological characteristics of different soil types, waste materials and geosynthetic
materials (such as liners), or again empirical data can be substituted if known. Subsurface layers
can be accommodated, and seasonal differences in weather patterns are also included. In fact, the
model simulates daily water movement into, through and out of the impacted soils. The model
includes changes in infiltration capacity when frozen conditions are predicted, and changes in the
energy balance caused by the presence of snow at the surface, and snow melting with and
without rain on a surface snow layer. The HELP model also calculates changes in
evapotranspiration due to the presence and health of vegetation at the site surface, and accounts
for such factors as topography and vegetation on runoff and interception of precipitation.
KEY INPUT PARAMETERS
Thickness of affected soil zone
Cap thickness
Weather data
Soil data
Permeability
Snow melt
Leakage
Soil storage
Evapotranspiration
Runoff
Leachate recirculation
Unsaturated vertical flow
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
APPLICABILITY
The HELP model is easy to use and adaptable to a range of site-specific parameters. It has a
long history of field validation, ease of use, and broad acceptance of the approach and results.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
USAGE - Waterways Experiment Station, Vicksburg, Mississippi
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
13
-------
VLEACH
MODEL OPERATION
VLEACH is a one-dimensional finite difference vadose zone leaching model. The model
estimates impact to groundwater due to the mobilization and migration of organic contaminates
in the vadose zone. The model describes the movement of an organic contaminant within and
between three phases: liquid (dissolved phase), vapor, and absorbed (solid phase). VLEACH
employs a number of simplifying assumptions:
Instantaneous equilibrium occurs between the three phases in each vertical cell.
The moisture content profile within the vadose zone is constant.
Liquid phase dispersion is not considered.
No degradation or in situ production occurs.
Homogeneous soil conditions are assumed.
Volatilization is either completely unimpeded or completely restricted.
Non-aqueous phase liquid or variable density flow is not considered.
SENSITIVE INPUT PARAMETERS
Organic carbon distribution coefficient
Effective porosity
Soil organic carbon content
Initial contaminant concentration
KEY INPUT PARAMETERS
Solubility in water
Recharge rate
Henry's law constant
Air diffusion coefficient
Dry bulk density
Number of model cells
Upper boundary conditions for vapor
Volumetric water content
Time step
Lower boundary conditions for vapor
APPLICABILITY
VLEACH can be used as a screening model due to conservative assumptions.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm), Technical
Background Document
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-
3041
Phone:(703)620-9214
Fax: (703) 620-6793
www.scisoftware.com
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.hrml
14
-------
SUTRA
MODEL OPERATION
SUTRA is a two-dimensional model simulating flow and transport (of energy or dissolved
substances) in the subsurface (Voss, 1984). It was developed by the U.S. Geological Survey, and
is available in the public domain. It operates under the DOS environment on IBM-PC or
compatible computers.
SUTRA uses hybrid finite-difference and finite-element methods to simulate flow and
transport in the subsurface, under both saturated and unsaturated conditions. The model allows
sources, sinks and boundary conditions to be time-dependent, which is a more realistic approach
than simpler models. It also allows simulation of the complete subsurface environment (i.e., it
links both unsaturated leaching and saturated ground water flow). SUTRA also calculates fluid
pressures over time and distance, and is one of the few public-domain programs capable of
simulating flow under variable-density conditions.
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
KEY INPUT PARAMETERS
Thickness of affected soil zone
Hydraulic conductivity
Specific yield
Pumping wells
Bulk density
Volumetric water content
Volumetric air content
Henry's Law Constant
Transmissivity
Boundary conditions
Recharge from precipitation, rivers, drains
Dilution factor
Darcy groundwater velocity
Mixing zone depth
Infiltration rate
Source width parallel to the groundwater flow
APPLICABILITY
Relatively complex site-specific model. Requires experienced user and reviewer.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
Scientific Software Group
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
U.S. Geological Survey
water.usgs.gov/sofrware
15
-------
JURY - TINSATURATED
MODEL OPERATION
Although designed for estimating chemical flux volatilizing from the soil to air, the Jury
model also predicts concentrations within the aqueous phase and can be used to estimate
contaminant mass loading through the unsaturated, or vadose, zone to groundwater over time.
The hydrology portion of the model is very simple to use and uniform and steady infiltration is
assumed.
Other assumptions to consider include the assumption of homogeneous and isotropic soil
(without depth variation), uniform chemical distribution within the source area, and
compositional equilibrium between all phases at all times. These assumptions limit the model's
usefulness. The model is most appropriate for simulating time-varying volatile flux from soil but
it may also be used for initial-tier evaluations of mass loading to groundwater. In such cases, the
infiltration rate is a sensitive parameter and the results should be compared to other screening-
level model predictions.
SENSITIVE INPUT PARAMETERS
Total soil porosity
Source area concentration
Soil-water sorption coefficient
Fraction of organic carbon
Carbon-water sorption coefficient
KEY INPUT PARAMETERS
Effective solubility
Retardation factor
Unsaturated hydraulic conductivity
First order decay rate
Volumetric air content in vadose zone soil
Soil bulk density
Volumetric water content in vadose zone soil
Henry's law constant
Dilution factor
Mixing zone depth
Source width parallel to groundwater movement
APPLICABILITY
Tested model which is very simple to operate.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
16
-------
MOFAT
MODEL OPERATION
Features are:
Simulate multiphase transport of up to five non-inert chemical species.
Model flow of light or dense organic liquids in three fluid phase systems.
Handles cases in which gas and/or NAPL phase are absent in part or all of the domain
at any given time.
Solve flow equations for phases exhibiting transient behavior using the ASD method.
Simulate dynamic or passive gas as a full three-phase flow problem.
Use a three-phase van Genuchten model for saturation-pressure-permeability
relations.
Handle flux type, specified head, specified concentration or mixed type boundary
conditions.
Consider hysteresis in oil permeability due to fluid entrapment.
Model water flow, transport, coupled oil-water flow, or water-oil-gas flow.
KEY INPUT PARAMETERS SENSITIVE INPUT PARAMETERS
Fluid properties Initial contaminant concentrations
Boundary condition data Equilibrium partition coefficients
Porous media dispersivities Soil hydraulic properties
Diffusion coefficients
Mass transfer coefficients
Time integration parameters
Mesh geometry
Initial water phase concentrations
Component densities
First-order decay coefficients
APPLICABILITY
Applicable for multi-phase flow and transpot of three fluid phases. Written in DOS.
ADDITIONAL INFORMATION MODEL OPERATION
Scientific Software Group
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-
3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.html
17
-------
VS2DT
MODEL OPERATION
VS2DT is a U.S.G.S. program for flow and solute transport in variably saturated, single-
phase flow in porous media. A finite-difference approximation is used to solve the advection-
dispersion equation. Simulated regions include one-dimensional columns, two-dimensional
vertical cross sections, and axially symmetric, three-dimensional cylinders. Program options
include backward or centered approximations for both space and time derivatives, first-order
decay, equilibrium adsorption (Freundlich or Langmuir) isotherms, and ion exchange. Nonlinear
storage terms are linearized by an implicit Newton-Raphson method. Relative hydraulic
conductivity is evaluated at cell boundaries using full upstream weighting, arithmetic mean or
geometric mean. Saturated hydraulic conductivities are evaluated at cell boundaries using
distance-weighted harmonic means.
KEY INPUT PARAMETERS
Thickness of affected soil zone
Dispersivities
Hydraulic conductivity
First-order decay rate
APPLICABILITY
SENSITIVE INPUT PARAMETERS
Source area concentration
Soil-water sorption coefficient
Total soil porosity
Organic carbon content
Carbon-water sorption coefficient
This model was developed and tested by the U.S.G.S., not widely used.
ADDITIONAL INFORMATION MODEL OPERATION
Scientific Software Group
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone:(703)620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center U.S. Geological Survey
" " water.usgs.gov/software
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
American Petroleum Institute
www.api.org/ehs
18
-------
-------
GROUNDWATER TO AMBIENT AIR
-------
-------
FARMER
MODEL OPERATION
The model that is used in the ASTM approach to estimate the contaminant flux term from
groundwater to ambient air is the Farmer model. It assumes that that the contaminated
groundwater is located at some depth beneath the ground surface. The model also assumes that
the concentration in the groundwater and the depth to the groundwater do not change with time.
This is equivalent to assuming that the groundwater represents an infinite source for
contamination. The model assumes that the water in the capillary fringe is "clean." The capillary
fringe is assumed to have a relatively high moisture content and a relatively low air-filled
porosity. The effect of this capillary fringe is to reduce the diffusion coefficient. It can be seen
that a relatively thin capillary fringe can significantly reduce the rate of vapor diffusion to the
ground surface.
KEY INPUT PARAMETERS
Source area concentration
Diffusion coefficient in air
Diffusion coefficient in water
Fraction of organic carbon
Henry's Law constant
Carbon-water sorption coefficient
Soil-water sorption coefficient
Total soil porosity
APPLICABILITY
Soil bulk density
Depth to groundwater contamination
Thickness of groundwater contamination
Averaging time for fluxes
Soil intrinsic permeability
Volumetric water content in vadose zone soil
Volumetric air content in vadose zone soil
This model is highly tested and used especially for screening purposes dues to its
conservative assumptions.
ADDITIONAL INFORMATION
ASTM 173 9-95 Risk-based Corrective Action Guidance
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
19
-------
-------
GROUNDWATER TO INDOOR AIR
-------
-------
FARMER
MODEL OPERATION
The contaminant flux term for migration from groundwater to an enclosed space is based on
the same model that is used to simulate migration from groundwater to ambient air (i.e., the
Farmer model). The equations for estimating the flux from groundwater to enclosed spaces
include the effects of degradation.
The flux is an average over time. The effects of a capillary fringe are also included through
the modified diffusion coefficient, Dwseff.
KEY INPUT PARAMETERS
Source area concentration
Fraction of organic carbon
Henry's Law constant
Thickness of groundwater contamination
Soil-water sorption coefficient
Volumetric air content in vadose zone soil
Effective diffusion coefficient through the crack
Effective radius of crack
Thickness of the foundation or floor of the enclosed
space or building
Area of cracks through which vapor enter the enclosed
space or building
Soil bulk density
Depth to groundwater contamination
Carbon-water sorption coefficient
Averaging time for fluxes
Soil intrinsic permeability
Floor/wall seam perimeter
Viscosity of gas
Total soil porosity
Building under pressure
Depth of crack below ground surface
Diffusion coefficient in air
Diffusion coefficient in water
SENSITIVE INPUT PARAMETERS
Enclosed space volume/infiltration area ratio
Ventilation rate for the enclosed space or building
APPLICABILITY
This model is used especially for screening purposes dues to its conservative assumptions.
ADDITIONAL INFORMATION
ASTM 1739-95 Risk-based Corrective Action Guidance
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet environment.
Computer programs for the model are currently not available from common sources.
20
-------
JOHNSON/ETTINGER (modified)
MODEL OPERATION
The Johnson/Ettinger Model is modified to include migration of contaminants from
groundwater sources. The model consists of five fundamental steps:
1. Calculation of the ratio of the soil vapor phase concentration to total concentration at
the source.
2. Calculation of the effective diffusion coefficient.
3. Calculation of the infiltration rate of contaminant vapors into the building.
4. Calculation of the building vapor concentration to groundwater vapor source
concentration ratio.
5. Back-calculation of the generic groundwater to indoor air inhalation criteria.
The model incorporates the following assumptions:
Soil is homogenous such that the effective diffusion coefficient is constant.
Contaminant loss from leaching downward does not occur.
Source degradation and transformation is not considered.
Concentration at the soil particle surface/soil pore air space interface is zero.
Convective vapor flow near the building foundation is uniform.
Contaminant vapors enter the building through openings in the walls and foundation
at or below grade.
Convective vapor flow rates decrease with increasing contaminant source-building
distance.
All contaminant vapors directly below the building will enter the building, unless the
floor and walls are perfect vapor barriers.
The building contains no other contaminant sources or sinks; well mixed air volume.
KEY INPUT PARAMETERS
Effective diffusion coefficient through the crack
Effective diffusion coefficient through capillary fringe
Effective diffusion coefficient through vadose zone
Thickness of vadose zone below enclosed space floor
Thickness of capillary fringe
Building foundation thickness
Crack depth below grade to bottom of enclosed floor
space
Crack radius
Depth below grade to bottom of enclosed space floor
Building floor length/width/height
SENSITIVE INPUT PARAMETERS
Ventilation rate for the enclosed space or building
Vapor flow rate into the building
Source-building separation distance for groundwater
APPLICABILITY
This model is widely tested and used especially for screening purposes due to its conservative
assumptions.
ADDITIONAL INFORMATION
Michigan department of Environmental Quality
SOURCES
Model is in the form of equations which are typically executed in a spreadsheet
environment. Computer programs for the model are currently not available from common
sources.
21
-------
GROUNDWATER TRANSPORT
-------
-------
DOMENICO
MODEL OPERATION
The Domenico Model is a mathematical solution of the advection-dispersion equation using
many simplifying assumptions. Several of the simplifying assumptions are:
ground-water transport is one-dimensional along the center-line, between the source and the
receptor
dispersion is quantified in three-dimensions
the solution includes error functions that provide approximate solutions for ground-water
transport equations across the site, over time
source area concentrations are constant
aquifer is initially clean.
The Domenico equation error functions are used to approximate the integration of the
ground-water transport differential equation. In order to solve this equation, an integration
scheme such as the Gauss-Legendre quadrature method could be used (Ungs, 1997).
SENSITIVE INPUT PARAMETERS
Source concentration
Retardation coefficient
Enclosed space volume/infiltration area ratio
Distance to receptor
KEY INPUT PARAMETERS
Source width
Source depth
First order decay rate
Longitudinal dispersivity
Transverse-horizontal dispersivity
Transverse-horizontal dispersivity
APPLICABILITY
The Domenico Model is a straight forward mathematical solution of the advection-dispersion
equation using many simplifying assumptions. The models AT123D and VADSAT also satisfy
the conditions of one direction uniform advection, three dimensional dispersion, and first-order
decay.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
ASTM RBCA guidance
GSI Tier 2 Tool Kit
SOURCES
Groundwater Services, Inc.
2211 Norfolk, Suite 1000
Houston, Texas 77098-4044
Phone: (713) 522-6300
Fax:(713)522-8010
22
-------
FATES
MODEL OPERATION
FATE 5 is a modification of the Domenico analytical groundwater transport model. The model
allows calibration to site conditions and both prediction of down gradient concentration and back
calculation of SSTLs. Key assumptions of the model are;
The aquifer and flow field are homogeneous and isotropic.
Groundwater flow is fast enough that molecular diffusion can be ignored,
Adsorption is a linear, reversible process.
Assumes simple groundwater flow conditions.
Based on steady-state formulation of the Domenico model.
Not applicable where vertical gradients affect contaminant transport.
Assumes simple first-order decay.
KEY INPUT PARAMETERS SENSITIVE INPUT PARAMETERS
Source width Source concentration
Source depth Retardation coefficient
First order decay rate Enclosed space volume/infiltration area ratio
Longitudinal dispersivity Distance to receptor
Transverse-horizontal dispersivity
Transverse-horizontal dispersivity
APPLICABILITY
FATE 5 is designed to predict the extent of contaminant plumes in the absence of further
source control and to determine the site specific steady-state rate of chemical decay.
ADDITIONAL INFORMATION
Groundwater Services, Inc. -
SOURCES
Groundwater Services, Inc.
2211 Norfolk, Suite 1000
Houston, Texas 77098-4044
Phone:(713)522-6300
Fax:(713)522-8010
23
-------
DISPERSE
MODEL OPERATION
Disperse is an advection/dispersion model developed to predict the size and duration of methyl
tertiary butyl ether (MTBE) and tertiary butyl alcohol (TEA) plumes. The model is conservative
and represents the potential worst case scenario. The model assumes:
Finite source, contaminate introduced as a slug
Contaminant does not degrade
Contaminant does not absorb to soil
Aquifer is horizontal and homogenous
Velocity is constant
Dispersion coefficients are constant and proportional to velocity
SENSITIVE INPUT PARAMETERS
Distance to exposure point parallel to direction of flow
Initial concentration
Groundwater velocity
KEY INPUT PARAMETERS
Rate of discharge
Period of discharge
Mass discharge
Longitudinal dispersivity
Transverse dispersivity
Time
Distance to exposure point perpendicular to direction of
flow
APPLICABILITY
The model provides an analytical solution of the classic dispersion equation for bi-
dimensional flow in a horizontal aquifer.
ADDITIONAL INFORMATION
New Jersey Department of Environmental Protection
SOURCES
Software available from New Jersey Department of Environmental Protection.
24
-------
SOLUTE
MODEL OPERATION
SOLUTE is a set of five programs based on analytical solutions of the advective-dispersive
transport equation for solutes. All SOLUTE programs facilitate menu-driven, interactive data
entry and editing, and results are given tabular and graphic form, including contour plots and line
graphs.
The five programs include one dimensional and radial symmetric models to simulate the
effects of a single source of contaminants, and two- and three-dimensional models that support
multiple point sources using the principal of superposition to calculate the accumulated effects of
various sources or to represent line (strip) or areal (patch) sources. These multiple sources may
have a different starting time and may be of limited duration. All models support advection and
dispersion, and the one-, two-, and three-dimensional models support retardation and decay. The
radial symmetric models handle only retardation. The programs use either consistent metric
units or a system of English units. The individual programs are:
ONED-1: One-dimensional solute transport in a semi-infinite area with constant
concentration as inlet boundary condition.
ONED-2: Same as ONED-1 with decaying source as inlet boundary condition.
ONED-3: Same as ONED-1 with concentration-dependent mass flux as inlet
boundary condition.
PLUME-2D: Two-dimensional areal or cross-sectional transport of a plume from one
or more limited duration point sources in a uniform groundwater flow field.
PLUME-3D: Same as PLUME -2D for three-dimensional transport
SLUG-2D: Two-dimensional areal or cross-sectional transport of a slug caused by
one or more instantaneous point sources in a uniform groundwater flow field.
SLUG-3D: Same as SLUG-2D for three-dimensional transport.
RADIAL: Solute transport in a plane radial flow field.
LTIRD: Same as RADIAL but no retardation.
KEY INPUT PARAMETERS
Longitudinal, transverse, and vertical dispersivity
Aquifer thickness
SENSITIVE INPUT PARAMETERS
Groundwater seepage velocity
Contaminant concentration at the source
Duration of solute pulse
First-order decay rate
Retardation factor
APPLICABILITY
The model has been thoroughly tested with accurate results.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
Scientific Software Group
SOURCES
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
25
-------
MULTIMED
MODEL OPERATION
MULTIMED, Multimedia Assessment Model, is a user-friendly model which simulates the
fate and transport of contaminants leaching from a waste disposal facility into the multimedia
environment. Release to either air or soil, including the unsaturated and saturated zone, and
possible interception of the subsurface contaminant plume by a surface stream are included in the
model. The model includes two options for simulating leachate flux. Either the infiltration rate to
the unsaturated or saturated zone can be specified directly or a landfill module can be used to
estimate the infiltration rate. The landfill module is one-dimensional and steady-state, and
simulates the effect of precipitation, runoff, infiltration, evapotranspiration, barrier layers (which
can include flexible membrane liners), and lateral drainage.
A steady-state, one-dimensional, semi-analytical module simulates flow in the unsaturated
zone. The output from this module, water saturation as a function of depth, is used as input to the
unsaturated zone transport module. The latter simulates transient, one-dimensional (vertical)
transport in the unsaturated zone and includes the effects of longitudinal dispersion, linear
adsorption, and first-order decay. Output from the unsaturated zone modules is used to couple the
unsaturated zone transport module with the steady-state or transient, semi-analytical saturated
zone transport module. The latter includes one-dimensional uniform flow, three-dimensional
dispersion, linear adsorption, first-order decay, and dilution due to direct infiltration into the
groundwater plume. Contaminant of a surface stream due to the complete interception of a
steady-state saturated zone plume is simulated by the surface water module. The air emissions
and the atmosphere dispersion modules simulate the movement of chemicals into the air.
SENSITIVE INPUT PARAMETERS
Saturated hydraulic conductivity
Hydraulic gradient
Sorption coefficients
Initial concentration
Well distance from the site
Organic carbon content
KEY INPUT PARAMETERS
Porosity
Depth of unsaturated zone
Residual water content
Biological decay rate
Soil bulk density
Recharge rate
Area of waste unit
Infiltration rate
Duration of pulse
Source decay rate
Number and thickness of each layer
Dispersivities
APPLICABILITY
The model has been thoroughly tested with accurate results.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
Scientific Software Group
SOURCES
Groundwater Services, Inc.
2211 Norfolk, Suite 1000
Houston, Texas 77098-4044
Phone: (713) 522-6300
Fax:(713)522-8010
26
-------
SUMMERS
MODEL OPERATION
SUMMERS is a screening level interactive computer program for estimating soil cleanup
levels. The model assumes that a percentage of rainfall at a polluted site will infiltrate and desorb
contaminants from the soil based on equilibrium soil-water partitioning. Using a mass balance
approach and assuming equilibrated, complete mixing in the aquifer, the soil cleanup level is
calculated from the original soil concentration, the concentration of the infiltrating water, and an
equilibrium coefficient.
The public domain SUMMERS model was developed to estimate when contaminant
concentrations in the soil will produce aquifer contaminant concentrations above acceptable
levels. The resulting soil concentrations can then be used as guidelines in estimating boundaries
or extent of soil contamination by applying the derived maximum soil contaminant concentration
level to the observed concentration in the soil at the site.
KEY INPUT PARAMETERS
Target concentration in groundwater
Downward porewater velocity
Void fraction
Width of spill perpendicular to flow
Equilibrium partition coefficient
Volumetric infiltration rate into aquifer
Horizontal area of spill
Darcy velocity in aquifer
Volumetric groundwater flow rate
APPLICABILITY
Highly used and simple model for screening purposes.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
SOURCES
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
SENSITIVE INPUT PARAMETERS
Initial concentration
Groundwater seepage velocity
27
-------
BIOSCREEN
MODEL OPERATION
BIOSCREEN is an easy-to-use screening model which simulates remediation through natural
attenuation (RNA) of dissolved hydrocarbons at petroleum fuel release sites. The software,
programmed in the Microsoftฎ Excel spreadsheet environment and based on the Domenico
analytical solute transport model, has the ability to simulate advection, dispersion, adsorption,
and aerobic decay, as well as anaerobic reactions that have been shown to be the dominant
biodegradation processes at many petroleum release sites. BIOSCREEN includes three different
model types: 1) solute transport without decay; 2) solute transport with biodegradation modeled
as a first order decay process (simple, lumped-parameter approach), and 3) solute transport with
biodegradation modeled as an "instantaneous" biodegradation reaction (approach used by
BIOPLUME models). The model is designed to simulate biodegradation by both aerobic and
anaerobic reactions. It was developed for the Air Force Center for Environmental Excellence
(AFCEE) Technology Transfer Division at Brooks Air Force Base by Groundwater Services,
Inc., Houston, Texas.
KEY INPUT PARAMETERS
Depth below water table
Lateral distance from center line of plume
Specific discharge
Porosity
Dissolved oxygen
Saturated thickness
Transmissivity
Leakance, between aquifer layers, vertical conductivity
Storativity, storage coefficient
Recharge
APPLICABILITY
Longitudinal dispersivity
Transverse dispersivity
Vertical dispersivity
Anions/cations
First-order degradation constant
SENSITIVE INPUT PARAMETERS
Source area contaminant concentrations
Saturated hydraulic conductivity
Distance along the center line from downgradient edge
of dissolved plume source zone
Easy screening tool, can be used for natural attenuation simulations.
ADDITIONAL INFORMATION
EPA Soil Screening Guidance (find at http://www.ntis.gov/search.htm)
SOURCES
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.html
28
-------
VADSAT
MODEL OPERATION
The VADSAT model is a 3-D transport model which simulates contaminant leaching and
volatilization in the vadose zone and advective/dispersive transport in the saturated zone. The
model considers:
A well-mixed finite-mass source zone
Pseudo steady-state volatilization and diffusive transport from the source to ground surface
Leaching from the source zone to groundwater
Dissolved-phase advection and dispersion in groundwater
Adsorption
First-order decay in the leachate
Van Genucten's algorithm to estimate moisture content
Simulate transport of individual contaminants that are part of a mixture
Presence of residual level hydrocarbons
Ability to make both deterministic and Monte Carlo simulations
KEY INPUT PARAMETERS
Porosity
Van Genucten's n parameter
Soil bulk density
Molecular weight of chemical and TPH mixture
Organic carbon partition coefficient for chemical
Henry's Law constant
Irreducible water content
Fraction organic carbon
Diffusion coefficients in air and water
Degradation rate
SENSITIVE INPUT PARAMETERS
Hydraulic conductivity
APPLICABILITY
Tested model which is very simple to operate.
ADDITIONAL INFORMATION
API's VADSAT Manual
BP RISC Manual, as incorporated in RISC has the extended capability to consider a lens
between the source and ground surface with difference soil properties.
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
Environmental Systems & Technologies, Inc. American Petroleum Institute
2608 Sheffield Drive www.api.org/ehs
Blacksburg, VA 24060
Phone: (540) 552-0685
Fax:(540)951-5307
www.esnt.com
29
-------
MODFLOW
MODEL OPERATION
MODFLOW is the name that has been given the USGS Modular Three-Dimensional Flow
Model. Because of its ability to simulate a wide variety of systems, its extensive publicly
available documentation, and its rigorous USGS peer review, MODFLOW has become the
worldwide standard ground-water flow model. It is a flow model only with no mass transport
component. It is used to simulate systems for water supply, containment remediation and mine
dewatering. When properly applied, it is the recognized standard model used by courts,
regulatory agencies, universities, consultants and industry.
The main objectives in designing MODFLOW were to produce a program that can be readily
modified, is simple to use and maintain, can be executed on a variety of computers with minimal
changes, and has the ability to manage the large data sets required when running large problems.
Ground-water flow within the aquifer is simulated using a block-centered finite-difference
approach. Layers can be simulated as confined, unconfined, or a combination of both. Flows
from external stresses such as flow to wells, areal recharge, evapotranspiration, flow to drains,
and flow through riverbeds can also be simulated. MODFLOW is most appropriate in those
situations where a relatively precise understanding of the flow system is needed to make a
decision. MODFLOW was developed using the finite-difference method. The finite-difference
method permits physical explanation of the concepts used in construction of the model.
Therefore, MODFLOW is easily learned and modified to represent more complex features of the
flow system.
To use MODFLOW, the region to be simulated must be divided into cells with a rectilinear
grid resulting in layers, rows and columns. Files must then be prepared that contain:
KEY INPUT PARAMETERS
Specific yield Recharge from precipitation, rivers, drains
Pumping wells
Initial groundwater heads SENSITIVE INPUT PARAMETERS
Transmissivity
Boundary conditions Hydraulic conductivity
APPLICABILITY
The most widely used groundwater flow model in the world.
ADDITIONAL INFORMATION
International Groundwater Modeling Center.
SOURCES
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.html
30
-------
PLASM
MODEL OPERATION
PLASM, Prickett Lonnquist Aquifer Simulation Model (PLASM) was first published in 1971
by the Illinois State Water Survey. It consists of three finite-difference simulation programs and
a preprocessor. The programs simulate two-dimensional nonsteady flow of ground-water in
heterogeneous anisotropic aquifers under water table, nonleaky, and leaky confined conditions.
Included are options for time-varying pumpage from wells, induced infiltration from streams or
shallow aquifers, and water-table-depth-dependent evapotranspiration. The finite-difference
equations are solved using a modified alternating direction method.
Transverse dispersivity
Vertical dispersivity
First-order degradation constant
Time since release
Source width
Source depth
SENSITIVE INPUT PARAMETERS
Source area concentration
Hydraulic gradient
Distance along the center line from downgradient edge
of dissolved plume source zone
KEY INPUT PARAMETERS
Volumetric water content in saturated zone
Depth below water table
Lateral distance from center line of plume
Specific discharge
Saturated hydraulic conductivity
Porosity
Saturated thickness
Transmissivity
Storativity, storage coefficient
Leakance, between aquifer layers, vertical conductivity
Recharge
Longitudinal dispersivity
APPLICABILITY
Tested and validated but not as widely used due to development of more advanced numerical
models like MODFLOW.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
SOURCES
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
31
-------
MOC
MODEL OPERATION
This model simulates solute transport in flowing ground water. The model is both general
and flexible in that it can be applied to a wide range of problem types. It is applicable for one- or
two-dimensional problem involving steady state or transient flow. The model computes changes
in concentration over time caused by the processes of convective transport, hydrodynamic
dispersion, and mixing (or dilution) from fluid sources. The model assumes that gradients of
fluid density, viscosity and temperature do not affect the velocity distribution. However, the
aquifer may be heterogeneous and/or anisotropic. The model is based on a rectangular, block-
centered, finite-difference grid. It allows the specification of injection or withdrawal wells and of
spatially varying diffuse recharge or discharge, saturated thickness, transmissivity, boundary
conditions and initial heads and concentrations. MOC incorporates: first-order irreversible rate-
reaction; reversible equilibrium controlled sorption with linear, Freundlich, or Langmuir
isotherms; and reversible equilibrium-controlled ion exchange for monovalent or divalent ions.
The model couples the ground-water flow equation with the solute-transport equation. The
program uses an alternating-direction implicit procedure to solve a finite-difference
approximation to the ground-water flow equation, and it uses the method of characteristics to
solve the solute-transport equation. The latter uses a particle tracking procedure to represent
convective transport and a two-step explicit procedure to solve a finite-difference equation that
describes the effects of hydrodynamic dispersion, fluid sources and sinks, and divergence of
velocity. This explicit procedure has several stability criteria, but the consequent time-step
limitations are automatically determined by the program.
KEY INPUT PARAMETERS
Specification of injection or withdrawal wells
Saturated thickness
Boundary conditions
Specification varying diffuse recharge or discharge
Transmissivity
APPLICABILITY
Limited application and cumbersome to use. However, verified and tested by U.S.G.S.
ADDITIONAL INFORMATION ' .
International Groundwater Modeling Center
SENSITIVE INPUT PARAMETERS
Initial concentrations
Initial heads
Scientific Software Group
P.O. Box 23041
Washington, B.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center U.S. Geological Survey
Colorado School of Mines water.usgs.gov/software
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
32
-------
BIOPLUME II/III
MODEL OPERATION
BIOPLUME II is a two-dimensional model the simulates the transport of contaminants in
groundwater under conditions of oxygen limited biodegradation. The model provides for
convective transport, dispersion, fluid source or sinks, chemical (nitrate, iron, sulfate) and
physical reactions (first order decay), and three potential sources of oxygen. BIOPLUME III
simulates the biodegradation of organic contaminants using a number of aerobic and anaerobic
electron acceptors: oxygen, nitrate, iron (III), sulfate, and carbon dioxide. The model solves the
transport equation six times to determine the fate and transport of the hydrocarbons and the
electron acceptors/reaction by-products. For the case where iron (II) is used as an electron
acceptor, the model simulates the production and transport of iron (II). BIOPLUME III runs in a
Windows 95ฎ environment whereas BIOPLUME II was mainly developed in a DOS
environment.
SENSITIVE INPUT PARAMETERS
Groundwater velocity
Contaminant concentration
Contaminant retardation factor
Natural organic carbon concentration
KEY INPUT PARAMETERS
Oxygen concentration
Contaminant utilization rate
Contaminant half saturation constant
First order decay rate
Microbial concentration
Microbial yield coefficient
Ratio of oxygen to contaminant consumed
Oxygen half saturation constant
Microbial decay rate
APPLICABILITY
An extremely versatile model which allows the simulation of hydrocarbon plumes
undergoing biodegradation.
ADDITIONAL INFORMATION
CSMoS
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-
3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.html
International Groundwater
Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
33
-------
RANDOM WALK
MODEL OPERATION
RANDOM Walk is a generalized FORTRAN computer code for simulation of two-
dimensional ground-water flow and solute transport, written by T.A. Prickett, etal. and released
in 1981 by the Illinois State Water Survey (ISWS). Ground-water flow is simulated using either
analytical solutions or a two-dimensional version of the PLASM finite difference model. The
solute transport portion of the code is based on a particle-in-a-cell technique for the convective
mechanisms and a random-walk technique for the dispersion effects. The model also handles
first-order decay, linear equilibrium sorption (retardation), and zero-order production.
RANDOM WALK is a DOS-based program that can simulate two-dimensional
nonsteady/steady flow problems in heterogeneous aquifers under water table and/or artesian or
leaky artesian conditions. Furthermore, the program covers time-varying pumpage or injection
by wells, natural or artificial recharge, the flow relationships between surface water and ground-
water, evapotranspiration, conversion of storage coefficients from artisan to water table
conditions, and flow from springs. The program allows injection of solute by wells, leachate
entering the aquifer from landfills or surface spills, location of a vertically averaged solute front
representing salt water intrusion, leakage of water from overlying source beds with different
water quality than the aquifer, and specification of concentrations along surface water boundaries
to reflect their water quality.
KEY INPUT PARAMETERS
Volumetric water content in saturated zone
Depth below water table
Lateral distance from center line of plume
Specific discharge
Porosity
Saturated thickness
Transmissivity
Storativiry, storage coefficient
Leakance, between aquifer layers, vertical conductivity
First-order degradation constant
Time since release
Source width
Source depth
Recharge
Longitudinal dispersivity
Transverse dispersivity
Vertical dispersivity
SENSITIVE INPUT PARAMETERS
Source area concentration
Saturated hydraulic conductivity
Hydraulic gradient
Distance along the center line from downgradient edge
of dissolved plume source zone
APPLICABILITY
Tested and validated but not as widely used due to development of more advanced numerical
models like MT3D.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
SOURCES
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
34
-------
MT3D
MODEL OPERATION
The most current version of MT3D96 is a numerical simulation code that models the fate and
transport of dissolved, single-species contaminants in saturated ground-water systems. MT3D96
calculates concentration distributions, concentration histories at selected receptor points and
hydraulic sinks (for example, extraction wells), and the mass of contaminants in the ground-
water system. The code can simulate three-dimensional transport in complex steady-state and
transient flow fields and can represent anisotropic dispersion, source-sink mixing processes, first-
order transformation reactions and linear and nonlinear sorption.
KEY INPUT PARAMETERS
Depth below water table
Lateral distance from center line of plume
Specific discharge
Saturated thickness
Transmissivity
Leakance, between aquifer layers, vertical conductivity
Storativity, storage coefficient
Recharge
First-order degradation constant
APPLICABILITY
Longitudinal dispersivity
Transverse dispersivity
Vertical dispersivity
SENSITIVE INPUT PARAMETERS
Source area concentration
Saturated hydraulic conductivity
Porosity
Distance along the center line from downgradient
edge of dissolved plume source zone
MT3D96 is widely accepted by regulators and the ground-water consulting and research
communities and has been used to model thousands of sites.
ADDITIONAL INFORMATION
Scientific Software Group
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-
3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
Robert S Kerr Environmental Research Center
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, Oklahoma 74821-1198
Phone: (580) 436-8586
Fax:(580)436-8718
www.epa.gov/ada/models.html
International Groundwater
Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
35
-------
AT123D
MODEL OPERATION
AT123D, analytical, transient One-, Two-, and Three-Dimensional Model, is an analytical
ground-water transport model. AT123D computes the spatial-temporal concentration distribution
of wastes in the aquifer system and predicts the transient spread of a contaminant plume through
a ground-water aquifer. The fate and transport processes accounted for are advection, dispersion,
adsorption, and decay. AT123D estimates all the above components at a user defined time
interval for up to 99 years of simulation time.
AT123D can be used as an assessment tool to help the user estimate the dissolved
concentration of a chemical in three-dimensions in ground water resulting from a mass release
over a source area. AT123D can handle: two kinds of source releases-instantaneous, continuous
with a constant loading or time-varying releases; three types of waste-radioactive, chemicals,
heat; four types of source configurations-a point source, a line source parallel to the x-, y-, z-axis,
and-area source perpendicular to the z-axis, a volume source; four variations of the aquifer
dimensions-finite depth and finite width, finite depth and infinite width, infinite depth and finite
width, infinite depth and infinite width.
KEY INPUT PARAMETERS
Bulk density
Dispersivities in x, y, and z directions
First-order decay rate
Molecular diffusion coefficient
Heat exchange
APPLICABILITY
Widely used and U.S.G.S. approved.
ADDITIONAL INFORMATION
International Groundwater Modeling Center
Scientific Software Group
SENSITIVE INPUT PARAMETERS
Hydraulic conductivity
Porosity
Hydraulic gradient
Sorption coefficients
Distance to receptor
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, B.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
36
-------
MODPATH
MODEL OPERATION
MODPATH is a particle tracking post-processing package that was developed to compute
three-dimensional flowpaths using output from steady-state or transient ground-water flow
simulations by MODFLOW. MODPATH uses a semi-analytic particle tracking scheme that
allows an analytical expression of the particle's flow to be obtained within each finite-difference
grid cell. Particle paths are computed by tracking particles from one cell to the next until the
particle reaches a boundary, an internal sink/source, or satisfies some other termination criterion.
Data input for MODPATH is a combination of data files and interactive keyboard input.
Output from steady-state or transient MODFLOW simulations is used in MODPATH to
compute paths for imaginary "particles" of water moving through the simulated ground-water
system. In addition to computing particle paths, MODPATH keeps track of the time of travel for
particles moving through the system. By carefully defining the starting locations of particles, it is
possible to perform a wide range of analyses such as delineating capture and recharge areas or
drawing flow nets.
KEY INPUT PARAMETERS
Lateral distance from center line of plume
Specific discharge
Transmissivity
Leakance, between aquifer layers, vertical conductivity
Depth below water table
Saturated thickness
Storativity, storage coefficient
Longitudinal dispersivity
Transverse dispersivity
Vertical dispersivity
APPLICABILITY
Tested and validated by U.S.G.S.
ADDITIONAL INFORMATION
Scientific Software Group
International Groundwater Modeling Center
Recharge
First-order degradation constant
SENSITIVE INPUT PARAMETERS
Source area concentration
Saturated hydraulic conductivity
Porosity
Distance along the center line from downgradient
edge of dissolved plume source zone
SOURCES
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
Phone: (703) 620-9214
Fax: (703) 620-6793
www.scisoftware.com
International Groundwater Modeling Center
Colorado School of Mines
Golden, Colorado 80401-1887
Phone:(303)273-3103
Fax: (303) 384-2037
www.mines.edu/igwmc
37
-------
A , * * * V * ,
_ 1
-------
------- |