c/EPA
United States
Environmental Protection
Agency
Office of Solid Waste
and Emergency Response
(OS-110)
EPA-500-B-92-006
October 1992
Ground-Water Modeling
Compendium
Model Fact Sheets, Descriptions, Applications
and Assessment Framework
OSWER Information Management
Rscydsdfflscyclabls
Printed on paper that contains
at toot 50% rwvctad «>«
-------
Purpose of This Compendium
The Ground-Water Modeling Compendium increases the reader's
awareness of ground-water models and modeling in general. The
Compendium also provides a convenient source of information on
overseeing modeling projects.
The Compendium provides an Assessment Framework for planning
and assessing modeling applications, summary descriptions of
model applications that were used to test the Assessment Framework |
and summary and detailed descriptions of four ground-water
models.
The use of this Compendium by technical support staff and remedial
project managers will help to promote the appropriate use of models
and therefore sound and defensible modeling within the hazardous
waste/Superfund programs.
Who May Benefit From This Compendium
This Compendium is intended for use by:
Technical Support Staff
Remedial Project Managers
This Compendium may also help:
OSWER and Regional Management
EPA Contractors
Other Consultants
U.S. Environmental Protection Agency
?5g,!°n 5, Library (PL-12J)
77 West Jackson Bpulevard, 12th
Chicago, IL 60604-3590
-------
Table of Contents
Section 1.0 Introduction
Introduction 1-1
Section 2.0 Assessment Framework for Ground-Water Model
Applications
Introduction 2-1
Assessment Criteria 2-2
Footnotes 2-15
Glossary of Technical Terms 2-17
EPA Publications Related to Ground-Water Modeling 2-24
Section 3.0 - Model Applications
Introduction 3-1
Summary Description #1 3-3
Summary Description #2 3-11
Summary Description #3 3-25
Summary Description #4 3-39
Section 4.0 Model Descriptions
Introduction 4-1
Model Fact Sheets 4-3
Detailed Model Descriptions 4a
MOC 4a-l
MODFLOW 4b-l
PLASM 4c-l
RANDOM WALK 4d-l
-------
1.0 Introduction
-------
1.0
Introduction
Background
This Ground-Water Modeling Compendium has been prepared as part of the
Models Management Initiative being conducted by EPA's Office of Solid Waste and
Emergency Response (OSWER). OSWER's Resource Management and Information
Staff (RMIS) is directing this effort in order to promote improved usage of
environmental models initially focusing on hazardous waste/Superfund
programs, and in the future, supporting Agency-wide efforts.
Objective and Intended Use
During FY 1991-92, OSWER has been conducting a pilot project on ground-
water modeling with the primary objective of providing useful information on
existing modeling practices and models to EPA staff, contractors, and the regulated
community. OSWER recognizes that ground-water models are being used in a
variety of ways and under different circumstances and constraints. Models can be
used to guide and complement field investigations, thereby improving the
understanding of the consequences of site-specific hydrogeologic conditions.
However, models should not be used in lieu of field investigations and care must be
taken to ensure that models are not misused. The intention of this Compendium is
to:
O promote the appropriate use of models by increasing users' awareness
about the strengths, weaknesses and inherent uncertainties associated
with ground-water models and modeling in general; and
O support model users and decision-makers by providing a convenient
source of information on how to oversee modeling projects, how certain
models have been applied in the context of hazardous waste/Superfund
programs, and the characteristics of four specific ground-water models.
The contents of the Compendium have been reviewed extensively by ground-
water modeling experts both within and external to EPA. OSWER recommends this
document as a reference source for promoting sound and defensible modeling
methods and approaches. This document does not, however, constitute official EPA
guidance, nor should it be construed as an endorsement of specific models.
Organization and Structure
The Compendium is organized into four distinct sections designed to meet the
needs of various audiences, including project/site managers, technical reviewers,
and model users. The four sections of the Compendium are:
Page 1-1
-------
Section 1s Introduction
O 1.0, this section, is the Introduction describing the Compendium's
purpose and organization.
G 2.0, Assessment Framework for Ground-Water Model Applications,
provides a framework for planning and assessing model applications. The
framework is organized into eight categories (e.g., Project Management,
Model Setup and Input Estimation). Each category contains a series of
criteria that can be used to assess a model application after the fact (i.e., by
using existing documentation and the knowledge of the participants to
determine how actual modeling practices compare to the criteria), or to
guide a team in an ongoing or future application (i.e., by conducting the
project in the manner indicated by the criteria). This framework has been
developed by the pilot project team in conjunction with recognized
modeling experts. Also included in Section 2.0 are footnotes for some of
the criteria, a glossary of technical terms, and a list of potential sources of
information on EPA modeling guidance.
CJ 3.0, Model Applications, provides summary descriptions of the model
applications used to test the assessment framework. These tests were
performed to gain insight into the effectiveness of the assessment
framework in different modeling situations. Each description discusses
the decision objective, regulatory context, site characteristics (e.g., geology,
ground-water contamination), modeling activities and results, interesting
features of the application, and the names of EPA staff to contact for more
details. This section is intended to help readers understand the variety of
modeling applications used to test the assessment framework.
O 4.0, Model Descriptions, provides summary and detailed descriptions for
the four selected models. The summary descriptions are contained on a
series of Fact Sheets, each of which is a double-sided encapsulation of the
important characteristics of the model and sources of additional
information. The detailed model information contained in the latter part
of the section includes names of the developers, names and addresses of
the custodian, technical features, and solution methods. This information
has been extracted from a database of ground-water models developed and
maintained by the International Ground Water Modeling Center
(IGWMC).
Future Plans
This Compendium may be expanded as additional information is collected,
analyzed, and organized. This edition focuses on ground-water modeling, reflecting
the scope of the OSWER pilot project. Later editions could be expanded or modified
in order to:
CJ enhance the assessment framework;
O include case studies of model applications that illustrate both appropriate
and less appropriate modeling methodologies;
Page 1-2
-------
Section 1: Introduction
G include summary and detailed descriptions for additional models;
G address other modeling domains; and
O modify the format or contents to meet specific requests for information.
If you have any comments about this edition or would like to identify sources
of additional information, please contact Mary Lou Melley at OSWER/RMIS (202-
260-6860).
Acknowledgements
The OSWER Pilot Project has benefitted since its inception from input by a
team of ground-water modeling experts that includes Dr. Mary Anderson,
University of Wisconsin, Dr. Paul van der Heijde, International Ground Water
Modeling Center, and Dr. Luanne Vanderpool, EPA Region V. These individuals
were instrumental in developing the assessment framework provided in Section 2
of the Compendium, and their insights led to many improvements in the other
sections.
Darcy Campbell and Jon Lutz, of Region Vm, contributed to the contents of the
"EPA Publications Related to Ground-Water Modeling" in Section 2.0.
A committee, representing model users in the EPA Regional offices, in
OSWER, and other program offices, has been serving as a review group and
providing suggestions for the Compendium. Its members include:
Office of Solid Waste and Emergency Response
David Bartenfelder
Randall Breeden
Dorothy Canter
Matthew Charskey
Lynn Deering
Subijoy Dutta
Loren Henning
Tony Jover
Zubair Saleem
Richard Steimle
Richard Willey
Allison Barry
Fred Luckey
Nancy Cichowicz
Mike Arnett
Region I
Region II
Region III
Region IV
Christos Tsiamis
Page 1-3
-------
Section 1: Introduction
Region VI
Shawn Ghose Michael Hebert
Region VII
Mary Bitney
Region VIII
Darcy Campbell
Region IX
Richard Freitas
Region X
Glenn Bruck
Office of Research and Development
Bob Ambrose Will LaVeille*
David Burden Ron Wilhelm
Carl Enfield Darwin Wright*
Amy Mills*
Office of Air and Radiation
Robert Dyer* Kung-Wei Yeh*
Joe Tikvart*
Office of Water
Marilyn Ginsberg*
Office of Information Resources Management
Dwight Clay*
* Ex-Officio Members
Page 1-4
-------
Section 2.0 - Assessment Framework for
Ground-Water Model Applications
-------
2.O
Assessment Framework for Ground-Water Model Applications
Introduction
This section provides a framework for assessing ground-water model
applications. The framework contains a series of assessment criteria, grouped into
eight categories:
CJ Modeling Application Objectives
G Project Management
O Conceptual Model Development
O Model (code) Selection
G Model Setup and Input Estimation
O Simulation of Scenarios
O Post Simulation Analysis
O Overall Effectiveness.
The objective of the assessment framework is to support the use of models as
took for aiding decision-making under conditions of uncertainty. A manager,
reviewer, or modeler can use these criteria to assess modeling work that has already
been performed, or they can use the criteria to guide a current or future effort.
These criteria cover a wide range of conceivable technical and management issues
that might be encountered in a variety of modeling applications. For certain types of
problems, some of the criteria may not be applicable. For the more complex
modeling applications, some criteria may have to be modified or expanded.
The criteria address the activities and thought processes that should be part of a
modeling application and the subsequent documentation of that activity or process.
Consequently, some of the following criteria are preceded with an asterisk "*"
indicating that the analysis, the process, or the data referred to in the question
should be documented to assure the defensibility of the modeling application.
Some of the criteria are followed by footnotes referencing other sources of
information. The criteria also contain numerous technical terms that may require
additional explanation. These terms are italicized the first time they appear. A
glossary that follows the criteria contains the definitions for these terms in the
context in which they are used in the criteria. In other contexts, alternative
definitions of these terms may be more appropriate.
Page 2-1
-------
Section 2* A*»0ssm*nt Frmmowork Assessment Criteria
Assessment Criteria
Modeling Application Objectives
1 n *Management's decision objectives and the modeling objectives should be
clearly specified up-front.
2 D *Management's decision objectives should be based upon existing
information about the physical characteristics of the site (e.g., hydrogeologic
system) and the source and location of the contamination.
3 n The function of the model (e.g., data organization, understanding the
system, planning additional field characterization or evaluation of
remediation alternatives) should be defined during the development of the
modeling objectives.
4 n The potential solutions to be evaluated (e.g., containment and remediation
solutions) should be identified prior to the initiation of the modeling.
5 D The level of analysis required (e.g., numerical model, analytical model or
graphical techniques) should be determined during the definition of the
modeling objectives.
6 D 'Management, in consultation with a professional ground-water scientist,
should specify the time period (e.g., one year, ten years or hundreds of years)
for which model predictions are required.
7 D The level of confidence (quantitative or qualitative) required of the
modeling results should be specified.
8 D Performance targets (e.g., allowable head error) for the model application
should be specified up-front.
9 D *An analysis should be performed of the incremental costs associated with
expanding these study objectives (e.g., expanding the size of the study area,
the number of remedial technologies modeled or the performance targets of
the model) and the consequent incremental improvement in supporting
management's decision objectives.
10 D Management's decision objectives should be reaffirmed throughout the
modeling process.
11 D The modeling objectives should be reviewed, after the development of the
conceptual model and prior to the initiation of the modeling, to ensure that
they support management's decision objectives.
12 D The level of analysis required should be reviewed during the course of the
project and if necessary modified.
13 D The function of the model should be reviewed during the course of the
project and if necessary modified.
Page 2-2
-------
Sect/on 2r Am*»»»m0nt Frmmowork Assessment Criteria
Project Management
14 D The individuals who are actually performing the modeling, managing the
modeling effort and performing the peer review should have the ground-
water modeling experience required for the project. Specifically, for their
role on the project, each should have the appropriate level of:
Formal training in modeling and hydrogeology
Work experience modeling physical systems
Field experience characterizing site hydrogeology
Modeling project management experience.
15 D These individuals should be organized as a cohesive modeling team with
well defined roles, responsibilities and level of participation.
16 D The organization of the team should be appropriate for the application.
17 D *An independent quality assurance (QA) process should be established at
the beginning of the project.
18 D This QA process should include ongoing peer review of the:
Modeling objectives development
Conceptual model development
Model code selection
Model setup and calibration
Simulation of scenarios
Post simulation analysis.
19 D This QA process should be implemented.
20 D A procedure should be established up front for documenting the model
application.
21 D The documentation should include a discussion of the:
General setting of the site
Physical systems of interest
Potential solutions to be evaluated
Modeling objectives and time frame for model predictions
Quality assurance and peer review process
Composition of the modeling team
Data sources and data quality
Conceptual model
Hydrostratigraphy
- Ground-water flow system
- Hydrologic boundaries
Hydraulic properties
- Sources and sinks
Page 2-3
-------
Sect/on 2* Assessment Framework Assessment Criteria
Contaminant source, loading and areal extent
Contaminant transport and transformation processes
Selection of the computer code
Description of the code
- Reliability
- Usability
Transportability
Performance
- Public Domain vs. Proprietary Models
Limitations
Related Applications
Ground-water model construction
- Code modifications
- Model grid
Hydraulic parameters
Boundary conditions
Simplifying Assumptions
Calibration, sensitivity analysis, and verification
Predictive simulations
- Scenarios
Implementation of the scenarios
- Discussion of the results of each run
Uncertainty analysis
Discussion of results related to management's information needs as
formulated in the decision objectives
Executive summary (in terms of the decision objectives)
References
Input and output files.
See Footnote 1.
22 D The documentation should provide the information required for an
independent reviewer to complete a post application assessment.
Page 2-4
-------
Section 2t A**0**m0nt Framework Assessment Criteria
Conceptual Model Development
23 D An initial conceptual model of both the local and regional hydrogeological
system should be developed prior to any computer modeling.
24 D The conceptual model should be based upon a quantification of field data
as well as other qualitative data that includes information on the:
Aquifer system (Distribution and configuration of aquifer and
aquitard units)
Thickness and continuity of relevant units
Areal extent
- Interconnections between units
Hydrologic boundaries
Physical extent of the aquifer system
Hydrologic features that impact or control the ground-water
system
- Ground-water divides
- Surface water bodies
Hydraulic properties (Including, where relevant, homogeneous and
isotropic characteristics)
Transmissivity
Porosity
Hydraulic conductivity
Storativity
Specific yield
Sources and Sinks
- Recharge to the aquifer (e.g., Infiltration)
Evapotranspiration
Drains
- Ground-water discharges (e.g., flow to surface water bodies)
- Wells (e.g., water supply, injection or irrigation wells)
Fluid Potential (i.e., the potentiometric surface, the magnitude and
direction of the hydraulic gradient within each model layer)
Contaminant
Source
- Loading
Areal extent
Physical properties
- Chemical interactions
Biotransformations
Soils.
See Footnote 2.
Page 2-5
-------
Section 2s Assessment Framework Assessment Criteria
25 D *The quantity, quality and completeness of the field data should be analyzed
as part of the development of the conceptual model.
26 D If there are data gaps (e.g., missing water level or hydraulic conductivity
information), additional field work and other attempts to fill in these gaps
should be documented.
27 D *If there are data gaps, the tradeoff should be analyzed between the cost of
acquiring additional data and the consequent improvement in meeting
management's decision objectives.
28 D The data sources should be documented.
29 D The quality of the data should be examined and documented and the
influence of their quality on the project's results should be assessed.
30 D *Any and all potential interactions with other physical systems (e.g., surface
water systems or agricultural systems) should be evaluated, prior to the
beginning of the modeling, by means of a water budget, a chemical mass
balance or other analytical techniques.
31 D The manner in which existing and future engineering (e.g., wells or slurry
walls) must be represented in the numeric or analytic model should be
explicitly incorporated into the conceptual model.
32 D Sufficient contaminant sources should be identified to account for the
contaminant mass in the plume.
33 D *A clear statement of the location, type and state of the boundary conditions;
justification of their formulation; and the source(s) of information on the
boundary conditions should be included as part of the conceptual model.
34 D *A11 conceptual model parameters and reasonable parameter ranges should
be specified prior to beginning the calibration of the numerical model.
Page 2-6
-------
Section 2t Assessment Framework Assessment Criteria
Model (code) Selection
35 D The selected model (code) should be described with regard to its flow and
transport processes, mathematics, hydrogeologic system representation,
boundary conditions and input parameters.
36 D The reliability of the model (code) should be assessed including a review of:
Peer reviews of the model's theory (e.g., a formal review process by
an individual or organization acknowledged for their expertise in
ground-water modeling or the publication of the theory in a peer-
reviewed journal)
Peer reviews of the model's code (e.g., a formal review process by an
individual or organization acknowledged for their expertise in
assessing ground-water computer models)
Verification studies (e.g., evaluation of the model results against
laboratory tests, analytical solutions or other well accepted models)
Relevant field tests (i.e., the application and evaluation of the model
to site specific conditions for which extensive data sets are available)
The model's (code) acceptability in the user community as evidenced
by the quantity and type of use.
37 D The usability of the model (code) should be assessed including the
availability of:
The model binary code
The model source code
Pre and post processors
Existing data resources
Standardized data formats
Complete user instruction manuals
Sample problems
Necessary hardware
Transportability across platforms
User support.
38 D The tradeoff should be analyzed between model (code) performance (e.g.,
accuracy and processing speed) and the human and computer resources
required to perform the modeling.
39 D The model (code) should be in the public domain or at least readily
accessible to all interested parties. If not, the modelers should explain how
the inaccessibility would not detract from the study objectives and the
regulatory process.
40 D The assumptions in the model (code) should be analyzed with regard to
their impact upon the modeling objectives.
Page 2-7
-------
Sectfon 2r A*»09*mont Framaworfc Assessment Criteria
41 D *Any and all discrepancies between the modeling requirements (i.e., as
indicated by the decision objectives, conceptual model and available data)
and the capabilities of the selected model should be identified and justified.
The modelers should explain why the modeling objectives and/or the
conceptual model did or did not need to be modified. For example, the
implications of the selected code supporting one, two or three dimensional
modeling; providing steady versus unsteady state modeling; or requiring
simplifications of the conceptual model should be discussed.
*
42 D *If the modeling objectives are modified due to such discrepancies, those
modifications should be documented.
43 n *If the model source code is modified, the following tests should be
performed and the testing methodology and results should be justified:
Reliability testing (See criteria #36)
Usability evaluation (See criteria #37)
Performance testing.
See Footnote 3.
Page 2-8
-------
Sect/on 2t A*90**m0nt Framework Assessment Criteria
Model Setup And Input Estimation
44 n The overall grid resolution (e.g., average spacing of 100 feet versus 1000
feet) should be analyzed with respect to the dependent variable accuracy
required to meet management's objectives. For example, the grid should be
fine enough to allow the hydraulic gradient to be accurately represented.
See Footnote 7.
45 D The finite element or finite difference grid design should be analyzed with
respect to the modeling objectives such as the need to locate or model wells,
existing and future engineering or contaminant sources and plumes.
See Footnote 7.
46 D The grid should be designed with respect to the physical system. For
example:
The main grid orientation should be aligned with the principal
directions of hydraulic conductivity and/or transmissivity.
A finer grid should be used in areas where results are needed (e.g., in
the area of highest pollution or drawdown) or areas having large:
Changes in transmissivity
- Changes in hydraulic head
- Concentration gradients
A coarser grid should be used where data are scarce and for those
parts of the study area that are not of particular interest.
See Footnotes 4 and 7.
47 D The grid spacing and time step size should be analyzed with respect to
numerical accuracy. For example:
If a finite difference model with variable grid spacing was used the
grid should be expanded towards distant boundaries by less than a
factor of 2.
If a finite element model was used the following should be analyzed
with regard to their impact on the numerical accuracy of the model
application:
- Length to width ratio of each element
Size difference between neighboring elements
- The Peclet number (Pe = v AX/D).
The Courant number (Cr = v At/AX) should be < 1
Where:
D = Dispersion Coefficient (!2/t)
At=Time Step
V = Velocity (1/t)
AX = Grid spacing (1).
Page 2-9
-------
Sect/on 2s Assessment Framework Assessment Criteria
If Random Walk particle tracking was used the grid spacing and
particle mass should be analyzed with respect to the contaminant
resolution required.
See Footnotes 5 and 7.
48 D The mapping of the location of the boundary conditions on the grid should
be evaluated. For example:
The boundaries should be located far enough away from the areas of
interest to dampen any instability in the model.
The manner in which the boundaries are represented in the grid
should ensure the fineness of the grid, the accuracy of the geometry
and the accuracy of the boundary conditions.
For finite element grids, internal and external boundaries should
coincide with element boundaries.
See Footnotes 6 and 7.
49 D Well nodes should be located near the physical location of the wells.
50 D *The data sources, the data collection procedures and the data uncertainty
for the model input data should be evaluated and documented in the
project report or file.
51 D *A11 model inputs should be defined as to whether they are measurements,
estimates or assumptions including:
The constitutive coefficients and parameters (i.e., parameters that are not
generally observable but must be inferred from observations of other
variables, for example the distribution of transmissivity and specific
storage)
The forcing terms (e.g., sources and sinks of water and dissolved
contaminants)
The boundary conditions
The initial conditions.
52 D The input estimation process whereby data are converted into model inputs
(e.g., spatial and temporal interpolation and extrapolation or Kriging) should
be described and the spatial location and the associated values of the data
used to perform the interpolation should be shown on a map or provided
in a table.
See Footnote 7.
53 D The uncertainty associated with the input estimation process should be
specified, explained and documented.
54 D The model should be calibrated.
Page 2-10
-------
Section 2s Assessment Framework Assessment Criteria
55 D *If the model is not calibrated, the reasoning for not calibrating the model
should be explained.
56 D The criteria being used to terminate the calibration process (i.e., the
definition of an adequate match between observed and modeled values)
should be justified with regard to the modeling objectives.
See Footnote 7.
57 D "The calibration should be performed in a generally acceptable manner.
Specifically:
A sensitivity analysis should be performed to determine the key
parameters and boundary conditions to be investigated during
calibration.
The calibration should include a calculation of residuals between
simulated and measured values.
The calibration should include an evaluation of both spatial and
temporal residuals.
The calibration should be performed in the context of the physical
features (e.g., were residuals analyzed with respect to the pattern of
ground-water contours including mounds or depressions or
indications of surface water discharge or recharge).
See Footnote 8.
58 D If a water budget is developed, the results and their use in calibrating the
model should be explained.
59 D *A11 changes in initial model parameter values due to calibration should be
justified as to their reasonableness.
60 D *Any discrepancies between the calibrated model parameters and the
parameter ranges estimated in the conceptual model should be justified.
61 D *If the conceptual model is modified as a result of the model calibration, all
changes in the conceptual model should be justified. Whenever feasible,
the calibrated model should be verified with an independent set of field
observations.
Page 2-11
-------
Section 2r Assessment Framework Assessment Criteria
Simulation Of Scenarios
62 D *F°r each modeling scenario, the model inputs and the location of features
in the model grid should be justified. For example:
If a pumping well was not located at a node, the allocation of well
discharges among neighboring nodes should be justified.
If a slurry wall is a remedial alternative, the representation in the
model of the wall's geometric and hydraulic properties should be
justified.
If cleanup times are calculated, all assumptions about the location,
quantity and state of the contaminants should be justified.
When a remedial action, such as extraction wells, affects the flow,
such effects should be determined, including the downgradient
distance to the stagnation point and the lateral reach of each
modeled extraction well.
Page 2-12
-------
Section 2r Assessment Framework Assessment Criteria
Post Simulation Analysis
63 D The success of the model application in simulating the site scenarios
should be assessed.
64 n This assessment should include an analysis of:
Whether the modeling simulations were realistic
Whether the simulations accurately reflected the scenarios
Whether the hydrogeologic system was accurately simulated
Which aspects of the conceptual model were successfully modeled.
65 D The sensitivity of the model results to uncertainties in site specific
parameters and the level of error in the model calibration should be
examined and quantified. For example the modeling scenarios should be
simulated for the range of possible values of the more sensitive
hydrogeologic parameters. Moreover, the range of error in the model
calibration should be considered when drawing conclusions about the
model results.
66 D The post-processing should be analyzed to ensure that it accurately
represents the modeling results and interpolation and smoothing methods
should be documented where appropriate.
67 D The post-processing results should be analyzed to ensure that they support
the modeling objectives.
68 D The final presentation should effectively and accurately communicate the
modeling results.
69 D When feasible, a post audit of the model should be carried out or planned for
in the future.
See Footnote 9.
Page 2-13
-------
Section 2r Assessment Framework Assessment Criteria
Overall Effectiveness
70 D *Any difficulties encountered in the model application should be
documented.
71 D The model application should provide the information being sought by
management for decision making.
See Footnote 10.
72 D The model application results should be acceptable to all relevant parties.
73 D The model application should support a timely and effective regulatory
decision.
74 n Those aspects of the modeling effort that in hindsight might have been
done differently should be documented.
Page 2-14
-------
Section 2r Assessment Frmmework Footnotes
Footnotes
1. For more information on documentation see the Draft ASTM Standard Section
D-18.21.10, "Guide for Application of a Ground-Water Flow Model to a Site-
Specific Problem" and "Standards For Mathematical Modeling of Ground
Water Flow and Contaminant Transport at Hazardous Waste Sites," Chapter 4
of Volume 2 of Scientific and Technical Standards For Hazardous Waste Sites -
Draft. Department of Health Services, State of California.
2. For more information on data requirements for conceptual model
development see the Draft ASTM Standard Section D-18.21.10, "Guide for
Application of a Ground-Water Flow Model to a Site-Specific Problem;"
"Standards For Mathematical Modeling of Ground Water Flow and
Contaminant Transport at Hazardous Waste Sites," Chapter 4 of Volume 2 of
Scientific and Technical Standards For Hazardous Waste Sites - Draft.
Department of Health Services, State of California, pg. 4; Ground Water
Models. Scientific and Regulatory Applications, Committee on Ground-Water
Modeling Assessment, National Research Council, National Academy of
Sciences, 1990, pgs. 221 - 230; and Applied Ground Water Modeling: Simulation
Of Flow And Advective Transport. Anderson, M. and Woessner, W. W.,
Academic Press, 1992.
3. For more information on the testing of model codes see Groundwater
Modeling: An Overview and Status Report, van der Heijde, P. K., El-Kadi, A. I.
and Williams, S. A., International Ground Water Modeling Center, Colorado
School of Mines, Golden, Colorado, 1988, pgs. 27 - 33; and "Testing and
Validation of Ground Water Models," van der Heijde, P. K., Huyakorn, P.S.
and Mercer, J.W., Proceedings. NWWA/IGWMC Conference on Practical
Applications of Groundwater Models. Columbus, Ohio, August 19-20, 1985,
National Water Well Association, Dublin, Ohio.
4. For more information on the design of the grid see Groundwater Modeling: An
Overview and Status Report, van der Heijde, P. K., El-Kadi, A. I. and Williams,
S. A., International Ground Water Modeling Center, Colorado School of Mines,
Golden, Colorado, 1988, pgs. 45-48 and Applied Ground Water Modeling:
Simulation Of Flow And Advective Transport. Anderson, M. and Woessner
W. W., Academic Press, 1992.
5. For information on how grid design and time step size can affect the numerical
accuracy of a model see Computational Methods in Subsurface Flow.
Huyakorn, P. S. and Pinder, G. F., Academic Press, New York, New York, 1983,
pgs. 206 and 392; Groundwater Modeling: An Overview and Status Report, van
der Heijde, P. K., El-Kadi, A. I. and Williams, S. A., International Ground
Water Modeling Center, Colorado School of Mines, Golden, Colorado, 1988,
pgs. 45 - 48; and Applied Ground Water Modeling; Simulation Of Flow And
Advective Transport. Anderson, M. and Woessner, W. W., Academic Press,
1992.
Page 2-15
-------
Section 2t Assessment Framework Footnotes
6. For information on properly locating and representing boundary conditions see
Definition of Boundary and Initial Conditions in the Analysis of Saturated
Ground-Water Flow Systems - An Introduction, Franke, O. L., T. E. Reilly and
G. D. Bennett, Open-File Report 84-458, U. S. Geological Survey, Reston,
Virginia, 1984; and Applied Ground Water Modeling; Simulation Of Flow And
Advective Transport. Anderson, M. and Woessner, W. W., Academic Press,
1992.
7. For information on model setup, input estimation and criteria for the
termination of the calibration process see: Applied Ground Water Modeling;
Simulation Of Flow And Advective Transport. Anderson, M. and Woessner,
W. W., Academic Press, 1992.
8. For more information on the calibration of ground-water models see: Applied
Ground Water Modeling; Simulation Of Flow And Advective Transport.
Anderson, M. and Woessner, W. W., Academic Press, 1992; Groundwater
Modeling: An Overview and Status Report, van der Heijde, P. K., El-Kadi, A. I.
and Williams, S. A., International Ground Water Modeling Center, Colorado
School of Mines, Golden, Colorado, 1988; Draft ASTM Standard Section D-
18.21.10, "Standard Guide for Comparing Ground-Water Flow Model
Simulations to Site-Specific Information;" Subpart 6.5; and "Standards For
Mathematical Modeling of Ground Water Flow and Contaminant Transport at
Hazardous Waste Sites," Chapter 4 of Volume 2 of Scientific and Technical
Standards For Hazardous Waste Sites - Draft. Department of Health Services,
State of California, Section 3.3.2.4.
9. For more information on post audits see; "The Role of the Postaudit in Model
Validation", Anderson, M. P. and Woessner, W. W., submitted to Advances in
Water Resources. October, 1991
10. For more information on decision making under conditions of uncertainty see:
"Hydrogeological Decision Analysis: 1. A Framework," Freeze, R. A.,
Massmann, J., Smith, L., Sperling,T. and James, B., Ground Water 1990, 28(5),
738 - 766.
Page 2-16
-------
Section 2* Assessment Framework Glossary of Technical Terms
Glossary of Technical Terms
This glossary provides definitions for some of the technical terms used in
Section 2.0, Assessment Framework, of the Compendium of Modeling Information.
Words appearing in italics are defined elsewhere in the glossary. Numbers in
parentheses following each definition correspond to the reference source for the
definition. A complete list of references is included on the last page of the glossary.
Analytical model - mathematical expression used to study the behavior of
physical processes such as ground-water flow and contaminant transport. This type
of model is generally more economical and simpler than a numerical model,
but it requires many simplifying assumptions regarding the geologic setting
and hydrologic conditions. In comparison with a numerical model, however,
an analytical model provides an exact solution of the governing partial
differential equation instead of an approximate solution. (9, 14)
Aquitard - a geologic unit with low values of hydraulic conductivity which
allows some movement of water through it, but at rates of flow lower than
those of adjacent aquifers. An aquitard can transmit significant quantities of
water when viewed over a large area and long time periods, but its
permeability is not sufficient to justify production wells being placed in it. It
may serve as a storage unit, but it does not yield water readily. (1,11,12,15)
Boundary conditions - mathematical expressions specifying the dependent
variable (head) or the derivative of the dependent variable (flux) along the
boundaries of the problem domain. To solve the ground-water flow equation,
specification of boundary conditions, along with the initial conditions is
required. Ideally, the boundary of the model should correspond with a physical
boundary of the ground-water flow system, such as an impermeable body of
rock or a large body of surface water. Many model applications, however,
require the use of non-physical boundaries, such as ground-water divides and
aquifer underflow. The effect of non-physical boundaries on the modeling
results must be tested. (3)
Calibration - a procedure for finding a set of parameters, boundary conditions,
and stresses that produces simulated heads and fluxes that match field-
measured values within a pre-established range of error. (3)
Capture Zone - steady state: the region surrounding the well that contributes
flow to the well and which extends up gradient to the ground-water divide of
the drainage basin; travel time related: the region surrounding a well that
contributes flow to the well within a specified period of time. (14)
Conceptual model - an interpretation or working description of the
characteristics and dynamics of a physical system. The purpose of building a
conceptual model is to simplify the field problem and organize the field data so
the system can be analyzed more readily. (3,9)
Constitutive coefficients and parameters - type of model input that is not
directly observable, but, rather, must be inferred from observations of other
model variables; for example the distribution of transmissivity, specific storage,
porosity, recharge, and evapotraspiration. They are difficult to estimate because
Page 2-17
-------
Section 2* Assessment Frmmowork Glossary of Technical Terms
they vary and can not be observed, particularly when field measurements are
limited. (13)
Containment - action(s) undertaken, such as constructing slurry trenches,
installing diversionary booms, earth moving, plugging damaged tank cars, and
using chemicals to restrain the spread of the substance, that focus on
controlling the source of a discharge or release and minimizing the spread of
the hazardous substance or its effects. (18)
Contaminant source, loading and area! extent - the physical location of the
source contaminating the aquifer, the rate at which the contaminant is entering
the ground-water system, and the surface area of the contaminant source,
respectively. In order to model fate and transport of a contaminant, the
characteristics of the contaminant source must be known or assumed. (2)
Contaminant transformation - chemical changes and reactions that change the
chemical properties of the contaminant. (2)
Contaminant transport - flow and dispersion of contaminants dissolved in
ground water in the subsurface environment. (13)
Evapotranspiration - a combined term for water lost as vapor from a soil or
open water surfaces, such as lakes and streams (evaporation) and water lost
through the intervention of plants, mainly via the stomata (transpiration).
Term is used because, in practice, it is difficult to distinguish water vapor from
these two sources in water balance and atmospheric studies. Also known as fly-
off, total evaporation and water loss. Losses from evapotranspiration can occur
at the water table. (1,2)
Field characterization - a review of historical, on- and off-site, as well as surface
and sub-surface data, and the collection of new data to meet project objectives.
When possible, aerial photographs, contaminant source investigations, soil
and aquifer sampling, and the delineation of aquifer head and contaminant
concentrations should be reviewed. Field characterization is a necessary
prerequisite to the development of a conceptual model. (2)
Finite difference model - a type of numerical model that uses a mathematical
technique called finite-difference to obtain an approximate solution to the
partial differential ground-water flow equation. Aquifer heterogeneity is
handled by dividing the aquifer into homogeneous rectangular blocks. An
algebraic equation is written for each block, leading to a set of equations which
can be input into a matrix and solved numerically. This type of model has
difficulty incorporating irregular and uneven boundaries. (2,6,9,10)
Finite element model - a type of numerical model that uses the finite-element
technique to obtain an approximate solution to the partial differential ground-
water flow equation. To handle aquifer heterogeneity, the aquifer can be
divided into irregular homogeneous elements, usually triangles. This type of
model can incorporate irregular and curved boundaries, sloping soil, and rock
layers more easily than a finite difference model for some problem types. This
technique, like finite-difference, leads to a set of simultaneous algebraic
equations which is input into a matrix and solved numerically. (2,6,9,10)
Page 2-18
-------
Section 2t Assessment Framework glossary of Technical Terns
Fluid potential - mechanical energy per unit mass of fluid at any given point
in space and time with regard to an arbitrary state and datum. (6,16)
Forcing terms - type of model input included in most ground-water models to
account for sources and sinks of water or dissolved contaminants. They may be
measured directly (e.g., where and when contaminants are introduced into the
subsurface environment), inferred from measurements of more accessible
variables, or they may be postulated (e.g., effect of proposed cleanup strategy).
(13)
Ground-water divides - ridges in the water table or potentiometric surface from
which ground water moves away in both directions (14); an imaginary
impermeable boundary at the crest or valley bottom of a ground-water flow
system across which there is no flow. (6)
Ground-Water flow system - movement of water through, and the collective
hydrodynamical and geochemical processes at work in, the interconnected
voids in the phreatic zone (the zone of saturation). (1,16)
Hydraulic conductivity - the ability of a rock, sediment, or soil to permit water
to flow through it. The scientific definition is the volume of water at the
existing kinematic viscosity of the medium that will move in unit time under
a unit hydraulic gradient through a unit area measured at right angles to the
direction of flow. (14)
Hydraulic properties - those properties of a rock, sediment, or soil that govern
the entrance of and the capacity to yield and transmit water (e.g., porosity,
effective porosity, specific yield, and hydraulic conductivity). (2,22)
Hydrologic boundaries - boundary conditions which relate to the flow of water
in an aquifer system. (2)
Hydrostratigraphy - a sequence of geologic units delimited on the basis of
hydraulic properties. (5)
Infiltration - flow of water downward from the land surface into and through
the upper soil layers. (5)
Kriging - an interpolation procedure for estimating regional distributions of
ground-water model inputs from scattered observations. (13)
Model grid - a system of connected nodal points superimposed over the aquifer
to spatially discretize the aquifer into cells (finite difference method) or elements
(finite element method) for the purpose of mathematically modeling the aquifer
(21,22) ° n
Model representation - a conceptual, mathematical or physical depiction of a
field or laboratory system. A conceptual model describes the present condition
of the system. To make predictions of future behavior, it is necessary to
develop a dynamic model, such as physical scale models, analog models, or
mathematical models. Laboratory sand tanks simulate ground-water flow
directly. The flow of ground water can be implied by using an electrical analog
model. Mathematical models, including analytical, analytic element, finite
difference, and finite element models are more widely used because they are
easier to develop and manipulate. (3,5).
Page 2-19
-------
Section 2i Assessment Framework Glossary of Technical Terms
Modeling objectives - the purpose of the model application. The objectives
should direct model selection and level of effort for the modeling study. (2,3)
Numerical model - a mathematical model that allows the user to let the
controlling parameters vary in space and time, enabling detailed replications of
the complex geologic and hydrologic conditions existing in the field.
Numerical models require fewer restrictive assumptions, and are potentially
more realistic and adaptable than analytical models, but provide only
approximate solutions for the governing differential equations. (2,10,13)
Peer review - a process by which a panel or individual is charged to review and
compare the results of modeling efforts and to assess the importance and
nature of any differences which are present. The review may examine, for
example, the scientific validity of the model, the mathematical code,
hydrogeological/chemical/biological conceptualization, adequacy of data, and
the application of the model to a specific site. (2,13)
Performance target - a measure of model accuracy. (2)
Performance testing - determining for the range of expected uses, the efficiency
of the model in terms of the accuracy obtained versus the human and
computer resources required by comparing model results with predetermined
benchmarks. (20)
Porosity - total volume of void space divided by the total volume of porous
material. The term, "effective porosity," is related. It is the total volume of
interconnected void space divided by the total volume of porous material.
Effective porosity is used to compute average linear ground-water velocity. (2,5)
Post audit - comparison of model predictions to the actual outcome measured
in the field. Used to determine the success of a model application as well as the
acceptability of the model itself. (13)
Potentiometric surface - a surface that represents the level to which water will
rise in tightly cased wells. The water table is a particular potentiometric surface
for an unconfined aquifer. (5)
Reliability - the probability that a model will satisfactorily perform its intended
function under given circumstances. It is the amount of credence placed in a
result. (15)
Remediation - long-term action that stops or substantially reduces and prevents
future migration of a release or threat of hazardous substances that are a
serious but not an immediate threat to public health, welfare, or the
environment. (17)
Residuals - the differences between field measurements at calibration points
and simulated values. (8)
Sensitivity analysis - process to identify the model inputs that have the most
influence on model predictions, at least over a specified range. (2,13)
Sources and Sinks - gain or loss of water or contaminants from the system. In
a ground-water flow system, typical examples are pumping or injection wells.
(2,13)
Page 2-20
-------
Section 2r Assessment Framework Glossary of Technical Terms
Specific yield - quantity of water that a unit volume of aquifer, after being
saturated, will yield by gravity (expressed as a ratio or percentage of the volume
of the aquifer). (15)
Storativity - volume of water given per unit horizontal area of an aquifer and
per unit decline of the water table or potentiometric surface. Also known as
storage coefficient. (1,6)
Surface water bodies - all bodies of water on the surface of the earth. (15)
Transmissivity - the rate at which ground-water of a prevailing density and
viscosity is transmitted through a unit width of an aquifer or confining bed
under a unit hydraulic gradient. It is a function of the properties of the liquid,
porous media, and the thickness of the porous media. Often expressed as the
product of the hydraulic conductivity and the full saturated thickness of the
aquifer. (1,14)
Uncertainty analysis - the quantification of uncertainty in the spatially
distributed values of input hydraulic properties within an aquifer system. (7)
Verification study - consists of the verification of governing equations
through laboratory or field tests, the verification of model code through
comparison with other models or analytical solutions, and the verification of
the model through tests independent of the model calibration data. (3,4,19)
Page 2-21
-------
Section 2g Assessment Framework glossary of Technical Terms
Sources
Definitions are directly drawn from, and based upon the following sources:
1. Allaby, Ailsa and Michael. The Concise Oxford Dictionary of Earth Sciences.
Oxford: Oxford University Press. 1990.
2. Anderson, Mary P., based wholly or in part on written comments provided on
the initial draft version of the glossary (May, 1992).
3. Anderson, Mary P., and William W. Woessner. Applied Groundwater Modeling -
Simulation of Flow and Advective Transport. New York, Academic Press, Inc. 1992.
4. Belgin, Milovan S. Testing, Verification, and Validation of Two-Dimensional Solute
Transport Models, International Ground Water Modeling Center, December,
1987.
5. Fetter, C.W. Applied Hydrogeology - 2nd Edition Columbus, Merrill Publishing
Company, 1988.
6. Freeze, R. Allen and John A. Cherry. Groundwater. Englewood, New Jersey:
Prentice-Hall. 1979.
7. Freeze, Allen, Joel Massman, Leslie Smith, Tony Sperling, and Bruce James.
Hydrogeological Decision Analysis, Ground Water, 1990, 28(5)
8. Golden Software, Inc. Surfer Reference Manual Golden Colorado.
9. Istok, Jonathan. Groundwater Modeling by the Finite Element Method.
Washington, D.C.: American Geophysical Union. 1989.
10. Javandel, Iraj, Christine Doughty, and Chin-Fu Tsang. Ground Water Transport:
Handbook of Mathematical Models. Washington, D.C.: American Geophysical
Union. 1984.
11. Kruseman, G.P. and N.A. de Ridder, Analysis and Evaluation of Pumping Test
Data. The Netherlands, International Institute for Land Reclamation and
Improvement, 1990.
12. Lohman, S.W., Definitions of Selected Ground-Water Terms - Revisions and
Conceptual Refinements, Geological Survey Water-Supply Paper 1988. Washington,
D.C.: United States Government Printing Office, 1972.
13. National Research Council. Ground Water Models, Scientific and Regulatory
Applications. Washington, D.C.: National Academy Press. 1990.
14. The Ohio State University, Department of Geological Sciences Capzone Users
Manual, Columbus, Ohio, September, 1991.
Page 2-22
-------
Section 2r Assessment Framework Glossary of Technical Terms
15. Parker, Sybil P. McGraw-Hill Dictionary of Scientific and Technical Terms, Fourth
Edition. New York: McGraw-Hill Book Company. 1989.
16. Subsurface-Water Glossary Working Group, Ground Water Subcommittee,
Interagency Advisory Committee on Water Data, Subsurface-Water and Solute
Transport Federal Glossary of Selected Terms. August 1989.
17. United States Environmental Protection Agency. Office of Public Affairs.
Glossary of Environmental Terms and Acronym List. Washington, D.C. August
1988.
18. United States Environmental Protection Agency. Technical Assistance Team
(TAT) Contract Users Manual. Washington, D.C. October 1981.
19. van der Heijde, Paul, K.M. Quality Assurance and Quality Control in Groundwater
Modeling. International Ground Water Modeling Center, November, 1989.
20. van der Heijde, Paul, K.M. Identification and Compilation of Unsaturated/Vadose
Zone Models Possibly Applicable to Setting Soil Remediation Levels at Superfund Sites,
Report to the EPA Kerr Lab, Ada, Oklahoma, Draft, July 1992.
21. van der Heijde, Paul, K.M., Aly I. El-Kadi, and Stan A. Williams Groundwater
Modeling: An Overview and Status Report Indianapolis, Indiana, International
Ground Water Modeling Center, December 1988.
22. Wang, Herbert F. and Mary P. Anderson. Introduction to Groundwater Modeling,
Finite Difference and Element Methods. San Francisco, California: W.H. Freeman
and Company. 1982.
Page 2-23
-------
EPA Publications Related
Section 2t Assessment Framework to Ground-Water Modeling
EPA Publications Related to Ground-Water Modeling
CONTAMINANT FATE AND TRANSPORT
USEPA. Grove, D.B. and Rubin, J. 1976. Transport and Reaction of Contaminants
in Ground Water Systems, Proceedings of the National Conference on Disposal
of Residues on Land. Office of Research and Development, pp. 174-178.
USEPA. 1989. Determining Soil Response Action Levels Based on Potential
Contaminant Migration to Ground Water: A Compendium of Examples.
EPA/540/2-89/057.
USEPA. 1989. Laboratory Investigations of Residual Liquid Organics from Spills,
Leaks, and Disposal of Hazardous Wastes in Groundwater. EPA/600/6-90/004.
USEPA. Schmelling, Stephen, et al. 1989. Contaminant Transport in Fractured
Media: Models for Decision Makers: Superfund Ground Water Issue, Ada,
Oklahoma.
USEPA. 1990. Subsurface Contamination Reference Guide. EPA/540/2-90/Oil.
USEPA. 1990. Groundwater, Volume I: Ground Water and Contamination.
EPA/625/6-90/016a.
DNAPLs
USEPA. 1992. OSWER Directive No. 9355.4-07. Estimating the Potential for
Occurrence of DNAPL at Superfund Sites.
USEPA. 1992. OSWER Directive No. 9283.1-06. Considerations in Ground Water
Expediation at Superfund Sites and RCRA Facilities Update.
USEPA. Dense Nonaqueous Phase Liquids - A Workshop Summary.
GROUND-WATER ISSUE PAPERS
USEPA. Puls, R.W., Barcelona, M.J. 1989. Ground Water Sampling for Metals
Analysis. EPA/540/4-89/001.
USEPA. Lewis, T.Y.; Crockett, R.L.; Siegrist, R.L.; Zarrabi, K. 1991. Soil Sampling
and Analysis for Volatile Organic Compounds. EPA/450/4-91/001.
USEPA. Breckenridge, R.P.; Williams, J.R.; Keck, J.F. 1991. Characterizing Soils for
Hazardous Waste Site Assessments. EPA/540/4-91 /003.
USEPA. Schmelling. S.G.; Ross, R.R. 1989. Contaminant Transport in Fractured
Media: Models for Decision Makers. EPA/540/4-89/004.
USEPA. Huling, S.G. 1989. Facilitated Transport. EPA/540/4-89/003.
Page 2-24
-------
EPA Publications Related
Section 2t A»*essment Framework to Ground-Water Modeling
USEPA. Piwoni, M.D.; Keeley, J.W. 1990. Basic Concepts of Contaminant Sorption
at Hazardous Waste Sites. EPA/540/4-90/053.
USEPA. Huling, S.G.; Weaver, J.W. 1991. Dense Nonaqueous Phase Liquids.
EPA/540/4-91/002.
USEPA. Keely, JJF. 1989. Performance Evaluations of Pump-and Treat
Remediations. EPA/540/4-89/005.
USEPA. Sims, J.L.; Suflita, J.M.; Russell, H.H. 1991. Reductive Dehalogenation of
Organic Contaminants in Soils and Ground Water. EPA/540/4-90/054.
USEPA. RusseU, H.H.; Matthews, J.E.; Sewell, G.W. 1992. TCE Removal from
Contaminated Soil and Ground Water. EPA/540/S-92/002.
USEPA. Palmer, C.D.; Fish, W. 1992. Chemical Enhancements to Pump-and-Treat
Remediation. EPA/540/S-92/001.
USEPA. Sims, J.L.; Suflita, J.M.; Russell, H.H. 1992. In-Situ Bioremediation of
Contaminated Ground Water. EPA/540/S-92/003.
USEPA. 1990. Colloidal-Facilitated Transport of Inorganic Contaminants in Ground
Water: Part 1, Sampling. EPA/600/M-90/023.
USEPA. Puls, R.W.; Powell, R.M.; Clark, D.A.; Paul, C.J. 1991. Facilitated Transport
of Inorganic Contaminants in Ground Water: Part 2, Colloidal Transport.
EPA/600/M-91/040.
GROUND-WATER MONITORING AND WELL DESIGN
USEPA. Denit, Jeffrey. 1990. Appropriate Materials for Well Casing and Screens in
RCRA Ground Water Monitoring Networks.
USEPA. Aller, L.; Bennett, T.W.; Hackett, G.; Denne, J.E.; Petty, R.J. 1990. Handbook
of Suggested Practices for the Design and Installation of Ground Water
Monitoring Wells. EPA/600/4-89/034.
USEPA. 1992. Chapter Eleven of SW-846, Ground Water Monitoring. Office of
Solid Wastes, Washington, DC. Pre-Publication.
HYDROGEOLOGY
USEPA. 1985. Issue Papers in Support of Groundwater Classification Guidelines.
EPA/440/6-85/001.
USEPA. 1986. Criteria for Identifying Areas of Vulnerable Hydrogeology Under
RCRA. Appendix D: Development of Vulnerability Criteria Based on Risk
Assessments, Office of Solid Waste and Emergency Response, Washington, DC.
Page 2-25
-------
EPA Publications Related
Section 2r Assessment Framework to Ground-Water Modeling
USEPA. 1990. A New Approach and Methodologies for Characterizing the
Hydrogeologic Properties of Aquifiers. EPA/600/2-90/002.
USEPA. Molz, F.J.; Oktay, G.; Melville, J.G. Auburn University. 1990.
Measurement of Hydraulic Conductivity Distributions: A Manual of Practice.
USEPA. 1990. Ground Water, Volume H: Methodology. EPA/625/6-90/016b.
MODELING
USEPA. Pettyjohn, W.A., Kent, D.C., Prickett, T.A., LeGrand, H.E. (No date.)
Methods for the Prediction of Leachate Plume Migration and Mixing, Office of
Research Laboratory, Cincinnati.
USEPA. Bond, R; Hwang, S. 1988. Selection Criteria for Mathematical Models Used
in Exposure Assessments: Ground Water Models. EPA/600/2-89/028.
USEPA. 1988. OSWER Directive No. 9355.0-08. Modeling Remedial Actions at
Uncontrolled Hazardous Waste Sites.
USEPA. Bear, J.; Geljin, M.; Ross, R. 1992. Fundamentals of Ground Water
Modeling for Decision Makers. EPA/540/S-92/005.
RCRA
USEPA. 1984. OSWER Directive No. 9504.01-84. Enforcing Groundwater
Monitoring Requirements in RCRA Part B Permit Applications.
USEPA. 1984. OSWER Directive No. 9481.06-84. Clarification of the Definition of
Aquifer in 40 CFR 260.10.
USEPA. 1984. OSWER Directive No. 9481.02-84. ACL Demonstrations, Risk Levels,
and Subsurface Environment.
USEPA. 1985. OSWER Directive No. 9481.02-85. Ground Water Monitoring Above
the Uppermost Aquifier.
USEPA. 1985. OSWER Directive No. 9481.05-85. Indicator Parameters at Sanitary
Landfills.
USEPA. 1985. OSWER Directive No. 9931.1. RCRA Ground Water Monitoring
Compliance Order Guidance.
USEPA. 1985. OSWER Directive No. 9476.02-85. RCRA Policies on Ground Water
Quality at Closure.
USEPA. 1985. OSWER Directive No. 99050.0. Transmittal of the RCRA Ground
Water Enforcement Strategy.
Page 2-26
-------
EPA Publications Related
Sect/on 2» Assessment Framework to Ground-Water Modeling
USEPA. 1986. OSWER Directive No. 9950.2. Final RCRA Comprehensive Ground
Water Monitoring Evaluation (CME) and Guidance.
USEPA. 1986. Leachate Plume Management. EPA/540/2-85/004.
USEPA. 1987. OSWER Directive No. 9950.1. RCRA Ground Water Monitoring
Technical Enforcement Document.
USEPA. 1987. OSWER Directive No. 9481.00-10. Implementation Strategy for
Alternate Concentration Levels.
USEPA. 1988. OSWER Directive No. 9476.00-10. Ground Water Monitoring at
Clean Closing Surface Impoundment and Waste Pile Units.
USEPA. 1988. OSWER Directive No. 9950.3. Operation and Maintenance
Inspection Guide (RCRA) Ground Water Monitoring Systems.
RISK ASSESSMENT
USEPA. 1988. Bond, R; Hwang, S. 1988. Selection Criteria for Mathematical
Models Used in Exposure Assessments: Ground Water Models. EPA/600/8-
88/075.
USEPA. 1988. Superfund Exposure Assessment Manual. EPA/540/I-88/001.
USEPA. 1989. Risk Assessment Guidance for Superfund, Vol. I: Human Health
Evaluation Manual. EPA/540/1-89/002.
USEPA. 1989. Risk Assessment Guidance for Superfund, Vol. II: Environmental
Evaluation Manual. EPA/540/1-89/001.
SAMPLING AND DATA ANALYSIS
Bauer, K.M.; Glauz, W.D.; Flora, J.D. 1984. Methodologies in Determining Trends
in Water Quality Data.
USEPA. 1986. NTIS No. PG-86-137-304. Practical Guide for Ground Water
Sampling.
USEPA. 1987. OSWER Directive No. 9355.0-14. A Compendium of Superfund Field
Operation Methods, Volumes 1 and 2.
USEPA. 1988. Field Screening Methods Catalog: User's Guide. EPA/540/2-88/005.
USEPA. 1989. Data Quality Objectives for Remedial Response Activities:
Development Process. EPA/540/G-87/003.
USEPA. 1991. OSWER Directive No. 9355.4-04FS. A Guide: Methods for
Evaluating the Attainment of Clean-Up Standards for Soils and Solid Media.
Page 2-27
-------
EPA Publications Related
Section 2s Assessment Framework to Ground-Water Modeling
USEPA. 1991. OSWER Directive No. 9360.4-06. Compendium of ERT Ground
Water Sampling Procedures.
REMEDIAL INVESTIGATION/FEASIBILITY STUDY
USEPA. 1988. Guidance on Conducting Remediation Investigation and Feasibility
Studies (RI/FS) Under CERCLA. EPA/540/G-89/004.
USEPA. 1988. OSWER Directive No. 9283.1-02. Guidance on Remedial Action for
Contaminated Ground Water at Superfund Sites.
USEPA. 1989. OSWER Directive No. 9355.4-03. Considerations in Ground Water
Remediation at Superfund Sites.
USEPA. 1989. Example Scenario: RI/FS Activities at a Site with Contaminated Soil
and Ground Water. EPA/540/G-87/004.
USEPA. 1989. OSWER Directive No. 9835.8. Model Statement of Work for a
Remedial Investigation and Feasibility Study Conducted by a Potentially
Responsible Party.
USEPA. 1991. Conducting Remedial Investigation/Feasibility Studies for CERCLA
Municipal Landfill Sites. Office of Emergency Remedial Response,
Washington, DC.
USEPA. 1991. OSWER Directive No. 9835.3-2a. Model Administrative Order on
Consent for Remedial Investigation/Feasibility Study.
USEPA. 1992. Site Characterization for Subsurface Remediation. EPA/625/4-
91/026.
REMEDIAL DESIGN/REMEDIAL ACTION
USEPA. 1986. OSWER Directive 9355.0-4A. Superfund Remedial Design and
Remedial Action Guidance.
USEPA. 1988. OSWER Directive No. 9355.0-08. Modeling Remedial Actions at
Uncontrolled Hazardous Waste Sites.
USEPA. 1988. OSWER Directive No. 9283.1-02. Guidance on Remedial Action for
Contaminated Ground Water at Superfund Sites.
USEPA. 1989. OSWER Directive No. 9355.4-03. Considerations in Ground Water
Remediation at Superfund Sites
USEPA. 1990. OSWER Directive No. 9355.0-27FS. A Guide to Selecting Superfund
Remedial Actions.
USEPA. 1990. Guidance on Expediting Remedial Design and Remedial Action.
EPA/540/G-90/006.
Page 2-28
-------
EPA Publications Related
Section 2i Assessment Framework to Ground-Water Modeling
USEPA. 1990. Guidance on EPA Oversight of Remedial Designs and Remedial
Actions Performed by Potentially Responsible Parties. Interim Final.
EPA/540/G-90/001.
USEPA. Ross, R. General Methods for Remedial Operations Performance
Evaluations.
USEPA. 1990. OSWER Directive No. 9355.4-01FS. Guide on Remedial Actions at
Super-fund Sites with PCB Contamination.
USEPA. 1990. OSWER Directive No. 9833.0-2b. Model Unilateral Order for
Remedial Design and Remedial Action.
USEPA. 1991. OSWER Directive No. 9835.17. Model CERCLA RD/RA Consent
Decree.
RECORD OF DECISION
USEPA. 1990. OSWER Directive No. 9283.1-03. Suggested ROD Language for
Various Ground Water Remediation Options.
USEPA. 1991. OSWER Directive No. 9355.3-02FS-3. Guide to Developing
Superfund No Action, Interim Action, and Contingency Remedy RODs.
USEPA. 1991. OSWER Directive No. 9355.7-02. Structure and Components of Five
Year Reviews.
CLEAN-UP STANDARDS
USEPA. 1990. OSWER Directive No. 9234.2-11FS. ARARs Q's and A's: State
Ground Water Antidegradation Issues.
USEPA. 1990. OSWER Directive No. 9234.2-09FS. ARARs Q's and A's: Compliance
with Federal Water Quality Criteria.
USEPA. 1990. OSWER Directive No. 9234.2-06FS. CERCLA Compliance with Other
Laws Manual: CERCLA Compliance with the Clean Water Act (CWA) and the
Safe Drinking Water Act (SDWA).
USEPA. 1991. OSWER Directive No. 9355.4-04FS. A Guide: Methods for
Evaluating the Attainment of Cleanup Standards for Soil and Solid Media.
PUMP AND TREAT REMEDIATION
USEPA. 1989. Evaluation of Ground Water Extraction Remedies, Volume I:
Summary Report. EPA/540/2-89/054a.
USEPA. 1989. Evaluation of Ground Water Extraction Remedies, Volume II: Case
Studies 1-19. EPA/540/2-89/054b.
Page 2-29
-------
EPA Publications Related
Section 2r Assessment Framework to ground-Water Modeling
USEPA. 1989. Evaluation of Ground Water Extraction Remedies, Volume HI:
General Site Data, Data Base Reports. EPA/540/289/054c.
USEPA. 1989, OSWER Directive No. 9355.0-28. Control of Air Emissions from
Superfund Air Strippers at Superfund Ground Water Sites.
USEPA. Mercer, J.W.; Skipp, D.C; Giffin, D. 1990. Basics of Pump and Treat
Ground Water Remediation Technology. EPA/600/8-90/003.
USEPA. 1990. Saunders, G.L. 1990. Comparisons of Air Stripper Simulations and
Field Performance Data. EPA/450/1-90/002.
USEPA. 1989. Forum on Innovative Hazardous Waste Treatment Technologies:
Domestic and International. EPA/540/2-89/056.
USEPA. 1990. OSWER Directive No. 9355.0-27FS. A Guide to Selecting Superfund
Remedial Actions.
IN SITU REMEDIATION
Middleton, A.C.; Hiller, D.H. 1990. In Situ Aeration of Ground Water: A
Technology Overview.
USEPA. 1990. OSWER Directive No. 9355.0-27FS. A Guide to Selecting Superfund
Remedial Actions.
USEPA. 1990. Emerging Technologies: Bio-Recovery Systems Removal and
Recovery of Metal Ions from Ground Water. EPA/540/5-90/005a.
USEPA. 1989. Forum on Innovative Hazardous Waste Treatment Technologies:
Domestic and International. EPA/540/2-89/056.
LANDFILLS AND LAND DISPOSAL
USEPA. Kirkham, R.R., et al. 1986. Estimating Leachate Production from Closed
Hazardous Waste Landfills. Project Summary, Cincinnati
USEPA. Mullen, H. and Taub, S.I. 1976. Tracing Leachate from Landfills, A
Conceptual Approach, Proceedings of the National Conference on Disposal of
Residues on Land. Environmental Quality Symposium, Inc., pp. 121-126.
USEPA. 1990. OSWER Directive No. 9481.00-11. Status of Contaminated Ground
Water and Limitations on Disposal and Reuse.
Federal Register, Part H, 40 CFR Parts 257 and 258, Wednesday, October 9, 1991. Solid
Waste Disposal Facility Criteria: Final Rule.
USEPA. 1989. NTISNo. PB91-921332/CCE. Applicability of Land Disposal
Restrictions to RCRA and CERCLA Ground Water Treatment Reinjection,
Superfund Management.
Page 2-30
-------
EPA Publications Related
Section 2? Assessment Framework to Ground-Water Modeling
USEPA. 1991. Conducting Remedial Investigation/Feasibility Studies for CERCLA
Municipal Landfill Sites. Office of Emergency Remedial Response,
Washington, DC.
MISCELLANEOUS GROUND-WATER PUBLICATIONS
USEPA. 1986. Background Document Groundwater Screening Procedure. Office of
Solid Waste.
USEPA. 1990. Continuous Release - Emergency Response Notification System and
Priority Assessment Model: Model Documentation, Office of Emergency and
Remedial Response, Washington, DC.
USEPA. 1991. Protecting the Nation's Ground Water: EPA's Strategy for the 1990s;
The Final Report of the EPA Ground Water Quality Task Force. Office of the
Administrator, Washington, DC.
USEPA. 1990. ORD Ground Water Research Plan: Strategy for 1991 and Beyond.
EPA/9-90/042.
DRINKING WATER SUPPLY
USEPA. 1988. Guidance on Providing Alternate Water Supplies. EPA/540/G-
87/006.
CATALOGS
USEPA. 1992. Catalog of Superfund Program Publications. EPA/540/8-91/014.
To obtain these documents, contact the EPA Regional Libraries, the EPA Regional
Records Center, or the Center for Environmental Research Information
(513-569-7562)
Page 2-31
-------
Section 3.0 - Model Applications
-------
3.0
Model Applications
Introduction
This section provides summary descriptions of the model applications
used to test the assessment framework. These descriptions are presented to
help the reader understand the variety of model applications that the
framework was tested against. These descriptions are not intended to be
examples of the application of the framework but simply summaries of the
test cases. Each description discusses the:
O Decision objective;
O Regulatory context;
O Site characteristics (e.g., geology, ground-water contamination);
O Modeling activities and results;
G Interesting features of the application; and
G Names of EPA staff to contact for more details about the test case.
Page 3-1
-------
Section 3t Model Applications Introduction
(THIS PAGE INTENTIONALLY LEFT BLANK.)
Pago 3-2
-------
3.1 Summary Description #1
-------
Section 3i Model Applications Summary Description ft
3.1
Model Applications
Summary Description #1
Decision Objective:
This RCRA site was modeled to determine if the effects of a hydraulic barrier
and hydraulic head maintenance system would meet hydraulic gradient
performance standards set forth in a Consent Decree. This was an enforcement lead
site where contractors for the responsible party performed the modeling. The
Region reviewed the modeling effort and results to ensure that the site was
modeled as accurately as possible and that the model results indicated that the
proposed corrective action would meet the Consent Decree performance standards.
Background:
The site is located in an area of industrial, commercial and warehousing
operations on a peninsula in Baltimore Harbor. A significant portion of the
perimeter of the site abuts the harbor. Chromium ore was processed at the site from
1845 to 1985. This was a waste intensive operation that generated large quantities of
process residuals containing soluble chromium. Historically, these process residuals
were used as fill material at the plant site and in other areas around Baltimore
Harbor. (See Figure 3.1-1.)
The contamination of soil and ground water with elevated levels of chromium
creates the potential for human and environmental exposure. The risk of such
exposure, although small, will potentially exist as long as the site remains
uncontrolled. To minimize this risk of exposure, a combination of a low
permeability cap, a hydraulic barrier and ground-water extraction wells was
proposed as a containment strategy.
Remedial investigation and subsequent ground-water modeling of this site
began in October 1985, with further site investigations occurring through 1989 as
potential remedial actions were evaluated and the ground-water model was refined.
The ground-water modeling that is the focus of this case study began with the
refinement of the prior model to increase modeling accuracy. This work was
performed in 1989 and 1990.
Geologic Summary:
The site is located within the Coastal Plain Physiographic Province, an area that
is characterized by an unconsolidated sediment wedge that thickens toward the
Atlantic coast. In the vicinity of the site, coastal plain sediments are on the order of
70 ft thick. This site is underlain by artificial fills, recent organic silty day deposited
in the harbor, Pleistocene sediments, lower Cretaceous sediments and bedrock.
Page 3-3
-------
Section Si Model Applications Summary Description *1
The deepest geologic unit investigated consists of gneiss bedrock which is
composed of three distinct strata. The lowest stratum is weathered rock composed
of fine to medium grained gneiss that is often broken and fractured. Overlaying this
lowest stratum is a second stratum with an average thickness of eighteen feet. This
stratum consists of fine to coarse sand within a white clay-silt matrix that includes a
high feldspar content and a fine to coarse sand containing trace to some silt with a
high biotite content. The upper bedrock stratum consists of clayey fine to coarse
sand and has an average thickness of ten feet.
Overlaying the bedrock are Lower Cretaceous sediments which vary in
thickness from 12 to 40 feet. These sediments range from a silty white sand to white
sand and gravel to a white clayey silt. Blanketing the Lower Cretaceous sediments
are Pleistocene deposits composed of silty day, sand, gravel and cobbles which range
from 5 to 15 feet in thickness. Above the Pleistocene deposits black organic silty
clays and silty fine sand are found along the bulkhead margins of the site. The
organic silty clays have an average thickness of 20 feet and the silty fine sand varies
between 5 and 10 feet in thickness.
The fill materials, which created the level surface on which the site facilities
were built, consist of silty sands, micaceous sandy silts and poorly to well graded
sands as well as construction debris, bricks and wood fragments. The fill material
occurs at every location across the site and can be as much as 30 feet thick.
Ground-Water Contamination Summary:
The chromium contamination at the site was found to be present in two
ground-water yielding layers, one shallow, a water table aquifer, and one deep, a
regional aquifer system. The shallow ground water, which lies principally above the
low permeability silts near the site perimeter, flows radially off the site through the
bulkheads to the north, west and south. The shallow ground-water contamination,
which occurs chiefly near major source areas was found to have chromium levels
from 0.01 mg/L to 14,500 mg/L.
Deep ground-water contamination within the Cretaceous sands was also
highest near the source areas. Regionally, flow in the deep ground water originates
from the northwest of the site and flows towards the southeast. Locally, the deep
ground water flows radially away from the central portion of the site, with localized
small upward and downward flow gradients between the deep and shallow ground
water.
Modeling Summary:
The U. S. Geological Survey's three dimensional flow code, MODFLOW, was
used to model this site in a steady state. Six model layers, each of which was
constructed with variable thicknesses and bottom elevations, were used to simulate
the flow of ground water under the site. A 24 row by 24 column finite-difference
grid was constructed. (See Figure 3.1-2.) Grid cell dimensions varied between 75 and
150 feet with the smallest cells located directly over the study site. Boundary
conditions modeled included the flow directions and gradients of the water table
and regional aquifer systems, leakage from the harbor and precipitation recharge.
Page 3-4
-------
Section 3i Model Applications Summary Description #1
Horizontal hydraulic conductivity was determined for each layer by averaging
hydraulic conductivities from field tests. Vertical hydraulic conductivities were
initially assumed to equal the horizontal hydraulic conductivities and were adjusted
during calibration of the model.
During the calibration of the model, water levels in the observation wells were
used as targets and compared to simulated water levels. The hydraulic parameters
were adjusted to provide the best possible match between measured and calculated
heads at the calibration targets. Residual analysis was used to measure the
effectiveness of the calibration. The residual sum of squares was equal to 32.6 ft2 and
the residual mean was 0.0053 ft.
Additional field testing was performed to refine the calibration of the model.
Aquifer tests were used to improve hydraulic conductivity estimates. Additional
piezometers were also installed to better understand hydraulic gradients. The
model was then recalibrated with these new data.
Three hydraulic barrier scenarios were evaluated through model simulations.
All scenarios assumed site closure with the emplacement of a low-permeability cap
over the entire site. The first scenario simulated a "deep soil mixing" wall that
extended one foot into the uppermost stratum of the bedrock, a clayey fine to coarse
sand. This scenario was modeled in combination with ground-water extraction
wells and with perimeter trench drains placed at different depths.
The second scenario simulated a shallow slurry wall that extended two feet into
the clayey silt stratum of the Lower Cretaceous sediments. This scenario was
modeled in combination with alternative configurations of ground-water extraction
wells and with perimeter trench drains placed at different depths.
The third scenario simulated only ground-water extraction wells on the site.
Modeling Results Summary:
Model simulations of the first scenario indicated that the use of a deep
hydraulic barrier without pumping or trench drains would not meet the head
difference performance standards of the Consent Decree. However, the use of 12
extraction wells pumping a total of 2100 gal/day in combination with the wall
would meet these performance standards. The model simulations also indicated
that perimeter drains in combination with the deep hydraulic barrier would meet
the performance standards. When drains were placed at -05 ft msl. and 0.0 ft msl.
the model simulations indicated that 13500 gal/day and 3500 gal/day of ground-
water extraction would occur respectively.
Model simulations of the second scenario indicated that the shallow hydraulic
barrier without pumping or drains would not meet the head difference performance
standards of the Consent Decree. Moreover, the modeling indicated that the use of
the shallow hydraulic barrier instead of the deep hydraulic barrier would require
significantly increased ground-water extraction. If twelve extraction wells were used
a total pumpage of 27500 gal/day would be needed. If only eight wells were used,
the pumpage requirements would increase to 28,000 gal/day. If perimeter trench
Page 3-5
-------
Section 3i Model Applications Summary Description #1
drains were used in lieu of wells the ground-water extraction rate would vary
between 17,440 and 13,500 gal/day depending on the elevation of the drains.
Model simulations of the third scenario indicated that the head difference
performance standards of the Consent Decree could be met by pumping alone with
no hydraulic barrier. The total pumpage, however, required to meet the
performance standard would be 92,580 gal/day distributed over twelve extraction
wells.
In summary, model simulations indicated that the Consent Decree hydraulic
head difference performance standards could be met by each of the three modeling
scenarios. The first scenario, a deep wall and pumping resulted in minimum
extraction rates. Pumping wells alone were able to meet the performance standards;
however, in the absence of a hydraulic barrier, the amount of pumpage necessary to
maintain the performance standards was more than an order of magnitude greater
than that of the first scenario.
Strengths And Interesting Features Of This Application:
In this case study, an existing model of a site was refined and recalibrated. The
refinement of the model reflected the need for increased simulation accuracy as the
regulatory process moved from site characterization, to technology screening, to the
design of a specific containment technology and management strategy.
This modeling application was used to predict the performance of alternative
corrective actions and the impact of those alternatives upon the hydrogeologic
system. It should be noted that the quantity and quality of site field data in this case
study are atypical of most sites. These data allowed the development of a model
with considerable hydrogeologic specificity.
This application provided the information necessary to justify the need for
both a slurry wall and ground-water pumping inside the wall. Originally the
responsible party had proposed just a slurry wall with no pumping. This study
demonstrated the problems with such an approach and the effectiveness of
combining the slurry wall with pumping. For example, this study dramatized that
required pumping rates varied by over an order of magnitude with and without the
slurry wall. Moreover, this study helped to determine the design capacity of a water
treatment system by determining the volume of ground-water extraction for each
scenario.
This study demonstrated that the hydrogeologic system would be modified by
the proposed corrective action and provided a mechanism for predicting these
impacts. The Region noted that they will monitor the effectiveness of the selected
corrective action and if necessary the corrective action and /or the model application
will be modified as additional information is gained. This type of review, often
called a post audit, is strongly encouraged by ground-water modeling experts.
The calibration report provides an example of the use of residual analysis to
evaluate the model calibration.
Areas that should be covered in model calibration documentation include:
Pag*
-------
Section 3s Model Applications Summary Description §1
1. Explicitly establishing and documenting a reasonable range of
parameter values to be used in the calibration of the model;
2. Documenting the model residuals and the residual analysis;
3. Documenting the sensitivity of the model to variations in model
parameters;
4. Documenting the impact of grid spacing and time step on the
numerical accuracy of the model;
5. Documenting the criteria being used to terminate the calibration
process; and
6. Documenting the evaluation of the spatial and temporal distribution
of residuals.
Contacts:
For further information about this ground-water study please contact:
Joel Hennessy Region 3 Tel. # (215) 597-7584
Nancy Cichowicz Regions Tel.# (215)597-8118
Modeling Documents:
Please contact the above people for specific modeling documents.
Page 3-7
-------
Section 3* Model Applications
Summary Description ft
Figure 3.1-1
Site Location
Page 3-3
-------
Section 3t Model Applications
Summary Description tl
o
CM
Figure 3.1-2
Finite Difference Model Grid
\
J
s
7
\/
Zi
^N
/\
a
I
I
N>55
o
-------
Section 3i Model Applications Summary Description
(THIS PAGE INTENTIONALLY LEFT BLANK.)
Page 3-10
-------
3.2 Summary Description #2
-------
3.2
Model Applications
Summary Description #2
Decision Objective:
The United States Army's (Army) Aberdeen Proving Ground O-Field site was
modeled in the early phases of a site investigation to assist in screening corrective
actions for the site. Specifically, the objective of the modeling was to:
1. Provide a framework for the characterization of contaminant releases
and plumes;
2. Determine if contamination was migrating to other aquifers or
surrounding surface water bodies; and
3. Predict the probable hydrologic and chemical effects of relevant
remedial actions.
This modeling was performed as a requirement of a RCRA permit that EPA
issued to the U.S. Department of Army. In turn, the U.S. Army Environmental
Management Office of the Aberdeen Proving Ground contracted with the U. S.
Geological Survey (USGS) to conduct the modeling that is the focus of this case
study. The EPA Region reviewed the modeling results to ensure that the site was
modeled as accurately as possible and that the above decision objectives were met.
Background:
O-Field site is a 259 acre area located within the 79,000 acre Aberdeen Proving
Ground Army installation and is surrounded by Army testing ranges. O-Field lies
on a neck that extends into the Chesapeake Bay and is directly bordered on the west
by the Gunpowder River and on the north and east by a tributary of the Gunpowder
River, Watson Creek. (See Figures 3.2-1 and 3.2-2.) Watson Creek, which is better
described as a pond, discharges into the Gunpowder River through a man-made
culvert that restricts tidal flushing and therefore causes high organic loading in the
creek. To the south of O-Field lie other Army testing ranges. The site topography is
relatively flat with the highest elevation being about 19 feet above sea level.
O-Field includes two sub areas, Old O-Field and New O-Field. Old O-Field was
periodically used from the late 1930's into the 1950's for the disposal of munitions
and chemical-warfare agents. Disposal at New O-Field began in 1950 and continued
for an unspecified period. Disposal materials at New O-Field included ordnance,
contaminated material, laboratory quantities of chemical-warfare agents and dead
animals. The primary activity in later years at New O-Field was the destruction of
material by burning.
At both New and Old O-Field, containerized and uncontainerized material was
disposed in unlined and uncovered trenches, pits and directly on the ground.
Beginning in 1949, sporadic cleanup efforts were initiated with the goal of destroying
Page 3-11
-------
Section 3i Model Applications Summary Description *2
some of the explosives. Periodically during the cleanup operations, explosions
raptured container casings and directly exposed contaminants. Today, most of the
pits and trenches have been covered with soil.
In 1976, the Army recommended an assessment of the Aberdeen Proving
Ground to determine the potential for off-post migration of chemical contaminants.
Observation wells installed at O-Field in 1978 showed the presence of arsenic and
chlorinated organic solvents in the ground water. Analysis of surface water and soil
samples indicated that the ground water was transporting arsenic into Watson
Creek. A limited resampling of ground and surface water in 1984 confirmed these
findings.
An observation well network was developed in 1985 when eleven existing
wells were supplemented with 21 additional wells installed at eight locations. All
well drilling was performed using a remote control drill, bombproof shelters and
other extraordinary safety and security procedures due to the possibility of
encountering buried ordnance or chemical-warfare agents. An additional five wells
were installed in 1987 when quarterly sampling indicated the possibility that
contaminants might be present beyond the area covered by the original observation
well network.
Geologic Summary:
O-Field is located on unconsolidated sand, clay and silt of the Atlantic Coastal
Plain. Beneath the Coastal Plain sediments lies a basement complex of Precambrian
to Paleozoic crystalline rocks and Mesozoic rift-basin sedimentary rocks. The depth
to the pre-Cretaceous basement rocks at O-Field is approximately 650 feet.
The site hydrogeology was investigated to only a 200 hundred foot depth
because of the difficulties associated with remote drilling operations and the
improbability that contamination had extended to this depth. Four aquifers were
discovered but only the upper three were investigated. These three consisted of a
water table aquifer and two confined aquifers with the lowest confined aquifer
occurring at a depth of 70 to 90 feet. (See Figure 3.2-3.)
The uppermost soils at O-Field are silt to silty lean clay and below this layer lies
a low permeability, tan to grey sand with some silt. Lenses of gray clay underlie the
sand followed by silty sand where the water table begins. Extensive excavation and
explosions have probably destroyed much of the natural strata of the site to a depth
of 10 to 12 feet.
The water table aquifer lies 9 to 15 feet below ground surface with an average
saturated thickness of ten feet that varies seasonally by as much as three feet. This
aquifer is present across the site and is composed of brown to reddish-brown quartz
sand interbedded with discontinuous silt and clay layers. The sand is medium
grained in the central areas of the site and becomes finer to the east and north and
coarser to the northeast. The water table aquifer is underlain by a confining layer
composed of highly plastic black to gray or greenish gray clay. The depth of this
confining layer is 11 to 30 feet below ground surface and ranges in thickness from 05
Page 3-12
-------
Section 3i Model Applications Summary Description *2
to 5 feet. The presence of contamination below this layer indicates that the
confining layer is either leaky or discontinuous.
The upper confined aquifer is below this confining layer. This aquifer lies 20 to
30 feet below the ground surface with a thickness that varies from 13 feet in the east
and south to less than a foot as it nears the Gunpowder River to the west. The
aquifer probably remains confined beneath the Watson Creek shoreline but loses its
confining bed beneath the deeper parts of the creek. This aquifer is composed of
dark grey to brown, medium to coarse-grained sand interbedded with gravel and
discontinuous clay lenses. This aquifer is underlain by a confining layer composed
of dense, black to dark grey clay. The depth of this confining layer is 20 to 39 feet
below ground surface and ranges in thickness from 43 to 60 feet. The extent,
thickness and low permeability of this confining bed are probably adequate to
prevent contaminant migration or significant water movement from the upper
confined aquifer to the lower confined aquifer.
The lower confined aquifer lies approximately 80 feet below ground surface and
is 10 to 20 feet thick. There is little information on the extent and lithology of this
aquifer. Downhole gamma, spontaneous-potential and resistance logs as well auger
behavior during drilling suggests that it is composed of a highly permeable, gravel
like material.
Ground-Water Hydrology Summary:
Water flow in the aquifers underlying this site is complex due to the surface-
water interactions and lagging tidal cycles in Watson Creek. Ground water in the
water-table and upper confined aquifer generally flows from south to
north/northeast towards Watson Creek. A ground-water divide that lies to the west
of Old O-Field causes a portion of the ground water to bypass Old O-Field and flow
into the Gunpowder River. (See Figure 3.2-4.) A gross estimate of the ground-water
flow rate is 50 feet/year.
The water-table aquifer derives most of its recharge from vertical infiltration of
precipitation. Additional recharge occurs by lateral movement of ground water as
discussed above and periodically by Watson Creek, which overflows its banks
during periods of high tides. Discharge is primarily to Watson Creek and the
Gunpowder River.
Recharge to the upper confined aquifer is by downward leakage from the
overlying water-table aquifer. Discharge is by slow upward leakage through the
confining bed to the water table aquifer in down gradient areas and by leakage to
surface water bodies where the confining bed has eroded.
The hydraulic gradient indicates that the ground water in the lower confined
aquifer flows toward the west-northwest, possibly discharging into the Gunpowder
River. The heads in this aquifer are typically higher than those in the upper
confined aquifer. Thus, in the unlikely event that there are pathways for downward
contaminant migration between the upper and lower confined aquifers, the
hydraulic gradient would oppose all such flow except for the density driven
migration of free product.
Page 3-13
-------
Section 3t Model Applications Summary Description *2
Ground-Water And Surface Water Contamination Summary:
The ground water in both the water table and the upper confined aquifer at O-
Field contains inorganic and organic contaminants. Inorganic contaminants
include arsenic, iron, manganese, zinc, boron, antimony, cadmium and chloride.
For example, arsenic concentrations range from 1.96 parts per million (ppm) in the
water-table aquifer to 0.0016 ppm in the upper confined aquifer. Dominant organic
contaminants are chlorinated aliphatic hydrocarbons, aromatic hydrocarbons and
chemical-warfare degradation products which contain sulfur and phosphorus.
Concentrations of thiodiglycol, a degradation product, ranged from 1000 ppm to 5
ppm in the water-table aquifer. In general the highest contaminant concentrations
are measured in the water-table aquifer, although higher concentrations of boron
and 1,1,2,2-tetrachloroethane are present in the upper confined aquifer than in the
water-table aquifer. The arsenic and cadmium concentrations found exceed EPA
drinking water maximum contaminant levels. The concentrations of chloride,
iron, manganese and zinc exceed 1987 EPA secondary maximum contaminant
levels.
The distribution of individual dissolved contaminants varies areally in the
water table and upper confined aquifer. For example, the concentrations of arsenic
and organic contaminants are highest along the northeastern side of Old O-Field.
However, iron is present as two distinct plumes, one along the eastern side and one
along the northeastern side of Old O-Field. The areal distribution of contaminants
at New O-Field could not be evaluated because of the limited number of wells at the
site.
Although low concentrations of organic contaminants have been detected in
water samples from the lower confined aquifer, hydraulic gradients and the
lithology and thickness of the overlying bed make it unlikely that O-Field
operations have contaminated the lower confined aquifer.
A surface water quality study in 1985 of Watson and nearby creeks found
unusually high organic loading in Watson Creek and dissolved inorganic
constituents that exceed EPA chronic toxicity levels for freshwater and saltwater
aquatic life. The lateral migration of ground-water contaminants into Watson
Creek is thought to be partially responsible for the surface water contamination.
Ultimately, this surface water contamination may migrate into the Gunpowder
River.
Modeling Summary:
The U. S. Geological Survey's (USGS) three dimensional flow code,
MODFLOW, was used to model this site in a quasi three dimensional, steady state
mode. Two model layers were used, one layer simulating the water-table aquifer
and the second layer simulating the underlying confining layer and the upper
confined aquifer. The lower boundary of the second layer coincides with the top of a
fifty foot thick layer of dense day below the upper confined aquifer. The low
permeability and continuity of the clay was thought to justify its use as a no-flow
boundary for this model application and thus eliminated the need to model the
Page 3-14
-------
Section 3i Model Applications Summary Description #2
lower confined aquifer. The model was found to be relatively insensitive to this
assumption as discussed below.
A 66-row by 61-column finite-difference grid was constructed and aligned with
the principal direction of flow. The grid extends to the south and considerably
beyond O-Field to coincide with surface water bodies that could be modeled under
steady state conditions as constant head boundaries. Grid cell dimensions varied
between 20 and 2375 feet with the smallest cells located directly over the areas of
interest. Changes in cell size were limited to no more than 1.5 times the size of
adjacent cells.
Boundary conditions modeled included surface water bodies, no-flow regions,
drains, and leakage between the layers. Vertical hydraulic conductivities were based
upon typical values for similar soil types, limited field data and model calibration.
The water-table aquifer was bounded at most locations by constant head boundaries
which represented the shoreline of surface water bodies. At the two areas where the
water-table aquifer extended beyond the modeled area the ground-water flow was
parallel to the model boundaries and thus these areas were modeled as no-flow
boundaries.
The lateral boundaries of the upper confined aquifer were specified as no-flow
boundaries. However, a relatively high vertical hydraulic conductivity was
specified for those portions of the overlying confining layer that lay beneath surface
water bodies and thus these areas of the modeled layer responded almost as if they
had been established as constant head boundaries. Consequently, as the modeled
portion of the upper confined aquifer is almost entirely surrounded by surface water
bodies this layer was effectively modeled with constant head boundaries with one
exception. That exception was determined to be far enough from O-Field that the
no-flow boundary had negligible effects on simulations within O-Field.
Preliminary calibration of the steady-state model was achieved by setting
average annual recharge constant and adjusting model coefficients, primarily the
horizontal hydraulic conductivity of the water-table aquifer and the transmissivity
of the upper confined aquifer. These coefficients were adjusted within a range of
reasonable values based upon field measurements. The calibration was considered
acceptable if the simulated and observed average annual heads agreed within 0.5
feet. The model was then recalibrated against observed head data for a period of
elevated ground-water levels. During this second calibration the horizontal
hydraulic conductivity and transmissivity values obtained from the first calibration
were held constant and recharge was uniformly increased until the predicted heads
acceptably matched the observed heads Horizontal hydraulic conductivity and
transmissivity were then adjusted slightly to improve the match. These new values
were then used to once again calibrate the model against the average annual
observed heads. This process was repeated until the model met the calibration
criteria under both hydrologic scenarios.
Because of uncertainties about the actual recharge and hydraulic conductivities
at O-Field, several alternative solutions to the steady state model were then
generated by varying both recharge and hydraulic conductivity. Based upon
recharge estimates and field measurements of horizontal hydraulic conductivity,
Page 3-15
-------
Section 3s Model Applications Summary Description *2
the recharge was varied between 9 and 16 inches per year and the hydraulic
conductivity was uniformly adjusted by the same percentage. The resulting head
configurations still closely matched the observed heads. These alternative
hydrogeologic system representations were used to model the system responses to
the remedial alternatives evaluated also. Thus, the impact of hydrogeologic
uncertainty upon the results of the remedial alternatives modeled could be
quantified.
A sensitivity analysis of the calibrated flow model was performed for a range of
recharge, hydraulic conductivity and transmissivity values. This analysis indicated
that heads in the water-table aquifer were relatively insensitive to variations in the
vertical hydraulic conductivity of the confining layer or transmissivity of the
confined aquifer. Heads in the confined aquifer were somewhat more sensitive to
these variations. The sensitivity analysis also indicated that simulated heads in
both aquifers were very sensitive to changes in the horizontal hydraulic
conductivity and recharge of the water-table aquifer.
The assumption that the lower confined aquifer did not need to be modeled
was also examined during the sensitivity analysis. Simulations of a modified
version of the model that included the lower confined aquifer as a third layer
demonstrated that significant head variations in the lower confined aquifer had a
minimal impact on the head in the upper two aquifers. This confirmed that
simulation of the ground-water system as a two layer system was adequate for the
purposes of this modeling effort.
Five remedial actions and a sixth no-action scenario were evaluated. These
remedial actions were evaluated on the basis of their ability to lower ground-water
levels within the disposal areas and to limit lateral or vertical movement of water
through the disposal areas. The remedial actions evaluated included: installation of
an impermeable cap; installation of subsurface barriers; installation of a ground-
water drain; ground-water pumping to control water levels; ground-water pumping
to recover contaminants; and no action. All remedial actions were also simulated
with the alternative model configurations described above to quantify the impact of
hydrogeologic uncertainty upon each remedial alternative's predicted effectiveness.
The impermeable cap remedial alternative was simulated by establishing areas
of no ground-water recharge. The subsurface hydraulic barriers were simulated by
reducing horizontal hydraulic conductances in the model cells representing the
barriers. These conductances were calculated to represent a 5 foot thick barrier with
a horizontal hydraulic conductivity of 0.001 feet per day and accounted for the fact
that the barriers did not comprise the entire cell. The ground-water drain was
simulated by lowering the surface elevation of an existing natural drain to 15 feet
above sea level during the period of record.
Ground-water pumping to control water levels was simulated with the
addition of three pumping wells upgradient from Old O-Field and two wells
upgradient from New O-Field. Each of these wells were simulated at pumping rates
of 2,900,5,800,10,000 and 21,600 gallons per day (gal/d) but Old O-Field and New Old
Held wells were simulated separately. The pumping was simulated with and
without impermeable covers. Similarly, ground-water pumping to recover
Page 3-16
-------
Section 3i Model Applications Summary Description #2
contaminants (pump-and-treat) was simulated with the addition of two wells
adjacent to the southeastern side of Old O-Field and two wells adjacent to the
northeastern side of New O-Field. Again, each of these wells was pumped at rates of
2,900,5,800,10,000 and 21,600 gal/d; Old O-Field and New O-Field wells were
simulated separately; and the pumping was simulated with and without
impermeable covers on the respective disposal areas.
Modeling Results Summary:
The simulations indicated that covering Old O-Field with an impermeable cap
would lower water levels beneath the site by less than 1 foot. An impermeable cap
over New O-Field would be even less effective and reduce water levels by only 0.3
feet. Ground-water velocities appear to be sufficient to compensate for the loss of
recharge water intercepted by the caps. However, the reduction of precipitation
infiltration into the unsaturated zone at the fill could decrease the amount of
contaminant leaching.
Subsurface hydraulic barriers upgradient from Old O-Field resulted in water
level declines below Old O-Field of about 1 foot but produced increases in water
levels at New O-Field. Thus while potentially reducing contaminant leaching at
Old O-Field the barriers may also increase leaching at New O-Field. Complete
encapsulation of Old O-Field with hydraulic barriers and an impermeable cap was
also simulated and was shown to provide short term aquifer protection. However,
the concentrations of some contaminants would likely increase in solution and
should the encapsulating walls fail at some future point, the concentrations of some
contaminants in the ground water would then increase.
The simulation of subsurface barriers at the New Old-Field reduced water
levels at the disposal trenches by 25 feet but increased water levels by 2 feet on the
upgradient side. These reductions increased to 35 to 4.5 feet with the addition of an
impermeable cap.
The simulation of deepening an existing natural drain lowered water levels in
the water table aquifer beneath both Old and New O-Field. In turn, a ground-water
divide developed between the drain and Old O-Field and contaminant movement
from Old O-Field towards the drain probably would not occur. However, during
periods of low ground-water levels or high surface water levels, brackish surface
water would enter the drain and recharge parts of the water-table aquifer with
brackish surface water.
Pumping to manage water levels at Old O-Field indicated that the water-table
aquifer would be drawn down by 0.7 to 1.0 feet when the wells were pumped at 5800
gal/d per well. When an impermeable cap was added the drawdown increased by
0.5 feet. However, some simulations within the range of hydrogeologic uncertainty
demonstrated that a pumping rate of 5800 gal/d per well or more could induce
contaminants to migrate to previously uncontaminated areas.
Pumping rates of 10,000 and 21,600 gal/d per well were required at New O-Field
to reduce ground-water levels by 1.0 and 2.3 to 3.5 feet respectively. However, both
Page 3-17
-------
Section 3: Model Applications Summary Description #2
rates would produce hydraulic gradient reversals extending to the disposal trenches
and thus induce the movement of contaminants toward the wells.
The simulations predicted that pump-and-treat at a rate of 2900 gal/d per well
in combination with an impermeable cap would intercept the bulk of ground-water
contamination at Old O-Field in the spring, summer and fall. This rate would have
to be increased to 5800 gal/d in the winter when ground-water levels are higher.
However, this higher rate could induce the movement of water from Watson Creek
into the aquifer if it is maintained throughout the year. Pump-and-treat at New O-
Reld resulted in drawdowns of 1.1 to 1.8 feet for pumpage of 5800 gal/d and 2.2 to 35
feet for pumpage of 10,000 gal/d. As little is known about the extent of
contamination under New O-Held, no conclusions could be drawn as to whether
this would intercept the bulk of the contamination. New O-Field simulations at
these rates did not show inducement of infiltration of creek water but the proximity
of the hydraulic grade reversal to the creek implies that infiltration would probably
be induced if a 5800 gal/d per well pumping rate was maintained during dry
summer months.
If no remedial actions are taken at O-Field, the simulations indicated that
mobilization and transport of organic and inorganic contaminants will continue,
primarily because the seasonal and recharge induced water levels rise above the base
of the buried contamination. The ground water contaminants, in turn, will
continue to discharge to Watson Creek. If the contaminants in the creek attain
sufficient concentrations, then the depletion mechanisms in the creek may be
inadequate to prevent contaminant migration from the creek into the Gunpowder
River.
Strengths And Interesting Features Of This Study:
In this case study, a model was applied in the early phases of a site investigation
to assist in characterizing the contaminant releases and plumes, the future
migration of contaminants and to explore and screen potential remedial
alternatives. Because of the hazards associated with sampling at this site, only
limited hydrogeologic data was available when the modeling began. Perturbations
in the ground-water system caused by tidal induced changes in the elevations of the
surrounding surface water bodies further complicated the modeling. Nevertheless,
data acquired after this modeling was completed confirmed the general flow field
and the surface water and ground-water interactions predicted by the model.
A major effort was made to ensure that the model application would accurately
predict the system response during both the average and the extreme hydrogeologic
conditions found at the site. Specifically, the model was calibrated against two
different sets of observed conditions; one representing average annual head
observations, the other representing a .period of time when significantly higher than
average heads were observed. The calibration continued until the model
application accurately predicted system response under both sets of conditions.
This modeling effort quantified the impact of the hydrogeologic uncertainty
upon the results of the simulations. Alternative solutions to the steady state model
application were developed by increasing and decreasing recharge and horizontal
Page 3-18
-------
Section 3i Model Applications Summary Description #2
hydraulic conductivity within ranges defined by field measurements and
professional judgement These alternative solutions closely matched the observed
heads, and therefore, were judged to bound the set of reasonable representations of
the ground-water system. In turn, these alternative representations of the system
were used to bracket the range of system responses to the remedial alternatives
evaluated.
An analysis of the sensitivity of the calibrated model application to changes in
hydrogeologic parameters was also performed. As a part of this analysis the
sensitivity of the model predictions to modeling the site as a two layer system was
investigated. The sensitivity analysis indicated that the two-layer representation
was sufficient for the purposes of the study.
The model application provided the EPA Region with the framework for
selecting the most appropriate remedial alternative. It provided guidance in
determining the goals and objectives of the remediation effort. Based upon the
results of the modeling described in this case study, ground-water extraction was
selected as the preferred remedial action. The model application was subsequently
used to evaluate alternative extraction systems.
Contacts:
For further information about this ground-water study please contact:
Steven Hirsh Regions Tel.# (215)597-0549
Nancy Cichowicz Region 3 Tel. # (215) 597-8118
Cindy Powels Aberdeen Tel.# (410)671-4429
Proving Ground
Modeling Documents:
Please contact the above people for specific modeling documents.
Pago 3-19
-------
Section 3f Model Applications
Summary Descriotion #2
Figure 3.2-1
Site Location
PENNSYLVANIA
76°15'
Washington
VIRGINIA
PROVING AGROUND
39°20'
Base from US. Geological Survey. 1:100.000.
2 MILES
0123 KILOMETERS
Page 3-20
-------
Sect'on 3t Model Applications
Summary Description 02
Figure 3.2-2
Detailed Site Location
EXPLANATION
S ObMrv«iion-w«M MineJna, Me and identification
number n O-FieW.
ObMTvanon-wel duster sit* and identMcation
number n O-FMd.
r w*l in H-FMd.
&' TeeeqraoMe comae Interval 6 feet.
39'20' 30'
39'2tf IS'
3920*
o 100 «M too ico
f tit i *'
9 M 110 HO 2«0
USGS, 1991
-------
Section 3i Model Application*
Summary Description #2
Figure 3.2-3
Sample Hydrogeologic Sections
Sit* OF« SHe OF 17 Sit* Oral SM* OF20
(AltHud* ! «.7 F**O (Altitude is «.2 F**|) (Altitude It «.a (Altitude ! S.«
confined aquifer
0 70 4d ( M WT
Lower confined
HYOROGEOLOGIC SECTION A-A'
vertical £i*«o*r«iion x 4
B*
FEET
30
L-J^---.------'-----^---.-.^-.-.".
<^$<<<<<<<<<<<$-z<:
HYOROGEOLOGIC SECTION B-8
, vertical EnaggereMon x 10
Page 3.22
-------
Section 3t Model Applications
Summary Description #2
Figure 3.2-4
Site Water-Table Contours
Page 3-23
-------
Section 3t Model Applications Summary Doscriotion #2
(THIS PAGE INTENTIONALLY LEFT BLANK.)
Page 3-24
-------
3.3 - Summary Description #3
-------
3.3
Model Applications
Summary Description #3
Decision Objective:
This fund-lead site was modeled during the Remedial Design (RD) to ensure an
appropriate design while minimizing design costs. Specifically, additional
information on the ground-water flow and contaminant fate and transport was
required to design a ground-water extraction and collection system that had been
specified in the Record of Decision (ROD). The EPA contractor proposed additional
field investigations including an extensive drilling program to obtain the needed
design information. EPA suggested that much of the required information could be
obtained through the use of ground-water modeling, thus significantly reducing
design costs. The contractor agreed to the modeling effort and specific modeling
objectives were established. These objectives were to:
1. Delineate the maximum possible extent of the contaminant plume in
January 1993, when the pump and treat system is scheduled to begin
operation;
2. Estimate the mass of contamination per unit volume of the aquifer;
3. Conceptualize extraction well alignment designs; and
4. Evaluate the design alternatives.
An EPA contractor working under an Alternative Remediation Contracting
System contract did the modeling that is the subject of this case study. The EPA
project manager reviewed the modeling work.
Background:
This is a six acre site located in a 250 home residential area with homes within
100 feet of the site. (See Figure 3.3-1.) Other land use in the area includes some
commercial development along a state highway southeast of the site and
agricultural activities southeast of the same highway. The site is an inactive
manufacturing facility and is currently used for minor non-production activities,
primarily warehousing. Located on the site are process, office and warehouse
buildings. (See Figure 3.3-2.) A tank farm and a laboratory have been removed
from the site. The site topography is relatively flat but exhibits occasional low rises
and gentle depressions. The average annual precipitation is approximately 35
inches and the mean monthly temperatures range from 26° F to 72° F.
This site was used to manufacture and repackage non-lubricating automotive
fluids from the early 1960s to 1978. During the facility's operation, a number of
releases and a major fire contributed to the site contamination. Undocumented
releases of chlorinated hydrocarbons into the vadose zone occurred south of the
process buildings. Documented releases of diethyl ether occurred in 1972 when an
underground pipeline was ruptured during excavation. Chlorinated organics,
Page 3-25
-------
Section 3t Model Applications Summary Description #3
benzene and other solvents were released during a two day fire in 1978. During this
fire numerous tanks and drums ruptured and their contents were spread onto
unpaved areas by the water used to fight the fire. The unsaturated zone soil was
thought to be the primary remaining source of contamination at the site.
History of Investigation
The owner of the site installed six on-site monitoring wells in 1972.
Contaminant levels detected in these wells and nearby residential wells indicated
that the contamination was moving off-site. Consequently, under an agreement
with the state agency responsible for environmental protection, the owner began
supplying bottled water to homes with contaminated wells in 1973. After the 1978
fire, the owner began removing underground tanks. Then in 1980,15 more
monitoring wells were installed at 9 locations. In 1982, the state agency initiated
legal actions against the site owner to force site remediation. The following year the
owner began operating a ground-water treatment system.
The site was placed on the National Priorities list (NPL) in 1984 and at that time
EPA assumed the lead enforcement role. In 1985, the owner agreed to fund a
remedial investigation and feasibility study (RI/FS) under an Administrative Order
of Consent. The completion of that study was funded by EPA when the owner filed
for bankruptcy in 1986. The state, in 1986, extended a public water supply to the
residents in the area near the site. In 1988, the ground-water treatment system was
closed down by the site owner due to financial problems.
In 1988, the Record of Decision (ROD) was issued. The ROD specified soil and
ground-water remediation including a contaminated soil flushing system, a ground-
water extraction and collection system and a ground-water treatment (air stripping)
and discharge system. The ROD further specified that the aquifer within the 1.5
parts per billion contaminant concentration isopleth would have to be restored to
drinking water standards.
Geologic Summary:
This site is located in an area dominated by glacial sediments deposited in a
northeast-southwest trending belt. The two major types of deposits are glacial
outwash with post glacial alluvium and ice-contact outwash deposits. Both types of
deposits contain fine sand to coarse gravel with occasional large cobbles and are very
poorly sorted. Sixteen borings made in the vicinity of this site indicate that the
glacial sediments extend to an average depth of 117 feet below the surface.
Underlying the site are clay, silty sand, sand and gravel fades which laterally
intergrade with one another and result in laterally discontinuous layers.
The uppermost soils at the site consist of a 5 to 10 foot upper layer of fine sand
and a lower 115 to 125 foot sand layer.
-------
Section 3i Model Applications Summary Description #3
just east of the site 50 feet below the ground surface. This lense extends over 3000
feet to the west and rises to thirty feet below the surface. It reaches a maximum
thickness of 20 feet. Several small gravel beds where detected also in the lower parts
of the lower sand layer.
Ground-Water Hydrology Summary:
The site hydrogeology is composed of three distinct hydrogeologic units. The
uppermost unit is an unconfined aquifer that is encountered at approximately 30
feet below the ground surface with a saturated thickness of 70 to 90 feet in the site
area and increasing to a thickness of 135 feet just over a mile and a half to the west.
This aquifer corresponds to the lower sand layer discussed previously. The second
hydrologic unit, a confining layer of sandy day that varies between 7 and 22 feet in
thickness near the site but thins to a two foot thickness a mile to the west, lies below
the aquifer. The third hydrogeologic unit, a confined aquifer, whose confined head
is approximately one and one half feet lower than the head of the upper aquifer lies
below this confining layer. This lower aquifer corresponds to the layer of sands and
gravel with occasional minor clay lenses discussed above. The thickness of this
confined aquifer is unknown as none of the borings reached bedrock.
The upper unconfined aquifer is no longer used as a drinking water supply in
the immediate site area because of contamination. Two miles to the west of the site
are three municipal wells which produce more than a million gallons per day from
this aquifer. The cone of depression caused by the operation of these wells may
eventually enhance the off-site movement of contaminated ground-water towards
the west. The direction of flow in this aquifer is generally west-southwest with a
velocity of approximately 0.5 feet per day. Both pump and slug test data were used
to estimate the hydraulic properties of this aquifer. The pump test data indicated an
average aquifer hydraulic conductivity and transmissivity of 0.0775 cm/sec, and
148,000 gpd/ft, respectively, assuming an average aquifer thickness of 90 feet. The
slug test data indicated an average aquifer hydraulic conductivity and transmissivity
of 0.0489 cm/sec, and 98,000 gpd/ft, respectively.
The lower confined aquifer was not extensively studied for the purposes of
remediating this site because of: (1) the presence of the confining layer above the
aquifer, (2) prior sampling which indicated no presence of contamination in the
lower aquifer and (3) the relatively small difference in head between the two
aquifers.
Ground-Water Contamination Summary:
Eighteen contaminants were found at the site, with ten being selected as
indicator compounds. These induded 1,1-DCE, 1,1-DCA, 1,1,1-TCA and Benzene.
The Remedial Investigation (RI) identified the unsaturated zone underlying two
areas of the site as the major sources of contamination. The RI also speculated that
there was potentially one other on-site source and several additional off-site sources
of contamination based upon the location and discontinuity of the contaminant
plume. This plume begins beneath the site and extends west-southwest beyond the
site property boundaries. The plume may not be continuous within its boundaries
and the thickness of the plume diminishes as it extends off-site. The contaminated
Page 3-27
-------
Section 3i Model Applications Summary Description #3
zone reaches the bottom of the upper aquifer at the site and at a minimum extends
1500 feet downgradient.
The on-site ground-water contamination was being partially contained by two
purge wells that operated from 1983 to the spring of 1988. These wells had a
combined pumping rate of 200 gallons per minute which resulted in a maximum
drawdown of about one foot. When these wells were operating they apparently
contained the plume within the western and northern site boundaries. Since they
ceased operation, significant additional contaminant migration has occurred.
Modeling Summary:
Analytic and analytic/numeric models and graphical techniques were used to
investigate the site and to evaluate the effectiveness of ground-water extraction
alternatives. This approach was chosen after a review of project objectives, the
limited quantity and poor quality of the available field data, the modeling budget
and project time constraints.
The modeling was divided into three phases. The objective of the Phase One
modeling was to determine the maximum extent of contamination in the upper
aquifer. An analytical function driven variation of the fate and transport model
Random Walk and the analytic transport model PLUME were used in this phase.
The objective of the Phase Two modeling was to conceptualize remedial design
alternatives. Graphical Javandel type curve analysis procedures were used to
determine the minimum number of pumping wells, discharge rates and recovery
well locations under different pumping scenarios. Then time related zones of
capture for each alternative were determined using the the U.S. EPA Wellhead
Protection Area Model (WHPA). Finally, the aquifer drawdown resulting from each
alternative was simulated using the analytic ground-water flow model WELFLO.
The objective of the Phase Three modeling was to evaluate the most promising
remedial design alternatives. Again, Random Walk was used in this phase to: (1)
estimate the discharge rates, average concentration and mass of contaminants in the
extraction well discharge; (2) evaluate and compare the remedial alternatives and (3)
identify areas of uncertainty in design features. This case study will focus on the use
of the Random Walk model in phases one and three.
The analytical function driven variation of the Random Walk model was
chosen, based on the advice of an outside consultant, to verify the results of the
PLUME model. Random Walk was suggested because of the modeling objectives,
data limitations and the outside consultant's detailed familiarity with the model.
This version of the Random Walk model utilizes an analytic function based upon
the Theis equation to generate the flow field. The numeric Random Walk model is
applied to simulate the fate and transport of the contaminants.
The simplifying assumptions associated with the analytic flow component of
the Random Walk model required that the aquifer be treated as homogeneous,
isotropic and infinite in areal extent. Thus, no site-specific boundary conditions or
spatial variability in aquifer characteristics could be modeled. Moreover,
unidirectional, steady state ground-water flow had to be assumed. The modeling
team established a uniform hydraulic conductivity, saturated thickness,
Page 3-28
-------
Section 3* Model Applications Summary Description *3
transmissivity, hydraulic gradient and aquifer porosity for the modeled area based
upon information obtained from the RI report and field data.
For the Phase One modeling, three contaminants were originally selected: the
most toxic (1,1-DCE), the most mobile (1,1-DCA) and the most pervasive (1,1,1-TCA).
However, initial modeling efforts indicated that insufficient soil and ground-water
data were available to develop or calibrate contaminant transport models for 1,1-
DCE or 1,1-DCA. As 1,1,1-TCA had migrated and persisted farther downgradient
than the other contaminants the modeling team felt that focusing solely on 1,1,1-
TCA was valid and appropriately conservative.
The site conceptual model developed by the modeling team hypothesized that
the downgradient plume developed primarily through the dissolution of 1,1,1-TCA
retained in aquifer pores at residual saturation. This was in contrast to prior studies
of the site which suggested that the major contaminant source lay in the
unsaturated zone and was percolating into the saturated zone. Initially, the
modeling team had accepted this latter hypothesis and developed a batch flushing
model to simulate the migration of the contaminant into the saturated zone.
However, when the results of the batch flushing model were compared with
observed phenomena, there were major discrepancies. For example, only 2 to 4
percent of the contamination in the aquifer could be accounted for under this
hypothesis.
After a careful review and some initial skepticism about the validity of the
batch flushing model results, the modeling team hypothesized that the source term
lay in the saturated zone. They searched peer reviewed ground-water literature for
examples of similar conceptual models and found three. Then, additional site
sampling including a soil gas survey was initiated. The results of this sampling
indicated very limited soil contamination, thus further supporting the new
hypothesis that the source term lay in the saturated zone.
The particle tracking component of the Random Walk model was then used to
model the 1,1,1-TCA dissolution from the 1,1,1-TCA mass stored at residual
saturation in the saturated zone. This included the rate of loading of 1,1,1-TCA
from the unsaturated zone by percolation to the ground water, minus the mass
removal of 1,1,1-TCA from the ground-water system, during the historic operation
of the purge wells described previously. This approach resulted in a source term
that varied as a function of time and dropped to zero during the historical operation
of the purge wells. The longitudinal and transverse dispersivities, background
concentrations, and organic carbon partition coefficients utilized in the model were
based upon field data and established EPA methods. All ground-water
contamination was assumed to exist in the dissolved phase; and, consequently,
dense, nonaqueous phase liquid migration was not simulated. Biodegradation and
volatilization was also ignored.
The calibration of the Random Walk model was restricted to the contaminant
fate and transport component. This version of Random Walk assumes a
unidirectional flow field which can not be directly calibrated against observed
piezometric heads. Field data from the RI report was used to establish the flow field
parameters. Calibration of the contaminant transport model was achieved by
Page 3-29
-------
Section 3t Model Applications Summary Description 93
adjusting the flow direction, flow velocity and dispersivity until predicted
concentration values matched observed concentrations. These modifications were
within reasonable parameter ranges and in the case of the flow direction were based
upon additional field data. The source release term was held constant during the
calibration process.
The calibration of the model presented a number of challenges. First, the data
used to calibrate the model was obtained from prior studies conducted by the site
owner's contractors. A review of that data indicated numerous data quality
problems including transcription errors, false positive and false negative
indications, conflicting sampling dates and surface maps of conflicting scales.
Second, data from nested monitoring wells indicated the presence of a vertical
contaminant concentration gradient. However, the direction and magnitude of this
gradient varied inconsistently from one monitoring location to another.
The modeling team established a two-tier calibration target after analyzing
these constraints and the primary model objective, which was to predict the extent
of the contaminant plume at the beginning of operation of the extraction system.
For those wells where the field data indicated no contamination, the model would
have to indicate no contamination. For those wells where field data indicated
contamination, the model concentrations would have to be within one half an
order of magnitude of the measured values at that location. Furthermore, at nested
monitoring wells, the average of the measured values would be used for model
calibration. The model was then calibrated against three different sampling events
spanning an 8 year period utilizing field data from up to 19 wells.
Recognizing the limitations of the calibration data and in turn the calibration
process, the contractor initiated a new round of sampling at 23 monitoring wells
under stringent quality assurance procedures. This data was not available until the
calibration of the model had been completed. The modeling team, however, was
able to use this data to verify the model calibration. The success of this verification
surprised the modeling team. With only one exception the model accurately
simulated all wells where no contamination was detected. At those wells where
field data indicated evidence of contamination, the residuals between the simulated
and observed values were much closer than the one half order of magnitude
calibration target used. Moreover, at the downgradient monitoring well that lay
directly on the center line of the plume, the measured and simulated concentration
value varied by only one part per billion.
Modeling Results Summary:
The Phase One modeling results indicated that the 1,1,1-TCA plume will have
migrated 625 meters downgradient and 225 meters laterally by 1993. This migration
has occurred in spite of the operation of two purge wells from 1983 to 1988. The
modeling team was careful to note that they were more confident of the delineation
of the plume boundary than the predicted concentrations within the plume because
of model and data limitations.
Based upon these results, 15 extraction well alternatives were developed in
Phase Two using Javandel type curve analysis, the GPTRAC module of the U.S. EPA
Well Head Protection Area Model (WHPA), and the analytic model WELFLO. This
Page 3-3O
-------
Section 3s Model Applications Summary Description #3
analysis indicated that under ideal conditions only one extraction well discharging
at 100 gallons per minute would be required. However, recognizing the limitations
associated with the Phase One modeling and the models used in this second phase,
two additional alternatives utilizing three and five wells respectively were chosen
for further consideration in the Phase Three modeling. The location of the capture
zone of one alternative is shown in Figure 3.3-5.
The efficiency, flexibility and the mass of contaminants in the extraction well
discharges for these three alternative were then evaluated in Phase Three using the
Random Walk model. The efficiency of each alternative was defined as a function
of the pumping rate, the volume of water requiring treatment and the time required
to restore the ground water to clean-up goals. While the pumping rates for the
three alternatives were found to be very similar, the five well alternative was found
to meet the cleanup goals in one quarter the time of the other two alternatives. The
five well alternative, moreover, would significantly reduce the volume of water
requiring treatment.
The flexibility of each alternative was defined as the ability of the alternative to
operate efficiently when the actual aquifer response to the pumping varied
significantly from the response predicted by the modeling. In particular, the
modeling team was concerned about the possibility of unmodeled heterogeneity in
aquifer properties and the distribution of contaminant concentrations within the
plume. The five well alternative was again found to be the preferred alternative
because it allowed the operation of each of the five extraction wells to be tailored to
the specific characteristics of that part of the plume and aquifer within which the
well was located.
EPA has awarded a contract for the construction of the five well ground-water
extraction system developed as part of this modeling effort. This system is expected
to be in place in early 1993. A ground-water monitoring program has been proposed
which will allow the monitoring of the remediation progress and the ongoing
validation of this ground-water modeling application.
Strengths And Interesting Features Of This Study:
In this case study, a model was utilized as part of the remedial design process.
The modeling was initiated at the suggestion of EPA in order to reduce the
magnitude and cost associated with additional site sampling. The model chosen
represented a compromise between the level of detail and accuracy desired by the
designers, the available data and the modeling budget. In fact, during the
development of the modeling objectives, the modeling team attempted to use both
a simple analytic and a finite difference contaminant transport model before
abandoning the former for its lack of specificity, and the latter because of the limited
site data available.
Page 3-31
-------
Section 3i Model Applications Summary Description #3
One of the more interesting aspects of this case is that both EPA and the
contractor agreed that the modeling reduced the design costs by $450,000. Part of this
cost reduction was due to the use of the model to delineate the plume boundary in
lieu of locating the boundary by means of an extensive drilling program. The other
part of the cost reduction occurred as a result of the modeling process which led to
the identification of an error in the source characterization.
Previously it had been hypothesized that the source lay in the unsaturated
zone. Consequently, soil flushing had been specified as a remedial action in the
ROD. While trying to develop the source term for the model, the modeling team
began to question this hypothesis. Based upon the results of a batch flushing model,
a search of peer-reviewed literature for similar cases, and additional field sampling,
the modeling team determined that the source term most probably lay in the
saturated zone. Thus, the need for the design and implementation of the soil
flushing system specified in the ROD was eliminated. This resulted in considerable
design cost savings and will result in considerable additional construction and
operational cost savings.
Another interesting aspect of this modeling was that the Phase One modeling
was actually performed twice using two different models. First the site was modeled
using PLUME. Then at the suggestion of an outside consultant the site was
remodeled using Random Walk. There was a high degree of correlation between
the model results. However, such correlation is not necessarily evidence that a site
was modeled properly. Rather it demonstrates that for this same conceptual model,
these two models produce very similar results.
Both EPA and the contractor noted that this case is another example of how the
use of modeling early in the remedial investigation process could have improved
the entire remediation effort. Specifically, it was noted that a very simple analytic
model could have improved the sampling plan and monitoring well locations in
prior site studies. This in turn would have increased the possibility that a more
accurate numerical model could have been used during the design process, possibly
resulting in a more efficient design.
Contacts:
For further information about this ground-water study please contact:
Bob Whippo Region 5 Tel. # (312) 886-4759
Modeling Documents:
Please contact the above person for specific modeling documents.
Page 3-32
-------
Section 3t Model Applications
Summary Descrtotfon #3
Figure 3.3-1
Site and Contaminant Plume Location
Page 3-33
-------
Section 3* Model Applications
Summary Description #3
Figure 3.3-2
Detailed Site Location and Monitoring Wells
I
X
!?
o
<
%
«p
§
a
*rf K
8
-------
Section 3t Modal Applications
Summary Description #3
Figure 3.3-3
Sample Hydrogeologic Section
IJ £ f i
ill
9
i:
.nil
\l
A « *
M!
I
-------
Section 3i Modal Applications
Summary Description #3
Figure 3.3-4
Location of Sample Hydrogeologic Section
Page 3-J6
-------
Section 3* Model Applications
Summary Description *3
Figure 33-5
Capture Zone Analysis
4
N
*
YonK«r St. Business Rout* 60
NOTE; Capture ton* of only tho two downqradtant
««! i* mown to niuctroto tho sy«twn«
eesoelllty to eoptur* th« eantominatM plum*.
Addltlonot w«H at ^t* (net mown)
ill winanc* piumo capture.
Page 3-37
-------
Section 3i Model Applications Summary Descrfntfon 93
(THIS PAGE INTENTIONALLY LEFT BLANK.)
Page 3-38
-------
3.4 Summary Description #4
-------
3.4
Model Applications
Summary Description #4
Decision Objective:
This enforcement lead site was modeled by EPA during the remedial design
(RD) to ensure an appropriate RD and remedial action (RA). Specifically,
information on the ground-water flow and contaminant fate and transport was
required to support RD and RA negotiations. As this was an enforcement lead site,
the objective of EPA in conducting this modeling was to gather sufficient
information to be able to review the responsible party's characterization of the
nature and extent of contamination at the site and the remedial design.
The modeling was deemed necessary because it was recognized in the Record of
Decision (ROD) that the design objectives and remedial action specified in the ROD
were based upon incomplete knowledge of both the sources and the actual extent of
the contamination and the contaminant plume. Consequently, EPA determined
that a limited modeling effort would assist in better understanding the site and the
effectiveness of the proposed pump and treat remedial design. In support of these
decision objectives the following four modeling objectives were established:
1. Identify potential off-site contaminant migration;
2. Examine the proposed pump and treat extraction well alignment;
3. Improve (if necessary) the extraction well alignment design; and
4. Determine the time required to clean the site.
The modeling that is the subject of this case study was performed by an EPA
contractor under a Technical Enforcement Support at Hazardous Waste Sites
contract. The modeling work was reviewed by both the EPA project manager and an
EPA hydrogeologist.
Background:
This site is comprised of 18 acres located in an industrial corridor adjacent to an
interstate in a major western metropolitan area. The site abuts a major railroad line
and is surrounded by other small industries. (See Figure 3.4-1.) Located on the site
are abandoned and active tank farms and a filled and abandoned sump. There is
also a capped evaporation pond, the contents of which are unknown. The site
topography is generally flat and low lying with a vacant swampy area to the south.
The nearest residential area is within a quarter mile of the site and the total
population within a one mile radius of the site is less than 5000.
Operations at the site began in 1957 and included the production of herbicides
and pesticides; the production of sodium hypochlorite; refilling and distributing
chlorine and ammonia cylinders; and the packaging and distribution of acids,
caustics and organic solvents. Information related to the historical operations at the
Page 3-39
-------
Section 3t Model Applications Summary Description #4
site indicates that the uncontrolled release of contaminates may have begun in the
very first year of operation and may have continued through 1989. These releases
are believed to have been the result of both disposal practices and spills.
Examples of disposal practises that led to releases include: the use of unlined
settlement and evaporation ponds for process wastewater discharge; the discharge of
industrial and process waste materials to an on-site septic tank and drain field; and
the dumping of wastewater, pesticide and herbicide tank wash water on the ground.
Examples of spills included a 4000 to 8000 gallon spill of muriatic acid that occurred
during the unloading of a rail tanker and a hydrochloric acid spill due to a tank
rupture. Sources of contamination include the former evaporation pond which was
filled with earthen materials and capped with concrete in 1980, a process drain
system and sump, settlement ponds (location unknown), a leach field, dioxin
removal wastes and contaminated soil.
History of Investigation
Beginning in 1980, the site was operated as a RCRA Interim Status Hazardous
Waste Storage Facility. Between 1983 and 1989, the owner/operator of the site was
cited for several violations by a state agency responsible for RCRA enforcement. In
1984, that agency advised the owner/operator of an alleged release from the property
to the environment and then initiated a Preliminary Assessment and a follow-up
Site Investigation. Additional field investigations were conducted by the state
agency between 1985 and 1987.
In 1986, CERCLA enforcement activities were initiated. This led to an
emergency action by EPA later that year to remove drums, cylinders and
contaminated material from the site. In 1987, EPA proposed that the site be placed
upon the National Priorities List. In 1988, the owner/operator agreed to undertake a
Remedial Investigation/Feasibility Study (RI/FS) which was completed in March
1990. In 1989, the owner/operator notified the relevant agencies of its intent to close
its RCRA Part A Interim Status Storage Facility. In 1990, additional site sampling
was initiated under EPA's "Make Sites Safe" initiative. This sampling found
evidence of high levels of dioxin on that part of the site where contaminated
materials had been removed as part of EPA's 1986 emergency action. Consequently,
actions to stabilize the contaminants on the site pending remediation were initiated.
In March 1991, the Record of Decision (ROD) was issued. In the ROD, EPA with
the concurrence of the state, specified soil and ground-water remediation measures
at the site but reserved the right to further modify the ground-water remedy because
the ground-water contamination had not been fully characterized. The ground-
water remedy specified consisted of a pump and treat system utilizing approximately
ten wells to capture and treat the contaminant plume. These wells were proposed to
be located along the northern and western boundaries of the site and to be operated
at an extraction rate of two gallons per minute.
Geologic Summary:
This site is located in a valley that was part of a Pleistocene great basin lake.
The surficial geology of the valley is the result of successive expansions and
Page 3-40
-------
Section 3t Model Applications Summary Description #4
contractions of that great basin lake and glacial surges and retreats. The valley
consists of alluvial deposits which overlie and merge with deep, unconsolidated
lacustrine sediments. The alluvial deposits typically consist of silt and sand in the
first several feet and mostly fine pebble gravel to a depth of 5 feet or more. These
deposits are of a deltaic type and are associated with the development of the valley
drainage network. The underlying lacustrine deposits consist of clay, silt, sand and
gravel facies that laterally intergrade with one another and result in laterally
discontinuous layers.
The uppermost soils at the site consist of a 2 to 6 foot layer of mixed fill
material and a 2 to 5 foot layer of clay to silty clay material beneath the fill material.
Underlying the clay material is a sand to clayey to silty sand layer that extends to a
depth of 15 to 18 feet and below this point a clayey layer begins. Two borings to
depths of 27 and 50 feet indicated that this clayey layer extends to at least a depth of
50 feet and exhibits interfingering (clay, silt and sand) characteristic of the general
geology of the valley. A fluvial paleochannel consisting of fine to coarse grained
sand appears to meander across the center of most of the site. The top of this
channel lies 7 to 9 feet below the surface of the site and the channel is approximately
6 to 8 feet deep.
Ground-Water Hydrology Summary:
There is some disagreement regarding the ground-water hydrology. Prior
studies by the responsible party's consultants indicated that the site hydrogeology is
consistent with the regional hydrogeology and is composed of three distinct
hydrogeologic units. These studies indicated that the uppermost unit is a shallow
unconfined aquifer that begins 3 to 5 feet beneath the site and extends to a depth of
15 to 18 feet. This aquifer corresponds to the sand to clayey to silty sand layer
discussed in the above paragraph. Below this shallow aquifer lies the second
hydrogeologic unit, a relatively impermeable confining layer which acts as a single
confining bed that ranges from approximately 40 to 100 feet in thickness. This unit
corresponds to the clayey layer described in the above paragraph. Below this
confining unit lies a confined aquifer that consists of Quaternary deposits of day,
silt, sand and gravel. The maximum thickness of this aquifer is greater than 1000
feet.
The state agency responsible for environmental enforcement disagrees with the
concept that there are two completely distinct aquifers underneath the site. They
take the position that the second hydrologic unit is not truly impermeable as there
are interfingerings of sand and silt in the clay layer. Consequently, they believe that
there is communication between the upper and lower hydrologic unit and all three
units should be considered as one aquifer with shallow and deep portions. The
modeling team did not feel it was necessary to resolve this discrepancy in the
conceptual model given EPA's decision objectives and the fact that the modeling
focused solely on the upper 13.5 feet of the shallow aquifer.
The shallow unconfined aquifer is not used as a drinking water supply because
of its poor quality (e.g. high total dissolved solids, sulfide and chloride) and the low
yields to wells. The direction of flow in this aquifer is generally to the west-
northwest, which is consistent with the regional flow direction. In the late summer
there are several localized deviations in the flow pattern as several ground-water
Page 3-4*
-------
Section 3s Model Applications Summary Description #4
mounds build and dissipate. The transmissivity of this aquifer is estimated to be 77
ft2 per day. The recharge to this aquifer is primarily from upward flow through the
confining layer from the underlying confined aquifer and infiltration from
irrigation, precipitation and a nearby drainage ditch. This drainage ditch which runs
adjacent to the site is hydraulically connected to the shallow unconfmed aquifer. In
the spring this ditch appears to recharge the aquifer while during the summer that
portion of the aquifer underlying the site appears to contribute 1.1 to 1.4 gallons per
minute to the flow in the ditch.
The deep confined aquifer is a primary source of drinking water for the
surrounding metropolitan area. Pumping tests from other studies indicate that the
transmissivity of this aquifer ranges from 4000 to 10,000 ft2. This aquifer was not
extensively studied for the purposes of managing this site because of the presence of
a thick confining layer above the aquifer and the upward movement of water from
this aquifer to the shallow aquifer.
Ground-Water Contamination Summary:
A wide variety of contaminants was found at the site including VOCs, SVOCs,
Pesticides, Herbicides, Dioxins and Furans. The major sources of contamination on
the site include a process drain system, a former evaporation pond, a yard drain
system and a septic system. The primary contaminants of concern in the ground
water are the indicator chemicals TCE, PGE, PCP and 2,4-D. Ground-water samples
from monitoring wells indicated that these contaminants exceed maximum
contaminant levels (MCLS) as established under the Safe Drinking Water Act by up
to three orders of magnitude. These contaminants are widely dispersed horizontally
and vertically in the shallow aquifer underlying the site. (See Figure 3.4-1.)
However, ground-water samples from monitoring wells below the shallow aquifer
do not indicate that the contamination has extended into the confining layer
separating the shallow aquifer from the deep confining aquifer.
At the time the ROD was issued, the extent and origin of the contaminated
water found on the northern portion of the site had not been fully characterized.
The ROD anticipated that further investigations and subsequent ground-water
remediation decisions would have to be made prior to the initiation of remedial
actions.
Modeling Summary:
A series of analytic and analytic/numeric models were used to investigate the
site. An analytic modeling approach was chosen as a function of EPA's decision
objectives and resource constraints. As this was an enforcement lead site, EPA's
responsibility was to review the responsible party's remediation plan and activities
to ensure they met regulatory and legal requirements. Thus, EPA's objective in
conducting this modeling was not to develop a definitive remedial design but to
gather sufficient information to be able to review the responsible party's
characterization of the site and the RD.
New information, received since the issuance of the ROD, suggested the need
for the RD/RA to address ground-water contamination not previously thought to be
Page 3-42
-------
Section 3: Model Applications Summary Description #4
connected with the site. Evidence suggested that the plume in the shallow aquifer
may have migrated off-site. EPA desired to explore this possibility and to examine
its impact upon the proposed remedial design. As these were review and not design
objectives, these decision objectives were not considered sufficient to warrant the
costs associated with a full fledged numerical modeling effort at this stage in the
remediation process. Moreover, EPA and their contractor believed that current data
on the site was not sufficient to support a numerical model. Consequently, an
analytic modeling approach was chosen.
The modeling was divided into two phases. The objective of the phase one
modeling was to determine the maximum extent of chloride and TCE
contamination in the shallow aquifer. An analytical function-driven variation of
the fate and transport model Random Walk was used in this phase. The objective
of the phase two modeling was to examine the effectiveness of proposed and
alternative extraction well designs. Two models were used in this phase. The
analytic flow model THWELLS was used to determine well field drawdown. Then
the capture zones for the ground-water remedial alternatives were estimated using
the particle tracking module of the U.S. EPA Wellhead Protection Area (WHPA)
model called GFTRAC This case study will focus on phase one of the modeling
which is the use of the Random Walk model to delineate the contaminate plume
and potential sources.
The analytical function-driven variation of the Random Walk model was
chosen for the phase one modeling because of its relative simplicity, prior contractor
experience and prior contractor verification of the model through a comparison of
its results with that of another well-respected model. This version of the Random
Walk model utilizes an analytic function based upon the Theis equation to generate
the flow field. Then the numeric Random Walk model is applied to simulate the
fate and transport of the contaminants.
The simplifying assumptions associated with the analytic flow component of
the Random Walk model required that the aquifer be treated as homogeneous,
isotropic and infinite in areal extent. Thus, no site specific boundary conditions or
spatial variability in aquifer characteristics could be modeled. Moreover,
unidirectional, steady state ground-water flow had to be assumed. The modeling
team established a uniform hydraulic conductivity, saturated thickness,
transmissivity, hydraulic gradient and aquifer porosity for the modeled area based
upon information obtained from the RI report and the ROD. They assumed that
there was no precipitation recharge to the aquifer.
Chloride was chosen as one of the two contaminants to be modeled because it is
mobile and non-retarded. Thus, its plume would represent the outermost limits of
the plumes of the other contaminants of interest. Moreover, since chloride is a
conservative substance, it was hypothesized that the successful replication in the
model of the existing chloride plume would be evidence that the simplifying
assumptions of the analytic flow model did not unduly compromise the model
results given EPA's decision objectives. TCE was selected as the other contaminant
to be modeled because it is relatively toxic and ubiquitous to the site.
Page 3-43
-------
Section 3r Model Applications Summary Description #4
The particle tracking component of the Random Walk model was used to
model the chloride releases as slugs that began in 1982 at four potential sources. The
TCE releases were modeled as continuous releases that began in 1957 at eight
potential sources. The location of the sources and the timing of the releases were
inferred from soil and ground-water chemistry, information in the RI and the ROD
and model calibration. The longitudinal and transverse dispersivities, background
concentrations and organic carbon partition coefficients were based upon field data
and established EPA methods. All ground-water contamination was assumed to
exist in the dissolved phase and consequently dense, nonaqueous phase liquid
migration was not simulated. Biodegradation and volatilization were also ignored.
The calibration of the Random Walk model was restricted to the contaminant
fate and transport component. This version of Random Walk assumes a
unidirectional flow field which can not be directly calibrated against observed
piezometric heads. Reid data from the RI report was used to establish the flow field
parameters. Moreover, as was noted above, it was thought that the successful
calibration of the chloride plume would serve as an indication of the
appropriateness of the assumed flow field. Calibration of the contaminant transport
model was achieved by adjusting the number, location and release terms of the
sources until simulated concentrations matched field observations for twelve on-
site monitoring wells. The calibration process was terminated when:
The mean model error and standard deviation of errors were less than 10
percent of the largest observed field concentration;
The regression analyses of field data versus model results yielded a
correlation coefficient squared (r2) value greater than 0.90, (an r2 value of
1.0 indicates perfect correlation between the field data and the model data);
and
The shapes of the simulated plumes approximated the shapes of the
plumes observed in the field.
The calibration of the contaminant transport model required that the grid
nodes in the model be close to the physical location of the monitoring wells on the
site. Considerable effort was made to do this, but in order to determine the predicted
contamination concentrations at the monitoring wells, linear interpolation routines
ultimately had to be developed In retrospect, the modeling team realized that by
varying the location of the origin of the model grid, the grid nodes could have been
located precisely over the monitoring well locations thus eliminating the need to
interpolate the contaminate concentrations and possibly improving the calibration.
It was noted, however, by the modeling team that the shifting of the grid would
become very time consuming with a large number of wells.
Difficulties were encountered calibrating the model. These difficulties caused
the modeling team to explore the possibility of the existence of undocumented on
and off-site sources. In an iterative fashion, sources were added to the model until
all calibration targets were met, the simulated plumes approximated the field
observable plumes and r2 values of 0.92 and 0.99 were obtained for the chloride and
TCE models respectively. (See Figures 3.4-2 and 3.4-3.) The reasonableness of the
location of these hypothetical contaminate sources was examined using site
information and aerial photographs of the site taken in the 1960's and 1980's.
Page 3-44
-------
Section 3; Model Applications Summary Description #4
Interestingly, these photographs had not been available to the modeling team
during the calibration process, yet wherever a hypothetical source had been
introduced into the model in order to effect a better calibration, the aerial
photographs provided confirming evidence of the existence of a potential source at
that location.
Because of budget limitations a sensitivity analysis was not performed to
determine the potential variations in the results as a function of changes in key
parameters or the removal of hypothetical sources. In retrospect, the modeling
team felt additional resources should have been requested to perform a sensitivity
analysis.
Modeling Results Summary:
The phase one modeling results indicated that the ground-water
contamination moved much further off-site than was estimated in the ROD.
Moreover, the model simulations suggested the possibility of off-site sources of
chloride and TCE contamination. (See Figure 3.4-4.) Specifically, the model
indicated that the plume had migrated almost 450 feet north and a 150 feet west of
the site and covered a surface area over 1/3 larger than the contaminant plume
defined in the ROD.
Four potential chloride sources and eight potential TCE sources were identified.
One of the chloride sources and three of the TCE sources were located off-site. None
of these potential off-site sources had been identified in the ROD. Consequently,
EPA's concerns about the completeness of the field characterization were supported
by the modeling results.
The phase one results were then used to evaluate the proposed and alternative
pump and treat extraction well designs. This evaluation (not part of this case study)
determined that the proposed design of the pump and treat extraction wells would
not fully capture the contaminant plume because the plume had migrated so far off-
site. Moreover, it was determined that the proposed pumping rate would de-water
the aquifer.
Based upon the information developed in this modeling effort, EPA directed
the Responsible Party to initiate further site characterization including a search for
additional off-site sources. The Agency also requested that the design of the
extraction wells be reviewed. However, the Agency and the modeling team were
careful to not assign certitude to the model results. As was stated in the modeling
report, the simplifications associated with the use of analytic models and the limited
field data available required that the results of this modeling effort be used with
great care. Thus the Agency limited the use of the model results to:
1. Estimate the maximum dowrigradient and lateral extent of ground-water
contamination;
2. Focus future field activities;
3. Provide insight into the location of additional potential sources; and
Page 3-45
-------
Section 3s Model Applications Summary Description #4
4. Provide preliminary ideas regarding the conceptual design of pump and
treat alternatives.
Strengths And Interesting Features Of This Study:
In this case study, a model was applied late in the remedial process to review
the completeness and accuracy of the site characterization and remedial design
specified in the ROD. The decision and modeling objectives of this model
application were carefully limited so as not to exceed the limitations of the model
and available data.
Perhaps the most striking aspect of this case was that the entire modeling effort
required only 180 hours. This was possible because the decision objectives of the
modeling effort were limited in such a way that relatively simple models and
simplifying assumptions could be used. Nonetheless, the information gained from
the modeling could be used to guide field characterization activities and to raise
significant concerns about the characterization of the site and the design of the
proposed extraction wells. The ultimate consequences may be a considerable cost
savings through the implementation of a more appropriate remedial design.
Another interesting aspect of this study was the use of the model to identify the
location of previously undocumented potential sources of contamination and the
post modeling corroboration through the use of aerial photographs. This is an
example of the power of relatively simple models when they are carefully used as
preliminary investigation tools.
In retrospect, the Region noted that it might have been useful to conduct this
type of modeling during the RI/FS phase of the cleanup. The Region also noted that
additional modeling might be required later, once the additional site
characterization activities are completed. The Region suspected that should
additional modeling be performed, the level of certainty they would then require
would necessitate the use of a more sophisticated model and significantly more data.
Contacts:
For further information about this ground-water study please contact:
Bert Garcia Region 8 Tel. # (303) 293-1526
Modeling Documents:
Please contact the above person for specific modeling documents.
Page 3-46
-------
Section 3t Model Applications
Summary Description
Figure 3.4-1
Chloride Concentrations
1
PI
I *
*
0 £ 2
x * X
Page 3-47
-------
Section 3s Model Applications
Summary Description
Figure 3.4-2
Location of Suspected Chloride Sources
5
*<
a 2 ii
* » is
u -*
i Ii
J 2
si
£x
3g
O*
Page 3-4S
-------
Section 3t Model Applications
Summary Description #4
Figure 3.4-3
Model Calibration - Chloride
3O
3U
H5
*f
!S
s I ! §1
Page
-------
Section 3: Model Applications
Summary Description
Figure 3.4-4
Estimated Maximum Extent of Contamination
81
1
f,
H
r
"^ i
;
M
i " ! .f"1 '*
ri cf^l i
i j ii
i 33
Page
-------
4.0 - Model
-------
4.0
Model Descriptions
Introduction
This section provides summary and detailed descriptions for four ground-
water models that were selected as the initial set of models to be considered as part
of this pilot project. The four models are:
O MOC
O MODFLOW
O RANDOM WALK
O PLASM
While these models are generally well known and have been used often in
EPA programs, the inclusion of them in the Compendium does not represent an
endorsement of them by OSWER for any specific purpose. In the future, OSWER
may develop guidelines for application of models under certain conditions or for
specified types of analyses, and add other models to the Compendium. At this time,
the descriptions are provided as general reference information only.
Fact Sheets
The beginning of this section contains a two-sided Fact Sheet for each of the
models. These were developed based on comments received from EPA Regional
office staff, who identified a need to have quick access to some key model
descriptors. The Fact Sheets are designed to provide an "at a glance" overview of
the models' characteristics, scope, and applicability, as well as the name of a contact
for technical support and more information.
Model Descriptions
The latter part of this section contains more detailed information on the same
set of four models. This information was extracted from a computerized database
maintained by the International Ground Water Modeling Center (IGWMC) in
Golden, Colorado. The information has been re-formatted, and in some cases, sub-
sections have been re-numbered and re-ordered, but the information itself has been
modified only slightly. IGWMC's database contains information on many more
ground-water models. Using data that comes directly from the IGWMC database
will maintain the integrity of the information, if for example, OSWER wishes to
expand this portion of the Compendium in the future by taking larger extracts from
the IGWMC database.
-------
Section 4: Model Descriptions Introduction
Modeling Data Requirements
There is often confusion about how modeling data requirements are
determined and the relationship between data requirements and the model (code).
It is important to understand that the determination of a modeling application's
general data requirements should precede the selection of a model. Furthermore,
the determination of the type, quantity, and accuracy of the data required for the
modeling application can be complex because it is dependent upon:
1. Management's reasons for performing the modeling and the subsequent accuracy
and level of detail required of the modeling results. For example, a preliminary
screening of alternative remedial technologies conducted as part of a
remedial investigation and feasibility study can usually be modeled with
less detail than a remedial design being modeled as part of the remedial
design phase.
2. The physical processes and the characteristics of the site being modeled. For
example, chemical fate and transport processes often require considerably
more types of data than flow processes. Moreover, the quantity and the
accuracy required of the data are often a function of the types of chemicals
being modeled or the level of heterogeneity of the site.
3. The engineering objectives of the modeling. For example, the types and
quantity of data required to determine the effectiveness of a deep hydraulic
barrier design are often different from the data required to analyze a
network of ground-water extraction wells.
Once the data requirements have been determined, a model can be selected
which will support these data requirements and provide management with the
information they need to make informed decisions. Thus, the decision objectives,
process and site characteristics, and engineering objectives are the factors which
primarily determine a model application's data requirements and, in turn, govern
the selection of a particular model.
To determine if a model will support an application's data requirements, the
reader is referred to the model descriptions that follow and the model
documentation that is usually available from the model's authors or the
International Ground Water Modeling Center.
Page 4-2
-------
Model Fact Sheets
For
MOC
MODFLOW
PLASM
RANDOM WALK
SOURCE:
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Model Type: Solute Transport
MOC
(USGS-2D-TRANSPORT, KONBRED)
Version 3.0
Release Date: 11/89
Summary Updated 5/29/92
MOC is a ground-water flow and mass transport model. It provides capabilities for two dimensional simulation of
non-conservative solute transport in heterogeneous, anisotropic aquifers. It computes changes in time in the spatial
concentration distribution caused by convective transport, hydrodynamic dispersions, mixing or dilution from
recharge, and chemical reactions. The chemical reactions include first-order sorption irreversible rate reaction (e.g.
radioactive decay), equilibrium-controlled sorption with linear, Freundlich or Langmuir isotherms, and monovalent
and/or divalent ion-exchange reactions. MOC solves the finite difference approximation of the ground-water flow
equation using iterative ADI and SEP. It uses the method of characteristics followed by an explicit procedure to solve
the transport equation. MOC uses a subgrid of the flow grid for simulation of containment transport.
Note: A version of MOC called MOC Dense simulates two dimensional density dependent flow and transport in a
crossectional plane. The specific characteristics of MOC Dense are not incorporated in this summary.
Technical
Assistance:
Dr. David S. Burden, Director
Center for Subsurface Modeling Support (CSMoS)
Robert S. Kerr Environmental Research Laboratory (RSKERL)
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
(405) 332-8800
Availability
Usability
I
I
Reliability
I
s
1
I
I
cu
w
en
§
I
Distributed By:
International Ground Water Modeling Center
Colorado School of Mines
Golden, CO 80401, USA
(303) 273-3103
U.S. Geological Survey
WRD WGS - Mail Stop 433
National Center
Reston, Virginia 22092
Scientific Software Group
P.O. Box 23041
Washington D.C. 20026-3041
(703) 620-9214
Geraghty & Miller, Inc.
Modeling Group
1895 Preston Drive, Suite 301
Reston, Virginia 22091
(703) 476-0335
Scope
Remedial Design Feature
bO
5 O
II
60 SJ
.5 5
o
i
TD
C
I
en
I
i
£
a>
I
W
Contaminant Source Type
Solid Waste
Disposal
S,
§
o
8,
M
i
I
ra
Liquid Waste
Disposal
J3
I
O
00
i
I
73
|
I
1
Leakage
a
I
i
Cfi
s
I
CD
o
H
8
!
I
Likely
Possibly
Not Likely
O
Not
Applicable
N/A
Page 4-3
-------
Technical Characteristics
Model
Processes
o
E
/
Transport
/
Transfer
/
Flow
Boundary
Conditions
01
3
m
T?
Ol
1
in
/
X
E
4>
a
Cu
/
X
3
E
Ol
o
*
Ol
3
CQ
/
Flow
System
^
tr
"bb
in
/
ultiple Aquifer
2
iturated Zone
in
/
Transport
Boundary
Conditions
Ol
3
(S
n
01
8.
in
/
X
E
"S
vC
'0
en
/
X
E
Ol
o
c
o
01
3
us
saturated Zone
d
Aquifer
Type
Confined
/
Unconfined
Flow
Output
1
01
C-
01
l-l
3
£
TJ
S
/
.8
o
_o
;^
X
3
E
/
-4_,
60
03
Ol
1
/
Flow
Characteristics
(Saturated)
Dimension
2
Steady State
/
Transient
'
Transport
Processes
Advection
/
Transport
Output
ra
%
a-
^*
G
u
C
Q
U
/
c
2
O
1
ij
I
c
o
Ol
u
c
<8
tS
(A
1
/
Diffusion
/
Dispersion
/
Fate&
Transfer
Processes
Ol
2
/
Transfer
/
Flow
Solution
Technique
o
<
nite Difference
E
'
Output
Format
a
Q
F^
<
B
a
o
c
d
P
C
nJ
;f
O
E
P
?
2
Ol
0
e
^
p
«
"S
r^
p
inite Element
Hnmoi?cneous
Ub
/
Parameter
Representation
-- o
leterogeneous
>JU
' ^
u
'S,
8
"5
'
Anisotropic
'
Kgy
Tk * _ J _|
Model
M
Pre or Post
Processor
P
Both
B
Neither
X
Deffm'tions
Boundary Conditions
Specified Value - Values of head, concentration or temperature are specified along the boundary
(Dirichlet Condition)
Specified Flux - Row rate of water, contaminent mass or energy is specified along the boundary.
(Neumann Condition)
Value Dependent Flux - A specified flux is given for a specified value. (Cauchy Condition)
Hardware Prime Preprocessors: PREMOC Postprocessors:
Platforms: DEC VAX MODELCAD
IBM PC/XT/AT
IBM 80386/486
Apple Macintosh
POSTMOC
MOCGRAF
Detailed information on this model, sources of distribution, and postprocessors and preprocessors
is available in The Ground-Water Modeling Compendium.
Page 4-4
-------
Model Type: Saturated Flow
MODFLOW
Summary Updated: 5/29/92
Version 3.2
Release Date: 10/89
MODFLOW is a modular, block-centered finite difference model for the simulation of two dimensional and quasi-
or fully-three-dimensional, transient ground-water flow in anisotropic, heterogeneous, layered aquifer systems. It
calculates piezometric head distributions, flow rates and water balances. The model includes modules of flow
towards wells, through riverbeds, and into drains. Other modules handle evapotranspiration and recharge. Various
textual and graphic pre-and postprocessors are available.
Note: Several particle tracking programs including ModPath utilize Modflow output to simulate contaminant
transport. The specific characteristics of these particle tracking programs are not incorporated in this summary.
Technical
Assistance:
Dr. David S. Burden, Director
Center for Subsurface Modeling Support (CSMoS)
Robert S. Kerr Environmental Research Laboratory (RSKERL)
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
(405) 332-8800
Availability
Usability
'
XI
cu
S.
£
I
JS
c-
1
a
Reliability
§
a
I
I
I
3!
tJU
a
2
CD
$
Distributed By:
International Ground Water Modeling Center
Colorado School of Mines
Golden, CO 80401, USA
(303) 273-3103
U.S. Geological Survey
WRD WGS - Mail Stop 433
National Center
Reston, Virginia 22092
Scientific Publications Co.
P.O. Box 23041
Washington D.C. 20026-3041
(703) 620-9214
Geraghty & Miller, Inc.
Modeling Group
1895 Preston Drive, Suite 301
Reston, Virginia 22091
(703) 476-0335
Scope
Remedial Design Feature
60
.5 s
-a o
q o
il
D.
co
U
60
T3
I
U
|
.1
,5
-------
Technical Characteristics
Model
Processes
0
E
/
Transport
Transfer
Flow
Boundary
Conditions
3
S
1
CD
/
X
E
"8
1
^
X
3
4-*
1
g.
-------
Model Type: Saturated Flow
PLASM
Summary Updated: 5/29/92
Version - Illinois State Water Survey
Release Date: 1985
PLASM (Prickett Lonnquist Aquifer Simulation Model) is a finite difference model for simulation of transient,
two-dimensional or quasi-three-dimensional flow in a single or multi-layered, heterogeneous, anistropic aquifer
system. The original model of 1971 consisted of a series of separate programs for various combinations of
simulation options. Later versions combined most of the options in a single code, including variable pumping
rates, leaky confined aquifer conditions, induced infiltration from a shallow aquifer or a stream, storage coefficient
conversion between confined and waterable conditions, and evapotranspiration as a function of depth to
watertable. The model uses the iterative alternating implicit method (IADI) to solve the matrix equation.
Technical
Assistance:
Dr. David S. Burden, Director
Center for Subsurface Modeling Support (CSMoS)
Robert S. Kerr Environmental Research Laboratory (RSKERL)
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
(405) 332-8800
Availability
Usability
'(0
6
u
.n
a
cu
Reliability
I
i
§
I
Distributed By:
International Ground Water Modeling Center
Colorado School of Mines
Golden, CO 80401, USA
(303) 273-3103
Thomas A. Prickett
6 G.H. Baker Drive, Urbana IL 61801
(217) 384-0615
Illinois State Water Survey
P.O. Box 232
Urbana, Illinois 61801
Ann Koch
2921 Greenway Drive
Ellicott City, Maryland 21043
(301)461-6869
Geraghty & Miller, Inc., Modeling Group
1895 Preston Drive, Suite 301
Reston, Virginia 22091
(703)476-0335
Scope
Remedial Design Feature
bo
a.
<3
U
bo
-------
Technical Characteristics
Model
Processes
o
E
/
Transport
Transfer
Flow
Boundary
Conditions
a
u
S.
in
/
3
E
1
'u
8L
/
X
3
E
Ol
o
Ol
Q
3
a
/
How
System
ingle Aquifer
tn
/
ultiple Aquifei
M
X
jturated Zone
rn
/
Transport
Boundary
Conditions
o>
3
«
3
S.
U5
E
1
8.
C/5
X
3
rr
b
8.
Q
01
3
<3
O)
saturated Zon
c
D
Aquifer
Type
Confined
/
Unconfined
/
Flow
Output
1
S
o
-^
0)
u
3
C_
T3
S
X
/
01
0
o
£
en
3
E
a!
M)
a
t£>
f?
/
Flow
Characteristics
(Saturated)
Dimension
3
Steady State
/
Transient
/
Transport
Processes
Advection
Transport
Output
.-
C/)
0
g
u
"cu
£
c
S
o
a
<0
S
4^
g
Q
8
.c
§
o>
u
C
B
10
Diffusion
Dispersion
Fate&
Transfer
Processes
-------
Model Type: Solute Transport
RANDOM WALK
Version - Illinois State Water Survey
Release Date: 7/81
Summary Updated: 5/29/92
RANDOM WALK/TRANS is a numerical model to simulate two-dimensional steady or transient flow and
transport problems in heterogeneous aquifers under water table and/or confined or leaky confined conditions.
The flow is solved using a finite difference approach and the iterative alternating direction implicit method. The
advective transport is solved with a partide-in-cell method, while the dispersion is analyzed with the random
walk method.
Note: A number of other versions of Random Walk including analytical function driven versions are available.
The specific characteristics of these other versions including output capababilities vary considerably.
Technical
Assistance:
Dr. David S. Burden, Director
Center for Subsurface Modeling Support (CSMoS)
Robert S. Kerr Environmental Research Laboratory (RSKERL)
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
(405) 332-8800
Availability
Usability
I
en
I
I
Reliability
§
a
I
1
=3
I
Distributed By:
International Ground Water Modeling Center
Colorado School of Mines
Golden, CO 80401, USA
(303) 273-3103
Thomas A. Prickett
6 G. H. Baker Drive
Urbana,IL 61801
(217) 384-0615
Bob Sinclair, Director of Computer Service
Illinois State Water Survey
Box 5050, Station A
Champaign, IL 61820
(217) 3334952
Geraghty & Miller, Inc., Modeling Group
1895 Preston Drive, Suite 301
Reston, Virginia 22091
(703) 476-0335
Scope
Remedial Design Feature
8
bo
60
S o
O
bo
T3
I
u
2
o
1
bo
is
1
en
bo
I
o
J3
(0
8
X
w
Contaminant Source Type
Solid Waste
Disposal
S,
CO
S
vg
I
en
o
nj
6
ra
Liquid Waste
Disposal
i
a
M
I
w
1
iif
8
1
(A
I
*
o
bo
i.
1
T3
£
I
1
en
01
r
S
Leakage
Is
(2
en
CD
a
s
en
1
I
en
Key.
Likely
Possibly
Not Likely
O
Not
Applicable
N/A
Pmge 4-9
-------
Technical Characteristics
Model
Processes
1
E
/
Transport
/
Transfer
Flow
Boundary
Conditions
8
"S
8.
en
S
E
«
'0
8.
C/)
/
X
3
E
01
o
8.
01
D
_3
>
/
Flow
System
"3
J-
"5b
in
/
Multiple Aquifer
Saturated Zone
/
Transport
Boundary
Conditions
/
E
1
C/i
'
X
3
1
!L
01
_3
>
Unsaturated Zone
Aquifer
Type
Confined
/
Unconfined
/
Flow
Output
"c
Ol
b
c-
Ol
i.
3
£
o
S
X
s
ft
o
o
V
$
X
3
E
/
bC
o
03
_0)
^
Flow
Characteristics
(Saturated)
Dimension
2
S
v5
o
1
/
Transient
/
Transport
Processes
Advection
/
Transport
Output
a
C/3
"p
0*
_c
0
s
U
/
=3
£
.5
in
C
0
£
m
!
£
o
U
/
$
c
o
/
Anisotropic
/
Key
Model
M
Pre or Post
Processor
P
Both
B
Neither
X
Definitions
Boundary Conditions
Specified Value - Values of head, concentration or temperature are specified along the boundary.
(Dirichlet Condition)
Specified Flux - Flow rate of water, containment mass or energy is specified along the boundary.
(Neumann Condition)
Value Dependent Flux - A specified flux is given for a specified value. (Cauchy Condition)
Hardware Cyber 175 Preprocessors: PREWALK Postprocessors:
Platforms: VAX 11/780 MODFJLCAD
IBM PC/XT/AT
POSTWALK
Detailed information on this model, sources of distribution, and postprocessors and preprocessors
is available in The Ground-Water Modeling Compendium.
Page 4-10
-------
Detailed Model Descriptions
For
MOC
MODFLOW
PLASM
RANDOM WALK
SOURCE:
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Model Description For
MOC
(USGS-2D- TRANSPORT/KONBRED)
SOURCES
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Table of Contents
PAGE
1. Model Identification 4a-1
1.1. Model Name(s) 4a-l
1.2. Date of First Release 4a-l
1.3. Current Version 4a-l
1.4. Current Release Date 4a-l
1.5. Author 4a-l
2. Model Information.......... 4a-1
2.1. Model Category 4a-l
2.2. Model Developed For 4a-l
2.3. Units of Measurement Used 4a-l
2.4. Abstract 4a-l
2.5. Data Input Requirements 4a-2
2.6. Versions Exist For The Following Computer Systems 4a-2
2.7. System Requirements 4a-2
2.8. Graphics Requirements 4a-2
2.9. Program Information 4a-2
3. General Model Capabilities................... ........ ........ 4a-2
3.1. Parameter Discretization 4a-2
3.2. Coupling 4a-2
3.3. Spatial Orientation 4a-3
3.4. Types of Possible Updates 4a-3
3.5. Geostatistics and Stochastic Approach 4a-3
3.6. Comments 4a-3
4. Flow Characteristics[[[ 4a-3
4.1. Flow System Characterization 4a-3
4.2. Fluid Conditions 4a-4
4.3. Boundary Conditions 4a-4
4.4. Solution Methods for Flow 4a-4
4.5. Grid Design 4a-5
4.6. Row Output Characteristics 4a-5
5. Mass Transport Characteristics...........................................4a-5
5.1. Water Quality Constituents 4a-5
5.2. Processes Included 4a-5
5.3. Boundary Conditions 4a-6
5.4. Solution Methods for Transport 4a-6
5.5. Output Characteristics for Transport 4a-6
6. Evaluation[[[4a-6
-------
PACE
7. Documentation and Support........ 4a-7
7.1. Documentation Includes 4a-7
7.2. Support Needs 4a-7
7.3. Level of Support 4a-7
8. Availability........................... 4a-7
8.1. Terms 4a-7
8.2. Form 4a-7
9. Pre and Post Processors....... 4a-8
9.1. Data Preprocessing 4a-8
9.2. Data Postprocessing 4a-8
10. Institution of Model Development.................................. 4a-8
10.1. Name 4a-8
10.2. Address 4a-8
10.3. Type of Institution 4a-8
11. Remarks[[[ 4a-8
-------
MOC
1. Model Identification
1.1. Model Name(s)
USGS-2D-TRANSPORT
MOC
KONBRED
1.2. Date of First Release
11/76
1.3. Current Version
3.0
1.4. Current Release Date
11/89
1.5. Author
1. Konikow, L.F.
2. Bredehoeft, J.D.
2. Model Information
2.1. Model Category
ground-water flow
mass transport
2.2. Model Developed For
general use (e.g. in field applications)
2.3. Units of Measurement Used
SI system
metric units
US customary units
any consistent system
2.4. Abstract
MOC is a two-dimensional model for the simulation of non-conservative
solute transport in heterogeneous, anisotropic aquifers. It computes changes
in time in the spatial concentration distribution caused by convective
transport, hydrodynamic dispersion, mixing or dilution from recharge, and
chemical reactions. The chemical reactions include first-order irreversible
rate reaction (e.g. radioactive decay), equilibrium-controlled sorption with
linear, Freundlich or Langmuir isotherms, and monovalent and/or divalent
ion-exchange reactions. MOC solves the finite difference approximation of
the ground-water flow equation using iterative ADI and SIP. It uses the
method of characteristics followed by an explicit procedure to solve the
transport equation.
Pago 4m-1
-------
Section 4: Model Descriptions Model Description for HOC
2.5. Data Input Requirements:
Data input requirements are provided in the model documentation and are
discussed in the introduction to Section 4.0 of the Compendium.
2.6. Versions Exist For The Following Computer Systems
minicomputer
workstations
mainframe
microcomputer
Make/Model
~ Prime
~ DEC VAX
-IBM PC/XT/AT
- operating system
MS DOS
- IBM 80386/486
- operating system
MS DOS
OS/2
Unix
~ Apple Macintosh
2.7. System Requirements
core memory (RAM) for execution (bytes)
- 640Kb (standard IBM PC version)
mass storage (disk space in bytes)
- at least 2Mb for data files
numeric/math coprocessor
(for micro computers)
compiler required
(for mainframe)
2.8. Graphics Requirements
none
2.9. Program Information
programming language/level
- Fortran 77
number of program statements (total)
- 2000
3. General Model Capabilities
3.1. Parameter Discretization
distributed
3.2. Coupling
none
Page 4a-2
-------
Sect/on 4: Model Descriptions Model Description for HOC
3.3. Spatial Orientation
saturated flow
2D-horizontal
2D-vertical
3.4. Types of Possible Updates
parameter values
boundary conditions
3.5. Geostatistics and Stochastic Approach
none
3.6. Comments
This model has restart capability
4. Flow Characteristics
4.1. Flow System Characterization
Saturated Zone
System
single aquifer
Aquifer Type(s)
- confined
- semi-confined (leaky-confined)
Medium
porous media
Parameter Representation
homogeneous
- heterogeneous
isotropic
anisotropic
Flow Characteristics (Saturated Zone)
laminar flow
linear (Darcian flow)
steady-state
- transient
Flow Processes Included
areal recharge
induced recharge (from river)
Changing aquifer conditions
in space
variable thickness
Page
-------
Section 4i Model Descriptions Model Description for HOC
Well Characteristics
none
Unsaturated Zone
none
4.2. Fluid Conditions
Single Fluid Flow
- water
Flow of Multiple Fluids
none
Fluid Properties
constant in time/space
4.3. Boundary Conditions
First Type - Dirichlet
head/pressure
Second type - Neumann (Prescribed Flux)
injection/production wells
no flow boundary
areal boundary flux
ground-water recharge
Third Type - Cauchy
head/pressure-dependent flux
4.4. Solution Methods for Flow
General Method
Numerical
Spatial Approximation
finite difference method
block-centered
Time-Stepping Scheme
- Crank-Nicholson
Matrix-Solving Technique
- SIP
- iterative ADIP
Page 4a-4
-------
Section 4: Model Descriptions Model Description for MQC
4.5. Grid Design
Cell/Element Characteristic
- constant cell size
variable cell size
Possible Cell Shapes
2D-square
- 2D-rectangular
Maximum Number of Nodes
- 2000
4.6. Flow Output Characteristics
Simulation Results
Head/Pressure/Potential
- ASCII file (areal values)
- ASCH file (hydrograph)
Water Budget Components
- ASCH file (global total area)
5. Mass Transport Characteristics
5.1. Water Quality Constituents
any component(s)
single component
total dissolved solids (TDS)
inorganics
organics
radionuclides
5.2. Processes Included
(Conservative) Transport
advection
dispersion (isotropic; anisotropic)
diffusion
Phase Transfers
- Solid <-> Liquid
- sorption equilibrium isotherm
- linear
- Langmuir
- Freundlich
Fate
- first-order radioactive decay (single mother/daughter decay)
- first-order chemical decay
first-order microbial decay
-------
Sect/on 4i Model Descriptions Model Description for MOC
5.3. Boundary Conditions
First Type - Dirichlet
- Chemical processes embedded in transport equation
- concentration
Second Type - Neumann (Prescribed Solute Flux)
areal boundaries
injection wells
point sources
line sources
areal sources
5.4. Solution Methods for Transport
General Method
numerical
uncoupled flow and transport equation
Spatial Approximation
finite difference method
block-centered
particle-tracking
Time-Stepping Scheme
- fully explicit
Matrix-solving Technique
- method of characteristics
5.5. Output Characteristics for Transport
Simulation Results
Concentration in Aquifer/Soil
- ASCII file (areal values)
- ASCH file (time series)
Mass Balance Components
- ASCH file (global total area)
6. Evaluation
6.1. Verification/Validation
verification (analytic solutions)
laboratory data sets
field datasets (validation)
synthetic datasets
code intercomparison
6.2. Internal Code Documentation (Comment Statements)
incidental
Page 4a-6
-------
Section 4: Model Descriptions Model Description for HOC
6.3. Peer (Independent) Review
concepts
theory (math)
coding
accuracy
7. Documentation and Support
7.1. Documentation Includes
model theory
user's instructions
example problems
code listing
7.2. Support Needs
Can be used without support
Support is available
from author
from third parties
7.3. Level of Support
limited
support agreement available
8. Availability
8.1. Terms
available
public domain
proprietary
8.2. Form
source code only (tape/disk)
source and compiled code
compiled code only
Pago 4m-7
-------
Section 4: Model Descriptions Model Description lor HOC
9. Pre and Post Processors
9.1. Data Preprocessing
name: PREMOC
separate (optional) program
generic (can be used for various models)
- textual data entry/editing
name: MODELCAD
separate (optional) program
generic (can be used for various models)
- textual data entry/editing
- graphic data entry/modification (e.g manual grid design, arrays)
data reformatting (e.g. for GIS)
- error-checking
help screens
9.2. Data Postprocessing
name: POSTMOC
separate (optional) program
reformatting (e.g. to standard formats)
10. Institution of Model Development
10.1. Name
U.S. Geological Survey
10.2. Address
National Center
Reston, Virginia
10.3. Type of Institution
federal/national government
11. Remarks
The MOC package distributed by the International Ground Water Modeling Center
includes a preprocessor (PREMOC) to prepare input files, a postprocessor
(POSTMOC) to reformat parts of the output file to allow the import the results in
graphic display programs, and two versions of the MOC simulation program,
MOCADI and MOCSIP. MOCADI and MOCSIP are identical apart from the methods
used to solve the finite difference flow equations. The IGWMC distributes both
IBM-PC and mainframe versions.
Page 4a-8
-------
Section 4: Model Descriptions Model Description for MOC
Contact the International Ground Water Modeling Center for latest information:
IGWMC USA: Inst. for Ground-Water Res. and Educ.,
Colorado School of Mines,
Golden, CO 80401, USA.
IGWMC Europe: TNO Institute of Applied Geoscience,
P.O. Box 6012, 2600JA Delft,
The Netherlands.
IBM-PC, Macintosh and mainframe versions of the MOC code are also available
from:
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
(703) 620-9214
MOCGRAP is a program developed by TECSOFT, Inc. to provide graphics capability
to MOC. It uses the output from MOC to contour heads and concentrations and to
plot velocity vectors. It supports a variety of graphic screen formats, printers and
plotters. MOCGRAF requires TECSOFTs TRANSLATE program. MOCGRAF is
available from Scientific Software Group
MACMOC is the implementation of the USGS Method of Characteristics
Solute Transport Model (MOC) for the Apple Macintosh. The data input
editor, simulation code and output postprocessor are integrated in a single
application. Graphic output includes head and concentration contouring,
and velocity vector plotting. It requires a Macintosh Plus with System 6.02
and Finder 6.1 or higher and at least 2Mb RAM. MACMOC is available
from the Scientific Software Group.
Notes on computer program updates have been published by the USGS, Reston,
Virginia, on the following dates:
1. May 16,1979 7. Jul. 26,1985 13. Oct. 20,1986
2. Mar. 26,1980 8. Jul. 31,1985 14. Mar. 2,1987
3. Dec. 4,1980 9. Aug. 2,1985 15. Mar. 5,1987
4. Aug. 26,1981 10. Aug. 8,1985 16. Jan. 29,1988
5. Oct. 12,1983 11. Aug. 12,1985 17. Nov. 21,1988
6. Jun. 10,1985 12. Jul. 2,1986 18. Jul. 20,1989
A modification of this model to track representative water of tracer particles initially
loaded along specific lines has been developed by Garabedian and Konikow (1983):
-------
Section 4s Model Descriptions Model Description for MOC
TRACK (see IGWMC key # 0741).
The code has been modified by Hutchinson to allow head-dependent flux as a
boundary condition:
y
Hutchinson, C.B. et al. 1981. Hydrogeology of Well-field Areas near Tampa,
Florida. USGS Open-File Report 81-630.
A modification to allow linear and non-linear sorption isotherms and first-order
decay was introduced in 1982:
Tracy, J.V. 1982. Users Guide and Documentation for Adsorption and Decay
Modifications of the USGS Solute Transport Model. NUREG/CR-2502, Div.
of Waste Management, Off. of Nuclear Material Safety and Safeguard, U.S.
Nuclear Regulatory Comm., Washington, D.C.
MODELCAD is a graphical oriented, model-independent preprocessor to prepare
and edit input files for two- and three-dimensional ground-water models, including
aquifer properties, boundary conditions, and grid dimensions. The program
prepares input files for MODFLOW, MOC, PLASM and RANDOM WALK, among
others. File formatting routines for other models are available upon request.
Contact:
Geraghty & Miller Modeling Group
1895 Preston White Drive
Suite 301, Reston, VA 22091
(703) 476-0335
Strecker,E.W., W-S. Chu, and D. P. Lettenmaier. 1985. Evaluation of Data
Requirements for Groundwater Contaminant Transport Modeling. Water
Resources Series, Techn. Rept. 94, Univ. of Washington, Seattle, Wash.
In this study a parameter identification algorithm was used together with the USGS-
MOC code and applied to two synthetic aquifers, evaluating the effects of data
availability and uncertainty on ground-water contaminant transport prediction.
The parameter identification algorithm is based on constrained least-squares
minimization.
Also:
Strecker, E.W., and W-s. Chu. 1986. Parameter Identification of a Ground-Water
Contaminant Transport Model. Ground Water, Vol. 24(1), pp. 56-62.
Page 4a-f 0
-------
Sect/on 4s Model Descriptions Model Description for MOC
A modified version of the 1978 version of the USGS MOC model has been presented
by Kent et Al. (1983; see references). Modifications include water-table option for
flow and non-linear sorption. The same authors developed a menu-driven,
preprocessor for their version of MOC. For more information contact Dr. Douglas C.
Kent, School of Geol., Oklahoma State University, Stillwater, Oklahoma.
A stochastics-based analysis of the performance of the MOC model in remedial
action simulations is discussed in:
El-Kadi, A.I. 1988. Applying the USGS Mass-Transport Model (MOC) to
Remedial Actions by Recovery Wells. Ground Water, Vol. 26(3), pp. 281-288.
An IBM PC/386 extended memory version of this model is also available from:
Geraghty & Miller, Inc.
Modeling Group
1895 Preston Drive, Suite 301
Reston, VA 22091
tel.: 703/476-0335
fax: 703/476-6372
Page 4a-11
-------
Sect/on 4: Model Descriptions Model Description for HOC
12. References
Finder, G.F. and H.H. Cooper. 1970. A Numerical Technique for Calculating the
Transient Position of the Saltwater Front. Water Resources Research, Vol.
6(3), pp. 875-882.
Bredehoeft, J.D., and G.F. Finder. 1973. Mass Transport in Flowing
Groundwater. Water Resources Research, Vol. 9(1), pp. 194-210.
Konikow, L.F., and J.D. Bredehoeft. 1974. Modeling Flow and Chemical Quality
Changes in an Irrigated Stream-Aquifer System. Water Resources Research,
Vol. 10(3), pp. 546-562.
Konikow, L.F., and J.D. Bredehoeft. 1978. Computer Model of Two-Dimensional
Transport and Dispersion in Ground Water. USGS Techniques of Water
Resources Investigations, Book 7, Chapter 2, U.S. Geological Survey, Reston,
Virginia.
Kent, D.C., J. Alexander, L. LeMaster, and J. Wagner. 1983. Interactive
Preprocessor Program for the U.S.G.S. Konikow Solute Transport Model.
Oklahoma State University, School of Geology, Stillwater, Oklahoma.
Kent, D.C., M.M. Hoque, L. LeMaster, and J. Wagner. 1983. Modifications to the
U.S.G.S. Solute Transport Model. School of Geology, Oklahoma State
University, Stillwater, Oklahoma.
Goode, D.J., and L.F. Konikow. 1989. Modification of a Method-of-Characteristics
Solute Transport Model to Incorporate Decay and Equilibrium-Controlled
Sorption or Ion Exchange. USGS Water Resources Investigations Report 89-
4030, U.S. Geological Survey, Reston, Virginia.
Page 4a-12
-------
Section 4; Model Descriptions Model Description for MOC
13. Users
Spinazola, J.M., J.B. Gillespie, and RJ. Hart. Ground-Water Flow and Solute
Transport in the Equus Beds Area, South Central Kansas, 1940-79. Water-
Resources Investig. Rept. 85-4336, Lawrence, Kansas.
The transport model was used to study the movement of chloride in the past
and under the various proposed pumping development schemes. The
sources of the chloride is oilfield brine moving towards the wellfields and the
Arkansas river.
Robertson, J.B. 1974. Digital Modeling of Radioactive and Chemical Waste
Transport in the Snake River Plain Aquifer at the National Reactor Testing
Station, Idaho. USGS Open-File Report IDO-22054. U.S. Geological Survey,
Boise, Idaho.
Konikow, L.F. 1977. Modeling Chloride Movement in the Alluvial Aquifer at
the Rocky Mountain Arsenal, Colorado. USGS Water Supply Paper 2044, U.S.
Geological Survey, Denver, Colorado.
Sohrabi, T. 1980. Digital Transport Model Study of Potential Nitrate
Contamination from Sceptic Tank Systems near Edmond, Oklahoma. Rept.
80-38, National Center for Ground Water Research, University of Oklahoma,
Norman, Okla.
Sophocleous, M.A. 1984. Groundwater Flow Parameter Estimation and Quality
Modeling of the Equus Beds Aquifer in Kansas, USA. Journ. of Hydrology,
Vol. 69, pp. 197-222.
In an accompanying Technical Completion Report of the Kansas Water
Resources Research Institute at the University of Kansas, Lawrence,
Sohocleous et al. compares the MOC code with Finder's ISOQUAD-4 code and
Grove's finite difference solute transport code.
Freeberg, K.M. 1985. Ground Water Contaminant Modeling Applied to Plume
Delineation and Aquifer Restoration at an Industrial Site. M.Sc. Thesis, Dept.
of Env. Sciences and Eng., Rice Univ., Houston, Texas.
Groschen, G.E. 1985. Simulated Effects of Projected Pumping on the Availability
of Fresh Water in the Evangeline Aquifer in an Area Southwest of Corpus
Christi, Texas. USGS Water Resources Investigations Report 85-4182, Austin,
Texas.
Chapelle, F.H. 1986. A Solute Transport Simulation of Brackish Water Intrusion
near Baltimore, Maryland. Ground Water, Vol. 24(3), pp 304-311.
The model was used to estimate the future movement of the brackish-water
plume in the coarse sands and gravels of the Patuxent Formation based on
alternative strategies of aquifer use. (see also Maryland Geological Survey
Report of Investig. 43 (1986)).
""~"~~'~~1"~~~"^ Page 4»-13
-------
Sect/on 4: Model Descriptions Model Description for MOC
Davis, A.D. 1986. Deterministic Modeling of Dispersion in Heterogeneous
Permeable Media. Ground Water, Vol. 24(5), pp. 609-615.
Perez, R. 1986. Potential for Updip Movement of Saline Water in the Edwards
Aquifer, San Antonio, Texas. USGS Water Resources Investigations Report
86-4032, U.S. Geological Survey, Austin, Texas.
Bond, L.D., and J.D. Bredehoeft. 1987. Origins of Sea water Intrusion in a Coastal
AquiferA Case Study of the Pajaro Valley, California. Journ. of Hydrology,
Vol. 92, pp.363-388.
Mixon, P.O., A.S. Damle, R.S. Truesdale, and C.C. Allen. 1987. Effect of
Capillarity and Soil Structure on Flow in Low permability Saturated Soils at
Disposal Facilities. EPA/600/2-87/029, Hazardous Waste Eng. Research Lab.,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
Patterson, C-V. 1987. Risk Analysis of Annular Disposal of Oil and Gas Well
Brines. M.Sc. Thesis, Dept. of Systems Eng., Case Western Reserve
University, Cleveland, Ohio.
Al-Layla, R., H. Yazicigil, and R. de Jong. 1988. Numerical Modeling of Solute
Transport Patterns in the Damman Aquifer. Water Resources Bulletin, Vol.
24(1), pp. 77-85.
Used to predict the extent of the saline intrusion in this carbonate rock aquifer
in the Kingdom of Saudi Arabia and studying the effects of changes in the
hydrologic regime on TDS concentration. The model was modified to
include the effects of salt gradients resulting from vertical leakage into the
host aquifer.
Satkin, R.L., and P.b. Bedient. 1988. Effectiveness of Various Aquifer Restoration
Schemes under Variable Hydrogeologic Conditions. Ground Water, Vol.
26(4), pp. 488-498.
Yager, R.M. and M.P. Bergeron. 1988. Nitrogen Transport in a Shallow Outwash
Aquifer at Olean, Cattaraugus County, New York. Water-Resources
Investigations Report 87-4043, U.S. Geological Survey, Ithaca, New York.
Page 4a-14
-------
Model Description For
MODFLOW
SOURCE:
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Table of Contents
PAGE
1. Model Identification ......... .......................... . .......... ........ ..... 4b-1
1.1. Model Name(s) [[[ 4b-l
1.2. Date of First Release [[[ 4b-l
1.3. Current Version # [[[ 4b-l
1.4. Current Release Date [[[ 4b-l
1.5. Authors [[[ 4b-l
2. Model Information ..... .......... ....... .. ................ ...................... 4b-1
2.1. Model Category [[[ 4b-l
2.2. Model Developed For [[[ 4b-l
2.3. Units of Measurement Used [[[ 4b-l
2.4. Abstract [[[ 4b-l
2.5. Data Input Requirements [[[ 4b-l
2.6. Versions Exist for the Following Computer Systems ................ 4b-2
2.7. System Requirements [[[ 4b-2
2.8. Graphics Requirements [[[ 4b-2
2.9. Program Information [[[ 4b-2
3. General Model Capabilities................................................ 4b*2
3.1. Parameter Discretization [[[ 4b-2
3.2. Coupling [[[ 4b-2
3.3. Spatial Orientation [[[ 4b-2
3.4. Types of Possible Updates [[[ 4b-3
3.5. Geostatistics and Stochastic Approach .......................................... 4b-3
3.6. Comments [[[ 4b-3
4. Flow Characteristics[[[ 4o-3
4.1. Row System Characterization [[[ 4b-3
4.2. Fluid Conditions [[[ 4b-4
-------
PAGE
8. Pre and Post Processors 4b-6
8.1. Data Preprocessing 4b-6
8.2. Data Postprocessing 4b-7
9. Institution of Model Development.. 4b-7
9.1 Name 4b-7
9.2. Address 4b-7
9.3. Type of Institution 4b-7
10. Remarks 4b-7
11. References................ 4b-1 1
12. Users 4b-12
Page 4b-ii
-------
MOD FLOW
1. Model Identification
1.1. Model Name(s)
MODFLOW
1.2. Date of First Release
6/83
1.3. Current Version #
3.2
1.4. Current Release Date
10/89
1.5. Authors
1. McDonald, M.G.
2. Harbaugh, A.W.
2. Model Information
2.1. Model Category
ground-water flow
2.2. Model Developed For
general use (e.g. in field applications)
research (e.g. hypothesis/theory testing)
demonstration/education
2.3. Units of Measurement Used
SI system
metric units
any consistent system
2.4. Abstract
MODFLOW is a modular, block-centered finite difference model for the
simulation of two-dimensional and quasi- or fully-three-dimensional,
transient ground-water flow in anisotropic, heterogeneous, layered
aquifer systems. It calculates piezometric head distributions, flow rates
and water balances. The model includes modules for flow towards
wells, through riverbeds, and into drains. Other modules handle
evapotranspiration and recharge. Various textual and graphic pre- and
postprocessors are available.
2.5. Data Input Requirements:
Data input requirements are provided in the model documentation
and are discussed in the introduction to Section 4.0 of the Compendium.
Page 4b-1
-------
Section 4: Model Descriptions Model Description for MODFLOW
2.6. Versions Exist for the Following Computer Systems
supercomputer
minicomputer
workstations
mainframe
microcomputer
Make/Model
- IBM-PC/XT/AT
- operating system
DOS 2.1 or later
- DEC VAX 11/780
- PRIME 750
- Macintosh
- Intel 80386/80486 based computers
2.7. System Requirements
core memory (RAM) for execution (bytes)
- 640Kb (standard IBM PC version)
mass storage (disk space in bytes)
- at least 2Mb for data files
numeric/math coprocessor
(for micro computers)
compiler require
(for mainframe)
2.8. Graphics Requirements
none
2.9. Program Information
programming language /level
- Fortran 77
number of program statements (total)
- 5000 lines
size of runtime (compiled) version (bytes)
- 580Kb (IBM-PC version)
3. Genera/ Model Capabilities
3.1. Parameter Discretization
distributed
3.2. Coupling
N.A.
3.3. Spatial Orientation
saturated flow
2D-horizontal
2D-vertical
3D-layered (quasi 3D)
Page 4b-2
-------
Section 4i Model Descriptions Model Description for iiODFLOW
3.4. Types of Possible Updates
parameter values
boundary conditions
3.5. Geostatistics and Stochastic Approach
optional
3.6. Comments
This model has restart capability.
4. Flow Characteristics
4.1. How System Characterization
Saturated Zone
System:
single aquifer
single aquifer/aquitard system
multiple aquifer/aquitard systems
Aquifer Type(s)
confined
semi-confined (leaky-confined)
- unconfined (phreatic)
Medium:
porous media
Parameter Representation
homogeneous
heterogeneous
isotropic
- anisotropic
Flow Characteristics (Saturated Zone)
laminar flow
linear (Darcian flow)
steady-state
transient
Flow Processes Included
areal recharge
induced recharge (from river)
evapotranspiration
Changing Aquifer Conditions
in space
- variable thickness
- confined/unconfined
- pitching aquitard
-------
Sect/on 4s Model Descriptions Model Description for MODFLOW
in time
- desaturation(saturated/unsat.)
- confined/unconfined
- resaturation of dry cells
Well Characteristics
partial penetration
4.2. Fluid Conditions
Single Ruid Flow
water
Flow of Multiple Fluids
- N.A.
Fluid Properties
constant in time/space
4.3. Boundary Conditions
First Type - Dirichlet
head/pressure
Second Type - Neumann (Prescribed Flux)
- injection/production wells
no flow boundary
areal boundary flux
ground-water recharge
seepage face
springs
induced infiltration
Third Type - Cauchy
- head/pressure-dependent flux
free surface (steady-state; movable)
4.4. Solution Methods for Flow
General Method
- Numerical
Spatial Approximation
- finite difference method
block-centered
Time-Stepping Scheme
fully implicit
Matrix-Solving Technique
- Iterative
- SIP
- LSOR
Page 4b-4
-------
Section 4i Model Descriptions Model Description lor iiODFLOW
4.5. Grid Design
Cell/Element Characteristic
constant cell size
variable cell size
- 3D-hexahedral
Possible Cell Shapes
2D-square
2D-rectangular
- 3D-cubic
Maximum Number of Nodes
- 9999
4.6. Row Output Characteristics
Simulation Results
Head / Pressure /Potential
- binary file (areal values)
- ASCII file (areal values)
- binary file (hydrograph)
- ASCH file (hydrograph)
Fluxes/Velocities
- binary file (areal values)
- ASCH file (areal values)
- binary file (temporal values)
- ASCII file (temporal values)
Water Budget Components
- ASCn file (cell-by-cell values)
- ASCH file (global total area)
5. Evaluation
5.1. Verification/Validation
verification (analytical solutions)
synthetic datasets
code intercomparison
5.2. Internal Code Documentation (Comment Statements)
sufficient
5.3. Peer (Independent) Review
concepts
theory (math)
coding
accuracy
documentation
usability
Pag* 4b-3
-------
Section 4i Model Descriptions Model Description lor MODFLOW
6. Documentation and Support
6.1. Documentation Includes
model theory
user's instructions
example problems
code listing
program structure and development
verification/validation
6.2. Support Needs
Can be used without support
Support is available
from author
from third parties
6.3. Level of Support
limited
support agreement available
7. Availability
7.1. Terms
available
public domain
proprietary
7.2. Form
source code only (tape/disk)
source and compiled code
compiled code only
paper listing of source code
8. Pro and Post Processors
8.1. Data Preprocessing
name: MODELCAD
- separate (optional) program
- graphic data entry/modification (e.g manual grid design,
arrays)
- (semi-) automatic grid generation
- data reformatting (e.g. for GIS)
error-checking
- help screens
name: PREMOD
- part of model package (dedicated)
- textual data entry/editing
Page 46-6
-------
Section 4i Model Descriptions Model Description for MODFLOW
8.2. Data Postprocessing
name: POSTMOD
- part of model package (dedicated)
- textual data display on screen/printer
- reformatting (e.g. to standard formats)
9. Institution of Model Development
9.1 Name
U.S. Geological Survey
9.2. Address
Ground Water Branch
WRD WGS - Mail Stop 433
National Center
Reston, Virginia 22092
9.3. Type of Institution
federal/national government
10. Remarks
The code is available from the U.S.G.S. on tape ($40). Contact Arlen
Harbaugh (see contact address). The documentation (paper copy $69.95,
microfiche $3.50) is available from:
U.S. Geological Survey
Open-File Service Section
Branch of Distribution
Box 25425, Federal Center
Denver, CO 80225
Various MODFLOW implementations for IBM-PC and main frame systems
are also available from the International Ground Water Modeling Center (see
also MARS # 3983, 3984, 3985, and 3986), including MODFLOW PC for IBM
PC/XT/AT (640K), and MODFLOW PC/EXT for extended memory Intel
80386/80486 based computers (2Meg,4Meg) which includes a preconditioned
conjugate gradient solver and a stream-flow routing package.
IGWMC USA: Inst. for Ground-Water Res. and Educ.,
Colorado School of Mines,
Golden, CO 80401, USA.
IGWMC Europe: TNO Institute of Applied Geoscience,
P.O. Box 6012, 2600JA Delft,
The Netherlands.
Page 46-7
-------
Section 4i Model Descriptions Model Description for MODFLOW
Wagner, Heindel and Noyes Inc. has implemented MODFLOW on a Hewlett-
Packard microcomputer (series 200). Contact Jeffrey E. Noyes, Geologist:
Wagner, Heindel and Noyes, Inc.
285 North St.
Burlington, Vermont 05401
phone: (802) 658-0820.
Various implementations of MODFLOW are available from:
Scientific Publications Co.
P.O. Box 23041
Washington, DC 20026-3041
phone: 703/620-9214
fax: 703/620-6793.
Versions included:
MODFLOW/PC for IBM PC/XT/AT (640K)
MODFLOW/EM for extended memory Intel 80386/80486 based
computers (2Meg,4Meg) which includes a preconditioned conjugate
gradient solver and a stream-flow routing package.
MOCGRAF is a program developed by TECSOFT, Inc. to provide graphics
capability to MODFLOW. The menu-driven MODGRAF program uses the
output from MODFLOW to automatically contour heads and drawdowns
from each layer, stress period and time step and superpose velocity vectors on
the head contour plots. The program requires TECSOFTs TRANSLATE
program. A special 80386/486 version is available. Contact:
Scientific Software Group
P.O. Box 23041
Washington, D.C. 20026-3041
phone 703/620-9214.
MODPATH: a particle-tracking program developed by the USGS for use with
the MODFLOW model. MODPATH calculates the path a particle would take
in a steady-state three-dimensional flow field in a given amount of time. It
operates as a post-processor for MODFLOW using heads, cell-by-cell flow
terms and porosity to move each particle through the flow field. The
program handles both forward and backward particle tracking. (See also
IGWMC Key 3984).
MODPATH-PLOT: a graphic display program for use with MODPATH-PC It
uses the Graphical Kernal System (GKS) to produce graphical output on a
wide range of commonly used printers and plotters. MODPATH-PLOT comes
with MODPATH.
Page 4b-8
-------
Section 4i Model Descriptions Model Description tor MODFLOW
MACMODFLOW: this is the Macintosh implementation of the MODFLOW
model. It supports the standard Macintosh user-interface. The simulation
code is integrated with the input data editor and the graphic post-processor.
Extensive data-checking is employed and simulation stops are trapped with
control returning to the program. Graphic functions include contouring of
heads, drawdowns, hydraulic conductivity, transmissivity, and plots of the
finite difference grid. This version of MODFLOW is available from Scientific
Software Group.
PATH3D is a general particle tracking program for calculating ground-water
paths and travel times. The program uses the head solution of the USGS
modular finite difference model MODFLOW. The program is available from:
S.S. Papadopulos and Assoc., Inc
12250 Rockville Pike
Suite 290
Rockville , Maryland 20852
MODELCAD is a graphical oriented, model-independent pre- processor to
prepare and edit input files for two- and three-dimensional ground-water
models, including aquifer properties, boundary conditions, and grid
dimensions. The program prepares input files for MODFLOW, MOC, PLASM
and RANDOM WALK, among others. File formatting routines for other
models are available upon request. Contact:
Geraghty & Miller Modeling Group
1895 Preston White Drive
Suite 301
Reston, VA 22091
Phone: (703) 476-0335.
A related program contains a preconditioned conjugate gradient method for
the solution of the finite difference approximating equations generated by
MODFLOW (Kuiper 1987; see references). Five preconditioning types may be
chosen: three different types of incomplete Choleski, point Jacob! or block
Jacobi. Either a head change or residual error criteria may be used as an
indication of solution accuracy and interation termination. A later version of
this solver is included in the extended memory PC version from IGWMC,
Scientific Software and Geraghty & Miller Modeling Group.
A computer program to summarize the data input and output from
MODFLOW is described by Scott (1990; see references). This program, the
Modular Model Statistical Processor, provides capabilities to easily read data
input to and output from MODFLOW, calculate descriptive statistics, generate
histograms, perform logical tests using relational operators, calculate data
arrays using arithmic operators, and calculate flow vectors for use in a
graphical display program. The program is written in Fortran 77 and tested on
a Prime 1/model 9955-11.
-------
Section 4i Model Description* Mode/ Description for MODFLOiy
A computer program for simulating aquifer-system compaction resulting
from ground-water storage changes in compressible beds has been published
by Leake and Prudic (1988; see references). This program can be incorporated
in MODFLOW as the INTERBED-STORAGE-PACKAGE. (see also MARS
3985).
STR1 is a computer program written for use in MODFLOW to account for the
amount of flow in streams and to simulate stream-aquifer interaction. The
program is known as the Streamflow Routing Package (Prudic 1989; see
references.) (see also MARS 3896).
PCG2 (Hill 1990; see references) is a numerical code to be used with the USGS
MODFLOW model. It uses the preconditioned conjugate-gradient method to
solve the equations produced by the MODFLOW model. Both linear and
nonlinear flow conditions may be simulated. PCG2 includes two
preconditioning options: modified Cholesky preconditioning and polynomial
preconditioning. Convergence of the solver is determined using both head-
change and residual criteria. Non linear problems are solved using Picard
iterations. This solver is included in various extended memory PC versions.
An IBM PC/386 extended memory version of this model is also available
from:
Geraghty & Miller, Inc.
Modeling Group
1895 Preston Drive
Suite 301
Reston, VA 22091
tel.: 703/476-0335
fax: 703/476-6372
Page 4b-10
-------
Section 4s Model Descriptions Model Description tor MOOFLOW
11. References
McDonald, M.G. and A.W. Harbaugh. 1983. A Modular Three-
Dimensional Finite-Difference Ground Water Model. Open-File Report
83-875, U.S. Geological Survey, Reston, Virginia.
Kuiper, L.K. 1987. Computer Program for Solving Ground-Water Flow
Equations by the Preconditioned Conjugate Gradient Method. Water-
Resources Investigations Report 87-4091, U.S. Geological Survey,
Austin, Texas.
Leake, S.A., and D.E. Prudic. 1988. Documentation of a Computer Program
to Simulate Aquifer-System Compaction using the Modular Finite-
Difference Ground-Water How Model. Open-File Report 88-482, U.S.
Geological Survey, Tuscon, Arizona.
McDonald, M.G., and A.W. Harbaugh. 1988. A Modular Three-
Dimensional Finite-Difference Ground-Water Flow Model. Book 6,
Modeling Techniques, Chapter Al, U.S. Geological Survey, Reston,
Virginia.
Prudic, D.E. 1989. Documentation of a Computer Program to Simulate
Stream-Aquifer Relations using a Modular, Finite-Difference, Ground-
Water Row Model. Open-File Report 88-729, U.S. Geological Survey,
Carson City, Nevada.
Hill, M.C. 1990. Preconditioned Conjugate Gradient 2 (PCG2), A Computer
Program for Solving Ground-Water Flow Equations. Water Resources
Investigations Report 90-4048, U.S. Geological Survey, Denver,
Colorado.
Scott, J.C. 1990. A Statistical Processor for Analyzing Simulations Made
Using the Modular Finite-Difference Ground-Water Flow Model.
Water-Resources Investigations Report 89-4159, U.S. Geological
Survey, Oklahoma City, Oklahoma.
Pmge 4b-11
-------
Section 4i Model Descriptions Mod*l Description for liODFLOW
12. Users
Brahana, J.V., and T.O Mesko. Hydrogeology and Preliminary Assessment
of Regional Flow in the Upper Cretaceous and Adjacent Aquifers in the
Northern Mississippi Embayment. Water-Resources Investig. Rept. 87-
4000, U.S. Geolog. Survey, Nashville, Tennessee.
McDonald, M.G., 1984. Development of a Multi-Aquifer Well Option for a
Modular Ground Water Flow Model. In: Proceedings Practical
Applications of Ground Water Models, Columbus, Ohio, August 15-17,
1984, pp. 786- 796. National Water Well Association, Dublin, Ohio.
Bailey, Z.C., 1985. Hydrologic Effects of Ground and Surface Water
Withdrawals in the Howe Area, LaGrange County, Indiana. Water
Resources Investig. Report 85-4163, U.S. Geological Survey,
Indianapolis, Indiana.
Grannemann, N.G. and F.R. Twenter, 1985. Geohydrology and Ground
Water Flow at Verona Well Fields, Battle Creek, Michigan. Water
Resources Investig. Report 85-4056, U.S. Geological Survey, East
Lansing, Michigan.
Kuniansky, E.L., 1985. Complex Depositional Environment at Baton
Rouge, Louisiana, Requires Innovative Model Layering. In:
Proceedings Practical Applications of Ground Water Models,
Columbus, Ohio, pp. 122-143. National Water Well Association,
Dublin, Ohio.
Lindgren, H.A., 1985. Hydrologic Effects of Ground and Surface Water
Withdrawals in the Milford Area, Elkart and Kosciusko Counties,
Indiana. Water Resources Investig. Report 85-4166, U.S. Geological
Survey, Indianapolis, Indiana.
Spinazola, J.M., J.B. Gillespie and R.J. Hart, 1985. Ground Water Flow and
Solute Transport in the Equus Beds Area, South-Central Kansas, 1940-
79. Water Resources Investig. Report 85-4336, U.S. Geological Survey,
Lawrence, Kansas.
Trudeau, D.A. and A. Buono, 1985. Projected Effects of Proposed Increased
Pumpage on Water Levels and Salinity in the Sparta Aquifer Near
West Monroe, Louisiana. Technical Report No. 39, Louisiana Dept. of
Transportation and Development, Baton Rouge, Louisiana.
Weatherington - Rice, J.P., 1985. A Three-Dimensional Ground Water
Modeling Study for Development of the New Well Field at London
Correctional Institute in Madison County, Ohio. In: Proceedings
Practical Applications of Ground Water Models, Columbus, Ohio, pp.
197-211. Nat. Water Well Association, Dublin, Ohio.
Page 4b-12
-------
Sect/on 4: Model Descriptions Model Description for MODFLOW
Yager, R.M. 1985. Simulation of Ground Water Flow Near the Nuclear
Fuel Reprocessing Facility at the Western New York Nuclear Service
Center, Cattaraugus County, New York. Water Resources Investig.
Report 85-4308, U.S. Geological Survey, Albany, New York.
Delin, G.N. 1986. Hydrogeology of Confined-Drift Aquifers near the
Pomme De Terre and Chippewa Rivers, Western Minnesota. Water
Resources Investig. Report 86-4098, U.S. Geological Survey,
Minneapolis, Minnesota.
Mandle, R.J. and A.L. Kontis, 1986. Directions and Rates of Ground Water
Movement in the Vicinity of Kesterson Reservoir, San Joaquin Valley,
California. Water Resources Investig. Report 86-4196, U.S. Geological
Survey, Sacramento, California.
Martin, P. and C. Berenbrock, 1986. Ground Water Monito ring at Santa
Barbara, California: Phase 3 Development of a Three-Dimensional
Digital Ground Water Flow Model for Storage Unit I of the Santa
Barbara Ground- Water Basin. Water-Resources Investig. Report 86-
4103, U.S. Geological Survey, Sacramento, California.
Tucci, P., 1986. Ground Water Flow in Melton Valley, Oak Ridge
Reservation, Roane County, Tennessee - Preliminary Model Analysis.
Water Resources Investig. Report 85-4221, U.S. Geological Survey,
Nashville, Tenn.
Zitta, V.L. and T.K. Pang, 1986. Application of a Groundwater Model to
Critical Areas in Northeast Mississippi. Department of Civil
Engineering, Water Resources Research Institute, Mississippi State
University, Mississippi State, Mississippi.
Bergeron, M.P. 1987. Effect of Reduced Industrial Pumpage on the
Migration of Dissolved Nitrogen in an Outwash Aquifer at Olean,
Cattaraugus County, New York. Water-Resources Investig. Report 85-
4082, U.S. Geological Survey, Ithaca, New York.
Kernodle, J.M. R.S. Miller and W.B. Scott. 1987. Three- Dimensional
Model Simulation of Transient Ground Water Flow in the
Albuquerque-Belen Basin, New Mexico. Water Resources Investig.
Report 86-4194, U.S. Geological Survey, Albuquerque, New Mexico.
Kidd, R.E., and W.S. Mooty. 1987. Simulation of the Flow System in the
Shallow Aquifer, Dauphin Island, Alabama. In: E.J. Edwards (ed.),
Proceed. 17th Mississippi Water Resources Conference, Jackson,
Mississippi, March 25-26, 1987, pp. 99-107. Water Resources Research
Inst, Mississippi State Univ., Mississippi State, Mississippi.
Kyllonen, D.P., and K.D. Peter. 1987. Geohydrology and Water Quality of
the Inyan Kara, Minnelusa, and Madison Aquifers of the Northern
Black-Hills, South Dakota and Wyoming, and Bear Lodge Mountains,
Wyoming. Water-Resources Investigations Report 86-4158, U.S.
Geological Survey, Rapid City, South Dakota.
Page 4b-13
-------
Section 4i Model Descriptions Model Description for MODFLOW
Phillips, S.W. 1987. Hydrogeology, Degradation of Ground- Water Quality,
and Simulation of Infiltration from the Delaware River into the
Potomac Aquifers, Northern Delaware. Water-Resources Investig.
Rept. 87-4185, U.S. Geol. Survey, Towson, Maryland.
Pool, D.R. 1987. Hydrogeology of McMullen Valley, West- Central
Arizona. Water-Investigations Report 87-4140, U.S. Geological Survey,
Tuscon, Arizona.
Risser, D.W. 1987. Possible Changes in Ground-Water Flow to the Pecos
River Caused by Santa Rosa Lake, Guadalupe County, New Mexico.
Water-Resources Investigations Report 85-4291, U.S. Geological
Survey, Albuquerque, New Mexico.
Weiss, E. 1987. Ground-Water Flow in the Navajo Sandstone in Parts of
Emery, Grand, Carbon, Wayne, Garfield, and Kane Counties, Southeast
Utah. Water-Resources Investigations Report 86-4012, U.S. Geological
Survey, Denver, Colorado.
Zitta, V.L., and T-K. Pang. 1987. Application of a Layered Groundwater
Model to Critical Areas in Northeast Mississippi. In: E.J. Hawkins (ed.),
Proceedings 17th Mississippi Water Resources Conference, Jackson,
Mississippi, March 25-26,1987, pp, 89-94. Water Resources Research
Inst., Mississippi State Univ., Mississippi State, Mississippi.
Aucott, W.R. 1988. The Predevelopment Ground-Water Flow System and
Hydrologic Characteristics of the Coastal Plain Aquifers of South
Carolina. Water-Resources Investigations Report 86-4347, U.S.
Geological Survey, Columbia, South Carolina.
Bailey, Z.C. 1988. Preliminary Evaluation of Ground- Water Flow in Bear
Creek Valley, The Oak Ridge Reservation, Tennessee. Water-
Resources Investigations Report 88-4010, U.S. Geological Survey,
Nashville, Tennessee.
Brown, J.G., and J.H. Eychaner. 1988. Simulation of Five Ground-Water
Withdrawal Projections for the Black Mesa Area, Navajo and Hopi
Indian Reservations, Arizona. Water-Resources Investigations Report
88-4000, U.G. Geological Survey, Tuscon, Arizona.
Buckles, D.R., and K.R. Watts. 1988. Geohydrology, Water Quality, and
Preliminary Simulations of Ground-Water Flow of the Alluvial
Aquifer in the Upper Black Squirrel Creek Basin, El Paso County,
Colorado. Water-Resources Investigations Report 88-4017, U.S.
Geological Survey, Denver, Colorado.
Danskin, W.R. 1988. Preliminary Evaluation of the Hydrogeologic System
in Owens Valley, California. Water- Resources Investigations Report
88-4003, U.S. Geological Survey, Sacramento, California.
Page 4b-14
-------
Section 4: Model Descriptions Model Description for MODFLOW
Davies-Smith, A., E.L. Bolke, and C.A. Collins. 1988. Geohydrology and
Digital Simulation of the Ground- Water Row System in the Umatilla
Plateau and Horse Heaven Hills Area, Oregon and Washington.
Water- Resources Investigations Report 87-4268, U.S. Geological
Survey, Portland, Oregon.
Emmons, P.J. 1988. A Digital Simulation of the Glacial- Aquifer System in
Sanborn and Parts of Beadle, Miner, Hanson, Davison, and Jerauld
Counties, South Dakota. Water-Resources Investigations Report 87-
4082, U.S. Geological Survey, Huron, South Dakota.
Fenemor, A.D. 1988. A Three-Dimensional Model for Management of the
Waimea Plains Aquifers, Nelson. Publication 18, Hydrology Centre,
Dept. of Scientific and Industrial Research, Christchurch, New Zealand.
Fretwell, J.D. 1988. Water Resources and Effects of Ground-Water
Development in Pasco County, Florida. Water-Resources
Investigations Report 87-4188, U.S. Geological Survey, Tallahassee,
Florida.
Hamilton,F.A., and J.D. Larson. 1988. Hydrogeology and Analysis of the
Ground-Water Flow System in the Coastal Plain of Southeastern
Virginia. Water- Resources Investigations Report 87-4240, U.S.
Geological Survey, Richmond, Virginia.
Hansen, D.S. 1988. Appraisal of the Water Resources of the Big Sioux
Aquifer, Moody County, South Dakota. Water-Resources
Investigations Report 87-4057, U.S. Geological Survey, Huron, South
Dakota.
Johnson, M.J., C.J. Londquist, J. Laudon, and H.T. Mitten. 1988.
Geohydrology and Mathematical Simulation of the Pajaro Valley
Aquifer System, Santa Cruz and Monterey Counties, California.
Water-Resources Investigations Report 87-4281, U.S. Geological
Survey, Sacramento, California.
Laczniak, R.J., and A.A. Meng HI. 1988. Ground-Water Resources of the
York-James Peninsula of Virginia. Water-Resources Investig. Rept. 88-
4059, U.S. Geol. Survey, Richmond, Virginia.
McAda, D.P., and M. Wasiolek. 1988. Simulation of the Regional
Geohydrology of the Tesque Aquifer System near Santa Fe, New
Mexico. Water-Resources Investigations Report 87-4056, U.S.
Geological Survey, Albuquerque, New Mexico.
Morgan, D.S. 1988. Geohydrology and Numerical Model Analysis of
Ground-Water Flow in the Goose Lake Basin, Oregon and California.
Water-Resources Investigations Report 87-4058, U.S. Geological
Survey, Portland, Oregon.
Page 4b-15
-------
Section 4s Model Descriptions Model Description for MODFLOW
Yates, E.B. and J.H. Wiese. 1988. Hydrogeology and Water Resources of
the Los Osos Valley Ground-Water Basin, San Luis Obispo County,
California. Water-Resources Investigations Report 88-4081, U.S.
Geological Survey, Sacramento, California.
Ackerman, D.J. 1989. Hydrology of the Mississippi River Valley Alluvial
Aquifer, South-Central United States --A Preliminary Assessment of
the Regional Flow System. Water-Resources Investig. Rept. 88-4028,
U.S. Geol. Survey, Little Rock, Arkansas.
Coen HI, A.W. 1989. Ground-Water Resources of Williams County, Ohio,
1984-86. Water-Resources Investigations Report 89-4020, U.S.
Geological Survey, Columbus, Ohio.
Duwelius, R.F., and T.K. Greeman. 1989. Geohydrology, Simulation of
Ground-Water Flow, and Ground-Water Quality at Two Landfills,
Marion County, Indiana. Water-Resources Investigations Report 89-
4100, U.S. Geological Survey, Indianapolis, Indiana.
Freckleton, J.R. 1989. Geohydrology of the Foothill Ground-water Basin
near Santa Barbara, California. Water-Resources Investig. Report 89-
4017, U.S. Geological Survey, Sacramento, California.
Imes, J.L. 1989. Analysis of the Effect of Pumping on Ground-Water Flow
in the Springfiled Plateau and Ozark Aquifers near Springfield,
Missouri. Water-Resources Investigations Report 89-4079, U.S.
Geological Survey, Rolla, Missouri.
A model of 42 by 42 mile area was calibrated under predevelopment
and transient pumping conditions. The model then was used to
quantify hydraulic properties of the aquifer and confining unit and the
hydrologic budget.
Lima, V. de, and J.C. Olimpio. 1989. Hydrogeology and Simulation of
Ground-Water Flow at Superfund-Site Wells G and H, Woburn,
Massachusetts. Water-Resources Investigations Report 89-4059,
U.S.Geological Survey, Boston, Massachusetts.
(The calibrated three-dimensional ground-water flow model of the
stratified drift area was designed as a tool for predicting the steady-state
and transient effects of pumpage in the vicinity of public supply wells.)
Martin, Jr., A., and C.D. Whiteman, Jr. 1989. Geohydrology and Regional
Ground-Water Flow of the Coastal Lowlands Aquifer System in Parts
of Louisiana, Mississippi, Alabama, and FloridaA Preliminary
Analysis. Water-Resources Investigations Report 88-4100, U.S.
Geological Survey, Baton Rouge, Louisiana.
Page 4b-16
-------
Sect/on 4s Model Descriptions Model Description for MODFLOW
Summer, D.M., B.E. Wasson, and SJ. Kalkhoff. 1989. Geohydrology and
Simulated Effects of Withdrawals on the Miocene Aquifer System in
the Mississippi Gulf Coast. Water-Resources Investig. Rept. 87-4172,
U.S. Geological Survey, Jackson, Mississippi.
Watson, L.R., R.J. Shedlock, K.J. Banaszak, L.D. Arihood, and P.K. Doss.
1989. Preliminary Analysis of the Shallow Ground-Water System in
the Vicinity of the Grand Calumet River/Indiana Harbor Canal, North-
western Indiana. Open-File Report 88-492, U.S. Geological Survey,
Indianapolis, Indiana.
Yobbi, D.K. 1989. Simulation of Steady-State Ground Water and Spring
Flow in the Upper Floridan Aquifer of Coastal Citrus and Hernando
Counties, Florida. Water-Resources Investig. Rept., U.S. Geol. Survey,
Tallahassee, Florida.
Young, H.W., and G.D. Newton. 1989. Hydrology of the Oakley Fan Area,
South Central Idaho. Water- Resources Investigations Rept. 88-4065,
U.S. Geolog. Survey, Boise, Idaho.
Bair, E.S., R.A. Sheets, and S.M. Eberts. 1990. Particle-Tracking Analysis of
Flow Paths and Traveltimes from Hypothetical Spill Sites within the
Capture Area of a Wellfield. Ground Water, Vol. 28 (6), pp. 884-892.
Velocity fields computed from simulated flows from a steadystate,
three-dimensional numerical model of a glacial drift/carbonate-
bedrock aquifer system were used in conjunction with a particle
tracking program to delineate traveltimes related capture areas of a
municipal wellfield and compute flow paths from a hypothetical spill
site.
Delin, G.N. 1990. Geohydrology and Water Quality of Confined-Drift
Aquifers in the Brooten-Belgrade Area, West-Central Minnesota.
Water-Resources Investigations Report 88-4124, U.S. Geological
Survey, St. Paul, Minnesota.
(The quasi-three dimensional model was constructed to improve
understanding of the movement of water in a complicated system of
confined drift aquifers.)
Dickerman, D.C., E.G. Todd Trench, and J.P. Russell. 1990. Hydrogeology,
Water Quality, and Ground-Water Development Alternatives in the
Lower Wood River Ground-Water Reservoir, Rhode Island. Water-
Resources Investigations Report 89-4031, U.S. Geological Survey,
Providence, Rhode Island.
Eimers, J.L., W.L. Lyke, and A.R. Brockman. 1990. Simulation of Ground-
Water Flow in Aquifers in Cretaceous Rocks in the Central Coastal
Plain, North Carolina. Water-Resources Investigations Report 89-4153,
U.S. Geological Survey, Raleigh, North Carolina.
Page 4b-17
-------
Section 4s Model Descriptions Model Description for UODFLOW
Lum II, W.E., J.L. Smoot, and D.R. Ralston. 1990. Geohydrology and
Numerical Model Analysis of Ground-Water Flow in the Pullman-
Moscow Area, Washington and Idaho. Water-Resources
Investigations Report 89-4103, U.S. Geological Survey, Tacoma,
Washington.
Mahon, G.L. and A.H. Ludwig. 1990. Simulation of Ground- Water Flow
in the Mississippi River Valley Alluvial Aquifer in Eastern Arkansas.
Water-Resources Investigations Report 89-4145, U.S. Geological
Survey, Little Rock, Arkansas.
Ohland, G.L. 1990. Appraisal of the Water Resources of the Skunk Creek
Aquifer in Minnehaha County, South Dakota. Water-Resources
Investigations Report 87- 4156, U.S. Geological Survey, Huron, South
Dakota.
Tucci, P., D.W. Hanchar, and R.W. Lee. 1990. Hydrogeology of a
Hazardous-Waste Disposal Site near Brentwood, Williamson County,
Tennessee. Water-Resources Investigations Report 89-4144, U.S.
Geological Survey, Nashville, Tennessee.
Zomorodi, K. 1990. Experiences in using MODFLOW on a PC. In: E.B.
Janes and W.R. Hotchkiss (eds.), Transferring Models to Users, Denver,
Colorado, November 4-8, 1990, pp. 351-356. American Water Resources
Assoc., Bethesda, Maryland.
Page 4b-18
-------
Model Description For
PLASM
SOURCE:
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Table of Contents
PAGE
1. Mode/ Identification 4c-1
1.1. Model Name(s) 4c-l
1.2. Date of First Release 4c-l
1.3. Current Release Date 4c-l
1.4. Authors 4c-l
2. Model Information 4c-1
2.1. Model Category 4c-l
2.2. Model Developed For 4c-l
2.3. Units of Measurement Used 4c-l
2.4. Abstract 4c-l
2.5. Data Input Requirements 4c-2
2.6. Versions Exist for the Following Computer Systems 4c-2
2.7. System Requirements 4c-2
2.8. Graphics Requirements 4c-2
2.9. Program Information 4c-2
3. General Model Capabilities... ................................... 4c-2
3.1. Parameter Discretization 4c-2
3.2. Spatial Orientation 4c-2
3.3. Types of Possible Updates 4c-2
3.4. Geostatistics and Stochastic Approach 4c-2
3.5. Comments 4c-3
4. Flow Characteristics........................ 4c-3
4.1. Row System Characterization 4c-3
4.2. Fluid Conditions 4c-4
4.3. Boundary Conditions 4c-4
4.4. Solution Methods for Flow 4c-4
45. Grid Design 4c-4
4.6. Flow Output Characteristics 4c-5
5. Evaluation[[[ 4c-5
5.1. Verification/Validation 4c-5
5.2. Internal Code Documentation (Comment Statements) 4c-5
5.3. Peer (Independent) Review 4c-5
6. Documentation and Support.............................................4c-5
6.1. Documentation Includes 4c-5
6.2. Support Needs 4c-5
6.3. Level of Support 4c-6
7. Availability. . 4c-6
7.1. Terms 4c-6
7.2. Form 4c-6
-------
PACE
9. Institution of Model Development... ...4c-7
9.1. Name 4c-7
9.2. Address 4c-7
9.3. Type of Institution 4c-7
10. Remarks 4c-7
11. References... 4c-9
12. Users 4c-10
Page 4c-»
-------
PLASM
1. Model Identification
1.1. Model Name(s)
PLASM
1.2. Date of First Release
1971
1.3. Current Release Date
1985
1.4. Authors
1. Prickett, T.A.
2. Lonnquist, C.G.
2. Model Information
2.1. Model Category
ground-water flow
2.2. Model Developed For
general use (e.g. in field applications)
research (e.g. hypothesis/theory testing)
demonstration/education
2.3. Units of Measurement Used
SI system
metric units
US customary units
any consistent system
2.4. Abstract
PLASM (Prickett Lonnquist Aquifer Simulation Model) is a finite
difference model for simulation of transient, two-dimensional or
quasi-three-dimensional flow in a single or multi-layered,
heterogeneous, anisotropic aquifer system. The original model of 1971
consisted of a series of separate programs for various combinations of
simulation options. Later versions combined most of the options in a
single code, including variable pumping rates, leaky confined aquifer
conditions, induced infiltration from a shallow aquifer or a stream,
storage coefficient conversion between confined and watertable
conditions, and evapotranspiration as a function of depth to water
table. The model uses the iterative alternating implicit method (IADI)
to solve the matrix equation.
Pmg* 4c-1
-------
Section 4s Model Descriptions Model Description for PLASM
2.5. Data Input Requirements:
Data input requirements are provided in the model documentation
and are discussed in the introduction to Section 4.0 of the Compendium.
2.6. Versions Exist for the Following Computer Systems
minicomputer
workstations
mainframe
microcomputer
make/model
- IBM 360,370
- DEC VAX
- IBM PC/XT/AT
2.7. System Requirements
core memory (RAM) for execution (bytes)
- 640K (for IBM PC version)
mass storage (disk space in bytes)
- 1M
numeric/math coprocessor
(for micro computers)
compiler required
- (for mainframe)
2.8. Graphics Requirements
none
2.9. Program Information
programming language/level
Fortran IV
3. Genera/ Model Capabilities
3.1. Parameter Discretization
distributed
3.2. Spatial Orientation
saturated flow
2D-horizontal
3D-layered
3.3. Types of Possible Updates
parameter values
boundary conditions
3.4. Geostatistics and Stochastic Approach
none
Page 4c-2
-------
Section 4i Model Descriptions Model Description for PLASM
3.5. Comments
This model has restart capability.
4. Flow Characteristics
4.1. Flow System Characterization
Saturated Zone
System
single aquifer
single aquifer/aquitard system
multiple aquifer/aquitard systems
Aquifer Type(s)
confined
semi-confined (leaky-confined)
- unconfined (phreatic)
Medium
porous media
Parameter representation
homogeneous
heterogeneous
isotropic
anisotropic
Flow characteristics (saturated zone)
laminar flow
linear (Darcian flow)
- steady-state
- transient
Flow processes included
areal recharge
induced recharge (from river)
evapotranspiration
Changing aquifer conditions
in space
- variable thickness
- confined/unconfmed
- in time
- confined/unconfined
Well characteristics
none
Pag* 4c-3
-------
Sect/on 4: Model Descriptions Model Description tor PLASM
4.2. Fluid Conditions
Single Fluid Flow
water
Flow of Multiple Fluids
- N.A.
Fluid Properties
constant in time/space
4.3. Boundary Conditions
First Type - Dirichlet
head/pressure
Second Type - Neumann (Prescribed Rux)
- injection/production wells
no flow boundary
areal boundary flux
ground-water recharge
induced infiltration
Third Type - Cauchy
head/pressure-dependent flux
4.4. Solution Methods for Row
General Method
Numerical
Spatial Approximation
finite difference method
node-centered
Time-Stepping Scheme
fully implicit
Matrix-Solving Technique
Iterative
- iterative ADIP
- Direct
Gauss elimination
4.5. Grid Design
Cell/Element Characteristic
- constant cell size
variable cell size
Possible Cell Shapes
2D-square
ID-rectangular
Page 4c-4
-------
Section 4: Model Descriptions Model Description for PLASM
Maximum Number of Nodes
- 5000
4.6. Flow Output Characteristics
Simulation Results
- head/pressure/potential
- ASCII file (areal values)
- ASCII file (hydrograph)
water budget components
- ASCH file (cell-by-cell values)
- ASCII file (global total area)
5. Evaluation
5.1. Verification/Validation
verification (analytical solutions)
field datasets (validation)
synthetic datasets
code intercomparison
5.2. Internal Code Documentation (Comment Statements)
incidental
5.3. Peer (Independent) Review
concepts
theory (math)
coding
accuracy
documentation
usability
efficiency
6. Documentation and Support
6.1. Documentation Includes
model theory
user's instructions
example problems
code listing
program structure and development
verification/validation
6.2. Support Needs
Can be used without support
Support is available
from author
- from third parties
Page 4c-5
-------
Sect/on 4: Model Descriptions Model Description for PLASM
6.3. Level of Support
limited
7. Availability
7.1. Terms
available
public domain
proprietary
restricted public domain
purchase
license
Form
source code only (tape/disk)
source and compiled code
compiled code only
paper listing of source code
(depends on version)
8. Pre and Post Processors
8.1. Data Preprocessing
name: PREPLASM
part of model package (e.g. under a shell; dedicated)
textual data entry/editing
error-checking
name: MODELCAD
separate (optional) program
- graphic data entry/modification (e.g manual grid design,
arrays)
- (semi-) automatic grid generation
- data reformatting (e.g. for GIS)
- error-checking
help screens
Page 4c-6
-------
Section 4: Model Descriptions Model Description for PLASM
9. Institution of Model Development
9.1. Name
Illinois State Water Survey
9.2. Address
P.O. Box 232
Urbana, Illinois 61801
9.3. Type of Institution
state/provincial government
fO. Remarks
A modified version of PLASM to analyze hydrologic impacts of mining is
documented in in a report of the U.S. Office of Surface Mining (1981; see user
references).
An extended and updated single aquifer version of PLASM for the IBM-PC is
available from:
T.A. Prickett
6 G.H. Baker Drive
Urbana, IL 61801
phone: 217/384-0615.
This version is the same as published in Bulletin 55, except for the multi-
layered option and the confined/unconfined storage conversion.
The IGWMC distributes a mainframe version of PLASM running on the
DEC/VAX 11 series. The Center distributes also an IBM-PC version. This
latter version comes as a program package including separate codes for
confined (CONPLASM) and watertable (UNCPLASM) conditions and a
textual preprocessor for input data preparation (PREPLASM). Contact:
IGWMC USA: International Ground Water Modeling
Center,
Colorado School of Mines, Golden, Colorado,
USA, or
TNO Inst. of Applied Geoscience, P.O. Box
6012, 2600JA
Delft, The Netherlands.
IGWMC Europe: TNO Institute of Applied Geoscience,
P.O. Box 6012, 2600JA Delft,
The Netherlands.
For earlier microcomputer versions of PLASM see IGWMC Key # 6010 and
6011 (MARS historic data base).
Pag* 4e-7
-------
Section 4: Model Descriptions Model Description for PLASM
MODELCAD is a graphical oriented, model-independent pre-processor to
prepare and edit input files for two- and three-dimensional ground-water
models, including aquifer properties, boundary conditions, and grid
dimensions. The program prepares input files for MODFLOW, MOC, PLASM
and RANDOM WALK, among others. File formatting routines for other
models are available upon request. Contact:
Geraghty & Miller Modeling Group
1895 Preston White Drive
Suite 301
Reston, VA 22091
Phone: (703) 476-0335.
The IBM-PC version of this program is also available from:
National Water Well Association
Ground Water Bookstore
P.O. Box 182039
Dept. 017
Columbus, Ohio 43218
Tel. (614) 761-1711
The version available from the National Water Well Association runs on
IBM PC or compatible and has been prepared by Koch and Associates and
includes interactive data preparation. The code is directly available from
Koch and Associates. Contact:
Ann Koch
2921 Greenway Drive
Ellicot City, Maryland 21043
tel. 301/461-6869.
Page 4c-8
-------
Section 4: Model Descriptions Model Description for PLASM
11. References
Prickett, T.A. and C.G. Lonnquist. 1971. Selected Digital Computer
Techniques for Groundwater Resource Evaluation. Bulletin 55.
Illinois State Water Survey, Urbana, 111.
Texas Water Development Board, Systems Engineering Division. 1974.
GWSIM - Groundwater Simulation Program, Program Documentation
and User's Manual. Texas Dept. of Water Resourc, Austin, Texas.
Prickett, T.A. and C.G. Lonnquist. 1976. Methods de Ordenador para
Evaluacion de Recursos Hidraulicos Subterraneas. Boletin 41.
(Spanish version of Bulletin 55, ISWS). Ministerio de Obras Publicas,
Direction General de Obras Hidraulicos, Madrid, Spain.
Texas Dept. of Water Resources. 1979. Program Description of GWSIM.
In: Mathematical Simulation Capabilities in Water Resources Systems
Analysis, Eng. and Environm. Systems Section, Report LP-16, revised
November 1979.
Institute Geologico Y Minero de Espana. 1981. Modelos Multicapa
Tomo II: Listados de Programas, Ministreio de Industria Y Energia,
Madrid, Spain.
Institute Geologico Y Minero de Espana. 1981. Modelos Multicapa
Tomo I: Manuales de Utilization. Ministerio de Industria Y Energia,
Madrid, Spain.
Institute Geologico Y Minero de Espana. 1982. Modelos Monocapa en
Regimen Transitorio Tomo II: Listados de Ordenador. Ministerio de
Industria Y Energia, Madrid, Spain.
Institute Geologico Y Minero de Espana. 1982. Modelos Monocapa en
Regimen Transitorio Tomo I: Manuales de Utilization. Dirretion de
Aguas Subterraneas y Geotecnia, Ministerio de Industria y Energia,
Madrid, Spain.
4c-9
-------
Sect/on 4s Model Descriptions Model Description for PLASM
12. Users
T.R. Knowles developed two models based on PLASM for the Texas Dept.
of Water Resources, Austin, Texas. See models GWS1M (IGWMC key
# 0681) and GWSIM II (IGWMC key # 0680).
Beltran, J.M.A., L.L. Garcia, and M.R.L. Madurga. 1973. Estudio de la
Explotacion de los Aquiferos del Llano de Palma de Mallorca Mediante
un Modelo Digital Simplificado. Boletin 38, Dirrecion General de
Obras Hidraulicas, Ministerio de Obras Publicas, Madrid, Spain.
Karanjac, J., M. Altunkaynak, and G. Ovul. 1976. Mathematical model of
Elazig-Uluova Plain, Republic of Turkey. Ministry of Energy and
Natural Resources. Geotechnical Services and Groundwater Division,
Ankara, Turkey.
Naymik, T.G. 1979. The Application of Digital Model for Evaluating the
Bedrock Water Resources of the Maumee River Basin, Northwestern
Ohio. Lawrence Livermore Laboratory. University of California.
Livermore, California.
U.S. Department of the Interior. 1981. Office of Surface Mining -
Reclamation and Enforcement. Ground Water Model Handbook (H-
D3004-021-81-1062D). Office of Surface Mining. Western Technical
Service Center. Denver, Colorado.
Yazicigil, H. and L.V.A. Sendlein. 1981. Management of Ground Water
Contaminated by Aromatic Hydrocarbons in the Aquifer Supplying
Ames, Iowa. Groundwater, Vol. 19(6), pp. 648-665.
Grosser, P.W. 1984. Application of Ground Water Models to the
Identification of Contaminant Sources. In: Proceed. Conf. on Practical
Applications of Ground Water Models, Columbus, Ohio, pp. 452-461.
Nat. Water Well Assoc., Dublin, Ohio.
Khaleel, R. and D.M. Peterson. 1984. A Quasi Three- Dimensional
Ground Water Flow Model for the Mesilla Bolson, New Mexico. In:
Proceed. Conf. on Practical Applications of Ground Water Models,
Columbus, Ohio, pp. 52-67. Nat. Water Well Assoc., Dublin, Ohio.
Martin, P.L. 1984. The Use of a Numerical Model in Predicting the
Effectiveness of a Dewatering System. In: Proceed. Conf. on Practical
Applications of Ground Water Models, Columbus, Ohio, pp. 759-777.
Nat. Water Well Assoc., Dublin, Ohio.
Osiensky, J.L., G.V. Winter, and R.E. Williams. 1984. Monitoring and
Mathematical Modeling of Contaminated Ground Water Plumes in
Fluvial Environments. Ground Water, Vol. 22(3), pp. 298-306.
Page 4c-10
-------
Section 4i Model Descriptions Model Description tor PLASM
Potter, S.T. and W.J. Gburek. 1984. Seepage Face Simulation Using the
Illinois State Water Survey Ground Water Model. In: Proceed. Conf.
on Practical Applications of Ground Water Models, Columbus, Ohio,
pp. 674-700. Nat. Water Well Assoc., Dublin, Ohio.
Ritchey, J.D. 1984. Method for Including the Effects of River Stage and
Precipitation on Ground Water Levels in an Aquifer Model. In:
Proceed. Practical Applications of Ground Water Models, Columbus,
Ohio, pp. 649-673. Nat. Water Well Assoc., Dublin, Ohio.
Sevi, E.A., T.M. Cosgrave, and F.L. Fox. 1984. Sequential Horizontal and
Vertical Cross-Sectional Modeling of a Small Site. In: Proceed. Conf. on
Practical Applications of Ground Water Models, Columbus, Ohio, pp.
621-633. Nat. Water Well Assoc., Dublin, Ohio.
Singh, R., S.K. Sondhi, J. Singh, and R. Kumar. 1984. A Ground Water
Model for Simulating the Rise of Water Table under Irrigated
Conditions. Journ. of Hydrol., Vol. 71, pp. 165-179.
Kalinski, R.J. 1985. Investigation and Computer Simulation of Stream-
Aquifer Relationship at Elm Creek, Nebraska. In: Proceed. Practical
Applications of Ground Water Models, Columbus, Ohio, pp. 146-161.
Nat. Water Well Assoc., Dublin, Ohio.
Song, S. and A.O. Akan. 1985. Ground Water Management by Linear
Programming with Sample Applications to the Eastern Shore of
Virginia Aquifer. In: Proceed. Conf. on Practical Applications of
Ground Water Models, Columbus, Ohio, pp. 90-107. Nat. Water Well
Assoc., Dublin, Ohio.
Contractor, D.N., S.M.A. El-Didy, and A.S. Ansary. 1986. Numerical
Modeling of Groundwater Flow and Water Quality at Underground
Coal Gasification Sites. DOE/LC/11053-2151. Office of Fossil Energy,
U.S. Dept. of Energy, Laramie, Wyoming, (pp.7-50).
4c-11
-------
Section 4s Model Descriptions Model Description for PLASM
(THIS PAGE INTENTIONALLY LEFT BLANK)
Page 4c-12
-------
Model Description For
RANDOM WALK
SOURCE:
INTERNATIONAL GROUND WATER
MODELING CENTER
(IGWMC)
-------
Table of Contents
PAGE
1. Model Identification ........................... . ............................... 4d-1
1.1. Model Name(s) [[[ 4d-l
1.2. Date of first release [[[ 4d-l
1.3. Current Release date [[[ 4d-l
1.4. Authors [[[ 4d-l
2. Model Information ........................... ........................ ...... ..... 4d-1
2.1. Model Category [[[ 4d-l
2.2. Model Developed For [[[ 4d-l
2.3. Units of Measurement Used [[[ 4d-l
2.4. Abstract [[[ 4d-l
2.5. Data Input Requirements [[[ 4d-l
2.6. Versions Exist for the Following Computer Systems ................ 4d-2
2.7. System Requirements [[[ 4d-2
2.8. Graphics Requirements [[[ 4d-2
2.9. Program Information [[[ 4d-2
3. General Model Capabilities. .............................................
3.1. Parameter Discretization [[[ 4d-2
3.2. Coupling [[[ 4d-2
3.3. Spatial Orientation [[[ 4d-2
3.4. Geostatistics and Stochastic Approach .......................................... 4d-2
4, Flow Characteristics[[[ 4d-2
4.1. Flow System Characterization [[[ 4d-2
4.2. Fluid Conditions [[[ 4d-3
4.3. Boundary Conditions [[[ 4d-4
4.4 Solution Methods for Flow [[[ 4d-4
-------
PACE
8. Availability. ....................... - ................................. - ................ 4d-7
8.1. Terms [[[ 4<*-7
8.2. Form [[[ 4
-------
RANDOM WALK
1. Model Identification
1.1. Model Name(s)
RANDOM WALK
1.2. Date of first release
7/81
1.3. Current Release date
7/81
1.4. Authors
Prickett, T.A.
Naymik, T.G.
Lonnquist, C.G.
2. Model Information
2.1. Model Category
ground-water flow
mass transport
2.2. Model Developed For
general use (e.g. in field applications)
research (e.g. hypothesis/theory testing)
demonstration/education
2.3. Units of Measurement Used
metric units
US customary units
any consistent system
2.4. Abstract
RANDOM WALK/TRANS is a numerical model to simulate two-
dimensional steady or transient flow and transport problems in
heterogeneous aquifers under water table and/or confined or leaky
confined conditions. The flow is solved using a finite difference
approach and the iterative alternating direction implicit method.
The advective transport is solved with a particle-in-a-cell method,
while the dispersion is analyzed with the random walk method.
2.5. Data Input Requirements
Data input requirements are provided in the model documentation
and are discussed in the introduction to Section 4.0 of the
Compendium.
Page 4d-1
-------
Section 4; Model Descriptions Model Description for Random Walk
2.6. Versions Exist for the Following Computer Systems
minicomputer
workstations
mainframe
microcomputer
make/model
- Cyber 175
- VAX 11/780
- IBM PC/XT/AT or compatible
2.7. System Requirements
core memory (RAM) for execution (bytes)
- 640K (for IBM PC version)
mass storage (disk space in bytes)
- at least 2Mb for data files
numeric/math coprocessor
(for micro computers)
compiler required
(for mainframe)
2.8. Graphics Requirements
none
2.9. Program Information
programming language/level
Fortran rV
3. General Model Capabilities
3.1. Parameter Discretization
distributed
3.2. Coupling
none
3.3. Spatial Orientation
saturated flow
2D-horizontal
3.4. Geostatistics and Stochastic Approach
random walk
4. Flow Characteristics
4.1. Flow System Characterization
Saturated Zone
Page 4d-2
-------
Section 4t Model Descriptions Model Description for Random Walk
System
single aquifer
- single aquifer/aquitard system
Aquifer Type(s)
confined
- semi-confined (leaky-confined)
unconfined (phreatic)
Medium
porous media
Parameter representation
homogeneous
heterogeneous
isotropic
anisotropic
Flow characteristics (saturated zone)
laminar flow
linear (Darcian flow)
steady-state
transient
Flow processes included
areal recharge
- induced recharge (from river)
evapotranspiration
Changing aquifer conditions
in space
- variable thickness
- confined/unconfined
in time
- confined/unconfined
Well characteristics
none
4.2. Fluid Conditions
Single Fluid Flow
water
Flow of Multiple Fluids
- N.A.
Fluid Properties
- constant in time/space
Page 4d-3
-------
Section 4: Model Descriptions Model Description for Random Walk
4.3. Boundary Conditions
First Type - Dirichlet
head/pressure
Second Type - Neumann (Prescribed Flux)
injection/production wells
no flow boundary
areal boundary flux
ground-water recharge
springs
- induced infiltration
Third Type - Cauchy
head/pressure-dependent flux
4.4. Solution Methods for Flow
General Method
Numerical
Spatial Approximation
finite difference method
node-centered
Time-Stepping Scheme
- fully implicit
Matrix-Solving Technique
Iterative
- iterative ADIP
Direct
Gauss elimination
4.5. Grid Design
Cell/Element Characteristic
constant cell size
variable cell size
Possible Cell Shapes
- 2D-square
ID-rectangular
Maximum Number of Nodes
- 5000
Page 4d-4
-------
Section 4: Model Descriptions Model Description for Random Walk
4.6. Flow Output Characteristics
Simulation Results
- head/pressure/potential
- ASCn file (areal values)
- ASCII file (hydrograph)
- fluxes/velocities
- ASCII file (areal values)
water budget components
- ASCn file (cell-by-cell values)
- ASCII file (global total area)
5. Mass Transport Characteristics
5.1. Water Quality Constituents
any component(s)
single component
total dissolved solids (TDS)
inorganics
organics
radionuclides
5.2. Processes Included
(Conservative) Transport
- advection
dispersion
- isotropic
anisotropic
diffusion
Phase Transfers
equilibrium
isotherm
Fate
first-order radioactive decay (single mother/daughter decay)
first-order chemical decay
5.3. Boundary Conditions
First Type - Dirichlet
- Chemical processes embedded in transport equation
concentration
Pag* 4d-S
-------
Section 4: Model Descriptions Model Description for Random Walk
Second Type - Neumann (Prescribed Solute Flux)
areal boundaries
injection wells
point sources
line sources
areal sources
5.4. Solution Methods for Transport
General Method
numerical
uncoupled flow and transport equation
Spatial Approximation
finite difference method
node-centered
particle-tracking
Time-Stepping Scheme
- fully implicit
Matrix-Solving Technique
- Random Walk
direct
Gauss elimination
5.5. Output Characteristics for Transport
Simulation Results
concentration in aquifer/soil
- ASCE file (areal values)
- ASCII file (time series)
- concentration in well
- ASCH file (time series)
- velocities (from given heads)
- ASOI file (areal values)
mass balance components
- ASCE file (cell-by-cell values)
- ASCH file (global total area)
6. Evaluation
6.1. Verification/Validation
verification (analytic solutions)
code intercomparison
6.2. Internal Code Documentation (Comment Statements)
incidental
Page 4d-
-------
Section 4s Model Descriptions Model Description for Random Walk
6.3. Peer (Independent) Review
concepts
theory (math)
7. Documentation and Support
7.1. Documentation Includes
model theory
user's instructions
example problems
program structure and development
code listing
verification/validation
7.2. Support Needs
Can be used without support
Support is available
from author
from third parties
7.3. Level of Support
limited
8. Availability
8.1. Terms
available
public domain
proprietary
purchase
8.2. Form
source code only (tape/disk)
source and compiled code
compiled code only
paper listing of source code
9. Pr0 and Post Processors
9.1. Data Preprocessing
name: PREWALK (IGWMC)
- separate (optional) program
- textual data entry/editing
- data reformatting
error-checking
Page 4d-7
-------
Sect/on 4: Model Descriptions Model Description for Random Walk
name: MODELCAD
separate (optional) program
generic (can be used for various models)
- textual data entry/editing
graphic data entry/modification (e.g manual grid design,
arrays)
data reformatting (e.g. for GIS)
error-checking
help screens
9.2. Data Postprocessing
name: POSTWALK (IGWMC)
separate (optional) program
textual data display on screen/printer
reformatting (e.g. to standard formats)
1O. Institution of Model Development
10.1. Name
Illinois State Water Survey
10.2. Address
P.O. Box 5050
Sta.A
Urbana, IL 61820
10.3. Type of Institution
state/provincial government
11. Remarks
Various microcomputer versions are available among others from the
International Ground Water Modeling Center:
IGWMC USA: Inst. for Ground-Water Res. and Educ.,
Colorado School of Mines,
Golden, CO 80401, USA.
IGWMC Europe: TNO Institute of Applied Geostience,
P.O. Box 6012
2600JA Delft
The Netherlands.
Page 4d-8
-------
Section 4: Model Descriptions Model Description for Random Walk
Code is available from:
Bob Sinclair
Director of Computer Service
Illinois State Water Survey
Box 5050
Station A
Champaign, IL 61820
telephone (217) 333-4952
at cost of magnetic tape, copying and postage.
Code is also available from:
Thomas A. Prickett
6 G.H. Baker Drive
Urbana, IL 61801
Phone 217/384-0615
A modified version of PLASM and RANDOM WALK to analyze hydrologic
impacts of mining is documented in U.S. Office of Surface Mining (1981; see
user references). Program codes are available through Boeing Computer
Network.
MODELCAD is a graphical oriented, model-independent pre-processor to
prepare and edit input files for two- and three-dimensional ground-water
models, including aquifer properties, boundary conditions, and grid
dimensions. The program prepares input files for MODFLOW, MOC, PLASM
and RANDOM WALK, among others. File formatting routines for other
models are available upon request. Contact:
Geraghty & Miller Modeling Group
1895 Preston White Drive
Suite 301, Reston, VA 22091
(703) 476-0335
Page 4d-9
-------
Section 4: Model Descriptions Model Description for Random Walk
12. References
Prickett, T.A. and C.G. Lonnquist. 1971. Selected Digital Computer
Techniques for Groundwater Resource Evaluation. Bulletin 55, Illinois
State Water Survey, Champaign, Illinois.
Navmik, T.G. and M.J. Barcelona. 1981. Characterization of a
Contaminant Plume in Groundwater, Meredesia, Illinois. Ground
Water, Vol. 19(5), pp. 517-526.
Prickett, T.A., T.G. Naymik and C.G. Lonnquist. 1981. A Random-Walk
Solute Transport Model for Selected Groundwater Quality Evaluations.
Bulletin 65, Illinois State Water Survey, Champaign, Illinois.
Naymik, T.G. and M.E. Sievers. 1983. Groundwater Tracer Experiment (II)
at Sand Ridge State Forest. Illinois State Water Survey Division,
Illinois Dept. of Energy and Natural Resources, Report 334, Champaign,
Illinois, pp. 1-105.
Page 4d-10
-------
Sect/on 4s Model Descriptions Model Description for Random Walk
13. Users
U.S. Office of Surface Mining. 1981. Groundwater Model Handbook.
Kept. H-D3004-021-81-1062D, U.S. Dept. of Interior, Western Technical
Service Center, Denver, Colorado.
Royce, B., M. Garrell, A. Kahn and E. Kaplan. 1983. A Methodology for
Modeling the Migration of Eor Chemicals in Fresh Water Aquifers.
Prepared for Bartlesville Energy Technology Center, Order No. DE-
API 9-82BC99996, U.S. Dept. of Energy, West Virginia.
Johnson, R. and S.A. Dendrou. 1984. Evaluation of the Potential Water
Supply Impacts Posed by the Transgulf Pipeline, Broward County,
Florida. In: Proceed. Practical Applications of Ground Water Models,
Columbus, Ohio, August 15-17, 1984, pp. 584-601. Nat. Water Well
Assoc., Dublin, Ohio.
Wright, S.J. 1984. Modeling Contaminant Migration Under the Influence
of a Series of Pumping Wells. In: Proceed. Practical Applications of
Ground Water Models, Columbus, Ohio, August 15-17, 1984, pp. 462-
477. Nat. Water Well Assoc., Dublin, Ohio.
Herweijer, J.C., G.A. Van Luijn and C.A.J. Appelo. 1985. Calibration of a
Mass Transport Model Using Environmental Tritium. Journ. of
Hydrology, Vol. , pp.
Naymik, T.G. and M.E. Sievers. 1985. Characterization of Dyetracer
Plumes: In Situ Field Experiments. Ground Water, Vol. 223(6), pp. 746-
752.
American Petroleum Institute. 1986. Cost Model for Selected
Technologies for Removal of Gasoline Components from
Groundwater. API Publication No. 4422, American Petroleum
Institute, Washington D.C.
Anderson, M.P. and C.J. Bowser. 1986. The Role of Groundwater in
Delaying Lake Acidification. Water Resources Research, Vol. 22(7), pp.
1101-1108.
Taylor, M.D., M.K. Leslie, and R.C. Johnson. 1989. Groundwater Modeling
Key to Isolating Contamination. Water and Waste Water
International, February 1989.
Page 44-11
------- |