GWMI 87-08
ig/wc
international ground water modeling center
Quality Assurance in Computer Simulations
of Groundwater Contamination
Paul K.M. van der Heijde
Holcomb Research Institute
Butler University
Indianapolis. Indiana 46200
USA
TNO-DGV Institute
of Applied Geoscience
PO Bo< 285, 26OU *G D«lrt
Th« Nethoilands
TK*
-------
IGUHC GROUNDWATER MODELING REPRINT
Quality Assurance In Computer Simulations of
Groundwater Contamination
Paul K.M. van der Heijde
International Ground Water Modeling Center
reprint
Environmental Software, 1987, Vol 2, No. 1
GWMI 87-08
INTERNATIONAL GROUND WATER MODELING CENTER
Holcomb Research Institute
Butler University
Indianapolis, Indiana 46208
-------
Groundwater Contamination: P.K.M. van der Heijde
Quality Assurance in Computer Simulations of
Groundwater Contamination
Paul K.M. van der Heijde
International Ground Water Modelling Center, Holcomb Research Institute, Butler University,
Indianapolis, IN 46208, USA
ABSTRACT
In the development of policies and, regulations for groundwater protection, In permitting, and 1n planning monitoring
and remedial actions, the role of mathematical models 1s growing rapidly. Because water-resource management decisions
should be based on technically and scientifically sound methods, quality assurance (QA) needs to be applied to groundwater
modeling, both 1n model development and field studies, and should also play an Important part 1n model selection.
Important aspects of QA 1n groundwater model development are peer review, and verification and validation of the
computer code and Us underlying theoretical principles. This paper discusses the role of review and testing as part of an
overall QA approach, and addresses QA 1n model selection and field application.
Key Words: groundwater, mathematical models, quality assurance,
INTRODUCTION
The science of groundwater flow and contaminant
transport 1s not yet an exact field of knowledge.
Although the physical processes Involved obey known
mathematical and physical principles, exact aquifer and
contaminant characteristics are hard to obtain and often
make even plume definition a difficult task. However,
where these characteristics have been reasonably estab-
lished, groundwater models may provide a viable, 1f not
the only, method to predict contaminant transport, to
locate areas of potential environmental risk, and to
assess possible remediation/corrective actions |lj.
Mathematical models are used to help organize the
essential details of complex groundwater management
problems so that reliable solutions are obtained.
Applications Include a wide range of technical, economic,
and sociopolitical aspects of groundwater supply and
protection |2. 3. 4, 5).
A groundwater protection policy based on monitoring
1s by Us very nature always reactive, not preventive;
however, model-based policies and regulations can be both
preventive and reactive. Because adequate on-s1te moni-
toring 1s not always feasible due to costs, available
manpower, or site accessibility, models can provide a
viable and effective alternative. An optimal approach to
the management of groundwater resources Includes the
Integrated use of modeling and monitoring strategies.
The role of groundwater-flow and contaminant-trans-
port models 1n the development of policies and regula-
tions, 1n permitting, and 1n planning of monitoring and
remedial action, 1s continuing to grow. Some of the prin-
cipal areas where mathematical models can now be used to
assist In the management of groundwater protection pro-
grams are |6):
development of regulations and policies
planning and design of corrective actions and waste
storage facilities
problem conceptualization and analysis
development of guidance documents
0266-9838/87/010019-07 $2.00
1987 Computational Mechanics Publications
Paper received on December 8, 1986 Referee: Dr. Paolo Zannetti
model validation, pollution, model selection
design and evaluation of monitoring and data
collection strategies
enforcement
Specifically, groundwater modeling plays or can play a
role 1n:
determining or evaluating the need for regulation
of specific waste disposal, agricultural, and
Industrial practices
analyzing policy Impacts such as evaluating the
consequences of setting regulatory standards and
banning rules, and of dellsting actions
assessing exposure, hazard, damage, and health
risks
evaluating reliability, technical feasibility and
effectiveness, cost, operation and maintenance, and
other aspects of waste-disposal facility designs
and of alternative remedial actions
providing guidance 1n siting of new facilities and
1n permit issuance and petitioning
detecting pollutant sources
developing aquifer or well-head protection zones
assessing liabilities such as post-closure
liability for disposal sites
These activities can be broadly categorized as either
site-specific or generic modeling efforts, and these cate-
gories can be further subdivided into point-source or non-
point-source problems. The success of these modeling
efforts depends on the accuracy and efficiency with which
the natural processes controlling the behavior of ground-
water, and the chemical and biological species it trans-
ports, are simulated. The accuracy and efficiency of the
simulations, 1n turn, depend heavily on the applicability
of the assumptions and simplifications adopted 1n the
model(s), on the availability of reliable data, and on
subjective judgments made by the modeler and management.
If litigation 1s Involved, the model code Itself and
Its theoretical foundation may become contested. There-
ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1 19
-------
Groundwater Contamination: P.K.M. van der Heijde
fore, adequate guidelines should be developed for selec-
tion of simulation codes to be used under such circum-
stances. Such guidelines should cover code review, vali-
dation, and documentation and should be widely accepted.
It 1s of the highest Importance that water resource
management decisions be based on the use of technically
and scientifically sound data collection, Information
processing, and Interpretation methods. Quality Assurance
(QA) provides the mechanisms to ensure that decisions are
based on the best available data and analyses. This paper
discusses QA guidelines applicable to groundwater modeling
and the role of QA 1n the model selection process.
QUALITY ASSURANCE IN GROUNOWATER MODELING
Quality assurance in groundwater modeling is the
procedural and operational framework put In place by the
organization managing the modeling study, to assure tech-
nically and scientifically adequate execution of all pro-
ject tasks Included in the study, and to assure that all
modeling-based analysis is verifiable and defensible [7|.
QA in groundwater modeling should be applied to both model
development and model application and should be an Inte-
gral part of all projects. The two major elements of
quality assurance are quality control (QC) and quality
assessment. Quality control refers to the procedures that
ensure the quality of the final product. These procedures
include the use of appropriate methodology, adequate vali-
dation, and proper usage.
To monitor the quality control procedures and to
evaluate the quality of the products of field studies,
quality assessment 1s applied. It consists of two ele-
ments: auditing and technical review. Audits are pro-
cedures designed to assess the degree of compliance with
QA requirements, commensurate with the level of QA pre-
scribed for the project. Compliance 1s measured in terms
of traceability of records, accountability (approvals from
responsible staff), and fulfillment of commitments 1n the
QA plan. Technical review consists of independent evalua-
tion of the technical and scientific basis of a project
and the usefulness of Us results.
QA is the responsibility of both the project team
(quality control and Internal evaluation) and the con-
tracting or supervising organization (quality assessment).
QA should not drive or manage the direction of a project
nor Is QA intended to be an after-the-fact filing of
technical data.
tives. Major elements of such a QA plan are: (1) formu-
lation of QA objectives and required quality level 1n
terras of validity, uncertainty, accuracy, completeness,
and comparability; (2) development of operational proce-^
dures and standards for performing adequate model1r
studies; (3) establishing a paper trail for QA activities'
in order to document that standards of quality have been
maintained; and (4) Internal and external auditing and
review procedures. The QA plan should also specify
Individual responsibilities for achieving these goals.
Model Development
Ideally, QA should be applied to all codes currently
1n use and yet-to-be-developed codes. Relevant QA proce-
dures include such aspects as the verification of the
mathematical framework, field validation, benchmarking,
and code comparison. A detailed discussion of model
testing and review 1s described in the second part of this
paper.
QA for code development and maintenance should
Include complete record-keeping of the model development,
of modifications made In the code, and of the code-valida-
tion process. The paper trail for QA in model development
consists of reports and files on the development of the
model. The reports should Include a description of:
assumptions
parameter values and sources
boundary and initial conditions
nature of grid and grid design justification
changes and verification of changes made in code
actual input used
output of model runs and Interpretation
validation (or at least calibration) of model
In addition, depending on the level of QA required
the following files may be retained (in hard-copy and, '
higher levels, 1n digital form):
version of source code used
verification Input and output
validation input and output
application Input and output
If any modifications are made to the model coding for
a specific problem, the code should be tested again; all
QA procedures for model development should again be
applied. Including accurate record keeping and reporting.
All new Input and output files should be saved for Inspec-
tion and possible reuse.
Various phases of quality assessment exist for both
model development and application. First, review and
testing 1s performed by the author, and sometimes by other
employees not Involved In the project, or by Invited
experts from outside the organization. Also to be con-
sidered 1s the quality assessment by the organization for
which the project has been carried out. Again, three
levels can be distinguished: project or product review or
testing by the project officer or project monitor, by
technical experts within the funding or controlling
organization, and by an external peer review group.
Decisions by natural resources and environmental
managers rest on the quality of environmental data and
data analysis; therefore, program managers In regulatory
agencies should be responsible for: (1) specifying the
quality of the data required from environmentally related
measurements and for the level of problem-solving data
analysis; and (2) providing sufficient resources to assure
an adequate level of QA.
QA procedures should be contained 1n a QA plan to be
developed for each modeling study. The plan lists the
measures required to achieve prescribed quality objec-
Model Application
QA in model application should address all facets of
the model application process:
correct and clear formulation of problems to be
solved -
project description and objectives
modeling approach to the project
1s modeling the best available approach and if so,
is the selected1 model appropriate and cost-
effective?
conceptualization of system and processes,
Including hydrogeologic framework, boundary
conditions, stresses, and controls
explicit description of assumptions and
simplifications
data acquisition and Interpretation
model selection, or justification for choosing to
develop a new model ~
model preparation (parameter selection, data ent
or reformatting, grldding)
the validity of the parameter values used In the
model application
protocols for parameter estimation and model call-
20 ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1
-------
Groundwater Contamination: P.K.M. van der Heijde
bratlon to provide guidance, especially for sensi-
tive parameters
level of Information 1n computer output (variables
and parameters displayed; formats; layout)
Identification of calibration goals and evaluation
of how well they have been met
sensitivity analysis
postslmulatlon analysis (Including verification of
reasonablHty of results, interpretation of
results, uncertainty analysis, and the use of
manual or automatic data processing techniques, as
for contouring)
establishment of appropriate performance targets
(e.g., 6-foot head error should be compared with a
20-foot head gradient or drawdown, not with the
250-foot aquifer thickness!); these targets should
recognize the Units of the data
presentation and documentation of results
evaluation of how closely the modeling results
answer the questions raised by management
A major problem in model use 1s model credibility.
In the selection process special attention should be given
to ensure the use of qualified models that have undergone
adequate review and testing.
As 1s the case with model development QA, all data
files, source codes, and executable versions of computer
software used 1n the modeling study should be retained for
auditing or postproject re-use.
MODEL REVIEW AND TESTING
Before a groundwater model 1s used as a planning and
decision-making tool, its credentials must be established.
Independently of its developers, through systematic
testing and evaluation of the model's characteristics.
Code testing is generally considered to encompass verifi-
cation and validation of the model |8|. To evaluate
groundwater models in a systematic and consistent manner,
the International Ground Water Modeling Center (IGWMC) has
developed a model review, verification, and validation
procedure (5). Generally, the review process is
qualitative in nature, while code testing results can be
evaluated by quantitative performance standards.
Model Review
model. Such a procedure determines whether the concepts
of a model adequately represent the nature of the system
under study, and identifies the processes and actions
pertinent to the model's Intended use. The examination
also determines whether the equations representing the
various processes are valid within the range of the
model's applicability, whether these equations conform
mathematically to the intended range of the model's use,
and whether the selected solution approach is the most
appropriate. Finally, model examination determines the
appropriateness of the selected Initial and boundary
conditions and establishes the applicability range of the
model.
For complex models, detailed examination of the
implemented algorithms is required to determine whether
appropriate numerical schemes, 1n the form of a computer
code, have been adopted to represent the model I10|. This
step should disclose any inherent numerical problems such
as non-uniqueness of the numerical solution, Inadequate
definition of numerical parameters. Incorrect or nonopti-
mal values used for these parameters, numerical disper-
sion, numerical Instability such as oscillations or diver-
gent solution, and problems regarding conservation of
mass.
In addition, the specific rules for proper appli-
cation of the model should be analyzed from the perspec-
tive of Us Intended use. These rules include data
assignment according to node-centered or block-centered
grid structure for finite-difference methods; size and
shape of elements 1n integrated finite-difference and
finite-element methods; grid size variations; treatment of
singularities such as wells; approach to vertical aver-
aging in two-dimensional horizontal models or ' layered
three-dimensional models; inclusion of partial solutions
1n analytical element; methods; and treatment of boundary
conditions. Consideration is also given to the ease with
which the mathematical equations, the solution procedures,
and the final results can be physically interpreted.
Evaluation of Model Documentation
Model documentation is evaluated through visual
inspection, comparison with existing documentation
standards and guidelines, and through*its use as a guide
in preparing for and performing verification and
validation runs.
A complete review procedure comprises examination of
model concepts, governing equations, and algorithms
chosen, as well as evaluation of documentation and general
ease-of-use, and examination of the computer coding (5,
9|. If the model has been verified or validated by the
author, the review procedure should Include evaluation of
this process.
To facilitate thorough review of the model, detailed
documentation of the model and its developmental history
1s required. In addition, to ensure Independent evalua-
tion of the performed verification and validation, the
computer code should be available or at least accessible
for Implementation on the reviewer's computer facilities,
together with a file containing the original test data
used in the code's verification and validation.
Review should be performed by experienced modelers
knowledgeable in theoretical aspects of groundwater
modeling. Because review 1s rather subjective 1n nature,
selection of the reviewers 1s a sensitive and critical
process.
Good documentation includes a complete treatment of
the equations on which the model is based, of the under-
lying assumptions, of the boundary conditions that can be
Incorporated 1n the model, of the method used to solve the
equations, and of the limiting conditions resulting from
the chosen method. The documentation must also Include a
user's manual containing instructions for operating the
code and preparing data files, example problems complete
with input and output, programmer's Instructions, computer
operator's instructions, and a report of the initial code
verification.
Evaluating Ease of Use
The data files provided by the model developer are
used to evaluate the operation of the code and the user's
guide through a test-run process. In this stage special
attention is given to the rules and restrictions
("tricks," e.g., to overcome restrictions in applic-
ability) necessary to operate the code, and to the code's
ease-of-use aspects (111.
Model Examination
Computer Code Inspection
Model examination determines whether anything funda-
mental was omitted 1n the Initial conceptualization of the
Part of the model review process is the inspection of
the computer code. In this Inspection attention is given
ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1 21
-------
Groundwater Contamination: P.K.M. van der Heijde
to the manner in which modern programming principles have
been applied with respect to code structure, optimal use
of the programming language, and internal documentation.
This step helps reveal undetected programming or logic
errors, hard to detect 1n verification runs.
MODEL VERIFICATION
The objective of the the verification process 1s
twofold: (1) to check the accuracy of the computational
algorithms used to solve the governing equations, and (2)
to assure that the computer code 1s fully operational.
To check the code for correct coding of theoretical
principles and for major programming errors ("bugs"), the
code is run using problems for which an analytical solu-
tion exists. This stage is also used to evaluate the
sensitivity of the code to grid design, to various domi-
nant processes, and to a wide selection of parameter
values 19, 12, 13. 14|.
Although testing numerical computer codes by com-
paring results for simplified situations with those of
analytical models does not guarantee a fully debugged
code, a we 11-selected set of problems ensures that the
code's main program and most of Its subroutines. Including
all of the frequently called ones, are being used in the
testing. In the three-level test procedure developed by
the International Ground Water Modeling Center (IGWMC),
this type of testing Is referred to as level I 115).
Hypothetical problems are ^used to test special
features that cannot be handled by simple close-form
solutions, as in testing irregular boundary conditions and
certain heterogeneous and anisotropfc aquifer properties;
this is the IGWMC level II testing.
For both level I and level II testing, sensitivity
analysis is applied to further evaluate code characteris-
tics.
MODEL VALIDATION
Model validation or field validation is defined as
the comparison of model results with numerical data Inde-
pendently derived from laboratory experiments or obser-
vations of the environment |10|. Complete model valida-
tion requires testing over the full range of conditions
for which the model is designed. Model development Is an
evolutionary process responding to new research results,
developments in technology, and changes 1n user require-
ments. Model review and validation needs to follow this
dynamic process and should be applied each time the model
Is modified.
The objective of model validation is to determine how
well the model's theoretical foundation describes the
actual system behavior in terms of the "degree of correla-
tion" between model calculations and actual measured data
for the cause-and-effect responses of the system.
Obviously, a comparison with field data is required. Such
a comparison may take either of two forms. One form,
calibration, is sometimes considered the weaker form of
validation insofar as it tests the ability of the code
(and the model) to fit the field data, with adjustments of
the physical parameters |13|. Some researchers prefer to
classify calibration as a form of verification rather than
a form of validation.
The other form of validation 1s that of prediction.
This is a test of the model's ability to fit the field
data with no adjustments of the physical parameters. In
principle, this is the correct approach to validation.
However, unavailability and inaccuracy of field data often
prevent such a rigid approach-. Typically, a part of the
field data 1s designated as calibration data, and a cali-
brated site-model 1s obtained through reasonable adjust-
ment of parameter values. Another part of the field data
is designated as validation data; the calibrated site
model is used in a predictive mode to simulate similar
data for comparison. The quality of such a test Is there-
fore determined by the extent to which the site model is
"stressed beyond" the calibration data on which It 1s
based |13|. In the IGWMC testing procedure, this approach
is referred to as level III testing.
For many types of groundwater models, a complete set
of test problems and adequate data sets for the described
testing procedure 1s not yet available. Therefore,
testing of such models is generally limited to extended
verification, using existing analytical solutions, and to
code 1ntercompar1son.
Whether a model Is valid for a particular application
can be assessed by performance criteria, sometimes called
validation or acceptance criteria. If various uses 1n
planning and decision making are foreseen, different
performance criteria might be defined. The user should
then carefully check the validity of the model for the
Intended use.
Three levels of validity can be distinguished |10|:
(1) Statistical Validity: using statistical
measures to check agreement between two differ-
ent distributions, the calculated one and the
measured one; validity is established by using
an appropriate performance or validity
criterion.
(2) Deviative Validity: if not enough data are
available for statistical validation, a
deviation coefficient D can be established,
e.g.,
0 = l(x-y)/x|100X
where x = predicted value and y = measured
value. The deviation coefficient might be
expressed as a summation of relative devla-'
tions. If ED is a deviattve validity criterion
supplied by subjective judgment, a model can
considered to be valid 1f D < ED.
(3) Qualitative Validity: using a qualitative scale
for validity levels representing subjective
judgment: e.g., excellent, good, fair, poor,
unacceptable. Qualitative validity is often
established through visual inspection of graphic
representations of calculated and measured data
116).
The aforementioned tests apply to single variables
and determine Iocal-or-s1ngle variable validity; if more
than one variable is present in the model, the model
should also be checked for global validity and for
validity consistency [10|. For a model with several
variables to be globally valid, all the calculated outputs
should pass validity tests. Validity consistency refers
to the variation of validity among calculations having
different Input or comparison data sets. A model might.be
judged valid under one data set but not under another,
even within the range of conditions for which the model
has been designed or is supposedly applicable. Validity
consistency can be evaluated periodically when models have
seen repeated use.
Often, the data used for field validation are not
collected directly from the field but are processed In an
earlier study. Therefore, they are subject to inaccur-
acies, loss of information, Interpretive bias, loss of
22 ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1
-------
Groundwater Contamination: P.K.M. van der Heijde
precision, and transmission and processing errors,
resulting In a general degradation of the data.
As noted earlier, for many types of groundwater
models no field data sets are available to execute a
complete validation. One approach sometimes taken Is that
of code Intercomparlson, where a newly developed model 1s
compared with existing models designed to solve the same
type of problems as the new model. If the simulation
results from the new code do not deviate significantly
from those obtained with the existing code, a relative or
comparative validity 1s established. It 1s obvious that
as soon as adequate data sets become available, all the
Involved models should be validated with those data.
Further development of databases for field validation
of solute transport models 1s necessary. This 1s also the
case for many other types of groundwater'models. These
research databases should represent a wide variety of
hydrogeologlcal situations and should reflect the various
types of flow, transport, and deformation mechanisms
present 1n the field. The databases should also contain
extensive Information on hydrogeological, soil, geochemi-
cal, and cl1matolog1cal characteristics. WHh the devel-
opment of such databases and the adoption of standard
model-testing and validation procedures, the reliability
of models used 1n field applications can be Improved
considerably.
Validation Scenarios
Often, various approaches to field validation of a
model are viable. Therefore, the validation process
should start with defining validation scenarios. Planning
and conducting field validation should include the
following steps |17|:
(1) Define data needs for validation and select an
available data set or arrange for a site to
study.
(2) Assess the data quality in terms of accuracy
(measurement errors), precision, and
completeness.
(3) Define model performance or acceptance criteria.
(4) Develop strategy for sensitivity analysis.
(5) Perform validation runs and compare model
performance with established acceptance
criteria.
Sensitivity Analysis
An Important characteristic of a model is Its
sensitivity to variations or uncertainty in input para-
meters. Sensitivity analysis defines quantitatively or
sem1quant1tatively the dependence of a selected model per-
formance assessment measure (or an Intermediate variable)
on a specific parameter or set of parameters 118]. Model
sensitivity can be expressed as the relative rate of
change of selected output caused by a unit change in the
Input. If the change in the input causes a large change
1n the output, the model is sensitive to that Input.
Sensitivity analysis is used to identify those parameters
most Influential in determining the accuracy and precision
of model predictions. This Information is of Importance
to the user, as he must establish required accuracy and
precision 1n the model application as a function of data
quantity and quality 117). In this context the use of a
sensitivity Index as described by Hoffman and Gardner [19)
1s of Interest. It should be noted that 1f models are
coupled, as in multimedia transport of contaminants, the
propagation of errors and the Increase in uncertainty
through the subsequent simulations must be analyzed as
part of the sensitivity analysis.
MODEL SELECTION
Using models to analyze alternative solutions to
groundwater problems requires a number of steps, each of
which should be taken conscientiously and reviewed care-
fully. After the decision to use an existing model has
been made, the selection process is Initiated. As model
credibility is a major problem in model use, special
attention should be given in the selection process to
ensure the use of qualified models that have undergone
adequate review and testing. Selecting an appropriate
model 1s crucial to the success of a modeling project.
Model selection 1s the process of matching a detailed
description of the modeling needs with well-defined,
quality-assured characteristics of existing models, while
taking into account the objectives of the study and the
limitations 1n the personnel and material resources of the
modeling team. In selecting an appropriate model, both
the model requirements and the characteristics of existing
models must be carefully analyzed. Major elements in
evaluating modeling needs are: (1) formulation of the
management problems to be solved and the level of analysis
sought; (2) description of the system under study; and (3)
analysis of the constraints in human and material
resources available for the study. Model selection is
partly quantitative and partly qualitative. Many subjec-
tive decisions must be made, often because there are
Insufficient data in the selection stage of the project to
establish the Importance of certain characteristics of the
system to be modeled.
Definition of modeling needs is based on the manage-
ment problem at hand, questions asked by planners and
decision makers, and on the understanding of the physical
system, including the pertinent processes, boundary condi-
tions, and system stresses. The major criteria in selec-
ting a model are: (1) that the model Is suited for the
Intended use; (2) that the model is thoroughly tested and
validated for the Intended use; and (3) that the model
code and documentation are complete and user-friendly.
Regardless of whether problem-solving performance
standards are set, management-oriented criteria need to be
developed for evaluating and accepting models. Such a set
of scientific criteria should include:
trade-offs between costs of running a model and
accuracy
profile of model user and definition of required
user-friendliness
accessibility 1n terms of effort, cost, and
restrictions
acceptable temporal and spatial scale and level of
aggregation
If different problems must be solved, more than one
model might be needed or a model might be used 1n more
than one capacity. In such cases, the model requirements
for each of the problems posed have to be clearly defined
at the outset of the selection process. To a certain
extent this is also true for modeling the same system in
different stages of the project. Growing understanding of
the system and data availability might lead to a need for
a succession of models of increasing complexity. In such
cases, flexibility of the model or model package might
become an important selection criterion.
It should be realized that a perfect match rarely
exists between desired characteristics and those of
available models. Many of the selection criteria are
subjective or weakly justified. If a match is hard to
obtain, reassessment of these criteria and their relative
weight 1n the selection process is necessary. Hence,
model selection 1s very much an iterative process.
In standardizing model selection, three major
approaches are employed in characterizing the validation
of numerical models. In one, the model is tested
ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1 23
-------
Groundwater Contamination: P.K.M. van der Heijde
according to established procedures; when accepted, the
model Is prescribed In federal or state regulations for
use 1n cases covered by those regulations. This approach
does not leave much flexibility for Incorporating the
advances of recent research and technological development.
The second approach Includes the establishment of a list
of groundwater simulation codes as "standard" codes for
various generic and site-specific management purposes. To
be listed, a code should pass a widely accepted review and
test procedure such as that described in a previous part
of this paper. This approach Is suggested 1n a recent
evaluation of the role of modeling in the U.S. Environ-
mental Protection Agency |6|. It should be noted that
establishing "standard" models will not prevent discussion
of the appropriateness of a selected model for analysis of
a specific problem nor of its proper use 1n a particular
decision-making process. In considering these two
approaches, questions have been raised such as [6]:
Are there legal liabilities for setting up certain
models as acceptable? (For Instance, if an enforcement
agency certifies a model for use, can that agency no
longer criticize an Industry's use of that model?)
Does certification squelch the development of new,
better models?
What balance should there be between using the newer,
faster models and using mature models already subjected
to peer review?
A third approach is to prescribe a review-and-test
methodology in regulations of enforcement agencies, and
require the model development team to show that the model
code satisfies the requirements. This approach leaves
room to update the codes as long as each version is ade-
quately reviewed and tested. An example 1s the quality
assurance program for models and computer codes of the
U.S. Nuclear Regulatory Commission [20).
extensive need for these models in assessing current and
potential water quality problems has resulted in two
groups of modelers: (1) model developers who are research-
oriented and who generally apply models only for*verifica-
tion and validation purposes, and (2) model users who
apply models routinely to actual generic or site-specific
groundwater problems. The economic consequences of model
predictions and the potential liabilities incurred by
their use have brought quality .guarantees and code
credibility to the forefront as major issues 1n ground-
water modeling. Hence, quality assurance (QA) needs to be
defined for both model development and model application.
There is a significant difference between these two: the
first 1s designed to result 1n a reliable code, and the
second to interpret correctly the simulation results.
Both require stringent QA procedures to be established and
enforced. As model credibility has become a major con-
cern, model selection should focus on those codes that
have undergone adequate review and testing. To further
Increase the applicability of the models, good documenta-
tion and user-friendliness of the computer coding Involved
should receive proper attention.
ACKNOWLEDGEMENT
The research described in this publication has been funded
in part by the U.S. Environmental Protection Agency
through Cooperative Agreement 0CR-812603 with the Holcomb
Research Institute. It has not been subjected to the
Agency peer and policy review and therefore does not
necessarily reflect the views of the Agency, and no
official endorsement should be Inferred.
In any case, a general framework of nondlscrlmlnatory
criteria should be established [6]. These criteria should
include:
publication and peer review of the conceptual and
mathematical frame-work
full documentation and visibility of the assumptions
testing of the code according to prescribed methods;
this should Include verification (checking the accuracy
of the computational algorithms used to solve the
governing equations), and validation (checking the
ability of the theoretical foundation of the code to
describe the actual system behavior)
trade secrets (unique algorithms that are not
described) should not be permitted if they might affect
the outcome of the simulations; proprietary codes are
already protected by the copyright law
Finally, as model selection is very closely related
to system concep-tualizatlon and problem solving, "expert
systems" integrating system conceptualization and model
selection on a problem-oriented basis promise to be
valuable tools.
Further Information on groundwater model selection Is
presented 1n [21, 22, 23, 24|.
SUMMARY
During the 1970s a rapidly Increasing awareness of
the threat posed to groundwater resources by human-induced
chemical and biological pollution has accelerated the
development of sophisticated simulation models. These
models are based on mathematical descriptions of the
physical, chemical, and biological processes that take
place in a complex hydrogeologlcal environment. The
REFERENCES
HI van der Heijde, P.K.M. 1986. Modeling contaminant
transport in groundwater: approaches, current
status, and needs for further research and
development. In Groundwater Contamination,
proceedings conf,, Santa Barbara, CA, November
11-16, 1984.
[2] Mercer, J.W., and C.R. Faust. 1981. Ground-water
Modeling. National Water Well Association, 60 pp.
[3| U.S. Office of Technology Assessment. 1982. Use of
Models for Water Resources Management, Planning, and
Policy. OTA Report 0-159, August, Congress of the
United States, 233 pp.
[41 Javandel, I., C. Doughty, and C.F. Tsang. 1984.
Croundvater Transport: Handbook of Mathematical
Models. Water Resources Monograph Series 10,
American Geophysical Union, Washington, DC.
[51 van der Heijde, P.K.M., Y. Bachmat, J. Bredehoeft,
B. Andrews, D. Holtz, and S. Sebastian. 1985.
Croundvater Management: The Use of Numerical Models,
2nd ed. Water Resources Monograph 5. Washington,
DC: American Geophysical Union. 180 pp.
[6] van der Heijde, P.K.M.. and R.A. Park. 1986. U.S.
EPA Groundwater Modeling Policy Study Group, Report
of Findings and Discussion of Selected Groundwater
Modeling Issues. EPA/CER1, Cincinnati, OH (In
press).
[7| Taylor, J.K. 1985. What Is quality assurance? in
J.K. Taylor and T.W. Stanley (eds.), Quality Assur-
ance for Environmental Measurements. ASTM Special
Technical Publication 867, Am. Soc. for Testing and
Materials, Philadelphia, PA, pp. 5-11.
[8| Adrion, W.R., M.A. Branstad, and J.C. Cherntasky.
1982. Validation, verification, and testing of
computer software. ACM Computing Surveys,
Vol.14(2), pp. 159-192.
24 ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1
-------
Groundwater Contamination: P.K.M. van der Heijde
|9] Huyakorn, P.S., A.G. Kretschek. R.W. Broome, J.W.
Mercer, and B. H. Lester. 1984. "Testing and vali-
dation of models for simulating solute transport 1n
groundwater: development, evaluation, and
comparison of benchmark techniques." GWMI 84-13,
International Ground Water Modeling Center, Holcomb
Research Institute, Butler University, Indianapolis,
IN, 420 pp.
|10| ASTM. 1984. Standard practices for evaluating
environmental fate models of chemicals. Annual book
of ASTM standards, E 978-84, Am. Soc. for Testing
and Materials, Philadelphia, PA.
|11| van der Heijde, P.K.M. 1984. Availability and
applicability of numerical models for groundwater
resources management. In Practical Applications of
Ground water Hodeis, proceedings NWWA/IGWMC conf.,
Columbus, OH, August 15-17, 1984.
(12) Sykes, J.F.. S.B. Pahwa, D.S. Ward, and R.B.
Lant2. 1983. The validation of SWENT, a geosphere
transport model. In Scientific Computing, R.
Stepleman et al. (eds.). New York: IMACS/North-
Holland Publishing Company.
113] Ward, D.S., M. Reeves, and L.E. Duda. 1984.
Verification and field comparison of the Sandia
Waste-Isolation Flow and Transport Model (SWIFT).
NUREG/ CR-3316, U.S. Nuclear Regulatory Commission.
Washington, DC.
I14| Gupta, S.K., C.R. Cole, F.W. Bond, and A.M. Monti.
1984. Finite-element three-dimensional ground-water
(FE3DGW) flow model: Formulation, computer source
listings, ' and user's manual. ONW1-548, Battelle
Memorial Inst., Columbus, OH.
[15] van der Heijde, P.K.M., P.S. Huyakorn, and J.W.
Mercer. 1985. Testing and validation of
groundwater models. In Practical Applications of
Croundvater Modeling, proceedings NWWA/IGWMC COnf.,
Columbus, OH, August 19-20, 1985.
116] Cleveland, W.S.. and R. McGlll. 1985. Graphical
perception and graphical methods for analyzing
scientific data, science 229:828-833.
|17] Hern. S.C., S.M. Melancon, and J.E. Pollard.
1985. Generic steps in the field validation of
vadose zone fate and transport models. in Hern,
S.C., and S.M. Melancon (eds.), Vadose Zone Modeling
of Organic Pollutants. Chelsea, MI: Lewis
Publishers, Inc., pp. 61-80.
|18] Intera Environmental Consultants, Inc. 1983. A
proposed approach to uncertainty analysis. ONW1-
488, Battelle Memorial Inst., Columbus, OH, 68 pp.
|19| Hoffman,- F.D., and R.H. Gardner. 1983. in J.E.
Till and H.R. Meyer (eds.), Hadioiogicai
Assessment. NUREG/CR-3332, ORNL-5968, U.S. Nuclear
Regulatory Commission, Washington, DC.
|20) Silling, S.A. 1983. Final technical position on
documentation of computer codes for high-level waste
management. NUREG/CR-0856, U.S. Nuclear Regulatory
Commission, Washington, DC 11, pp.
|21| Rao, P.S.C., R.E. Jessup, and A.C. Hornsby. 1981.
Simulation of nitrogen in agro-ecosystems: criteria
for model Selection and use. In Nitrogen Cycling in
Ecosystems of Latin Aaerica and tne Caribbean.
Proceeding Internal. Workshop, Call, Colombia, March
16-21, 1981, pp. 1-16.
122] Kincaid, C.T., J.R. Morrey, and J.E. Rogers.
1984. Geohydrological models for solute migration;
Vol. 1: Process description and computer code for
selection. EA 3417.1, Electric Power Research Inst.,
Palo Alto, CA.
|23| Boutwell, S.H., S.M. Brown, B.R. Roberts, and D.F.
Atwood. 1985. Modeling remedial actions at uncon-
trolled hazardous waste sites. EPA/540/2-85/001,
U.S. Environmental Protection Agency, OSWER/ORD,
Washington, DC.
|24] Simmons, C.S., and C.R. Cole. 1985. Guidelines for
selecting codes for groundwater transport modeling
of low-level waste burial sites. Volume
1Guideline approach. PNL-4980 Vol. 1. Battelle
Pacific NW Laboratory, Richland, WA.
ENVIRONMENTAL SOFTWARE, 1987, Vol 2, No. 1 25
------- |