Subsurface Modeling
August 13-16, 1996
Ada, Oklahoma
3
a
Sponsored by the
Robert S. Kerr Environmental Research Laboratory
United States Environmental Protection Agency
-------
EPA/96-002
Subsurface Modeling
August 13-16, 1996
Ada, Oklahoma
Sponsored by the
Subsurface Protection and Remediation Division
National Risk Management Research Laboratory
United States Environmental Protection Agency
-------
Subsurface Modeling
1. Introduction
2. Saturated Flow in Aquifers
3. Introduction to Modeling
4. References
5. Availability of Models
6. Center for Subsurface Modeling Support
7. Practical Applications of MODFLOW
8. Ground Water Flow Modeling with the Wellhead
Analytical Element Model (WhAEM)
9. Solute Transport Modeling
10. Vadose Zone Flow
11. Introduction to Nonaqueous Phase Liquids
12. The Hydrocarbon Spill Screening Model
13. Computer Familiarization
-------
Introduction
-------
1. Introduction
-------
Subsurface Modeling
August 13-16, 1996
U.S. Environmental Protection Agency
Subsurface Protection and Remediation Division
National Risk Management Research Laboratory
Ada, Oklahoma
Purpose
This 3-1/2 day training course will include an introduction to the process and philosophy of
modeling, and a discussion of the availability of models. Several site-specific ground water flow
and contaminant transport problems, chosen from Superfund, RCRA, Wellhead protection and UST
programs, will be discussed. Site data will be used to guide the application of a model to answer
specific questions about the site. The models presented will include the Wellhead Analytical
Element Model (WhAEM) for capture zone delineation, the Hydrocarbon Spill Screening Model
(HSSM) for petroleum product releases, and the two layer model (TWOLAY) for free product
recovery. Each student in the course will have the opportunity to run the simulation models and
study various problem outcomes.
Special Notice
For the August Subsurface Modeling Course we will have two special sessions on MODFLOW
drawn from the US EPA MODFLOW Instructional Manual. These sessions will be taught by
Brad Hill of CDSl/CSMoS.
US EPA Instructors
Dr. David Burden Dr. Jim Weaver
Dr. Stephen Kraemer Mr. Joseph Williams
-------
Subsurface Modeling Course Agenda
Tuesday 8:00 - 8:15 Welcome, Introductions, and Introduction to RSKERL
8:15 -10:30 Purpose of the class and Introduction to modeling
10:45 -11:45 Computer familiarization
11:45-1:15 Lunch
1:15- 2:00 Introduction to modeling (Continued)
2:00 - 2:30 Availability of models
2:30 - 3:00 Resources of the Center for Subsurface Modeling Support (CSMoS)
3:00-5:00 1st Special Session on MODFLOW
Wednesday 8:00 -10:30 Groundwater flow modeling for well head protection (WhAEM)
10:45 -12:00 Groundwater flow modeling for well head protection (WhAEM)
12:00-1:00 Lunch
1:00 - 3:00 Groundwater flow modeling for well head protection (WhAEM)
3:15 - 5:00 Groundwater flow modeling for well head protection (WhAEM)
6:30 - 8:30 2nd Special Session on MODFLOW
Thursday 8:00 -10:30 Modeling Solute Transport
10:45 -11:45 Introduction to NAPLs
11:45-1:15 Lunch
1:15 - 3:00 NAPL flow and transport at a LUST site (HSSM)
3:15 - 4:00 Introduction to Free-Product Recovery (TWOLAY)
4:00 - 5:00 Computer Lab Session or Optional Lab Tour
6:30 - 9:00 Optional Problem Session (HSSM, TWOLAY, WhAEM, MODFLOW)
Friday
8:00 -10:30 NAPL flow at an Air Force base (TWOLAY)
10:45 -11:45 Computer Lab Session
11:45-12:00 Course wrap-up
-------
Robert S. Kerr Environmental Research Laboratory
o United States Environmental Protection Agency
3 P.O. Box1198
" Ada, Oklahoma 74820
RSKERL Modeling Contacts
Superfund Technology Support Center
Dave Burden
Dom Digiulio
Randall Ross
Joe Williams
GIS applications
Air/Vapor Flow
MODFLOW/GIS
Unsaturated Zone
405-436-8606
405-436-8607
405-436-861 1
405-436-8608
Center for Subsurface Modeling Support (GSA IAG/ Contract with CDSI)
Brad Hill
Dan West
Model Distribution and
support, MODFLOW
GIS Applications
405-436-8586
405-436-8717
Subsurface Systems Branch
Jong Cho
Steve Kraemer
Tom Short
Jim Weaver
Vapor extraction
Regional ground water
flow
Solute Transport
Multiphase Flow
405-436-8547
405-436-8549
405-436-8544
405-436-8545
-------
The Ada Community
Ada is the county seat of Pontotoc County, consisting of 714 square miles of rolling terrain with a mean elevation of
about 1,058 feet. The City is located in Southeastern Oklahoma about 80 miles southeast of Oklahoma City and 180
miles north of Dallas. The population of the City is almost 18,000 and that of the County is slightly less than 35,000.
Ada is noted for its high quality water which comes from Byrds Mill Spring located south of Ada and originating in
the Arbuckle Uplift. The spring, named after Benjamin Franklin Byrd who operated a grist mill nearby, flows up to
eighteen million gallons daily. The attractions for the spring as a water supply for Ada included its location, tremen-
dous quantity, excellent quality, and the possibility for gravity flow to the City.
Ada is the center of an area well balanced between industry and agriculture. It is the heart of a strong retail trade area
and a center for higher education, medical care, and research. The area is also one of Oklahoma's richest oil produc-
ers.
Beef cattle are one of the largest economic industries in Pontotoc County and the trade area. It boasts some of the
state's finest native and improved pasture lands and serves as an excellent source of foundation herds for all breeds
of beef cattle and horses. National and regional sales of many popular breeds of horses and livestock are held
annually.
Oil and gas remains a tremendous factor in the area's economy. This community is assured of a supply of natural gas
with its source originating in the Anadarko Basin. There is an underground compressor station facility that provides
for the storage of over fourteen billion cubic feet of gas. It is possible for a heavy user of natural gas to develop their
own gas field in the area near Ada.
East Central University, one of Oklahoma's finest institutions of higher education by scholastic standards, is located
in Ada. It has a department of environmental science which has established a state and national stature. In addition
to a department of business administration which trains students in advance skills of accounting, computer science,
and other related subjects, it has a nursing program and other para-medical courses. Also located on the campus is
one of the finest officer's training programs in the nation. Handicapped students are enrolled in all departments at
the University and architectural barriers have been removed from learning facilities as well as living quarters for
their benefit.
The Robert S. Kerr Environmental Research Laboratory is an Environmental Protection Agency facility which was
founded in 1966. With nearly 200 federal, contract, student, and visiting scientists working at the Laboratory, it has
proven to be a world leader in the development of science and technology pertaining to ground-water protection and
restoration.
The Social Services Center Complex houses state agencies which include State Board of Public Affairs; District
Health Office; Regional Guidance Center; Tag Agency; Pontotoc County Health Department; Visual Services; Evalu-
ation Center; Total Learning Center; Social Security; Oklahoma Employment Security Commission; Department of
Public Safety; Oklahoma Department of Institutions, Social and Rehabilitative Services; Oklahoma State University
District Extension Office; and the American Red Cross.
Ada is the center for many Native American activities including the Indian Housing Authority, which serves eighteen
counties, and a thirteen million dollar Indian Hospital. The Carl Albert Indian Health Facility is a 75-bed, solar
powered, general medical and surgical hospital serving eligible Indian residents in south central Oklahoma. The
-------
physician staff of 14 includes specialists in the areas of internal medicine, pediatrics, general surgery, obstetrics/
gynecology, radiology, and family practice. The hospital and its outreach health centers located in Shawnee,
Tishomingo, and Wewoka are fully accredited by the Joint Commission for the Accreditation of Hospitals and are
part of the nationwide Indian health service system.
Valley View Regional Hospital, which was opened in 1986, is a 200-bed regional health center which is fully accred-
ited by the Joint Commission for Accreditation of Hospitals. Valley View, serving a ten-county area of south central
Oklahoma, is highly specialized in several areas of patient care including: coronary intensive care, surgical intensive
care, neonatal intensive care, ultrasound, CT scan, special procedures, nuclear medicine, xerography, an out-patient
surgery center, and a cancer treatment center. Ambulance units, specially trained emergency medical technicians,
emergency nurses and emergency medicine physicians are on duty 24 hours a day serving Pontotoc County resi-
dents. Valley View is a remote cardiac monitoring center, providing heart monitoring service for patients of smaller
hospitals in surrounding areas. It is also the center for a talk-back network which supplies continuing education
programs to more than 46 hospitals across Oklahoma. Valley View is the clinical site for East Central University's
Schools of Medical Technology, Registered Nursing, Registered Records Administration and Radiology Technol-
ogy. The hospital also serves as a clinical site for the University of Oklahoma's physical therapy students, Oklahoma
State University registered dietician students, and the Byng Vocational Center's licensed practical nursing program.
Through the Oklahoma State Health Department, Valley View also offers a registered emergency medical technician
program.
City of Ada's Recreation Department has the facilities to provide assistance to the community in a number of areas
including family reunions, dances, singing events and other activities for groups of up to 500 persons. Ada is blessed
with a variety of recreational activities which include a strong tennis program, little league teams, four softball fields,
two golf courses, a twenty-four lane bowling alley, skeet range, the home of the Robert S. Kerr Activities Center on
the campus of East Central University. The Ada area is nationally recognized as a coon hunters' paradise with
several different hunts scheduled each year. Ada takes great pride in Wintersmith Park consisting of something for
the entire family including walking trails, fish, games, rides, a zoo, birds, flowers, and picnic areas.
The Area Youth shelter is a non-profit youth service organization which has been in operation since 1971. The
services offered are: emergency shelter, individual counseling, family counseling, tutoring and parent training. In
1977, a mental health program was established for all citizens in the ten county Southern Oklahoma Development
Association. Catchment centers were established in Ada, Durant and Ardmore. Located in the beautiful area south
of Ada near the Kerr Research Laboratory is the Baptist Retirement Home. McCall's chapel is a community village
for exceptional children located just east of Ada.
Rolling Hills Psychiatric Hospital is a 40-bed facility which began admitting patients in 1988. The hospital provides
treatment for adults, adolescents and preadolescents in the areas of psychiatric illness and chemical dependency.
The state of the art facility ensures a private atmosphere for its patients while the hospital's staff and program provide
the best quality of care available to its patients.
There is an active lake country association which promotes interesting activities in the area such as peach festivals,
watermelon seed spitting contests, a sand bass festival, and old time arts and crafts fairs. The locally grown pecans
are called 'nectar of Oklahoma' because of their wonderful flavor.
Ada is a natural center for rock hounds far and wide. It is within driving distance of the rose rocks, trilobites,
brachiopods, crincio bulbs, as well as petrified wood and several different varieties of fossils. Ada is near Indian
arrow chipping grounds.
There is an active artists association which promotes two or three art shows annually as well as several craft shows.
Many of the artists and craftsmen are recognized in their respective fields regionally and statewide and devote much
of their time to developing new artists.
-------
East Central Callixylon is
Largest Example Known to Exist
The giant remnant of Callixylon that stands at the end of East Main Street (directly in front of Science Hall) at
East Central University Campus is the "champ" of its kind.
In fact, it brought about naming of the type of petrified material in 1934, several years after it was discovered
in an abandoned hog lot of a former ranch in the isolated hills near Clarita, Oklahoma in Southeast Pontotoc
County.
Not true wood, this ancestor of modern trees was a fibrous material with nodes rather similar to those of a
cornstalk. Callixylon extends as far back as knowledge of seed plants. Being highly specialized, it is believed
to show evidence of being the termination of an evolutionary time.
It is the largest example of callixylon, or primitive wood, from the Devonian period known to exist, part of
what was once a gigantic semi-tree. Estimates of geologists set the period at about 350,000,000 years ago.
The largest piece is almost five feet in diameter and more than eight feet long. Other large pieces lay nearby
and there were hundreds of smaller chunks scattered about. In its petrified state, it weighs many tons.
Some idea of the original mighty plant may be had from the estimate that the massive relics put together at
East Central University were at least 16 feet above the ground.
Prior to the mid-1920's, only small twigs and fragments had been found anywhere in the world. The late John
Pitts, a local geologist, saw the big fragments in the hog lot and obtained possession of them.
In 1930, the late Dr. David White, of the Smithsonian Institute's Department of Paleobotany, was shown the
fragments, along with Churchill W. Thomas, then geology teacher at East Central. Dr. White recognized what
they were and began to try to raise money to move the fossil to Washington, D.C. However, he died before he
could raise the funds. John Fitts then donated the fossilized tree to East Central University in September,
1935.
Churchill W. Thomas supervised removal of the callixylon to Ada. With student help, he 'built' the pieces
into a section of semblance of the original plant.
Dedication of the geological oddity to Dr. White came in late 1935, with many eminent geologists present.
East Central President Linscheid formally accepted the donation to the University in May, 1936. Since its
dedication, the callixylon has been visited by thousands of persons, from this area and from far and wide.
It continues to be the champion, although a slightly smaller specimen was found near Wapanucka, Oklahoma
in 1958.
Its scientific name? An authority from Michigan University studied the fragments in 1934 and gave it the title
CALLIXYLON WHITENAUM honoring Dr. White.
-------
Byrd's Mill Spring
This tremendous spring was known to the Indians before Europeans came to the continent. The cold clear
water was preferred by the Choctaws and Chickasaws, and the area became a choice camp site. The spring
bears the name of one of the last and best known Chickasaw Chieftains, Governor William L. Byrd, who
operated a grist mill nearby.
The City of Ada was experiencing water troubles, both in supply and threatened contamination, before the
town was ten years old. The spring's tremendous supply and the possibility of gravity flow made Byrd's Mill
a natural place to turn. The Superintendent of Kansas City's waterworks, William G. Goodwin, was brought
here on October 5,1910, to test the spring. He found the spring flowing 18 million gallons per day.
Gabriel Brown, a restricted Indian, owned the spring site. Finally, on December 8,1910, a deed for the spring
site land was executed at Atoka. The City of Ada paid $7,500.00. The City didn't have the cash at the time, but
local citizens signed notes and the money was advanced by the First National Bank, M & P Bank, Ada Na-
tional Bank, Oklahoma State Bank, and the Oklahoma Portland Cement Company.
The spring is located at a point higher than Ada's reservoir, but the main is constructed on a principle that
permits it to carry the water up and down hill without pumping to the high reservoirs south of Ada. In the early
days, water was pumped at the spring.
Upon installation of its present water system, the City installed a 36" concrete conduit for about 3,700 feet
from the spring, then a 24" cast iron water line for a considerable distance and finally a 20" cast iron line
running into the City reservoir. The concrete reservoir is located south of the City of Ada and approximately
15 miles North of Byrd's Mill Spring. The underground storage capacity is 7,000,000 gallons; the 2,000,000
gallon reservoir was constructed in 1920 and improved in 1927 while the 5,000,000 gallon reservoir was
completed about 1937. The water is pumped from the two concrete reservoirs at the pump station located at
the reservoirs and the pumping equipment consists of four pumps. The pumping operation is entirely manual,
there being no automatic control. The water reaches the distribution system from the pump station through
two lines, one being 14 inches and the other 12 inches in diameter.
Byrd's Mill Spring has supplied the City of Ada with water since 1912. It was almost 20 years later that the
present 24" gravity flow line was constructed to bring in an estimated 7,500,000 gallons daily when flowing at
capacity.
The elevated water towers have a total capacity of 1,000,000 gallons storage; the capacity of one being 600,000
gallons and of the other 400,000 gallons.
The spring is located south of Ada in Section 34, Township 2 North, Range 6 East. It lies 2 miles due west of
Highway 99.
The tunnel section of the main water line from Byrd's Mill extends 1,240 feet with average depth of around 45
feet.
Total hardness of Byrd's Mill water is about 17 grains.
-------
CITY OF ADA
Arby's
Brawn's
Brawn's
Burger King
Dairy Lou
Dairy Queen
DJ.'s Drive-In
Hardee's
Kentucky Fried
Long John Silver's
McDonald's
McDonald's (Inside WalMart)
Schlotsky's
Sonic Drive-In
Subway
TacoBell
Taco Mayo
Taco Tico
Domino's Pizza
Campus Comer
McCortney's Coffee Shop
Dandee Donut Shop
Sunrise Donut Shop
Daylight Donut Shop
Daylight Donut Shop
Aldridge Coffee Shop
Bandana's
Blue Moon Cafe
Folger's
Gasoline Alley
Golden Corral
Italian Village
Jack Sprats's (Low Fat)
J.D.'s Cafe
Ken's Pizza
Liberty Cafe
Mazzio's Pizza
Oscar's Chinese
Pied Piper Pizza
Pizza Hut
Polo's Mexican Rest.
Village Restaurant
FAST FOOD
400 N. Mississippi
601 N. Mississippi
830 N. Country Club
609 N. Mississippi
903 W. Main
1001 E. Arlington
831 W. Main
200 S. Mississippi
501 N. Mississippi
930 Arlington
818 N. Country Club
1601 Lonnie Abbott Rd.
North Hills Centre
415 N. Mississippi
812 N. Country Club
7th & Mississippi
1400 Cradduck Rd.
300 S. Mississippi
908 Arlington Center
SNACK SHOPS
401 S. Mississippi
100 W. Main
109 S. Mississippi
414 E. Main
1101 Cradduck Rd.
1200 Arlington
RESTAURANTS
200 S. Broadway
1600 N. Mississippi
North Hills Centre
406 E. Main
1212 N. Broadway
1620 Arlington
620 S. Mississippi
123 W. Main
911 N, Broadway
711 E. 9th
215 W. Main
905 Lonnie Abbott Rd.
618 N. Mississippi
231 S. Mississippi
1239 N. Mississippi
219 W. Main
1121 Cradduck Rd.
436-0730
436-3055
436-1860
436-1138
332-6276
436-0166
436-5500
332-1600
332-7544
436-1131
436-4216
332-2212
332-6868
436-1484
332-2030
436-3930
332-1060
332-4556
436-5361
332-1078
436-2244
332-7700
332-7274
332-6745
332-8579
332-5818
332-2583
332-4477
332-9808
436-6699
332-5016
332-0130
436-4974
332-9750
332-1910
436-6161
436-3323
436-4838
436-3636
332-5662
332-2710
332-9841
-------
Saturated Flow
in Aquifers
-------
2. Saturated Flow in Aquifers
-------
w w w W w w
Precipitation
(on land)
Sublimation
Transpiration
Y Y Y Y
Precipitation
(on the ocean)
Groundwater table
ET - Evapotranspiration
E = Evaporation
SR = Surface Runoff
/ = Infiltration
Freshwater-salt water
interface
Groundwater flow
(saturated flow)
Evaporation
Figure 2-1. Schematic diagram of hydrologic cycle.
-------
Aquifer B
i Phreatic (Confined i Leaky i
i ^T< ^r*: >i<— Artesian
i Recharge' i
Ground gurface
Flowing
Well
Confined
Leaky
Perched Water
Piezometric surface (B)
. r -i>^
^J— Piezometric surface (C)
-Water Table
Sea
Impervious Stratum
Semipervious Stratum
Aquifer C
Figure 2-4. Types of aquifers.
-------
Ground Surface
Vadose Zone
Water Table
Aquifer
-------
Saturated Flow in Aquifers
Darcy's Law
Q = 0
x,
X
X2
Q
h2
The hydraulic head, h, is the sum of the elevation head and the pressure
head
h = z +
(i)
z is the elevation above the elevation datum [L]
p is the pressure [M/LT2]
p is the density of water [M/L3]
g is the acceleration of gravity [L/T2]
h = z +
(2)
is the pressure head
-------
(atm)
V
h
z = 0
Jl
Hydrostatic
Pressure
-------
Darcy's Law
Q .
A
- /T —
y
K(h2 - V
\ x - x
I ^i.o -A--I
} - Kdh
dx
(3)
where Q is the volumetric flow rate [L3/T]
A is the area of the column [L2]
q is the darcy flux (volume/area/time) [L/T]
K is the hydraulic conductivity [L/T]
h is the hydraulic head [L]
x is the distance coordinate [L]
Using cartesian coordinates
dx
dh
dz
where x,y and z are the coordinate directions [L]
QI is the darcy flux in the ith direction [L/T]
Kj is the hydraulic conductivity in the ith direction [L/T]
h is the hydraulic head [h]
-------
Flow in Vertical Columns
X2
JO.
h2
X.,
-------
Mass Conservation
(5)
where T| is the porosity.
V
d(pn)/dt
^(out)
r
or for steady state, constant density flow in a non-deforming medium
+JL(K—) +JL(K—) =0
»g**-v^S ' ^S * z^ '
—
\ •''•v / »g**-v^S ' z
ox ox oy y oy oz oz
(6)
-------
GROUNDWATER FLOW EQUATION
Rate of change
of mass of
fluid in
reference volume
per time unit
Rate of flow
of fluid mass
into reference
volume
Rate of flow
of fluid mass
out of reference
volume
Water
Mass
Balance
Groundwater
Flow
Equation
15. Formulation of the groundwater flow equation.
-------
Two Dimensional Planar Models
Fully Penetrating Wells
Successful for Water Supply Problems
Commonly Used in Modeling
-------
(o)
%
' T (ft /«) «* ; : i . : i : i ; . : ; . > : i
^T: ± :::;;:* 3c :
• 1.54 i:la±:i:J
o 0.386 E::3ii::^
B 0.0774 :::::::*:::
::: ii: i
:s;::;:::;;:|:::;i:^
(b)
T(ft2/s)
a 0.274
O 0 137
o 0 0685
ffl
tea
(c)
Musquoduboit River
Figure 8^7 Numerical simulation of aquifer performance at Musquoduboh
Harbour, Nova Scotia (after Finder and Bredehoeft 1968).
357
-------
Transient Two-Dimensional Flow
Confined Aquifer
--
dx\ dx
dy\ dy
w
= S
dh
dt
(7)
Where x and y are the coordinate directions [L]
t is the time [T]
h is the hydraulic head [L]
T is the transmissivity [L2/T]
w is the leakage rate per unit area [L/T]
S is the storativity or storage coefficient [*]
-------
Steady Two-Dimensional Flow
Confined Aquifer
oy\ dy I
= o (8)
With constant transmissivity (constant K and b)
dx2 8y
with no recharge:
dx2 dy2
LaPlace Equation
0 (10)
-------
Steady Two-Dimensional Flow
Unconfined Aquifer
dx\ dx } By\ dy
= 0
With constant conductivity
-------
Figure 3.1 Capture zones generated for a well discharge of 1500 ms/
-------
Introduction to
Modeling
-------
3. Introduction to Modeling
-------
Introduction to Modeling
What is a model ?
The mode) is a simplified version of a real-world ground-water system
that includes:
D Understanding of a site,
D Conceptualization,
D Site data,
D Computer code,
D Results,
D and Interpretation of the results
The model of the site allows our understanding of the site to be
quantified.
Simplifications
D geometry of the subsurface system
D treatment of heterogeneities
D porous vs fractured media
D isotropic vs anisotropic
D fluid properties
D flow regime
D source of contamination
D transformation of contaminants
-------
What normally limits application of models to the field ?
D Limited understanding of the site.
D Limited data.
D Limited capability of models.
Bear.J., Beljin, M.S., and R.R. Ross, 1992, Fundamentals of Ground-Water
Modeling, Ground Water Issue, US EPA, EPA/540/S-92/005.
despite limitations "...there is no alternative"
Uses of models:
D Quantify our understanding of multiple processes at a site.
D Evaluate our conceptualization.
D Design and/or evaluation of pump and treat systems
D Evaluation of slurry wall or hydraulic containment effectiveness
D Analysis of "no action" alternatives
D Evaluation of past migration patterns
D Assessment of contamination
3-3.
-------
The Modeling Process
(Tsang, 1991)
D Review of the Available data
D Development of a conceptual model
D Establishment of performance criteria
D Construction of calculational models
D Evaluate the calculated results
D If uncertainties are too great, repeat
-------
Formulation of Objectives
1
i
Ir
Review and Interrelation 1
of Available Data |
i
\
Model Conceptuallxatfon
>
f
I<
* '
j Code Selection |
Mora Data
Needed
Field Data Collection
>
Input Data
i t
i
Preparation
t
1 Calibration and Sensitivity An*
\
r
Predictive Runs
\
'
Uncertainty Analysis
(„
>
I
(Yd* • i
1
1
Improve
Conceptual
Modal
Flgursl. KoovlAppOcstlonProotss.
-------
Conceptualization
Understanding of the site must be translated into the model
From data at the site, the basic framework of the modeling effort is
developed
-------
Generally, the conceptual model incorporates information on
D structure geometric structure
physical boundaries
type of media (porous, fractured)
heterogeneity
dimensions
contaminated region
D processes physical and chemical phenomena
flow of water
aquifer(s)/vadose zone
type of aquifer
recharge
chemical transport
chemical transformations
immiscible flow
D boundary conditions conditions applied to the boundaries over
time
sources and sinks of water
water levels
D initial conditions conditions applied to the whole domain
at a specific time
water levels
source of contamination
initial contaminant levels
-------
Availability of data
Oversimplification may lead to a model that lacks required information
Undersimplication may result in a costly model or
lack of parameter data
-------
Conceptualization:
Approaches to Subsurface Contaminant Modeling
1.) Traditional Solute Transport Modeling /
By assuming that the solute does not affect the flow of the water,
the ground water flow and solute transport equations can be solved
separately.
Typically it is assumed that any changes in the head field come to
steady state much more rapidly than the time scale at which transport
is occurring.
Example Code: USGS MOC code for solute transport (2D),
USGS MODFLOW/MT3D (3D)
-------
2.) NAPL Modeling
Flow of the NAPL is concurrent with the dissolution of chemicals into the
aquifer.
Example Code: Hydrocarbon Spill Screening Model (HSSM)
Multiphase Organic Flow and Transport (MOFAT)
Flow of the NAPL occurs prior to the period of interest
Example Code: Regulatory Investigative Treatment Zone Model(RITZ)
NAPL Recovery from a water table aquifer.
Example Code: Two Layer Model for Free Product (TWOLAY)
ARMOS
-------
Dimensions
The subsurface is three dimensional, but is not necessarily modeled in three-
dimensions.
The dimensionality of the model depends on:
D the problem,
D the modeling objectives,
D the available data, and
D the available models.
-------
Recharge
Water Table
• Aquifer • .
- •.;.-.'.;; -.'•.;.;.•.'.• •.•.7°^.
////As ////A/////A//////
'/Cupper Confining nod-
*y»^^™^^^*^^^1 • * — •!«< iv^^"^— * •»• * ^^•^•••^••^•^••.^•^^^•^^l^^
Mlde
."!.''.*'.. • • , - . , '. *,*.'•*.*.' '* .'r* *ij' ' '.' ' '_' ' • -•*.'-*' L ' '*-' , ' '-''^"'i7 Vf •---*-.-*'*'-'.'''', fl ' 'l ' !_''.'.''.'*-1'** f *V
I
ow«r Conflnlno
yi^y../K /• / > / ^
I ' ^-r
Lower Confined J
LAqulfer
I" I
±
1
JL
_L
.^Conlufloa-f
I
I
I I I I _ _ I _ _J _ i_J _ ^41 I*
Bedrock — -*•— — — — — — — — — Mlllonla- —
Fig. 3. Schematic overview of groundwater resident times In large regional
systems (after van der Meljde 1988).
-------
Example 2 JP-4 motion and recovery at Hill Air Force Base Utah
D Data is sparse
D Hydraulic conductivities are uncertain and disputed
D The volume of the release is unknown
D The rate, timing and duration of the release(s) are unknown
Modeling in 3D vs modeling in 2D
-------
Boundary and Initial Conditions
Solutions of partial differential equations requires the specification of
boundary and initial conditions that "fix" the solution in space and time.
D Initial conditions are required for time-dependent problems (Initial Value
Problems, IVP)
D Boundary conditions are required for space-dependent problems (Boundary
Value Problems, BVP)
D Initial and boundary conditions are required for time- and space-dependent
problems (Initial-Boundary Value Problems IBVP)
-------
REGIONAL
Fig. 5. Conceptual diagram of the telescopic mesh refinement modeling approach (from Ward et aL,
1987).
3-/S*
-------
O.L. Franke, I.E. Reilly, and G.D. Bennett, Definition of Boundary and Initial Conditions
in the Analysis of Saturated Ground-Water Flow Systems~An Introduction,
Techniques of Water-Resources Investigations of the United States Geological
Survey, Book 3, Chapter B5.
Ground Water Flow Boundary Conditions
Boundary Condition
1 . Constant head
2. Specified head
3. Streamline or stream-
surface boundary
4. Specified flux
5. Head dependent flux
6. Free-surface
7. Seepage face
h = constant
(Supplies unlimited amount of
water)
h = f(x,y,t)
(Supplies unlimited amount of
water)
no flow, q = 0
q = f(x,y,t)
see below
Movable liquid boundary,
h = z, h = f(z)
Interface between
saturated media and the
atmosphere, h=z
Geological example
Aquifer out-cropping
beneath a lake
Stream whose stage is
independent of ground-
water seepage
Impermeable boundary
Areal recharge entering
the aquifer
Upper surface of an
aquifer overlain by a
semiconfining layer,
overlain by a surface
water body
The water table, sea
water/fresh water
interface
Seepage face on a
earthen dam
-------
Wctwttbto
Figure 1.—Row net within thre« different hydraulic settings:
A, through and beneath an earth dam underlain by sloping
bedrock; B, beneath a vertical impermeable wall; and C,
beneath an impermeable dam and a vertical impermeable •
wall.
Piezometer*
A
_
1 1
i
•
I
7 h
B
„
v
«P-7»8 !
*
i
» h.
* A ; * °
; : 1
! 1 i
i i ZB i
• : . 1
* i 1
C
7
Surface of fluid
subject to
atmospheric pressure^
I
: i
i
' i
i 1
(P/7)—
h_ •
.
! i
1 •
i ! .
f i
1 « x ! ! «2r '
T T ^ T T • C »
2«0
'•Body of stationary fluid
(Datum)
Figure 2.—Piezometers at different depths demonstrating
that the total head at all depths in a continuous body of
stationary fluid is constant
-------
Head dependent flux boundary
(see Figure 4 of Franke et al, Definition of Boundary and Initial Conditions...)
„ H - h
q - ~K'-
The flux through the confining layer (i.e., the boundary condition for the underlying
aquifer) is q.
Kc = the hydraulic conductivity of the confining layer
H = the head in the overlying aquifer
h = the head in the underlying aquifer
bc = thickness of the confining layer
When the underlaying aquifer dewaters the head for the flux calculation becomes the
elevation of the top of the underlying aquifer.
-------
Surface water body
Constant head on upper surfSce
of confining Ded
Leaky confining bed
Aquifer
!B^^iJ£&^£#J
Impermeable layer
Figure 4.—A leaky aquifer system.
3-11
-------
Specification of Boundary Conditions
® Determine the physical boundary conditions
i.e. Are there no flow boundaries ?
® Sensitivity analysis for boundary conditions
Do the stresses imposed on the model (i.e., pumping rates)
affect the domain near the boundaries ?
Ground-water divides are features of the head distribution that can be expected
to change with stresses.
The location of the boundary condition may depend on the stresses to the aquifer.
-------
Hypothetical Examples
1) Aquifer bounded by medium-sized stream in humid environment
Stress (pumpage)
Small
Large
Distance
far
near
Boundary Condition
specified head defined by the
streambed
if the stream dries up
specified head is not appropriate
2) Aquifer bounded by leaky confining beds above and below and by a freshwater-
saltwater interface
Small
Large
far
near
fixed stream-surface (no-flow)
fixed stream-surface not appropriate if
interface moves with stressing
3-jLf
-------
C StartJ
( Sttrt J
initial parameter estimates
model specification
initial parameter
estimates
•
E
Model
specification
TRIAL AND ERROR
AUTOMATIC
Fig. 10. History matching/calibaration using trial and error and .automatic
procedures (after Mercer and Faust 1981).
-------
Evaluation and Testing of Models
C. F. Tsang, 1991, The Modeling Process and Model Validation, Ground
Water, 29(6) 825-831.
Standard Guide for Developing and Evaluating Ground-Water Modeling
Codes, Second Draft ASTM Standard D-18.21.10, April 6, 1994.
N. Oreskes, K. Shrader-Frechette, K. Belitz, 1994, Verification, validation,
and confirmation of numerical models in the Earth sciences, Science, Vol.
263, 641-646.
Terminology-common difinitions
Verification Does the code solve the mathematical model correctly?
mass conservation
comparision with analytical solutions
Validation Does the model represent the physical system correctly?
comparision with field data
comparision of predictions with data
controversial-can a subsurface model be validated?
The meanings of these terms vary, so knowing the definition that a
person is using is critical for communication
-------
Other definitions from Tsang (1991)
IAEA, 1982
A conceptual model and the computer code derived from it are
validated when is confirmed that the conceptual model and the computer
code provide a good representation of the actual processes occurring in the
real system.
-------
Tsang (1991)
... a model, including the conceptualization and the code, can be said
to be validated with respect to
D a process
or
D a site-specific system
It is not logical to refer to a validated model in the generic sense.
-------
Mathematical Models
Types of Mathematical Models
Class
Analytical
Semi-Analytical
Numerical
Characteristics
Solution expressed
solely in terms of
mathematical
functions.
A mathematical
solution reduces the
complexity of
obtaining the result,
but requires some
numerical evaluation.
Solution determined
from "numerical"
approximations.
Example
Theis Equation
HSSM, WhAEM
TWOLAY,
MODFLOW
3-37
-------
An example Analytical Solution
The Theis (1935) Solution
For radial, two-dimensional flow in a confined aquifer:
a2 h + _! dji = S dh
dr2 r dr T dt
Initial Condition:
h(r,Q) = h0
Boundary Conditions:
A («f t) = h0
]h\
~i)
lim/r|M = _Q_ fort>0
r = radius [L]
h = head [L]
S = storativity [*]
T = transmissivity [L2/T]
-------
The Theis (1935) Solution
The solution for all radii and all times is given by:
where u = (r2 S)/(4 T t)
-------
General Features of Solution Types
Class
Advantages
Limitations
Analytical
1. One solution.
2. Easy evaluation.
3. Check on
numerical error in
numerical model
1. Severely restricted
geometry
(homogeneous
media, regular
domains, limited
boundary conditions).
2. Limited processes.
Semi-Analytical
1. Easy evaluation.
2. Less restrictive than
analytical models.
3. Low CPU.
1. Restricted
geometry.
2. Limited processes.
Numerical
1. "Any" processes.
2. "Any" geometry.
1. CPU intensive.
2. Data intensive.
3. Hard to learn
4. Numerical errors
5. Complex set up
6. Complex to
interpret results
-------
Numerical Models
Computers can: add, subtract, multiply, and divide (+r,x,-r)
Subsurface Flow and Transport Equations:
partial differential equations (PDE)
The process:
reduce the partial differential equations that govern subsurface flow
and transport problems
to approximations that can be evaluated on a computer.
Steps Common to All Models
D Divide the problem domain into a grid
D Approximate the PDE at each node in the grid
(generates a system of algebraic equations)
D Solve the algebraic equations
-------
Basic Finite Difference Approximations
AX AX
Taylor Series Approximations
2 x (31)
.. ;°L(Ax)3+. . .
b
df(x) . i 62f(x)
2 dX2 (32)
!^!£Uo! ...
6 ax3
First derivative
dx 2 Ax
Second derivative
f(x+Ax) - .CA3 (33)
6x2 (Ax)
{34)
-------
Two-dimensional steady flow
dx
dy'
0
(1)
h = hydraulic head
Kx = x component of hydraulic conductivity
Ky = y component of hydraulic conductivity
qx = vxn = -
dx
qy = vyn = -
(2)
-
qx = x component of the Darcy flux
qy = y component of the Darcy flux
vx = x component of the seepage velocity
v = y component of the seepage velocity
Approximation of the x term:
-h±)
dx xdx'
O)
By using the individual Kx and x values (i.e., Kxi+1/2j, xi+1ijl etc), the model
allows for K to vary at each node and the node spacing to vary across the grid.
-------
0
Or
V.
i-1/2j
x
. ij+1/2
k i r\
r ^-
ij i+1/2j '
. ij-1/2
O ij-1
3-3S
-------
Collecting the unknown h terms
a , dh)a
K~
h<. „•-
-------
Likewise for the y direction terms
-A
ay
2 (6)
.. ( - _^ - - - _^ - : - ] +
—
_
Simplifying
-------
(ny-1)nx+1
j
nx«1
1
-------
Combining the contributions of both the x and y terms gives an equation that relates the
head at the point xy to the heads at each of its four direct neighbors (xMj, xi+1j,xiH,xij+1)
This is the finite difference representation of the two-dimensional, steady flow equation
with non-uniform conductivities.
+ (8)
= 0
The equation is applied to each node in the grid in turn to generate a system of equations
that will be solved for the unknown heads.
Notice that the system of equations is linear
(no head value multiplies or divides another head term).
There has to be special treatment of the nodes on the edge of the domain.
-------
IJ
•+1J
-------
Constant Flux Boundary Condition
Discretized Darcy flux on the left hand edge of the domain
Applying the flux boundary condition where the flux is known, Fv
, , -•
Fv = -Kx *J _ *"J (10)
Solving for the hMj node gives
Since the hMj node doesn't exist on the left edge of the domain
(where the flux boundary condition is applied),
the boundary condition substitutes for the fictitious point hMj in the equation for the head
at point Xij
12 >
-------
Constant Head Boundary Condition
hn = hn (13)
(l)hi:J=hQ (14)
Either the known head value is used in the system of equations
or the point is considered known and the point dropped from the system of equations
-------
0
0
h
1 '
'nnode
B,
62
3-
-------
nx
-------
Solving systems of equations
Direct methods based on Gaussian Elimination
Gaussian Elimination with Partial Pivoting
L-U decomposition
Cholesky decomposition
LAPACK
E. Anderson et al., 1995, LAPACK User's Guide Second Edition,
Society for Industrial and Applied Mathematics,
Philadelphia
software available via internet:
send the message "send index from lapack" to netlib@ornl.gov
-------
Indirect methods based on Iteration
Jacobi
Gauss-Siedel
Conjugate Gradient
ITPACK
Kincaid, D.R., T.C. Oppe, J.R. Respess, D.M. Young, 1984,
ITPACKV 2C User's Guide, Center for Numerical Analysis,
The University of Texas at Austin, CNA 191.
T.C. Oppe, W.D. Joubert, and D.M. Young, 1988, NSPCG User's
Guide, Version 1.0, A Package for Solving Large Sparse
Linear Systems by Various Iterative Methods, Center for
Numerical Analysis, The University of Texas at Austin,
CNA 216.
software available via internet:
send message "send index from itpack" to netlib@ornl.gov
-------
References
Anderson E.et al., 1995, LAPACK User's Guide Second Edition, Society for Industrial and
Applied Mathematics, Philadelphia
Baehr, A.L. and M.Y. Corapciogliu, 1987, A compositional multiphase model for
groundwater contamination by petroleum products, 2. numerical solution, Water
Resources Research, 23(1), 201-214.
Bear, J., 1979, Hydraulics of Groundwater, Mcgraw-Hill.
Bear, J., Beljin, M.S., and R.R. Ross, 1992, Fundamentals of Ground-Water
Modeling, Ground Water Issue, US EPA, EPA/540/S-92/005.
Cline, P.V., J.J. Delfino, and P.S.C. Rao, 1991, Partitioning of aromatic constituents into
water from gasoline and other complex solvent mixtures, Environemental Science
Technology, 25(5), 914-920.
de Marsily, G., 1986, Quantitative Hydrogeology, Academic Press.
Demond, A.M., and P.V. Roberts, 1987, An examination of relative permeability
relations for two-phase flow in porous media, Water Resources Bulliten, 23(4), 617-
626.
Franke, O. L., T.E. Reilly, and G.D. Bennett, Definition of Boundary and Initial Conditions
in the Analysis of Saturated Ground-Water Flow Systems--An Introduction,
Techniques of Water-Resources Investigations of the United States Geological
Survey, Book 3, Chapter B5.
Freeze, R. A. and J.A. Cherry, 1979, Groundwater, Prentice-Hall.
Freyberg, D., 1988, An Exercise in Ground-Water Model Calibration and Prediction,
Ground Water, 26(3), 350-360.
Gelhar, L.W., C. Welty, and K.R. Rehfeldt, 1992, A critical review of data on field-scale
dispersion in aquifers, Water Resources Research, 28(7), 1955-1974.
Hampton D.R. and P.D.G. Miller, 1988, Laboratory investigation of the relationship
between actual and apparent product thickness in sands, in Proceedings of the
Conference on Petroleum Hydrocarbons and Organic Chemicals in Ground Water,
NWWA, November 9-11.
Hill, M.C. , 1992, A Computer Program (MODFLOWP) for Estimating Parameters of a
-------
Transient, Three-Dimensional, Ground-Water Flow Model Using Nonlinear
Regression, USGS, Open-File Report, 91-484.
Kincaid, D.R., T.C. Oppe, J.R. Respess, D.M. Young, 1984, ITPACKV 2C User's
Guide, Center for Numerical Analysis, The University of Texas at Austin, CNA
191.
Kemblowski, M.W., and C.Y. Chiang, 1990, Hydrocarbon thickness fluctuations in
monitoring wells, Groundwater, 28(2), 244-252.
Konikow, L.K., 1986, Predictive accuracy of a ground-water model-lessons from a
postaudit, Ground Water, 24(2), 173-184.
Konikow, L.F., and M. Person, 1985, Assessment of Long-Term Salinity Changes in
an Irrigated Stream-Aquifer System, Water Resources Research, 21(11), 1611-
1624.
Kueper, B.H., W. Abbott, and G. Farquhar, 1989, Expermimental observation of
multiphase flow in heterogeneous porous media, J. Contaminant Hydrology, 5, 83-
95.
Nofziger, D.L, K. Rajender, S.K. Nayudu and P.Y. Su, 1989, CHEMFLO One-
Dimensional Water and Chemical Movement in Unsaturated Soils, USEPA,
EPA/600/8-89/076.
Oppe T.C. , W.D. Joubert, and D.M. Young, 1988, NSPCG User's Guide, Version 1.0,
A Package for Solving Large Sparse Linear Systems by Various Iterative Methods,
Center for Numerical Analysis, The University of Texas at Austin, CNA 216.
Oreskes, N., K. Shrader-Frechette, K. Belitz, 1994, Verification, validation, and
confirmation of numerical models in the Earth sciences, Science, Vol. 263, 641-
646.
Schwille, F., 1971, Groundwater pollution by mineral oil products, Groundwater pollution
Symposium, IASH-AISH Publication No. 103, 1975, 226-240.
Schwille, F., 1988, Dense Chlorinated Solvents in Porous and Fractured Media-Models
Experiments, Lewis Publishers, 146 pp.
Standard Guide for Developing and Evaluating Ground-Water Modeling Codes,
Second Draft ASTM Standard D-18.21.10, April 6, 1994.
United States Environmental Protection Agency, 1992, Dense Nonaqueous Phase
Liquids-A Workshop Summary, EPA, EPA/600/R-92/030.
-------
United States Environmental Protection Agency, 1999, Estimating the Potential for
Occurrence of DNAPL at Superfund Sites, EPA, OSWER, 9355.4-07FS.
Tsang, C.F., 1991, The Modeling Process and Model Validation, Ground Water, 29(6)
825-831.
Wilson, J.L., S.H. Conrad, W.R. Mason, W. Peplinski, and E. Hagan, 1990, Laboratory
Investigation of Residual Liquid Organics from Spills, Leaks, and the Disposal of
Hazardous Wastes in Groundwater, EPA, Robert S. Kerr Environmental Research
Laboratory, EPA/600/6-90/004.
Zheng, C., 1990, MT3D A Modular Three-Dimensional Transport Model for Simulation
of Advection, Dispersion and Chemical Reactions of Contaminants in Groundwater
Systems, S.S. Papadopulos and Associates, Rockville, MD.
-------
References
-------
4. References
-------
United States
Environmental Protection
Agency
Office of
Research and
Development
Office of Solid Waste
and Emergency
Response
EPA/540/4-91-002
March 1991
&EPA Ground Water Issue
DENSE NONAQUEOUS PHASE LIQUIDS
Scott G. Huling* and James W. Weaver**
Background
The Regional Superfund Ground Water Forum is a group of
EPA professionals representing EPA's Regional Superfund
Offices, committed to the identification and the resolution of
ground water issues impacting the remediation of Superfund
sites. The Forum is supported by and advises the Superfund
Technical Support Project. Dense nonaqueous phase liquids is
an issue identified by the Forum as a concern of Superfund
decision-makers. For further information contact Scott G.
Huling (FTS:743-2313), Jim Weaver (FTS:743-2420), or
Randall R. Ross (FTS: 743-2355).
Introduction
Dense nonaqueous phase liquids (DNAPLs) are present at
numerous hazardous waste sites and are suspected to exist at
kmany more. Due to the numerous variables influencing DNAPL
Iransport and fate in the subsurface, and consequently, the
ensuing complexity, DNAPLs are largely undetected and vet
are likely to be a significant limiting factor in site remediation.
This issue paper is a literature evaluation focusing on DNAPLs
and provides an overview from a conceptual fate and transport
point of view of DNAPL phase distribution, monitoring, site
characterization, remediation, and modeling.
A nonaqueous phase liquid (NAPL) is a term used to describe
the physical and chemical differences between a hydrocarbon
liquid and water which result in a physical interface between a
mixture of the two liquids. The interface is a physical dividing
surface between the bulk phases of the two liquids, but
compounds found in the NAPL are not prevented from
solubilizing into the ground water. Immiscibility is typically
determined based on the visual observation of a physical
interface in a water- hydrocarbon mixture. There are numerous
methods, however, which are used to quantify the physical and
chemical properties of hydrocarbon liquids (31).
Nonaqueous phase liquids have typically been divided into two
general categories, dense and light. These terms describe the
specific gravity, or the weight of the nonaqueous phase liquid
relative to water. Correspondingly, the dense nonaqueous
phase liquids have a specific gravity greater than water, and
the light nonaqueous phase liquids (LNAPL) have a specific
gravity less than water.
Several of the most common compounds associated with
DNAPLs found at Superfund sites are included in Table 1.
These compounds are a partial list of a larger list identified by a
national screening of the most prevalent compounds found at
Superfund sites (65). The general chemical categories are
halogenated/non-halogenated semi-volatiles and halogenated
volatiles. These compounds are typically found in the following
wastes and waste-producing processes: solvents, wood
preserving wastes (creosote, pentachlorophenol), coal tars,
and pesticides. The most frequently cited group of these
contaminants to date are the chlorinated solvents.
DNAPL Transport and Fate - Conceptual Approach
Fate and transport of DNAPLs in the subsurface will be
presented from a conceptual point of view. Figures have been
selected for various spill scenarios which illustrate the general
behavior of DNAPL in the subsurface. Following the
conceptual approach, detailed information will be presented
explaining the specific mechanisms, processes, and variables
which influence DNAPL fate and transport. This includes
DNAPL characteristics, subsurface media characteristics, and
saturation dependent parameters.
Unsaturated Zone
Figure 1 indicates the general scenario of a release of DNAPL
into the soil which subsequently migrates vertically under both
the forces of gravity and soil capillarity. Soil capillarity is also
responsible for the lateral migration of DNAPL. A point is
reached at which the DNAPL no longer holds together as a
continuous phase, but rather is present as isolated residual
globules. The fraction of the hydrocarbon that is retained by
capillary forces in the porous media is referred to as'residual
* Environmental Engineer," Research Hydrologist, U.S.
Environmental Protection Agency, Roberts. Kerr Environmental
Research Laboratory, Ada, Oklahoma.
Superfund Technology Support Center for
Ground Water
Robert S. Kerr Environmental
Research Laboratory
Ada, Oklahoma
Technology fem&vatton Office
Office of S&fid Waste and Emergency
Response, US EPA, Washington, D.C,
•, ?N,D,
' iS- Printed on Recycled Paper
-------
Residual Saturation of
DNAPL in Soil From Spill
Plume of Dissolved
• Contaminants
Figure 1. The entire volume of DNAPL is exhausted by residual
saturation in the vadose zone prior to DNAPL reaching
the water table. Soluble phase compounds may be
leached from the DNAPL residual saturation and
contaminate the ground water.
saturation. In this spill scenario, the residual saturation in the
unsaturated zone exhausted the volume of DNAPL, preventing
it from reaching the water table. This figure also shows the
subsequent leaching (solubilization) of the DNAPL residual
saturation by water percolating through the unsaturated zone
(vadose zone). The leachate reaching the saturated zone
results in ground-water contamination by the soluble phase
components of the hydrocarbon. Additionally, the residual
| saturation at or near the water table is also subjected to
leaching from the rise and fall of the water table (seasonal, sea
level, etc.).
Increasing information is drawing attention to the importance of
the possibility that gaseous-phase vapors from NAPL in the
unsaturated zone are responsible for contaminating the ground
water and soil (18,47). It is reported that the greater "relative
vapor density" of gaseous vapors to air will be affected by
gravity and will tend to sink. In subsurface systems where
lateral spreading is not restricted, spreading of the vapors may
occur as indicated in Figure 2. The result is that a greater
amount of soils and ground water will be exposed to the
DNAPL vapors and may result in further contamination. The
extent of contamination will depend largely on the partitioning
of the DNAPL vapor phase between the aqueous and solid
phases.
DNAPL Phase Distribution - Four Phase System
It is apparent from Figures 1 and 2 that the DNAPL may be
present in the subsurface in various physical states or what is
referred to as phases. As illustrated in Figure 3, there are four
possible phases: gaseous, solid, water, and immiscible
hydrocarbon (DNAPL) in the unsaturated zone. Contaminants
associated with the release of DNAPL can, therefore, occur in
four phases described as follows:
1.
2.
Air phase - contaminants may be present as vapors;
Solid phase - contaminants may adsorb or partition onto
the soil or aquifer material;
3. Water phase - contaminants may dissolve into the water
according to their solubility; and
DNAPL Gaseous
Vapors
\
Residual
Saturation of
DNAPL in
Vadose Zone
Infiltration, Leaching and
Mobile DNAPL Vapors
Plume From DNAPL
Soil Vapor
Groundwater
Plume From DNAPL Flow
Residual Saturation
After, Waterloo Centre for Gioundwater ReMaioX 1989.
Figure 2. Migration of DNAPL vapors from the spill area and
subsequent contamination of the soils and ground
water.
4.
Immiscible phase - contaminants may be present as
dense nonaqueous phase liquids.
The four phase system is the most complex scenario because
there are four phases and the contaminant can partition
between any one or all four of these phases, as illustrated in
Figure 4. For example, TCE introduced into the subsurface as
a DNAPL may partition onto the soil phase, volatilize into the
soil gas, and solubilize into the water phase resulting in
contamination in all four phases. TCE can also partition
between the water and soil, water and air, and between the soil
and air. There are six pathways of phase distribution in the
unsaturated zone. The distribution of a contaminant between
these phases can be represented by empirical relationships
referred to as partition coefficients. The partition coefficients, or
the distribution of the DNAPL between the four phases, is
highly site-specific and highly dependent on the characteristics
of both the soil/aquifer matrix and the DNAPL. Therefore, the
distribution between phases may change with time and/or
location at the same site and during different stages of site
remediation.
Solid
Water
DNAPL
Figure 3. A DNAPL contaminated unsaturated zone has four
physical states or phases (air, solid, water, Immiscible).
The contaminant may be present in any one, or all four
phases.
-------
Four Phase System
Partition Coefficients
K = Soil-water partition coefficient
KH = Henry's Constant
K' = DNAPL-water partition coefficient
K" = DNAPL-air partition coefficient
Water
after DiGiulio, 1990(9)
Figure 4. Distribution of DNAPL between the four phases found
in the vadose zone.
Figure 5. DNAPL spilled into fractured rock systems may
follow a complex distribution of the preferential pathways.
The concept of phase distribution is critical in decision-
making. Understanding the phase distribution of a DNAPL
introduced into the subsurface provides significant insight in
determining which tools are viable options with respect to site
characterization and remediation.
DNAPL represented by residual saturation in the four phase
diagram is largely immobile under the usual subsurface
pressure conditions and can migrate further only: 1) in water
according to its solubility; or 2) in the gas phase of the
unsaturated zone (47). DNAPL components adsorbed onto the
soil are also considered immobile. The mobile phases are,
therefore, the soluble and volatile components of the DNAPL
in the water and air, respectively.
The pore space in the unsaturated zone may be filled with one
or all three fluid phases (gaseous, aqueous, immiscible). The
presence of DNAPL as a continuous immiscible phase has the
potential to be mobile. The mobility of DNAPL in the
subsurface must be evaluated on a case by case basis. The
maximum number of potentially mobile fluid phases is three.
Simultaneous flow of the three phases (air, water, and
immiscible) is considerably more complicated than two-phase
flow (46). The mobility of three phase flow in a four-phase
system is complex, poorly understood, and is beyond the
scope of this DNAPL overview. The relative mobility of the two
phases, water and DNAPL, in a three-phase system is
presented below in the section entitled "Relative Permeability."
Generally, rock aquifers contain a myriad of cracks (fractures)
of various lengths, widths, and apertures (32). Fractured rock
systems have been described as rock blocks bounded by
discrete discontinuities comprised of fractures, joints, and
shear zones which may be open, mineral-filled, deformed, or
any combination thereof (61). The unsaturated zone overlying
these fractured rock systems also contain the myriad of
preferential pathways. DNAPL introduced into such formations
(Figure 5) follow complex pathways due to the heterogeneous
distribution of the cracks, conduits, and fractures', i.e.,
preferential pathways. Transport of DNAPL may follow non-
Darcian flow in the open fractures and/or Darcian flow in the
'porous media filled fractures. Relatively small volumes of
NAPL may move deep, quickly into the rock because the
retention capacity offered by the dead-end fractures and the
immobile fragments and globules in the larger fractures is so
small (32). Currently, the capability to collect the detailed
information for a complete description of a contaminated
fractured rock system is regarded as neither technically
possible nor economically feasible (61).
Low permeability stratigraphic units such as high clay content
formations may also contain a heterogeneous distribution of
preferential pathways. As illustrated in Figure 6, DNAPL
transport in these preferential pathways is correspondingly
complex. Typically, it is assumed that high clay content
formations are impervious to DNAPL. However, as DNAPL
spreads out on low permeable formations it tends to seek out
zones of higher permeability. As a result, preferential pathways
allow the DNAPL to migrate further into the low permeable
formation, or through it to underlying stratigraphic units. It is
apparent from Figures 5 and 6 that the complexity of DNAPL
transport may be significant prior to reaching the water table.
Saturated Zone
The second general scenario is one in which the volume of
DNAPL is sufficient to overcome the fraction depleted by the
residual saturation in the vadose zone, as illustrated in Figure
7. Consequently, the DNAPL reaches the water table and
contaminates the ground water directly. The specific gravity of
DNAPL is greater than water, therefore, the DNAPL migrates
into the saturated zone. In this scenario, DNAPL continues the
vertical migration through the saturated zone until the volume
is eventually exhausted by the residual saturation process or
until it is intercepted by a low permeable formation where it
begins to migrate laterally.
DNAPL Phase Distribution - Three Phase System
Due to the lack of the gaseous phase, the saturated zone
containing DNAPL is considered a three-phase system
consisting of the solid, water, and immiscible hydrocarbon
(Figure 8). Contaminant distribution in the three-phase systenj
is less complex than the four-phase system. Again, this is
highly dependent on the characteristics of both the aquifer
-------
Water
DNAPL
Solid
Figure 6. DNAPL spilled into a low permeable formation may
follow a complex distribution of preferential pathways.
The volume of DNAPL is exhausted in the vadose zone
prior to reaching the water table.
matrix and the DNAPL. Figure 9 indicates the three phases
and the transfer of the mass of contaminant between the
phases. In this scenario, there are only three pathways of
phase distribution in the saturated zone.
Note that when the DNAPL is represented by residual
saturation in the three-phase system, the mobile phase of the
contaminant is the water soluble components of the DNAPL
and the immobile phases are the residual saturation and the
adsorbed components of the DNAPL associated with the
aquifer material. The main mobilization mechanism of the
residual saturation is removal of soluble phase components
into the ground water. When the DNAPL is present as a
continuous immiscible phase, it too is considered one of the
mobile phases of the contaminant. While the continuous phase
DNAPL has the potential to be mobile, immobile continuous
phase DNAPL may also exist in the subsurface. Although the
saturated zone is considered a three-phase system, gaseous
vapors from DNAPL in the unsaturated zone does have the
Residual
Saturation of
DNAPL in Soil
From Spill
Plume of Dissolved
• Contaminants •
^ Groundwater
Flow
Residual
Saturation in Saturated Zone
After. Waterloo Centre lor Groundwater Research. 1989.
| Figure 7. The volume of DNAPL is sufficient to overcome the
residual saturation in the vadose zone and
consequently penetrates the water table.
Figure 8. A DNAPL contaminated saturated zone has three
phases (solid, water, Immiscible). The contaminant
may be present in any one, or all three phases.
potential to affect ground-water quality, as was indicated earlier
in Figure 2.
Assuming the residual saturation in the saturated zone does
not deplete the entire volume of the DNAPL, the DNAPL will
continue migrating vertically until it encounters a zone or
stratigraphic unit of lower permeability. Upon reaching the zone
of lower permeability, the DNAPL will begin to migrate laterally.
The hydraulic conductivity in the vertical direction is typically
less than in the horizontal direction. It is not uncommon to find
vertical conductivity that is one-fifth or one-tenth the horizontal
value (4). It is expected that DNAPL spilled into the subsurface
will have a significant potential to migrate laterally. If the lower
permeable boundary is "bowl shaped", the DNAPL will pond as
a reservoir (refer to Figure 10). As illustrated in Figure 11, it is
not uncommon to observe a perched DNAPL reservoir where a
discontinuous impermeable layer; i.e., silt or clay lens,
intercepts the vertical migration of DNAPL When a sufficient
volume of DNAPL has been released and multiple
discontinuous impermeable layers exist, the DNAPL may be
present in several perched reservoirs as well as a deep
Three Phase System
Water
K1 = DNAPL-water partition coefficient
K = Soil-water partition coefficient
Figure 9. Distribution of DNAPL between the three phases found
in the saturated zone.
-------
Figure 10. Migration of DNAPL through the vadose zone to an
impermeable boundary.
Figure 12. Perched and deep DNAPL reservoirs.
reservoir (refer to Figure 12). Lateral migration continues until
either the residual saturation depletes the DNAPL or an
impermeable depression immobilizes the DNAPL in a reservoir
type scenario. Soluble-phase components of the DNAPL will
partition into the ground water from both the residual saturation
or DNAPL pools. The migration of DNAPL vertically through
the aquifer results in the release of soluble-phase components
of the DNAPL across the entire thickness of the aquifer. Note,
that ground water becomes contaminated as it flows through,
and around, the DNAPL contaminated zone.
As indicated earlier, DNAPL will migrate laterally upon
reaching a stratigraphic unit of lower permeability. Transport of
DNAPL will therefore be largely dependent on the gradient of
the stratigraphy. Occasionally, the directional gradient of an
impermeable stratigraphic unit may be different than the
direction of ground-water flow as illustrated in Figure 13a. This
may result in the migration of the continuous phase DNAPL in
a direction different from the ground-water flow. Nonhorizontal
stratigraphic units with varying hydraulic conductivity may also
convey DNAPL in a different direction than ground-water flow,
and at different rates (refer to Figure 13b). Determination of the
direction of impermeable stratigraphic units will therefore
provide useful information concerning the direction of DNAPL
transport.
DNAPL
Pool
DNAPL Residual
Saturation -,
Dissolved
Contaminants
Low Permeable
Stratigraphic Unit
CLAY
Groundwater
Flow
Similar to the unsaturated zone, the saturated zone also
contains a complex distribution of preferential pathways from
cracks, fractures, joints, etc. DNAPL introduced into such
formations correspondingly follow the complex network of
pathways through an otherwise relatively impermeable rock
material. Other pathways which may behave as vertical
conduits for DNAPL include root holes, stratigraphic windows,
disposal wells, unsealed geotechnical boreholes, improperly
sealed hydrogeological investigation sampling holes and
monitoring wells, and old uncased/unsealed water supply wells
(72). Transport of the DNAPL may migrate very rapidly in these
open conduits or follow Darcian flow in the surrounding porous
media or porous media filled fractures. A relatively small
volume of DNAPL can move deep into a fractured system due
to the low retentive capacity of the fractured system.
Consequently, fractured clay or rock stratigraphic units, which
are often considered lower DNAPL boundary conditions, may
have preferential pathways leading to lower formations, as
depicted in Figure 14. Careful inspection of soil cores at one
Superfund site indicated that DNAPL flow mainly occurred
through preferential pathways and was not uniformly
distributed throughout the soil mass (8). Due to the complex
Figure 11. Perched DNAPL reservoir.
Figure 13a. Stratigraphic gradient different from ground water
gradient results in a different direction of flow of the
ground water and continuous phase DNAPL.
-------
Where Kx2>Ki(,>Kx3
Kx = Horizontal Hydraulic Conductivity
Groundwater
Flow
Figure 13b. Non-horizontal stratigraphic units with variable
hydraulic conductivity may convey DNAPL in a
different direction than the ground water flow
direction.
-distribution of preferential pathways, characterization of the
volume distribution of the DNAPL is difficult.
Important DNAPL Transport and Fate Parameters
There are several characteristics associated with both the
subsurface media and the DNAPL which largely determine the
fate and transport of the DNAPL. A brief discussion of these
parameters is included to help identify the specific details of
DNAPL transport mechanisms. Several of the distinctive
DNAPL phenomena observed on the field-scale relates back to
phenomena at the pore-scale. Therefore, it is important to
understand the principles from the pore-scale level to develop
an understanding of field-scale observations, which is the scale
at which much of the Superfund work occurs. A more
complete and comprehensive review of these parameters is
available (2,36,71).
DNAPL Characteristics
Density
Fluid density is defined as the mass of fluid per unit volume,
i.e. g/cm3. Density of an immiscible hydrocarbon fluid is the
parameter which delineates LNAPL's from DNAPL's. The
property varies not only with molecular weight but also
molecular interaction and structure. In general, the density
varies with temperature and pressure (2). Equivalent methods
of expressing density are specific weight and specific gravity.
The specific weight is defined as the weight of fluid per unit
volume, i.e. Ib/ft3. The specific gravity (S.G.) or the relative
density of a fluid is defined as the ratio of the weight of a given
volume of substance at a specified temperature to the weight
of the same volume of water at a given temperature (31). The
S.G. is a relative indicator which ultimately determines whether
the fluid will float (S.G.< 1.0) on, or penetrate into (S.G.>1.0)
the water table. Table 1 contains a list of compounds with a
density greater than one that are considered DNAPL's. Note,
however, that while the specific gravity of pentachlorophenol
and the non-halogenated semi-volatiles is greater than 1.00,
these compounds are a solid at room temperature and would
not be expected to be found as an immiscible phase liquid at
wood preserving sites but are commonly found as contami-
nants. Pentachlorophenol is commonly used as a wood
preservant and is typically dissolved (4-7%) in No. 2 or 3 fuel
oil.
Viscosity
The viscosity of a fluid is a measure of its resistance to flow.
Molecular cohesion is the main cause of viscosity. As the
temperature increases in a liquid, the cohesive forces
decrease and the absolute viscosity decreases. The lower the
viscosity, the more readily a fluid will penetrate a porous
media. The hydraulic conductivity of porous media is a function
of both the density and viscosity of the fluid as indicated in
equation [1]. It is apparent from this equation that fluids with
either a viscosity less than water or fluids with a density greater
than water have the potential to be more mobile in the
subsurface, than water.
7s ///////
Impermeable Boundary
K = k p g where,
Figure 14. DNAPL transport in fracture and porous media
stratigraphic units.
K = hydraulic conductivity [1]
k = intrinsic permeability
p = fluid mass density
g = gravity
u. = dynamic (absolute) viscosity
Results from laboratory experiments indicated that several
chlorinated hydrocarbons which have low viscosity (methylene
chloride, perchloroethylene, 1,1,1-TCA, TCE) will infiltrate into
soil notably faster than will water (47). The relative value of
NAPL viscosity and density, to water, indicates how fast it will
flow in porous media (100% saturated) with respect to water.
For example, several low viscosity chlorinated hydrocarbons
(TCE, tetrachloroethylene, 1,1,1-TCA, Methylene Chloride,
Chloroform, Carbon Tetrachloride, refer to Table 1) will flow
1 .5-3.0 times as fast as water and higher viscosity compounds
including light heating oil, diesel fuel, jet fuel, and crude oil (i.e.
LNAPL's) will flow 2-10 times slower than water (45). Both coal
tar and creosote typically have a specific gravity greater than
one and a viscosity greater than water. It is interesting to note
-------
that the viscosity of NAPL may change with time (36). As fresh
crude oils lose the lighter volatile components from
evaporation, the oils become more viscous as the heavier
components compose a larger fraction of the oily mixture
resulting in an increase in viscosity.
Solubility
When an organic chemical is in physical contact with water, the
organic chemical will partition into the aqueous phase. The
equilibrium concentration of the organic chemical in the
aqueous phase is referred to as its solubility. Table 1 presents
the solubility of several of the most commonly found DNAPL's
at EPA Superfund Sites. The solubility of organic compounds
varies considerably from the infinitely miscible compounds,
including alcohols (ethanol, methanol) to extremely low
solubility compounds such as polynuclear aromatic
compounds.
Numerous variables influence the solubility of organic
compounds. The pH may affect the solubility of some organic
compounds. Organic acids may be expected to increase in
solubility with increasing pH, while organic bases may act in
the opposite way (31). For example, pentachlorophenol is an
acid which is ionized at higher pH's. In the ionized form,
pentachlorophenol would be more soluble in water (59).
Solubility in water is a function of the temperature, but the
strength and direction of this function varies. The presence of
dissolved salts or minerals in water leads to moderate
decreases in solubility (31). In a mixed solvent system,
consisting of water and one or more water-miscible
compounds, as the fraction of the cosolvent in the mixture
increases, the solubility of the organic chemical increases
exponentially (12). In general, the greater the molecular weight
and structural complexity of the organic compound, the lower
the solubility.
Organic compounds are only rarely found in ground water at
concentrations approaching their solubility limits, even when
organic liquid phases are known or suspected to be present.
The observed concentrations are usually more than a factor of
10 lower than the solubility presumably due to diffusional
limitations of dissolution and the dilution of the dissolved
organic contaminants by dispersion (74). This has also been
attributed to: reduced solubility due to the presence of other
soluble compounds, the heterogeneous distribution of DNAPL
in the subsurface, and dilution from monitoring wells with long
intake lengths (10). Detection of DNAPL components in the
subsurface below the solubility should clearly not be
interpreted as a negative indicator for the presence of DNAPL.
In a DNAPL spill scenario where the DNAPL or its vapors are
in contact with the ground water, the concentration of the
soluble phase components may range from non-detectable up
to the solubility of the compound. The rate of dissolution has
been expressed as a function of the properties of the DNAPL
components (solubility), ground water flow conditions,
differential between the actual and solubility concentration, and
the contact area between the DNAPL and the ground water
(10). The contact area is expected to be heterogeneous and
difficult to quantify. Additionally, as the time of contact
increases between the DNAPL and the water, the
concentration in the aqueous phase increases.
Vapor Pressure
The vapor pressure is that characteristic of the organic
chemical which determines how readily vapors volatilize or
evaporate from the pure phase liquid. Specifically, the partial
pressure exerted at the surface by these free molecules is
known as the vapor pressure (30). Molecular activity in a liquid
tends to free some surface molecules and this tendency
towards vaporization is mainly dependent on temperature. The
vapor pressure of DNAPL's can actually be greater than the
vapor pressure of volatile organic compounds. For example, at
20 C, the ratio of the vapor pressures of TCE and benzene is
1.4(1).
Volatility
The volatility of a compound is a measure of the transfer of the
compound from the aqueous phase to the gaseous phase. The
transfer process from the water to the atmosphere is
dependent on the chemical and physical properties of the
compound, the presence of other compounds, and the physical
properties (velocity, turbulence, depth) of the water body and
atmosphere above it. The factors that control volatilization are
the solubility, molecular weight, vapor pressure, and the nature
of the air-water interface through which it must pass (31). The
Henry's constant is a valuable parameter which can be used to
help evaluate the propensity of an organic compound to
volatilize from the water. The Henry's law constant is defined
as the vapor pressure divided by the aqueous solubility.
Therefore, the greater the Henry's law constant, the greater the
tendency to volatilize from the aqueous phase, refer to Table 1.
Interfacial Tension
The unique behavior of DNAPLs in porous media is largely
attributed to the interfacial tension which exists between
DNAPL and water, and between DNAPL and air. These
interfacial tensions, result in distinct interfaces between these
fluids at the pore-scale. When two immiscible liquids are in
contact, there is an interfacial energy which exists between the
fluids resulting in a physical interface. The interfacial energy
arises from the difference between the inward attraction of the
molecules in the interior of each phase and those at the
surface of contact (2). The greater the interfacial tension
between two immiscible liquids; the less likely emulsions will
form; emulsions will be more stable if formed, and the better
the phase separation after mixing. The magnitude of the
interfacial tension is less than the larger of the surface tension
values for the pure liquids, because the mutual attraction of
unlike molecules at the interface reduces the large imbalance
of forces (31). Interfacial tension decreases with increasing
temperature, and may be affected by pH, surfactants, and
gases in solution (36). When this force is encountered between
a liquid and a gaseous phase, the same force is called the
surface tension (66).
The displacement of water by DNAPL and the displacement of
DNAPL by water in porous media often involves a phenomena
referred to as immiscible fingering. The lower the interfacial
tension between immiscible fluids, the greater the instability of
the waterDNAPL interface and thus the greater the immiscible
fingering (27). The distribution of the fingering effects in porou^
media has been reported to be a function of the density,
viscosity, surface tension (27) and the displacement velocity
-------
(13) of the fluids involved as well as the porous media
heterogeneity (28).
.Wetlability
Wettability refers to the relative affinity of the soil for the
various fluids - water, air, and the organic phase. On a solid
surface, exposed to two different fluids, the wettability can be
inferred from the contact angle (66), also referred to as the
wetting angle, refer to Figure 15. In general, if the wetting angle
is less than 90 degrees, the fluid is said to be the wetting fluid.
In this scenario, water will preferentially occupy the smaller
pores and will be found on solid surfaces (14). When the
wetting angle is near 90 degrees, neither fluid is preferentially
attracted to the solid surfaces. If the wetting angle is greater
than 90 degrees, the DNAPL is said to be the wetting fluid. The
wetting angle is an indicator used to determine whether the
porous material will be preferentially wetted by either the
hydrocarbon or the aqueous phase (71). Wettability, therefore,
describes the preferential spreading of one fluid over solid
surfaces in a two-fluid system. The wetting angle, which is a
measure of wettability, is a solid-liquid interaction and can
actually be defined in terms of interfacial tensions (71).
Several methods have been developed to measure the wetting
angle (36,71). In most natural systems, water is the wetting
fluid, and the immiscible fluid is the non-wetting fluid. Coal tar
may be the exception (i.e. contact angle greater than 90
degrees), which is mainly attributed to the presence of
surfactants (70). The wetting fluid will tend to coat the surface
of grains and occupy smaller spaces (i.e. pore throats) in
porous media, the non-wetting fluid will tend to be restricted to
the largest openings (47).
The wetting angle depends on the character of the solid
surface on which the test is conducted. The test is conducted
on flat plates composed of minerals which are believed
representative of the media, or on glass. Contact angle
measurements for crude oil indicates that the wetting angles
vary widely depending on the mineral surface (53). Soil and
aquifer material are not composed of homogeneous mineral
composition nor flat surfaces. The measured wetting angle can
only be viewed as a qualitative indicator of wetting behavior.
The reader is recommended to refer to reference No. 31 for
review of the basic principles and for various techniques to
measure the following DNAPL parameters: density, viscosity,
interfacial tension, solubility, vapor pressure, and volatility.
e»90°
Wetting Fluid: DNAPL
Water
« < 90 °
Wetting Fluid: Water
Water
Fluid Relationships:
System Wetting Fluid
airwater water
air: DNAPL DNAPL
waterDNAPL water
air:DNAPL:water water>organic>air(l
(1) Wetting fluid order
Non-Welting Fluid
air
air
DNAPL
Centre for
-ch. 1989.
C Alter, Waterloo Cen
Groundwater Resear
Figure 15. Wetting angle and typical wetting fluid relationships.
Subsurface Media Characteristics
Capillary Force/Pressure
Capillary pressure is important in DNAPL transport because it
largely determines the magnitude of the residual saturation that
is left behind after a spill incident. The greater the capillary
pressure, the greater the potential for residual saturation. In
general, the capillary force increases in the following order;
sand, silt, clay. Correspondingly, the residual saturation
increases in the same order. Capillary pressure is a measure
of the tendency of a porous medium to suck in the wetting fluid
phase or to repel the nonwetting phase (2). Capillary forces are
closely related to the wettability of the porous media. The
preferential attraction of the wetting fluid to the solid surfaces
cause that fluid to be drawn into the porous media. Capillary
forces are due to both adhesion forces (the attractive force of
liquid for the solids on the walls of the channels through which
it moves) and cohesion forces (the attraction forces between
the molecules of the liquid) (32). The capillary pressure
depends on the geometry of the void space, the nature of
solids and liquids, the degree of saturation (2) and in general,
in-creases with a decrease in the wetting angle and in pore
size, and with an increase in the interfacial tension (71). All
pores have some value of capillary pressure. Before a
nonwetting fluid can enter porous media, the capillary pressure
of the largest pores (smallest capillary pressure) must be
exceeded. This minimum capillary pressure is called the entry
pressure.
In the unsaturated zone, pore space may be occupied by
water, air (vapors), or immiscible hydrocarbon. In this scenario,
capillary pressure retains the water (wetting phase) mainly in
the smaller pores where the capillary pressure is greatest. This
restricts the migration of the DNAPL (non-wetting phase)
through the larger pores unoccupied by water. Typically,
DNAPL does not displace the pore water from the smaller
pores. It is interesting to note that the migration of DNAPL
through fine material (high capillary pressure) will be impeded
upon reaching coarser material (low capillary pressure).
The capillary fringe will obstruct the entry of the DNAPL into
the saturated zone. When a sufficient volume of DNAPL has
been released and the "DNAPL pressure head" exceeds the
water capillary pressure at the capillary fringe (entry pressure),
the DNAPL will penetrate the water table. This is why DNAPL
is sometimes observed to temporarily flatten out on top of the
water table. Similarly, laboratory experiments have been
conducted in which DNAPL (tetrachloroethylene) infiltrating
through porous media was found to flow laterally and cascade
off lenses too fine to penetrate (28), (refer to Figure 11). This
was attributed to the inability of the DNAPL to overcome the
high capillary pressure associated with the lenses. Logically,
when "DNAPL pressure head" exceeds the capillary pressure,
the DNAPL will penetrate into the smaller pores. These
laboratory experiments are important because they illustrate
that small differences in the capillary characteristics of porous
media can induce significant lateral flow of non-wetting fluids.
A comprehensive investigation of capillary trapping and
multiphase flow of organic liquids in unconsolidated porous
media revealed many intricacies of this process in the vadose
and saturated zone (66). An important note is that while
capillary pressure is rarely measured at hazardous waste sites,
. 9 .
-------
the soil texture (sand, silt, clay) is usually recorded during
drilling operations and soil surveys. This information, along with
soil core analyses will help to delineate the stratigraphy and
the volume distribution of NAPL.
Pore Size Distribution/Initial Moisture Content
In natural porous media, the geometry of the pore space is
extremely irregular and complex (2). The heterogeneity of the
subsurface environment i.e. the variability of the pore size
distribution, directly affects the distribution of the capillary
pressures along the interfaces between the aqueous and
immiscible phases (50). In saturated column experiments, it
was observed that NAPL preferentially traveled through strings
of macropores, almost completely by-passing the water filled
micropores (66). In the same study, a heterogeneous
distribution of coarse and fine porous material was simulated.
Most of the incoming organic liquid preferentially traveled
through the coarse lens material.
In short term column drainage experiments, results indicated
that the particle grain size is of primary importance in
controlling the residual saturation of a gasoline hydrocarbon
(19). Fine and coarse sands (dry) were found to have 55%
and 14% residual saturation, respectively. The finer the sand,
the greater the residual saturation. During these experiments,
the residual saturation was reduced 20-30% in a medium
sand and 60% in a fine sand when the sands were initially wet.
Soil pore water held tightly by capillary forces in the small
pores will limit the NAPL to the larger pores, and thus, result in
lower residual saturation. In a similar laboratory (unsaturated).
column study, the smaller the grain size used in the
experiment, the greater the residual saturation of the NAPL
(74). The residual saturation in the saturated column
experiments was found to be greater than the unsaturated
columns and was independent of the particle size distri-
bution.
These observations follow traditional capillary force theory.
Residual saturation resulting from a DNAPL spill in the
unsaturated zone is highly dependent on the antecedent
moisture content in the porous media. When the moisture
content is low, the strong capillary forces in the smaller pores
will tenaciously draw in and hold the DNAPL. When the
moisture content is high, the capillary forces in the smaller
pores will retain the soil pore water, and DNAPL residual
saturation will mainly occur in the larger pores. Therefore,
greater residual saturation can be expected in dryer soils.
Correspondingly, NAPL will migrate further in a wetter soil,
and displacement of NAPL from small pores is expected to
be more difficult than from large pores.
Stratigraphic Gradient
DNAPL migrating vertically will likely encounter a zone or
stratigraphic unit of lower vertical permeability. A reduction in
the vertical permeability of the porous media will induce lateral
flow of the DNAPL The gradient of the lower permeable
stratigraphic unit will largely determine the direction in which
the DNAPL will flow. This is applicable to both the saturated
and unsaturated zones. As depicted in Figures 13a and 13b,
the lateral direction of DNAPL flow may be in a different
direction than ground-water flow.
Ground Water Flow Velocity
The ground water flow velocity is a dynamic stress parameter
which tends to mobilize the hydrocarbon (39). As the ground
water velocity increases, the dynamic pressure and viscous
forces increase. Mobilization of DNAPL occurs when the
viscous forces of the ground water acting on the DNAPL,
exceeds the porous media capillary forces retaining the
DNAPL.
Saturation Dependent Functions
Residual Saturation
Residual saturation is defined as the volume of hydrocarbon
trapped in the pores relative to the total volume of pores (38)
and therefore is measured as such (74). Residual saturation
has also been described as the saturation at which NAPL
becomes discontinuous and is immobilized by capillary forces
(36). The values of residual saturation vary from as low as 0.75
-1.25% for light oil in highly permeable media to as much as
20% for heavy oil (50). Residual saturation values have also
been reported to range from 10% to 50% of the total pore
space (39,74). Other researchers reported that residual
saturation values appear to be relatively insensitive to fluid
properties and very sensitive to soil properties (and
heterogeneities) (66). Laboratory studies conducted to predict
the residual saturation in soils with similar texture and grain
size distribution yielded significantly different values. It was
concluded that minor amounts of clay or silt in a soil may play
a significant role in the observed values.
In the unsaturated zone during low moisture conditions, the
DNAPL residual saturation will wet the grains in a pendular
state (a ring of liquid wrapped around the contact point of a
pair of adjacent grains). During high moisture conditions, the
wetting fluid, which is typically water, will preferentially occupy
the pendular area of adjacent grains and the hydrocarbon will
occupy other available pore space, possibly as isolated
droplets. In the saturated zone, the DNAPL residual saturation
will be present as isolated drops in the open pores (47).
Furthermore, results of laboratory experimentation indicated
that residual saturation increased with decreasing hydraulic
conductivity in both the saturated and unsaturated zones and
that the residual saturation is greatest in the saturated zone.
Laboratory experiments indicated that vadose zone residual
saturation was roughly one third of the residual saturation in
the saturated zone (66). The increase in residual saturation in
the saturated zone is due to the following: [1] the fluid density
ratio (DNAPL:air versus DNAPL:water above and below the
water table, respectively) favors greater drainage in the vadose
zone; [2] as the non-wetting fluid in most saturated media,
NAPL is trapped in the larger pores; and, [3] as the wetting
fluid in the vadose zone, NAPL tends to spread into adjacent
pores and leave a lower residual content behind, a process
that is inhibited in the saturated zone (36). Thus, the capacity
for retention of DNAPLs in the unsaturated zone is less than
the saturated zone.
Relative Permeability
Relative permeability is defined as the ratio of the permeability
of a fluid at a given saturation to its permeability at 100%
saturation. Thus it can have a value between 0 and 1 (71).
10
-------
Figure 16 illustrates a relative permeability graph for a two fluid
phase system showing the relationship between the observed
permeability of each fluid for various saturations to that of the
observed permeability if the sample were 100% saturated with
that fluid (73). The three regions of this graph are explained as
follows (71): Region I has a high saturation of DNAPL and is
considered a continuous phase while the water is a
discontinuous phase, therefore, water permeability is low.
Assuming the DNAPL is the non-wetting fluid, water would fill
the smaller capillaries and flow through small irregular pores. In
Region II, both water and DNAPL are continuous phases
although not necessarily in the same pores. Both water and
Kr = relative permeability
Q
0)
1
• Increasing DNAPL Saturation
Increasing Water Saturation
c
Mar William ml Wilder, I9?t
Figure 16. Relative permeability graph.
NAPLflow simultaneously. However, as saturation of either
phase increases, the relative permeability of the other phase
correspondingly decreases. Region III exhibits a high
saturation of water while the DNAPL phase is mainly
discontinuous. Water flow dominates this region and there is
little or no flow of DNAPL.
Both fluids flow through only a part of the pore space and thus
only a part of the cross section under consideration is available
for flow of each fluid. Therefore, the discharge of each fluid
must be lower corresponding to its proportion of the cross
sectional area (46).
Figure 17 is another relative permeability graph which
demonstrates several points. Small increases in DNAPL
saturation results in a significant reduction in the relative
permeability of water. However, a small increase in water
saturation does not result in a significant reduction in DNAPL
relative permeability. This figure identifies two points, SO1 and
SO2, where the saturation of the DNAPL and the water are
greater than 0 before there is a relative permeability for this
fluid. The two fluids hinder the movement of the other to
different degrees and both must reach a minimum saturation
before they achieve any mobility at all (47). These minimum
"saturations, for the water and DNAPL, are identified as
irreducible and residual saturation, respectively.
100%
i 100%
10
^ Alter Schwllle, 1988^)
Figure 17. The relative permeability curves for water and a
DNAPL In a porous medium as a function of the pore
space saturation.
Site Characterization for DNAPL
Characterization of the subsurface environment at hazardous
waste sites containing DNAPL is complex and will likely be •
expensive. Specific details associated with the volume and
timing of the DNAPL release are usually poor or are not
available and subsurface heterogeneity is responsible for the
complicated and unpredictable migration pathway of
subsurface DNAPL transport. As discussed previously, slight
changes in vertical permeability may induce a significant
horizontal component to DNAPL migration.
Site characterization typically involves a significant investment
in ground-water analyses. Although analysis of ground water
provides useful information on the distribution of the soluble
components of the DNAPL, the presence of other phases of
the DNAPL may go unrecognized. The investigation must,
therefore, be more detailed to obtain information concerning
the phase distribution of the DNAPL at a site. Site
characterization may require analyses on all four phases
(aqueous, gaseous, solid, immiscible) to yield the appropriate
information (refer to Table 2). In brief, data collected on the
various phases must be compiled, evaluated and used to help
identify: where the contaminant is presently located; where it
has been; what phases it occurs in; and what direction the
mobile phases may be going. A comprehensive review of site
characterization for subsurface investigations is available (68).
Development of monitoring and remediation strategies can be
focused more effectively and efficiently after a clear definition
Of the phase distribution has been completed.
Ground Water
Ground water analyses for organic compounds, in conjunction
with ground water flow direction data, has repeatedly been
used to: delineate the extent of ground water contamination
from DNAPL; determine the direction of plume migration; and
11
-------
Table 2 - Phase Distribution of DNAPL in the Subsurface
MATRIX
PHASE
1. ground water aqueous - soluble components of DNAPL
2. soil/aquifer solid - adsorbed components of DNAPL
material on solid phase material
3. DNAPL immiscible - continuous phase (mobile),
residual saturation (immobile)
4. soil gas gaseous - volatile components
to identify probable DNAPL source area(s). While this
approach has been used successfully to characterize the
distribution of contaminants in the subsurface, there are
limitations. For example, since DNAPL and ground water may
flow in different.directions, as indicated in Figures 13a and 13b,
ground water analyses may not necessarily identify the
direction of DNAPL migration.
Ground water analyses may be useful to identify probable
DNAPL source areas, but, estimating the volume of DNAPL in
the subsurface is limited using this approach. Soluble phase
components of DNAPL are rarely found in excess of 10% of
the solubility even when organic liquids are known or
suspected to be present. The concentration of soluble DNAPL
components in the ground water is not only a function of the
amount of DNAPL present, but also the chemical and physical
characteristics of the DNAPL, the contact area and time
between the ground water and DNAPL, and numerous
transport and fate parameters (retardation, biodegradation,
dispersion, etc.). One technique has been developed using
chemical ratios in the ground water as a means of source
identification and contaminant fate prediction (18).
Soil/AquiferMaterial
Exploratory Borings
Physical and chemical analyses of soil and aquifer material
(drill cuttings, cores) from exploratory borings will provide
useful information in the delineation of the horizontal and
vertical mass distribution of DNAPL. While simple visual
examination for physical presence or absence of contamination
might seem like a worthwhile technique, it can be deceiving
and does nothing to sort out the various liquid phases and their
relationship to each other (71). A quantitative approach is
necessary to determine DNAPL distribution.
Drill cuttings or core material brought to the surface from
exploratory borings can be screened initially to help delineate
the depth at which volatile components from the various
phases of the hydrocarbon exists. The organic vapor analyzer
and the HNU are small portable instruments that can detect
certain volatile compounds in the air. These methods are used
to initially screen subsurface materials for volatile components
of DNAPL. Identification of individual compounds and their
concentrations may be confirmed by other, more precise,
analyses.
Analysis of the soil or aquifer material by more accurate
means, such as gas chromalography or high pressure liquid
chromatography, will take longer but will provide more specific
information on a larger group of organic compounds, i.e.,
volatile/non-volatile, and on specific compounds. This
information is necessary to help fix the horizontal and vertical
mass distribution of the contaminant and to help delineate the
phase distribution. These analyses do not distinguish between
soluble, sorbed or free-phase hydrocarbon, however; a low
relative concentration indicates that the contaminant may
mainly be present in the gaseous or aqueous phases; and a
high relative concentration indicates the presence of sorbed
contaminant or free phase liquid either as continuous-phase or
residual saturation. A more rigorous set of analyses is required
to distinguish between the various phases.
Additional tests to identify the presence of NAPL in soil or
aquifer core sample are currently undeveloped and research in
this area is warranted. Squeezing and immiscible displacement
techniques have been used to obtain the pore water from
cores (40). Other methods of phase separation involving
vacuum or centrifugation may also be developed for this use. A
paint filter test was proposed in one Superfund DNAPL field
investigation where aquifer cores were placed in a filter/funnel
apparatus, water was added, and the filtrate was examined for
separate phases. These core analysis techniques have
potential to provide valuable field data to characterize NAPL
distribution.
Cone Penetrometer
The cone penetrometer (ASTM D3441-86)(69) has been used
for some time to supply data on the engineering properties of
soils. Recently, the application of this technology has made the
leap to the hazardous waste arena. The resistance of the
formation is measured by the cone penetrometer as it is driven
vertically into the subsurface. The resistance is interpreted as
a measure of pore pressure, and thus provides information on
the relative stratigraphic nature of the subsurface. Petroleum
and chlorinated hydrocarbon plumes can be detected most
effectively when the cone penetrometer is used in conjunction
with in-situ sensing technologies (48). Features of the cone
penetrometer include: a continuous reading of the stratigraphy/
permeability; in-situ measurement; immediate results are
available; time requirements are minimal; vertical accuracy of
stratigraphic composition is high; ground-water samples can be
collected in-situ; and the cost is relatively low.
Data from the cone penetrometer can be used to delineate
probable pathways of DNAPL transport. This is accomplished
by identifying permeability profiles in the subsurface. A zone of
low permeability underlying a more permeable stratigraphic
unit will likely impede vertical transport of the DNAPL. Where
such a scenario is found, a collection of DNAPL is probable
and further steps can be implemented to more accurately and
economically investigate and confirm such an occurrence.
This general approach has successfully been implemented at
one Superfund site (8).
DNAPL
Well Level Measurements
In an effort to delineate the horizontal and vertical extent of the
DNAPL at a spill site, it is important to determine the elevation
12
-------
£if DNAPL in the subsurface. Monitoring DNAPL elevation over
•me will indicate the mobility of the DNAPL. There are several
Methods that can be used to determine the presence of
DNAPL in a monitoring well. One method relies on the
difference in electrical conductivity between the DNAPL and
water. A conductivity or resistivity sensor is lowered into the
well and a profile is measured. The interface of the DNAPL is
accurately determined when the difference in conductivity is
detected between the two fluids. This instrument may also be
used to delineate LNAPL. A transparent, bottom-loading bailer
can also be used to measure the thickness (and to sample) of
DNAPL in a well (36). The transparent bailer is raised to the
surface and the thickness of the DNAPL is made by visual
measurement.
Several laboratory and field studies have been performed
which investigate the anomaly between the actual and
measured LNAPL levels in ground-water wells (15,16,24,25).
The anomaly between actual and measured NAPL thickness in
the subsurface is also applicable to DNAPL, but for different
reasons. The location of the screening interval is the key to
understanding both scenarios. First, if the well screen interval
is situated entirely in the DNAPL layer, and the hydrostatic
head (water) in the well is reduced by pumping or bailing, then
to maintain hydrostatic equilibrium, the DNAPL will rise in the
well (36,44,71) (refer to Figure 18). Secondly, if the well screen
extends into the barrier layer, the DNAPL measured thickness
will exceed that in the formation by the length of the well below
the barrier surface (36) (refer to Figure 19). Both of these
scenarios will result in a greater DNAPL thickness in the well
and thus a false indication (overestimate) of the actual DNAPL
mickness will result. One of the main purposes of the
monitoring well in a DNAPL investigation is to provide
information on the thickness of the DNAPL in the aquifer.
Therefore, construction of the well screen should intercept the
ground waterDNAPL interface and the lower end of the screen
should be placed as close as possible to the impermeable
stratigraphic unit.
Measured > Actual
DNAPL Pool
Impermeable Boundary
Ifigure 18. A well screened only in the DNAPL in conjunction
with lower hydrostatic head (i.e. water) in the well
may result in an overestimation of DNAPL thickness.
Measured > Actual
Jf_
DNAPL Pool
Impermeable /
Boundary
Figure 19. A well screened into an impermeable boundary
may result in an over-estimation of the DNAPL
thickness.
DNAPL Sampling
Sampling of DNAPL from a well is necessary to perform
chemical and physical analyses on the sample. Two of the
most common methods used to retrieve a DNAPL sample from
a monitoring well are the peristaltic pump and the bailer. A
peristaltic pump can be used to collect a sample if the DNAPL
is not beyond the effective reach of the pump, which is typically
less than 25 feet. The best method to sample DNAPL is to use
a double check valve bailer. The key to sample collection is
controlled, slow lowering (and raising) of the bailer to the
bottom of the well (57). The dense phase should be collected
prior to purging activities.
Soil-Gas Surveys
A soil-gas survey refers to the analysis of the soil air phase as
a means to delineate underground contamination from volatile
organic chemicals and several techniques have been
developed (34,52). This investigative tool is mainly used as a
preliminary screening procedure to delineate the areal extent
of volatile organic compounds in the soil and ground water.
This method is quick, less expensive than drilling wells and can
provide greater plume resolution (33).
Data from a soil-gas survey is a valuable aid in the
development of a more detailed subsurface investigation
where ground water monitoring wells and exploratory borings
are strategically located for further site characterization. There
are limitations to soil-gas surveys (26,52) and data
interpretation must be performed carefully (35,49). Soil-gas
investigations have mainly been conducted to identify the
location of the organic contaminants in ground water. At the
time of this publication, the scientific literature did not contain
information specifically applicable to the delineation of DNAPL
from soil-gas survey data. However, it is surmisable that soil-
gas surveys can be used to help delineate DNAPL residual
saturation in the unsaturated zone or the location of perched
DNAPL reservoirs.
13
-------
Miscellaneous
Pumping Systems
The vertical migration of DNAPL in the saturated zone will
eventually be challenged by a low permeability stratigraphic
unit. According to the principles of capillary pressure, the lower
permeability unit will exhibit a greater capillary pressure.
Displacement of water by DNAPL requires that the hydrostatic
force from the mounding DNAPL exceed the capillary force of
the low permeability unit. The Hobson formula is used to
compute the critical height calculation to overcome the
capillary pressure under different pore size conditions (70).
In an effort to minimize further DNAPL contamination as a
result of drilling investigations, precautionary steps should be
taken. Penetration of DNAPL reservoirs in the subsurface
during drilling activities offers a conduit for the DNAPL to
migrate vertically into previously uncontaminated areas. It is
very easy to unknowingly drill through a DNAPL pool and the
bed it sits on, causing the pool to drain down the hole into a
deeper part of the aquifer or into a different aquifer (32).
Special attention to grouting and sealing details during and
after drilling operations will help prevent cross-contamination.
Precautionary efforts should also be considered when a
DNAPL reservoir is encountered during drilling operations. The
recommended approach is to cease drilling operations and
install a well screen over the DNAPL zone and cease further
drilling activities in the well. If it is necessary to drill deeper,
construction of an adjacent well is recommended. Alternatively,
if it is not necessary to screen off that interval, it is
recommended to carefully seal off the DNAPL zone prior to
drilling deeper.
Well construction material compatibility with DNAPL should be
investigated to minimize down hole material failure. A
construction material compatibility review and possible testing
will prevent the costly failure of well construction material. The
manufacturers of well construction material are likely to have
the most extensive compatibility data and information
available.
Remediation
Remediation of DNAPL mainly involves physical removal by
either pumping or trench-drainline systems. Removal of
DNAPL early in the remediation process will eliminate the main
source of contaminants. This step will substantially improve the
overall recovery efficiency of the various DNAPL phases
including the long term pump and treat remediation efforts for
soluble components. Remediation technologies such as
vacuum extraction, biodegradation, ground water pumping,
and soil flushing is mainly directed at the immobile DNAPL and
the various phases in which its components occur. Physical
barriers can be used in an effort to minimize further migration
of the DNAPL
Clean-up of DNAPL can involve sizable expenditures: they are
difficult to extract and the technology for their removal is just
evolving (43). Historically, field recovery efforts usually proceed
with a poor understanding of the volume distribution of the
DNAPL. This reflects the difficulties involved in adequate site
characterization, poor documentation of the release, and the
complexity associated with the DNAPL transport in the
subsurface.
Pumping represents an important measure to stop the mobile
DNAPL from migrating as a separate phase by creating a
hydraulic containment and by removal of DNAPL (44). Very
simply, DNAPL recovery is highly dependent on whether the
DNAPL can be located in the subsurface. The best recovery
scenario is one in which the DNAPL is continuous and has
collected as a reservoir in a shallow, impermeable subsurface
depression. Once the DNAPL has been located and recovery
wells are properly installed, pumping of pure phase DNAPL is
a possible option but depends largely on site specific
conditions which include, but are not limited to: DNAPL
thickness, viscosity, and permeability.
Many DNAPL reservoirs in the subsurface are of limited
volume and areal extent. Therefore, it can be expected that
both the level of DNAPL (saturated thickness) in the well will
decline from the prepumping position and the percentage of
DNAPL in the DNAPL:water mixture will decrease rather
rapidly. Correspondingly, DNAPL recovery efficiency
decreases. Field results indicate that recovery wells screened
only in the DNAPL layer will maintain maximum DNAPL:water
ratios (102). Well diameter was not found to influence long
term DNAPL recovery; however, large diameter wells allow
high volume pumping for short durations; and small diameter
wells result in lower DNAPL:water mixtures and greater
drawdown.
An enhanced DNAPL recovery scheme may be used to
improve recovery efficiency. An additional well is constructed
with a screen interval in the ground water zone located
vertically upward from the DNAPL screen intake. Ground water
is withdrawn from the upper screen which results in an
upwelling of the DNAPL (70), refer to Figure 20. The upwelling
of the DNAPL, coal tar in this case, improved the rate (twofold)
at which the coal tar was recovered resulting in a more efficient
operation. The ground water withdrawal rate must be carefully
determined; too much will result in the coal tar from rising
excessively and being either mixed (emulsions) with or
suppressed by the higher water velocity above; too low will not
High Level
High Level
Storage
Treatment
i
*^
- w
J
Static Ground Water Level
Hydraulically Induced
DNAPL Level
Static DNAPL Level
Sand
After J.F. VHIaume. al.al. 1983
Figure 20. A DNAPL recovery system where deliberate
upwelling of the static coal-tar surface Is used to
Increase the flow of product into the recovery wells.
14
-------
paused upwelling. An estimate of this upwelling can be
Calculated using the simplified Ghyben-Herzberg Principle
'under ideal conditions (4). Laboratory studies indicated that
dimethyl phthalate (1.19 g/cc) recovery rate was doubled or
tripled over the conventional, non-upconing, recovery scheme
(75). A similar application of this technique was used to
increase the level of DNAPL (solvents) in a sandstone bedrock
formation (11). Other enhanced DNAPL recovery techniques
were implemented utilizing both water flooding and wellbore
vacuum. Essentially, this minimized drawdown, allowing a
maximum pumping rate of the DNAPL:water mixture. Both
techniques offered significant advantages in terms of the rate
and potential degree of DNAPL removal (8).
The highly corrosive nature of some DNAPL's may increase
maintenance problems associated with the recovery system. A
design consideration during any DNAPL recovery program
should include a material compatibility review to minimize
downhole failures. This is applicable to the well construction
material and the various appurtenances of the recovery
system. Manufacturers of the construction material would
most likely have the best compatibility information available.
While most scientists agree that the residual saturation of
immiscible hydrocarbon droplets in porous media are
immobile, researchers have investigated the mobility of
residual saturation in porous media for enhanced oil recovery
and for NAPL remediation at spill sites. Specifically, this
includes a complex interplay between four forces (viscous,
gravity, capillary, buoyancy). These forces are dependent on
both the chemical and physical characteristics of the DNAPL
kand porous media. The mobilization of residual saturation
Inainly hinges on either increasing the ground water velocity
"which increases the viscous forces between the residual
saturation and the ground water, or decreasing the interfacial
tension between the residual saturation and the ground water
which decreases the capillary forces.
The capillary number is an empirical relationship which
measures the ratio between the controlling dynamic stresses
(absolute viscosity and ground water velocity) and static
stresses (interfacial tension) of the residual saturation (39). The
former are the viscous stresses and the dynamic pressure in
the water which tend to move the oil. The latter are the
capillary stresses in the curved water/oil interfaces which tend
to hold the oil in place. As the capillary number is increased,
the mobility of the residual saturation increases. In a laboratory
column study, the capillary number had to be increased two
orders of magnitude from when motion was initiated to
complete displacement of the hydrocarbon in a sandstone core
(74). In a glass bead packed column, only one order of
magnitude increase was required. However, a higher capillary
number was required to initiate mobility. The difference in
mobility between the two columns was attributed to the pore
geometry, i.e. size, shape.
There are limitations to residual saturation mobilization. The
ground water gradient (dh/dl) necessary to obtain the critical
capillary number to initiate blob mobilization would be 0.24. To
obtain complete NAPL removal would require a gradient of 18
(3). Ground water gradients of this magnitude are unrealistic.
Another estimate of the gradient necessary to mobilize carbon
tetrachloride in a fine gravel and medium sand was 0.09 and
9.0 respectively (74). The former gradient is steep but not
unreasonable and the latter gradient is very steep and
impractical to achieve in the field. The same researchers
concluded from more recent, comprehensive studies, that the
earlier predictions were optimistic, and that the gradient
necessary to mobilize residual organic liquid is clearly
impractical (66). Another limitation is that along with residual
saturation mobilization, the NAPL blobs disperse into smaller
blobs and that the blob distribution was dependent on the
resulting capillary number (6). Recovery of the NAPL residual
saturation by pumping ground water may be more feasible
where the porous media is coarse and capillary forces are low,
i.e. coarse sands and gravel. However, even in this scenario, it
is expected that the radius of residual saturation mobilization
would be narrow.
It is held in petroleum engineering theory that the only practical
means of raising the capillary number dramatically is by
lowering the interfacial tension (39) and that this can be
achieved by using surfactants (66). Surfactants reduce the
interfacial tension between two liquids, and therefore, are
injected into the subsurface for enhanced recovery of
immiscible hydrocarbons. In laboratory experiments, surfactant
flushing solutions produced dramatic gains in flushing even
after substantial water flushing had taken place (54).
Unfortunately, surfactants can be quite expensive and cost
prohibitive in NAPL recovery operations. Surfactants are
usually polymeric in nature and a surfactant residue may be
left behind in the porous media which may not be
environmentally acceptable. Additionally, surfactants may be
alkaline and thus affect the pH of the subsurface environment.
It has been suggested that such a surfactant may inhibit
bacterial metabolism and thus preclude subsequent use of
biological technologies at the site. Significant research in this
area is currently underway which may uncover information
improving the economics and feasibility of this promising
technology.
In summary, practical considerations and recommendations
concerning the mobilization and recovery of residual saturation
include the following: greater effectiveness in very coarse
porous media i.e. coarse sands and gravel; recovery wells
should be installed close to the source to minimize flow path
distance; a large volume of water will require treatment/
disposal at the surface; compounds with high interfacial
tension or viscosity will be difficult to mobilize; and implemen-
tation of linear one-dimensional sweeps through the zones of
residual saturation (74) and surfactants will optimize recovery.
Pumping the soluble components (aqueous phase) of DNAPL
from the immiscible (continuous and residual saturation), solid
(sorbed), and gaseous phases has been perhaps one of the
most effective means to date to both recover DNAPL from the
subsurface and to prevent plume migration. Recovery of
soluble components quite often has been the only remediation
means available. This is largely attributed to the inability to
locate DNARL pools and due to low, DNAPL yielding
formations. The basic principles and theory of pump and treat
technology and the successes and failures have been
summarized in other publications (64,67) and is beyond the
scope of this publication.
Pumping solubilized DNAPL components from fractured rock
aquifers historically has been plagued with a poor recovery
efficiency. Although the rock matrix has a relatively small
intergranular porosity, it is commonly large enough to allow
dissolved contaminants from the fractures to enter the matrix
15
-------
Ground Surface
Ox*id Wdaf Suflace
DNAPL Surface
Oil Distribution
• DNAPL denser than ground water,
has accumulated at Ihe basa of the
alluvium.
Ground Water Surtat
DNAPL 5urtace_
Ground Surface
Ground Waar Sudace
. CTMPl Surface^Z>-
•" ~~
DNAPL Mounding
• Drawdown of Ihe cvarlying water
table by pumping the water drain!ne
results hi mounding of the DNAPL
DNAPL Recovery
• Pumping from both Ihe water and
DNAPL drainline induces increasing
DNAPL low lo Ihe DNAPL drainline.
• Separate production of DNAPL and
ground water reduces above ground
separation requirements.
• A flow path of maximum formation
permeability to DNAPL Is established
at lie base of Ihe aNuvium.
AHer, Sato 91«!.. 19S6
Figure 21. Trench recovery system of DNAPL utilizing the dual
drainline concept.
by diffusion and be stored there by adsorption (32). The
release of these components is expected to be a slow diffusion
dominated process. This is because little or no water flushes
through dead-end fracture segments or through the porous,
impervious rock matrix. Therefore, clean-up potential is
estimated to be less than that expected for sand and gravel
aquifers.
Trench Systems
Trench systems have also been used successfully to recover
DNAPL and are used when the reservoir is located near the
ground surface. Trench systems are also effective when the
DNAPL is of limited thickness. Recovery lines are placed
horizontally on top of the impermeable stratigraphic unit.
DNAPL flows into the collection trenches and seep into the
recovery lines. The lines usually drain to a collection sump
where the DNAPL is pumped to the surface. Similar to the
pumping system, an enhanced DNAPL recovery scheme may
be implemented using drain lines to improve recovery
efficiency. This "dual drain line system" (41) utilizes a drain line
located in the ground water vertically upward from the DNAPL
line. Ground water is withdrawn from the upper screen which
results in an upwelling of the DNAPL which is collected in the
lower line, refer to Figure 21. This increases the hydrostatic
head of the DNAPL. Excessive pumping of either single or dual
drain line systems may result in the ground water "pinching off"
the flow of DNAPL to the drain line. An advantage of the dual
drain system is that the oihwater separation requirements at
the surface are reduced.
Vacuum Extraction
Soil vacuum extraction (SVE) is a remediation technology
which involves applying a vacuum to unsaturated subsurface
strata to induce air flow. Figure 22 illustrates that the volatile
contaminants present in the contaminated strata will evaporate
and the vapors are recovered at the surface and treated.
Common methods of treatment include granular activated
carbon, catalytic oxidation, and direct combustion. SVE can
effectively remove DNAPL present as residual saturation or its
soluble phase components in the unsaturated zone. In general,
vacuum extraction is expected to be more applicable for the
chlorinated solvents (PCE, TCE, DCE) than the polycyclic
aromatic compounds (wood preserving wastes, coal tars, etc.).
When DNAPL is present in perched pools (Figure 12) it is more
effective to remove the continuous phase DNAPL prior to the
implementation of SVE. The same strategy is applicable in the
saturated zone where DNAPL removal by SVE is attempted
concomitantly with lowering the water table. Upon lowering the
water table, SVE can be used to remove the remnant volatile
wastes not previously recovered. Often, the precise location of
the DNAPL is unknown; therefore, SVE can be used to
remediate the general areas where the presence of DNAPL is
suspected. Removal of DNAPL by SVE is not expected to be
as rapid as direct removal of the pure phase compound. One
advantage of SVE however, is that the precise location of the
DNAPL need not be known.
Important parameters influencing the efficacy of SVE concern
both the DNAPL and porous media. Porous media specific
parameters include: soil permeability, porosity, organic carbon,
moisture, structure, and particle size distribution. DNAPL
specific parameters include: vapor pressure, Henry's constant,
solubility, adsorption equilibrium, density, and viscosity (20).
These parameters and their relationships must be evaluated
on a site specific basis when considering the feasibility of
vacuum extraction and a practical approach to the design,
construction, and operation of venting systems (22).
Additionally, soil gas surveys which delineate vapor
concentration as a function of depth is critical in locating the
contaminant source and designing an SVE system.
Historically, SVE has been used to remove volatile compounds
from the soil. Recently it has been observed that SVE
enhances the biodegradation of volatile and semivolatile
organic compounds in the subsurface. While SVE removes
volatile components from the subsurface, it also aids in
supplying oxygen to biological degradation processes in the
unsaturated zone. Prior to soil venting, it was believed that
biodegradation in the unsaturated zone was limited due to
inadequate concentrations of oxygen (17). In a field study
where soil venting was used to recover jet fuel, it was observed
that approximately 15% of the contaminant removal was from
the result of microbial degradation. Enhanced aerobic
biodegradation during SVE increases the cost effectiveness of
the technology due to the reduction in the required above
ground treatment.
Vacuum extraction is one form of pump and treat which occurs
in the saturated zone where the fluid is a gas mixture.
Therefore, many of the same limitations to ground water pump,
and treat are also applicable to vacuum extraction. While the
application of vacuum extraction is conceptually simple, its
success depends on understanding complex subsurface
16
-------
Figure 22. Vacuum extraction of DNAPL volatile components
in the unsaturated zone. As shown here, vapors are
treated by thermal combustion or carbon adsorp-
tion and the air is discharged to the atmosphere.
chemical, physical, and biological processes which provide
|[nsight into factors limiting its performance (9).
Biodegradation
The potential for biodegradation of immiscible hydrocarbon is
highly limited for several reasons. First, pure phase
hydrocarbon liquid is a highly hostile environment to the
survival of most microorganisms. Secondly, the basic
requirements for microbiological proliferation (nutrients,
electron acceptor, pH, moisture, osmotic potential, etc.) is
difficult if not impossible to deliver or maintain in the DNAPL. A
major limitation to aerobic bioremediation of high
concentrations of hydrocarbon is the inability to deliver
sufficient oxygen. A feasible remediation approach at sites
where immiscible hydrocarbon is present is a phased
technology approach. Initial efforts should focus on pure phase
hydrocarbon recovery to minimize further migration and to
decrease the Volume of NAPL requiring remediation.
Following NAPL recovery, other technologies could be phased
into the remediation effort. Bioremediation may be one such
technology that could be utilized to further reduce the mass of
contaminants at the site. NAPL recovery preceding
bioremediation will improve bioremediation feasibility by
reducing the toxicity, time, resources, and labor.
Similar to other remediation technologies, a comprehensive
feasibility study evaluating the potential effectiveness of
bioremediation is critical and must be evaluated on a site
specific basis. A comprehensive review of biodegradation of
surface soils, ground water, and subsoils of wood preserving
pastes, i.e. PAH's (29,37,51,62,63) are available. A
Comprehensive review of microbial decomposition of
chlorinated aromatic compounds is also available (58).
Soil Flushing
Soil flushing utilizing surfactants is a technology that was
developed years ago as a method to enhance oil recovery in
the petroleum industry. This technology is new to the
hazardous waste arena and available information has mainly
been generated from laboratory studies. Surfactant soil
flushing can proceed on two distinctly different mechanistic
levels: enhanced dissolution of adsorbed and dissolved phase
contaminants, and displacement of free-phase nonaqueous
contaminants. These two mechanisms may occur
simultaneously during soil flushing (42).
Surfactants, alkalis, and polymers are chemicals used to
modify the pore-level physical forces responsible for
immobilizing DNAPL In brief, surfactants and alkalis reduce
the surface tension between the DNAPL and water which
increases the mobility. Polymers are added to increase the
viscosity of the flushing fluid to minimize the fingering effects
and to maintain hydraulic control and improve flushing
efficiency. Based on successful laboratory optimization studies
where an alkali-polymer-surfactant mixture was used, field
studies were conducted on DNAPL (creosote) which resulted
in recovery of 94% of the original DNAPL (42). Laboratory
research has also been conducted which indicated that
aqueous surfactants resulted in orders of magnitude greater
removal efficiency of adsorbed and dissolved phase
contaminants than water flushing (55).
Depth to contamination, DNAPL distribution, permeability,
heterogeneities, soil/water incompatibility, permeability
reduction, and chemical retention are important factors when
considering soil flushing (42). Prior to this technology being
cost effective in the fiold, surfactant recycling will be necessary
to optimize surfactant use (55). Soil flushing is complex from a
physical and chemical point of view; is relatively untested in the
field; and will likely be challenged regulatorily. Considerable
research currently being conducted in this area may result in
the increased use of this technology to improve DNAPL
recovery in the future.
Thermal methods of soil flushing involve injecting hot water or
steam in an effort to mobilize the NAPL. The elevated
temperature increases volatilization and solubilization and
decreases viscosity and density. A cold-water cap is used to
prevent volatilization. The mobile phases of the DNAPL are
then recovered using a secondary approach, i.e. pumping,
vacuum extraction etc. This approach (Contained Recovery of
Oily Wastes) to enhance recovery of DNAPL is currently under
EPA's Superfund Innovative Technology Evaluation Program
and a pilot-scale demonstration is forthcoming (21). A
limitation in the use of thermal methods is that the DNAPL may
be converted to LNAPL due to density changes (36). The
adverse effects from this are that the DNAPL, existing as a thin
layer, becomes buoyant and mobilizes vertically resulting in a
wider dispersal of the contaminant. Other limitations involve
the high energy costs associated with the elevated water
temperature and the heat loss in the formation (36).
Physical Barriers
Physical barriers may be used to prevent the migration of
DNAPL's in the subsurface and are typically used in
conjunction with other recovery means. One feature of physical
17
-------
barriers is the hydraulic control it offers providing the
opportunity to focus remediation strategies in treatment cells.
Unfortunately, physical barriers, while satisfactory in terms of
ground water control and containment of dissolved-phase
plumes, may contain small gaps or discontinuities which could
permit escape of DNAPL (7). Chemical compatibility between
physical barriers and construction material must agree to
insure the physical integrity of the barrier. The history of the
performance of these containment technologies is poorly
documented and is mainly offered here for completeness of
review. A more complete review of these physical barriers is
available (5,56).
Sheet piling involves driving lengths of steel that connect
together into the ground to form an impermeable barrier to
lateral migration of DNAPL. Ideally, the bottom of the sheet pile
should be partially driven into an impermeable layer to
complete the seal. Slurry walls involve construction of a trench
which is backfilled with an impermeable slurry (bentonite)
mixture. Grouting is a process where an impermeable mixture
is either injected into the ground or is pumped into a series of
interconnected boreholes which together form an impermeable
boundary. Again, the main feature of these techniques is to
physically isolate the DNAPL.
In summary, site characterization and remediation options for
sites containing DNAPL are limited. Field data from site
characterization and remediation efforts are also limited. This
is largely due to the complexity of DNAPL transport and fate in
the subsurface, poorly developed techniques currently
available to observe and predict DNAPL in the subsurface, and
to the fact that this issue has not been widely recognized until
recently. Clearly, there is a growing realization within the
scientific and regulatory community that DNAPL is a significant
factor in limiting site remediation. Correspondingly, current
research efforts within the private, industrial, and public sectors
are focusing on both the fundamentals and applications
aspects of DNAPL behavior in subsurface systems.
Additionally, the number of field investigations reflecting an
increased awareness of DNAPLs, is growing.
DNAPL Modeling
A modeling overview report identified nineteen (numeric and
analytic) multiphase flow models which are currently available
(60). Most of these models were developed for salt water
intrusion, LNAPL transport, and heat flow. Four models are
qualitatively described as immiscible flow models but do not
specifically indicate DNAPL. A more recent model has been
developed which simulates density driven, three phase flow,
that is capable of modeling DNAPL transport (23). Presently,
very little information is available on DNAPL modeling in the
scientific literature.
Multiphase flow modeling involves modeling systems where
more than one continuous fluid phase (NAPL, water, gaseous)
is present. Modeling any subsurface system requires a
conceptual understanding of the chemical, physical, and
biological processes occurring at the site. Modeling of
simultaneous flow of more than one fluid phase requires a
conceptual understanding of the fluids and the relationship
between the fluid phases. The significance of multiphase flow
over single phase flow is the increased complexity of fluid flow
and the additional data requirements necessary for modeling.
As presented earlier, numerous variables strongly influence
DNAPL transport and fate, and consequently, the
mathematical relationship of these variables is complex.
Therefore, it follows that DNAPL modeling presents paramount
technical challenges.
Presently, it is exceedingly difficult to obtain accurate field data
which quantitatively describes DNAPL transport and fate
variables within reasonable economic constraints. DNAPL
transport is highly sensitive to subsurface heterogeneities
(8,27,28) which compounds the complexity of modeling.
Heterogeneities are, by nature, difficult to identify and quantify
and models are not well equipped to accommodate the
influence of heterogeneities. Additionally, relative permeability
and capillary pressure functions must be quantified to identify
the relationship between fluids and between the fluids and the
porous media. Unfortunately, these parameters are very
difficult to measure, particularly in three phase systems. Prior
to an investment of time and money to model a given site, a
careful evaluation of the specific objectives and the confidence
of the input and anticipated output data should be performed.
This will help illuminate the costs, benefits, and therefore, the
relative value of modeling in the Superfund decision making
process.
In summary, DNAPL modeling at Superfund sites is presently
of limited use. This is mainly due to: the fact that very little
information is available in the scientific literature to evaluate
previous work; accurate and quantitative input data is expected
to be costly; the sensitivity of DNAPL transport to subsurface
heterogeneities; and, the difficulty in defining the
heterogeneities in the field and reflecting those in a model.
However, multiphase flow models are valuable as learning
tools.
References
1. Baehr, A.L, Selective Transport of Hydrocarbons in the
Unsaturated Zone Due to Aqueous and Vapor Phase
Partitioning, Water Resources Research. Vol. 23, No. 10,
pp. 1926-1938, 1987.
2. Bear, J., 1972, Dynamics of Fluids in Porous Media.
American Elsevier Publishing Co., New York, 763 p.
3. Bouchard, D., Contaminant Transport in the Subsurface:
Sorption Equilibrium and the Role of Nonaqueous Phase
Liquids, in, Intermedia Pollutant Transport: and Field
Measurement. (David T. Allen, Yoram Cohen and Isaac
R. Kaplan, Eds.), New York, Plenum Pub. Corp., pp. 189-
211.
4. Bower, H., Groundwater Hydrology. McGraw-Hill Book
Co., 1978, 480pp.
5. Canter, L. W. and R. C. Knox, Ground Water Pollution
Control, Lewis Publishers Inc., Chelsea, Mich., 1986, 526
PP.
6. Chatzis, I., M.S. Kuntamukkula, and N.R. Morrow, Blob-
size Distribution as a Function of Capillary Number in i
Sandstones, Paper 13213, Presented at SPE Annual
Tech. Conference and Exhibition. Houston, TX, 1984.
18
-------
<^. Cherry, J.A., S. Feenstra, B.H. Kueper and D.W.
McWhorter, "Status of In Situ Technologies for Cleanup
of Aquifers Contaminated by DNAPL's Below the Water
Table," in, International Specialty Conference on How
Clean is Clean? Cleanup Criteria for Contaminated Soil
and Groundwater. Air and Waste Management
Association, pp. 1-18, November 6-9, 1990.
8. Connor, J.A., C.J. Newell, O.K. Wilson, Assessment,
Field Testing, Conceptual Design for Managing Dense
Nonaqueous Phase Liquids (DNAPL) at a Superfund
Site, in, Proceedings of Petroleum Hydrocarbons and
Organic Chemicals in Ground Water: Prevention.
Detection, and Restoration. A Conference and
Exposition, The Westin Gaileria, Houston, TX, Vol. 1, pp.
519-533,1989.
9. DiGiulio, D.C. and J.S. Cho, Conducting Field Tests for
Evaluation of Soil Vacuum Extraction Application, in,
Proceedings of the Fourth National Outdoor Action
Conference on Aquifer Restoration. Ground Water
Monitoring, and Geophysical Methods. Las Vegas, NV,
May 14-17, 1990, pp. 587-601.
10. Feenstra, S., Evaluation of Multi-Component DNAPL
Sources by Monitoring of Dissolved-Phase
Concentrations, in, Proceedings of the Conference On
Subsurface Contamination by Immiscible Fluids.
International Association of Hydrogeologists, Calgary,
Alberta, April 18-20, 1990.
$ 1. Ferry, J.P. and P.J. Dougherty, Occurrence and
Recovery of a DNAPL in a Low-Yielding Bedrock Aquifer,
in, Proceedings of the NWWA/API Conference on
Petroleum Hydrocarbons and Organic Chemicals in
Ground Water - Prevention. Detection and Restoration.
Nov. 12-14, Houston, TX., 1986, pp. 722-733.
12. Fu, J.K. and R.G. Luthy, Effect of Organic Solvent on
Sorption of Aromatic Solutes onto Soils, Journal of
Environmental Engineering. Vol. 112, No. 2, pp. 346-366,
1986.
13. Glass, R.J., T.S. Steenhuis, and J.Y. Parlange,
Mechanism for Finger Persistence in Homogeneous
Unsaturated Porous Media: Theory and Verification, Soil
Science. 148(1), pp. 60-70,1989.
14. Hall, A.C., S.H. Collins, and J.C. Melrose, Stability of
Aqueous Wetting Films, Society of Petroleum
Engineering Journal. 23(2), pp. 249-258, 1983.
15. Hall, R.A., S.B. Blake, and S.C. Champlin, Jr.,
Determination of Hydrocarbon Thickness in Sediments
Using Borehole Data, in, Proceedings of the 4th National
Symposium and Exposition on Aquifer Restoration and
Ground Water Monitoring. Columbus, OH, pp. 300-304,
May 23-25, 1984.
16. Hampton, D.R., and P.D.G. Miller, Laboratory
Investigation of the Relationship Between Actual and
Apparent Product Thickness in Sands, in, Proceedings of
Petroleum Hydrocarbons and Organic Chemicals in
Ground Water: Prevention. Detection, and Restoration. A
Conference and Exposition, The Westin Gaileria,
Houston, Texas, Vol. 1, pp. 157-181, November 9-11,
1988.
17. Hinchee, R.E., D.C. Downey, R.R. Duppnt, P. Aggarwal,
and R.N. Miler, Enhancing Biodegradation of Petroleum
Hydrocarbon Through Soil Venting, Journal of Hazardous
Materials, (accepted) 1990.
18. Hinchee, R.E. and H.J. Reisinger, A Practical Application
of Multiphase Transport Theory to Ground Water
Contamination Problems, Ground Water Monitoring
Review, pp. 84-92, Winter, 1987.
19. Hoag, G.E. and M.C. Marley, Gasoline Residual
Saturation in Unsaturated Uniform Aquifer Materials,
Journal of Environmental Engineering. Vol. 112, No. 3,
pp. 586-604, 1989.
20. Hutzler, N.J., B.E. Murphy, and J.S. Gierke, Review of
Soil Vapor Extraction System Technology, Presented a!
Soil Vapor Extraction Technology Workshop. June 28-29,
1989, Edison, New Jersey.
21. Johnson, L.A. and F.D. Guffey, "Contained Recovery of
Oily Wastes, Annual Progress Report." Western
Research Institute, Laramie, Wyoming, June, 1989.
22. Johnson, P.C., C.C. Stanley, M.W. Kemblowski, D.L.
Byers, and J.D. Colthart, A Practical Approach to the
Design, Operation, and Monitoring of In Situ Soil-Venting
Systems, Ground Water Monitoring Review, pp. 159-178,
Spring 1990.
23. Katyal, A.K., J.J. Kaluarachchi, and J.C. Parker, MOFAT:
A Two-Dimensional Finite Element Program for
Multiphase Flow and Multicomponent Transport.
Program Documentation, Version 2.0, Virginia
PolyTechnic Institute and State University, 58 pp.,
August, 1990.
24. Kemblowski, M.W. and C.Y. Chiang, Analysis of the
Measured Free Product Thickness in Dynamic Aquifers,
in, Proceedings of Petroleum Hydrocarbons and Organic
Chemicals in Ground Water: Prevention. Detection, and
Restoration. A Conference and Exposition, The Westin
Gaileria, Houston, Texas, Vol. 1, pp. 183-205, November
9-11, 1988.
25. Kemblowski, M.W. and C.Y. Chiang, Hydrocarbon
Thickness Fluctuations in Monitoring Wells, Ground
Water. Vol. 28, No. 2, pp. 244-252, 1990.
26. Kerfoot, H.B., Is Soil-Gas Analysis an Effective Means of
Tracking Contaminant Plumes in Ground Water? What
are the Limitations of the Technology Currently
Employed? Ground Water Monitoring Review, pp. 54-57,
Spring 1988.
27. Kueper, B.H. and E.O. Frind, An Overview of Immiscible
Fingering in Porous Media, Journal of Contaminant
Hydrology. Vol. 2, pp. 95-110, 1988.
28. Kueper, B.H., W. Abbott, and G. Farquhar, Experimental
Observations of Multiphase Flow in Heterogeneous
Porous Media, Journal of Contaminant Hydrology. Vol. 5,
pp. 83-95, 1989.
19
-------
29. Lee, M.D., J.M. Thomas, R.C. Borden, P.B. Bedient, J.T.
Wilson, and C.H. Ward, Biorestoration of Aquifers
Contaminated with Organic Compounds, National Center
for Ground Water Research, CRC Critical Reviews in
Environmental Control. Vol. 18, Issue 1, pp. 29-89, 1988.
30. Lindeburg, M.R., 1986, Civil Engineering Reference
Manual. 4th edition, Professional Publications Inc.
Belmont, CA.
31. Lyman, W.J., W.F. Reehl, and D.H. Rosenblatt,
Handbook of Chemical Property Estimation Methods.
McGraw-Hill Book Company, 1982.
32. Mackay, D.M. and J.A. Cherry, Ground-Water
Contamination: Pump and Treat Remediation,
Environmental Science & Technology. Vol. 23, No. 6, pp.
630-636, 1989.
33. Marrin, D.L and G.M. Thompson, Gaseous Behavior of
TCE Overlying a Contaminated Aquifer, Ground Water.
Vol. 25, No. 1, pp. 21-27, 1987.
34. Marrin, D., Kerfoot, H, Soil-gas surveying techniques
Environmental Science & Technology. Vol. 22, No. 7, pp.
740-745, 1988.
35. Marrin, D.L., Soil-Gas Sampling and Misinterpretation,
Ground Water Monitoring Review, pp. 51-54, Spring
1988.
36. Mercer, J.W. and R.M. Cohen, A Review of Immiscible
Fluids in the Subsurface: Properties, Models,
Characterization and Remediation, Journal of
Contaminant Hydrology. Vol. 6, pp. 107-163,1990.
37. Mississippi Forest Products Laboratory. Proceedings of
the Bioremediation of Wood Treating Waste Forum.
Mississippi State University, March 14-15,1989.
38. Morrow, N.R., Interplay of Capillary, Viscous and
Bouyancy Forces in the Mobilization of Residual Oil, The
Journal of Canadian Petroleum. Vol. 18, No. 3, pp. 35-46,
1979.
39. Ng, K.M., H.T. Davis, and L.E. Scriven, Visualization of
Blob Mechanics in Flow Through Porous Media,
Chemical Engineering Science. Vol. 33, pp. 1009-1017,
1978.
40. Patterson, R.J., S.K. Frape, LS. Dykes, and R.A.
McLeod, A Coring and Squeezing Technique for the
Detailed Study of Subsurface Water Chemistry,
Canadian Journal Earth Science. Vol. 15, pp. 162-169,
1978.
41. Sale, T., CH2M Hill, and Kuhn, B., Recovery of Wood-
Treating Oil from an Alluvial Aquifer Using Dual
Drainlines, in, Proceedings of Petroleum Hydrocarbons
and Organic Chemicals in Ground Water: Prevention.
Detection, and Restoration. A Conference and
Exposition, The Westin Galleria, Houston, Texas, Vol. 1,
pp. 419-442, November 9-11,1988.
42. Sale, T., K. Piontek, and M. Pitts, Chemically Enhanced
In-Situ Soil Washing, in Proceedings of the Conference
on Petroleum Hydrocarbons and Organic Chemicals in
Ground Water: Prevention. Detection, and Restoration.
Houston, TX, November 15-17, 1989.
43. Schmidtke, K., E. McBean, and F. Rovers, Drawdown
Impacts in Dense Non-Aqueous Phase Liquids, in
NWWA Ground Water Monitoring Symposium. Las
Vegas, Nevada, pp. 39-51, May, 1987.
44. Schmidtke, K., E. McBean, and F. Rovers, Evaluation of
Collection Well Parameters for DNAPL, Journal of
Environmental Engineering, accepted, August, 1990.
45. Schwille, F., Groundwater Pollution in Porous Media by
Fluids Immiscible With Water, The Science of the Total
Environment. Vol. 21, pp. 173-185, 1981.
46. Schwille, F., Migration of Organic Fluids Immiscible with
Water in the Unsaturated Zone, in, Pollutants in Porous
Media: The Unsaturated Zone Between Soil Surface and
Groundwater. (B. Yaron, G. Dagan, J. Goldshmid, Eds.)
Springer-Verlag, New York, pp. 27-48, 1984.
47. Schwille, F.. Dense Chlorinated Solvents in Porous and
Fractured Media: Model Experiments (English
Translation), Lewis Publishers, Ann Arbor, Ml 1988.
48. Seitz, W.R., !n-Silu Detection of Contaminant Plumes in
Ground Water, Special Report 90-27, U.S. Army Corps of
Engineers, Cold Regions Research & Engineering
Laboratory, August 1990, 12 pp.
49. Silka, L, Simulation of Vapor Transport Through the
Unsaturated Zone - Interpretation of Soil-Gas Surveys,
Ground Water Monitoring Review, pp. 115-123, Spring
1988.
50. Sitar, N., J.R. Hunt, and K.S. Udell, Movement of
Nonaqueous Liquids in Groundwater, in, Proceedings of
a Speciality Conference. Geotechnical Practice for Waste
Disposal '87. University of Michigan, Ann Arbor, Ml, pp.
205-223, June 15-17, 1987.
51. Sims, R., Soil Remediation Techniques at Uncontrolled
Hazardous Waste Sites, Air & Waste Management
Association. Vol. 40, No. 5, pp. 704-732, May 1990.
52. Thompson, G., and Marrin, D., Soil Gas Contaminant
Investigations: A Dynamic Approach, Ground Water
Monitoring Review, pp. 88-93, Summer, 1987.
53. Treiber, L.E., D.L. Archer, and W.W. Owens, A
Laboratory Evaluation of Wettability of Fifty Oil-Producing
Reservoirs, Society of Petroleum Engineering Journal.
12(6), 531-540.
54. Tuck, D.M., P.R. Jaffe, and D.A. Crerar, Enhancing
Recovery of Immobile Residual Non-Wetting
Hydrocarbons from the Unsaturated Zone Using
Surfactant Solutions, in, Proceedings of Petroleum j
Hydrocarbons and Organic Chemicals in Ground Water.
Prevention. Detection, and Restoration. A Conference
20
-------
and Exposition, The Westin Galleria, Houston, Texas,
Vol. 1, pp. 457-478, November 9-11,1988.
55. U.S. EPA,, Treatment of Contaminated Soils with
Aqueous Surfactants, EPA/600/2-85/129, NTIS PB86-
122561,84pp., 1985.
56. U.S. EPA, Handbook Remedial Action at Waste Disposal
Sites, EPA/625/6-85/006, October, 1985.
57. U.S. EPA, RCRA Ground-Water Monitoring Technical
Enforcement Guidance Document (TEGD), OSWER
Directive 9950.1, 1986c.
58. U.S. EPA, Microbial Decomposition of Chlorinated
Aromatic Compounds, EPA/600/2-86/090, September
1986.
59. U.S. EPA, Characterization and Laboratory Soil
Treatability Studies for Creosote and Pentachlorophenol
Sludges and Contaminated Soil, EPA/600/2-88/055 or
NTIS Publication #PB89-109920, 138 p., 1988.
60. U.S. EPA, Ground Water Modeling: An Overview and
Status Report, EPA/600/2-89/028, December, 1988.
61. U.S. EPA, Contaminant Transport in Fractured Media:
Models for Decision Makers, EPA/600/SF-88/002,
October, 1988.
62. U.S. EPA, Characterization and Laboratory Soil
Treatability Studies for Creosote and Pentachlorophenol
Sludges and Contaminated Soil, EPA/600/2-88/055,
September 1988.
63. U.S. EPA, Bioremediation of Contaminated Surface
Soils, EPA-600/9-89/073, 23 pp., August 1989.
64. U.S. EPA, Performance Evaluations Of Pump And Treat
Remediations, Superfund Ground Water Issue, EPA/540/
4-89/005, 19 pp.,1989.
65. U.S. EPA, Subsurface Contamination Reference Guide,
EPA/540/2- 90/011, October, 1990.
66. U.S. EPA, Laboratory Investigation of Residual Liquid
Organics from Spills, Leaks, and Disposal of Hazardous
Wastes in Groundwater, EPA/600/6-90/004, April, 1990.
67. U.S. EPA, Basics of Pump and Treat Ground Water
Remediation Technology, EPA-600/8-90/003, 31 pp.,
March 1990.
68. U.S. EPA, Site Characterizations for Subsurface
Remediations, EPA/625/ - /(in press) 1990.
69. U.S. Federal Highway Administration, Guidelines for
Cone Penetration Test: Performance and Design,
FHWA-T5-78-209 (TS 78 No. 209) February, 1977.
70. Villaume, J.F., P.C. Lowe, and D.F. Unites, Recovery of
Coal Gasification Wastes: An Innovative Approach, in,
Proceedings Third National Symposium on Aquifer
Restoration and Ground Water Monitoring. National
Water Well Association, Worthington, OH, pp. 434-445,
1983.
71. Villaume, J.F., Investigations at Sites Contaminated with
Dense, Non-Aqueous Phase Liquids (NAPLs), Ground
Water Monitoring Review. Vol. 5, No. 2, pp. 60-74,1985.
72. Waterloo Centre for Ground Water Research, University
of Waterloo Short Course, "Dense Immiscible Phase
Liquid Contaminants in Porous and Fractured Media,"
Kitchener, Ontario, Canada, Nov. 6-9, 1989.
73. Williams, D.E. and D.G. Wilder, Gasoline Pollution of a
Ground- Water Reservoir - A Case History, Ground
Water. Vol. 9, No. 6, pp. 50- 54,1971.
74. Wilson, J.L and S.H. Conrad, Is Physical Displacement
of Residual Hydrocarbons a Realistic Possibility in
Aquifer Restoration?, in, Proceedings of the NWWA/API
Conference on Petroleum Hydrocarbons and Organic
Chemicals in Ground Water—Prevention. Detection, and
Restoration. The Intercontinental Hotel, Houston, Texas,
pp. 274-298, November 5-7,1984.
75. Wisniewski, G.M., G.P. Lennon, J.F. Villaume, and C.L.
Young, Response of a Dense Fluid Under Pumping
Stress, in, Proceedings of the 17th Mid-Atlantic Industrial
Waste Conference. Lehigh, University, pp. 226-237,
1985.
21
.S. GOVERNMENT PRINTING OFFICE: 1991 - 548-187/25613
-------
United States
Environmental Protection
Agency
Office of
Research and
Development
Office of Solid Waste
and Emergency
Response
EPA/540/S-92/005
April1992
&EPA Ground Water Issue
Fundamentals of Ground-Water Modeling
Jacob Bear8, Milovan S. Beljinb, and Randall R. Rossc
Ground-water flow and contaminant transport modeling has
been used at many hazardous waste sites with varying
degrees of success. Models may be used throughout all
phases of the site investigation and remediation processes.
The ability to reliably predict the rate and direction of ground-
water flow and contaminant transport is critical in planning
and implementing ground-water remediations. This paper
presents an overview of the essential components of ground-
water fbw and contaminant transport modeling in saturated
porous media. While fractured rocks and fractured porous
Emay behave like porous media with respect to many
ind contaminant transport phenomena, they require a
ate discussion and are not included in this paper.
Similarly, the special features of flow and contaminant
transport in the unsaturated zone are also not included. This
paper was prepared for an audience with some technical
background and a basic working knowledge of ground-water
flow and contaminant transport processes. A suggested
format for ground-water modeling reports and a selected
bibliography are included as appendices A and B,
respectively.
For further information, contact David Burden (FTS 700-743-
2294), Randall Ross (FTS 700-743-2355), or Joe Williams
(FTS 700-743-2312) at (405) 332-8800.
Modeling as a Management Tool
The management of any system means making decisions
aimed at achieving the system's goals, without violating
specified technical and nontechnical constraints imposed on
it. In a ground-water system, management decisbns may be
related to rates and location of pumping and artificial
recharge, changes in water quality, location and rates of
pumping in pump-and-treat operations, etc. Management's
objective function should be to evaluate the time and cost
necessary to achieve remediation goals. Management
decisions are aimed at minimizing this cost while maximizing
the benefits to be derived from operating the system.
The value of management's objective function (e.g., minimize
cost and maximize effectiveness of remediation) usually
depends on both the values of the decision variables (e.g.,
area! and temporal distributions of pumpage) and on the
response of the aquifer system to the implementation of
these decisions. Constraints are expressed in terms of future
values of state variables of the considered ground-water
system, such as water table elevations and concentrations of
specific contaminants in the water. Typical constraints may
be that the concentration of a certain contaminant should not
exceed a specified value, or that the water level at a certain
location should not drop below specified levels. Only by
comparing predicted values with specified constraints can
decision makers conclude whether or not a specific constraint
has been violated.
An essential part of a good decision-making process is that
the response of a system to the implementation of
contemplated decisions must be known before they are
implemented.
In the management of a ground-water system in which
decisions must be made with respect to both water quality
and water quantity, a tool is needed to provide the decision
maker with information about the future response of the
" Technion - Israel Institute of Technology
''University of Cincinnati
CU.S. EPA, Roberts. Kerr Environmental Research Laboratory
Superfund Technology Support Center for
Ground Water
Robert S. Kerr Environmental
Research Laboratory
Ada, Oklahoma
Tac*i¥»0gy Innovation Offioa
Qfficaaf $of«J Waste and Etnergency
WaltsrW.KovaScMr,.
Printed on Recycled Paper
-------
system to the effects of management decisions. Depending
on the nature of the management problem, decision
variables, objective functions, and constraints, the response
may take the form of future spatial distributions of
contaminant concentrations, water levels, etc. This tool is the
model.
Examples of potential model applications include:
• Design and/or evaluation of pump-and-treat systems
• Design and/or evaluation of hydraulic containment
systems
• Evaluation of physical containment systems (e.g.,
slurry walls)
• Analysis of "no action" alternatives
• Evaluation of past migration patterns of contaminants
• Assessment of attenuation/transformation processes
• Evaluation of the impact of nonaqueous phase liquids
(NAPL) on remediation activities (dissolution studies)
What Is a Ground-Water Model?
A model may be defined as a simplified version of a real-
world system (here, a ground-water system) that
approximately simulates the relevant excitation-response
relations of the real-world system. Since real-world systems
are very complex, there is a need for simplification in making
planning and management decisions. The simplification is
introduced as a set of assumptions which expresses the
nature of the system and those features of its behavior that
are relevant to the problem under investigation. These
assumptions will relate, among other factors, to the geometry
of the investigated domain, the way various heterogeneities
will be smoothed out, the nature of the porous medium (e.g.,
its homogeneity, isotropy), the properties of the fluid (or
fluids) involved, and the type of flow regime under
investigation. Because a model is a simplified version of a
real-world system, no model is unique to a given ground-
water system. Different sets of simplifying assumptions will
result in different models, each approximating the
investigated ground-water system in a different way. The
first step in the modeling process is the construction of a
conceptual model consisting of a set of assumptions that ,
verbally describe the system's composition, the transport
processes that take place in it, the mechanisms that govern
them, and the relevant medium properties. This is envisioned
or approximated by the modeler for the purpose of
constructing a model intended to provide information for a
specific problem.
Content of a Conceptual Model
The assumptions that constitute a conceptual model should
relate to such items as:
• the geometry of the boundaries of the investigated
aquifer domain;
• the kind of solid matrix comprising the aquifer (with
reference to its homogeneity, isotropy, etc.);
• the mode of flow in the aquifer (e.g., one-dimensional,
two-dimensional horizontal, or three-dimensional);
• the flow regime (laminar or nonlaminar);
• the properties of the water (with reference to its
homogeneity, compressibility, effect of dissolved solids
and/or temperature on density and viscosity, etc.);
• the presence of assumed sharp fluid-fluid boundaries,
such as a phreatic surface;
• the relevant state variables and the area, or volume,
over which the averages of such variables are taken;
• sources and sinks of water and of relevant
contaminants, within the domain and on its boundaries
(with reference to their approximation as point sinks
and sources, or distributed sources);
• initial conditions within the considered domain; and
• the conditions on the boundaries of the considered
domain that express the interactions with its
surrounding environment.
Selecting the appropriate conceptual model for a given
problem is one of the most important steps in the modeling
process. Oversimplification may lead to a model that lacks
the required information, while undersimplification may result
in a costly model, or in the lack of data required for model
calibration and parameter estimation, or both. It is, therefore,
important that all features relevant to a considered problem
be included in the conceptual model and that irrelevant ones
be excluded.
The selection of an appropriate conceptual model and the
degree of simplification in any particular case depends on:
• the objectives of the management problem;
• the available resources;
• the available field data;
• the legal and regulatory framework applying to the
situation.
The objectives dictate which features of the investigated
problem should be represented in the model, and to what
degree of accuracy. In some cases averaged water levels
taken over large areas may be satisfactory, while in others
water levels at specified points may be necessary. Natural
recharge may be introduced as monthly, seasonal or annual
averages. Pumping may be assumed to be uniformly
distributed over large areas, or it may be represented as
point sinks. Obviously, a more detailed model is more costly
and requires more skilled manpower, more sophisticated
codes and larger computers. It is important to select the
appropriate degree of simplification in each case.
Selection of the appropriate conceptual model for a given
problem is not necessarily a conclusive activity at the initial
stage of the investigations. Instead, modeling should be
considered as a continuous activity in which assumptions are
reexamined, added, deleted and modified as the
investigations continue. It is important to emphasize that the
availability of field data required for model calibration and
parameter estimation dictates the type of conceptual model
to be selected and the degree of approximation involved.
The next step in the modeling process is to express the
(verbal) conceptual model in the form of a mathematical
model. The solution of the mathematical model yields the
required predictions of the real-world system's behavior in
response to various sources and/or sinks.
Most models express nothing but a balance of the considered
extensive quantity (e.g., mass of water or mass of solute). \m
the continuum approach, the balance equations are written
"at a point within the domain," and should be interpreted to
-------
ean "per unit area, or volume, as the case may be, in the
:in'rty of the point." Under such conditions, the balance
es the form of a partial differential equation. Each term in
that equation expresses a quantity added per unit area or per
unit volume, and per unit time. Often, a number of extensive
quantities of interest are transported simultaneously; for
example, mass of a number of fluid phases with each phase
containing more than one relevant species. The
mathematical model will then contain a balance equation for
each extensive quantity.
Content of a Mathematical Model
The complete statement of a mathematical model consists of
the following items:
• a definition of the geometry of the considered domain
and its boundaries;
• an equation (or equations) that expresses the balance
of the considered extensive quantity (or quantities);
• flux equations that relate the flux(es) of the considered
extensive quantity(ies) to the relevant state variables
of the problem;
• constitutive equations that define the behavior of the
fluids and solids involved;
• an equation (or equations) that expresses initial
conditions that describe the known state of the
considered system at some initial time; and
• an equation (or equations) that defines boundary
conditions that describe the interaction of the
considered domain with its environment.
Ill the equations must be expressed in terms of the
Dependent variables selected for the problem. The selection
of the appropriate variables to be used in a particular case
depends on the available data. The number of equations
included in the model must be equal to the number of
dependent variables. The boundary conditions should be
such that they enable a unique, stable solution.
The most general boundary condition for any extensive
quantity states that the difference in the normal component of
the total flux of that quantity, on both sides of the boundary,
_ is equal to the strength of the source of that quantity. If a
source does not exist, the statement reduces to an equality
of the normal component of the total flux on both sides of the
boundary. In such equalities, the information related to the
external side must be known (Bear and Verruijt, 1987). It is
obtained from field measurements or on the basis of past
experience.
The mathematical model contains the same information as
the conceptual one, but expressed as a set of equations
which are amenable to analytical and numerical solutions.
Many mathematical models have been proposed and
published by researchers and practitioners (see Appendix B).
They cover most cases of flow and contaminant transport in
aquifers encountered by hydrologists and water resource
managers. Nevertheless, it is important to understand the
procedure of model development.
jlhe following section introduces three fundamental
Assumptions, or Hems, in conceptual models that are always
made when modeling ground-water flow and contaminant
transport and fate.
The Porous Medium as a Continuum
A porous medium is a continuum that replaces the real,
complex system of solids and voids, filled with one or more
fluids, that comprise the aquifer. Inability to model and solve
problems of water flow and contaminant transport within the
void space is due to the lack of detailed data on its
configuration. Even if problems could be described and
solved at the microscopic level, measurements cannot be
taken at that level (i.e., at a point within the void space), in
order to validate the model. To circumvent this difficulty, the
porous medium domain is visualized as a continuum with fluid
or solid matrix variables defined at every point. Not only is
the porous medium domain as a whole visualized as a
continuum, but each of the phases and components within it
is also visualized as a continuum, with all continua
overlapping each other within the domain.
The passage from the microscopic description of transport
phenomena to a macroscopic one is achieved by introducing
the concept of a representative elementary volume (REV) of
the porous medium domain. The main characteristic of an
REV is that the averages of fluid and solid properties taken
over it are independent of its size. To conform to this
definition, the REV should be much larger than the
microscopic scale of heterogeneity associated with the
presence of solid and void spaces, and much smaller than
the size of the considered domain. With this concept of an
REV in mind, a porous medium domain can be defined as a
portion of space occupied by a number of phases: a solid
phase (i.e., the solid matrix), and one or more fluid phases,
for which an REV can be found.
Thus, a macroscopic value at a point within a porous medium
domain is interpreted as the average of that variable taken
over the REV centered at that point. By averaging a variable
over all points within a porous medium domain, a continuous
field of that variable is obtained.
By representing the actual porous medium as a continuum,
the need to know the detailed microscopic configuration of
the void space is circumvented. However, at the macroscopic
level, the complex geometry of the void-solid interface is
replaced by various solid matrix coefficients, such as
porosity, permeability and dispersivrty. Thus, a coefficient
that appears in a macroscopic model represents the relevant
effect of the microscopic void-space configuration.
In practice, all models describing ground-water flow and
contaminant transport are written at the continuum, or
macroscopic level. They are obtained by averaging the
corresponding microscopic models over the REV. This
means that one must start by understanding phenomena that
occur at the microscopic level, (e.g., on the boundary
between adjacent phases) before deriving the macroscopic
model. For most models of practical interest, this has
already been done and published.
Horizontal Two-Dimensional Modeling
A second fundamental approximation often employed in
dealing with regional problems of flow and contaminant
transport is that ground-water flow is essentially horizontal.
The term "regional" is used here to indicate a relatively large
aquifer domain. Typically, the horizontal dimension may be
-------
from tens to hundreds of kilometers with a thickness of tens to
hundreds of meters.
In principle, ground-water flow and contaminant transport in a
porous medium domain are three-dimensional. However,
when considering regional problems, one should note that
because of the ratio of aquifer thickness to horizontal length,
the flow in the aquifer is practically horizontal. This
observation also remains valid when small changes exist in
the thickness of a confined aquifer, or in the saturated
thickness of an unconfined aquifer. On the basis of this
observation, the assumption that ground-water flow is
essentially horizontal is often made and included in the
conceptual model. This leads to an aquifer flow model
written in horizontal two dimensions only. Formally, the two-
dimensional horizontal flow model is obtained by integrating
the corresponding three-dimensional variable over the
aquifer's thickness. This procedure is known as the hydraulic
approach. The two-dimensional horizontal flow model is
written in terms of variables which are averaged over the
vertical thickness of the aquifer and thus are a function of the
horizontal coordinates only.
In the process of transforming a three-dimensional problem
into a two-dimensional one, new aquifer transport and
storage coefficients (e.g., aquifer transmissivity and
storativity) appear. In addition to the advantage of having to
solve a two-dimensional rather than a three-dimensional
mathematical model, fewer field observations may be
required for the determination of these coefficients.
Whenever justified on the basis of the geometry (i.e.,
thickness versus horizontal length) and the flow pattern, the
assumption of essentially horizontal flow greatly simplifies the
mathematical analysis of the flow in an aquifer. The error
introduced by this assumption is small in most cases of
practical interest.
The assumption of horizontal flow fails in regions where the
flow has a large vertical component (e.g., in the vicinity of
springs, rivers or partially penetrating wells). However, even
in these cases, at some distance from the source or sink, the
assumption of horizontal flow is valid again. As a general
rule, one may assume horizontal flow at distances larger than
1.5 to 2 times the thickness of the aquifer in that vicinity
(Bear, 1979). At smaller distances the flow is three-
dimensional and should be treated as such.
The assumption of horizontal flow is also applicable to leaky
aquifers. When the hydraulic conductivity of the aquifer is
much larger than that of the semipermeable layer, and the
aquifer thickness is much larger than the thickness of the
aquitard, it follows from the law of refraction of streamlines
("tangent law") that the flow in the aquifer is essentially
horizontal, while it is practically vertical in the aquitards (de
Marsily, 1986).
When considering contaminant transport in aquifers, the
model user must be cautious in attempting to utilize a two-
dimensional model, because in most cases the hydraulic
approach is not justified. The contaminant may be trans-
ported through only a small fraction of the aquifer's thickness.
In addition, velocities in different strata may vary appreciably
in heterogeneous aquifers, resulting in a marked difference in
the rates of advance and spreading of a contaminant.
Momentum Balance
The third concept relates to the fluid's momentum balance.
In the continuum approach, subject to certain simplifying
assumptions included in the conceptual model, the
momentum balance equation reduces to the linear motion
equation known as Darcy's law. This equation is used as a
flux equation for fluid flow in a porous medium domain. With
certain modifications, it is also applicable to multiphase flows
(e.g., air-water flow in the unsaturated zone).
Major Balance Equations
The following examples of major balance equations constitute
the core of models that describe fbw and contaminant
transport in porous medium domains. A number of
simplifying assumptions must be stated before any of these
equations can be written. Although these assumptions are
not listed here, they must be included in the conceptual
model of the respective cases.
Mass balance for 3-D saturated flow In a porous medium
domain:
d(p
dt
• (K- V
(x,y,z,t).
Thus, equation (1) states that the excess of inflow over
outflow of water in a unit volume of porous medium, per unit
time, at a point, is equal to the rate at which water volume is
being stored, where storage is due to fluid and solid matrix
compressibilities.
Mass balance for 2-D saturated flow In a confined
aquifer:
V« (T- V
-------
T - aquifer transmissivity tensor
P (x,y,t ) * rate of pumping (per unit area of aquifer)
R (x.y.t ) - rate of recharge (per unit area of aquifer)
The storativrty. S, is defined as the volume of water added to
storage in a unit area of aquifer, per unit rise of piezometric
head. Hence, the left side of equation (2) expresses the
volume of water added to storage in the aquifer, per unit area
per unit time. The divergence of a flux vector, («-T • Vq>),
expresses the excess of outflow over inflow per unit area, per
unit time. Note that here, the two operators are in the two-
dimensional horizontal coordinates, and the variable to be
solved is
*« Ve, J*= -D« Vc.
.where D - coefficient of dispersion
2? * - coefficient of molecular diffusion
in a porous medium
Model Coefficients and Their Estimation
In passing from the microscopic level of describing transport to
the macroscopic level, various coefficients of transport and
storage are introduced. The permeability of a porous medium,
aquifer transmissivity, aquifer storativrty, and porous medium
dispersivrty, may serve as examples of such coefficients.
Permeability and dispersivrty are examples of coefficients that
express the macroscopic effects of microscopic configuration
of the solid-fluid interfaces of a porous medium. They are
introduced in the passage from the microscopic level of
description to the macroscopic, continuum, level. The
coefficients of aquifer storativrty and transmissivity are
introduced by the further averaging of the three-dimensional
macroscopic model over the thickness of an aquifer in order to
obtain a two-dimensional model. All these coefficients are
coefficients of the models, and therefore, in spite of the
similarity in their names in different models, their interpretation
and actual values may differ from one model to the next.
This point can be illustrated by the following example. To
obtain the drawdown in a pumping well and in its vicinity, one
employs a conceptual model that assumes radially
converging flow to an infinitesimally small well in a
homogeneous, isotropic aquifer of constant thickness and of
infinite areal extent. The same model is used to obtain the
aquifer's storativity and transmissivity by conducting an
aquifer pumping test and solving the model's equation for
these coefficients, ft is common practice to refer to these
coefficients as the aquifer's coefficients and not as
coefficients of the aquifer's model. However, it is important to
realize that the coefficients thus derived actually correspond
to that particular model. These coefficients should not be
employed in a model that describes the flow in a finite
heterogeneous aquifer with variable thickness and with non-
radial flow in the vicinity of the well. Sometimes, however,
there is no choice because this is the only information
available. Then, the information must be used, keeping in
mind that when coefficients are derived by employing one
model in another model for a given domain, the magnitude of
the error will depend on the differences between the two
models. In principle, in order to employ a particular model,
the values of the coefficients appearing in it should be
determined using some parameter identification technique for
that particular model.
Obviously, no model can be employed in any specified
domain unless the numerical values of all the coefficients
appearing in it are known. Estimates of natural recharge and
a priori location and type of boundaries may be included in
the list of model coefficients and parameters to be identified.
The activity of identifying these model coefficients is often
referred to as the identification problem.
In principle, the only way to obtain the values of the
coefficients for a considered model is to start by investigating
the real-world aquifer system to find a period (or periods) in
the past for which information is available on (/') initial
conditions of the system; (//) excitations of the system, as in
the form of pumping and artificial recharge (quality and
quantity), natural recharge introduction of contaminants, or
changes in boundary conditions, and (///') observations of the
response of the system, as in the form of temporal and spatial
distributions of water levels and solute concentrations. If such
-------
a period (or periods) can be found, one can (/) impose the
known initial conditions on the model, (//') excite the model by
the known excitations of the real system, and (Hi) derive the
response of the model to these excitations. Obviously, in order
to derive the model's response, one has to assume some trial
values for the coefficients and compare the response
observed in the real system with that predicted by the model.
The sought values of the coefficients are those that will make
the two sets of values of state variables identical. However,
because the model is only an approximation of the real
system, one should never expect these two sets of values to
be identical. Instead, the "best fit" between them must be
sought according to some criterion. Various techniques exist
for determining the "best" or "optimal" values of these
coefficients (i.e., values that will make the predicted values
and the measured ones sufficiently close to each other).
Obviously, the values of the coefficients eventually accepted
as "best" for the model depend on the criteria selected for
"goodness of fit" between the observed and predicted values
of the relevant state variables. These, in turn, depend on the
objective of the modeling.
Some techniques use the basic trial-and-error method
described above, while others employ more sophisticated
optimization methods. In some methods, a priori estimates
of the coefficients, as well as information about lower and
upper bounds, are introduced. In addition to the question of
selecting the appropriate criteria, there remains the question
of the conditions under which the identification problem, also
called the inverse problem, will result in a unique solution.
As stressed above, no model can be used for predicting the
behavior of a system unless the numerical values of its
parameters have been determined by some identification
procedure. This requires that data be obtained by field
measurements. However, even without such data, certain
important questions about the suitability of the model can be
studied. Sensitivity analysis enables the modeler to
investigate whether a certain percentage change in a
parameter has any real significance, that is whether it is a
dominant parameter or not. The major point to be established
from a sensitivity analysis is the relative sensitivity of the
model predictions to changes in the values of the model
parameters within the estimated range of such changes.
A successful model application requires appropriate site
characterization and expert insight into the modeling process.
Figure 1 illustrates a simple diagram of a model application
process. Each phase of the process may consist of various
steps; often, results from one step are used as feedback in
previous steps, resulting in an iterative procedure (van der
Heijdeetal.. 1989).
Formulation of Objective*
Review and Interpretation
of Available Data
Model Conceptualization
Code Selection
More Data
Needed
Reid Data Collection
Input Data Preparation
Calibration and Sensitivity Analysis
Predictive Runs
>
r
Uncertainty Analysis |
Improve
Conceptual
Model
Figure 1. Model Application Process.
-------
Methods of Solution
'Once a well-posed model for a given problem has been
.constructed, including the numerical values of all the
coefficients that appear in the model, it must be solved for
any given set of excitations (i.e., initial and boundary
conditions, sources and sinks). The preferable method of
solution is the analytical one, as once such a solution is
derived, it can be used for a variety of cases (e.g., different
values of coefficients, different pumping rates, etc). However,
for most cases of practical interest, this method of solution is
not possible due to the irregularity of the domain's shape, the
heterogeneity of the domain with respect to various
coefficients, and various nonlinearrties. Instead, numerical
models are employed.
Although a numerical model is derived from the mathematical
model, a numerical model of a given problem need not
necessarily be considered as the numerical method of
solution, but as a model of the problem in its own right. By
adding assumptions to the conceptual model of the given
problem (e.g., assumptions related to time and space
discretization) a new conceptual model is obtained which, in
turn, leads to the numerical model of the given problem.
Such a model represents a different approximate version of
the real system.
Even those who consider a numerical model as a model in its
own right very often verify it by comparing the model results
with those obtained by an analytical solution of the
corresponding mathematical model (for relatively simple
cases for which such solutions can be derived). One of the
main reasons for such a verification is the need to eliminate
errors resulting from the numerical approximations alone.
Until the early 1970s, physical (e.g., sand box) and analog
(e.g., electrical) laboratory models were used as practical
tools for solving the mathematical models that described
ground-water flow problems. With the introduction of
computers and their application in the solution of numerical
models, physical and analog models have become
cumbersome as tools for simulating ground-water regimes.
However, laboratory experiments in soil columns are still
needed when new phenomena are being investigated and to
validate new models (i.e., to examine whether certain
assumptions that underlie the model are valid).
Analytical Models
During the early phase of a ground-water contamination
study, analytical models offer an inexpensive way to evaluate
the physical characteristics of a ground-water system. Such
models enable investigators to conduct a rapid preliminary
analysis of ground-water contamination and to perform
sensitivity analysis. A number of simplifying assumptions
regarding the ground-water system are necessary to obtain
an analytical solution. Although these assumptions do not
necessarily dictate that analytical models cannot be used in
"real-life" situations, they do require sound professional
judgment and experience in their application to field
situations. Nonetheless, it is also true that in many field
situations few data are available; hence, complex numerical
models are often of limited use. When sufficient data have
jbeen collected, however, numerical models may be used for
'predictive evaluation and decision assessment. This can be
done during the later phase of the study. Analytical models
should be viewed as a useful complement to numerical
models.
For more information on analytical solutions, the reader is
referred to Bear (1979), van Genuchten and Alves (1982),
and Walton (1989).
Numerical Models
Once the conceptual model is translated into a mathematical
model in the form of governing equations, with associated
boundary and initial conditions, a solution can be obtained by
transforming it into a numerical model and writing a computer
program (code) for solving it using a digital computer.
Depending on the numerical technique(s) employed in solving
the mathematical model, there exist several types of
numerical models:
• finite-difference models
• finite-element models
• boundary-element models
• particle tracking models
- method-of-characterislics models
- random walk models, and
• integrated finite-difference models.
The main features of the various numerical models are:
1. The solution is sought for the numerical values of state
variables only at specified points in the space and time
domains defined for the problem (rather than their
continuous variations in these domains).
2. The partial differential equations that represent
balances of the considered extensive quantities are
replaced by a set of algebraic equations (written in
terms of the sought, discrete values of the state
variables at the discrete points in space and time).
3. The solution is obtained for a specified set of
numerical values of the various model coefficients
(rather than as general relationships in terms of these
coefficients).
4. Because of the large number of equations that must
be solved simultaneously, a computer program is
prepared.
In recent years, codes have been developed for almost all
classes of problems encountered in the management of
ground water. Some codes are very comprehensive and can
handle a variety of specific problems as special cases, while
others are tailor-made for particular problems. Many of them
are available in the public domain, or for a nominal fee. More
recently, many codes have been developed or adapted for
microcomputers.
Uncertainty
Much uncertainty is associated with the modeling of a given
problem. Among them, uncertainties exist in
• the transport mechanisms;
• the various sink/source phenomena for the considered
extensive quantity;
• the values of model coefficients, and their spatial (and
sometimes temporal) variation;
-------
• initial conditions;
• the location of domain boundaries and the conditions
prevailing on them;
• the meaning of measured data employed in model
calibration; and
• the ability of the model to cope with a problem in which
the solid matrix heterogeneity spans a range of scales,
sometimes orders of magnitude apart.
The degree of uncertainty is increased in most cases by the
lack of sufficient data for parameter estimation and model
validation. Errors in observed data used for parameter
identification also contribute to uncertainty in the estimated
values of model parameters.
Various methods for introducing uncertainty into the models
and the modeling process have been proposed. For
example, one approach is to employ Monte Carlo methods in
which the various possibilities are represented in a large
number of simulated realizations. Another approach is to
construct stochastic models in which the various coefficients
are represented as probability distributions rather than
deterministic values.
Often the question is raised as to whether, in view of all these
uncertainties, which always exist in any real-world problem,
models should still be regarded as reliable tools for providing
predictions of real-world behavior--f/?ere is no alternative!
However, the kind of answers models should be expected to
provide and the very objectives of employing models, should
be broadened beyond the simple requirement that they
provide the predicted response of the system to the planned
excitations. Stochastic models provide probabilistic
predictions rather than deterministic ones. Management must
then make use of such predictions in the decision-making
process. Methodologies for evaluating uncertainties will have
to be developed; especially methods for evaluating the worth
of data in reducing uncertainty. It then becomes a
management decision whether or not to invest more in data
acquisition.
In view of the uncertainty involved in modeling, models
should be used for additional roles, beyond predicting or
estimating the deterministic or probabilistic responses to
planned excitations. Such roles include:
• predicting trends and direction of changes;
• providing informatbn on the sensitivity of the system
to variatbns in various parameters, so that more
resources can be allocated to reduce their uncertainty;
• deepening our understanding of the system and of the
phenomena of interest that take place within it; and
• improving the design of observation networks.
Many researchers are currently engaged in developing
methods that incorporate the element of uncertainty in both
the forecasting and the inverse models (Freeze et al., 1989;
Gelhar, 1986; Yeh, 1986; Neuman et al., 1987, and others).
Model Misuse
As stated above, the most crucial step in ground-water
modeling is the development of the conceptual model. If the
conceptual model is wrong (i.e., does not represent the
relevant flow and contaminant transport phenomena), the
rest of the modeling efforts — translating the conceptual
model into mathematical and numerical models, and solving
for cases of interest — are a waste of time and money.
However, mistakes and misuses may occur during any phase
of the modeling process (Mercer, 1991).
Following is a list of the more common misuses and mistakes
related to modeling. They may be divided into four
categories (Mercer and Faust, 1981):
1. Improper conceptualization of the considered problem:
• improper delineation of the model's domain;
• wrong selection of model geometry: a 2-D
horizontal model, or a 3-D model;
• improper selection of boundary conditions;
• wrong assumptions related to homogeneity and
isotropy of aquifer material;
• wrong assumptions related to the significant
processes, especially in cases of contaminant
transport. These may include the type of sink/source
phenomena, chemical and biological transformations,
fluid-solid interactions, etc.; and
• selecting a model that involves coefficients that vary in
space, but for which there are insufficient data for
model calibration and parameter estimation.
2. Selection of an inappropriate code for solving the
model:
• selecting a code much more powerful/versatile than
necessary for the considered problem;
• selection of a code that has not been verified and
tested.
3. Improper model application:
• selection of improper values for model parameters and
other input data;
• misrepresentation of aqurtards in a multi-layer system;
mistakes related to the selection of grid size and time
steps;
• making predictions with a model that has been
calibrated under different conditions;
• making mistakes in model calibration (history
matching); and
• improper selection of computational parameters
(closure criterion, etc.).
4. Misinterpretation of model results:
• wrong hydrobgical interpretation of model results;
• mass balance is not achieved.
Sources of Information
In selecting a code, its applicability to a given problem and its
efficiency in solving the problem are important criteria. In
evaluating a code's applicability to a problem and its
efficiency, a good description of these characteristics should
be accessible. For a large number of ground-water models,
such information is available from the International Ground
Water Modeling Center (IGWMC, Institute for Ground-Water
Research and Education, Colorado School of Mines, Golden,
Colorado 80401), which operates a clearinghouse service for
-------
Information and software pertinent to ground-water modeling.
Information databases have been developed to efficiently
organize, update and access information on ground-water
models for mainframe and microcomputers. The model
annotation databases have been developed and maintained
over the years with major support from U.S. EPA's Robert S.
Kerr Environmental Research Laboratory (RSKERL), Ada,
Oklahoma.
The Center for Subsurface Modeling Support (CSMoS),
located at the RSKERL (P.O. Box 1198, Ada, OK 74820),
provides ground-water and vadose zone modeling software
and services to public agencies and private companies
throughout the nation. CSMoS primarily manages and
supports ground-water models and databases resulting from
research at RSKERL. CSMoS integrates the expertise of
individuals in all aspects of the environmental field in an effort
to apply models to better understand and resolve subsurface
problems. CSMoS is supported internally by RSKERL
scientists and engineers, and externally by the IGWMC,
National Center for Ground Water Research and numerous
ground-water modeling consultants from academia and the
private consulting community.
The National Ground Water Information Center (NGWIC,
6375 Riverside Drive, Dublin, Ohio 43017) is an information-
gathering and dissemination business that performs
customized literature searches on various ground-water-
related topics, and locates and retrieves copies of available
documents. The center maintains its own on-line databases.
Appendix A: Suggested Format for a
Ground-Water Modeling Report
Following is a suggested standardized format for a report that
involves modeling and analysis of model results. The emphasis is
on the modeling efforts and related activities. It is not an attempt
to propose a structure for a project report. The Ground Water
Modeling Section (D-18.21.10) of the American Society of Testing
and Materials (ASTM) Subcommittee on Ground Water and
Vadose Zone Investigators is in the process of developing
standards on ground-water modeling. Additionally, specific infor-
mation regarding the content of ground-water modeling studies is
addressed by Anderson and Woessner (1992, Chapter 9).
Introduction
The introduction may start with a description of the problem that
lead to the investigations. The description will include the domain in
which the phenomena of interest take place, and what decisions
are contemplated in connection with these phenomena The
description should also include information about the geography,
topography, geology, hydrology, dimate, soils, and other relevant
features (of the domain and the considered transport
phenomena). Sources of information should be given. The
description of the problem should lead to the kind of information
mat is required by management/decision maker, which the
investigations described in the report are supposed to provide.
This section should continue to outline the methodology used for
obtaining the required information. In most cases, a model of the
problem domain and the transport (i.e., flow and contaminant)
phenomena will be the tool for providing management with the
required information. On the premise that this section concludes
that such a model is needed, the objective of the report is to
describe the construction of the model, the model runs, and the
results leading to the required information.
Previous Studies
This section should contain a description of earlier relevant studies
in the area, whether on the same problem or in connection with
other problems. The objective of this section is to examine the
data and conclusions in these investigations, as far as they relate
to the current study.
The Conceptual Model
Because the previous section concluded that a model is required,
the objective of this section is to construct the conceptual model of
the problem, including the problem domain and the transport
phenomena taking place within it. The content of a conceptual
model has been outlined in the text. However, the importance of
the conceptual model cannot be overemphasized. It is possible
that the existing data will indicate more than one alternative model,
if the available data (or lack of it) so dictates.
The Mathematical Model
The conceptual model should be translated into a complete, well-
posed mathematical one. At this stage, the various terms that
appear in the mathematical model should be analyzed, with the
objective of deleting non-dominant effects. Further simplifying
assumptions may be added to the original conceptual model at
this stage.
If more than one conceptual model has been visualized, a
corresponding mathematical model should be presented for each.
This section should conclude with a list of coefficients and
parameters that appear in the model. The modeler should then
indicate for which coefficients values, or at least initial ones, are
available (including the actual numerical value and the source of
information), and for which coefficients the required information is
missing. In addition, the kind of field work or exploration required
to obtain that information should be reported. If possible, an
estimate should be given for the missing values, their possible
range, etc. At this stage, it is important to conduct and report a
sensitivity analysis in order to indicate the significance of the
missing information, bearing in mind the kind of information that
the model is expected to provide.
Selection of Numerical Model and Code
The selected numerical model and the reasons for preferring it
over other models (public domain or proprietary) should be
presented. Some of the questions that should be answered are:
Was the code used as is, or was it modified for the purpose of the
project? What were the modifications? If so stated in the contract,
the modified code may have to be included in the appendix to the
report. The full details of the code (name, version, manual, author,
etc.) should be supplied. This section may include a description of
the hardware used in running the code, as well as any other
software (pre- and post-processors). More information about
model selection can be found in Simmons and Cde (1985), Beljin
and van der Heijde (1991).
-------
Model Calibration
Every model must be calibrated before it can be used as a tool for
predicting the behavior of a considered system. During the
calibration phase, the initial estimates of model coefficients may be
modified. The sensitivity analysis may be postponed until a
numerical model and a code for its solution have been selected.
In this section objectives of the calibration or history matching, the
adjusted parameters/coefficients, the criterion of the calibration
(e.g., minimizing the difference between observed and predicted
water levels), the available data, the model calibration runs, etc.,
should be described.
The conclusions should be the modified set of parameters and
coefficients to be used in the model.
Model Runs
Justification and reasoning for the various runs.
Model Results
This section includes all tables and graphic output. Ranges and
uncertainties in model results should be indicated. Results of
sensitivity analysis and the significance of various factors should
also be discussed.
Conclusions
The information required by the decision maker should be clearly
outlined.
Appendices
Tables and graphs, figures, and maps not presented in the body of
the report, along with a list of symbols, references, codes, etc.,
should be included.
Appendix B: Selected Bibliography
Anderson, M.P. 1979. Using Models to Simulate the
Movement of Contaminants through Ground Water Flow
Systems. Critical Reviews in Environmental Control,
v. 9, no. 2, pp. 97-156.
Anderson, M.P. 1984. Movement of Contaminants in
Groundwater: Groundwater Transport - Advection and
Dispersion. In Groundwater Contamination, Studies in
Geophysics, National Academy Press, Washington, D.C.
Anderson, M.P. and W.W. Woessner. 1992. Applied
Groundwater Modeling. Academic Press, Inc. San
Diego, California.
Appel, C.A. and T.E. Reilly. 1988. Selected Reports That
Include Computer Programs Produced by the U.S.
Geological Survey for Simulation of Ground-Water Flow
and Quality. WRI 87-4271. U.S.G.S., Reston. Virginia.
Bear, J. 1972. Dynamics or Fluids in Porous Media.
American Elservier, New York, New York.
Bear, J. 1979. Hydraulics of Groundwater. McGraw-Hill,
New York.
Bear, J. and A. Verruijt. 1987. Modeling Groundwater Flow
and Pollution. Kluwer Academic Publishers, Hingham,
Massachusetts.
Bear, J. and Y. Bachmat. 1990. Introduction to Modeling of
Transport Phenomena in Porous Media. Kluwer
Academic Publishers, Hingham, Massachusetts.
Beck, M.B. 1985. Water Quality Management: A Review of
the Development and Application of Mathematical
Models. NASA 11, Springer-Verlag, Berlin, West
Germany.
Beljin, M.S. and P.K.M. van der Heijde. 1989. Testing,
Verification, and Validation of Two-Dimensional Solute
Transport Models. In (G. Jousma et al., eds.)
Groundwater Contamination: Use of Models in Decision-
Making. Kluwer Academic Publishers, Hingham,
Massachusetts.
Beljin, M.S. and P.K.M. van der Heijde. 1991. Selection of
Groundwater Models for WHPA Delineation. Proc. the
AWWA Computer Conference, Houston, Texas.
Boonstra, J. and N.A. De Ridder. 1981. Numerical Modelling
of Groundwater Basins, International Institute for Land
Reclamation and Improvement, Wageningen, The
Netherlands.
Boutwell, S.H., S.M. Brown, B.R. Roberts, and D.F. Atwood.
1985. Modeling Remedial Actions of Uncontrolled
Hazardous Waste Sites. EPA 540/2-85/001, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
de Marsily, G. 1986. Quantitative Hydrogeology. Academic
Press, Inc., Orlando, Florida.
Domenico, P.A. 1972. Concepts and Models in Groundwater
Hydrology. McGraw-Hill, New York, New York.
Freeze, R.A., G. DeMarsily, L. Smith, and J. Massmann.
1989. Some Uncertainties About Uncertainty. In (B.E.
Buxton, ed.) Proceedings of the Conference
Geostatistical, Sensitivity, and Uncertainty Methods for
Ground-Water Flow and Radbnuclide Transport
Modeling, San Francisco, California, CONF-870971,
Battelle Press, Columbus, Ohio.
Gelhar, L.W., A.L. Gutjahr, and R.L Naff. 1979. Stochastic
Analysis of Macrodispersion in Aquifers. Water
Resources Research, v. 15, no. 6, pp. 1387-1397.
Gelhar, L.W. 1984. Stochastic Analysis of Flow in
Heterogeneous Porous Media. In (J.Bear and M.Y.
Corapcioglu, eds.) Fundamentals of Transport
Phenomena in Porous Media, Marinus Nijhoff Publishers,
Dordrecht, The Netherlands.
Gelhar, L.W. 1986. Stochastic Subsurface Hydrology from
Theory to Applications. Water Resources Research,
v. 22, no. 9, pp. 135S-145S.
Gorelick, S.M. 1983. A Review of Distributed Parameter
Groundwater Management Modeling Methods. Water
Resources Research, v. 19, no. 2, pp. 305-319.
Grove, D.B. and K.G. Stollenwerk. 1987. Chemical Reactions
Simulated by Ground-Water Quality Models. Water
Resources Bulletin, v. 23, no. 4, pp. 601-615.
Herrling, B. and A. Heckele. 1986. Coupling of Finite Element
and Optimization Methods for the Management of
Groundwater Systems. Advances in Water Resources,
v. 9, no. 4, pp. 190-195.
Hunt, B. 1983. Mathematical Analysis of Groundwater
Resources. Butterworths Publishers, Stoneham,
Massachusetts.
Huyakorn, P.S. and G.F. Pinder. 1983. Computational
Methods in Subsurface Flow. Academic Press,
New York.
10
-------
.Huyakorn. P.S.. B.H. Lester, and C.R. Faust. 1983. Finite
Element Techniques for Modeling Ground Water Flow in
Fractured Aquifers. Water Resources Research, v. 19,
no. 4, pp. 1019-1035.
Istok, J. 1989. Groundwater Modeling by the Finite-Element
Method. AGU Water Resources Monograph 13, American
Geophysical Union, Washington, D.C.
Javandel, I., C. Doughty, and C.F. Tsang. 1984.
Groundwater Transport: Handbook of Mathematical
Models. AGU Water Resources Monograph 10, American
Geophysical Union, Washington, D.C.
Keely, J.F. 1987. The Use of Models in Managing Ground-
Water Protection Programs. EPA/600/8-87/003.
Kinzelbach, W. 1986. Groundwater Modeling: An Introduction
with Sample Programs in BASIC. Elsevier Publication
Company, Amsterdam, The Netherlands.
Konikow, L.F. and J.D. Bredehoeft. 1978. Computer Model of
Two-Dimensional Solute Transport and Dispersion in
Ground Water. USGS Techniques of Water-Resources
Investigations. Book 7, Chap. C2, 90 pp.
Ligget, J.A. and P.L-F. Liu. 1983. The Boundary Integral
Equation Method for Porous Media Flow. Allen and
Unwin, Inc., Winchester, Massachusetts.
Mercer, J.W. and C.R. Faust. 1981. Ground-Water Modeling.
National Water Well Association (NWWA), Worthington,
Ohio.
Mercer, J.W., S.D. Thomas, and B. Ross. 1982. Parameters
and Variables Appearing in Repository Siting Models.
Report NUREG/CR-3066, U.S. Regulatory Commission,
Washington, D.C.
Mercer, J.W. 1991. Common Mistakes in Model Applications.
Proc. ASCE Symposium on Ground Water, Nashville,
Tennessee, July 29 - August 2,1991.
Rfaore, J.E. 1979. Contribution of Ground-water Modeling to
Planning. Journal of Hydrology, v. 43 (October), pp.121-
128.
Narasimhan, T.N. and P.A. Witherspoon. 1976. An Integrated
Finite-Difference Method for Analyzing Fluid Flow in
Porous Media. Water Resources Research v. 12, no. 1,
pp. 57-64.
Narasimhan, T.N. 1982. Numerical Modeling in
Hydrogeology. In T.N. Narasimhan (ed.), Recent Trends
in Hydrogeology, pp. 273-296, Special Paper 189
Geological Society of America, Boulder, Colorado.
National Research Council (NRC). 1990. Ground Water
Models: Scientific and Regulatory Application. National
Research Council, Water Science and Technology
Board, Washington, D.C.
Neuman, S.P., C.L. Winter, and C.M. Newman. 1987.
Stochastic Theory of Field-Scale Fickian Dispersion in
Anisotropic Porous Media. Water Resources Research,
v. 23, no. 3, pp. 453-466.
Pickens, J.F. and G.E. Grisak. 1981. Modeling of Scale-
Dependent Dispersion in Hydrologic Systems. Water
Resources Research, v. 17, no. 6, pp. 1701-1711.
Pinder, G.F. and J.D. Bredehoeft. 1968. Application of the
Digital Computer for Aquifer Evaluation. Water
Resources Research, v. 4, no. 5, pp. 1069-1093.
Pinder, G.F. and W.G. Gray. 1977. Finite Element Simulation
In Surface and Subsurface Hydrology. Academic Press,
New York, 295 pp.
.Pinder, G.F. and L. Abriola. 1986. On the Simulation of
Nonaqueous Phase Organic Compounds in the
Subsurface. Water Resources Research, v. 22, no. 9,
pp. 1092-1192.
Prickett, T.A. 1975. Modeling Techniques for Groundwater
Evaluation. In Van Te Chow (ed.), Advances in
Hydroscience, Vol. 10.
Prickett, T.A. and C.G. Lonnquist. 1982. A Random-Walk
Solute Transport Model for Selected Groundwater
Quality Evaluations. Bull. No. 65, Illinois State Water
Survey, Urbana, 105 pp.
Remson, -I., G.M. Hornberger, and F.J. Molz. 1971.
Numerical Methods in Subsurface Hydrology. John Wiley
and Sons, New York, New York.
Rushton, K.R. and S.C. Redshaw. 1979. Seepage and
Groundwater Flow: Numerical Analysis by Anabg and
Digital Methods. John Wiley and Sons, Chichester, U.K.
Schmelling, S.G. and R.R. Ross. 1989. Contaminant
Transport in Fractured Media: Models for Decision
Makers. USEPA Superfund Ground Water Issue Paper.
EPA/540/4-89/004.
Simmons, C.S. and C.R. Cole. 1985. Guidelines for Selecting
Codes for Ground-Water Transport Modeling of Low-
Level Waste Burial Sites. PNL-4980, Volume I and II,
Battelle Pacific Northwest Labs, Richland, Washington.
Strack, O.D.L. 1989. Groundwater Mechanics. Prentice Hall,
Englewood Cliffs, New Jersey.
Trescott, P.C., G.F. Pinder, and S.P. Larson. 1976. Finite-
Difference Model for Aquifer Simulation in Two-
Dimensions with Results of Numerical Experiments.
U.S.G.S. Techniques of Water Resources Investigatbn,
Book 7, Chap. C1, 116pp.
U.S. Office of Technology Assessment. 1982. Use of Models
for Water Resources Management, Planning, and Policy.
U.S. OTA for Congress of the United States, U.S.
Government Printing Office, Washington, D.C.
van der Heijde, P.K.M. et al. 1985. Groundwater
Management: The Use of Numerical Models. 2nd edition,
AGU Water Resources Monograph no. 5, Washington,
D.C.
van der Heijde, P.K.M. and M.S. Beljin. 1988. Model
Assessment for Delineating Wellhead Protection Areas.
EPA/440/6-88-002. Office of Ground Water Protection,
U.S. Environmental Protection Agency, Washington,
D.C.
van der Heijde, P.K.M., A.I. El-Kadi, and S.A. Williams. 1989.
Groundwater Modeling: An Overview and Status Report.
EPA/600/2-89/028. R.S. Kerr Environmental Research
Lab, U.S. Environmental Protection Agency, Ada,
Oklahoma.
van Genuchten, M.Th. and W.J. Alves. 1982. Analytical
Solutions of the One-Dimensional Convective-Dispersive
Solute Transport Equation. U.S. Dept. of Agriculture,
Tech. Bull. No. 1661.
Walton, W. 1985. Practical Aspects of Ground Water
Modeling. Lewis Publishers, Chelsea, Michigan.
Walton, W. 1989. Analytical Ground Water Modeling. Lewis
Publishers, Chelsea, Michigan.
Wang, H.F. and M.P. Anderson. 1982. Introduction to
Groundwater Modeling. W.H. Freeman and Co., San
Francisco, California.
Yeh, W. W-G. 1986. Review of Parameter Identification
Procedures in Groundwater Hydrology: The Inverse
Problem. Water Resources Research, v. 22, no. 2,
pp. 95-108.
-14
. GOVERNMENT PRINTING OFFICE: 1*93 • S5
-------
United States R.S. Kerr • Office of Solid Waste Publication: 9355.4-07FS
Environmental Environmental and Emergency January 1992
Protection Agency Research Laboratory Response
CPA Estimating Potential for Occurrence
of DNAPL at Superfund Sites
Office of Emergency and Remedial Response
Hazardous Site Control Division (OS-220W) Quick Reference Fact Sheet
GOALS
The presence of Dense Nonaqueous Phase Liquids (DNAPL) in soils and aquifers can control the ultimate success or failure
of remediation at a hazardous waste site. Because of the complex nature of DNAPL transport and fate, however, DNAPL
may often be undetected by direct methods, leading to incomplete site assessments and inadequate remedial designs. Sites
affected by DNAPL may require a different "paradigm," or conceptual framework, to develop effective characterization and
remedial actions (2).
To help site personnel determine if DNAPL-based characterization strategies should be employed at a particular site, a
guide for estimating the potential for DNAPL occurrence was developed. The approach,- described in this fact sheet,
requires application of two types of existing site information:
• Historical Site Use Information • Site Characterization Data
By using available data, site decision makers can enter a system of two flowcharts and a classification matrix for estimating
Lite potential for DNAPL occurrence at a site. If the potential for DNAPL occurrence is low, then conventional site
^Kessment and remedial actions may be sufficient. If the potential for DNAPL is moderate or high, however, a different
Henceptual approach may be required to account for problems associated with DNAPL in the subsurface.
BACKGROUND
DNAPLs are separate-phase hydrocarbon liquids that are denser than water, such as chlorinated solvents (either as a single
component or as mixtures of solvents), wood preservative wastes, coal tar wastes, and pesticides. Until recently, standard
operating practice in a variety of industries resulted in the release of large quantities of DNAPL to the subsurface. Most
DNAPLs undergo only limited degradation in the subsurface, and persist for long periods while slowly releasing soluble
organic constituents to ground water through dissolution. Even with a moderate DNAPL release, dissolution may continue
for hundreds of years or longer under natural conditions before all the DNAPL is dissipated and concentrations of soluble
organics in ground water return to background levels.
DNAPL exists in the soil/aquifer matrix as free-phase DNAPL and residual DNAPL. When released at the surface, free-
phase DNAPL moves downward through the soil matrix under the force of gravity or laterally along the surface of sloping
fine-grained stratigraphic units. As the free-phase DNAPL moves, blobs or ganglia are trapped in pores and/or fractures by
.capillary forces (7). The amount of the trapped DNAPL, known as residual saturation, is a function of the physical
properties of the DNAPL and the hydrogeologic characteristics of the soil/aquifer medium and typically ranges from 5% to
50% of total pore volume. At many sites, however, DNAPL migrates preferentially through small-scale fractures and
heterogeneities in the soil, permitting the DNAPL to penetrate much deeper than would be predicted from application of
typical residual saturation values (16).
Once in the subsurface, it is difficult or impossible to recover all of the trapped residual DNAPL. The conventional aquifer
E'"*"?diation approach, ground water pump-and-treat, usually removes only a small fraction of trapped residual DNAPL
21, 26). Although many DNAPL removal technologies are currently, being tested, to date there have been no field
onstrations where sufficient DNAPL has been successfully recovered from the subsurface to return the aquifer to
king water quality. The DNAPL that remains trapped in the soil/aquifer matrix acts as a continuing source of dissolved
I contaminants to ground water, preventing the restoration of DNAPL-affected aquifers for many years.
Printed on Recycled'Paper..,«
-------
DNAPL TRANSPORT AND FATE - CONCEPTUAL APPROACHES
The major factors controlling DNAPL migration in the subsurface include the following (5):
• the volume of DNAPL released;
• the area of infiltration at the DNAPL entry point to the subsurface;
• the duration of release;
• properties of the DNAPL, such as density, viscosity, and interfacial tension;
• properties of the soil/aquifer media, such as pore size and permeability;
• general stratigraphy, such as the location and topography of low-permeability units;
• micro-stratigraphic features, such as root holes, small fractures, and slickensides found in silt/clay layers.
To describe the general transport and fate properties of DNAPL in the subsurface, a series of conceptual
models (24) are presented in the following figures:
Case 1: DNAPL Release to Vadose Zone Only
After release on the surface, DNAPL moves
vertically downward under the force of gravity
and soil capillarity. Because only a small amount
of DNAPL was released, all of the mobile DNAPL
is eventually trapped in pores and fractures in the
unsaturated zone. Infiltration through the
DNAPL zone dissolves some of the soluble
organic constituents in the DNAPL, carrying
organics to the water table and forming a
dissolved organic plume in the aquifer. Migration
of gaseous vapors can also act as a source of
dissolved organics to ground water (13).
DNAPL
Gaseous
Vapors^
Residual
Saturation of
DNAPL in
Vadose Zone
v
Infiltration, Leaching
and Mobile DNAPL
Vapors
Dissolved Contaminant Plume
From DNAPL Soil Vapor
Ground Water
"* Flow
Dissolved Contaminant
Plume From DNAPL
Residual Saturation
After, Waterloo Centre for Groundwater Research, 1989.
Case 2: DNAPL Release to Unsaturated and
Saturated Zones
If enough DNAPL.is released at the surface, it can
migrate all the way through the unsaturated zone
and reach a water-bearing unit. Because the
specific gravity of DNAPL is greater than water, it
continues downward until the mobile DNAPL is
exhausted and is trapped as a residual
hydrocarbon in the porous media. Ground water
flowing past the trapped residual DNAPL
dissolves soluble components of the DNAPL,
forming a dissolved plume downgradient of the
DNAPL zone. As with Case 1, water infiltrating
down from the source zone also carries dissolved
constituents to the aquifer and contributes further
to the dissolved plume.
Residual
Saturation of
DNAPL in Soil
From Spill
7
Dissolved
Contaminant Plume
Ground Water
— Flow
Residual
Saturation in Saturated Zone
After, Waterloo Centre for Groundwater Research, 1989.
-------
CONCEPTUAL APPROACHES - Continued
|se 3: DNAPL Pools and Effect of Low-
Permeability Units
Mobile DNAPL will continue vertical migration
until it is trapped as a residual hydrocarbon (Case
1 and Case 2) or until low-permeability
stratigraphic units are encountered which create
DNAPL "pools" in the soil/aquifer matrix. In this
figure, a perched DNAPL pool fills up and then
spills over the lip of the low-permeability
stratigraphic unit. The spill-over point (or points)
can be some distance away from the original
source, greatly complicating the process of
tracking the DNAPL migration.
Residual
DNAPL
Dissolved
Contaminant
Plume ,
Low Permeable
^Stratigraphic Unit
Sand
Ground Water
Flow
V/7/7/////////////A
Clay
After, Waterloo Centre for Groundwater Research, 1989.
Case 4: Composite Site
In this case, mobile DNAPL migrates vertically
downward through the unsaturated zone and the
first saturated zone, producing a dissolved
constituent plume in the upper aquifer. Although
a DNAPL pool is formed on the fractured clay
unit, the fractures are large enough to permit
srtical migration downward to the deeper
lifer (see Case 5, below). DNAPL pools in a
^graphic low in the underlying impermeable
unit and a second dissolved constituent plume is
formed.
Residual
DNAPL
Dissolved
Contaminant
Plumes
Fractured
Clay
Residual DNAPL
DNAPL Pool
y///////////////////////.
Clay
After, Waterloo Centre for Ground Water Research, 1989.
Case 5: Fractured Rock or Fractured Clay System
DNAPL introduced into a fractured rock or
fractured clay system follows a complex pathway
based on the distribution of fractures in the
original matrix. The number, density, size, and
direction of the fractures usually cannot be
determined due to the extreme heterogeneity of a
fractured system and the lack of economical
aquifer characterization technologies. Relatively
small volumes of DNAPL can penetrate deeply
into fractured systems due to the low retention
capacity of the fractures and the ability of some
DNAPLs to migrate through very small (<20
microns) fractures. Many clay units, once
considered to be relatively impermeable to
§PL migration, often act as fractured media
preferential pathways for vertical and
ontal DNAPL migration.
Fractured
Rock or
Fractured
Clay _
After, Waterloo Centre for Ground Water Research, 1989
-------
Does Historical Site Use Information Indicate Presence of DNAPL?
YES
U
6
O
• i—(
C/l
• I—(
u
01
O
QJ
O)
5-H
U
U
O
Does the
industry type suggest a high
probability of historical
DNAPL release?
(see Table 1)
Does a
process or waste
practice employed at the site
suggest a high probability of
historical DNAPL release?
(see Table 2)
Were any
DNAPL-related chemicals
used in appreciable quantities at the site?
(> 10-50 drums/year)
(see Table 3)
Go To Next Page
INSTRUCTIONS
1. Answer questions in Flowchart 1
(historical site use info. - page 4).
2. Answer questions in Flowchart 2
(site characterization data - page 5).
3. Use "Yes," "No,"and "Maybe"
answers from both flowcharts and enter
Occurrence of DNAPL matrix
(page 6).
TABLE 1
Industries with high probability
of historical DNAPL release:
• Wood preservation (creosote)
• Old coal gas plants
(mid-1800s to mid-1900s)
• Electronics manufacturing
• Solvent production
• Pesticide manufacturing
• Herbicide manufacturing
• Airplane maintenance
» Commercial dry cleaning
• Instrument manufacturing
• Transformer oil production
• Transformer reprocessing
• Steel industry coking
operations (coal tar)
• Pipeline compressor stations
TABLE 2
Industrial processes or waste
disposal practices with high
probability of historical DNAPL
release:
• Metal cleaning/degreasing
• Metal machining
• Tool-and-die operations
• Paint removing/stripping
• Storage of solvents in
underground storage tanks
• Storage of drummed solvents
in uncontained storage areas
• Solvent loading and unloading
• Disposal of mixed chemical
wastes in landfills
• Treatment of mixed chemical
wastes in lagoons or ponds
TABLE 3 DNAPL-Related Chemicals (20):
Note:
The potential for DNAPL release increases with the size
and active period of operation for a facility, industrial
process, or waste disposal practice.
Halogenated Volatiles
Chlorobenzene
1,2-Dichloropropane
1,1 -Dichloroethane
1,1-DichIoroethylene
1,2-Dichloroethane
Trans-1,2-Dichloroethylene
Cis-1,2-Dichloroethylene
1,1/1 -Trichloroethane
Methylene Chloride
1,1,2-Trichloroethane
Trichloroethylene
Chloroform
Carbon Tetrachloride
1,1,2,2-Tetrachloroethane
Tetrachloroethylene
Ethylene Dibromide
Halogenated
Semi-Volatiles
1,4-Dichlorobenzene
1,2-Dichlorobenzene
Aroclor 1242,1254,1260
Chlordane
Dieldrin
2,3,4,6-Tetrachlorophenol
Pentachlorophenol
Non-Halogenated
Semi-Volatiles
2-Methyl Napthalene
o-Cresol
p-Cresol
2,4-Dimethylphenol
m-Cresol
Phenol
Naphthalene
Benzo(a) Anthracene
Fluorene
Acenaphthene
Anthracene
Dibenzo(a,h) Anthracene
Fluoranthene
Pyrene
Chrysene
2,4-Dinitrophenol
Miscellaneous
Coal Tar
Creosote
Note: Many of these
chemicals are found
mixed with other chemicn
or carrier oils.
c-n/q
-------
( Do Site Characterization Data Indicate Presence of DNAPL?^
CM
-M
5-<
OS
^H
U
c
o
• I—I
en
• r-i
u
OJ
Q
r 1
Q
M-H
o
O)
cu
3
8
Has DNAPL
been found in monitoring wells,
observed in soil cores, or physically
observed in the aquifer?
(see Table 4)
Do chemical
analyses of ground water
or soil indicate the possible presence of
DNAPL at the site?
(see Table 5)
(Standard
Field
Program)
Is it likely that
the existing field program could
miss DNAPL at the site?
(see Table 6)
(Extensive Field Program)
INSTRUCTIONS
1. Answer questions in Flowchart 1
(historical site use info. - page 4).
2. Answer questions in Flowchart 2
(site characterization data - page 5).
3. Use "Yes," "No,"and "Maybe"
answers from both flowcharts and enter
Occurrence of DNAPL matrix
(page 6).
Go To Next Page
c
NO
( MAYBE j( YES
TABLE 4
Methods to confirm DNAPL in wells:
• NAPL/water interface probes that signal a
change in conductivity of the borehole fluid
• Weighted cotton string lowered down well
• Pumping and inspecting recovered fluid
• Transparent bottom-loading bailers
• Mechanical discrete-depth samplers.
In general, the depth of DNAPL accumulation
does not provide quantitative information
regarding the amount of DNAPL present (24).
Methods to confirm DNAPL in soil samples:
Visual examination of cores or cuttings may not
be-effective for confirming the presence of
DNAPL except in cases of gross DNAPL
contamination. Methods for enhancing visual
inspection of soil samples for DNAPL include:
• Shaking soil samples in a jar with water to
separate the DNAPL from the soil (14).
Performing a paint filter test, in which soil is
>laced in a filter funnel, water is added, and the
kter is examined for separate phases (20).
TABLE 5
Conditions that indicate potential for
DNAPL at site based on laboratory data:
Condition 1:
Concentrations of DNAPL-related chemicals
(see pg. 3) in ground water are > 1% of pure
phase solubility or effective solubility,
(defined in Worksheet 1, pg. 7) (25).
Condition 2:
Concentrations of DNAPL-related chemicals
on soils are > 10,000 mg/kg (equal to 1% of
soil mass) (6).
Condition 3:
Concentrations of DNAPL-related chemicals
in ground water calculated from water/soil
partitioning relationships and soil samples
are > pure phase solubility or effective
solubility(see Worksheet 2, pg. 7).
Condition 4:
Concentrations of DNAPL-related chemicals
in ground water increase with depth or
appear in anomalous upgradient/across
gradient locations (25).
Note: This procedure is designed primarily for hydrogeologic settings comprised of gravel, sand, silt, or
clay and may not be be applicable to karst or fractured rock settings.
TABLE 6
Characteristics of extensive field
programs that can help indicate the
presence or absence of DNAPL (if
several are present, select "NO"):
• Numerous monitoring wells, with
wells screened in topographic lows
on the surface of fine-grained,
relatively impermeable units.
> Multi-level sampling capability.
• Numerous organic chemical analyses
of soil samples at different depths
using GC or GC/MS methods.
• Well-defined site stratigraphy, using
numerous soil borings, a cone
penetrometer survey, or geophysics.
• Data from pilot tests or "early action"
projects that indicate the site
responds as predicted by
conventional solute transport
relationships, rather than responding
as if additional sources of dissolved
contaminants are present in the
aquifer (11, 25).
-------
Potential for Occurrence of DNAPL at Superfund Sites
DNAPL Category
^
-------
Worksheet 1: Calculation of Effective Solubility (from Shiu, 1988; Feenstra, Mackay, & Cherry, 1991)
L For a single-component DNAPL, the pure-phase solubility of the organic constituent can be used to estimate the theoretical
Supper-level concentration of organics in aquifers or for performing dissolution calculations. For DNAPLs comprised of a
Fmixture of chemicals, however, the effective solubility concept should be employed:
e
S j = the effective solubility (the theoretical upper-level dissolved-phase concentration
e I of a constituent in ground water in equilibrium with a mixed DNAPL; in mg/1)
i i i I Where ^ _ ^ mole fraction of component i in the DNAPL mixture (obtained from a lab
analysis of a DNAPL sample or estimated from waste characterization data)
S [ - the pure-phase solubility of compound i in mg/1 (usually obtained from
literature sources)
For example, if a laboratory analysis indicates that the mole fraction of trichloroethylene (TCE) in DNAPL is 0.10, then the
effective solubility would be 110 mg/1 [pure phase solubility of TCE times mole fraction TCE: (1100 mg/1) * (0.10) = 110
mg/1]. Effective solubilities can be calculated for all components in a DNAPL mixture. Insoluble organics in the mixture
(such as long-chained alkanes) will reduce the mole fraction and effective solubility of more soluble organics but will not
contribute dissolved-phase organics to ground water. Please note that this relationship is approximate and does not account for
non-ideal behavior of mixtures, such as co-solvency, etc.
Worksheet 2: Method for Assessing Residual NAPL Based on Organic Chemical
Concentrations in Soil Samples (From Feenstra, Mackay, and Cherry, 1991)
To estimate if NAPLs are present, a partitioning calculation based on chemical and physical analyses of soil samples from
the saturated zone (from cores, excavations, etc.) can be applied. This method tests the assumption that all of the urganics
in the subsurface are either dissolved in ground water or adsorbed to soil (assuming dissolved-phase sorption, not the
presence of NAPL). By using the concentration of organics on the soil and the partitioning calculation, a theoretical pore-
water concentration of organics in ground water is determined. If the theoretical pore-water concentration is greater than
the estimated solubility of the organic constituent of interest, then NAPL may be present at the site. A worksheet for
jrforming this calculation is presented below; see Feenstra, Mackay, and Cherry (1991) for the complete methodology.
Step 1: Calculate Sj , the effective solubility of organic constituent of interest. [See Worksheet 1, above7|
Step 2: Determine Koc, the organic carbon-water partition coefficient from one of the following:
A) Literature sources (such as 22) or
B) From empirical relationships based on Kow, the octanol-water partition coefficient, which is also found in the
literature (22). For example, Koc can be estimated from Kow using the following expression developed for
polyaromatic hydrocarbons (8):
Log Koc = 1.0 * Log Kow - 0.21 I Other empirical relationships between Koc
••^•••••Mn^HMHiMHHj ami Kow are presented in refs. 4 and 15.
Step 3: Determine foe, the fraction of organic carbon on the soil, from a laboratory analysis of clean soils from the site.
Values for foe typically range from 0.03 to 0.00017 mg/mg (4). Convert values reported in percent to rng/mg.
Step 4: Determine or estimate pb, the dry bulk density of the soil, from a soils analysis. Typical values range from 1.8 to 2.1
g/ml (kg/1). Determine or estimate (pw, the water-filled porosity.
Step 5: Determine Kd, the partition (or distribution) coefficient between I Kd - K *f I
the pore water (ground water) and the soil solids: ~
Step 6: Using Ct, the measured cone, of the organic compound in saturated soil in mg/kg,
calculate the theoretical pore water cone, assuming no DNAPL (i.e., Cw in mg/1):
Cw =
(Ct * pb)
(Kd*pb •
Q
I
tep 7: Compare Cw and Sj (from Step 1):
Cw > S | suggests possible presence of DNAPL
Q
Cw < S | suggests possible absence of DNAPL
-------
GLOSSARY (adapted from Cherry, 1991):
DNAPL: A Dense Nonaqueous Phase Liquid. A DNAPL can be either a single-component DNAPL (comprised of only
one chemical) or a mixed DNAPL (comprised of several chemicals). DNAPL exists in the subsurface as free-phase DNAPL
or as residual DNAPL (see following definitions). DNAPL does not refer to chemicals that are dissolved in groundwater.
DNAPL ENTRY LOCATION: The area where DNAPL has entered the subsurface, such as a spill location or waste pond.
DNAPL SITE: A site where DNAPL has been released and is now present in the subsurface as an immiscible phase.
DNAPL ZONE: The portion of a site affected by free-phase or residual DNAPL in the subsurface (either the unsaturated
zone or saturated zone). The DNAPL zone has organics in the vapor phase (unsaturated zone), dissolved phase (both
unsaturated and saturated zone), and DNAPL phase (both unsaturated and saturated zone).
DISSOLUTION: The process by which soluble organic components from DNAPL dissolve in ground water or dissolve in
infiltration water and form a ground-water contaminant plume. The duration of remediation measures (either clean-up or
long-term containment) is determined by 1) the rate of dissolution that can be achieved in the field, and 2) the mass of
soluble components in the residual DNAPL trapped in the aquifer.
EFFECTIVE SOLUBILITY: The theoretical aqueous solubility of an organic constituent in ground water that is in
chemical equilibrium with a mixed DNAPL (a DNAPL containing several organic chemicals). The effective solubility of a
particular organic chemical can be estimated by multiplying its mole fraction in the DNAPL mixture by its pure phase
solubility (see Worksheet 1, page 7).
FREE-PHASE DNAPL: Immiscible liquid existing in the subsurface with a positive pressure such that it can flow into a
well. If not trapped in a pool, free-phase DNAPL will flow vertically through an aquifer or laterally down sloping fine-
grained stratigraphic units. Also called mobile DNAPL or continuous-phase DNAPL.
PLUME: The zone of contamination containing organics in the dissolved phase, The plume usually will originate from
the DNAPL zone and extend downgradient for some distance depending on site hydrogeologic and chemical conditions.
To avoid confusion, the term "DNAPL plume" should not be used to describe a DNAPL pool; "plume" should be used only
to refer to dissolved-phase organics.
POOL and LENS: A pool is a zone of free-phase DNAPL at the bottom of an aquifer. A lens is a pool that rests on a fine-
grained stratigraphic unit of limited areal extent. DNAPL can be recovered from a pool or lens if a well is placed in the
right location.
RESIDUAL DNAPL: DNAPL held in soil pore spaces or fractures by capillary forces (negative pressure on DNAPL).
Residual will remain trapped within the pores of the porous media unless the viscous forces (caused by the dynamic force
of water against the DNAPL) are greater than the capillary forces holding the DNAPL in the pore. At most sites the
hydraulic gradient required to mobilize all of the residual trapped in an aquifer is usually many times greater than the
gradient that can be produced by wells or trenches (26).
RESIDUAL SATURATION: The saturation (the fraction of total pore space containing DNAPL) at which DNAPL
becomes discontinuous and is immobilized by capillary forces (14). In unsaturated soils, residual saturation typically
ranges from 5% to 20% of total pore volume, while in the saturated zone the residual saturation is higher, with typical
values ranging from 15% to 50% of total pore volume (14,17). At many sites, however, DNAPL migrates preferentially
through small-scale fractures and heterogeneities in the soil, permitting the DNAPL to penetrate much deeper than would
be predicted from application of typical residual saturation values (16).
T~\ £• J A L I-\TVT A TJT c>*i / , ~", Dissolved-Phase PLUME
Ueiined Areas at a UlSAl L bite (contains free-phase DNAPL in pools or
lenses and /or residual DNAPL)
DNAPL ENTRY LOCATION
(such as a former waste pond)
Ground Water Flow Direction
-------
References
•. Anderson, M.R., R.L. Johnson, and J.F. Pankow, The Dissolution of Residual Dense Non-Aqueous Phase Liquid (DNAPL) from a
T Saturated Porous Medium, Proc.: Petrol. Hcarb. and Org. Chemicals in Ground Water, NWWA, Houston, TX, Nov., 1987.
2. Cherry, J. A., written communication to EPA DNAPL Workshop, Dallas, TX, R. S. Kerr Environmental Research Laboratory, U.S.
EPA, Ada, OK., Apr. 1991.
3. Connor, J.A., C.J. Newell, and O.K. Wilson, Assessment, Field Testing, and Conceptual Design for Managing Dense Nonaqueous
Phase Liquids (DNAPL) at a Superfund Site, Proc.: Petrol. Hcarb. Org. Chemicals in Ground Water, NWWA, Houston, TX, 1989.
4. Domenico, P.A. and F. W. Schwartz, Physical and Chemical Hydrogeology, Wiley, New York, NY, 1990.
5. Feenstra, S. and J.A. Cherry, Subsurface Contamination by Dense Non-Aqueous Phase Liquids (DNAPL) Chemicals, International
Groundwater Symposium, International Assoc. of Hydrogeologists, Halifax, N.S., May 1-4,1988.
6. Feenstra, S., D. M. MacKay, and J.A. Cherry, A Method for Assessing Residual NAPL Based on Organic Chemical Concentrations in
Soil Samples, Groundwater Monitoring Review, Vol. 11, No. 2,1991.
7. Hunt, J.R., N. Sitar, and K.D. Udell, Nonaqueous Phase Liquid Transport and Cleanup, Water Res. Research, Vol. 24 No. 8,1991.
8. Karickhoff, S.W., D.S. Brown, and T.A. Scott, Sorption of Hydrophobic Pollutants on Natural Sediments, Water Res. R., Vol. 3,1979.
9. Keller, C.K., G. van der Kamp, and J.A. Cherry, Hydrogeology of Two Saskatchewan Tills, T. of Hydrology, pp. 97-121,1988.
10. Kueper, B.H. and E. O. Frind, An Overview of Immiscible Fingering in Porous Media, T. of Cont. Hydrology, Vol. 2,1988.
11. Mackay, D.M. and J.A. Cherry, Ground-Water Contamination: Pump and Treat Remediation, ES&T Vol. 23, No. 6,1989.
12. Mackay, D.M., P.V. Roberts, and J.A. Cherry, Transport of Organic Contaminants in Ground Water, ES&T, Vol. 19, No. 5,1985.
13. Mendoza, C.A. and T. A. McAlary, Modeling of Ground-Water Contamination Caused by Organic Solvent Vapors, Ground
Water, Vol. 28, No. 2,1990.
14. Mercer, J.W. and R.M. Cohen, A Review of Immiscible Fluids in the Subsurface: Properties, Models, Characterization and
Remediation, I. of Cont. Hydrology, Vol. 6,1990.
15. Olsen, R.L. and A. Davis, Predicting the Fate and Transport of Organic Compounds in Groundwater, HMC, May /June 1990.
16. Poulson, M. and B.H. Kueper, A Field Experiment to Study the Behavior of Perchloroethylene in Unsaturated Porous Medium.
Submitted to ES&T, 1991.
17. Schwille, F., Dense Chlorinated Solvents in Porous and Fractured Media: Model Experiments (English Translation), Lewis
Publishers, Ann Arbor, MI, 1988.
18. Shiu, W.Y., A. Maijanen, A.L.Y. Ng, and D. Mackay, Preparation of Aqueous Solutions of Sparingly Soluble Organic Substances:
II. Multicomponent System - Hydrocarbon Mixtures and Petroleum Products, Environ. Toxicology & Chemistry, Vol. 7,1988.
19. Sitar, N., J.R. Hunt, and J.T. Geller, Practical Aspects of Multiphase Equilibria in Evaluating the Degree of Contamination, Proc. of
the Int. Asso. of Hydrog. Conf. on Subsurface Cont. by Immiscible Fluids, April 18 - 20, Calgary, Alb., 1990.
20. U.S. EPA, Dense Nonaqueous Phase Liquids, EPA Ground Water Issue Paper. EPA/540/4-91-002.1991.
21. U.S. EPA, Evaluation of Ground-Water Extraction Remedies, Volume 1 (Summary Report), EPA/540/2-89/054,1989.
22. Verschueren, K., Handbook of Environmental Data on Organic Chemicals, Van Nostrand Reinhold, New York,_NY, 1983.
23. Villaume, J.F., Investigations at Sites Contaminated with Dense Non-Aqueous Phase Liquids (NAPLs), Ground Water Monitoring
Review, Vol. 5, No. 2,1985.
24. Waterloo Centre for Ground Water Research, University of Waterloo Short Course, Dense Immiscible Phase Liquid Contaminants
in Porous and Fractured Media, Kitchener, Ont., Oct., 1991.
25. Waterloo Centre for Ground Water Research, University of Waterloo Short Course, Identification of DNAPL Sites: An Eleven
Point Approach, Kitchener, Ont., Oct., 1991.
26. Wilson, J.L. and S.H. Conrad, Is Physical Displacement of Residual Hydrocarbons a Realistic Possibility in Aquifer Restoration?,
Proc.: Petrol. Hcarb. and Org. Chemicals in Ground Water, NWWA, Houston, TX, NWWA, Nov. 5-7,1984.
NOTICE: The policies and procedures set out in this document are intended solely as guidance. They are not intended, nor can they
be relied upon, to create any rights enforceable by any party in litigation with the United States. EPA officials may decide to follow
the guidance provided in this memorandum, or to act at variance with the guidance, based on an analysis of specific site
circumstances. The Agency also reserves the right to change this guidance at any time without public notice.
For more information, contact: Randall R. Ross
R. S. Kerr Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Ada, Oklahoma 74820
Authors: Charles J. Newell, Groundwater Services, Inc., Houston, Texas
Randall R. Ross, R. S. Kerr Environmental Research Laboratory
-------
Techniques of Water-Resources Investigations
of the United States Geological Survey
DEFINITION OF BOUNDARY AND INITIAL
CONDITIONS IN THE ANALYSIS OF
SATURATED GROUND-WATER FLOW
SYSTEMS—AN INTRODUCTION
BOOK 3
CHAPTER B5
-------
Techniques of Water-Resources Investigations
of the United States Geological Survey
Chapter B5
DEFINITION OF BOUNDARY AND INITIAL
CONDITIONS IN THE ANALYSIS OF
SATURATED GROUND-WATER FLOW
SYSTEMS—AN INTRODUCTION
By O. Lehn Franke, Thomas E. Reilly, and Gordon D. Bennett
Book3
APPLICATIONS OF HYDRAULICS
-------
PREFACE
The series of manuals on techniques describes procedures for planning and executing
specialized work in water-resources investigations. The material is grouped under
major subject headings called books and further subdivided into sections and chapters;
section B of book 3 is on ground-water techniques.
The unit of publication, the chapter is Emited to a narrow field of subject matter. This
format permits flexibility in revision and publication as the need arises. Chapter 3B5
deals with the definition of boundary and initial conditions in the analysis of saturated
ground-water flow systems.
Provisional drafts of chapters are distributed to field offices of the U.S. Geological
Survey for their use. These drafts are subject to revision because of experience in use
or because of advancement in knowledge, techniques, or equipment. After the
technique described in a chapter is sufficiently developed, the chapter is published and
is for sale from U.S. Geological Survey, Books and Open-File Reports Section, Federal
Center, Box 25425, Denvei; CO 80225.
Reference to trade names, commercial products, manufacturers, or distributors in
this manual constitutes neither endorsement by the Geological Survey nor recommen-
dation for use.
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS OFTHEf
UNITED STATES GEOLOGICAL SURVEY
The U.S. Geological Survey publishes a series of manuals describing, procedures for planning and conducting
specialized work in water-resources investigations. The manuals published to date are listed below and may be ordered
by mail from the U.S. Geological Survey Books and Open-File Reports, Federal Center, Box 25425, Denver, Colorado
80225 an authorized agent of the Superintendent of Documents, Government Printing Office).
Prepayment is required. Remittance should be sent by check or money order payable to U.S. Geological Survey. Prices
are not included in the listing below as they are subject to change. Current prices can be obtained by writing to the
USGS, Books and Open File Reports. Prices include cost of domestic surface transportation. For transmittal outside the
U.S.A. (except to Canada and Mexico) a surcharge of 25 percent of the net bill should be included to cover surface
transportation. When ordering any of these publications, please give the title, book number, chapter number, and "U.S.
Geological Survey Techniques of Water-Resources Investigations."
TWI l-Dl. Water temperature—influential factors, field measurement, and data presentation, by H.H. Stevens, Jr., J.F. Ficke, and G.F. Smoot,
1975, 65 pages.
TWI 1-D2. Guidelines for collection and field analysis of ground-water samples for selected unstable constituents, by W.W. Wood. 1976. 24
pages.
TWI 2-D1. Application of surface geophysics to ground water investigations, by A-A-R. Zohdy, G.P. Eaton, and D.R. Mabey. 1974. 116 pages.
TWI 2-E1. Application of borehole geophysics to water-resources investigations, by W.S. Keys and L.M. MacCary. 1971. 126 pages.
TWI 3-A1. General field and office procedures for indirect discharge measurement, by M-A. Benson and Tate Dalrymple. 1967. 30 pages.
TWI 3-A2. Measurement of peak discharge by the slope-area method, by Tate Dalrymple and M.A. Benson. 1967. 12 pages.
TWI 3-A3. Measurement of peak discharge at culverts by indirect methods, by G.L. Bodhaine. 1968. 60 pages.
TWI 3-A4. Measurement of peak discharge at width contractions by indirect methods, by H.F. Matthai. 1967. 44 pages.
TWI 3-A5. Measurement of peak discharge at dams by indirect methods, by Harry Hulsing. 1967. 29 pages.
TWI 3-A6. General procedure for gaging streams, by R.W. Carter and Jacob Davidian. 1968. 13 pages.
TWI 3-A7. Stage measurements at gaging stations, by TJ. Buchanan and W.P. Somers. 1968. 28 pages.
TWI 3-A8. Discharge measurements at gaging stations, by TJ. Buchanan and W.P. Somers. 1969. 65 pages.
TWI 3-A9. Measurement of time of travel and dispersion in streams by dye tracing, by E.P. Hubbard, F.A. Kilpatrick, L-A. Martens, and
J.F. Wilson, Jr. 1982. 44 pages.
TWI 3-A10. Discharge ratings at gaging stations, by EJ. Kennedy. 1984. 59 pages.
TWI 3-A11. Measurement of discharge by moving-boat method, by G.F. Smoot and C.C. Novak. 1969. 22 pages.
TWI 3-A12. Fluorometric procedures for dye tracing, Revised, by James F. Wilson, Jr., Emest D. Cobb, and Frederick A. Kilpatrick. 1986. 41
pages.
TWI 3-A13. Computation of continuous records of streamflow, by Edward J. Kennedy. 1983. 53 pages.
TWI 3-A14. Use of flumes in measuring discharge, by F.A. Kilpatrick, and V.R. Schneider. 1983. 46 pages.
TWI 3-A15. Computation of water-surface profiles in open channels, by Jacob Davidian. 1984. 48 pages.
TWI 3-A16. Measurement of discharge using tracers, by F.A. Kilpatrick and E.D. Cobb. 1985. 52 pages.
TWI 3-A17. Acoustic velocity met6er systems, by Amonius Laenen. 1985. 38 pages.
TWI 3-B1. Aquifer-test design, observation, and data analysis, by R.W. Stallman. 1971. 26 pages.
TWI 3-B2. Introduction to ground-water hydraulics, a programmed text for self-instruction, by G.D. Bennett. 1976. 172 pages. Spanish
translation TWI 3-B2 also available.
TWI 3-B3. Type curves for selected problems of flow to wells in confined aquifers, by J.E. Reed. 1980. 106 p.
TWI 3-B6. The principle of superposition and its application in ground-water hydraulics, by Thomas E. Reilly, O. Lehn Franke, and Gordon D.
Bennett. 1987. 28 pages.
TWI 3-C1. Fluvial sediment concepts, by H.P. Guy. 1970. 55 pages.
TWI 3-C2. Field methods of measurement of fluvial sediment, by H.P. Guy and V.W. Norman. 1970. 59 pages.
TWI 3-C3. Computation of fluvial-sediment discharge, by George Porterfield. 1972. 66 pages.
TWI 4-A1. Some statistical tools in hydrology, by H.C Riggs. 1963. 39 pages.
TWI 4-A2. Frequency curves, by H.C Riggs, 1968. 15 pages.
TWI 4-B1. Low-flow investigations, by H.C Riggs. 1972. IS pages.
TWI 4-B2. Storage analyses for water supply, by H.C Riggs and C.H. Hardison. 1973. 20 pages.
TWI 4-B3. Regional analyses of streamflow characteristics, by H.C. Riggs. 1973. 15 pages.
TWI 4-D1. Computation of rate and volume of stream depletion by wells, by C.T. Jenkins. 1970. 17 pages.
TWI S-A1. Methods for determination of inorganic substances in water and fluvial sediments, by M.W. Skougstad and others, editors. 1979. 616
pages.
TWI 5-A2. Determination of minor elements in water by emission spectroscopy, by P.R. Bamett and E.C. Mallory, Jr. 1971. 31 pages.
-------
TV/I 5-A3. Methods for the determination of organic substances in water and fluvial sediments, edited by R.L. Wershaw, M J. Fishman, R.R.
Grabbe, and L.E. Lowe. 1987. 80 pages. This manual is a revision of "Methods for Analysis of Organic Substances in Water" by
Donald F. Coerlitz and Eugene Brown, Book 5, Chapter A3, published in 1972.
TWI 5-A4. Methods for collection and analysis of aquatic biological and microbiological samples, edited by P.E Greeson, T.A. Ehlke, G.A.
Irwin, B.W. Hum, and K.V. Slack. 1977. 332 pages.
TV/I 5-A5. Methods for determination of radioactive substances in water and fluvial sediments, by L.L. Thatcher, VJ. Janzer, and K.W.
Edwards. 1977. 95 pages.
TV/I 5-A6. Quality assurance practices for the chemical and biological analyses of water and fluvial sediments, by L.C. Friedman and D.E.
Erdmann. 1982.181 pages.
TV/I S-C1. Laboratory theory and methods for sediment analysis, by H.P. Guy. 1969. 58 pages.
TV/I 7-C1. Finite difference model for aquifer simulation in two dimensions with results of numerical experiments, by P.C Trescott, G.F.
Finder, and S.P. Larson. 1976. 116 pages.
TWI 7-C2. Computer model of two-dimensional solute transport and dispersion in ground water, by L.F. Konikow and J.D. Bredehoeft. 1978.
90 pages.
TWI 7-C3. A model for simulation of flow in singular and interconnected channels, by R.W. Schaffranek, RA. Baltzer, and D.E. Goldberg.
1981.110 pages.
TWI 8-A1. Methods of measuring water levels in deep wells, by M.S. Garber and F.C Koopman. 1968. 23 pages.
TWI 8-A2. Installation and service manual for U.S. Geological Survey monometers, by J.D. Craig. 1983. 57 pages.
TWI &-B2. Calibration and maintenance of vertical-axis type current meters, by G.F. Smoot and CE. Novak. 1968. 15 pages.
-------
CONTENTS
r*t*
Abstract 1
Introduction 1
Boundary conditions 1
Principal types of boundary conditions 2
Some important aspects of specifying boundary conditions in ground-water models-——1— 6
Model boundaries versus physical boundaries •• . 6
Selection of boundary conditions in relation to system stress 8
Boundary conditions in steady-state models 8
The water table as a boundary 9
Reference elevation in ground-water models 9
Concluding remarks •• • • •• 10
Exercises 10
Initial conditions 10
Concept of initial conditions : 11
Specifying initial conditions in models 11
Example of specifying initial conditions in a field situation __________ 13
Concluding remarks 12
Acknowledgments 13
References • 13
Appendix: Discussion of the solution of differential equations and the role of boundary
conditions .... . 14
FIGURES
1. Flow net within three different hydraulic settings: Through and beneath an earth dam underlain by sloping bedrock; beneath a
vertical impermeable wall; and beneath an impermeable dam and a vertical impermeable wall ------------ 3
2-7. Diagrams of:
2. Piezometers at different depths demonstrating that the total head at all depths in a continuous body of stationary fluid is
constant ---------------------- 3
3. Flow pattern in uniformly permeable material with constant areal recharge and discharge to symmetrically placed streams --- 4
4. A leaky aquifer system ------------------------ 5
5. Flow pattern in a permeable dam having vertical faces --------- ; - • -------- 7
6. Flow pattern near a discharging well in an unconfincd aquifer ------------- 7
7. Flow pattern near a seawater-freshwater interface ----------------------------- 8
8. Hydrograph of well N1614 tapping the upper glacial aquifer in central Nassau County, New York ----------------- 12
9. Example of solutions to a differential equation: Diagram of idealized aquifer system, and two of the family of curves solving the
general differential equation for the idealized aquifer system ------------------------------------- 15
TABLE
1. Common designations for several important boundary conditions
VII
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS IN
THE ANALYSIS OF SATURATED GROUND-WATER FLOW
SYSTEMS-AN INTRODUCTION
By 0. Lehn Franke. Thomas E. Reilly, and Gordon D. Bennett
Abstract
Accurate definition of boundary and initial conditions is an essential
part of conceptualizing and modeling ground-water flow systems. This
report describes the properties of the seven most common boundary
conditions encountered in ground-water systems and discusses major
aspects of their application. It also discusses the significance and
specification of initial conditions and evaluates some common errors in
applying this concept to ground-water-system models. An appendix is
included that discusses what the solution of a differential equation
represents and bow the solution relates to the boundary conditions
defining the specific problem. This report considers only boundary
conditions that apply to saturated ground-water systems.
Introduction
The specification of appropriate boundary and initial
conditions is an essential part of conceptualizing and
modeling1 ground-water systems and is also the part most
subject to serious error by ground-water hydrologjsts.
Although some excellent discussions of these topics are
provided in a few readily available texts (for example, Bear,
1979, p. 94-102, 116-123; and Rushton and Redshaw,
JThe word "model" is used in several different ways in this report
and in ground-water hydrology. A general definition of model is a
representation of some or all of the properties of a system. Developing
a "conceptual model" of the ground-water system is the first and
critical step in any study, particularly studies involving mathematical-
numerical modeling. In this context, a conceptual model is a clear,
qualitative, physical description of the operation of the natural system.
A "mathematical model" represents the system under study through
mathematical equations and procedures. The differential equations
that describe in approximate terms a physical process (for example,
ground-water flow and solute transport) are a mathematical model of
that process. The solution to these differential equations in a specific
problem frequently requires numerical procedures (algorithms),
although many simpler mathematical models can be solved analytically.
Thus, the process of "modeling" usually implies developing either a
conceptual model, a mathematical model, or a mathematical-numerical
model of the system or problem under study. The context will suggest
which meaning of "model" is intended.
1979, p. 153-156, 182-184), most standard texts on
ground-water hydrobgy do not thoroughly discuss these
topics from the standpoint of ground-water flow modeling.
The purpose of this report is to provide a concise
introduction to these topics to give ground-water hydrol-
ogists the information necessary to successfully apply
these concepts in conceptualizing ground-water systems
for the purpose of modeling.
The report consists of three parts. The first part
provides the definition and properties of the most common
boundary conditions in ground-water systems and dis-
cusses their application in special situations. The second
part explains the concept of initial conditions and discusses
some common pitfalls in specifying initial conditions in
models of ground-water systems. The third part is an
appendix that discusses what the solution of a differential
equation represents and how the solution relates to the
boundary conditions defining the specific problem. The
report considers only boundary conditions that apply to
saturated ground-water systems.
Boundary Conditions
Quantitative modeling of a ground-water system entails
the solution of a boundary-value problem—a type of
mathematical problem that has been extensively studied
and has applications in many areas of science and tech-
nology. The flow of ground water is described in the
general case by partial differential equations. A ground-
water problem is "defined" by establishing the appropriate
boundary-value problem; solving the problem involves solv-
ing the governing partial differential equation in the flow
domain whOe at the same time satisfying the specified
boundary and initial conditions. In ground-water prob-
lems, the solution is usually expressed in terms of head (h);
that is, head is usually the dependent- variable in the
governing partial differential equation. The solution to a
simple boundary-value problem in ground-water flow is
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
given in the appendix and serves as an example of a formal
solution to this type of problem.
Defining a specific ground-water problem in prepara-
tion for subsequent quantitative modeling requires a clear
concept of how the ground-water flow system under study
functions. Various representations of a ground-water flow
system are possible, depending on one's objectives and
point of view. In this discussion, the term "flow system"
refers to the part of the ground-water regime that has
been isolated for study and implies the following:
1. A three-dimensional body of earth material is satu-
rated with flowing ground water;
2. The region containing (he ground water is bounded by
a dosed surface called the "boundary surface" of the
flow system;
3. Under natural (unstressed) conditions, average flow in
the system, as well as average ground-water levels,
normally fluctuate around a mean value;
4. Inflow (continuous or intermittent) of water to the
system and outflow from it occur through at least
part of the boundary surface.
In ground-water investigations, the system or subsystem
under study ideally should be enclosed by a boundary
surface that corresponds to identifiable hydrogeobgic
features at which some characteristic of ground-water
flow is easily described; examples are a body of surface
water, an almost impermeable surface, a water table, and
so on. In many studies, however, some part of the bound-
ary surface must be chosen arbitrarily, often in ways that
depend on the proposed modeling strategy. The position of
the three-dimensional boundary surface in nature
(regardless of the extent to which it has been arbitrarily
specified) defines the "external geometry" of the ground-
water flow system.
Specifying the boundary conditions of the ground-water
flow system means assigning a boundary type (usually one
or a combination of the types listed in the following
paragraphs) to every point on the boundary surface.
The selection of the boundary surface and boundary
conditions is probably the most critical step in conceptu-
alizing and developing a model of a ground-water system.
Improper selection of these components may result in a
failure of the modeling effort, with the result that the
model's response to an applied stress bears little relation
to the corresponding response in the real system.
Usually the selection of boundary conditions for a
conceptual or numerical model involves considerable sim-
plification of actual hydrogeologic conditions. To avoid
serious error, the assumptions underlying such simplifica-
tions must be clearly understood and their effect on model
response critically evaluated.
Principal types of
boundary conditions
This section describes the pertinent characteristics of
seven types of boundaries—constant head, specified head.
streamline (or stream surface), specified flux, heao
dependent flux, free surface, and seepage surface.
1. Constant-head (surface or line) boundary—Hydraulic
head (h) in a ground-water system is the sum of elevation
head (z) and pressure head (p/j), where p is gage
pressure and y is the unit weight of wa'-r. Elevation head
represents the potential energy of a water particle due to
its vertical position above some datum, and pressure head
represents pressure measured in terms of the height of a
column of water in a piezometer. Physically, hydraulic head
represents the water level above datum in a piezometer or
observation well open only to the point in question.
A surface of equal head is an imaginary surface having
the same head value at all points. Thus, all piezometers
open to different points on a surface of equal head will
show exactly the same water level in reference to a
common datum. In a two-dimensional problem2, the con-
cept of a line (rather than a curving surface) of equal head
is used—that is, a line along which all points have the same
head value.
A constant-head boundary3 occurs where a part of the
boundary surface of an aquifer system coincides with a
surface of essentially constant head. (The word "con-
stant," as used here, implies a value that is uniform at all
points along the surface as well as through time.) An
example is an aquifer that crops out beneath a lake in
which the surface-water stage is nearly uniform over all
points of the outcrop and does not vary appreciably with
time. Other examples of a constant-head boundary are
shown in figure \A (lines ABC, EG), figure IB (lines BA,
CD), and figure 1C (lines AB, CD), all of which depict
two-dimensional steady-state ground-water seepage
beneath engineering structures that are bounded in part
by surface-water bodies. The artificial lateral boundaries
in figure 1 (dashed) are discussed later in the section
"Model Boundaries Versus Physical Boundaries."
Let us consider the boundary ABC in figure IA. The
question sometimes arises as to whether line BC on the
submerged side of the dam in fact represents a uniform
constant head (or constant potential) along its entire
length. Obviously, thepressure varies with depth along this
surface. We assume that the surface water behind the dam
is essentially static; the rule is that within such a body of
stationary fluid, the fatal head is a constant at ever)' point.
including points along the boundary surface between the
fluid body and the ground-water system, regardless of the
surface configuration. To demonstrate this concept, con-
sider piezometers at various depths in the body of a
In a two-dimensional problem, the components of ground-water-
velocity vectors can be designated by two coordinate axes. Two-
dimensional now is planar. The illustrations in this report depict
two-dimensional problems; that is, ground-water flows only in the
plane of the illustration.
'Also referred to as constant-potential or equipoicniial boundary.
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
Impermeable l«yer (Bedrock)
B
Imptrmeibfe will
Impermeiblt Ityer
EquipountiM lint'
Figure 1.—Flow net within three different hydraulic settings:
A, through and beneath an earth dam underlain by sloping
bedrock; B. beneath a vertical impermeable wall; and C.
beneath an impermeable dam and a vertical impermeable
wall.
stationary fluid (fig. 2). At the surface of the fluid body
(piezometer A), where the fluid is in contact with the
atmosphere, h = z because p/f-0. As one moves the
piezometer downward from the fluid surface (piezome-
ters B and C), the increase in pressure head (p/-y) is
exactly balanced by a decrease in elevation head (z); thus,
h remains constant.
2. Specified-head boundary.—A. more general type of
boundary condition, of which the constant-head boundary
is actually a special case, occurs wherever head can be
specified as a function of position and time over a part of
Piezometers
A
^
i i
** "*
B
„
V
4 1
'""•
f h
i
I I !
t V ^ »
a
r
C
7
Surface of fluid
subject to
•tmospheric pressure,
x
i t
i
C hc
I
!*<
2-0
NBodv of stationary fluid
(Djiuml
Figure 2.—Piezometers at different depths demonstrating
that the total head at all depths in a continuous body of
stationary fluid is constant
the boundary surface of a ground-water system. An
example of the simplest type might be an aquifer that is
exposed along the bottom of a large stream whose stage is
independent of ground-water seepage. As one moves
upstream or downstream, the head changes in relation to
the slope of the stream channel. If changes in head with
time are not significant, the head can be specified as a
function of position alone (h = f(x,y) in a two-dimensional
analysis) at all points along the streambed. In a more
complex situation, in which stream stage varies with time,
head at points along the streambed would be specified as
a function of both position and time, h = f(x,y,t). In either
example, heads along the streambed are specified accord-
ing to circumstances external to the ground-water system
and maintain these specified values throughout the prob-
lem solution, regardless of the stresses to which the
ground-water system is subjected.
Both specified-head and constant-head boundaries (or
nodes, in a discretized analysis) have an important "phys-
ical" characteristic in models of ground-water systems—in
effect, they can provide an inexhaustible source of water.
No matter how much water is "pumped" from a system
model, the specified-head boundaries will continue to
supply the required amount, even if that amount is not
physically reasonable in the real system. This aspect of
specified-head boundaries should be considered carefully
whenever this boundary type is selected for simulation and
also when any model result or prediction is evaluated.
3. Streamline or stream-surface boundary (no-flow).—A.
streamline is a curve that is tangent to the flow-velocity
vector at every point along its length; ihus, no flow
components exist normal to a streamline and no flow
crosses a streamline. Because a stream surface is a
continuous three-dimensional surface made up entirely of
streamlines, it follows that no flow crosses a stream
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
LjndsurfKt, .-Wjttf IlDle Strewn —
Strevnlir*-' Equipotentiilline-~
Figure 3.—Flow pattern in uniformly permeable material with
constant areal recharge and discharge to symmetrically
placed streams. (Modified from Hubbert. 1940.)
surface. In a steady flow, that is, a flow that does not
change with time, streamlines and stream surfaces remain
constant, whereas in nonsteady or transient flow— flow
that changes with time— the streamlines and stream sur-
faces in the interior of the flow may differ from one instant
to the next. Even in nonsteady flows, however, it is common
for some parts of the boundary to consist of stream
surfaces that remain fixed with time. An example is an
impermeable boundary. Natural earth materials are never
completely impermeable. Howevei; they may sometimes
be regarded as effectively impermeable for modeling pur-
poses if the hydraulic conductivities of the adjacent mate-
rials differ by several orders of magnitude.
For an isotropic medium, the flow per unit area from a
boundary into an aquifer is given by Carey's law4 as
where
q is specific discharge (L/T),
K is hydraulic conductivity of the aquifer (L/T),
h is hydraulic head (L), and
n is distance normal to the boundary (L).
The condition that q be equal to zero, as required for
no-flow boundaries, can be satisfied only if dh/dn, the head
gradient normal to the boundary, is also zero. Thus, a
simple formulation of the no-flow condition in terms of
head is possible.
An example of a boundary that is effectively imperme-
able is the contact between unweathered granite and
permeable unconsolidated material; another is a sub-
4 Hydraulic conductivity in the example above is specified as isotro-
pic 10 simplify the form of Darcy's law that is used. In anisot topic
systems, the direction normal to the boundary (designated n) must
coincide with a major axis of the hydraulic-conductivity tensor (repre-
sented geometrically by the hydraulic conductivity ellipsoid) to enable
use of the simple form of Darcy's law above in each coordinate
direction. Otherwise, the off-diagonal terms of the
hydraulic-conductivity tensor are not zero and must be used to
calculate the flu* in the specified coordinate direction (Lohoian and
others, 1972).
merged sheetpfle wall3. Some examples of boundaries thi
are assumed to be no-flow (stream-surface) boundaries
are depicted in figure IA, line HI; figure 15, Knes AEC,
FG; and figure 1C, lines BGHC, EF.
4. Specified-flux boundary.—Another general type of
boundary, of which the stream-surface (or no-flow)
boundary is a special case, is found wherever the flux
across a given part of the boundary surface can be
specified as a function of position and time. (The term
"flux" as used in this discussion refers to the volume of
fluid crossing a unit cross-sectional surface area per unit
time.) In the simplest type of specified-flux boundary, the
flux across a given part of the boundary surface is
considered uniform in space and constant with time; this
assumption is often made, for example, with respect to
areal recharge crossing the upper surface of an aquifer. A
flow net depicting constant areal recharge to a water table
is shown in figure 3. Boundaries of this type are termed
"constant-flux" boundaries, and the stream-surface
boundary can be considered a special case in which the
constant flux is zero. In a more general case, the flux
might be constant with time but specified as a function of
position: q = f(x,y,z) over the part of the boundary surface
in question. In the most general case, flux is specified as a
function of time as well as position: q = f(x,y,z,t).
In all three examples, flux across the boundary ir
specified—that is, it is established in advance and is not
affected by events within the ground-water system; more-
over, it may not deviate from its specified values during
problem solution.
If the direction normal to the boundary coincides with a
major axis of hydraulic conductivity, the expression
obtained above for flux from the boundary,
q = -K,,(3h/3n), can be used to provide a statement of the
boundary condition in terms of head. For the constant-flux
boundary we have dh/dn = constant, and for the two more
general cases we have dh/dn = f(x,y,z) and
dh/dn = f(x,y,z,t), respectively.
5. Head-dependent fitx boundary.—In some situations,
flux across a part of the boundary surface changes in
response to changes in head within the aquifer adjacent to
the boundary. In these situations, the flux is a specified
function of that head and varies during problem solution as
the head varies. An example of this type of boundary is the
upper surface of an aquifer overlain by a sera icon fin ing
bed that is in turn overlain by a body of surface water. This
type of boundary is illustrated by line BC in figure 4. The
head in the surface-water body remains constant, and the
1 A continuous wall of driven piles, generally made of thick planks or
corrugated sheet steel.
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
flux, q, across the semiconfining bed is given by Darcy's
law as
q=-K'
H-h
b' '
where
K' is the hydraulic conductivity of the semiconfin-
ing bed;
b' is its thickness;
H is the head in the surface-water body, and
h is the head in the aquifer.
Thus, flux is a linear function of head in the aquifer—as
head falls, flux across the semiconfining bed increases,
and as head rises, flux decreases.
Inherent in most head-dependent boundary situations is
a practical limit beyond which changes in head cease to
cause changes in flux. In the example cited above, this limit
will be reached where the head within the aquifer falls
below the top of the aquifer, so that the aquifer is no longer
confined at that point, but rather is locally under an
unconfined or water-table condition, while the semiconfin-
ing unit above remains saturated from top to bottom.
Under these conditions, the bottom of the semiconfining
bed becomes locally a seepage face (discussed later) in
the sense that it responds to atmospheric pressure in the
unsaturated aquifer immediately beneath it. Thus, with
atmospheric pressure considered to be zero, the head at
the base of the semiconfining unit is simply the elevation
(z,) of that point above datum, and no matter how much
additional drawdown now occurs in the underlying aquifer;
the flux through the semiconfining bed remains constant,
as given by
q=-K'-
b'
Thus, in this hypothetical case, flux through the confining
bed increases linearly as the head in the aquifer decreases
until the head reaches the level z,, after which flux remains
constant.
This behavior; or some form of it, is characteristic of
almost all head-dependent flux boundaries; for example,
evapotranspiration from the water table is often repre-
sented as a flux that decreases linearly with the depth of
the water table below land surface and becomes zero when
the water table reaches some specified "cutoff" depth,
such as 8 ft below land Surface. In terms of water-table
elevation or head above datum, this is equivalent to a flux
that is zero whenever head is below the specified cutoff
level and that increases linearly as head increases above
that level.
Common designations for the five boundary conditions
described above are summarized in table 1. The last two
boundary types—free surface and seepage surface—are
unique to liquid-flow systems governed by the gravity force
Surface water body
..Constant head on uooer surface
/ ol conlimno. oeo
B
ft/ Leaky confining bed
Aquifer
Impermeable layer
Figure 4. —A leaky aquifer system.
and have no counterpart in systems involving heat flow or
flow of electrical current.
6. Free-surface boundary (h =2 or, more generally
h =f(z)).—The most common free-surface boundary is
the water table, which is the boundary surface between the
saturated flow field and the atmosphere (capillary zone
not considered). An important characteristic of this
boundary is that its position is not fixed—that is, it may
rise and fall with time. In some problems, for example,
analysis of seepage through an earth dam (fig. L4), the
position of the free surface is not known beforehand but
must be found as part of the problem solution, which
complicates the problem solution considerably.
The pressure at the water table is atmospheric. If we
imagine a hypothetical piezometer with its bottom at the
water table, we see that pressure head equals zero (p/*y
= 0, no fluid in the piezometer). Thus, the total ground-
water head at points along the water table is just equal to
the elevation head, or h=z. This situation is analogous to
the head at the surface of a static fluid body, as discussed
previously. (See fig. 2 and related discussion.)
Examples of "top boundary" free surfaces that may be
treated as water tables are line CD in figure L4, the top
boundary in figure 3, line CO in figure 5, and line AJB in
figure 6. In all these examples the position of the water
table partly determines the geometry of the saturated flow
system. Furthermore, the position of the water table in
these systems could change significantly from (he positions
illustrated through a change in the absolute head value at
constant-head boundaries (figs. L4, 5, and 6) or in the
quantity of areal recharge to the water table (fig. 3).
Because these changes in heads and fluxes at boundaries
alter the geometry of the flow system, the relationship
between changes at boundaries and changes in heads and
flows must be nonlinear. This nonlinear relationship is an
important characteristic of ground-water systems with
free-surface boundaries.
Another example of a free-surface boundary .is the
transition between freshwater and underlying sea water in
a coastal aquifer. If we neglect diffusion and assume the
salty ground water seaward of (he interface (o be static.
the freshwater-saltwater transition zone can be treated as a
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
Table 1.—Common designations for several important boundary conditions
Boundary condition
name and reference
number In this
report
Boundary type
and
general name1
Formal
namel
Mathematical
Designation
Constant head (1)
and
specified head (2}
Type 1
(specified head)
Dlrlchlet
constant.
Streamline or
streaa surface (3)
and
specified flux (4)
Type 2
(specified flux)
dh
Neumann
constant.
Head-dependent
flux (5)
Type 3
(nixed boundary
condition)
Cauchy
dh
+ ch • constant.
(where c Is also
a constant)
1 See Bear, 1979, p. 96-98
sharp interface and can be taken as the bounding stream
surface (no-flow boundary) of the fresh ground-water flow
system. It is not difficult to show that, under these
conditions, the freshwater head at points on the interface
(or within the saltwater body) varies only with the eleva-
tion, z (Bennett and Giusti, 1971), and that the freshwater
head at any point on this idealized stream-surface bound-
ary is thus a linear function of the elevation of that point, or
h = f(z). Line CD in figure 7 is an example of this
"idealized" boundary condition, which is both a free-
surface and a no-flow boundary.
Because of the inherent difficulty in modeling ground-
water systems with free-surface boundaries, the represen-
tation of such systems is sometimes facilitated through a
set of simplifying assumptions proposed by Dupuit (the
"Dupuit assumptions") in the 19th century. A list and
discussion of these assumptions can be found in most
textbooks on ground-water hydrology. (See, for example,
Freeze and Cherry, 1979, p. 188,189.)
7. Seepage-surface or seepage-face boundary (h =z).—A.
surface of seepage is a boundary between the saturated
flow field and the atmosphere along which ground water
discharges, either by evaporation or movement "downhill"
along the land surface as a thin film in response to the
force of gravity. The location of this type of boundary is
generally fixed, but its length is dependent on other system
boundaries. A seepage surface is always associated with a
free surface (boundary condition 6). The junction point
(or line in three dimensions) of the seepage face and the
free surface (position of junction point determines the
length of the seepage face) is generally not known during
formulation of a problem but must be determined as a
part of the solution. The situation is in that sense analo-
gous to the free-surface boundary, and the equation
expressing the seepage-surface boundary condition is also
analogous to that for a free surface: h = z along a seepage
face.
Examples of seepage faces are represented by line DE
in figure L4, line BC in figure 5, and line BC in figure 6.
Study of the flow nets in these figures shows that the
seepage face is neither an equipotential line nor a stream-
line but a surface of discharge, as mentioned previously.
As the illustrations indicate, seepage faces may be associ-
ated with individual wells or with earth dams or embank-
ments. Seepage faces are often neglected in models of
large aquifer systems because their effect is often insig-
nificant at a regional scale of problem definition. However,
in problems defined over a smaller area, which require
more accurate system definition (for example, those
depicted in the illustrations cited above), they must often
be considered.
Some important aspects of
specifying boundary conditions
in ground-water models
The preceding sections give a basic introduction to the
boundary conditions most commonly used in modeling
ground-water systems. The following discussion provides
additional information on boundaries and the specification
of boundary conditions in ground-water models.
Model boundaries versus
physical boundaries
It is useful to distinguish conceptually between three
classes of boundary conditions—those associated with
anarytical solutions of boundary-value problems, those
associated with ground-water-system models (digital for
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
the most part, but also analog and other types), and those
associated with natural (real-world) ground-water sys-
tems. The first two classes of boundary conditions are
virtually the same except fhat analytical solutions may
involve an unbounded region. Far example, in the Theis
well solution (see any basic text on ground-water hydrol-
ogy, for example, Freeze and Cherry, 1979, p. 343), the
confined aquifer extends laterally to infinity. Assuming an
infinite boundary sometimes simplifies the analytical solu-
tion or is necessary in obtaining an analytical solution.
Obviously, infinite aquifer dimensions do not occur in
natural systems or in numerical, analog, or physical mod-
els of them.
In formulating a ground-water modeling problem, it is
essentiaJ to distinguish carefully between the "physical"
boundaries of the natural system and the boundaries of
the model. Unfortunately, they are often not the same. To
ensure that the proposed model boundaries will have the
same effect as the natural system boundaries, the follow-
ing procedure is recommended:
1. Identify as precisely as possible the natural "physical"
boundaries of the system, even if they are distant
from the area of concern;
2. Wherever the proposed model boundaries differ from
the natural system boundaries, prepare a careful
justification (both conceptually and in the written
report of the investigation) to show that the proposed
model boundary is appropriate and will not cause the
model solution to differ substantially from the
response that would occur in the real system.
Simple examples of model boundaries that do not
correspond to physical boundaries are lines AI and GH in
figure 1A, BF and DG in figure IB, and DE and AF in
figure 1C. In all three examples, the natural flow system
extends beyond these boundaries, perhaps for a consider-
able distance. Thus, to model the flow system in the region
of interest—that is, cbse to the engineering structures-
it is necessary to establish lateral model boundaries some-
where near the structures. The question of where these
boundaries should be located and what conditions should
be assigned to them is critical to the success of (he model.
Experience with many solutions to this general type of
problem (two-dimensional seepage flow beneath engineer-
ing structures in Vertical cross section) indicates that if the
distance to the lateral boundaries is at least three times
the depth of the flow system, further increases in the
distance have.only a slight influence on the potential
distribution near the structure. In these problems, water
flows from and toward the nearly horizontal constant-head
boundaries, and the lateral boundaries are usually desig-
nated as bounding streamb'nes. The many available solu-
tions to problems of this type provide a kind of "sensiuv ity
analysis" on the position of the lateral boundaries. In
modeling ground-water systems whose boundaries are
more complicated and whose geometries arc less regular
i.o
Impermeable layer
Figure 5.—Flow pattern in a permeable dam having vertical
faces. (From Wyckoff and Reed. 1935.)
Impermeable layer
Figure 6.—Flow pattern near a discharging well in an
unconfined aquifer. (From Nahrgang. 1954.)
than in these examples, a sensitivity analysis may be
needed to select an appropriate boundary position and
type for a model boundary where no corresponding
physical boundary exists. These tests should, of course, be
made in the early stages of the investigation.
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
(« vj
Figure 7.—Flow pattern near a seawater>freshwater interface. (From Glover. 1964.)
In some modeling studies, ground-water divides have
been chosen as lateral model boundaries in the belief that
they represent a physical boundary in the natural system.
Ground-water divides are not boundaries in the sense
used in this report; rather, they are features of the
potential distribution that can be expected to change or
disappear when stresses are introduced. Their use as
boundaries can sometimes be justified without a sensitivity
analysis if the only objective of simulation is to gain an
understanding of the natural flow system in its unstressed
condition; when the objective goes beyond this, however,
positioning a model boundary at a ground-water divide
requires the same process of justification, which should
include sensitivity tests, as the positioning of any other
model boundary that does not coincide with a physical
boundary.
Selection of boundary conditions in
relation to system stress
An important consideration in selecting model bound-
ary conditions is that the choice often depends on the
location and magnitude of the stresses applied to the
system. As a first example, consider an aquifer bounded by
a small to medium-sized stream in a humid environment.
If the stress is small and is some distance from the stream,
the streambed may be treated in a model as a spatially
varyingspecified-head boundary. If, on the other hand, the
stress is close to the stream and so large that it causes part
of the stream to dry up, treating the stream as a specificd-
head boundary is no longer physically reasonable because
a specified-head boundary is capable of supplying an
unlimited quantity of water in a model analysis.
As a second example, consider an aquifer bounded by
leaky confining beds above and below and by a freshwater-'
saltwater interface on one side. If the stress is small ana
far enough from the interface, the interface may be
simulated as a fixed stream-surface boundary (no-flow
boundary). If, on the other hand, the stress is great enough
to cause appreciable movement of the interface, use of the
fixed stream-surface boundary would not be appropriate.
If the boundary conditions in a ground-water model are
stress dependent, the model cannot be considered a gen-
eral, all-purpose tool for investigating any stress on the
system because it wfll give valid results only for the specific
stresses it was designed to investigate. The study of a new
stress on the same system may require the development of
a completely new model.
Boundary conditions in
steady-state models
Steady-state simulation has many applications in hydro-
logic investigations. It is used to "analyze the natural
(predevelopment) flow system as well as any new equilib-
rium conditions that have been attained during the course
of development. Calibration of steady-state models pro-
vides information on hydraulic conductivity and transmis-
sivity. Because storage effects are not involved in steady-
state modeling, the results of steady-state calibrations are
often less subject to ambiguity than those of transient-
state calibrations. Steady-state analysis can also be a rapid
method of evaluating new equilibrium conditions that may
develop in response to future stresses.
Often, investigators use a different set of boundary
conditions for unstressed steady-state models than for
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
stressed steady-state or transient-state models of the same
system. These "substitute" boundary conditions may offer
such advantages as (1) easier interpretation or manipula-
tion of model results, (2) ability to model only a part of the
flow system, as opposed to the entire system, or (3) easier
input to a digital model.
As an example, consider a shallow ground-water system
discharging to a small stream. In an unstressed steady-
state model of this system, the stream might be treated as
a specified-head boundary. Flow to the stream in the
model can be calculated from computed heads and com-
pared with field measurements of stream gains. If the
model is stressed, however; a more complex simulation of
the stream may be required, particularly if the stream
stage changes in response to the stress. As a second
example, the water table is sometimes treated in three-
dimensional steady-state models as a specified-head
boundary. Inflow to the model through this boundary may
be calculated from the model results and compared with
field information. Also, treating the water table in this way
may permit simulation of only a part of the flow system
instead of the entire flow system. In this example the
model of the unconfined aquifer behaves as a confined
linear system.
In conclusion, these "substitute" boundary conditions
usually can be employed only in unstressed steady-state
models and, furthermore, they must be compatible with
the investigator's concept of the natural flow system.
The water table
as a boundary
Because of the water table's importance in ground-
water systems and, therefore, in system models, the vari-
ous ways of treating the water table as a boundary that have
been discussed are summarized below for reference.
1. The water table is usually conceptualized as a free-
surface recharge boundary—either where recharge
equals zero and the water table is a stream surface
(as in line CD in fig. IA, b'ne CD in fig. 5, and b'ne
AB in fig. 6) or where recharge equals a specified
value (as in fig. 3) and the water table is neither a
potential surface nor a stream surface.
2. Sometimes the water table acts as a discharge bound-
ary, particularry where it is near land surface and
thus is subject to bsses by evaporation and transpi-
ration. The discharge from the water table in this
case is usually conceptualized as a function of the
depth of the water table below land surface—that is,
as a function of the water-table altitude. Thus, in a
model simulation, the water table is treated as a
head-dependent flux boundary. (See the discussion
of this boundary condition under "Principal Types of
Boundary Conditions.")
3. As discussed in the preceding section, the water table
may also be treated as a specified-head boundary in
unstressed steady-state models; that is, the position
of the water table is fixed as part of the problem
definition.
One way in which the water table differs from other
boundaries is that it acts as a source or sink of water in
transient-state problems because its position is not fixed.
Because the storage coefficient associated with uncon-
fined, or water-table, storage is large, significant quantities
of water are released from storage during a decline in the
water table, and, likewise, significant quantities must be
supplied for a rise in the water table to occur.
Because the water table is so important in natural
systems, and because it has characteristics not common to
other system boundaries and may be simulated by bound-
ary conditions that differ significantly from one another in
their characteristics, the role of the water table in a specific
problem requires special consideration, and its simulation
requires particular care.
Reference elevation in
ground-water models
In all ground-water models (steady state or transient,
absolute head or superposition) a reference ele\-tmon to
which all heads in the model relate is required so that the
model algorithm can calculate one particular solution to
the governing differential equation and associated bound-
ary conditions defining the problem from the existing
family of solutions. (See discussion of solution of differen-
tial equations in the appendix.) In other words, a reference
elevation is needed to define a unique solution to the
differential equation governing the problem.
A fixed reference ele\'adon6 is required in steady-state
ground-water models. In all types of steady-state models
(as well as transient-state models), constant-head or
specified-head boundaries (constant-head or specified-
head nodes in discretized systems), usually associated with
bodies of surface water, automatically provide a fixed ref-
erence elevation. Because most ground-water models have
a constant-head boundary somewhere, the question of a
reference elevation usually need not be considered explic-
itly. Some systems, however, for example a desert valley
having internal drainage and a playa on the valley (Toot, do
not have surface-water bodies associated with them. In
this example the water table beneath the playa. which is
discharging water by evapotranspiration, may be treated as
a head-dependent flux boundary, with the rate of discharge
from the water table by evapotranspiration defined as a
function of the depth of the water table below land surface.
*As an example of a steady-slate problem for which a fucd
reference elevation is not specified, consider a system that is com-
pletely bounded laterally by constant-flux boundaries and has a
pumping well within the flow domain whose discharge equals the
boundary flux. Because no reference elevation is specified, this prob-
lem has a family of solutions—all with equal gradients hui with
differing absolute heads—but no unique solution.
-------
10
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
In this example, the reference elevation for the ground-
water model becomes the land-surface elevation.
In transient-state models of ground-water systems with-
out constant-head boundaries, the initial heads in the
model (the initial conditions) at the beginning of the
simulation provide sufficient reference to establish a
unique solution to the problem. In a sense, as new sets of
head values are calculated for each time step, the refer-
ence heads continually change and equal the calculated
heads at the end of the preceding time step. In another
sense, the heads at the end of any time step are indirectly
related to the initial heads in the model.
In summary, a reference elevation is necessary in all
types of models to obtain a unique solution to the differ-
ential equation governing the problem being simulated.
The only case in which the reference elevation in a
ground-water-system model requires explicit consideration
is a steady-state model without a constant-head boundary.
Concluding remarks
The discussions presented herein emphasize the impor-
tance of selecting appropriate boundary conditions for
models of ground-water systems. If the boundary condi-
tions used during model calibration are not realistic, the
calibration exercise will generally result in erroneous
values of transmissivity and storage to represent the
system, and predictions made by the model may bear little
relation to reality. Even if the transmissivity and storage
distributions have been correctly determined, incorrect
boundary representation in itself can render the model
predictions meaningless.
The selection of boundary conditions is often the most
important technical decision made in a modeling project.
Alternatives should be considered carefully, sensitivity
analysis should always be used, and investigators should
always be ready to revise their initial assumptions regard-
ing boundaries.
Exercises
1. Choose from a set of colored pencils a color for each
type of boundary condition and trace the extent of
each boundary type in figures 1-7. Upon completion,
the colored lines in each figure should form a
continuous, closed curve that outlines each ground-
water flow system. Note that the lateral boundaries in
figures 1,3, and 4 and the bottom boundary in figure
3 are not physical flow-system boundaries and,
therefore, will remain uncolored.
2. Make a sketch and designate the boundary conditions
of the hypothetical ground-water systems repre-
sented by the following well-known formulas:
a. Thcis noncquilibrium formula: The assumptions of
this formula are listed in many books; careful
consideration should reveal that several of these
"assumptions" in fact describe the boundary
conditions of the hypothetical ground-water sys-
tem. How would you set up a numerical model of
this problem?
b. Dupuit formula for radial flow under water-table
conditions:
In r2/r,
c. Thiem equation for flow to a well in a confined
aquifer written in terms of head (h):
2TrKm(h,-h,)
In r2/r,
Consider the various possible relationships
between these "model'* boundary conditions and
the boundary conditions in field situations.
3. Make a sketch in plan view and in cross section of the
following types of ground-water systems and desig-
nate appropriate boundary conditions. Each system
may be represented in several different ways, but
most ground-water hydrologists will probably treat
some boundary conditions in these systems in the
same way.
a. An oceanic island in a humid climate; permeable
materials are underlain by relatively impermeable
bedrock.
b. An alluvial aquifer associated with a medium-sized
river in a humid climate; the aquifer is underlain
and bounded laterally by bedrock of low hydraulic
conductivity.
c. An alluvial aquifer associated with an intermittent
stream in an arid climate; the aquifer is underlain
and bounded laterally by bedrock of low to inter-
mediate hydraulic conductivity.
d. A westers valley with internal drainage in an arid
region; intermittent streams flow from surround-
ing mountains toward a valley floor; part of valley
floor is playa.
e. A confined aquifer bounded above and below by
leaky confining beds.
Initial Conditions
The results that one obtains from a quantitative model
of a ground-water flow system (head values for various
points and times) represent a particular solution to some
form of the ground-water flow equation. Ground-water
flow equations represent general rules on how ground
water flows through saturated earth material. These equa-
tions have an infinite number of solutions. An individual
ground-water problem must be defined carefully so (hat
the particular solution corresponding to that problem can
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
11
be obtained. (See appendix.) Definition of a specific
problem always involves specification of boundary condi-
tions, and in transient-state (time-dependent) problems,
the initial conditions must be specified as well.
Concept of initial conditions
Definition of initial conditions means specifying the
head distribution throughout the system at some particular
time. These specified heads can be considered reference
heads; calculated changes in head through time will be
relative to these given heads, and the time represented by
these reference heads becomes the reference time. As a
convenience, this reference time is usually specified as
zero, and our time frame (expressed in seconds, days,
years) is reckoned from this initial time.
In more formal terms, an initial condition gives head as
a function of position at t = ,0; that is,
h=f(x,y,z;t=0).
(1)
This notation suggests that, conceptually, initial conditions
may be regarded as a boundary condition in time.
In formal presentations dealing with the solution of
differential equations, boundary conditions and initial con-
ditions are usually discussed together. Problems requiring
their specification are known as boundary-value problems
and initial-value problems. Analytical solutions are avail-
able for a relatively small number of boundary-value and
initial-value problems dealing only with simple system
geometries (for example, spheres, cylinders, rectangles)
and aquifer characteristics that are constant or that vary in
a simple way. For the vast majority of these problems,
approximate solution techniques based on numerical
methods (simulation) must be used.
Analytical solutions are often expressed in terms of
drawdowns, riot absolute heads, and use the principle of
superposition. (See Reflly and others, in press). Absolute
heads (h = p/Y + z) relate to a specific datum of elevation
such as in a water-table map, whereas drawdowns are not
related to a datum but represent the difference in head
between two specific water-level surfaces. // we can use
drawdowns rather than absolute heads and use the principle
of superposition in solving a specific problem, we simplify
the task of defining initial conditions in either analytical or
modeling problems.
Specifying initial conditions
in models
This section discusses two problems in specifying initial
conditions in absolute-head models.
The first problem relates to the use of field-measured
head values, obtained at a time when the natural ground-
water system is at equilibrium, to specify initial conditions
in a model. To use these field values of head, the various
natural hydrologic inputs (recharge and discharge) and
field system parameters (hydraulic conductivity and stor-
age coefficients) that caused the observed distribution of
heads must be represented exactly in the model—which is
virtually impossible to achieve in practice. Therefore, in a
transient-state problem, the initial conditions should be
determined through a steady-state simulation of the flow
system at equilibrium. After appropriate adjustments of
model hydrologic inputs and parameters (process of model
calibration), an acceptably close, although not exact, cor-
respondence between model heads and field heads is
obtained, and the model-generated heads should then be
used as initial conditions for subsequent transient-state
model investigations. Use of the model-generated head
values ensures that the initial head data and the model
hydrologic inputs and parameters are consistent. If the
field-measured head values were used as initial conditions.
the model response in the early time steps would reflect
not only the model stress under study but also the adjust-
ment of model head values to offset the lack of correspon-
dence between model hydrologic inputs and parameters
and the initial head values.
The second problem in defining initial conditions is in
the simulation of systems that are not in equilibrium,
where the objective is to predict the effects of an addi-
tional stress on the system at some future time and where
absolute heads, rather than superposition, are to be used.
In this case the simulation strategy would involve the
following steps: (1) identify a period in the past during
which the system was in equilibrium7; (2) carry out a
steady-state simulation for that period to obtain computed
water levels that are acceptably cbse to measured water
levels; (3) use these simulated heads as initial conditions;
and (4) model all intervening stresses, including the new
stress for which the effects are required, to the specified
time in the future.
Of course, if we are interested only in the effect of the
additional specified stress on a linear system and are not
concerned with predicting absolute heads, we can employ
superposition as the simulation strategy (see Reflly and
others, in press). The problem of defining initial conditions
then disappears because the initial conditions are defined
as zero drawdown (or change in head) everywhere in the
system.
7 If a certain pattern of stress on the ground-water system remains
unchanged for a sufficiently long period, the system may achieve
equilibrium with this stress. Thus, system equilibrium can, but does not
necessarily, imply predevelopment conditions.
-------
TECHNIQUES OF WATER-RCSOUKCES INVESTIGATIONS
Estimated avenge ground water level before iewer.no
Estimated average ground-water level liter hydrologic tvitem
has completely adjusted to the effects of sewering
1940
1945 1950 1955 1960 1965 1970 1975
Figure 8.—Hydrograph of wall N1614 tapping the upper glacial aquifer in central Nassau County. N.Y.
Example of specifying
initial conditions in a
field situation
Some of the issues concerning the specification of initial
conditions can be discussed in reference to the well
hydrograph in figure 8, in which the water level in the well
indicates the water-table altitude. The well is in southwest-
ern Nassau County on Long Island, N.Y, where a sewer
system began operation in the early 1950's and, by elimi-
nating recharge to the water table through septic systems,
constituted a stress on the hydrologic system. The sewer
system achieved close to maximum discharge by the
mid-1960's. The upper horizontal line (h equals about 69
ft) represents an "average" water-table altitude at the well
(a point) before sewering. The fluctuations in water level
around the average value represent a response to the
annual cycle of recharge and evapotranspiration and quan-
titative differences in this cycle from year to year. The
lower horizontal line (h equaJs about 52 ft) represents the
average ground-water level after the hydrologic system had
completely adjusted to the effects of sewering. By about
1966, the hydrograph seems to "level off at this elevation,
indicating that the regional system had attained a new
equilibrium with the stress of sewering. The water-level
fluctuations around the lower horizontal line again reflect
only natural recharge and cvapotranspiration cycles.
If we were studying a stress in addition to the sewering,
and if that stress began in 1975, the lower horizontal line
in figure 8 could be taken as the reference level, or initial
Condition. If we were studying a stress that began in 1963,
TiowcvCr, we would have to take the upper horizontal line
as the initial condition and represent the entire sewering
operation, as well as the new stress, in the simulation
because adjustment of the water levels to sewering would
still be in progress in 1963. The decline in head still to
occur after 1963 as a result of sewering would be unknown
and could be predicted with confidence only by including
the sewering stress in the model study.
Concluding remarks
The most important concepts in the application of initial
conditions are as follows:
1. Proper specification of initial conditions in a model of a
natural ground-water system at equilibrium requires
initial hydrologic inputs consistent with the initial
water levels. To achieve this, it is often necessary to
carry out a steady-state simulation of the prepump-
ing condition and to use the results as the initial
condition for the transient-state simulation.
2. If the natural system is not in equilibrium, a previous
period of equilibrium must be identified to specify
initial conditions, and all subsequent stresses must
be included in the model simulation to predict the
absolute head values that wfll occur at a given future
time.
3. Using superposition as part of the modeling strategy
simplifies or avoids the need to specify initial condi-
tions. However, superposition modeling predicts only
water-level changes related to the specific stress
under study and does not predict absolute heads.
Furthermore, superposition may be applied only to
systems (hat exhibit a linear (or almost linear)
response to stress.
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
13
Acknowledgments
The authors are deeply indebted to the other instructors
of the course "Ground-Water Concepts," given at the U.S.
Geological Survey's National Training Center. Over the
years, the training materials for this course, including
this report, have been generated through a melding of the
ideas and work of many individuals, including Eugene
Patten, Ren Jen Sun, Edwin Weeks, Herbert Buxton, and
others.
References
Bear, Jacob, 1979, Hydraulics of groundwater: New York, McGraw-
Hill, 567 p.
Bennett, G.D., 1976, Introduction to ground-water hydraulics: U.S.
Geological Survey Techniques of Water-Resources Investigations,
Book 3, Chapter B2, 172 p.
Bennett, G.D, and Giusti, EV, 1971, Coastal ground-water (low near
Ponce, Puerto Rico: U.S. Geological Survey Professional Paper
750-D. p. D206-D211.
Freeze, RJV, and Cherry, J.A, 1979, Groundwater Englewood Cliffs,
NJ, Prcnticc-Hall, 604 p.
Glover, R.E., 1964, The pattern of fresh-water flow in a coastal
aquifer, in Cooper, H.H, cod others, Sea water in coastal aquifers:
U.S. Geological Survey Water-Supply Paper 1613-C, p. O2-C35.
Hubbert, M.K_ 1940, The theory of ground-water motion: Journal of
Geology, v. 48, no. 8, p. 785-944.
Lohman, S.W, and others, 1972, Definitions of selected ground-water
terms—Revisions and conceptual refinements: U.S. Geological Sur-
vey Water-Supply Paper 1988, 21 p.
Nahrgang, Gunther, 1954, ZurTheorie des vollkommenen und unvoll-
kommencn Brunnens: Berlin, Springer Verlag, 43 p.
Reilly, T.E, Franke, O.L-, and Bennett, G.D., in press, The principle of
superposition and its application in ground-water hydraulics: U.S.
Geological Survey Techniques of Water-Resources Investigations,
Book 3, Chapter B6.
Rushton, K.R., and Redshaw, S.C, 1979, Seepage and groundwater
How: New York, John Wiley, 339 p.
Wyckoff, R.D., and Reed, D.W, 1935, Electrical conduction models
for the solution of water seepage problems: Physics, v. 6, p. 395—401.
-------
14
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
Appendix: Discussion of
the Solution of Differential
Equations and the Role of
Boundary Conditions
The solution of a differential equation describing
ground-water flow provides a distribution of hydraulic
head over the entire domain of the problem, fbr simple
problems, this distribution of hydraulic head can be
expressed formally by a statement giving head as a function
of the independent variables. Fbr one independent space
variable, we may express this statement in general mathe-
matical notation as
h = f(x). (1)
This function, f(x), when substituted into the differential
equation, must satisfy the equation—that is, the equation
must be a true statement. The function f(x) usually
contains arbitrary constants and is called the general
solution of the differential equation.
The solution must also satisfy the boundary conditions
(and the initial conditions for time-dependent problems)
that have been specified for the flow region. To satisfy the
boundary conditions, the arbitrary constants in the general
solution must be defined, resulting in a more specific
function, fp(x), which is called the particular solution to the
differential equation. Thus, a particular solution of a
differential equation is the solution that solves the partic-
ular problem under consideration, and the general solution
of a differential equation is the set of all solutions. The
following example from Bennett (1976, p. 34—44) helps
develop these concepts by using the differential form of
Darcy's law as the governing differential equation in a
specific problem.
An idealized aquifer system (fig. 9A) consists of a
confined aquifer of thickness b, which is cut completely
through by a stream. Water seeps from the stream into the
aquifet The stream level is at elevation h0 above the head
datum, which is an arbitrarily chosen level surface. The
direction at right angles to the stream axis is denoted the
x direction, and x equals 0 at the edge of the stream. We
assume that the system is in steady state, so that no
changes occur with time. Along a reach of the stream
having length w, the total rate of seepage from the stream
(in ffVs, for example) is denoted 2Q. Because only half of
this seepage occurs through the right bank of the stream.
the amount entering the part of the aquifer shown in our
sketch is Q. This seepage moves away from the stream as
a steady flow in the x direction. The resulting distribution
of hydraulic head within the aquifer is indicated by the
dashed line marked "potentiomctric surface." This sur-
face, sometimes also referred to as the "piezometnc
surface," actually traces the static water levels in wells or
pipes tapping the aquifer at various points. The differen-
tial equation applicable to this problem is obtained by
apph/ing Darcy's law to the flow, Q, across the cross-
sectional area, bw, and may be written
dh___Q_
dx~ KA'
(2)
where K is the hydraulic conductivity of the aquifer and A
is the cross-sectional area perpendicular to the direction
of flow; in this problem, A is equal to bw.
Integration of the previous equation gives the general
solution, f(x), as simply
(3)
where C is an arbitrary constant. Two particular solutions
from the family of general solutions are shown in figure
9B, one where the arbitrary constant equals zero (eq. a),
and one where the arbitrary constant equals h,, (eq. b).
The differential equation (Darcy's law) states that if head
is plotted with respect to distance, the slope of the plot will
be constant—that is, the graph will be a straight line. Both
of the b'nes in figure 9B are solutions to the different)-'
equation. Each is a straight b'ne having a slope equal tc
KA- r>
The intercept of equation a on the h axis is h = 0, whereas
the intercept of equation b on the h axis is h = ha. These
intercepts give the values of h at x = 0 and thus provide the
reference points from which changes in h are measured.
The particular solution for the ground-water system
depicted in figure 9A is obtained when the boundary
conditions are considered. In this problem, the head in the
stream, which is represented at x = 0, is designated as the
constant h0. Thus, the line in figure 9B that has an h axis
intercept of ho is the particular solution to the problem as
posed. Therefore, the particular solution, fp(x), of the
governing differential equation in this problem is
(5)
h = h°-RAX'
This solution satisfies the boundary condition at x = 0.
An accurate description of boundary conditions in
obtaining a particular solution to any ground-water prob-
lem is of critical importance. In multidimensional prob-
lems, boundaries are just as important as in the example
above, although their effect on the solution may not always
be as obvious. Assuming incorrect or inappropriate
boundary conditions for a modeling study must inevitably
generate an incorrect particular solution to the problem.
-------
DEFINITION OF BOUNDARY AND INITIAL CONDITIONS
15
- Datum
Figure 9.—Example of solutions to a differential equation: A, idealized aquifer system; B. two of the family of curves solving the
general differential equation for the idealized aquifer system.
In summary, a particular solution to a differential simulate the differential equation by a set of simultaneous
equation is a function that satisfies the differential equation algebraic equations, the concepts are analogous, although
and its boundary conditions. In numerical models that the solution is not a continuous function.
-------
Availability of
Models
-------
5. Availability of Models
-------
"If you're studying the eruption of a
volcano, a computer model is better
than being there'."
Joseph Dcken, "The Electronic College"
-------
Aspects in GW Modeling
• Objectives/goals
• Conceptualization
• Input parameter values
• Code (program) selection
• Calibration/verification
• Sensitivity/uncertainty analysis
• Interpretation of modeling results
' • Report organization
'Model selection is not normally treated
us a process of natural selection."
Philosophy of Model Selection
Familiarity and bias concerning
a specific code can drive the
conceptual model development.
Thus, conceptual model
development becomes dictated
by model selection.
Summary
Problems observed in 20 model applieations
• Input parameter values 12.6/5.0)
• Sensitivity/Uncertainty analysis (2.7/5.11)
• B.C. / I.C. (.1.0/5.U)
• Conceptualization & result interpretation (3.1/5.0)
• Calibration/Verification /5,o)
• Report organization (3.5/s.o)
• Modeling objective (4.0/5.0)
-------
-------
Conceptualization must be
independent of the code
selection process.
Code selection must be based
on conceptual model
development.
The "Goldilocks " Approach
Complex
Simple
Just Right
-------
Model Education
Distribution Centers
Training Courses
Scientific Publications
Technology Transfer Documents
Distribution Centers
AVAILABLE,
ACCESSIBLE..
BUT
SUPPORTED?
Proprietary:
"If an institution owns the copyright,
trademark or patent of the software,
or distributes it solely under license
agreements, the code is considered
\ to be proprietary."
Public Domain:
"A code is considered to be public domain,
when its development has been supported
through public funds, and no distribution
restrictions, copyrights, or patents apply."
Good Model...
Bad Model...
User-Friendly!
-------
To stamp a model as
is to become arrogant regarding our
understanding of natural processes.
'Approved'' Models
We do not know, with absolutes, the
processes that drive contaminant
fate and transport in the subsurface.
-------
Acceptance:
Quality Assurance
Technical & Scientific
Soundness
Accountability
_j
Useability:
Preprocessor, Postprocessor,
User Instructions, Sample Problems,
Hardware Dependence, Support
Reliability:
Peer-Reviewed Theory,
Peer-Reviewed Coding, Verified,
Field Tested, Model Users
Quality Assurance
& Quality Control
• ASTM Standards
• EPA Guidance
• Reference Books
-------
Review Criteria
• Modeling objectives/goals
• Conceptualization
• Input parameter values
• Code selection (numerical/analytical)
• Calibration/validation
• Sensitivity/uncertainty
• Results interpretation
• Report organization
Recommendation
Public Education
• Dissemination of the GW Model Guidelines
• Short courses
• Conferences/workshops
QA/QC or Model Testing
-------
Model Distribution & Support Centers
American Petroleum Institute (API)
1220 "L" Street
Washington, DC 20005
Phone: (202)682-8345
Analytic & Computational Research, Inc. (ACRi)
1931 Stradella Road
Bel Air, CA 90077
Phone: (310)471-3023
FAX: (310)471-0797
Contact: Dr. Akshai K. Runchal
AScI Corporation (AScI)
987 Gaines School Road
Athens, GA 30605
Phone: (706)353-8718
FAX: (706)353-7461
Contact: Dr. Robert W. Schottman
Center for Exposure Assessment Modeling (CEAM)
Environmental Research Laboratory
U.S. EPA
Athens, Georgia 30613
Phone: (706)546-3130
Contact: Dermont Bouchard
Center for Environ. Research Information (CERI)
U.S. EPA
26 West Martin Luther King Drive
Phone: (513)569-7272
BBS: (513)569-7610
NASA Computer Software Technology Transfer
Center (COSMIC)
COSMIC Customer Support
The University of Georgia
382 Broad Street
Athens, Georgia 30602-4272
Phone: (706)542-3265
Contact: John Gibson
Center for Subsurface Modeling Support (CSMoS)
U.S. EPA/NRMRL/SPRD
P.O.Box 1198
Ada, Oklahoma 74820
Phone: (405)436-8586
FAX: (405)436-8718
Website: http://www.epa.gov/ada/kerrlab.html
Colorado State University (CSU)
Ground-Water Program
Civil Engineering Department
Engineering Research Center
Fort Collins, Colorado 80523
Phone: (303)491-8381
Contact: Dr. James Warner
Draper Aden Environmental Modeling, Inc. (DAEM)
2206 S. Main St.
Blacksburg, VA 24060
Phone: (540)961-3236
FAX: (540)552-0291
Contact: Mark Kay nor
Electric Power Software Center (EPSC)
Electric Power Research Institute (EPRI)
11025 N. Torrey Pines Rd., Suite 120
LaJolla,CA 92037
Phone: 1-800-763-3772
FAX: (614)453-4495
Electric Power Research Institute (EPRI)
3412 Hillview Avenue
P.O.Box 10412
Palo Alto, CA 94303
Phone: (415)855-2974
FAX: (415)855-2954
Contact: Ms. Tamsen Gagnon
Environmental Monitoring Systems Lab-Las Vegas
(EMSL-LV, EAD)
P.O. Box 93478
Las Vegas, Nevada 89193-3478
Phone: (702)798-2100
Contact: Evan Englund
Environmental Protection Agency (EPA/UST)
Office of UST
Phone: (703)308-8877
Contact: David Wiley
(General questions for Hyperventilate)
Environmental Simulations, Inc.
2997 Emerald Chase Drive, Suite 100
Herndon.VA 22071
Phone: (703)834-3054
Contact: Jim Rumbaugh
J.R. Williams (August 7, 1996)
-------
Energy Science and Technology Software Center
(ESTSC)
P.O.Box 1020
Oak Ridge, Tennesse 37831
Phone: (615)576-2606
Contact: Ed Kidd
(Distributes DOE Software)
Environmental Systems and Technologies, Inc.
(ES&T)
2608 Sheffield Drive
Blacksburg, VA 24060-6326
Phone: (540)552-0685
(703) 552-0685
FAX: (703)951-5307
Geraghty and Miller, Inc. (G&M)
Modeling Group
10700 Parkridge Boulevard, Suite 600
Reston, Virginia 22091
Phone: (703)758-1200
FAX: (703)758-1201
BBS: (703)758-1203
Internet: p01019@psilink.com
Contact: Karen Crow
Colder Associates, Inc.
4104-148th Avenue, N.E.
Redmond, WA 98052
Phone: (206)883-0777
FAX: (206)882-5498
Contact: William S. Dershowitz
Dr. Charles R. Fitts
Dept. of Geosciences
University of Southern Maine
Gorham, ME 04038
Phone: (207)780-5351
(TWODAN Distribution & Support)
GAEA
1575 Lyons Ave.
Windsor, Ontario
N9J 3K4 Canada
Phone: (800)880-6731
FAX: (519)978-0815
Contact: Lorraine Fraser
General Sciences Corporation (GSC)
6100 Chevy Chase Drive
Laurel, MD 20707-2929
Phone: (301)953-2700
FAX: (301)953-1213
GeoTrans, Inc. (GeoTrans)
46050 Manekin Plaza, Suite 100
Sterling, Virginia 22170
Phone: (703)444-4400
FAX: (703)444-1685
Internet: geotrans@access.digex.net
Contact: David S. Ward
Hydrogeologic, Inc. (Hydrogeologic)
1165 Hemdon Parkway, Suite 900
Herndon, Virginia 22070
Phone: (703)478-5186
Contact: Jan Kool
Illinois State Water Survey (ISWS)
2204 Griffith Drive
Champaign, Illinois 6820
Phone: (217)333-4968
Contact: Marvin Clevenger
International Ground-Water Modeling Center
(IGWMC)
Institute for Ground-Water Research and Education
Colorado School of Mines
Golden, Colorado 80401-3278
Phone: (303)273-3103
FAX: (303) 273-3278
Contact: Forest Arnold
Midwest Research Institute (MWRI)
401 Harrison Oaks Blvd.
Gary, NC 27513
Phone: (919) 677-0249 ext. 5270
Contact: Tom Carson
National Ground Water Association
6375 Riverside Dr.
Dublin, OH 43017
Phone: (614)761-1711
1-800-551-7379
FAX: (614)761-3446
-2-
J.R. Williams (August 7, 1996)
-------
National Technical Information Service (NTIS)
Superintendent of Documents
P.O. Box 371954
Pittsburgh, PA 15250-7954
Phone: (202)783-3238
Office of Exposure Assessment (OEA)
U.S. EPA
401 M Street, SW
Washington, DC 20460
Phone: (202)260-3930
Contact: Annette Nold
Office of Solid Waste (OSW)
Modeling Group OS-331
U.S. EPA
401 M Street, SW
Washington, DC 20460
Phone: (202)260-4765
Office of Toxic Substances (OTS)
U.S. EPA
401 M Street, S.W.
Washington, DC 20460
Phone: (202)382-3926
Oklahoma State University (OSU)
Department of Agronomy
265 Agriculture Hall
Stillwater, Oklahoma 74078-0507
Phone: (405)744-9592
Contact: Dr. David Nofziger
Pacific NW Lab (PNL)
Richland, Washington 99352
Pennsylvania State University (PSU)
212SackettBldg.
University Park, PA 16802
Phone: (814)863-2931
Contact: Dr. George Yeh
Princeton Groundwater
P.O. Box 263033
Tampa, FL 33685
Phone: (813)855-6898
FAX: (813)855-6390
Scientific Software Group (SSG)
P.O. Box 23041
Washington, DC 20026-3041
Phone: (703)620-9214
FAX: (703) 620-6793
S.S. Papadopulos & Associates, Inc. (SSP)
12250 Rockville Pike, Suite 290
Rockville, Maryland 20852
Phone: (301)718-8900
FAX: (301)718-8909
Contact: Chris Neville
University of Florida (UF)
IFAS Software Support
P.O.Box 110340
Gainesville, Florida 32611-0340
Phone: (904)392-7853
Contact: Dennis Watson
University of Wisconsin - Milwaukee
Ctr for Continuing Engineering Education
College of Engineering & Applied Science
929 North Sixth Street
Milwaukee, WI 53203
Phone: (414)227-3200
1-800-638-1828
FAX: (414)227-3146
(Model & Risk Assessment Training)
U.S. Geological Survey (USGS)
437 National Center
12201 Sunrise Valley Drive
Reston, Virginia 22092
Phone: (703)648-6978
Contact: Oliver Holloway
U.S. Salinity Lab (USSL)
4500 Glenwood Drive
Riverside, California 92501
Phone: (909)369-4847
FAX: (909)369-4818
Contact: Martinus Th. van Genuchten
Waterloo Hydrogeologic Software (WHydro)
200 Candlewood Crescent
Waterloo, Ontariom Canada N2L 5Y9
Phone: (519)746-1798
FAX: (519)746-1798
Contact: Dr. Nilson Guiger
-3-
J.R. Williams (August 7, 1996)
-------
Suggested Reading
ASTM. 1996. ASTM1996 Annual Book of Standards, Section 4. Volume 4.09, Soil and Rock (H):
D4943 - latest; Geostnthetics.
Breckenridge, R.P., J.R. Williams, and J.F. Keck. 1991. Ground-Water Issue: Characterizing Soils for
Hazardous Waste Site Assessments. EPA/540/4-91/003, March 1991.
El-Kadi, A.I. (ed). 1995. Groundwater Models for Resources Analysis and Management. CRC Press
Inc./Lewis Publishers, Boca Raton, Florida.
Hem, S.C. and S.M. Melancon. 1986. Vadose Zone Modeling of Organic Pollutants. Lewis Publishers,
Inc.
Luckner, L. and W.M. Schestakow. 1991. Migration Processes in the aoii und Groundwater Zone.
Lewis Publishers, Inc.
National Research Council. 1990. Ground Water Models: Scientific and Regulatory Applications.
National Academy Press, Washington, DC.
National Research Council. 1993. Ground Water Vulnerability Assessment: Contamination Potential
Under Conditions of Uncertainty. National Academy Press, Washington, DC.
Nofziger, D.L., J-S. Chen, and C.T. Haan. 1993. Evaluation of Unsaturated/Vadose Zone Models for
Superfund Sites. EPA/600/R-93/184, May 1993.
Spitz, K. And J. Moreno. 1996. A Practical Guide to Groundwater and Solute Transport Modeling.
John Wiley & Sons, Inc., New York, New York.
Stephens. D.B. 1996. Vadose Zone Hydrology. CRC Press, Inc./Lewis Publishers, Boca Raton,
Florida.
U.S.EPA. 1994. Assessment Framework for Ground-Water Model Applications. OSWER Directive
#9029.00, EPA/500/B-93/003, July 1994.
van der Heijde, P.K.M. 1994. Identification and Compilation of Unsaturated/Vadose Zone Models.
EPA/600/R-94/028
van der Heijde, P.K.M. and O. Elnawawy. 1992. Quality Assurance and Quality Control in the
Development and Application of Ground-Water Models. EPA/600/R-93/011, September 1992.
van der Heijde, P.K.M. and O. Elnawawy. 1993. Compilation of Ground-Water Models. EPA/600/R-
93/118, May 1993.
- 7 - J.R. Williams (August 7, 1996)
-------
Van der Heijde, P.K.M. 1996. Compilation of Saturated and Unsaturated Zone Modeling Software
(Update ofEPA/600/R-93/118 and EPA/600/R-94/028). EPA600/R-96/009, March 1996 or
NTISPB96167606.
Weaver, J., C.G. Enfield, S. Yates, D. Kraemer, and D.White. 1989. Predicting Subsurface
Contaminant transport and Transformation: Considerations for Model Selection and Field
Validation. EPA/600/2-89/045, August 1989.
Wilson, L.G., L.G. Everett, and S.J. Cullen (eds.). 1995. Handbook of Vadose Zone Characterization
and Monitoring. CRC Press Inc./Lewis Publishers, Boca Raton, Florida.
- 8 - J.R. Williams (August 7, 1996)
-------
Top Ten : August http://www.epa.gov/epahome/current_topl0.html
HomePa°e Comments Search Index
The top 10 data area retrievals from the U.S. EPA's WWW server for the month of August.
1. EPA Main Home Pages (total hits: 182.555)
2. Office of Water (total hits: 83,777)
3. Office of Air and Radiation (total hits: 50.096)
4. Office of Policy. Planning, and Evaluation Home Page (total hits: 47,439)
5. EPA and Ozone Depletion (total hits: 33,509)
6. The Federal Register (FR) Environmental Subset (total hits: 30,423)
7. Air Pollution Prevention Division (total hits: 23,506)
8. Envirofacts (total hits: 21,677)
9. Great Lakes National Program Office (GLNPO) (total hits: 21,512)
10. National Risk Management Research Laboratory - Subsurface Protection and Remediation Division
(total hits: 21,191)
Total Completed Requests (www.epa.gov): 998,571
Last updated: Sun Aug 11 07:05:011996.
\ EPA Home Page I Comments I Search I Index ]
internet support@unixmail.rtpnc.epa.gov
REVISED: August 11,1996
URL: http://www.epa.gov/epahome/current_topl0.html
of 1 08/11/96 16:07:4
-------
Robert S. Kerr Environmental Research Center
P.O.Box 1198
Ada, Oklahoma 74820
(405) 436-8500
Welcome to the U.S. EPA's Kerr Lab Home Page
Resources brought to you by Kerr Lab:
•Mission Statement for this Laboratory
•Center for Subsurface Modeling Support CSMoS * Free Software *
•Technology Support Center * Weekly Highlights have moved to this page *
• Subsurface Remediation Information Center (SRIC) *** * Project Summaries added *
^Research Projects at Kerr Lab ;m
Opportunities at Kerr Lab:
•Employment Announcements :*«
« Research Opportunities »*"
•Kerr Lab Guest Book Let us know what you think about these pages!
-------
Links to other Related Home Pages:
® Environmental Protection Agency WWW Server
®NERL/ERD/Athens Extramural Research Programs Home Page
»USEPA Center for Exposure Assessment Modeling (CEAM)
^International Ground Water Center, Colorado School of Mines
aGroundwater Modeling Links
^Richard B. Winstons Home Page
«ENVision's list of quality sites on Earth Sciences and Geology
»GWRTAC Ground-Water Remediation Technologies Analysis Center**
This page is maintained by Dan West (CDSI) at NRMRL - Subsurface Protection & Remediation
Division, Ada, OK.
Comments, suggestions, or questions about these pages may be sent to:
Dan West, dwest@ad3100.ada.epa.gov
or to the EPA Coordinator:
Joe Williams, williams@ad3100.ada.epa.gov
Graphics Support provided by Martie Williams (CDSI)
I accesses since October 1, 1995
Last updated July 22, 1996.
-------
US EnmtnnKitil Pctfrtion Agency
Cento foi Subsuibcc Modeling Stfpait
Rittoral RUkMxajnnckl RtsaiAi Litaratay
SufcsuzfKf Protection itdRemrdbrlon DMjtai
F.O. Box 1198
Welcome to the Center for Subsurface Modeling Support
The Center for Subsurface Modeling Support, or CSMoS, provides a source for publicly available
ground-water and vadose zone modeling software and services. CSMoS was established in 1989 to
provide a focal point for the distribution of models and databases developed through in-house and
extramural research activities, to provide technical support for these models and databases, and to
provide review of model applications at hazardous waste sites. CSMoS is located in Ada, Oklahoma as a
function of ORD's Subsurface Protection and Remediation Division of the National Risk Management
Research Laboratory, or NRMRL/SPRD. NRMRL/SPRD is the former Robert S. Kerr Environmental
Research Laboratory. As an integral part of NRMRL/SPRD's Technology Support Center, CSMoS
consists of in- house research and technical support personnel, and is supported in its' day-to-day
activities through contractor personnel located both on-site and off-site.
CSMoS provides assistance in the following areas of modeling: conceptualization, development,
application, distribution, training, and education. Since inception in 1989, CSMoS has distributed over
8000 copies of models and databases, along with their associated user's manuals. Members of the
CSMoS team have provided model application review support to over 100 sites nationwide through the
Technology Support Center. Shortly after the start-up of CSMoS, the NRMRL/SPRD began activities in
the area of geographic information systems (GIS). Since that time, GIS activities have focussed on the
better presentation of site characterization information and the integration of model utilization with GIS
packages. NRMRL/SPRD researchers are utilizing models and GIS to enhance research efforts in
analytic element methods, bioremediation simulation, and visualization of site characteristics.
CSMoS is co-directed by Dr. David S. Burden (Soil Scientist/Hydrologist) and Mr. Joe R. Williams (Soil
Scientist) in the Technical Assistance and Technology Transfer Branch. Software and user's manuals can
be obtained through this Web Page, through the anonymous FTP (ftp.epa.gov /pub/gopher/ada/models),
and through a bulletin board at (405)436-8506.
Ground-Water and Vadose Zone Models/Manuals
Add your name to the CSMoS mailing list
Home Page Links:
-------
This page is maintained by Dan West (CDSI) at NRMRL - Subsurface Protection & Remediation
Division, Ada, OK.
Comments, suggestions, or questions about these pages may be sent to:
Dan West, dwest@ad3100.ada.epa.gov
or to the EPA Coordinator:
Joe Williams, williams@ad3100.ada.epa.gov
Last updated April 16, 1996.
-------
CSMoS Home Page http://www.epa.gov/adii/models2.htm
v^eo Sr^
y
U.S. Ercnionmtnul Piouaion Ajtruy
Onto foi Subsurfice Modtlici; Sifpon _
0 Nltiorol Rnk MniJfmint Rtstuch Libcoltoy C
T Subsujftct Piouaion ind Rtmrtotjon Divisicn Q
P.O. Box 1196
Adj,0kl*iorta 74820 ^
**'•>.
Ground-Water and Vadose Zone Models/Manuals available from CSMoS
Brief Description of All CSMoS Models (35,936 bytes) m
Note: The above model list file and all documentation for
CSMoS models are in Adobe Acrobat (PDF) format:
The Adobe Acrobat Reader is available Free from Adobe
Click here for Instructions from Adobe on configuring you web browser for PDF files
Important! You will need the following instructions to install the software after you download it:
^^Installation Instructions for all CSMoS Software (9683 bytes)
•B1OPLUME II
•BIOSCREEN *Coming Soon* iw*g
-------
HSSM (Ver 1.10 - Apr '94)
Hydrocarbon Spill Screening Model
Application: Simulates flow of the LNAPL phase and transport of
a chemical constituent of the LNAPL
from the surface to the water table; radial
spreading of the LNAPL phase at the water table,
and dissolution and aquifer transport of the
chemical constituent.
Processes: One-dimensional in the vadose zone, radial in the
capillary fringe, two-dimensional vertically
averaged analytical solution of the
advection-dispersion equation in the saturatead
zone.
Miscellaneous: Model is based on the KOPT, OJLENS and TSGPLUME
models.
•Download Installation Instructions for all CSMoS Software (9,683 bytes)
^Download HSSM for Windows Installation Disk (607,954 bytes)
® Download HSSM for DOS Installation Disk (831,398 bytes)
» Download HSSM Volume 1 : Users Guide (1,205,858 bytes) jfi
•Download HSSM Volume 2: Theoretical Background and Source Codes (1,61 1,238
•Download HSSM Volume 2: Parameter Variation Data Sets (44,579 bytes)
IIDownload HSSM Readme File (30,974 bytes) jg
Note: The above documentation files are in Adobe Acrobat (PDF) format: The Adobe Acrobat Reader is
available Free from Adobe
• Example applications of HSSM to field data ^
Back to the List of Models
This page is maintained by Dan West (CDSI) at NRMRL - Subsurface Protection & Remediation
Division, Ada, OK.
Comments, suggestions, or questions about these pages may be sent to:
Dan West, dwest@ad3100.ada.epa.gov
or to the EPA Coordinator:
Joe Williams, mlliams@ad3100.ada.epa.gov
Last updated July 18, 1996
-------
U.S. EPA CEAM Software (DOS) ftp://ftp.epa.gov/epa_ceam/wwwhtml/software.htm
CENTER FOR EXPOSURE ASSESSMENT MODELING
Software, Descriptions, Abstracts (DOS)
[This page best viewed within Netscape 2.0. A text only version of this page is available.]
DOS releases of selected CEAM software are available through the World Wide Web (WWW).
Individual CEAM software products are packaged in unique files with a file name of the form
INSTALxx.EXE where "xx" is a 2 character abbreviation for the software product name (e.g.,
INSTALQ2.EXE for QUAL2EU model system).
These files are compressed, full-screen, interactive, self-extracting, installation programs with
on-line context sensitive help. Executable, full documentation (where noted), test input and
example output, and FORTRAN source code file(s) included. These software products are
designed for DOS based microcomputer systems, NOT UNIX systems.
Iof3 08/07/9612:12:39
-------
U.S. EPA CEAM Software (DOS)
ftp://ftp.epa.gov/epa_ceam/wwwhtrnl/software.htrn
CEAM Software
File Name/Size (MB)
i . .
INSTALAN.EXE /1. 28
INSTALCI.EXE / .5
INSTALCM.EXE /1. 03
INSTALCF.EXE 74.35
INSTALEX.EXE /1. 46
INSTALFG.EXE /1. 07
INSTALFW.EXE /1. 05
INSTALHS.EXE / 7.46
!
INSTALMM.EXE /N/A
INSTALMT.EXE / .89
I
INSTALTD.EXE / .29
INSTALOF.EXE / .34
INSTALP2.EXE / 2.76
INSTALPL EXE / 1 44
INSTALPT.EXE / 5.43
INSTALO2.EXE 72.21
. _ „.. - ..
INSTALSW.EXE 7 1.6
INSTALSX.EXE 7 .39
INSTALWP.EXE 7 3. 14
Description/ Abstract
ANNIE-IDE tool kit
CEAM information system
CORMIX model 7 documentation
CQRMIX User Guide Figures
EXAMS model 7 documentation
FGETS model system
""
FEMWATER model 7 documentation
HSPF model 7 documentation
MULTIMED model 7 documentation
MINTEO model svstem
MINTEO documentation
Sample ANNIE-IDE application
PRZM2 model 7 documentation
PLUMES model 7 documentation
PATRIOT model 7 documentation
OUAL2EU model svstem
SWMM model svstem
.
SMPTOX3 model 7 documentation
WASP model 7 documentation
Ver. No.
1.14
3.21
,2.10
1.00
2.96
3.0.18
1.00
10.11
2.00
3.11
3.U ""
1.61
2.00
3.00
1.20
3.22 ™
4.30
2.01
5.10
Rel. Date
Sep91
May 95
Oct93
Jun94
Mar 96
Sep94
Jul93
Apr 95
Sep95
Dec 91
Jan 93
Sep91
Oct94
Dec 94
Nov94
May 96
May 94
Feb93
Oct93
Other files, documents, and/or software for DOS are available through the World Wide Web
(WWW). Files and/or software with a ".ZIP" file name extension are compressed and packaged
using PKZIP (tm) and require PKUNZIP (tm) version 1.10 or greater to un-compress.
2 of 3
08/07/96 12:12:40
-------
U.S. EPA CEAM Software (DOS)
ftp://ftp.epa.gov/epa_ceam/wwwhtml/software.htm
Other CEAM Software
File Name/Size (MB)
NPSMODEL.ZIP / .28
SUBD.DEF / .01 :
SWDATA.EXE / .41
PRZM2Q1B.ZIP / .43
READ201B.ME/.01 '
Description/Abstract
Compressed Modeling Non-Point Source Water Quality Documents
Default data file for MULTIMED Subtitle D applications
Additional data sets to test sections of SWMM v. 4.20
PRZM2 version 2.01 Beta executable file and information
READ.ME file latest information on PRZM2 model v. 2.01 Beta
[Home | Top | Comment/Feedback | Mailing List | Search | Index]
Mailing Address
Webmaster: ceam(2)epamail.epa.gov
Revised July 22,1996
URL: ftp://ftp.epa.gov/epa_ceam/wwwhtml/software.htm
3 of 3
08/07/96 12:12:40
-------
U.S. EPA CEAM Software (UNIX)
ftp://ftp.epa.gov/epa_ceam/ww\vhtml/softunix.htm
CENTER FOR EXPOSURE ASSESSMENT MODELING
Software, Descriptions, Abstracts (UNIX)
[This page best viewed within Netscape 2.0. A text only version of this page is available.]
UNIX releases of selected CEAM software are available through the World Wide Web (WWW).
Individual CEAM software products are packaged in unique files with a file name of the form
xxxxxxxx.tar.Z where "xxxxxxxx" is a 1 to 8 character abbreviation for the software product
name (e.g., przm2.tar.Z for PRZM2 model system). To install a software product file set
down load the product compressed distribution file
uncompress the distribution file (e.g., uncompress przm2.tar.Z)
untar the software product file (e.g., tar xvf przm2.tar)
change to the product sub-directory (e.g., cd przm2)
execute "make" command file to build executable file (e.g., make)
execute or start an xterm session (e.g., xterm)
execute the software product (e.g., ./przm2)
These are compressed, tar files. Full documentation (where noted), test input and example output,
FORTRAN source code file(s), and "make" command files included. These software products are
designed for UNIX based systems.
CEAM Software
• File Name/Size (MB) • Description/Abstract Ver. No. j Rel. Date
'I przm2.tar.Z/2.30 I PRZM2 model /documentation I 2.01 «« I Jul 96
przm2.read.me / 0.03 I PRZM2 Model / documentation ' 2.01 NEW j Jul 96
[Home | Top | Comment/Feedback | Mailing List | Search | Index]
Mailing Address
Webmaster: ceam@CDamail.cna.gov
1 of 2
08/07/96 12:13:09
-------
Robert S. Kerr Environmental Research Center
P.O. Box 1198
Ada, Oklahoma 74820
(405) 436-8500
Welcome to the U.S. EPA's Kerr Lab Subsurface Modeling Links Page
Center for Subsurface Modeling Support CSMoS
http://www.epa.gov/ada/csmos.html
Environmental Protection Agency WWW Server
http://www.epa.gov
NERL/ERD/Athens Extramural Research Programs Home Page
http://www.epa.gov/docs/AthensR/
USEPA Center for Exposure Assessment Modeling (CEAM1
ftp: //ftp. epa.go v/epa_ceam/wwwhtml/ceamhome. htm
International Ground Water Center. Colorado School of Mines
http://igwmc.mines.Colorado.edu:3851/I /
* ASCE Seepage/Groundwater Modeling Links
http://www.et.byu.edu/~asce-gw
Richard B. Winstons Home Page
http://aapg.geol.lsu.edu/rbwinsto.htm
ENVision's list of quality sites on Earth Sciences and Geology
http://www.envision.net/osites/geology/geology.html
* GWRTAC Ground-Water Remediation Technologies Analysis Center
http://www.gwrtac.org
This page is maintained by Dan West (CDSI) at NRMRL - Subsurface Protection & Remediation
Division, Ada, OK.
Comments, suggestions, or questions about these pages may be sent to:
Dan West, dwesl(^,ad31 OO.ada. epa.gov
or to the EPA Coordinator:
Joe Williams, williamsffiadS 1 OO.ada.epa.gov
Last updated August 06, 1996.
-------
ASCE
Geotechnical Engineering
Seepage/ Groundwater Modeling Software
Welcome to the ASCE Geotechnical Engineering Seepage/Groundwater Modeling Software web pages.
These pages are a part of the ASCE Geotechnical Engineering Division World Wide Web Site. Any
comments or suggestions to improve these pages are welcomed.
These pages have been designed to allow any interested party to create a link pointing to web pages that
describe seepage/groundwater modeling software. If you do not have any existing web pages describing
the software, feel free to use the create link form to compose a web page that will be stored here. After a
link has been created, you can edit a link, or delete a link as you please.
The group moderator reserves the right to delete any link at his/her own discretion.
DISCLAIMER: The data contained in this web site is provided for information only. ASCE does not
recommend or endorse any of the software that is listed on this site.
Seepage/ Groundwater Modeling Software
* Department of Defense Groundwater Modeling System (GMS)
URL: http://www.et.byu.edu/~geos/software/gms/gms.html
Contact: Russell J. Berrett
A comprehensive graphical user environment for numerical modeling. Provides tools for site
characterization, model conceptualization, mesh and grid generation, geostatistics, and
sophisticated tools for graphical visualization. The current version of GMS provides a
complete interface for the codes MODFLOW, MODPATH, MT3D and FEMWATER.
* FastSeep (Automated Seepage Modeling)
URL: http://www.et.byu.edu/~geos/software/fastseep/fastseep.html
Contact: Norm Jones
Graphical User Interface for SEEP2D. Available for both Windows and UNIX environments.
GCT/Partnership in Computation Sciences
URL: http://www.isc.tamu.edu/PICS/
Contact: Alan Peery
GCT, for "Groundwater Contaminant Transfer", is the result of a modeling effort carried out
by the broad-based Partnership in Compuational Sciences. Version 1.3, under development in
1995, uses mixed methods to model contaminant flow, bioremediation, and chemical reaction.
The 3D logically rectangular domain is computed using domain decomposition methods.
* Draper Aden Environmental Modeling, Inc.
URL: http://www.daem.com/daem/
Contact: Douglas Garnett-Deakin
This home page contains information on models written by DAEM for multiphase flow and
multicomponent transport in the subsurface. DAEM provides consulting and specialized
software development for advanced numerical models, as well as database management and
3-D graphics and animation.
* Hydrology Web
URL: http://terrassa.pnl.gov:2080/EESC/resourcelist/hydrology.html
Contact: Tim Scheibe
Pacific Northwest Laboratory's Earth and Environmental Sciences Center (EESC) hydrology
resource list.
* Groundwater Remediation Project
URL: http://gwrp.cciw.ca/
-------
Contact: Andrew Piggott
This World-Wide Web server is maintained by the Groundwater Remediation Project of the
National Water Research Institute of Environment Canada, and is located at the Canada Centre
0 for Inland Waters in Burlington, Ontario.
Groundwater Information System
URL: http://wtrwww.env.gov.bc.ca/gwis.html
Contact: Rodney Zimmerman
Brought to you by The Groundwater Section, of the Water Management Program. The
Groundwater Program is delivered throughout British Columbia by Water Management
Regional Staff.
USGS Water Resources of the United States
URL: http://h2o.er.usgs.gov/
Contact: Ken Lanfear
The U.S. Geological Survey is the Nation's largest earth-science agency and has the principal
responsibility within the Federal government for providing hydrologic information and for
appraising the Nation's water resources. Hydrologic data and other data are used in research
and hydrologic studies to describe the quantity, quality, and location of the water resources of
the U.S.
Universities Water Information Network
URL: http://www.uwin.siu.edu/
Contact: UWIN staff
This Web site is a product of the Universities Council on Water Resources, provided through a
grant from the U.S. Geological Survey.
* Environmental Professional's Guide to the Net
URL: http://www.geopac.com/
Contact: Greg Arnold
This homepage grew out of a need for quickly locating technical information on the Internet
related to site characterization and cleanup projects.
* Environmental Simulations, Inc.
URL: http://www.us.net/envisim/welcome.html
Contact: James Rumbaugh
Environmental Simulations, Inc. (ESI) was established to provide government and industry
with superior groundwater modeling software, services, and training. Groundwater modeling
is now a tool routinely employed in environmental projects; however, it takes a skilled analyst
to make the most of this technology.
The SSEC Visualization Project
URL: http://ssec.wisc.edu/~billh/vis.html
Contact: Bill Hibbard
The Visualization Project at the Space Science and Engineering Center (SSEC) of the
University of Wisconsin-Madison focuses on making advanced visualization techniques useful
to Earth scientists in their daily work. We accomplish this goal by making two scientific
visualization systems, named VisSD and VisAD, freely available over the Internet, and by
using these systems as testbeds for exploring and evaluating new techniques.
* BOSS International
URL: http://www.bossintl.com/
Contact: Chris Maeder
Distributes groundwater related software including: DoD Groundwater Modeling System,
FastSeep , Visual MODFLOW, Flowpath, AirFlow/SVE, Spatial Explorer, Flotrans, and
Prince.
EarthSoft
URL: http://www.earthsoft.com/
Contact: Chris Kutler
EarthSoft markets a number of environmental software products for data management,
visualization and modeling.
* International Ground Water Modeling Center
URL: http://igwmc.mines.colorado.edu:3851/
Contact: IGWMC
-------
International Groundwater Modeling Center at Colorado School of Mines.
ADEPT
URL: http://www.us.net/adept/welcome.html
Contact: ADEPT
9 A Program for Aquifer Data Evaluation
Subsurface Modeling Group
URL: http://riceinfo.rice.edu/projects/subsurf/home.html
Contact: Mary F. Wheeler
The Subsurface Modeling Group (SMG) investigates the use of high-performance parallel
9 processing as a tool to simulate the behavior of petroleum reservoirs and groundwater aquifers.
Developing Software for Earth Scientists
URL: http://l30.223.104.15/software.html
Contact: Yvan Pannatier
This page provides links to various software developed in the department of Earth Sciences,
University Lausanne, Switzerland.
E03/6
URL:http://www.llnl.gov/IPandC/opportunities93/03-ENVIRONMENT/EQ_3_6.shtml
Contact: Joe Milner
EQ3/6 can model the complex geochemical processes that result from interactions among the
host rock,nuclear waste, groundwater, air, and waste.
* PEST
URL: http://gil.ipswichcity.qld.gov.au/comm/pest/index.html
Contact: PEST
Parameter Estimation for any Model. PEST is a nonlinear parameter estimation "shell" which
communicates with any existing model through the model's own input and output files.
Free Groundwater Modeling Software
URL: http://aapg.geol.lsu.edu/rbwinsto.htm
Contact: Richard B. Winston
DOS executable versions of MODFLOW, MODFLOWP, SUTRA, MODFE, VS2DT and
other groundwater modeling programs can be downloaded. Source code is included. There are
also links to other web pages dealing with groundwater modeling, hydrology, and wetlands.
Natural Neighbor Interpolation
URL: http://maths.uwa.edu.au/~watson/homepage.html
Contact: Dave Watson
Natural neighbor interpolation is the most robust and reliable method for interpolation of
sparse scattered data in 2D, 3D, or higher dimensions. It is an appropriate technique for
generating models of groundwater aquifers and contaminant plumes.
Starpoint Software
URL: http://sage.cc.purdue.edu/~rpitteng
Contact: Starpoint Software
Windows based aquifer test analysis software. Commercial programs are available for the
analysis of aquifer pumping tests, slug tests, and step-drawdown tests.
* U.S. EPA Groudwater Modeling Software and Information
URL http://www.epa.gov/ada/kerrlab.html
Contact: Joe Williams
This is the home page for the U.S. EPA's National Risk Management Research Laboratory /
Subsurface Protection and Remediation Division (formerly known as the R.S. Kerr
Environmental Research Laboratory. The sub-page for the Center for Subsurface Modeling
Support (CSMoS) has groundwater modeling software available for download.
GAEA Environmental Engineering Ltd.
URL: http://www.wincom.net/gaea
Contact: Mike Fraser
Software for Environmental and Geotechnical Engineering. Including POLLUTEv6 and
MIGRATEv9 contaminant transport modelling software.
Ecological Data Consultants. Inc.
URL: http://gnv.ifas.ufl.edu/~rsm/wetlands.htm
Contact: WETLANDS GROUP: R.S. Mansell. Ali Fares. S.A. Bloom
-------
WETLANDS is a multi-dimensional water flow solute transport variably saturated
mathematical model. It dynamically links a surface water body with subsurface groundwater.
Potential evapotranspiration can be estimated for as many as three different plant spcecies
given a minimum weather data. WET-PREP is a menu-drive program. It is used to construct
input files for WETLANDS. It includes utilities to estimate parameters of soil water release
curves using different analytical models (van Genuchten,...). WET-PLOT, an interactive
visualization plotting program, is included in the package.
Environmental Systems & Technologies. Inc.
URL: http://www.esnt.com
Contact: Rick Parrish
Assessment, Remedial Strategies & Litigation Support, and Advanced Computer Applications.
ES&T carries over 40 programs for groundwater modeling, including multiphase flow and
transport, data management and visualization. ES&T offers expert modeling, visualization,
animation, multimedia and Internet services, in addition to training and tech support. Check
out our online newsletter for info on our upcoming short course in Orlando! New additions to
our free FTP software area. (540) 552-0685 e-mail: admin@www.esnt.com
Argus Numerical Environments/Argus Interware. Inc.
URL: http://www.argusint.com
Contact: Sharon Grossman
Graphical pre/post-processors for numerical modelers, combining GIS, CAD, Math, data
import, auto mesh/grid generation, relational data base, visualization, and export to all models.
Runs on Mac/PowerMac, PC and UNIX. Widely used by groundwater modelers.
AquaLogic
URL: ftp://ftp.ccnet.com/users/aqualog
Contact: Nicolas Spycher
Home of PUMPIT, a practical analytical radial flow model to estimate well drawdown,
pumping rate, and spacing of groundwater recovery and injection wells. It is useful to generate
groundwater flow maps and data tables, and to delineate capture zones, injection fronts, and
migration paths of plumes and fronts of particles.
S.S. Papadopulos & Associates. Inc.
URL: http://access.digex.net/~sspa
Contact: Charlie Andrews
Environmental and Water-Resource Consultants (Ground-Water Investigations, Remediation,
Transport Modeling and Software).
Earthware Water Well and AOuifer Test Analysis Textbook.
URL: ./htdocs/aaaaez.html
Contact: Earthware Phil Hall
This book is published by Water Resources Publications and includes software.
ParFlow Project at Lawrence Livermore National Laboratory
URL: http://www.llnl.gov/CASC/ParFlow/
Contact: Steven F. Ashbv
Enabling detailed simulations of groundwater flow and contaminant migration through the use
of high performance computing technologies.
* The Modflow Help File
URL: http://aapg.geol.lsu.edu/modhelp.htm
Contact: Richard B. Winston
The Modflow help file is an online guide to Modflow and provides a list of Modflow-related
internet resources. Version 2.0.1 available now.
University of Manitoba Groundwater Shareware Database and FTP site !!
URL: http://www.umanitoba.ca/geo_eng/Groundwater/
Contact: Greg Desrosiers
Groundwater Shareware available. Links to related site can be found here !
ToxSpill Predictive GW Contamination Software
URL: http://ourworld.compuserve.com:80/homepages/jim_wilkins
Contact: Jim Wilkins
Soil & groundwater modeling software package designed and published by HydroScience Inc.
Hvdrogeology Group at University of Alabama
-------
URL: http://hydro.geo.ua.edu
Contact: Chunmiao Zheng
The Hydrogeology program offers interdisciplinary curricula leading to an M.S. or Ph.D.
degree in the Department of Geology. The hydrogeology group is actively engaged in research
on a wide range of issues of both scientific and practical implications on the nation's
groundwater resources. This site includes downloadable software developed by the research
0 group.
Ground-Water Remediation Technologies Analysis Center (GWRTAC)
URL: http://www.gwrtac.org
Contact: Turgut T. Onay or Kathy Jacox
The GWRTAC was established through a cooperative agreement between the U.S. EPA and
the Center for Hazardous Materials Research (CHMR) to facilitate in the transfer of innovative
ground water remediation technology information. The role of GWRTAC is to collect,
compile, assimilate and provide current information concerning innovative ground water
remediation technologies to the ground water community. Check out our on-line technical
documents on innovative ground water remediation technologies, and vendor information
. databases.
Geotechnical & Geo-environmental Software Directory
URL: http://www.ibmpcug.co.uk/~bedrock/gsd/
Contact: Tim Spink
Details several hundred programs, software publishers and suppliers in the fields of
Hydrogeology, Geo-environmental Engineering, Geotechnical Engineering, Engineering
Geology, Data Analysis and Data Visualisation. Also lists other WWW pages with related
software.
SEEP/W
URL: http://www.geo-slope.com/seepw01 .htm
Contact: J. Paul Brvden
SEEP/W, for Microsoft Windows, is a finite element software product for analyzing
groundwater seepage and excess pore-water pressure dissapation problems. The
comprehensive formulation of SEEP/W makes it possible to consider analyses ranging from
simple saturated steady-state problems to sophisticated saturated / unsaturated time-dependent
problems.
Water Resources Publications, LLC
URL: http://www.waterplus.com/wrp
Contact: Water Resources Publications, LLC
Water resources and related publications & software
* Research Center for Groundwater Remediation Design (RCGRD)
URL: http://www.rcgrd.uvm.edu
Contact: David E. Dougherty
The Research Center for Groundwater Remediation Design focuses on the development of
least-cost optimization for groundwater contamination characterization, cleanup, and
monitoring design.
Eclipse Technology
URL: http://www.eclipsetech.com
Contact: David Gottholm
Eclipse provides environmental software products that enable an organization of any size to
manage and integrate their environmental data more effectively. Eclipse also provides
consulting services for customized software design and development.
GeoTrans Groundwater Specialist
URL: http://www.geotrans.com
Contact: David S. Ward
GeoTrans is a comprehensive, technical environmental consulting firm offering groundwater
flow and transport codes including: ModelGIS, SiteGIS, SWIFT/486, MODFLOWT, BIO1D
and others.
* ChemPath — Windows-based Solute Transport Modeling Package
URL: http://www.baseline-concept.com
Contact: Raiesh Lalwani
-------
ChemPath has an innovative technique for 3-D solute transport modeling. It is extremely easy
to use. Runs in Windows 3.1 and Windows 95. Comprehensive online help. Perfect
companion for PATH3D, MODPATH, Micro-Fern, and FLOWPATH. Runs in minutes as
compared to hours taken by other 3-D transport packages. Powerful post-processing. You can
9 also export the output in SURFER or DXF format.
Porous Media Modelling Group
URL: http://www.informatik.unibw-muenchen.de/instl/infl .html
Contact: Ulrich Hornung
The group works on modelling of flow and transport through porous media. Special topics are
9 optimization of soil vapor extraction, capillary barriers, and homogenization.
Ground Water Software & Publications
URL: http://www.scisoftware.com
Contact: Susan Hardy
Scientific Software Group announces its 1995-96 Environmental Software & Publications
Catalog. SSG carries mostly hydrology products, but has geology and other environmental
ones as well.
VENT2D
URL: http://www.hydro.unr.edu/homepages/benson/v2d.html
Contact: David Benson
This free software simulates advection and diffusion of up to 60 volatile compounds. The
mobile chemical vapors may also partition into the dissolved, adsorbed, and NAPL phases.
The model is used to simulate vapor extraction of multicomponent spills from the vadose
zone.
* GIS/KEY from GIS Solutions
URL: http://www.gisedm.com\gisedni\
Contact: Robert Romzick (510)-602-9208
Link to GIS/Solution's Home Page. A relational database with a graphical interface...Providing
management, analysis, manipulation, validation, and more for all environmental data. Exports
data to GMS.
Micro-Fern Groundwater Flow Modeling
URL: http://www.xs4all.nl/~microfem/index.html
Contact: Micro-Fern
Micro-Fern is an integrated large-capacity finite-element microcomputer program for
multiple-aquifer steady-state and transient groundwater flow modeling
EarthVision« by Dynamic Graphics
URL: http://www.dgi.com
Contact: Heather Kellev
Integrated software products for 3-D geospatial analysis.
PPATH (Soil Interconnection Simulation Model)
URL: http://CC.USU.EDU/~slvdO/ppath.html
Contact: Alaa All
PPATH Model simulates flow preferential pathways by searching for the highest conductivity
interconnected paths in 3-D dual porous media.
* DLOG3D (Bore hole data interpretation Model)
URL: http://cc.usu.edu/~slvdO/dlog3d.html
Contact: Alaa AH
This model uses nonparametric techniques to characterize and map subsurface soil variability
in a nonstationary depositional environment.
* GIS-Ground Water Flow and Contaminant Fate and Transport Models and Multimedia Analytic
Contaminant Transport Simulation Models (ACTS)
URL:http://www.ce.gatech.edu/Projects/WaterResources/aral/edrp_hom.html
Contact: Mustafa M. Aral
PC based ARC/INFO Geographic Information Systems integrated ground water flow and
contaminant transport modeling software for multilayer, nonhomogeneous, anisotropic
confined/unconfined aquifers. User friendly interface on PC Arc/Info platform through
ArcView. Also, (ACTS) a multimedia (air/ground water/surface water) analytic contaminant
transport analysis models. Uncertainty analysis and graphics packages included with a user
-------
friendly interphase.
[ Create II Edit || Delete ]
Norman L. Jones
norm(q).byu.edu
Access Statistics
-------
ASTM GUIDES RELATED TO GROUND WATER MODELING
Volume 04.09: Soil & Rock (II)
D 5447-93 Guide for Application of Ground-Water Flow Model Simulation to a Site
D 5490-93 Standard Guide for Comparing Ground-Water Flow Model Simulations to Site-
Specific Information
D 5609-94 Standard Guide for Defining Boundary Conditions in Ground-Water Flow
Modeling
D 5610-94 Standard Guide for Defining Initial Conditions in Ground-Water Flow Modeling
D 5611-94 Standard Guide for Conducting a Sensitivity Analysis for a Ground-Water Flow
Model Application
D 5718-95 Standard Guide for Documenting a Ground-Water Flow Model Application
D 5719-95 Standard Guide for Simulation of Subsurface Air Flow using Ground Water Flow
Modeling Codes
Volume ???:
D 5880-95 Standard Guide for Subsurface Flow and Transport Modeling
Volume 11.05: Pesticides; Resource Recovery; Hazardous Substances & Oil Spill Responses;
Waste Management; Biological Effects.
E 978-92 Standard Practice for Evaluating Mathematical Models for the Environmental
Fate of Chemicals
E 1689-95 Standard Guide for Developing Conceptual Site Models for Contaminated Sites
E 1739-95 Standard Guide for Risk-Based Corrective Action Applied at Petroleum Release
Sites
ASTM Standards - 1
-------
Standards currently in draft or at Subcommittee or Committee/Society level ballot:
Standard Guide for Conceptualization and Characterization of Ground-Water Flow
Systems
Standard Guide for Calibrating a Ground-Water Flow Model Application
Standard Guide for Application of a Solute Transport Model to a Site-Specific Ground-
Water Problem
Standard Guide for Developing and Evaluating Ground-Water Modeling Codes
Also available is a compilation of modeling and aquifer test analysis standards, Standards on
Analysis of Hydrologic Parameters and GW Modeling; ASTM Publication Code Number 03-
418096-38; $59 non-members, $53 members; contains 14 standards on aquifer analysis, and 9
standards on ground-water modeling.
ASTM Standards - 2
-------
GROUND WATER MODELING PUBLICATIONS
SYSTEMATIC EVALUATION AND TESTING OF
GROUNDWATER MODELING CODES
by
Paul K.M. van der Heijde
International Ground Water Modeling Center
Colorado School of Mines
Preprint
IAHS/1AH/IGWMC InternatConf. ModelCARE'96
Golden, Colorado, USA
Sept. 24-26,1996
IGWMC-GWMI 96-03
February 1996
ig/wc
INTERNATIONAL GROUND WATER MODELING CENTER
Colorado School of Mines
Golden, CO 80401, USA
-------
SYSTEMATIC EVALUATION AND TESTING OF GROUNDWATER MODELING CODES
PAUL K.M. VAN DER HEIJDE
International Ground Water Modeling Center, Colorado School of Mines, Golden, Colorado 80401, United
States.
Abstract Effective use of groundwater simulation codes as management decision tools requires the
establishment of their functionality and performance characteristics, and evaluation of their applicability to the
problems at hand This is accomplished.through systematic code-testing using informative evaluation techniques,
and through careful code selection. In this paper, a two-component protocol is introduced consisting of
functionality analysis and performance evaluation. Functionality analysis is the description and measurement
of the capabilities of a simulation code; performance evaluation concerns the appraisal of a code's operational
characteristics (e.g.. computational efficiency, sensitivity for problem layout and parameter selection, and
reproducibility). The results of code testing are analyzed using standardized statistical and graphical techniques.
INTRODUCTION
Reliability of groundwater model predictions typically depends on the correctness of the conceptual model, the
availability and quality of model data, and the adequateness of the predictive tools, hi groundwater modeling,
the predictive tools consist of one or more computer codes for data analysis, system simulation, and presentation
of results. This paper focuses on the testing of the computer codes used in predicting groundwater responses.
The importance of this aspect of groundwater modeling is illustrated by the efforts currently underway within the
American Society for Testing and Materials (ASTM) to codify the systematic description and the testing of the
capabilities of groundwater modeling codes and within the American Society of Civil Engineers (ASCE) to
provide guidance on this issue.
The development of a groundwater modeling code may be part of a research or development project, based
on an existing mathematical model, or derived from an existing set of modeling codes. Code development in
groundwater modeling is often part of research aimed at acquiring new, quantitative knowledge about nature
through observation, hypothesizing, and verifying deduced relationships, leading to the establishment of a
credible theoretical framework for the observed phenomena. The resulting research model represents a
fundamental understanding of the studied groundwater system. The object for model research in groundwater
is a subset of the hydrologic system, called the reference system. It contains selected elements of the global
hydro-logic system. The selection of a particular reference system is influenced by regulatory and management
priorities, and by the nature of the hydrologic system (Fig. 1). The conceptual model of the selected reference
system forms the basis for quantifying the causal relationships among various components of this system, and
between this system and its environment. These relationships are defined mathematically, resulting in a
mathematical model. If the solution of the mathematical equations is complex or when many repetitious
calculations are required, the use of computers is essential. This requires the coding of the solution to the
mathematical problem in a programming language, resulting in a computer code. The conceptual formulations,
mathematical descriptions, and computer coding constitute the (generic) model (Fig. 1). Attributing the
parameters and stresses in the generic model results in an operational model of the reference system .sometimes
referred to as a groundwater model application.
The development of a groundwater modeling code typically consists of: 1) definition of design criteria and
determination of applicable software standards and practices; 2) the development of algorithms and program
structure; 3) computer programming; 4) preparation of documentation; 5) code testing; and 6) independent review
of scientific principles, mathematical framework, software and documentation.
When the development of a groundwater modeling code is initiated, procedures are formulated to ensure that
the final product conforms with the design objectives and specifications, and that it correctly performs the
-------
incorporated functions. These procedures cover the formulation and evaluation of the code's theoretical
foundation and code design criteria, the application of coding standards and practices, and the establishment of
the code's credentials through review, systematic testing of its functional design, and evaluation of its
performance characteristics. The two major approaches to achieve acceptance of a groundwater modeling code
are: 1) the evaluation or (peer) review process covering all phases of the code development process; and 2)
quantitative comparison with independently obtained data for the reference groundwater system.
CODE TESTING
A systematic approach to code testing combines elements of error-detection, evaluation of the operational
characteristics of the code, and assessment of its suitability to solve certain types of management problems, with
dedicated test problems, relevant test data sets, and informative performance measures. The results of code
testing are expressed in terms of correctness (e.g., in comparison with a benchmark), reliability (e.g.,
reproducibility of results, convergence and stability of solution algorithms, and absence of terminal failures),
efficiency of coded algorithms (in terms of numerical accuracy versus code execution time, and memory and mass
storage requirements), and resources required for model setup and analysis (e.g., input preparation time, effort
needed to make output ready for graphic analysis) (van der Heijde and Kanzer, 1995).
The code-testing protocol described in this paper is applied in a step-wise fashion. First, the code is analyzed
with respect to its simulation functions and operational characteristics. Potential code performance issues are
identified, based on analysis of simulated processes, mathematical solution methods, computer limitations and
execution environment. This is followed by the formulation of a test strategy, consisting of design or selection
of relevant test problems. The set of test problems is chosen such that all code functions and features of concern
are addressed. Results of the testing are documented in tables and matrices providing an overview of the
completeness of the testing, in various types of informative graphs, and with a set of statistical measures. The
actual testing may take the form of benchmarking using known, independently derived solutions, intra-
comparison using different code functions inciting the same system responses, inter-comparison with comparable
simulation codes, or comparison with field or laboratory experiments. It is important that each test is documented
with respect to test objectives, model setup for both the tested code and the benchmark, if applicable (structure,
discretization, parameters), and results for each test (for both the tested code and the benchmark).
FUNCTIONALITY ANALYSIS AND PERFORMANCE EVALUATION
Functionality of a groundwater modeling code is defined as the set of functions and features the code offers the
user in terms of model framework geometry, simulated processes, boundary conditions, and analytical and
operational capabilities. The code's functionality needs to be defined in sufficient detail for potential users to
asses the code's utility, as well as to enable the code developer to design a meaningful code testing strategy.
Functionality analysis involves the identification and description of the code's functions, and the subsequent
evaluation of each code function or group of functions for conceptual correctness and computational accuracy
and consistency (e.g., reproducibility and numerical stability) (van der Heijde and Kanzer, 1995). Comprehensive
testing of a code's functionality and performance is accomplished through a variety of test methods. Determining
the importance of the tested functions and the ratio of tested versus non-tested functions provides an indication
of the completeness of the testing.
The information generated by functionality analysis is organized into a summary structure, or matrix, that
brings together the description of code functionality, code-evaluation status, and appropriate test problems. This
functionality matrix is formulated by combining a complete description of the code functions and features with
the objectives of targeted test problems. The functionality matrix illustrates the extent of the performed
functionality analysis.
Performance evaluation is aimed at characterizing the operational characteristics of the code in terms of: 1)
correctness; 2) overall accuracy; 3) reliability; 4) sensitivity for grid orientation and resolution, and for time
-------
discretization; 5) efficiency of coded algorithms (including bandwidth, rate of convergence, memory usage, and
disk I/O; and 6) level of effort and resources required for model setup and simulation analysis (van der Heijde
and Kanzer, 1995). Results of the performance evaluation are expressed both quantitatively and qualitatively
in checklists and in tabular form. Reporting on performance evaluation should provide potential users
information on the performance as a function of problem complexity and setup, selection of simulation control
parameters, and spatial and temporal discretization. The functionality matrix and performance tables, together
with the supporting test results and comments, should provide the information needed to select a code for a site-
specific application and to evaluate the appropriateness of a code used at a particular site.
CODE TESTING STRATEGY
A code testing strategy is developed before actual code testing takes place. Such a code testing strategy should
consist of: 1) formulation of test objectives (as related to code functionality and test issues), and of test priorities
(Tables 1 and 2); 2) selection and/or design of test problems and determination of type and extent of testing for
selected code functions; 3) determination of level of effort to be spent on sensitivity analysis for each test
problem; 4) selection of the qualitative and quantitative measures to be used in the evaluation of the code's
performance; and 5) determination of the level of detail to be included in the test report and the format of
reporting (Table 3).
The test procedure includes the three levels of testing defined in van der Heijde and Elnawawy (1992). At
Level I, a code is tested for correctness of coded algorithms, code logic and programming errors by: 1) conducting
step-by-step numerical walk-throughs of the complete code or through selected parts of the code; 2) performing
simple, conceptual or intuitive tests aimed at specific code functions; and 3) comparing with independent,
accurate benchmarks (e.g., analytical solutions). If die benchmark computations themselves have been made using
a computer code, this computer code should in turn be subjected to rigorous testing by comparing computed
results with independently derived and published data.
At Level II, a code is tested to: 1) evaluate functions not addressed at Level I; and 2) evaluate potentially
problematic combinations of functions. At this level, code testing is performed by intracomparison (i.e.,
comparison between runs with the same code using different functions to represent a particular feature), and
intercomparison (i.e., comparison between different codes simulating the same problem). Typically, synthetic
data sets are used representing hypothetical, often simplified groundwater systems.
At Level III, a code (and its underlying theoretical framework) is tested to determine how well a model's
theoretical foundation and computer implementation describe actual system behavior, and to demonstrate a code's
applicability to representative field problems. At this level, testing is performed by simulating a field or
laboratory experiment and comparing the calculated and independently observed cause-and-effect responses.
Because measured values of model input, system parameters and system responses are samples of the real system,
they inherently incorporate measurement errors, are subject to uncertainty, and may suffer from interpretive bias.
Therefore, this type of testing will always retain an element of incompleteness and subjectivity.
The test strategy requires that first Level I testing is conducted (often during code development), and, if
successfully completed, this is followed by Level 2 testing. The code may gain further credibility and user
confidence by subjecting it to Level 3 testing (i.e., field or laboratory testing). Although, ideally, code testing
should be performed for the full range of parameters and stresses the code is designed to simulate, in practice this
is often not feasible due to budget and time constraints. Therefore, prospective code users need to assess whether
the documented tests adequately address the conditions expected in the target application(s). If previous testing
has not been sufficient in this respect, additional code testing may be necessary.
CODE TESTING EVALUATION CRITERIA
Evaluation of code testing results should be based on: 1) visual inspection of the graphical representation of
variables computed with the numerical model and its benchmark; and 2) quantitative measures of the goodness-
-------
of-fit. Such quantitative measures or evaluation or performance criteria, characterize the differences between
the results derived with the simulation code and the benchmark, or between the results obtained with two
comparable simulation codes.
Graphical measures are especially significant for test results that do not lend themselves to statistical
analysis. For example, graphical representation of solution convergence characteristics may indicate numerical
oscillations and instabilities in the iteration process. Practical considerations may prevent the use of all data-pairs
in the generation of graphical measures. Thus, a subset of data-pairs may be selected for use with graphical
measures. The selection of a set of representative sample data-pairs may be based on symmetry considerations,
model domain areas with potential higher deviations, or on specific interest in subdomains (i.e., vertical or
horizontal slices of the model domain). There are five types of graphical evaluation techniques particular suited:
1) X-Y plots or line graphs of spatial (e.g., distance) or temporal behavior of dependent variable and other
computed entities (Fig. 2 and Fig. 3); 2) one-dimensional column plots or histograms (specifically to display test
deviations); 3) combination plots of line graphs of dependent variable and column plots of deviations (Fig. 4);
4) contour and surface plots of the spatial distribution of the dependent variable and the residuals; and 5) three-
dimensional, isometric, column plots or three-dimensional histograms (Fig. 5). The conclusions from visual
inspection of graphic representations of testing results may be described qualitatively (and subjectively) by such
attributes as "poor," "reasonable," "acceptable," "good," and "very good".
There are three general procedures, coupled with standard linear regression statistics and estimation of error
statistics, to provide quantitative goodness-of-fit measures (Donigjan and Rao, 1986): 1) paired-data performance
« the comparison of simulated and observed data for exact locations in time and space; 2) time and space
integrated, paired-data performance — the comparison of spatially and temporally integrated or averaged
simulated and observed data; and 3) frequency domain performance — the comparison of simulated and observed
frequency distributions. The organization and evaluation of code inter-comparison results can be cumbersome
due to the potentially large number of data-pairs to be analyzed if every computational node is included. This
can be mitigated by analyzing smaller, representative sub-samples of model domain data-pairs. The
representativeness of the selected data-pairs is often a subjective judgement. For example, in simulating one-
dimensional, uniform flow, the data pairs should be located on two lines parallel to the flow direction, one in the
center of the model domain and one at the edge. Another example is the simulation of the Theis problem; here,
two lines of data pairs should be chosen parallel to the two horizontal principal hydraulic conductivity axes, while
a third set of data pairs should be on a line at 45 degrees to these axes. Test cases that are symmetrical can be
analyzed for a smaller portion of domain based upon the type of symmetry present. For .example, test cases that
have radial symmetry can be divided into four equal representative radial slices; this significantly decreases the
required number of data pairs in the analysis and considerably reduces the evaluation effort.
Useful quantitative evaluation measures for code testing include (van derHeijde and Kanzer, 1995): DMean
Error (ME), defined as the mean difference (i.e., deviation) between the dependent variable calculated by the
numerical model and the benchmark value of the dependent variable; 2) Mean Absolute Error (MAE), defined
as the average of the absolute values of the deviations; 3) Positive Mean Error (PME) and Negative Mean Error
(NME), defined as the ME for the positive deviations and negative deviations, respectively; 4) Mean Error Ratio
(MER), a composite measure indicating systematic overpredicting or underpredicting by the code; 5) Maximum
Positive Error (MPR) and Maximum Negative Error (MNE), defined as the maximum positive and negative
deviation, respectively, indicating potential inconsistencies or sensitive model behavior; and 6) Root Mean
Squared Error (RMSE), defined as the square root of the average of the squared differences between the
dependent variable calculated by the numerical model and its benchmark equivalent.
Various computed variables may be the focus of graphic or statistical comparison:
For saturated flow codes - hydraulic heads (in space and time), head gradients, global water balance, internal
and boundary fluxes, velocities (direction and magnitude), flow path lines, capture zones, travel times, and
location of free surfaces and seepage surfaces;
For unsaturated flow codes ~ hydraulic heads or suction heads, water content or saturation, head gradients,
global water balance, and internal and boundary fluxes;
-------
For solute transport codes - concentrations, mass fluxes, global mass balance (per species), and breakthrough
curves at observation points and sinks (wells, streams).
CONCLUSIONS
Historically, reporting on simulation code testing has been limited to the use of author-selected verification
problems. Few studies have focused on author-independent evaluation of a code, or at code intercomparison.
Main deficiencies in reported code-testing efforts include incompleteness of the performed testing, absence of
discussion regarding tested code functions as compared with available code functions and features, and lack of
detail in test problem implementation making it difficult to recreate the data sets for additional analysis. The
protocol developed by van der Heijde and Kanzer (1995) aims to address these issues, hi addition, the protocol
covers many other test issues, ranging from performance and resource utilization to usefulness as a decision-
making support tool.
The code testing protocol is designed to be applicable to all types of simulation codes dealing with fluid flow
transport phenomena in the unsaturated and saturated zones of the subsurface. Selection and implementation of
test problems will differ for the different types of codes. However, evaluation approaches and techniques are in
principle independent of code type. Test results are presented in a form which is unbiased by the requirements
posed by specific applications. It aims to provide enough detail to establish confidence in the code's capabilities
and to determine its applicability to specific field problems. Because users of code testing results may differ in
terms of objectives, the protocol leaves it to the users to determine if a tested code is suitable to their needs. The
most critical element of the code testing protocol is the functionality analysis (including elements of what is often
called "code verification"). Many different test configurations can be used for each type of code and, for some
code types, a large number of benchmark solutions may be available. For other code types, inter-comparison
using specially designed hypothetical problems or simple field cases may be the only available option. Selection
of test problems and benchmarks should be guided by test objectives and in the context of the completeness of
the testing exercise. Protocol tools such as functionality tables and functionality matrices are effective aids in
the design of test problems. Well-designed tests not only identify code functionality problems, but may also
provide important information on correct implementation of code features. Functionality analysis may be limited
because not all code features can be adequately addressed using benchmark solutions. Often, code
intracomparison, code intercomparison and conceptual testing are required, resulting in a more subjective
assessment of code accuracy and operational constraints.
The functionality analysis, performance evaluation and applicability assessment protocol, presented in this
paper provides a comprehensive framework for systematic and in-depth evaluation of a variety of groundwater
simulation codes. While allowing flexibility in implementation, it secures, if properly applied, addressing all
potential coding problems. It should be noted that the protocol does not replace scientific review nor the use of
sound programming principles. Most effectively, the code testing under the protocol should be performed as part
of the code development process. Additional testing in accordance with the protocol may be performed under
direction of regulatory agencies, or by end-users. If properly documented, code testing in accordance with the
protocol supports effective independent review and assessment for application suitability. As such, the protocol
contributes significantly to improved quality assurance in code development and use in groundwater modeling.
REFERENCES
Donigian, Jr., A. S., & Rao, P. S. C. (1986) Example model testing studies. In: Vadose Zone Modeling of
Organic Pollutants (ed. by S. C. Hern and S. M. Melancon), Lewis Publishers, Chelsea, Michigan.
van der Heijde, P. K. M., El-Kadi, A. I. & Williams, S. A. (1988) Groundwater modeling: an overview and status
report. EPA/600/2-89/028, U.S. Environmental Protection Agency, Ada, Oklahoma.
-------
van der Hcijde, P. K. M., and Elnawawy, O. A. (1992) Quality assurance and quality control in the development
and application of ground-water models. EPA/600/R-93/011, U.S. Environmental Protection Agency, Ada,
Oklahoma.
van der Heijde, P. K. M., and Kanzer, D. A. (1995) Ground-Water Model Testing: Systematic Evaluation and
Testing of Code Functionality and Performance. Submitted to U.S. Environmental Protection Agency under
CR-818719, International Ground Water Modeling Center, Golden, Colorado.
Table 1. Functionality issues for advective and dispersive solute transport
Functionality Issue
Test Objective
Type of Test
Advection-dominatcd transport often creates
numerical problems in the vicinity of the
solute front.
To determine accuracy in terms of
concentrations and mass balance, to evaluate
stability and the occurrence of oscillations and
numerical dispersion, and to perform sensitivity
analysis with respect to transport parameter
values, and spatial and temporal discretization.
steady-state uniform flow
transient transport
benchmark, Level IB
Accuracy of simulation of dispersive transport
is dependent on grid orientation. Inclusion of
cross-terms of the dispersion coefficient may
improve accuracy.
To determine sensitivity of concentration
distribution and mass balance for grid
orientation.
steady-state uniform flow
transient transport
benchmark, Level IB
Accuracy of dispersive transport may be
influenced by the contrast in the main
directional components of the dispcrsivity,
especially when using non-optimal grid
orientation.
To determine accuracy of concentration
distribution and mass balance for different
ratios for the dispcrsivity components.
steady-state uniform flow
transient transport
benchmark. Level IB
Sometimes, advective-dispersive transport is
negligible and molecular diffusion is
prominent.
To determine accuracy in terms of
concentrations and mass balance when
molecular diffusion is important
transient
benchmark, Level 1A
-------
Table 2. Major test issues for three-dimensional finite-difference saturated groundwater flow
and solute transport codes
General Features
• mass balances (regular versus irregular grid)
• variable grid (consistency in parameter and stress allocation)
Hvdrogeologic Zoning. Parameterization, and Flow
Characteristics
• aquifer pinchout, aquitard pinchout
• variable thickness layers
• storativity conversion in space and time (confined-
unconfined)
• anisotropy
• unconfmed conditions
• dewatering
• sharp contrast in hydraulic conductivity
Boundary Conditions for Flow
• default no-flow assumption
• areal recharge in top active cells
• induced infiltration from streams (leaky boundary) with
potential for dewatering below the base of the semi-pervious
boundary
• drain boundary
• prescribed fluid flux
• irregular geometry and internal no-flow regions
Transport and fate Processes
• hydrodynamic dispersion (longitudinal and transverse)
• advection-dominated transport
• retardation (linear and Freundlich)
• decay (zero and first-order)
• spatial variability of dispersivity
• effect of presence or absence cross-term for dispersivity
Boundary Conditions for Solute Transport
• default zero solute-flux assumption
• prescribed solute flux
• prescribed concentration on stream boundaries
• irregular geometry and internal zero-transport zones
• concentration-dependent solute flux into streams
Sources and Sinks
• effects of time-varying discharging and recharging wells on
flow
• multi-aquifer screened wells
• solute injection well with prescribed concentration (constant
and time-varying flow rate)
• solute extraction well with ambient concentration
Table 3. Elements of a test report (van der Heijde and Kanzer, 1995).
Program name, title, version, release date, authors, custodian
Reviewer (name, organization)
Detailed program description (functionality)
Computer/software requirements
Terms of availability and support
Overview of testing performed by authors
Overview of additional testing performed
Discussion of specific test results (illustrating strength and
weaknesses)
Elements of a Test Report
Discussion of completeness of testing (functionality matrix)
Representative performance information
Conclusions on test results
Comments on installation, operation and documentation
List of main documentation and review references
Tables providing overview of performed tests and
performance information
Figures illustrating key results
general problem description (including assumptions,
limitations, boundary conditions, parameter distribution,
time-stepping, figures depicting problem situation)
test objectives (features of simulation code, specifically tested
by test problem)
benchmark reference
if feasible, benchmark solution (e.g., analytical solution)
reference to benchmark implementation (hand calculation,
dedicated software, generic mathematical software, etc.)
Reported Test DetaUs
test data set
model setup, discretization, implementation of boundary
condition, representation of special problem features (for
both tested code and benchmark code; electronic input files)
results (table of numerical and benchmark results (if
available) for the dependent variable at selected locations and
times; mass balances; statistical measures and supporting
figures; electronic results files)
sensitivity analysis strategy and results
discussion of results
-------
management
problem
global
hydro logic
system
system
characterization
applicability
assessment
(code
selection)
A
V
AIT
reference system
representative subset of
hydrologic system
e.g., ground-water system
generic model
rnnrpntual
__ _ -j_i
model
i
mathematical
^
<_
^-
model
computer code
examination
verification
(functionality &
performance)
parameterization <
operational model
validation
Fig.1
Fig. \. Model development and code testing concepts (after van der Heijde et al., 1988).
-------
lOWMC Tntlng of FTWOUK v.2.18
tad 1-2: cuinjonof hod tfunjeswlfi JtUm^ twi
ttp ctangt boundary tar t» 1 .52
-------
HE = MAE = PME = 0.62
NME; MER <= MM
RMS = 0.9ff
Avc. DC = 2.8%
20
100
200
300 400 SOO 600 700
Distance Along Model Center Line In FMt
600
900
Fig. 4. Combination plot of X-Y graph of dependent variable and column plot
of residuals (van der Heijde and Kanzer, 1995)
Y-direction
1 6 e is 17 21 26 29 33 37 41 X-direction
Fig. 5. Quasi-three-dimensional graphic representation of computed
heads (van der Heijde and Kanzer, 1995).
11
-------
United States
Environmental Protection
Agency
Solid Waste And
Emergency Response
(5103)
OSWER Directive #9029.00
EPA 500-B-94-003
July 1994
Assessment Framework For
Ground-Water Model Applications
OSWER Information Management
-------
\ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
? WASHINGTON. D.C. 20460
^ ' ? |CCy
OFFICE OF
SOLID WASTE AND EMERGENCY
RESPONSE
MEMORANDUM
SUBJECT: Assessment Framework for Ground-Water Model
Applications - Directive Noy 9029.00
FROM: Elliott P. Laws
Assistant Administrator
TO: OSWER Office Directors
Regional Waste Management Division Directors
- This memorandum establishes the guidance, "Assessment
Framework for Ground-Water Model Applications", as an OSWER
Directive 9029.00.
The Framework provides guidance for planning and evaluating
ground-water flow and advective transport model applications..
The set of criteria helps guide current or future modeling by
assessing modeling activities, thought processes, and
documentation needs. It is intended for EPA technical support
staff and remedial project managers, as well as program managers
and contractors who support EPA's waste management program.
Purpose of the Guidance
The purpose of this guidance is to promote the appropriate
use of ground-water models in EPA's waste management programs.
More specifically, the objectives of the "Assessment Framework
for Ground-Water Model Applications" are to:
Support the use of ground-water models as tools for aiding
decision-making under conditions of uncertainty;
Guide current or future modeling;
Assess modeling activities and thought processes; and
Identify model application documentation needs.
Recycled/Recyclable
Printed with Soy/Canola Ink on paper thai
contains at least 50*4 recycled liber
-------
Background
OSWER management has been addressing the issue of the
increasing use of environmental regulatory models first by
reviewing modeling use and support in the waste management
program. In 1992, OSWER produced the Ground-Water Modeling
Compendium to increase the awareness of ground-water models which
are available and supported by EPA. The Compendium contained the
first version of the. "Assessment Framework for Ground-Water Model
Applications", which was developed by a group of experts from
within and outside of EPA and reviewed by a cross-Agency ad hoc
group of scientists.
The Science Advisory Board (SAB) then reviewed the Framework
and suggested that it be distributed apart from the Compendium.
The SAB's comments are reflected in the attached version.
The 1994 edition of the Ground-Water Modeling Compendium,
containing additional model information and cost guidelines, will
soon be available for distribution from the National Center for
Environmental Publications and Information (NCEPI), 500-B-94-003.
In addition, in 1992, OSWER and the Office of Research and
Development, requested the Deputy Administrator to establish a
temporary Agency Task Force on Environmental Regulatory Modeling
IATFERM) to address modeling issues across the Agency. ATFERM
has produced a report which it presented to the Science Policy
Council in July. The Deputy Administrator has distributed the
"Guidance for External Peer Review of Environmental Regulatory
Modeling" which was developed by ATFERM and is in their report.
In addition, the Science Policy Council agreed to the
establishment of a permanent, Agencywide group to provide a focus
for modeling issues and information.
In summary, OSWER and Regional managers should., encourage the
use of the Framework to ensure that sound and defensible models
are being chosen, that these models are being applied in a
reasonable manner, and that management's decision objectives are
incorporated into the modeling objectives for each application.
Attachment - Assessment Framework for Ground-Water Model
Applications - OSWER Directive 9029.00
(EPA500-G-94-004)
-------
OSWER Directive f9029.00
Assessment Framework Introduction
Assessment Framework for Ground-Water Model Applications
Introduction
The Assessment Framework addresses the use and review of primarily
ground-water flow and advective transport model applications. The criteria in this
Assessment Framework focus upon the activities and thought processes that should
be part of a model application and the subsequent documentation of that activity or
process. The Assessment Framework is a "living document" which may be
expanded as additional information is collected, analyzed, and organized.
The intended primary users of this framework are U.S. Environmental
Protection Agency (EPA) technical support staff and remedial project managers. The
secondary users of the framework are Office of Solid Waste and Emergency
Response (OSWER) and Regional management, EPA contractors, and other
consultants. However, this Assessment Framework is not a substitute for modeling
education and experience. The framework should not be used to promote modeling
by inexperienced people, nor should it be relied upon to supplant experienced
professional judgment or measurement.
The objectives of the Assessment Framework are to:
D Support the use of ground-water models as tools for aiding decision-making
under conditions of uncertainty;
O Guide current or future modeling;
O Assess modeling activities and thought processes; and
O Identify model application documentation needs.
Modeling is often used for prediction and/or evaluation of alternative
remedial schemes. While prediction is often the endpoint of the modeling process,
the value of modeling is not limited to this goal. The modeling process can
enhance one's understanding of the natural system, help in the refinement of a
conceptual model, facilitate hypothesis testing, help check consistency of data sets,
help identify critical processes, and aid in the planning of site characterization.
Modeling is not a linear process but instead is an iterative evolutionary
approach to the refinement of our understanding of a natural system.
The framework contains a series of assessment criteria, grouped into eight
categories:
O Modeling Application Objectives
O Project Management
Page 1
-------
OSWER Directly* •9029.00
Assessment Framework Introduction
O Conceptual Model Development
D Model (code) Selection
D Model Setup and Input Estimation
O Simulation of Scenarios
O Post Simulation Analysis
O Overall Effectiveness.
Figure 1 clarifies how the criteria in the Assessment Framework apply to the
ground-water modeling process. The figure also indicates key points in the
modeling process where prior decisions or assumptions should be reviewed and
adjusted if necessary. These review steps are emphasized because modeling is
naturally an iterative process and is analogous to the scientific method of
formulating a hypothesis and testing it. If the hypothesis, or for example,
conceptual model, is shown at a later stage in the modeling process to be incorrect or
incomplete, another hypothesis needs to be formulated and the modeling process
started again. To optimize or streamline this process, a model should be prepared at
the beginning of a project (for example, during site characterization) so that it can be
refined and used during the entire length of a project (for example, for feasibility
studies, remedial design, remedial action, and eventually for site closure). Because
Figure 1 is generic and may not apply to all sites, professional advice and experience
should be utilized in the application of models.
A user with the appropriate training and experience may apply these criteria
at various stages in the modeling process. For example, when modeling is initially
proposed the user may apply the "Modeling Application Objectives" and "Project
Management" criteria to help determine the applicability of the modeling to the
specific situation. During the early application of the model a user may apply the
"Conceptual Model Development," "Model (code) Selection," and "Model Setup
and Input Estimation" criteria to help guide the modeling process. Upon
completion of the model application the user may apply the "Simulation of
Scenario," "Post Simulation Analysis," and "Overall Effectiveness" criteria to help
assess the results of the modeling and to guide future efforts.
OSWER and Regional managers may encourage the use of the criteria to
ensure that sound and defensible models are being chosen and that these models are
being applied in a reasonable manner. The manager may also encourage the use of
the criteria to ensure that management decision objectives are incorporated into the
modeling objectives.
Page 2
-------
A9*es»ment Framework
Introduction
ESTABLISH MODELING OBJECTIVES
- Establish decision objectives
- Determine the necessity of ground-water modeling
- Determine the level of model complexity
ESTABLISH PROJECT MANAGEMENT PLAN
COLLECT. ORGANIZE. AND INTERPRET AVAILABLE
DATA
I
PREPARE A CONCEPTUAL MODEL
SELECT A SUITABLE MATHEMATICAL CODE
SET UP THE MODEL AND PERFORM INPUT
ESTIMATION
NOTES:
* Includes calibration
sensitivity analysis
** Includes predictive
sensitivity analysis
COMPARE WITH FIELD
DATA
CALIBRATE THE MODEL
ARE CALIBRATION
TARGETS MET?
HISTORY MATCHING
SIMULATE THE SCENARIOS
PERFORM POST-SIMULATION ANALYSIS
EVALUATE THE OVERALL EFFECTIVENESS
ARE THE MODELING
OBJECTIVES MET?
COMPLETE
FIGURE 1 • Diagram of the ground-water modeling process.
Page 3
-------
OSWER Directive «9029.00
Assessment Framework Introduction
These criteria will generally apply to most modeling applications, however,
in certain instances some of the criteria may not be applicable or some of the criteria
may be applicable at different stages in the modeling process. In other instances,
some criteria may have to be modified or expanded.
Documentation of the modeling process is crucial for assuring the
defensibility of the modeling application. Consequently, some of the following
criteria are preceded with an asterisk (*) indicating that the analysis, process, or data
referred to in the question should be documented. Some users may find it useful to
reference additional information when applying the criteria. Therefore, some of the
criteria are followed by endnotes which provide further explanation and reference
additional sources of information. The criteria also contain numerous technical
terms that may require additional explanation. These terms are italicized
throughout the document. A glossary that follows the criteria contains the
definitions for these terms in the context of the criteria.
Page 4
-------
OSWER Directive «9029.00
Assessment Framework Assessment Criteria
Assessment Criteria
Modeling Application Objectives
1 'Management's decision objectives should be dearly specified up front,
considering applicable regulatory and policy issues.
2 The role and need for a modeling study in the pursuit of management's
decision objectives should be established.
3 *Management's decision objectives should be translated into mode ling
objectives up front.
4 Modeling objectives should be based upon existing information about the
physical characteristics of the site (e.g., hydrogeologic system) and the source,
location, and nature of the contamination.
5 *A11 assumptions incorporated within the modeling objectives should be
reviewed with respect to reality and their potential impacts on
management's decision objectives.
6 The purpose of the model application (e.g., data organization,
understanding the system, planning additional field characterization,
prediction, or evaluation of remediation alternatives) should be defined
during the development of the modeling objectives.
7 The purpose of the model application should be reviewed during the course
of the project and, if necessary, modified.
8 Potential solutions to be evaluated (e.g., containment and/or remediation
solutions) should be identified prior to the initiation of the modeling.
9 The level of model complexity and, in turn, the type of model required
(e.g., numerical model, analytical model, or graphical techniques) should be
determined during the definition of the modeling objectives.
See Endnote 1.
10 This level of model complexity should be reviewed as a better
understanding of the site/problem/data is developed.
11 "Management, in consultation with a professional ground-water scientist,
should specify the time period (e.g., 1 year, 10 years, or hundreds of years) for
which model predictions are intended.
Page 5
-------
OSWER Dfredlve f9029.00
Assessment Framework Assessment Criteria
12 'Calibration targets (e.g., multiple criteria such as heads and ground-water
discharge) for the model application should be specified up front.
13 If, during the modeling process, it is determined that the original calibration
targets cannot be met, the modeling objectives should be reviewed.
14 *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 calibration targets of
the model) and the consequent incremental improvement in supporting
management's decision objectives.
15 Management's decision objectives should be reaffirmed throughout the
modeling process.
16 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.
Page 6
-------
OSWER Directive 419029.00
Assessment Framework Assessment Criteria
Project Management
17 *A financial budget appropriate to the modeling objectives, level of analysis,
available data, and available resources should be established at the inception
of the project.
18 The individuals who are actually managing or performing the modeling
should participate in the development of the financial budget.
19 The project organization should be designed to facilitate the interative
nature of the modeling process.
20 *The individuals who are actually performing the modeling, managing the
modeling effort, or 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 mathematics, physics, chemistry, soil science,
fluid -mechanics, geology, hydrogeology, and modeling
• Work experience in modeling physical systems, preferably with
the type of model being used on the project
• Field experience characterizing site hydrogeology
• Modeling project management experience.
See End note 2.
21 These individuals should be organized as a cohesive modeling team with
well defined roles, responsibilities, and level of participation.
22 The organization of the team should be appropriate for the application.
23 A documentation procedure should be established up front to assure that
an independent reviewer can duplicate the modeling results or perform a
postapplication assessment using the documentation.
24 *The documentation should include a discussion of the following:
General setting of the site
Physical systems of interest
Management's decision objectives
Role of modeling study
Potential solutions to be evaluated
Modeling objectives and timeframe for model predictions
Level of model complexity
Calibration targets
Quality assurance and peer review process
Composition of the modeling team
Page 7
-------
OSWER Directive *9029.00
Assessment Framework Assessment Criteria
• Data sources, quality, and completeness
• Conceptual model
- Hydrostratigraphy
- Ground-water flow system
— Hydrologic boundaries
- Hydraulic properties
- Water sources and sinks
— Contaminant identity, source, loading, and areal extent
- Contaminant transport and transformation processes
- Background chemical quality
— Boundary conditions
• Selection of the computer code
— Description of the code and documentation
- Reliability
— Usability
- Transportability
- Performance
- Access to source code
— Limitation
- Related Applications
• Ground-water model construction
- Code modifications
— Geologic representation
— Flow representation
- Data averaging procedures
— Input estimation procedures
— Model grid
- Hydraulic parameters
- Chemical parameters
— Boundary conditions
- Water budget
— Simplifying assumptions
• Uncertainty analysis
• Calibration and calibration sensitivity analysis
• Predictive simulations
— Scenarios
— Implementation of the scenarios
- Discussion of the results of each run
— Predictive sensitivity analysis
— Postprocessing
• Modeling study scope, conceptual model, and model code
assumptions with respect to reality and their impacts on the
modeling results
• Results related to management's information needs as formulated
in the decision objectives
Page 8
-------
OSWER Directive #9029.00
Assessment Framework Assessment Criteria
• Executive summary (in terms of the decision objectives)
• References
• Input and output files.
See Endnote 3.
25 An independent quality assurance (QA) process not involving staff
assigned to any aspect of the project should be established at the beginning
of the project.
See Endnote 4.
26 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
Postsimulation analysis.
Page 9
-------
OSWER Directive 19029.00
Assessment Framework Assessment Criteria
Conceptual Model Development
27 An initial conceptual model of both the local and regional hydrogeological
system should be developed prior to any computer modeling.
28 The conceptual model should be based upon a quantification of field data
as well as other qualitative data that includes information on the nature
and variability of the:
• Aquifer system (distribution and configuration of aquifer and
confining formations)
• Hydrologic boundaries
• Hydraulic and chemical properties of formations
• Piezometric head and hydraulic gradient (i.e., magnitude and
direction of flow within each model layer)
• Fluid properties
• Contaminant sources and properties
• Fluid sources and sinks.
See Endnote 5.
29 The quantity, quality, and completeness of the data should be analyzed
with regard to their impact on the overall success of the model application.
30 *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.
31 If there are data gaps (e.g., missing water level or hydraulic conductivity
information), any additional field work and other attempts to fill in these
gaps should be documented.
32 The data sources should be documented.
33 *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.
34 The manner in which existing and future engineering (e.g., wells or slurry
walls) must be represented in a numerical or analytical model should be
explicitly incorporated into the conceptual model.
35 A mass balance of the contaminant should be developed as part of the
conceptual model.
Page 10
-------
OSWER Directive §9029.00
Assessment Framework Assessment Criteria
36 The conceptual model should include a clear statement of the location,
type, and state of boundary conditions; justification of their formulation;
and source(s) of information used to develop the boundary conditions.
37 *A11 assumptions incorporated within the conceptual model should be
reviewed with respect to reality and their potential impacts on the modeling
objectives.
Page 11
-------
OSWER Directive #9029.00
•Assessment Framework Assessment Criteria
Model (code^ Selection
38 *The selected model (code), as distinguished from the model application,
should be described with regard to its flow, contaminant transport and
transformation processes, mathematics, hydrogeologic system
representation, boundary conditions, and input parameters.
39 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.
See Endnote 6.
40 The usability of the model (code) should be assessed including the
availability of:
The model binary code
The model source code
Pre- and postprocessors
Existing data resources
Standardized data formats
Complete user instruction manuals
Sample problems
Necessary hardware
Transportability across platforms
User support
Key assumptions.
41 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.
Page 12
-------
OSWER Directive *9029.00
Assessment Framework Assessment Criteria
42 If the ground-water model results will be used to form expert opinions, all
parties should have access to the foundation of those opinions, including
the source code and an executable image of the version used, if present. In
addition, the ground-water model code should meet the following criteria:
• Publication and peer review of the model's conceptual and
mathematical framework, including the model's underlying
assumptions
• Full model documentation
• Publication and peer review of model code testing.
See Endnote 7.
43 The assumptions in the model (code) should be analyzed with regard to
their impact upon the modeling objectives and site-specific conditions.
44 *Any and all discrepancies between the modeling requirements (i.e., as
indicated by management's decision objectives, conceptual model, and
available data) and the capabilities of the selected model should be identified
and justified. For example, the implications of the selected code supporting
1-, 2- or 3-dimensional modeling; providing steady versus unsteady state
modeling; or requiring simplifications of the conceptual model should be
discussed.
45 *If the modeling objectives are modified due to such discrepancies, those
modifications should be documented.
46 *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 Criterion #38)
• Usability evaluation (See Criterion #39)
• Performance testing.
See Endnote 8.
Page 13
-------
OSWER CHredlve 19029.00
Assessment Framework Assessment Criteria
Model Setup and Input Estimation
47 *When the dimensional aspects of the geology at the site are simplified in
the model representation (e.g., representing a multilayer aquifer with a
single layer), the impact of this simplification on the modeling results
should be evaluated.
48 *When data averaging procedures are used to represent the site conditions
in the model, the impact of the averaging upon the modeling results should
be evaluated.
See Endnote 9.
49 *When flow representations in the model are assumptions or
simplifications of site conditions (e.g., only horizontal flow is considered,
thus ignoring the impact of vertical flow components), the impact of these
assumptions and simplifications on the modeling results should be
evaluated.
50 For numerical models, generally acceptable rules of grid design and time
step selection should be applied to meet the modeling objectives.
See Endnotes 10 and 19.
51 *When a numerical model is used, the mapping of the location of the
boundary conditions and other geometric details (e.g., wells, slurry walls,
and contaminant sources) on the grid should be evaluated. For example:
• 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 Endnotes 11 and 12.
52 *If arbitrary or artificial boundaries are used, justification for their use
should be given and evidence presented to demonstrate that their use does
not adversely impact the model results within the area of interest.
53 *When an analytical model \s> used, the following boundary conditions
should be evaluated:
• Where infinite boundaries are used, the engineering feature being
modeled should not impact actual physical boundaries at the site
during the timeframe of interest.
Page 14
-------
OSWER Directive «9029.00
Assessment Framework Assessment Criteria
• Where image wells are used, general rules of imaging should be
followed.
See Endnote 13.
54 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.
55 *A11 model inputs should be defined as to whether they are measurements,
interpretations, 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.
56 "The input estimation process whereby data are converted into model
inputs (e.g., spatial and temporal interpolation, extrapolation or Kriging, or
averaging) should be described. This description should include a map or
table containing the spatial location and the associated values of data used to
perform the interpolation.
See Endnote 12.
57 The uncertainty associated with the input estimation process should be
specified, explained, and documented.
58 The model should be calibrated.
59 *If the model is not calibrated, the rationale for not calibrating the model
should be explained.
See Endnote 14.
60 The criteria (i.e., calibration targets) used in the termination of the
calibration process should be justified with regard to the modeling
objectives.
See Endnote 12.
Page 15
-------
OSWER Directive #9029.00
Assessment Framework Assessment Criteria
61 The calibration should be performed in a generally acceptable manner.
Specifically:
• A calibration sensitivity analysis should be performed to
determine the key parameters and boundary conditions to be
investigated during calibration.
• The calibration should include an evaluation of spatial residuals
between simulated and measured values. If transient data are
available, an evaluation of spatial residuals at selected time steps
should also be performed.
• The calibration should be performed in the context of the physical
features (e.g., residuals should be analyzed with respect to the
pattern of ground-water contours including mounds, depressions,
or indications of surface water discharge or recharge).
See Endnote 15.
62 A water budget for the natural aquifer system based upon measurements
and/or estimates should be developed and used to create a mass balance for
the model.
63 *If a water budget is not developed, the reasoning for not developing a
water budget should be explained.
64 *If actual measurements of components of the water budget are available,
they should be used to calibrate the model.
65 *A11 changes in initial model parameter values due to calibration should be
justified as to their reasonableness.
66 *Any discrepancies between the calibrated model parameters and the
parameter ranges estimated in the conceptual model should be justified.
67 *If the conceptual model is modified as a result of the model calibration, all
changes in the conceptual model should be justified.
Page 16
-------
OSWER Directive #9029.00
Assessment Framework Assessment Criteria
Simulation of Scenarios
68 *For each modeling scenario, the model inputs and the location of features
in the model grid should be justified. For example:
• fat finite element models, 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 extent of the
capture zone.
Page 17
-------
OSWER Directive §9029.00
Assessment Framework Assessment Criteria
Postsimulation Analysis
69 The success of the model application in simulating the site scenarios
should be assessed.
70 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.
71 The sensitivity of the model results to uncertainties in site-specific
parameters (predictive sensitivity analysis) and the level of error in the
model calibration (calibration sensitivity analysis) 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.
See Endnote 16.
72 The modeling results should be consistent with available data.
73 The postprocessing, including the use of interpolation and smoothing,
should be analyzed and documented to ensure that it accurately represents
the modeling results.
74 The postprocessing results should be analyzed to ensure that they support
the modeling objectives.
75 The final presentation should effectively and accurately communicate the
modeling results.
76 When feasible, a postaudit of the model should be carried out or planned
for the future.
See Endnote 17.
Page 18
-------
OSWER Olmcllve 19029.00
Assessment Framework Assessment Criteria
Overall Effectiveness
77 *Any difficulties encountered in the model application should be
documented.
78 The model application should provide the information being sought by
management for decision making.
See Endnote 18.
79 The model application results should be acceptable to all relevant parties.
80 The model application should support a timely and effective regulatory
decision process.
81 Those aspects of the modeling effort that in hindsight might have been
done differently should be documented.
Page 19
-------
OSWER Directive §9029.00
Assessment Framework Endnotes
Endnotes
A complete list of references follows the Glossary of Technical Terms.
1. Some of the factors which might influence the level of model complexity and,
in turn, the selection of a particular type (e.g., numerical model, analytical
model, or graphical techniques) of model include:
• The importance of the decisions which will be influenced by the
model results
• The sensitivity of these decisions to the range of possible or likely
outcomes of the modeling
• The availability of time and resources for the modeling application
• The complexity of site characteristics.
See Anderson and Woessner 1992a, and other standard modeling references.
2. See USOTA 1982 and van der Heijde and Park 1986.
3. For more information on documentation see ASTM and CADHS.
4. QA should assure that:
• The project is staffed with qualified people
• There is appropriate documentation of the modeling process
• Peer review of the modeling process and project deliverables is
performed.
It should be noted that there is no way to assure that a specific model of a
physical system can ever be completely verified. Thus, the QA process helps to
build confidence in a model application, but following the QA process does not
guarantee accurate predictions. See Konikow and Bredehoeft 1992.
5. For more information on data requirements for conceptual model
development, see ASTM; CADHS, page 4; NCR 1990, pages 221-230; and
Anderson and Woessner 1992a.
6. Sources of information on model codes include USEPA 1992; Bond and Hwang
1989; the International Ground Water Modeling Center (IGWMC) data base; the
Integrated Model Evaluation System (IMES); and numerous ground-water
modeling texts.
7. The Special Master in the case of United States of America, et al.vs. Hooker
Chemicals & Plastics Corporation, et al.(Love Canal) ruled on November 30,
1989 that if models are to be relied upon to form expert opinions in litigation,
all relevant parties will be permitted access to the foundation of those opinions,
including the source code. At the same time, the Special Master granted a
Page 20
-------
OSWER Directive #9029.00
Assessment Framework Endnotes
Protective Order so that the code could not be used by the opposing side for any
purpose other than the trial. (See United States of America, et al. vs. Hooker
Chemicals & Plastics Corporation, et al. (Love Canal) on November 30,1989.)
For further information on the use of ground-water models in litigation, see
Kezsbom and Goldman 1991. For more information on model code
documentation, see van der Heijde and Elnawawy 1992.
8. For more information on the testing of model codes, see van der Heijde and
Elnawawy 1992, Section 3.
9. Contaminant transport is affected by "nonaverage" conditions, with
contaminant plumes following preferential flow paths. See Anderson and
Woessner 1992a, page 326; and Fetter 1988, page 395.
10. For example, the grid should be fine enough in the area of interest to produce
accurate results and nodes should coincide with physical features, remediation
wells, and contamination sources as much as possible. Grid orientation, grid
expansion factors, and aspect ratios should also meet general modeling
standards. For more information, refer to Anderson and Woessner 1992a; and
van der Heijde, El-Kadi, and Williams 1988, pages 45-48.
11. For information on properly locating and representing boundary conditions,
see Franke, Reilly, and Bennett 1984; and Anderson and Woessner 1992a.
12. For information on model setup, input estimation, and criteria for the
termination of the calibration process, see Anderson and Woessner 1992a.
13. For more information on imaging, see Freeze and Cherry 1979, page 330.
14. The model should be calibrated, especially if it is used for predictive purposes.
For interpretive or generic models, calibration is encouraged but not required
(Anderson and Woessner 1992a). Analytical models should also be calibrated,
if possible (e.g., in the case of flow, to transient data; or in the case of transport,
to plume data).
15. For more information on the calibration of ground-water models, see
Anderson and Woessner 1992a; van der Heijde, El-Kadi, and Williams 1988;
ASTM, Subpart 6.5; and CADHS, Section 3.3.2.4.
16. When a physical system is subject to new stresses (as during the application of a
remedial strategy), errors in the conceptual model which had little impact
during the calibration phase may become dominant sources of error for the
prediction phase. Because a specific model of a physical system can never be
completely "verified," it becomes important to identify uncertainties in model
input parameters and conceptual assumptions and to explore the implications
Page 21
-------
OSWER Dtocthw €8029.00
Assessment Framework Endnotes
of these uncertainties on model predictions. For a more complete discussion,
see Konikow and Bredehoeft 1992, pages 75-83.
17. For more information on post audits, see Anderson and Woessner 1992b.
18. For more information on decision making under conditions of uncertainty, see
Freeze, Massmann, Smith, Sperling, and James 1990, pages 738-766.
19. For information on how time-step size can affect the numerical accuracy of a
model, see Huyakorn and Finder 1983, pages 206 and 392; van der Heijde, El-
Kadi, and Williams 1988, pages 45-48; and Anderson and Woessner. 1992a.
Page 22
-------
OSWER Directive §9029.00
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 follows 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 a solution of the
governing partial differential equation at any location and/or time instead
of approximate solutions at discreet locations and moments in time. (15, 22)
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 nonphysical boundaries,
such as ground-water divides and areas of aquifer underflow. The effect of
nonphysical boundaries on the modeling results must be tested. (4)
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 an acceptable range of error. (4)
Calibration sensitivity analysis - a procedure to establish the effect of
uncertainty on the calibrated model. The calibrated model is influenced by
uncertainty owing to the inability to define the exact spatial (and temporal)
distribution of parameter values in the problem domain. There is also
uncertainty over the definition of boundary conditions and stresses. (4)
Calibration target - a preestablished range of allowable error between heads
and fluxes and field measured values. (4)
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. (22)
Page 23
-------
OSWER Directive 19029.00
Assessment Framework Glossary of Technical Terms
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. (4,15)
Confining bed - 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. A confining bed 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. (1,19,20,23)
Constitutive coefficients and parameters - type of model input that is not
directly observable but must be inferred from observations of other model
variables; for example, the distribution of transmissivity, specific storage,
porosity, recharge, and evapotraspiration. These are difficult to estimate
because they vary spatially and may vary temporally as well. (21)
Containment - action(s) undertaken, such as constructing slurry trenches,
installing diversionary booms, earth moving, plugging damaged tank cars,
and using chemical retardants. These actions focus on controlling the
source of a discharge or release and minimizing the spread of the hazardous
substance or its effects. (28)
Contaminant source, loading, and areal 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.
(3)
Contaminant transformation - chemical changes, reactions, and biological
transformations that change the chemical properties of the contaminant. (3)
Contaminant transport - flow and dispersion of contaminants dissolved in
ground-water in the subsurface environment. (21)
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
Page 24
-------
OSWER CHredlve 19029.00
Assessment Framework Glossary off Technical Terms
known as fly-off, total evaporation, and water loss. Losses from
evapotranspiration can occur at the water table. (1,3)
Field characterization - a review of historical, on- and offsite, as well as
surface and subsurface 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. (3)
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 and transport
equations. 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. (3,11, 15, 16)
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 and transport equations. 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. (3,11,15,16)
Fluid potential - mechanical energy per unit weight of fluid at any given
point in space and time with regard to an arbitrary state and datum. (11, 24)
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). (21)
Ground-water divides - ridges in the water table or potentiometric surface
from which ground-water moves away in both directions (14); a hydraulic
boundary at the crest or valley bottom of a ground-water flow system across
which there is no flow. (11)
Page 25
-------
OSWEH Orally* M029.00
Assessment Framework Glossary of Technical Terms
Ground-water flow system - a rather vague designation pertaining to a
ground-water reservoir that is more or less separate from neighboring
ground-water reservoirs. Ground-water reservoirs can be separated from
one another by geologic or hydrologic boundaries. In some ground-water
modeling studies, artificial or arbitrary boundaries may be applied. Water
moves or "flows" through the ground-water reservoir through openings in
sediment and rock. (10)
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
that will move in unit time under a unit hydraulic gradient through a unit
area measured at right angles to the direction of flow. (22)
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). (3,33)
Hydrologic boundaries - boundary conditions which relate to the flow of
water in an aquifer system. (3)
Hydrostratigraphy - a sequence of geologic units delimited on the basis of
hydraulic properties. (10)
Kriging - an interpolation procedure for estimating regional distributions of
ground-water model inputs from scattered observations. (21)
Management's decision objectives - the information needs required to
identify courses of action necessary for reaching environmental and
regulatory goals.
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. (31, 33)
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, fin it e
difference, and finite element models are more widely used because they are
easier to develop and manipulate. (4,10)
Page 26
-------
OSWER Directive #9029.00
Assessment Framework Glossary of Technical Terms
Modeling objectives - the information that the modeling application is
expected to provide so that management can evaluate potential courses of
action. (4)
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 at discreet locations and moments in time for
the governing differential equations. (3,16, 21)
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. (3, 21)
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. (30)
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.
(3,10) .
Postaudit - 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. (21)
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. (10)
Predictive sensitivity analysis - a procedure to quantify the effect of
uncertainty in parameter values on the prediction. Ranges and estimated
future stresses are simulated to examine the impact on the model's
prediction. (4)
Reliability - the probability that a model will satisfactorily perform its
intended function under given circumstances. It is the amount of credence
Page 27
-------
OSWER Diredlve 19029.00
Assessment Framework Glossary of Technical Terms
placed in a result. Measures of reliability include peer review of model
theory and code; evaluation of the model results against laboratory tests,
analytic solutions, or other well accepted modes; field testing; and user
acceptability. (23,30)
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. (25)
Residuals - the differences between field measurements at calibration points
and simulated values. (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. (3,21)
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). (23)
Surface water bodies - all bodies of water on the surface of the earth. (23)
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, 22)
Uncertainty analysis - process to identify uncertainties in model input
parameters and conceptual assumptions, and the implications of these on
the uncertainty in model predictions, including potential impacts on the
decisions which will be made based on these predictions. (26)
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. (4, 7,
30)
Water budget - the sources and outflow of water to the system, which may
include ground-water recharge from precipitation, overland flow, recharge
from and discharge to surface water bodies, springflow, evaportranspiration,
or pumping. (4)
Page 28
-------
OSWER Directive t9029.00
Assessment Framework Sources
Sources
Definitions and endnotes are directly drawn from and based upon the following
sources:
1. Allaby, Ailsa and Michael. 1990. The Concise Oxford Dictionary of Earth
Sciences. Oxford: Oxford University Press.
2. American Society of Testing and Materials (ASTM). (No date.) "Guide for
Application of a Ground-Water Flow Model to a Site-Specific Problem." Draft
ASTM Standard Section D-18.21.10.
3. Anderson, Mary P. 1992. Based wholly or in part on written comments
provided on the initial draft version of the glossary.
4. Anderson, Mary P., and William W. Woessner. 1992a. Applied Groundwater
Modeling - Simulation of Flow and Advective Transport. New York:
Academic Press, Inc.
5. Anderson, Mary P., and William W. Woessner. 1992b. "The Role of the
Postaudit in Model Validation." Advances In Water Resources.
6. California Department of Health Services (CADHS). (No date.) "Standards for
Mathematical Modeling of Ground Water Flow and Contaminant Transport at
Hazardous Waste Sites." Chapter 4, Volume 2 of Scientific and Technical
Standards for Hazardous Waste Sites-Draft.
7. Belgin, Milovan S. 1987. Testing. Verification, and Validation of Two-
Dimensional Solute Transport Models. Golden, Colorado: International
Ground Water Modeling Center, Colorado School of Mines.
8. Bond, F., and S. Hwang. 1989. "Selection Criteria for Mathematical Models
Used in Exposure Assessments: Ground Water Models." Office of Health and
Environmental Assessment (OHEAI Manual. EPA/600/2-89/028. U.S.
Environmental Protection Agency.
9. Franke, O.L., T.E. Reilly, and G.D. Bennett. 1984. Definition of Boundary and
Initial Conditions in the Analysis of Saturated Ground-Water Flow Systems -
An Introduction. Open-File Report 84-458. Reston, Virginia: U.S. Geological
Survey.
10. Fetter, C.W. 1988. Applied Hydrogeology - 2nd Edition. Columbus: Merrill
Publishing Company.
Page 29
-------
OSWER Directive 19029.00
Assessment Framework Sources
11. Freeze, R Allen, and John A. Cherry. 1979. Groundwater. Englewood, New
Jersey: Prentice-Hall.
12. Freeze, Allen, Joel Massman, Leslie Smith, Tony Sperling, and Bruce James.
1990. "Hydrogeological Decision Analysis." Ground Water 28(5^.
13. Golden Software, Inc. September 1990. Surfer Version 4 Reference Manual.
Golden, Colorado.
14. Huyakorn, P.S., and G.F. Finder. 1983. Computation Methods in Subsurface
Flow. New York: Academic Press.
15. Istok, Jonathan. 1989. Groundwater Modeling by the Finite Element Method.
Washington, D.C.: American Geophysical Union.
16 Javandel, Iraj, Christine Doughty, and Chin-Fu Tsang. 1984. Ground Water
Transport: Handbook of Mathematical Models. Washington, D.C.: American
Geophysical Union.
17. Kezsbom, A., and A.V. Goldman. 1991. "The boundaries of groundwater
modeling under the law: Standards for excluding speculative expert
testimony." Tort and Insurance Law Tournal. Vol. XXVII, No.l.
18. Konikow, Leonard F., and John D. Bredehoeft. 1992. "Ground-water Models
Cannot Be Validated." Advances in Water Resources 15.
19. Kruseman, G.P., and N.A. de Ridder. 1990. Analysis and Evaluation of
Pumping Test Data. The Netherlands: International Institute for Land
Reclamation and Improvement.
20. Lohman, S.W. 1972. Definitions of Selected Ground-Water Terms - Revisions
and Conceptual Refinements. Geological Survey Water-Supply Paper 1988.
Washington, D.C.: U.S. Government Printing Office.
21. National Research Council (NRC) Committee on Ground-Water Modeling
Assessment. 1990. Ground Water Models. Scientific and Regulatory
Applications. Washington, D.C.: National Academy Press.
22. The Ohio State University, Department of Geological Sciences. 1991. Capzone
Users Manual. Columbus, Ohio.
23. Parker, Sybil P. 1989. McGraw-Hill Dictionary of Scientific and Technical
Terms. Fourth Edition. New York: McGraw-Hill Book Company.
Page 30
-------
OSWER Dtrective «9029.00
Assessment Framework Sources
24. Subsurface-Water Glossary Working Group, Ground Water Subcommittee,
Interagency Advisory Committee on Water Data. 1989. Subsurface-Water and
Solute Transport Federal Glossary of Selected Terms.
25. U.S. Environmental Protection Agency (USEPA) Office of Public Affairs. 1988.
Glossary of Environmental Terms and Acronym List. Washington, D.C.
26. U.S. Environmental Protection Agency (USEPA) Science Advisory Board. 1993.
Review of the Assessment Framework for Ground-water Model Applications.
Washington, D.C.
27. U.S. Environmental Protection Agency (USEPA). 1992. Ground-Water
Modeling Compendium - Draft. Washington, D.C.
28. U.S. Environmental Protection Agency (USEPA). 1981. Technical Assistance
Team (TAT) Contract Users Manual. Washington, D.C.
29. U.S. Office of Technology Assessment (USOTA). 1982. Use of Models for
Water Resources of the United States. Washington, D.C.: U.S. Government
Printing Office.
30. van der Heijde, Paul, K.M., and O.A. Elnawawy. 1992. Quality Assurance and
Quality Control in the Development and Application of Groundwater Models.
EPA/600/R-93/011. U.S. Environmental Protection Agency Office of Research
and Development.
31. van der Heijde, Paul, K.M., Aly I. El-Kadi, and Stan A. Williams. 1988.
Groundwater Modeling: An Overview and Status Report. EPA/600/2-89/028.
U.S. Environmental Protection Agency.
32. van der Heijde, Paul K.M., and Richard A. Park. 1986. Ground-water Modeling
Policy Study Group Report. Report for the U.S. EPA Office of Research and
Development. Golden, Colorado: International Ground Water Modeling
Center, Colorado School of Mines.
33. Wang, Herbert F., and Mary P. Anderson. 1982. Introduction to Groundwater
Modeling. Finite Difference and Element Methods. San Francisco, California:
W.H. Freeman and Company.
Page 31
-------
OSWER Directive #9029.00
EPA Publications Related
Assessment Framework To Ground-Water Modeling
EPA Publications Related to Ground-Water Modeling
CONTAMINANT FATE AND TRANSPORT
USEPA. Grove, D.B., and J. Rubin. 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, pages 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/540/2-90/011.
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. (No date.) Dense Nonaqueous Phase Liquids - A Workshop Summary.
GROUND-WATER ISSUE PAPERS
USEPA. Puls, R.W., and M.J. Barcelona. 1989. Ground Water Sampling for Metals
Analysis. EPA/540/4^89/001.
USEPA. Lewis, T.Y., R.L. Crockett, R. L. Siegrist, and K. Zarrab. 1991. Soil Sampling
and Analysis for Volatile Organic Compounds. EPA/450/4-91/001.
Page 32
-------
OSWER Directive #9029.00
EPA Publications Related
Assessment Framework to Ground-Water Modeling
USEPA. Breckenridge, R.P., J.R. Williams, and J.F. Keck. 1991. Characterizing Soils
for Hazardous Waste Site Assessments. EPA/540/4-91/003.
USEPA. Schmelling, S.G., and R.R. Ross. 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.
USEPA. Piwoni, M.D., and J.W. Keeley. 1990. Basic Concepts of Contaminant
Sorption at Hazardous Waste Sites. EPA/540/4-90/053.
USEPA. Huling, S.G., and J.W. Weaver. 1991. Dense Nonaqueous Phase Liquids.
EPA/540/4-91/002.
USEPA. Keely, J.F. 1989. Performance Evaluations of Pump-and Treat
Remediations. EPA/540/4-89/005.
USEPA. Sims, J.L., J.M. Suflita, and H.H. Russell. 1991. Reductive Dehalogenation
of Organic Contaminants in Soils and Ground Water. EPA/540/4-90/054.
USEPA. Russell, H.H., J.E. Matthews, and G.W. Sewell. 1992. TCE Removal from
Contaminated Soil and Ground Water. EPA/540/S-92/002.
USEPA. Palmer, C.D., and W. Fish. 1992. Chemical Enhancements to Pump-and-
Treat Remediation. EPA/540/S-92/001.
USEPA. Sims, J.L., J.M. Suflita, and H.H. Russell. 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., R.M. Powell, D.A. Clark, and CJ. Paul. 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., T.W. Bennett, G. Hackett, J.E. Denne, and R.J. Petty. 1990.
Handbook of Suggested Practices for the Design and Installation of Ground
Water Monitoring Wells. EPA/600/4-89/034.
Page 33
-------
OSWER Directive «9029.00
EPA Publications Related
Assessment Framework To Ground-Water Modeling
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.
USEPA. 1990. A New Approach and Methodologies for Characterizing the
Hydrogeologic Properties of Aquifers. EPA/600/2-90/002.
USEPA. Molz, F.J., G. Oktay, and J.G. Melville. 1990. Measurement of Hydraulic
Conductivity Distributions: A Manual of Practice. Auburn University.
USEPA. 1990. Ground Water, Volume II: Methodology. EPA/625/6-90/016b.
MODELING
USEPA. Pettyjohn, W.A., D.C. Kent, T.A. Prickett, and H.E. LeGrand. (No date.)
Methods for the Prediction of Leachate Plume Migration and Mixing. Office of
Research Laboratory, Cincinnati.
USEPA. Bond, F., and S. Hwang. 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., M. Geljin, and R. Ross. 1992. Fundamentals of Ground Water
Modeling for Decision Makers. EPA/540/S-92/005.
•
USEPA. van der Heijde, Paul K.M., Aly I. El-Kadi, and Stan A. Williams. 1988.
Groundwater Modeling: An Overview and Status Report. EPA/600/2-89/028.
USEPA. van der Heijde, Paul K.M., and O.A. Elnawawy. 1992. Quality Assurance
and Quality Control in the Development and Application of Groundwater
Models. EPA/600/R-93/011. Office of Research and Development.
Page 34
-------
OSWER Directive 49029.00
EPA Publications Related
Assessment Framework to Ground-Water Modeling
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 Aquifer.
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.
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.
Page 35
-------
OSWER Directive *9029.00
EPA Publications Related
Assessment Framework To Ground-Water Modeling
USEPA. 1988. OSWER Directive No. 9950.3. Operation and Maintenance
Inspection Guide (RCRA) Ground Water Monitoring Systems.
RISK ASSESSMENT
USEPA. Bond, F., and S. Hwang. 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/540/1-88/001.
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., W.D. Glauz, and J.D. Flora. 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.
USEPA. 1991. OSWER Directive No. 9360.4-06. Compendium of ERT Ground
Water Sampling Procedures.
REMEDIAL INVESTIGATION/FEASIBILITY STUDY
Page 36
-------
OSWER Diredtve #9029.00
EPA Publications Related
Assessment Framework to Ground-Water Modeling
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.
Page 37
-------
OSWER Directive 19029.00
EPA Publications Related
Assessment Framework To Ground-Water Modeling
USEPA. 1990. Guidance on Expediting Remedial Design and Remedial Action.
EPA/540/G-90/006.
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 (No date.) General Methods for Remedial Operations
Performance Evaluations.
USEPA. 1990. OSWER Directive No. 9355.4-01FS. Guide on Remedial Actions at
Superfund 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.
CLEANUP 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.
Page 38
-------
OSWER Directive t9029.00
EPA Publications Related
Assessment Framework to Ground-Water Modeling
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.
USEPA. 1989. Evaluation of Ground Water Extraction Remedies, Volume ffl:
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., D.C. Skipp, and D. Giffin. 1990. Basics of Pump and Treat
Ground Water Remediation Technology. EPA/600/8-90/003.
USEPA. 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., and D.H. Killer. 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, Ohio.
Page 39
-------
OSWER Directive #9029.00
EPA Publications Related
Assessment Framework To Ground-Water Modeling
USEPA. Mullen, H., and S.I. Taub. 1976. Tracing Leachate from Landfills, A
Conceptual Approach, Proceedings of the National Conference on Disposal of
Residues on Land. Environmental Quality Symposium, Inc. pages 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. NTIS No. PB91-921332/CCE. Applicability of Land Disposal
Restrictions to RCRA and CERCLA Ground Water Treatment Reinjection,
Superfund Management.
USEPA. 1991. Conducting Remedial Investigation/Feasibility Studies for CERCLA
. Municipal Landfill Sites. Office of Emergency Remedial Response,
Washington, D.C.
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, D.C.
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, D.C.
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.
Page 40
-------
OSWER Directive «9029.00
EPA Publications Related
Assessment Framework to Ground-Water Modeling
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 41
-------
CSMoS
-------
6. Center for Subsurface
Modeling Support (CSMoS)
-------
Center for Subsurface Modeling Support
(CSMoS)
Directors
David S. Burden, Ph.D.
Joe R. Williams
Center for Subsurface Modeling Support
R.S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
Ada, Oklahoma
-------
RSKERL-ADA
Technology Support Center
->
XX
In-House
Researchers
CORE
TEAM
Extramural
Experts
Ground-Water
Remediation
Technologies
Analysis Center
(GWRTAC)
Center for Subsurface
Modeling Support
(CSMoS)
Site Specific
Technical
Support
Technology Transfer
Activities
Subsurface
Remediation
Information
Center
6-2
-------
RSKERL Technology Support
Center provides assistance to
T
U)
Superfund
EPA Headquarters
EPA Regional Personnel
State Personnel
Underground Storage Tanks
Pesticides
Underground Injection Control Programs
Wellhead Protection Programs
-------
Bioremediation
Modeling
Leaching
Soil Gas Extraction
Chemistry
Metals
Pump & Treat
Treatability
Hydrology
Principal Areas of Technical Assistance Requests
-------
tn
The growth in the use of computer
models in the U.S. stems from a
series of stringent and
comprehensive environmental
statutes developed since the early
1970s.
-------
These include:
a*
Comprehensive Environmental
Response, Compensation, and
Liability Act (CERCLA)
Resource Conservation and Recovery
Act (RCRA)
Safe Drinking Water Act (SDWA)
National Environmental Policy Act
(NEPA)
-------
Estimated Numbers of Contamination Problems
that Need to be Addressed Under Various Statutes
NPL Sites 951
RCRA Hazardous Waste Facilities
Operating Landfills 393
Closing Landfills 1,095
RCRA Non-Hazardous Waste Facilities 70,419
to 261,930
Mining Waste Sites 22,339
Underground Storage Tanks 10,820
Underground Injection Wells
Class I, II, III, IV, V 200,204
-------
Modeling may be used in the
Superfund program to
CO
Guide placement of monitoring wells
Predict concentrations in ground water for
an assessment of the risks at the site
Assess the feasibility and efficacy of
remedial alternatives
Predict the concentration for an assessment
of the risk after implementation of
the preferred remedial action
Apportion liability among responsible parties
-------
Center for Subsurface
Modeling Support (CSMoS)
vO
Provides ground-water and vadose
zone modeling software and services
to public agencies and private
companies throughout the nation.
-------
CSMoS Functions
>-•
o
Model Distribution
Model Technical Support
Model Applications Review
Model Development / Testing
Model Education / Training
Database Development / Support
Geographic Information Systems (GIS) Support
-------
Support for EPA 9s
Center for Subsurface Modeling Support
U.S.EPA/RSKERL
Scientific & Support Staff
Contractor Support
Computer Data Systems,
Incorporated (CDSI)
Mechanism: lAGwithGSA
. Client Representative: Joe Williams /
Dynamac Corporation
Mechanism: Off-Site Contract
Project Officer: David Burden
-------
Additional Support via Dynamac Corporation
GeoTrans
Ground-water modeling consultants from
academia and the private consulting community
-------
CSMoS FY1992 - Present
i m
Model Distribution by Client
Federal
State
Local
Private
Academic
Foreign
8%
1%
10%
o
Total Clients: 8006
65%
100
-------
CSMoS
Cumulative Client Model Requests
October 1991 - April 1995
10,000-
,2 8000--
CD
cr
4)
u
0)
A
I
I
6000--
4000--
2000--
8006
0
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
FY
92
FY
93
FY
94
FY
95
FY
96
-------
Center for Subsurface
Modeling Support
Ground-Water and Vadose Zone
Models Available:
- BIOPLUME H
- CHEMFLO
- GEOPACK
- HSSM
- MODFLOW
Instructional Manual
- MOFAT
- MT3D
- OASIS
- PESTAN
- RETC
- RITZ
- STF Database
- VLEACH
- WhAEM
- WHPA
6-15
-------
As ground-water model usage has
increased, a shortage of qualified
staff capable of appropriately
applying models has been identified.
-------
I
H-»
-^1
"There is no model that will adequately
describe all ground water quality problems
because the assumptions and simplifications
generally associated with models do not
adequately mimic all the processes that
influence the movement and behavior of the
water and/or the chemicals of interest."
(National Research Council, 1988).
-------
U.S. Environmental Protection Agency NRMRL/SPRD
Office of Research & Development Ada, Oklahoma
National Risk Management Research Laboratory
Center for Subsurface Modeling Support (CSMoS)
Ground-Water and Vadose Zone Models / Manuals ,
June 1996
Ordering Instructions
Preferred Methods
Download models and manuals via:
Homepage http://www.epa.gov/ada/kerrlab.html
Anonymous FTP ftp.epa.gov /pub/gopher/ada/models
Bulletin Board 405-436-8506 14,400 baud - 8 bits - 1 stopbit -no parity
Note: Manuals are in .pdf format. Must download Adobe Acrobat Reader program to read and print
.pdf files.
Alternate Method
• Request specific model(s) on company letterhead (Specify DOS or WIN >vhen appropriate)
• Include appropriate number of required new, pre-formatted, 3.5" high density diskettes
All models and manuals will be distributed on diskettes (Adobe Acrobat Reader included)
• All models are free of charge
Send requests to: NRMRL/SPRD
Center for Subsurface Modeling Support (CSMoS)
P.O.Box 1198
Ada, OK 74821-1198
405-436-8594 or 8656 or 8586
FAX 405-436-8718
BIOPLUME n (Version 1.1 - Oct '89)
Application: Two dimensional contaminant transport under the influence of oxygen limited biodegradation.
Processes: Advection, dispersion, sorption, biodegradation (aerobic and anaerobic) and reaeration.
Miscellaneous: Model is based on the 2-D solute transport code, USGS-MOC.
CHEMFLO (Version 1.30 - Aug '89)
Application: Simulation of 1-D water and chemical movement in vadose zone.
Processes: Advection, dispersion, first-order decay and linear sorption.
Miscellaneous: Numerical solution, handles a variety of boundary conditions.
c:\pm5Ncsmos\fliersNmodlist.pm5
-------
GEOPACK (Version l.O.e • Jan '90)
Application: Geostatistica! analysis of spatially correlated data.
Components: Basic statistics, variography, linear and nonlinear estimation (kriging).
HSSM (Version 1.10 - Apr '94) - Specifiy DOS or WIN Version
Application: Simulates flow of the LNAPL phase and transport of a chemical constituent of the LN APL from the
surface to the water table; radial spreading of the LNAPL phase at the water table, and dissolution and
aquifer transport of the chemical constituent.
Processes: One-dimensional in the vadose zone, radial in the capillary fringe, two-dimensional vertically averaged
analytical solution of the advection-dispersion equation in the saturatead zone.
Miscellaneous: Model is based on the KOPT, OILENS and TSGPLUME models.
MODFLOW Instructional Manual (Feb '93)*
Application: Instructional manual (study guide) for the USGS MODFLOW Model.
Components: A series of twenty problem sets that illustrate by example the use of MODFLOW including modeling
principles, input/output specifics, available options, rules of thumb, and common modeling mistakes,
etc.
*Miscellaneous: Manual does not require diskettes. Also, manual is not available on Homepage, FTP, or BBS. Must
submit request letter.
MOFAT (Version 2.0a - May '91)
Application: Two-dimensional flow and transport of three fluid phases: water, nonaqueous phase liquid, and gas.
Processes: Advection, dispersion, diffusion, sorption, decay, mass transfer.
Miscellaneous: Finite-element, numerous parameter required.
MT3D (Version 1.11 - Jan '92)
Application: Three dimensional contaminant transport in the saturated zone.
Processes: Advection, dispersion, non-linear sorption, first-order irreversible
decay, and biodegradation.
Miscellaneous: Numerical solution, uses hybrid method of characteristics, particle tracking, handles a variety of
discretization schemes and boundary conditions, includes MODFLOW program.
OASIS (Macintosh Version 2.1 only - May '94)
Application: A decision support system for contaminant transport modeling.
Components: A hydrogeologic database, two chemical databases, Bioplume II and a few other simple hydrogeologic
models.
PESTAN (Version 4.0)
Application: Vadose zone modeling of the transport of organic (pesticide) contaminants.
Processes: Advection, dispersion, first-order decay and linear sorption.
Miscellaneous: Screening model, few parameters required.
RETC (Version 1.1 - Nov '94)
Application: Estimates soil-water retention curve, unsaturated hydraulic conductivity or soil model parameters.
Processes: Uses the parametric equations of Brooks-Corey, van Genuchten and the pore-size distribution models
of Maulem and Burdine.
Miscellaneous: Analytical model, requires relatively few parameters.
-------
RITZ (Version 2.12 - Jan '88)
Application: Vadose zone modeling of the transport of contaminants associated with oily wastes.
Processes: Water movement, volatilization, degradation, sorption and leaching.
Miscellaneous: Analytical model, requires relatively few parameters.
STF (Version 2.0 • Jun '91)
Application: Database providing information concerning the behavior of organic and a few inorganic chemicals in
the soil environment.
Components: Degradation, transformation, toxicity, bioaccumulation and partitioning.
Miscellaneous: RITZ and VIP models included.
VLEACH (Version 2.2a - Jun '96)
Application: Simulation of 1 -D water and chemical movement in vadose zone.
Processes: Advection, sorption, vapor-phase diffusion, three-phase equilibration.
Miscellaneous: Numerical solution, screening model.
WhAEM (Version l.OOa - Oct '95)
Application: Delineates capture zones and isochrones of ground-water residence time for the purpose of "wellhead
protection."
Processes: Steady-state, homogeneous, isotropic, advection and dispersion.
Miscellaneous: Consists of two independent executables: G AEP (Geographic Analytic Element Preprocessor) and
CZAEM (Capture Zone Analytic Element Model).
WHPA (Version 2.2 - Sep '93)
Application: Simulates captures zones for pumping wells.
Processes: Steady-state, horizontal flow.
Miscellaneous: Particle-tracking with analytical, semi-analytical, numerical modules.
-------
Model Distribution Summary
MODEL NAME
BIOPLUME II
CHEMFLO
GEOPACK
HSSM(DOSorWIN)
MFINSTR MANUAL
MOFAT
MT3D
OASIS (Macintosh)
PESTAN
RETC
RITZ
STF
VLEACH
WHAEM
WHPA
MODEL DISKS
1
1
2
2
0
1
2
6
1
1
1
2
1
1
2
MANUAL DISKS
1
2
1
2
HARD COPY
1
3
HARD COPY
1
2
1
1
1
2
3
*Note: Add 1 disk for the Adobe Acrobat Reader per request.
Example 1
Requested Geopack - 4 disks
2 disks - model
1 disk - manual
1 disk - reader
Example 2
Requested Geopack and Pestan - 6 disks
2 disks - model Geopack
1 disk - model Pestan
1 disk - manual Geopack
1 disk - manual Pestan
1 disk - reader
-------
CSMoS
Software Distribution Request Form
Send to:
Name:
Company:
Address:
Phone:
FAX:
EMAIL
Requestor.
(if different)
Phone:
Date:
Q BIOPLUMEn(ldsk)
Q CHEMFLO (1 dsk)
Q GEOPACK (2 dsks)
Q HSSM DOS (1 dsk)
Q HSSM WIN (1 dsk)
Q MODFLOW
Instructional Manual
Q MOFAT(ldsk)
Q MT3D (2 dsks)
Q OASIS (6 Macintosh dsks)
Q PESTAN (1 dsk)
Q RETC (1 dsk)
Q RTTZ (1 dsk)
Q STF DATABASE (2 dsks)
Q VLEACH (1 dsk)
Q WHPA (2 dsks)
Q Modeling Information
; All.mocleis; arefre«of ^chargel-On^ompanyletterhead^ request specific mode1(sj and include
;appropjia1^;number:oif^prerfprmalted, 3.§™high densi^ diskettes.'All models will be copied
onto diskettes and returned with user's manual (The MODFLOW Instructional Manual does
1 •••• y sv >..y&> *?VA* w--v*'- ^«.",!^'. —^* ,, *w>J> «^vs« -""f--*.-. ,•**"&- ,4" *^ - . „ ,>, ? ^ .^ v'. '. . . - %".^v
no
nter for Subsurface Modeling Support
- * -''1 •! - -- - '.^.f
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box1198
Ada, OK 74820
January 1992
Center for Subsurface Modeling Support
SEPA BIOPLUMEII:
Two-Dimensional Contaminant Transport
Under the Influence of Oxygen Limited
Biodegradation in Ground Water
(Version 1.0)
Introduction
Biotransformation processes are potentially important in the
restoration of aquifers contaminated with organic pollutants.
As a result, these processes require evaluation in remedial
action planning studies associated with hydrocarbon
contaminants. BIOPLUMEII is a versatile two-dimensional
numerical model that simulates the transport of dissolved
hydrocarbons under the influence of oxygen-limited
biodegradation in homogeneous and heterogeneous media.
It is designed to aid environmental managers, regulators.
and planners in evaluating the movement of a hydrocarbon
plume in the saturated zone. BIOPLUMEII predicts changes
in the contaminant plume due to convection, dispersion,
mixing, and biodegradation.
BIOPLUME II is based on the USGS solute transport code
MOC. The model solves the solute transport equation twice
for each time step, once for hydrocarbon concentration and
onceforoxygenconcentration. Nonphotosyntheticmicrobes
obtain energy for growth by oxidation of substrates
(hydrocarbons) in the presence of an electron acceptor,
such as oxygen under aerobic conditions, and nitrate,
sulfate, and carbon dioxide under anaerobic conditions.
Oxygen is assumed to be the sole electron acceptor in
BIOPLUME II. To simulate the biodegradation process, the
model assumes an instantaneous reaction between oxygen
and hydrocarbon. Independent mass balances are calculated
for oxygen and hydrocarbon and are adjusted to account for
the mass loss due to biodegradation.
BIOPLUME II is a particularly useful tool for remedial action
planning. The model simulates three natural oxygen sources:
existing dissolved oxygen concentration; reaeration from
the vadose zone; and recharge of dissolved oxygen from
ground-water inflow. The model allows for the injection of
oxygen from wells associated with biorestoration. As a
result, BIOPLUME II can be utilized to optimize the efficiency
and effectiveness of remedial designs. BIOPLUME II has
been previously used in conjunction with remedial activities
at two Superfund sites.
Input/Output
A menu-driven preprocessor allows the user to easily enter
and edit data files in BIOPLUME II. The model requires input
parameters that define the hydrogeologic and chemical
conditions. The hydrogeologic parameters include
potentiometric elevation, transmissivity, thickness, effective
porosity, recharge, leakance and pumping/injection rates.
Chemical parameters include contaminant and oxygen
concentrations, half-life of solute, anaerobic decay, and
reaeration. Sensitivity analyses indicate the biodegraded
mass is particularly sensitive to the hydraulic conductivity,
anaerobic decay, and reaeration parameters.
Model output is provided as an oxygen and hydrocarbon
distribution matrix. However, the output can be plotted as
contour diagrams using the SURFER (Golden Software)
graphics package. The preprocessor includes a graphics
option that will transform the oxygen and hydrocarbon
matrices for direct use in SURFER.
Documentation
The BIOPLUME II software package includes a useful,
comprehensive user's manual. The manual presents an
overview of the model and its operation in a clear, concise
format. The theoretical processes on which the model is
based and a discussion of the model development are also
given. In addition, the manual documents the results of a
sensitivity analysis and provides example problems, which
aid in the use of the model.
Software/Hardware Requirements
The software/hardware requirements for BIOPLUME II are:
• IBM PC-AT. PS/2 or compatible computer
• Minimum 640K RAM
• 80287 or 80386 math coprocessors
• A graphics card is highly recommended if graphics are
to be used
• The operating system should be MS-DOS 2.0 or later
• SURFER graphics software from Golden Software is
recommended
Availability
To obtain a copy of the BIOPLUME II software package and
the user's manual, send one (1) pre-formatted, high density,
3.5 inch diskette in a floppy disk mailer to the following
address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada, OK 74820
January 1992
Center for Subsurface Modeling Support
®EPA CHEMFLO:
One-Dimensional Water
and Chemical Movement
in Unsaturated Soils
(Version 1.30)
Introduction
CHEMFLO is a one-dimensional model that simulates the
effects of soil hydraulic properties on the movement of water
and chemicals in the unsaturated zone. The software was
developed to assist regulators, environmental managers,
consultants, scientists, and students in understanding
unsaturated flow and transport processes. Water movement
and chemical transport are modeled using the Richards and
the convection-dispersion equations, respectively. The
equations are solved numerically using finite differences.
The CHEMFLO software package is user friendly. Only
limited technical terminology is used, and an understanding
of the model's mathematical basis is not required. From the
main menu, the user selects the desired simulation, water
movement or both water and solute movement. Four
different soils are included in the model to run the simulation,
but additional soil types and properties can also be entered.
The model can simulate both drainage and drying of a soil.
General assumptions in the model include homogeneous
soil properties, negligible hysteresis in the wetting and
drying process, and chemical partitioning is instantaneous
and reversible.
Input/Output Parameters
The model requires input parameters defining the water
system, the chemical character, and the soil conditions.
Several different conditions can define the upper and lower
boundaries of the water system; these include constant
matric suction, constant flux, constant rainfall, and mixed
type. In addition the matric potential can be defined
through-out the soil column. Constant chemical
concentrations are assumed to occur at the soil boundaries
but heterogeneous conditions can exist within the soil
column. Additional chemical parameters needed include
the soil-water partitioning coefficient, dispersivity, and
degradation rates. The program allows the user to choose
from a variety of moisture retention characteristic curves
such as the Brooks-Corey, Brutsaert, and van Genuchten
models. Similar models are also used to describe the
hydraulic conductivity-moisture content relationship.
Required soil parameters include bulk density, saturated
hydraulic conductivity, saturated water content (porosity),
residual water content, and various conductivity and water
content parameters.
CHEMFLO presents the modeling output in both graphical
and tabular form. Graphical output includes numerous
displays such as matric potential, driving force, conductivity,
and flux density versus time or distance. The model allows
the display of three graphs per screen.
Documentation
The CHEMFLO user's guide is well documented. The
mathematical equations, on which the model is based, are
presented as well as the limitations of the model. Menu
items, input screens, and output displays are illustrated and
described. Furthermore, a set of numerical experiments are
presented to enable the user to gain an understanding of the
dynamic processes involved in water movement and chemical
transport in soils.
Software/Hardware Requirements
The software/hardware requirements are:
• IBM PC, AT, PS/2 (or a compatible) computer
• Minimum 640K RAM
• Two floppy disk drives or one floppy disk drive and one
fixed disk
• A graphics card compatible with IBM, CGA, EGA, or
VGA graphics, and a compatible monitor
• An 80x87 math coprocessor is highly recommended
• The operating system must be MS-DOS or PC-DOS
2.01 or later
Availability
To obtain a copy of the CHEMFLO software package and the
user's manual, send one (1) pre-formatted, high density, 3.5
inch diskette in a floppy disk mailer to the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box1198
Ada. OK 74820
(405)436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada, OK 74820
January 1994
Center for Subsurface Modeling Support
dEPA GEOPACK:
Geostatistlcs for Waste
Management
(Version 1.0.e)
Introduction
Geostatistics is a useful group of statistical methods that can
be applied where a sample value is expected to be affected
by its position and its relationship with its neighbor. Although
the application of geostatistics has historically been utilized
in the mining industry, it is beginning to be utilized effectively
in the environmental field since it is often required that
chemical and/or physical parameters be estimated at specific
locations where data is absent.
GEOPACK is a useful, comprehensive software program
that conducts both standard statistical and geostatistical
analyses. Basic statistics such as mean, median, variance,
standard deviation, skewness, kurtosis, linear regression,
polynomial regression, and the Kolomogorov-Smirnov test
can be calculated. More importantly, GEOPACK includes
geostatistical analyses of the spatial variability of one or
more random functions. Semivariograms and cross-
semivariograms for combined random functions for two-
dimensional spatially dependent random functions can be
determined. In addition, linear and nonlinear estimations
can be calculated using kriging and cokriging estimators as
well as disjunctive kriging and disjunctive cokriging
techniques.
GEOPACK is a user friendly software package that can be
used by scientists, engineers or regulators having little
experience in geostatistical techniques. The program
employs an integrated system which frees the user from
excessive file editing and program manipulation. In addition,
the system is adaptable to the incorporation of additional
programs without having to alter previous programs or
recompile the entire system. Furthermore, GEOPACK
includes an on-line help screens concerning the operation of
the system, its capabilities and limitations, as well as
programming conventions and definitions. Finally, example
problems are provided to aid the user in executing GEOPACK.
Input/Output
To operate GEOPACK, a data file is created using a text
editor or word processor. The data must contain one or more
spatially-dependent random variable (i.e., soil moisture, K,
concentration, etc.) and an X and Y coordinate position for
each value. Output is presented in tabular and graphical
output forms. Various graphics capabilities are included with
GEOPACK such as linear or logarithmic line plots and
contour and block diagrams. Device drivers for dot matrix
printers, HP plotters, and HP Laser Jet printers are included
with GEOPACK. The graphics programs produce
intermediate quality output, however high quality graphics
can be easily developed by interfacing with commercially
available packages.
Documentation
The GEOPACK user's manual presents a description of the
software and operation instructions. Aspects of the program
system, data files, as well as potential program enhancements
are described. In addition, each of the ten menus of the
model are illustrated and documented. The manual does not
discuss either the conceptual or mathematical basis of the
geostatistical techniques utilized, however references and
suggested literature are given.
Software/Hardware Requirements
The software/hardware requirements are:
• IBM PC-AT or compatible computer
• Minimum 640K RAM
• Hard disk storage of about 4 Mbytes
• A graphics card compatible with CGA, EGA, VGA or
Hercules graphics and a compatible monitor
• A math coprocessor is not required but will be used by
the model if available
• The operating system should be MS-DOS 3.30 or greater
Availability
To obtain a copy of the GEOPACK software package and
the user's manual, send two (2) pre-formatted, high density,
3.5 inch diskettes in a floppy disk mailer to the following
address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O.Box 1198
Ada, OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box1198
Ada. OK 74820
June 1994
Center for Subsurface Modeling Support
SEPA A Manual of
Instructional Problems for
the U.S.G.S. MODFLOW
Manual
Introduction
A Manual of Instructional Problems for the USGS MODFLOW
Model is a practical self study guide illustrating the application
of the USGS modular ground-water flow model, MODFLOW.
The manual was developed as a text for modeling and
ground-water hydrology courses and as a self study guide to
aid regulators, modelers, and educators in learning
MODFLOW. The information in the manual should be of
interest to both the beginner and advanced modeler for
hands-on experience with the practical application of
MODFLOW. In addition the problems in the manual can be
used to verify the correct installation of the MODFLOW code
on a particular computer system and as a benchmark for the
verification of other flow models.
The Manual
The manual addresses the principles of ground-water flow
modeling and model options associated with MODFLOW. It
provides a series of twenty problem sets that illustrate by
example the use of MODFLOW including modeling principles,
input/output specifics, available options, rules of thumb, and
common modeling mistakes. For each problem, the manual
presents a problem statement and setup, input data sets,
model output, and a discussion of results. Model input data
sets and output files are included on diskettes available in
the manual.
Specific issues addressed in the manual include radial flow
to a well, anisotropic conditions, artesian-water table
conversion, steady-state simulations, grid and time-stepping
considerations, transient calibration and prediction,
representation of aqurtards and leaky aquifers, solution
techniques and convergence, head dependent (third-type)
boundary conditions, cross-sectional simulations, application
to a water supply problem, and application to a hazardous
waste problem. All of the packages in MODFLOW are
utilized at least twice in the series of problems.
Data set preparation time and execution time have been
minimized by simplifying the problems to a small size and to
focus only on specific aspects. Model grids are generally
smaller and more homogeneous than would be used in
practice; however, the intent and result of each exercise are
not compromised by the simplification.
The user of the manual should attempt to solve the problems
as described in the problem statement portion of each
exercise. The model setup can be checked in the data set
listing given in the model input section of each problem.
Results can be checked by the pertinent portions given in the
model output section. Some training on the structure and
input of MODFLOW as well as some training on the theory of
ground-water modeling is assumed. Abbreviated input
instructions are given in the manual as an Appendix; however,
the user will need to refer to the MODFLOW manual on some
occasions.
A secondary function of the manual is for verification purposes.
Although the MODFLOW code has been extensively applied,
very little documentation of its testing and verification is
available in the literature. To address this situation, where
possible, model generated results are compared to analytical
solutions, results of other models, or to simulations with
alternative boundary conditions or configurations. In addition
to providing informal benchmarking of MODFLOW, these
problems can be used to verify the correct installation of the
code on a particular computer system or to verify that certain
user modifications have not altered the integrity of the
program.
Documentation and Software
A Manual of Instructional Problems forthe USGS MODFLOW
Model includes hard copy text and three diskettes. The text
introduces each problem, describes the available input data,
outlines the output, and discusses the results. Each problem
is supported by several figures and tables to illustrate and
clarify the model scenario. The diskettes contain input and
output data sets for each problem. A copy of MODFLOW is
not included in the manual. It is assumed the user has
obtained a copy of MODFLOW and has the necessary
computer hardware to execute the program.
Availability
To obtain a copy of the Manual of Instructional Problems for
the USGS MODFLOW Model, send a letter of request to the
following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O.Box 1198
Ada, OK 74820
Phone: (405)436-6500
FAX: (405)436-8529
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada. OK 74820
July 1994
Center for Subsurface Modeling Support
EPA MOFAT:
A Two-Dimensional Finite Element
Program for Multiphase Flow and
Multicomponent Transport
(Version 2.0a)
Introduction
The flow and transport of multiple fluid phases, such as water,
hydrocarbons, and air, are common conditions in soils and aquifers
at contaminated sites. In addition, the need to evaluate potential
impacts derived from multi-component contaminants and multiphase
conditions is increasing. However, simulation of these conditions
remains a difficult problem within the environmental field. In
particular, few microcomputer (PC)-based models describe such
phenomena, and data to support these modeling efforts are difficult
to obtain.
MOFAT is a two-dimensional, multiphase, mutti-component finite
element model that is useful in simulating the transport of nonaqueous
phase liquids under both saturated and nonsaturated conditions. In
particular, MOFAT simulates the movement of three fluid phases:
water, nonaqueous phase liquids (NAPLs), and gas. The code will
simulate flow or coupled flow and transport. The flow module can be
used to evaluate two-phase flow of water and NAPL in a system of
constant gas pressure or explicit three-phase flow of water, NAPL,
and gas at variable pressures. The transport module can incorporate
five partition components that simulate the redistribution of the
contaminant between water, NAPL, gas, and solid phases.
Due to its ability to simulate multiphase flow and multicomponent
transport, MOFAT has the flexibility to handle the conditions expected
in a field scenario. For example, the flow of a gas phase can be
modeled directly, and its impact on volatilization can be assessed.
Likewise, NAPL flow and transport can be evaluated from a leaking
underground storage tank or an accidental spill.
Input/Output
The flexibility of MOFAT to simulate field conditions is highly
dependent on the availability of input data. Required input for flow
analyses consists of initial conditions, soil hydraulic properties, fluid
properties, time integration parameters, boundary condition data
and mesh geometry. Three-phase permeability-saturation-capillary
pressure relations are defined by an extension of the van Genuchten
model, which considers the effects of oil entrapment during periods
of water imbibition. Time-dependent boundary conditions for the
flow analysis may involve user-specified phase heads at nodes or
phase fluxes along a boundary segment. For transport analyses,
additional input data are porous media dispersivities, initial water
phase concentrations, equilibrium partition coefficients, component
densities, diffusion coefficients, first-order decay coefficients, mass
transfer coefficients (for nonequilibrium analyses) and boundary
condition data. Initial conditions in a transport simulation are
specified in terms of equilibrium water phase concentrations of each
partitionable component. The time dependent boundary conditions
may be stipulated as equilibrium water phase concentrations in the
porous medium, as prescribed fluxes defined in terms of a specified
concentration in the influent liquid, or by a zero dispersive flux. The
MOFAT code contains a pre-processor to input this data.
Program output consists of basic information on input parameters,
mesh details and initial conditions as well as pressure heads,
saturations, and velocities for each phase at every node for specified
output intervals. The total volume or mass of each phase, time-step
size and number of iterations are printed at each time-step. For
transport analyses, the phase concentrations at each node are
output at each print-out interval. In addition, the program allows the
user to restart simulations from a previous simulation using an
auxiliary data file so that initial conditions for the restart problem do
not have to be duplicated. Output from MOFAT can be exported to
a commercial graphics package to plot graphs or concentration
diagrams.
Documentation
The MOFAT software package includes a comprehensive user's
manual. The manual fully describes the mathematical models for
multiphase flow and transport and the numerical models for the flow
and transport equations. The basic features and operation of the
model are discussed. In addition, the manual presents information
on the estimation of soil, bulk fluid, and component properties.
Furthermore, three example problems a presented in the manual.
References are also provided to assist the user in further research.
Software/Hardware Requirements
The software/hardware requirements for MOFAT are:
IBM-PC, XT. AT or compatible computer
Approximately 4 MB RAM
A hard disk
One floppy disk drive
DOS 3.0 or higher
A 80386 or 80486 math coprocessor is required
A commercial graphics software package is
recommended
Availability
To obtain a copy of the MOFAT software package, send one (1) pre-
formatted, high density, 3.5 inch diskette in a floppy disk mailer to the
following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada. OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O.Box 1198
Ada. OK 74820
June 1992
Center for Subsurface Modeling Support
AEPA MT3D:
A Modular Three-Dimensional
Transport Model
(Version 1.3)
Introduction
Numerical modeling of contaminant transport, especially in three
dimensions, is difficult and vulnerable to numerical errors such as
numerical dispersion and artificial oscillation as well as requiring
considerable memory and execution time. These factors, along with
the difficulty involved in characterization and data procurement,
have made three-dimensional transport modeling impractical for
many field applications, particularly in the microcomputer (PC)
environment.
MT3D, a Modular Three-Dimensional Transport Model, is a useful
tool in simulating advection, dispersion, and chemical reactions of
dissolved constituents in ground-water systems in either two or
three dimensions. It resolves many of the typical problems of
numerical transport modeling by utilizing a mixed Eulerian-Lagrangian
approach based on a combination of the method of characteristics
(MOC) and the modified method of characteristics (MMOC). This
hybrid method of characteristics (HMOC) approach combines the
strength of the MOC for eliminating numerical dispersion in the
presence of sharp concentration fronts and the computational
efficiency of MMOC. In addition, MT30 uses a modular structure
that makes it possible to simulate advection, dispersion, source/sink
mixing, or chemical reactions independently without reserving
computer memory space for unused options. Furthermore, new
options can readily be added to the model without having to alter the
existing code.
MT3D can be utilized for a variety of hydrologic settings. The model
allows for several different spatial discretization schemes and
transport boundary conditions. These include: (1) confined,
unconfined or variably confined/unconfined aquifer layers; (2) inclined
model layers; (3) variable cell thickness within the same layer; (4)
specified concentration or mass flux boundaries; and (5) the solute
transport effects of external sources and sinks such as wells, drains,
rivers, area! discharge and evapotranspiration.
In the MT3D code, changes in concentration of a dissolved
contaminant in ground water can be simulated due to advection,
dispersion, and chemical reactions. The chemical processes included
in the model are simple, equilibrium-controlled linear or non-linear
sorption and first-order irreversible decay or biodegradation.
However, the code will allow more sophisticated chemical reactions
if the user desires. It is assumed by the model that changes in the
concentration field will not significantly affect the flow field, thereby
enabling the transport problem to be solved independent of the flow
problem.
Input/Output
When using MT3D the advection term in the contaminant transport
problem has to be evaluated using a flow model. Hence, the MT3D
transport model can be used with any block-centered finite-difference
flow model, for example MODFLOW. As a result, after a flow
modeling simulation has been conducted and calibrated, the data,
such as heads and flow terms, determined from the flow modeling
effort can be saved in a binary file, which can then retrieved by
MT3D. Thus, MT3D allows simulation of contaminant transport
without having to learn a new flow model or to modify an existing flow
model to fit the transport model. Likewise many transport simulations
can be performed while the flow solution remains constant.
MT3D includes post-processing programs to facilitate output use.
The programs can extract the simulated concentrations at desired
transport steps and save the data in such a form that it can be used
by most commercially available graphic software packages to
generate contour maps and othertypes of plots. The post-processing
programs also can extract the concentration data at the last step and
save it in a separate file as the starting concentration for a continuation
run.
Documentation
The MT3D software package includes a comprehensive user's
manual. The manual fully describes the modified approach on which
the program is based as well as the framework and the assumptions
of the model. Instructions regarding the use of the model is well
documented. In addition, example problems including input and
output data sets are presented. References are also provided to
assist the user in further research.
Software/Hardware Requirements
The software/hardware requirements for MT3D are:
IBM PC, XT, AT, (or a compatible) computer
Minimum 3.5 MB Extended Memory
An 80386 or 80486 math coprocessor is required
The operating system must be PC-DOS or MS-
DOS version 2.0 or greater
• A graphics card is highly recommended if
commercial plotting programs are to be used
Availability
To obtain a copy of the MT3D software package, send two (2) pre-
formatted, high density, 3.5 inch diskettes in a floppy disk mailer to
the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada. OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada. OK 74820
Center for Subsurface Modeling Support
EPA OASIS:
Parameter Estimation System
for Aquifer Restoration Models
(Version 2.1)
April 1994
Introduction
OASIS is a graphical decision support system for ground-
water modeling. It was designed to provide scientists and
modelers with a collection of tools to assess and analyze
ground-water contamination problems. OASIS consists
of several different components including reference
documentation, databases, and ground-water transport
models. The software system operates on a Macintosh
computer platform using HyperCard. OASIS is extremely
user-friendly as the HyperCard system allows one to
navigate through large amounts of information using a
series of mouse clicks on active buttons.
Components
Useful reference documentation related to the ground
water field is provided in OASIS. The documentation
consists of information on contaminant source by industry,
contaminant source by zone, a glossary of terms,
remediation techniques, and parameters used in ground-
water modeling (the Rokey database). As a result, the
user can obtain a wealth of information regarding potential
contaminants and possible remediation techniques from
the reference documentation component.
OASIS contains an extensive amount of information in the
database component of the system to facilitate most
ground-water modeling efforts. The OASIS chemical
database consists of over 130 compounds with extensive
information available for about 20 of the more common
compounds. The chemical database includes such
information as specific gravity, vapor pressure, and water
solubility. In addition to the chemical database, OASIS
includes a hydrogeologic database. This database
consists of information obtained from over 400 field sites
across the United States and utilizes the concept of
hydrogeologic settings developed for the E P A's D R ASTIC
system.
Two transport models, ODAST and BIOPLUME II are
included in the OASIS software system. ODAST is a one-
dimensional analytical model that considers advection,
dispersion, solute decay, source decay and adsorption. It
is a useful tool for providing preliminary estimates of
solute transport. BIOPLUME II is a two-dimensional
numerical model that simulates the transport of dissolved
hydrocarbon under the influence of oxygen-limited
biodegradation. The model is based on the USGS solute
transport code MOC. BIOPLUME II is versatile in that it
can simulate natural biodegradation processes, retarded
plumes, and in-situ biorestoration scenarios. Simulations
that demonstrate the importance of well placement and
source term definition using BIOPLUME II are presented.
Documentation
The user's manual for OASIS provides limited but ample
description in utilizing the software system. However, the
OASIS program is user-friendly as it provides extensive
on screen help and documentation. Throughout the
system, icons lead the user to further discussion and help,
if needed. As a result, the OASIS system can be operated
without support from cumbersome paper documentation.
Example problems are included in the manual to assist
the user in running the software.
Software/Hardware Requirements
The software/hardware requirements are:
• Apple Macintosh II, llx, Ilex, SE/30, or SE (limitations
exist with the SE model)
• Minimum 1 Mb of RAM
• Minimum of 9 Mb of Hard Disk Storage
Availability
To obtain a copy of the OASIS software system and the
user's manual, send six (6) pre-formatted, high density,
3.5-inch diskettes in a floppy disk mailer to the following
address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box1198
Ada, OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada, OK 74820
May 1992
Center for Subsurface Modeling Support
SEPA PESTAN:
Pesticide Analytical Model
(Version 4.0)
Introduction
PESTAN version 4.0, the Pesticide Analytical Model, is a
computer code for estimating the transport of organic solutes
through soil to ground water. H is a one dimensional, unsatur-
ated flow and transport model that is based on a closed-form
analytical solution of the advective-dispersive-reactive
transport equation. The original code was developed by
Enfield et a), in 1982 and has since been used by the EPA
Office of Pesticides Program (OPP) for initial screening
assessments to evaluate the potential for ground-water
contamination of pesticides. PESTAN has been tested under
both field and laboratory conditions. The recent version of the
code is user-friendly, menu-driven, and well documented.
The vertical transport of dissolved pollutants through the
vadose zone is simulated in PESTAN as a "slug* of contami-
nated water that migrates into a homogeneous soil. The
concentration of the slug equals the solubility of the pollutant in
water. The pollutant begins to enter the soil at the first
precipitation/irrigation event at a rate equal to the pore water
velocity. PESTAN assumes steady flow conditions through the
soil domain according to the relationship developed by
Campbell (1974), which relates water content of the soil to
hydraulic conductivity. Once the slug enters the soil, the
pollutant transport is influenced by sorption and dispersion.
Linear isotherms describe the partitioning of the pollutant
between liquid and soil phases, and local or instantaneous
equilibrium between these phases is assumed. Mass of the
pollutant can be lost via first-order liquid-phase decay or by
means of migration out of the soil domain.
Input/Output Parameters
The model requires input parameters describing the pollutant
properties, soil character, and environmental conditions. The
pollutant parameters include water solubility, sorption constant,
and solid- and liquid-phase decay rates. Parameters that
define the soil character are bulk density, saturated water
content, saturated hydraulic conductivity, dispersion coeffi-
cient, and the characteristic curve coefficient. Environmental
parameters include recharge rate, pollutant application rate,
and depth to ground water. The user's guide provides
considerable information in the form of tables and references
regarding suggested values for many of the parameters.
PESTAN presents the modeling output in tables and files.
which can be used to plot graphs using common commercial
graphics packages. The tabular output includes an input
summary, calculated water and pollutant conditions, the
pollutant concentration profile in the soil, and mass balance
results. Three different graphs can be constructed using the
output files: a breakthrough curve, a pollutant flux curve, and a
soil-depth pollutant concentration profile. The time and
location for the curves can be selected by the user.
Documentation
The PESTAN model includes a well documented user's
manual. The manual describes the conceptual framework and
assumptions of the model as well as the mathematical basis
for the code. The input and output parameters are fully
described. In addition, a sensitivity analysis is presented so
the user can evaluate the potential effects of given model
parameters. Numerous appendices are included in the guide
that present values for various pollutant and soil parameters.
Furthermore, a model data sheet is given to record information
regarding the values for a model simulation. References are
also provided to assist the user in further research.
Software/Hardware Requirements
The software/hardware requirements for PESTAN are:
IBM-PC or compatible computer
Minimum 256K RAM
Color Graphic Adapter (CGA) board
One floppy disk drive (MS/PC) DOS 2.0 or higher
Recommended hardware and software include:
A math coprocessor
A hard disk
A commercial graphics software package
Further Information
To obtain a copy of the PESTAN software package and the
user's manual, send one (1) pre-formatted, high density. 3.5
inch diskette in a floppy disk mailer to the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory .
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box 1198
Ada. OK 74820
May 1992
Center for Subsurface Modeling Support
SEPA RETC:
The Retention Curve
Computer Code
(Version 1.0)
Introduction
Computer models are becoming widely utilized in the environmental
industry to simulate the movement of water and chemicals into
and through the unsaturated zone. These simulations aid in our
understanding of subsurface processes and. in many cases.
provide a basis for important management decisions. To be
utilized properly these computer models require knowledge of the
hydraulic properties of the soil. Unfortunately, the measurement
of many soil hydraulic properties is costly and difficult, which
limits the usefulness as well as the accuracy of these computer
simulations.
The RETC (Retention Curve) computer code provides an
alternative to direct measurements of the soil water retention
curve and the hydraulic conductivitiy curve by using theoretical
methods to predict these properties from more easily measured
and available soil water retention data. Specifically, RETC
utilizes several analytical models to estimate water retention,
unsaturated hydraulic conductivity or soil water diffusivity for a
given soil. The analytical models contained in RETC include the
parametric equations of Brooks-Corey and van Genuchten, which
are used in conjunction with the theoretical pore-size distribution
models of Mualem and Burdine to predict the unsaturated hydraulic
conductivity from observed soil water retention data. The RETC
code can be used in two different modes, forward and fitting. The
forward problem estimates the soil-water retention curve and
hydraulic conductivity from the analytical model parameters.
Conversely, the fitting or parameter estimation mode determines
the analytical model parameters from soil-water retention and/or
hydraulic conductivity versus suction data.
RETC provides assistance to the user in defining the input model
parameters and in evaluating the model results. The model
parameters can be obtained from nonlinear regression or can be
constrained by the user when knowledge regarding these
parameters is known. In addition, the model provides default
values for the parameters to be estimated; however, the user can
modify the initial estimates if proper converged values are not
obtained. The code also provides a comprehensive statistical
description of the converged parameter estimates, including
standard error, 95 percent confidence limits, and correlation
matrix coefficients. These are very useful in assessing the quality
of the parameter estimates. In addition, the maximum number of
iterations required for parameter convergence can also be adjusted
by the user.
Input/Output Parameters
To use RETC, either hydraulic properties of the soil or the
analytical model parameters are entered as input into the program.
The user can then select the desired analytical method from a
menu screen to derive the output. The resulting program output
is presented in a file designated by the user. If the output is soil
water retention and/or hydraulic conductivity/diffusivrty data then
these values can be written to separate files for plotting. This data
is saved as a standard ASCII text file and can be plotted using
various commercial off-the-shelf graphing packages.
Documentation
The RETC software package includes a users manual that
thoroughly describes the theoretical basis for the analytical
solutions used in the model. The manual only briefly discusses
the operational aspects of the code, such as the input.output, and
menu screens. However, the menu screens are well documented
in the program so that the operation of the code is not difficult.
Example input and output files are provided in the appendix of the
RETC code.
Software/Hardware Requirements
The software/hardware requirements for RETC are:
IBM PC or compatible computer
Minimum 256K RAM
• One floppy disk drive
An 8087 or 80287 math coprocessor is
recommended
The operating system should be MS-DOS
version 2.0 or greater
Availability
To obtain a copy of the RETC code and the user's manual, send
one (1) pre-formatted, high density. 3.5 inch diskette in a floppy
disk mailer to the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada. OK 74820
(405) 43&4500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box1198
Ada, OK 74820
January 1992
Center for Subsurface Modeling Support
SEPA RITZ:
Regulatory and Investigative
Treatment Zone Model
(Version 2.12)
Introduction
RITZ, the Regulatory and Investigative Treatment Zone
model, is a useful tool for predicting fate and transport
potentials of hazardous organic constituents contained in
contaminated soils. It is a one dimensional, unsaturated flow
and transport model. The model provides an estimate of the
amount of each organic constituent which will be volatilized,
transformed, leached, and retained in a defined zone of the
soil. One of the principal features of RITZ is its ability to
account for effects of the presence of oil in the waste-soil
matrix. Hydrophobic compounds, in general, tend to be
highly soluble in oil and only sparingly soluble in water; thus,
these constituents will commonly associate with any oil
phase that is present in the matrix. RITZ accounts for the
effect of sorption of a hazardous constituent into the oil and
biodegradation of both the hazardous constituent and the oil
phase.
RITZ was developed to aid regulators, environmental
managers, and researchers in making decisions involving
the movement and transformation of hazardous chemicals
in the vadose zone. Potential applications for RITZ include
land treatment evaluations, site closure permitting, and
remedial action planning.
The model describes the subject site as consisting of two
zones: the plow zone where the contaminated material is
applied; and the treatment zone where the contaminant is
transported and transformed. General assumptions in the
model include uniform soil properties, homogeneous waste
mixing, insignificant hydrodynamic dispersion, immobile oil
phase, and first order degradation and linear sorption.
Input/Output Parameters
The model requires input parameters describing the soil
character, oil and pollutant properties, and operational and
environmental factors. The soil parameters include fraction
organic content, bulk density, saturated water content, and
saturated hydraulic conductivity. Most of these factors can
be obtained from U.S. Soil Conservation reports or can be
estimated fairly accurately. A weighted averaging feature is
provided in the model to calculate uniform soil conditions
from heterogeneous soils. The input parameters describing
the oil and pollutant properties may not be readily available
from most references and must be estimated. Likewise,
some of the operational and environmental factors, recharge
rate, evaporation rate, and temperature, are generally difficult
to obtain. However, the user's guide provides valuable
information as to how these parameters can be obtained.
RITZ presents the modeling output in numerous well de-
signed formats. Model results describe the pollutant
concentration and fluxes in various phases as a function of
depth and time. The user can select the depths and times
of interest as well as the desired format and output device.
Graphical output includes line, bar, and pie chart formats.
Results can also be obtained in tabular form.
Documentation
The RITZ model includes a well documented user's manual.
The manual describes the conceptual framework and
assumptions of the model as well as the requirements for
model use. The input and output parameters are fully
described. In addition, the computational equations of the
model are presented. References are also provided to assist
the user in further research.
Software/Hardware Requirements
The software/hardware requirements for RITZ are:
• IBM PC, XT, AT, (or a compatible) computer
• Minimum 256K RAM
• One floppy disk drive
• An 8087 or 80287 math coprocessor
• An IBM color/graphics board and a compatible monitor
• The operating system must be PC-DOS or MS-DOS
version 2.0 or greater
Availability
To obtain a copy of the RITZ software package and the
user's manual, send one (1) pre-formatted, high density, 3.5
inch diskette in a floppy disk mailer to the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, OK 74820
(405) 436-8500
-------
United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
P.O.Box 1198
Ada, OK 74820
April 1994
Center for Subsurface Modeling Support
&EPA VLEACH:
A Vadose Zone Leaching Model
(Version 2.1)
Introduction
VLEACH is a relatively simple one-dimensional finite
difference model designed to simulate the leaching of a
volatile, sorbed contaminant through the vadose zone. It is
a useful screening and educational tool for regulators,
environmental managers, and researchers in evaluating the
migration of contaminants in soil and the potential effects of
contaminants to the ground water.
VLEACH can be used to simulate the transport of any non-
reactive chemical that displays linear partitioning behavior.
In particular, VLEACH simulates four transport-related
processes: (1) liquid-phase advection, (2) solid-phase
sorption, (3) vapor phase diffusion, and (4) three-phase
equilibrium. The contaminant mass within each model cell
is partitioned among liquid (dissolved in water), vapor, solid
phases. During each discrete time step, which can be
defined by the user, VLEACH simulates three separate
processes; these include: the downward migration of the
contaminant in the liquid phase, the gas diffusion of the
contaminant in the vapor phase, and the equilibration of the
contaminant according to the distribution coefficients within
each model cell.
Numerous conditions are assumed in VLEACH. The model
assumes a homogeneous porous medium with steady flow
and no dispersion. In addition, partitioning is linear; there is
no in-situ production or degradation; and free product is not
present. These assumptions over-simplify most vadose
zone transport problems, and as a result, constrain VLEACH
as a screening tool.
Input/Output
VLEACH is easy to use as it requires input data that is,
generally, readily known or accessible. Specifically, the
input data includes information regarding chemical, soil, and
site conditions. The required chemical data include the
organic carbon coefficient (Kx), Henry's Law constant (KJ,
the aqueous solubility and the free air diffusion coefficient.
The input soil properties are dry bulk density, total porosity,
volumetric water content and organic carbon fraction, and
the input site parameters are recharge rate and depth to
ground water.
VLEACH output consists of mass balance calculations and
ground-water impact estimates. The mass balance
calculations compare the change in mass within the profile
to the calculated boundary fluxes, while the ground-water
impact calculations are based on the total downward flux of
the contaminant at the water table. In addition, the vertical
concentration profiles for the vadose zone for all three
phases can be retrieved at user-selected intervals throughout
the model simulation.
Documentation
The VLEACH software package contains a user's manual
that describes all the principal aspects of the model and its
use. The manual includes a description of the model theory
as well as a discussion of the chemical processes involved
in the model. The manual clearly describes the form of the
model input and output and presents a sample input file. In
addition, the source code is documented in the manual.
Software/Hardware Requirements
The software/hardware requirements for VLEACH are:
• IBM PC, XT, AT, (or a compatible)
computer
Minimum 256K RAM
One floppy disk drive
An 8087 or 80287 math coprocessor
The operating system must be PC-DOS or
MS-DOS version 2.0 or greater
Availability
To obtain a copy of the VLEACH software package and the
user's manual, send one (1) pre-formatted, high density, 3.5
inch diskette in a floppy disk mailer to the following address:
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
Center for Subsurface Modeling Support
P.O. Box 1198
Ada, OK 74820
(405) 436-8500
-------
PRACTICAL APPLICATIONS
OF MODFLOW
Robert S. Kerr Environmental Research Center
Ada, Oklahoma
by
Bradley M. Hill, R.G.
Computer Data Systems, Inc.
-------
SUBSURFACE MODELING COURSE
National Risk Management Research Laboratory
Robert S. Kerr Environmental Research Center
Ada, Oklahoma
Practical Applications of MODFLOW
August 13-16,1996
by
Bradley M. Hill, R.G.
Computer Data Systems, Inc.
I. MODFLOW Overview
Objective: Head (water level), flow rates, velocities
Input: Hydraulic parameters, initial & boundary conditions, stresses
Hydraulic Parameters: transmissivity, hydraulic conductivity (KH & Kv), specific yield,
storativity
Ouput: Head (water level)
Modular format (BASIC, BCF, STR, WEL...)
Finite-difference numerical approximation
Block-centered (Head calculated at center of model cells)
Heterogeneous, anisotropic, 2D, 3D or quasi-3D
Layers must be continuous throughout model domain
Conductance: Darcy's law for flow between cells: (q) = KA * dh
81
where: K = hydraulic conductivity A = cell area
81 = length of flow path 8h = change in head across flow path
• Numerical approximation, equations are solved through an iterative process using various
methods: SIP, SSOR, PCG2
• Time and Space Descritization:
- Stress Period: length of time where all model input stresses are constant
- Time Steps: length of time to solve heads within each stress period
- Closure (error) criterion: max. head difference between each time step
(typically lOx-lOOx smaller than desired level of accuracy)
- Grid design: cell dimensions and grid orientation
\MODCIASS\MODFIX>W2.DOC
-------
Types of Boundary Model Cells
- Type 1: Specified-Head (Dirichlet): where head is specified in advance (e.g.,
lake, river, ocean)
- Type 2: Specified-Flux (Neumann): where flux is specified in advance. A no-
flow boundary is where the flux is specified to be zero (e.g., underflow,
impermeable hydrogeologic barrier).
- Type 3: Head-Dependant (Cauchy): where the flux across the boundary is
calculated given a boundary head value (e.g., river, drains, springs, ET)
II. Discussion of Selected MODFLOW Modules
A. Basic (BAS)
overall model setup and execution, time discretization, starting heads, output
control for printing.
B. Block Center Flow (BCF)
calculates terms of finite-difference equations, space discretization (i.e., grid
dimensions), aquifer geometry (top and bottom elevations), aquifer properties
(i.e., horizontal and vertical hydraulic conductivity, storage).
C. Recharge (RCm
simulates areally distributed infiltration from precipitation, lakes, agriculture, etc.
Requires infiltration rate per unit area.
D. Wells
simulates features such as wells that withdraw water (or add to it) at specified rate
at specified locations both areally and vertically.
t
E. River (RIV> / Streamflow Routing (STR1
simulates surface water/aquifer interactions, river level elevations, river bottom
elevation and stream bed conductance. Calculates stream elevations based on
stream dimensions (rectangular ) and Manning's coefficient, accounts for stream
flows including tributaries and diversions.
F. Solver (SIP)
method of solving simultaneous linear differential equations.
VMODCLASSMHODFLOWIDOC
-------
ni. PRE and POST PROCESSORS for MODFLOW
A. Examples of Pre Processors
• GMS (U.S. Department of Defense)
• Visual MODFLOW (Waterloo Hydrogeologic)
• Modelcad (Geraghty & Miller)
• MODIME (S.S. Papadopulos & Assoc.)
• Groundwater Vistas (Environmental Simulations, Inc)
• MR (U.S.Geological Survey)
B. Examples of Post Processors
• "Postmod"
• Surfer™ (contouring software)
• Zonebud (cell-by-cell flow terms)
• Hydrograph Package (Plato, 1993)
IV. FUNDAMENTALS OF DEVELOPING A MODFLOW MODEL
(after Anderson and Woessner, 1992, ASTM standards, Zheng & Bennett, 1995)
1. Purpose and Objective of Model
2. Develop Conceptual Model of Hydrogeologic System
A. Define hydro-stratigraphic units
B. Identify boundary conditions
C. Define water budget inflow and outflow components
D. Water level data both areally and vertically (if appropriate)
E. Identify hydrologic stresses on system
3. Model Design
A. Model grid size & orientation, units of length & time
B. Initial conditions, boundary types (e.g., specified-head, specified-flux)
C. Aquifer parameters, hydrologic stresses (descritization of the physical system)
D. Definition of simulation stress periods and time-steps
E. Steady-state or Transient-state
4. Model Calibration (iterative process)
A. Methods:
• Trial and Error
• Automated (MODINV or MODFLOWP)
B. Traditional Measures of Calibration:
• Comparison of measured versus simulated heads (e.g., plan view
or well hydrographs)
\MODCLASS\MODFLOW2.DOC
-------
• Statistical analysis - residuals
• Comparison of measured versus simulated water budgets
C. Establish Calibration Targets
• Pre-determine maximum head or water balance error
(common rule: 10% of total head loss over model domain)
5. Sensitivity Analysis of Calibration
A. Purpose to define uncertainty within model solution both areally and vertically
B Establish the effects of uncertainty on the model solution by varying input
parameters over reasonable range, typically 1 parameter at time.
C. Commonly presently with some statistical analysis, for example, residual mean
error (ME), residual mean absolute error (MAE), or residual root mean squared
error (RMSE)
6. Verification (if possible)
A. Data permitting, use calibrated model to reproduce a second set of data
7. Prediction
A. Quantify a response on the hydrologic system to possible future stresses (e.g.,
pump and treat remediation alternatives)
B. Addition of contaminant transport simulation/particle tracking capabilities (e.g.,
MT3D, MODPATH, PATH3D)
V. INTRODUCTION TO USING MODFLOW
(Using USEPA Instructional Manual Problems, EPA/600/R-93/010)
Example Problem No. 4: Steady-state
VI. AVAILABLE MODFLOW MODULES
Basic (BAS1
overall model setup and execution, time discretization, starting heads, output control for printing.
Block Center Flow (BCF)
calculates terms of finite-difference equations, space discretization (i.e., grid dimensions), aquifer
geometry (top and bottom elevations), aquifer properties (i.e., horizontal and vertical hydraulic
conductivity, storage).
Block Center Flow fBCF2 and BCF31 (McDonald, et. al., 1991; Goode and Appel, 1992)
same as above plus permits conversion of no-flow cells to variable-head cells (e.g., re-wetting of
dry cells) and allows for a more accurate calculation of aquifers having smoothly varying
transmissivity, respectively.
VMODCLASSVMODFLOW2.DOC
-------
River (RIV)
simulates surface water/aquifer interactions, requires constant river level elevation, river bottom
elevation and stream bed conductance.
Streamflow Routing Package (SIR) (Prudic, 1989)
simulates surface water/aquifer interactions, calculates river level elevations and streamflow
within river, accounts for diversions or tributaries to streamflow, requires river geometry
dimensions including bottom elevation and width, stream bed conductance, and manning's
coefficient. Modified version; trapezoidal shape to river geometry (Catherine Kraeger-Rovey).
Rechare
simulates areally distributed infiltration from precipitation, lakes, agriculture, etc. Requires
infiltration rate per unit area.
Well (WED
simulates features such as wells that withdraw water (or add to it) at specified rate at specified
locations both areally and vertically.
Drain (DRN)
simulates features such as agricultural drains or natural springs that remove water from an
aquifer, requires drain elevation and conductance.
Evapotranspiration (EVT)
simulates the effects of plant transpiration by removing water from the groundwater table.
Requires land or water table surface, maximum ET rate, and maximum depth of ET influence.
Interbed-Storage Package (IBS) (Leake and Prudic, 1991)
simulates the storage change from elastic and inelastic compaction of aquifer materials where
specific storage is a function of head. Requires a preconsolidation head, elastic and inelastic
storage factors, and compaction values.
Transient-Leakage Package (TLKn (Leake, et. al., 1994)
simulates transient leakage in confining units between model two saturated model layers.
Requires vertical hydraulic conductivity, thickness and specific storage of confining units.
Horizontal Row Barrier (HFB) (Hsieh and Freckleton, 1993)
simulates thin, vertical low-permeability geologic features (e.g., faults or grout curtains) that
impede horizontal flow. Requires location and hydraulic characteristic of the flow barrier.
General Head Boundary (GHB)
boundary type that simulates flow into or out of a model cell. Requires a reference boundary
head and conductance.
\MODCLASS\MODFLOW2XIOC
-------
Strongly Implicit Procedure fSIP)
method of solving simultaneous linear differential equations.
Slice-Successive Overrelaxation Package (SSOR)
another method of solving simultaneous linear differential equations.
Preconditioned Conjugate-Gradient Package (PCG21 (Hill, 1991)
a method of solving simultaneous linear and non-linear differential equations.
**A11 simulations MUST have at least a BASIC, BCF and solver to run.
VII. REFERENCES
ASTM 5447-93, Standard guide for application of a ground-water flow model to a site-specific
problem: ASTM Annual Book of Standards, Volume 04.09: Soil and Rock (II).
ASTM 5490-93, Standard guide for comparing ground-water flow model simulations to site-
specific information: ASTM Annual Book of Standards, Volume 04.09: Soil and Rock
(II).
ASTM 5609-94, Standard guide for defining boundary conditions in ground-water flow
modeling: ASTM Annual Book of Standards, Volume 04.09: Soil and Rock (n).
ASTM 5610-94, Standard Guide for Defining Initial Conditions in Ground-Water Flow
Modeling: ASTM Annual Book of Standards, Volume 04.09: Soil and Rock (H).
ASTM 5611-94, Standard guide for conducting a sensitivity analysis for a ground-water flow
model application: ASTM Annual Book of Standards, Volume 04.09: Soil and Rock (II).
Anderson, M.P. and W.W. Woessner, 1992, Applied groundwater modeling, simulation of flow
and advective transport: Academic Press, 381 p.
Goode, D.J., and C.A. Appel, 1992, Finite-difference interblock transmissivity for unconfined
aquifers and for aquifers having smoothly varying transmissivity: U.S. Geological
Survey, WRI92-4124, 79 p.
Harbaugh, A.W., 1994, A data input program (MFI) for the U.S. Geological Survey modular
finite-difference ground-water flow model: U.S. Geological Survey, OFR 94-468, 24p.
Hill, M.C., 1991, Preconditioned conjugate-gradient 2 (PCG2), a computer program for solving
ground-water flow equations: U.S. Geological Survey, WRI 90-4048, 43p.
Hsieh, P.A. and J.R. Freckleton, 1993, Documentation of a computer program to simulate
horizontal-flow barriers using the U.S. Geological Survey's modular three-dimensional
finite-difference ground-water flow model: U.S. Geological Survey, OFR 92-477, 32 p.
Leake, S.A., Leahy, P.P. and A.S. Navoy, 1994, Documentation of a computer program to
simulate transient leakage from confining units using the modular finite-difference
ground-water flow model: U.S. Geological Survey, OFR 94-59,70p.
XMODCLASS\MODFLOW2JX)C
-------
Leake, S.A. and D.E. Prudic, 1991, Documentation of a computer program to simulate aquifer-
system compaction using the modular finite-difference ground-water flow model: U.S.
Geological Survey, Techniques in Water Resources Investigations, Book 6, Chapter A2,
68p.
McDonald, M.G., Harbaugh, A.W., 1988, A modular three-dimensional finite-difference ground-
water flow model: U.S. Geological Survey, Techniques in Water-Resources
Investigations, Book 6, Chapter Al
McDonald, M.G., Harbaugh, A.W., Orr, B.R., and D.A. Ackerman, 1991, A method of
converting no-flow cells to variable-head cells for the U.S. Geological Survey modular
finite-difference ground-water flow model: U.S. Geological Survey, OFR 91-536,99 p.
Plato, P.R., 1993, Hydrograph Package for MODFLOW (SHYDG.FOR): Journal of
Groundwater, Volume 31, Number 6, Nov-Dec. 1993, pp. 1025-1028
Prudic, D.E., 1989, Documentation of a computer program to simulate stream-aquifer relations
using a modular, finite-difference, ground-water flow model: U.S. Geological Survey,
OFR 88-729,113 p.
Zheng, C., and G.D. Bennett, 1995, Applied contaminant transport modeling, theory and
practice: Van Nostrand Reinhold, 440p.
VMODCLASS\MODFLOW2JX>C
-------
MODFLOW
Modular Format
(BASIC, BCF, WEL, RIV...)
Finite-Difference Numerical Approximation
(Iterative Solver)
Block-Centered
(Head calculated at center of cell)
#&
-------
MODFLOW
HETEROGENEOUS
ANISOTROPIC
3-DIMENSIONAL
(VCONT option)
MODEL LAYERS MUST BE
CONTINUOUS
-------
MODFLOW
DARCY'S LAW
(Conductance)
DISCRETIZATION
Time
Space
MODEL BOUNDARY CELL TYPES
Type 1. Specified-Head (Dirichlet)
Type 2. Specified-Flux (Neumann)
Type 3. Head-Dependant (Cauchy)
-------
DARCY'S LAW
q =
Flow in Between Cells
CONDUCTANCE
KA
*9H
3L
Surface / Groundwater
i_
1
y////,
^-^c
Model Cell
Conductance
-------
DISCRETIZATION
TIME
STRESS PERIOD: Time Where All Input is
Constant (e.g., Seasonal)
TIME STEPS: Time to Solve Heads within
each STRESS PERIOD
CLOSURE CRITERION: Maximum Head Difference
between each Time Step
(Iterative Solution)
lOx - lOOx ««« Desired Level of Accuracy
-------
DISCRETIZATION
TIME
O»t
O
CO
Stress . Stress . Stress
Period 1
V-r tJL V U U • K^ l-A XX U U
I Period 2 I Periods
Time Time Time Time Time Time Time
Step 1 Step 2 Step 1 Step 1 Step 2 Step 3 Step 4
O
O
a e
CO
-------
DISCRETIZATION
SPACE
(MODEL GRID DESIGN)
CELL DIMENSIONS: Fine v. Coarse
GRID ORIENTATION: One Axis MUST be Aligned
with Principal Direction of the
Hydraulic Conductivity Tensor
-------
WATER LEVEL ELEVATIONS
?^"?>*""f™%!!«
-------
MODEL GRID
-------
MODEL BOUNDARY CELL TYPES
Specified-Head (Type i)
(e.g., constant = lake, river, ocean)
Specified-Flux (Type 2)
(e.g., no-flow = impermeable hydrogeologic boundary)
(e.g., underflow = known flow rate)
Head-Dependant Flow (Type 3)
(e.g., river, drains, springs, etc.)
-------
MODFLOW MODULES
BASIC
• Overall Model Setup & Execution
*
• Cell Type Designation
• Time Discretization
• Starting Heads
• Output Control for Printing
-------
MODFLOW MODULES
BCF
(Block Centered Flow)
Calculates Terms for Finite Difference Eq.
Space Discretization (Column & Row width)
Layer Designation (Confined, Unconfined)
Aquifer Geometry
(Top & Bottom Elevations)
Aquifer Properties
(K., K , S )
n v y
-------
VCONT
THK.
THK,
VCONT =
2
THK
\/
1
THK, THK2
K
+
VI
JL lk.
V2
-------
MODFLOW MODULES
WELL
Simulates Pumpage or Injection
Specified Rate (L3 / T) per Cell
Location: Horizontally (Cell) and
Vertically (Layer)
Layer 1
Layer 2
-------
MODFLOW MODULES
RIVER / STREAM-ROUTING
Surface / Groundwater Interaction
River Stage and Bottom Elevations
Streambed Conductances (Kv, Dimensions)
Calculates Stage Elevations, Permits River to go Dry
Accounts for Streamflows
Diversions and Tributaries
-------
A
Stream-Aquifer System
Low Permeability
Streambed Material
Cell Boundary
MODFLOW Approximation
Impermeable •
Walls
River Surface (HRTV)
Head
in Cell (H)
v^* ,", , Low Permeability
Material '
-\
Cell Boundary
-------
Pre & Post Processors
Pre I CMS (Dept. of Defense)
Visual MODFLOW (Waterloo Hydrogeologic)
ModelCAD (Geraghty & Miller)
MODME (SSP&A)
Groundwater Vistas (ESI)
MFI (U.S.G.S.)
Post! "Postmod" = converts binary files to XYZ
SURFER™
Zoriebud (U.S.G.S.)
Hydrograph Package (Plato, 1993)
••'"
-------
FUNDAMENTALS
MODFLOW MODELING
1. PURPOSE & OB JECTIVES
A. Dictates Model Design
B. Calibration Targets
C. Data Collection
2. CONCEPTUAL MODEL
A. Hydro-Stratigraphic Units
B. Boundary Conditions
C. Define Water Budget
D. Water Level Data (Horizontally & Vertically)
E. Hydrologic Stresses (e.g., Pumpage, ET, Irrigation ...)
-------
YUMA AREA
— - INTERNATIONAL ft STATE BOUNDARY UK
RIVERS
VALUY/MESA ESCARPMENT
1A6UNA DAM
CRYSTAUJJE HARDROCX AREAS
AREAS ELEVATED ABOVE FLOOOPIAIN
ALFORNA
PILOT KNOB
HACIENDA
ESTATES
YUMA VALLEY
-------
GILA MOUNTAIN
PEDIMENT SLOPE
YUMA VALLEY
YUMA MESA
Groundwater
200-
Clayey-silt Layer
^'-:-;-;''-;->'-v^
Model Layer 3
Coarse-Gravel and
Upper Wedge Zones
Model
loyer 4
10
15
Vertical Exaggeration 170:1
DISTANCE (MILES)
20 25 30
NOTE: Geology east of groundwater mound inferred for
modeling purposes
-------
RIVERS
VALLEY/MESA ESCARPMENT
LACUNA DAM
CRYSTALLJNE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
YUMA
AGENDA
ESTATES
ALAMD DEL NDRT
I
SOUTHERN MODEL BOUNDARY SHDVN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
-------
Large, Unlined Canal
Water Table
Recharge from
Irrigated Agriculture
Small, Unlined Canal
V
Clayey-silt layer \
Lower, Coarse-gravel zone
note: drawing not to scale
-------
CONCEPTUAL WATER
BUDGET 1983-1989
DATA
INFLOW
AG RECHARGE
CANAL SEEPAGE
RIVER LOSS TO AQUIFER
UNDERFLOW
TOTAL
OUTFLOW
PUMPAGE
DRAINS
RIVER GAIN FROM AQ.
ET
UNDERFLOW
TOTAL
CHANGE IN STORAGE
395,000
218,000
205,000
2,000
820,000
400,000
30,000
200,000
35,000
125,000
790,000
+30,000
-------
RIVERS
VALLEY/MESA ESCARPMENT
LACUNA DAM
CRYSTALLINE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
GROUNDWATER FLOW DIRECTION
LACUNA
'////,
MTN
PILOT KNOB
-.-.• ox,- • Y.Y.Y • • """"lAWfc^iim; •
• • • • O -v*i.- ^V_;.4
i»» ^»;i»«ri>*ri»lirW
ESTATES '
YUMA VALL
NDTEi SOUTHERN MODEL BOUNDARY SHOWN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
mxs
-------
FUNDAMENTALS
MODFLOW MODELING
3. MODEL DESIGN
A. Grid Size & Orientation
B. Units of Length & Time
C. Initial Conditions, Boundary Types
D. Aq. Parameters, Hydrologic Stresses
(discretization of Physical System)
E. Stress Period-Time Step Length
R Steady-state or Transient-state
-------
10
COLUMNS
19 30 40 50 56
YU.MA AREA
GROUND WATER FLOW
MODEL
ARIZONA-CALIFORNIA-MEXICO
FIGURE 30
Model Cell Types and Locations
CONSTANT HEAD CELLS
MODEL BOUNDARY
RIVERS
VALLEY/MESA ESCARPMENT
HARD ROCK AREAS
J
sou
HUNBMrWU
-------
FUNDAMENTALS
MODFLOW MODELING
4. Model calibration (iterative process)
A. Methods
- Trial & Error
- Automated (MODFLOWP)
B. Measures of calibration
- Simulated v. measured heads
(plan view - qualitative)
(hydrographs - quantitative)
- Statistical analysis of residuals
- Water balance components
i
C. Calibration targets
- Pre-determined
- Head error (maximum difference)
(common rule: 10% head loss)
- Water balance error
-------
Calibration is an Iterative Process
Define Objectives
Develop Conceptual Model
Develop Mathematical Model
Develop Numerical Model
i
Select Boundary Conditions and
Hydraulic Parameters
i
Run Simulation
Compare Results with Observations
4
no
Acceptable Match ?
y yes
Apply Model
-------
100
WELL ID # 12 S 10 W
T — i — i — i — i — i — i — c— i — i — i — i
i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i
95-
90-
85-
I
80-
75-
70
SUBREGION: YUMA VALLEY, GROUND ELEVATION « 96.5 FEET
ROW - 64, COL - 10, LAYER 3
SIMULATED WATER LEVELS - SOLID LINE (MARCH, SEPT.)
MEASURED WATER LEVELS « DASHED LINE (MARCH, JUNE, SEPT., DEC.)
i — r
T10S R25W See I4dd
i i i r T T r i
78 79 80
• t
82 I 84 85
MEASUREMENT DATE
86
87
88
-------
T15S
CAUFORNIA CADASTRAL
R21E R22E
T7S
R23W
ARIZONA CADASTRAL
YUMA AREA
GROUND WATER FLOW
MODEL
ARBONA-CAIJFORNIA-MEXICO
FIGURE 40
MAP OF CALIBRATION ERROR
POST-COLORADO RIVER FLOODING
MODEL LAYER 4
(ABOUJTE VAUUt)
mmmm
HARD ROCK AREAS
MODEL BOUNDARY
RIVERS
VALLEY/MESA ESCARPMENT
£3 GREATER THAN 5 FT.
0 - 5 FT.
I
ecus
-------
CALIBRATION
MODEL-WIDE HEAD RESIDUALS
Absolute Difference (feet)
LAYER MEAN MEDIAN STD. DEV
1
2
3
4
6.9
3.4
3.4
3.8
5.0
2.0
3.0
3.0
6.8
4.2
3.4
3.8
ZONED HEAD RESIDUALS
ZONE 1 (Yuma Mesa)
LAY1 4.8 3.6 4.0
ZONE 2 (Yuma Valley)
LAY 2 0.9 0.9 1.1
LAY 3 1.1 1.0 1.2
-------
10
19 30
COLUMNS
40 50 58
1-
10-
YUMA AREA
GROUND WATER FLOW
MODEL
ARIZONA-CALIFORNIA-MEXICO
FIGURE 36
Zones for Calibration and
Sensitivity Analysis
ZONES
MODEL BOUNDARY
RIVERS
VALLEY/MESA ESCARPMENT
HARD ROCK AREAS
I
ecus
-------
WATER BALANCE COMPARISON
CONCEPTUAL V. MODEL
INFLOW Concept.
AG 395,000
CANAL 218,000
RIVER LOSS 205,000
STRM LEAK
UNDERFLOW 2,000
CONSTANT HD.
OUTFLOW
PUMPAGE 400,000
DRAINS 30,000
RIVER 200,000
STRM LEAK
ET 35,000
UNDERFLOW 125,000
CONSTANT HD
A STORAGE +30,000
Model
380,000
376,000
2,000
397,000
245,000
44,000
140,000
+31,000
Error
-4%
-11%
-1%
+ 7%
+ 26%
+ 12%
+ 3%
-------
COMPARISON
CONCEPTUAL V. MODEL
SURFACE WATER
RIVERS
COLORADO
GILA
Concept.
842,000
339,000
Model
825,000
337,000
Error
-2%
-1%
CANALS
ALL-AMERICAN
GILA GRAVITY
EAST MAIN
2,139,000
166,000
45,000
2,225,000
165,000
44,000
+ 4%
0%
-2%
DRAINS
YUMA MAIN
YUMA MESA
75,000
32,000
77,000
32,000
+ 3%
'0
-------
FUNDAMENTALS
MODFLOW MODELING
5. SENSITIVITY Analysis
A. Purpose - Define Uncertainty both Areally & Vertically
B. Establish Effects of Uncertainty
C. Vary Input Components over Range
D. Statistical Analysis
6. VERIFICATION
A. Data Permitting
B. Use Calibrated Model
C. Reproduce Second Set of Data
-------
FUNDAMENTALS
MODFLOW MODELING
7. Scenarios
A. Prediction of future
B. Develop "realistic" picture
C. Quantify a response to future stresses
(e.g., pump & treat remediation)
D. Addition of contaminant transport/particle tracking
(e.g., MT3D, PATH3D, MODPATH)
-------
RIVERS
VALLEY/MESA ESCARPMENT
LACUNA DAM
'/A CRYSTALLINE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
Boundary
fie Foot
Groundwate
Level
Decline
YUMA
*•* ^^fvfmmmmm'm
WELLS DV14 AND DV15
LOCATED NEAR
SUBDIVISION
Boundary of
Groundwater
VALLEY
NDTEi SOUTHERN MODEL BOUNDARY SHOWN HERE FDR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
liTTJS
-------
RIVERS
VALLEY/MESA ESCARPMENT
LACUNA DAM
CRYSTALUNE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
LAGUNA
'////>.
MTN
Boundary of One-Half Foot
Groundwater Level Decline
PILOT KNOB
LYUMA
Boundary of One Foot
Groundwater Level Decline
YUMA VALLEY
S
SOUTHERN MODEL BOUNDARY SHOWN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
-------
RIVERS
VALLEY/MESA ESCARPMENT
LACUNA DAM
CRYSTALLINE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
LACUNA
'////,
MTN
Boundary of One Foot
Groundwater Level Decline
PILOT KNOB
Boundary of Five
Foot Groundwater
Level Delclne
A:
«" \rmmmmmmm
Boundary of Seven
Foot Groundwater
Level Decline
YUMA VALLEY
NDTEi SOUTHERN MODEL BOUNDARY SHOWN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
HOES
-------
LACUNA DAM
CRYSTALLINE HARDROCK AREAS
AREAS ELEVATED ABOVE FLOODPLAIN
LACUNA
'/////
MTN
Quechan Indian
Reservation
LYUMA1
Boundary of One Foot
Groundwater Level Sr
Decline
Boundary of Nine Foot
Groundwater Level
Decline
Boundary of Five Foot
Groundwater Level Decline
YUMA VALLEY
SOUTHERN MODEL BOUNDARY SHDVN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH.
MODEL BOUNDARY
-------
Use of a Numerical Model for Management
of Shallow Ground-Water Levels
in the Yuma, Arizona Area
by Bradley M. Hfll'
Abstract
The Yuma area has experienced agricultural development since the late 1890s and ground-water levels have risen over 70 feet
due to recharge from heavy application of irrigation water, unlined canals and flooding along the Colorado and Gila Rivers. The
resulting shallow water levels have seriously impacted residential areas and prime agricultural land.
The Arizona Department of Water Resources in conjunction with Yuma County Flood Control District developed a regional
three-dimensional ground-water flow model of the Yuma area. The purpose of the model is to assist local agencies in evaluating
.remedial water management alternatives to mitigate the shallow ground-water level problems.
The model domain incorporates over 900 mi2 of Arizona, California, and Mexico and simulates ground-water pumpage, deep
percolation from agricultural irrigation, evapotranspiration from phreatopbytes and flow in 12 canals, 16 drains, and the Colorado
and Gfla Rivers. The model contains four layers with over 30,000 model cefls ranging in size from 40 acres to 640 acres.
Different model scenario simulations were conducted to evaluate the effectiveness of proposed water management alternatives
on lowering ground-water levels within the northern portion of Yuma Valley. These scenarios include lining a portion of the East
Main canal and pumping two drainage wells, lining the Ail-American canal, and simulating a decrease in deep percolation from
agricultural irrigation on the Yuma Mesa and northern portion of Yuma Valley.
Introduction
The Yuma area is located geographically at the downstream
end of the Colorado River basin and is comprised of portions of
the United States and the Republic of Mexico (Figure 1). This
region has experienced agricultural development since the late
1890s and ground-water levels have risen since the early 1900s.
Ground-water levels have risen due to the heavy application of
irrigation water and leakage from unlined canals, creating shal-
low water levels which have waterlogged residential areas and
prune agricultural land. Ground-water levels beneath the Yuma
Mesa have risen over 70 feet since the beginning of this century
(Olmsted et al., 1973). The first of many open drainage canals
was constructed in 1916 with drainage wells being first installed
in the early 1920s in an attempt to lower the shallow water levels
(lakisch and Sweet, 1948).
The Colorado and Gila Rivers flow through the Yuma area
and are extensively regulated by upstream dams (Figure 1). The
rivers flooded the region between 1983 through 1986, which
compounded the existing shallow ground-water level problem.
Yuma County Flood Control District approached the Arizona
Department of Water Resources (ADWR) in 1988 to assist in
addressing the problems associated with rising ground-water
levels.
'Arizona Department of Water Resources, 500 N. Third St.,
hoenix, Arizona 85004 (presently with Computer Data Systems Inc.,
Robert S. Kerr Environmental Research Laboratory, P.O. Box 1198,
Ada, Oklahoma 74820).
Received November 1993, revised June 1994, December 1994, and
April 1995, accepted May 1995.
The purpose of this investigation was to develop a regional
three-dimensional ground-water flow model that would be use-
ful for Yuma County Flood Control District in its evaluation of
remedial water management alternatives to mitigate the shallow
ground-water level problems. The objective was to effectively
simulate the regional hydrologic regime and test various water
management scenarios to determine which approach would be
the most effective in lowering the shallow water levels in northern
Yuma valley near the Hacienda Estates subdivision (Figure 1).
Hydrogeologic Regime
The Yuma area located within the Sonoran desert is charac-
terized by northwest-trending, elongated, low, rugged moun-
tains consisting of low-permeability crystalline rocks which are
separated by extensive broad desert plains. The broad alluvial
plains are comprised of thousands of marine and nonmarine
sediments (Olmsted et al., 1973).
Three lithologic water-bearing zones were identified within
the upper several hundred feet of sediments by the U.S. Geologi-
cal Survey (Olmsted et al., 1973). These layers are extremely
transmissive and consist of basin-filling fluvial and deltaic sedi-
ments deposited by the Colorado and Gila Rivers. From lowest
to uppermost, these are the wedge, lower coarse-gravel, and
upper, fine-grained zones. A generalized geologic cross section
illustrates the relationship between each of these zones (Figure
2). The Yuma valley floor is comprised of a thin layer of clayey
silt which overlies the upper, fine-grained and lower, coarse-
gravel zones. This clayey silt layer is laterally continuous with a
clay layer beneath the Yuma Mesa identified informally by
Olmsted et al. (1973) as Clay B. This clay layer is thought to play
an integral part in creating the ground-water mound that exists
VoL 34, No. 3—GROUND WATER—May-June 1996
397
-------
Fie. '• Yuma area location.
beneath the Yuma Mesa by inhibiting the vertical movement of
ground water. The sediments above Clay B are comprised of
coarse sand and fine gravel which permit the vertical movement
of irrigation water downward.
Hydraulic characteristics of each of these water-bearing
zones were quantified and several multiple-well aquifer pumping
tests were conducted on the upper, fine-grained zone as part of
this study. Horizontal hydraulic conductivity was estimated to
range between 50 and 500 ft/day for the upper, fine-grained zone
and a maximum of 1,300 ft/day for the lower, coarse-gravel zone
(Hill, 1993). The cumulative transmissivity of the upper, several
hundred feet of saturated sediments ranges between zero at the
basin margins to over 140,000 ft2/day in the valley portions of the1
Yuma area (Olmsted et al., 1973).
A detailed analysis of the hydrology and water resources
was conducted for the Yuma area by Hill (1993). Ground-water
underflow enters the region between the Gila and Laguna Moun-
tains and exits the United States into the Republic of Mexico to
the west beneath the Colorado River and south across the
Southern International Border. Regionally, ground water flows
from the north to south-southwest, except beneath the Yuma
Mesa where ground water flows radially away from a prominent
ground-water mound (Figure 3). This ground-water mound has
been created since the early 1900s due to agricultural irrigation
on the sandy soils on Yuma Mesa (lakisch and Sweet, 1948).
The Yuma area has an extremely complex surface-water
system consisting of the Colorado and Gila Rivers, 12 primary
canals including the large unlined Ail-American, Gila Gravity
Main and Yuma Main, and the smaller unlined East and West
Main canals (Figure 4). An extensive network of 16 primary
open drainage canals and approximately 70 drainage wells were
constructed throughout the Yuma area in an attempt to lower
water levels within the agricultural areas. Ground-water flow is
locally influenced by these features. Figure 5 conceptually illus-
trates the interrelationship between the rivers, canals, open
drains, and drainage wells with the ground-water system. This
figure illustrates the Colorado River as a "gaining" river. During
flood flows the river becomes a "losing" river contributing signif-
icant amounts of water to the ground-water system.
GILA MOUNTAIN
PEDIMENT SLOPE
YUMA VALLEY
YUMA MESA
Model Layer 2
Voter Table
Clayey-silt Layer
i
, S^fSO*
Model Layer 3
Coarse-Gravel and
Upoer Vedqe Zones
-40
10
15
Vertical Exaggeration 170:1 DISTANCE (MIL£S)
Generalized geologic cross section of the Yunu are*.
NOTE: Geology not of groundwofer mound Inferred for
modeling purposa
398
-------
__.
noon«iD woo 011. cm
atom KVWL <*a nmi a
MODEL BOUNDARY
MODEL BOUNDARY
Fij. 3. Measured {round-water elevations for 1989.
Fif. 4. Primary canals within Yiuna area.
Data collected to quantify the relationship of each surface-
water feature with the ground-water system included monthly
flow gaging records for all surface-water features, reported seep-
age losses from canals within each irrigation district, as-built
construction characteristics for each primary canal and drain,
and cross-sectional profiles of the Colorado River. All of these
data were used during the development and calibration of the
numerical model.
Water Budget Components
The primary components of inflow to the ground-water
system in the Yuma area are recharge from irrigated agriculture,
fcanal seepage, river losses, and underflow from adjacent basins.
•Tlie primary components of outflow from the ground-water
system include pumpage, aquifer discharge to drains and rivers,
evapotranspiration from phreatophyte plant growth, and under-
flow into the Republic of Mexico. Data were compiled and
analyzed for the time period between 1978 and 1989.
Recharge from agricultural irrigation for each irrigation
district was estimated using monthly reported water applied to
farms, annual total acres of each crop type grown, and estimated
consumptive use for each crop type. Water application rates
averaged 4.3 acre-feet/acre in the valley regions and almost 12
acre-feet/acre on Yuma Mesa. This difference is primarily
Large, Unlined Canal
Recharge from
Water Table Irrigated Agriculture
Small, Unlined Canal
note drawing not to scale
Fij. 5. Conceptual illustration of the surface-water/ground-water
interrelationship in the Yuma area.
attributed to the different soil types between the two regions.
Yuma Mesa is comprised primarily of coarse sand and fine gravel
while the valley regions are composed of clayey silt and fine sand
(Olmsted et al., 1973). Estimates of annual recharge volumes also
vary depending on location and averaged 128,000 acre-feet/year
(1.3 acre-feet/acre) for the valley regions and 151,000 acre-
feet/year (7.7 acre-feet/acre) for Yuma Mesa between 1978 and
1989 (Hill, 1993).
Canal seepage is another major component of inflow to the
ground-water system. Annual canal seepage losses for each irri-
gation district are reported to the U.S. Bureau of Reclamation.
Annual seepage volumes for each canal within the Yuma area
were estimated to range from 112,000 acre-feet /year for the
unlined Ail-American canal (Olmsted et al., 1973) to approxi-
mately 43,000 acre-feet/year for the combined total from the
unlined, smaller East and West Main canals (Hill, 1993).
When the Colorado and Gila Rivers are at flood stage, they
can also be another major component of inflow to the ground-
water system. However, the relative accuracy of stream gaging
stations along each river limited the quantification of the hydro-
logic connection between the rivers and ground-water system.
Gaging stations along the Colorado and Gila Rivers are reported
to be accurate within 10-15% (Boner et al., 1991; Curt Webb,
U.S. Geological Survey, personal communication). The average
annual river loss to the aquifer was estimated at 154,000 acre-
feet/year which is well within the error of the accuracy of each
gaging station and was used with some uncertainty.
Ground-water underflow from adjacent basins is a minor
component of inflow to the Yuma area and was estimated at
approximately 3,400 acre-feet/year between the Laguna and
Gila Mountains beneath the Colorado and Gila Rivers (Olmsted
et al., 1973).
A major component of ground-water outflow is from pump-
age for agricultural and municipal water supply, agricultural
drainage, or domestic purposes. The U.S. Bureau of Reclama-
tion and U.S. Geological Survey have historically kept detailed
records of individual well pumpage within the United States
portion of the Yuma area (U.S. Bureau of Reclamation, 1991).
Total annual average pumpage between 1978 and 1989 is approx-
imately 266,000 acre-feet/year in the United States while Mexico
reported an average of 642,000 acre-feet/year for the same time
period (I.B.W.C., 1989).
399
-------
Aquifer discharge to open drains is another component of
ground-water outflow and varies widely depending upon drain
location and size. The greatest discharge to an open drain occurs
to the Yuma Main drain located in Yuma Valley and to the
Reservation Main drain located between the Ail-American canal
and the Colorado River in California. Annual average aquifer
discharge to the drains was estimated at 150,000 acre-feet/year
and ranges from a low of approximately 5,000 acre-feet/year for
the North Gila drain located near the Laguna Mountains to a
maximum of 99,000 acre-feet/year for the Yuma Main drain
(Hill, 1993).
Under normal flow conditions the Colorado and Gila
Rivers gain water from the underlying aquifer and are therefore
a major component of outflow from the ground-water system
(Figure 5). However, as previously mentioned, the relative accu-
racy of stream gaging stations along each river limits the quanti-
fication of the hydrologic connection between the rivers and
ground-water system.
Evapotranspiration (ET) from phreatophytes was also con-
sidered as a major outflow from the ground-water system. ET
was estimated using areal plant survey conducted by Younker
and Anderson (1986) and an estimated consumptive use for each
plant type (U.S. Department of Agriculture, 1982). The annual
ET between Morelos and Laguna Dams was estimated to range
between 33,000 and 45,000 acre-feet/year (Hill, 1993).
Underflow out of the United States flows west from the
Yuma Valley into Mexico and flows south from Yuma Mesa into
Mexico (Figures I and 3). Underflow into Mexico for the upper
several hundred feet of saturated sediments was calculated to
average 39,000 acre-feet/year to the west and 102,000 acre-
feet/year to the south (Hill, 1993).
Numerical Modeling
A regional model was constructed using the U.S. Geologi-
cal Survey computer code MODFLOW (McDonald and
Harbaugh, 1988). This model simulated ground-water flow from
April 1978 through March 1989. The active model domain
encompasses 900 square miles including portions of Arizona,
California, Sonora, and Mexicali Valley, Mexico. The model has
approximately 30,000 model cells distributed among four layers,
each layer simulating a distinct hydrogeologic unit. Model cells
range in size from 40 acres to 640 acres. The east side of the
model is bounded by the crystalline Gila and Laguna Mountains
which create a hydrologically impermeable boundary. The
northern boundary is controlled by seepage from the All-
American canal which has altered ground-water flow creating a
constant source of underflow into the Yuma area since its com-
pletion in 1940. The southern and western boundaries were
arbitrarily selected to be located several miles into Mexico away
(OVERS
VAUIY/MESA ESCARPMENT
CRYSTAUJNC HAROROCK AREAS
AREAS ELEVATED ABOVE FIOOOPLAM
WELLS DVI4 AND IMS
LOCATED NEAR
SUBDIVISION
Boundary of
Groundwatw
NOTE' SOUTHERN MODEL BOUNDARY SHOWN HERE FOR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH
MODEL BOUNDARY
MM9
Fig. 6. Simulated ground-water level change after four yean of lining the East Main Canal and pumping drainage wells DW14 and DW15.
400
-------
HVERS
VAUCT/HESA ESCARPMENT
CRYSTAUMC HARDROCK AREAS
AKAS EUVATED ABOVE FUOOftAM
Boundary of Om Fool
Groundwafcr Lml D*dM
Boundary of Fhw
Foot Graundwatar
Boundary of Stvon
Fool Groundwatv
Lml
YUMA VALLEY
NOTE> SOUTHERN MOKL BOUNDARY SHDVN HERE FDR
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL HUES FURTHER SOUTH
MODEL BOUNDARY
tig. 7. Simulated ground-water level change after four years of reducing agricultural recharge on the Yuma Mesa.
from the primary area of interest in northern Yuma Valley and
simulate ground-water underflow out of the model.
Model Layer One corresponds to the coarse-grained sedi-
ments beneath the Yuma Mesa and the Ail-American canal.
Layer Two corresponds to the thin silt and clay layer that
composes the river valley floor and day B beneath the Yuma
Mesa. Model Layer Three corresponds to the upper, fine-
grained zone beneath the clayey silt layer throughout the Yuma
area and Layer Four is a combination of the lower, coarse-gravel
zone and the upper portion of the wedge zone (Figure 2). Limited
data exist to the east of the ground-water mound beneath Yuma
Mesa, and the hydrogeology was inferred for modeling purposes.
To simulate the complex surface-water system, an extensive
application of the streamflow-routing package was employed.
The streamflow-routing package was used to simulate 12 canals,
16 drains, and the Colorado and Gila Rivers using 70 segments
and over 800 reaches. The streamflow-routing package devel-
oped by Prudic (1989) permits leakage to or from the aquifer
from each surface-water feature, accounts for streamflow within
each stream reach including any diversions or additions to flow
from tributaries, and calculates stream stage elevations within
^ach stream reach. A primary assumption in Prudic (1989) is
leakage to and from the aquifer from each surface-water feature
is computed using Darcy's Law as follows:
where Q is the leakage to or from the aquifer through the
streambed (L3/T), Hs is the head in the stream (L), H A is the head
in aquifer side of streambed (L), and CSTR is the conductance of
the streambed (L2/T), which is the hydraulic conductivity of the
streambed times the product of the width of the stream and its
length divided by the thickness of the streambed. A second
primary assumption is the method of accounting for streamflow,
which assumes that flow in a stream reach is instantly available in
downstream reaches for each stress period in the model simula-
tion. This assumption is reasonable since ground-water flow
rates are typically slow relative to changes in streamflow. A third
primary assumption is how stream stage is calculated for each
stream reach. Prudic (1989) uses the Manning equation and
assumes incompressible steady flow in the stream at constant
depth and a rectangular channel where the width is much wider
than depth. Prudic (1989) presents the equation to compute
stream depth as follows:
Q = CSTR(HS - HA)
(1)
(2)
where d is the depth of the water in the stream (L), Q is the
leakage to or from the aquifer through the streambed (L3/T), n is
Manning's roughness coefficient (dimensionless), C is a constant
(Ll/3/T), which is 1.486 for cubic feet per second and 1.0 for cubic
401
-------
meters per second, w is the width of the channel (L), and S is the
slope of the stream channel per model cell (L/L).
The model was calibrated for steady-state and transient-
state ground-water flow conditions. The Summer of 1978 (Le.,
April through September) was assumed to be representative of
steady-state conditions because the volume of ground water in
storage remained relatively constant between 1970 and 1978
since flows in the Colorado River and ground-water pumpage
were relatively constant (Hill, 1993). The primary model input
component calibrated was aquifer transmissivities which was
accomplished by varying the magnitude and distribution of
hydraulic conductivity throughout the model domain.
The transient-state model was calibrated in two parts: pre-
Colorado River flooding (October 1978 through March 1983)
and Colorado River flooding and post-flooding (April 1983
through March 1989). The temporal nature of inflows and out-
flows (e.g., river flows, water applied for irrigated agriculture and
pumpage) dictated the need to simulate winter and summer
seasonal stresses. The primary input component calibrated dur-
ing transient-state simulations was vertical hydraulic conductiv-
ity of the streambed for each surface-water feature in the
streamflow-routing package. The calibration included, in part,
matching gaged stream flows to simulated flows within each
river, canal, and drain, matching conceptual fluxes to and from
the aquifer between each of the surface-water features and
matching actual well hydrographs from observation wells with
simulated water levels. This was an important aspect of the
calibration process since the success of modeling the Yuma area
was dependent on effectively simulating the complex surface-
water/ground-water interrelationship.
Results of the transient-state calibration indicate that the
model replicates measured water level elevations, ground-water
flow directions, and surface-water flow measurements very well.
The model simulates flow in the rivers, canals, and drains within
1% to 10% of conceptual estimates based on stream gaging data
and the model simulates fluxes to and from the aquifer within
10% to 30% of conceptual estimates. The model also simulated
ground-water levels within an average of one foot as compared
to measured water levels near the Hacienda Estates subdivision
and an average of five feet beneath the Yuma Mesa (Hill, 1993).
Model Scenario Simulations
Four model scenario simulations were conducted to deter-
mine which water management alternative would be most effec-
tive in lowering the shallow water levels within the northern
portion of Yuma Valley near the Hacienda Estates subdivision.
These simulations included: (1) lining four miles of the East
Main canal and pumping the new drainage wells DW 14 and
RIVERS
VAUEY/MESA ESCARPMENT
CRYSTAUJNE HARDROCK AREAS
AREAS ELEVATED ABOVE FIOOOPIAK
Boundary of Om-Nalf Fool
Groundwatcr Uvil DtcfiM
Boundary of Orw Foot
Groundwator Lml Dwlnt
YUMA VALLEY
NOTE' SOUTHERN MODEL BOUNDARY SHOVN HERE FTK
PRESENTATION PURPOSES ONLY. ACTUAL BOUNDARY
EXTENDS SEVERAL WLES FURTHER SOUTH
MODEL BOUNDARY
Fit. 8. Simulated ground-water level chance after four years of reducing agricultural recharge in Yuma VaOey.
402
-------
OVERS
VAU£Y/MESA ESCARPMENT
CTOTAUK HARDROCK AREAS
AREAS E1TVATED ABOVE FLMOTIAM
Quochon Indian
Boundary of Ono Fool
Boundary of MM Foot \S Greundvattr Uwl
Groundvahr Lml f DwlM
Oodno
Boundary of Fin Foot
Groundwattr LM( Dodlno
YUMA VALLEY
NOTEi SOUThCRN MODEL HUNDARY StCVN ICRE FDR
PRESENTATION PURPOSES OLY. ACTUAL BOUNDARY
EXTENDS SEVERAL MILES FURTHER SOUTH
MODEL BOUNDARY
MTM
9. Simulated {round-water level change after four years of lining the All-American Canal within the Yuma area.
DW IS located near the subdivision; (2) reducing deep percola-
tion recharge from agricultural irrigation by 25 percent within
the northern portion of Yuma Valley; (3) reducing deep percola-
tion recharge from agricultural irrigation by 25 percent on the
Yuma Mesa; and (4) lining the All-American canal within the
Yuma area.
The results of lining four miles of the East Main canal and
pumping the new drainage wells D W14 and DW 15 adjacent to
the Hacienda Estates had a significant impact on lowering
ground-water levels. Leakage from the East Main canal was
simulated to decrease from 2,800 acre-feet/year to zero (i.e., no
seepage loss) while the combined pumpage from the two drain-
age wells was simulated at 8,200 acre-feet/year. This scenario
predicted that water levels would decline a maximum of six feet
near the subdivisions with a one foot decline radiating outward
for several miles after four years (Figure 6). This simulation
predicted that pumping the drainage wells would greatly over-
shadow any effects from lining the canal since the wells remove
almost three times more water from the ground-water system
that would have leaked from the unlined canal.
Reducing deep percolation from intensive agricultural irri-
•mtion by 25 percent also had a significant impact on lowering
water levels near the Hacienda Estates subdivision. Alternative
examples of reducing recharge from irrigated agriculture may be
increased farm efficiencies, decreased fanned acres due to
urbanization, or change in crop types away from water intensive
vegetables. Reducing recharge by 25 percent on Yuma Mesa
resulted in a decrease of 50,000 acre-feet/year of inflow to the
ground-water system. This scenario predicted water levels would
decline adjacent to the subdivision over one to two feet, while
water levels on the ground-water mound beneath Yuma Mesa
decline over seven feet after four years (Figure 7).
Reducing recharge by 25 percent in the northern portion of
Yuma Valley resulted in a decrease of 3,600 acre-feet/year of
inflow to the ground-water system. This scenario predicted water
levels would decline by approximately one-half foot after four
years near the subdivision with isolated declines of over one foot
(FigureS).
The two scenarios that address agricultural recharge illus-
trate the potential influence agricultural irrigation has on
ground-water levels within the Yuma area, especially the impacts
from the ground-water mound. However, the model could not
accurately simulate the maximum localized effects from deep
percolation recharge since the area! location of each crop type
and the water applied could not be identified precisely.
Lining the All-American canal within the Yuma area had a
significant impact on lowering water levels on the north side of
the Colorado River near the Quechan Indian Reservation in
California. Seepage from the canal was reduced from 112,000
acre-feet/year to zero (i.e., no seepage loss), and ground-water
403
-------
Table 1. Comparison Between Each Water Management Alternative Simulated Water Level and Water Budget Change
Between 1989 and 1997 for the Northern Portion of Yuma Valley ;•_
Water management scenario
A. Lining 4 miles of East Main Canal and
pumping drainage wells DW14 and DW15
B. Reducing recharge by 25% on Yuma Mesa
C. Reducing recharge by 25% in Yuma Valley
D. Lining All-American Canal
Maximum water
level change
-6.1 feet
-2.4 feet
-1.0 feet
-1.6 feet
Average water
level change
-1.8 feet
-0.7 feet
-0.5 feet
-0.5 feet
*<*:
V
£f:
Water budget
change
11.000
50,000
3,600
112,000
'Total simulated reduction of inflow to ground-water system (acre-feet/year).
levels were predicted to decline up to nine feet adjacent to the
canal and range between one and live feet within the reservation
area after four years (Figure 9). This simulation predicted that
water levels also would decline in the northern portion of Yuma
Valley a maximum of less than two feet after four years. The
impacts from lining the canal on ground-water levels south of the
Colorado River in northern Yuma Valley can be attributed to
ground water flowing southward beneath the river since the river
only partially penetrates the aquifer in the Yuma area.
Table 1 presents a comparison of the maximum and average
ground-water level decline between measured 1989 and simu-
lated 1997 water levels for the northern portion of Yuma Valley.
Table 1 also presents a comparison of the simulated water budget
change for each simulation.
Conclusions
Shallow ground-water levels that exist in the Yuma area are
a cumulative result of several factors which include deep percola-
tion from intense agricultural irrigation, seepage from unlined
canals, and to a lesser degree the short-term effects from the
historical flooding of the Colorado River. Each of these factors
has some impact on creating shallow ground-water levels in
northern Yuma Valley near the Hacienda Estates subdivision.
Utilization of a numerical model as a tool in evaluating the
impacts of various water management alternatives was very
useful in permitting the ability to predict the cumulative effect of
each alternative on the ground-water and surface-water systems
and in identifying the greatest effects on lowering ground-water
levels. The extensive application of the streamflow-routing
package was successful in simulating the complex interaction
between surface-water features and the ground-water system.
The model predicted that the water management alterna-
tive of pumping drainage wells DW14 and DW15 and lining four
miles of the East Main canal would have the greatest impact on
lowering ground-water levels in the vicinity of the Hacienda
Estates subdivision. However, the sumulation also predicted that
pumping the drainage wells would greatly overshadow any
effects from lining the canal. In addition, the model predicted
that reducing deep percolation from agricultural irrigation
within the Yuma Valley and on Yuma Mesa had a significant
impact on lowering ground-water levels.
Acknowledgments •; , „.
I would like to recognize Frank G. Putman and Dale, A.
Mason with the ADWR Hydrology Division, Roger Schoenherr
with Yuma County Flood Control District;and Earl Burnett and
Fred Croxen from the U.S. Bureau of Reclamation for their
assistance and direction regarding this project. Stan A. Leake
and David E. Prudic from the U.S. Geological Survey should
also be recognized for their technical oversight and assistance in
designing the application of the streamflqw-routing package ;to
the complex surface-water system in the Yuma area. . • -,
References
Boner, F. C., R. G. Davis, and N. R. Duet. 1991. Water Resources
Data, Arizona, Water Year 1991. U.S. Geological Survey Water-
Data Report AZ-91-1. 411 pp.
Hill, B. M. 1993. Hydrogeology, Numerical Model and Scenario Simuj
lations of the Yuma Area Groundwater Flow Model, Arizona!
California and Mexico. Arizona Depl. of Water Resources—
Hydrology Division, Modeling Report No. 7. .113 pp.
lakisch, J. R. and C. L. Sweet. 1948. Report on Drainage, Valley
Division—Yuma Project, Arizona. U.S. Department of Interior
and U.S. Bureau of Reclamation. February 1948,55 pp.
I.B.W.C. 1989. Comision NacionaJ del Agua, Bistrito de Riego No. 14,
Rio Colorado. International Boundary fiid Water Commission-
United States and Mexico, Department^ State. > 'V-
McDonald, M. G. and A. W. Harbaugh. 1988. A modular three-
dimensional finite-difference ground-crater flow model. UiS.
Geological Survey Techniques of Water-Resources Investiga-
tions Book 6, Chapter A1. '•'••" ' '•
Olmsted, F. H., O. J. Loeltz, and B. Irelan. 1973. §eohydrology of (he
Yuma area, Arizona and California. U.S. Geological Survey
Professional Paper 486-H. 227 pp., 17 pfates.
Prudic, D. E. 1989. Documentation of a computer program to simulate
stream-aquifer relations using a modular, finite-difference,
ground-water flow model U.S. Geological Survey, Open-File
Report 88-729. 113 pp. :
U.S. Bureau of Reclamation. 1991. Ground-Water Status Report*-
1989. Yuma Area—Arizona, California. U.S. Bureau of Reclp-
mation, Yuma Projects Office, Yuma, Arizona. December 1991.
U.S. Department of Agriculture. 1982. Conservation Research Report,
No. 29. U.S. Department of Agriculture. May 1982,40 pp. •..
Younker, G. L. and G. W. Anderson. 1986. Mapping methods and
vegetation changes along the lower Colorado River between
Davis Dam and the border of Mexico. AAA Engineering and
Drafting Inc., Salt Lake City. Contract No. 6-C5-30-38000.
404
-------
WhAEM
-------
8. Ground Water Flow Modeling
with the Wellhead Analytical
Element Model (WhAEM)
-------
Center for Subsurface Modeling Support
Robert S. Kerr Environmental Research Laboratory
.S. Environmental Protection Agency
J.O. Box 1198
Ada, Oklahoma 74820
USEPA Region VI
1445 Ross Avenue
12th Floor Suite 1200
Dallas, Texas 75202-2733
Introduction to the
Wellhead Analytic Element Model
August 14,1996
Ada, Oklahoma
-------
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/R-94/210
December 1994
SEPA
W/7AEM: Program
Documentation for the
Wellhead Analytic
Element Model
-------
W/?AEM:
PROGRAM DOCUMENTATION for the
WELLHEAD ANALYTIC ELEMENT MODEL
by
Hendrik M. Haitjema
i Jack Wittman
Vic Kelson
Nancy Bauch
School of Public and Environmental Affairs
Indiana University, Bloomington
CR 818029
Project Officer
Stephen R. Kraemer
Processes and Systems Research Division
Robert S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 74820
-------
NOTICE
The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency (USEPA) under assistance agreement number CR-818029 to
Indiana University. It has been subjected to the Agency's peer and administrative review, and it
has been approved for publication as an USEPA document.
All research projects making conclusions or recommendations based on environmentally related
measurements and funded by the USEPA are required to participate in the Agency Quality
Assurance Program. This project did not involve environmentally related measurements and did
not involve a Quality Assurance Plan.
The material introduced in this document should be fully understood prior to the application of
the computer codes in WfcAEM to field problems. Both the creation of the conceptual model and
the interpretation of the program's output require an understanding of the Analytic Element
Method and its implementation in WAAEM. Interpretation of the output generated by the
programs is the sole responsibility of the user.
The software described in this document is available on an "as-is" basis without guarantee or
warranty of any kind, expressed or implied. Neither the United States government (USEPA,
Robert S. Kerr Environmental Research Laboratory), Indiana University, the University of
Minnesota, nor any of the authors accept any liability resulting from the use of the codes.
Mention of trade names or commercial products does not constitute endorsement or
recommendations for use.
-------
FOREWORD
USEPA is charged by Congress to protect the Nation's land, air, and water systems. Under a
mandate of national environmental laws focused on air and water quality, solid waste
management, and the control of toxic substances, pesticides, noise, and radiation, the Agency
strives to formulate and implement actions which lead to a compatible balance between human
activities and the ability of natural systems to support and nurture life.
The Robert S. Kerr Environmental Research Laboratory is the Agency's center for expertise for
investigation of the soil and subsurface environment. Personnel at the Laboratory are responsible
for management of research programs to: (a) determine the fate, transport, and transformation
rates of pollutants in the soil, the unsaturated zones of the subsurface environment; (b) define the
processes to be used in characterizing the soil and subsurface environments as a receptor of
pollutants; (c) develop techniques for predicting the effect of pollutants on ground water, soil,
and indigenous organisms; and (d) define and demonstrate the applicability of using natural
processes, indigenous to the soil and subsurface environment, for the protection of this resource.
Computer modeling has many uses in environmental research and development, including:
representing our degree of understanding of subsurface processes in comparison to field and
laboratory observations; and assisting in the design and predicting the performance of aquifer
remediation and protection strategies. Occasionally, a new numerical solution technique comes
along that is particularly well suited for environmental application. This research report explores
the application of the analytic element method for the modeling of capture zones of pumping
wells, and in particular in the design of wellhead protection areas. The Wellhead Analytic
Element Model (W/iAEM) should be considered a demonstration of a promising technology.
Clinton W. Hall, Director
Robert S. Kerr Environmental
Research Laboratory
-------
ABSTRACT
The Wellhead Analytic Element Model (WfiAEM) demonstrates a new technique for the
definition of time-of-travel capture zones in relatively simple geohydrologic settings. The
\WtAEM package includes an analytic element model that uses superposition of (many) analytic
solutions to generate a ground-water flow solution. W/iAEM consists of two executables: the
preprocessor GAEP, and the flow model CZAEM. WAAEM differs from existing analytical
models in that it can handle fairly realistic boundary conditions such as streams, lakes, and
aquifer recharge due to precipitation.
The preprocessor GAEP is designed to simplify input data preparation; specifically it facilitates
the interactive process of ground-water flow modeling that precedes capture zone delineation.
The flow model CZAEM is equipped with a novel algorithm to accurately define capture zone
boundaries by first determining all stagnation points and dividing streamlines in the flow domain.
No models currently in use for wellhead protection contain such an algorithm.
-------
TABLE OF CONTENTS
CHAPTER 1. INTRODUCTION 1
System Requirements 2
Hardware Requirements 1 3
3
Software Requirements 3
Installation Procedure 3
Installing WhAEM on Your Hard Disk 3
Digitizer Configuration 4
Printer Configuration 4
WhAEM as a DOS application in Windows 5
CHAPTER 2. MODELING GROUND-WATER FLOW 6
Steady State Dupuit-Forchheimer Flow 6
Steady State Flow 6
Dupuit-Forchheimer Flow 7
Analytic Element Method 7
Hypothesis Testing 8
CHAPTER 3. WhAEM TUTORIAL 10
Load and Start the GAEP Program 11
An Example Digital Map 12
Prepare a Base Map for Digitizing 15
Four Step Mark-Up 16
Digitize Features from the Vincennes Quad 17
Associate Map Coordinates with Digitizer Coordinates 17
Digitize Two Sections of a Single Stream 18
Digitize the Elevations 18
Digitize Point Sets for Wells and Other Features 19
Digitize Roads and City Limits 20
View the Digitized Features 20
File Operations in GAEP 21
Edit Features in GAEP 21
View the Data -'.' 22
Join Features 22
Create an Input File for CZAEM 23
Creating Analytic Elements 24
Creating Line-sinks: 26
Creating Wells 27
Aquifer Module 28
Saving the CZAEM Input Data on Disk 31
Imaging Line-sinks Generated in GAEP 31
Model the Site with CZAEM 36
-------
Read CZAEM Input File: VINCENNE.DAT 37
Check the Input File 37
Turning map on and off 39
Changing the Window 39
Solve the Ground-Water Row Problem 40
Generate Grid for Contouring Heads 41
View Piezometric Contours in PLOT Module 41
Use Pathline Tracing in TRACE Module 42
Initial Capture Zone Analysis 42
Hypothesis Testing 44
Read in New File: VINIMAGE.DAT 45
Create Final Capture Zones 45
Generate a Subzone Around the Well 46
Generate Isochrones Around the Well 47
Sending Graphics to the Printer 49
Exit CZAEM 51
REFERENCES 52
APPENDIX A. GAEP REFERENCE 54
Concepts 55
Files 55
Digital Map Files 55
Analytic Element Files 55
Digital Map Features 56
Stream Features 56
Curve Features 56
Point Set Features 56
Measurement Units 57
CZAEM Units 57
UTM Coordinates 57
Conversion of Latitude-Longitude to UTM Coordinates 57
Coordinate Origins 57
Digitizer Origin 58
Model Origin 58
Digitizer Origin 58
Construction of No-Flow Boundaries 59
Use of GAEP 60
Menus 60
Special Keys 61
Procedure for Using GAEP 62
Create the Digital Map (Digitize menu) 62
Save the Digital Map (File menu) 62
Create Analytic Elements (Element menu) 62
Save the Analytic Element Data File (File menu) .'62
-------
Detailed GAEP Command Descriptions 62
File Menu 63
Aquifer Menu 65
Digitize Menu 67
Digitize/Edit Submenu 70
Element Menu 71
Options Menu 73
Utility Menu 74
UTM / Latitude-Longitude Utility 75
APPENDIX B. CZAEM REFERENCE 78
CZAEM Main Module 78
AQUIFER 82
GIVEN 83
REFERENCE 84
WELL 85
LINESINK 86
CHECK 88
GRID 90
PLOT 92
TRACE 92
CAPZONE in 96
CURSOR 101
PSET 105
STOP 106
APPENDIX C. TABLET CONFIGURATION GUIDE 107
Introduction 107
Installation of GAEP 107
Digitizer Configuration 107
How Do I Configure My Digitizer for GAEP? 108
General 108
Step-By-Step 108
Digitizer Protocols 110
Formatted ASCII Protocol 110
SummaGraphics MM Binary Protocol Ill
SummaGraphics MM ASCII Protocol 112
SummaGraphics Bit Pad Plus Protocol 112
Microsoft Mouse 113
Keyboard Data Entry (For Systems Without Digitizers) 113
Program TABTEST 114
Commands 115
Tested Configurations for Various Digitizers 118
Appendix C Bibliography 120
Appendix C Acknowledgments 120
-------
ACKNOWLEDGMENTS
The authors express their appreciation to the volunteer group of "guinea pigs" who did the beta
testing of the W/zAEM codes. The codes and documentation were improved after anonymous
technical review, as well as the many hours of review by graduate students at the Indiana University
Groundwater Modeling Lab. Chursey Fountain of the Kerr Lab provided editorial comments and
review.
IX
-------
CHAPTER 1. INTRODUCTION
W/zAEM is a steady state ground-water flow modeling package designed to delineate capture zones
and isochrones of ground-water residence times for the purpose of "wellhead protection." The
package has been designed under USEPA assistance (CR # 818029) between the Robert S. Ken-
Environmental Research Laboratory, Indiana University, and the University of Minnesota. The
principal investigators were:
Hendrik M. Haitjema (Associate Professor) Otto D. L. Strack (Professor)
School of Public and Environmental Affairs Department of Civil and Mineral
Indiana University Engineering
Bloomington University of Minnesota
Minneapolis
W/fAEM consists of two independent executables:
• GAEP (Geographic Analytic Element Preprocessor) developed at Indiana University
• CZAEM (Capture Zone Analytic Element Model) developed at the University of Minnesota
The foundation of the package is CZAEM, which solves the ground-water flow problem and
generates the desired capture zones and isochrones of ground-water residence times. CZAEM is a
single-layer analytic element model for steady state (regional) ground-water flow modeling (USEPA,
1994). The analytic element method employs superposition of elementary analytic solutions to
ground-water flow features (Strack and Haitjema, 1981, Strack, 1989). The CZAEM program is an
extension of the public domain code SLWL and is designed specifically to solve ground-water flow
problems near well fields. New in CZAEM is the logic to automatically generate capture zones and
isochrones of ground-water residence times (Bakker, Strack, in preparation), as well as the
generation of isochrones for "contaminant fronts" using a new expanded transport equation (Strack,
1992).
The program GAEP handles data management for CZAEM (Kelson et al., 1993). Data preparation
with GAEP is accomplished in two steps:
1) Create a digital map of hydrography using one or more U.S. Geological Survey (USGS)
topographic maps. This procedure is greatly facilitated by use of a digitizer.
2) Create or edit an input data file for CZAEM using the digital map as a template for model
building.
-------
The purpose of GAEP is to ease the burden of input data preparation, in particular, to facilitate the
editing of input data to improve CZAEM's ground-water flow solutions. Figure 1 shows a cartoon
of the W/jAEM assisted modeling process. Creating and editing CZAEM input data files with
GAEP is done graphically on screen by use of a mouse and appropriate key strokes. By making the
process of modifying ground water models more efficient, GAEP allows the user to quickly evaluate
a site several different ways. This iterative process of modeling is described in Chapter 2 of this
document as the "hypothesis testing" approach to understanding ground-water flow.
GAEP
WhAEM
Model Building
hypothesis
testing
j'-Ftow
, ^ Modeling
hypothesis
forming
CZAEM
Figure 1 The modeling process using W/jAEM.
System Requirements
The ground-water flow modeling system, W/zAEM , has minimum hardware and software
requirements described below. The ground-water flow model CZAEM has been written in Lahey
Fortran1 and runs on IBM-PC compatible personal computers in extended memory. The
geographically-oriented preprocessor, GAEP, is written in Borland C/C-H2 and runs in conventional
memory.
Lahey is a trademark of Lahey Computer Systems, Inc.
2 Borland is a trademark of Borland International
2
-------
Hardware Requirements
80386- or 80486 based PC
At least 1.5 megabytes of available extended memory
At least 2 megabytes of available hard disk space
Numeric Data Processor (except when using a 80486DX system)
Microsoft, PS/2, or compatible mouse device
Digitizer (optional)
Printer (optional)
Software Requirements
• PC-DOS or MS-DOS Version 5.0 or higher
• DOS extender bound with CZAEM. EXE (provided)
• ANSI. SYS driver (supplied with DOS)
• Windows 3.1 (optional)
Installation Procedure
See the file README on the distribution diskette for the most current instructions.
Installing W/iAEM on Your Hard Disk
The installation procedures for W/zAEM are facilitated by the INSTALL.EXE program on the
distribution disk. The installation program will locate potential target drives on the system, install
the products in a target directory specified by the user and, when necessary, make any necessary
changes to your CONFIG.SYS and AUTOEXEC.BAT files.
• To install, place the distribution disk in drive A: or B: and switch your logged drive to the
floppy drive:
C:\> A:
A;\>
or, if using drive B:,
C:\> B:
-------
• Next, run the installation program:
A:\> INSTALL
or
B:\> INSTALL
The INSTALL program will determine which available drives can be used for the installation and
request an installation directory (default is C:\WHAEM). Once all the files are unpacked, you will
have the opportunity to let INSTALL make changes to your AUTOEXEC.BAT and CONFIG.SYS
files. The changes to these files can be made directly, in which case the old versions will be backed
up, or changes may be saved to a new file and implemented in AUTOEXEC.BAT and CONFIG.SYS
by the user. If changes are required, W/iAEM will not operate properly until they are made and the
system is rebooted with the changes made, in order for them to take effect.
Digitizer Configuration
Geographic data entry into ground water models is accomplished by transferring map data into
digital form. This can be done manually, or through the assistance of a digitizing tablet or digitizer.
GAEP facilitates both approaches. Configuration of the tablet is described in detail in Appendix
C, "Tablet Configuration Guide." A utility program TABTEST is provided to simplify the process
of setting up a digitizer to use with GAEP. The INSTALL program places a batch file
TABSETUP.BAT in the \WHAEM installation directory. This file contains the DOS commands
required to configure the digitizer. Program TABTEST allows the user to adjust the parameters for
the digitizer driver allowing use of a standard digitizing tablet as the digitizing device. When the
parameters are set properly, exit TABTEST to save the parameters to TABSETUP.BAT. In
Appendix C, the supported tablet configurations are documented, including instructions for several
popular digitizer models.
Printer Configuration
The printer is configured using a batch file. From the DOS prompt:
To create postscript files:
• Type: "printer portrait" This sets the CZAEM initialization up to send
graphics to a postscript file (POST.PS) with ^.portrait
orientation.
• Type: "printer landscape" This sets the CZAEM initialization up to send
graphics to a postscript file (POST.PS) with a
landscape orientation.
-------
To print directly on an HP laser III compatible printer:
• Type: "printer hplaser" This sets CZAEM initialization up so that graphics
will be printed on an HP laserjet III.
Another option for those without a Postscript printer is to use a public domain Postscript interpreter,
such as Ghostscript, which is available by anonymous ftp over the Internet (for example, address:
ftp.cica.indiana.edu[ 129.79.26.27], directory: pub/pc/win3/util/gs*.zip).
W//AEM as a DOS application in Windows
Both programs, GAEP and CZAEM run as DOS applications under windows, with some rare,
system-specific limitations. Run this way there is a familiar set of tools available to Windows users
for capturing graphics screens, text, and other data generated while modeling.
-------
CHAPTER 2. MODELING CROUD-WATER FLOW
When using W/iAEM to delineate "time-of-travel capture zones" for well fields, you are engaging
in ground-water flow modeling, which is an art just as much as it is a science. The opinions about
the value of ground-water flow modeling are diverse: some profess an almost religious belief in
model predictions; others call them a fraud. We believe that ground-water flow modeling can be a
valuable tool to gain insight into the often complex subsurface flow processes which are otherwise
hidden from the eye. The success of ground-water flow modeling, however, depends less on the
"model" than on the professional skills and creativity of the hydrogeologist or engineer .who performs
the modeling. In designing W/zAEM, we have tried to lower the barriers that can make ground-water
flow modeling a slow, costly, and frustrating experience. We have not tried to create an "expert
system" which claims to make human expertise obsolete.
There are several books that deal specifically with ground-water flow modeling (e.g. Bear and
Verruijt, 1987 and Anderson and Woessner,1992). We assume that you are familiar with the basics
of ground-water modeling in general, but not necessarily with the analytic element method employed
by W/zAEM. Therefore, we are providing a brief overview of some basic concepts that underlie the
useofW/zAEM.
Steady State Dupuit-Forchheimer Flow
Real world aquifer systems are usually far too complex to reproduce in a computer model.
Consequently, several simplifications are made which are designed to make ground-water flow
modeling feasible while maintaining the essential characteristics of the real world flow regime. Two
of the most important simplifications are discussed below.
Steady State Flow
Although in reality the ground-water flow patterns vary in time, W/zAEM deals only with steady
state flow. Consequently, the modeling results reflect average conditions, which, depending on
circumstances, may differ significantly from the actual conditions at any one time. Yet, there are
important reasons to limit the initial study to steady state ground-water flow modeling, such as:
• Transient models require significantly more input data than steady state models (e.g. aquifer
storativity, initial conditions).
• The transient modeling procedure is much more complex than steady state modeling.
Consequently, it requires more expertise and more resources.
-------
• Many more field data are required to properly calibrate a transient model as compared to a
steady state model.
Instead of attempting transient ground-water flow modeling for delineating time-of-travel capture
zones, it seems more practical to model average conditions. Deviations from average conditions,
such as summer conditions and winter conditions, may be estimated by steady state modeling using
summer and winter data, respectively. Such steady state representations of summer or winter
conditions tend to overestimate the real transient solution; they bracket the actual transient flow
patterns. In this manner, some insight is obtained into the relative importance of transient effects
without engaging in a full blown transient modeling exercise.
Dupuit-Forchheimer Flow
In principle, ground-water flow is three-dimensional in nature. On a regional scale, however,
horizontal flow is found to be far more important than vertical flow components (Dupuit,! 863 and
Forchheimer, 1886). In W/zAEM, we employ the Dupuit-Forchheimer assumption, ignoring
resistance to vertical flow. This assumption means that the piezometric head in the aquifer does not
vary with depth, although vertical components of flow are still possible and may be estimated from
continuity of flow (Strack, 1984). It appears that the Dupuit-Forchheimer assumption is acceptable
as soon as boundary conditions in the aquifer are more than two aquifer thicknesses apart, or when
areal recharge zones are more than two or three aquifer thicknesses in size (Haitjema, 1987).
Otherwise, a fully three-dimensional analysis is warranted (Haitjema, 1985).
Limiting our modeling efforts to steady state Dupuit-Forchheimer flow greatly reduces the amount
of field data required and enhances the efficiency of the modeling process. When done properly,
most of the limitations that result from our simplifications can be overcome, at least in view of our
objective: delineating time-of-travel capture zones. However, before discussing modeling
procedures we will briefly introduce the analytic element method.
Anal/tic Element Method
The analytic element method was developed at the end of the 1970's by Otto Strack at the University
of Minnesota (Strack and Haitjema, 1981). For a detailed description of the method refer to
Groundwater Mechanics (Strack, 1989). A brief review of the method follows.
This new method avoids the discretization of a ground-water flow domain by grids or element
networks. Instead, only the surface water features in the domain are discretized, broken up in
sections, and entered into the model as input data. Each of these stream sections or lake sections are
represented by closed form analytic solutions: the analytic elements. The comprehensive solution
to a complex, regional ground-water flow problem is then obtained by superposition of all analytic
-------
elements in the model.
Traditionally, superposition of analytic functions was considered to be limited to homogeneous
aquifers of constant transmissivity. However, by formulating the ground-water flow problem in
terms of appropriately chosen discharge potentials, rather than piezometric heads, the analytic
element method becomes applicable to both confined and unconfined flow conditions as well as to
heterogeneous aquifers (Strack and Haitjema, 1981b).
The analytic elements are chosen to best represent certain hydrologic features. For instance, stream
sections are represented by line-sinks; lakes or wetlands are represented by areal sink distributions.
Streams and lakes that are not fully connected to the aquifer are modeled by area sinks with a bottom
resistance. Discontinuities in aquifer thickness or hydraulic conductivity are modeled by use of line
doublets (double layers). Specialized analytic elements may be used for special features, such as
drains, cracks, slurry walls, etc.
Hypothesis Testing
The ground-water flow solution obtained with W/zAEM depends on many parameters, such as
aquifer hydraulic conductivity, aquifer depth, areal recharge, and the interaction between the ground-
water flow regime and ditches, streams, lakes, and wetlands. Many of these parameters are, in
general, not very well known. In addition, there are real world complexities which cannot be
included in our simplified model, such as local inhomogeneities, variations in recharge, aquifer
stratification, etc. Instead of trying to remove uncertainties in parameterization through extensive
data acquisition efforts prior to the modeling, we suggest a procedure of hypothesis testing.
The strategy is to model time-of-travel capture zones for various bounding values of the uncertain
parameters, for instance, the lowest and highest hydraulic conductivities reported from pumping
tests, or the low and high values on recharge, porosity, and pumping rates (summer and winter). The
procedure accomplishes two things: (1) it will become apparent which parameters most affect the
time-of-travel capture zones, and (2) a selection of time-of-travel capture zones for various extreme
parameter choices may be overlain so that an envelope of time-of-travel capture zones can be
constructed which incorporates uncertainties in the data.
The parameter sensitivity issue may be illustrated by the following example. Assume that the
inclusion of a particular stream hi the solution causes a major shift in the capture zone of the well
field. In that case it may not be appropriate to include the stream hi the envelope of time-of-travel
capture zones. Instead, it is necessary to determine what the actual role of that stream is in the flow
system. The model may provide clues in this case. For instance, if the stream is indicated on the
USGS quad map as ephemeral (dash-dot line), while the model predicts that it loses large amounts
of water to the aquifer, it is doubtful whether it is in contact with the aquifer. If the stream or ditch
is known to carry water year round, it very well may be a boundary condition for flow in the regional
aquifer. Hypothesis testing allows you to distinguish between those data which can be accepted and
-------
incorporated in the modeling results and those that require further study. In this manner, your
modeling guides data acquisition and analysis activities, and not only vice versa.
The details of the step-wise approach to modeling with W/zAEM, shown in Figure 2, will be
introduced through a tutorial in the next chapter.
START
MAP PREPARATION
DIGITIZE MAP
FEATURES
EDIT & JOIN
FEATURES
ASSOCIATE
ELEVATION WITH
FEATURES
SAVE DIGITAL MAP
FILE
CREATE CZAEM INPUT
FILE
RUN CZAEM USING
INPUT FILE CREATED
BY GAEP
<-HYPO-
THESIS
TESTING
EVALUATE SOLUTION
RUN GAEP TO ADJUST
INPUT FILE FOR
CZAEM
HYPO-
THESIS
TESTING
CREATE FINAL
CAPTURE ZONES
Figure 2 Hypothesis testing with W/zAEM.
-------
CHAPTER 3. WMM TUTORIAL
This chapter of the manual is a tutorial for the operation of the Geographic Analytic Element
Preprocessor (GAEP) software package and the Capture Zone Analytic Element Model (CZAEM)
software package for ground-water flow studies.
Data prepared for a wellhead protection demonstration project in the City of Vincennes, Indiana, will
be used to illustrate the operation of W/zAEM. The tutorial data set includes rivers, streams, and
landmark features such as the well locations, municipal boundaries, and roads. Working through
the following exercises should acquaint the new user with many of the key features of the programs
and the basic concepts involved.
The purpose of the tutorial is to illustrate steps in delineating capture zones with W/zAEM. These
steps are:
1. Create a digital map file from a base map.
2. Use the digital map file to create an input data file for program CZAEM.
3. Read this file into CZAEM and solve the ground-water flow problem.
4. Generate graphical output and interpret modeling results.
5. Conduct hypothesis testing.
6. Delineate capture zones.
The reader is advised to work through this tutorial using a computer. We strongly suggest that as
questions come up about specific commands you also take the time to read the CZAEM and GAEP
reference guides in the appendices.
As discussed in Chapter 1, the two programs GAEP and CZAEM are in some important respects
very different. The descriptions provided in this manual reflect that difference. Moving through the
commands in GAEP is done either by clicking on the command with the mouse or typing in the first
letter of the command from the keyboard. Throughout this chapter, the following convention is
followed:
• Keys to be pressed are enclosed in angle brackets, e.g., for the enter key.
• Text to be typed is shown in quotation marks.
• System prompts are in underlined bold face courier font, e.g., C;\WHAEM>.
CZAEM is a command line program which requires direct input of commands using the keyboard.
General keyboard control procedures for moving around within the GAEP program are shown in
Figure 3. We will start with the creation of a digital map file in GAEP.
10
-------
Function-key control-
- context sensitive help.
- accept the data as entered.
- quit this menu (or exit at the main menu).
Cursor control -
Be aware that the cursor is moved by either the digitizer puck or the
mouse depending on the digitizer settings and the type of data being
entered. Element creation and feature selection are always done with
the mouse.
ZOOM in and out of any graphics image:
= ZOOM OUT
= ZOOM IN
= PAN
Figure3 General GAEP hints.
Load and Start the GAEP Program
To start GAEP, set your working directory to the C:\WHAEM* directory. Type "GAEP" and press
the key:
C;\>CD \whaem
C;\WHAEM> gaep
The GAEP program will start, and the introductory menu will appear. Press any key to enter the
main GAEP menu. The main menu screen of GAEP is divided into two parts: an upper line listing
Throughout this tutorial, we assume that you have installed W/iAEM on the C: drive. If
otherwise, replace C: by the proper drive letter.
11
-------
the various modules of the program and the remainder of the screen, which indicates current settings.
The settings can each be changed in one of the modules of the program. Each module can be entered
by one of two ways: 1) selecting the module with the mouse by clicking the word on the menu bar
once with the left mouse button, or 2) pressing the key for the highlighted letter in the module name.
Appendix A of this document provides a complete description of all GAEP menus, settings, and
options. Figure 4 shows the main menu screen with default settings.
Zile aquifer fiigitize Element Options Utility fluit
GAEP Release 1.0
Indiana University
SPEA Groundwater Modeling Laboratory
WhAEM Version
Current Directory:
C i \WHAEM\DAT
Current Map File:
Current Element File;
Memory available: 347248
Option Settings:
Unit Conversion:
Digitizer Mode:
Video graphics mode:
M->FT
MOUSE
COLOR
help return to previous menu 07/Aug/94 01:00 PM
Figure 4 GAEP Main Menu screen.
When you run GAEP for the first time, prior to digitizer setup, the "Digitizer Mode:" will be set to
"MOUSE." If a digitizer has been connected and configured for GAEP, as described in Appendix
C, "Digitizer Mode:" will be set to "DIGITIZER." You can toggle this setting by selecting twice (DigitizerMode, Digitizer). Press to return to the main menu,
which should now look like the menu in Figure 4. Note: the "Memory Available" and the drive
letter before the \WHAEM are, of course, dependent on your particular system.
An Example Digital Map
Before creating a digital map yourself, it may be helpful to take a look at an example. A digital map
file has been prepared for this tutorial and was copied into your \WHAEM\DAT directory as part of the
12
-------
installation. The file VINCENNE.DM contains hydrologic features, roads and domestic wells (the
file extension .DM is used on all GAEP digital map files). In this session you will read the digital
map, view the contents, digitize a new stream in two separate sections, join the sections, and then
create analytic elements for that stream. If you do not have a tablet, you can skip the exercises on
digitizing.
At the main menu:
• Select the File module from the main menu, either with the mouse or the key, and the
File Submenu will appear (Figure 5).
New ReadDM griteDM LpadElem gaveElem Map C_hangeDir setDrive Quit
Current Directory:
C: \WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 347248
Option Settings:
Unit Conversion:
Digitizer Mode:
Video graphics mode:
M->FT
DIGITIZER
COLOR
help return to previous menu 07/Aug/94 01:16 PM
Figure 5 GAEP File Submenu.
Read in the tutorial digital map file:
• Select (for ReadDM).
• Type "VINCENNE" and press .
This option allows the user to read files from disk. The
program prompts for the file name.
The program displays the number of features read. The
file-name extension .DM is assumed
13
-------
• Press the key (or for Quit). Returns control to the main menu.
View the data:
• Press in the main GAEP menu.
Enter the DIGITIZE module (Figure 6).
Origin Stream Beads Curve goints J£iew Edit Quit
Current Directory:
C: \HHAEM\DAT
Current Map File:
Current Element File:
Memory available: 347248
Option Settings:
Unit Conversion:
Digitizer Mode:
Video graphics mode:
M->FT
DIGITIZER
COLOR
help return to previous menu 07/Aug/94 02:25 PM
Figure 6 GAEP Digitize Submenu.
• Press to view.
The Vincennes digital map will appear on the screen, as shown in Figure 7. Individual features can
be highlighted by moving the mouse; its color changes to yellow, and its name appears at the top
of the screen. All lines are composed of a string of x,y coordinate pairs. Light blue lines are streams
which have associated water table elevations along their reach. Dark blue lines are stream features
which have no associated elevations (not in this file). The light blue lines through the middle of the
study area are the banks of the Wabash River. Other map features are shown in red; roads are
shown as straight and curved lines; production wells and observation well locations (piezometers)
are marked by plus signs.
While viewing features in GAEP, you may change scale or shift the window on the screen as
follows:
Press
Zoom in.
14
-------
• Press
• Press the arrow keys
Zoom out.
Pan across the map.
Press ENTER to return to Menu
Figure 7. Prepared digital map of the Vincennes area.
When you are finished viewing, press to return to the DIGITIZE menu and again to
return to the main GAEP menu. Then press (for Quit) to exit the GAEP program.
Prepare a Base Map for Digitizing
For this exercise you will be learning about the general operation of the program GAEP by building
an example digital map file using a small portion of the USGS 7.5 minute Vincennes quad (included
as Plate 1). The completed Vincennes digital map file will be used later in the tutorial to build a
CZAEM input data file and to demonstrate CZAEM functions. Digitizing is greatly assisted through
use of a digitizing tablet, but can be also be done "by-hand" (see Appendix C, keyboard entry
protocol). If you would like to skip the digitizing discussion in the tutorial, proceed to section
"Create an Input File for CZAEM."
15
-------
Note:
The following exercises in the tutorial are designed to give the user practice in creating a GAEP digital map file. It is
not necessary to create a complete file during this exercise; a digital map file VNCENNE.DM is available for later use.
Four Step Mark-Up
Digitizing is tedious but important work. The best rule to apply to the task is DO IT RIGHT THE
FIRST TIME. One of the ways to assure success is to take the time to mark up the project maps
before you digitize a single point.
1) Mark the intersection of the topographic contour lines and the streams.
Label each contour-stream intersection with the contour elevation on each of the streams on
the map. For this particular exercise we will use Swan Pond Ditch, a small stream south of
the City of Vincennes (located in the SE comer of the Vincennes, Indiana, quad sheet). In
order to proceed, you should have all of the contours on this stream marked (contour levels
410- 440 ft),
2) If there is no UTMgrid on the map, draw and label this grid.
GAEP has been designed to use the Universal Transverse Mercator (UTM) coordinate
system. For more detailed information on UTMs, refer to the UTM Coordinates section of
Appendix A, the GAEP Reference. UTM coordinates are marked along the margins of the
topographic map. Unlike the other coordinates along the border of a USGS 7.5 minute map,
UTMs appear in black type adjacent to light blue tick marks at 1000 meter intervals. Use
these to draw straight lines across the map in a few locations to geo-reference the features
digitized from the map. Be sure to include eastings 4276 000m, 4280 000m, and 4281 000m,
and northings 450 000m and 456 000m. These will be used to define "digitizer origin points"
within the digitizer surface area. An alternate way to get UTM coordinates is to convert
from the latitude and longitude listed for a point on the USGS map using the GAEP
UTM/LatLong conversion option in the Utility Menu (refer to Appendix A).
3) Define a model origin for the project.
The numeric values of UTM coordinates are large because the origin is at the equator (e.g.,
x-coordinates around 456,000 m and y-coordinates around 4,280,000 m near Vincennes).
These large values may cause round-off errors in CZAEM. In order to minimize the
dimensions of the model coordinates, GAEP shifts the origin from the equator to a spot
within the area of interest. This is accomplished by defining a "Model Origin" (also referred
to as the "Local Origin"). This model origin should be used for the duration of the project.
For convenience, the model origin may be located west of the well field where UTM grid
lines cross on the map. For instance, the point (450000, 4280000) may be used which is
located just south of the Wabash River. The well field is located just south of the Wabash
16
-------
River, west of town, near the sewage treatment area (no comments, please).
4) Mark the locations and -water levels of observed piezometric heads on the map from water
levels in domestic wells, observation wells, or piezometers.
These heads can be used to calibrate the model. For this tutorial we will not be entering
these data.
Digitize Features from the Yincennes Quad
The next few sections will give you practice building a digital map file with GAEP digitizing
utilities. It is assumed that a digitizer (minimum 12 inches by 12 inches) has been connected to the
computer and tested, as described in Appendix C.
Run GAEP by typing "GAEP" at the C:\WHAEM> prompt. From the main menu enter the DIGITIZE
module:
• Press The DIGITIZE menu will appear
i Origin Stream Heads Curve Points, View Edit Quit
Associate Map Coordinates with Digitizer Coordinates
• Secure the map sheet onto the digitizer. Position the lower right sections (T2N.R10W) on the
active area of the tablet.
• Press for Origin . Defining an origin registers the map to the digitizer.
You are prompted to move the puck to the first
reference point
• Move the puck to the UTM coordinates
450000, 4276000 and press button one. A prompt will appear for the coordinates of the first
reference point.
• Type in the UTM coordinates "450000,4276000",
press . You will be prompted to move the puck to the second
reference point.
• Move the puck the UTM coordinates
456000, 4281000 and press . A prompt will appear for the coordinates ofthe second
17
-------
reference point.
• Type in "456000, 4281000" and press . The map coordinates are now entered in
GAEP, and you will automatically be retimed
to the DIGITIZE menu, ready to digitize.
NOTE:
Referencing the map to the digitizer needs to be repeated when you move the map on the digitizer, or when digitizing
a different map using new reference points.
Digitize Two Sections of a Single Stream
From the DIGITIZE Menu:
• Select Stream or type .
• At the prompt type: "Swan Pond Ditch". Press . The program responds with a default
abbreviation for the feature (SPD ). To accept
this abbreviation press .
• Place the puck at the south end of Swan Pond Ditch.
• Press button one on the puck. Repeatedly press button one as you move along the
stream reach. Each point should appear on the screen
as a dark blue plus sign, hi general these points should
be nearly the same distance apart along the stream.
Stop digitizing at the intersection of Swan Pond Ditch
and the railroad (where the stream turns sharply to the
east), press .
• If you make an error while digitizing, press
-------
GAEP as "SPD _ ", and "SPD _ A", respectively. Next we will add surface water elevation
data to these stream sections.
• Select Head or press . Digitized features appear on the screen; in our case,
only two sections of Swan Pond Ditch.
• Select the southern segment of the stream
with the mouse. Move the mouse (not the digitizer puck), so that the
first segment is highlighted. The segment will appear
yellow and its name will appear at the top of the sceen:
[SPD _ ]Swan Pond Ditch. You may want to use
the to abort.
• When the surface water elevations along that reach have been digitized, press to save
and exit the Head module.
Repeat the process for the other stream reach. Notice that when entering the graphics screen
the stream section with digitized water levels (heads) is now colored light blue.
Digitize Point Sets for Wells and Other Features
The locations of the wells are digitized from within the DIGITIZE menu:
• Select Point or press
. GAEP prompts you for the name of the feature to be
entered.
• Type: "VINCENNES WELLFIELD". Press twice. Thedigitizing screen will appear
on the monitor.
19
-------
• Place the digitizer puck on one of the well marks (the light blue circles along
the river, west of the city).
Press puck button one.
If you make an error, press to abort.
Press when you are finished.
Move the puck to another well and repeat until all five
wells have been digitized. You will see purple plus
signs on the screen for the points entered. to improve the viewing.
You may repeat the process of digitizing points to
represent the contamination sites and the known head
locations (domestic wells, etc.).
• Contamination sites may be digitized as a single, named, point set. Known well/piezometers
elevations may be digitized as another. Non-point sources can be outlined using the "curve"
command discussed below.
Digitize Roads and City Limits
The curve command is used to digitize a set of points that define a curve for non-hydrologic
landmark features such as roads and city limits. From the Main Menu select Curve or press .
At the prompt enter the curve name:
Type: "City Limits".
Place digitizer puck on city limits.
Press twice. The digitizing screen will appear
on the monitor.
Press puck button one. Continue to press puck button
one as you move along the city limits, selecting points
farther apart where the boundary is straight, and closer
together where it is more curved. Plus signs will appear
on the screen for the points entered.
• If you make an error, press .
• Repeat the procedure to digitize roads. Select a single road for the exercise.
View the Digitized Features
20
-------
After digitizing Swan Pond Ditch, the city limits, a road, and the well field, select the view command
by pressing , in the DIGITIZE module to see the results. You may have to zoom out
or shift the figure using the arrow keys to see all the features entered. Press .
ReadDM WriteDM LoadElem SaveEIera Map ChangeDir setDrive, Quit ^ :
• Select WriteDM or press . The system prompts for a file name, e.g., your name (no
extension needed). Make sure to give a name other than
VINCENNE! We do not want to overwrite the existing file
VINCENNE.DM.
• At the title prompt type: type a title, or press .
• The file will be saved with the extension ".DM" .
• Press to return to the main GAEP menu.
Edit Features in GAEP
You have digitized sections of the same stream from a single quad map. These two sections could
have come from a stream which crossed from one map to another. In.fiither case, you may wish to
connect these two sections into one feature.
Using the EDIT menu, you can join these features into a single stream. Commands to edit digitized
features are available in the EDIT menu of the DIGITIZE module.
From the main menu, select the Digitize option and the Edit command, press followed by .
21
-------
sloin View Delete Rename Quit
View the Data
Select the view command . When the graphics screen appears, move the mouse around to
highlight the various features. Note that the stream has two sections and each has a separate name.
Our next step will be to join these features together under a single name; press to return to
the Edit menu.
Join Features
The Join command connects stream or curve segments. The feature that results will have the name
of the segment selected first.
• Press to enter the graphics screen.
• Highlight one of the two digitized segments by moving the mouse cursor.
• Press the left mouse button to select segment. The stream segment color will change from
yellow to red.
• Highlight the other segment.
• Press the left mouse button. The color will again change to red. You will be
prompted to confirm that these are the correct segments
to join.
• Press and . The entire stream reach will appear in red and is now
one feature. A prompt will again appear to let you join
another segment to these. Press twice to return to
the EDIT menu.
• If you make an error, press . You will then be returned to the EDIT menu.
• Revised digital map can be viewed by pressing .
• When you are finished, press . This will to return to the DIGITIZE menu and
again to return to the main GAEP menu.
The same join procedure can also be used to join "curves." The revised digital map file can be saved
on disk. From the Main Menu:
• Press (File) and then (WriteDM).
• At the prompt, type in a title or press .
22
-------
Type in a file name (do not use vincenne!), and press or accept the current file name
by pressing . When the file has been saved to disk, press to
exit.
Create an Input File for CZAEM
Up to this point, you have used GAEP to build and edit features of a digital map. The map, however,
is far from complete; only a few features were entered for demonstration purposes. For the
remainder of the tutorial, you will be working with a complete digital map of the Vincennes area:
VINCENNE.DM, which is provided in the \WHAEM\DAT directory.
Relevant information about the aquifer, the well field, and local geology are needed for a ground-
water modeling project. Because CZAEM is a steady state model, the input parameters should be
thought of as average annual values. The following aquifer and well data are required:
• Pumping rates and radii for the wells in the well field
• Elevation of the aquifer base
• Average aquifer thickness
• Hydraulic conductivity (permeability)
• Porosity
• Average head in the study area (called reference head)
• Average areal recharge rate
Much of this information can be acquired from state and U.S. Geological Survey reports, well field
data, and domestic well logs from state natural resource or environmental agencies.
GAEP allows the modeler to quickly build an initial model of an area covered by a digital map. It
also makes modification of input data fast and efficient. These quick and easy changes in the
CZAEM input file are important for hypothesis testing, as discussed in Chapter 2.
23
-------
Creating Analytic Elements
The Element module in GAEP will be used to create and edit CZAEM input files using
VINCENNE.DM. It is in this step of the process that hydrological features (streams, lakes, wells)
are going to be represented by analytic elements used in CZAEM to solve the ground-water flow
problem. This section of the tutorial outlines the necessary steps to build a basic CZAEM model
containing line-sinks and wells.
First a word about data resolution. A decrease in model "resolution" is apparent moving outward
from the area of interest (see Figure 8). The "near field" features are those in the immediate area
of the well field. These features should be defined with the highest resolution (shortest) line-sinks.
Figure 8 Relative positions of near-field and far-field features in an analytic element model.
Notice the difference in resolution of the line-sinks.
Out in the "far field", hydrologic features can be represented much more coarsely. The purpose of
24
-------
these "far field" features is to control the ground-water flow into (or out of) the study area of the
model.
Users unfamiliar with the analytic element method and the process of representing streams as
discrete line segments frequently ask, "How many line-sinks are needed and how precisely should
the segments match the actual streams?" Representing streams by an appropriate string of line-sinks
requires experience and trial-and-error runs with CZAEM. The length of the line-sinks determines
the effective resolution with which you can represent a stream. The fact that a stream may have a
long straight section does not mean one long line-sink will adequately represent that section. The
discharge along the section may vary significantly, .in which case several (constant strength) line-
sinks are required to properly model that discharge distribution. In general, shorter line-sinks are
used for streams close to the well field (the area of interest). Line-sink length should gradually
increase or decrease with changes in resolution because adjoining line-sinks of significantly different
length may cause numerical inaccuracies. When creating line-sinks, you do not need to precisely
match the stream location or shape, certainly not in the far field. The reader is referred to the
CZAEM User's Guide for exercise in creating line-sink representation of streams (USEPA, 1994).
The Element module requires that streams and wells to be included in the model to already be
digitized and read into GAEP. A complete digital map has been installed in the \WHAEM\DAT
directory for this exercise.
NOTE:
The following section of the tutorial is designed to give the user practice in creating a CZAEM input data file. Howver,
it is not necessary to create a complete file during this exercise. A completed input file (VINCENNE.DAT) has been
included for later use in CZAEM.
Start GAEP at the DOS prompt:
C:\WHAEM> gaep
From the main menu:
• Select File or press .
• Select ReadDM or press .
• Type: "VINCENNE". Press to import the file. Press to return to the main menu.
Enter the Element module:
• Select Element or press . The following set of choices will appear:
k;> \ I? I Linesink Well Delete Itserwindow Image View Quit : ~ «,
25
-------
Creating Line-sinks:
Select the Linesink command .
GAEP will display all presently defined map features.
Only streams which have water levels defined (features
displayed in light blue) can be used to generate line-
sinks. In VINCENNE.DM all features are light blue.
For this exercise, we will create line-sinks for Kelso Creek, located east of the well field and the City
of Vincennes, as shown in Figure 9. Prior to selecting the stream, adjust the figure so that the
House button 1: add point Keyboard F3: Done; ESC: CANCEL
Figure 9 Kelso Creek with elevation marks where contours cross the stream.
stream fits on the screen using the zoom or and the arrow keys to adjust the
viewing window.
• Select the stream, using the left mouse button. The stream and the water levels at known
points will be displayed on the screen in red
To create the line-sinks along Kelso Creek, place the mouse pointer at the 400 foot elevation point
and press the left mouse button. Move the mouse along the stream (the pointer on the screen will
move) and press the left mouse button where end points are to be created. When subsequent points
26
-------
are selected, line-sinks appear in yellow with calculated heads printed at their centers. If an error
is made while creating line-sinks, press to abort. Press the key when you are finished
creating line-sinks for Kelso Creek. The line-sink string appears in yellow with crosses to mark end
points. Figure 10 shows an example of line-sink representation of Kelso Creek.
When finished creating line-sinks, press
-------
Wrong: Permeability = 3.527E-2 (cm/sec)
Recharge = 4.0 (inches/year)
fiase Jhick permeability pfirosity IJniformFlow re£erence gain fluit
Aquifer Parameter Settings:
Base Elevation: 0.0
Thickness: 100.0
Permeability: 100
Porosity: 0.2
Uniform Flow: QO 0
Current Directory:
C. \WHAEM\D AT
Current Map File:
Vincenne.dm
Vincennes
Current Element File:
alpha 0.00
Memory available:
368800
Reference point:
X 0.0
Y 0.0
Head 0.0
Rain element is not defined
Option Settings:
Unit Conversion:
Digitizer Mode:
Video graphics mode:
help return to previous menu
M->FT
DIGITIZER
COLOB
Figure 11 GAEP Aquifer Submodule.
Enter the aquifer data for Vincennes as defined in Table 1. Most of the parameters are self-
explanatory. As a first approximation, the reference point is assigned the average head in the study
area, and located far away from the study area. Coordinates for this point are entered as UTM values
(meters); they will be transformed to coordinates with respect to the "model origin" when creating
a CZAEM input file. See the CZAEM User's Guide (USEPA, 1994) for additional discussion of the
reference point.
The thickness parameter can be used to indicate the presence of a confining layer and subsequently
define the top of the aquifer. In an unconfined aquifer, the thickness parameter is often set
artificially high to assure that flow is unconfined throughout the domain of interest. This does not
affect the computations; see Strack( 1989), sections, for a full discussion. For the outwash aquifer
near Vincennes, the outwash is less than 100 feet thick.
29
-------
Table 1 Aquifer data for Vincennes case study.
Base Elevation
Thickness
Permeability (hydr.cond.)
Porosity
Reference Point X:
Y:
Head:
330
100
350
0.20
0
656160
410
The "rain element" is the only item in the aquifer module which is defined graphically. To define
a "rain element" (circular recharge area):
• Press at the AQUIFER menu. The Vincennes digital map will appear on the screen.
You may have to zoom out and shift the figure to get all elements on the screen.
Use the mouse (which controls the cursor) to define the rain circle as follows:
• Place the cursor at the center of the near-field (City of Vincennes).
• Click the left mouse button.
• Move the cursor away from the center. The rain circle expands as the cursor moves.
• Click the left mouse button when all of the line-sinks are inside the rain circle. In our case
only the line-sinks on Kelso Creek are present, but imagine that they cover other streams.
• Answer "yes" to accept. You are now prompted for the recharge rate.
• Type "0.0032" for a recharge of 0.0032 ft/day (14 in/yr). Press . You will be
returned to the AQUIFER menu. If an error is made (before pressing or ), press
-------
Saving the CZAEM Input Data on Disk
The analytic elements and aquifer data created above need to be saved to disk as an input file for
CZAEM. File operations take place within the file menu of the main GAEP menu.
Once you are in the file menu:
• Select SaveElem to save the elements and aquifer data to disk. The analytic element file will
contain line-sinks, well elements, aquifer properties, the recharge rate, and (optionally) map
features such as roads.
• At the prompt, change the name of the file by typing in a new filename and press .
DO NOT use the name vincenne.dat! This would overwrite the existing data file.
• You will then be prompted for the UTM coordinates of the model origin.
• Type in "450000 4280000" and press . You will be returned to the main menu in
GAEP.
The coordinates of all elements saved to the file will be in feet from the model origin. If you want
the coordinates to be in meters, select Options from the main menu. Then select to change the
units to metric. See the GAEP Reference Manual for more details on the Options menu. Exit the
program.
Imaging Line-sinks Generated in GAEP
In some cases, the assumption of a homogeneous aquifer of infinite lateral extent is so far from
reality that the modeler needs to consider other options, hi WTzAEM , the method of images can be
used to model a no-flow boundary condition along an outwash valley wall (Figure 12). For a
complete discussion of image theory, see Strack (1989), pages 27-33.
Note:
The following section is designed to give the user practice in creating an image data file; however, it is not necessary
to complete the exercise. A complete file VINIMAGE.DAT is included on your disk.
31
-------
no -flow boundary
Figure 12 Geologic setting of an outwash aquifer.
Near the city of Vincennes, this linear no-flow feature can be used to model a sandstone
bedrock/outwash interface that occurs to the south and east of town. In GAEP, the method of images
is implemented by defining an image line through two points on the no-flow boundary of the aquifer,
as shown in Figure 13. The program then images the existing line-sinks across that boundary and
repositions the rain circle so that it is centered on the image line. Any line-sinks occuring on the east
side (the image side) of the line are removed prior to imaging. The application of this imaging
technique (introducing a no flow boundary) is illustrated by the following steps, from the main menu
in GAEP:
• Select File or press .
• Select ReadNew or press .
• Type: "VINCENNE".
• Press .
• Select Element or press .
We will first load the completed CZAEM input data file (without
images) -VINCENNE.DAT.
This will bring you back to the Main Menu of GAEP.
32
-------
Go into View and identify Mantle Ditch on the southeast side ot the city.
The bedrock/outwash boundary is nearly parallel to this stream just east of the ditch. We will create
an image line along this stream by the following steps:
Select Image or press .
The screen will shift to a graphics image of the element file
(VINCENNE.DAT). The following promptwill be displayed at
the top of the screen:
point #2
image line to be'aefined
Figure 13 Outwash boundary within regional layout.
Select image origin and press mouse button
• Move the cursor south of town, on Mantle Ditch. Press the left mouse button to select the
first image line point. The following prompt will appear at the top of the screen:
33
-------
Select the second oint on
line and ress left mouse button
Move mouse to the NE and notice the GAEP image line .
Rotate the line so that it reasonably approximates the position of the rock outcrop (roughly
parallel to Mantle Ditch and Wabash River).
Click the mouse button to define the image line, as shown in Figure 14.
DofiD* ing* lisa, clicking Out waaa* batten at two points.
Figure 14 Defining the image line with mouse control.
Type: "yes".
Image line OK?
Next you will see a new rain circle in red, centered at the first image line
point.
Use the mouse to open the circle so that it covers the image domain and the "real" linesinks
to the left of the line, as shown in Figure 15.
Click left mouse button.
34
-------
rttrtrtng Uu> i
Figure 15 Defining rain circle with mouse control.
Rain circle OK?
• Type: "yes".
• Use same recharge rate as before: (0.0032)
Enter reference head;
• As before, use average heads in modeled area: (405)
Now you are back in the Element menu. Press to View the "image" and "real" linesinks. The
image linesinks are plotted in light gray. From here you may return to the main GAEP menu, press
(File) and save the data in a new input file. We will omit these steps, because a complete file
with images has already been provided for use in the \WHAEM\DAT subdirectory named
VINIMAGE.DAT. Do not replace this file by saving new data to this file name. Exit the program.
35
-------
Model the Site with CZAEM
You are now ready to run the Capture Zone Analytic Element Model (CZAEM) to determine the
capture zones for the Vincennes City well field. The tutorial will take you through basic operations
of using CZAEM for wellhead protection. A more complete discussion is found in the CZAEM
User's Guide (USEPA, 1994). A reference guide to CZAEM commands can be found in Appendix
B.
There are eight steps to performing a wellhead protection capture zone analysis with CZAEM:
1. Read the input file: VDsfCENNE.DAT.
2. Visually check the input file.
3. Solve the ground-water flow problem.
4. Generate a grid for contouring piezometric heads.
5. Evaluate the ground-water
solution.
6. Create capture zones.
7. Save solution to disk.
8. Test hypotheses by adjusting
input data and start over.
To start the CZAEM program, change
your working directory to the
\WHAEM directory:
C:\?CD \WHAEM
C:\WHAEM5CZ
At the introductory CZAEM
information screen, press to
enter the main module of CZAEM.
When the main module command line
appears on the screen, you are ready
to read in an input file produced with
GAEP. Like GAEP, CZAEM is a
modular program. The most
important operational difference is the
user interface. CZAEM is a
command line program and requires
direct keyboard input. CZAEM
commands can be abbreviated to the
first few letters (enough to make them
General CZAEM Hints
All commands may be abbreviated to the first few
letters.
< ? > context sensitive help.
"return" previous menu, exit graphic screen.
previous command.
Cursor control -
The mouse is the default cursor control device.
The mouse buttons are not active.
will delete commands when not in
graphics mode.
< (the left angle bracket key) > will delete commands
while in graphics mode.
36
-------
unique). This W/zAEM tutorial is only a brief introduction to the program. For more information
about the use of CZAEM, the reader is referred to the more extensive discussion in the separate
CZAEM Users Guide (USEPA, 1994).
Read CZAEM Input FOe: VINCENNE.DAT
The CZAEM input file created by GAEP is an ASCII file which contains the data describing the flow
problem as well as the CZAEM commands needed to process the data. The file is read into CZAEM
through the SWITCH module. GAEP has added the instructions to the VINCENNE.DAT file to
solve the ground-water flow problem, create a grid for contouring the piezometric surface, and to
define some parameters for capture zone delineation.
Files created by GAEP are stored in the c:\WHAEM\DAT\ subdirectory.. To read in the tutorial
CZAEM input file:
• Type: "SWI VINCENNE.DAT ".
As the file is being read in, the data will scroll across the screen quickly. When the end of the file
is reached control will be returned to the keyboard (console) and the Main Command Menu (Figure
16) will appear on the screen.
\\\ Module=MAIN MENU Level=0 Routine=INPUT ///
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)///]
-------
Type: "WINDOW ALL".
Type: "LAY ".
buffer.
This defines a new window including all elements.
The user can examine the distribution of linesinks, as well as the road and well field locations for
the area being modeled. The screen image should look like Figure 17.
Figure 17 Layout of the entire VINCENNE.DAT data file - window all. (Large
overlapping plus signs indicate the well field; they will appear smaller on the screen).
The solid white lines and the small white square are the linesinks and the well element defined
earlier in GAEP. The dark red dashed-dotted lines are roads and the city boundary. to
return to command mode.
38
-------
Turning map on and off
You may want to turn off the plotting of the roads and wells to reduce visual clutter on the screen.
To do this, at the MAIN MENU:
• Type: "MAP". This will place you in the MAP module.
• Type: "PLOT OFF ". Plotting of the road and well file is turned off
• Type: "RET ". You are back at the MAIN MENU.
• Type: "LAY ". Again, check the image. To turn the map plotting routine back
on, type "PLOT ON" in the MAP module.
Changing the Window
The "window" is the area displayed on the screen and is defined by two coordinate pairs: the lower
left and the upper right comer of the area to be displayed. A new window may be defined by
typing: "WIN xl, yl, x2, y2" where the first coordinate pair is the lower left corner, and the second
coordinate pair is the upper right hand corner of the domain.
The WINDOW PUSH command will save the current window. The WINDOW POP command
will retrieve the last window pushed into memory. It is not saved! Only the windows pushed into
memory can be retrieved (popped). We advise the user to select a set of windows for the project
and take the time to write these coordinates down.
The initial window can be restored with GAEP's User Window utility:
• Type: "WIN POP ". This retrieves the last window in memory (selected in GAEP).
You will be restoring a window with lower left coordinates of (4752, -9327) and upper right
coordinates of (21625, 5545).
• Type: "WIN". The current window coordinates are printed on the screea
• Type: "LAY ". To see a close-up of the well field, as in Figure 18.
• Press again. You will be back at the main menu.
In order to be complete, this tutorial includes instructions for interactively issuing the solve
command (even though this command has been included in the Vincennes file) as well as
instructions and explanation of the grid routine. Because this has been automated, the reader may
39
-------
elect to skip these topics. You should be aware that any time you change the number of elements,
or aquifer properties, or the reference point, you will need to re-solve and re-grid. Any time you
change the window, or desire a different resolution on a contour plot, you need to re-grid.
\
V
Figure 18 Zoomed-in view of layout of VINCENNE.DAT file, including map
features.
Solve the Ground-Water Flow Problem
(Optional: already done in vincenne.dat!)
Once the input data appears correct, you proceed to solve the ground-water flow problem. At the
MAIN MENU:
Type: "SOLVE ".
When SOLVE is complete, the MAIN MENU will appear on
the screen.
40
-------
Generate Grid for Contouring Heads
(Optional: already done in vincenne.dat)
In order to generate a contour plot of piezometric heads, we will calculate the head at a grid of evenly
spaced points in the current window (domain seen in layout). At the MAIN MENU:
• Type: "GRID 30 ".
Important:
Anytime you change the window, you will have to repeat the grid procedure if you want to view
contours inside the new window. This can be done by typing "GRID 30" at the MAIN MENU after
the new window has been defined. The parameter "30" specifies 30 grid points along the horizontal
axis of the plot and an appropriate number of points along the vertical axis to obtain an even grid
spacing. You may create higher resolution plots by increasing this number, but this will take a little
more time.
View Piezometric Contours in PLOT Module
To inspect the computed piezometric head surface, at the MAIN MENU:
• Type: "PLOT ". This will place you in the PLOT module. You will see the following
prompt:
EFAULT [NUMBER OF LEVELS] AYOUT
(MIN LEVEL [INCREMENT [>0]][MAX LEVEL]
(MAX LEVEL [DECREMENT [<0]][MIN LEVEL]
MIN. LEVEL= 3.963030E+02 MAX. LEVEL= 4.448840E+02
• The "min. level" and "max. level" values on your screen may differ somewhat from those
shown here.
• You can either type, "D" , to accept the default contour interval or adjust the settings
by supplying the minimum contour level to be plotted, the desired contour interval (optional),
and the maximum contour level to be plotted (also optional). For example:
• Type: "400 5 ". The number of contours will appear on the screen, along with the
following message: THERE ARE 9 LEVELS: PRESS ENTER.
• Press . The screen will display a contour plot of piezometric heads plotted from
the lowest to highest head.
41
-------
Use Pathline Tracing in TRACE Module
While potentiometric contours are the most familiar image of a ground-water flow solution, an
alternative way to evaluate flow is to use the pathline tracing functions of CZAEM to see how
ground-water moves throughout the model domain. This is done in the TRACE module, which
allows the modeler to view a layout of the study area (with or without piezometric contours) and
trace pathlines from any point in the user window. From the main menu:
• Type: "TRACE ". The TRACE menu appears on the screea
• Type: "LAY ". A "layout" will appear of the current window without the
piezometric contours. The linesinks, well, and cursor will be
displayed on the screen along with menu selections at the top.
To draw pathlines place the cursor near the upper end of the Mantle Ditch linesinks (the lower left
white line on the screen):
• Type: "TRACE 380 ". A pathline will be drawn in purple, with the pathline starting at
the cursor point (x,y) defined by the cursor location, and
elevation equal to 380 feet. Markers cross the pathline at one
year intervals.
Place the cursor between City Ditch (the white line straight south of the well) and the well field.
• Type: "TRACE ". A pathline will again be drawn, this time starting at the default
value of the top of the aquifer at the cursor location x,y. If you
wish to do so, continue to move the cursor around the domain,
starting pathlines in various locations to "gel a feel" for how
ground water is moving. Type "RET" to return to the main
module.
NOTE:
The tic marks on the streamlines indicate ONE YEAR ground-water travel time intervals (defined earlier in
VINCENNE.DAT).
Initial Capture Zone Analysis
It is recommended to use the option WGEN (Well Generate) in the TRACE module to perform
initial analysis of capture zones. The WGEN routine is relatively fast and can be used to get a clear
picture of ground-water flow patterns near the well. After modeling the site and selecting the most
important scenarios, final capture zones can be prepared using the SUBZONE and TIMEZONEs
commands in CZAEM. These commands are explained later in this section.
42
-------
At the MAIN MENU:
• Type: "TRACE" .
• Type: "SET" .
• Type: "BACK ON" .
You will be placed in the TRACE module.
This moves you into the SET submodule of TRACE and allows
you to alter the tolerance and other settings.
This changes the setting of the tracing routine to trace against the
direction of flow. A message will tell you that the tracing
direction is set to backward.
Type:"MAXSTEP 100" . Sets the step size along a pathline to 100 days.
Type: "MARK TIME 730:.
Changes the time markers on the streamlines from one
year to two years (730 days).
Type: "TERM TIME 3650" . This will terminate the pathline trace after 3650days or
10 years.
Type: "RET" .
Type: "LAY" .
Type: "WGEN 16" .
You will be returned to the main menu of the TRACE
module.
Menu choices and the layout of the element features
will appear on the screen, with the cursor located near
the middle of the window. Move the cursor to thewell.
This will trace 16 pathlines from the well and will
define the shape of the capture zone. The pathlines will
show up in purple on the screen. While the lines are
being traced the cursor will disappear. When all of the
lines are complete the cursor reappears. See Figure 17.
Note:
The well receives part of its water from the Wabash River. The tick marks indicate 2-year increments. This set of
stream line and tick marks provide an approximation of a more detaled 10-year time-of-travel capture zone that will be
generated later with the SUBZONE and TIMEZONE commands.
Type: "MENU" .
You will be returned to the TRACE module screen
43
-------
.
\ \
\ \
\
\
Figure 19 Steady state pathlines backward traced from the Vincennes wellfield
(WGEN 16 command in TRACE module; tic marks every two years).
Hypothesis Testing
During the initial stages of modeling the user is comparing modeled piezometric heads with heads
observed in the field. Evaluating the differences between modeled and observed heads is the first
step in model calibration and is one way to judge the adequacy of the ground-water flow model.
This procedure requires good hydrologic insight and substantial modeling skills. Developing these
skills is outside the scope of this tutorial.
The current CZAEM model of the Vincennes area (VINCENNE.DAT) predicts heads to be higher
than have been observed. In addition, it lacks the necessary degree of realism because it does not
include the transition in hydraulic conductivity from the highly permeable channel deposits along
the Wabash River to the sandstone bedrock outcrop east of town. Previous USGS modeling efforts
44
-------
in this area dealt with this transition as a no-flow boundary (Shedlock, 1980). As discussed earlier
in this tutorial, GAEP allows the modeler to introduce no-flow boundaries by applying the method
of images. The file VINIMAGE.DAT includes an image line that approximates this rock outcrop.
The solution to VINIMAGE.DAT is in much better agreement with field data than
VINCENNE.DAT which ignores this feature.
We will continue our tutorial using the file VINIMAGE.DAT.
Read in New File: VINIMAGE.DAT
• Return to the main menu by typing "RET".
• Type: "RESET" .
• Answer "Y" .
• At the main menu type: "SWI VINIMAGE.DAT" .
Create Final Capture Zones
When a set of satisfactory models have been obtained for the ground-water flow around the
Vincennes well field, you can generate capture zones. For this example we will use both of the two
different commands to delineate a capture zone. The first command, "subzone," delineates the
complete capture zone of a well. The second command, "timezone," will draw isochrones within
the capture zone. The water in the area bounded by the isochrone is captured by a well within a
specified time period. Each command can be used to identify a wellhead protection area for a well
field. A solution to the ground-water flow problem must be in memory before creating capture
zones.
The first step in creating a capture zone is to make sure that the well generates at least one stagnation
point within the current window. The subzone routine that is used to create a capture zone searches
the window for a stagnation point generated by the well. Use the WGEN command to verify that
stagnation point(s) are inside the current window. For this purpose repeat the steps under the
heading "Initial capture zone analysis" (See Figure 20).
Compare new streamline patterns with the previous WGEN without the image line.
45
-------
7 \
/ /''s. \V\M
-L ! \ \ \ V\
\ \ \ \x \
\ V •- > /
\ X \/ !
\ \ /
/
\ V- x \ • ••
y> x x \\\\,*
/\ \ \ \ \ '$ \ \
/ xXvv\vvv/ \ s\ N
^ xx\ \ \ y^r \ V \ / \ //
x s
\ \
/ \
\^
\
\
Figure 2to Modeled steady state path lines from the Vincennes wellfield (WGEN 16
command and the input data VINIMAGE.DAT).
Generate a Subzone Around the Well
For our case the window is suitable (includes all stagnation points for the well) so that we may
proceed delineating the capture zone. This will be accomplished by first entering the CAPZONE
submodule of TRACE and then typing SUBZONE. In the TRACE module:
Type: "CAPZONE" .
Information about piezometric head levels will appear on the
screen.
Type: "400 5" .
Again, press .
This will plot contour levels in 5 foot intervals beginning at the
400 foot level. You will be told how many contours will be
plotted.
An image of the elements and contour lines will appear on the
46
-------
Position the cursor on the well.
Type: "SUBZONE" .
screen along with the menu choices for the CAPZONE
submodule.
The program goes through a three phase calculation procedure.
This takes a few minutes..
If problems occur at this point (for example, no stagnation point is found), messages will appear on
the screen indicating the nature of the problem. If you receive an error message, type:" COM" and
press to get the menu for the CAPZONE submodule. If the subzone is ready, it will be
plotted on the screen. The subzone you have created for the Vincennes well field should look like
Figure 21.
\ y. \ i i
Figure 21 Modeled subzone for VINIMAGE.DAT showing source areas for the
wellfield.
Generate Isochrones Around theWell
After the total capture zone has been delineated with the SUBZONE routine, you can compute steady
state time-of-travel capture zones (TIMEZONES) for the well. Place the cursor on the well.
47
-------
Type: "TIME" .
You will be prompted to enter a minimum time, time step, and
maximum time, or redraw last time zone, calculate the default
time zone, or exit
• Type: "730 730 3650" . Start at 2 years, step 2 years, stop at 10 years.
The time zone calculations may take a while to compute. When the solution is complete, the
timezones will appear on the screen. The steady state time-of-travel (TOT) capture zones for the
well field are illustrated Figure 22.
Figure 22 Steady state time-of-travel capture zones (2-yr intervals) including no-flow
boundary.
Type: "RET" .
Type: "RET" .
To clear the screen and return to the TRACE MENU.
To return to the MAIN MENU.
48
-------
Sending Graphics to the Printer
In Chapter 1, it was explained that a small batch file needs to be run outside of the program (from
the DOS prompt) which sets the system up for producing hard-copy output. Before generating the
capture zone printouts, be sure that the map file shows all of the features that you need for
interpretation of the results. (Be sure that the PRINTER.BAT file has been run during set-up.)
After using WGEN to identify the solutions which are most useful, you should be ready to generate
pictures of the time zones, subzones or both. There are two ways of doing this in CZAEM:
1) Recalculate the time zones and subzones with graphics output routed to the printer (or a print
. file).
2) Save the time zone buffers to disk (binary file) and read them back when you are ready to
generate graphics on the printer.
The following discussion outlines the steps needed to generate images on the printer using each
approach. We assume that you are currently in the CAPZONE module.
To regenerate capture zones and then send graphics to the printer:
•
•
Type:"RETURN" .
Type:"PSET" .
Type:"PRINTER" .
Type:"RETURN" .
Type:"CURSOR OFF".
Type:"CAPZONE" .
Type the contour levels: "400 5 ".
Press twice.
Type:"8549 2234 TIMEZONE".
This will put you at the TRACE module.
The PSET module allows you to control various
settings relating to the graphics displayed by CZAEM.
This redirects graphics output to the printer.
You are now back in the TRACE module.
This is very important! This is required in order to go
through the next steps.
No graphic appears. The program waits for the
coordinates ofthe well followed by the TIMEZONE"
command.
The well is centered at the point (8549,2234). Wait for
the calculations to be complete, then enter the
49
-------
minimum, increment, and maximum time:
• Type:"730 730 3650" Wait for the calculations to be completed
• Type:"RETURN" This closes the Postscript file (PLOT.PS) which may be
copied to any Postscript printing device or it sends the
file to the HP Laser Jet III printer.
Alternately, if you would like to save the image buffers and then read them back later, you would
have to do the following: (This assumes that you are looking at a graphics screen in the CAPZONE
module.)
• Type: "CSAVE". To save the subzone and timezone buffers to disk.
(This command is not documented on the menu.)
• Enter a filename, (e.g., "vincenne.cap" ).
• Type: "RETURN". To get to the TRACE module.
• Type: "PSET". To re-route the graphics to the printer.
• Type: "PRINTER".
• Type: "RETURN". You are now back at the trace module.
• Type: "CURSOR OFF". Do not forget this!!!
• Type: "CAPZONE". At the prompt enter the contour levels.
• Type: "400 5" (or ). To either print the contours or simply the layout
• Type: "CREAD".
• To re-load the buffers enter the filename at the prompt (e.g., "vincenne.cap") (This command
is not documented on the menu.)
• Type: "8549 2234 TIMEZONE".
• Type: "730 730 3650". To contour from the two to the ten year time-of-travel
capture zones in two year intervals.
• Type: "RETURN". To plot to file PLOT.PS or to send the graphics directly
to your HP Laser Jet printer, depending on your setup.
50
-------
Exit CZAEM
When you are finished with your work in CZAEM, return to the MAIN MENU.
• Type: "STOP" .
51
-------
REFERMI
Anderson, M.P., and W.W. Woessner, 1992. Applied Groundwater Modeling: Simulation of Flow
and Advective Transport Academic Press, San Diego, CA.
Bakker, M., and O.D.L. Strack, in preparation. Capture zone delineation with analytic element
models, Water Resources Res.
Bear, J. and A. Verruijt, 1987. Modeling Groundwater Flow and Pollution, Reidel, Dordrecht.
Dupuit, J., 1863. Etudes Theoriques et Practiques sur le Mouvement des Eaux dans les Canaux
Decouverts et a Trovers les Terrains Permeables. Dunod, Paris, 2nd edition.
Forchheimer, P., 1886. Ueber die ergiebegdeit von brunnen-anlagen und sickerschlitzen.
Architekt. Ing. Ver. Hannover, 32:539-563.
Haitjema, H.M., 1985. Modeling three-dimensional flow in confined aquifers by superposition of
both two- and three-dimensional analytic functions. Water Resour. Res., 21(10): 1557-
1566.
Haitjema, H.M., 1987. Comparing a three-dimensional and a Dupuit-Forchheimer solution for a
circular recharge area in a confined aquifer. J. Hydrology, 91:83-101.
Haitjema, H.M., in preparation. Groundwater Modeling Using the Analytic Element Method.
Academic Press.
Kelson, V.A., H.M. Haitjema, and S.R. Kraemer, 1993. GAEP: A geographic pre-processor for
groundwater flow modeling. Hydrological Science and Technology, 8(1-4): 74-83.
Shedlock, R.J., 1980. Saline water at the base of glacial outwash aquifer near Vincennes, Knox
County, Indiana, USGS Water Resources Investigation 80-65, 52pp.
Strack, O.D.L., 1984. Three-dimensional streamlines in Dupuit-Forchheimer models. Water
Resour. Res.,20(7):* 12-822.
Strack, O.D.L., 1989. Groundwater Mechanics. Prentice Hall, Englewood Cliffs, New Jersey.
Strack, O.D.L., 1992. A mathematical model for dispersion with a moving front in groundwater,
Water Resources Res., 28(11):2973-2980.
52
-------
Strack, O.D.L., and H.M. Haitjema, 198la. Modeling double aquifer flow using a comprehensive
potential and distributed singularities. 1. Solution for homogeneous permeabilities. Water
Resour. Res., 17(5): 1535-1549.
Strack, O.D.L., and H.M. Haitjema, 1981b. Modeling double aquifer flow using a comprehensive
potential and distributed singularities 2. Solution for inhomogeneous permeabilities. Water
Resour. Res., 17(5)1551-1560.
USEPA, 1994. CZAEM User's Guide: Modeling Capture Zones of Ground-Water. Wells Using
Analytic Elements, by O.D.L. Strack, E.I. Anderson, M. Bakker, W.C. Olsen, J.C. Panda,
R.W. Pennings, D.R. Steward, Robert S. Kerr Environmental Research Lab Research
Report, EPA/600/R-94/174.
53
-------
APPENDIX A. GAEP
Written by:
Victor A. Kelson
Groundwater Modeling Laboratory
School of Public & Environmental Affairs
Indiana University, Bloomington
Copyright (c) 1993,1994
This. Chapter of the W/iAEM Manual describes the operation of the GAEP program with a number
of different hardware configuration systems. The GAEP program was developed by Vic Kelson
at the SPEA Groundwater Modeling Laboratory, Indiana University, Bloomington. The author
acknowledges Phil DiLavore for his work on the initial design of GAEP. Thanks also to Jack
Wittman of IU and Dr. Stephen R. Kraemer of the USEPA/RSKERL-Ada for assistance and
guidance with this work.
CalComp is a trademark of CalComp Inc.
Summagraphics, SummaSketch, MM1812, and Bit Pad Plus are trademarks of Summagraphics
Corporation.
Microsoft and MS-DOS are trademarks of Microsoft Corporation
80386 and 80486 are trademarks of Intel, Inc.
54
-------
Concepts
This Appendix describes the program GAEP (as implemented for the WzAEM product). It is not
intended as a complete tutorial for GAEP, but as a user reference once the printed tutorial has been
studied. GAEP (Geographic Analytic Element Preprocessor) is a program which greatly speeds
and simplifies the process of developing ground-water flow models using the analytic element
method. The purpose of GAEP is not to supplant the use of GIS tools for ground-water
management, but to provide a specific set of functions which streamline the modeling process.
GAEP allows the modeler to create and manage a digital map of the hydrography of his study
region, irrespective of any planned modeling work. The digital map area used should be large
enough to cover any intended modeling work in a region. Once a digital map is prepared, it can
be used on a variety of modeling projects, with the modeler creating his model from the digital map
using the mouse on his computer. Other considerations, such as aquifer properties, can also be
managed with GAEP. Once the modeler has defined his elements and aquifer properties, a data
file for the CZAEM modeling program can be written, freeing the modeler from the process of
manually editing CZAEM input files.
Files
GAEP manages two different types of data files, "Digital Map" files (*.DM) which are specifically
defined for GAEP's use, and "Analytic Element" files (*.DAT), which are in a format compatible
with CZAEM, but include additional information required for GAEP's interpretation.
Digital Map Files
GAEP creates and manages digital maps in a simple ASCII file format. These "Digital Map" files
contain the locations of streams, lakes and background map features and the elevations of surface
hydrologic features for use in creating analytic elements for CZAEM.
DIGITAL MAP FILES SHOULD NOT BE MODIFIED BY THE USER EXCEPT BY THE USE
OF THE GAEP PROGRAM. USE OF OTHER METHODS FOR EDITING THESE FILES MAY
RESULT IN LOSS OF DATA OR SEVERE MODELING ERRORS.
Analytic Element Files
Once the modeler has completed the definition of analytic elements and aquifer properties for a
55
-------
model run, GAEP saves the elements in a CZAEM-compatible "Analytic Element Data" file.
Included in the analytic element file are a set of commands which solve the model, create a grid
of heads for contour plotting and create a background map of geographic features. A view window
selected in GAEP by the user is optionally set in the analytic element data file, as well.
ANALYTIC ELEMENT FILES SHOULD NOT BE MODIFIED BY THE USER EXCEPT BY
THE USE OF THE GAEP PROGRAM. USE OF OTHER METHODS FOR EDITING THESE
FILES MAY RESULT IN LOSS OF DATA OR SEVERE MODELING ERRORS.
Digital Map Features
GAEP allows the user to create a variety of digital map features which represent hydrologic
features (which can then be used to create line-sink analytic elements) and other geographic
features such as roads and political boundaries.
Stream Features
A stream feature represents a body of surface water, either a river or the perimeter of a lake.
Stream features are used as the basis for the creation of LINESINK analytic elements. Stream
features are entered as a set of points which define the geographic extent of the feature and a set
of points where the water elevation is known (usually where a topographic contour crosses the
feature). It is important for the user to determine which stream reaches should be included in the
digital map.
Curve Features
A curve feature represents a road, geographic boundary, geological feature, contaminant site or
other linear feature which will ultimately be used in CZAEM to orient the viewer. Typically, only
major roads will be digitized, to ease the interpretation of the model without making the CZAEM
screen too cluttered.
Point Set Features
A point set represents a set of wells, homes, locations of known water levels or other point features
which will ultimately be used in CZAEM to orient the viewer. Point sets are also used as the basis
for the creation of WELL analytic elements. The user should create a point set for each wellfield,
set of known heads or other set of features.
56
-------
Measurement Units
CZAEM Units
CZAEM works in a "dimensionless coordinate space"; that is, the modeler is responsible for
maintaining a consistent set of units throughout a project. Before beginning a project, you should
determine which unit of length to use (GAEP supports feet and meters) and which time unit to use
(GAEP assumes days). Once the set of units has been decided upon, the modeler must maintain
them in all his work.
UTM Coordinates
GAEP, however requires that geographic input data be entered in standard georeferenced
coordinates. For convenience and consistency, the developers of the W/iAEM package have
selected the UTM coordinate system for use in GAEP. The UTM system breaks the globe into
"zones," each of which has a central meridian. Within each zone, a set of coordinates is assigned
to each point, measured (in meters) relative to the central meridian (UTM X value at the meridian
is 500000) and to the equator (UTM Y value at the equator is 0). Over a relatively small
geographic area, the coordinates can be considered to be Cartesian. This provides a simple X-Y
coordinate system in data units for modeling work.
Conversion of Latitude-Longitude to UTM Coordinates
For users' convenience, a facility is included in GAEP for converting between UTM coordinates
and latitude-longitude coordinates, given the number of the UTM zone. This feature simplifies the
process of locating digitizer origin points.
Users who are unfamiliar with the UTM coordinate system may wish to investigate this topic
further. A good reference into geographic coordinate systems should be available in your local
library.
Coordinate Origins
The GAEP user is required to set two different types of coordinate origins when managing a
modeling project with GAEP. The first, the "Model Origin," is consistent throughout all GAEP
and CZAEM operations and should be set at the beginning of the project. The second, the
"Digitizer Origin," is set whenever a map is mounted on the digitizing tablet. It is important that
57
-------
the modeler not confuse the meanings of these two terms.
As discussed above, CZAEM works in a dimensionless coordinate space. GAEP, however,
assumes that its data are measured in world coordinates which may have a base length unit of
meters or feet. A conversion may be performed by GAEP to change the X-Y coordinates from
meters to feet when creating an analytic element data file if the modeler desires.
Digitizer Origin
Whenever a map is mounted on the digitizer, the GAEP user must tell GAEP how to convert
digitizer coordinates (typically measured in inches or millimeters internally) to "real-world"
coordinates. This task is performed by the use of "Digitizer Origins", which are points marked on
the map for which the world coordinates are known. GAEP requires that the user locate these
points with the digitizer and then enter the world coordinates from the keyboard (Origin command
in the Digitize menu).
A pair of digitizer origin points are required each time the map is mounted on the digitizer and
must fit on the digitizer surface (of course). This means that for a small digitizer, several sets of
digitizer origins may be required on each topographic map. A convenient way to enter these is by
the use of the UTM conversion utility (see above), converting the latitude-longitude points on the
edges of the map, and writing the corresponding UTM coordinates in the map margin.
Model Origin
The numeric values of world coordinates are often so large (particularly in the Y direction for UTM
coordinates) that numerical errors can occur in CZAEM if the geographic coordinates are simply
used directly from GAEP. To prevent this, GAEP requires that the modeler enter a "Model Origin"
in world coordinates that will be the "zero point" for CZAEM's computations. The model origin
should be maintained throughout a particular modeling project, and is included in the CZAEM
input data files prepared by GAEP. To select a Model Origin, simply choose a point near the
model study region and record its world coordinates. It is particularly convenient to mark and label
this point on your maps as well. When prompted by GAEP for a Model Origin, enter the
appropriate world coordinates. GAEP will convert the X-Y coordinates of all element features to
either feet or meters (depending on the "metric output files" option setting) from the model origin.
Digitizer Origin
Whenever a map is mounted on the digitizer, the GAEP user must tell GAEP how to convert
digitizer coordinates (typically measured in inches or millimeters internally) to "real-world"
coordinates. This task is performed by the use of "Digitizer Origins", which are points marked on
the map for which the world coordinates are known. GAEP requires that the user locate these
points with the digitizer and then enter the world coordinates from the keyboard (Origin command
in the Digitize menu).
58
-------
A pair of digitizer origin points are required each time the map is mounted on the digitizer and
must fit on the digitizer surface (of course). This means that for a small digitizer, several sets of
digitizer origins may be required on each topographic map. A convenient way to enter these is by
the use of the UTM conversion utility (see above), converting the latitude-longitude points on the
edges of the map, and writing the corresponding UTM coordinates in the map margin.
Construction of No-Flow Boundaries
Implementation of aquifer heterogeneity is considered to be beyond the scope of CZAEM. GAEP,
however, allows the modeler to use the "method of images" to create a single, linear no-flow
.boundary in a model. This feature provides a simple, foolproof technique for generating a "worst
case" simulation of this specific case of aquifer heterogeneity. The intent of this feature is to allow
the modeler to determine whether additional modeling with a model code which supports
heterogeneous aquifers is necessary. If the image result is substantially different from the "no
image" result, additional modeling work with a more powerful program is warranted. For a
mathematical discussion of the Method of Images, see Groundwater Mechanics (Strack , 1989)
pages 28-29.
The image line can be used to simulate the aquifer interface between a highly permeable alluvial
channel or glacial outwash zone and a much less permeable aquifer outside the channel or outwash.
The user is cautioned that this feature allows only an analysis of two extremes; one with no aquifer
heterogeneity, and another with a no-flow boundary. The analysis of image-based, no-flow
boundaries should be used only to evaluate the potential effect of a nearby aquifer inhomogeneity
in order to determine whether further modeling with a more sophisticated modeling program is
warranted.
The steps in creating an image region were illustrate earlier in the 'WTzAEM tutorial. The user first
selects the "image origin" — the center of the "image axis." This center also is the center of the
rain circle for imaged solutions. The "left-hand rule" is used to determine the region in which
elements are to be imaged, or mirrored across the image line. This mirroring of elements across
the image line generates a mathematical no-flow boundary along the image line.
Once the user has defined the image line, GAEP prompts for a point on the perimeter of the rain
circle, the recharge rate in the rain circle and the head at the reference point. GAEP will
automatically locate the reference point far from the study region along the image axis.
This method for creating no-flow boundaries requires that a regional solution with recharge be
used. Use of uniform flow in an imaged solution is not allowed.
59
-------
Use of GAEP
GAEP is designed to be very easy to understand and use. Commands are set out in a logical set of
menus, accessible either from the keyboard or with the mouse. To simplify the discussion of the
various GAEP commands, this manual will first describe the use of the menu system, and the use
of certain "special keys," which are used consistently throughout GAEP.
Menus
The basic GAEP screen is shown in Figure 23. The screen menu shows the current settings of
GAEP options and a list of available commands are arrayed across the top of the screen. GAEP
commands are accessed by either pressing the "hot" key for the command (shown on the menu in
upper case - in red on color monitors) or by placing the mouse cursor on the desired command and
pressing the left mouse button. For example, the F (file), A (aquifer), D (digitize), E (element), U
(utilities), O (options) and Q (quit) "hot" keys are available either by pressing the appropriate letter
on the keyboard or by using the mouse.
File Aquifer Digitize Element Options Utility Quit
GAEP Release 1.0
25 January 1994
Indiana University Current Directory:
SPEA Groundwater Modeling Laboratory C:\WHAEM\DAT
Current Map File:
WhAEM Version
Current Element File:
Memory available: 309920
Option Settings:
Metric Output Files: NO
Use digitizer: YES
Video graphics mode: COLOR
Figure 23 GAEP Main Menu.
60
-------
Special Keys
In addition to the "hot" keys available from the GAEP menu system, some additional special keys
are used:
Fl Key - Help!
Whenever you are at a GAEP menu, press the Fl key to display a help screen which describes the
commands available from that menu. This feature is available from menus only, not while entering
data or while digitizing.
Esc Key - Go Back!
At all times while using GAEP, the Esc key (upper left on most keyboards) aborts the current
command and returns to the previous menu. This key is active at all times while using GAEP to
abort the current command. The Esc key is commonly known as the "escape" key.
F3 Kev - I'm Done!
While performing certain functions, the F3 key is used to tell GAEP that an operation is complete
(examples are completing entry of a stream with the digitizer or creating line-sink elements). This
key is used consistently as the "Successful Completion" command while performing data entry.
Page Up / Page Down / Home - Zoom In / Zoom Out
Whenever graphics screen is active, the Page Up and Page Down keys zoom in or out on the area
around the center of the screen. To zoom in, press Page Down. To zoom out, press Page Up. The
Home key zooms out to a window which shows the entire extent of the digital map data presently
loaded. GAEP remembers where you have set the view window for future commands.
Arrow Keys - Scrolling
Whenever graphics screen is active, the left, right, up and down arrow keys scroll the graphics
image. GAEP remembers where you have set the view window for future commands.
Data Entry Considerations - Free-Form Input
Whenever more than one value is requested by GAEP, the user may use either spaces, commas or
other common punctuation characters (except decimal points and minus signs, of course). All
punctuation marks are ignored. For example, if the latitude of a point 37 degrees, 30 minutes, 15
seconds is requested, the user may enter:
61
-------
37 30 15 or 37,30,15
Procedure for Using GAEP
The user should apply the following procedure when using GAEP on a modeling project:
Create the Digital Map (Digitize menu)
This step involves the entry of hydrologic features using the digitizer, including the points of
known head, and the maintenance of the digital map using GAEP's editing features.
Save the Digital Map (File menu)
The digital map is saved to a Digital Map File (see above). It is STRONGLY recommended that
the user perform save operations regularly during map creation and editing to prevent loss of data.
Always back up your digital map files to floppy disks for safekeeping!
Create Analytic Elements (Element menu)
Once a digital map is complete, the modeler uses GAEP to create analytic elements (line-sinks and
wells) and to set the various aquifer properties.
Save the Analytic Element Data File (File menu)
After element creation is finished, the user saves the analytic element file for use in CZAEM.
GAEP also allows the modeler to re-load the analytic element file into GAEP for editing and
modifications.
Detailed GAEP Command Descriptions
The remainder of this document describes the commands available from each menu in the GAEP
program in detail, organized by menu. Each GAEP command menu is described separately.
File [F] - File operations
Selects the "File" menu (see Figure 24).
62
-------
Aquifer [A] - Set aquifer properties
Selects the "Aquifer" menu (see Figure 25).
Digitize [D] - Digital map creation/editing
Selects the "Digitize" menu (see Figure 26).
Element [E] - Analytic element creation/editing
Selects the "Element" menu (see Figure 28).
Options [O] - Set GAEP options
Selects the "Options" menu (see Figure 29).
Utility [U] - Run GAEP utilities
Selects the "Utilities" menu (see Figure 30).
Quit [Q] - Exit program
Exits GAEP. If changes to the digital map file or analytic element data file have been made but
not saved to disk, GAEP will ask if the user really wishes to leave the program. Answer "YES"
to exit GAEP without saving to disk.
File Menu
This menu provides access to the various file management facilities in GAEP.
New [N] - Clear program memory
Clears all GAEP's memory, both the digital map and any elements which have been created. This
is functionally equivalent to leaving GAEP and re-entering the program.
ReadDM [R] - Read a digital map
Reads a digital map file from the current directory (defaults to x:\WHAEM\DAT after installation,
where x: is the drive where you installed W/zAEM). GAEP will prompt for the name of the file
to be read from disk. Type the name of the file, followed by the key, or press to
abort the command. If a digital map is already loaded, this command will add the newly read map
to the already loaded map. To remove the previously loaded map, either re-start GAEP or use the
New command (see above). Note: A digital map file MUST be loaded before certain functions
can be performed. It is impossible to create or to view analytic elements unless a digital map is
loaded.
63
-------
Mew ReadDM ffiriteDM LoadElem SaveElem Map ChangeDir setDnve Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help return to previous menu 25/Jan/94 12:56 PM
Figure 24 GAEP File Menu.
WriteDM [W] - Write a digital map
Writes the current digital map to a digital map file in the current directory (defaults to x :
\WHAEM\DAT after installation, where x: is the drive where you installed W/zAEM). GAEP will
prompt for the name of the file to be written. The current version of GAEP does not enforce file
extensions, so any filename and extension are allowed, but it is useful to choose a consistent
naming convention. The "unofficial" standard for digital map file names is to use the extension
.DM. Type the name of the file, followed by the key, or press to abort the
command.
LoadElem [L] - Read an analytic element data file
Reads an analytic element data file from the current directory (defaults to x:\WHAEM\DAT after
installation, where x: is the drive where you installed WTzAEM). GAEP will prompt for the name
of the file to be read from disk. Type the name of the file, followed by the key, or press
to abort the command.
64
-------
SaveElem [S] - Write an analytic element data file
Writes the currently defined analytic elements and aquifer properties to an analytic element data
file in the current directory (defaults to x:\WHAEM\DAT after installation, where x: is the drive
where you installed W/jAEM). GAEP will prompt you for the Model Origin for the project (see
above discussion of the Model Origin). GAEP will then prompt for the name of the file to be
written. The current version of GAEP does not enforce file extensions, so any DOS filename and
extension is allowed, but it is useful to choose a consistent naming convention. The "unofficial"
standard for analytic element data file names is to use the extension .DAT. Type the name of the
file, followed by the key, or press to abort the command.
Note: If you have defined any GAEP digital map features for "backgroind" maps in CZAEM (that is, curves or point
sets), GAEP will automatically put the background map into your CZAEM-compatible analytic element datafile. You
will be asked if you wish to leave out the map features (this is discouraged). If so, answer "YES" to the question,
"Exclude background map?"
Map [M] - Create a background map file
Writes any currently defined "background map features" (that is, curves and point sets) to a
CZAEM-compatible data file in the current directory (defaults to x:\WHAEM\DAT after
installation, where x: is the drive where you installed W/iAEM). GAEP will prompt you for the
Model Origin for the project (see above discussion of the Model Origin). GAEP will then prompt
for the name of the file to be written. The current version of GAEP does not enforce file
extensions, so any DOS filename and extension is allowed, but it is useful to choose a consistent
naming convention. The "unofficial" standard for analytic element data file names is to use the
extension .MAP.
ChangeDir [C] - Change the current directory
Changes the default working directory for GAEP. GAEP will prompt for the new working
directory, which will be the "permanent" directory for future GAEP sessions until changed again.
The default working directory is x:\WHAEM\DAT where x: is the drive where you installed
W/iAEM, after installation. Since this command does not change the INITAEM.DAT file (see
CZAEM documentation), use of this command is discouraged. If you do wish to change the default
directory, please remember to edit INITAEM.DAT so that the DATA directory prefix matches the
new directory.
Quit [Q] - Return to the main menu
Aquifer Menu
This menu allows the specification of aquifer properties.
Base [B] - Set the aquifer base.
Sets the elevation of the base of the aquifer. GAEP will request the elevation, in units consistent
with the desired project units. Enter the value and press . Press to abort.
65
-------
Base .Thick Eermeability pQrosity JUniforaiFlow reference Rain Quit
Aquifer Parameter Settings: Current Directory:
C:\WHAEM\DAT
Base Elevation: 0.0 Current Map File:
Thickness: 100.0
Permeability: 100
Porosity: 0.2 Current Element File:
Uniform flow: QO 0
alpha: 0.00 Memory available: 297632
Reference point: X 0.0
Y 0.0
Head 0.0
Rain element is not defined Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help return to previous menu 25/Jan/94 06:39 PM
Figure 25 GAEP Aquifer Menu.
Hick [T] - Set the aquifer thickness.
Sets the thickness of the aquifer. GAEP will request the thickness, in units consistent with the
desired project units. Enter the value and press . Press to abort.
Permeability [P] - Set the aquifer hydraulic conductivity.
Sets the permeability (hydraulic conductivity) of the aquifer. GAEP will request the permeability,
in units consistent with the desired project units (feet per day or meters per day are commonly
used). Enter the value and press . Press to abort.
Porosity [O] - Set the aquifer porosity.
Sets the porosity of the aquifer. GAEP will request the porosity, as a fraction between 0.0 and 1.0.
Enter the value and press . Press
-------
project units. Enter the value and press . Next the orientation of the discharge vector (in
degrees) is requested. Enter the value and press . Press to abort.
NOTE: When uniform flow is used, the uniform flow discharge rate is .not recomputed when the user changes the
aquifer geometry or permeability. It is up to the user to recompute the uniform flow discharge when quifer properties
or geometries are adjusted.
Reference [F] - Set the reference point.
Allows the user to define a reference point for the model (see the modeling discussion in the main
W/jAEM manual). GAEP will request the location of the reference point in UTM coordinates.
Enter the location and press . GAEP will then request the reference head. Enter the value
and press . Press to abort at any point.
NOTE: When using imaging, the location is determined by GAEP; only the head will be requested.
Rain Circle [R] - Define a rain recharge circle.
Allows the user to define the rain (areal recharge) element (see the modeling discussion in the
WAAEM tutorial). GAEP will display the digital map and request the center of the rain circle.
Select the center for the rain circle and press the left mouse button. GAEP then requests that a
point on the perimeter of the circle be selected! Select the point and press the left mouse button.
GAEP then requests that a recharge rate be entered, in the same units as the permeability. Enter
the value and press . Press
-------
Qrigin Stream Jleads Curve joints Yiew Edit Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode:COLOR
help return to previous menu 25/Jan/94 12:58 PM
Figure 26 GAEP Digitize Menu.
for the UTM coordinates of the first point. Next, GAEP will repeat the process for the second
origin point. Once the digitizer origin is set, it is possible to digitize hydrologic and background
map features. Note: If the digitizer option is set to "mouse markup" mode, you may digitize
onscreen at any time after a digital map is loaded. See the "Options" menu discussion of the digizer
setting.
Stream [S] - Digitize a stream
Allows the user to digitize the location of a stream. GAEP will prompt for the name of the feature;
enter the name and press . GAEP will now design an abbreviation for the name and allow
you to change it by entering a new abbreviation (up to 9 characters), or just press to use
the one GAEP designs. Now, the graphics screen will show the extent of the digitizing area. Select
points on the stream from the map and press the first digitizer button to enter the stream's location.
When complete, press , or press to abort digitizing the feature.
While digitizing, GAEP will display the points entered as dark blue "plus" signs. The points are
connected into a stream feature after the user presses the F3 key. Once entered, the stream will be
displayed as a continuous curve.
68
-------
Heads [H] - Digitize heads on a stream
Allows the user to digitize the locations of known elevations on a stream. GAEP will show the
digital map onscreen. Select the feature to have heads added and press the left mouse button.
Now, the graphics screen will show the extent of the digitizer. Select points where elevation
contours cross the stream from the map and press the first digitizer button. GAEP will request the
elevation. Enter the elevation in units consistent with the desired model units (feet or meters).
Repeat the process for all possible elevations. When complete, press , or press to abort
digitizing the heads. If an elevation is entered incorrectly, you should press . GAEP will now design an abbreviation for the name and allow you to
change it by entering a new abbreviation (up to 9 characters), or just press to use the one
GAEP designs. Now, the graphics screen will show the extent of the digitizer. Select points on
the curve from the map and press the first digitizer button to enter the curve's location. When
complete, press , or press . GAEP will now design an abbreviation for the name and allow you to change it by
entering a new abbreviation (up to 9 characters), or just press to use the one GAEP designs.
Now, the graphics screen will show the extent of the digitizer. Select points from the map and
press the first digitizer button to enter each point's location. When complete, press , or press
to abort digitizing the feature.
View [V] - View the digital map
Shows the current digital map on the screen. Streams which have heads associated will be in bright
blue (or white on monochrome systems). Streams without heads are dark blue (or dashed).
Background map curves and points are dark purple (or dotted).
Edit [E] - Edit the background map
Enters the Digitize/Edit submenu (see Figure 27).
Quit [Q] - Return to the main menu
69
-------
Digitize/Edit Submenu
Join Vjew Delete Rename Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help
-------
button. GAEP then requests a new name; enter the new name or simply hit to keep the old
name. GAEP then requests a new abbreviation; enter the new abbreviation or simply hit
to keep the old abbreviation.
Delete [D] - Delete a feature
Allows a digital map feature to be deleted. GAEP requests that the feature be selected from the
graphics screen. Select the feature with the mouse and press the left mouse button. GAEP asks,
"Are you sure?" Press Y to delete the feature.
Quit [Q] - Return to the Digitize menu
Element Menu
This menu allows for creation of analytic elements.
Linesink [L] - Create line-sink elements
Allows the user to create line-sinks for a stream. GAEP requests that the stream be selected from
the graphics screen. Select the stream with the mouse and press the left mouse button. GAEP will
then show a screen which shows only the feature selected. Select the end points of line-sink
elements, pressing the left mouse button at each point. Line-sink elements will appear with heads
computed by GAEP. Press to complete the creation of line-sinks or to abort. To
create additional line-sinks for the s tream, repeat the procedure.
Note: the first line-sink endpoint entered will not appear on the screen until a second point is entered. It does not matter
to CZAEM whether line-sinks are entered "heading upstream" or "heading downstream," but the user might wish to
work in a consistent manner, creating line-sinks in the same direction for all features.
Well [W] - Create well elements
Allows the user to create wells for a point set. GAEP requests that the "background map point set"
(see the "Digitize/Point" command discussion above) be selected from the graphics screen. Select
the point set with the mouse and press the left mouse button. GAEP will then show a screen which
shows only the feature selected. Select the location of a well and press the left mouse button.
GAEP will prompt for the discharge (in units consistent with the ones determined by the user) and
radius of the well. Repeat the procedure for each well. Well elements will appear onscreen. Press
to complete the creation of wells or to abort.
Delete [D] - Delete elements associated with a digital map feature
Allows elements to be deleted. GAEP requests that the feature be selected from the graphics
screen. Select the feature with the mouse and press the left mouse button. GAEP asks, "Are you
sure?" Press Y to delete all elements associated with the digital map feature.
71
-------
Linesink Well Delete Userwindow Image Yiew Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help
-------
(in the same units as permeability) and press . Finally, the head at the reference point will
be requested. Enter the reference head and press . To abort the imaging procedure, press
at any time in the procedure.
Please see the discussion of imaging in the introductory portion of this manual.
View [V] - View the digital and element map
Shows the current digital map on the screen, with all elements. Streams which have heads
associated will be in bright blue (or white on monochrome systems). Streams without heads are
dark blue (or dashed). Background map curves and points are dark purple (or dot-dashed). Line-
sink elements which have been created by the user will appear in green, those which were created
by imaging (if any) in white. The image line (if any), rain circle and user window will be shown
in green.
Quit [Q] - Return to the main menu
Options Menu
All of the commands listed here set program option settings and perform no direct action on the
digital maps or analytic elements in use. Once a setting is modified, it is saved to the disk and
remains set until changed, regardless of whether GAEP is re-started.
UnitConv [U] - Set the unit conversion mode
Allows the user to change the way unit conversions are handled when reading or writing analytic
element files. Three modes are supported; "feet-to-meters," "none" and "meters-to-feet." If your
digital map was digitized in UTM, you would use the "meters-to-feet" setting to make analytic
element files in feet or the "none" setting for analytic element files in meters.
When this command is selected, GAEP cycles through the unit conversion options. Continue until
the desired mode is shown on the screen.
Digitizer [D] - Select the digitizer mode
The digitizer mode can have any of three states: "Digitizer", "Mouse Markup", and "Direct." In
"Digitzer Mode," all coordinate input is performed using the presently configured digitizer. In
"Mouse Markup Mode," the mouse may be used to enter points directly on a displayed digital map.
It is NOT RECOMMENDED for general purpose data entry. "Direct" mode allows the user to
enter the coordinate locations (in world coordinates) from the keyboard, such as for wells and
piezometers of known location.
NOTE: Previous versions of GAEP used this command to support the "keyboard digitizer" for users who had no
digitizers. This function is now handled by the digitizer driver as the "keyboard"protocol; see the Appendix C Tablet
Installation Guide for details.
73
-------
UnitConv Digitizer YideoMode Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help return to previous menu 251 Jan/94 01:01 PM
Figure 29 GAEP Options Menu.
VideoMode [V] - Switch video modes
Switches the video from color to monochrome or back.
Quit [Q] - Return to the main menu
Returns to the main menu. All option settings will be maintained for future GAEP sessions until
changed again.
Utility Menu
This menu provides several useful utility functions.
UTM/LatLong [U] - Enter the UTM/Latitude-Longitude conversion utility
Enters the UTM/LatLong submenu (see Figure 31).
Dos [D] - Run MS-DOS
Starts the MS-DOS shell specified in the COMSPEC environment variable (see MSDOS manual).
74
-------
To return to GAEP, type EXIT at the DOS prompt.
Quit [Q] - Return to the main menu
Htm/latlong Dos Quit
Current Directory:
C:\WHAEM\DAT
Current Map File:
Current Element File:
Memory available: 309920
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help return to previous menu 25/Jan/94 01:02 PM
Figure 30 GAEP Utility Menu.
UTM / Latitude-Longitude Utility
This menu allows the user to convert UTM coordinates to latitude-longitude coordinates or
latitude-longitude coordinates to UTM coordinates, given a known UTM zone.
Zone [Z] - Enter the UTM zone
Allows the user to define the UTM zone for conversions. GAEP will request that the zone number
be entered. Enter the desired zone (look in the lower left corner of the topographic map for the
zone number for you map) and press . Press to abort.
75
-------
Zone lalitdue loMgitdue Htm .Quit
Geographic Location: Current Directory:
C:\WHAEM\DAT
UTMZone: 0 Current Map File:
Latitude: 00 d 00 m 00 s
Longitude: 00 d 00 m 00 s Current Element File:
UTM coordinates: X 0.0 Memory available: 297632
Y 0.0
Option Settings:
Unit Conversion: FT->M
Digitizer Mode: DIGITIZER
Video graphics mode: COLOR
help return to previous menu 25/Jan/94 06:56 PM
Figure 31 GAEP Lattidude-Longiture/UTM Utility Menu.
Latitude [T] - Enter the latitude, convert to UTM
GAEP requests the latitude of the point to be converted. Enter the degrees, minutes and seconds
of latitude of the point and press (do not use any text labels such as "degrees" in the data
entry, enter only the numerical values in the specified order). Press to abort. GAEP will
report the UTM coordinates of the point entered in latitude-longitude coordinates.
Longitude [O] - Enter the longitude, convert to UTM
GAEP requests the longitude of the point to be converted. Enter the degrees, minutes and seconds
of longitude of the point and press (do not use any text labels such as "degrees" in the data
entry, enter only the numerical values in the specified order). Press . Press to abort. GAEP will report the latitude-longitude coordinates of the
point, based on the UTM zone specified.
76
-------
Quit [Q] - Return to the utility menu.
77
-------
REFERENCE
There are two levels of help in CZAEM: 1) brief, context-sensitive help available by typing a
command followed by a question mark (e.g., "trace ?"); and 2) more extensive help available by
typing "help" while in one of the modules. This chapter of the W/zAEM manual is a printout of
these more detailed help files located in the \WHAEM\CZAEM\HELP directory.
CZAEM Main Module
\\\ Module=MAIN MENU Level=0 RoutIne=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(XI,Yl,X2,Y2)777]
-------
(TOL)
CHECK MODULE:
TRACE MODULE:
maximum number of iterations is set.
The program will abort the solving process whenever it detects that
control points are too close together. It will tell you which control
points on which elements are too close. You can then modify your
input accordingly.
Display of all control points that are within TOL from one another,
if the conditions applied at these points are in conflict. The value
of TOL will be used by the solve routine. Thus, you can force the
program to solve problems with control points that are very close
together (but do not coincide) by setting TOL to zero in this
command. Whenever control points have been detected that are too
close, they will be marked on the layout. If the optional argument
TOL is not entered, then the program will display the current value
prior to checking control points.
This module is used for retrieval of data in numeric form.
This module is used to generate pathlines. This module also allows
access to the capture zone routines.
CONTOURING MODULE:
STAND-ALONE
COMMANDS:
This module is used for grid scalar values for contouring.
(NX)
Contour rectangular grid of NX intervals horizontally. The default
of the function to be contoured is piezometric head.
Will cause a layout to be displayed.
79
-------
Will cause all screens to be cleared. On printers the previous page
will be ejected.
Suspends execution of the program. Availability of commands
depends upon the memory available with the program installed.
The extended memory versions do not support a meaningful use of
the PAUSE command.
Initiates contour plotting. The program will display the maximum
and minimum levels encountered in the grid to be displayed, and
will ask you to supply starting levels and a contour increment.
After you enter this command, the machine will prompt you for a
title of less than 17 characters. If you press ENTER without text,
the current title is displayed. The title will be displayed on the plots
and on printed output.
(FACTOR)[11,12]
This command allows you to reduce the plots in size and to move
it to a new position. FACTOR must be <=1. The values of II and 12
are also <=1, and represent the distance as a fraction of the display
width and height over which the plot is to be moved right and up,
respectively.
[(X1 ,Y1 ,X2,Y2)]/]
sets the window for subsequent plotting. X1,Y1 and
X2,Y2 represent the coordinates of lower left and upper right
corners of the window, respectively. Typing WINDOW ALL will
change the window size to be large enough to include all elements
in the model. IT IS IMPORTANT THAT YOU ENTER THE
WINDOW BEFORE ENTERING AQUIFER DATA, BECAUSE
MANY DEFAULT VALUES ARE SET AS A FRACTION OF
THE WIDTH OF THE CURRENT WINDOW.
80
-------
Resets all values in the program to what they are when the program
is loaded into memory.
(FILENAME)
Will cause the program to read further input from the file FILE-
NAME; the last command in the file must be SWITCH CON.
Stops the program.
SERVICE MODULES: /
To store or retrieve solutions and grids in binary form; to subtract
grids.
Data retrieval by means of the cursor. Changing of well and
line-sink data.
Input and output re-direction.
To direct graphics output, and to set plotting attributes.
-------
AQUIFER
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
^TFAO(FACTOR)
Set the factor by which time entries must be multiplied in order to
obtain units as used in the coefficient of permeability and rainfall,
e.g., if permeability is in m/s and times are given in days, then
FACTOR=3600*24=86400.
Reset all parameters in the aquifer module to their default values.
Return control to the main menu.
GIVEN
The word "given" refers to the already known or "given" average extraction or infiltration rate of
a hydrologic feature such as wells, line-sinks, or ponds.
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
measured over the entire thickness of the aquifer.
(X,Y,RADIUS,INFILTRATION RATE)
Specify the rainfall entering the top of the aquifer. INFILTRATION
RATE is positive for water entering the aquifer and is measured in
volume per unit area (L/T). Only one region of infiltration due to
rainfall may be specified.
Reset all parameters in the module GIVEN to their default values.
Return control to main menu.
REFERENCE
The reference point is a point in the domain where the head is "known" or assumed. It is usually
treated like the average head in the study area placed at some arbitrary point outside the domain
of interest. It is entered directly from the main command line.
\\\ Module=MAIN MENU Level=0 RoutIne=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
WELL
Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
entered between brackets. The program will expect continued entry
of head specified wells until a command word is entered. You can
return to the well input menu by entering COMMAND.
(NUMERICAL VALUE)
Specifying a factor by which each discharge value which is as typed
in is to be multiplied to convert it to the units used in the program;
e.g., to convert from GPM to cubic feet/min.
Cause all parameters in the module WELL to be reset to their
default values.
Return control to main menu.
LINESINK
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
be done with these line-sinks entered as elements of given strength. In general, boundaries of
specified head will be modeled more accurately when more elements are used to discretize these
boundaries.
The command words are:
Enter line-sinks of constant rate of extraction. The program
prompts for input of coordinates and extraction rate as follows:
(X1 ,Y 1, X2,Y2, EXTRACTION RATE)[[label]]
The program will expect such lines of input to continue until
another command word is entered. An optional label may be
entered between brackets. You may return to the linesink menu by
entering COMMAND.
Enter line-sinks of constant head. The program prompts for input
of coordinates and head as follows:
(XI, Yl, X2,Y2, HEAD)[[label]]
The program will expect such lines of input to continue until
another command word is entered. An optional label may be
entered between brackets. You may return to the linesink menu by
entering COMMAND.
[EL.NR.]
[/]
[TOL]
Change the location and head of the head-specified linesink
element number EL.NR.
This command allows the user to specify whether linesink elements
will be linked together into strings of elements. When line-sinks are
linked into strings, the start nodes of linesink elements on the same
string will be moved to the location of the closest end node on the
string. Typing any command or specifying a start node which is at
least the TOLERANCE distance away from all end nodes on the
string will cause a new string of linesink elements to be started.
This command allows the tolerance for connecting linesink
elements into strings of elements to be displayed and/or specified.
If the start node of the linesink element that is being entered is
within the TOLERANCE distance of the end node of any other
87
-------
linesink element on the string, then the element will be linked to the
closest end node on the string.
This command causes all parameters in the linesink module to be
reset to their default values.
Returns control to the main menu.
CHECK
Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[ (X1,Y1,X2,Y2)777]
-------
The RANGE command, common to nearly all of the check routines, is explained here. This
command sets the type and range of elements to which all subsequent data refers.
The range command words are:
(TYPE,NR1 )[NR2]
TYPE is a word signifying the type of element to be considered (the
types are for WELL: GIVEN, HEAD, or TIME; for LINESINK:
GIVEN or HEAD). The value NR1 is the starting number of the
element to be considered (this number is determined by the
sequence in which the elements of TYPE were entered).
For example, if 10 wells are entered as type=GIVEN, the com-
mand:
RANGE GIVEN 4,8
specifies that subsequent data should pertain to the wells entered as
4,5,6,7, and 8. If NR2 is omitted, the default is to set NR2 equal to
NR1.
Other command words are:
Display the conditions at the control points for each of the modules.
In general, the numbers in the last two or three columns should be
identical. Differences in these values indicate inaccuracies for a
given problem. Module specific information for this command is
available by invoking in the appropriate module.
(X,Y)[MEASURED HEAD]
Display the head computed at point (X,Y). If an optional value for
MEASURED HEAD is entered, the program will also display the
measured head and the difference between the measured and
computed heads. This command is useful for comparing observed
and predicted heads at observation points. An input file with this
command, followed the X,Y location and the associated measured
head is recommended for model calibration. The input file is
subsequently read using the command in the module
SWITCH.
(X,Y)
Display the complex potential, i.e. the value of the potential and the
89
-------
value of the stream function, at point (X,Y).
(X,Y)[DELTA]
Display the components Qx and Qy of the discharge vector (units of
LA2/T) for a point (X,Y). If the optional DELTA is entered,
numerical differentiations are performed over an interval DELTA
in the x and y directions. The numerically computed discharge
components are displayed below those obtained analytically.
-------
The command words are:
Set the piezometric head as the function to be used for future GRID
commands.
Set the potential as the function to be used for future GRID
commands.
Set the stream function as the function to be used for future GRID
commands.
Note that the contours of the stream function may show branch cuts
generated by features that remove water from the aquifer.
The routine will issue a warning if the window has not been set.
Set the phreatic surface as the function to be used for future GRID
commands.
Set the aquifer bottom as the function to be used for future GRID.
(X1 ,Y1 ,X2,Y2)
Set the window boundary, where X1,Y1 are the coordinates of the
lower left corner and X2,Y2 those of the upper right comer.
(NUMBER OF INCREMENTS)
Set the number of increments in the X direction of the grid to be
contoured.
Use this command to specify a title of your problem. The program
will prompt you to enter a title of less than 17 characters, or to press
ENTER to display the current title.
Use this command to have the program write the grid information
to a file in the form: x,y,item, where item is whatever has been
gridded the last time. This file can then be used later in a contouring
91
-------
package, or other program for display. The program will prompt for
the name of the file to be written to.
Returns control to main menu.
PLOT
The plot module is entered AFTER a grid has been generated in the GRID module. In the plot
module contours are generated based on the grid data.
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777]
-------
The command words are:
Set device for numerical output. If OUTPUT=CON, then numerical
data will be printed on graphics screen.
Provide access to the routine for setting tracing parameters.
-------
[TOL]
without any arguments will display the current window settings.
Sets the tolerance. Entering the command without argument will
cause the current value to be displayed. The default value is
automatically scaled to the current window.
Will save a plan view of pathlines on a file. The program will
prompt for a filename. If this command is not given, then upon
entering the routine TRACE, any plot of pathlines produced will be
saved on a file with the name CZDFLT.PTH. Because the tracing
routine uses the same memory locations as those reserved for
storing grids, the current grid will be saved upon entering trace
under the filename SLDFLT.GRD.
Will read a plan view of pathlines and produce the plot when either
the command PLOT or the command LAYOUT is entered. The
program prompts for a filename (which may be CZDFLT.PTH)
I i i I i I i I i I i i i I I i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
CAUTION: This will occur, for example, if the program attempts
to read a grid file for a pathline plot.
To help avoid this, it is suggested that the files be given easily
distinguishable extensions, such as GRD, and PTH.
i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
Provides access to the routine for setting plot options.
Provides access to the routine for re-directing I/O.
Return to the main menu.
CURSOR ACTIVITY
Entering one of the commands PLOT, LAYOUT, or START, will cause the cursor to appear on
the graphics screen. Once the cursor appears, move it to the desired position and enter any one of
94
-------
the commands listed below. Appearance of the cursor may be suppressed, while retaining the
graphics capabilities of the TRACE module. This is done by entering the command CURSOR OFF
from the main TRACE INPUT menu. In this case, enter manually a pair of coordinates before
giving the appropriate command.
[ELEV]
Print the base elevation.
Print the aquifer upper boundary elevation. For unconfined flow,
this boundary is the phreatic surface elevation; for confined flow it
is the elevation of the confining layer.
Print the values of both the potential and the stream function.
Print coordinates of cursor location.
Record set of coordinates for lower left corner of the new WIN-
DOW. Move the cursor to the upper right corner of the window and
press enter. The screen will be cleared and the new WINDOW will
be activated.
Start tracing of pathline, beginning at cursor location at elevation
ELEV. If ELEV is omitted, the pathline starts at either aquifer top
or phreatic surface.
Set the tolerance. Entering this command followed by moving the
cursor will cause a temporary line to be drawn from the position of
the cursor at the time the command was given to the current cursor
position. Pressing the enter key will cause the tolerance to be stored
as the length of the displayed line. This value will be displayed on
the screen. If you press ENTER without moving the cursor, the
current tolerance will be displayed. Note that the value set for the
tolerance using the cursor is temporary; it is valid only while in
cursor mode. To re-set the tolerance permanently, enter the value
after issuing the command MENU.
(# LINES)[ELEV]
95
-------
Generate pathlines starting at the radius of the well at which the
cursor is located. The pathlines will be generated by backwards
tracing at equally spaced intervals around the well. The elevation
at which the pathlines are started may be specified, otherwise the
pathlines will be started at the bottom of the aquifer.
(/)
Enables or disables backward tracing. Once set, backward tracing
will remain into effect until disabled by this command. Note that
the clock will run backward while backward tracing occurs.
Causes the command line to be displayed on the screen.
-------
The capture zones are created by moving the cursor close to any discharge well and then typing in
either SUBZONE or TIMEZONE.
The capture zone routines use a set of internal buffers to create the various capture zone diagrams.
These buffers are set large enough to handle all practical cases. If the buffers are filled during the
creation of capture zones, messages will be displayed to suggest how the capture zones may be
created. The following variables may be changed in order to create the capture zones properly:
- the minimum step, maximum step, and neighborhood in the tracing module,
- the window size,
- the number of time zones being created,
- the initial number of pathlines used to define the capture zones (NLINE).
The command words are:
Create subzones for the well at which the cursor resides.
Create time zones for the well at which the cursor resides. The user
will be queried for the following information:
ENTER [TIME STEP][MAXIMUM TIME], OR
EDRAW LAST TIME ZONES, OR EFAULT TIME ZONES, ORXIT
MINIMUM AND MAXIMUM TIMES FOR CAPTURE ZONE: (mln time) (max time)
Specify the times for which time zones are to be created. Entering 'R' will redraw the last time
zones that were created. The minimum and maximum time it takes to reach the boundary of the
time zone is also displayed as an indication of what times to enter.
Display the percentage of water going to the well from all the
different subzone sources for the last well at which subzones were
calculated. The subzones are displayed in the counter-clockwise
direction, starting with the first subzone which begins at an angle
greater than zero degrees.
(LINES)
Specify the initial number of pathlines used to define the capture
zones. There must be enough pathlines specified so that water
from two different sources will not both flow between any two
adjacent pathlines. The default number of lines is adequate for
97
-------
[(X 1 ,
most cases.
Print coordinates of cursor location.
Print the elevation of the base of the aquifer at the cursor location.
Print the surface elevation at the cursor location.
I ,X2,Y2)///]
Change the window size. When the window size is changed, the
capture zones which were in the previous window are redisplayed.
Typing WINDOW ALL will change the window size to be large
enough to include all elements in the model. Specifying WIN-
DOW PUSH will save the window setting to a stack in memory.
Specifying the WINDOW POP command will retrieve the window
settings that were saved via the PUSH option, in the order in
which the window settings were saved. Typing WINDOW without
any arguments will display the current window settings.
Change the window size by specifying the new window coordi-
nates via the cursor. This command records the set of coordinates
for lower left corner of the new WINDOW. Move the cursor to the
upper right corner of the new window and press enter. Note that
this command is not allowed when CURSOR is OFF.
Allows access to the switch routines.
Clear the screen and layout the elements. This command erases
the BSAVE or BREAD file which was currently being used.
Causes the command line to be redisplayed.
Causes this help file to be displayed.
98
-------
Returns control to the tracing routines.
[[VELOCITYFACTOR]/]
Specify whether to create timezones for the mean velocity of the
contaminant, or for the front of the contaminant. Specifying
will create timezones for the mean travel time of the
contaminant. Specifying along with the VELOCITY
FACTOR will create timezones for the front of the contaminant
using the specified velocity factor. The velocity factor specifies
how much faster the front of the contaminant is traveling than the
mean travel time; e.g. specifying a VELOCITY FACTOR of 2.0
will mean that the front of the contaminant travels twice as fast as
the average ground-water flow. Note that pathlines in the tracing
module are created using these velocities also.
(# LINE)
Generate pathlines which go backwards in time and start at the
radius of the well at which the cursor is located. The pathlines will
be generated by tracing backwards at equally spaced intervals
around the well at the bottom of the aquifer.
[COLOR1 ] [2] [3]
Specify the color numbers for the different line types:
- COLOR 1 is the color of the subzone dividing streamlines
- COLOR2 is the color of the timezones
- COLORS is the color of the subzone outer envelopes
(FILE)
Specify a file to which all subsequent capture zone boundaries will
be saved. This file will be closed when any of the following
occurs: a new file is specified to save capture zones, a capture
zone file is read in via the BREAD command, the PAGE com-
mand is specified, or the RETURN command is specified. The
default file to which capture zones are saved is CZDFLT.CZ.
(FILE)
Read and plot the capture zone boundaries which were previously
saved to a file via the BSAVE command.
(FILE)
Specify a file to which capture zone buffers will be saved. This
command will save all internal buffers which are used by this
99
-------
(FILE)
capture zone program to create capture zones. The buffers will only
be saved for the last well for which capture zones were created,
prior to issuing the CSAVE command. This command is to be used
in conjunction with the CREAD command to redraw capture zones
without having to recalculate the capture zone buffers.
Read the capture zone buffers for a well that was previously saved
via the CSAVE command. After reading these buffers, capture
zones may be redrawn for the well.
Note: If any changes to the model of the aquifer was made between
when the capture zone buffers were saved via CSAVE and when
they were read via CREAD, then the capture zones will not be an
accurate representation of the true capture zones for the aquifer.
The subzones and time zones should be recalculated whenever
changes are made to the model of the aquifer. Likewise, the buffers
were created for a particular window, and hence the window setting
which is used during the CREAD command must be the same
window setting which was used during the CSAVE command.
Note: The capture zone buffers which are read via the CREAD
command were created using either the mean velocity of
ground-water or using the front velocity of the contaminant.
Therefore, when buffers are read via the CREAD command, the
variables specified by the FRONT command will also be changed
to correspond to the values which were used when the capture zone
buffers were created.
NOTE1:
The time zone and subzone capture zone routines all use the same buffers, and these buffers are
filled for the well whose capture zone is currently being drawn. Therefore, it is faster to create all
subzone and time zone contours for the given well before creating capture zones for the next well.
NOTE2:
If the buffers are getting full while creating time zones, then try creating just one time zone instead
of multiple time zones at once. Changing the window size may also result in being able to create
time zones more efficiently. Typing in the PAGE command or creating a capture zone at a new
well will clear the internal buffers so that new time zones may be created at the well.
NOTES:
When a capture zone file is read via the BREAD command, the file remains open and any further
capture zones which are created will also be saved in the same file.
100
-------
NOTE4:
Calling the PAGE command will erase the capture zone file that is currently being filled.
NOTES:
The window size is not saved in the capture zone file. When the file is read it uses the current
window size.
NOTE6:
The layout of the elements is not saved in the capture zone files.
NOTE7:
The capture zones will not be created correctly if there is a stagnation point too close to the well.
If the capture zone doesn't look correct, then the WGENERATE command may be used to get an
approximate idea of where the stagnation points should be for the well. If the stagnation point is
too close to the well, then either the window size may be changed or the tracing maximum step size
may be changed to correctly create the subzones.
CURSOR
\\\ Module=MAIN MENU Level=0 RoutIne=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777] [FILE]
(NUMBER OF POINTS)
cur
\\\ Module=CURSOR Level=l Routine=INPUT 777
(TOL)
This module allows access to cursor activity.
The command words are:
Allow setting device for numerical OUTPUT. If OUTPUT=CON,
then numerical data will be printed on graphics screen
101
-------
Display the layout prior to appearance of the cursor.
Display both the layout and contour plots prior to the appearance of
the cursor.
Clear both the text and graphics screens.
Will cause the cursor to be displayed.
(TOL)
Set the tolerance for subsequent MOVE commands (see below) to
TOL.
Return to the main menu.
Once in cursor mode, enter any of the following commands.
Print the head.
Print both the potential and stream function. (Stream function has
meaning only if the infiltration rate at position of cursor is zero).
Print the two components of the discharge.
Print coordinates of the cursor location. A line is drawn from the
previous point to the current point. You may prevent this line from
being drawn by pressing enter twice. You should do this prior to
entering any other command.
[(X1,Y1,X2,Y2)///]
Change the window size. When the window size is changed, the
capture zones which were in the previous window are redisplayed.
102
-------
Typing WINDOW ALL will change the window size to be large
enough to include all elements in the model. Specifying WINDOW
PUSH will save the window setting to a stack in memory. Specify-
ing the WINDOW POP command will retrieve the window settings
that were saved via the
PUSH option, in the order in which the window settings were saved.
Typing WINDOW without any arguments will display the current
window settings.
Record set of coordinates for lower left corner of the new window.
Move the cursor to the upper right corner of the window and press
enter. You may abandon by entering any key other than enter; the
cursor will then reappear ready for the next command. The screen
will be cleared and the new window will be activated.
-------
[STRENGTH]
maximum of 50 are recorded. Plotting is deferred until the com-
mand ENDISCHARGE is given. A scale factor is then computed
such that the largest vector is equal to SIZE. Both tangential and
normal components of the discharge vector are recorded on the
OUTPUT device, as well as the coordinates of the points in question
and the scale factor. The normal component of discharge is taken
positive if pointing to the left with respect to an arrow pointing from
the first to the second point entered.
Terminate the entry of points for plotting normal components of
discharge.
Make it possible to change the locations of endpoints of linesinks,
and to change their strength. Move the cursor within a distance of
TOL from an endpoint, then move the cursor to the desired location
and press ENTER. If you don't move the cursor, the current linesink
attributes will be displayed. You may change the strength of the
linesink by entering the strength as an optional parameter. If you
move the cursor, a line will appear connecting the old and new
locations of the endpoint. To facilitate updating of input files, both
the old and new locations of the endpoint as well as the linesink
number (according to its position in the input file) are printed on the
OUTPUT device. It is recommended to echo OUTPUT to a file by
the use of the SWITCH command.
[DISCHARGE]
Use this command to move wells. It works in the same way as
LSMOVE above.
Mark a point on the screen with a marker; this marker is temporary
and will not appear on subsequent plots.
Return to the main menu.
104
-------
PSET
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777] [FILE]
(NUMBER OF POINTS)
pset
\\\ ROUTINE SET PLOT MODE 777
(NUMBER) (7)
This module will set the program for various types of hardware
configurations.
(CODE)
Selects one out of 4 color palettes; CODE may be 1,2,3, or 4.
Causes all subsequent plots to be sent to the printer. Note that the
resolution on most printers will be much higher for plots generated
in this way than by using screen dumps.
Causes all subsequent plots to be sent to the screen.
Please see read.me file.
Transfers control back to main program unit.
105
-------
STOP
\\\ Module=MAIN MENU Level=0 Routine=INPUT 777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2,Y2)777] [FILE]
(NUMBER OF POINTS)
stop
106
-------
APPENDIX 0. TABLET CONFIGURATION GUIDE
This appendix contains a description of the digitizing tablet driver provided as part of GAEP and
details about configuration. A description of the program TAUTEST, a tool for testing digitizer
installation is included. A number of digitizers have been tested with GAEP. The configurations
for those tablets are also included here.
Introduction
This document describes the configuration procedure for the program GAEP (as implemented for
the W/zAEM product). It also describes the configuration of the digitizer driver, the devices
supported, and gives setup instructions for several digitizers. It is intended that this manual will
be expanded as additional digitizers are tested with GAEP.
Installation of GAEP
GAEP installation was performed as part of the install procedure for the W/zAEM product. A
default GAEP configuration based upon no digitizer being connected to the system was installed.
If you have not yet installed GAEP, refer to the installation information in the W/?AEM
documentation. This manual presumes that you have already installed GAEP on your system.
Digitizer Configuration
The tablet driver included as part of the W/zAEM product supports four common digitizing tablet
protocols:
• Formatted ASCII protocol (digitizer writes digitizer inches)
• SummaGraphics MM ASCII protocol
• SummaGraphics MM Binary protocol
• SummaGraphics Bit Pad Plus protocol
It is expected that one or more of these protocols will work with nearly any digitizer on the market.
Both binary mode and ASCII mode protocols are available. A support program, TABTEST, is
provided to assist with digitizer configuration and testing. It is easy to configure GAEP for any of
the protocols by modifying the \WHAEM\TABSETUP.BAT file. Protocol selection for GAEP is
done by setting the environment variable TABLET and appropriately setting the port parameters
in the \WHAEM\TABSETUP.BAT batch file. For example,
107
-------
SET TABLET=ASCII COM1 12124
MODE COM l:9600,E/7,2
will select a formatted ASCII mode tablet connected to port COM1 at 9600 bps, even parity, 7 data
bits and 2 stop bits. The tablet is 12" x 12" and has a 4-button puck. TABTEST will modify the
file TABSETUP.BAT once digitizer configuration is complete. The following several pages
document the available protocols.
How Do I Configure My Digitizer for GAEP?
General
Digitizer configuration can be a difficult and frustrating process. GAEP uses a digitizer driver
which was written particularly for its use. As part of the digitizer driver, a support program
(TABTEST) is provided which may be used to ensure that the digitizer and the software are
communicating properly.
This section outlines the basic steps which the user needs to execute to configure the GAEP
digitizer driver for a particular digitizer.
Note:
If your system does not have a digitizer, you may wish to coifigure GAEP to use a Microsoft (or compatible) mouse or
to request direct keyboard entry of coordinates. The digitizer driver supports these also; no hardware isting is required.
See the appropriate protocol discussion in the "Digitizer Protocols" section of this appendix. TABTEST does not test
these protocols.
Step-By-Step
To configure and test your digitizer with the GAEP digitizer driver, the following step- by-step
process should be performed. Detailed discussions of the options are to be found in the "Digitizer
Protocols," "Program TABTEST" and "Tested Configurations" sections of this appendix.
• If necessary, unpack and install your digitizer and cable it to your computer. Place the
digitizer puck on the active digitizing surface.
• Locate your digitizer's reference manual and have it handy before beginning the configura-
tion process.
• Examine your digitizer manual and the "Digitizer Protocols" section of this appendix. Select
a protocol to be used. Our experience has shown that the ASCII protocols are easier to test,
108
-------
because the tablet transmits readable characters. In some cases, however, only binary
protocols are available. If you have a SummaGraphics Bit Pad Plus, only the Bit Pad Plus
protocol may be used.
• Select a transmission baud rate, parity and character format for serial communications.
Recommended settings are:
ASCII Protocols: 9600 bps, even parity, 7 data bits, 2 stop bits
Binary Protocols: 9600 bps, no parity, 8 data bits, 1 stop bit
• Configure your digitizer for the desired protocol. This will require that you follow the
instructions in your digitizer manual carefully. This may include setting of hardware
switches in your digitizer or running a DOS-based configuration program, or both.
Note: Software-based digitizer configuration
Depending on your digitizer model, you may need to run a program from DOS to set up the digitizer protocol. Beaware
that many applications which use your digitizer may transmit configuration information prior to their execution. If a DOS
command is needed to configure your digitizer, you will need to manually modify the file TABSETUP.BAT in the
WAAEM installation directory to execute the proper configuration command, once the digitizer communications have
been tested.
Note: Tested Digitizers
Some digitizers have already been fully tested with GAEP. Check the "Tested Configurations"section of this appendix
to see if your digitizer has been previously tested.
• Once the digitizer has been configured, run the program TABTEST to test the communica-
tions with the digitizer. Set the TABTEST driver, port (COM1 or COM2) and the
communications settings (baud rate, etc.), then use the "CommTest" command (see the
"Program TABTEST" section of this appendix for details).
• Once the communications test is successful, use TABTEST's "DriverTest" command
to ensure that the digitizer driver is working and that puck coordinates are being read in
inches from the lower left comer of the digitizer (see the "Program TABTEST" section of
this appendix for details).
• Once all of the tests are complete, use the command to leave TABTEST, and tell
TABTEST to write the current settings to TABSETUP.BAT in the \WHAEM installation
directory.
• If your digitizer required the execution of a DOS program to set up the digitizer configura-
tion, you will now need to modify the TABSETUP.BAT file in your \WHAEM installation
directory to include the command(s) required to configure the digitizer. Place the
configuration commands at the beginning ofTABSETUP.BAT. The TABSETUP.BAT file
is run automatically prior to each execution of GAEP.
109
-------
Digitizer Protocols
The following section of this appendix contains a technical description of the protocols supported
by GAEP. Four digitizer protocols, plus the use of a Microsoft (or compatible) mouse or direct
keyboard entry are supported.
Formatted ASCII Protocol
This is the favored protocol, when possible. The digitizer transmits the puck coordinates directly
as a formatted string, in inches from the lower left corner of the tablet. This protocol is the easiest
to debug in most cases. (CalComp digitizers usually refer to this protocol as mode 8.)
To configure the formatted ASCII driver, the TABLET environmental variable should be set as
follows:
SET TABLET=ASCII <# BUTTONS>
where:
• Is the serial port used. Only COM1 and COM2 are supported.
• Is the size of the digitizer in the X direction, in inches.
• Is the size of the digitizer in the Y direction, in inches.
• <# BUTTONS> Is the number of buttons on the puck.
The configuration used is described in CalComp 2500 Series User's Manual as follows:
Mode Run - the tablet transmits continuously
Commands Enabled (optional - the driver doesn't use commands)
Transmit rate 50 points per second. Can be set as you like; this setting works well
with the 2500 and an 80386 or 80486 system.
Line feed Disable (required)
Out of proximity Enable (optional)
Margin data Disable (optional)
Resolution 1000 Ipi (optional - may be set as desired)
ASCn Format 8 Required. This format transmits "XXXX.X YYYY. Y CC TO ",
where XXXX.X is the X position in inches, YYYY.Y is the Y position in inches, CC is a
two-character code for the puck button currently pressed, TO is the digitizer status setting
(ignored) and CR is a carriage return.
110
-------
NOTE:
The number of significant digits transmitted is dependent on the digitizer resolution; check your digitizer manual for
details.
Baud rate (optional)
Data bits 7 (required)
Stop bits (optional)
Parity (optional)
Echo Disabled
Handshake Enabled
Cursor buttons Set according to your hardware
Beeper Disabled (GAEP beeps when points are entered)
SummaGraphics MM Binary Protocol
To configure the SummaGraphics MM Binary driver, the TABLET environmental variable should
be set as follows:
SET TABLET=MMBINARY <#BUTTONS>
where:
Is the serial port used. Only COM1 and COM2 are supported.
Is the size of the digitizer in the X direction, in inches.
Is the size of the digitizer in the Y direction, in inches.
<# BUTTONS> Is the number of buttons on the puck.
Is the number of lines per inch on the digitizer.
The configuration used is described in the SummaGraphics MM1812 Technical Reference as follows
(all settings are selected by switches on the tablet, except as noted):
• Mode Stream - the tablet transmits continuously. Driver sends the "@"
command to select this mode.
• Transmit rate 110 points per second. This setting works well with the MM 1812 and
an 80386 or 80486 system.
Report format Binary
Resolution 500 Ipi
Baud rate 9600
Data bits 8 (required)
Stop bits 1 (required)
Parity Odd
111
-------
SummaGraphics MM ASCII Protocol
To configure the SummaGraphics MM ASCII driver, the TABLET environmental variable should
be set as follows:
SET TABLET=MMASCD <# BUTTONS>
where:
Is the serial port used. Only COM1 and COM2 are supported.
Is the size of the digitizer in the X direction, in inches.
Is the size of the digitizer in the Y direction, in inches.
<# BUTTONS> Is the number of buttons on the puck.
Is the number of lines per inch on the digitizer.
The configuration used is described in the SummaGraphics MM 1812 Technical Reference as follows
(all settings are selected by switches on the tablet, except as noted):
• Mode Stream - the tablet transmits continuously. Driver sends the "@"
command to select this mode.
• Transmit rate 110 points per second. This setting works well with the MM 1812 and
an 80386 or 80486 system.
Report format Binary
Resolution 500 Ipi
Baud rate 9600
Data bits 7
Stop bits 2
Parity Odd (by default)
SummaGraphics Bit Pad Plus Protocol
The Bit Pad Plus uses a different protocol than other SummaGraphics digitizers. It is not
configurable, so simply setting the TABLET environmental variable is sufficient:
SET TABLET=BITPAD <# BUTTONS>
where:
• Is the serial port used. Only COM1 and COM2 are supported.
• Is the size of the digitizer in the X direction, in inches.
• Is the size of the digitizer in the Y direction, in inches.
• <# BUTTONS> Is the number of buttons on the puck
112
-------
• Is the number of lines per inch (200 on the Bit Pad Plus)
The configuration used is described in the SummaGraphics Bit Pad Plus Technical Reference. No
tablet settings are required.
• Baud rate 9600
• Data bits 8 (required)
• Stop bits 1 (required)
• Parity Odd (required)
Microsoft Mouse
GAEP supports the use of the mouse for digitizing in two manners. First is the "Mouse Markup"
mode, which allows the user to add features to an existing map. This mode is available regardless
of the TABLET setting and is selected by commands in GAEP. An alternative use of the mouse is
to use the absolute mouse cursor position as a digitizer, so that the position of the mouse on the
screen can be scaled as you desire. It is anticipated that this has little use in the context of GAEP
and its application is discouraged.
The use of the mouse as a "digitizer" is supported in the digitizer drivers and is documented here
only for completeness.
To use the absolute mouse position for digitizing, the TABLET environmental variable should be
set as follows:
SET TABLET=MOUSE
• No options are required.
Note:
You will need to set the digitizer origin in GAEP, just as if you had a digitizer (see GAEP manual).
Keyboard Data Entry (For Systems Without Digitizers)
GAEP supports digital map data entry without the use of a digitizer by making the user's keyboard
into a "digitizer." The user can use a quadruled sheet (8 squares to the inch vellum works quite
well), and trace the features to be digitized onto the sheet, along with geo-referenced origin locations.
When GAEP requests data from the "digitizer," the following message appears on-screen :
[KEYBOARD DIGITIZER; FI - BUTTON I, F2 - BUTTON 2]
113
-------
To "digitize," you simply press the Fl key for "tablet button 1" or the F2 key for "tablet button 2."
The digitizer driver will then ask for the coordinates of the point to be entered from your grid sheet.
The default is to enter the data in inches, but you may use any grid coordinates you wish.
To use the keyboard data entry method for digitizing, the TABLET environmental variable should
be set as follows:
SET TABLET=KEYBOARD
where:
• Is the size of the digitizer in the X direction, in inches.
• Is the size of the digitizer in the Y direction, in inches.
Note:
If the X and Y are maximum vahes, the driver defaults to a 20" (X direction) by 24" (Y direction) space, calibrated in
inches. You are not restricted to any particular grid coordinate system for data entry; for example, you might have ajrid
sheet calibrated in grids that was 500 grids on the X axis by 400 on the Y axis. To tell the driver this, you can set the
TABLET environmental variable as follows:
SET TABLET=KEYBOARD 500 400
and GAEP will work properly, showing the grid sheet extent while performing data entry.
Note:
You will need to set the digitizer origin in GAEP, just as if you had a digitizer (see GAEP manual).
Program TABTEST
Testing serial communications devices such as digitizers can be time consuming and frustrating due
to a lack of standards and because digitizers usually make no directly visible signals. TABTEST is
designed to facilitate this process by allowing you to experiment with parameter settings and
instantly monitor the effect. The configuration process with TABTEST is subdivided into three
steps:
1. Configure the driver for the communications port (COM1 or COM2), baud rate, parity,
and number of data and stop bits ("Port" and "Baud, Etc." commands).
2. Establish basic communications with the digitizer ("CommTest" command).
3. Test proper functioning of the digitizer driver ("DriverTest" command).
114
-------
Fl-Help
F6-Tablet
Tablet:
Port:
Baud:
F2-CommTest
F7-Port
F3-Dri veriest
F8-Baud,etc.
FlO-Quit
Formatted ASCII (+x.xxx,+y.yyy,bb,s)
COM1
9600 Parity: Even Data bits:? Stop bits:2
Current setup lines for TABSETUP.BAT:
MODE: COM1:9600,E,7,2
SET TABLET=ASCII COM1 12 12 4
TABTEST vO.2- Graphics Tablet Setup/Test Program
Copyright (c) 1994 V.A. Kelson
Figure 32 TABTEST Menu
Commands
Help
• Displays a help screen
CommTest
• Tests low-level communications, displaying results one byte at a time.
• For ASCII mode drivers fASCII and MMASCID The tablet response as discussed in the
driver reference (see above) will be printed on the screen, with continuous update. The user
should be able to easily read the puck coordinates and button status. When test is complete,
press to return to the TABTEST menu.
• For binary mode drivers (MMBINARY and BITPAD) The tablet response will be displayed
as five 2-digit hexadecimal numbers, continuously updated. Since it is not easy to read these
numbers to ensure they are correct, the user can only look to ensure that they remain constant
115
-------
when the puck is motionless on the tablet and that they change in a regular pattern when the
puck is moved. When test is complete, press
• Tests the tablet driver. Prints puck location in inches from the lower left corner of the tablet.
• The results of this test are the same for all tablet drivers. The current puck coordinates will
be printed and continuously updated. The values printed should be the x- and y- coordinates
of the puck in inches from the lower left comer of the tablet.
• During the use of the driver test, you may need to experiment with the selected protocol
selection and the number of lines per inch (LPI) setting. A common problem is that the test
looks fine, but the number of inches is off by some factor. This usually indicates that the LPI
setting is wrong.
• Again, be aware of the "Common Blunders" mentioned above.
Tablet
• Selects the tablet driver. A menu is displayed, showing the digitizer driver choices.
Note:
After the tablet driver choice is made, it is usually necessary to set the port parameters (see below).
Port
• Chooses the serial port (COM1 or COM2) where the tablet is connected. A menu of choices
is printed; choose the appropriate port, depending upon your cabling.
Note:
After the port choice is made, it is usually necessary to set the port parameters (see below).
116
-------
Baud, etc.
Sets up the COM port parameters for serial communications with the digitizer. The choices are:
• Baud rate: 300,600,1200,2400,4800,9600 bps
• Parity: Odd, Even or None
• Data Bits: 7 or 8
• Stop Bits: 1 or 2
Notes:
• Binary mode tablet drivers (MMBINARY and BITPAD) MUST use 8 data bis. ASCII mode
drivers (ASCII and MMASCII) usually use 7 data bits.
• It is usually best to use the default communication parameters for your tablet; check your
tablet's reference manual for details.
• Many digitizer models require the execution of a DOS program which sends command to the
tablet, initializing the protocol, baud rate, etc., prior to use. It is most critical that the user
check the digitizer manual THOROUGHLY and perform any necessary tasks prior to
attempting to use TABTEST.
Quit
• Exits TABTEST. TheuserispromptedwhetherornottoupdateTABSETUP.BAT. If the
user wishes, the updated settings for the TABLET environmental variable and a DOS MODE
command will be placed in \WHAEM\TABSETUP.BAT for execution at startup for future
GAEP and TABLET runs.
Note:
TABTEST does not retain any of the settings internally after termination. Each TABTEST run starts "from scratch."
In most cases, the user will only run TABTEST once when installing WAAEM, and the tablet will work thereafter.
117
-------
Tested Configurations for Various Digitizers
During the development of GAEP, several digitizers have been tested, and their known configura-
tions are shown here. This information is provided as an example of the way one might use the
system.
Note:
If you have a digitizer not shown here and you are successful in using the device with one of the supported protocols,
it would be much appreciated if you would record the tablet and software configuration information and send it to the
author:
Vic Kelson cc:=> Steve Kraemer
Indiana University USEPA/RSKERL
SPEA Groundwater Modeling Laboratory P.O. Box 1198
PV418 Ada, OK 74820
Bloomington IN 47405 Internet: kraemer@ad3100.ada.epa.gov
Internet: vkelson@ucs.indiana.edu
CalComp 2500 n2"x!2";>
Soft switch settings:
Bankl: 00000001
Bank 2: 10110001
Bank 3: 01101000
Bank 4: 00100001
Bank 5: 01110010
Environmental variable settings:
SET TABLET=ASCII COM1 12 12 4 for COM1: connection
MODE COM 1:9600,E,7,2
SET TABLET=ASCII COM2 12 12 4 for COM2: connection
MODE COM2:9600,E,7,2
CalComp 9500 (48"x36"l
Soft switch settings:
Area l(left to right): 0100000001101000
Area 2 (Port A): 001000101010
Area 4 (top to bottom): 0010000000011000
Environmental variable settings:
SET TABLET=ASCII COM1 48 36 16 for COM1: connection
MODE COM 1:9600,E,7,1
SET TABLET=ASCII COM2 46 36 16 for COM2: connection
MODECOM2:9600,E,7,1
CalComp DrawingBoard II (48"x36")
118
-------
Soft switch settings (18 switches per bank, 2 banks):
Bank A: 110 001 Oil 100 100 000
BankB: 001 001 010 000 000 000
Environmental variable settings:
SET TABLET=ASCII COM1 48 36 4 for COM1: connection
MODE COM 1:9600,EJ,2
SET TABLET=ASCn COM2 48 36 4 for COM2: connection
MODE COM2:9600,E,7,2
It is expected that other DrawingBoard II models will use the same soft switch settings.
Calcomp Estimat G6" x 30"^
Soft Switch Settings
Bank A: 110001011100100000
Bank B: 001001010000000000
Environmental variable settings:
SET TABLET=ASCII COM1 36 30 4 for COM1: connection
MODE COM 1:9600,E,7,1
SET TABLET=ASCII COM2 36 30 4 for COM2: connection
MODE COM2:9600,E,7,1
SummaGraphics SummaSketch Professional (18"xl2")
DIP Switch Settings (0=off, l=on):
Bankl: 1 1 100000
Bank2: 00000000
Bank3: 00000000
Environmental variable settings:
SET TABLET=MMBINARY COM1 18124 500 for COM1: connection
MODE COM 1:9600,0,8,1
SET TABLET=MMBINARY COM2 18124 500 for COM2: connection
MODE COM2:9600,O,8,1
SummaGraphics Bit Pad Plus (12"xl2")
No switch settings required.
Environmental variable settings:
SET TABLET=BITPAD COM1 12 12 4 200 for COM1: connection
MODE COM1:9600,N,8,1 (REQUIRED)
SET TABLET=BITPAD COM2 12 12 4 200 for COM2: connection
MODECOM2:9600,N,8,1 (REQUIRED)
Summagraphics SummaGrid IV (24"x36")
(Model CEM2436). Dip switch settings (0=off, l=on):
Bank A: 10011100
Bank B: 10000100
Bank C: 00000010
Environmental variable settings:
119
-------
SET TABLET=MMBINARY COM1 24 36 4 500 for COM1: connection
MODE COM 1:9600,0,8,1
SET TABLET=MMBINARY COM2 24 36 4 500 for COM2: connection
MODE COM2:9600,O,8,1
It is expected that these settings will be the same for other SummaGrid models.
Note:
Before using the SummaGrid, the DOS program supplied with the digitizer must be run to set the appropriate tablet
protocols. Check the tablet manual prior to testing.
Appendix C Bibliography
CalComp Inc. CalComp 2500 Series User's Manual.
Kelson, V. A, H.M. Haitjema, and S.R. Kraemer, GAEP: A Geographic Preprocessor for
Groundwater Flow Modeling, Hydrological Science and Technology, 8(l-4):74-83,
1993.
Summagraphics Corporation. Bit Pad Plus User's Guide. 1987.
Summagraphics Corporation. MM1812 Data Tablet (MM Format) Technical Reference. 1986.
Appendix C Acknowledgments
This document describes the operation of the GAEP program with a number of different hardware
configuration systems. The GAEP program was developed by Vic Kelson at the SPEA Groundwater
Modeling Laboratory, Indiana University. The author acknowledges Phil DiLavore for his work on
the initial design of GAEP. Thanks also to Jack Wittman of IU and Dr. Stephen R. Kraemer of the
USEPA for assistance and guidance with this work.
CalComp is a trademark of CalComp Inc.
Summagraphics, SummaSketch, MM1812, and Bit Pad Plus are trademarks of Summagraphics
Corporation.
Microsoft and MS-DOS are trademarks of Microsoft Corporation 80386 and 80486 are trademarks
of Intel, Inc.
120
-------
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/R-94/174
September 1994
v>EPA CZAEM User's Guide
Modeling Capture
Zones of Ground-Water
Wells Using Analytic
Elements
-------
EPA/600/R-94/174
September 1994
CZAEM USER'S GUIDE
MODELING CAPTURE ZONES OF GROUND-WATER WELLS USING ANALYTIC ELEMENTS
O.D.L. Strack - Principal Investigator
E.I. Anderson
M. Bakker
W.C. Olsen
J.C. Panda
R.W. Pennings
D.R. Steward
University of Minnesota
Minneapolis, Minnesota 55455
CR-818029
Project Officer
Stephen R. Kraeraer
Processes and Systems Research Division
Robert S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
This study was conducted in cooperation with
Indiana University
School of Public and Environmental Affairs
Bloomiugton, Indiana 47405
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 74820
Printed on Recycled Paper
-------
NOTICE
The work presented in this document has been funded wholly (or in part) by the U.S. Envi-
ronmental Protection Agency under cooperative agreement CR-818029 to Indiana University. It
has been subjected to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
All research projects making conclusions or recommendations based on environmentally related
measurements and funded by the U.S. Environmental Protection Agency are required to participate
in the Agency Quality Assurance Program. This project did not involve environmentally related
measurements and did not involve a Quality Assurance Plan.
The material introduced in the User's Guide should be fully understood prior to the application
of the computer program CZAEM to field problems. Both the creation of the conceptual aquifer
model and the interpretation of this program's output require an understanding of the Analytic
Element Method and its implementation in CZAEM. Interpretation of the output generated by
the CZAEM program is the sole responsibility of the user.
-------
FOREWORD
EPA is charged by Congress to protect the Nation's land, air, and water systems. Under a
mandate of national environmental laws focused on air and water quality, solid waste management,
and the control of toxic substances, pesticides, noise, and radiation, the Agency strives to formulate
and implement actions which lead to a compatible balance between human activities and the ability
of natural systems to support and nurture life.
The Robert S. Kerr Environmental Research Laboratory is the Agency's center for expertise
for investigation of the soil and subsurface environment. Personnel at the Laboratory are responsi-
ble for management of research programs to: (a) determine the fate, transport, and transformation
rates of pollutants in the soil, the unsaturated and the saturated zones of the subsurface environ-
ment; (b) define the processes to be used in characterizing the soil and subsurface environments
as a receptor of pollutants; (c) develop techniques for predicting the effect of pollutants on ground
water, soil, and indigenous organisms; and (d) define and demonstrate the applicability of using
natural processes, indigenous to the soil and subsurface environment, for the protection of this
resource.
The Capture Zone Analytic Element Model (CZAEM) is a practical, PC-based ground-water
analysis tool that allows for the definition of the areas contributing recharge to pumping wells,
including the influence of rivers, streams, and other surface water bodies. The solution is based on
a new technique for ground-water modeling known as the analytic element method. Capture zone
definition is fundamental in the design of remediation systems for source containment or pump-
and-treat of contaminated ground water, and also in the delineation of protection areas around
drinking water wells.
^LJZ^
Clinton W. Hall, Director
Robert S. Kerr Environmental Research Laboratory
111
-------
ABSTRACT
The computer program CZAEM is designed for elementary capture zone analysis, and is based
on the analytic element method. CZAEM is applicable to confined and/or unconfined flow in
shallow aquifers; the Dupuit-Forchheimer assumption is adopted. CZAEM supports the following
analytic elements: uniform flow, uniform infiltration over a circular area, wells, and line-sinks. The
line-sinks can be used to simulate streams and the boundaries of lakes and rivers.
The program will generate and plot the envelopes of capture zones, the boundaries of capture
zones corresponding to different times, dividing streamlines including stagnation points, stream-
lines, and piezometric contours.
A tutorial is provided to introduce the user to CZAEM and consists of two parts, each with
three examples. Part A is concerned with introducing the user to the primary capabilities of
the program along with elementary modeling techniques. Part B is aimed at advanced modeling
techniques and commands. It is explained at the end of part B how to obtain hardcopy output
from the program.
IV
-------
TABLE OF CONTENTS
Notice ii
Foreword iii
Abstract iv
Figures vii
Acknowledgment viii
Introduction 1
Background 1
The Analytic Element Method — 1
The Computer Program CZAEM ! 2
Installation 3
CZAEM TUTORIAL - PART A
Example 1. Uniform Flow with a Well 4
Entering the program CZAEM 5
Entering aquifer data 6
Solution and generation of contour plots 8
Entering and analyzing the proposed well 10
Exiting the program CZAEM 12
Example 2. Well near a River 12
Entering line-sinks 13
Calculating the head at any point 15
Entering the well 16
Saving a solution 17
Influence of the reference point 17
Determining a well's water source using pathlines .. 18
Example 3. Critical Pumping Level for a Well 21
Creating capture zones 21
Interpretation of Figure 3.1 24
Determining a well's water source using capture zones 24
Summary of Part A 26
CZAEM TUTORIAL - PART B
Example 4. Contaminant Pumpout System 29
Obtaining results using CHECK 31
Determining capture zones for multiple wells 35
Well water travel times 36
Moving wells in graphics mode 38
Example 5. Data Manipulation and Model Refinement 40
Using input files 40
Saving grid files 41
Comparing grids 42
Obtaining results using the cursor in CAPZONE 44
Validity of Solutions 45
-------
Example 6. Data File and Graphics Control 45
Accessing multiple data files contiguously 45
Entering rainfall 46
Window manipulation and saving capture zone and time zone boundaries .46
Obtaining a hardcopy of graphical output 49
References 52
Command Summary 53
VI
-------
FIGURES
Figure 1.1 Site map for Example 1 4
Figure 1.2 Conceptual model of the aquifer 5
Figure 1.3 Contour plot of the phreatic surface 9
Figure 1.4 Piezometric contours with the well present 11
Figure 1.5 The water beneath the field does not enter the well 12
Figure 2.1 Site map for Example 2 13
Figure 2.2 Conceptual model of the aquifer 14
Figure 2.3 Plot of piezometric contours, well not present 16
Figure 2.4 Contours with the well present, reference point at (-2000,4000) 18
Figure 2.5 Contours with the well present, reference point at (1,1) 19
Figure 2.6 Contours with the well present, reference point at (—2000,5000) 20
Figure 2.7 Several pathlines from the well generated by < WGENERATE > begin at the
well and end at the line-sink, showing that the well does capture river water 21
Figure 3.1 Capture zones generated for a well discharge of 1500 m3/day 23
Figure 3.2 Capture zones generated for a well discharge of 1000 m3/day 25
Figure 3.3 Capture zones generated for a well discharge of 970 m3/day 27
Figure 3.4 Capture zones generated for a well discharge of 990 m3/day 27
Figure 4.1 Site map for Example 4 29
Figure 4.2 Conceptual model of the aquifer 30
Figure 4.3 Existing conditions: uniform flow with reference head of 129.84 meters at
(-750, -875) 31
Figure 4.4 Existing conditions: uniform flow with reference head of 133.586 meters at
(-2000, -2000) 33
Figure 4.5 Contours with the well present 35
Figure 4.6 Subzones drawn for the well 36
Figure 4.7 Subzone curves for both wells 37
Figure 4.8 Twenty year time zones for Wells 1 and 2 38
Figure 4.9 Twenty year time zones for the fronts, with a front velocity factor of 1.1 39
Figure 5.1 Case of Example 3 with refined line-sinks 43
Figure 5.2 Refined line-sink grid minus the original line-sink grid 44
Figure 6.1 Model of existing conditions (plot; d 10) 46
Figure 6.2 Model of proposed conditions (plot; 780 5) 47
Figure 6.3 Well-field subzones 49
vii
-------
ACKNOWLEDGMENTS
The authors express their appreciation to the volunteer group of "guinea pigs" who did the
beta testing of the code. The code and documentation were much improved after technical reviews
provided by Mr. Paul van der Heijde, Dr. Jeffrey Johnson, and Dr. Randall Charbeneau. Ms.
Chursey Fountain provided editorial review of the documentation.
This report was typeset by AMS-l^Y, the TfpC macro system of the American Mathematical
Society.
vin
-------
INTRODUCTION
The Wellhead Analytic Element Model (WhAEM) package, a capture zone delineation tool for
wellhead protection and pollution containment, is the result of a cooperative agreement between the
USEPA, the Indiana University at Bloomington, and the University of Minnesota at Minneapolis.
The package consists of two executables: the first contains the graphical pre-processor GAEP along
with a manual describing its use; the second contains the computer program CZAEM, Capture
Zone Analytic Element Model. The latter program's intended use is for the modeling of problems
where the flow is generated by few (say 50) features (small-scale problems), either together with
GAEP, or by itself. The integrated WhAEM package is described in USEPA (1994).
The presented tutorial deals with the use of CZAEM as a stand-alone program, along with
some applications to elementary problems.
BACKGROUND
The objective of the project was to develop algorithms for determining, in relatively simple
settings, the envelopes of capture zones of wells. The capture zone envelope of a well contains
all water that will reach the well, given an infinite time period. Capture zones are divided into
sub-capture zones (called subzones in the program), according to the source of the water in the
zone (for example, a section of a river or a recharge well.) Also of interest are capture zones for a
certain time period, called time zones in the programs. The capture zone for a given time period
of length t is defined as that portion of a capture zone containing all water that reaches the well
within a time period of t. As far as the authors know, the program CZAEM is the first program
that fully computes and displays the boundaries of the capture zone envelope, subzones, and time
zones, including dividing streamlines and stagnation points, for any well in the flowfield.
CZAEM is intended primarily for small-scale applications that can be handled with little
effort and without the need for an advanced computer model. It will generate useful information
for such problems, mostly in graphical form, and hopefully will increase the understanding of the
shape and extent of capture zones. As CZAEM was not written with complex problems in mind,
it is recommended that applications be limited to relatively simple problems. Complex problems
should be dealt with by professionals with access to more powerful computer programs. This
document contains a brief description of the method on which CZAEM is based, the analytic
element method, a brief description of the program, and the tutorial.
THE ANALYTIC ELEMENT METHOD
The analytic element method is based on the superposition of analytic functions. Each of these
functions satisfies the fundamental equations for ground-water flow exactly and has properties that
make it suitable to model a certain feature of the aquifer. The method is described in detail by
Strack (1989).
Although simple in principle, the analytic element method has grown into a complex framework
of many specially developed functions, eliminating most of the limitations that used to be associated
with analytical models. The analytic element method differs fundamentally from most of the
classical numerical techniques. Important differences are:
1. The aquifer is unbounded in the horizontal plane.
2. The solution is analytical, and therefore no interpolation is required for computing heads or
velocities. This allows the user to create contour plots and streamlines for any part of the
aquifer, varying in size from several square feet to hundreds of square miles.
-------
3. There is no numerical dispersion. Inaccuracies, for example in capture zone boundaries, are
due solely to approximations made in the conceptual model and its implementation in the
program.
The application of the method in CZAEM is elementary, and contains only a few analytic
elements. These elements can be used to simulate river boundaries, streams, lakes, wells, uniform
flow, and uniform infiltration over a circular area. The elements used to model river boundaries,
streams, and lakes are called line-sinks. Line-sinks are mathematical functions designed such that
they simulate a constant rate of extraction along a line between the end points of the line-sink.
The line-sinks may be used to model flow from a stream into an aquifer, with the ground-
water table below the stream bottom. In this case the strength of the line-sink (defined as the
extraction rate per unit length) can be estimated from the head in the stream and the. resistivity
of the stream bottom. The stream is then divided into sections, chosen such that the infiltration
approximates the computed infiltration rate of the stream.
Another application of line-sinks is to model constant-head boundaries of rivers, lakes, or
streams. In this application, the model is approximate in that the program computes the strength
of each line-sink section to match the value of head entered. A fine subdivision in line-sink
segments will render a better approximation of the real extraction rate along the stream than a
coarse one.
The well function is used to model wells with either given head or given discharge. The
solution generated by these well functions is accurate, even in the neighborhood of the well, as
there is no numerical discretization.
The uniform flow function adds a uniform component to the far-field (this is the flow pattern
far away from the area modeled.) Finally, the function for radial infiltration (called the rain
function in the program) may be used to simulate uniform infiltration inside a circle.
THE COMPUTER PROGRAM CZAEM
The computer program CZAEM is an enhanced version of the computer program SLWL
(Strack, 1989) with the capability added to generate capture zones of wells. The program is fully
modular; the main modules are the following:
1. AQUIFER, for the input of aquifer data.
2. GIVEN, for the input of uniform flow and rain.
3. REFERENCE, for the input of the head at one point in the aquifer.
4. WELL, for the input and implementation of wells.
5. LINESINK, for the input and implementation of line-sinks.
6. GRID, for the generation of grids of values of piezometric heads, to be contoured.
7. PLOT, for the generation of piezometric contours.
8. TRACE, for the generation of streamlines.
9. CAPZONE, for the generation of capture zones.
10. CURSOR, for the retrieval of data using graphical input.
11. CHECK, for the retrieval of data using keyboard input.
12. IO, for the binary input and output of solutions
All of these modules, with the exception of CAPZONE, existed prior to the present project.
The module CAPZONE is written in FORTRAN. The listing of this module is available separately
-------
from the USEPA. In order to facilitate implementation in other computer programs, it is indicated
in the listing where and how functions providing necessary data to the module are to be called.
The user interface of CZAEM (modeled after SLWL) is a command-line interface. Such an
interface differs from many of the current mouse-driven interfaces. It has the advantage of flexibility,
but requires a longer learning period. Prospective users of CZAEM unfamiliar with the analytic
element method or with the command-line interface of SLWL are urged to take the time to follow
the examples in the tutorial. These examples are designed specifically to guide the user through
the use of CZAEM, progressively introducing several aspects of the program. A study of the first
three examples is considered pre-requisite (preferably by hands-on application) before application
of CZAEM. The final two examples are used to introduce some of the more advanced features of
the program.
In order to simplify the user interface, the command lines and menus are reduced to contain
only the commands used in this tutorial. Other commands, useful for more advanced applications,
are available but are left out of the menus. All commands, however, are listed in the help files
provided with the program. Everywhere hi CZAEM the user may obtain a brief help summary for
a command by typing the command word in the current menu followed by a space and a question
mark. A list of the commands that are available in a module will be displayed by entering the
command word help.
INSTALLATION
An installation batch file called czinst. bat is supplied with the program CZAEM. To install
the program onto a hard drive, change to the disk drive containing the diskette and type
czinat a c
where a represents the disk drive containing the diskette and c represents the hard drive destina-
tion. Further information regarding installation of the program and supported printer devices is
contained in the file read.me. It is recommended to read this file prior to installation.
-------
CZAEM TUTORIAL - PART A
EXAMPLE 1. UNIFORM FLOW WITH A WELL
A farmer wishes to engage in organic farming to complement his regular farming practices. In
order to comply with regulations, he must irrigate his organic crops with ground water that does
not contain the chemicals that are placed on his other fields. It is estimated that a minimum of 60
cubic meters of water per day is necessary to make the venture profitable. Therefore the following
is required:
1. Field characterization to determine aquifer parameters.
2. A model to represent existing flow conditions.
3. An analysis of flow conditions due to the addition of the irrigation well.
Four monitoring wells were installed to determine the aquifer parameters (Figure 1.1). Soil
tests indicate that the permeability and porosity are 6 m/day and 0.3, respectively. Surface
elevations are relatively uniform at 250 m above sea level. Boring logs show the distance to a
confining soil layer from the surface to be 100 m. These values, including ground-water elevations
from the monitoring wells, allow for a model conceptualization of the aquifer (Figure 1.2). It is
necessary to enter all values in consistent length and time measurements (i.e., if the discharge were
given in units of gallons per minute, it would have to be converted to cubic meters per day prior
to entering). The next step is to implement this representation in CZAEM.
Monitoring Well 1
4i = 200.5 m
(-350,350) [ (-150,350)
Monitoring Well 2
(-250,250)
APPLIED
:HEMICALS
= 200.0 m
(250,250)
(-350,150) (-150,150)
V
Proposed Well
Q = 60 m3/day
(-250,-250) (250,-250)
• •
Monitoring Well 3 Monitoring Well 4
3 = 200.5 m 4 = 200.0 m
Figure 1.1 Site map for Example 1.
4
-------
MV
*t =
IOC
^
Im
f#l
200.5m
ELEV = 250 m
1 • /W— -—
r MV
*» =
T$— »• Q = 60 m3/day
permeability = 6 m/day
porosity = 0.3
'#2
200.0m
D:
0:\ >CD CZAEM
Where D: represents the disk drive where CZAEM is installed. Enter CZAEM by typing
D:\CZAEM>CZAEM
Before entering any data, a brief description of CZAEM's structure is in order. On your screen
is the MAIN menu of command words
\\\ Module-MAIN MENU Level-0 Routine-INPUT ///
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1,Y1,X2.Y2)///] , [FILE]
(NUMBER OF FOOTS)
Words in angular brackets, '< >', are commands; words in parentheses, '( )', are required argu-
ments; words in square brackets,'[ ]', are optional arguments; and a slash,'/', indicates alternatives.
Type only the command word and arguments, not the brackets. Some of the commands perform
a function immediately, and some access other modules. If you enter another module, you may
return to the previous module by typing
return
-------
followed by an enter. Only the initial characters of the command words need to be entered, as
many as required to be unique (four is the maximum you may have to enter, but you can enter
more if you wish). In this tutorial, the command is commonly abbreviated
ret
Throughout this first problem, the full names of the bracketed terms shall be typed out; in subse-
quent examples in this tutorial, only the significant letters shall be used.
Entering aquifer data.
We enter the aquifer parameters in the module AQUIFER. Enter the module with the com-
mand . Note that all data must be input with consistent units. Enter
aquifer
CZAEM responds with
\\\ Nodule-AQUIFER Level-1 Routine-INPUT ///
(PERM) (THICK) (ELEVATION) (POROSITY)
The permeability and porosity may be entered directly:
permeability 6
porosity 0.3
Note that the numbers and letters need only be separated by a space. This is also true for
commands that require two values or more. In addition, values may be entered using exponential
form (e.g., 0.6el and 300e-4).
The command requires as the argument the actual vertical extent of the
aquifer (Figure 1.2). is the elevation of the bottom of the aquifer above sea level (or any
datum you chose). Enter
thickness 100
base 150
Note that BASE and POROSITY have default values of 0 and 0.3, respectively. Default values
are used if the user does not enter parameter values. To return to the MAIN module, enter
return
CZAEM will respond by displaying the MAIN menu
\\\ Module=HAIN MENU Level=0 Routine=INPUT ///
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
[(X1.Y1.X2,Y2)///] [FILE]
(NUMBER OF POINTS) '
..
Next enter the module GIVEN
given
CZAEM responds with
\\\ Hodule-CIVEN Level«l Routine-INPUT ///
(DISCHARGE) [ANGLE]
-------
This module is where certain elements with given strengths (i.e., known discharges) are defined.
The only element we can calculate at this point is uniform flow, or UNTFLOW. This value represents
the amount of ground water flowing per unit length of aquifer. To compute the uniform flow
components, use three of the four monitoring well water elevations and locations in conjunction
with the Dupuit formulae for unconfined flow
2(«a-*i)
\Qa\ =
where
Qx = magnitude of flow in the x-direction
Qv = magnitude of flow in the y-direction
Qa = magnitude of flow in the direction of angle a
k = permeability
4>\ = head at location (zi,yi)
>2 = head at location (x2,t/2)
4>3 = head at location (13, ya)
6 = base elevation
a = angle between direction of flow and the x axis (0 to 360°)
(see Strack [1989]). UNIFLOW consists of a constant discharge per unit width of aquifer (mag-
nitude) and a direction (angle). The magnitude is the resultant of Qx and Qv and the angle is
measured from 0 to 360 degrees where 0 is the positive x-axis on the standard coordinate system
(0 is due East). Here we have the following:
6[(200.5-150)2-(200-150)2]
Q* = - 2[250-(-250)] -
2 - - 2
6[(200 - ISO)2 - (200 - ISO)2]
Qv ~ 2(250 - (-250)]
|QQ| = \/0.30152 + 02 = 0.3015
To enter, type:
uniflow 0.3015 0
Note that the term angle in the command line is surrounded by [ ]; this indicates that it is an
optional value and has a default. The default value is 0. Therefore, you could have just typed
UNIFLOW 0.3015. Return to the MAIN module:
-------
return
The amount of information entered thus far can adequately describe the shape of the phreatic
surface. We must fix the elevation of that surface at some point by specifying a known head at a
known location. We shall use monitoring well number 3 as our reference point as follows
reference
CZAEM responds with
(X.Y.REFERENCE HEAD)
You must enter the coordinates and ground-water elevation at the reference point. From Figure
1.1, enter
-250 -250 200.5
More thought must go into the choice of the reference point than appears in this first example.
Its effect is to control the amount of water that comes from infinity (i.e., very far away). When
fixing the reference point, the value of head at this location shall remain unchanged no matter what
other elements are placed in proximity. This issue will become more clear in following examples.
Note for now that a reference point is required to solve any ground-water flow problem with
CZAEM.
Solution and generation of contour plots.
The final step in completing the model of existing flow conditions is to solve for all unknowns.
To do this, enter
solve
CZAEM responds with
ITERATION 1
SOLVING 1 EQUATIONS
10 .
Head and discharge may now be explicitly determined everywhere. We will make a contour
plot of the head on the screen. Our region, or window, shall be four square kilometers centered
around the proposed well. Enter the coordinates of the lower left- and upper right-hand corners
as follows:
window -2000 -2000 2000 2000
We choose the number of points, distributed uniformly within this window, at which to compute
heads. Higher numbers yield higher resolution from the contouring routine used for interpolation.
This grid should not be confused with the mesh in a numerical technique such as finite elements
or finite differences. It is best to specify the grid with values between 20 and 50 (maximum 150).
Enter
grid 50
CZAEM responds with
10
20
30
40
50
60 .
-------
A beep indicates completion of calculations. To view the flow field, enter
plot
Note that a solution, window, and grid must be specified prior to any plot. Plot options and limits
are displayed
EFAULT DTOMBER OF LEVELS] AYOUT
OUH LEVEL [IHCREMENT (X)}] DUX LEVEL]
(MAX LEVEL [DECREMENT {
-------
Entering and analyzing the proposed well.
Enter the module WELL from MAIN
Mil
CZAEM responds with
\\\ Hodnle-VELL Level«l Rootine-Mm ///
The problem specifies a given discharge. Figure 1.1 indicates that the well is placed at the
center of the coordinate system and Figure 1.2 shows the well to have a radius of 0.15 m. We wish
to enter a well with a known discharge.
giren
CZAEM responds with
\\\ Hodnle-WELL Level-1 Routlne-HELL CIVEH ///
(X.Y.DISCHARGE) [RADIUS] [[LABEL]]
Following the instructions of the command line, enter
0 0 60 0.16
return
After entering a new element (the well), we must find a new solution, and again grid and plot
to see the effects. We shall also zoom in on the area of interest. It is important to enter the new
window size prior to gridding, otherwise the plot for the previous window will be placed over the
new window giving erroneous results.
solve
Bindov -500 -500 500 500
grid 50
plot
d 20
[enter]
Notice the drawdown around the well (Figure 1.4). Press [enter] to return to the MAIN
command line, and enter the module TRACE
[enter]
trace
CZAEM responds with the command line
\\\ Module-TRACE; Level=l Routine=INPUT ///
[ (XI. Yl. X2. Y2)///] [TOLERANCE] (/)
First plot the phreatic surface again
plot
d 20
[enter]
To generate a streamline, move the cursor anywhere on the screen using the arrow keys and type
trace
The direction in which the streamline progresses is the direction of flow. Repeat the above proce-
dure by moving the cursor and typing trace again. Note that you can reduce the cursor step size
by pressing the insert key. Generate several streamlines and see that the streamlines never cross.
10
-------
Figure 1.4 Piezometric contours with the well present.
Assuming that the mass of the chemicals applied to the crops is negligible with respect to the
original flow, any streamlines passing under the crops represents the path of chemicals through the
aquifer. It would be beneficial to identify the boundary of the crops in question. To draw a map
of the field boundary, return to the MAIN module and enter the module MAP
return
map
CZAEM responds with the command line
\\\ Hodule=MAP Level=l Routine=INPUT ///
(ON/OFF)
To view the map on the screen with each plot, set the option on
plot on
We draw the crop boundary using the command . Start at one corner of the field and
draw a line from corner to corner, completing the drawing by entering the first corner again. MAP
will prompt for each set of coordinates.
curve
-350 150
-350 350
-150 350
-150 150
-350 150
return
Return to the module TRACE and move the cursor along the boundary to see if any chemically
influenced water enters the well (Figure 1.5).
11
-------
: ?~--Sr.r.r.1—i~
I (&'• i
Figure 1.5 The water beneath the field does not enter the well.
Exiting the program CZAEM.
To exit CZAEM, return to the MAIN module and type
stop
This returns control to DOS.
EXAMPLE 2. WELL NEAR A RIVER
A high capacity well with a diameter of 0.6 m and a pumping rate of 1500 m3/day is proposed
to be placed near a river at a fixed location. We are asked to determine whether or not the well will
capture river water. The hydrogeologic information is shown in Figures 2.1 and 2.2. The aquifer
parameters are the following: permeability 5 m/day, thickness 50 m, base elevation 0 m, porosity
0.25, uniform flow 0.5 m3/(m day) at 30°. In addition, piezometric head measurements are known
at different points along the river.
In CZAEM, enter the following data using consistent units from the MAIN command line
aqui
perm 5
thick 50
base 0
poro 0.25
ret
given
uni 0.5 30
ret
12
-------
32m
(-1500.1500)
(-600.1300)
,33m
35m
(500,200)
Uniform Flow
Qa = 0.5 m'/(m day)
Proposed Well
Q = 1500 m3/day
(500. -800)
(800,-1
(1100, -1000)
39
m
(1500,-1800)
Figure 2.1 Site map for Example 2.
Entering line—sinks.
A river is simulated in CZAEM by a series of line-elements called line-sinks. We first need
to break a river into straight line segments. Each line segment will be entered in the model as a
line-sink. River discharge into or out of the aquifer is assumed to be constant along each segment.
The discharge is either known a priori or is calculated by specifying the head at the center of the
line-sink resulting in given or head-specified line-sinks, respectively. A given line-sink extracts
a fixed amount of water per unit length of line-sink without placing any restriction on the head
distribution along the line-sink; a head-specified line-sink also extracts a fixed amount per unit
length but creates a control point at its center which makes it possible to solve for the discharge
such that the head at the control point equals that entered. Given line-sinks are entered in
a similar fashion as wells. The input of head-specified line-sinks is outlined below. Enter the
module LINESINK
linesink
CZAEM responds with
13
-------
Uniform Flow
Qa = 0.5 m3/(m day)
= 1500 m'/dmy
River
x^W"
50m
permeability = 5 m/day
porosity = 0.25
ELEV = 0 m
Well radius = 0.10 m
Figure 2.2 Conceptual model of the aquifer.
\\\ Module«LINE-SINK Level«1 Routine-INPUT ///
[/] [TOL]
Line-sinks with known strengths are entered through the command , and those with
known heads through the command . We have head-specified line-sinks; type
head
CZAEM responds with
\\\ Module=LINE-SINK Level«l Routine=LINE-SINK HEAD ///
(XI,Yl, X2.Y2, HEAD)[[LABEL]]
CZAEM requests the coordinates of the starting point (XI,Yl), the coordinates of the endpoint
(X2,Y2), and the head at the midpoint of each line-sink (HEAD), in that order. For clarity, start
at any one end of a series of contiguous line-sinks and enter them in order of occurrence. We start
from the north end of the river and enter the first line-sink
-1500 1500 -600 1300 32
After a line-sink is entered, CZAEM prompts for another line-sink.
\\\ Module«LINE-SINK Level=l Routine-LINE-SINK HEAD ///
(XI.Yl, X2.Y2), HEAD)[[LABEL]]
Although the command line is not displayed, it is still active, and we may at any time enter
to begin a head specified line-sink, to see the command line, or
to return to the MAIN module. Enter the remaining line-sinks and return to the
MAIN module
-600 1300 -200
-200 900 200
900 33
500 34
14
-------
200 600 500 200 35
600 200 600 -800 37.5
600 -800 800 -1000 38
800 -1000 1100 -1000 39
1100 -1000 1500 -1800 40
r«t
We have returned to the MAIN module. A visual check of our data entries is possible with
the command . In this case we choose a window with a lower left corner at (-1500,
-1500) and the upper right corner at (1500,1500).
window -1500 -1500 1500 1500
To view a layout of all the elements, type
layout
When finished viewing, press [enter] to return to MAIN.
The last piece of information required is the reference point. In this example the reference
point is used as a calibrating parameter, rather than as a point of given location and head as in
Example 1. Analytic element models such as CZAEM do not require a bounded model but deal
with an infinite domain. The flow of water from far away (infinity) can be used to approximate
physical sources and sinks that are present in the real aquifers but not represented in the model
explicitly. This flow of water is regulated via the head specified at the reference point. The more
the physical sources or sinks are included in the model the less the influence of the reference point.
Since the reference point is often used to approximate complex features that are far away and
left out of the model, its use as a calibration tool requires some experience with analytic element
models. The reference point should be chosen far enough away from the area of interest that the
head is not expected to change appreciably due to the introduction of any new element (e.g., a
well). We first use a point with coordinates (-2000,4000) and a head of 40 m as the reference point.
re*
-2000 4000 40
Solve, grid, and plot the solution (Figure 2.3).
solve
grid SO
plot
d
[enter]
We assume for now that the model closely matches observed heads.
Calculating the head at any point.
We can determine the head at any point in the aquifer through the module CHECK. We type
check
CZAEM responds with
\\\ Hodule=CHECK Level-1 Routine-INPUT ///
(X. Y) (X, Y)
This is the CHECK module menu. Enter
head 1 1
CZAEM responds with
15
-------
Figure 2.3 Plot of piezometric contours, well not present.
X Y READ
1.OOOOOOE+00 1.OOOOOOE+00 3.TI8284E+01
followed by the command line. Enter
head -2000 50OO
CZAEM responds with
X Y HEAD
-2.000000E+03 5.000000E+03 3.975401E+01
Note the heads at these two points for future use. With only uniform flow present, entering
either of these heads at their respective locations as the REFERENCE would result in identical
solutions. Return to the MAIN module.
ret
Entering the well.
We now solve the problem with the original reference point and the well present. First, we
will reduce the window size so that we may examine more closely the changes due to the well.
window -1000 -1000 1000 1000
To enter the well, type
well
given
0 0 1500 0.3
ret
16
-------
Since the well affects the flow field, we must solve again
•olve
Saving a solution.
It is often desirable to save a solution for later use. The command is used for that
purpose and stores all current information in a binary file. We save this solution for retrieval later;
enter
CZAEM responds with
<80LOTIOVXGRn»
To select SOLUTION, type
Ml
CZAEM will request a file name
{to abort}
Enter the filename
ex2.iol
If the file ex2. sol does not already exist, CZAEM responds with
SOLUTION FILE HAS BEEN WRITTEN
PROBLEM: UNHAMED
If the file already exists, CZAEM will offer you a prompt to overwrite the file or abort the command.
The name and extension are arbitrary. The directory under which the file ex2.sol is placed is
CZAEM unless otherwise specified.
In order to view the new solution, enter
grid SO
plot
d
[enter]
Observe the changes in the piezometric contours due to the well (Figure 2.4). Press [enter] to
return to the MAIN module.
Influence of the reference point.
We will now examine the effect of the reference point on the solution. First choose the reference
point at (1,1) with a head of 37.1828 (these are values that we determined using CHECK).
ref
1 1 37.1828
solve
grid 60
plot
34 4
[enter]
This plot is shown in Figure 2.5. Comparing Figures 2.4 and 2.5, we see that the influence of the
reference point on the piezometric contours is major, because we entered a well near the reference
point.
17
-------
Figure 2.4 Contours with the well present, reference point at (-2000,4000).
Now we see what happens when we enter (-2000,5000) as our reference point with a head of
39.7540 (the second point determined in CHECK)
ret
-2000 5000 39.7540
solve
grid SO
plot
26 2
[enter]
This plot is reproduced in Figure 2.6 and is nearly identical to the one in Figure 2.4 even though
the reference points are different; the influence of the well is insignificant at both reference points.
This confirms that the reference point must always be chosen sufficiently far away so that elements
in the model do not influence the head at the reference point significantly.
Determining a well's water source using pathlines.
We are now prepared to answer the question: will the well capture the river water? We retrieve
the original solution and data saved in the file ex2.sol by the use of the command .
From the MAIN command line, enter
read
CZAEM responds with
18
-------
Figure 2.5 Contours with the well present, reference point at (1,1).
Enter
solution
CZAEM responds with
PLEASE ENTER FILENAME; TO ABORT
Enter
ez2.sol
CZAEM responds with
SOLUTION FILE HAS BEEN READ
PROBLEM: UNNAMED
CZAEM has read in the binary file with all the elements entered at the time of saving. Since
the problem was solved prior to saving, the parameter values have also been read in. CZAEM
returns to the MAIN menu after reading in the file.
There are several ways to determine whether the well draws river water in CZAEM; here we
shall use the technique of tracing the particle pathlines in a backward fashion starting at the well.
Enter TRACE by typing
trace
Streamlines are traced in the direction of flow by default. We set it to backward tracing with the
command . Backward tracing from the well is achieved by the command
19
-------
Figure 2.6 Contours with the well present, reference point at (—2000,5000).
. Note that to return to forward tracing one must type .
We must first enter PLOT, or LAYOUT; here we choose PLOT as it will produce the piezometric
contours. We enter PLOT exactly as we did from the MAIN menu.
plot
d
[enter]
CZAEM plots the picture on the screen and gives the following menu:
\\\ Module=TRACE; Level=l Routine=INPUT ///
[ELEVATION] ]
(» LINES) [ELEVATION]
-------
proposed discharge is not feasible without drawing river water. To return to the MAIN menu and
exit CZAEM, enter
ret
stop
Figure 2.7 Several pathlines from the well generated by < WGENERATE > begin at
the well and end at the line-sink, showing that the well does capture river
water.
EXAMPLE 3. CRITICAL PUMPING LEVEL FOR A WELL
This example expands on the problem presented in Example 2 and introduces the module
CAPZONE. We will determine the critical pumping level of the well, defined here as the largest
discharge not capturing river water. In Example 2, the well is entered with a discharge of 1500
m3/day, and is drawing water from the river. We will use to view the current
solution, then adjust the well discharge until the critical level is reached.
Since the current problem has few elements, it will be sufficient to use from
the outset to evaluate each well pumping level. For larger problems, it may be more efficient to
use or for the first iterations.
Creating capture zones.
Begin by entering CZAEM and reading in the binary file describing Example 2. From the
MAIN command line enter
read
solution
ez2.sol
21
-------
The binary file with all elements entered in Example 2 and the parameter values determined by
has been read in. Enter
grid 60
trace
CZAEM responds with
\\\ Modul«-TRACE; Uval«l Routin«-I«PUT ///
t (XI , Tl , X2 . T2) ///] [TOLERANCE] (/)
We enter the module CAPZONE from within TRACE and type
CUkZOlM
CZAEM responds with
EFAOLT DnmBER OF LEVELS] AYOUT
(HIM LEVEL [INCREMENT (X)}) [MAX LEVEL]
(MAX LEVEL [DECREMENT {<0}] [MM LEVEL]
NIN. LEVEL- 2.606169E+01 Nil. LEVEL- 4.22S666E+01
We can enter the desired levels, request default levels , or just get a layout . For this
example enter
d
[«nt«r]
We are now in the CAPZONE module, and the menu is
\\\ Module-CAPTURE ZONE; Uv«l-l Routine-INPUT ///
[(XI ,Y1 ,X2.Y2)///] (LINES)[ [VELOCITY FACTOR] /] (t LINESXCOLOR> [COLOR1] [2] [3]
(FILE) (FILE) {TO BACKSPACE. PRESS < }
CZAEM is in graphics mode, and the backspace key is no longer active. It is replaced by the
less-than sign (<). We are about to let CZAEM determine the capture zone envelopes for the well
and streamlines dividing the capture zone into subzones. Before discussing the meaning of these
curves, we will generate them on the screen. Move the cursor over the well and enter the command
. Since the well is centered in the current window and therefore already coincides
with the cursor, we enter
aubzone
CZAEM responds with
UNCONFINED: X.Y. PHREATIC SURFACE 4.066071E-02 4.066071E-02 2.551642E+01
CALCULATING SUBZONES PHASE 1: CREATING INITIAL PATHLINES FROM THE WELL
10 ..........
20 ..........
30 .
CALCULATING SUBZONES PHASE 2: DETERMINING LOCATION OF STAGNATION POINTS
10 ..........
20 ..........
30 .
CALCULATING SUBZONES PHASE 3: FILLING SUBZONE BUFFERS
10 ..........
22
-------
Figure 3.1 Capture zones generated for a well discharge of 1500 m3/day.
CZAEM computes the capture zone envelope curves and plots them along with all dividing stream-
lines (Figure 3.1).
Dividing streamlines either pass through stagnation points (points where no flow occurs in
any direction) and end at the well, or separate unique source areas for the selected well. The
envelope curves bound the entire area supplying water to the well. CZAEM distinguishes each
individual element (e.g., a line-sink in a river stretch) as a unique source of water. CZAEM's
23
-------
search for dividing streamlines, envelope curves, and distinct sources terminates at the current
window boundary. In order to obtain meaningful results within CAPZONE, the window must be
chosen large enough to include all of the important sources and the stagnation points.
The window boundary is identified by CZAEM as the source of any water entering the well
originating from far away.
Interpretation of Figure 3.1.
The capture zone envelope for the well consists of envelope curve A, the line-sink, and curve
D below (Figure 3.1). The source area is divided into two subzones, one whose source is the
window edge, and one whose source is the line-sink. The source areas are delineated by dividing
streamlines B, C, and Z>. Curve C and the upper part of curve V are intended to be a single
dividing streamline distinguishing the two source zones, so why are they plotted distinctly rather
than as one? CZAEM tries to plot dividing streamlines starting from stagnation points, which
worked for line B. However, the stagnation point which should have been used for drawing lines C
and T> is discarded by CZAEM because it is within a tolerance distance of the line-sink; CZAEM
plots the nearest dividing streamlines from each source instead. These are separated by 1/200'th
of the well's discharge, which is generally well within modeling precision. The lower part of D is
thus inside of the actual envelope curve by less than 1/200'th of the well's discharge and can be
taken as the working envelope curve.
Determining a well's water source using capture zones.
The amount of water supplied by the river can be found using the command
CZAEM responds with
SOURCE DISTRIBUTION FOR WELL NUMBER 1
SUBZONE NUMBER SOURCE TYPE SOURCE NUMBER X OF WATER
1 WINDOW BOUNDARY 79.5
2 LINE-SINK STRING 5 20.1
PRESS THE ENTER KEY TO CONTINUE
The current pumping level is too high; 20 percent of the well water originates from the river.
We will next try a pumping level of 1000 m3/day. Return to the Well module, reset, enter the well
with the new strength, solve, grid, and regenerate the capture zones. Note that erases
all input data within the current module. This command requires confirmation.
ret
ret
veil
reset
yes
given
0 0 1000 0.3
ret
sol
grid 50
tra
cap
d
[enter]
sub
24
-------
Figure 3.2 Capture zones generated for a well discharge of 1000 m^/day.
At this pumping level, the capture zone still intersects the river (Figure 3.2). Enter
source
CZAEM will report that about 1.5 percent of the water is coming from the line-sink. We next
repeat the above procedure to reset the pumping level to 970 m3/day. Enter
ret
ret
well
reset
yes
given
0 0 970 0.2
ret
solve
grid 50
trace
capzone
d
[enter]
sub
At this pumping level, the capture zone just misses the river (Figure 3.3).
Next try a pumping level of 990 m3/day. Enter
ret
ret
well
reset
yes
25
-------
given
0 0 990 0.2
ret
•olve
grid 60
trace
c&pzone
d
[enter]
•ub
The capture zone boundary now ends at the line-sink (Figure 3.4); The command
reports that all the water is coming from uniform now. Enter
•ource
What does it mean that the envelope ends at the line-sink but the line-sink is not identified
as a source? Recall that CZAEM determines the contributions from each source with a precision of
1/200'th of the well's discharge; flow from the line-sink must therefore be less than that amount. No
dividing streamlines are drawn to demarcate the source zones since only one source is found. The
envelope curve is drawn from the stagnation point toward the line-sink and stops there because
CZAEM detects the change in flow direction at the line-sink. As in the first plot, there is no
stagnation point recorded at the line-sink, thus the envelope curve is not continued.
We conclude that the critical pumping level is between 970 and 990 m3/day. Note that this
solution is based on an oversimplified representation of the river reach supplying water to the well.
In the advanced tutorial lessons (Example 5) we will see that refining the model can change this
estimate. The modeling process normally includes successive refinement of the model until changes
in the results are within modeling accuracy.
SUMMARY OF PART A
In the first three tutorial exercises we have introduced elementary modeling techniques and
the display of capture zones with the following CZAEM commands
AQUIFER:
GIVEN:
REFERENCE
WELL:
LINESINK:
SOLVE
CHECK:
WINDOW:
NAP:
LAYOUT
GRID
PLOT:
LAYOUT
TRACE:
SAVE:
READ:
PERMEABILITY. POROSITY. THICKNESS, BASE
UNIFLOW
GIVEN
HEAD
HEAD
XI, Yl X2.Y2
CURVE, POINT, PLOT ON
D. L
TRACE
PLOT: UGENERATE. BACKWARDS
CAPZONE: SUBZONE, SOURCE
SOLUTION
SOLUTION
ON/OFF
RESET
STOP
26
-------
Figure 3.3 Capture zones generated for a well discharge of 970 m3/day.
Figure 3.4 Capture zones generated for a well discharge of 990 m3/day.
27
-------
In the remaining examples, we will introduce more advanced modeling techniques, and the
following CZAEM commands:
GIVEN: RAIN
LINESINK: GIVEN. STRING ON/OFF
CHECK: AQUIFER, GIVEN. REFERENCE. DISCHARGE, CONTROL, SUMMARY
WELL: RANGE
LINESINK: RANGE, STRING. ENDS. BVAL, DISCHARGE
WINDOW: ALL. POP, PUSH
TRACE: CURSOR ON/OFF
SET: MAXSTEP. FRONT ON/OFF. MARKER, TIME
LAYOUT: BASE. SURFACE. COORDINATE. WLL. TOL. MENU
CAPZONE: COORDINATE, BASE. SURFACE. TIMEZONE. KLINE
PAGE. COLOR. BSAVE. BREAD
CURSOR: LAYOUT: WLMOVE, LSMOVE
SWITCH: PREFIX. INPUT/OUTPUT/MESSAGES/ERROR. LOG ON/OFF. CALL.
BACK
SAVE: GRID. BOTH
READ: GRID. BOTH, DIFGRID
PSET: PRINTER. SCREEN, PALETTE, MOUSE ON/OFF
We strongly encourage you to complete the advanced tutorial lessons, but you should now be
able to apply CZAEM effectively to many small-scale practical problems as a stand-alone program.
28
-------
CZAEM TUTORIAL - PART B
EXAMPLE 4. CONTAMINANT PUMPOUT SYSTEM
A contaminant plume has been identified in a confined aquifer upgradient of a rural subdivi-
sion. To avoid contamination of private wells, the contaminant is to be pumped out of the aquifer.
Field studies have identified the plume limits and estimated aquifer properties. Three monitoring
wells have been installed to evaluate local ground-water flow conditions. A proposed pumpout
system is to place a well at coordinates (100,20) of discharge rate 220 m3/day (Figure 4.1).
Monitoring Well 2
-------
2. Adding a discharge well or wells downgradient of the contaminant, strong enough to capture
the entire plume.
3. Determining time capture zones to estimate the pumping time required to capture the entire
plume assuming no longitudinal dispersion.
The strength of the initial uniform flow may be determined from the monitoring wells as in
Example 1. The results follow:
Qx = 0.0966 m3/(m day)
Qv = 0.0259 m3/(rn day)
Qa = 0.100 m3/(m day)
(a = 15°)
To model the existing site conditions, input the aquifer parameters (Figure 4.2) and given
strength elements (i.e., uniform flow). Using consistent units of measure, enter
MW# I
= 129.84 m
Q = 220 m3/day MW # 3
,~ft__± <*3= 126.58m
P
permeability = 2 in/day
porosity = 0.25
Well radius = 0.10 m
Figure 4.2 Conceptual model of the aquifer.
aquifer
perm 2
thick 20
base 100
poro 0.25
ret
giv
uni 0.1 15
ret
To complete the model of existing conditions, a reference point where the head is known must
be entered. Here, as in Example 1, a good choice is one of the monitoring wells. We will choose
MW#1.
30
-------
ref
-750 -875 129.84
Vie must solve, set a window size, grid, and plot the existing conditions to view the results
(Figure 4.3).
solve
window -1000 -1000 1000 1000
grid 40
plot
d 20
[enter]
Figure 4.3 Existing conditions: uniform flow with reference head of 129.84 meters
at (-750, -875).
Obtaining results using CHECK.
At this point it is useful to test our model to make sure that it reflects observed conditions.
Modules CHECK and CURSOR provide two means of testing results. Here we will introduce
CHECK; CURSOR will be discussed later. The module CHECK allows the user to check input
data as well as model results, including point values of head and discharge. We will begin by
checking point values. Enter the module by typing
check
CZAEM responds with .
\\\ Module=CHECK Level=l Routine-INPUT ///
(X, Y) (X. Y)
31
-------
A good test for the model is to see whether it reproduces the observed heads at the monitoring
wells. Type "head" followed by the coordinates of MW#2:
head -542 750
CZAEM responds with
X Y BEAD
-5.4200OOE+02 7.500000E+02 1.282863E+02
Now, check the head at MW#3:
head 500 -500
CZAEM responds with
X Y HEAD
5.000000E+02 -6.000000E+02 1.265788E+02
We can also check the discharge at any point hi the aquifer. As uniform flow is the only element
contained in the current model, discharge will be the same throughout the flowfield. Check the
discharge at the origin.
discharge 0 0
CZAEM responds with
X . Y O.OOOOOOE+OO 0.OOOOOOE+00
QX . QY 9.659258E-02 2.B88191E-02
The results are consistent with the field data. Commands < REFERENCED ,
, , , and in module CHECK allow the user to
check input data. Enter each command to check your input. Only the first four commands apply
to the current model. When you are finished, return to the MAIN menu.
ret
The model of existing conditions is now complete. To test the proposed pumpout system
design, a well must be added near the plume. Ideally, the reference point should be far enough
away from the area of interest so that elements added to the model (in this case a well) have a
minimal effect on the head at the reference point. Here, the reference point must be moved away
from the area of interest. The problem is easily handled in this simple case. Use the above model
to check the head far from the plume. A reasonable choice here may be (-2000,-2000).
che
head -2000 -2000
CZAEM computes the head at the entered coordinates and responds
X Y HEAD
-2.000000E+03 -2.000000E+03 1.335864E+02
Return to the MAIN menu.
ret
Use the results to set a new reference point at -2000, -2000.
ref
-2000 -2000 133.586
CZAEM stores only one reference point. Adding the reference point at -2000.-2000 replaces
the previous reference point. Solve, grid, and plot the revised solution (Figure 4.4).
32
-------
•olv*
grid 40
plot
d 30
[«nt«r]
Figure 4.4 Existing conditions: uniform flow with reference head of 133.586 meters
at (-2000, -2000).
The solution should be exactly the same as the prior solution. To verify this, enter the check
module and once again check the heads at the monitoring wells using the same command sequence
as before. If the data has been entered correctly, the results will be consistent with the field data.
Now use MAP to identify visually the plume and monitoring wells on the screen. The com-
mand is used to show the locations of the monitoring wells (Figure 4.1). Field data
provide coordinates of points on the perimeter of the plume (Figure 4.1), which are plotted with
.
map
plot on
point
-750 -875
-542 750
500 -500
curve
-500 568
-300 0
-583 -891
-800 -200
-500 568
33
-------
r*t
Check the locations of the input data using the layout command from the MAIN menu.
layout
The layout of the various elements of the model will appear on the screen without the head contours.
The proposed pumpout system includes a well at coordinates (100, 20) withdrawing 220
m3/day. To check that the system captures the entire plume, add the well to the model from the
MAIN menu by entering
mil
Riv
100 20 220 0.1
r«t
Solve, and grid the results.
•olva
grid 40
Several methods may be used to check the adequacy of the well. The simplest method is to
use the command in the module TRACE to draw forward pathlines from the plume
boundary to the well. A second approach is to use in the module TRACE to
draw backward pathlines from the well. The approach we will use here is to enter CAPZONE from
the module TRACE. Once in module CAPZONE use the command to generate the
capture zone envelopes for the well. Input the following sequence
trace
capzone
d
[enter]
The contoured solution will be plotted on the screen (Figure 4.5).
Move the cursor to the discharge well. Use the [insert] key to reduce the cursor step size.
Create the subzones for the well
subzone
The capture zone envelope will be displayed on the screen, Figure 4.6. The entire plume is captured
by the well. The pumpout system appears to be adequate, but we will check conditions at the well
for any possible problems. Return to the MAIN module and enter the CHECK module.
ret
ret
check
head 100 20
CZAEM responds with
X Y HEAD
l.OOOOOOE+02 2.000000E+01 1.181211E+02
Note that the head at the well is below the elevation of the confining unit; flow near the well
is unconfined. CZAEM handles cases of combined confined/unconfined flow directly—the solution
is correct. However, you may wish to maintain confined conditions at the well. To achieve this,
the discharge of the well must be reduced. The plume may still be captured while maintaining
confined conditions by adding a second discharge well downgradient from the first. We will reduce
34
-------
t .
; \
/:
/ •:.
\ i
••*••»
i
I ',
\v
Figure 4.5 Contours with the well present.
the discharge of Well 1 to 110 ms/day and add a second well discharging at the same rate 200
meters downgradient from the first. Return to the MAIN menu and reset Well 1;
r«t
well
reset
y
giv
100 20 110 0.1
Well 2 may be added directly at this point
293.2 71.76 110 0.1
ret
Determining capture zones for multiple wells.
We again wish to use the module CAPZONE to determine the capture zone envelopes for the
two discharge wells using the command , but this time we will use layout and not
plot the contours.
solve
tra
cap
1
[enter]
Move the cursor to the leftmost well (from now on we will refer to the leftmost discharge well as
Well 1 and the rightmost discharge well as Well 2).
sub
35
-------
Figure 4.6 Subzones drawn for the well.
The capture zone envelope for Well 1 will be recomputed and displayed. Move the cursor to Well 2.
sub
The capture zone envelope for Well 2 will be displayed (Figure 4.7). Note that the entire plume
lies within the combined capture zone envelope for both wells. Return to the MAIN menu and
enter module CHECK to see that both wells remain in confined conditions.
Well water travel times.
Additional pertinent information may be obtained from CZAEM. Next, we will generate time
capture zones for each discharge well to determine how long the wells must operate for their capture'
zones to reach the plume. A time zone provides the zone of water that a well will capture if operated
for a specified period of time. For example, the water at the edge of a five-year time zone will be
captured by the well if it pumps continuously for five years. Return to module CAPZONE from
the MAIN module.
trace
cap
1
[enter]
Move the cursor to Well 1 and enter the command
time
CZAEM responds with
CALCULATING SUBZONES PHASE 1: CREATING INITIAL PATHLIKES FROM THE WELL
10
36
-------
figure 4.7 Subzone curves for both wells.
20
3d .
CALCULATING SUBZONES PHASE 2: DETERMINING LOCATION OF STAGNATION POINTS
10
20
30 .
CALCULATING SUBZONES PHASE 3: FILLING SUBZONE BUFFERS
10
ENTER [TIME STEP][MAXIMUM TIME], OR
EDRAW LAST TINE ZONES, OR EFAULT TIME ZONES. ORXIT
MINIMUM AND MAXIMUM TIMES FOR CAPTURE ZONE: O.OOOOOE+00 4.78038E+04
You may enter for default to generate ten equal increment time zones on the screen.
CZAEM determines the increment based on window size. Here we will enter a starting time zone
and five increments of twenty years each (7300 days)
7300 7300 36500
CZAEM computes and draws the capture zones in 20-year increments. Note that for Well 1,
more than 20 years of continuous pumping are required to reach the plume. Move the cursor to
Well 2 and repeat
time
7300 7300 36500
Twenty-year time zones will be displayed for Well 2 (Figure 4.8). We see that 60 years of continuous
pumping are required to reach the plume. The time zones computed so far are based on water
velocity. CZAEM allows the user to input a contaminant front velocity as a factor of the water
velocity. The velocity factor is capable of describing hydrodynamic dispersion (Strack, 1992) and
37
-------
sorption and must be determined by field studies. Here we will assume a factor of 1.1 has been
determined. While in module CAPZONE, set a front velocity factor.
front on 1.1
Figure 4.8 Twenty year time zones for Wells 1 and 2.
Now move the cursor to Well 1 and enter the command.
time
7300 7300 36500
New time zones are computed based on the velocity of the front and plotted over the previously
computed time zones. Move the cursor to Well 2 and repeat (Figure 4.9). Note that time zones
with or without a front factor do not provide information as to how long pumps must operate to
capture all of a contaminant; only the time required for the contaminant front to reach the well is
provided.
Moving wells in graphics mode.
The pumpout system described here requires long periods of continuous pumping before ever
reaching the contaminant. To refine the design, the user may wish to move the discharge wells closer
to the plume and/or examine different combinations of wells and discharges. This may be done by
resetting the wells as was previously done, or it may be done directly on the graphics screen using
the command in module CURSOR. Exit both CAPZONE and TRACE, return to
the MAIN command line, enter CURSOR, and draw a layout as follows:
ret
ret
cursor
38
-------
Figure 4.9 Twenty-year time zones for the fronts, with a front velocity factor of 1.1.
lay
[enter]
Move the cursor to Well 1
vlmove
CZAEM responds with
PLEASE RE-POSITION CURSOR AND PRESS ENTER
Move the cursor to the location where you wish to place the well. If the cursor was not originally
close to the well, CZAEM will prompt the user to set a new tolerance.
WELL NOT FOUND; HOVE CURSOR CLOSER OR RESET TOLERANCE
Enter
TOL
CZAEM responds with
PLEASE RE-POSITION CURSOR AND PRESS ENTER
Move the cursor one step and enter
[enter]
A new tolerance is now set and CZAEM responds
RTOL» 5.000000E+01
Now move the cursor to the well and enter
39
-------
CZAEM responds with
PLEASE RE-POSITION CURSOR AHD PRESS ENTER
Move the cursor to a new position where you wish to place the well and press enter.
[enter]
The well is moved to the new position and CZAEM responds with
DISCHARGE-SPECIFIED HELL IR 1
POSITION CHANGED FROM
0.100000E+03 0.200000E+02
TO
-0.166633E+03 -0.832961E+02
CURSOR POSITION (X.T): -0.166633E+03 -0.832961E+02
Solve, grid and plot to check the results. Note that allows the user to change
the well location and discharge simultaneously simply by entering the new discharge following
. This process may be repeated until an optimal design is obtained.
EXAMPLE 5. DATA MANIPULATION AND MODEL REFINEMENT
Example 5 will build on the model created in Examples 2 and 3. The model will be refined
and entered via a data file instead of via the keyboard.
Using input files.
The input file example5.dat is included in the CZAEM directory and is listed below:
•input echo off
ret
win -1000 -1000 1000 1000
•qui
perm 5
thick SO
base 0
por 0.25
ret
giv
unl 0.6 30
rat
line
head
-1500 1500 -600 1300 32
-600 1300 -200 900 33
-200 900 200 500 34
200 500 500 200 35
500 200 500 -800 37.5
600 -800 800 -1000 38
800 -1000 1100 -1000 39
1100 -1000 1500 -1800 40
ret
ref
-2000 4000 40
well
given
0 0 1000 0.3
ret
solve
swi
40
-------
•nd
The asterisk (*) indicates a comment statement and CZAEM shall disregard all information to
the right of the asterisk on that line. The data file applies to Examples 2 and 3 with a well
discharging 1000 m3/day. To confirm this, start CZAEM and enter module SWITCH from the
MAIN command line.
•witch .
CZAEM responds with
\\\ ROUTME SWITCH ///
[ (PREFIX)]
[ [FILE HAKE][LOGICAL OTIT)
(FILENAME)
The command PREFIX sets the DOS directory where files are either read or sent. The CZAEM
directory is currently the default and is specified by the file initaem.dat. The second set of
command words in SWITCH dictates how and where to send input data and program feedback.
Further information on these features is contained in the help file in this module. When a file is
read in, the input and any CZAEM error messages will scroll quickly up the monitor. To record
this information to a file, enter
log on example5.log
This creates a transcript of all information displayed on the screen (aside from graphics) which
may be consulted after exiting or when using PAUSE from the MAIN menu. If the command LOG
ON is not followed by a file name, the information is sent to the file log. dat by default. To read
in the data file, enter either
call exaopleS.dat
or
ret
sui example5.dat
Both of these command sequences accomplish the same. It is important to do only one or the
other, otherwise the data will be superimposed. We must RESET from the MAIN command line
before calling in the same data file the second time. Try reading in the data both ways. Also,
remove the asterisks on the first two lines and read in the file to see the effects of the command
INPUT ECHO OFF. Note that INPUT ECHO OFF is disabled after reading each file and only
the input is not displayed (the solve response is still shown on the monitor).
Saving grid files.
Enter 50 and plot. Notice that the results are the same as in Example 3 where the
data were entered manually. After viewing, enter ; save the current grid by typing
save
grid
CZAEM responds with
{to abort}
Enter the filename
BOll.gSO
41
-------
The filename is arbitrary and the extension (.gSO) reflects the number of grid points chosen.
Now we will refine the model elements. Recall that head-specified line-sinks approximate
a constant head boundary along each line segment by specifying the head at the midpoint and
determining a constant discharge rate along the segment. The results are approximate as the
head matches only at the midpoint. To refine the model we divide the long line-sink nearest the
well into several smaller head-specified line-sinks. This provides more control points and a better
approximation along the bank. Field data provides the new information. Exit CZAEM and replace
the line-sinks by editing the data file example5.dat with the following:
liM
string on
h«»d
-1500 1500 -600 1300 32
-600 1300 -200 900 33
•200
200
400
600
600
600
600
470
480
600
530
600
900
600
300
-100
-200
-325
-400
-500
-600
-700
-800
-900
200
400
500
500
500
600
470
460
600
630
600
800
500
300
-100
-200
-325
-400
-500
-600
-700
-800
-900
-1000
34
34.7
35.5
36.0
36.5
37.0
37.2
37.3
37.4
37.8
38.0
38.2
800 -1000 1100 -1000 39
1100 -1000 1500 -1800 40
ret
Save the file with the new data, enter CZAEM, and read in the file with the command.
switch eiample5.dat
A command present in the data file which has not yet been explained is the
command in the module LINESINK. Line-sinks may act as sources to a well (Example 3) and
subzones will be computed for each line-sink segment. This requires much computational time
and generates data which may not be of interest. For example, each line segment will be identified
as an individual source; often a user will only be interested in the total amount of water pumped
from a river, not the amount pumped from small segments. The command is used
to link line-sink segments together which will then act as a single source in subzone computations.
Comparing grids.
Enter 50 and plot the results in the module CAPZONE. Create subzones and note
that the well no longer draws river water at a discharge of 1000 m3/day (Figure 5.1). Refining the
line-sinks has improved our model and we can now determine a safe pumping level more accurately.
The subzone boundary has changed significantly due to the refinements, but the user may wish to
know the extent to which heads have changed in the refined model.
Return to the command line of the MAIN module and enter the module READ. We will
contour the difference between the original model and the refined model by entering
read
CZAEM responds with
42
-------
Figure 5.1 Case of Example 3 with refined line-sinks.
enter
difgrid
CZAEM responds with
PLEASE ENTER FILENAME; TO ABORT
enter the filename of the grid previously saved
soll.gSO
CZAEM responds with
GRIDFILE HAS BEEN SUBTRACTED FROM CURRENT GRID
and control returns to the MAIN command line. Enter the module PLOT and CZAEM responds
with
EFAULT [NUMBER OF LEVELS] AYOUT
(MIN LEVEL [INCREMENT {>0}][MAX LEVEL]
(MAX LEVEL [DECREMENT {<0}][MIN LEVEL]
MIN. LEVEL= -9.516754E-01 MAX. LEVEL" 9.SS543SE-02
These numbers represent the minimum and maximum head differences between the two models.
Contour the difgrid by entering
d 10
and observe the graphical results (Figure 5.2). We see that the greatest difference in head occurred
at the river where the contour lines are concentrated the most (here the head is approximately 1
43
-------
Figure 5.2 Refined line-sink grid minus the original line-sink grid.
meter less in the refined model than in the original model). When using DIFGRID to examine
head differences, both the number of grid points and the window size must be the same between
models.
Obtaining results using the cursor in CAPZONE.
Several additional options exist within module CAPZONE to check results. Regrid the original
solution from the MAIN command line and enter the module CAPZONE. Position the cursor over
the well and enter
coordinate
surface
base
and CZAEM responds with
2.235174E-05 2.2351T4E-05
UNCONFINED: X.Y. PHREATIC SURFACE 2.235174E-OS 2.235174E-OS 2.924427E+01
X, Y. BASE 2.23S174E-05 2.23S174E-OS O.OOOOOOE+00
respectively. Now return to TRACE and enter
cursor off
This moves control from the cursor to the keyboard. Enter CAPZONE and generate a plot of
the solution (grid again if you wish to see piezometric contours instead of difference contours).
SURFACE and BASE may still be used to check data, but the commands must be followed by the
coordinates of the point to be checked. Check the phreatic surface elevation at coordinates (500,
500). Enter
44
-------
•urface 500 500
CZAEM responds with
UNCONFINED: X.Y. PHREATIC SURFACE O.OOOOOOE+OO O.OOOOOOE+OO 2.924426E+01
Note that SUBZONE, TIMEZONE, and WGENERATE may also be used in CURSOR OFF mode
by following the command with coordinates of the well of interest.
Validity of Solutions.
Sometimes a solution appears to be correct when observing the graphics, but may not be
valid. A clear example of this is a well which pumps the water table below the aquifer base or
below the actual vertical extent of the well. Checking the validity of a solution is always necessary.
CHECK, CURSOR, SURFACE, BASE, and COORDINATE have already been introduced as
methods of checking a solution. Here we will introduce CONTROL hi the module CHECK. Note
that CONTROL appears in check levels 1 and 2. CONTROL allows the user to check input and
computed values at control points. Entering CONTROL from check level 1 will produce a listing
of all control point coordinates, specified heads, and computed heads. Large differences between
specified and computed values indicate erroneous results. Return to the MAIN command line and
enter the module CHECK. Enter CONTROL and observe CZAEM's response. Now, enter check
level 2 by entering LINESINK and CONTROL. Only the line-sink control points are listed. The
CONTROL command is also located within REFERENCE in the CHECK module.
EXAMPLE 6. DATA FILE AND GRAPHICS CONTROL
A city located adjacent to a large river is expanding its corporate boundaries and developing
an industrial park. The city maintains two water supply wells each operating at 100 million gallons
per year; the existing wells are inadequate to handle new demands. Three new wells are proposed,
and their locations have been determined previously. The city wishes to enforce land-use zoning
near the proposed well-field to protect its water supply from contamination. The existing and
proposed well-fields are to be modeled and time zones delineated to aid in zoning decisions.
Accessing multiple data files contiguously.
Data files have been compiled for the model using an ASCII editor and are included in the
CZAEM directory. While data files can always be created in this manner, they can also be produced
with the assistance of the Geographic Analytic Element Pre-processor (GAEP) developed by Kelson
et. al., 1993. The data files and descriptions follow:
map.dat contains township and range lines, corporate boundaries, and proposed
industrial park limits;
line.dat contains line-sinks which model the major river and tributary near the city;
exist.dat contains aquifer parameters, rain, and existing well data; and
well. dat contains proposed well information.
call.dat contains call commands to the preceding data files.
Use of the data files will be taught by example. The user is encouraged to examine and
evaluate each data file line by line and to run the example by using call. dat and by calling each
data file individually. The model area is larger than previous examples, consisting of 12 townships.
Township boundaries are included in the map.dat file. Consistent English units (feet, days) are
used in this problem.
From the MAIN command line, enter the module SWITCH and read in the data
45
-------
call call.dat
The call.dat file includes calls to map.dat, line.dat, and exist.dat. All data will scroll past
the screen and control will return to the keyboard at the MAIN command line. All elements for
reproducing existing conditions have been read in. Solve, grid the solution, and plot the model of
existing conditions (Figure 6.1).
Figure 6.1 Model of existing conditions (plot; d 10).
Entering rainfall.
All elements in the model of the existing conditions have been discussed previously except for
< RAIN>. The RAIN element models a constant infiltration rate over a circular area; the user
must provide centroid coordinates, a radius over which the infiltration acts, and an infiltration
rate in units of length per time. The circle in Figure 6.1 shows the area of infiltration.
is contained in the module GIVEN. The command for the present model may be found
in data file exist.dat. Note that by itself creates a mound of water in the northwest
portion of the model.
Window manipulation and saving capture zone and time zone boundaries.
The proposed conditions may now be evaluated. Return to the module SWITCH and call the
data file containing the proposed well information.
call
well.dat
The data are read and control is returned to the keyboard at the MAIN command line. Solve and
grid the solution (Figure 6.2).
46
-------
Figure 6.2 Model of proposed conditions (plot; 780 5).
Within module CAPZONE move the cursor to the northernmost well and enter .
CZAEM responds with
CALCULATING SUBZONES PHASE 1: CREATING INITIAL PATHLINES FROM THE WELL
CALCULATING SUBZONES PHASE 2: DETERMINING LOCATION OF STAGNATION POINTS
CALCULATING SUBZONES PHASE 3: FILLING SUBZONE BUFFERS
CAPTURE ZONES CAN NOT BE CREATED:
NO STAGNATION POINTS FOUND IN THE WINDOW, CHANGE WINDOW SIZE
The scale of the current window is too large to evaluate the stagnation point caused by the well.
To produce a subzone, we must reduce the window size. First, store the current window plot with
the command. Enter
win push
The command stores the current window in a stack; the command
recalls the last window which has been stored in the stack. Now, reduce the window size with the
command. Position the cursor at the lower left corner of the township containing the
well. Enter
wll
CZAEM responds with
PLEASE REPOSITION CURSOR AND PRESS ENTER OR ANY OTHER KEY TO ABORT
Move the cursor upward and to the right. The cursor will drag a box with a lower left hand corner
at the initial position of the cursor. Move the cursor until the box encloses the area of interest
(the area in which you would expect the subzone to be created - in this case, the entire township
47
-------
containing the well) and press enter. The new window will be zoomed in on and a layout will be
displayed. Before creating the subzone with the command, enter
bsave
The command opens a file which stores all computed subzone and time zone boundaries.
CZAEM responds with
to abort
Enter the filename
well.tod
All subzones created will be saved in well.bnd until the user leaves module CAP ZONE or enters
the command. Leaving CAPZONE closes the file; entering erases the file.
The saved file may be recalled by the command. Now, move the cursor to the well
and create the subzone with the command. When the subzone is created, return to the
large scale window by entering
win pop
The original large-scale window will appear with the subzone drawn around the northernmost
well. Now, instead of using to create a smaller window around the remaining four wells,
use the command. Enter
window 34000 22000 87000 75000
The new window containing the four remaining wells will appear with a layout. Create subzones for
each well. This time we will return to a large-scale window by using the command.
Enter
win all
A large-scale window which includes all elements will appear with the subzones included (Figure
6.3). The shapes of the subzones appear in teardrop form and are finite. In this model, rain is the
only source of water in the area of interest.
Recall that subzone computations end at the border of the window in which they are computed.
As a result, subzone boundaries may be incomplete when viewed within a larger window.
Recall that was entered before creating any subzones, but we have not yet needed
. Computed boundaries will remain on screen until the module CAPZONE is left or
the command is used. To demonstrate the use of exit CAPZONE and
TRACE. Reset the window size
window 0 0 130000 130000
Enter CAPZONE and plot the layout. Note that the proper layout appears, but the subzones are
no longer present. To reproduce the subzones, read the boundary file well.bnd.
bre&d
CZAEM responds with
to abort
Enter the filename
well.bud
The subzones are displayed on the screen. Now create time zones for each well on which the city
will base land-use planning.
48
-------
Figure 6.3 Well-field subzones.
Obtaining a hardcopy of graphical output.
To obtain a hardcopy of Figure 6.1, we must route the output to a printer instead of the
screen. When CZAEM was installed on your computer, you were prompted for the type of printer
device you use. It is assumed here that you chose a postscript device and therefore will create a
postscript file instead of directly accessing a printer. This operation is done in the module PSET.
Return to the MAIN module and enter PSET.
ret
ret
pset
49
-------
CZAEM responds with the command line
\\\ ROUTINE SET PLOT NODE ///
(NUMBER) (/)
To print, type
printer
ret
plot
d 10
[enter]
[enter]
[enter]
This will create a postscript file of the plot called plot.ps. CZAEM sends all the graphics to the
file instead of the screen. To redirect graphical output to the screen, enter
pset
screen
To produce the hardcopy, exit CZAEM and print by typing
ret
•top
print plot.ps
Note that no plot appears on the screen as graphical output is redirected to the file. This poses a
difficulty in creating plots in TRACE and CAPZONE, where the cursor is used to identify wells
or starting points of streamlines.
To obtain hardcopies of streamline traces and capture zones, we first create the plots with the
screen as the graphical output device and record our input onto a file. We then direct graphical
output to the printer and retrace our steps to produce the plot using the recorded input file as a
guideline. In this way we can locate the cursor at desired locations without seeing it on the screen.
We turn the cursor off prior to creating the printer file, and manually enter the coordinates of
where to begin trace lines or capture zones.
To create the plot on the printer of the subzone boundaries for the well at (6.4e4,4.3e4) for
the window 34000 22000 87000 75000, enter the following commands:
window 34000 22000 87000 75000
grid 60
pset
printer
ret
trace
cursor off
capzone
d
[enter]
[enter]
6.4*4 4.3*4 subzone
ret
ret
•top
Entering the coordinates 6.4e4,4.3e4 in front of the subzone command has the same effect as moving
the cursor to the well at that location. The hardcopy is produced in the same manner as before.
The remaining commands in PSET are as follows. The command followed by
the number 1, 2, 3, or 4 results in different combinations of line colors in graphical output. The
50
-------
command will enable or disable the use of a mouse for cursor movement in graphics
mode. If your PC supports a mouse, CZAEM defaults the command MOUSE to ON unless
otherwise specified in the file initaem. dat. For consistency with this tutorial, the line 'mouse off'
was placed in the file. To default to MOUSE ON, remove the indicated line in the initaem.dat
file. The command is explained in the ASCII file read.me in the CZAEM directory.
51
-------
REFERENCES
Kelson, V.A., H.M. Haitjema, S.R. Kraemer, 1993. GAEP: a geographic preprocessor for ground-
water flow modeling, Hydrological Science & Technology, 8(1-4): 74-83.
Strack, O.D.L. Groundwater Mechanics, Prentice Hall, Englewood Cliffs, N.Y., 1989.
Strack, O.D.L., 1992: A mathematical model for dispersion with a moving front in groundwater,
Water Resources Research, 28 (11), 2973-2980.
USEPA, 1994. Program documentation for WhAEM (Wellhead Analytic Element Model), Robert
S. Kerr Environmental Research Laboratory, Ada, OK, in press.
52
-------
Command
Aquifer
permeability
thickness
base
porosity
reset
Given
help
return
i
uniflow
rain
reset
help
return
Reference
Well
given
reset
help
return
Linesink
given
head
string
tolerance
reset
help
return
Solve
Check
aquifer
summary
return
given
summary
uniflow
rain
COMMAND SUMMARY
Description
Input module for aquifer parameters
Hydraulic conductivity in [L/T]; default is 1.0
Thickness of the aquifer; default is 1.0
Elevation of the aquifer base; default is 0.0
Effective porosity; default is 0.3
Clears all the input data in module AQUIFER; resets to
default values
Extended help for the module AQUIFER
Exit to the MAIN menu
Input module for uniform flow and infiltration
Uniform far field component
Infiltration or evaporation rate at the top of the aquifer
in [L/T]
Clears all the input data in module GIVEN
Extended help for the module GIVEN
Exit to the MAIN menu
Enter the reference point parameters
Input module for wells
Well with given discharge
Clear all the well input data
Extended help for the module WELL
Exit to the MAIN menu
Input module for line-sink
Line-sinks with given discharge (per unit length of the
line-sink)
Line-sinks with head specified at the midpoint
Make a series of line-sinks to be treated as one source in
CAPZONE
Tolerance used for joining the nodes of line-sinks when
STRING is ON
Clear all the line-sink input data
Extended help for the module LINESINK
Exit to the MAIN menu
Solve the current problem
Check the solution and the input data
Check module for aquifer parameters
General information on aquifer
Exit to the CHECK menu
Check module for uniform flow and infiltration parameters
General information on uniform flow and infiltration
General information on uniform flow
General information on rainfall
Page(s)
6,12,30
6
6
6
6
24
5-6
6-7
7
46
24
5-6,7-8
8,15,17-18,32-33
10,16,34-35
10,16
24,35
5-6,10
13-15,42
14
14
42
39
24
5-6,14
8,10,17
15,31,45
32
5-6,32
32
53
-------
Check
help
return
reference
control
well
return
summary
range
input
control
help
return
linesink
summary
range
string
ends
bval
discharge
control
help
return
head
discharge
control
summary
help
return
Window
Map
curve
point
plot
reset
Extended help for GIVEN commands
Exit to the CHECK menu 5-6,32
Check module for reference point parameters 32-33,45
Comparison of the condition at the reference point with
the computed value 45
Exit to the CHECK menu 5-6
Check module for well parameters 32
General information on wells
Specify the start and end well numbers to be checked
Display locations and radii of all wells
Display the control point conditions and the computed,,
values for the wells 45
Extended help for WELL commands
Exit to the CHECK menu 5-6
Check module for line-sink parameters 32
General information on line-sinks 45
Specify the start and end line-sink numbers to be checked
Displays all the string information
Display end coordinates of line-sinks
Display boundary conditions of line-sinks
Display discharges of the line-sinks
Display the control point conditions and the computed
values for the line-sink 45
Extended help for LINESINK commands
Exit to the CHECK menu 5-6
Display the head value at a point 15,32,34
Display the discharge components at a point 32
Comparison of conditions at the control points with
computed values 32,45
General information about all the modules
Extended help for the module CHECK
Exit to MAIN menu 5-6,32
Set the viewing area; 8,10,46-48
WINDOW ALL sets the viewing area to include ALL
elements; 48
WINDOW PUSH saves the window setting and 47
WINDOW POP retrieves the window settings 47,48
(in the order in which they were saved via PUSH);
WINDOW without any options displays the current
viewing area
Input module for a map or diagram of the modeled area 11,33
Begin entry of curve coordinates 11,33
Begin entry of point coordinates 33
Turn display of the map ON or OFF 11
Clear all the MAP input 24
54
-------
Map
help
return
Layout
Grid
Plot
Trace
window
tolerance
cursor
switch
prefix
input
output
messages
error
log
call
back
help
return
set
maxstep
backward
front
marker
help
return
plot
base
surface
coordinate
trace
backward
wgenerate
wll
tol
command
Extended help for the module MAP
Exit to MAIN menu 5-6
Plot all the elements within the current window on the
screen 15,34
Compute head values at the nodes of a mesh; used by
PLOT to create head contours 8
Plot the contours computed by grid on the screen 9
Determine streamlines or capture zones 10,19
See extended help for this command
Tolerance used for determining which well the cursor is on 39
See extended help for this command 44
Input/output operations 40-41,42,45
Specify DOS path for the input and output files 41
Read a data file. With ECHO, copy input to a file 41
Write output to a file. With ECHO copy output to a file
Write messages to a file. With ECHO copy messages to a
file
Write errors to a file. With ECHO copy errors to a file
Create a log of all input/output operations 41
Read a data file 41,45-46
Return control to an input file from SWITCH
Extended help for the module SWITCH 41
Exit from SWITCH module 5-«
Set TRACE options
Set the maximum step size for tracing the particle
pathlines
Set pathlines to be trace in the backward direction 19
Activate computation of solute front; requires an optional
velocity factor multiplied by the average velocity to
compute the front position. 37-38
See extended help for this command
Extended help for SET commands
Exit to the TRACE menu 5-6
Plot the piezometric contours and allow user to trace
pathlines 10,20
Display base of the aquifer at the cursor location 44
Display surface of the aquifer at the cursor location 44
Display coordinates of the cursor location *' 44
Determine and display streamline through cursor location 10,34
Enable backward tracing of pathlines 19-20
Generate a specified number of pathlines from a well by
backward tracing 19-20,34
Set the lower left corner of the new window 47
Set the tolerance for well identification graphically 39
Display command words 14
55
-------
Trace
menu
return
layout
base
surface
coordinate
trace
backward
wgenerate
wll
tol
command
menu
return
capzone
coordinate
base
surface
window
wll
subzone
timezone
source
nline
page
help
command
return
front
wgenerate
color
bsave
bread
help
return
Exit to the TRACE menu
Exit to the MAIN menu
Display layout of elements and allow user to trace
pathlines
Display base of the aquifer at the cursor location
Display surface of the aquifer at the cursor location
Display coordinates of the cursor location
Determine and display streamline through cursor location
Enable backward tracing of pathlines
Generate a specified number of pathlines from a well by
backward tracing
Set the lower left corner of the new window
Set the tolerance for well identification graphically
Display command words
Exit to the TRACE menu
Exit to the MAIN menu
Draw capture zones for a well
Display coordinates of the cursor location
Display base of the aquifer at the cursor location
Display surface of the aquifer at the cursor location
See extended help for this command
Set the lower left corner of the new window
Create subzones for the well at the cursor position
Create time zones for the well at the cursor position
List the sources contributing water to the well
Specify number of pathlines to be used to determine the
capture zones
Clear the screen and erase BSAVE file contents
Extended help for CAPZONE commands
Display the CAPZONE commands
Exit to the TRACE menu
Set the velocity factor for the solute front used in drawing
the time zones
Generate a specified number of pathlines from a well by
backward tracing
Specify colors for different line types;
COLOR1: dividing streamlines
COLOR2: time zones,
COLORS: subzone envelopes.
Specify a file into which subsequent capture zone
boundaries will be saved
Clear screen, Read BSAVEd file, and draw capture zone
boundaries
Extended help for the module TRACE
Exit to the MAIN menu
5-6
5-6,21
20
44
44
44
10-11,34
19-20
19-20,34
47
39
14
5-6
5-6,21
22,34,35,44
44
44
44
47
22,34,35-36,46-48
36-38
24,26
14
5-6
37-38
19-20,34
48
48
5-6
56
-------
Cursor
tolerance
switch
plot
prefix
input
output
messages
error
log
call
back
help
return
coordinates
head
discharge
tolerance
whnove
Ismove
wll
command
menu
return
layout
coordinates
head
discharge
tolerance
wlmove
Ismove
wll
command
menu
return
help
return
Help
Switch
prefix
Enter CURSOR module for graphical data retrieval 31,38
Set the tolerance with which the cursor can detect an
element 39
Input/output operations 40-41,42,45
Specify DOS path for the input and output files 41
Read a data file. With ECHO, copy input to a file 41
Write output to a file. With ECHO copy output to a file
Write messages to a file. With ECHO copy messages to a
file
Write errors to a file. With ECHO copy errors to a file
Create a log of all input/output operations 41
Read a data file 41,45,46
Return control to an input file from SWITCH
Extended help for the module SWITCH 41
Exit from SWITCH module 5-6
Plot the piezometric contours and activate cursor
Display the coordinates of the cursor location 44
Display the head at the cursor location
Display the discharge at the cursor location
Set the tolerance with which the cursor can detect an
element 39
Move a well and optionally change its discharge; also
identify a well by its number 38-40
See extended help for this command
Set the lower left corner of the new window 47
Display the command words in the module CURSOR 14
Exit to the CURSOR menu 5-6
Exit to the MAIN menu 5-6
Display layout of elements and activate cursor
Display the coordinates of the cursor location 44
Display the head at the cursor location
Display the discharge at the cursor location
Set the tolerance with which the cursor can detect an
element 39
Move a well and optionally change its discharge; also
identify a well by its number 38-40
See extended help for this command
Set the lower left corner of the new window 47
Display the command words in the module CURSOR 14
Exit to the CURSOR menu 5-6
Exit to the MAIN menu 5-6
Extended help for CURSOR commands
Exit to the MAIN menu 5-6
Extended help for the command words
Enter SWITCH module 40-41,42,45
Specify DOS path for the input and output files 41
57
-------
Switch
input
output
messages
error
log
call
back
help
return
Save
Read
Pause
Reset
Pset
printer
screen
driver
palette
mouse
help
return
Stop
Read a data file. With ECHO, copy input to a file 41
Write output to a file. With ECHO copy output to a file
Write messages to a file. With ECHO copy messages to a
file
Write errors to a file. With ECHO copy errors to a file
Create a log of all input/output operations 41
Read a data file 41,45,46
Return control to an input file from SWITCH
Extended help for the module SWITCH 41
Exit from SWITCH module 5-6
Save a solution or grid in binary format for future use 17,41
Reset the program and retrieve a solution or grid 18-19,42-43
Pause from CZAEM to access DOS 41
Clears the program of all the input data 24,41
Sets the graphical output 49-51
Sends graphical output to the printer or a file 50
Sends graphical output to the screen 50
see read.me file for details
Sets the color attributes of the screen (NUMBER = 1,2,3 or 4)
Turns the mouse on or off 51
Extended help for the module PSET
Exit from the PSET module 5-6
Exit the program
12
58
•&U.S. GOVERNMENT PUNTING OFFICE: l»*4 - 5S040I/OOIM
-------
Solute
Transport
Modeling
-------
9. Solute Transport Modeling
-------
Solute Transport
The Transport Equation
General mass conservation
O
where J is the mass [M]
Jm is the mass flux [M/L2T]
t is the time [T]
V« is the divergence operator
For the contaminant in the ground water at concentration C
+ V- ( gC - 0£ VC) =0 (2)
ot
where 0 is the porosity [L3/L3]
C is the concentration [M/L3]
q is the darcy flux [L/T]
D is the dispersion coefficient [L2/T]
V is the gradient operator
-------
MASS BALANCE FOR SPECIES
Rate of change
of mass in control,
volume per time
unit
Rate of transport
of mass into and
out of control,
volume per time
unit
» •
o
TRANSPORT TERM
- inflows
- outflows
Rate of
transformation
of mass in
control, volume
per time unit
O
TRANSFORMATION
TERM
- biological reaction
- chemical reaction
- physical change
Transport Term= Dispersive Flux + Advective Flux
23. Formulation of the solute transport equation.
-------
Dimensions
Each term in the transport equation has the same units:
i M V i V /A M 1 V T 2 i M
JL M yw -L vyl **• ijx -L / vw\ L, ± Lix ,,.
~T~V~V ' ~L T ~V ' ~L^~V'~T~L~V
where L is length
T is time
Vw is the volume of water
VT is the total volume of the media (voids + solids)
Mx is the mass of chemical x
for each term
—— (4)
VTT
-------
Advective and Dispersive Fluxes
The flux is composed of an advective flux
qC (5)
which gives the contribution to contaminant transport due to the flow of the
water, and
a dispersive flux,
F= -
i. e.
(6)
ac
where F is the diffusive and dispersive flux [M/L2T]
D is the dispersion coefficient [L2/T]
C is the concentration [M/L3]
which gives the contribution to transport due to diffusion and "dispersion"
The dispersive flux is assumed to follow Pick's law of diffusion.
-------
Dispersion
As flow occurs, a solute or tracer gradually spreads and occupies an
ever-increasing portion of the flow domain, beyond the region it would
occupy due only to the average flow.
D = a v + D* (7)
where D, is the coefficient of hydrodynamic dispersion in the direction i [L2/T]
tt| is the dispersivity in direction i [L]
v is the seepage velocity
D* is the coefficient of molecular diffusion
Gelhar, L.W., C. Welty, and K.R. Rehfeldt, 1992, A critical review of data on
field-scale dispersion in aquifers, Water Resources Research, 28(7),
1955-1974.
The Peclet number represents the ratio of advective to dispersive fluxes
-------
(A)
Hypothetical
Velocity
Distribution
Tracer at
Time*0
fmm
max
H
(B)
Tracer Distribution
at Time
Scale Dispersion
Multiple of
Figure AITJ-1. Part (A) shows a hypothetical velocity distribution and an initial distribution of
tracer while part (B) shows how the tracer would be dispersed by the moving
groundwater at several different scales. Three common mechanisms of pore
scale dispersion (velocity variation within a pore (a); flow path tortuosity (/J), and
molecular diffusion due to concentration differences (7)) are illustrated also.
-------
(A)
Tracer at
time=0
B*_
.2 1
a>
u
c
o
u
Tracer at later
times t, and tt>0
(B)
Gradient (slope
f curve)
Displacement
Figure 1-4. Schematic diagram showing the inherent lack of vertical contaminant
concentration structure that results from the solution of vertically-averaged
transport models.
-------
c
o
s'i-o
ii-
O> O <->
O
1
0.5
0
Tracer front if
diffusion only
Distance x
v position of input
water at time t
Dispersed
tracer front
Figure 9.2 Schematic diagram showing the contribution of molecular diffu-
sion and mechanical dispersion to the spread of a concentration
front in a column with a step-function input.
-------
Advective-Dispersive Transport
CO
d
o
"S
CD
O
d
O
O
0.0m
1.0m
5.0m
10. m
0
0
Time (d)
7-7
-------
1968
GELHAX ET AL.: FIELD-SCALE DISTEKSION IN AQUIFERS
io4r—
103
H"2
1 10°
ficr1
1CT
o •
£3go 00 O
o °
RELIABILITY
• low
o intermediate
O high
10'1 10° 101 102 103 104 105 106
Scale (m)
Fig. 2. Longitudinal dispersivity versus scale with data classified by reliability.
10
103
i
|to1
a
I 10°
MB
§
fio-1
I
ID'2
-•-3
11111
RELIABILITY • .,
• tow
o irtermediatB
O HO*1 . -^
• * .
c-o
• -
o 0
O
0 -
10"1 10° 101 102 103 104 105
Scale (m)
Fig. 4. Horizontal transverse dispersivity as a function of obser-
vation scale.
-------
GELHAX ET AL.: FIELD-SCALE DBTEXSION IN AQUIFERS
1971
10
RELIABILITY
. tow
A intermediate
A
101
102
104
103
Scale (m)
Fig. 5. Vertical transverse dispersivity as a function of observation scale.
105
= io~
e
i
CD
o
S "2
0
01
f 1.1
c
i
° 10°
B
i
^
-------
Hydrophobic Theory of Sorption
By assuming a linear equilibrium isotherm
M V M
LAX W L*X
M M V
s s w
(11)
where Cs is the solids concentration of the chemical
kd is the partition coefficient [V^IVy
C is the aqueous concentration of the chemical
Mx is the mass of chemical
Ms is the mass of soil
Vw is the volume of water
The sorption coefficient
Kd is the sorption coefficient
Koc is the organic carbon partition coefficient
foc is the fraction of organic carbon in the soil
-------
Hydrophobic Theory of Sorption
Sorption was noted not to vary by chemical type but rather by the amount
of organic carbon present. So the sorption (assuming hydrophobic
partitioning) is occurring to the organic carbon present in the system.
Hydrophobic theory valid only for foc > 0.001; otherwise sorption of organic
compounds on nonorganic solids can become significant.
The linear isotherm is valid only if the solute concentration remains below
about one-half of the aqueous solubility of the compound.
A variety of empirical equations have been developed to estimate Koc.
These are largely unneeded now because of the number of measured Koc's
For example the octanol/water partition coefficient
Ko» = — (13)
where Kow is the dimensionless octanol/water partition coefficient
c0 is the concentration of chemical in octanol
cw is the concentration of chemical in water
Koc = 0.411^ (14)
-------
Solubility, K^, and mobility class for common organic pollutants
Compound
1,4-dioxane
4-hydroxy-4-methyl-2-pentanone
acetone
tetrahydrofuran
N,N'-dimethylformamide
N, N '-dimethylacetamide
2-methyl-2-butanol
2-butanol
ethyl ether
cyclohexanol
3-methylbutanoic acid
benzyl alcohol
aniline
2-hexanone (butylmethyllcetone)
2-hydroxy-triethylamine
2-methylphenol (o-cresol)
2-methyl-2-propanol
4-methylphenol (p-cresol)
pentanoic acid
cyclohexanone
4-methyl-2-pentanone
2,4-dimethyl phenol
4-methyl-2-pentanol
methylene chloride
isophorone
phenol
2-chlorophenol
hexanoic acid
chloroform
1 ,2-dichIoroethane
1 ,2-trans-dichloroethene
chloroethane
5-methyl-2-hexanone
chloromethane
1 , 1 -dichloroethane
1,1,2-trichloroethane
1,2-dichloropropane
benzoic acid .
octanoic acid
heptanoic acid
1 , 1 ,2,2-tetrachIoroethane
benzene
Solubility
(ppm)
miscible
miscible
miscible
miscible
.
140000.
125000.
84300.
56700.
42000.
40000.
34000.
35000.
31000.
24000.
24000.
23000.
19000.
17000.
17000.
13200.
12000.
82000.
11087.
11000.
7840.
8450.
6300.
5700.
5400.
5380.
5100.
4420.
3570.
2900.
2500.
2410.
3230.
1780.
*.
1
1
1
1
1
2
6
6
8
10
12
12
13
14
15
15
16
17
17
18
20
21
21
25
26
27
27
28
34
36
39
42
43
43
45
49
51
64
70
71
88
97
Mobility
Class
very high
very high
very high
very high
very, high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
very high
high
high
high
high
high
high
-------
Mass conservation
(QC+pbkdC) = -V-(gC-6£>VC) (17)
Expanding the derivatives by the chain rule
0 £ + c + * = -<2V-C ~ cV-g - 6V-DVC ( is >
Collecting terms
(0 + p^d) + C ( + V-g) = -gV-C + 6V-DVC (19)
Noticing the mass conservation equation for the water and dividing by the
porosity
o ot \j
(20)
-------
Mass Conservation
Writing in terms of the retardation coefficient, Rd and the seepage
velocity, v
V-DVC (21)
dt
Writing out by components (x, y, z):
dt x dx y dy z dz
lD l£\ __
dx\x dx) dyy dy dz z dz
(22)
<>\ dD
-------
Non Linear Partitioning
Solute concentrations are high, activity coefficient not constant
Mechanjsm other than hydrophobic sorption
M.L. Brusseau, P.C.S. Rao, 1989, Sorption Nonideality during organic
contaminant transport in porous media, Critical Reviews in
Environmental Control, 19(1), 33-99
-------
40 Critical Reviews in Environmental Control
-0.8
0.6
o
0.4-
0.2-
H-1.3
P-SO
R-2
PORE VOLUMES
FIGURE 4. Breakthrough curves for linear and nonlinear torption itotherms.
42 Critical Reviews in Environmental Control
2 4 6
PORE VOLUMES
8
FIGURE 6. Equilibrium and nonequilibrium breakthrough curves.
9-AI
-------
Boundary Conditions in Solute Transport
Zheng, C., 1990, MT3D A Modular Three-Dimensional Transport Model for
Simulation of Advection, Dispersion and Chemical Reactions of
Contaminants in Groundwater Systems, S.S. Papadopulos and
Associates, Rockville, MD.
also
Nofziger, D.L., K. Rajender, S.K. Nayudu and P.Y. Su, 1989, CHEMFLO
One-Dimensional Waterand Chemical Movement in Unsaturated Soils,
USEPA, EPA/600/8-89/076.
Boundary Condition
Form
Example
Specified
concentration
(Dirichlet)
C(x,y,z,t) = Cc
Source of constant
concentration
Specified gradient
(Neumann)
,y,z, t)
Along impermeable
boundaries q = 0
At outflows q = 0
allows the chemical to
flow out of the
domain
Specified
concentration and
gradient
(Mixed)
dx
=g(x,y,z, t)
Along impermeable
boundaries g = 0
Customarily -vC = g
is used on inflow or
outflow boundaries
-------
An example Analytical Solution
Governing equation:
R
d_c
dt
dx'<
-v
dc
dx
c
v
D
R
solute concentration [M/L3]
seepage velocity [L/T]
Darcy velocity divided by the porosity (v
Dispersion coefficient [L2/T]
Retardation coefficient [*]
Initial Condition
c(x,Q)
0
Boundary Conditions
c(0,t) =
'*>. = 0
dx
c(0,t) =
dc/dt = 0
c(x,0) = 0
-------
Solute Transport Examples
Parameter values
v = 10.0 m/d
D = 10.0 m2/d (a= 1.0 m)
Rd=1.0
Initial Condition: C = 0.0 mg/L
Boundary Condition: C = 1.0 mg/L
Pulse Duration = 10.0 Days
-------
Effect of Dispersion at 10.0 Meters from the Source
en
c:
o
CD
O
d
O
O
A—A a = 0.1 m
a—a a = 1.0 m
o—o a = 10. m
Note: D = a Ivl
Seepage Velocity
(m/d)
10
10
10
Dispersivity
(m)
0.1
1.0
10.
Dispersion
Coefficient
(m2/d)
1.0
10.
100.
-------
Effect of Dispersion on Pulse Sources
Pulse duration = 1.0 days
O)
o
"•a
O
c
o
0-8
0.6
0.2
0
0
A—A oc = 0.1 m
oc= 1.0 m
2 3
Time (d)
-------
Effect of Retardation at 10.0 Meters from the Source
0
234
Time (d)
?-**
-------
Effect of Retardation on Pulse Sources
Pulse duration = 1.0 days
0
1 2 3
Time (d)
4
-------
Accurate analytical
solution
Figure 7-8 Schematic illustrations
of oscillations in numerical solu-
tions or an advancing front (after
Bender tt al.. 1975 and ran Ce-
nuchien. 1976).
-------
Case Study
K.M. Freeberg, P.B. Bedient, and J.A. Connor, 1987, Modeling of TCE
Contamination and Recovery in a Shallow Sand Aquifer, Ground
Water, 25(1), 70-80.
D 5000 gallon tank used to store trichloroethene (TCE) and toluene
D periodic accidental spillage and dissolution of the seams of the tank
D Estimated total release of TCE of 8 kg (176 ug/l in 12.3 x 106 gallons)
D Sand deposits 10 to 35 feet below the ground surface
D Clay layers cause the contamination to be in a confined unit
D study area of 700 by 700 feet, boreholes to 37 feet
D 1.5 year period of monitoring water levels and chemical concentrations (for
calibration)
water levels fluctuate as much as two feet over a one-year period;
the direction of the gradient is mostly constant
Data from constant-discharge pump test
D Wells were screened over the entire thickness of the aquifer; thus the
concentrations are vertically averaged.
-------
A recovery system was designed to:
Dcontain the plume
Dminimize the amount of water pumped
Dminimize costs
-». four wells to be pumped at 30 gpm
Goal of the modeling:
to predict how long the system would have to run to achieve the
desired cleanup
-------
M M H H 69 H
69 69 H 69 M 64
M M 64 M *» 64
U M 68 68 « «
U M 68 M 68 M
M M M U M U
U U U U M M
U U U U M M
M U U U M M
(7 (7 iT i7 (7 tT
67 *T *7 67 t7 *7
67 17 67 (7 (7 (7
67 (7 47 67 i7 67
67 67 67 67 67 67
67 67 67 67 67 67
67 67 67 67 67 67
67 67 67 67 67 67
66 66 66 66 66 66
M 66 66 66 66 M
66 66 66 66 66 66
66 66 66 66 66 66
66 66 66 66 66 66
66 66 66 66 66 66
64 64~«9 64 MJU U
64 64 69 69 [a 68 H
.** **
U 68
68 68
68 68
68 68
68 68
68 68
67 67
67 67
67 67
67 67
67 67
67 67
67 67
66 66
66 66
*%>/"
68 68 68 68 68
68 68 68 68 68
68 68 68 68 68
68 68 68 68 68
68 68 68 68 68
68 68 68 68 68
67 67 67 67 67
67 67 67 67 67
67 67 67 67 67
67 67 67 67 67
67 67 67 67 67
67 67 67 67 67
67 67 67 67 67
66 66 66 [ 67 67
66 66 66 66 66 !
68
U
68
68
68
68
68
68
67
67
67
67
67
67
67
67
67
66 66 66 66 66 66
66 66^6 66 66 66 66
66 66
66 66
66 66
66
66 66 66 66 66 66
66 66 66 U 66 66
66 66 66 66 66 66
U
68
68
68
68
68
68
67
67
67
67
67
67
67
67
67
67
67
66
66
66
66
U H U
U U U
a 68 a
68 68 68
68 68 68
68 68 68
68 68 68
67 67 67
67 67 67
67 67 IT
67 67 67
67 67 67
67 67 67
67 67 67
67 67 67
67 67 67
67 67 67
67 67 67
67 67 67
66 | 67 67
66 M] 67
66 66 66
•»^
MW1
(68.51)«> CQ KH-
_w—-v OO.OU
(68.49) »MW4
(68.44)
MW7
- -C67
•MW5
(68.24)
*MW6
(68.16)
/ «r
.84)
(66.93)
.MW10
--_ ,67.00--
(66.93)
WATER TABLE ELEVATION IN FEET
(a)
Fig. 3. (a) Simulated head distribution map.
scale i
(b)
100
200
lecl
(b) Potentiometric surface contours constructed from measured data.
-------
Computer Code
DUSGS MOC model
2D confined aquifer, solute transport
steady ground water flow, transient solute transport
D20 by 35 cell grid (50 by 50 ft)
D15 observation wells
inconstant head boundaries
(the MOC code places no-flow boundaries around the domain)
D the ditch
D the up gradient boundary to drive flow and "produce the
observed migration of the plume"
-------
•-axis •
123466789 10 11C O M6«t7B«20
i
>.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
KEY
§ constant head node
with potantiometrfc
value indicated in
box
@ monitoring well
Fig. 2. Grid of study area.
Rl-4e recovery well
Ol-5=
designated observation
well in simulations
-------
Calibration parameters
Parameter
Effective porosity
Longitudinal dispersivity
aT/aL
' y\r ' xx
Saturated thickness
Background concentration
Range of Values or Value
0.3
10 -20 feet
0.1 to 0.3
1.0
25ft
0
Transmissivity 0.003 ft2/s varied by 23%
-------
10
0 1 4 7 12 12 t
> 4 11 21 17 15 1C
) 10 2* »2 105 10*
(.•HI
22 tO lit 2)1
1 '*
)> 112
10 45 It7 )lt 744
251 410 4M 17» 45
'Tj :[T1
2tt t7 IS
f 4) 17? 450 II* ?|* It' t« it
741 12) I? 14
S 11 lit 155
ilOll
•>«?
mi
It 71 21) tS* tOO 24f tT
t 11 Itt 472 417 It! J5
2 3
2 0
1 0
1 0
1 0
*l
2 !12 7J 2*1 27J If 17
101
000
^^ Imioi
0 0 0 0 ) 21 144 144 44
0 0.0 0^0
o^r o
78
21
21 40 10
2 0 0
1 0 S
000
000
toit* indic*tt Monitoring veil location*.
•old figure* rtpr«s*nt ob»«rvtO concentration*
of TCE lua/n. provided (or coap«ri»on to
50 100 !••!
Fig. 4. TCE plume predicted at end of monitoring period.
•ourt*
A) 0.5 YEAR
R1
0 100 R
C) 1.5 YEAR
,R2
B) 1.0 YEAR
8eurl«*R1
o won
0) 2.0 YEAR
•R2
Fig. 5. Predicted concentrations of TCE (pg/I) «t points in
time after start-up of pumping at wells R1-R4.
-------
Contaminant source:
D Rate of release of TCE is unknown
D Source was modeled as an injection well so that 8 kg of TCE entered
the aquifer
1500 ug/l at 0.012 cfs for 1/2 year
Application of the Model
Model was run to generate the plume
Calibration with data from before the recovery system was started (one
year) more than 100 runs of the model were required
Remedial activities were simulated
The hydraulic gradient and contaminant loading were found to be the
most important parameters.
Attempted to minimize error between predicted and measured
concentrations, the monitoring wells, and match the peak concentration
-------
Model was tested at two times:
D 0.16 years and 0.47 years after pumping began
D Prediction of 2.0 years to reduce TCE concentrations to 6 ug/l or less
-------
SOUKS
*«•
0 0 0 0 1 4 10 21 41
j 144
0 0 0 1 * 14 It 171 1 171
0 0 0 1 8 3» 102 2St S21
0 0 0 I « 27 100 40S t4*
0 0 0 0 i ' 24 17« S24
^~\ i i'J i
0 0 0 0 j 0 : 1 it 74 402
^s"\ '
0090004 u m
o o o jo o^o^; u . 102
A ^\ 'tun
Botff* indicttt .oAitorino, v«H locations.
of TCZ Iva/tl. provided (er »ap«riion to
IS • 2 0 «
19*
41 21 t 1 0
ten
120 SS It 3 0
2IS US 2* ' 1
)47 1S1 41 1 1
47J 21* fl 12 2
S27 2tS tl IS 2
144 117 JO > 1
271 7» It 2 0
1S4 4* 6 1 3
1 71 1) 4 0 0
j
\
\
. , ,
1
I
d
V
SOUKS
rn
Lad
* i
rr— Jfiaj
4 o o j : i i i i .7 ii « j i o
Jfili' m -n m
i>»» . . l**\* !>WM
1 *•
ri
^^^^N* n
^"*
k>». ixdictt* Kx.torinq v.U loc.Uo/u. .
f TCt (wq/1). prov;d«d (or ;o*p«ri«oA to
• IIM* predicted «cro«s th« arid.
Fig. 6. TCE plume predicted after 0.16 years pumping.
Fig. 7. TCE plume predicted after 0.47 years pumping.
-------
Sources of Error (Freeberg et al., J 987)
Initial Period ,.:,...
Release rate is unknown (actual varying release over 15 years)
Inaccuracies between model and actual well locations
(up to 15 feet) .
Water discharge at MW-13
Recovery Period
Malfunction of pumping at MW-11
Declining pumping rates in the field
Unaccounted-for heterogeneity
-------
Other Comments
D Declining concentrations could rebound after pumpage ends
D TCE introduced as a NAPL
the model cannot account for a NAPL
the injection rate of
D TCE degradation products found at the site . . .
(probably would not have been recognized in 1987)
D Model is for a confined aquifer
how does pollutant enter ?
how is the ditch a physical boundary ?
D Ground water contours developed from sparse data set
-------
-------
Vadose Zone
Flow
-------
10. Vadose Zone Flow
-------
Vadose Zone Flow
Vadose Zone <-> Unsaturated Zone
D Subsurface region between the ground surface and the water table
Darcy's law is applied to vadose zone flow:
D The hydraulic conductivity depends on the amount of water in the
pore space
K=K(QW) (14)
D The pressure depends on the amount of water in the pore space
Pc = P,,., - P,
note
and
h
if the air phase pressure is assumed to be atmospheric (Pair = 0)
c
/£>-/
-------
Vadose Zone Hydrostatics
(atm)
V
Negative
Hydrostatic
Pressure
h
Positive
Hydrostatic
Pressure
-------
Capillary Fringe
'w
-------
Unsaturated Hydraulic Conductivity
Vadose Zone conductivity depends on the mositure content
Ks'= K(0)
-300
-250
fi -200
1
-150
-so
-300
-250
fc -200
E -150
-100
-50
10 20 30
Moisture content, 0 (percent)
2X10-4 3X10"4 4X10-4
Hydraulic conductivity (cm/sec)
FIGURE 4.24. Idealized curves showing relationships of soil moisture, 9. hydraulic
conductivity, K, and soil-moisture tension head,
-------
Richard's Equation (1931)
The unsaturated Darcy's Law and Mass Conservation:
dt
- K( 0 ) \z + -2-
\ 99
= 0
One equation for vadose zone flow
(flow of the air phase is not modeled)
air is included implicitly in the
unsaturated conductivity K(0) and
capillary pressure (Pw = Pc - Pa)
/€>-<>
-------
Moisture content. 9
0 0.1 0.2 0.3 0 0.1 0.2 0.3
"T o
-Porosity
Capillary
fringe
\
Water
table
0 0.1 0.2 0.3 0 0.1 0.2 0.3
-i 1 1—t 0
FIGURE 4.25. Moisture profiles showing the downward passage of a wave of infil-
trated water. The soil is saturated at a moisture content of 0.29 and has a field capacity
moisture content of 0.06. SOURCE: Modified from Agronomy Monograph 17, "Drain-
age for Agriculture," 1974, pp. 359-405. Used with permission of the American Society
of Agronomy.
-------
NAPLs
-------
11. Introduction to Nonaqueous
Phase Liquids (NAPLs)
-------
Multiphase Flow of Contaminants
Many organic contaminants are found as water-immiscible liquids
;
Nonaqueous Phase Liquid (NAPL)
A liquid subsurface contaminant that is immiscible with water.
D Darcy's Law is applied to each phase.
Vadose Zone: Three Mobile Phases: Water, Organic Liquid.Air
Aquifer: Two Mobile Phases: Water, Organic Liquid
D Capillary pressures are defined for each pair of phases.
Pc(wa) ~ Pa Pw
P = P — P
Jrc(wo) o w
P = P - P
^c(oa) ^a ^o
(19)
-------
Introduction to Nonaqueous Phase Liquids
(NAPLs)
A NAPL is a subsurface liquid contaminant which is immiscible
with water.
(i.e., the NAPL doesn't mix with water as there exists an interfacial
tension between the NAPL and the water)
Typical NAPLs less dense than water (L-NAPL)
Crude Oil
Gasoline
Diesel
Jet Fuels
Kerosene
Other petroleum products
Typical NAPLs more dense then water (D-NAPL)
Organic Solvents: Perchloroethylene (PCE)
Trechloroethylene (TCE)
Carbon Tetrachloride
Creosote
Coal Tar
Other Chemical Mixtures
-------
Significance of NAPLs
One of the major reasons why pump and treat remediation schemes
fail to meet their cleanup goals in a reasonable time is that undetected or
ignored NAPLs are present.
Why?
Although the NAPLs are immiscible with water, chemicals dissolve out
of the NAPL phase and contaminate subsurface water
Hydrophobic contaminants reside in the NAPL phase with only
a small fraction of the contaminant mass residing in the water phase
NAPL phase which are themselves compounds of regulatory
interest dissolve slowly into the water.
-------
Factors Affecting NAPL Phase Flow
Fluid Properties
Density and Viscosity
Capillary Pressure
Trapping & Entry Phenomena
Effective Hydraulic Conductivity
Multiple fluids in the pore space reduces the conductivity
to each
-------
Density and Viscosity
Saturated Conductivity
Ks = saturated conductivity (media and fluid property)
k = intrinsic permeability (media property only)
p = fluid density
M = dynamic viscosity
Q = gravity
-------
Density and Viscosity
Liquid
Methylene
Chloride
TCE
PCE
Gasoline
Carbon
Tetrachloride
Water
No. 2 Fuel Oil
Transmission
Fluid
Aroclor 1254
Density
g/cc
1.33
1.47
1.60
0.74
1.59
1.00
0.87
0.89
1.51
Viscosity
cP
0.426
0.566
0.900
0.45
0.970
1.00
5.9
80
2050
Density/
Viscosity
3.12
2.60
1.78
1.64
1.64
1.00
0.15
0.011
0.00074
(properties at 20° C)
-------
Interfacial Tension
Differences in molecular forces of attraction cause there to be
interfacial tensions between the phases.
Gas
/ / / /
Solid
-------
Capillary Pressure
Because of the differing forces of attraction between immiscible fluids and
the matrix solids, there is a pressure difference between wetting and
nonwetting fluids, which is called the capillary pressure, Pc
P = P - P
1 c ' nw w
Pnw = pressure in the nonwetting phase
Pw = pressure in the wetting phase
In a single, uniform, circular capillary tube of radius, r:
2o cos6
Pc = Capillary Pressure
a = Interfacial (or surface) tension
9 = Contact Angle
r = tube radius
Note: Pc <* a cos9
Pc oc 1/r
The capillary pressure is related to the capillary rise or capillary head,
PC = (Pw - Pnw) 9 hc
-------
V
hc= Capillary Rise (Capillary Head)
r - Tube Radius
-------
Capillary Pressure in Porous Media
In a porous medium the capillary pressure also depends on
1) the geometry of the pore space
2) the amount of fluids present
There is not a simple equation for the capillary pressure in a
porous medium.
PC = PC(S)
S = Fluid Saturation
(fraction of the pore space filled by a fluid)
//-/o
-------
Water/NAPL Capillary Pressure Curve in a Fine
Sand
1. Residual Saturations
2. Nonwetting Phase Entry Pressure (Head)
Oil Creek Sand
Modified Su and Brooks Method
Kater/Soltrol
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-0
-------
Water/NAPL Capillary Pressure Curve Showing
Hysteresis
100
60
Capillary
Bead
cm Kater
Oil Creek Sand
Modified Su and Brooks Method
Weter/Soltrol
; -0.0 0.1 0.2 0.3 0.4 OJS 0:€ 0.7 0.8 0*9 1.0
Water Saturation
-------
Example Calculation
Vadose Zone Residual NAPL Saturation
Porosity
Disposal Area
Water table Depth
0.05
0.35
25m2
10 m
If:
1) the medium is uniform
2) the NAPL is uniformly distributed
3) there are no macropores, preferential flow paths, or fractures
then
the maximum amount of NAPL retained in the pore space is
(25m2) (10m) (0.35) (0.05) = 4.4m3
Disposal volumes in excess of 4.4 m3 (21 drums) reach the water table.
J/-/3
-------
Effective Conductivity
When two or more fluids fill the pore space the conductivity to
any one of the fluids is reduced because of the presence of the other
fluids. (This effect is in addition to density and viscosity effects)
Kei = Ksikr(S) =
Kei = effective conductivity to fluid i
Ksj = saturated conductivity to fluid i
kr(S) = relative permeability to fluid i
k = intrinsic permeability
Pi = density of fluid i
Uj = dynamic viscosity of fluid i
g = gravity
-------
Relative Permeability
Relative permeability
Incorporates multiphase phenomena into the conductivity
blocking of pores
increased tortuosity
Ranges from 0 to 1
0 at residual saturation = no flow
1 at full saturation == saturated conductivity
100%
Fie. 9.3-3. tffect of hysteresis on relative permeability.
Bear, 1972
-------
Multiphase Flow Equations
Darcy's Law for Multiphase Flow
kk
qf = flux of fluid i
k = intrinsic permeability
krj = relative permeability
pj = density of fluid i
Uj = viscosity of fluid i
g = gravity
z = elevation coordinate
Pi = pressure in fluid i
Mass Conservation for Each Phase
dt
t = time
pi = density of fluid i
r| = porosity
S, = saturation of fluid i
v«pigi = 0
-------
Land Surfac*
-------
LNAPLs
PRODUCT
SOURCE
INACTIVE
TOP or
CAPIUURY
nuMct
J2i£SSifiB
• . - - . ...< \ PRODUCT
/-• ':./:• :\^ AT RESIDUAL
-.-,,«. V:.- :->^ SATURATION
jr*«4'v^A sy>p.Tjjj^;'^x^c.'-v,/y;:T/jn»
g^^^^Ey5Jrf^5.ji»i«S^
CTOUNT)»tTtR
PRODUCT /
AT RESIDUAL —'
SATURATION
ru>*
ru*
LNAPLs
PRODUCT
SOURCE
INACTIVE
TOP or
CAPILURY
FRINGE
\.
-V.
PRODUCT
AT RESIDUAL
SATURATION
now
PRODUCT
AT RESIDUAL
SATURATION
n.ow
-------
6 toll
water and oil
Figure 3.9. The effect of water table fluctuations on pollutant
distributions (Schwille, 1967).
-------
Effect of Layering
Lower permeability layers Impede flow of wetting and nonwetting phases.
Nonwetting fluids do not enter a wetting-fluid saturated medium until the
threshold capillary pressure is exceeded.
The capillary fringe can act as a barrier to nonwetting fluids.
A wetting fluid may not flow into a nonwetting-fluid saturated medium until
the capillary pressures equalize (end effect).
SOURCE
4 4 4
Fig. 11. Tetrachloroelhylene-water displacement experiment, t = 313 seconds.
Kueper et al. (1989)
-------
Monitor Well Thicknesses
Thicknesses observed in wells are not representative of the thicknesses in
the formation.
Correction methods have been developed for hydrostatic distributions of
fluids.
Hampton and Miller (1988) used a physical model to observe formation
and well thicknesses.
2.16 cm hydrocarbon measured in formation:
Predictions from 6 "models" from 0 to 12 cm.
Non-hydrostatic conditions invalidate approximate corrections (Klemblowski
and Chiang, 1990)
-------
Summary
1) Capillary forces cause the trapping of NAPLs,
2) Trapped NAPLs serve as long term sources of contamination by
hydrophobic chemicals.
3) Flow of NAPLs is largely controlled by the pattern of heterogeneities.
4) NAPLs may be present at a site, but not directly detected in a well.
5) Aqueous concentrations below aqueous solubility do not preclude the
existence of NAPL.
6) The mass of NAPL at a site is normally unknown.
-------
Saturation and Concentration
Definitions
Saturation, S
= fraction of the pore space occupied by a fluid
0
-------
Example: Mass of Oil in the Pore Space
Property
Porosity/n
Bulk Density, pb
Oil (NAPL) saturation,
S0
Sample volume, SV
Oil Density, p0
Value
0.30 (cm3 voids) / (cm3 total)
1 .86 g/cm3 = 2.65 g/cm3 (1 - TI)
0.10 (cm3 oil) / (cm3 voids)
1 cm3
0.74 (g oil) / (cm3 oil)
Quantity
1. Void Volume in Sample
SV XT]
2. Oil Volume in Sample
V0 = SVxTixS0
3. Oil Mass in Sample
M0 = V0xPo
4. Porous Media Mass in Sample
Mp = SV x Pb
5. Oil Concentration in Sample
C0 = MyMp
Example Values
1 cm3 xO.30 = 0.30 cm3
0.30 cm3 x 0.1 0 = 0.03 cm3
0.03 cm3 x 0.74 g/cm3 = 0.022 g
1 cm3 x 1 .86 g/cm3 = 1 .86 g
(0.022 g oil) / (1.86 g porous media) =
0.01 2 g/g
12,000 mg oil / kg porous media
-------
Equilibrium Chemical Partitioning in a Multiphase
System
Vadose Zone: Water, Soil, NAPL, Air
Aquifer: Water, Soil, NAPL
Fluid Balance: (i.e., the fluids fill the pore space)
Sw = water saturation
Sa = air saturation
S0 = NAPL saturation
(Saturation = fraction of pore space occupied by a fluid)
Mass Balance: (i.e., the chemical partitions among the phases)
m = TI (Swcw + Saca + S0cJ +Pbq
m = bulk concentration (mass per total volume)
r| = porosity
GJ = concentration in phase j (mass per volume of phase j)
pb = bulk density
q = sorbed concentration (mass sorbed per mass of soil)
-------
Local Equilibrium
Soil
t \
Water
Air
/ t
\ \
•+* NAPL
The partition coefficients relate the soil, NAPL and air phase concentrations
to the water phase concentration:
ca - Kh cw
= K c
w
=Kdcw
Kh = Henry's Law coefficient
K0 = NAPL-water partition coefficient
Kd = soil-water distribution coefficient
-------
Henry's Law Coefficient
Ca - Kh Cw
1.) Kh is a dimensionless constant or
2.) Henry's Law Coefficients are often tabulated as
Kh' which has the units of atm-m /mol K
Kh = Kh'/R T
R = The universal gas constant (8.2 x 10'5 atm-m3/mol K)
T = absolute temperature (degrees Kelvin, K)
(@ 20° C Kh =41.62 Kh')
-------
Soil-Water Distribution Coefficient
q =Kdcw
q = sorbed concentration (mg / kg (soil))
cw = aqueous phase concentration ( mg / L (water))
Kd = soil-water distribution coefficient ( L (water) / kg (soil))
Kd is a function of
1) hydrophobic character of organic compound
2) amount of organic matter present
= Koc foc / pw
Koc = organic carbon partition coefficient
foc = fraction of organic carbon in soil matrix
pw = density of water
-------
NAPL - Water Partition Coefficient
= K0 cw
K0 = dimensionless NAPL-water partition coefficient
Raoult's Law for ideal solutions,
Effective Solubility = Aqueous Solubility x mole fraction
cw = s, X,
Estimate the K,,:
1000p,
,, C0 MW°
K0 = «
MW° is the average molecular weight of the NAPL phase
for gasoline (100 to 105)
Sj is aqueous phase solubility of the chemical (mol/L)
p0 is the NAPL phase density (g/mL)
-------
Raoults Law
To account for the detailed composition of the mixture:
ECoi
— -
K -
C0j = molecular weight of the jth constituent (g/mol)
coj = cone, of the jth constituent in the NAPL phase (g/L)
Sk = solubility of constituent k in water (g/L)
yk = activity coefficient (1.0 for ideal solutions)
-------
Equilibrium Phase Partitioning Example
For benzene composing 1.15% by mass of Gasoline,
what is the distriubtion of the mass of benzene in an aquifer among the
water, porous media and NAPL phases?
Properties
Property
Porosity,t|
Bulk Density, pb
Oil (NAPL) saturation,
S0
Sample volume, SV
Oil Density, p0
Value
0.30 (cm3 voids) / (cm3 total)
1.86 g/cm3 = 2.65 g/cm3 (1 - T\)
0.10 (cm3 oil) /(cm3 voids)
1 cm3
0.74 (g oil) / (cm3 oil)
-------
Benzene Concentration in the Gasoline, Cb,0):
'b(o)
mg benzene
L gasoline
*b
mass benzene mass gasoline
mass gasoline volume gasoline
1.15%
100
07/1 9
3
C/773
1000cm3
L
1000/770
9 .
Mass of Benzene in the Gasoline, Mb,0):
'b(o)
0 255 mg benzene
cm3 total
0.10
'b(o)
cm3 oil
cm3 voids
Q5<\Qm9 benzene
cm3 total
Z.o/7
1000 C/773
/A? 1 cm3 of sample there are 0.255 milligrams of benzene
-------
Benzene Concentration in Sample CWs):
In 1 cm3 of the sample there are 1.86 grams of porous media:
t QC g porous media
1.00 ° .
cm3 volume total
1 cm*
So the benzene concentration in the sample from the gasoline phase is:
'b(s-o)
mg benzene
kg porous media
0.255 /n<7 benzene
1.86 gr porous media
-------
Pseudo-Gasoline Mixture (Baehrand Corapciogliu, 1987)
Constituent
benzene
toluene
xylene
1-hexene
cyclohexane
n-hexane
other aromatics
other paraffins C4 - C8
heavy ends (> C8)
Coi
(gm/cc)
0.0082
0.0436
0.0718
0.0159
0.0021
0.0204
0.0740
0.3367
0.1451
percent by
mass
1.14
6.07
10.00
2.22
0.29
2.84
10.31
46.91
20.21
molecular
weight
(g/mol)
78
92
106
84
84
86
106
97
128
Density = 0.72 gr/cc
Approximated Partition Coefficients:
Constituent
benzene
toluene
o-xylene
Partition
Coefficient
K0
310 (350)a
1200 (1250)a
4240 (3630)a
avalues measured by Cline et al. (1991) on samples of gasoline obtained
in Florida
-------
Water Phase Concentration of Benzene:
If the Gasoline fills 10% of the pore space, then water fills 90% (Sw = 0.90)
'b(w)
mg benzene
L water
'b(o)
8510 m9benzene
Loil
312
Mass of benzene in the water phase M
b(w)
'b(w)
0.0073 mg benzene
CAT?3 total
'b(w)
27-
n qn cm3 water
\j.v\j
cm3 voids
mg benzene
L water
nvnCm3 voids
cm3 total
r L 1
1000 C/773
In 1 cm3 of sample there are 0.0073 milligrams of benzene.
-------
Benzene Concentration in Sample Cb(s.w):
So the benzene concentration in the sample from the water phase is:
'b(s-w)
o a mg benzene
kg porous media
0.0073 mg benzene
1.86 g porous media
1000g
-------
Sorbed Concentration of Benzene, Cb(s):
The Koc for benzene is 312 -- (literature/Raoult's Law Calculation)
Assuming that the fraction of organic carbon, foc is 0.001
'b(s)
m9 benzene
kg porous media
83
L water
kg porous media
[0.001]
27-
mg
L water
Sorbed mass of benzene in a 1 cm3 sample:
Cb(S<)
0.0041 mg benzene
cm3 total
^b(s-s) Pb
2 g mg benzene
kg porous media
-86 g
cm
1000 g.
In 1 cm3 of sample there are 0.0041 milligrams of benzene.
-------
Summary
Phase
Gasoline
Water
Porous Media
Total
Concentration
of Benzene
mg/kg
137
3.9
2.2
143.1
Mass of
Benzene in 1
cm3 of sample
(mg)
0.225
0.0073
0.0041
0.2364
Fraction
0.952
0.031
0.017
1.000
-------
Comments
1. Observed aqueous phase concentrations which result from mixtures are
less than aqueous phase solubilities.
2. Low aqueous phase concentrations do not preclude the existence of
NAPLs.
3. Hydraulic mixing can cause concentrations to be lower still.
(i.e., mixing of clean formation water with contaminated water).
-------
Non Equilibrium Partitioning
Laboratory and field experiments have shown that the local equilibrium
assumption (LEA) is invalid for certain cases.
Local equilibrium assumes that the
rate of sorption is fast relative to:
advection and
dispersion
-------
Non Linear Partitioning
Solute concentrations are high, activity coefficient not constant
Mechanism other than hydrophobic sorption
M.L Brusseau, P.C.S. Rao, 1989, Sorption Nonideality during organic
contaminant transport in porous media, Critical Reviews in
Environmental Control, 19(1), 33-99
-------
Non Equilibrium Mass Transfer
During disolution the water phase concentration of NAPL may be mass
transfer limited
Linear mass transfer models
Single component NAPL, 1D column experiments
where p0 = NAPL density [M/L3]
r| = porosity [*]
S0 = NAPL saturation [*]
K1 = mass transfer rate coefficient [T1]
=k13na , ,
K! = mass transfer coefficient [L T ]
ana = interfacial area between phases [L2]
Cs = solubility limit of the NAPL [M/L3]
C = aqueous concentration [M/T]
-------
Estimation of the mass transfer rate coefficient, K,
Empirical fitting of data
Relations between the
Sherwood Number
Reynolds Number
NAPL saturation
Others
-------
Sherwood Number, S
sh
d50 = mean grain diameter [L]
Dm = molecular diffusion coefficient
m
Reynolds Number, Re
Re
U = Darcy flux [L/T]
p = density of water [M/L3]
u = dynamic viscosity of water [ML'1T1]
T| = porosity [*]
S0 = NAPL saturation
Imhoff et al., 1994, An experimental study of complete dissolution of a
nonaqueous phase liquid in saturated porous media, Water Resources
Research, 30(2), 307-320
Powers et al., 1994, An experimental investigation of nonaqueous phase
liquid dissolution in saturated subsurface systems: Transient mass
transfer rates, Water Resources Research, 30(2), 321-332.
-------
312
IMHOFF ET AL.: COMPLETE DISSOLUTION OF A NONAQUEOUS PHASE LIQUID
K»tlMI
1« 111 CM Ui Ml «.U Ml IM
IM 111 111
8
Figure 3. Changing TCE saturation profile during experiment 6. Clean water enters the column at x =
0 mm (upgradient edge of the topmost 1.0 mm scanned region). After 25 pore volumes (PV = 25) a
dissolution front of length x* = 11 mm has formed. (See text for definition of dissolution front.) This front
lengthens to x* = 21 mm at PV = 120.
-------
Comments
Non equilibrium in 1D columns
vs
Non equilbrium in unconstrained systems .
Water does not have to flow through the NAPL zone
R. L. Johnson and J. F. Pankow, 1992, Dissolution of dense chlorinated
solvents into groundwater. 2. Source functions for pools of solvent,
Environmental Science and Technology, 26(5), 896-901
-------
HSSM
-------
12. The Hydrocarbon Spill Screening
Model (HSSM)
-------
The Hydrocarbon Spill Screening Model (HSSM)
Jim Weaver
Robert S. Kerr Environmental Research Laboratory
United States Enivonmental Protection Agency
Ada, Oklahoma 74820
405-436-8545
Randall Charbeneau
Center for Research in Water Resources
The University of Texas at Austin
Austin, Texas 78712
Objectives:
1. Introduce multiphase (NAPL) modeling.
2. Highlight important parameters.
3. Describe readily available models.
4. Run the HSSM.
-------
Modeling of NAPLs
NAPL-Nonaqueous Ehase Liquid
NAPLs are immiscible with water
lighter than water-gasoline, diesel, etc
denser than water-TCE, PCE,CCI4,Creosote
Models must incorporate:
- flow of the NAPL as a liquid distinct from water
- dissolution of NAPL constituents into the groundwater:
Constituent(s)
Benzene,
Toluene,
Ethylbenzene,
Xylenes
PCE
PAHs
NAPL
Gasoline
PCE
Creosote
-------
Land Surf ace
Figure 1 Schematic view of NAPL release
-------
The Hydrocarbon Spill Screening Model (HSSM)
1. Simple to use (the numerical models are difficult to use)
2. Low computational time (we can run realistic simulations during the class)
3. Illustrates character of hydrocarbon spill episodes
Hydrocarbon (Oil, Fuel, L-NAPL) Release Characteristics
Downward Transport Through the Vadose Zone
Hydrocarbon Spreading in the Capillary Fringe .
Contamination of the Aquifer Through Dissolution of the Hydrocarbon
From the point of view of aquifer contamination:
What amount of mass enters the aquifer ?
What is the size of the source at the water table ?
-------
Land Surf ace
VadoseZone
4 NAPL
\7
Aquifer
Figure 2 Schematic view of idealized NAPL release that is used in HSSM
Land Surface
KOPT
VadoseZone
.NAPL
.OlLENS
Aqueous
Contamination
7
TSGPLUME
Aquifer
Figure 3 HSSM schematic showing the use of each module
-------
Components of the Model
The model consists of separate modules for the vadose zone, capillary
fringe, and the water table aquifer.
Module
KOPT
Kinematic Oily Pollutant Transport
OILENS
TSGPLUME
Transient Source Gaussian Plume
Model
Region
Vadose Zone
Capillary Fringe
Water Table Aquifer
-------
KOPT and OILENS Source Condition at Release Point
TSGPLUME Source Condition at Water Table
Figure 9 Coordinate systems for the KOPT, OILENS and TSGPLUME Modules of HSSM
(D
O
c
o
O
At
ABC
Time
D
Figure 10 Schematic representation of a TSGPLUME concentration history
-------
Purpose
The Hydrocarbon Spill Screening Model (HSSM) is intended for use as a
screening model to estimate the impacts of petroleum hydrocarbons releases
on ground water. By using a suite of simplifying assumptions, the model is
designed to execute rapidly on small computers.
The HSSM serves as a screening model to
1 evaluate the effects of various input parameters on
hydrocarbon transport, especially soil properties and NAPL
compositions.
2. simulate hydrocarbon spills in idealized geologic
environments.
The model is geared toward applications where data is limited because of
time, economic, safety, regulatory or administrative reasons:
1 Underground storage tank program
2. Emergency response
3. Initial phases of site investigation
4. Development of regulations
-------
C once ptua lization
The petroleum hydrocarbon is assumed to consist of a two-component
mixture; the first component is an inert "oil" phase which is transported
under gravity and pressure gradients. The second component is a chemical
constituent of the oil phase which is the compound of environmental
concern. For example, with gasoline as the oil phase, benzene may be the
chemical constituent of concern.
The oil phase is assumed to be released in sufficient quantity to persist as
a separate phase in the subsurface. The chemical constituent is released
slowly into the vadose zone water and the aquifer. Concentrations of the
constituent are estimated for down gradient receptor points under natural
gradient conditions.
In order to develop a model which executes rapidly on PCs, the domain is
assumed to be homogeneous and other assumptions are made concerning
the flow and transport phenomena. As a result each part of the model is in
the form of an analytic or semi-analytic solution.
-------
HSSM Module Features and Limitations
General
The number of phenomena which can be included in a simplified model is
limited. HSSM is not suited for simulation of complex hydrological or
geologic regimes.
Two component NAPL
1. inert "oil" phase,
2. partitionable constituent
Transient flow of NAPL and the constituent
Raoult's Law Partitioning of the constituent between the water, NAPL
and soil
Linear equilibrium partitioning
Uniform porous media
-------
HSSM Module Features and Limitations (Continued)
KOPT (Kinematic Oily Pollutant Transport)
1D Vadose Zone Transport
Uniform Water Saturation
Advective Transport of the Chemical Constituent
Capillary Gradient Neglected During Redistribution
Green-Ampt Model for NAPL Infiltration
OILENS
Radial Flow in the Capillary Fringe
Static Water Table
Mass Transfer Limited Dissolution into the Aquifer
Lens Shape determined by Dupuit Assumptions of Horizontal Flow and
Ghyben-Herzberg Density Relations.
Trapping of Residual Hydrocarbons in the Vadose Zone and Aquifer
TSGPLUME
Analytic Solution
Convolution for Variable Mass Flux Input
2D Planar Aquifer Model
Penetration Thickness for z-direction
Natural Gradient Flow
-------
User Interfaces
The model runs under both a simple MS-DOS interface and also a MS-
Windows 3.x interface. Both interfaces allow input data sets to be entered
and edited, and allow the automatic plotting of the simulation results.
Function
Data Input
KOPT/OILE
TSGPLUME
Displaying Results
MS-DOS Interface
PREHSSM
HSSM-KO
HSSM-T
HSSM-PLT
MS-WINDOWS
Interface
HSSM-WIN
HSSM-KO
HSSM-T
HSSM-WIN
NS
-------
Table 2 Comparison of MS-DOS and MS-Windows Interfaces
Interface
Advantages
Disadvantages
DOS
1. The fastest performance of model
calculations is achieved (for any given
computer) under the DOS interface.
2. DOS interface can run on a
machine with limited processing power
and limited RAM. The code will
execute, albeit slowly, on a 286
machine with 640 kiloBytes of RAM.
1. The DOS preprocessor is
interactive but not graphical.
Windows
1. A single shell program performs all
necessary functions of the model.
2. Data are entered directly on
graphical screens.
3. Simultaneous display of all model
output.
4. Simultaneous display of output from
simulations with different parameter
values.
5. Ability to cut and paste to other
Windows applications.
1. The calculations performed by
HSSM-KO and HSSM-T are slower
under the Windows interface due to
Windows overhead.
2. Requires a machine with enough
processing power and memory to run
Windows effectively. Typically this
would be a 386 or higher with at least
4 megaBytes of RAM.
3. Requires a certain level of
expertise with Windows.
4. More system memory is consumed
by Windows than by DOS.
Table 3 HSSM Data Calculation Utilities
Parameter(s)
Soil Hydraulic Properties
Equilibrium NAPL/water partition coefficients
Average NAPL saturation for OILENS
Utility Program Name
SOPROP
RAOULT
NTHICK
-------
Table 21 HSSM Graphics
Title
Saturation Profiles
NAPL Lens Profiles
NAPL Lens Radius
History
Contaminant Mass
Flux History
NAPL Lens
Contaminant Mass
Balance
Receptor
Concentration
Histories
HSSM Module
KOPT
OILENS
OILENS
OILENS
OILENS
TSGPLUME
Description
Vadose Zone Liquid Saturations from the
Surface to the Water Table
Cross-section of the NAPL lens on the water
table
History of the radius of the NAPL lens and the
effective radius of the contaminant
History of the mass flux from the NAPL lens to
the aquifer
History of the mass in the NAPL lens
History of the contaminant concentrations at the
receptor points
-------
Saturation Profiles
Benzene transport from 1500 gal gasoline spill
0.00 Depth (m)
3.00
6.00
25.000 d
50.000 d
75.000 d
100.00 d
125.00 d
150.00 d
200.00 d
0.20 0.40 0.60 0.80
Total liquid saturation
Figure 23 Typical saturation profiles
1.00
••• **• t-ero numex
^*™ Benzene bancport bom 1500 gal gasoline tpffl
3.00 Depth (m)
125.00 days
9.50
10.0
10.5
11.0L.
0.00
5.00 10.0
Radius (ra)
Figure 24 Typical NAPL tens profile
15.0
Contaminant Mass Flux
Benzene transport from 1500 gal gasoline spill
0.075 Moss ""* (*g/<0
0.060
0.045
0.030
0.015
0.000
0.00 1.00 2.00 3.00 4.00
Time (yr)
Radius Histories
Benzene transport from 1500 gal gasoline spill
20.00 Radius (m)
15.00
10.00
5.000
0.000
NAPL
Contaminant
0.00 1.00 2.00 3.00 4.00
Time(yt)
Figure 25 Typical contaminant mass flux history Figure 26 Typical tyAPL tens radius history
Contaminant Mass in Lens
Benzene transport irom 1500 gal gasoline spill
25.00 ^Mass (kg)
Mass in lens
Mass dissolved
20.00
15.00
10.00
5.000
0.000
Receptor Well Concentrations
Benzene transport from 1500 gal gasoline spill
15 00 Concentration (mg/L)
X(m) Y(m)
25.000 0.0000
50.000 0.0000
75.000 0.0000
100.00 0.0000
10.00
s.ooo
0.000
0.00
125.00 0.0000
150.00 0.0000
1.50 3.00 4.50
Time (yr)
6.00
0.00 1.00 2.00 3.00 4.00
Time (yr)
Figure 27 Typical NAPL tens contaminant mass Figure 28 Typical receptor concentration histories
balance
-------
HSSM Major Model Parameters
Saturated hydraulic conductivity
Brooks and Corey Capillary pressure curve parameters
Release details
geometry
flux, volume/area, or ponding depth
NAPL/Water partitioning coefficient (from Raoult's Law)
NAPL fluid properties: density, viscosity, surface tension
Sorption coefficient
Aquifer thickness
Aquifer properties-dispersivities, seepage velocity
Receptor Location
HSSM simulation control parameters
-------
Appendix 13 HSSM-WIN Input Data Templates
The following figures are to be used as input data templates for the MS-Windows interface (HSSM-
WIN). Each input dialog box in HSSM-WIN is represented by a template. These pages are intended as
aids in preparing input data sets.
General Model Parameters
Run "Titles:
| Cancel |
"Erinting switches
O Create output files
O Echo print data only
O Run models
'Module switches
ORunKOPT
D Run OILENS
D Write HSSM-T input file
"File names*
NOTE: These filenames will be used if the data file
is saved under a new name with the "SoveAs" option.
HSSM-KO input file
HSSM-KO output file
HSSM-KO plot file 1
HSSM-KO plot file 2
HSSM-KO plot file 3
HSSM-T input file
HSSM-T output file
HSSM-T plot file
[Appendix 13 HSSM-WIN Data Templates]
-------
Hydroloyic Parameters
HYDROLOGIC PROPERTIES
Data file:
Water dynamic viscosity (cp)
Water density (g/cm*)
Water surf, tension (dyne/cm)
Maximum krw during infiltration
•Recharge
O Average recharge rate (m/d)
O Saturation
'Capillary pressure curve model
O Brooks and Corey
Qvan Genuchten
Brooks and Corey's lambda .
Air entry head (m)
Residual water saturation ...
van Genuchten's alpha (1/rn)
van Genuchten's n
E Enable range checking
j Cancel |
POROUS MEDIUM PROPERTIES
Sat*d vert, hydraulic cond. (m/d)
Ratio of horz/vert hyd. cond
Porosity
Bulk density (g/cm*)
Aquifer saturated thickness (m)
Depth to water table (m)
Capillary thickness parameter (m) ..
Groundwater gradient (m/m)
Longitudinal dispersrvity (m)
Transverse dispersrvity (m)
Vertical dispersrvity (m)
[Appendix 13 HSSM-WIN Data Templates]
-------
Hydrocarbon Phase Parameters
HYDROCARBON PHASE PROPERTIES
Data file:
NAPL density (g/cm*)
NAPL dynamic viscosity (cp)
Hydrocarbon solubility (mg/L) .
Aquifer residual NAPL saturation...
Vadose zone residual NAPL set'n..
Soil/water partition coeff. (L/kg)
NAPL surface tension (dyne/cm)
DISSOLVED CONSTITUENT PROPERTIES
O DJssolved constituent exists
Initial constit cone, in NAPL (mg/L)..
NAPL/water partition coefficient.
Soil/water partition coeff. (L/kg)....
Constituent solubility (mg/L)
O Constit VHife in aquifer (d)
Cancel
El Enable range checking
[-HYDROCARBON RELEASE
QiSpecrfied flux!
O Specified volume/area
O Constant head ponding
O Variable ponding after const head period
NAPL flux (m/d) ..
Beginning time (d)
Ending time (d)
Ponding depth (m)
NAPL volume/area (m)
Lower depth of NAPL zone (m)
[Appendix 13 HSSM-WIN Data Templates]
-------
Simulation Parameters
SIMULATION CONTROL PARAMETERS
Radius of NAPL lens source (m)...
Radius multiplication factor
Max NAPL saturation in NAPL lens .
Simulation ending time (d)
Maximum solution time step (d)
Minimum time between printed time
steps (d)
"OILENS Simulation ending criterion
O User-specified time
O NAPL lens spreading stops
O Max contaminant mass flux into aquifer
O Contaminant leached from lens
Fraction of moss remaining ....
HSSM-T MODEL PARAMETERS
Percent max. contam't radius (5{)
Minimum output conc'n (mg/L)
Beginning time (d)
Ending time (d)
Time increment (d)
Data file:
Kl Enable range checking
NAPL LENS PROFILES
Enter time (d) for
each of up to
10 profiles
Number of
profiles
| Cancel |
1
2
3
4
5
6
7
e
9
10
RECEPTOR WELL
LOCATIONS
Enter coordinates
for each of up to
Swells
X(m)
Y(m)
Number of wells
1
2
3
4
5
6
[Appendix 13 HSSM-WIN Data Templates]
-------
200
150
hce(cm) 100
50
\°
OA \°an
A AA\ ° a,
— Sand (Brakensiek et al.,1981)
—- Sand(Carseletal.,1988)
A C109
o 20/30
a TCS
• Lincoln
Oil Creek
Texas
Dx a
A°
o o
A
2 ° A° OA° DA
a o o o o o <
, A * * "A;-*—i—x ,-^-^-A-
0.2
0.4
0.6
0.8
Saturation
Figure 43 Comparison of average capillary pressure curves with measured data
-------
Brooks and Corey (1964)
Sw -
Sw = water saturation = fraction of the pore space filled by water
hc = head capillary
Swr = residual water saturation
hce = non-wetting phase entry head
A, = pore size distribution index
-------
van Genuchten (1980)
0W = volumetric water content (saturation x porosity = content)
hc = capillary head
0wr = residual water content
a = a parameter
n = a parameter
m = a parameter (usually m = 1 - 1/n)
if 6m = porosity
-------
Table VI Descriptive Statistics from Carsel
and Parrish (1988) Data Set
Soil type
Clay*
Clay Loam
Loam
Loamy
Sand
Silt
Silt Loam
Silty Clay
Silty Clay
Loam
Sand
Sandy Clay
Sandy Clay
Loam
Sandy
Loam
Hydraulic Conductivity Ks, (m/d)
sample size
114
345
735
315
88
1093
126
592
246
46
214
1183
mean
0.048
0.062
0.25
3.5
0.060
0.11
0.0048
0.017
7.1
0.029
0.31
1.1
standard
deviation
0.10
0.17
0.44
2.7
0.079
0.30
0.026
0.046
3.7
0.067
0.66
1.4
* The clay type represents an agricultural soil with clay content of 60% or less.
'3-47
-------
Cline, P.V., J.J. Delfino, and P.S.C. Rao, 1991, Partitioning of aromatic constituents
into water from gasoline and other complex solvent mixtures, Environemental
Science Technology, 25(5), 914-920.
Demond, A.M., and P.V. Roberts, 1987, An examination of relative permeability
relations for two-phase flow in porous media, Water Resources Bulliten, 23(4),
617-626.
Mercer J.W. and R. M. Cohen, 1990, A review of Immiscible Fluids in the Subsurface:
Properties, Models, characterization and remediation, Journal of Contaminant
Hydrology, 6, 107-163.
Schwille, F., 1988, Dense Chlorinated Solvents in Porous and Fractured Media--
Models Experiments, Lewis Publishers, 146 pp.
United States Environmental Protection Agency, 1992, Dense Nonaqueous Phase
Liquids-A Workshop Summary, EPA, EPA/600/R-92/030.
United States Environmental Protection Agency, 1992, Estimating the Potential for
Occurrence of DNAPL at Superfund Sites, EPA, OSWER, 9355.4-07FS.
United States Environmental Protection Agency, 1992, DNAPL Site Evaluation, EPA-
600/R-93/002. • *
Wilson, J.L., S.H. Conrad, W.R. Mason, W. Peplinski, and E. Hagan, 1990, Laboratory
Investigation of Residual Liquid Organics from Spills,"Leaks, and the Disposal of
Hazardous Wastes in Groundwater, EPA, Robert S. Kerr Environmental
Research Laboratory, EPA/600/6-90/004.
-------
Application of the Hydrocarbon Spill Screening Model to Field Sites
James W. Weaver -
Proceedings of the American Society of Civil Engineers
Conference on Non-aqueous Phase Liquids in the Subsurface Environment:
Assessment and Remediation
(To appear)
November 12-14, 1996
Washington, D.C.
Weaver
-------
Application of the Hydrocarbon Spill Screening Model
to Field Sites
James W. Weaver1
Abstract
The Hydrocarbon Spill Screening Model, HSSM, was developed for esti-
mating the impacts of petroleum hydrocarbon contamination on subsurface
water resources. The model simulates the release of the hydrocarbon at the
ground surface, formation of lens in the capillary fringe, dissolution of con-
stituents of the gasoline, and transport to a receptor in the aquifer. Field data
from two case histories were used to develop input parameter sets for HSSM.
In one case there were aqueous concentration data from an extensive monitor-
ing network. In the second case the monitoring network was small, but the
date and volume of the release could be estimated. Both of these cases have
features that are well suited for testing of the model. In both cases the model
was able to reproduce the trends in the data set and the concentrations to
within an order of magnitude.
Introduction
The Hydrocarbon Spill Screening Model (HSSM) was intended as a simpli-
fied model for estimating the impacts of petroleum hydrocarbons on subsurface
water resources (Weaver et al., 1994 and Charbeneau et al., 1995). In this pa-
per the model was applied to two sites with releases from leaking underground
storage tanks. The data were drawn from State Agency case files and were
not intended for research purposes. The objectives of the work were to de-
termine if HSSM could reproduce the observed contaminant distributions and
to demonstrate the effect of data gaps on model results. These applications
demonstrate the effects of parameter uncertainty on model results, because in
1 National Risk Management Research Laboratory, United States Environ-
mental Protection Agency, Ada, Oklahoma 74820
1 Weaver
-------
each case some information for running the model was not available.
Field Data Sets
Contamination from underground storage tank releases is usually charac-
terized by contaminant concentrations in the ground water and soils. Nor-
mally, data are collected for benzene, toluene, ethylbenzene and the xylenes
(BTEX), and total petroleum hydrocarbons (TPH). Water samples give con-
centrations that in effect are related through time and space by the transport
equation:
dc
r,R— = v^Vc-9'Vc- ^Rc + J(t) (1)
where 77 is the porosity, R is the retardation factor, c is the contaminant
concentration in the ground water, q is darcy velocity, D is the dispersion
constant, A is a first order decay constant, J(i) is the amount of mass per
unit volume of aquifer added per unit time. Each term on the right hand
side of equation 1 can cause the concentration to change. Apparent dilution
along the length of contaminant plumes is characterized by the dispersion
constant (term 1: y • D V c) > which is assumed to depend upon the seepage
velocity and inherent dispersivity of the aquifer. Water flowing along divergent
streamlines can cause reduction in concentration (term 2: q- Vc)- In equation 1
biodegradation is assumed to follow first order decay (term 3: XrjRc). The last
term, J(t), is the source/sink term which can include losses due to hydrolysis,
extraction or volatilization, for example. It may also include gains due to
loading from the contaminant source.
For leaks from underground storage tanks the source of contamination is
the released hydrocarbon liquid. Normally the volume and timing of the re-
lease is unknown. The magnitude of the source term in equation 1, however,
depends directly upon the volume, rate and timing of the release. Typical
data from a site that reflect the hydrocarbon liquid are free product levels
observed in wells, and concentrations from soil samples. Free product levels
can not provide a reliable estimate of the release volume, because there is not
a clear relationship between the free product levels in wells and the amount of
free product in the formation (Kemblowski and Chiang, 1990). Free product
recovery data can obviously give a lower bound on the release volume. Core
extracts could provide more reliable estimates of the product volume, but they
are only taken upon installation of wells and may not be analyzed over their
full length. Hydrogeologic data normally consists of information on regional
hydrology and geology from published sources, well logs, and measured or
estimated hydraulic conductivities. Well logs define the vertical and spatial
stratigraphy.
Weaver
-------
The Hydrocarbon Spill Screening Model
The Hydrocarbon Spill Screening Model (HSSM) was intended as a screen-
ing model for estimating the impacts of hydrocarbon (or light nonaqueous
phase liquid, LNAPL) releases to the subsurface (Weaver et al., 1994, and
Charbeneau et al., 1995). The model consists of three modules that treat
transport in the vadose zone, the formation and decay of an oil lens in the
capillary fringe, and transport of soluble constituents of the LNAPL in the
aquifer to receptor locations. The model uses semi-analytical solutions of the
transport equations which include many of the important physical and chem-
ical processes. It does not include all processes which may be important, and
because of the usage of semi-analytical solutions, it does not account directly
for heterogeneity.
There are 36 physical and chemical parameters required to run HSSM.
Each of these has an impact on the model results. Depending on the specific
model scenario, some of the parameters are far more important than others.
Of the entire suite of parameter values, only a small number of the most im-
portant parameters were varied to achieve the fits to the data described below
(Table 2).
Features of the data sets
Both of the data sets simulated below were accepted by the State Agency
that was responsible for managing the site and thus met the regulatory require-
ments for assessment of contamination. In neither of these cases was modeling
of the spill considered an essential or integral assessment activity. Table 1 lists
some features of each spill. The data used in this paper came from excerpts
of state agency case files (Weaver et al., 1996, and State of Utah, 1996). Each
data set has unique features that led to its inclusion in this study. Hager-
man Avenue has a large number of monitoring wells and Mountain Fuels has
estimates of the date and volume of release. For each case model results for
benzene are discussed below. These results are intended to illustrate certain
features of the cases and the ability of the model to duplicate them.
Mountain Fuels, Salt Lake City, Utah
At the Mountain Fuels site in Salt Lake City, Utah, approximately 7500
gallons of gasoline leaked from a tank that was punctured in 1979 (State of
Utah, 1996). Four monitoring wells were sampled from June 1991 to December
1994 to characterize the resulting aquifer contamination (Figure 1). Thus
these data date from 12 to 15 years after the release. Concentrations in two
of the four monitoring wells were always below the detection limits. One of
3 Weaver
-------
Item
Release Date
Release Volume
Composition of Fuel
Mass in the Ground Water
Cores Analyzed
Monitor Wells
Vertical Plume Definition
Sample Rounds
Data Points per Sample Round
Slug Tests
Pump Tests
Hagerman Ave.
E. Patchogue,
New York
Unknown
Unknown
Unknown
Est.
30
48<°>
Yes
3
210
13
1
Mountain Fuels
Salt Lake City,
Utah
Known
est. 7500 gal
Unknown
Unknown
-
4
No
8<»)
4
Unknown (c)
0
(°) 26 multilevel samplers and 22 screened wells
(*) 6 samples were taken from MW-2
(c) number not given in case file excerpt
Table 1: General features of the two data sets
the remaining wells is clearly directly in the path of the contaminant plume
(MW-2). The fourth well (MW-1) appears to be on the edge of the plume,
both because of its geographic location and its concentration history.
Parameter values given in Table 2 were used in the simulation. The hy-
draulic conductivity and gradient were estimated from the field data. The
other parameters listed in the table were estimates that resulted in order of
magnitude matches to the observed data. At the furthest down gradient well,
MW-2, the concentrations decline steadily. This implies that for all the sim-
ulations and the field data that the peak concentration has already passed
this receptor. Table 3 lists the simulated peak concentrations, maximum mass
fluxes to the aquifer, and times of their occurrence. With increasing conduc-
tivity, the benzene is released sooner at a higher maximum rate. The peak
concentrations in the aquifer occur sooner, but with higher velocities the peak
concentrations may decline due to the effect of increasing dispersion (which is
proportional to the velocity).
The simulated concentration distributions at MW-2 shown in Figure 2
decline uniformly. The concentrations are seen to be dependent upon the
hydraulic conductivity because of their dependence upon the release rate of
benzene from the gasoline and advective-dispersive transport in the aquifer.
The value of concentration can also be adjusted by changing the degradation
rate constant as noted in the figure. Both the simulation with a Ka = 1.5 m/d
and half life of 30 days and that with K3 = 0.75 m/d and half life of 69.3 days
(loss rate of 0.01% per day) give a similar match to the field data. The breaks
4 Weaver
-------
Figure 1: Mountain Fuels, Salt Lake City, Utah
in the concentration curves for Ks = 1.5 m/d and Ks = 3.0 m/d occurring at
about January 1, 1990 and January 1, 1996, respectively, are caused by the
ending criterion used in HSSM (Weaver et al., 1994).
Observation well MW-1 was located within the oil lens generated by HSSM.
Therefore no contaminant concentrations were calculated for this location by
the aquifer module. The concentrations in the ground water below the lens,
however, give an indication of concentrations that would be observed in MW-
1. Figure 3 shows a comparison of the model and the data for this well. Three
values of hydraulic conductivity were used in the simulations: the reported
average of 1.5 m/d, and half and twice this value. As the hydraulic conductiv-
ity increased, the rate of release of benzene to the aquifer increased. Thus the
benzene concentrations for the higher conductivity simulations decrease more
rapidly than for the low conductivity cases. The case with the lowest estimate
of conductivity (0.75 m/d) falls through the scatter of the field data, but the
variation in concentration observed in the monitor well cannot be matched by
the model. Both MW-1 and MW-2 are best fit by the simulation with con-
ductivity of 0.75 m/d and half life of 69.3 days.
Hagerman Avenue, East Patchogue, New York
The gasoline spill at East Patchogue, New York is described in detail in
Weaver
-------
25O
= O.75 m/d. h.l. = 69.3 m/d
- 1.5 m/d. h.l. - 69.3 d
•Kg - 3.0 m/d. h.l. — 69.3 days
• K., = 1.5 m/d. h.l. = 3O days
Field Data MW-2
O
03 co co co co co
Date
Figure 1: Model results and Field data for MW-2 Mountain Fuels, Salt Lake
City, Utah
2000
= O.75 m/d
- 1.5 m/d
= 3.O m/d
Date
Figure 3: Model results and Field data for MW-1 Mountain Fuels, Salt Lake
City, Utah
Weaver
-------
Item Hagerman Ave. Mountain Fuels
E. Patchogue, Salt Lake City
New York Utah
Hydraulic Conductivity 43, 94, 149 m/d 0.75, 1.5, 3.0 m/d
Hydraulic gradient 0.0013 0.02
Water Table Fluctuation 0.1 m 0.1 m
Initial Benzene Concentration 8200 mg/L 4100 mg/L
Half Life 2250 days 69.3, 30 days
Longitudinal Dispersivity 10, 15, 20 m 15 m
Table 2: Main Adjusted Parameters for the Cases
Conductivity Time Maximum Time Maximum
Concentration Mass flux
m/d d /^g/L d kg/d
968
839
390
0.1281
4.7276
2.9253
731
502
306
0.0066
0.0262
0.0742
0.75
1.5
3.0
Table 3: Mountain Fuels Simulated Peak Concentrations, MW-2
another paper in this proceedings (Weaver et al, 1996). In that paper, the
extensive data set was used to estimate the ground water flow velocity, the
volume of gasoline released and the mass of BTEX and methyl £eri-butyl ether,
MTBE, released to the aquifer.
The data from the site suggest a release volume of at least 13,200 gallons,
which contains 420 kg of benzene (Weaver et al., 1996). The release likely
occurred as a series of continuing leaks over several years. The tanks at the
service station were removed in 1988, so any releases ended that year. Since
some fraction of the gasoline contained MTBE, that gasoline was released after
1979 when MTBE was approved for use as an octane enhancer. The MTBE
in the aquifer traveled from the source zone to its center of mass in 1994 and
1995 in 16 years or less. Using the centers-of-mass calculated by Weaver et
al. (1996), the MTBE plume traveled at the rates listed in Table 4. The
rates show remarkable consistency suggesting that the average transport time,
averaged over the duration of the contamination event, is nearly constant for
distances between 1387 m and 1583 m from the suspect source. The rate would
have been 0.65 m/d if the entire release occurred on December 31, 1988 and
7 Weaver
-------
the 0.25 m/d if the release began on January 1, 1979.
The ground water flow velocity influences advective transport in the aquifer
and the rate of release of mass from the oil lens. These two processes must
both be consistent with the data from the field for the simulation to be appro-
priate. Parameter values for the simulation are listed in Table 2. In separate
simulations the hydraulic conductivity was taken as the average and the aver-
age plus or minus one standard deviation of values determined from slug tests
performed at the site. The water table fluctuation was determined from wells
near in the source zone. The other parameters listed in Table 2 were taken
as reasonable estimates that resulted in order of magnitude matches to the
observed data.
Sample Date Distance Estimated Velocities (m/d)
Round Release Date Scenarios
Late Early
days since rate days since rate
m Dec 31, 1988 m/d Jan 1, 1979 m/d
1
2
3
Dec 16, 1994
April 16, 1995
Oct 17, 1995
1387
1557
1583
2176.25
2297.5
2481.5
0.64
0.68
0.64
5828.75
5950.0
6134.0
0.24
0.26
0.26
Table 4: Rates of Movement of the MTBE Plume
The critical parameters for simulating the Hagerman Avenue site were the
rate and duration of the release, ground water flow velocity, degradation rate
constant, and water table fluctuation. The ground water velocity was selected
to be 0.40 m/d which is within the range given in Table 4. The release was
assumed to occur continuously from January 1, 1979 to December 31, 1988
at a rate that resulted in 50 m3 (13200 gallons) of gasoline in the aquifer.
This release scenario was used because the true rates and timings of releases
are unknown and because it was found necessary to generate relatively low
concentrations in the ground water to match the monitor well data. From
sample round one and two water level data near the source, the water table
fluctuation was approximately 0.33 ft (0.10 m). Figure 4 shows a comparison
of the vertically averaged measured concentration in MW-13, MW-12 and
MW-1,4 with the HSSM model results (see Weaver et al., 1996, Figure 1).
The highest concentrations were simulated at the up gradient monitoring
well (MW-13). The observed concentrations at this well (open triangles on
Figure 4) decline over the sampling period indicating that the peak concentra-
tion has passed this well. MW-12 shows the highest observed concentrations
(open squares) that decline with time, in contrast to the model result that
8 Weaver
-------
indicates relatively constant concentration over this period. Being down gra-
dient of MW-13 the concentrations would be expected to decrease if advective-
dispersive transport is occurring in a uniform aquifer. Figure 3 of Weaver et
al. (1996) shows that the benzene distribution in the aquifer is unsmooth, sug-
gesting that there is preferential flow through certain regions. The observed
concentration in MW-1,4 increased from sample rounds one to two (open cir-
cles). This behavior is not matched by the model which indicates that the
peak concentration at this receptor has yet to arrive. This result is consistent
with the field data (Weaver et al., 1996, Figure 3) which show that the ben-
zene plume is beginning to pass MW-1,4 during the sampling interval. Despite
variability, each of the model results was within an order of magnitude of the
field data.
Figure 5 shows the effect of hydraulic conductivity variation on model
results at MW-13. As expected, the benzene arrives sooner and at higher
concentration as the ground water flow velocity increases. In sample round
one the field data fall near the average curve (Ks = 94 m/d) while later data
fall near the low conductivity result (Ks = 43 m/d). These results suggest
that with a steeper front one curve may fit data from all three sample rounds.
Figure 6 shows the effect of varying dispersivity on the results for MW-13. De-
creasing dispersivity over the range shown here sharpens the front somewhat,
but does not force an exact match to the data. The parameter values could
be further adjusted to try to match the field data at this well, but other wells
would likely remained unmatched as shown in Figure 4.
Conclusions
Uncertainty exists in model parameters for both data sets. In a general
sense HSSM was able to reproduce the trends in the monitoring wells. These
trends are largely related to the hydraulics of the system, which apparently
are relatively well matched by the model. At the Mountain Fuels site, the
concentrations in MW-2 decline continually, which matches the model result
that indicates that the peak concentration already passed the well. At Hager-
man Avenue the concentration peak had not yet reached MW-1,4 as noted on
Figure 3 of Weaver et al. (1996) which was reflected in the simulation results.
For both cases the data were collected more than ten years after the release
may have occurred so that the measured concentrations are below a few hun-
dred micrograms per liter. The higher concentration data which could provide
a better test of the model are not available. As noted at Hagerman Avenue,
heterogeneity plays an important role in determining contaminant concentra-
tions. Since HSSM cannot include heterogeneity, a possible use for the model
in some situations is to use its capability for generating a mass input func-
tion for the aquifer, but then to simulate the aquifer with a numerical solute
9 Weaver
-------
cu
2
o
O
1500
1000
500
0
* Model Result MW-13
• Model Result MW-12
• Model Result MW-1.4
A Field Data MW-13
CD Field Data MW-12
O Field Data MW-1.4
to
O5
en
Date
Figure 4: Model results and field data Hagerman Avenue, East Patchogue,
New York
8
1000
800
I
600
400
200
» K. = 43 m/d
• K, = 96 m/d
• KJ = 149 m/d
A Field Data MW-13
5
C7>
Date
Figure 5: Effects of hydraulic conductivity variation at MW-13, Hagerman
Avenue, East Patchogue, New York
10
Weaver
-------
o
I
8
c§
1000
800
i
600
400
200
0
ce,_ — 10 m
15 m
• ce,_ — 20 m
A Field Data MW-13
rarafflmraajosaJCT
Date
Figure 6: Effects of dispersivity variation at MW-13, Hagerman Avenue, East
Patchogue, New York
transport model.
Relatively close approximations of the concentrations could be obtained
for wells at these sites. In each case it was necessary to include degrada-
tion of benzene to achieve an order of magnitude estimate of concentration.
This reflects the common occurrence of benzene degradation and the power
of the decay constant in reducing concentrations. Where there was unsmooth
variation in concentration as at Hagerman Avenue and MW-1 of Mountain
Fuels, the model did not capture the fluctuation and is not capable of doing
so. Hagerman Avenue was simulated with relatively low values of dispersivity
(Gelhar et al., 1992) and degradation, while Mountain Fuels was simulated
with relatively high values of both parameters. These suggest, tentatively,
that dispersion and degradation are less important at Hagerman Avenue than
Mountain Fuels.
The examples presented in this paper showed that parameter values could
be selected from field data sets so that HSSM matched the data to within an
order-of-magnitude for Hagerman Avenue and Mountain Fuels. The model
results may or may not be predictive of future conditions at the sites because
of parameter uncertainty, model assumptions, and that the values used were
fitting parameters. The two sites selected for this study are unique as they
have unusual amounts of data (Hagerman Avenue) or estimates of the time
11
Weaver
-------
and volume of the release (Mountain Fuels). These cases represent unusual op-
portunities for testing the HSSM against field data. The cases, also, illustrate
that for many fuel spills, site data is a limiting factor in testing or applying
models.
A cknowledgement
The information in this document has been funded wholly or in part by
the United States Environmental Protection Agency. It has been subjected
to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for
use. The author thanks Joseph Haas, New York State Department of Envi-
ronmental Conservation for providing the E. Patchogue data set; and Robin
Jenkins, Utah Department of Environmental Quality for providing the Salt
Lake City, Utah data set.
References
[1] R. J. Charbeneau, J. W. Weaver, B. K. Lien, 1995, The Hydrocarbon Spill
Screening Model (HSSM) Volume 2: Theoretical Background and Source
Codes, US EPA, EPA/600/R-94/039b.
[2] L. W. Gelhar, C. Welty, K. R. Rehfeldt, 1992, A critical review of data on
field-scale dispersion in aquifers Water Resources Research, 28(7), 1955-
1974.
[3] M. W. Kemblowski and C. Y. Chiang, 1990, Hydrocarbon thickness fluc-
tuations in monitoring wells, Ground Water, 28(2), 244-252.
[4] State of Utah, 1996, Unpublished case file.
[5] J. W. Weaver, R. J. Charbeneau, J. D. Tauxe, B. K. Lien and
J. B. Provost, 1994, The Hydrocarbon Spill Screening Model (HSSM) Vol-
ume 1: User's Guide, US EPA, EPA/600/R-94/039a.
[6] J. W. Weaver, J. E. Haas, J. T. Wilson, 1996 Analysis of the Gasoline
Spill at East Patchogue, New York, Proceedings of the Conference on
Non-aqueous Phase Liquids in the Subsurface Environment: Assessment
and Remediation, American Society of Civil Engineers, November 14-16,
Washington, D.C.
12 Weaver
-------
Analysis of the Gasoline Spill at East Patchoque, New York
James W. Weaver, Joseph E. Haas, John T. Wilson
Accepted for the Proceedings of the American Society of Civil Engineers
Conference on Non-aqueous Phase Liquids in the Subsurface Environment:
Assessment and Remediation
November 12-14, 1996
Washington, B.C.
Weaver et al.
-------
Analysis of the Gasoline Spill at
East Patchogue, New York
James W. Weaver1 Joseph E. Haas2 John T. Wilson3
Abstract
Gasoline containing methyl ierf-butyl ether (MTBE) was released from a
service station in East Patchogue, Long Island, New York. The resulting plume
of contaminated ground water was over 1800 m (6000 feet) long, and resulted
in the closing of private water supply wells. Data from a three-dimensional
monitoring network were used to estimate the mass and position of the center
of mass of benzene, toluene, ethylbenzene, xylenes and MTBE contaminant
plumes. The monitoring network was sampled on three occasions so temporal
information on the evolution of the plume was available. By estimating the
moments of the contaminant distributions for each of the sample rounds, the
loss of mass of each contaminant was estimated, as was the rate of migration
of the center of mass. An estimate of the volume of gasoline released was made
from plausible estimates of the gasoline composition.
Introduction
Over 300,000 releases from leaking underground storage tanks have been
reported to state regulatory authorities (USEPA, 1995). Depending on a num-
ber of factors, chemicals which compose fuels may form contaminant plumes
in the ground water. Field and laboratory investigations have established that
the most important of these (benzene, toluene, ethyl benzene and the xylenes
1 National Risk Management Research Laboratory, United States Environ-
mental Protection Agency, Ada, Oklahoma 74820
2 Division of Spills Management, New York State Department of Environ-
mental Conservation, Stony Brook, New York 11790
3 National Risk Management Research Laboratory, United States Environ-
mental Protection Agency, Ada, Oklahoma 74820
1 Weaver et al.
-------
[BTEX]) are degradable under most conditions (see e.g., Rifai et al., 1995).
Oxygenated additives that are used as octane enhancers and as mandated by
the Clean Air Act have also found their way into subsurface water supplies
(Squillace et al., 1995). The purpose of this paper is to describe a gasoline
release that occurred in East Patchogue, New York and to use chemical data
collected from the aquifer to estimate the mass and location of center of mass
of each constituent, the gasoline release volume, and the ground water flow ve-
locity. A companion paper describes simulation modeling of the site (Weaver,
1996b).
Background
Published studies of groundwater flow on Long Island indicate that a re-
gional ground water divide lies along the length of the island and to the north
of the geographic centerline (Eckhardt and Stackelberg, 1995). South of the di-
vide, flow is generally toward the Atlantic Ocean. Buxton and Modica (1993)
estimate that the hydraulic conductivity of the upper glacial aquifer is on
the order of 8.1 x 10~2 cm/sec (230 ft/day) in the outwash section near the
southern shore, with estimated ground water velocites of 3.5 x 10~4 cm/sec
(1 ft/day) or greater.
Table 1 lists the density, /?, solubility, 5, organic carbon partition coefficient
Koc, fuel/water partition coefficient, K0, and the mass fraction in gasoline, x,
of MTBE and the BTEX compounds. Koc values were taken from Mercer and
Cohen, (1990) and US EPA (1990). The fuel/water partition coefficient and
mass fraction data were measured by Cline et al. (1991) on 31 samples of
gasoline from Florida. The range reported covers the variation in measured
mass fractions in samples from other parts of the continent and from lists
of typical gasoline compositions (see e.g., Cline et al., 1991, Corapcioglu and
Baehr, 1987).
The usage of methyl tert-b\iiyl ether, MTBE, began on Long Island in the
late 1970s, after EPA approved its usage as an octane enhancer. Initial usage
of MTBE on Long Island was likely in the range of 5% by volume. Oxygenated
additives were mandated to reduce carbon monoxide emmissions in 39 cities,
including New York City and Long Island, by the 1990 admendments to the
Clean Air Act. State of New York regulations have required use of fuel with
oxygen content between 2.7% and 2.9% in the winter months since 1992 (State
of New York, 1995). The most commonly used oxygenated additive is MTBE,
which provides the required oxygen content at about 15% MTBE by volume.
The subsurface behavior of MTBE is notable for two reasons. First, MTBE
is highly water soluble. As a measure of the solubility, the fuel/water partition
coefficient for MTBE is about 23 times lower than that for benzene and 280
times lower than those for m- or p-xylene (Table 1). The release of MTBE
2 Weaver et al.
-------
Chemical
MTBE
benzene
toluene
ethyl benzene
m-xylene
p-xylene
o-xylene
density
g/mL
0.74
0.88
0.87
0.87
0.86
0.86
0.88
solubility
mg/L
48000
1750
535
152
130
196
175
L/kg
11.2
83 (65)
300 (257)
1100 (676)
982 (691)
870 (691)
830 (691)
K°(6)
15.5
350
1250
4500
4350
4350
3630
% (mass)
1.73 (0.7-3.8)
9.51 (4.5-21.0)
1.61 (0.7-2.8)
00
(<0
2.33 (1.1-3.7)
(") Organic carbon partition coefficient reported by Mercer and Cohen (1990), the second
value (in parenthesis) from US EPA (1990).
(') Fuel/water partition coefficient reported by Cline et al. (1991).
(c) Mass fraction of chemical in gasoline reported by Cline et al. (1991).
(d) m- and p-xylene were not differentiated they composed 5.95% with range of 3.7% to
14.5%.
Table 1: Chemical parameter values
from gasoline, therefore, is expected to be much more rapid than the release
of BTEX.
The second notable fact about MTBE is its recalcitrance to biodegrada-
tion. Microcosm studies conducted with three soils showed no degradation
of MTBE over a 250 day study period under anaerobic conditions (Yeh and
Novak, 1994). Degradation was induced under anaerobic conditions with the
addition of nutrients, a hydrogen source and molybdate in an organic-poor soil.
In organic rich soils degradation of MTBE could not be induced. Horan and
Brown (1995) concluded MTBE degradation might occur at a very low rate,
however, under aerobic conditions. In a controlled field study, gasoline with
10% MTBE, and an 85% methanol/15% gasoline blend were released in the
same aquifer (Hubbard et al., 1994). MTBE was found to be recalcitrant to
degradataion, while methanol and BTEX were degraded. Further, the MTBE
had no measurable effect on the degradation of the other compounds.
Site History
Subsurface contamination was detected at E. Patchogue, New York when
water from a residential well on Hagerman Avenue became undrinkable. The
site investigation began at the well and expanded through the drilling of mon-
itoring wells in the up-gradient and down-gradient directions (Figure 1). The
3 Weaver et al.
-------
Figure 1: Hagerman Avenue Site Plan.
purpose of the drilling was to delineate the extent of contamination and locate
the suspected source. Ultimately, the source was traced back to an aban-
doned service station approximately 1200 m (4000 ft) up-gradient from the
Hagerman Avenue residence. Soil borings in the area of the service station
confirmed the presence of hydrocarbon contamination. The service station's
tanks are believed to have been removed in 1988, which is the latest date that
gasoline could have been released. In 1994 and 1995, the contaminant plume
was mapped from samples taken from 26 multilevel samplers and 22 monitor-
ing wells. Water samples from three sample rounds were analyzed for BTEX
and MTBE. Total organic carbon contents were determined on 11 clean core
samples.
Moments Analysis
The relatively large number of monitoring wells and multilevel samplers
generated a three-dimensional data set, which were analyzed by calculating
the moments of each concentration distribution. The moments, M^k, are
Weaver et al.
-------
defined by
Mijk = I xiyjzk n C(x, y, z) dx dy dz (1)
where z, y, and z are the moment arms, n is the porosity, C(x,y,z) is the
concentration. These moments can be used to estimate the mass of the con-
taminant distribution, given by the zeroth moment, M°°°. Likewise the first
moments can be used to determine the center of mass of the distribution:
M100 M010 M001
Xc = ~OOO 2/c = zc = -55^ (2)
where xc, t/c, and zc are the x, y, and z coordinates of the center of mass of
the distribution.
The challenge in applying equation 1 to field data is in evaluating the
integrals. The SITE-3D program developed by US EPA for visualization of site
data (Weaver, 1996a) was used to generate the moment estimates by dividing
the contaminant plume into a set of nearest-neighbor polygons. The polygons
represent zones of influence of each well. In essence, the polygons replace the
explicit interpolation schemes between sampling locations that have been used
in other analyses (Freyberg, 1986 among others).
For most of the plume, the wells cross the entire width of the plume. In
some locations, however, monitor wells with high contaminant concentrations
are located on the edge of the sampling network (MW-12, MW-30, MW-38,
MW-39). Therefore some of the contaminant mass is not included in the esti-
mates given below. Because the MTBE is located down-gradient of MW-30,
MW-38, and MW-39, its mass estimates were not greatly impacted.
Results
Table 2 shows the mass estimates and the distance of the center of mass
of the contaminant distribution from the contaminant source, dcom. The data
in sample round one were taken as the wells were installed from July 1994 to
March 1995. The average date of the first sample round, weighted by number
of samples taken, is December 16, 1994. Data from sample round two were
taken from April 11, 1995 to April 20, 1995 and those from sample round three
were taken from October 10, 1995 to October 24, 1995. Since the samples in
round one were taken over a long time period, contaminants sampled up-
gradient may have been transported to down-gradient receptor wells before
they were sampled. The order of sampling, however, proceeded up-gradient
from the discovery point (MW-1) to the suspected source, followed by the wells
down-gradient from MW-1.
Each of the chemicals listed in Table 2 has some tendency for sorption.
Since the chemical data come from water samples, the sorbed mass must be
5 Weaver et al.
-------
Chemical Sample Round One Sample Round Two Sample Round Three
Mxw '°' Mxa ' ) dcom \c' Mxw MX, dcom MXw Mx, dcom
kg kg m kg kg m kg kg m
MTBE
B
T
E
X
268
241
108
29
149
24
156 (122)
253 (217)
249 (153)
1041 (804)
1387
991
230
347
222
386
117
65
24
95
34
76 (59)
152 (130)
206 (127)
663 (513)
1557
1004
298
347
277
229
58
60
21
92
20
38 (29)
141 (120)
180(111)
643 (497)
1583
1061
306
326
272
(°) Mxw is the mass dissolved in ground water.
^ Mx, is the mass sorbed to the aquifer solids, estimated from the Koc reported by Mercer
and Cohen (1990) and, in parenthesis, from that reported by US EPA (1990).
^ dcom is the distance from the suspect source to the center of mass of the contaminant
distribution.
Table 2: Moment based estimates of mass and center of mass
estimated. Chemicals sorb in proportion to the fraction of organic carbon in
the aquifer material, /oc, and the chemical's organic carbon partition coeffi-
cient, Koc. Sorption was assumed to follow the linear equilibrium isotherm as
given by
LXS = -Kocfoc^xw . (3)
where Cxs is the sorbed concentration of contaminant x expressed per unit
mass of aquifer solids, and Cxw is dissolved concentration of chemical x. The
sorbed mass of contaminants was estimated from
Mxs = ^KocfocMxw (4)
n
where Mxa and Mxw are the respective sorbed and dissolved masses of chemical
x, and pi, is the bulk density. Organic carbon contents were determined for 11
uncontaminated samples taken from 4.88 m to 8.23 m (16 ft to 27 ft) below
the ground surface near the source region. The arithmetic average of foc was
0.126%, with range of 0.009% to 0.627% and standard deviation of 0.190%.
The porosity and solids density were assumed to equal 0.30 and 2.65 g/cm3,
respectively, giving a bulk density of 1.86 g/cm3. The Koc values were taken
from Table 1. Table 2 lists estimated sorbed masses for each chemical.
The estimated mass of benzene, toluene, ethyl-benzene and the xylenes
decreased between each sample round. Each of these compounds is expected
to undergo biodegradation in the aquifer, but each continued to dissolve into
the aquifer through from October 1995. The latter fact is established by the
persistence of BTEX concentrations near the source. The mass of MTBE,
6 Weaver et al.
-------
however, appeared to increase between the first two sample rounds; then de-
creased between the second and third sample rounds. The distribution of
MTBE was such that in all sample rounds, no MTBE was found between the
source and a point approximately 600 m (2000 ft) down-gradient (Figure 2).
Thus it appears that the MTBE was almost entirely leached from the gasoline
near the source.
A'
E. Patchogue, NY: Sample Round 2
MTBE (ppb)
40.00-
30.00-
-110.00
-120.00
5500
6000
6500 7000
Distance (ft)
Figure 2: Distribution of MTBE in sample round two
The average concentrations over the entire plume are given in Table 3.
These concentrations were calculated from the pore volume estimates and
measured concentrations at all points in the sampling network where concen-
trations were above the detection limit.
Estimation of the Mass of Gasoline Released
The mass of contaminants in the aquifer can be used to place bounds on
the volume of gasoline released. The estimated mass of MTBE in the aquifer
is 292 kg for sample round one and 420 kg for sample round two. Since the
gasoline must have been released before the Clean Air Act mandates, MTBE
was assumed to comprise 5% by volume of the gasoline. The corresponding
volume of gasoline for these estimated masses would be 7.89 m3 (2080 gallons)
Weaver et al.
-------
Chemical Sample Round One Sample Round Two Sample Round Three
Concentration Concentration Concentration
mg/L mg/L mg/L
MTBE
benzene
toluene
ethyl benzene
xylenes
421
749
419
153
746
789
275
225
134
558
806
148
293
223
516
Table 3: Concentrations averaged over the entire contaminant plume
and 11.35 m3 (2999 gallons). Because of the apparent complete leaching of
MTBE from the gasoline, this estimate would represent the entire volume of
MTBE enhanced gasoline released to the aquifer.
The BTEX data suggest the volumes of gasoline listed in Table 4, assum-
ing that the density of the gasoline was 0.72 g/cm3. In the absence of specific
knowledge concerning the composition of the released gasoline, the estimates
developed by Cline et al., (1991) (see Table 1) were used in estimating the
gasoline volumes in Table 4. Unlike MTBE each of the BTEX chemicals per-
sists in gasoline at the source (Figure 3). More of the benzene orginally con-
tained in the gasoline, however, would be in the aquifer than any of the other
BTEX compounds because of benzene's lower fuel/water partition coefficient.
A greater fraction of each of T, E and X remain in the gasoline because of
their higher affinities for the gasoline phase (expressed in Table 1 by their lower
water solubilities and higher fuel/water partition coefficients). The benzene-
based gasoline volume estimate is the minimum estimate for these reasons. To
contrast with the Cline et al., (1991) average benzene mass fraction of 1.7%,
the often-used estimate of 1% by mass gives an estimated gasoline volume
of 55.1 m3 (14600 gallons) for a benzene mass of 397 kg and 50.4 m3 (13300
gallons) for a benzene mass of 363 kg.
Estimation of the Average Rate of Advance of the Contaminant Plumes
The average rate of advance of the contaminant plumes are related to the
ground water velocity and can be estimated from the position of the center of
mass of the distributions during each sample round. These moment-based es-
timates include concentration changes, presumably caused by transport, from
all sample locations, and thus average fast and slow moving portions of the
plume. Generally, in sample round one the data were collected from the down-
gradient wells first, followed by those nearer the source. Thus 101 days passed
between the average sample date for sample round one and sample round two.
The wells near the Hagerman Avenue residence, however, were sampled 216
8 Weaver et al.
-------
E. Patchogue, NY: Sample Round 2
Benzene (ppb)
-120.00
I
500
1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000
Distance (ft)
Figure 3: Distribution of benzene in sample round two
days before the sampling for sample round two. Sample rounds two and three
were separated by 186 days. Based upon these durations, average rates of
advance of the contaminant plumes are given in Table 5.
Because of the long duration of sample round one, the most reliable of
the results are those for sample round two to three. The aquifer properties
and gradients are presumed to vary with space, so the velocities may depend
on position. For example, the gradient is lower at the down-gradient end of
the plume, which may have caused the MTBE plume to move slower between
sample rounds two and three than did the benzene plume, because the cen-
ter of mass benzene plume was approximately 500 m up gradient from the
MTBE center of mass (Table 2). The increase in average MTBE concentra-
tion from sample rounds one to three supports this contention as mass could
be accumulating down-gradient, causing the average concentration to increase
(Table 3). Also, as shown in Figures 2 and 3, the vertical thickness of the
plume increases down-gradient. The velocity of the center of mass should de-
crease in the thicker part of the plume by mass balance, because the mass of
contaminants fills a larger volume. The average MTBE concentration would
be expected to decrease, which it does not (Table 3), however. The centers
of mass of both the ethyl benzene and xylenes plumes retreated from sample
round two to sample round three. This suggests that the rate of input of mass
Weaver et al.
-------
Chemical
benzene
toluene
ethyl benzene
xylenes
(«)
Sample Round One Gasoline Volume Estimates (gal)
mass mass fraction from Cline et al., 1991
kg low middle high
(«) 397
<»> 363
<°) 361
<*> 325
(«) 278
(») 182
<•> 1190
(*) 953
Mass estimate using Koc
20807
19025
2943
2650
14624
9539
9096
7284
of Mercer
8567
7834
1393
1255
6335
4148
5273
4223
and Cohen
3832
3505
631
568
3643
2385
2399
1921
(1990)
Moss estimate using Koc of USEPA (1990)
Table 4: Gasoline volume estimates from BTEX mass estimates
of these compounds to the aquifer is less than the rate of their loss.
Rate of Advance (m/d)
Chemical Sample Rounds
One to Two Two to Three
101 d 216 d 186 d
MTBE
benzene
toluene
ethyl benzene
xylenes
1.68
0.13
0.67
0
0.56
0.79
0.06
0.31
0
0.27
0.14
0.30
0.04
-0.11
-0.02
Table 5: Average rate of advance of the contaminant plumes
Conclusions
The extensive monitoring network at the Hagerman Ave site allows de-
termination of the mass and moments of the contaminant distributions. The
accuracy of the mass estimates depends upon the sampling network, duration
of sample events, and the accuracy of the procedure used for forming the es-
timates. Each of these introduces uncertainty into the estimates presented in
this paper.
The mass of each of the BTEX compounds appears to decrease over the
three sample rounds, indicating a net loss of mass in the aquifer. MTBE data
do not show a clear trend. The mass of each contaminant in the aquifer can
be used to give an estimate of the volume of the gasoline release. The mass
10
Weaver et al.
-------
of MTBE in the aquifer represents approximately 11.35 m3 (2999 gallons) of
MTBE-enhanced gasoline. Since MTBE was not used regularly before 1992,
this gasoline volume estimate is likely to be low. From sample round one, the
mass of benzene gives a maximum lower bound estimate of approximately 50
m3 (13200 gallons), if the benzene composed 1 % by mass of the gasoline. Since
the released gasoline composition is unknown, this estimate can be considered
a tentative estimate of the gasoline volume. The masses of the other BTEX
compounds gave lower gasoline volume estimates which are thereby consistent
with the estimate from the benzene. The rates of advance of the contaminant
plumes can give estimates of the ground water velocity. From sample rounds
two and three, MTBE advanced 0.14 m/d (0.5 ft/d) which gives a plausible
velocity for the down-gradient portion of the site.
The mass of the release influences all activities at a contaminated site.
Because the release or releases which occurred at Hagerman Avenue occurred
at unknown times and intervals, much about the contamination at the site
remains unknown. By studying the data from the gasoline release, estimates
of important quantities have been developed. Although, the values are not
completely certain, they represent plausible estimates for the site.
Disclaimer
The information in this document has been funded wholly or in part by the United
States Environmental Protection Agency. It has been subjected to Agency review
and approved for publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use. The authors express their
thanks to Sarah Hendrickson and Julia Mead of the United States Environmen-
tal Protection Agency's Environmental Research Apprenticeship Program, Charles
Sosik of Environmental Assessment and Remediation of Patchogue, New York, and
three anonymous ASCE reviewers.
References
[1] H. T. Buxton and E. Modica, 1993, Patterns and rates of ground-water flow
on Long Island, New York, Ground Water, 30(6), 857-866.
[2] P. V. Cline, J. J. Delfino, P. S. C. Rao, 1991, Partitioning of aromatic con-
stituents into water from gasoline and other complex solvent mixtures, Envi-
ronmental Science and Technology, 23, 914-920.
[3] M. Y. Corapcioglu and Baehr, 1987, A compositional multiphase model for
groundwater contamination by petroleum products, Water Resources Research
23, 201-243.
[4] D. A. V. Eckhardt and P. E. Stackelberg, 1995, Relation of ground-water quality
to land use on Long Island, New York, Ground Water, 33(6), 1019-1033.
11 Weaver et al.
-------
[5] D. L. Freyberg, 1986, A natural gradient experiment on solute transport in a
sand aquifer: 2. Spatial moments and the advection and dispersion of nonre-
active tracers, Water Resources Research, 22(13), 2031-2046.
[6] C. M. Koran, E. J. Brown, 1995, Biodegradation and inhibitory effects of
Methyl-Tertiary-Butyl Ether (MTBE) added to microbial consortia, Proceed-
ings of the 10
-------
Computer
Familiarization
-------
13. Computer Familiarization
-------
Computer Familiarization
DOS
To view directory:
To view file:
To edit file:
To delete a file:
To delete a directory:
To change directories
dir
dir Imore
type file.txt
type file.txt Imore
edit file.txt
del file.txt
rmdir dname
cd/
cd.
cd..
cddl
cd/d1
cd ../d2
go to root directory
show current directory
go up one level in directory
go to d1 (assumed below current directory)
go to d1 (under root)
go to d2 (assumed lateral to current
directory)
/3- /
-------
Windows Installation Procedure for the Windows Version
of the
Hydrocarbon Spill Screening Model
Start Windows
Launch Windows by entering the command
win
at the DOS prompt.
Copy the HSSM model onto the hard drive fc:)
Locate the Windows File Manager program
File Manager is usually found in the Main program group
File Manager's icon looks like a filing cabinet
Start File Manager by clicking twice on the icon
Open a window for the diskette drive (a:) by clicking twice on the drive letter at the top
of the file manager window
"Tile" the windows by selecting Windows on the file manager menu bar, and then
select "tile"
Now the screen should show directories for the hard drive and the diskette drive
For drive c:
Select the root directory by scrolling up the left hand portion of the window
and clicking on c:\
(The root directory is the highest level of directory)
Select File from the File Manager menu bar
Select Create directory from this menu
-------
In the dialog box type
HSSM
and click once on the OK button
Check to see that the directory HSSM has been created by scrolling through the directory
entries
Select the HSSM directory by clicking on HSSM
Select the root directory on the a: diskette drive
While holding the left mouse button down, drag the mouse to the newly created HSSM
directory
Release the button and the files will be copied
Close the File Manager
install the HSSM program
Within Program Manager:
Select File on the Program Manager menu bar
Select New from this menu
Select Program Group
Enter HSSM Model in the dialog box for the Description
Again select File on the Program Manager menu bar
Select New from this menu
Select Program Item
Enter HSSM in the dialog box for the Description
Enter c: \HSSM\HSSM-WIN.EXE for the Command line
Click twice on the icon to execute HSSM
/3-3
-------
Windows Installation Procedure for the Windows Version
of the
Hydrocarbon Spill Screening Model
® Start Windows
Launch Windows by entering the command
win
at the DOS prompt.
Copy the HSSM model onto the hard drive (c:)
Locate the Windows File Manager program
File Manager is usually found in the Main program group
File Manager's icon looks like a filing cabinet
Start File Manager by clicking twice on the icon
Open a window for the diskette drive (a:) by clicking twice on the drive letter at the top
of the file manager window
"Tile" the windows by selecting Windows on the file manager menu bar, and then
select "tile"
Now the screen should show directories for the hard drive and the diskette drive
For drive c:
Select the root directory by scrolling up the left hand portion of the window
and clicking on c:\
(The root directory is the highest level of directory)
Select File from the File Manager menu bar
Select Create directory from this menu
-------
In the dialog box type
HSSM
and click once on the OK button
Check to see that the directory HSSM has been created by scrolling through the directory
entries
Select the HSSM directory by clicking on HSSM
Select the root directory on the a: diskette drive
While holding the left mouse button down, drag the mouse to the newly created HSSM
directory
Release the button and the files will be copied
Close the File Manager
® Install the HSSM program
Within Program Manager:
Select File on the Program Manager menu bar
Select A/ewfrom this menu
Select Program Group
Enter HSSM Model in the dialog box for the Description
Again select File on the Program Manager menu bar
Select A/eivfrom this menu
Select Program Item
Enter HSSM in the dialog box for the Description
Enter c: \HSSM\HSSM-WIN. EXE for the Command line
® Click twice on the icon to execute HSSM
-------
Computer Familiarization
DOS
To view directory:
To view file:
To edit file:
To delete a file:
To delete a directory:
To change directories
dir
dir I more
dir/p
type file.txt
type file.txt \more
edit file.txt
del file.txt
rmdir dname
cd/
cd.
cd ..
cddl
cd/d1
cd ../d2
go to root directory
show current directory
go up one level in directory
go to d1 (assumed below current directory)
go to d1 (under root)
go to d2 (assumed lateral to current
directory)