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 [*]

-------
Confined Aquifer Parameters
Transmissivity

          T = Kb
          K = hydraulic conductivity [L/T]
          b = aquifer thickness [L]
Storativity

          S = S. b
              's
          Ss = specific storage [L1]
          b = aquifer thickness [L]
     Storativity is the volume of water that an aquifer releases from storage
     per unit surface area per unit decline in the hydraulic head.

     0.005 to 0.00005 (Freeze and Cherry, 1979)

     Water is released from storage in confined aquifers when the head
          falls because
          © of compaction of the porous medium
          © of expansion of the water

-------
Vertical Flow in Confined Aquifers
                               t

-------
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)

-------
Two-Dimensional Flow
Unconfined Aquifer

-------
Two-Dimensional Flow
Unconfined Aquifer
                             -
                           dy \    dy /
                                           w  =   S         (ID
Where    x and y are the coordinate directions [L]
          t is the time [T]
          h is the hydraulic head [L]
          K is the hydraulic conductivity [L/T]
          w is the leakage rate per unit area [L/T]
          Sy is the specific yield  [*]
     Specific yield is  the volume of water that  an unconfined  aquifer
releases from storage per unit surface area per unit decline in the hydraulic
head.

     Unconfined aquifers release water when the head declines because

     © of dewatering of media pores
     ® of compaction of the porous medium
     ® of expansion of the water

     0.01 - 0.30 (Freeze and Cherry, 1979)

-------
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 1  Solute transport at SL Joseph, Michigan

     D Data set is three dimensional (collected from slotted auger borings)
     D Flow is three dimensional
          A downward flow component ?
          Stratification of the aquifer
     D The aquifer thickness  is about 50 to 60 feet, the length of the flow
          system is about 1500 feet

Problems

     ®  The distribution of chemicals in the vertical is important for analysis
          of bioremediation.
     ®  The flux of chemicals  into Lake Michigan is needed.
Notes
     ©  MODFLOW and MT3D take approx. 36 hours on a 486-66MHz PC
          for a 23 year simulation.

-------
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).

-------
Calibration

From Freyberg, 1988,  An Exercise in Ground-Water Model Calibration and Prediction,
      Ground Water, 26(3), 350-360.
"The parameter identification problem may be defined simply as

      the solution of the ground-water flow equation for the values of its parameters
            using observations of the dependent variable (hydraulic head), aquifer
                  geometry, and boundary conditions."
D Parameter estimates are highly sensitive to noise
D Typical solutions are nonunique
Two approaches:


      © Trial and Error

      ® Automated

           M.C.  Hill,  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
                                    3-33

-------
 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)

-------
Example Finite Difference Application
Part 1  Ground Water Flow
©  Illustrate general construction of numerical model
©  Illustrate classical finite difference techniques

-------
 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

-------
Table 1.  Most prevalent chemical compounds at U.S. Superfund Sites (65) with a specific gravity
         greater than one.
Density
Compound [1]
Halogenated Semi-volatiles
1,4-Dichlorobenzene 1.2475
1 ,2-Dichlorobenzene 1.3060
Aroclor1242 1.3850
Aroclor1260 1.4400
Aroclor1254 1.5380
Chlordane 1 .6
Dieldrin 1.7500
2,3,4,6-Tetrachlorophenol 1.8390
Pentachlorophenol 1.9780
Halogenated Volatiles
Chlorobenzene 1 . 1 060
1,2-Dichloropropane 1.1580
1,1-Dichloroethane 1.1750
1,1-Dichloroethylene 1.2140
1 ,2-Dichloroethane 1 .2530
Trans-1 ,2-Dichloroethylene 1 .2570
Cis-1 ,2-Dichloroethylene 1 .2480
1,1,1-Trichloroethane 1.3250
Methylene Chloride 1 .3250
1,1,2-Trichloroethane 1.4436
Trichloroethylene 1 .4620
Chloroform 1 .4850
Carbon Tetrachloride 1.5947
1,1,2,2-Tetrachloroethane 1.6
Tetrachloroethylene 1 .6250
Ethylene Dibromide 2.1720
Non-halogenated Semi-volatiles
2-Methyl Napthalene 1.0058
o-Cresol 1 .0273
p-Cresol 1 .0347
2,4-Dimethylphenol 1.0360
m-Cresol 1 .0380
Phenol 1.0576
Naphthalene 1.1620
Benzo(a)Anthracene 1.1740
Flourene 1.2030
Acenaphthene 1 .2250
Anthracene 1 .2500
Dibenz(a,h)Anthracene 1.2520
Fluoranthene 1 .2520
Pyrene 1.2710
Chrysene 1 .2740
2,4-Dinitrophenol 1.6800
Miscellaneous
Coal Tar 1 .028<7>
Creosote 1 .05
[1] g/cc
Dynamic[2]
Viscosity

1 :2580
1 .3020



1.1040




0.7560
0.8400
0.3770
0.3300
0.8400
0.4040
0.4670
0.8580
0.4300
0.1190
0.5700
0.5630
0.9650
1 .7700
0.8900
1 .6760





21.0












1 8.98<"
1 .08<8>
Kinematic
Viscosity[3]

1.008
0.997



0.69




0.683
0.72
0.321
0.27
0.67
0.321
0.364
0.647
0.324
0.824
0.390
0.379
0.605
1.10
0.54
0.79





20
3.87













Water[4]
Solub.

8.0 E+01
1.0 E+02
4.5 E-01
2.7 E-03
1.2 E-02
5.6 E-02
1.86 E-01
1.0 E+03
1.4 E+01

4.9 E+02
2.7 E+03
5.5 E+03
4.0 E+02
8.69 E+03
6.3 E+03
3.5 E+03
9.5 E+02
1 .32 E+04
4.5 E+03
1 . 0 E+03
8.22 E+03
8.0 E+02
2.9 E+03
1.5 E+02
3.4 E+03

2.54 E+01
3.1 E+04
2.4 E+04
6.2 E+03
2.35 E+04
8.4 E+04
3.1 E+01
1.4 E-02
1.9 E+00
3.88 E+00
7.5 E-02
2.5 E-03
2.65 E-01
1 .48 E-01
6.0 E-03
6.0 E+03



Henry's Law
Constant[5]

1 .58 E-03
1 .88 E-03
3.4 E-04
3.4 E-04
2.8 E-04
2.2 E-04
9.7 E-06

2.8 E-06

3.46 E-03
3.6 E-03
5.45 E-04
1 .49 E-03
1.1 E-03
5.32 E-03
7.5 E-03
4.08 E-03
2.57 E-03
1.1 7 E-03
8.92 E-03
3.75 E-03
2.0 E-02
5.0 E-04
2.27 E-02
3. 18 E-04

5.06 E-02
4.7 E-05
3.5 E-04
2.5 E-06
3.8 E-05
7.8 E-07
1 .27 E-03
4.5 E-06
7.65 E-05
1.2 E-03
3.38 E-05
7.33 E-08
6.5 E-06
1.2 E-05
1 .05 E-06
6.45 E- 10



Vapor[6]
Pressure

6 E-01
9.6 E-01
4.06 E-04
4.05 E-05
7.71 E-05
1 E-05
1 .78 E-07

1.1 E-04

8.8 E+00
3.95 E+01
1 .82 E+02
5 E+02
6.37 E+01
2.65 E+02
2 E+02
1 E+02
3.5 E+02
1.88 E+01
5.87 E+01
1.6 E+02
9.13 E+01
4.9 E+00
1.4 E+01
1.1 E+01

6.80 E-02
2.45 E-01
1.08 E-01
9.8 E-02
1.53 E-01
5.293E-01
2.336E-01
1.16 E-09
6.67 E-04
2.31 E-02
1.08 E-05
1 E-10
E-02 E-06
6.67 E-06
6.3 E-09
1.49 E-05



[5] atm-m3/mol
[2] centipoise (cp), water has a dynamic viscosity of
[3] centistokes (cs)
[4] mg/l

1 cp at 20°C.



[6] mm Hg
[7] 45° F (70)


[8] 15.5°C, varies with creosote mix (62)

-------
                             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)

-------
                      Commonly Available Models
Model Name

3DFEMWATER

3DLEWASTE

AGU-10
AIRFLOW
ASM
AT123D
BALANCE
BEAVERSOFT
BIOPLUME H
CATTI
CFITIM

CHEMFLO
CMLS
COVAR
                      Distribution Center

                      CEAM
                      PSU
                      CEAM
                      PSU
                      IGWMC
                      WHydro
                      IGWMC
                      IGWMC
                      IGWMC
                      IGWMC
                      CSMoS
                      IGWMC
                      IGWMC
                      USSL
                      CSMoS
                      OSU
                      IGWMC
CTM (Chemical Transport Model)  PNL
                      CSU
                      IGWMC
                      API
                      CEAM
                      CEAM
                      IGWMC
                      OSW
                      OSW
                      CEAM
                      CEAM
                      IGWMC
                      WHydro
                      WHydro
                      WHydro
                      WHydro
                      EMSL-LV
                      CSMoS
                      CERI
                      IGWMC
                      CEAM
                      CSMoS
CSUGAS
CXTFIT
DSS/API
DYNHYD4
DYNTOX
EPA-VHS
EPACML
EPACMTP
EXAMS-H
FGETS
FP
FLOWCAD
FLOWNET
FLOWPATH
FLOWTRANS
GEOEAS
GEOPACK
GRIT/STAT
GWFLOW
HSPF
HSSM
Phone #

(706)546-3130
(814)863-2931
(706)546-3130
(814)863-2931
(303)273-3103
(519)746-1798
(303)273-3103
(405) 436-8586
(303)273-3103

(909) 369-4847
(405) 436-8586
(405) 744-9592
(303) 273-9592
(303)491-8381
(303)273-3103
(706)546-3130

(303)273-3103
(202) 260-4765

("-.it) 546-3130

(303)273-3103
(519)746-1798
(702)798-2100
(405) 436-8586
(513)569-7272
(303)273-3103
(706)546-3130
(405) 436-8586
                                     -4-
                                                       J.R. Williams (August 7, 1996)

-------
HST3D
HYDRUS
HYPERVENTILATE
ICE-1
INFIL
MICROFEM
MINTEQA1
MOC
MOCDENSE
MODFLOW
MODPATH
MOFAT
MOTRANS
MT3D
MULTMED
NETPATH
OASIS
ONED
ONESTEP
OPTP/PTEST
PAT
PESTAN
PESTRUN
PHREEQE
PLASM
PLUME
PLUME2D
PORFLOW
PRINCE
PRZM
PUMPTEST
QUAL2E
RADFLOW
RANDOM WALK

RETC
RITZ
RWH
SARAH
IGWMC
USSL, IGWMC
MWRI
NTIS
EPA
IGWMC
IGWMC
IGWMC
CEAM
USGS
IGWMC
G&M
SSG
USGS, IGWMC, G&M, SSG
IGWMC, G&M, SSG
IGWMC
CSMoS
ES&T
CSMoS
CEAM
IGWMC
CSMoS
IGWMC
IGWMC
IGWMC
IGWMC
CSMoS
IGWMC
IGWMC
ISWS
IGWMC
IGWMC
ACR
WHydro
CEAM
IGWMC
CEAM
IGWMC
ISWS
IGWMC
CSMoS
CSMoS
IGWMC
CEAM
(303)273-3103
(714)369-4847
(919)677-0249
(202) 783-3238
(703) 308-8877
(303) 273-3103
(706)546-3130
(703) 648-6978
(303)273-3103
(703)758-1200
(703)620-9214
(303)272-3103
(405) 436-8586
(800) 926-5923
(405) 436-8586
(706)546-3130
(303)273-3103
(405) 436-8586
(303)273-3103
(405) 332-8800
(303)273-3103

(217)333-4968
(303)273-3103

(310)471-3023
(519)746-1798
(706)546-3130
(303)273-3103
(706)546-3130
(303)273-3103
(217)333-4968
(303)273-3103
(405) 332-8800

(303) 273-8800
(706)546-3130
                                     -5-
                                  J.R. Williams (August 7, 1996)

-------
SESOIL

SOHYP
SOIL
SOILVENT
SOLUTE
ST2D
STF DATABASE
SUMATRA-1
SUTRA

SWACROP
SWANFLOW
SWICHA
SWIMM
TETRA
TGUESS
THCVFTT
THEISFTT
THWELLS
TIMELAG
TRAFRAP-WT
TSSLEAK
UNSAT1
USGS-2D-FLOW

USGS-3D-FLOW
VARQ
VENTING
VIRALT
VLEACH
WALTON35
WASP
WATEQ4F
WELL
WHAEM
WHPA
WQA
OEA

IGWMC
IGWMC
CSMoS-Cho
IGWMC
IGWMC
CSMoS
IGWMC
USGS
IGWMC
IGWMC
      IGWMC
IGWMC
CEAM
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
IGWMC
USGS
IGWMC, USGS
IGWMC
ES&T
IGWMC
CSMoS
IGWMC
      CEAM
IGWMC
IGWMC
CSMoS
CSMoS
CEAM
(202) 260-3930
(202) 382-3926
(303)273-3103

(405) 436-8547
(303)273-3103

(405) 332-8800
(303)273-3103
(703) 648-6978
(303)273-3103
(706)546-3130
(303)273-3103
(303)273-3103
(703) 648-6978
(800)926-5923
(303)273-3103
(405) 332-8800
(303)273-3103
     (706)546-3130
(303)273-3103

(405) 436-8586

(706)546-3130
                                     -6-
                                 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
January 1992
              Center for Subsurface Modeling Support
©EPA   STF:
              Soil Transport and  Fate Database
              and Model  Management  System
              (Version 2.0)
              Introduction

              The Soil Transport and Fate Database (STF) version 2.0
              presents quantitative and qualitative information concerning
              the behavior of organic and inorganic chemicals in soil
              environments. The STF Database system provides the user
              with recent information on chemical  properties,  toxicity,
              transformation and bioaccumulation for hundreds of chemical
              compounds. It is an extremely useful software package for
              environmental managers, scientists and regulators working
              in problems related to vadose contamination and remediation.

              Components

              The software consists of three major components: the STF
              Database; the Vadose Zone Interactive Processes (VIP) and
              Regulatory and Investigative Treatment Zone (RITZ) models
              and the VIP and RITZ model editors. The STF Database is
              divided  into seven different files:  (1) identification; (2)
              references; (3) chemical characteristics; (4) immobilization;
              (5) transformation/degradation;  (6) toxicity; and (7)
              bioconcentration.  The first two files assist the user in finding
              and accessing information regarding chemical compounds.
              Approximately 400 chemicals are included in the database,
              these compounds are identified by their chemical name as
              referenced in  40 CFR Part 261. the  Chemical Abstract
              Service (CAS) number, and the common chemical  name.

              The chemical characteristics file contains physical  and
              chemical information about the compounds. Fifteen different
              parameters are described such as viscosity, specific gravity,
              Henry's  Law Constant, and log octanol-water partition
              coefficient.  The immobilization file consists of information
              concerning  partitioning, immobility,  and  transport of the
              chemicals in a soil environment. This file contains three
              general categories, (1) soil type and conditions;  (2) soil
              properties; and (3) waste properties. The transformation/
              degradation file  contains  similar information  as the
              immobilization file but provides additional information on the
              reactivity and half-life of the compound.  Toxicological
              information concerning the compounds is provided in the
              toxicity file. This file includes such information as carcinogenic
              risk, the reference dose, LD50, and mutagenic dose. Finally,
             the bioconcentration file contains information regarding the
             bioaccumulation potential of the chemicals.

             The last two components of the STF software involve the use
             of the RITZ and VIP models, which are one-dimensional
             models that are useful in predicting the fate and transport of
                 hazardous organic constituents in the vadose zone. The VIP
                 and RITZ model editors aid in the creation of input files for the
                 models. The model editors are designed to interface with the
                 STF Database and allows the user to draw upon studies
                 published in the literature to assess potential transport and
                 fate of chemicals in  the vadose zone.   Input files are
                 developed in the RITZ and VIP model editors by using three
                 input sources: (1) site specific parameters entered directly by
                 the user; (2) literature-based parameters, entered from the
                 database; and (3) default parameters, provided as part of the
                 initial setup.

                 Documentation

                 The Soil Transport and  Fate  Database 2.0  and Model
                 Management System software  series  includes a well
                 documented user's manual. The manual fully describes the
                 use and contents of the database and model editors. In
                 addition, a tutorial is provided to introduce the user to each
                 of the operation modes in STF 2.0.

                 Software/Hardware Requirements

                 The software/hardware requirements for STF 2.0 and Model
                 Management System are:

                      IBM XT Compatible (AT or better is recommended)
                  •   640 K RAM Memory
                      Math Coprocessor  (for VIP and RITZ models only)
                      Hard Disk with 12.5 Mbytes free
                      Supports any display standard (MDA, CGA, EGA)

                 Availability

                 To obtain a copy of the STF 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.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

-------
               United States
               Environmental Protection
               Agency
Robert S. Kerr Environmental
Research Laboratory
P.O. Box1198
Ada, OK  74820
January 1993
               Center for Subsurface Modeling Support
©EPA   WHPA:
               A Modular Semi-Analytical  Model
               for the  Delineation of Wellhead
               Protection Areas
               (Version 2.2)
               Introduction

               WHPA, A Modular Semi-Analytical Model for the Delineation of
               Wellhead Protection Areas, Version 2.2, is a user-friendly
               computer model for the delineation of capture zones and
               contaminant fronts. The code was originally designed to assist
               federal, state, and local technical staff in the delineation of
               Wellhead Protection Areas as defined in the 1986 Amendments
               to the Safe Drinking Water Act, but the model can be applied to
               many different problems associated with wells.

               The WHPA code consists of four different particle-tracking
               modules: RESSOC. MWCAP. GPTRAC. and MOMTEC. Each of
               the modules has slightly different assumptions and requirements;
               however, all of the modules assume steady-state and horizontal
               flow in the aquifer. The RESSQC module delineates time-related
               capture zones around multiple pumping wells or contaminant
               fronts around multiple  injection wells in homogeneous aquifers of
               infinite areal extent with steady and uniform ambient ground-
               water flow. RESSOC  accounts for multiple well interference
               effects. MWCAP is similar to RESSOC but can incorporate
               stream or barrier boundary conditions for semi-infinite aquifers. It
               can be used to delineate steady, time-related, or hybrid capture
               zones.  The GPTRAC module contains two options: (1) semi-
               analytical, and (2) numerical.  The semi-analytical option is
               similar to RESSOC and MWCAP, but  can accomodate a wider
               range of aquifer and boundary conditions.  This option can
               simulate delineation in homogeneous  confined,  leaky-confined, or
               unconfined aquifers with areal recharge. The aquifer may be of
               infinite areal extent or may be bounded by one or two (parallel)
               stream and/or barrier (semi-permeable with ambient flow)
               boundaries. The numerical option performs particle tracking
               using a head field obtained from a numerical ground-water flow
               code and accounts for many types of boundary conditions as well
               as aquifer heterogeneities and anisotropies. Lastly, the
               MONTEC module performs uncertainty analysis (based on Monte
               Carlo techniques) for time-related capture zones for a single
               pumping well in a confined and leaky confined homogeneous
               aquifers of infinite areal extent.

               Input/Output

               WHPA requires relatively few input parameters, which makes it
               extremely useful as a screening tool in delineating capture zones
               or contaminant fronts.  The general parameters required by all
               the WHPA modules are transmissivity. porosity, saturated
               thickness and the rates of recharge or discharge from the
               simulated injection or pumping wells. Other parameters required
               for specific modules include location and type of boundaries, the
               areal recharge rate, confining bed hydraulic conductivity, and
               thickness of the confining bed.
                  WHPA presents the modeling output as a plot of the capture zone
                  and particle paths.  When simulating multiple wells, the respective
                  capture zones and particle paths will be shown in different colors.
                  The user can select the scale of the plot to any arbitrary ratio or to
                  one of five standard USGS topographic scales.  Plots of up to 15
                  different simulations can also be overiayed, one on top of the
                  other, for comparison analysis. This feature is especially useful in
                  conducting sensitivity analyses. A hard copy of the plot as well as
                  a tabulation of the data can be printed using most standard printer
                  or plotter output devices.  Plot files can be saved in HPGL format
                  or as an ASCII file that may be used as input to ARC/INFO or
                  other GIS proprietary software.

                  Documentation

                  The WHPA model includes a well documented users manual.
                  The manual describes approaches in simulating capture zone
                  problems, the mathematical basis of the code, the modules of the
                  model, and the pre- and post-processors of the code. Examples
                  using real data are presented for each module; these are
                  extremely helpful in understanding the application of the model to
                  real world conditions. The last chapter of the manual discusses
                  how the model can  be misapplied and can be very useful to the
                  inexperienced modeler.

                  Software/Hardware Requirements

                  The software/hardware requirements for WHPA are:

                          IBM PC. XT. AT (or compatible) computer

                          Minimum  640K RAM

                          Hard disk drive

                          CGA, EGA. VGA. or Hercules graphics capability

                          MS-DOS  version 2.1 or later operating system

                  Availability

                  To obtain a copy of  the WHPA 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 (or Subsurface Modeling Support
                                       P.O.Box 1198
                                      Ada. OK 74820
                                       (405) 436-8500

-------
MODFLOW

-------
7.   Practical Applications of
    MODFLOW

-------
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

-------
    MODELOW
Objective
    - Head (water level)
    - Flow rates
    - Velocities


-------
        MODFLOW
Input
     Hydraulic parameters (e.g., T, Kh, Kv, S  , S)
     Initial & boundary conditions
     Stresses (e.g., recharge, pumpage)

-------
    MODFLOW
Output
     Head (water levels)

-------
              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
       RECHARGE
  Simulates Areal Infiltration
  (e.g., Precipitation, Agricultural Irrigation)
    (Seepage from Lakes, etc.)
INFILTRATION RATE (Units = L/T)
                              """ ff	j

-------
 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

-------
 MODFLOW MODULES

           SOLVER
Method of Solving Simultaneous Linear
Differential Equations
  -  Strongly Implicit Procedure (SIP)
  -  Preconditioned Conjugate-Gradient (PCG2)
  -  Slice-Successive Overtaxation (SSOR)

-------
       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

-------
               SENSITIVITY
HEAD RESIDUALS (FEET) = SIMULATED - MEASURED

ZONE1
     PARAMETER         LAY 1        LAY 2
     < Recharge 10%        -3.00         -0.75
     > Recharge 10%         0.50         1.90

     < Trans x5             5.00         1.30
     > Trans x5            -4.50         -0.50

     < VCONT x2            **           **
     > VCONT x2          -6.70         -2.30

     < Specific Yield x2      -1.20         -0.50
     > Specific Yield x2       0.75         0.50

-------
   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 
-------
•      Press  or select Well.   The screen will show all presently defined map features.  The dark red
                                  plus signs are the points which can be selected for creation of well
                                  elements.   to get see the individual wells of the wellfield

•      Move the mouse cursor toward the plus sign representing the Vincennes wellfield.  The
       feature will be highlighted and its name will appear at the top of the screen.

•      Press the left mouse button.  You will be prompted to select the well position:   "mouse
                                  button 1: well	"

•      Place the cursor in the middle of the points and press the left mouse button. The well element
       will be added as a yellow plus sign. You will then be prompted for the well discharge.

•      Type: "467000".            A total pumping center discharge of 467,000  ft /day. A positive discharge
                                  takes water out of the aquifer.

•      Press .              You will be prompted for the well radius.

•      Type: "1" for 1 ft.           Data can be entered in any units, but they  must be  consistent.   Press
                                  <£nter> again.

•      If an error is made, press  when finished.   You will be prompted for another well feature.

•      Press  to return to the main GAEP menu.

Aquifer Module

The aquifer module is used to add aquifer parameters and a recharge rate to the CZAEM input data
file.  To access the_AQUIFER module, select the aquifer command  in the main GAEP menu
(Figure 11). Enter aquifer data by selecting the parameter in the AQUIFER menu and typing in the
value.  For example, to enter a base elevation of 330 ft, press .  You will be prompted for the
elevation of the  parameter.  Type "330" and press .  GAEP will update the value on the
screen. The units used for the aquifer parameters and the recharge must be consistent.  The length
unit used for the permeability (hydraulic conductivity) must be the same, as that used when digitizing
coordinates (usually feet). The time unit used to define the average areal recharge rate (usually days)
must be consistent with that used to define permeability. This version of GAEP assumes "days" are
used for the unit of time!

Right:        Permeability =100          (feet/day)
              Recharge = 0.000913        (feet/day)
                                            28

-------
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)///]   
                                              [FILE]
                                       
        (NUMBER  OF POINTS)                    
                                          
                                            
                                           
                                                         

Figure 16  CZAEM Main Menu.

Check the Input File

At this point we will inspect the input data file through on-screen graphics. The window size
(domain to be plotted on the screen) will be whatever was set as a User Window in GAEP (see
Appendix A, GAEP reference manual).  We will store the current window, expand the window
to include all elements in the file, and look at this model layout using the following sequence of
commands.

•      Type: "WINDOW PUSH".       This stores the current window settings into the

                                       37

-------
       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]


(NUMBER OF POINTS)




[FILE]






If you type help at the main menu,

help

the following help file will scroll across the screen:
NOTATION:
ANALYTIC ELEMENT
MODULES:
SOLVING:


           (PARAMETERS) [OPTIONAL] {explanation}
          

          These modules allow for the entry of aquifer parameters and
          boundary conditions.

          [TOL]
                        Initiate the process of solving the system of equations. If iteration is
                        necessary, this will be done with a relative accuracy of 0.01; no
                                     78

-------
 (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.

<VIEWPORT>(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.

<WINDOW>[(X1 ,Y1 ,X2,Y2)]/<ALL>]

<WINDOW> 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.

<RESET>

              80
</pre><hr><pre>
-------
                        Resets all values in the program to what they are when the program
                        is loaded into memory.

                        <SWITCH>(FILENAME)

                        Will cause the program to read further input from the file FILE-
                        NAME; the last command in the file must be SWITCH CON.

                        <STOP>

                        Stops the program.
SERVICE MODULES:  <SAVE><READxCURSORxSWITCH><PSET><MAPxFIPLOT>

                        <SAVE>/<READ>

                        To store or retrieve solutions and grids in binary form; to subtract
                        grids.

                        <CURSOR>

                        Data retrieval by means of the cursor. Changing of well and
                        line-sink data.

                        <SWITCH>

                        Input and output re-direction.

                        <PSET>

                        To direct graphics output, and to set plotting attributes.

                        <MAP>

                        To generate a background map and to activate plotting of the
                        background map.

                        <FIPLOT>

                        To record plots in binary form.
                                     81
</pre><hr><pre>
-------
                                                                     AQUIFER
\\\  Module=MAIN MENU           Level=0     Routine=INPUT              777
ENTER COMMAND  WORD FOLLOWED BY ? FOR BRIEF  HELP FROM ANY MENU
<AQUIFER>      <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]   <HELP>
<GIVEN>        <MAP>                                         <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                      <SAVE>  •
<WELL>         <GRID>(NUMBER OF POINTS)                      <READ>
<LINESINK>     <PLOT>                                        <PAUSE>
<SOLVE>        <TRACE>                                       <RESET>
<CHECK>        <CURSOR>                                      <PSET>
                                                             <STOP>
aquifer <Enter>

\\\  Module=AQUIFER             Level=l     RoutIne=INPUT              777
<PERMEABILITY>(PERM)<THICKNESS>(THICK)<BASE>(ELEVATION)<POROSITY>(POROSITY)
<RESETXHELPXRETURN>

This module allows input of aquifer parameters.

The command words are:

<PERMEABILITY>(PERMEABILITY)
                         Specify the hydraulic conductivity in units of L/T.

<THICKNESS>(THICKNESS)
                         Specify the aquifer thickness. If the aquifer is unconfined, choose a
                         value for THICKNESS large enough to ensure that the aquifer is
                         everywhere unconfined. Do not choose unnecessarily large values
                         to prevent loss of accuracy.  Transition from confined to unconfined
                         conditions is automatically taken care of by the program.

<BASE>(ELEVATION)
                         Specify the base of the aquifer. The program will refer all values of
                         head with respect to this base. For example, if ELEVATION equals
                         10, a head of 20 will correspond to a head of 10 with respect to the
                         base of the aquifer system. Unnecessarily large values for base may
                         lead to loss of accuracy.

<POROSITY>(POROSITY)
                         Set the porosity in the aquifer.

                                       82
</pre><hr><pre>
-------
^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>
                         Reset all parameters in the aquifer module to their default values.

<RETURN>
                         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
<AQUIFER>      <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>        <MAP>                                          <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                       <SAVE>
<WELL>         <GRID>(NUMBER OF POINTS)                       <READ>
<LINESINK>     <PLOT>                                         <PAUSE>
<SOLVE>        <TRACE>                                        <RESET>
<CHECK>        <CURSOR>                                       <PSET>
                                                              <STOP>
given <Enter>

\\\  Module=GIVEN               Level=l    RoutIne=INPUT              777
<UNIFLOW> (DISCHARGE) [ANGLE] <RAIN> (X, Y, RADIUS, RATE) <RESETXHELPXRETURN>


This module allows input of the following given functions: uniform flow,
and constant infiltration over a circular region of the aquifer.


The command words are:


<UNIFLOW>(DISCHARGE RATE)[ANGLE OF FLOW IN DEGREES]
                         Specify the discharge  rate and the angle of flow (optional) in
                         degrees between the direction of flow and the x-axis for uniform
                         flow. The discharge rate equals the amount of flow per unit width,


                                       83
</pre><hr><pre>
-------
                        measured over the entire thickness of the aquifer.

<RAIN>(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>
                        Reset all parameters in the module GIVEN to their default values.

<RETURN>
                        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
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>       <MAP>                                          <SWITCH>[FILE]
<REFER£NCE>   <LAYOUT>                                       <SAVE>
<WELL>        <GRID>(NUMBER OF  POINTS)                      <READ>
<LINESINK>    <PLOT>                                         <PAUSE>
<SOLVE>       <TRACE>                                        <RESET>
<CHECK>       <CURSOR>                         .              <PSET>
                                                             <STOP>
ref
(X, ^REFERENCE HEAD)


X,^REFERENCE HEAD : O.OOOOOE+00 O.OOOOOE+00 O.OOOOOE+00
                                      84
</pre><hr><pre>
-------
                                                                         WELL
     Module=MAIN MENU            Level=0    Routine=INPUT              777
ENTER COMMAND WORD FOLLOWED BY  ? FOR BRIEF HELP  FROM ANY MENU
<AQUIFER>      <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>        <MAP>                   .                       <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                       <SAVE>
<WELL>         <GRID> (NUMBER OF  POINTS)                       <READ>
<LINESINK>     <PLOT>                                         <PAUSE>
<SOLVE>        <TRACE>                                        <RESET>
<CHECK>        <CURSOR>                                       <PSET>
                                                              <STOP>
well <Enter>
     Module=WELL                 Level=l    RoutIne=INPUT              777
<GIVENXRESETXHELPXRETURN>

This module allows input of wells. The program currently supports
two types of wells: wells with given discharge and wells with the head
specified at the well boundary.

The command words are:

<GIVEN>
                         Enter wells of given discharge. The user will be prompted for well
                         coordinates and discharge as follows:

(XW,YW,Q)[RADIUS][[LABEL]]<COMMAND>
                         where XW,YW represent the  x and y coordinates of the center of
                         the well, and Q the discharge. RADIUS is the radius of the well and
                         is optional. The radius has a default value of .001 in absolute units.
                         The value of RADIUS is used  in the pathline tracing routine and in
                         the contouring routine. An optional label may be entered between
                         brackets. The program will expect continued entry of discharge
                         specified wells until a command word is entered. You can return to
                         the well input menu by entering COMMAND.

<HEAD>
                         Enter  wells of given head. The user will be  prompted for well
                         coordinates, radius, and head at the well radius as follows:

(XW,YW,HEAD,RADIUS)[[LABEL]]<COMMAND>
                         where XW,YW represent the  x and y coordinates of the center of
                         the well, HEAD represents the head of the well at the well radius,
                         and RADIUS is the radius of the well. An optional label may be

                                       85
</pre><hr><pre>
-------
                          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.

<FACTOR>(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.

<RESET>
                          Cause all  parameters in the module  WELL to be reset to their
                          default values.

<RETURN>
                          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
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<PDP>]   <HELP>
<GIVEN>        <MAP>                                           <SWITCH>[FILE]
<REFERENCE>   <LAYOUT>                                       <SAVE>
<WELL>        <GRID>(NUMBER OF POINTS)                       <READ>
<LINESINK>    <PLOT>                                          <PAUSE>
<SOLVE>        <TRACE>                                         <RESET>
<CHECK>        <CURSOR>                                       <PSET>
                                                               <STOP>
linesink <Enter>

\\\  Moaule=LINE-SINK            Level=l    Routine=INPUT              777
<GIVENXHEADXSTRING> [<ON>7<OFF>] <TOLERANCE> [TOL] <RESETXHELPXRETURN>


This module makes it possible to enter line-sinks. The line-sinks may be used to model narrow
creeks  that may be either above the groundwater table or intersecting it,  as  well as  closed
boundaries.


The line-sinks may be chained together. Either the rate of extraction or the head at the center of
the linesink may be specified. The rate of extraction is positive for removal of water from the
aquifer, and is measured in discharge per unit length of line-sink, i.e., in units of LA2/T. If the
line-sinks are used to model creeks in direct contact with the aquifer, i.e., in cases that the heads
are specified at the centers of the line-sinks, the user should verify, after solving the problem, that
indeed the heads below the linesink are at or above the streambed level. Otherwise, a run should


                                        86
</pre><hr><pre>
-------
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:
<GIVEN>
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]]<COMMAND>
                          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.
<HEAD>
                          Enter line-sinks of constant head. The program prompts for input
                          of coordinates and head as follows:
(XI, Yl, X2,Y2, HEAD)[[label]]<COMMAND>
                          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.
<FCHANGE>[EL.NR.]
<STRING>[<ON>/<OFF>]
<TOLERANCE>[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
</pre><hr><pre>
-------
                         linesink element on the string, then the element will be linked to the
                         closest end node on the string.

<RESET>
                         This command causes all parameters in the linesink module to be
                         reset to their default values.

<RETURN>
                         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
<AQUIFER>      <WINDOW>[ (X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>        <MAP>                                          <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                       <SAVE>
<WELL>         <GRID> (NUMBER OF  POINTS)                       <READ>
<LINESINK>     <PLOT>                                         <PAUSE>
<SOLVE>        <TRACE>                                        <RESET>
<CHECK>        <CURSOR>                                       <PSET>
                                                              <STOP>
check <Enter>
     Module=CHECK               Level=l    Routine=INPUT              777
<AQUIFER><GIVENXREFERENCE><WELLXLINESINK>
<HEAD> (X, Y) <DISCHARGE> (X, Y) <CONTROL><SUMMARYXHELPXRETURN>

The check routine is intended to provide access to data such as well locations and discharges,
aquifer thickness and hydraulic conductivity. It also provides a means to check that the solution
meets the conditions specified at the control points. It is important to note that errors caused by
loss of significant digits can always lead to a solution that is inaccurate at certain control points.
Check gives access to data in each of the modules, all of which have individual check routines and
relevant help screens via the <HELP> command.

The following commands give access to modules:

<REFERENCExAQUIFER><GrVEN><WELL><LINESINK>

Each of the modules has a command <CONTROL> which displays the conditions at the control
points along with the computed values. For example, the command <REFERENCE> followed by
the command <CONTROL> displays the coordinates at the reference point, the value of the head
specified, and the value of the head computed.
                                       88
</pre><hr><pre>
-------
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:

<RANGE>(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:

<CONTROL>
                         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 <HELP> in the appropriate module.

 <HEAD>(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 <INPUT> in the module
                         SWITCH.
<OMEGA>(X,Y)
                         Display the complex potential, i.e. the value of the potential and the

                                       89
</pre><hr><pre>
-------
                         value of the stream function, at point (X,Y).

<DISCHARGE>(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.

<TIME>[TIME]
                         Display the program time in user units and the value.of the time
                         factor TFAC. Entering the optional parameter TIME will cause the
                         time to be set to TIME.

<SUMMARY>
                         Display general information about all of the modules, including the
                         current  and maximum numbers of specific elements.  Press the
                         ENTER key after the summary of each module has been  displayed,
                         in order to display the next summary.
                                                                          GRID

The grid module can be executed from the main command line.
\\\  Module=MAIN  MENU           Level=0     RoutIne=INPUT              777
ENTER COMMAND  WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
<AQUIFER>      <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]   <HELP>
<GIVEN>        <MAP>                                         <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                      <SAVE>
<WELL>         <GRID>(NUMBER OF POINTS)                      <READ>
<LINESINK>     <PLOT>                                        <PAUSE>
<SOLVE>        <TRACE>                                       <RESET>
<CHECK>        <CURSOR>                                      <PSET>
                                                             <STOP>
grid <Enter>

\\\  Module=GRID                 Level=l     Routine=INPUT              777
<TITLE><HEADXPOTENTIALXPSI><LEAKAGE><SURFACE><BOTTOM><GRID>(NX)
<WINDOW> [XI, Yl, X2, Y2] <HELPXWRITEXSWRITEXRETURN>


This module will set the function to be gridded before contouring.
                                       90
</pre><hr><pre>
-------
The command words are:

<HEAD>
                         Set the piezometric head as the function to be used for future GRID
                         commands.
<POTENTIAL>
                         Set the potential as the  function to be used for future GRID
                         commands.

<PSI>
                         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.

<SURFACE>
                         Set the phreatic surface as the function to be used for future GRID
                         commands.

<BOTTOM>
                         Set the aquifer bottom as the function to be used for future GRID.

<WINDOW>(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.

<GRID>(NUMBER OF INCREMENTS)
                         Set the number of increments in the X direction of the grid to be
                         contoured.

<TITLE>
                         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.

<WRITE>
                         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
</pre><hr><pre>
-------
                        package, or other program for display. The program will prompt for
                        the name of the file to be written to.
<RETURN>
                        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
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]   <HELP>
<GIVEN>       <MAP>                                        <SWITCH>[FILE]
<REFERENCE>   <LAYOUT>                                     <SAVE>
<WELL>        <GRID>(NUMBER  OF POINTS)                      <READ>
<LINESINK>    <PLOT>                                        <PAUSE>
<SOLVE>       <TRACE>                                       <RESET>
<CHECK>       <CURSOR>                                     <PSET>
                                                            <STOP>
plot <Enter>
                                                                       TRACE


\\\  Module=MAIN MENU            Level=0    RoutIne=INPUT              777
ENTER COMMAND WORD  FOLLOWED BY ? FOR BRIEF HELP FROM ANY MENU
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]   <HELP>
<GIVEN>       <MAP>                                        <SWITCH>[FILE]
<REFERENCE>   <LAYOUT>                                      <SAVE>
<WELL>        <GRID>(NUMBER OF POINTS)                      <READ>
<LINESINK>    <PLOT>                                        <PAUSE>
<SOLVE>       <TRACE>                                       <RESET>
<CHECK>       <CURSOR>                                      <PSET>
                                                            <STOP>
trace <Enter>

\\\  Module=TRACE;                 Level=l    Routine=INPUT              777
<WINDOW>[(XI,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]<TOL>[TOLERANCE]<OJRSOR>(<ON>7<OFF>)
<SWITCH><SET><PLOTXLAYOUTXCAPZONE><HELP><RETURN>


This module allows tracing of pathlines and it allows access to the capture zone module.

                                      92
</pre><hr><pre>
-------
The command words are:

<SWITCH>
                         Set device for numerical output. If OUTPUT=CON, then numerical
                         data will be printed on graphics screen.
<SET>
                         Provide access to the routine for setting tracing parameters.

<TIME>[TIME]
                         Set the time, given in user units, to TIME. Entering the command
                         without argument will cause the current time to be displayed both
                         in user and program units. Note that TIME must be positive.

<LAYOUT>
                         Layout all elements prior to the appearance of the cursor.

<PLOT>
                         Layout all elements and draw contour plots prior to the appearance
                         of the cursor.

 <CAPZONE>
                         Allow capture zones to be created.

<CURSOR>(<ON>/<OFF>)
                         When entering the portion of the program for  input of starting
                         points of pathlines with CURSOR ON, coordinates will be entered
                         using the cursor.  If the CURSOR is OFF, coordinates must be
                         entered by typing in values.  These values must then precede the
                         command word.  You would set CURSOR OFF, for example, if
                         it is desired to produce the graphical output on the plotter (speci-
                         fied by entering the commands:  PSET and PLOTTER).  This
                         facility also makes it possible to start pathlines at  points with
                         specified coordinates, either typed in or read from a file, using the
                         SWITCH facility.

<WINDOW>[(X 1 ,Y 1 ,X2,Y2)/<ALL>/<PUSH>/<POP>]
                         Sets the window. 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.  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

                                       93
</pre><hr><pre>
-------
<TOLERANCE>[TOL]
<SAVE>
<READ>
 <PSET>
<SWITCH>
<RETURN>
                         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
</pre><hr><pre>
-------
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.
<BASE>
<SURFACE>
<POTENTIAL>
<COORDINATES>
<WLL>
<TRACE>[ELEV]
<TOLERANCE>
                          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.
<WGENERATE>(# LINES)[ELEV]
                                        95
</pre><hr><pre>
-------
                          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.

<BACKWARD>(<ON>/<OFF>)
                          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.

<COMMAND>
                          Causes the command line to be displayed on the screen.

<MENU>
                          Terminate cursor activity while remaining in the module.

<RETURN>
                          Returns control to main menu.
                                                       CAPZONE in <TRACE >


\\\  Module=CAPTURE  ZONE;                   Level=l     RoutIne=INPUT     777
<COORDINATEXBASEXSURFACEXWINDOW>[ (XI, Y1,X2, Y2) 7<ALL>7<PUSH>7<POP>] <WLL>
<SUBZONEXTIMEZONEXSOURCEXNLINE>( LINES )<PAGEXHELPXCOMMANDXRETURN>
<FRONT>[<ON>[VELOCITY FACTOR]7<OFF>]<WGENERATE>(# LINES)<COLOR>[COLORI][2][3]
<BSAVEXBREAD>{TO BACKSPACE, PRESS <  }

This module allows the creation of capture zones. There are two different types of capture zones
which may be created: subzones and time zones.

Subzones are  capture zones which delineate the source of all water which enters a well.  The
capture zone envelope is drawn for this diagram to show what water will enter the well. Dividing
streamlines are also drawn from the well to the capture zone envelope to show how much of the
water comes from different sources.

Time zones are capture zones which delineate how long it takes the water to reach the well.
Contours are drawn which indicate how long it will take  water from different locations in the
aquifer to flow to the well. The capture zone envelope is drawn for this diagram and it is also equal
to the time zone for very large times.  The dividing streamlines are not drawn in this diagram;
however, if it is required, then both time zones and subzones may be drawn for the same well.

                                        96
</pre><hr><pre>
-------
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:

<SUBZONE>
                          Create subzones for the well at which the cursor resides.

<TIMEZONE>
                          Create time zones for the well at which the cursor resides. The user
                          will be queried for the following information:
ENTER  <MINIMUM  TIME>[TIME  STEP][MAXIMUM TIME],  OR
   <R>EDRAW LAST  TIME ZONES,  OR <D>EFAULT TIME  ZONES, OR<E>XIT
   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.

<SOURCE>
                          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.

<NLINE>(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
</pre><hr><pre>
-------
<COORDINATES>
<BASE>
<SURFACE>
<WINDOW>[(X 1 ,
<WLL>
<SWITCH>
<PAGE>
 <COMMAND>
<HELP>
<RETURN>
                         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)/<ALL>/<PUSH>/<POP>]
       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
</pre><hr><pre>
-------
                         Returns control to the tracing routines.

<FRONT>[<ON>[VELOCITYFACTOR]/<OFF>]
                         Specify whether to create timezones for the mean velocity of the
                         contaminant, or for the front of the contaminant. Specifying
                         <OFF> will create timezones for the mean travel time of the
                         contaminant.  Specifying  <ON> 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.

<WGENERATE>(# 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.

<COLOR>[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

<BSAVE>(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.

<BREAD>(FILE)
                         Read and plot the capture zone boundaries which were previously
                         saved to a file via the BSAVE command.

<CSAVE>(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
</pre><hr><pre>
-------
<CREAD>(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
</pre><hr><pre>
-------
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
<AQUIFER>      <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>        <MAP>                                          <SWITCH>[FILE]
<REFERENCE>    <LAYOUT>                                       <SAVE>
<WELL>         <GRID>(NUMBER OF  POINTS)                       <READ>
<LINESINK>     <PLOT>                                         <PAUSE>
<SOLVE>        <TRACE>                                        <RESET>
<CHECK>        <CURSOR>                                       <PSET>
                                                              <STOP>
cur <Enter>

\\\  Module=CURSOR               Level=l    Routine=INPUT              777
<TOLERANCE> (TOL) <SWITCHXPLOT><LAYOUT><HELPXRETURN>


This module allows access to cursor activity.


The command words are:


<SWITCH>
                         Allow setting device for numerical OUTPUT. If OUTPUT=CON,
                         then numerical data will be printed on graphics screen
                                       101
</pre><hr><pre>
-------
<LAYOUT>
                         Display the layout prior to appearance of the cursor.

<PLOT>
                         Display both the layout and contour plots prior to the appearance of
                         the cursor.

<PAGE>
                         Clear both the text and graphics screens.

<START>
                         Will cause the cursor to be displayed.

<TOL>(TOL)
                         Set the tolerance for subsequent MOVE commands (see below) to
                         TOL.

<RETURN>
                         Return to the main menu.
Once in cursor mode, enter any of the following commands.

 <HEAD>
                         Print the head.

<POTENTIAL>
                         Print both the potential and stream function. (Stream function has
                         meaning only if the infiltration rate at position of cursor is zero).

<DISCHARGE>
                         Print the two components of the discharge.

<COORDINATES>
                         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.

<WINDOW>[(X1,Y1,X2,Y2)/<ALL>/<PUSH>/<POP>]
                         Change the window size. When the window size is changed, the
                         capture zones which were in the previous window are redisplayed.

                                      102
</pre><hr><pre>
-------
<WLL>
                          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.
<MENU>
<TOLERANCE>
<SIZE>
<NDISCHARGE>
                          Exit from graphics mode; return to main cursor menu.
                          Set the tolerance for MOVE commands; e.g., commands that will
                          allow the user to modify endpoints of line-sinks. 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.
                          Set the maximum size for the display of discharge vectors to the
                          distance between the previous cursor location and the current one.
                          Make it possible to plot the discharge normal to any given line. The
                          discharge is computed at the midpoint of the line between the
                          current cursor location and the next one, defined by entering a
                          carriage return. A rubberband line is displayed if the command is
                          entered properly. Discharge in this context is the component of the
                          discharge vector normal to the line.  Points and discharges up to a

                                        103
</pre><hr><pre>
-------
<ENDISCHARGE>
<LSMOVE>[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.
<WLMOVE> [DISCHARGE]
                          Use this command to move wells. It works in the same way as
                          LSMOVE above.
<MARK>
 <RETURN>
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
</pre><hr><pre>
-------
                                                                         PSET
\\\  Module=MAIN MENU           Level=0    Routine=INPUT              777
ENTER COMMAND WORD  FOLLOWED BY ? FOR BRIEF HELP FROM ANY  MENU
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>       <MAP>                                         <SWITCH>[FILE]
<REFERENCE>   <LAYOUT>                                       <SAVE>
<WELL>        <GRID>(NUMBER OF POINTS)                      <READ>
<LINESINK>    <PLOT>                                         <PAUSE>
<SOLVE>       <TRACE>                                       <RESET>
<CHECK>       <CURSOR>                                      <PSET>
                                                             <STOP>
pset <Enter>

\\\  ROUTINE SET PLOT MODE                                            777
<PRINTERXSCREENXDRIVERXPALETTE> (NUMBER) <MOUSE> (<ON>7<OFF>) <HELPXRETURN>


This module will set the program for various types of hardware
configurations.
<PALETTE>(CODE)

<PRINTER>



<SCREEN>

<DRTVER>

<RETURN>
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
</pre><hr><pre>
-------
                                                                       STOP
\\\  Module=MAIN MENU           Level=0    Routine=INPUT              777
ENTER COMMAND WORD FOLLOWED BY ? FOR BRIEF HELP FROM ANY  MENU
<AQUIFER>     <WINDOW>[(X1,Y1,X2,Y2)7<ALL>7<PUSH>7<POP>]    <HELP>
<GIVEN>       <MAP>                                        <SWITCH>[FILE]
<REFERENCE>   <LAYOUT>                                     <SAVE>
<WELL>        <GRID>(NUMBER OF POINTS)                      <READ>
<LINESINK>    <PLOT>                                       <PAUSE>
<SOLVE>       <TRACE>                                      <RESET>
<CHECK>       <CURSOR>                                     <PSET>
                                                           <STOP>
stop <Enter>
                                     106
</pre><hr><pre>
-------
                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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
       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 <F2> "CommTest" command (see the
       "Program TABTEST" section of this appendix for details).

•      Once the communications test is successful, use TABTEST's <F3> "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 <ESC> 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
</pre><hr><pre>
-------
                                                             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 <PORT> <X IN> <Y IN> <# BUTTONS>

where:

•    <PORT>            Is the serial port used. Only COM1 and COM2 are supported.
•    <XIN>             Is the size of the digitizer in the X direction, in inches.
•     <YIN>             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 <CR>",
      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
</pre><hr><pre>
-------
                                       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 <PORT> <X IN> <Y IN> <#BUTTONS> <LPI>

where:

      <PORT>            Is the serial port used. Only COM1 and COM2 are supported.
      <X IN>             Is the size of the digitizer in the X direction, in inches.
      <Y IN>             Is the size of the digitizer in the Y direction, in inches.
      <# BUTTONS>      Is the number of buttons on the puck.
      <LPI>              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
</pre><hr><pre>
-------
SummaGraphics MM ASCII Protocol

To configure the SummaGraphics MM ASCII driver, the TABLET environmental variable should
be set as follows:

SET TABLET=MMASCD <PORT> <X IN> <Y IN> <# BUTTONS> <LPI>

where:

      <PORT>            Is the serial port used.  Only COM1 and COM2 are supported.
      <X INCHES>       Is the size of the digitizer in the X direction, in inches.
      <Y INCHES>       Is the size of the digitizer in the Y direction, in inches.
      <# BUTTONS>      Is the number of buttons on the puck.
      <LPI>              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 <PORT> <X IN> <Y IN> <# BUTTONS> <LPI>

where:

•     <PORT>           Is the serial port used. Only COM1 and COM2 are supported.
•     <X IN>             Is the size of the digitizer in the X direction, in inches.
•     <YIN>             Is the size of the digitizer in the Y direction, in inches.
•     <# BUTTONS>      Is the number of buttons on the puck

                                       112
</pre><hr><pre>
-------
•      <LPI>              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
</pre><hr><pre>
-------
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 <MAXIMUM X> <MAXIMUM Y>

where:

•      <xinches>    Is the size of the digitizer in the X direction, in inches.
•      <y 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
</pre><hr><pre>
-------
              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  <F1>

•     Displays a help screen

CommTest  <F2>

•     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 <ESC> 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
</pre><hr><pre>
-------
       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 <ESO to return to the TABTEST menu.

•      During the communications test you may need to experiment with different drivers and the
       various communications parameters.

                                   Note: Common Blunders

The author has typically made two major blunders at this step in testing, both of which seem trivial. First, make surahat
you select the correct communications port (COM1 or COM2), and second, make sure that the digitizer puck is on the
active digitizing surface. The driver uses "RUN" mode to read coordinates, and points are only transmitted by most
digitizers when the puck is on the digitizer.

DriverTest <F3>

•      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      <F6>

•      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    <F7>

•      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
</pre><hr><pre>
-------
Baud, etc. <F8>

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    <ESC>

•      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
</pre><hr><pre>
-------
                                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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
            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
</pre><hr><pre>
-------
                                                       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
</pre><hr><pre>
-------
                                        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.
</pre><hr><pre>
-------
                                      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
</pre><hr><pre>
-------
                                      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
</pre><hr><pre>
-------
                               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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
                                        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
</pre><hr><pre>
-------
                                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
</pre><hr><pre>
-------
                                      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.
</pre><hr><pre>
-------
  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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
                             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
                       <t>3 = 200.5 m                   <t>4 = 200.0 m
                            Figure 1.1   Site map for Example 1.

                                             4
</pre><hr><pre>
-------
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
<W

                                    Well radius = 0.15 m
                         Figure 1.2   Conceptual model of the aquifer.


Entering the program CZAEM.
     Change to the CZAEM directory, which contains the executable file czaem.exe.

     A:\ >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
     <AQUIFER>     <VINDOW>[(X1,Y1,X2.Y2)/<ALL>/<PUSH>/<POP>]   <HELP>
     <GIVEN>       <MAP>   ,                                 <SWITCH>[FILE]
     <REFERENCE>    <LAYOUT>                                 <SAVE>
     <WELL>        <GRID>(NUMBER OF FOOTS)                   <READ>
     <LIHESINK>    <PLOT>                                   <PAUSE>
     <SOLVE>       <TRACE>                                  <RESET>
     <CHECK>       <CURSOR>                                 <PSET>
                                                          <STOP>

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
</pre><hr><pre>
-------
 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 <RETURN> 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 <AQUIFER>. Note that all data must be input with consistent units.  Enter

     aquifer

 CZAEM responds with

     \\\ Nodule-AQUIFER            Level-1    Routine-INPUT           ///
     <PERMEABILITY> (PERM) <THICKNESS> (THICK) <BASE> (ELEVATION) <POROSITY> (POROSITY)
     <RESET><HELP><RETURN>

 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 <THICKNESS> requires  as the argument the actual vertical extent of the
 aquifer (Figure 1.2). <BASE> 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
     <AQUIFER>     <WINDOW>[(X1.Y1.X2,Y2)/<ALL>/<PUSH>/<POP>]   <HELP>
     <GIVEN>      <MAP>                                    <SWITCH>[FILE]
     <REFERENCE>   <LAYOUT>                                 <SAVE>
     <VELL>       <GRID>(NUMBER OF POINTS)                   <READ>
     <LINESINK>    <PLOT>                                    <PAUSE>
     <SOLVE>      <TRACE>                                  <RESET>        '
     <CHECK>      <CURSOR>                                 <PSET> ..
                                                          <STOP>

Next enter the module GIVEN

     given

CZAEM responds with

     \\\  Hodule-CIVEN              Level«l    Routine-INPUT            ///
     <UNIFLOW> (DISCHARGE) [ANGLE] <RAINXX, Y.RADIUS,RATE) <RESETXHELPXRETURN>
</pre><hr><pre>
-------
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:
</pre><hr><pre>
-------
     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 .
</pre><hr><pre>
-------
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
     <D>EFAULT DTOMBER OF LEVELS] <L>AYOUT
     OUH  LEVEL [IHCREMENT (X)}] DUX LEVEL]
     (MAX  LEVEL [DECREMENT {<OU DOB LEVEL]
     «N.  LEVEL-   1.982092E+02 MAX. LEVEL-    2.022123E+02

At this time, we are not interested in specific contours, so we use the default option and ask for
twenty levels to be plotted between the extreme values. Enter
       d 20
CZAEM responds with
     START LEVEL   1.983000E+02 IHCREMEHT   2.000000E-01  PRESS ENTER
This shows the value of head at which contouring starts and the contour interval.  When using
default, the contours will be drawn from lower to higher head due to the positive increment. Press
[enter] to view the flow field.
       [enter]
                       Figure 1.3  Contour plot of the phreatic surface.

    The plot has straight contours as we move in the direction of now. The results reflect uniform
flow.  Press [enter] to return to the MAIN command line after viewing the model of existing flow
conditions.
</pre><hr><pre>
-------
 Entering and analyzing the proposed well.
     Enter the module WELL from MAIN

     Mil

 CZAEM responds with

     \\\  Hodnle-VELL               Level«l   Rootine-Mm            ///
     <GIVEIXBESETXHELPXRETURI>

     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]]<COMMAKD>

 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            ///
     <WINDOW> [ (XI. Yl. X2. Y2)/<ALL>/<PUSH>/<POP>] <TOL> [TOLERANCE] <CURSOR> (<ON>/<OFF>)
     <SWITCH><SET><PLOT><LAYOUT><CAPZO)IE><HELP><RETURH>

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
</pre><hr><pre>
-------
                   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            ///
    <CURVE><POINT><PLOT>(ON/OFF) <RESET><HELPXRETURN>

To view the map on the screen with each plot, set the <PLOT> option on

      plot on

We draw the crop boundary using the command <CURVE>. 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
</pre><hr><pre>
-------



                                          :  ?~--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
</pre><hr><pre>
-------
                                   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
</pre><hr><pre>
-------
    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           ///
     <GIVEN><HEADXSTRING> [<OH>/<OFF>] <TOLERANCE> [TOL] <RESETXHELP><RETURN>

Line-sinks with known strengths are entered through the command <GIVEN>, and those with
known heads through the command <HEAD>. 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]]<COHKAND>

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]]<COMMAND>

Although the command line is not displayed, it  is still active,  and we  may at any  time  enter
<HEAD>  to begin  a  head specified line-sink,  <COMMAND> to see the command line, or
<RETURN>  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
</pre><hr><pre>
-------
         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 <LAYOUT>. 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           ///
    <AQU1FER><GIVEN><REFERENCE><WELLXLINESINK>
    <HEAD> (X. Y) <DISCHARGE>(X, Y) <CONTROL><SUMMARY><HELP><RETUW(>

This is the CHECK module menu. Enter
    head 1  1

CZAEM responds with

                                             15
</pre><hr><pre>
-------
                  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
</pre><hr><pre>
-------
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 <SAVE> 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»<BOTBXRET(mi>

To select SOLUTION, type

     Ml

CZAEM will request a file name

     <FILEMMEXR>{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
</pre><hr><pre>
-------
         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  <READ>.
From the MAIN command line, enter

     read

CZAEM responds with

    <SOLUTIOH><GRID><BOTH><DIFCRIDXRETORN>

                                            18
</pre><hr><pre>
-------
             Figure 2.5  Contours with the well present, reference point at (1,1).
Enter
     solution

CZAEM responds with
     PLEASE ENTER FILENAME; <R> 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 ON>.  Backward tracing from the well is achieved by the command

                                            19
</pre><hr><pre>
-------
         Figure 2.6  Contours with the well present, reference point at (—2000,5000).

<WGENERATE>. Note that to return to forward tracing one must type <BACKWARD OFF>.
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            ///
    <BASEXSURFACE><COORDINATEXTRACE> [ELEVATION] <BACKWAH» [<ON/OFF>]
    <WGENERATE>(» LINES) [ELEVATION] <WLLXTOL><COMMAND><MENU><RETURN>
    (TO BACKSPACE. PRESS  < }
The cursor appears at  the center of the screen (in this case directly over the well).
    With the cursor positioned at the well, first switch backward tracing on
    backward on
Now the command <WGENERATE>  may be used. This command has one required parameter
and one optional parameter. The required parameter is the number of pathlines generated from the
well and the optional parameter  is the vertical elevation in the well from which they will originate.
The default elevation is the bottom of the aquifer.  We choose the  number of traces to be 20 by
typing
      ugen 20
You will see pathlines  starting from the well and going back toward their original source (Figure
2.7).  The well is  seen to be drawing  some of its water from the  river.   We conclude that the

                                            20
</pre><hr><pre>
-------
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 <CAPZONE> 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 <CAPZONE>  from
the outset to evaluate each well pumping level. For  larger problems,  it may  be more efficient to
use <WGEN> or <TRACE> 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
</pre><hr><pre>
-------
 The binary file with all elements entered in Example 2 and the parameter values determined by
 <SOLVE> has been read in. Enter

     grid 60
       trace

 CZAEM responds with

     \\\  Modul«-TRACE;                Uval«l    Routin«-I«PUT           ///
     <VXHDOV> t (XI , Tl , X2 . T2) /<ALL>/<POSH>/<POP>] <TOL> [TOLERANCE] <CDRSOR> (<OH>/<OFF>)
     <SVITCH><SET><PLOT><UTODT><CAPZOMEXHELPXRErDRR>

 We enter the module CAPZONE from within TRACE and type

         CUkZOlM

 CZAEM responds with
     <D>EFAOLT DnmBER OF LEVELS] <L>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 <D>, or just get a layout <L>. 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    ///
     <COORDINATE><BASE><SURFACE><WINDOW> [(XI ,Y1 ,X2.Y2)/<ALL>/<PUSH>/<POP>] <VLL>
     <SUBZONE><TIMEZONE><SOURCE><NLINE>(LINES)<PAGE><HELP><COKNAND><RETDRN>
     <FRONT>[<ON> [VELOCITY FACTOR] /<OFF>] <VGEKERATE>(t LINESXCOLOR> [COLOR1] [2] [3]
     <BSAVE> (FILE) <BREAD> (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
<SUBZONE>.  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
</pre><hr><pre>
-------
           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
</pre><hr><pre>
-------
 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 <SOURCE>
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 <RESET> 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
</pre><hr><pre>
-------
          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
</pre><hr><pre>
-------
      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 <SOURCE>
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
     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
</pre><hr><pre>
-------
                            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
                           <h - 128.29 m
                            (-542,750)
                                     (-500,568)
         (-800,-200)
(-300,0)
                                                      „   Proposed Well
                                                      \  Q = 220 m3/<Uy
                                                      1+ • (100,20)
                                                                    Monitoring Well 3
                                                                      <h = 126.58 m
                                                                           •
                                                                       (500, -500)
         Monitoring Well 1   -^   -
           4i = 129.84 m      ^  ./i
                        •    V<'
              (-750.-875)       (-583,-891)

                           Figure 4.1  Site map for Example 4-

    An approach for testing the pumpout system design might include:
 1. Modeling the existing local ground-water flow as uniform, based on monitoring well informa-
    tion.
                                           29
</pre><hr><pre>
-------
  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
</pre><hr><pre>
-------
     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            ///
     <AQUIFER><GIVEN><REFERENCE><WELL><LINESINK>
     <HEAD> (X, Y) <DISCHARGE> (X. Y) <CONTROL><SUMMARY><HELP><RETURM>


                                             31
</pre><hr><pre>
-------
     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 <AQUIFER>,
 <GIVEN>, <CONTROL>, <WELL>, and <LINESINK> 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
</pre><hr><pre>
-------
     •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 <POINT> 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
<CURVE>.

    map
      plot  on
      point
       -750 -875
       -542 750
        500 -500
      curve
       -500 568
       -300   0
       -583 -891
       -800 -200
       -500 568


                                             33
</pre><hr><pre>
-------
      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 <TRACE> in the module TRACE to draw forward pathlines from the plume
boundary to the well. A second approach is to use <WGENENERATE> 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 <SUBZONE> 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
</pre><hr><pre>
-------
                            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 <SUBZONE>, 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
</pre><hr><pre>
-------
                         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 <TIMEZONE>
    time

CZAEM responds with
    CALCULATING SUBZONES PHASE 1:   CREATING INITIAL PATHLIKES FROM THE WELL
       10 	
                                           36
</pre><hr><pre>
-------
                          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 <HINIMUH TIME>[TIME STEP][MAXIMUM TIME], OR
     <R>EDRAW LAST TINE ZONES, OR <D>EFAULT TIME ZONES. OR<E>XIT
     MINIMUM AND MAXIMUM TIMES FOR CAPTURE ZONE:  O.OOOOOE+00  4.78038E+04

     You may enter  <D>  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
</pre><hr><pre>
-------
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 <TIMEZONE> 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 <WLMOVE> 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
</pre><hr><pre>
-------
     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 <WLMOVE>

                                             39
</pre><hr><pre>
-------
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 <WLMOVE> allows the user to change
the well location and discharge simultaneously simply by entering the new discharge following
<WLMOVE>. 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
</pre><hr><pre>
-------
      •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> [<IHm/OUTPUT/READ/SAVE/HELP> (PREFIX)]
    <IHPUT/OUTPUT/MESSACES/ERMR> [<ECHO ON/OFF/APPEW»] (FILE NAME) [LOGICAL IWIT]
    <LOG ON/OFF> [FILE HAKE][LOGICAL OTIT)
    <CALL> (FILENAME) <BACKXBELPXRETURN>

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 <GRID> 50 and plot. Notice that  the results are the same as in Example 3 where the
data were entered manually.  After viewing, enter <SAVE>; save the current grid by typing
    save
      grid

CZAEM responds with
    <FILENAHE><R>{to abort}

Enter the filename
    BOll.gSO

                                           41
</pre><hr><pre>
-------
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 <SWITCH> command.
    switch eiample5.dat

    A command present in the data file which has not yet been explained is the <STRING ON>
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 <STRING ON> command is used
to link line-sink segments together which will then act as a single source in subzone computations.

Comparing grids.
    Enter <Grid> 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
    <SOLUTIOH><GRID><BOTH><DIFGRID><RETUW)>

                                           42
</pre><hr><pre>
-------
                    Figure 5.1  Case of Example 3 with refined line-sinks.

enter
     difgrid
CZAEM responds with
     PLEASE ENTER FILENAME;  <R> 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
     <D>EFAULT [NUMBER OF LEVELS] <L>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
</pre><hr><pre>
-------
             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
</pre><hr><pre>
-------
     •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
</pre><hr><pre>
-------
      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. <RAIN>
is contained in the module GIVEN. The <RAIN> command for the present model may be found
in data file exist.dat. Note that <RAIN> 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
</pre><hr><pre>
-------
                   Figure  6.2   Model of proposed conditions (plot; 780 5).

    Within module CAPZONE move the cursor to the northernmost well and enter <SUB>.
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 <WIN PUSH> command. Enter
    win push

The <WIN PUSH> command stores the current window in a stack; the <WIN POP> command
recalls the last window which has been stored in the stack.  Now, reduce the window size with the
<WLL> 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
</pre><hr><pre>
-------
 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 <SUB> command, enter
     bsave
 The <BSAVE> command opens a file which stores all computed subzone and time zone boundaries.
 CZAEM responds with
     <FILEHAME><R>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 <PAGE> command. Leaving CAPZONE closes the file; entering <PAGE> erases the file.
 The saved file may be recalled by the <BREAD> command. Now, move the cursor to the well
 and create the subzone with the <SUB> 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 <WLL> to create a smaller window around the remaining four wells,
 use the <WINDOW> 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 <WIN ALL> 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 <BSAVE> was entered before creating  any subzones, but we have not yet needed
 <BREAD>. Computed boundaries will remain on screen until the  module CAPZONE is left or
 the <PAGE>  command is used.  To demonstrate the use of <BREAD> 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
    <FILEHAME><R>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
</pre><hr><pre>
-------
                            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
</pre><hr><pre>
-------
 CZAEM responds with the command line

     \\\ ROUTINE SET PLOT NODE                                     ///
     <PRINTER><SCREEH><DRIVER><PALETTE> (NUMBER) <MOUSE> (<ON>/<OFF>) <HELP><RETORN>

 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 <PALETTE> followed by
the number 1, 2, 3, or 4 results in different combinations of line colors in graphical output.  The


                                            50
</pre><hr><pre>
-------
command <MOUSE> 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 <DRJVER> is explained in the ASCII file read.me in the CZAEM directory.
                                          51
</pre><hr><pre>
-------
                                    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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
  Solute
Transport
Modeling
</pre><hr><pre>
-------
9.   Solute Transport Modeling
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
       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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
 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.
</pre><hr><pre>
-------
 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

</pre><hr><pre>
-------
                                           (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.
</pre><hr><pre>
-------
                                  (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.
</pre><hr><pre>
-------
   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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
                           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
^<ft-i
5 10
10





—


r

— -••

r



-i
	 i • • •

ft A /A


A *»i '^^




••••••"•••••••
o




10°
1 ""1 ' ' ' '""I • • • "•"! 	 ••
4f
•i
•i
•I
A •'
A ::
f |i
i
o : •
OA 6 j . *
... A_o_ 	 •....Jl.^ 	

• 0




101 102 103 1(


'





• -

^B$+mmJL±m
O
•

"

)4 1
















o5
                                                   Scale (m)
Fig.  6.  Ratio of longiiudinal to horizontal and vertical transverse dispersivities; largest symbols are high reliability
and smallest symbols are low  reliability. Vertical dashed lines connecting two points indicate sites where all three
principal components of the dispersivity tensor have been measured. Horizontal dashed line indicates a ratio ofAJAj
m  1/3, which has been widely  used in numerical simulations.
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------










































Compound
diethyl phthalate
2-nonanol
bromodichloromethane
3-methylbenzoic acid
trichloroethene
1,1,1-trichIoroethane
di-/j-butyf phthalate
1 , 1 -dichloroethene
carbon tetrachloride
2-butanone (methylethylketone)
4-methyIbenzo?c acid
toluene
tetrachloroethylene
chlorobenzene
1 ,2-dichIorobenzene
o-xylene
1 ,2,2-trifluoro-1 , 1 ,2-trichIoroethane
styrene
1 ,3-dichIorobenzene
fluorotrichloromethane
4, 6-di n itro-2-methy (phenol
p-xylene
m-xylene
1 ,4-dichlorobenzene
ethyl benzene
pentachlorophenol
fV-nitrosodiphenylamine
3,5-dimethylphenoI
BHC-delta
2, 6-dimethylphenoI
1 ,2,4-trichIorobenzene
naphthalene
4-ethylphenol
dibenzofuran
hexachloroe thane
acenaphthene
tri-N-propylamine
BHC-alpha
BHC-beta
hexachlorobenzene
hexach lorobutadiene
Solubility
(ppm)
1000.
1000.
900.
850.
1100.
, 700.
400.
400.
800.
353.
340.
500.
200.
' 448.
148.
170.

162.
118.
110.

156.
146.
79.
150.
14.
35.1

31.5

30. .
31.7

10.
8.
7.4

8.5
2.7
0.035
3.2
««
123
123
131
136
152
155
217
217
232
235
240
242
303
318
343
363
372
380
463
476
477
552
588
594
622
900
982
1038
1052
1060
1080
1300
1986
2140
2450
2580
2610
2627
3619
3910
4330
Mobility
Class
high
high
high
high
„ moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
moderate
low
low
low
low
low
low
low
low
low
low
low
low
slight
slight
slight
slight
slight
slight
slight
slight
</pre><hr><pre>
-------
Compound
di-n-octy! phthalate
butyl benzyl phthalate
fluorene
2-methylnaphthaIene
bis(2-ethylhexyl)phthalate
toxaphene
heptachlor epoxide
endosulfan II
fluoranthene
1,2-diphenyIhydrazene (as azobenzene)
endosulfan suifate
phenanthrene
•• tj .
dieldnn
anthracene
BHC-gamma
decanoic acid
chlordane
pyrene
rt^*r> 4 ^^ M
PCB-1254
heptachlor
• •
endnn
benzo(a)anthracene
i • •
aldnn
4,4'-DDE
4,4'-DDT
4,4'-DDD
benzo(a)pyrene
PCB-1260 -
chrysene
benzo(fa)fluoranthene
benzoMfluoranthene
Solubility
(ppm)
3.
2.9
1.98
25.4
0.6
0.4
0.35
0.28
0.275
0.252
0.22
1.29
0.188
0.073
0.15
0.056
0.135
0.042
0.03
0.024
0.014
0.013
0.01
0.0017
0.005
0.0038
0.0027
0.022
*oc
4510
4606
5835
8500
12200
15700
17087
19623
19800
20947
22788
23000
25120
26000
28900
39610
53200
63400
63914
78400
90000
125719
132000
155000
238000
238000
282185
349462
420108
1148497
2020971
Mobility!
* Qass
slight
slight
slight
slight
slight
slight
slight
slight
slight
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
immobile
Source: R. A. Griffin, 198S. personal communication, and W.
in water-saturated soil materials." Environment*! Ceo/ogy and
R. Roy. and R. A. Griffin. "Mobility of organic solvents
Water Sconces. 7 (1985)341-47.
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
Nonlinear Sorption
Freundlich isotherm
Lanamuir isotherm
                       Cs   =   kdCb                   (15)
                              kC
                                                    (16)
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
Analytical Solution
  c(x, t)  =  —2
                      Rx - vt
Rx + vt
                   erf(z)
                   erfc(z)   =   1  - erf (z)
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
Effect of Retardation at 10.0 Meters from the Source
              0
234
Time (d)
                             ?-**
</pre><hr><pre>
-------
Effect of Retardation on Pulse Sources
Pulse duration = 1.0 days
               0
1       2       3
        Time (d)
4
</pre><hr><pre>
-------
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).
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
                    t
                /MW3  "

                                         MW1S

               I           ©    MW*4V *
                                 "IW1*
               ,MW12    MW6
               I •    •                I
                      MW13   »MW7   y


                 \            ©       '
                            MWS  x^
             *MW9        *MW10   ®MW11
 DITCH   ••
                       Scale
                                    200       400 ft
         KEY                                  j

          •      monitoring well                I
                                              i

          •      excavated storage tank


        — •—.___   contaminant  plume boundary



                 monitoring well  later
         ©
                utilized in recovery system
Fig. 1. Monitoring well configuration and contaminant
plume delineation at 0.5 year; in monitoring period.
</pre><hr><pre>
-------
Parameter
Transmissivity
Saturated thickness
Hydraulic conductivity
Storage coefficient
Effective porosity
Average hydraulic gradient
Seepage velocity
Value
2220 ft2/d
25 feet
1.02 x 10-3ft/s
0.0075
0.3 (estimated)
0.002 (estimated)
200 ft/y
Chemical
trichloroethylene
1,1,1-trichloroethane
1 , 1 -dichloroethylene
1 ,2-transdichloroethylene
Measured Concentration
ug/l
1383
322
294
305
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------

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.
</pre><hr><pre>
-------
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"
</pre><hr><pre>
-------
  •-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
</pre><hr><pre>
-------
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%
</pre><hr><pre>
-------
     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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
</pre><hr><pre>
-------
Vadose Zone
   Flow
</pre><hr><pre>
-------
10.  Vadose Zone Flow
</pre><hr><pre>
-------
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
                           /£>-/
</pre><hr><pre>
-------
Vadose Zone Hydrostatics
(atm)
           V
                                         Negative
                                         Hydrostatic
                                         Pressure
                h
                                           Positive
                                            Hydrostatic
                                             Pressure
</pre><hr><pre>
-------
Capillary Fringe
  'w
</pre><hr><pre>
-------
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, <li. The effect of hysteresis is included for
wetting and drying cycles.
</pre><hr><pre>
-------
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)
                            /€>-<>
</pre><hr><pre>
-------
                    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.
</pre><hr><pre>
-------
NAPLs
</pre><hr><pre>
-------
11.  Introduction to Nonaqueous
    Phase Liquids (NAPLs)
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
Interfacial Tension
     Differences in  molecular forces of  attraction cause there  to be
     interfacial tensions between the phases.
       Gas
   / / / /
                            Solid
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
                    V
hc= Capillary Rise (Capillary Head)
r - Tube Radius
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
Land Surfac*
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
                                    6 toll
                                      water and oil
Figure 3.9.  The  effect  of water table fluctuations  on pollutant
             distributions (Schwille, 1967).
</pre><hr><pre>
-------
       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)
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
Saturation  and  Concentration
Definitions
Saturation, S
= fraction of the pore space occupied by a fluid



0<S<1.0



= (volume fluid) / (Void Volume)
Porosity, T)
= (Void Volume) / (Total Volume)
Bulk Density, pb      = (Mass Solids) / (Total Volume)
Solids Density,ps      = (Mass Solids) / (Volume Solids)
                               P, (1  - T|)
                                       (25)
Fluid Density, PJ       = (Mass Fluid i) / (Volume Fluid i)



                    i = water, oil, air
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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')
</pre><hr><pre>
-------
                                                                          RJC  #13
                    Partitioning Characteristics for Selected Chemicals
                                   (from Mercer et al, 1990)
   Acetone
   Aldrin
   Atrazine
   Benzene
   Bis-(2-ethylhexyl)pmhaiaie
   Chlordane
   Chlorobenzene
   Chloroe thane
   DDT
   Diazinon
   Dibutyl phthalate
  1,1-Dichloroethane
  1,2-Dichloroethane
  1,1-Dichloroethene
  1,2-Dichloroethene (trans)
  Dieldrin
  Ethyl benzene
  Methylene chloride
  Methyl parathion
  Naphthalene
  Paxathion
. Phenol
 Tetrachloroethene (PERC)
                i
 Toluene
 Toxaphene
 1,1,1-Trichloroe thane
 Trichloroethene (TCE)
 TrichJoromethane (Chloroform)
 Vinyl chloride
o-XyJene
Water
Solubility
(mg/L)
infinite
1.80E-01
3.30E+01
1.75E*03
2.b5t-Oi
5.60E-01
4.66E+02
5.74E+03
S.OOE-03
4.00E+01
130E401
5.50E+Q3
8.52E+03
2^5E+03
6.30E+03
1.9SE-01
1.52E+02
2.00E+04
6.00E+01
3.17E+01
2.40E+01
9.30E+04 .
1.50E+02
SJ5E-f02
S.OOE-01
1.50E-K)3
1.10E+03
8.20E+03
2.67E+03
1.75E-I-02
Vapor
Pressure
(atm)
3^5E-01
7.S9E-09
L84E-09
L25E-01
^.O^C-iw
L32E-08
1.54E-02
1J2E+00
7.24E-09
i.Siii-u:
L32E-08
2J9E-01
&42E-02
7.89E-01
4J26E-01
2.34E-10
9.00E-03
4.76E-01
L28E-08
3.03E-04
4.97E-08
4.49E-04
230E-02
3.70E-02
5.26E-04
1.62E-01
7.60E-02
1.99E-01
3.50E+00
9.00E-03
Henry's Law
KH
(atmon3/mol)
2JD6E-QS
1.60E-05
2^9E-13
S^9E-03
i.6lH-CT
9.63E-06
3^2E-03
6.15E-04
5.13E-04
i.-iOE-Oo
2^2E-07
4J1E-03
9.78E-04
3.40E-02
6.56E-03
4.58E^)7
6.43E-03
2.03E-Q3
5.59E-08
1.15E-03
6.04E-07
4.54E-07
2J9E-02
6J7E-03
. 436E-01
1.44E-02
9.10E-03
2.87E-03
8.19E-02
5.10E-03 '
Organic Carbon
Koc
(ml/g)
2JOE-KX)
9.60E+C4
L63E-f€2
830E+01
5.5CZ-rC3
L40E+05
330E4C2
l^OE+01
2.43E+05
S30E+01
L70E-KJ5
3JWE+01
L40E-KJ1
&50E+C1
S50E401
1.70E+03
1.10E+03
8^0E-K)0
SJOE+03
130E-KJ3
1^)7E+04
1.42E+01
3.64E402
3.00E+02
9.64E402
152E+02
1^6E-KJ2
4.70E+01
5.70E+01
S30E+02
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
        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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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).
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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]
</pre><hr><pre>
-------
Estimation of the mass transfer rate coefficient, K,

Empirical fitting of data

Relations between the

     Sherwood  Number
     Reynolds Number
     NAPL saturation
     Others
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
HSSM
</pre><hr><pre>
-------
12.  The Hydrocarbon Spill Screening
    Model (HSSM)
</pre><hr><pre>
-------
  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.
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
                                                Land Surf ace
Figure 1  Schematic view of NAPL release
</pre><hr><pre>
-------
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 ?
</pre><hr><pre>
-------
                                                      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
</pre><hr><pre>
-------
                      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
</pre><hr><pre>
-------
                        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
</pre><hr><pre>
-------
                              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
</pre><hr><pre>
-------
                          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.
</pre><hr><pre>
-------
                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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
                                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
</pre><hr><pre>
-------
                   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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
                  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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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]
</pre><hr><pre>
-------
                                           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]
</pre><hr><pre>
-------
                                 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]
</pre><hr><pre>
-------
                                         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]
</pre><hr><pre>
-------
            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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
Average Soil Properties determined from Brakensiek et al. (1981)
Soil Texture Class
(number of samples)
Sand (19)
Loamy Sand (69)
Sandy Loam (166)
Loam (83)
Silt Loam (199)
Sand Clay Loam (129)
Clay Loam (112)
Silty Clay Loam (175)
Silty Clay (26)
Clay (108)
X
0.573
0.460
0.398
0.258
0.216
0.368
0.283
0.178
0.212
0.214
hce
(cm)
35.3
15.9
29.2
50.9
69.6
46.3
42.3
57.8
41.7
64.0
11
(porosity)
0.349
0.410
0.423
0.452
0.484
0.406
0.476
0.473
0.476
0.475
®wr
0.017
0.024
0.048
0.034
0.018
0.075
0.087
0.054
0.085
0.106
</pre><hr><pre>
-------
Table IV 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
Saturated Water Content 9S
sample
size
400
364
735
315
82
1093
374
641
246
46
214
1183
mean
0.38
0.41
0.43
0.41
0.46
0.45
0.36
0.43
0.43
0.38
0.39
0.41
standard
deviation
0.09
0.09
0.10
0.09
0.11
0.08
0.07
0.07
0.06
0.05
0.07
0.09
Residual Water Content 9r
sample
size
353
363
735
315
82
1093
371
641
246
46
214
1183
mean
0.068
0.095
0.078
0.057
0.034
0.067
0.070
0.089
0.045
0.100
0.100
0.065
standard
deviation
0.034
0.010
0.013
0.015
0.010
0.015
0.023
0.009
0.010
0.013
0.006
0.017
</pre><hr><pre>
-------
Table V 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
n
sample size
400
364
735
315
82
1093
374
641
246
46
214
1183
mean
1.09
1.31
1.56
2.28
1.37
1.41
1.09
1.23
2.68
1.23
1.48
1.89
standard
deviation
0.09
0.09
0.11
0.27
0.05
0.12
0.06
0.06
0.29
0.10
0.13
0.17
a, (nT1 )
sample size
400
363
735
315
82
1093
126
641
246
46
214
1183
mean
0.80
1.9
3.6
12.4
1.6
2.0
.50
1.0
14.5
2.7
5.9
7.5
standard
deviation
1.2
1.5
2.1
4.3
0.70
1.2
0.50
0.60
2.9
1.7
3.8
3.7
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
   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
</pre><hr><pre>
-------
           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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
              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
</pre><hr><pre>
-------
              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
</pre><hr><pre>
-------
                  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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
       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
</pre><hr><pre>
-------
        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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
       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.
</pre><hr><pre>
-------
                     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.
</pre><hr><pre>
-------
 [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.
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
                  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.
</pre><hr><pre>
-------
 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.
</pre><hr><pre>
-------
  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.
</pre><hr><pre>
-------
 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.
</pre><hr><pre>
-------
      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.
</pre><hr><pre>
-------
                           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.
</pre><hr><pre>
-------
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.
</pre><hr><pre>
-------
 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.
</pre><hr><pre>
-------
  [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</l Annual Conference on Hazardous Waste Research, May 23-24,
     Kansas State University, Manhattan, Kansas, 11-19.

  [7] C. E. Hubbard, J. F. Barker, S. F.  O'Hannesin, M. Vandegriendt, R. W. GUI-
     ham, 1994, Transport and Fate of Dissolved Methanol, Methyl-Tertiary-Butyl-
     Ether, and Monoaromatic Hydrocarbons in a Shallow Sand Aquifer, American
     Petroleum Institute, Washington D.C.,  Health  and Environmental  Sciences
     Publication 4601.

  [8] J. W. Mercer, R. M. Cohen, 1990, A review of immiscible fluids in the subsur-
     face: Properties, models, characterization and remediation Journal of Conta-
     minant Hydrology, 6, 107-163.

  [9] State of New York ,1995, Official Compilation of Codes, Rules and Regulations
     of the State of New  York, Title 6 Environmental Conservation Subpart 255-3.

[10] H. S. Rifai, R. C. Borden, J. T. Wilson, C. H. Ward, 1995, Intrinsic bioreme-
     diation for subsurface restoration, in Intrinsic Bioremediation, R. E. Hinchee,
     J. T. Wilson and D. C. Downey eds., Batelle Press, 3(1), 1-29.

[11] P. J. Squillace, J. S.  Zogorski, W. G. Wilber, C. V.  Price, 1995, A Preliminary
     Assessment of the Occurrence and Possible Sources  of MTBE in Ground Water
     of the United States, 1993-94 U. S. Geological Survey Open File Report 95-456.

[12] United States Environmental Protection Agency, 1995, UST Corrective Action
     Measures for Fourth Quarter FY 95, Unpublished Report.

[13] United  States Environmental Protection  Agency,  1990, Subsurface Remedi-
     ation Guidance  Table 3,  United States Environmental Protection  Agency,
     EPA/540/2-90/011b.

[14]  J. W. Weaver, 1996a, Animated Three-Dimensional Display of Field Data with
     SITE-3D  User's Guide for  Version  LOO, US EPA,  EPA/600/R-96/004.

[15]  J. W. Weaver, 1996b, Application of the Hydrocarbon Spill Screening Model to
     Field Sites, 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.

[16]  C. K. Yeh and J. T.  Novak, 1994, Anaerobic biodegradation of gasoline oxy-
     genates in soils, Water Environment Research, 66(5), 744-752.
                                     12                         Weaver et al.
</pre><hr><pre>
-------
  Computer
Familiarization
</pre><hr><pre>
-------
13.  Computer Familiarization
</pre><hr><pre>
-------
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- /
</pre><hr><pre>
-------
            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
</pre><hr><pre>
-------
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
</pre><hr><pre>
-------
              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
</pre><hr><pre>
-------
 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
</pre><hr><pre>
-------
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)
</pre><hr><pre>
-------
 PAGE NOT
AVAILABLE
DIGITALLY
</pre><hr><pre>
-------</pre></td></tr></table></body></html>