U.S.E.P.A.
        GROUNDWATER MODELING WORKSHOP
                   Coordinated by

                  PEI Associates, Inc.
                  11499 Chester Rd.
                 Cincinnati, OH  45246
               Contract No. 68-03-3413
              Work Assignment No. 0-20E
                    PN 3741-20E
              EPA Technical Project Monitor
                     Don Draper
       U.S. ENVIRONMENTAL PROTECTION AGENCY
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
                   P.O. BOX 1198
                   ADA, OK 74820

-------
                            U.S. EPA
                             RSKERL
                 Ground Water Modeling Workshop
                         Evaluation Form

In order to make this series of workshops and upcoming technology
transfer events as useful as possible, we ask that you please
complete this evaluation form.  Your comments will remain
confidential.  Feel free to use the back of the page for
additional comments.  Thank you for your honesty and cooperation.
Position
Division/Branch	
1-3     3-5     5-10
Years experience; 0-1     1-3	3-5     5-10     10+
Background: Geology	 Engineering    Other (specify)	
Name (optional)	

1.      Please rate this workshop for its value to you and
        the tasks you perform as an EPA employee.
          1
        (poor)
                   (excellent)
        Was the subject matter adequately covered during each
        workshop session?
        Geochemical Modeling        Yes
        Unsaturated Zone Modeling   Yes"
        Saturated Zone Modeling     Yes"
                   No
                   No"
                   No"
        Were the sessions well paced within allotted time?
        Geochemical Modeling        Yes
        Unsaturated Zone Modeling   Yes"
        Saturated Zone Modeling     Yes"
                   No
                   No"
                   No"
        Were the handouts relevant, and appropriate for the
        workshop?
        Geochemical Modeling        Yes
        Unsaturated Zone Modeling   Yes"
        Saturated Zone Modeling     Yes
                   No
                   No"
                   No"
        How would you rate this workshop in relation to other
        training programs you have attended?
        Excellent
        Good
        Average
        Fair
        Poor
                                   Unsaturated Saturated
                       Geochemical    Zone       Zone
                        Modeling    Modeling   Modeling

-------
6.      Please rate the speakers conducting this workshop.

        Geochemical Modeling
                   Content    Presentation
        Excellent  	      	'
        Good       	      	
        Average    	      	
        Fair       	      	
        Poor       	      	

        Unsaturated Zone Modeling
                   Content    Presentation
        Excellent  	      	
        Good       	      	
        Average    	      	
        Fair       	      	
        Poor       	      	•

        Saturated Zone Modeling
                   Content    Presentation
        Excellent  	      	
        Good       	      	•
        Average    	      	•
        Fair       	      	
        Poor       	      	

7.      To what extent will you make use of the content and
        materials presented, either directly or indirectly?

                                            Unsaturated Saturated
                                 Geochemical    Zone      . Zone
                                  Modeling    Modeling   Modeling
        Will use frequently       	    	   	
        May be useful             	    	   	
        Of limited use            	._    	
        Don't see any application 	    	
        Don't know                	    	   	

8.      What was the most valuable part of the workshop?
9.      What was the least valuable part of the workshop?
10.     Do you have any suggestions to improve the content and
        presentation of the workshop (please be as specific as
        possible)?

-------
                       Background
    The Superfund Amendments and Reauthorization Act of 1986
(SARA) directs the EPA, as part of the overall Superfund site
clean-up program, to conduct a program of research, evaluation
and demonstration of alternative or innovative technologies for
response actions that will achieve more permanent solutions. Due
to the magnitude of the problem and the potential economic and
environmental benefits of inplace reclamation, the Office of
Solid Waste and Emergency Response has identified the need to
have in operation a dedicated technical support program that can
provide a readily-available source of technology transfer
support, including up-to-date technical information plus
associated expert assistance and review services, to those
responsible for making remediation decisions at Superfund sites.
  Four technical support centers have been established at the
following locations:

   1.    Robert S. Kerr Environmental Research Laboratory
            (RSKERL)-Ada, Oklahoma, (Ground-Water/Soil Fate and
            Transport Technology)
   2.    Hazardous Waste Environmental Research Laboratory-
            Cincinnati, Ohio, (Engineering and Treatment
            Technology)
   3.    Environmental Monitoring Systems Laboratory-Las Vegas,
            Nevada, (Monitoring and Site Characterization
            Technology)
   4.    Environmental Research Laboratory-Athens, Georgia and
            Research Triangle Park, (Health Risk and Ecology)

    The Subsurface Remediation Support Program has been
established at RSKERL. The components of this program include:

    - Subsurface Remediation Technical Support Team — composed
of ten scientists and engineers to provide a readily available
source of technical assistance;

    - Subsurface Remediation Information Clearinghouse —
designed to provide the user community highly specialized, fate,
transport, and remediation information;

    - National Center for Ground Water Research — consortium of
Oklahoma, Oklahoma State and Rice Universities charged with
developing and conducting a long-range exploratory research

-------
program to help anticipate and solve the Nation's emerging
ground-water problems;

    - International Ground Water Modeling Center (IGWMC) --
Holcomb Research Institute Indianapolis, Indiana, clearinghouse
for ground-water modeling software, providing research short
courses, seminars and educational activities;

    - National Ground Water Information Center — National Water
Well Association, Dublin, Ohio, repository of ground water
quality information accessible to scientists, government
agencies, business and the public; and

    - RSKERL Extramural Research Program — expertise of
subsurface processes and systems from more than thirty
universities and research institutions.

    The EPA Ground Water Modeling Workshop results directly from
Superfund technology transfer efforts planned by RSKERL. This
series of introductory ground water modeling workshops is being
conducted to introduce EPA personnel to a selected group of
geochemical characterization models, and saturated and
unsaturated zone contaminant transport models. The overall
objectives of this workshop include:

    1.   Provide a general overview of the use of models and the
limitation of using model results. Emphasis will be placed on
both the identification of modeling parameters and on the
importance of understanding the assumptions and limitations of
specific models as applicable to Superfund sites;

    2.   Familiarization of selected models through hands-on
experience and with case studies. Workshop participants will be
provided with the software presented to keep for future use; and

    3.   Identification of data quality objectives during the
site characterization process to satisfy both modeling data input
and site characterization requirements.

    RSKERL has made arrangements with IGWMC to:  (1) provide EPA
Regional staff easy access to the groundwater modeling
information services available from IGWMC's Indianapolis office;
(2) demonstrate  type and level of assistance available from IGWMC
in the selecting and use of groundwater modeling software by EPA

-------
Regional staff; (3) provide, on a limited basis, assistance to
EPA Regions in improving usability of Agency-sponsored public
domain software, by installing user-friendly interfaces, model
demonstration software, and computer-aided instruction software;
(4) advise EPA regions regarding development of groundwater
modeling policies and preparation of RFP's and workplans for
groundwater modeling projects in the Regions; and (5) provide EPA
Regions with IGWMC experts for limited review of proposals,
workplans, software, and reports related to groundwater modeling
insofar as these activities are initiated and managed by EPA
Regions.
    For further IGWMC information, call AC317/283-9458. Specific
requests for assistance from IGWMC should be made through M.R.
Scalf at RSKERL, AC405/332-8800.
    Although the Ground Water Modeling Workshop has been funded
by the United States Environmental Protection Agency, the models
presented herein may not necessarily reflect the view of the
Agency and no official endorsement should be inferred.

-------
                               U.S.E.PA
                 GROUNDWATER MODELING WORKSHOP
                          EPA-RSKERL, Ada, OK
                        August  16,17 and 18,1988
                               AGENDA
DAY 1      Topic:      Geochemical Characterization Models

                      WATEVAL - Water analyses, rock/water
                      interactions, graphical methods

                      BALANCE - Mass balance calculations

           Speaker:   Dr. Arthur W. Hounslow
                      Oklahoma State University
DAY 2      Topic:      Unsaturated Zone Modeling

                      RITZ - Regulatory and Investigative
                      Treatment Zone Model

                      CHEMRANK - Ranks which contaminants
                       might reach groundwater first

                      CHEMFLOW - Yields actual concentrations for
                      contaminants

           Speaker:   Dr. David L. Nofziger
                      Oklahoma State  University
DAY 3      Topic:      Saturated Zone Solute Transport Modeling

                      MOC - Method of Characteristics

           Speaker:   Dr. Randy Charbeneau
                      University of Texas

-------
                            U.S.E.P.A.
              GROUNDWATER MODELING WORKSHOP
                        FALL/WINTER, 1988
      Name/Address
LIST OF INSTRUCTORS

                 Work /Research Interests
Dr. Arthur W. Hounslow
Oklahoma State University
Department of Geology
Stillwater, OK 74078
           Organic and environmental
           Geochemistry; organic pollutants
           in air/water/soil systems.
Dr. P. S. C. Rao
University of Florida
Science Department
2169 McCarty Hall
Gainesville, FL 32611
            Environmental chemodynamics of
            agrochemicals, nontoxic wastes, soil
            and toxic/hazardous wastes.
Mr. Mark L. Brusseau
University of Florida
Soil Science Department
2169 McCarty Hall
Gainesville, FL 32611
            Transport and fate of organic
            contaminants in the subsurface;
            experimental investigation of
            sorption dynamics.
Dr. David L. Nofziger
Oklahoma State University
Agronomy Department
265 Ag Hall
Stillwater, OK 74078
            Modeling of the fate and
            transport of chemicals in soils;
            modeling of water movement in
            soils.
Mr. Joe Williams
Soil Scientist
U. S. EPA- RSKERL
P.O. Box1198
Ada, OK  74820
            Modeling of organic contaminant
            transport and fate in soils; variability
            of physical properties of soils.

-------
                            U.S.E.P.A.
               GROUNDWATER MODELING WORKSHOP
                         FALL/WINTER, 1988
                  LIST OF INSTRUCTORS (Continued)

      Name/Address                      Work/Research Interests

Mr. Peter F. Anderson, P.E.           Saturated zone solute transport
Vice President, Herndon Office        modeling.
GeoTrans, Inc.
250 Exchange Place, Suite A
Herndon, VA 22070
Dr. Leonard F. Konikow              Saturated zone solute transport
U.S. Geological Survey              modeling.
Water Resources Division
431 National Center
Reston, VA 22092
Dr. Daniel J. Goode                 Saturated zone solute transport
U.S. Geological Survey              modeling.
Water Resources Division
431 National Center
Reston, VA 22092
Dr. Randy Charbeneau              Saturated zone solute transport
University of Texas                  modeling.
Ernest Cockrell Jr. Hall
Austin, TX 78712

-------
                                     D
                                     >
                                     -<
      DAY 1
  GEOCHEMICAL
CHARACTERIZATION
     MODELS
     WATEVAL


     BALANCE

-------
    CONTEMPORARY INTERPRETATION OF WATER QUALITY DATA


A Practical Geochemical Approach Using Personal Computers
                   Arthur W. Hounslow
                    School of Geology
                Oklahoma State University
             Environmental Protection Agency
         Regional Ground Water Modeling Workshop
                        Fall 1988

-------
         CONTEMPORARY INTERPRETATION OF WATER-QUALITY DATA

                         Arthur W. Hounslow

I.  WATER ANALYSES 	1
   A.  SOURCE
      1. RIVERS
      2. WELLS
      3. SPRINGS
   B.  SAMPLES
      1. COLLECTION
      2. CONTAINER
      3. PRESERVATION
   C.  FIELD DETERMINATIONS
   D.  ANALYSES
      1. SELF
      2. SUBMIT TO LABS
      3. PUBLISHED DATA
      4. DATA BANKS
      5. PARTIAL ANALYSES
   E.  INTERPRETATION
      1. INDIVIDUAL ANALYSES
      2. COLLECTIVELY
   F.  CALCULATIONS
      FIGURE 1. GROUND WATER GEOCHEMISTRY	3
II. UNITS	4
   A.  MASS UNITS
      i. g./i
      2. FRACTION
      3. PERCENT %
      4. SALINITY °/OQ
      5. PPM
   B.  MOLES
      1. MOLES AND ATOMIC WEIGHTS
      FIGURE 2. PERIODIC TABLE 	5
      2. MOLE FRACTION
      3. UNITS AND CONVERSIONS
      4. EQUIVALENTS
      EXERCISES 1-7 	9
III. COMMONLY DETERMINED CONSTITUENTS 	11
   A.  SOURCE OF MAJOR IONS IN WATERS
      1. SODIUM
      2. CHLORIDE
      3. POTASSIUM
      4. CALCIUM
      5. STRONTIUM
      6. SULFATE
      7. GYPSUM
      8. BARIUM
      9. MAGNESIUM
     10. CARBONATE / BICARBONATE
     11. CARBON DIOXIDE
                            11

-------
   B. MISCELLANEOUS DETERMINATIONS
      1. HARDNESS
      2. DISSOLVED SOLID CONTENT
      3. CONDUCTIVITY
      4. CALCULATED DENSITY
      FIGURE 3. IONIC VOLUMES 	19
      5. pH
      6. ALKALINITY AND ACIDITY
      FIGURE 4. CARBONATE SPECIES 	.22
      EXERCISES 8-18	24
IV. ANALYSIS INTERPRETATION	27
   A. SPOTTING QUESTIONABLE ANALYSES
      1. ANION - CATION BALANCE
      2. RELATIVE AMOUNTS OF IONS
      3. MISCELLANEOUS CHECKS
   B. COMPLETING PARTIAL ANALYSES
      1. GIVEN HARDNESS, Ca OR Mg
      2. TEMPORARY HARDNESS OR ALKALINITY
      3. CARBONATE / BICARBONATE
      4. MISSING VALUES
      EXERCISE 19 	30
V. ROCK - WATER INTERACTIONS 	31
   A. MINERALS
   B. BALANCING EQUATIONS
      EXERCISE 20 	35
   C. SOURCE ROCK DEDUCTION
      TABLE OF WEATHERING PRODUCTS 	38
      FIGURE 5. WEATHERING FLOWCHART 	39
      EXERCISE 21 	40
VI. GRAPHICAL METHODS 	41
   A. AREAL TRENDS
      1. ONE COMPONENT PLOTS
      2. MULTI-COMPONENT PLOTS
         a. BAR GRAPHS
         b. PIE DIAGRAMS
         C. RADIAL DIAGRAMS
         d. VECTOR DIAGRAMS
         6. KITE DIAGRAMS
         f. STIFF DIAGRAM
   B. CHEMICAL TRENDS
      1. PIPER DIAGRAMS
         a. PRECIPITATION OR SOLUTION
      FIGURE 6. TRIANGULAR DIAGRAMS 	43
      FIGURE 7. PIPER DIAGRAM	44
      FIGURE 8. DUROV DIAGRAM	45
      FIGURE 9. TRIANGULAR GRAPH PAPER	46
         b. MIXING
         C. ION EXCHANGE
         d. WATER TYPES
      2. DUROV GRAPHS
      EXERCISES 22	48
                            111

-------
VII. GEOCHEMICAL ENVIRONMENTS 	49
   A. pH BARRIERS
      1. STRONGLY ACID - pH < 4
      2. MODERATELY ACID - pH 4-6.5
      3. NEUTRAL pH 6.5-7.8
      4. MODERATELY ALKALINE pH 7.8-9
      5. STRONGLY ALKALINE - pH>ll
   B. ADSORPTION BARRIERS
      1. MONTMORILLONITE CLAYS
      2. KAOLINITE CLAY
      3. GOETHITE (FEOOH)
      4. NATURAL ORGANIC MATTER
   C. REDOX BARRIERS
      THE FEW ELEMENTS IN NATURAL WATERS
      IRON GEOCHEMISTRY
      CARBON GEOCHEMISTRY - REDOX "BUFFER"
      1. AEROBIC WATERS
      2. ANAEROBIC WATERS (1)
      3. ANAEROBIC WATERS (2)
      FIGURE 10. REDOX ZONES 	54
      FIGURE 11. SEDIMENTARY GEOCHEMISTRY	55
      FIGURE 12. FENCE DIAGRAMS 		56
      EXERCISE 23
VIII. MASS BALANCE MODELING	58
   A. COMPOSITIONAL CHANGES
      FIGURE 13. MASS BALANCE DIAGRAMS 	59
   B. MIXING
   C. COMPUTER CALCULATION
      EXERCISES 24 & 25
XI. WATER GEOCHEMISTRY - SELECTED BIBLIOGRAPHY	64
                            IV

-------
          CONTEMPORARY INTERPRETATION OF WATER-QUALITY DATA

                         Arthur W. Hounslow

                          I. WATER ANALYSES

A. SOURCE
   1. RIVERS
      Ground water contribution greatest under low flow conditions.
   2. WELLS
      Most difficult samples to avoid contamination and the change in
      parameters during collection of sample.
   3. SPRINGS
      Deposits around spring will indicate water conditions prior to
      being exposed to atmospheric temperatures and pressures.

B. SAMPLES
   1. COLLECTION
   2. CONTAINER
   3. PRESERVATION
           Other determinations are done in the laboratory but in
      order to reduce the effects of adsorption or biodegradation
      they must be collected and transported in special containers,
      preserved either in ice or by the addition of acid or some
      other preservative depending on the constituent being
      considered.
            For each constituent in a water analysis the E.P.A has a
      recommended preservation and analysis procedure which must be
      followed.

C. FIELD DETERMINATIONS
      Any determinations dependent on dissolved gases must be run as
   soon as possible after collection. For some ground waters that
   contain gases under pressure even this may be too late.
   Determinations usually done in the field include:
          temperature
          pH
          DO - dissolved oxygen
          Eh - redox potential
          alkalinity - dependent on C02

D. ANALYSES
   1. SELF
   2. SUBMIT TO LABS
   3. PUBLISHED DATA
   4. DATA BANKS
      US E.P.A - STORET,
      US Geol. Survey - WATSTORE
   5. PARTIAL ANALYSES - ESTIMATING MISSING PARAMETERS

-------
E. INTERPRETATION

   1. INDIVIDUAL ANALYSES
     " a. accuracy checks
      b. dissolved minerals
         (1) mass balance
             BALANCE
         (2) thermodynamics
             WATEQF
             (a) speciation
             (b) saturated or unsaturated with
                 respect to a mineral
             (c) redox

   2. COLLECTIVELY
      a. Graphical representation
         (1) areal plots
         (2) trends - graphs
      b. Statistical analysis
         (1) anomalous samples or values
              mixtures of 2 or more populations
              cumulative frequency plots
         (2) mixed sources -
             factor analysis

F. CALCULATIONS
   Tedious but must be understood
   Computer imperative

-------
§
50
M

M
O
58
O
td
o
td
o

s
                                       GROUND-WATER GEOCHEMISTRY
                         LITHOSPHERE
                                                                         ATMOSPHERE
                            Sulfldes


                          Carbonates
                                               HYDROSPHERE
                   Soluble  salts
                                                     Ionic strength
                                      ; Water



                                      movement
                    Porosity



                   Permeability
                                                                             BIOSPHERE
                                            SorptTon
                                                      Degradation
                                                                              Bacteria
                        Soli organic

                          matter
                                                 Ground water
                         Hydroxides
                                                                              Mlcroflora
                     Rocks  and minerals

-------
                              II. UNITS

A. MASS UNITS
                             MASS/VOLUME

   1. g./i
      small concentrations mg./l
      1 g. = 1,000 mg.

                              MASS/MASS

   2. fraction - part of total which equals 1
      g. per g.
      mole fraction - mole/mole

   3. percent % - parts/hundred
      g. per 100 g.

   4. salinity °/QO - parts/thousand
      g. per 1,000 g.

   5. ppm - parts/million
      g. per 1,000,000 g.
           or
      mg. per 1,000,000 mg.
      1,000,000 mg. = 1,000 g.
                    = 1 kg.
              1 ppm = 1 mg./kg.


B. MOLES

   1. MOLES AND ATOMIC WEIGHTS
        One mole of an item is defined as 6.023 * 1023 of that item.
      It applies to atoms, molecules,  ions, golf balls etc. This
      number is known as Avogadro's Number.

         The atomic weight of an atom is the mass of that atom
      compared to that of the carbon isotope   Cc, which is defined
      as exactly 12.00. One gram atomic weight of an element or one
      gram formula weight of a compound contains 6.023E23
      atoms(formulas) of that material. Thus one mole of a compound
      equals the atomic or molecular weight of that compound in
      grams.
                            4

-------
H
o
G
JO
w

to
•a
w
»
H
o
o
H
O
§
f
n
      (-.   PERIODIC TABLE OK THE ELEMENTS
V Group
X
PailodN

1


2


3


4


5


8


.7



1

•
II

*
.
"

• * •
.


1.14
5 t
' Na
22.11
•19
J K
31.10
}37
i Rb
15.47
•55
' 'Cs
132.11
a >*
is F'
1 (373)

B*
9.01
2
Mg
34.31
20
C.
40.08
36
Sr
17.62
56
Ba
137.34
86
R.
(326)
.
t
H
1.001

• •

. •
•
.
"
TRANSITION ELEMENTS .•-*••.

• 21
: se
44.16
•39
t yf
a •
61.11
57-71
•

tl-1C3
"
.
.
•ta'ntSanlde Series
•

22
Tl
47.10
40
Zr
• 1.32
72
HI
171.41
104
Ku
(361)

• .
• Ls
a
130.11

23
V
50.14
41
'Nb
12.11
73
TA
•
110.15
IDS
Hat
(3(1)
• .
56.
Ct
1W.11
**
Cr
51.00
42
Mo
15.14
74
w •
113.15
106 •

(It3)

,5S9
'•; Pr
i
140.11
•
t •"
• Mn
84.14
a ^
a Tc
(171
S 7*
' Rai
a •*•
116.2
•26
• F.
55.15
s44
• Ru
101.07
a "
i o,
110.2
.

a *• a
a Co '«
51.93
,:45 ,i
!; Rh "
102.91
I"
i; tr i
112.2


28 J
NI "
51.71
48 j
Pd ;
1C1.4
jm
tw
PI i
115.01


a
Cu
63.55
41
Afl
107.17
79
AU
116.17


30
Zn
65.37
,«
i Cd
112.40
•80

190.51

III
•

IV


v.


yi


VII

.



B
10.11
|J
Al
36.11
It
G,
61.72
«9
In
114.12
81-
!•
It Tl
!• §i
304.37


« J
c
13.01
'4
SI
31.09
32
Ge
72.59
50
Sn
111.69
• 82
a
i pk
a
207.19


* t
H
14.01
15
P
30.17
p3
As
74.13
it
Sb
121.7S
83
Bl
301.11


•
o
16.00
16
S
3104
34
S.
71.96
52
T.
127.60
64
PO
(210)


*
F
11.00
7
"
Cl
3S.4S
w
Br
71.10
a
1
134.10
as
At
(310)


0


H»
4.00
10
N*
30.11
18
Ar
3115
36
Kr
83.10
54
*9
131.30
86
Rn
(222)

: . AlemTc vat(Mt art fcaitd e» tatbon-12;
•
•60
'•• Nd
t
144.24

•j Pm
a
(147)
tatutt !n f aranlfciatt an fer M>t meal lUt't or !h« meat famlTtar tactop*.
• t Sjmfcot fa wr»oWiet*l
* « ' * "
«; Sm
a
150.35
•• Eu
111.11
i"
'.; Gd
i
137.25
\"n
151.11
» Dy
163.50
I"
« Ho
164.13
{68
117 J 4
'•; Tm
111.13
•70
« Yb
• 173.04
>• Lu
174.17
      •Acltnli!* Scrtts
t
II
SI
II
*
f
89
Ac
1227)

t
31
II
•i

90
Th
332.04
J19I
ii p.
•1 (131)

a

92
U
331.03

it
i
t

93
Np
(217)



a
ia
li

54
Pu
(144)

i

95
Am
(213)
i
u
a
S6
Cm
(247)

t
1

•7
BJc
(2«7)
I


;a
Cf
(151)
a
at
ti
a
99
Es
(154)

i
i

100
Frn
(157)
a
SI
a
101
Md
(251)
•1102
13 No
•l (351)
a
ia
tt
at
1
103
Lr
(160)

-------
2. MOLE FRACTION

    Ratio of the number of moles of the give constituent to the
  total number of moles of all constituents.
                   nl
      Nl =
                n2
  Mole fraction of NaCl in 1.0 molal solution
           1
    =	    =   0.0177
      1 + 55.51

  where 55.51 = # moles of water in 1000 g water
                          EXAMPLES

 1.       55.85 g. Fe contains 6.023E23 atoms of Fe
     and  55.85 g. Fe                  =1 mole of Fe
     and  173 g. Fe contains 173/55.85 =3.10 mole Fe


 2.       70.91 g. C12 contains 6.023E23 molecules of C12
     and  70.91 g. C12                 = 1 mole C12
     and  50 g. C12 contains 50/70.91  = .71 mole C12


 3.       18.015 g. H2O contains 6.023E23 molecules of H20
     and  18.015 g. H20                = 1 mole H20
     and  1 1.  of H20                  = 1,000 g. H2O contains
          1,000/18.012                 = 55.51 mole H20


 4.       44.009 g. C02 contains 6.023E23 molecules of C02
    and   44.009 g. C02                = 1 mole C02
    and   84 g. CO2 contains 84/44.009 = 1.91 mole CO2

-------
   3. UNITS AND CONVERSIONS

                     * D
      ppm
    (mg/Kg)
•>  mg/1
        /(M.Wt.*1000)
                                      / M.Wt.
               * Z
   mmole/1

       /1000
moles/Kg solution
  (formality - f)
        * X
moles/kg solvent
(molality - m)
                        /
               * Z
   moles/1
 (molarity - M)
                                                      meq/1

                                                          /1000
•>  equiv/1
 (normality - N)
                 D = Density
                 Z = Valence
                            wt. solution

                     (wt. solution - wt. solute)
                        1000
                     (1000 - TDSg)
                             EXAMPLE

   A brine has a density of 1.018 and contains 12,000 ppm
        dissolved solids of which 3700 ppm is sodium.

        Sodium concentrations:
            3700 ppm
            mg/1      = ppm * D = 3700 * 1.018 = 3767 mg/1
            formality = ppm/(1000*M.Wt.)
                      = .1609
                = 3700/(1000*23)
            mmoles/1  = mg/1 /M.Wt.  = 3767/23  = 163.7826 mmoles/1

            molarity  = mmoles/1 / 1000        = .1638 moles/1

            TDS(g)    = 12,000/1000 = 12 g./kg.

            molality  = formality * 1000 / (1000 - TDS g)
                      = 0.1609 * 1000 / ( 1000 - 12)
                      = .1629

-------
      4. EQUIVALENTS

         Equivalents /I       = normality N
         Normality (N)        = g/1 / Eq. Wt.
         Equivalents/ million = epm
         Milliequivalents/1   = meq/1
                          epm = meq/1 / D
        Calculation of equivalent weights based on:

        *    a. charge of an ion

                Eq. Wt. = M Wt. / charge
    e.g. Pe"""  + 3C1~  <==>  FeCl3; Eq. Wt. Fe"*"1"1" = 55.85/3 = 18.62


             b. # of electrons transferred in
                an oxidation - reduction reaction

                Eq Wt. = M Wt. / # electrons transferred
    e.g. Fe++ <==> Fe    + e~; Eq. Wt. Fe    = 55.85/1 = 55.85
             c. # of protons or hydroxyls transferred in
                an acid base reaction

                Eq Wt. = M Wt./# of protons or hydroxyls transferred
    e.g. H+  + Cl~  <==> HC1; Eq. Wt. Cl" = 35.45/1 = 35.45
             d. neutral salts

                Eq Wt. = M.Wt./# of H atoms equivalent
                         to total cations

    e.g. Eq. Wt. Ca(N03)2  = 164.088/2 =82.044


* Most common method of calculation.
                            8

-------
                              Exercises

1. How many moles of sulfur are needed to combine with 1 mole of
   iron to form pyrite (FeS2)?
2. How many moles of iron are needed to combine with 1.44 mole S to
   form pyrite?
3. How many mole of S are in 3 mole of pyrite?
4. How many moles of sulfur are in 1 mole of
5. How many moles of CO? would be liberated from 1 mole of limestone
   (CaC03)?

Answers: 1. 2, 2. 0.72, 3. 6, 4. 3, 5. 1

6. For the following analysis calculate mg/l,molarity,
   formality, molality and meq/1.

         ppm        mg/1      M         f         m        meq/1

Na+     51600

K+       2650

Ca++     1360

Mg++     1720

HC03~

C03=

S04=     3680

Cl~     86600

Density  1.11

-------
7. For each of the following reactions calculate the equivalent
   weight for the ion specified.
a. Ca    + C03=       — >   CaC03                   C03
b. HC1   + NaOH       -->   NaCl   +   H20          HC1


C. C03=  +  H2O       -->   HC03~  +   OH"          C03


d. Oo    + 4H"1" + 4e~  — >   2H0O                    00
                            10

-------
TDS= 147,610 ;  D=  1.11
X   = 1,0007 (1.000 - 147)= 1.1732


Na
K
Ca
Mg
SO 4
Cl
ppm

51,600
2,650
1.360
1.720
3.680
S6.600
mg/1
ppm*D
57,276
2.942
1,510
1,909
4,085
96,126
M
mg/1
/1000
/MWt
2.49
0.08
0.04
0.08
0.04
2.71
f
ppm
/1000
/MWt
2.24
0.07
0.03
0.07
0.04
2.44
m
f*X
2.63
0.08
0.04
0.08
0.04
2.87
meq/1
mg/1
/MWt
*Z
2491.2
75.2
75.3
157.1
83.2
2711.0

-------
                III. COMMONLY DETERMINED CONSTITUENTS

   The parameters usually measured in the field include
   temperature, pH, conductivity and alkalinity.
The ions commonly determined in a water sample include
                      l~, S04~, HC03~ and C03=.
   Na+, K+, Ca++, Mg++,
   Other determinations include  Si02, TDS (total dissolved solids)
   and hardness. Density should be measured on brines.

A. SOURCE OF MAJOR IONS IN WATERS

   1. SODIUM - Na+

      Sources                                      Sinks

      plagioclase (albite) NaAlSi3Og

      halite          NaCl

      nahcolite       NaHCOo
      nepheline
                   NaAlSiO,
      montmorillonite Na - clay

      2 Na-clay + Ca"1"1" —> Ca-clay + Na+

      sea spray

      brines

      hot springs


   2.  CHLORIDE Cl~

      Sources

      halite          NaCl

      sea spray

      brines

      hot springs
                                                Sinks
                            11

-------
3. POTASSIUM - K+

   Sources                                      Sinks

   potash feldspar KAlSi3O8              clays + K+ --> illite

                                                plants

   mica            KAl2(AlSi3)010(OH)2

   leucite         KAlSi206


4. CALCIUM Ca++

   Sources                                      Sinks

   plagioclase (anorthite) CaAl2Si2Og      calcite

   calcite         CaCOo                   gypsum
   aragonite       CaCO3                   montmorillonite
                                            (natural softening)
   dolomite
   gypsum          CaSO^.2H20

   anhydrite       CaSO^

   fluorite        CaF2

   pyroxene (diopside) CaMgSi2O6

   amphibole       NaCa2(Mg,Fe,Al)Sig022(OH)2


5. STRONTIUM Sr++

   Sources                                      Sinks

   strontianite    SrC03                        strontianite

   celestite       SrSO^                        celestite

   aragonite       CaC03

   (Sr"*"*" substitutes for Ca*+ in
   aragonite but not in calcite.)
                         12

-------
6. SULFATE S04=

   Sources                                      Sinks

   pyrite          FeS2                        pyrite

   gypsum          CaS04.21^0                  gypsum

  anhydrite       CaS04                        sulfate reduction

  organic sulfur compounds

  combustion of coal and
  petroleum

  smelting of sulfide ores

  geothermal waters


7. GYPSUM CaS04.2H20

       Saturated solution in water

            636 mg./l Ca"1"1"

           1600 mg./l S04=

       Increases with NaCl concentration


8. BARIUM Ba++

   Sources                                      Sinks

   barite BaS04                                 barite

   oil-field brines

        Solubility in water < 1 ppm.

        Brines may contain 10|S to 100|S ppm
                         13

-------
 9. MAGNESIUM - Mg++

    Sources

    olivine        (Mg,Fe)2Si04

    pyroxene (diopside) CaMgSi206

    amphibole       NaCa2(Mg,Fe/A

    mica            K(Mg,Fe)3(AlSi3)010(OH)2

    dolomite        CaMg( 003)2

10. CARBONATE / BICARBONATE  C03 = / HC03~

    Sources

    atmosphere      C02 ~

    [H20 + C02 <==> H2CO3 <==> H+ + HC03~]

    nahcolite       NaHCO3

    sulfate reduction

    [S04= + 2CH20 ==> H2S + 2HC03~


11. CARBON DIOXIDE C02

        Atmosphere 0.03%

        PC02 = 0*0003 atmospheres
                                        Sinks

                                        montmorillonite
                                        Sinks

                                        calcite
C02  -x  °  -x higher in soils because of decomposition
    of organic matter.
                 14

-------
B. MISCELLANEOUS DETERMINATIONS

   1. HARDNESS

       Hardness is the sum of the Ca and Mg concentrations expressed
      in terms of mg/1 of calcium carbonate.

      Hardness = Ca(mg/l) * (M.Wt. CaC03j/(At.  Wt.  Ca)

               + Mg(mg/l) * (M. Wt. CaC03j/(At. Wt. Mg)

      EXAMPLE :

            Ca = 200 mg/1 Ca"1"*"; Mg = 30 mg/1 Mg++

      Hardness = 200 * 100.088 / 40.08 + 30 * 100.088 / 24.312

              • =            499.4      +           123.5

               = 622.9 mg/1
      Ca and Mg form an insoluble residue with soap; bath tub ring.
      Detergents introduced to overcome this. It may be a major
      problem in boilers as CaCOg is precipitated, resulting in poor
      heat conduction. Hardness of streams may vary seasonally
      because of variation of ground-water / surface-water run-off.
      The ground water is more likely to be harder than the surface
      water.
      Temporary hardness - calcium and magnesium carbonates, removed
      by boiling with precipitation of CaC03.
      Permanent hardness - calcium and magnesium sul fates, not
      removed by boiling.
                            15

-------
2. DISSOLVED SOLID CONTENT
    (Often called TDS or Total Dissolved Solids.)

   Calculated by adding the mass of ions plus Si02 and
   subtracting losses due to C02 losses from carbonates. It is
   determined by evaporating to dryness a known volume of water
   at a specified temperature, usually 105 -180°C. During this
   heating bicarbonates are converted to carbonates in the solid
   phase:

                  2HC03~  —> C03= + C02 + H20

   Amount of carbonate formed

      = mg/1 HC03~ *M.Wt. C03=/(2*M.Wt. HC03)

   Amount of H20 and C02 lost

      = mg/1 HC03~ *M.Wt. H2C03/(2*M.Wt. HC03)

   Thus estimated TDS

      = sum of ions + Si02

                    - mg/1 HC03~ * M.Wt. H2C03/(2* M.Wt. HC03)

      = sum ions + Si02 - mg/1 HC03~ * 62.02 / 122.032

      = sum ions + Si02 - mg/1 HC03~ * 0.5082


   In waters with high calcium and sulfate the residue at 180°C
   may still be slightly hydrated, thus giving high results.


3. CONDUCTIVITY
     Also called B.C. - electrical conductivity, specific
   conductivity or conductance.
     Conductivity is the reciprocal of the resistance in ohms
   between the opposite faces of a 1 cm cube of an aqueous
   solution at a specified temperature (usually 25°C). It is
   temperature dependent. The units are mho's. As these are too
   big the units generally used are micromho's, i.e. mho's * 10°
   It is a good estimator of TDS.
   TDS (mg/1) approximately equals A * conductivity (micromho's)
   where A = 0.54 - 0.96  (usually 0.55 - 0.76)
                         16

-------
4. CALCULATED DENSITY

   Partial ionic volumes of dissolved constituents in water are
   used primarily to estimate the effects of pressure on
   solutions. They may also be used to estimate the density of a
   solution.

   The volume of ions in solution is the sum of the product of
   the number of moles/1 * the partial molar volume.
   where
V = sum (n± * V±)

     V is the molar volume

     V^ is the partial molar volume at 25°C

     n  is the concentration in moles/1 of ion i
   Thus:
      Total volume of solution =  1000  cmc
      Mass of water
     Mass of solids

     Mass of water + solids

     Density
                   = volume of water
                     * density of water

                  = (1000 - V) * 1 g.

                  = TDS(mg/l)  / 1000 g.

                  = (1000-V) + TDS/1000

                  = mass / volume

                  = [(1000-V)  + TDS/1000]  / 1000
                         17

-------
ion
K1
Ca
  ++
mg/1
S04=

HCOr
Total
Density =
                   mole/1
EXAMPLE

  partial
molar volume
molar volume
  ni*Vi
11,162
414
427
1339
20,059
2811
146
36,358
0.4855
0.0106
0.0107
0.0551
0.5658
0.0293
0.0024

[(1000cc-8.75cc)*1.0]
- 1.5
8.7
- 17.7
-20.9
18.1
14.5
24

g + [36358/1000]
0.7283
0.0922
0.1894
1.1516
10.2410
0.4249
0.0576
8.75 cc
g
	 i
                        lOOOcc
                            18

-------
§
M
i
§
f
M
V)
                                 TABLE 3
Partial metal ionic volumes aruf compressibilities at infinite dilution in water at 25°C.
                              and 1 atmosphere
                              (Relative to H+)
C ATI OK
H* 	
Ii+ 	
Na+ 	
K+ 	
Rb+.... 	
Ca+ 	
NH4+.... ...... 	
•*.! • : 	
Ag+ 	
Mg4^ 	
Ca4"1- 	
Sr4"1- 	 v.... 	
Ba4* 	
Be4-1".:.... 	 	
Cd4*. 	 .*..... 	
Cu4* 	
ZQ-H-.... 	 	
La44* 	
Ce44*. 	 i..

«
0
-1.0
-1.5
4-8.7
+13.7
+21.1
+17.9
-1.0
-20.9
-17.7
-18.2
-12.3

-13


-38.3


10<£j
0
—34
-42
—37

-27
—11

—83
—71

—99
—23
—57
—62
—70

—152

AX10X
i '
OH- 	
F- 	 	
ci- 	
Br" 	
I- 	 	 	
CHiCOr 	 	 	
NOr 	
CNS-... 	
HCOa~ 	 	
MnO4~ 	 	
do,- 	
BrOa- 	 	 	
CrOr~ 	
cor -. . . .
sor~ 	 	



* •
F;
—5.3
-2.1
+18.1
+25 0
+36.6
+41.5
+29 3

+24
+43
+46
+44
+19.7
—3.7
+14.5




io«je5
—44

-8
+2
T"
+18
i ***
-10
+7
i •
+15
i ***
+2
i *•

•

—85
-70





-------
5. pH
    pH is a measure of the hydrogen ion concentration [H+],  or
   "more correctly activity which will be discussed later.

                        PH = -log10[H+]
   Note:

   mmole/1 H+  approx. = mg/1 H+

   At pH = 10;

       [OH~] = 10~4 moles/1 = 10"1 meq/1 = 10"1 * 17 = 1.7 mg/1

   pH may be raised by adding a base or by removing CC>2 from a
   solution, e.g. by photosynthetic assimilation.
   There are three main sources of hydrogen ions in natural
   waters,

    a. hydrolysis:

       H20 --> H+ + OH~ ;   In pure water [H+] = 10~7

    b. dissociation:

       H2CO3 —> H+ + HC03~

    c. oxidation:

       2FeS2 + 7.5O2 + 7H20 ~> 2Fe(OH)3 + 8H+ + 4S04=

   The [H+] in an aqueous solution is controlled by chemical
   reactions that produce or consume hydrogen ions. One of the
   most important of these is that set of reactions initiated
   when CO2 is dissolved in water, i.e.

                  C02(g) + H20(l) --> H2C03(aq)

                        H2C03(aq) ~> H+ + HC03~

                            HC03~ --> H* + C03=

                      BUFFERED SOLUTIONS
       If, when acids and bases are added to a solution the pH
   changes very little,  these solutions are said to be buffered,
   e.g. if we add hydrogen ions to a solution of carbonate,  then

                       C03= + H+ --> HC03"

   or another ion, namely bicarbonate is formed that uses up the
   added hydrogen ions,  such that the pH does not decrease as it
   otherwise would.
   (This is an important concept when dealing with acid rain or
   acid mine drainage.)
                         20

-------
                 INTENSITY AND CAPACITY VALUES
       Most quantities in chemical analyses are "intensity"
   functions,  i.e.  actual concentrations of a constituent.  Thus
   pH measures the  concentration of hydrogen ions in a solution.

       There are certain properties of solutions that are
   "capacity"  functions, that measure the response of the
   solution to change. For example the capacity of a solution to
   neutralize acids or bases is called the buffering capacity of
   the solution.

      As an example, if we have two solutions A and B, both with
   a pH of 8.2, but A has no carbonate whereas B has a high
   concentration of carbonate. If we add acid to both solutions
   the pH of A will be lowered after only a few drops of acid
   whereas the pH of B will remain relatively constant until
   sufficient acid has been added to change all the carbonate to
   bicarbonate.
6. ALKALINITY AND ACIDITY

   Alkalinity and acidity are quantitative measurements of the
   capacity of a solution to react to acids and bases.

      a. Alkalinity
          The alkalinity of a solution is defined as the capacity
          of a solution to react with strong acid. It is
          determined by a titration to specific end-points,
          namely pH = 4.5 - methyl orange and pH = 8.3 —
          phenolphthalein. A measured volume V of the water is
          titrated with a strong acid such as HC1 having a
          normality N.
          Several different solute species contribute to the
          alkalinity of a natural water sample, and the titration
          with acid does not specifically identify them.
          Alkalinity may be reported in several ways, the most
          common being in terms of an equivalent amount of CaCOg,
          usually meq/1 CaCOg.

          where meq/1 = mg/1 CaCOg/50  ... 50 is Eq. Wt. of CaCOg

          In most natural waters alkalinity is produced by the
          dissolved CC>2 species, bicarbonate and carbonate.  Non-
          carbonate contributors to alkalinity include hydroxide,
          silicate, borate and the organic ligands,  especially
          acetate and propionate. The inclusion of these ions in
          the alkalinity figure will only effect the analysis if
          they are also included as separate ions.

          Carbonate species are the most important participants
          in reactions that control the pH of natural waters.
          These relations are often illustrated by a graph
          showing the % of each species present at a particular
          pH.


                         21

-------
ro
to
            M

            8
            50
            PJ
            00
            O


            1
            M

            CO
            t)
            w
            o
            H
            w
            w
                                                                                                                                                    12.0
                                                                                                                                                                 13.0
                                                  19.—FeruoUga of (oUl ilssotT»d ewboo dloild* ipecles la solution u » fonetloo of pH^li*C; prcuui* 1 ttaoepber*.

-------
                              EXERCISE
     Comment on the reasonableness of the following water analyses
using the carbonate/bicarbonate graph.

     # |  pH   |  mg/1 HC03" |  mg/1 C03=
1
2
3
4
5
6
7
8
10.8
5.9
4.2
11.7
6.5
12.6
2
7
51
23
151
162
2
170
10
500
0
23
2
0
121
20
10
50
         b. Acidity
             Acidity is the capacity of a solution to neutralize a
             strong base,  that is,  to react with hydroxyl ions,  and
             in doing so to convert all Carbonate species to
             carbonate. Measured by titrating a measured volume  of
             water (V) with a strong base (such as NaOH) with
             normality N.

             Origin of acidity
                volcanic gases
                acid rain
                oxidation of sulfide minerals
                     coal and metal mines
                dissolved undissociated CO2 (I^COg) in water
                     e.g.  160 mg/1  I^COg in a water has a pH = 5.2
                oilfield waters
                     often contain dissolved acetate
                natural dissolved organic matter
                large molecules with -COOH and -OH sites
                originate in vegetation rich areas
                usually strongly colored waters

             Note:
             A solution possessing caustic alkalinity (OH~) has  no
             acidity and a solution possessing mineral acidity has
             no alkalinity
                            23

-------
                              EXERCISES

8. Calculate the concentrations of Na+ and Cl~ in mg./l resulting
   from the solution of 1 g. NaCl in 1 liter of water.


9. Rain water percolates through gypsum beds to an unconfined
   aquifer. The sulfate content of the aquifer water was found to be
   760 mg./l SO^". What is the sodium content of this water after
   passing through an ion exchange column?

10. Calculate the hardness in terms of CaCOq of a water containing
   350 mg./l Ca++ and 125 mg./l Mg++.

11. Given that a water has a hardness of 560 mg./l CaCOg and a Mg++
    content of 72 mg./l. Calculate the concentration of Ca++ in
    mg./l.

12. A water has a TDS(180°C) of 570 mg./l. It contains 155 mg./l
   HC03. What is the true TDS of the water?

13. A water contains 800 mg./l SO^3 and 500 mg./l Ca++. The measured
   TDS(180°C) was found to be 1750 mg./l.  An XDR examination of the
   residue revealed the presence of CaS04.0.5H20 but not anhydrite
   nor gypsum. What is the true TDS content of the water?

14. Estimate the TDS of a solution having  a conductivity of  2730
   umho's.

15. Calculate the H+ concentration in mg./l in waters having pH's of
   2,  and 7.

16. A ground water is reported to have a hardness of 1000 mg/1
   expressed as calcium carbonate. The aquifer is known to be
   composed primarily of dolomite. What would you expect the Ca and
   Mg contents to be in mg/1?

17. Calculate the dissolved solid content  of the following water
   Concentrations in mg/1.

     Na = 2.14; Ca = 48;  Mg - 3.6, HCOo = 152; S04 = 3.2; Cl = 8.0;
     Si02 =8.6

18. Calculate the density of the Red Sea Brine whose analysis is
   given below (all values are in g/1):

     Na = 105, K = 3.61, Mg = 0.95, Ca = 6.44, SO* = 1.14, Cl = 195
     Measured density = 1.196. Comment on  the difference.
                            24

-------
                          ANSWERS
8. Na=1000*23/58.45=393mg/l; 01=1000*35.45/58.45=607mg/l.

9. 760 mg/1 S04= = 7.912 nunole S04= = 7.912 mmole Ca++

   = 15.82 mmole Na+ = 364 mg/1 Na+

10. 350/40.08*100.09 =  874 mg./l CaC03 +

   125/24.31*100.09 =  515 mg./l CaC03 «

   Hardness         = 1389 mg./l CaC03

11. Hardness = 560 mg./l CaC03 = 5.595 nunole CaC03

             = mmole(Ca++ + Mg++)

   72 mg./l Mg++ = 2.962 mmole Mg++;

   Ca++ = 5.595 - 2.962 = 2.663 mmole = 106 mg./l

12. TDS(180°C) = 570 mg/1; 155 mg./l HC03~ =2.54 mmole/1

                    2HC03" <==> C03= + H20 + C02

 2     mole bicarbonate loses 1     mole water + 1     mole  CC>2

 2.54 mmole bicarbonate loses 1.27 nunole water + 1.27 mmole  C02

                             22.9 mg.           55.9 mg.
 Total loss = 78.8 mg./l

 True TDS   = 570 + 79 = 649 mg./l
                                 OR

                    2HC03~  <==>  C03=  +  H20  +  C02

                   122.036        60.01  18.016   44.01
                                         	v	
                   122.036 mg  loses         62.026 mg.

  Loss is 50.83 % HC03~; Loss = .5083* 155 = 79 mg.

13. TDS(180°C) = 1750 mg./l;

   800 mg./l S04= = 8.33 mmole/1; 500 mg./l Ca++ = 12.48 mmole/1

   Limited by sulfate; CaS04.0.5H20 = 8.33 mmole/1

                   2CaS04.0.5H20 <==> 2CaS04 + H20

                      2 mmole                 1 mmole

                 8.33 mmole                   4.165 mmole
                                            = 75 mg./l H20


                            25

-------
   True TDS = 1750 -75 = 1675 mg./l



14. TDS - A * cond. ;  A = 0.55 - 0.76



     cond. o 2730; TDS - 1502 - 2075
15.  pH = 2;  [H+] = 10~2 mole/1 = 10   mmole/1 = 10.1  mg./l H+





     pH = 7;  [H+] = 10~7 mole/1 = 10~4 mmole/1 =  1E-4 mg./l H+
                            26

-------
                     IV. ANALYSIS INTERPRETATION

A. SPOTTING QUESTIONABLE ANALYSES

   1. ANION - CATION BALANCE
      The accuracy of many water analyses may be readily checked
      because the solution must be electrically neutral, and thus
      the sum of cations in meq/1 should equal the sum of the anions
      in meq/1.
                (meq/1 = mg/1 * valency / formula wt.)

      The result is usually expressed as a percentage,

          i.e. Balance = (C-A) / (C + A) * 100
               where C = sum of cation and A = sum of anions.

      If the balance is < 5% the analysis is likely to be good.

      If the balance is exactly 0% it is likely that the Na or Na+K
      were determined by difference,  especially in older
      analyses, i.e. prior to 1960.

      If the balance is much greater than 5% then

      a. the analysis is poor (inaccurate)
      b. other constituents are present that were not used to
         calculate the balance.
      c. the water is acid and the H* were not included
      d. organic ions are present in significant quantities,
         often indicated by colored waters.
   2. RELATIVE AMOUNTS OF IONS REPORTED
      (Units used are meq/1)

      The following is only a guide and it must be emphasized that
      many accurate analyses may reflect exceptions to these general
      statements, however, the objective is not to throw out all
      analyses that do not comply with these generalizations but
      rather to ensure their accuracy, by bringing potential
      inconsistencies to the attention of the investigator.

      a.   Na » K

      b.  Ca >= Mg  unless Ca removed by precipitation

      c. Ca >= S04  unless Ca removed by precipitation or ion
                    exchange

      d.  Na >= Cl  unless Na by ion exchange (reverse
                    softening)
                            27

-------
   3. MISCELLANEOUS CHECKS

      a. Calculated hardness should equal reported hardness.  A
         discrepancy may indicate incorrect copying of reported
         data.
      b. Calculated TDS should equal reported TDS Incorrect
         transcription is a frequent cause of non equivalence.
      c. TDS / Measured conductivity should be between 0.55 and
         0.76
      d. If C03= is absent pH should be < 8
      e. Conductivity /(sum of cation) approx. = 100
         Usual range 90-110.
      f . Temporary hardness usually equals bicarbonate


B. COMPLETING PARTIAL ANALYSES

   1. GIVEN HARDNESS, Ca OR Mg
      Given any two of the following — hardness, Ca or Mg — the
      other may be calculated.

   2. TEMPORARY HARDNESS OR ALKALINITY
      Temporary hardness is equivalent to alkalinity

   3. CARBONATE / BICARBONATE
      If HCOg~ or COg12 a
      reasonable values.
   If HCOg~ or CO12 are unreasonable recombine to obtain more
4. MISSING VALUES
   A single missing value may be obtained by subtracting the sum
   of cations from the sum of anions or vice versa.
                         28

-------
                               EXAMPLE





     permanent hardness =  75 mg/1 CaC03



     temporary hardness = 345 mg/1 CaCOg



      Mg = 6 mg/1; NaCl = 35 mg/1; pH = 7.1



Total hardness = 345+75  = 420 mg/1 CaC03 = 8.4 meq/1



            Mg = 6/12    = 0.5 meq/1



thus        Ca = 8.4-0.5 =7.9 meq/1      = 7.9*20 = 158 mg/1



Alkalinity      = temporary hardness  = 345 mg/1 CaC03



     At pH      =7.1 HCOg" is the only carbonate species present



thus alkalinity = [HC03~]    = 345/50*61     = 421 mg/1



As    log[CO3=] = log[HC03~] + log K2 + pH



                = -2.16      - 10.5   + 7.1  = -5.56



then      C03=  =0.17 mg/1



Permanent hardness  = 75 mg/1 CaC03 = 75/50 = 1.5 mmole CaCO3



     This is considered to be CaSO^



          thus S04= = 1.5 * 48      =72 mg/1 S04=



Given 35 mg/1 NaCl



          = 35/(23+35.5) = 0.598 mmole/1 NaCl



          = 0.598 * 23   = 13.8 mg/1 Na+



          = 0.598 * 35.5 = 21.2 mg/1 Cl
                            29

-------
                              EXERCISE

19. You are studying a limestone aquifer and collected a sample that
   was sent to a commercial lab. The results obtained from the lab
   are listed below. At the same time that you took the sample you
   obtained a pH of 6.9 and a total alkalinity of 1400 mg/1
   bicarbonate using simple field equipment.

                   mg/1

Na+                 130

K+                  130

Ca++                 57

Mg++                375

HC03""               645

C03=                360

S04=                121

Cl~                 156

Hardness (€3003)   1171

TDS (180°C)        1529

pH                    6.72
     Your Job if you decide to accept it is to list the suspect
numbers with all your evidence against them, and replace them with
more reasonable substitutes with appropriate documentation. Should
you be caught juggling the figures without the appropriate evidence
the Department will disavow all knowledge of you.

[ Note: Analytical errors as such are much less common than those
involving arithmetic, transposing of figures or columns, or slipping
a decimal point such as recording 67.0 as 6.70. ]
                            30

-------
                    V. ROCK - WATER INTERACTIONS
A. MINERALS

CARBONATES
     ARAGONITE
CLAYS
CALCITE

DOLOMITE

KAOLINITE

MONTMORILLONITE
     ILLITE
FELDSPARS
     PLAGIOCLASE
          ALBITE

          ANORTHITE

     POTASH FELDSPAR
FERROMAGNESIAN SILICATES
     AMPHIBOLES
          TREMOLITE
          HORNBLENDE
     MICAS
          BIOTITE

     OLIVINE
     PYROXENES
          DIOPSIDE

          AUGITE
HALIDES
     FLUORITE

     HALITE
OXIDES/HYDROXIDES
     BAUXITE

     HEMATITE

     LIMONITE
QUARTZ
SULFATES
     GYPSUM

     ANHYDRITE

     BARITE
SULFIDES
     PYRITE
CaC03

CaCOg

CaMg(C03)2

Al2Si205(OH)4

Al2Si4010(OH)2

KAl2(AlSi3010)(OH)2
                    NaAlSi308

                    CaAl2Si2Og

                    KAlSi308
                    Ca2(Mg,Fe)4AlSi7A1022(OH)2  X

                    NaCa2(Mg,Fe,Al)5Si8022(OH)2 X

                    K(Mg,Fe)3AlSi3010(OH)2

                    (Mg,Fe)2Si04

                    Ca(Mg,Fe)Si206

                    Ca(Mg,Fe,Al)(Al,Si)206

                    CaF2

                    NaCl

                    AIO(OH)

                    FE203

                    FeO(OH)
                    Si02

                    CaS04.2H20

                    CaS04

                    BaS04

                    FeSo
[






X
X
X
X
X
X
X
X
X
X



X


X
X
Mm Sed
X
X
X
X
X
X
X
X
X X
X
X
X
X
X
X

X
X
X
X
X X
X
X
X
X X
                            31

-------
B. BALANCING EQUATIONS
   Pertaining to surface mineral weathering reactions involving C02
   and H20

   1. Mineral + C02 + H20.

   2. Decide clay type to be formed.
      kaolinite or montmorillonite

   3. Balance aluminum.

   4. balance cations released.

   5. Set bicarbonate to balance cations released.

   6. Balance Si with H4SiO4.

   7. Set # C02 molecules to balance HC03~.

   8. Set # H20 molecules to balance # of H atoms in products.

   9. Count oxygen atoms on both sides to check balance.


                              EXAMPLES

Albite: — > kaolinite

2NaAlSi308 + 2C02 + 11H20 --> Al2Si205(OH)4 + 2Na+ + 2HC03~ + 4H4Si04

Albite: — > montmorillonite

2NaAlSi308 + 2C02 + 6H20 — > Al2Si4010(OH)2 + 2Na+ + 2HC03~ + 2H4Si02


Biotite: — > kaolinite

2KMg3AlSi301Q(OH)2 + 14C02 + 15H20 -->
                     Al2Si205(OH)4 + 2K+ + 6Mg++ + 14HC03~ + 4H4Si04
Biotite: — > montmorillonite

                   + 14C02 + 10H20 -->

                    Al2Si401Q(OH)2 + 2K+
                                                   14HC03~ + 2H4Si04
                            32

-------
                WEATHERING OF ORTHOCLASE TO KAOLINITE

 .  KAlSi3Og + ? C02   + ? H20   <===>
2.  CLAY TYPE
3. BALANCE Al

   2KAlSi308 +  ? C02


4. BALANCE CATIONS

   2KAlSi308 + •  ? C02
                                 < = = = >  KAOLINITE

                                 <===>  Al2Si205(OH)4



                            H20  <===>  Al2Si205(OH)4
                            H20  <===>  Al2Si205(OH)4

                             + 2K*
5. BALANCE CATIONS WITH BICARBONATE

   2KA1S1308 +  ? C02  +  ? H20  <===>  Al2Si205(OH)4

                             + 2K+ + 2HC0~
6. BALANCE Si
   2KA1S1308 +  ? C0
7. DETERMINE O>2

   2KA1S1308 +   2C02



8. DETERMINE H20

   2KA1S1308 +   2C02
                          ? H20  <===>  A12S1205(OH)4

                             + 2K+ + 2HC03~ + 4H4SiO4



                          ? H20  <===>  Al2Si205(OH)4

                             + 2K* + 2HC03~ •*• 4H4Si04



                          11H20  <===>  Al2Si205(OH)4

                            + 2K+ + 2HC03~ + 4H4Si04
9. CHECK OXYGEN
       16   +    4    +    11
                31
                                        9+6+16
                                           31
                            33

-------
              WEATHERING OF BIOTITE TO MONTMORILLONITE

1.  KMg3AlSi3010(OH)2 + 1 C02 + ? H2O <===>

2. CLAY TYPE
                                      <===> MONTMORILLONITE
                                             Al2Si4010(OH)2

3. BALANCE ALUMINA

   2KMg3AlSi3010(OH)2 + ? C02 + ? H20 <===>  Al2Si4010(OH)2

4. BALANCE CATIONS

   2KMg3AlSi3O10(OH)2 + ? C02 + ? H20 <===>  A12S14010(OH)2

                            + 2K+ + 6Mg++

5. BALANCE CATIONS WITH BICARBONATE

   2KMg3AlSi301Q(OH)2 + ? C02 + ? H20 <===>  Al2Si4010(OH)2

                           + 2K"1" + 6Mg++ + 14HC03~

6. BALANCE Si

   2KMg3AlSi3Ol0(OH)2 + ? C02 + ? H2O <===>  A12S±4O10(OH)2

                          + 2K+ + SMg"*"1" + 14HCO3~ + 2H4Si04

7. DETERMINE CO2

   2KMg3AlSi301Q(OH)2 + 14C02 + ? H20 <===>  A12S±4010(OH)2

                          •f 2K+ + 6Mg++ + 14HC03~ + 2H4SiO4

8. DETERMINE H20

   2KMg3AlSi301Q(OH)2 + 14C02 + 10H20 <===>  A12S±4010(OH)2

                          + 2K+ + 6Mg+-f + 14HC03" + 2H4Si04

9. CHECK OXYGEN
         24      +      28    +10    =12+42+8
                        62                       62
                            34

-------
                              EXERCISE
20. Balance the equations for the weathering of the following
   minerals. Try each with both clay minerals,  kaolinite and
   montmorillonite.
    a. anorthite

    b. tremolite

    c. diopside    CaMgSi206

    d. dolomite    CaMg(C03)2
                            35

-------
C. SOURCE ROCK DEDUCTION

     The purpose of this technique is to help the reader gain an
   insight into the possible origin of a typical water analysis. The
   approach used is not infallible but it can be very helpful. The
   method is a somewhat simplistic mass balance approach to water
   quality interpretation and is no way intended to replace the more
   sophisticated models such as WATEQ,  BALANCE etc.   It is based on
   the calculations and discussion in Garrels and MacKenzie,  1967. A
   flow chart of the technique is given in Figure 1.

      Step 1. If the pH of the water is less than about 6 abandon
      the project.   Acid waters cannot be interpreted this way
      without modifying the procedure.

      Step 2. The concentration of the various constituents,
      usually expressed as mg/1. must be changed to meq/1. in order
      to be able to combine the various ions in a chemically
      meaningful way.  Silica which exists as a neutral complex is
      converted to millimoles/ liter.  This is accomplished by
      dividing the ion concentration by its equivalent weight,  and
      silica by its molecular weight.

      Step 3. Sum the cations and the anions separately,  omitting
      Silica. Their totals should be within about 5% of each other;
      if not proceed with caution or throw the analysis out.

      Step 4. Compare the chloride and sodium contents.  We assume
      that the primary source of chloride in the water is from
      sodium chloride,  directly or indirectly from the ocean via
      precipitation. Sodium on the other hand can be derived from
      other sources, for example the solution of feldspars,  ion
      exchange etc. Thus,  if chloride > sodium then there is either
      an analytical error,  or the water derived its composition
      from the solution of evaporite minerals. In the latter case
      one would expect the dissolved solid content of the water to
      be high, at least over 500 mg/1.


      Step 5. Compare the sulfate and calcium contents.  The primary
      assumption is that sulfate is generally the result of direct
      dissolution of gypsum or the neutralization of acid waters by
      limestone or dolomite. In the latter case magnesium may be
      prominent. If sulfate > calcium then the inference is that
      calcium has been removed from solution, most likely by the
      precipitation of calcite. Calcium may also be removed from
      solution by ion exchange reactions such as:

                    Ca"1"*" + 2Na-CLAY  -->  2Na+ + Ca--CLAY

      Step 6. Compare bicarbonate with silica. Bicarbonate is formed
      when carbon dioxide and water react with various minerals in a
      process called acid hydrolysis.  Carbonates dissolve without
      releasing silica whereas albite releases considerable silica,
      and other silicates a much lower amount, Table 1.  An arbitrary


                            36

-------
division of silica/bicarbonate of 0.1 is used to indicate
silicate versus carbonate weathering.

Step 7. Compare silica with Ma"1" (-Cl~) + K+. We assume that
after subtracting the chloride from the sodium, then the
remaining sodium is due to the weathering of plagioclase, and
the potassium from the weathering of biotite and to a lesser
extent potash feldspar. If other ferromagnesian minerals are
present, silica will be present in a considerably excess over
the sodium plus potassium. It is also assumed that the solid
weathering product formed is either kaolinite or
montmorillonite; the former releasing more silica to the water
than the latter. Thus we may conclude that if:

   a. SiOo >  2 * (Ma* + K+ - Cl~)  then the minerals
   subjected to weathering contained a considerable quantity
   of ferromagnesian minerals, such as olivine, pyroxene or
   amphibole. Under these conditions the source of much of the
   calcium is probably plagioclase.

   b. SiC>2 < (Na+ + K+ -Cl~)  then cation exchange is probably
   the source of most of the excess sodium, in which case it
   is probable that calcium is less than the sulfate.

   c. (Na+ + K+ - Cl") < Si02 < 2 * (Na* + K+ - Cl~)  then Na-
   feldspar weathering is suspect, and the product is either
   kaolinite or montmorillonite.
   There are some occasions when the application of these
   calculations will lead to misleading conclusions, but I
   suspect that 90% of the time they will be meaningful and
   will lead to ideas for further investigation. .PA
                      37

-------
                             TABLE 1

             Weathering products of common minerals.

                    Cations and SiC>2 based on 100 meq HC03

                                                    Ratios
Mineral
j
albite
albite
diopside
tremolite
tremolite
forsterite
phlogopite
phlogopite
anorthite
calcite
dolomite
Clay Na+ K+ Ca++ Mg+
formed
(1)
K 100 -
M 100 - - -
25 25
K - 17 33
M - 17 33
50
K - 14 43
M 14 43
K - - 50 -
-. - 50
25 25
+ Si02
200
100
50
42
25
25
29
14
-
-
-
SiO2 I
HC03~
2
1
0.5
0.4
0.25
0.25
0.29
0.14
0
0
0
«a++K+
Si02
0.5
1
0
0
0
0
0.5
1
0
0
0
(1) K = kaolinite
    M=  montmorillonite
                            38

-------
FIGURE 5. WEATHERING FLOWCHART
         Flow chart illustrating  logic  of water quality
     interpretation using  simplified mass balance technique,

                     Water analysis

         Na+, K+, Ca++, Mg"1"1", HC03~,  C03=,  S04=, Cl~,  Si02
+                  '    —                    4- '    -•»
             rasa   s pi                   Ma   S04=                       Ca"*"1" <  S04=

                                       Ca"1"1" removal,  by
                                       calcite precipitation, or
                                       Ion exchange





HCO3~ > 10*Si02                     HC03~ < 10*Si02

Carbonate weathering


                             ,-—•----

                       Si02  >  (Na +K -Cl )      Si02

                                                 cation exchange
                                                 probable Ca++ < S04=
    Si02 > 2*(Na++K+-Cl~)

     Weathering of
     ferromagnesian minerals
     Calcium from feldspar
  Si02 < 2*(Na++K+-Cl~)

Granitic weathering
Calcium from feldspar
                            39

-------
                              EXERCISE
21. Interpret the origin of the waters listed below, values  in
   ing./I.
Na+
K+
Ca++
HC03'

S04=

Cl"
SiO2
1
9.5
1.4
27.0
6.2
93.0
32.0
5.2
39.0
2
0.2
0.0
2.5
7.7
44.0
0.0
0.7
16.0
3
24.0
7.0
74.0
9.5
277. 0
19.0
24.0
11.0
4
2.0
0.6
34.0
14.0
160.0
3.7
2.5
9.2
5
76.0
3.5
178.0
86.0
285.0
707.0
11.0
14.0
                            40

-------
                        VI. GRAPHICAL METHODS

     Graphical methods of illustrating water analyses have two main
     objectives:

A. AREAL TRENDS
   To enable analyses to be plotted as symbols on a map.
   These are shown on maps using a variety of plots to represent the
   analyses. They are of two general types.

   1. One component is usually plotted, generally TDS or
      conductivity although other elements of interest or even
      ratios of ions or elements may be plotted.  They are readily
      contoured and in some cases easily interpreted.
   2. Multi-component plots although commonly used are are extremely
      difficult to interpret because an interpretation involves an
      eyeball examination of various shapes and sizes on a map.

      The main types of such plots in common use are:

      a. Bar graphs; usually expressed in meg/1 as this enables the
         sum of anions and the sum of cations to be drawn the same
         length.

      b. Pie diagrams or circular diagrams. These may be drawn such
         that the diameters of the circles are proportional to the
         dissolved solid content.

      c. Radial diagrams. Concentrations usually in meq/1 or %
         meq/1. The arms of the plots are usually 60° apart and the
         ends connected to form a polygon.

      d. Vectors where along predefined directions the length is
         proportional to the concentration which is usually
         expressed in meq/1. The ends of the plot are not usually
         connected.

      e. Kite diagrams. This configuration is limited to four
         concentrations and the axes are Ca+Mg, Na+K, Cl+SO^,
                  in meq/1.
      f .  Stiff diagram. This uses four parallel horizontal axes
         extending on each side of a vertical zero axis. Four anions
         and four cations can thus be plotted on the left and right
         of the vertical respectively. The concentrations are in
         meq/1. The resulting points are connected to give an
         irregular polygonal pattern. The size of the pattern is
         approximately equal to the total ionic content.
                            41

-------
B. CHEMICAL TRENDS

   The determination of  any trends existing in a collection of
   water analyses is usually accomplished using either an x-y plot
   or some type of trilinear plot.

    Usually when considering plotting techniques only major
   components are plotted,  specifically Na (+K),  Ca,  Mg,  Cl,  SO^ and
   HCOo (+033). This approximation results in 6 major ions or
   combination of ions to be plotted.

   In contrast to the plotting of areal trends,  where individual
   analyses are each plotted in some way on a map,  chemical trend
   plots are usually done on one type of diagram on which all the
   analyses are plotted.

   The simplest type of such a plot is a simple x-y plot  of two ions
   or variables,  such as  Na versus  Ca. These are by far the simplest
   graphs to interpret and many computer statistical  packages such
   as SAS can be set up to do many  such plots.

   A more complex method of plotting is the use of  trilinear or
   triangular diagrams. These diagrams have a disadvantage that only
   three variables can be plotted on a triangle.  This is  overcome to
   some extent by plotting cations  and anions on separate triangles
   and arranging them is  such a way that they can be  related to one
   another. One major problem with  this type of plot  is that the
   analyses plotted are only ratios,  and the effect of dilution is
   not immediately apparent. The two main techniques  for  plotting
   water data are as follows:

   1. Piper diagrams where the sides of the anion and cation
      triangles are set at 60° apart and a diamond  shape  is used to
      replot them on one  diagram.

      a. PRECIPITATION OR SOLUTION
      If a series of water analyses are plotted on  a  Piper diagram
      and they lie on a straight line which when extrapolated passes
      through the corner of one or  both of the triangles  then it
      is possible that the trend is the result of the component at
      the corner of the triangle either being added or removed from
      solution.
       Examples include calcite precipitation or solution (calcium
      and bicarbonate) or gypsum precipitation or solution (calcium
      and sulfate). Mass  balance and solubility equilibrium may be
      used to further confirm or disprove this hypothesis.
                            42

-------
            FIGURE 14
            Determination of point C by the two-line method. Two line* urc drawn through C
            parallel to tiny iwu of the tide* at the iriuaiflc (here AY urul \'Z\. The
            interaction of thciic two lino wiih the third tide Uf/l divide* ilui Me into
            three pun* wltotc kntihi we proponional m the rclMive aniuunu
            A', f. uiul 2 ki pttint r.
100% X
                                    CO   60   50   S>   20    10
                                                                                                                      KXJ %/
                                                                                showing mcihixl of determining or pbuinu compoiiiion C
                                                             wiiliiu UM iliree-cumpoucnl kytlcm X-Y-Z.
        Tte poiAl C on irUnituliir graph paper. The pereeniuge» ofcomponcnu AT, Y. and
        2 krc rod directly from the numbered coordinate uxc*.
FIGURE  6.  TRIANGULAR  DIAGRAMS
                                                 43

-------
H
8
50
M
•a
H
•0
w
                         Permanent
                         hardness
                     Temporary
                      hardness
Saline
                                                      Cl
                                   Alkali
                                  carbonate

-------
               Ca5+ 50%
               HC0750V.
                                Ca  25%
                                Mg2+25%
                                           Ion exchange
                                            Reverse
                                          Ion exchange
 Fig. 6.19. Expanded Durov diagram with subdivisions and processes
                           demonstrated.
FIGURE 8. DUROV DIAGRAM
                            45

-------
8
»
w

NO
H
o
G
f<

g

O


I
33
w
S3
                             VWmVAVAVAAA
                                                                VvYWi;\ °i.'/VYYWVi1. . -W\ /WVv"

                                                                777?77^TO^A^W?90TO)TO
                                                                \7vAA *v^A /vA/VAJ* A\A A/\ A A.K 'v/vA A ^A A/v K 'v J
                 Vf)^yxVAV)5ggo'
                 ^\ww\^w\rv^w **
                 mx\\W)ftgS55g

                   WVWWW7OTS
                   fMxWxWW
yyvYYV^^TK>^rxVYxvi
A AT* 'V^v A '•. ^* 'VA '» A />> 'V A /\ ^\ '
                                           x
                                          vY/V)Vyr/v^>^yyy'i\YY) xVy^>^\Y/A\\ A\\^.V^\^vv^vy/^^^VA\\nA.V\\^a

-------
  b. MIXING
    If  two waters mix then the composition of the mixture will
    lie on a  straight line joining the two end members, and the
    relative  amount of each end member in the mixture is
    inversely proportional to the distance of the mixture from
    that end  member. If a water is strictly the result of
    mixing, without the addition or removal of any phase, then
    the mixture will exhibit exactly the same proportions
    between the end members on both cation and anion triangles
    as  well as on the diamond.

 C. ION  EXCHANGE
    The replacement of calcium and magnesium by sodium is a
    special case of addition and removal from solution. The
    line connecting water compositions changed by ion exchange
    starts parallel to constant magnesium and then curves down
    towards the sodium apex. This suggests that ore calcium is
    being exchanged than magnesium.

  d. WATER TYPES
    Waters may be divided into four basic types according to
    their placement near the four corners of the diamond. Water
    plotting  at the top of the diamond is high in both Ca+Mg
    and 804 resulting in an area of permanent hardness. The
    waters plotting near the left hand side corner are rich in
    Ca+Mg and HC03 and lie in the region of temporary hardness.
    Those waters plotting at the lower end of the diamond are
    primarily composed of alkali carbonates (Na+K and
    HCC^+COg). Those waters lying near the right hand side of
    the diamond may be considered saline (Na+K and Cl).

,  The  Durov or expanded Durov graphs are similar in that the
  analyses are plotted on separate anion and cation triangles,
  but  in this  case the sides of these triangles are 90° apart.
  Also in the  expanded Durov the three corners of each triangle
  are  physically separated from one another. The net result is a
  square plot  divided into nine areas characteristic of nine
  different water types.
                       47

-------
A
12.2
1.2
0.8
B
4.6
1.0
6.4
C
4.6
4.0
2.4
D
6.3
1.1
5.0
                              EXERCISES
           Analyses in meg/1
Ca
Na
Mg

Problem a.
     Plot analyses A, B, C & D.
     Label triangle LHS=Ca; RHS=Na; Vertex=Mg

Problem b.
     Remove 1/4 (E), 1/2 (F) & 3/4 (G) of Ca from analysis A,
and plot the results. This could result from calcite precipitation.
What is your conclusion? Can you explain analysis C?

Problem c.
     Combine analyses A & B in proportions 1:1 A:B (H) and
3:1 A:B (I). Use meq/1 not percentages.
What is your conclusion?  Can you explain analysis D?
                            48

-------
                    VII. GEOCHEMICAL ENVIRONMENTS

     Th.e rocks comprising an aquifer, through which ground water
flows, usually contain substances that provide sinks or sources for
hydrogen ions, electrons or soluble salts which in turn determine
the pH, redox potential or ionic strength of the water passing
through it.  Changes in these parameters in turn may change the
chemical composition of the water because of precipitation, solution
or change of valence. When such conditions exist, they have
generally been referred to as geochemical barriers. Some of the more
important barriers will be discussed below.


A. pH BARRIERS
   pH measures the ability of the environment to supply or remove
   hydrogen ions to(from) the solution.

   1. STRONGLY ACID - pH < 4

          Ore deposits; coal mines
              oxidation of pyrite

          FeS2  + 3.502 +    H20 <«>  Fe"*"1"  + 2S04= + 2H+

         2Fe++  + 0.502 +   2H20 <==>  Fe203         + 4H+

          Fe2O3         +    H20 <==> 2FeOOH
                                  OR
          FeS2 + 3.7502 + 2.5H20 <==>  FeOOH + 2S04  + 4H+

          Clays destroyed; aluminum mobile
          Many trace metals mobile (e.g.  Cu, Zn)
          Increase after rain,  varies seasonally
          Sulfate » chloride

   2. MODERATELY ACID - pH 4-6.5

          Carbonic acids

               C02 + H20 <==> H2C03   - pH = 5.6

               atmospheric C02
               oxidation of organic material
                    more C02 therefore higher pH
          Humic acids
          Partial breakdown of organic material
          Low chain organic acids
               oil field waters
          Replacement of cations by H+ ions
               cations in ground water
          Alteration of feldspars to clays
          Podzolic soils (leached A-horizon)
                            49

-------
   3. NEUTRAL pH 6.5-7.8

          Bicarbonate dominant
               cations mainly Ca and Mg
          Humid climate
               karst topography
               solution cavities
          Dry climate
               caliche layer in soil
               carbonate concretion
              + HoO
                             humid
                             	>
            H2C03 + CaC03           Ca++ + 2HC03

                              dry

          Mn often mobile as bicarbonate —> black stains
          Good buffer for acids

   4. MODERATELY ALKALINE pH 7.8-9

          Carbonates precipitated
          Many trace metals co-precipitated
          Measurable carbonate (C03=) in water
          Silica often mobile
               may replace carbon, e.g.  fossil wood

   5. STRONGLY ALKALINE - pH>ll

          Ca and Mg hydroxides
          Leaching of fresh cement; Ca(OH)2 in water
B. ADSORPTION BARRIERS

   1. MONTMORILLONITE CLAYS

          Ca/Na exchange - natural softening

                Na2-clay + Ca++ <==> Ca++-clay + 2Na+

          Reversed by saline solutions - regeneration

   2. KAOLINITE CLAY

          Anion exchange
               Phosphate, sulfate
               F~/OH~ exchange (acid)
                      reversed (neutral)
                            50

-------
   3. GOETHITE (FEOOH)

          Anion exchange
               selenate, molybdate
          May be reversed under reducing conditions


   4. NATURAL ORGANIC MATTER

          Bonding to humic/fulvic acids
               U02 > Hg > Cu > Pb > Zn > Ni > Co
          pH dependent
               Pb more strongly adsorbed under alkaline conditions
                    than under acid conditions

C. REDOX BARRIERS
   The ability of a natural environment to bring about an oxidation
   or reduction process is defined by what is called its redox
   potential. This measures the ability of the environment to supply
   electrons to an oxidizing agent or remove electrons from a
   reducing agent. Because many elements have more than one
   oxidation state and the stability of a particular oxidation state
   depends on the availability of electrons the ratio of two such
   oxidation states of a particular element in a water will also
   depend on the environment.

                 THE FEW ELEMENTS IN NATURAL WATERS

          0  - O2/H20

          N  - N03~/N02~;  N03~/N2;  N03~/NH4'f

          C  - C02/CH4

          Fe - Fe++/Fe+++;  Fe++/FeOOH,   Fe"l"l"/Fe(OH)3

          Mn - Mn"l"f/Mn02

          S  - S04S/H2S
                            51

-------
                       IRON GEOCHEMISTRY

       High natural abundance
       Ubiquitous
       Exists in two valence states
            Fe"*"*"  - ferrous (reduced)
            Pe"1"1"1" - ferric (oxidized)
       Low solubility oxides and hydroxides
       Low solubility sulfides
       At pH > 3 ferric iron insoluble
       Iron colloids have high adsorption capacity
            2Fe(OH)3 —> 2FeOOH + H20

                            Y
                          Fe203 + H20
             CARBON GEOCHEMISTRY - REDOX "BUFFER"

       organic carbon
            DO becomes depleted if NOM present
            DO may decrease from recharge to discharge in
                 aquifer
            Water may become anaerobic
                 highly reducing

                      H2S present

                      CH4 present (C —> CH4 + C02)

1. AEROBIC WATERS

       Measurable dissolved oxygen
       H2S absent
       Fe(OH)o solid and gelatinous
            adsorbed trace metals e.g. As

                 C + 02 —> C02

                 Fe"1"*"1"  --> FeOOH or Fe(OH)3

2. ANAEROBIC WATERS (1)

       Mildly reducing (Gley waters)
       Dissolved oxygen absent
       H2S absent
       Soluble Fe+-f
       Soluble arsenite (very toxic)
                         52

-------
3. ANAEROBIC WATERS (2)

       Strongly reducing
       H2S present
       Insoluble sulfides, e.g.
            Many co-precipitated heavy metal sulfides
       sulfate reduction

            2C + S04= + 2H2O <==> H2S + 2HC03~

                         Fe"1"1" —> FeS2
       fermentation

            2C + 2H20 <==> CH4 + C02

            Fe"1"1" + C02 — > FeC03
                         53

-------
             I"
FIGURE 10. REDOX ZONES
                            54

-------
                   Table 5-V11
Classification and composition t>J sediments
ui
in

M
1
M
£
*
CA
m
O
M
3
m
1
«!
o
M
0
33
2

CO
»
*^
















w
CLASS ' RESISTATES IIVDROLYSATEi
Elements SI . . A). Si, Fe
•
Minerals Quartz Clay minerals
Accessory minerals Al hydroxide*
Glauconite
Chamosite
. •






«<•*»

1.6
^
.** 1-6
'c •
^ 1.4
E
:0
^1.2
to*

•~ 0.8
V)
5
'•5 f.6
<
V
0 0.4
'c
o
o


'
._ • sofuble
•Cs
• ?b
_
•K »Ba

• 5r

• Nil' «Ca
Mn /
• /
•i; FcX
XA\g



„
•Be

*-— ^*"
OXIDATES
Fe, Mn

Hematite
Gocthite
Tyrolusite
Psilomelane









.
cations
•i
A
/
/
/
/
JbfTa.
kin • V ^— — *--" ~
^~~^~*~~'^
^ 	 '
	 soluble
* P cornplcx
an lops
: •N »S





















                                               01234-56

                                                                     ionic   charge


                                         Fig. 5-3   Ionic potential and the behavior of elements in sedimentary processes.

-------
                     CHEMICAL WEATHERING REACTIONS  211
Table I. Avenge Composition of Igneous and Some Types of Sedimentary Rocks [6J*
Element
Si
Al
Ft
Ca
Na
K
Mg
Ti
P
Mn
F
Ba
S
Sr
C
a
Cr
Rb
Zr
V
Ce
Cu
Ni
Zn
Nd
U
N
Y
Li
Co
Nb
Ca
Pr
Pb
Sm
Sc
Th
Cd
Dy
B
Yb
Cs
Hf
Be
Er
U
Sn
Ho
Br
Igneous
Rocks
285.000
79.500
42.200
36,200
28.100
25.700
17.600
4.830
1.100
937
715
595
410
368
320
305
198
166
160
149
130
97
94
80
56
48
46
41
32
23
20
18
17
16
16
IS
11
9.9
9.8
7J
4.8
4.3
3.9
3.6
3.6
2.8
23
2.4
2.4

Resistatet
(sandstone)
359,000
32.100
18,600
22.400
3.870
13,200
8,100
1.950
539
392
220
193
945
28
13,800
IS
120
197
204
20
ss
IS
2.6
16
24
19
-
16
IS
0.33
0.096
5.9
7.0
14
6.6
0.73
3.9
4.4
3.1
90
1.6
2.2
3.0
0.26
0.88
1.0
0.12
1.1
1.0
Sedimentary Rock*
Hydrotyzale*
(shale)
260.000
80.100
38.800
22.500
4.850
24.900
16.400
4.440
733
S75
560
250
1.850
290
15.300
170
423
243
142
101
45
45
29
130
18
28
600
20
46
8.1
20
23
5.S
80
5.0
10
13
4.1
4.2
194
1.6
6.2
3.1
2.1
1.8
4.5
4.1
0.82
4.3

Precipitates
(carbonates)
34
8.970
8.190
272,000
393
2,390
45.300
377
281
842
112
30
4,550
617
113400
305
7.1
46
18
13
11
4.4
13
16
8.0
9.4
_
15
5.2
0.12
0.44
2.7
1.3
16
1.1
0.68
0.20
0.77
0.53
16
0.20
0.77
0.23
0.18
0.45
2.2
0.17
0.18
6.6
212  CHEMISTRY OF NATURAL WATERS
                        Table I, continued

Element
Eu
Ta
Tb
As
W
Ce
Mo
Lu
Tl
Tm
Sb
I
Hg
Cd
In
Ag
Se
Au

Igneous
Rocks
2.3
2.0
.8
.8
.4 •
.4
.2
.1
.1
0.94
0.51
0.45
0.33
0.19
0.19
0.15
0.050
0.0036

Resistates
(sandstone)
0.94
0.10
0.74
1.0
1.6
0.88
0.50
0.30
1.5
0.30
0.014
4.4
0.057
0.020
0.13
0.12
0.52
0.0046
Sedimentary Rocks
Hydtolyzates
(shale)
1.1
3.S
0.54
9.0
1.9
1.3
4.2
0.28
1.6
0.29
0.81
3.8
0.27
0.18
0.22
0.27
0.60
0.0034

Precipitates
(carbonates)
0.19
0.10
0.14
1.8
0.56
0.036
0.7S
0.11
0.06S
0.07S
0.20
1.6
0.046
0.048
0.068
0.19
0.32
0.0018
                                                                            'Concentration in ppm by weight.

-------
            HI

            I
            to
            i
            o
            o
            H
            >
            o
en
                             +0.1
                    Hematite
                    Limonite
                    Mn oxides
                    Silica
                    Glsuconile
                                                                           8.0
                           Cslcite
                           Hematite
                           Limonite
                           Mn oiides
                                                                           fence
                               t
                               Ch
Peat
                                                                                   Cilcite
                                                                                   Organic
                                                                                    matter
Silica
Organic
 matter
Stterite
Rhodochrosite
Phosphorite
Glauconite
                                   Peat
                                   Marcasite
                             -OJ
                                                                                   CatcKe
                                                                                   Organic
                                                                                    matter
                   Organic
                    matter
                   Phosphorite
                   Pyr«e
                   Silica
                                                                                   Calcite
                                                                                   Organic
                                                                                  .  matter
   + 1.0


   +03


   +0.6


   +0.4


   +05


Eh     0





   -0.4


   -0.6


   -03


   -1.0
                                                                                             6      8
                                                                                                PH
                                                                                                                                                                  10
                                                                                                                                    12    14
                                                                 Figure 6.9  Approximate position of some natural environ-
                                                                             ments as characterized  by Eh and pH. (From
                                                                             Garrets and Christ, Solutions,  Mineralst and
                                                                             Equilibria. Courtesy Harper and Row)
                             Figure 6.71   Sedimentary associations in relation to environmental limita-
                                          tions imposed by oxidation potential and pH. (After Krvmbcin
                                          and Carrels. J. Ceol. £0,26.1952)

-------
                             EXERCISE 23
26. Arrange in order of decreasing pe (giving your reasons) the
    following environments:
     A. water from a small mountain stream
     B.  a swamp where bubbles of an odorless gas sporadically rise
         to the surface
     C. ground water containing 0.5 mg/1 Fe"1"1"
     D. water from a swamp in Northern Canada
     E. ground water smelling of hydrogen sulfide
     F. pore water from a deep lake sediment
     G. lake water above a thermocline
     H. lake water below a thermocline
                            57

-------
                     VIII. MASS BALANCE MODELING

     A mass balance is simply the sum of what was originally present
          plus whatever entered the system
          minus whatever left the system.
     For example, the number of people in a room at any instant in
time equals the number present in the room initially plus those that
entered minus those that left.

Three major processes are considered:
     1. mineral dissolution or precipitation.
          +ve results indicate dissolution
          -ve results indicate precipitation or loss
     2. variable fluxes of C02 gas
     3. the mixing of two end member waters

A. COMPOSITIONAL CHANGES

   A common application in water chemistry is the determination of
   the change in chemical composition of water samples between two
   points along a flow path. The program calculates the amounts of
   solid phases (minerals) entering (dissolving) or leaving
   (precipitating) the aqueous phase. The minerals to be considered
   as well as their chemical compositions must be specified by the
   user on the basis of geology, hydrology or mineralogy of the
   system. In addition to minerals, gases, ion exchangers and even
   other solutions may also be considered.  In order to solve the
   equations the number of phases must equal the number of elements.

   The objective in selecting phases is to provide a source or sink
   for each element in the initial and final solution. Although the
   calculated mass transfer for one or more phases might be zero,
   indicating that the phase(s) did not participate in the reaction,
   the phase(s) must still be included in order to perform the
   calculations.

   The inclusion of minerals whose composition can be derived by
   linear combinations of other minerals in the set will produce an
   error message.

   Thus if calcite,  magnesite and dolomite are all included,  the
   program will crash because CaCOg •*• MgCO3 —> CaMg(003)2

   The mass balance approach can never prove that a reaction has
   taken place, although it may indicate that a certain reaction
   could not happen as stated.

                               EXAMPLE
   In general terms:
          Consider elements x, y and z
          with concentrations in water of (x), (y) and (z)

          Also consider 3 solid phases A, B and C
          such that A = xz, B = xyz  and C = z
                            58

-------
                              MASS  BAI-ANCE
     IN
    	>
    OUT
                               WATER
DXSSOL.VE
                         COMROSITIOM
  F-RECXF1.
                               MIXIMG
WATER
    XlxlERAL-S
    I SSOL.VE
    <	
                     : COMPOSITION   < ——— >  GASES
MATER  4*2
s	>
*-M X NER Al_S
   R-RECIR-.
FIGURE 13. MASS BALANCE DIAGRAMS
                             59

-------
          The coefficients giving the number of atoms of each
          element in each phase are
                                             phases
                                    A          B        C
           water comp.    |          xz        xyz2      z
(x)
(Y)
(z)
i
0
i
i
i
2
0
0
1
          The mass balance for each element is then:

               (x) = 1*A + 1*B + 0*C
               (y) - 0*A + 1*B + 0*C
               (z) = 1*A + 2*B + 1*C
          or
               (x) =  A  +  B
               (Y) =        B
               (z) =  A  + 2B + C

          Solving for the amounts of A,  B and C

               B = y
               A = (x) - B
               C = (z) - A - 2B
B. MIXING
   In addition to the determination of possible compositional
   changes along a flow path the balance program may be used to
   determine the composition of a water resulting from the mixing of
   2 waters with or without the precipitation or solution of any
   other phases. In the simplest case it may be used to calculate
   the composition that could result from evaporation and
   precipitation, although care must be taken to avoid the
   production of a singular matrix.

C. COMPUTER CALCULATION

   BALANCE is a USGS FORTRAN computer program designed to define and
   quantify reactions between ground water and minerals.

   To aid in producing the required formatted file used by BALANCE
   another interactive program B-INPUT.EXE is used.

   Note that the units used are either moles/1 or mmoles/1,  it does
   not matter which but they must be consistent.

   The number of elements must equal the number of phases,  and the
   program will ask for the number of elements and then will expect
   the same number of phases to be input.
                            60

-------
1.
There are 3 input options:
       1. the difference between final and initial concentrations
          or one analysis
       2. the final and initial concentrations
       3. a final concentration and two end members
            Note: If the end member waters calculation ends up
            with values <0 or >1 then an impossible mix is
            indicated,  and other phases must then be chosen,
            e.g. if a± = 1.33 and a2 = -0.33 there is a problem!

                            EXAMPLES

  What minerals could dissolve to give the following water
  analysis and in what amounts?
mg/1
meq/1
mmole/1
            Ca

            32
             1.6
             0.8
                       Mg
5
0.4
0.2
       S04

       48
        1.0
        0.5
HC03

 61
  1.0
  1.0
     Consider the minerals calcite, dolomite and gypsum and CC>2 gas
          (4 phases and 4 elements)

     Once the analysis is converted to moles/1 or mmole/1 the next
step is to determine the coefficients for each phase. That is, the
number of input elements present in each phase. Thus, calcite has 1
Ca and 1 C; dolomite has 1 Ca, 1 Mg and 2 C; gypsum 1 Ca and IS;
CC>2-gas 1 C. Note that we are dealing with moles and that 1 mole
CO3= equals 1 mole C.
water composition
 (mmole/1)

0.8               Ca
0.2               Mg
0.5               S
1.0               C
                                    Phases
                         number of elements in each phase
calcite
   1
         dolomite
            1
            1
                                        gypsum
                                          1
               CC>2 gas
Sulfur balance
          gypsum   = 0.5
Magnesium balance
          dolomite = 0.2
Calcium balance
          calcite  = 0.8 - 0.2 (dolomite) - 0.5 (gypsum)
                   = 0.1
Carbon balance
          C02 gas  = 1.0 - 0.1 (calcite) - 2 * 0.2 (dolomite)
                   = 0.5
     Thus the above water could be obtained by dissolving
          0.5 mmole/1 gypsum
          0.1 mmole/1 calcite
          0.2 mmole/1 dolomite
                            61

-------
2.
     0.5 mmole/1 C02

What proportions of the two end member waters are necessary and
what minerals would dissolve or precipitate to give the
 following water composition. The analyses are given in
 mmoles/1.
Fina:

7
9
32
1
L

Ca
Mg
C
Mix
Init 1
Wl
8
6
28
1
Init 2
W2
9
10
38
1
Calcite
C
1
0
1
0
Dolomite
D
1
1
2
0
Mass balance for Ca
     7 = 8W1 + 9W2 + C + D         (1)
Mass balance for Mg
     9 = 6W1 + 10W2 + D            (2)
Mass balance for C
     32 = 28W1 + 38W2 + C + 2D     (3)
Mass balance for end members
     1 B HI •+ H2                   (4)

Thus
     from (4)                      W2 •
     substitute in (2)
          9 = 6W1 + 10 - 10W1 + D
     or                            D B 4W1 -  1
Substitute (5) and (6) in (1) and (3)
     (1)   7 = 8W1 + 9 - 9 Wl + 4W1 - 1 + C
          -1 = 3W1 + C   (7)
     (3)  32 = 28W1 + 38 - 38W1 + 8W1 - 2 + C
          -4 = -2W1 + C  (8)
                                        1 -  Wl
                                            (5)


                                            (6)
Subtracting (8) - (7)
           3 = 5W1
                    or Wl =
                       W2 =
                        D =
                                  0.6
                                  0.4
                                  1.4
                             C = -2.8
     Therefore final water could result from mixing 60% of water 1
and 40% of water 2,  dissolving 1.4 mmole/1  dolomite and
precipitating 2.8 mmoles/1 calcite.
                            62

-------
                             EXERCISES 24 & 25


23.  Calculate by hand the amounts of minerals which when dissolved
     would give a  water of the following composition.

          Na        K         Ba        Cl        HCOo
mg/1      46        20         7        71         37

     BaCC>3 is witherite, NaHCOg is nahcolite


24.  Three water sample are thought to be related by mixing. Verify
     or negate this hypothesis.

          Na     K       Ca      Mg      HCOo    S04     Cl
1. mg/1    5     .4      12      1       18      20       7
2. mg/1   30    8         61       61      20       4
3. mg/1   22    6         8      1       48      33       5

     In this example two end-member waters mix in unknown proportions
and, in addition, phases dissolve and precipitate to produce a final
water. The two initial waters are treated exactly like other phases
and ai is the fraction of solution 1 and a2 is the fraction of
solution 2, which combine, along with mineral reactions, to produce
the final solution. An additional equation is automatically included
to ensure that the 2 fractions are equal to I.

                i.e. a^ + &2 = 1*

As a result the number of phases that can be included in the
calculations (other than the solutions) is the number of elements
minus 1.

        In this example you may want to consider ion exchange;

                 NaX + Ca++ — > 2Na+ + CaX
        In this case the coefficients are,  Na =2, Ca = -1
   That is, two Na ions are added to the water for every Ca removed.
                            63

-------
           XI. WATER GEOCHEMISTRY - SELECTED BIBLIOGRAPHY

Drever, J. I., 1988. The Geochemistry of Natural Waters. Second
edition. Prentice Hall. 438 p.

Eriksson, E., 1985. Principles and Applications of
Hydrogeochemistry. Chapman and Hall. 187 p.

Faust, S. D. & O. M. Aly, 1981. Chemistry of natural waters. Ann
Arbor Science. 400 p.

Garrels, R. M. & C. L. Christ, 1965. Solutions, Minerals &
Equilibria. Freeman, Cooper & Company. 450 p.

Garrels, R. M. & F. T. MacKenzie, 1967. Origin of the chemical
compositions of some springs and lakes. In: Equilibrium concepts in
natural water systems. Am. Chem. Soc. Ser. 67, p. 222-242.

Gibbs, R. J., 1970. Mechanisms controlling world water chemistry.
Science 170, p. 1088-1090.

Goldschmidt, V. M., 1954. Geochemistry. Clarendon Press. 730 p.

Hem, J. D. 1985. Study and Interpretation of the Chemical
Characteristics of Natural Water. Third edition. U. S. Geological
Survey Water-Supply Paper 2254. 264 p.

Krauskopf, K. B., 1979. Introduction to geochemistry. Second
Edition. McGraw-Hill Book Co. 617 p.

Krumbein, W. C. and R. M. Garrels, 1952. Origin and classification
of chemical sediments in terms of pH and oxidation-reduction
potentials. The Journal of Geology, v.60, p. 1-33.

Levinson, A. A., 1974. Introduction to exploration geochemistry.
Applied Publishing Ltd. 614 p.

LLoyd, J. W. & J. A. Heathcote, 1985. Natural Inorganic
Hydrochemistry in Relation to Groundwater: An introduction.
Clarendon Press. 297 p.

Mason, B. & C. B. Moore, 1982. Principles of Geochemistry. John
Wiley & Sons. 344 p.

Parkhurst, D. L., L. N. Plummer and D. C. Thorstenson, 1982. BALANCE
- A computer program for calculating mass transfer for geochemical
reactions in ground water. U.S. Geological Survey, Water-Resources
Investigations 82-14.   p.

Perel'man, A. I., 1986. Geochemical barriers. Applied Geochemistry,
v. 1, p. 669-680.

iper, A. M., 1944. A graphical procedure in the geochemical
 interpretation of water-analyses. Amer. Geophys. Union Trans., v.
25, p. 914-923.


                            64

-------
Plummer, L. N., B. F. Jones & A. H. Truesdell, 1976. WATEQF- a
FORTRAN IV version of WATEQ, a computer program for calculating
chemical equilibrium in natural waters. U. S. Geol. Survey Water
Resour. Invest. 76-13, 73 p.

Rankama, K. & T. G. Bahama, 1950. Geochemistry. University of
Chicago Press. 912 p.

Stumm, W. and J. J. Morgan, 1981. Aquatic Chemistry: An introduction
emphasizing Chemical Equilibria in Natural Waters. Second edition.
Wiley-Interscience. 780 p.

Thurman, E. M., 1985. Organic Geochemistry of Natural Waters.
Martinus Nijhoff/DR W. Junk Publishers. 497 p.

Truesdell, A. H. and B. F. Jones, 1974. WATEQ, a computer program
for calculating chemical equilibria of natural waters. Jour.
Research U. S. Geol. Survey v. 2, p. 233-248.
                            65

-------
                        IT!
                   MG/L    .

SODIUM        5.0"    4.8    8.3     5.4
POTASSIUM    0.5    ..1.1    0.9     1.2
CALCIUM       0.6    19     33      1.5
MAGNESIUM    0.5     0.9     1.5     0.6
CHLORIDE      89     10      7
SULFATE       4544
ALKALINITY    2     47     85      7
SILICON      .  1.8     3.3    3.4     2.4
                   TABLE IV

-------
mm  O
  O
lO
     c/j
   N>
                   MW-6
                             PW   to OW-1 is
                             OW-1 to OW-2 is
                             PW   to OW-2 is
                           35 feet
                           15 feet
                           50 feet
            ESTIMATED EXTENT
            OF GROUNWATER
            CONTAMINATION PLUME
                 STREAM
              PW  to MW-4 is 190 feet
              MW-4 to OW-3 is 60 feet
              PW  to OW-3 is 250 feet

              PW  to OW-4 is 15 feet
              OW-4 to OW-5 is ^Wfee
              OW-5 to OW-6 is 65 feet
              PW  to OW-6 is 100 feet
•  MWM MONiTQRWG WELL

•  PW  , PUMRfNG WELL

A  OW-M O0SERVAT
-------
29
28'
27'
    EXPLANATION
           •
    Sampling point
    	75	
    tinei of «qoo/
      above sea /ere/,
      in mefers
      0 10 20 30 40 50 Km
      i	i	•_ L—i—•
              83

-------
                                        TABLE  1
Standard chemical analyses of water from the FloriJan aquifer. Wells are listed from north to south
Milligrams per liter
Well
location
Ocala 4
Wildwood 2
Grovel and
Polk City
Fort Meade
Wauchula
Arcadia
Cleveland
V}/VU
interval, Temperature
meters °C
30-115
39-82
40-180
6-172
127-291
1 14-245
100-151
39-152
24.5
23.8
23.7
23.8
26.6
25.4
26.3
26.7
Si02
10
10
11
0.2
16
18
31
18
c.«
96
51
42
34
56
66
106
114
-
15
2.6
4.1
5.6
17
29
60
82
-
7.8
4.7
3.6
3.2
6.1
8.3
21
283
K+
1.0
0.2
0.5
0.5
0.7
2.0
3.7
9.6
Field
HCO,
175
150
143
124
163
168
206
145
so;.
148
3.2
1.6
2.4
71
155
344
216
ci-
11
8.0
6.5
4.5
9.0
10
28
655
F-
0.3
0.2
—
0.1
0.4
0.7
2.2
0.9
Dissolved
solids,
residue
N0« at 180'C
1.6
3.8
0.1
O.I
0.1
—
—
0.1
420
158
148
138
272
392
762
1600

Field
PH,
±0.02
7.50
7.59
7.80
8.00
7.75
7.69
7.44
7.51

-------
 nesses ot
 e presence o
 for the
 i contrast,
  of the
• jor in the
 raulic
 circulation
 • within a
 i, the respective
 •scharge to
 ited near the
 ove river
 NGS
 that range
 ew gal/min
 s 30 and 34.5.
 jure 2 and
 ;gs become
 • high river
 10 ft3/sec
 of the
 in the
 1 and discharge
 vere first
 ange (1956,
     springs
    Structural
     Control
in
ro
e
(O
IO
 O
 to
 to
 10
Sompled Spring

    ®—
  River Mile


Rim of Canyon

	U^
      0
   Fault
                   -M-
                 Syncline

                   v*
                Dtp, Degrees
                                   0
                                   I
                                          Miles
                                     >
                                       North
                ,*VX':
                  'rt*
                                                       &
                                                  ^
                                \
                                y
 Fence Fault


-------
                  Table 1. Location, Estimated Discharge, and Geologic Setting of Springs Sampled
                            August 8, 1979, in the Vicinity of Vasey's Spring, Arizona


Approx.
No* Name River Mile"
1
2
3
4
5
6
7
8
E. Fence No. 1
E. Fence No. 2
W. Fence No. 1
W. Fence No. 2
Diagonal
Vasey's
Hanging No. 1
Hanging No. 2
30.1
30.2
30.2
30.7
30.9
31.9
34.4
34.5
Side of
Colo. River
East
East
West
West
West
West
West
West
Estimated
Discharge
(gal /m in)
500
6500
20
30
900
2500
30
10
Producing
Unitc
Mrm
Mrm
Mrm
Mrm
Mrm
Mrm
Mrw
Mrw
Structural
Control
Fence Fault
Fence Fault
Fence Fault
Fence Fault
Joint-Fence Fault
Joint
Joint
Joint
a  Numbers are identical to numbers used on Figure 2.
b  Distance measured from Lee's Ferry, Arizona.
c  Mrm = Mooney Falls Member, Redwall Limestone. Mrw = Whitmore Wash Member, Redwall Limestone.
   Nomenclature from McKee and Gutschick (1969).

              Table 2. Temperatures and Concentrations of the Major  Cations and Anions in the Water from
                          Springs in the Vicinity of Vasey's Spring, Marble Canyon, Arizona1
                                       Cations (meq/l)
                                     Anions (meq/l)
No.1
          Name
                     Temp.
Ca    Mg
Na
      Total
K    Cations
Cl
                        Total
F    SO4  CO3 HCO3  Anions
1
2
3
4
5
6
7
8
E. Fence #1
E. Fence #2
W. Fence #1
W. Fence #2
Diagonal
Vasey's
Hanging #1
Hanging #2
69
70
71
70
71
62
65
64
7.49
7.49
5.99
2.10
1.60
2.00
2.50
2.35
3.45
3.54
3.54
1.73
1.56
1.56
1.56
1.56
10.87
11.31
8.70
0.61
0.09
0.06
0.04
0.07
0.56
0.64
0.49
0.04
0.03
0.02
0.03
0.02
22.37
22.98
18.72
4.48
3.28
3.64
4.13
4.00
9.90
10.43
7.92
0.51
0.06
0.07
0.07
0.07
0.01
0.02
0.01
0.01
0.01
0.02
0.05
0.00
0.03
0.03
0.02
0.01
0.01
0.01
0.01
0.01
4.63
5.17
3.70
0.45
0.17
0.27
0.41
0.72
0
0
0
0
0
0.40
0
0
8.89
9.06
7.93
3.80
3.31
3.24
3.56
3.56
23.46
24.71
19.58
4.78
3.56
4.01
4.10
4.36
          es collected August 8, 1979. Chemical analyses byJ-ision of Laboratories, Wyoming Department of Agricuhure,
       amie, Wyoming.

-------
  T
  o
  t
  a
  1
  H
  a
  r
  d
  n
  e
  s
  s
Na

Ca
+
Mg



Permanent
Hardness

Temporary
Hardness
Cl
S04
C03
HC03



Alkalinity
T
o
t
a
1
H
a
r
d
n
e
s
s
Na
Ca
MB


Temporary
Hardness
Cl
S04
C03
HC03


Alkalinity

-------
       WATEVAL
                        DRIVE
MEMORY
^V \
\\1
	 ' lUIMU
X/FILE
SCREEN

B:
C:
          T ANALYSIS
       KEYBOARD
              WATEVAL
OUTPUT
   RAM - MEMORY - DEFAULT
   FILE
    SELECT OPTION
    CHOOSE DRIVE
    NAME FILE
    SAVE EACH ANALYSIS
              WATEVAL
INPUT
  KEYBOARD - DEFAULT
  RAM
     SELECT OPTION
     CHOOSE SAMPLE
              WATEVAL
INPUT
  FILE
    SELECT OPTION
    CHOOSE DRIVE
    SELECT FILE
    CHOOSE SAMPLE
                WATEVAL
  OUTPUT
    SAVE RAM TO FILE
       SELECT OPTION
       CHOOSE.DRIVE
       NAME FILE
       SAVE RAM
       CLOSE FILE
       RESET OPTION

-------
               CHLORIDE



Cl


1
I

Na Na Na
halite sink source
                      reverse  silicates
                     exchange
                               cation
                              exchange



s
SULFAT


[E
r


I
Ca Ca Ca
source sink
gypsum
         silicates   precip.
        carbonates exchange
              SODIUM
Na         Si       Si        Si

        Ab •> M   Ab - K   Or - K + M
                                                                     CALCIUM



la





Bi
Cc
An







Si Si Si
Di tremolite
no clay .. K -» M



la



SODIUM
1


Si
source
Fe/Mg
silicates

l'
Si
Na
source
cation
exchange
                                                                     BICARBONATE







1


1
Bi Ca Si Si
1
Si
                                                                calcite  albite diopside olivine
                                                                        k-spar

-------
    DAY 2
UNSATURATED
    ZONE
  MODELING
     RITZ

  CHEMFLOW

  CHEMRANK

-------
                    Unsaturated Zone Modeling
Introduction to Models

Water Movement in Unsaturated Soils
  - Introduction
  - Infiltration: Film of Physical Model from
                  Washington State Univ.
  - Infiltration: Simulation Using CHEMFLO
  - Redistribution: Simulation using CHEMFLO
  - Evaporation: Simulation using CHEMFLO

Water and Chemical Movement in Unsaturated Soils
  - Introduction
  - Non-adsorbed Chemicals Using CHEMFLO
      - Steady-state Water Flow
      - Transiant Water Flow
  - Adsorbed Chemicals Using CHEMFLO
  - Non-Uniform Initial Conditions Using CHEMFLO

Management Models:
  - CHEMRANK
  - RITZ

Concerns in Field Scale Use
  - Assessing Validity of Assumptions Within Models
  - Natural Variability within Soil
  - Uncertainty in Model Parameters

-------
                      Some Interesting Reading Material
Garden, W.H.  1977.  Historical highlight of American soil physics.  Soil
     Science Society of American Journal, Vol. 41, pp. 221-229.

Baveye, P. and G. Sposito.  1984.  The operational significance of the
     continuum hypothesis in the theory of water movement through soils
     and aquifers.  Water Resources Research, Vol. 20, pp. 521-530.

Donigian, Jr., A.S. and P.S.C. Rao.  1986.  Overview of terrestrial processes
     and modeling.  lr± "Vadose Zone Modeling of Organic Pollutants", S.C.
     Hern and S.M. Melancon (Eds.), Lewis Publishers, Inc.

Wagenet, R.J. and P.S.C. Rao.  1985.  Basic concepts of modeling pesticide
     fate in the crop root zone.  Weed Science, Vol. 33, Suppl. 2, pp. 25-32.

Rao, P.S.C. and R.J. Wagenet.  1985.  Spatial variability of pesticides in
     soils:  Methods for data analysis and consequences.  Weed Science,
     Vol. 33, Suppl. 2, pp. 18-24.

Rao, P.S.C., R.E. Jessup, and A.G. Hornsby.  1982.  Simulation of nitrogen in
     agro-ecosystems:  Criteria for model selection and use.  Plant and
     Soil, Vol. 67, pp. 35-43.

-------
     How   Water   Moves   in   the   Soil
                                                                      by Walter  H. Gardner
                                                                      Time-la pie  photograph y
                                                                      condenses hours of  water
                                                                      movement in  soil into a \cw
                                                                      minutes. This time reduction
                                                                      helps show important principles
                                                                      of water movement in different
                                                                      soils. It also reveals tchat
                                                                      cjjecls layers such as  claypans
                                                                      or coarse  sands have on water
                                                                      movement.  Farmers  should
                                                                      consider these principles in
                                                                      fertilizer  placement and
                                                                      irrigation practices  .  . .
 scs
                   Part  l-The  Basic   Concept
    HOW WATHR CONTINUALLY MOVJ-S  in the soil
    is one of the most basic -and important agricultural
concepts.
  It is from an understanding of water movement that we
learn how to solve such problems as supplying  crops with
adctjuate water and nutrients.
  Water itself is an important part of plants. It makes up
almost 85 percent of the fresh weight of growing plants.
Thus, water  is not only a plant constituent but  is a carrier
for mineral  nutrients anil gases entering the plant.
  Water is  nearly always moving in  the soil,  cither as a
liquid or a vapor. In addition to moving downward follow-
ing  rain  or  irrigation,  water moves upward to cvajwrate
from the soil surface, or into plant roots and  eventually
into the atmosphere through transpiration.  Under  many
conditions some water is lost from the root zone  through
deep drainage. Water and water vapor can move horizon-
                   The Author

  Walter II. Gardner is professor of soils .it Washington Sl.-ite
University,  Pullman. Wash., where he specialises in the field of
soil physics. The pictures  incluilccl with this article  are taken
from the time-lapse motion picture, "Water Movement in Soil."
This 26-minnte.  lo-niillimcter color film, with sound, was  pro-
duced at Washington  Stale University  hy Gardner and J. C.
Hsich. More than -10 copies of (his film have hccn sold in (his
country and abroad. It  may he purchased from the Washington
State University  Agronomy Club or rented through  numerous
audio-visual centers, or. in  (he West, through the NV'SU Audio-
Visual Center.

October  1962
 tally, upward or downward, depending on conditions. Only
 rarely is soil  water completely stationary.
  Water (lows through open pores between soil particles.
 In an ordinary silt loam the pore space makes  up  about
 one-half of the soil volume. For most crops, some air must
 be present  in the soil  pores. It must  be possible for fresh
 air rrom the surface to exchange with  carbon dioxide  laden
 air from the  root zone.
  Soils vary as to pore size and number. Silty and clayey
 soils generally have more total pore space but smaller  pores
 than sandy soils.  And,  when all of the pores are  filled with
 water, silly and clayey soils contain more total water than
 sandy  soils,  simply because the heavy soils have so  many-
 pores. Although much  of the water in soils with fine  pores
 is unavailable to plants, the  amount availtible is still greater
 than  that available to  plants in  soils with large pores.
  When a  soil is near saturation,  the larger the pores the
greater the rate of flow per unit  of  applied  force. How-,
ever,  when a soil is not saturated, large pores  contribute
 little to flow. Water moves on particle  surfaces  and through
the liner pores under these conditions.
  Two forces cause liquid  water  to  move through  these
soil pores:  Gravity and adhesion.  Gravity  causes a down-
ward  pressure on  water. This  force is most important in
saturated soil.  The second  force,  adhesion, is due to the
attraction of soil  particle surfaces for water and becomes
important in misaluntied soil. Adhesion—together with co-
hesion, which causes water molecules  to hang  together—is
responsible  for water  rise  in capillary tubes  and the ab-

-------
1                 UNIFORM  OR HOMOGENEOUS  SOIL—Water
                 added  to  the center  of  this  dry,  homogeneous  soil.
             Note that water moves out almost equally in all directions.
             Gravity has small effect  as  seen  by the slightly  greater
             downward  wetting. This illustrates wator mi>vctiu'iit under
             iintnlurnlcd conditions where  there  is no free water in the
             soil. Under inttiraleil  conditions, or  as  .saturation  is  ap-
             proached,  gravity begins  to  play  a much  greater role in
             water movement.
             r-;,-,  •   ,         .- '•••••• -
             *•'.!•-•.' '    •  .  • •  •*"  ~
                                                                                                                 J
                                                                           O  CI.AYI'AN I.AYKU—When water readies a claypan.
                                                                           •^  this  layer  with very  line  pores  resists water  lln«
                                                                           due  to Ihc  liny transmitting channels. Although clatpuns
                                                                           (In wet up, they transmit water so slowly that »ali-r lahli-s
                                                                           often build  up above  llu'in.  Some  plow pans  would  act
                                                                           similarly, creating  inlillraliun  problems.
il
sorptive properties  of  blotting paper nnd  other  porous
materials.
  When soil is very wet the gravitational force  predomi-
nates.  But in soil in which most  crops  grow, the  major
force causing water  to  move is due to adhesion.
  Water moves until these forces are balanced  in the soil.
Water films  on soil  particles will  be  uniform  throughout
any homogeneous  soil,  except for some vertical differences
that  exist because  of  gravitation. If  the soil is not uniform
or homogeneous, water  is retained  most strongly  by those
portions of the soil having the smallest pores.
  In stratified soil—that in which various  "layers" exist—
water flow is greatly affected  by the sizes of pores in strata
encountered by an advancing wetting-front. II  fine  mate-
rials  are encountered, the rate  of advance may be slowed
by resistance to Ilo\v in the extremely fine pores.  Bui wau-r.
nevertheless, docs continue to move. 11  coarse materaU arc-
encountered, water  movement stops until the soil becomes
nearly  saturated.  Any porosity change occurring  in  soil
alTects the rate of water (low.
  Water  retained in stratified  soil  for  plant use is  greatly
alTcctcd by the nature of the strata. Generally, then, strati-
fied  soils  hold more water for plant use than uniform soils
of similar nature  because of  the effect  of porosity crum:es
on water (low.
  The  pictures  accompanying this  article show  some basic
principles  of  water  movement  using artificial soil profiles.
These  principles,  though not always striking, may  be  ob-
served in  your field if you take time to examine soil soon
after water has been applied.
            8
                                                                                                           Crops i Soils

-------
3    SAND LAYER—When water passes throuch a Tine
    soil  and reaches  a layer of coarse sand it stops un-
 til  sufficient water accumulates to  nearly saturate the
 overlying fine Roil. Water will move readily through the
 large pores in tlic  sand only under near-saturation condi-
 tions. The saturation  of the overlying soil works mncli tlic
 same as adding  water to a piece of Moiling paper which
 must  become saturated before  it will drip.
                                                       JUi
                                                                         ?> \-£f'-':;  &-}'-'-\.;--. > .
4    COARSE  SAND OR GRAVEL  SUHS01I/—This  is
    the rflTi-cl  mi water movement when a line soil over-
 lies a coarse sand or gravel suhsoil. Note that G hours of
 wetting  are required before soil water moves down into
 the gravel.  The overlying soil must become very wet be-
 fore water will move down through the large pores in the
 gravel. Under  these conditions, the overlying soil will hold
 considerable water—up to 2 or 3 times as much  as would
 the same material if the  coarse  sand and gravel  were mil
 present.
U  LAYER OK COARSE SOIL AGGREGATES IN KINK
vJ SOIL—Any change  in  soil  porosity  encountered  by
a wetting-front affects  water movement. In these three.
photographs, a layer of coarse soil aggregates acts much
the  same  as a layer of sand, hut with one important  dif-
ference: Water can move through the interior of the  soil
aggregates  themselves.  Hut the  amount of water that
moves through the layer of soil  aggregates is limited  hy
the  relatively small number of contacts  between soil  ag-
gregates.  Water  moves rapidly  through  the soil aggre-
gate layer only under saturated conditions. In this lest,
such saturated conditions were not reached.
              Part  II—See  Next Month's  Issue

                                                                Jr 1  '  *i '- ) '"^ ' :>







                                                         4.  	TM^»F* ~  <-._^6-»'U**£U.\*—
 T~"TP*is
JL -. Jfeai
October 1962

-------
 How  Water   Moves   in   the   Soil
 by Walter H. Gardner

            The Author
  Walter H. Gardner is professor ol soils
 at Washington  Stale University.  Pullman.
 Wash. The first part ol the article was pub-
 lished in the October I9o: issue of  CHOI'S
 .v Sou., bejjinninc «n p.ij!i- 7. The  author
 extends credit to Robert II. Kunkvl, \VSU
 hortictitturjlt.tl.  win* developed  the new
 method lor  pl.miin.c  potatoes.

     WAT12R IS NEARLY ALWAYS
     moving in I he soil, as was pointed
 out in Part J of this article. Through
 time-lapse  photography, .scientists art-
 able to condense  hours of soil-water
 movement  into a  few minutes.  Their
 findings highlight  important principles
 that  farmers can use during irrigation,
 and  in seed and fertilizer placement.
  Part I pointed oul that when  water
 reaches a claypan. movement is slowed
 because of  small soil pores. Likewise.
 when water moves  through a line soil
 snd  reaches a  coarse saTilf or gravel
 layer, water movement is restricted by
 the coarse  layer.  In fact, movement
 stops until  sufficient water accumulates
 to nearly saturate  ihc overlaying fine
 soil.
  Straw or organic matter layers, and
 vertical mulched straw,  act much the
 same way  as a layer of coarse sand or
 soil aggregates when the mulch  is no
 longer  exposed to the soil surface or
 !o free water.
  Practical  applications   of  such re-
 search  can  he  observed. Radial, up-
 ward, and  downward  movement of
 soluble fertilizers may be followed by
VERTIC
                                                               L'SDA
        Part  ll-ln   The   Field
the use- of dye tracers. I'rom basic re-
search, loo. a new method lor planting
potatoes has been  developed  to  take
lull advantage ol available soil mois-
ture and to insure complete welting ol
the pot.Uo hill by irrigation.
  A press-wheel  method places jxitato
seed pieces ju.st below the surface of an
inverted  "V" shaped  mound.  The
usual way ol planting is  to put the
seed  piece in ihe  center  of a high,
wide ridge ol soil.
  The  press wheels  firm the coarse
soil aggregates, reducing the size ol
the pores and allowing water to move
into the ridge.  Without this compac-
tion,  the coarse soil aggregates often
stop water movement into the ridge,
leaving new tubers in dry soil.
  By knowing the principles  of soil-
water  movement,  farmers often  can
take advantage of  new techniques  that
lit their soil conditions and availability
of water.
6    VERTICAL MULCHING—Here, deep vertical chan-
    nels  arc cm in  the soil and Tilled with chopped  or-
ganic,  matter.  If  channels  remain  o|icn  (o  the surface,
large pores in  the organic nialorinl.s rc-flily lake the free
water  from rain or  irrigation and transmit it deeply into
I ho soil. Tin1 n. it is absorbed under nnsaturaled  condi-
tions.  Hut if  channels arc  not open to the  soil surface.
vertical mulching  docs little good.  Organic mailer chan-
nels, therefore, should be exposed  In Hie  source of  free
water—rain or irrigation.
  Holes  left in the  soil  by  angleworms, rodents, or  de-
caying crop  residue  behave exactly the. same as vertical
mulch  channels: if they remain open to the surface and
exposed to free water, they carry  water readily, lint these
holes cannot transmit water  unless  they arc  directly con-
nected to Ihc  surface  or n source  of  free water.  These
open channels  or holes also serve a useful purpose under
conditions where aeration may be poor. They permit  cx-
chnngc of gases between soil atmosphere  and air above.
November

-------
             WATUi AtVUCD TOO RAPIDLY
                            ,-'•-•.:•;:;,£.#. Y.1
                                                                  7STKAW  OR  ORGANIC  MATTEU LAYER—A layer
                                                                  or  straw or other coarse organic  matter acts much
                                                              the same as a layer of sand or  coarse  soil aggregate*.
                                                              Note  how  the  straw plowed under and left  in  a layer
                                                              forms a barrier to the downward movement of water, lint
                                                              if  straw is worked  into the soil, this helps to  promote
                                                              Rood soil aggregation and to maintain  :i  porous  structure
                                                              in  contact  with applied  water.  Downward movement of
                                                              rain or  irrigation water thus is aided.  Large pores—nude
                                                              by the incorporated  straw—speed  downward transmission
                                                              of  free  water.
8                                                                  SOIL  TEXTURE  AND  INr'ILTKATlON—Soil  tex-
                                                                  ture has a significant  effect on water movement. In
                                                              the  photo, water  was applied  :il a uniform rate and time
                                                              to  three soils. Note that  infiltration  and  advance of the
                                                              wetting front is more rapid in a sandy soil than in either
                                                              a  loam or  clav soil.
9                                                                 SOIL TEXTURE  AND  WATER-HOLDING  CAPAC-
                                                                 ITY—Clay  soils  will  hold   more  lulnl  water  than
                                                              either loams or sands,  however. In  this case,  the same
                                                              amount of water was applied  to each soil. Note  the  rela-
                                                              tively smaller portion of the clayey soil re(|iiired to hold
                                                              the water compared with sand.  Silt loam  and clay  loan:
                                                              soils  arc likely  to  be better  soils for  dryland  farming
                                                              than  the coarser sandy soils. This is because  of the larger
                                                              amount  of  water that can  be  retained  in  the  silly  and
                                                              clayey soils. Under  irrigation,  however, such soils are nut
                                                              as  good as sandier  soils  because of poor water-transmit-
                                                              ting  properties of heavier soils.
 1  A  UNEVEN  SURFACE  SOIL—Water  penetration
 J- \J  into soil  is  a (Tec led by  an  uneven  surface, such
as in rolling  or hilly terrain. Where water is applied more
rapidly than  it can infiltrate, it  runs off of high spots and
accumulates  in low  places. Here, the  greater amount  or
water  penetrates to  greater depths. Where soils are level
or where  the application rale does  not  exceed the infiltra-
tion  rate,  uniform wetting results.
10
                                                                                                        Crops  & Soil*

-------
 1 *~\ CLODDY  MOUNDS CUT WATER MOVEMENT—
 JL L* Coarse soil thrown up into hills between irrigation
 furrows  is often dillicult to wet  because of excessively
 large pores in  the  soil.  These pores move water slowly
 under unsaturated conditions.  Here,  in a potato Held, soil
 on tops of the  hills remains dry even though  water had
 been present  in the  furrows for several days. Under  these
 conditions, tubers must grow in dry  soil. An  inverted "V"
 shaped mound, compacted to reduce the pore si/.c, may help
 to solve this  water  movement  problem.                •*-
n                                                                    SOLUBLE FEKTILIZEKS MOVE WITH VVATf
                                                                    —Direction  of  water movement  near  irrigati-
                                                              furrows is indicated  by dye tracers. The movement
                                                              \vatcr and soluble fertilizers is almost radially away frr
                                                              the point  of  water  application in the furrows. After  t
                                                              wetting fronts join, direction of flow changes slightl
                                                              Above the water level  in furrows,  movement  is upwa
                                                              toward the drier soil.  Below the free water  level, mo\
                                                              mcnt of soluble materials is downward. Also, water cvapc
                                                              aline from Ihc soil surface causes an upward movement
                                                              soluble materials in the soil solution.
                                                                                                       Photo by Kun
             Potassium-Supplying  Power  Linked  to   Clay  in  Soil
  A better measure of potassium-sup-
plying power of soils now may be ob-
tained by considering total potassium
(K) of the clay fraction—in  addition
to  regular  soil  tests,  according  to
Michigan State University studies.
  Although soils often contain 33,000
to 50,000 pounds of K per acre in the
plow  layer, 90 to 98 percent is unavail-
able to plants. Most K occurs as part
of the primary soil minerals.  The bal-
ance exists in two forms: Readily avail-
able and  slowly available.  Doth are
measured by  availability  to plants.
  The readily-available portion, 1 to 2
percent of  the total, is the immediate
source of K for plant  nutrition. It  is
made  up  of the soluble K in soil water
solution  plus  the  K  attached to. the
surface of clay and organic matter par-
ticles.  This attached K often  is called
exchangeable potassium, since calcium
or magnesium may replace  it. Routine
soil tests measure readily-available  K.
  Slowly-available  K,  1 to  10 percent
of the total, is  an  important reserve
supply of K for  plants. But, soils men

November 1962
have long had great trouble in measur-
ing this  form.  It is thought  to  be
trapped in the crystal structure of clay
minerals,  and   is  often  called  non-
exchangeable potassium.
  As  plants use  readily-available  K,
exchangeable  K is released  gradually
to the exchangeable  or  soluble form.
Some  soils, however, release more of
it than others. And, this is an impor-
tant  factor  when  potassium fertilizer
recommendations are  made.
  Rough estimates of  noncxchange-
able  K arc made  by  washing the soil
sample with a salt solution to  remove
all exchangeable K. Then, the  soil is
cropped  for several months to deter-
mine  K  uptake by  plants  from  the
noncxchangcable portion. The total of
readily-available  and  slowly-available
K is then estimated as the potassium-
supplying power of the  soil.
  In Michigan research aimed at find-
ing a more accurate measure of potas-
sium-supplying  power,  clay and  silt
portions  in samples of six soils were
separated. Samples were  washed with
a magnesium chloride  solution  to  re-
move exchangeable  K  and  cropped
with wheat for 12 weeks.
  Afterwards plants were analyzed for
K. All K found in the wheat was con-
sidered  to  have  come  from the non-
exchangeable portion—that previously
trapped in clay crystals. There  was a
very close  relationship  between plant
uptake of K and total  K  of clay but
not of the  silt. Silt may be an impor-
tant  factor  in some soils.
  A better  understanding of the potas-
sium-supplying  power  of  soils now
may be obtained, when total K  in the
clay  fraction is known. This procedure
is much quicker and  cheaper than the
long-time  cropping method now used
to determine the  potassium-supplying
power of soils.
  Such  knowledge   greatly increases
the accuracy of fertilizer recommenda-
tions made from  soil  tests  for ex-
changeable potassium.—E.  C.  DOLL,
M.  M. MORTLAND, and K. LAVPTON,
Michigan State  University, East Lan-
sing, Mich.
                                                                                                               11

-------

-------
H  =
                                  — oo < h  <
Y-C«
                                     ^'t   'Pe>4<,^-ktJ2
    K =•
                                                  ^  K

-------
g
0)
4J
I
o
•H

       0.30-
       0.10
               Bulk Density  (g/cm  )
Cobb Loamy Sand       1.54
Cobb Sandy Loam       1.61
Teller Loam           1.67
Zaneis Loam           1.50
Port Sllty Clay       1.60
            Figure 18.
         40                 80                 120

                        Negative Pressure Head (cm)  H^X
    Soil-water characteristic curves fo-  "ive Oklahoma soils,
                                         —•  r
                                                                                       loU
/uu

-------
   Physical Characteristics of Soils
      of the Southern  Region-
Bethany, Konawa, and Tipton Series
                  D.L. Nofziger

                  J.R. Williams

                  A.G. Hornsby

                  A.L Wood
 Project Number S-124 "Movement and Retention of Water and Solutes in Selected Southern Region Field Soils'

-------
                                                                                                   45 CM
                                  V*.  SILT
                                  "A
                                                                 SILT   CLAY   /
                                                                                           v\     \
    •W  80  70  60   50
                                             360/
                            o  SILT   CLAY   X
* SILT   CLAY   X
                 SO  40  JO   ?0  10
                                                   SAND
                        SAND
Figure 1.  Particle size distributions for Bethany soil.

-------
                           30
                           60
                       u
                      Q.   90
                      LU
                      0
                          120
                          150
                             1.4
i.:
1.6
i.;
1.8      1.9
                                         BULK DENSITY, g/cm
                                                               3
Figure  2.  Mean bulk density as a function of depth for Bethany soil sites.
                                                   21

-------
          30
          60
          90
         120
         130
          30
      Q.
      Ul
      Q
          90
          120
          150
           30
           90
          120
          150
  BETHANY

  SITE tl
                                                       30
                                                       60
                                                       90
120
                    150
                                                                                 SITE «2
                                    SITE «3
                     30
                                                       60
                     90
                    120
                     150
                                /
                                     SITE «5
            .18    .24     .3      .36     .42    .48      .18     .24     .3     .36     .42    .48
            .18    .24     .3     .36    .42    .48      .18    .24     .3     .36    .42    .43
            .18     .24
                      90
                     120  -
                     150
.36    .42    .48      .13    .24     .3     .36
                                       WATER CONTENT,  cm3/cm
Figure  3. Water content profiles of Bethany soils before (open symbols) and after (solid symbols) drainage measured
         by neutron scattering.
                                                  22

-------
     .42
     .36
      .3
     .24
   (J
  01
      _,
   u  .36
   O
   O
       .
      .3
      .34
      18
      48
      .42
      .36
       .3
      .24
      .18
                          BETHflMY §1

                          13 cr»
   43 cw
             30    60
90
                           90 en
               Meutron Scottenng
                                     .48
              .42
              .36
               .3
              .24
                                 i     i

                                   30 cn
                                      .18
             30    60    90    120   150    0

                                      .48
                     30    60    90    120    ISO
                                      .42
              .36
               .3
              .24
                                                           60 cm
120   ISO  '  0

         .48
30    60    90    120   150
              .42
              .36
               .3
              .24
              .18
                                    120 en
            IH-HH
             30    60    90    120   150  "~~0    30    60    90    120    150


                         NEGATIVE PRESSURE  HEAD,  cm



Figure 4. Soil-water desorption curves for site 1 of Bethany soil.


                                    23

-------
.48
.42
.36
.3

.24
.18
0
.48
, .42
o
'B -36
«k
1—
z
^ -a
(_ .3
0
0
UJ .24
0 0
.48
.42
.36
.3

.24
* 'I8o
.48
.42
.36
.3
.24
.18
50
III!
30 en
•n •
Q
0 a
• U
.^^^7^^ .
4
"°\ o.
3D 60 90 120 IS
DD a a a 60 cw
"li- • i ~M
» I 1 -f
moo o * o
-

-

30 60 90 120 15
I i i I
120 CM
V-5 "-fl
.n-t-t-H
: + « * ****^
-*
1 30 60 90 120 15
                                  NEGATIVE  PRESSURE  HEAD,  cm




Figure  5.  Soil-water desorption curves for site 2 of Bethany soil.



                                                  24

-------

.42
.36
.3
.24
.18
0
.48
co .42
0
co
w .36
V-
LU
O
0
UJ 94
H~ " *-
3
.48
.42
.36
.3'
.24
.18
IIII
BETHANY §3
15 cn
£H^. •
. ^^rj-ssj^.
-
30 60 90 120 I!
•
oo o 45 cn
DO D D
++ + -t-fc^^.^

till
30 60 90 120 1!
iiii
90 cn
U—U-J
-+V* + V>^SH.
.°° 0 0 0
IIII
) 30 60 90 120 1
.to
.42
.36
.3
.24
.18
>0 0
.48
.42
.36
.3

.24
.18
SO G
.48
.42
.36
.3
.24
.18
JO C
IIII
30 cn
:is s 8 :
_^-*^~^ .
-%
30 60 90 120 131
iiii
60 cn
m 9m
/ V*+


30 60 90* 120 15
iiii
120 cn
JH-H-i
- ** ^ **'
+ Neutrr •, Scattering
So? c&r«
iiii
1 30 60 90 120 13
                                    NEGATIVE  PRESSURE  HEAD,  cm
Figure  6.  Soil-water desorption curves for site 3 of Bethany soil.




                                                  25

-------
  U
     .48
     .42
     .36
      .3
     .24
     .18
     .48
      .42
                                 BETHANY »4   -
                                 15 CM
                                                   .42
                                                   .36
                                                    .3
                                                .24
                                                                             30 en
               30      60      90     120    150
                                                             30      60      90      120
CO
  u   .36
UJ

o
0£
       .3
      .24
      .18
              •o-r
                                   45 CM
                        I
                                                   .48
                                                   .42
                                                   .36
                                                    .3
                                                   .24
                                                                                 60  en
      .42
      .36
       .3
      ..'24
      .18
                 30     60      90      120     150  'l 0
                                                   .48
                                                            30      60     90     120     150
                 Neutron Scattering
                   Cores          90
                                                   .42
                                                   .36
                                                    .3
                                                   .24
                                                   .18
                                                                               120 en
                 30      60      90     120    150    "0      30     60

                                 NEGATIVE  PRESSURE HEAD, cm

Figure  7. Soil-water desorption curves for site 4 of Bethany soil.

                                                 26
                                                                           90      120     150

-------
      .42
      .36
       .3
      .24
      .18
                  o-i-
                BETHftNY «5  -

                15 cr>
                30     60      90      120
                                                  .48
.42
.36
 .3
.24
.18
      .42
 PO
  o  .36
  o
  0
        _
        .3
       -24
       .18
       .48
       .42
       .36
        .3
       .24
       ,18
            •~^.
           O O   0    O    O    O
                                  45  en
                 30      60      90     120    150
.42
.36
 .3
 .24
.18
              D   a
•f Neutron Scattering



    Corzs        
-------
       .48
       .42
       .36
        .3
       .24
   .18
     I


   .48
       .42
                       i
   co
o  .36
   CO
    E
    o
    LU
    O
    <_)  •
        .42
        .36
         .3
        .24
        .18
                                BETHANY *6

                                15 en
                                 45 on
                30     60     90     120
                                  90 crt
       I!   *   $    *   •
        MT~T^T-T-.
                          Scot-ten ng
                                              .42
                                              .36
                                               .3
                                          .24
                              .18
30      60     90    120    150     0

                              .48
                                                        i       i
                                          .42
                                              .36
                                               .3
                                          .24
                                               .18
                                           .42
                                           .36
                                           .3
                                           .24
                                           .18
                                                                    30 cn
                                                       30     60     90     120
                                                                    60 ct'i
                                      150     0     30     60     90     120    150


                                          .48
                                  iLJfi   B   •
                                                                     120 cn
                30     60     90     120    150    "0      30     60     90     120    15u



                               NEGATIVE  PRESSURE HEAD,  cm
Figure  9. Soil-water desorption curves for site 6 of Bethany soil.



                                           28

-------
                                                             I      I
                                                   c  •

                                                   •'  •

                                                                         "&•
                                                                         a
                                                                                      10
      1-J
o
3
O
  .01



 .00!



. .1.101

   10
      .01
    . frr.if.1 1
                                                       y\     *('.     "in    \<*f\
                                                          •
                                                          r
                                                          m
                                                                                     .01
                                                                                     .001
                                                                                              30     60
                                                                                       10
                                                                                      .01
                                                                                     .001
          NEGATIVE PRESSURE HEAD,  cm
 Figure 10. Hydraulic conductivity as a function of negative pressure head for Bethany soil. '
                                                            29

-------
                                             10
      .001
     .0001
                                           .001
                                           .0001
                                                     1      1
                                                                   «
                                                      ••<*<
  —    .01
S   -001
     .0001
                                             .1
                                             .01
                                            .001
         .29
                                .44   .48
        10
       .01
      .001
     .0001
                            ?0-liO en
              .32   .36    .4    .44   .48

             WATER CONTENT, cn3/cm3
                                                    . 3i
                                                     SITE
                                                      1 O
                                                      3 «
                                                      3 a
                                                                  11-30 c«
                                                                                   10
                                                                                   .1
                                                                                  .01
                                                                                .0001
                                                                  60 -.*5 CP.
                                                                 .4    .44
                                                                                   . 1
                                                                                  .01
                                                                                 .001
                                                                                .0001
                                                                                    .23
                                                                                             *

                                                                 .4    .44    .48      .28    .3?    . 3t     .4
Figure 11.  Hydraulic conductivity as a function of volumetric water content for Bethany soil.
                                                         30

-------
TADLE 8.1  Volumetric w»t»r content of foil corts
           tt ••ltet*d pr«§«ur« htidt for fit* 1
           of B* thany toi1.
PRESSURE
HEAD
cm
DEPTH






_
-
-
-
•
-
•V
-
-
-
-
-
_
-8
-8
-8
-8
-8
-8
16
16
16
16
16
16
36
36
36
36
36
36
56
-56
-
-
-
-
_
-
-
56
56
56
56
76
76
76
-76
-
-
_
-
-
-
-
-
-1
-1
-1
-1
-1
-1
76
76
96
96
96
96
96
96
26
26
26
26
26
26
WATER
CONTENT
cm /em
PRESSURE
HEAD
cm
: 15 cm
.0
.0
.0
.0
.0
.0
. t
. 1
. 1
. 1
. 1
. 1
. 0
. 0
.0
.0
. 0
0
. 3
3
. 3
.3
.3
.3
.0
0
.0
.0
.0
0
. 0
. 0
.0
.0
.0
. 0
.0
. 0
. 0
. 0
.0
. 0
.361
.386
.372
.377
.282
.370
.361
386
.372
.375
.277
.370
.361
. 386
.371
.373
.276
.367
.354
.377
.361
. 373
. 267
.358
.341
359
.348
.348
.253
.343
.328
. 343
.336
.336
.240
.328
.314
.324
.322
.320
.240
.312






w
-
-
•»
-
-
_
-
-
-
-
-
_
-
-
-
-
-
-I
-I
-1
-1
-1
-1
-5
-5
-5
-5








-8
-8.
-8.
-8 .
-8
-8
16 .
16 .
16.
16.
16.
16.
55
55 .
55.
S3
55.
55 .
96 .
96 .
96.
96 .
96.
96 .
56.
56.
56 .
56.
56.
56 .
10.
10.
10.
10.








WATER
CONTENT
c.3/e.3
PRESSURE
HEAD
em
30 cm
9
9
9
9
9
9
4
4
4
4
4
4
6
6
6
6
6
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0








.452
.384
. 443
.402
.324
.281
.412
.372
.434
.397
.321
.273
.365
.345
399
.371
.290
247
.348
.336
.385
.360
.271
.238
.340
.330
.377
.353
.257
.222
300
.308
.333
.335














*
-
-
-
-
-
..
.
-
-
.
-8.
-8.
-8.
-8.
-8 .
-8.
16.
16.
16.
16.
16.
16.
56 .
56 .
56 .
56.
56.
-56.
_
.
-
.
-
-
-1
96 .
96.
96 .
96.
96.
96.
56 .
-156 .
-1
56 .
-156
-156 .
-1
-5
-5
-5
-5

-






$6.
10.
10.
10.
10.








WATER
CONTENT
em /cm
45 cm
0
0
0
0
0
0






0
0
0
0
0
0
t





0
0
0
0
0
0
0
0
0
0








.372
.441
.369
.405
.423
.395
.372
.441
.369
.405
.420
.395
.365
.420
.361
.396
.411
.384
.356
.403
.354
.389
.399
.378
.350
.379
.348
.384
.386
.372
.341
. 340
.343
.372








                                 94

-------
TABLE 8.1  Continued
PRESSURE
HEAD

cm
DEPTH: 15
-156 .2
-15* . 2
-154 . 2
-156.2
-156 .2
-156.2

PRESSURE
HEAD

em
DEPTH: 60
-8 . 0
-8 .0
-8 .0
-8 .0
-8.0
-8 . Q
-16 . 1
-16.1
-16.1 '
-161
-16 . I
-16.1
-56 .0
-56 . 0
-56 .0
-56. 0
-56 .0
-56. 0
-75 .9
-75. ?
-75 . ?
-75. 9
-75 . 9
-75. 9
-96 .0
-96.0
-96 .0
-76.0
-96 .0
-96.0
WATER
CONTENT
3 3
cm /cm
cm
.301
.306
.308
.304
.240
.296

WATER
CONTENT
3, 3
cm /cm
cm
.361
.361
.409
.369
.361
.412
355
.368
.400
359
.355
.401
.350
.354
.394
.351
.353
.394
.348
.350
. 391
.347
.353
. 392
.346
.347
.389
.344
.352
.390
PRESSURE
HEAD

em








PRESSURE
HEAD

cm

- .0
- .0
- .0
- .0
- .0
-8.0
-16. 3
-16.3
-16 . 3
-16 . 3
-16 .3
-16 . 3
-56.0
-56 .0
-56 .0
-56 .0
-56.0
-56 .0
-76 .0
-76 . 0
-76 . 0
-76 .0
-76 .0
-76.0
-96 . 0
-96 .0
-96.0
-96. 0
-96.0
-96 . 0
WATER
CONTENT
3 3
cm /cm
30 cm







WATER
CONTENT
3, 3
cm /cm
90 cm
.397
.396
.375
.377
.388
.380
. 392
.383
371
. 372
. 382
.374
. 379
.376
.361
.368
.377
.368
.372
.373
. 357
. 366
.375
366
.368
.370
.353
.364
.374
.364
PRESSURE
HEAD

cm
45







PRESSURE
HEAD

cm
120
- .0
- . 0
- .0
- .0
- .0
- .0
-16 .0
-16.0
-16 .0
-16 . 0
-16.0
-16. 0
-36 . 0
-36.0
-36 .0
-36.0
-36.0
-36.0
-56.0
-56 . 0
-56 .0
-56.0
-56 . 0
-56 .0
-76. 1
-76. 1
-76.1
-76. 1
-76. 1
-76. 1
WATER
CONTENT
3 3
cm /cm
cm







WATER
CONTENT
3, 3
cm /cm
cm
.425
.395
.410
.422
.382
.392
.405
.391
.401
.411
.376
.390
.403
.388
.407
.407
.374
.389
.401
.383
.403
.402
372
.388
.397
.379
.399
.398
.369
.383
                                  95

-------
TABLE  8.1   Continued
 PRESSURE
   HEAD
    cm
 WATER
CONTENT
  3,  3
cm /ca
PRESSURE
  HEAD
 WATER
CONTENT
  3,  3
cm /ca
PRESSURE
  HEAD
 WATER
CONTENT
CB /ca
 DEPTH:   40  em
                   »0 ca
                            120 ei
154 .0
154.0
154 0
154.0
154 .0
154 . 0
.343
.342
.384
.340
.350
.387
163. 0
143.0
143 . 0
163 0
143.0
143.0
.354
.361
.343
.354
. 367
354
                                              -94 .0
                                              •96. 0
                                              -94 .0
                                              •94 . 0
                                              -94 . 0
                                              •94.0
                                            .394
                                            .375
                                            .394
                                             394
                                            .344
                                            .380
  •510.0
  •510. 0
  •510 .0
  •5100
  .341
  .385
  .373
  .330
 •510.0
 •510.0
 •510.0
 •510.0
  .345
  .334
  .320
  .331
  •155
  •155
  •155
  •155
  -155
  •155
  •S10
  •510
  •510
                                            -510.0
  .385
  .345
  .387
  .383
  .357
  .370
  .324
  .340
  .363
  .357
                                        96

-------
TABLE 8.2  Volumetric water content of toil cores
           at (elected pressure heads for lite 2
           of Be thany toil.
PRESSURE WATER
HEAD CONTENT
cm cm ten
DEPTH:
-8.
-8.
-8 .
-8 .
-8 .
-8 .
-14.
-14.
-14.
-U.
-14.
-U
-34 .
-34
-34 .
-34
-34 .
-34
-54 .
-54.
-54 .
-54.
-54 .
-54.
-74 .
-74.
-74 .
-74
-74 .
-74.
-94 .
-94.
-94 .
-94
-94 .
-94.
-1 24 .
-124 .
-124 .
-124.
-124 .
-124

0
0
0
0
0
0
1
1
1
1
1
I
0
0
0
0
0
0
3
3
3
3
3
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
IS cm
.347
.284
.349
.374
.377
.385
. 347
284
.349
.374
.377
385
.343
.284
.349
.361
.340
. 372
. 352
.282
. 337
. 344
.338
348
.340
. 275
.325
.331
.323
.342
.332
.249
.314
. 320
309
.324
.322.
249
.305
304
.294
.309
PRESSURE VATER
HEAD CONTENT
cm cm /cm

-8
-8
-8
-8
-8
-8
-14
-14
-14
-14
-14
-14
-55
-55
-55
-55
-55
-55
-94
-94
-94
-94
-94
-94
-154
-154
-154
-154
-154
-154
-510
-510
-510
-510








30
. 9
. 9
.9
.9
. 9
.9

.

.
.
•
.4
4
.4
.4
.4
.4
.0
.0
.0
.0
. 0
.0
.0
. 0
.0
0
.0
.0
. 0
. 0
.0
. 0








cm
.413
.412
.418
.391
.219
.285
. 402
.404
. 409
.388
.217
.282
383
.375
. 380
343
.195
.243
.372
. 342
347
.351
. 184
.251
344
.351
.354
.340
. 177
.244
.334
.322
334
305
•







PRESSURE VATER
HEAD CONTENT
3, 3
cm cm /cm

-8
-8
-8
-8
-8
-8
-14
-14
-14
-14
-14
-14
-54
-54
-54
-54
-54
-54
-94
-94
-94
-94
-94
-94
-154
-154
-154
-154
-154
-154
-510
-510
-510
-510








45
0
.0
. 0
. 0
.0
. 0

.

,
.
.2
.0
.0
.0
.0
.0
.0





. 1
0
.0
. 0
.0
.0
.0
.0
0
.0
. 0








cm
.443
.440
.459
.459
.479
.454
.443
.440
.459
.459
.471
.454
445
434
.447
.458
.445
.449
.429
.429
.434
449
.444
.438
.415
.420
.423
.435
.432
.427
.405
.392
.389
.401








                                97

-------
TABLE 8.2  Continued
PRESSURE WATER
HEAD

ea
DEPTH:
-ISA .
-ISA.
-154 .
-154 .
-154
-154 .




2
2
2
2
2
2
CONTENT
3 3
CB /ca
IS CB
.311
.269
.295
.2?2
.281
.273
PRESSURE WATER PRESSURE WATER
HEAD CONTENT HEAD CONTENT
33 33
CB CB /CB CB CB / CB
30 CB 45 CB






PRESSURE WATER
HEAD CONTENT
3, 3
CB CB /CB
DEPTH
-
.
-
-
-
. -8
-14
-14
- 14
-14
-14
-14
-54
-54
-54
-54
-54
-54
-75
-75
-75
-75
-75
-75
-94
-96
-94
-94
-94
-94
: 40
. 0
.0
.0
. 0
0
. 0
1
. 1
. 1
. 1
. 1
. 1
. 0
.0
.0
.0
.0
. 0
. 9
9
. 9
. 9
. 9
. 9
. 0
.0
.0
.0
. 0
. 0
CB
.108
.425
.440
421
348
.413
404
415
453
.413
348
403
400
.410
449
408
.347
.399
.398
. 408
.447
404
.347
.398
397
.405
.444
. 404
.347
.394
PRESSURE WATER
HEAD CONTENT
3, 3
CB CB /CB

- .
- .
•^
- .
- .
-
-14 .
-14
-14 .
-14
-14 .
-14
-54 .
-54.
-54.
-54 .
-54
-54
-74.
-74
-74
-74 .
-74
-74.
-94.
-94.
-94.
-94 .
-94 .
-94.
90
0
0
0
0
0
0
3
3
3
3
3
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ca
347
.371
.377
.397
.404
344
361
348
372
.385
.392
358
357
.362
.363
377
. 386
354
355
340
360
375
.385
353
354
.358
.358
.373
394
352
PRESSURE WATER
HEAD CONTENT
CB CB /CB

-8
•8
-1
-8
-8
-8
-16
-16
- 16
-16
-16
' -16
-36
-36
-36
-36
-34
-36
.-54
-56
-56
-56
-54
-56
-74
-74
-74
-74
-74
-74
120
0
.0
.0
0
0
. 0
.0
. 0
0
.0
.0
0
.0
.0
. 0
0
0
0
.0
.0
0
.0
. 0
. 0
. 1
. 1
. 1
. 1
1
. 1
CB
.38}
.373
.394
.402
394
385
.383
.372
.393
.395
393
.383
381
.371
.390
.394
.391
.382
.380
.371
.389
393
390
.381
374
.370
.384
.392
.388
.380
                                   98

-------
TABLE 8.2  Continued
 PRESSURE   WATER     PRESSURE   WATER     PRESSURE   WATER
   HEAD     CONTENT      HEAD    CONTENT      HEAD    CONTENT
             3,  3                3,  3                33
    ca      cm /ca        ea     ca  'cm        ca     ea /cm

 DEPTH:  60 ea                ?0 ea               120 ea

  -154.0     .392        -U3.0     .346        -94.0     .373
  -154.0     .399        -143.0     .330        -94.0     .348
  -154.0     .441        -143.0     .348        -94.0     .383
  -154.0     .398        -143.0     .344        -94.0     .391
  -154.0     .344        -143.0     .374        -94.0     .384
  -154.0     .391        -143.0     .344        -94.0     .378

  -510.0     .341        -510.0     .344        -155.9     .344
  -510  0     .345        -510.0     .325        -155.9     .343
  -510.0     .425        -510.0     .325        -155.9     .374
  -510.0     .372        -510.0     .334        -155.9     .383
                                            -155.9     .379
                                            -155.9     .372
                                            -510.0     .354
                                            -510.0     .349
                                            -510.0     .350
                                            -510.0     .357
                                      99

-------
TABLE 8.3  Volumetric water content of soil core*
           at selected pressure heads for site 3
           of Be thany soil.
PRESSURE WATER
HEAD CONTENT
ea ca /ca
^ 	 	
DEPTH
-8
-8
-8
-8
-8
-8
-14
-14.
-14
-14.
-14
-14.
-34
-34.
-34
-34
-34
-34.
-54
-54
-54
-54
-54
-54
-74
-74
-74
-74
-74
-74
-94
-94
-94
-94
-94
-94
-124
-124
-124
-124
-124
-124
: 15
.0
.0
.0
.0
.0
0
1
1
1
1
. 1
1
0
0
0
0
0
o
.3
.3
. 3
3
.3
. 3
.0
.0
.0
.0
0
0
.0
0
.0
0
0
.0
.0
. 0
.0
.0
.0
.0
_ —
ca
.372
.343
.349
.343
.378
.344
.344
.343
.349
.343
.378
.344
.359
.342
.342
.340
.373
.357
.343
.349
.347
. 331
.341
. 344
.328
.334
.334
.321
.346
.327
.314
324
.323
. 31 1
334
. 315
.302
.311
.310
.301
.318
.301
,.,
PRESSURE WATER
HEAD CONTENT
3, 3
ca ca /ca
— ^ — ~ 	 *~ 	
-8
-8
-8
-8.
-8
-8.
-14.
-14 .
-14
-14.
-14
-14.
-55
-55.
-55
-55.
-55
-55
-94
-94
-94
-94
-94
-94
-154
-154
-154
-154
-154
-154
-510
-510
-510
-510








30
9
9
9
9
.9
9
4
4
4
4
.4
4
4
4
4
4
4
4
.0
0
.0
0
.0
0
.0
.0
.0
.0
.0
0
.0
0
.0
0








II. 	
ca
.399
.403
.394
.405
.234
.194
.391
.401
. 389
.401
.225
.190
.370
.392
.378
.388
. 199
175
.340
.380
.349
.377
. 184
. 144
.354
374
.342
349
. 174
.158
.329
.349
342
.338








PRESSURE WATER
HEAD CONTENT
3, 3
ca cm /ca

-8.
-8
-8 .
-8.
-8.
-8.
-U.
-14.
-14 .
-14.
-U .
-14.
-54.
-54.
-54.
-54.
-54
-54.
-94
-94.
-94
-94 .
-94
-94.
-154
-154
-154
-154
-154
-154
-510
-510
-510
-510








15
0
0
0
0
0
0
2
2
2
2
2
2
0
0
0
0
0
0
1
1
I
1
1
1
.0
0
.0
.0
. 0
0
.0
0
.0
0








ca
445
.421
.384
483
.425
410
.445
.421
.384
483
.425
.410
442
.421
.384
.474
425
.408
431
.421
.383
. 440
.417
. 399
.419
.415
.374
.448
.407
. 391
.384
.384
343
.389

«






                                100

-------
TABLE 8.3  Cont tnu«d
PRESSURE WATER
HEAD CONTENT
3 3
ea CB /CB
DEPTH: IS ca
-154.
-ISA.
• 156 .
-134.
-154 .
-156.
.288
.299
.298
292
.301
.292
PRESSURE WATER PRESSURE WATER
HEAD CONTENT HEAD CONTENT
3, 3 3, 3
CB CB /cm ca cm /cm
30 ea 45 ea






PRESSURE WATER
HEAD CONTENT
ca ca /ca
DEPTH:
-a
-8
-a
-a
-a
-a
-14
-14.
-14
-14
-14
-16
-54
-56
-54
-54
-54
-56
-75
-75
-73
-75
-75
-75
-94
-96
•94
-94
-94
-94
60
0
0
0
0
0
0
1
1
1
1
1
1
0
0
0
0
0
0
9
9
9
9
9
9
0
0
.0
0
.0
0
ca
.343
364
374
.359
.356
.35?
341
. 342
.372
. 353
.356
.353
.341
.361
.365
350
. 3S6
. 352
341
361
.361
350
356
.352
.340
.360
.360
.350
. 356
.352
PRESSURE WATER
HEAD CONTENT
ca ca /ca

-a
-a
-a
-a
-8
-8.
-16
-16 .
-14
-14.
-14
-14
-54
-54
-56
-56
-54
-54
-74
-74
-74
-74
-74
-74
-94
-94
-94
-94
-96
-96
90
0
0
0
0
0
0
3
3
3
3
3
3
0
0
0
0
0
0
0
0
0
0
0
o
0
0
.0
0
.0
0
ea
.343
337
354
349
254
347
341
.337
. 350
345
.252
.346
. 335
333
. 340
.333
.248
.338
333
331
.336
328
.244
335
. 331
.329
.333
325
.244
.333
PRESSURE WATER
HEAD CONTENT
ca ca /cm

-
- .
-
-
-
- .
-14
-1
-I
-1
. 1
-1
-34
-36
-36
-36
-36
-36
-56
-56
-56
-56
-56
-56
-76
-76
-74
-76
-76
-76
120
0
0
0
0
0
0
o
0
0
0
0
0
0
o
0
0
0
0
0
0
0
0
0
0






ca
.372
.407
.404
.391
.391
.381
.372
.404
.396
.384
.391
.377
.371
.402
.393
. 383
.391
.374
.370
. 397
390
.381
.390
.372
.369
393
.386
.378
.388
.368
                                101

-------
TABLE 8.3   Cont inucd
PRESSURE
HEAD
ca
DEPTH
-1
-I
-I
-1
-1
-1
-5
-5
-5
-5






54
54
34
Si
34
Si
10
10
10
10






WATER
CONTENT
ca /ca
PRESSURE
HEAD
CB
40 oa
0
0
0
o
0
o
0
0
0
0






.340
.358
358
.347
334
350
.348
342
.351
.337






-1
-1
-1
-1
-1
-1
-5
-5
-5
-5






43
43
43
43
43
43
10
10
10
10






WATER
CONTENT
ca /ca
PRESSURE
HEAD
ca
90 cm
0
0
0
0
0
0
0
o
0
0






. 325
.325
.324
314
.238
.325
. 225
311
.312
.301






—
-
-
-
-
-
-1
94
96
94
94
94
94
55
WATER
CONTENT
ca / ca
120 ea
0
0
0
0
0
0

-155.
-I
55

-155.
-I
-1
-5
-5
-5
-5
55
55
10
10
10
10

9
.0
0
.0
0
.364
.388
.381
.375
.385
.363
.359
.378
.372
.368
.379
.352
354
.323
.344
.345
                                 102

-------
Table 11-1.  Field Studies of Soil  Matrix and Water  Retention  Properties.
Parameter
Porosity

Porosity
Porosity

Porosity

Bulk Density
Bulk density
Bulk density


Bulk density
Bulk density
Bulk density
Bulk density
Bulk density
Bulk density
X sand/X clay
X sand/X clay
X sand/X clay
X sand/X clay
X sand/X clay
.1 bar water
content
.09 bar water
content
.1 bar water
content (9g)
.1 bar water
content
15 bar water
content
15 bar water
content
15 bar water
content
15 bar water
content
15 bar water
content (9g)
pH
pH
PH
PH ,
Kn (cm3/g)
Mean
0.45

0.37
0.53

0.42

1.36
1.30
1.20


1.47
1.26
1.47
1.65
1.59
1.20
24/45
17/32
59/12
83/9
65/28
.37

.37

.27

.45

.166

.041

.193

.074

.095

6.1
6.4
5.8
8.2
2.01
cv
(X)
11

11
7

10

7
7
26


9
6
6
3
6
15
15/33
32/16
37/53
3/34
8/18
4

17.6

20

15

14.4

45

14

19

33

15
7
9
2
31
Field
***•»*«
(ha)
150

0.8
.03

0.4

150
15
3.8


1.3
0.5
0.5
0.34
91.6
40
150
85
15
91.4
0.28
85

150

15
•
40

85

1.3

0.5

3.3

15

.04
.02
.04
85
0.64
Number
of
Soil RepH.
Texture cates
clay loam

sand
clay loam
•
loany sand

clay loam
sandy loam
sandy loam


sandy loam
sllty clay
silt loam
sand
sand
clay loam
clay
sllty clay loam
sandy loam
loamy sand
sandy clay loam
clay loam

clay loam

sandy loam

clay loam

clay loam

sandy loam

silty clay

silt laom

sandy loam

clay loam
loam
sandy loam
clay loam
loamy sand
120

120
20

12

120
64
30


192
144
72
5
5
36
480
100
64
5
35
100

120

64

36

900

172

144

192

64

1,040
640
208
100
36
Method
e
Measurement
Water content at
zero suction
Not given
Water content at
zero suction
Water content at
zero suction
Undisturbed cores
• Not given
Measure volume of
plastic-lined
hole
Undisturbed cores
Undisturbed cores
Undisturbed cores
Undisturbed cores
Undisturbed cores
Undisturbed cores
Hydrometer
Light scattering
Not given
Not given
Not given
Hanging water
table






Pressure plate

Pressure plate

Pressure plate

Pressure plate

Pressure plate

Pressure plate
Pressure plate
Pressure plate
Pressure plate
Batch equlibrlum
Reference
[15]

[16]
[17]

[18]

[15]
[19]

[20]

[21]
[21]
[21]
[22]
[22]
[23]
[15]
[24]
[19]
C22]
[25]
[24]

[15]
•
[19]

[23]

[24]

[21]

[21]

[21]

[19]

[26]
[26]
[26]
[24]
[27]

-------
Table 11-2.   Field Studies of Water Transport  Properties.


Parameter
(cm d'1)
Saturated K

Saturated K
Saturated K

Saturated K
Saturated K

Saturated K

Saturated K

Saturated K
Saturated' K
Saturated K
Saturated K
Saturated K
Saturated K
Saturated K

Infiltration
Infiltration
rate
Infiltration
rate
Infiltration
rate
Inf1 Itration
rate
Infiltration
rate
Infiltration
rate
Unsaturated
K(9)
(8[9-9o])P
KO
8
KO
3
KO
8
KO
8




Mean
20.6

168
316

84
3.6

18.9

11.0

6.9
28.1
55.6
71.2
98.5
24.1
203

14.6
16.3

6.6

8.5

8.5

47

263




22.5
14.6
4.6
89.1
9.6
65.4
4.0
32.9



CV
(*)
120

190
69

69
48

103

118

92
320
118
105
81
178
50

94
40

71

56

23

79

97




343
64
235
41
76
37
46
19



Soil
Texture
Clay loam

Sandy loam
Sand

Loamy sand
Silty clay
loam
Coarse

Fine

Silty clay
Very coarse
Coarse
Loamy sand
Loamy sand
Sandy laom
Silt loam

Clay loam
Loam

Silty clay
loam
Silty clay

Silty clay
loam
7 series

7 series




Clay loam

Clay loam

Silt laom

Loam



Field
Number
of
Size Measure-
(ha)
150

15
0.8

0.4
Composite
SCS date
for a given
soil series
in Imperial
Valley, CA




91.6
91.6
9.6

150
0.9

.004

.004

.004

100 ha

100 ha




150





.66


ments
120

64
90

12
33

330

287

339
36
352
121
5
5
26

20
1,280

625

125

25

20

15




20

20

611

24


Method
of
Measure-
ment
Steady infiltration
20 plots x 6 depths
Lab permeameter
In Situ air-entry
permeameter
Lab permeameter
Lab permeameter

Lab permeameter

Lab permeameter

Lab permeameter
Lab permeameter
Lab permeameter
Lab permeameter
Lab permeameter
Lab permeameter
0-30 cm infiltration
(double ring)
Steady state
Steady infiltration
(double ring)
Adjacent infiltro-
meters along
transect



Double ring infiltro-
neter
Inverse auger hole
method 150 cm



Instantaneous profile
method
Unit gradient method

Unit gradient method

Instantaneous profile
method 4 plots x 6
depths



Reference
[15]

[19]
[16]

[18]
[28]

[28]

[28]

[28]
[28!
[28!
[28!
[22!
[22:
[29]

[15]
[30]

[31]





[32]

[32]




[33]

[33]

[34]

[35]



-------
Table 11-3.'  Field Studies of Chemical  Concentrations.
Chemical
1. Chloride


2. Bromide


3. Bromide

4. Chloride

5. EC(l:l
extract)
6. EC(sat
extract)

7. Chloride



8. Chloride



9. Chloride


10. Chloride


Origin
of
Chemical
3 cm pulse


I cm pulse


0.5 cm pulse

Surface
application
Native
•
Sugarcane
Plantation

Fertilized
fields


Native
to


Manure
fertilized
fields
Irrigated-
fertilized
fields
Measure-
ment
Depth (m)
0.65
1.15

0.3
0.6
0.9
0.6

0.12
0.32
0-.025
075-.15
0-1
0-1
0-1
1.8-6
1.8-6
1.0-3.6
1.0-4.2
Surface
bedrock
•v20 cm

1.5-6.3
1.5-6.3
1.5-6.3
.6-1.2
.9-1.5
.6-1.2
Field
Soil Size
Texture (ha)
Loamy 0.64
sand

Loamy 0.64
sand

Sandy 0.2
loam
Loamy 0.3
sand
Mankos
shale
Clay loam- loam
Clay loam-loam
Clay loam-loam
Sandy loam- loan
Sandy loam-loam
Loam-sandy loam
Loam-sandy loam
Various



Sandy loam
Sandy loam
Clay loam
Loam
Fine sandy loam
Fine sandy loam
Number
of
Repli-
cates CV
"127
79
76
118
90
89
100

67
61
128
263
150 91
440 75
445 225
16 22
8 30
20 101
13 21
100-200 12-70



81 19
14 19
14 66
60
8 102
2 87
Method
of
Measure-
ment
Soil cores


Solution
samplers

Soil cores

Soil cores

Soil samples

Soil cores


Soil cores con-
solidated to
depth average
below root zone
Summary of 9
catchment areas
Average concen-
tration in core
Average concentra-
tion in entire
vertical profile
Soil samples
Soil samples
Soil samples
Reference
[36]


[36]


[37]

[38]

[39]

[40] :


[41] ?
«
c
c
[42]



[43]


[44]



-------
                                 SPATIAL VARIABILITY Of SOIL PKOff8TI£S  257
Table 11-5.  Sample Sizes Required to Have a 951 Probability of
             Detecting a Change of FX 1n the Mean Using a T-Test
             with a*5X (Sample CVS are the mean of all field
             studies).
Number of
Parameter Studies
Bulk density or
porosity
Percent sand or
clay
0.1 bar water
content
15 bar water
content
PH
Saturated K
Infiltration
rate
IC0 in K(9)
Ponded solute
velocity
Unsaturated
solute velocity
Solute concen-
tration
13
10
4
5
4
13
8
4
1
5
4
F: 20%
40% lOOt
Number of Samples
6
28
9
23
4
502
135
997
1,225
127
119-551
* *
9 *
* *
7 *
* *
127 22
36 8
251 42
308 51
33 7
32-140 7-24
Average
CVtSO
10 t 6
28 ± 18
14 ± 7
25 t 14
8 t 5
124 t 71
64 t 26
175 ± 139
194
52 t 9
60-130
(range)
 •Sample size estimates  are  less  than 5  and  should  not  be used.

-------
D
go

          /us/«y /*/ v-

-------
*      * Jly

-------
Co/u s er
                         9(

015
Ue  I;
                tS

-------
 Jarcy- Doc ki*>* n
          s
 u /c»
s

    s'+t*  u

-------
                  *, o?
                Li
                * . *  X
3

-------
              C
               '*>TC
            /*>
  3t
   £
   Sx
(«rtg>+£
       &/
(k(ti
                     2h
*/<
(5
TKi
s 
-------
T/.I
                        3
                                 /
he Ids    -f
                    c^   5
                       * h-
                       *  h -

-------
A; s e r
                 5

                    - CeS (X
is  e
           <9

-------
Soil Science Fact Sheet
Sept. 1983
                               SL 40 (Revised)
Pesticides and  Their  Behavior  in  Soil  and  Water
P.S.C. Rao, R.S. Mansell, L.B. Baldwin and M.F. Laurent*
Florid* Cooperative titeruion Service / Initltulr of Food »nd Apiculture Science* / Uoivcrtlty of Hortdi / J. 1. Woeitt, D«tn
 Concern for man himself and his late must always lorrr. the
 chief interest of all technical endeavor.
                                   Albert Einstein

   Pesticides  stand out  as one  of the  major
 developments of the twentieth century. During the
 past twenty years, however, concern has arisen as
 to the extent their presence in  the  environment
 poses a threat to wildlife  and mankind.
   Certainly, pesticides have improved  longevity
 and the quality of life, chiefly in the area of  public
 health.  Insect  control   programs  have  saved
 millions of  lives by combatting diseases such as
 malaria, yellow fever and  typhus.  The use of
 pesticides also constitutes an important aspect
of modern  agriculture, for without chemicals to
control various pests like insects, weeds,  plant
diseases,  worms and  rodents, our food supply
would  decrease and  prices  would  increase.
Florida's temperate to subtropical climate favors
growth  of  many  harmful  insects,  weeds  and
diseases,  thus  making this  state  particularly
dependent  on  pesticides for  economical  crop
management.
  Unfortunately, pesticides are poisons and can
be particularly dangerous when  misused.  Fish-
kills, reproductive failure m birds, and acute ill-
nesses  in  people  have all  been attributed to
exposure to or ingestion of pesticides — usually
         ?;. ;;c:i-'/"-iv'-^':';o'j^:'f'iZjjf-^ Ateo'v-'0''*-   ;c.f
  Figure 1: Pathways of pesticide lost. P'peiticide.

    Adapted from Herbifidt Injury Symptoms *nd Diannos:- Skroct. K'. A tnd Sfteen. T. J tfdi.f. North Ctrolint Agriculwrt' Exttnsio:
  Servict. AC-S5. Doc. 19S1.
         Pro<«».3- of So'.' Scitnct, Profwto; of Soil Sci»nci, Anoititt Prof«vor of Ae'iculturi' Engin«»rif>t »nd Anintr.i ir. Ed to-n

-------
•s § result of misapplication or careless disposal
of unused pesticides and  pesticide containers.
Pesticide  losses from areas of application and
contamination of non-target sites such as surface
and ground water  represent a monetary loss to
the farmer as well as a threat to the environment.
Thus careful management  of pesticides In order
to avoid environmental contamlnaion Is desired
by both farmers and the general public.
  The purpose of this fact sheet is to explain how
pesticides can  move from the area  In which they
are applied, and to show how this Information can
be used, along with other  factors,  to select the
proper pesticide.

      PATHWAYS OF PESTICIDE LOSS
  There are basically two  ways properly-applied
pesticides may reach surface and underground
waters — through  runoff and leaching.1 Runofl is
the physical  transport  of pollutants  over the
ground surlace by rainwater which  does not
penetrate the soil. Leaching is a process whereby
pollutants are flushed through the soil by rain or
Irrigation water as It moves downward. In many
areas of Florida soils are  sandy and permeable
and leaching Is likely to be a more serious prob-
lem than runoff. We now have technology to help
estimate  tne potential contamination  of  water
from  a  given pesticide.  To understand  this
technology,  It is  necessary to  know how  a
pesticide behaves in soil and water.
  Once applied to cropland, a number of things
may happen to a pesticide (Fig. 1). It may be taken
up by plants or Ingested by animals, Insects,
worms, or microorganisms  in the soil. It may move
downward In the soil and either adhere to par-
ticles or dissolve. The pesticide may vaporize and
enter the atmosphere, or break down via microbial
and chemical pathways Into other, less toxic com-
pounds. Pesticides may be  leached out of the root
zone by rain or irrigation water, or wash off the
surface of land. The fate of a pesticide  applied to
soil depends largely on two of Its properties, per-
sistence and solubility.

               PERSISTENCE
   Persistence  defines the "lasting-power" of a
pesticide.  Most pesticides break down or "de
grade"  over time as a result of several chemical
and micro-biological  reactions In soils. Sunlight
breaks  down  some  pesticides. Generally,
chemical pathways result in only partial deactiva
tion of  pesticides,  whereas soil microorganisms
can completely break down many  pesticides to
carbon dioxide, water and other Inorganic const*
tuents.  Some pesticides produce  Intermediate
substances,  called  "metabolites"  as  they
degrade.  The  biological  activity  of  these
substances may also have environmental signifi-
cance.  Because  populations  of  microbes
decrease rapidly beio* the root rone, pesticides
leached beyond this depth are less likely to be
degraded.  However, some pesticides will con-
tinue to degrade by chemical reactions alter they
have left the root zont.
  Degradation time is  measured  in "half-life."
Each half-life unit measures the amount of time  It
takes  for  one-halt tne  original amount  o1  a
pesticide  In soil to be deactivated.  Half-life is
sometimes  defined as  the time required for half
the amount of applied pesticide to be completely
degraded end released as carbon dioxide. Usu-
ally, the half-life of a pesticide measured by  the
latter basis is longer than that based on deactiva-
tion only. This is especially true If  toxic or non-
toxic metabolites  accumulate in the soil during
the  degradation. Table  1  groups  some of  the
pesticides  used  in Florida by persistency, or
length of half-life, on the basis of their deactiva-
tion in soils.
 Table 1: Grouping of pttticldes bated on p»r»l»tenct In tolls
                                   Ptrtittent
Non P«rtl«Unt Mo6>r*t«1y P«r»l»ttnt
(halt-lite less (hail-iile greater than
than 30 days) 30 days, tess than 100)
Aldicarb
Captan
Dalapon
Dlcamba
Malalhion
Methyl para
Ihion
Oxamyl
2.4-D
2, 4. 5-T
Aldiin
Atrazine
Carba'y!
Cerbofuran
Diajirt-r.
Endrir.
Fonofos
Glyphosate
Heptachlor
Llnuron
Parathion
Phorate
Simazine
Terbacil
TCA
(n, It-hie
Q'tater than
IOC days)
Btomacil
Chlordane
Line ane
Paraquat
Pidorarn
Tritluralm
        SOLUBILITY AND SORPT1ON

   Probably the single most important property in-
 fluencing a pesticide's movement with water is its
 solubility. Soil is a complex mixture of solids, li-
 quids and  gases  that provides the life support
 system for roots  of growing plants and micro
 organisms  such as bacteria. When  a  pesticide
 enters soil, some  of it will stick to soil panicles.
 particularly organic matter, through a process
 called adsorption  and some will dissolve and mi>
 with the water between soil particles, called "soil-
 water." As more water enters the soil through rain
 or irrigation,  the  adsorbed pesticide molecules
 may be detacher from soil particles through t
 process called desorption. The solubility  of e
  Two othc pcthwiyi of peslteidt Ion »'• throuB1' r«rnov>' in tht hirvvrtec! pltnt »-id by vcpo-Ujtior, (volitilititlon) Into ttu •tmosphfi
  Occurrence of p*rttcidt rtiidun in (dibit parti of pl«nu it lignihunt In term of burnt* «npo«-'». whilt pcrticidei rtlttated into the tTrrvoi
  pht:« hm «n imptrt on «lr qutlity »nd cr«U probl»-ni wher. *9ricullJr»< wo-V»-i «rt«- tht trtrtKJ §r*»i. Whil» th»M two pithwiyt 1-1
  imporum. thty will not b* considered furthe- in this 1»rt*.-.ec<.. which Is devot*-! tc p»r.i:id« b*".r->c.- in toil «r.d wttei.

-------
pesticide and Its sorption on soil are Inversely
related; that is, Increased solubility results In less
sorption.
  One of the most useful Indices for quantifying
pesticide  adsorption  on  soils Is the "partition
coefficient" (PC). The PC value Is defined as the
ratio of pesticide concentration In the adsorbed
state (that is, bound to  soil particles) and the
solution-phase   (that   Is.  dissolved  in  the
soil-water). Thus, for a given  amount of pesticide
applied, the smaller the PC value, the greater the
concentration of pesticide In solution. Pesticides
with small PC values are more likely to be leached
compared to those with large PC values.
   Partition coefficients of several chemicals are
shown in Table 2. Note the wide range of partition
coefficients.  Values  of  partition   coefficients
listed  in Table 2 are independent of soil type and
are characteristic of each pesticide. The partition
coefficient   is  determined  by  a  pesticide's
chemical   properties  such   as   solubility  and
melting point.
    1»blt 2: Partition coffficknn (PC) tor tckclron5 occu' *oo- i*tep
 to pollute  surface or ground water is  possible.
 Quantitative  prediction of  pesticide  loss  via
 runoff and leaching requires complex computer
 models which utilize site-specific  soil, crop, and
 climatological  information. This  would include
 the soil type, the date, amount and method of ap-
 plication, and the amount, frequency and duration
 of rain or irrigation following application.

-------
     PESTICIDE SELECTION AND USE
  Agricultural use of pesticides should be part of
an overall pest management strategy which in
eludes biological controls, cultural methods, pest
monitoring  and other  applicable  practices,  re
ferred to  altogether as Integrated  Pest Manage
menl or IPM. When a pesticide is needed its selec-
tion  should be  based on effectiveness, toxicity to
non-target species, cost, end site characteristics,
as well  as its solubility and persistence.
  Half-lives and  partition  coefficients  are par-
ticularly important when the application site of  a
pesticide is near  surface  waters or  is underlain
with permeable subsoil and a shallow aquifer.
Short half-lives and intermediate to large PC's are
best in  this situation.
  Many areas  of  Florida have impermeable sub-
soils which impede  deep leaching  of  soluble
pesticides. On such land, soluble pesticides with
low  PCs  and  moderate-to-long half-lives require
cautious application to prevent rapid transport in
drainage water to a nearby lake or stream. Non-
erosive soils are common to much of Florida  and
pesticides with large  PCs remain on the applica-
tion site for a long time. However, the user should
be  cautious of pesticides with long half-lives as
they are likely to  build up in the soil.
   In addition to the pesticide solubility and  soil
permeability it is important that  the pesticide's
toxicity to non-target  species  be considered.
Some of the pesticides listed in Tables 1 and 2
have severely  restricted use due to acute toxicity
or  long  half-life. An  important purpose  of the
pesticide container's  label is to instruct users  to
 apply  the  pesticide  safely  and   with  minimum
 threat  to non-target species, both on and off the
application site. Pesticide users assume  respon-
sibility to follow label Instructions. It  Is unsafe
and unlawful not to do so.
        NEED MORE INFORMATION?

   Pesticide recommendations for various crops
and pests may be obtained from the  Florida Co
operative Extension Service. Contact your count>
Extension office for this  information. For more
discussion of some of the ideas presented here.
consult these Extension publications.
   IF AS 76 A Clean Water Refresher... answers
frequently asked questions about water quality in
Florida and describes Florida's agricultural wale:
quality program.
   SP-19 Pollution Solutions for  Florida Farmers
... discusses agricultural water quality problems
in Florida and presents prevention measures.
   SL-14  The  Soil:  Our  Number  One   Waste
Disposal  System ... discusses  14  types  of
pollutants related to agriculture and explains soil
as a recycling system for treatment.
   Si.-37 Soil as a Porous  Medium ... part one in
 Basics of Soil  and Water Relationships series; ex-
 plains fundamentals of soil structure, particularly
 particle and pore size, total porosity and soil bulK
 density.
   SL-38 Retention of Water... part two in above
 series; explores the influence of soil structure on
 retention of water.
   SL-39  Movement of  Water... part three  in
 above  series;  deals with  fundamental principles
 of water  flow in soil.
   IPM-1 Integrated Pest Management Primer ...
 goals,  benefits and implementation of IPM as  ar.
 alternative to heavy use of pesticides.
                                                   I
                                                    Best
                                                    Management
                                                             ices
   This public document was promulgated at a cost ot $314.40. or 10.1 cents per copy. 10 help implement
   the agricultural element of the Florida wate: quality plan.  7-3-1M-84
 COOPERATIVE EXTENSION StRVICE. UMVERSITV oc  FLOPIDA. INSTITUTE O' FOOL) A','D AGRICULTURAL
 SCIENCES, K. ft. Ttfertllie1, director. Ir, cooperation wlir- the UniteC Slates DtPB'tment o' Ay'k^itu'e. publishes this Info1-
 matlor, to further the purpose of the May tanaJune 30.1(14 Acts of Congress. *nc Is authorises to provide r«March, •Oud-
 tloni! Information ane other Mrvlces only to Individual! anc Institutions that function without retire to race, color, e*> or
 national orlglr.. Single copies of Extension publications (*>dudlng I -H and Youtn publications) ft available fre* to Florida
 ras'deMi  from County Extension Oftlces  Information o* bulk  rates or copies fo> out-of-stete pjrcnat«r> Is available from I
 C rV. Hint or, PuMiu'io-.i DiKrit jtior. Cente-. IF AS Bui'J.r.j tf t , University o' Flcrlfli  Gaines. 'it. ftc'iOt 3 ?£ ! 1 . Before put I:: •>••>( if.,i
 put-i'tatlof.. e)dlto-s »^.c~-ic con la:1 tnis aocresi to oete'rrii* ».a liability.

-------
                                                 t
                           orous
                 4-41«i   (




                 vAirv^ATS

               Lratf-
                                                          .
|.
Uo
                          dU.-fjuil
-------
7k
V
WerljL
     X     t-S




       fVc/C JT
                        -fi'^f
      —      \/
                                                                                 &, A,
                                                                                           v
                                                                Jlts~#fa
                                             A
                                                      (
                                                                             a
                                                      iJU.

-------
e.CiT*
   U
           O
                         -fW»c^'ox\Jl

                                        £•
                                                      OC-  fi  I  )
      Ox^v
                                           "

-------
                              bores
                              r
L  --

-------
Ho
     ryy
j, rC n t^ j»U S
                CD
  vrL



-------
3-
  C -

                  t*
                                        — *o  £ X ^-4-<




                                        - O    *  X- < O
                              rVr
                       =  0
                                 r.Cx   t ^oy  x > o

      ( X > o )
/o«/C
                            xCUx.
                                              ( A ^  o )

-------
                   (11)
L
 ( /-
                 6°;  ^  0
                  *

-------

-   Hg-fcznJ
                     09.
                 - 0
               -t
      »vj    ZEva.

      3**      C  (X)  ;
Tka
       c«<\(3UL-ksv\   (_
Q
 ,
J
                                                     UA
                              AAftx   W
                                               (7
                                          0    *
                                       Jit
                                                            dvuz
                                                       (
                                                                    C/f)

-------
 f
J
                    /rau
  >  - (A, G) (
            A*C,  L
          L

-------
c *  f
  At  ^



5/1
j)'
(V/0



                             «• ; jjfTtJi ^ •
                             - ^
                           c.
                     T

-------
                               Diffusion Cell
        -18
  Df :0,i00
0 : ADD    1 :
    1,06
Esc = HEN
Length (CHS)
 fllphaA2 :G,5W
Tine =99999,

-------
•Z
   J.
                                      -
J*OL;
                     C * ^ ) •
                                              =  Mr"
                               \A
                              tf
                                 ML. r
                                        -j
                     J(koa

-------
                                    3X
         LJLTTWjp





          4
                                  «i
S
       ,   "TUusx    £, [z]
%  -
                                                 2. 'u4

-------
;. A/^J
ii
                        .     Lit   X;  cuJL -t,  W


                                       JrMiKLi^
^i


t,
                 =   C x - v.t
                  =   t
                        C(x/O &  CCx,/t,y
                                   at
         ^ I
ax
                            T
              ac
 ax,
                                  s>c
                                  a-t,
              ac
                   =  o

-------
                          = o
;|
l-i
             C Cx;t)   =    O  ;  X>v0t
                                                    BA3

-------
Jfct

-------
UtA
 Ae_
C   =
  n

 r
£*•/

                                 .  fe^COUK   V0

-------
                                         X
         Cj,



         YX
  10




   o
   2


   3
    *



   5
As
                    •/

-------
                                 SJMUOU.S  cuJ
I
ii   4«.Lbrt  d^c^iocJbL  -fco
               OU
                  C
                   
-------
'06
  V
   o

-------

                    J&JL   PlST&fO  blSf>*A*£N-E
-------
I   •
               (
         Jttvt
—   -JD
                             t •  "3;
                          60^* "^JUXTj
                 at:
                 jH*w -  Ji- I D0
                 at        ax

                                          «f
                                                   oo

-------
or
ty
         »r-
3tr
              - v
                                era
                            M IT3
                             LT
                                ~;

                             T

-------
A Microcomputer-Based  Management  Tool  for
Chemical  Movement  in Soil1
D.L. Nofzigrr
Associate Professo'
Agronomy Department
Oklahoma State L'r..\ersit\-
Stillwater. OK
A.G. Hornsby
Associate Proicsso-
Soil Science Depaimenl
University of Florida
Gainesville. K
Abstract. A management model i>
presented that can be used to make
informed derisions reparrimc the be-
havior of aprichemicaK in soiK. The
conceptual and mathematica1 ba>e-
of the model are disc ussed. ncludmc
assumption^ and limitation^ Graph-
ical outputs from the mode! illustrate
the impact that soil properties chem-
ical properties, and climatolocic tai-
lors have on chemical movement ir
soil. Data entrx routines permit ap-
plication to an\ area with appropriate
data This software require* an IBM-
PC or compalable computer with
23dK bvles random-access memo'x
one disk dn\e. and PC - or MS-DOi
version 2.0.
Introduction

The objectives of this paper are (I) to illustrate the
influence of soil properties, chemical properties.
and weather patterns upon the movement of organic
chemicals in soils. (2) to present a computer model
for estimating the movement of nonpolar organic
chemicals in soil, and (3) to present the mathemat-
ical basis of the model, assumptions inherent in it.
and its limitations.
  The depth to which chemicals move in soils de-
pends in a complex way upon a number of soil.
chemical, and weather factors. Soil properties that
influence chemical movement include: pH. bulk
density, "field capacity" and "permanent wilting
point" volumetric water contents, and soil organic-
carbon content. Properties of chemicals that influ-
ence  their movement include: partition coefficient
(Kd or Km). and degradation half-lives. Climatic and
cultural factors include plant root depth, daily rain-
fall and irrigation amounts,  and daily evapotrans-
piration amounts.
  Due to the  large number of chemicals, soils, and
weather patterns of interest, it was deemed desir-
able to develop a model for interactively simulating
the movement of chemicals. The model was to be
based on physical processes involved in the flow
process, to require information for chemicals and
soils  that are relatively easy to obtain, to be easy
to use.  and to display results in easily understood
 1 Approved for publication as Florida Agricultural Ex-
 periment Station Journal Series No. 6659.
  Address reprint requests to: A.G. Hornsby. 2169
 McCartv Hall. University of Florida. Gainesville. FL
 32611. U.S.A.
                graphical form. The following sections describe the
                model, assumptions in the model, use of the model.
                and required computer hardware.
                Description of Model

                Movement of Chemical

                The model used to estimate the position of the
                chemical in the soil is a modification of one pre-
                sented by Rao. Davidson, and Hammond [ I]. In this
                model, chemicals move only in the liquid phase in
                response to soil-water movement. In the present im-
                plementation, the soil is comprised of as many as
                25 layers or horizons. The soil properties may vary
                in different layers, but they are assumed  to be uni-
                form within each layer.
                  Let ds represent the depth of the solute front in
                a uniform  soil. The change in depth of the solute.
                A 0.

               if q « 0.
(1)
                 where q is the amount of water passing the depth
                 
-------
Chemical Movement in Soil
                                                                              51
|2) and Karickhoff |3. 4] have shown that ihc par-
tition coefficient for a particular organic chemical
in a soil divided by the organic-carbon content of
that soil is nearly constant for a wide range of soils.
Therefore, in this model the  partition coefficient is
given by
Klt
                           . or.
 where A'f)(  is  the linear sorption coefficient nor-
 malized by the organic carbon conteni (OC) of Ihc
 soil. The  use of A'()f as defined in equation (?) is
 applicable only to nonionic organic solutes.
   To use  equation (I) to predict the position of a
 chemical in a soil requires that the quantity of water
 passing the solute front be determined. The model
 estimates this  from  daily records of evapotranspi-
 ration demand and infiltration. The following steps
 must be carried out for each day in which flow is
 being simulated:

 1. Adjust water content in  the root zone for the
   evapotranspiration on that day.
 2. Adjust the water content in the root zone for any
   infiltration on that day and determine the quan-
   tity of water passing the solute depth.
 3. Determine the new solute depth.
 These steps are described below for the layered soil
 shown in Figure I. Notice that the soil contains two
 layers more than the  number of horizons. One of
 these  layers has a  lower boundary at the solute
 depth, d[s],  and an upper boundary at the  lower
 edge of the layer above the solute front. This layer
 changes in thickness and  position as the chemical
 moves. The  second layer has  a lower boundary at
 the bottom of the rooting zone. c/[RZ]. and an upper
 boundary at the lower edge  of the layer above this
 depth. These layers are inserted for computational
 convenience.

 Step 1
 Each layer in  the soil profile is assumed to be at
 "field capacity" at the beginning of day 1. As long
 as water in the root zone is available to plants, it is
 removed from the soil to meet the daily evapotrans-
 piration demand. Water is considered available if
 the  water content in any  layer of the root zone is
 above the "permanent wilting point" of the  soil in
 that layer. If Q(j) represents the volumetric water
 content of layer j. the available water in that layer.
 AWl/). is given by
           AW(/) =
                             (4)
 where t(/) is the thickness of the layer and 6PWP0)
 is the volumetric water content at the "permanent
                                    «[0]

                                    d[8]
                                               VXl*€TRlC WATER CONTENT
                                             0.1    0.2     0.1    0.4    0.5
                                                     d[2]	'	*—.—
                                 O
                                 OC
                                 a
S  d[3]
o
z
X
t-
Q.
u
o
                                    d[4]
                              •  Siiitiiirrt mint)
                                little it lirlui I
                                linn it rut tin

                              — Itttu itliriin J
                                                                 llttll it lulni 1
                                                               F.C.
                                 Lilt tirilli lltiid<
                                 ti iii| In* till
                                 Fig. 1, Representation of soil layers used in the compu-
                                 tational scheme. The solid vertical line represents the vol-
                                 umetric water conteni at lime /,.
                                 wilting point" of the layer. The total available water
                                 in the root zone. A\V,ota|. is the sum of the amounts
                                 available in each layer in the root zone. If the total
                                 available water is  greater than the evapotranspira-
                                 tion (ET) demand  for the day. the water content of
                                 each layer in the root zone is decreased in propor-
                                 tion to the amount of water available in that layer.
                                 That is.
     8l/)  =
                                                  - [ET * AW(/)]'
                                                                              (5)
where 8'(/) is the water content prior to adjustment.
If the total available water is less than the evapo-
transpiration demand.

                  ei./i  = ePWPo")               (6)
for all layers in the root zone.

Step 2

The water content of each layer in the root zone
must  be adjusted when an infiltration event occurs.
Starting with the la\er closest to the soil surface (j
= 1). the soil-water deficit for that layer. swd(/>. is
determined using the equation

                                             (7)

-------
52
                         D.L. Nofzipcr and A.G. Hon»>K
where flj.r(.») i«- the volumetric water content of the
layer at "field capaciu." If the infiltrating amount.
/(/). is greater than swd(/) then.
                 W.M = eK-i/).
and
            /

where time ^ the elapsed time since  the chemical
was applied ir*J half-life is the biologic degradation
half-life of the chemical.
Assumptions in Model

The following assumptions arc used in this model:

 1. All soil waier residing in pore spaces participates
   in the transport  process. Soil water initiallv
   present in the  profile  is completely displaced
   ahead of water entering at the soil surface. Rao
   and colleagues 11) present data from different re-
   searchers indicating  that these assumptions are
   valid  for many  soils. If they are not  valid and a
   portion of the soil water is bypassed during flow.
   this model would tend to  underestimate the
   depth of the chemical front.
 2. Water entering the soil redistributes instanta-
   neously to "field capacity." This assumption is
   approached for coarse textured soils.  If the
   water redistributes  more  slowly as  in fine tex-
   tured soils, the depths predicted here are likely
   to be associated with an elapsed time a few days
   later  than that specified.
 3. Water is removed by  evapotranspiration from
   each  layer in the root  zone in proportion to the
   relative amount of water available in that layer.
   A uniform root distribution  with depth is as-
   sumed. The validity of this  assumption will de-
   pend upon the root distribution in the soil. It will
   not be strictly  valid for many situations. More
   precise schemes for dealing with evapotranspi-
   ration would require information about the root
   distribution and the  soil hydraulic properties.
 4. Upward movement of water does not occur an>-
   where in the soil profile. Water is lost from the
   root  zone by evapotranspiration. but soil water
   in the roo; zone is not replenished from below.
 5. The adsorption process can be described b> a
    linear, reversible, equilibrium model. If the sorp-
   tion coefficient is described  by  nonlinear iso-
   therm, the partition coefficient decreases with
    increasing concentration of the chemical. Thus.
    the depth to which the chemical will be leached
    will depend upon the concentration. This aspect
    is probabh not significant for the concentration
    range of imerest in most agricultural application?-

-------
Chemical Mo\cmcni in Soil
   |5). When adsorption equilibrium is not instan-
   taneous, the chemical will be leached to a greater
   depth than predicted here. Irreversible sorption
   would result in less leaching.
6. The half-life for biologic degradation  of the
   chemical is constant with time and soil  depth.
   Degradation rale coefficients arc dependent
   upon a variety of environmental  factors,  pri-
   marily  temperature and  soil-water  contcni.
   Hence, seasonal changes in rale coefficient can
   be expected. Also, with decreasing microbial ac-
   tivity at greater soil depths, the degradation rate
   coefficient may decrease with  depth. Sufficient
   data are not available to formulate mathematical
   relationships to describe these  effects.
wilting point." and the depth of the bottom of the
horizon. Options are provided  in the software to
enable the user to enter these data in files, to modify
data previously stored in the files, and to displax
the contents of the file on the screen or printer. The
use of files enable the user to make repeated sim-
ulations without reentcring the data. Tables 1 and 2
illustrate the data displayed by the software for the
soils and chemicals used in Figure 2.
  The model also requires daily amounts of effec-
tive rainfall or infiltration and evapotranspiration
(Fig. 3). These data are stored in separate files. In-
formation can be entered in English or metric units.
Options  for entering, modifying, and displaying
data are  similar to those for the chemical and soil
data.
 Use of Model

 The computer software for this  model is menu
 driven with the following options:  (I) calculate the
 movement of a chemical in a soil. (2) enter, modify.
 or display data for soils or chemicals, and (3) enter.
 modify, or display daily effective rainfall and evapo-
 transpiration amounts. The steps involved in each
 of these options are discussed below.
   The calculation of chemical  movement requires
 the user to specify the chemical and soil of interest.
 the depth to the bottom of the root zone for the
 crop being grown, the names of data files containing
 weather data for the location of interest, the depth
 of application of the chemical,  the date of applica-
 tion, and the final data to be included  in this sim-
 ulation.  Depths  entered by the user and those cal-
 culated can be in either English or metric units. The
 calculated depth of the chemical front as a function
 of time  since application is displayed  graphically.
 Depths for several chemicals can  be displayed on
 one graph if desired. If the computer has high res-
 olution graphics, bar graphs of daily rainfall as a
 function of time are also displayed.  Calculated
 depths of the chemical front and the relative amount
 of chemical remaining in the soil profile can be dis-
 played in tabular form.
   For each chemical used in the model, the parti-
 tion coefficient normalized for organic carbon and
 the  half-life for  biologic degradation are required.
 The model includes these data for approximately 40
 chemicals. A soil name, identifer. and  the number
 of horizons in the profile are required for each soil.
 For each horizon, the model  requires  the percent
 organic carbon,  bulk density, and volumetric water
 contents at "field  capacity" and "permanent
Hardware and Software Requirements

This software requires an IBM PC or a compatable
computer  with 256K bytes  of random access
memory and one disk  drive. An IBM compatible
color/graphics card is very desirable, but it is not
essential. An 808" numeric coprocessor can be uti-
lized for enhanced speed. (Simulating movement for
one year and drawing the graphs will require ap-
proximately 10 s with the 8087  processor or ap-
proximately 1 min without it). The operating system
must be PC-DOS 2.0 or MS-DOS 2.0 (or a more
recent version). The software is available from the
Institute for Food and Agricultural Sciences. Uni-
versity of Florida.
 Use of the Model as a
 Management Tool

 Increasing the use efficiency of applied agrichemi-
 cals (fertilizers and pesticidesl requires a good un-
 derstanding of the factors that control  their fate in
 the environment. Soil-applied agrichemicals and
 others reaching the soil as overspray or foliar wash-
 off may leach through the soil and pose a threat to
 ground water quality, remain at or near the soil sur-
 face and be carried by surface runoff to streams or
 surface impoundments, or be metabolized by  soil
 microbes and'or plants, thereby posing no environ-
 mental impact.
   Agrichemicals are applied to the  soil to enhance
 crop production and economic return to investment
 of land and capital. Knowledge of the fate of applied
 agrichemicals can improve nutrient- and pesticide-

-------
54
                   D.L. Nof/ipci and A.G. Hormtn
Table I. Soil parameters for Tavares fine sand (Typic QuartzipsammcntM and Orangchurp fine sand\ loam (Typic
PaJeudults).
Soil name
Soil identifier


Number of horizons in profile

Horizon
1
t
3
4
5
6
Dcplh
(cm)
10.2
20.?
5?.?
106.7
121.9
203.2
Organic carbon
7,
0.55
0.42
O.oy
0.06
0.04
o.o:
Tax-are-, fine sand
S27-8-II-6I
6
Water content
ul -O.I bar
6.5
7.2
5.2
5.4
6.1
11.9


Water content Bulk dcn-.il>
ai - 15 bar* g cc
1.3
1.5
0.8
0.7
0.8
1.0
.42
.44
.5(1
.5(>
.56
.58
Soil name
Soil identifier
Oranpeburg fine sand> loam
S37-8-II-6)
Number of horizons in profile

Horizon
1
2
3
4
5
6
Depth
(cm)
i3.o
25.0
41.0
102.0
155.0
203.0
Organic carbon
T,
2.48
0.66
0.40
0.14
0.10
0.06
6
Water content
at -O.I bar
21.9
23.7
33.2
30.0
32.8
30.8

Water conten; Bulk density
at -15 bars p cc
9.4
11.5
12.2
14.1
19.3
18.0
.42
.58
.50
.53
.71
.71
 Table 2. Chemical data for Diuron and Picloram.

     Common name                   Diuron
     Trade name                      Karmex
     Trade name                      lirox D
     Trade name                      Direx 4L
     Trade name                      Diurol
     Partition coefficient (ml/g OC)        383

     Half-life (days)                   328
     Common name                   Picloram
     Trade name                      Tordon
     Trade name                      Tordon 22K
     Trade name                      Amdon
     Trade name                      Grazon
     Partition coefficient (ml/g OCi        26
     Half-life (davs)                   138
 use efficiency as well as reduce unwanted environ-
 mental consequences. The software described
 herein can provide insight and increased under-
 standing of the interactions that influence the fate
 of agrichemicals in soil-water-plant systems.
   With the exception of volatile compounds, such
 as NH3 and fumigants, chemicals move through soil
 dissolved in  soil water. Wherever the soil water
 moves, the dissolved chemicals move  with it. The
 chemical may lag behind the water front due to re-
 tardation as shown by Equation (2). Figure 2 shows
 a simulation of the depth distribution of three chem-
icals (diuron. picloram. and nitrate ion) in two con-
trasting soih (Tavares fine  sand and Orangeburg
fine sandy loam) under two different periods of rain-
fall and evapotransporation (Fig. 3). The differences
in depth of leaching in a given soil reflect differ-
ences in properties of the chemicals themselves:
namely, the partition coefficients as shown in Table
2. Nitrate nitrogen has a partition coefficient of 0:
that is. no sorption is expected. Comparing Figures
2A and 2B illustrates the significant impact that dif-
ferences in soil properties alone make in the depth
of leaching of a particular chemical. Soil properties
for the two soils used in this example are given in
Table 1.  Differences in  organic carbon and -0.01
MPa (-0.1 bar) water  contents  are the principal
contributors to this contrast.
  By comparing Figures 2A with 2C. and 2B with
2D, the effect of climatic parameters (rainfall and
evapotranspiration) can be seen for the Tavares and
Orangeburg soils, respectively.  The rainfall and
evapotranspiration distributions for both different
time periods are given in figures 3A and B and 3C
and D.  respectively.  The differences in depth of
leaching  reflect  the  differences in amount and
timing of rainfall versus evapotranspiration and the
resultant soil-water deficit.  The total rainfall rep-
resented in Figure 3A is 804.2 mm versus 65S.9 mm
in Figure 3C. Potential evapotranspiration for  the

-------
Chemical Movement in Soil
u
1.0

2.0

§ 3'°
(C
u.
-•an



~~ ^ 	 1






— 1
t


1 — v
\v
s
>\_
A




t-.
Sv
^

















-v<





!






TAVARES U,

g °0
£ o
u
u-
x -
0.
Ill
0 2.0
3£


I
30

i i 	 .3
60 90





IjO

2 0

3.0



-, — _,
.^
w^ —
L ^ \ . !
h ~^ 1
- C ;

• TAVARES
i i i i
120 0 30 60 90 120



^-i \ '




. .^


B
^-u
L_
> 	


















t
1







ORANGEBURG

0

60 -

~ 60-
E
£
j 40-
IL
Z
C 20-
n .








,
30
,



60 90




1.0

2.0
3.0

A A
'-^ 	 	 '
• »-, 	 * I
Ltgtnn tor ni 9niM , Fig. 2. Depth of chemica
i Diuron fronts as a function of time
2 p'cior«T. . since application date. Ap
3Nltrtt"on plication dates for Panels^
D Rcct z.»« and B and for Panels C anc
ORANGEBURG D were 2-1-83 and 4-1-83
i i . t i resncclivelv.
120 0 30 60 90 120
ELAPSED TIME (DAYS)
• A _ 	 . 	 r 	 	












1







|























|
















A C







|






60-




I
40- 1
2 0 *
i n i ill 1 1 ii jit
                        40
                                      80
                                                    120
                                                                          40
                                                                                        80
                                                                                                      120
     100 -r
                                                        100 T-
                                                         80
                                                         60
                                                         40
                                                         20
                                                    120
                                                                          40
                                                                                        80
                                                                                                       120
                     ELAPSED TIME , (DAYS)                             ELAPSED TIME . (DAYS)
 Fig. 3. Distribution of rainfall and evapotranspiraiion as a function of time since application. Panels A and B are for
 the period 2-1-83-5-31-83. Panels C and D are for the period 4-1-83-7-29-83.

-------
                                                                      D.L. Nof/ipcr and A.G. Hornxb)
same periods was 540.5 mm and M>2.4 mm. respec-
tively. Nci evapotranspiralion varied, due to differ-
ences in the water-holding capacities of the two
soils.
  Tabular outputs of data for date of rainfall, rainfall
amount,  depth of chemical front, and relative
amount of chemical remaining in the soil profile as
a function of elapsed time arc also provided as an
option in  the software. By examining the behavior
of selected alternative chemicals using  this pro-
gram, one can  make informed choices and thereby
maximize the efficacy of the chemical used.

References
 1.  Rao. P.S.C.. Davidson. J.M.. Hammond.  L.C. Esti-
    mation of nonreactive and reactive  solute front lo-
  cations in soils. In Proc. Hazard. Wastes Res. Symp.
  EPA-600 19-76-015. Tucson. AZ. 1976. pp. 235-241.
2. Hamaker. J.W.. Thompson. J.M. Adsorption. In
  Goring. C.A.I.. Hamaker. J.W. (eds.). Organic
  Chtmirah it: tin  Environment. New  York: Marcc!
  Dckker. 1972. pp. 49-14?.
3. Karickhoff. S.W. Semi-empirical estimation  of sorp-
  lion of hxdrophobic pollutants  on natural sediment*
  and soils. Chcmosphcrc  10:833-8-16.  I9SI.
4.  Karickhoff. S.W. Organic pollutant sorption in
  aquatic systems. J. Hydr. Eng.  110:707-735. I9M.
5. Rao. P.S.C.. Davidson. J.M. Estimation ol pesticide
  retention and transformation parameters required in
  nonpoint source pollution models. In Overcash.
  M.R.. Davidson. J.M. (eds.). Envintnnicnttil Imputi
  of .\onpoint Source Pollution.  Ann Arbor. Ml: Ann
  Arbor Science Publishing. 1980. pp. 23-67.

-------
        PISTON DISPLACEMENT OF WATER IN SOIL

   CONSIDER A SOIL WITH A BULK DENSITY OF 1.58
AND A SATURATED WATER CONTENT OF 0.4 CM3/CM3.

1. AT SATURATION, HOW MUCH WATER WILL BE STORED IN
   THE TOP 10 CM OF THIS SOIL? IF THE WATER CONTENT
   IS ONLY 0.35 CM3/CM3, HOW MUCH WATER IS STORED IN
   THE TOP 10 CM?

2. IF THE SOIL IS INITIALLY DRY AND 5 CM OF WATER IS
   APPLIED TO THE SURFACE, APPROXIMATELY HOW DEEP
   WILL THE WET FRONT BE AT IMMEDIATELY AFTER
   INFILTRATION STOPS? IF THE WATER REDISTRIBUTES TO
   A WATER CONTENT OF 0.3 CM3/CM3 VERY QUICKLY AFTER
   INFILTRATION, HOW DEEP WILL THE WET FRONT BE?

3. IF THE SOIL HAD AN INITIAL WATER CONTENT OF 0.15
   CM3/CM3, WHERE WILL THE WET FRONT BE IMMEDIATELY
   AFTER INFILTRATION? WHERE WILL THE WET FRONT BE
   AFTER THE WATER REDISTRIBUTES TO A WATER CONTENT
   OF 0.3 CM3/CM3?

4. WHERE is THE LEADING EDGE OF THE INFLOWING WATER
   IN CASES C AND D ABOVE?

-------
          PISTON DISPLACEMENT OF  WATER  IN  SOIL

   CONSIDER A SOIL WITH A BULK DENSITY  OF  1.58
   AND A SATURATED WATER CONTENT  OF 0.4 CM^/CM^.

1. AT SATURATION, HOW MUCH WATER  WILL BE STORED IN
   THE TOP 10 CM OF THIS SOIL? IF THE WATER CONTENT
   IS ONLY 0.35 CM3/CM3, HOW MUCH WATER IS STORED IN
   THE TOP 10 CM?

-------
          PISTON DISPLACEMENT OF WATER  IN SOIL

   CONSIDER A SOIL WITH A BULK DENSITY  OF  1.58
   AND A SATURATED WATER CONTENT OF 0.4 CM3/CM^.

2. IF THE SOIL IS INITIALLY DRY AND 5 CM OF WATER IS
   APPLIED TO THE SURFACE,  APPROXIMATELY HOW DEEP
   WILL THE WET FRONT BE AT IMMEDIATELY AFTER
   INFILTRATION STOPS? IF THE WATER REDISTRIBUTES TO
   A WATER CONTENT OF 0.3 CM^/CM^  VERY  QUICKLY  AFTER
   INFILTRATION, HOW DEEP WILL THE WET  FRONT BE?

-------
          PISTON DISPLACEMENT OF WATER  IN SOIL

   CONSIDER A SOIL WITH A BULK DENSITY  OF 1.58
   AND A SATURATED WATER CONTENT OF 0.4
3. IF THE SOIL HAD AN INITIAL WATER CONTENT OF 0.15
   CM^/CM^,  WHERE WILL THE WET FRONT BE IMMEDIATELY
   AFTER INFILTRATION? WHERE WILL THE WET FRONT BE
   AFTER THE WATER REDISTRIBUTES TO A WATER CONTENT
   OF 0.3 CM3/CM3?

-------
            PISTON DISPLACEMENT OF WATER IN SOIL

     CONSIDER A SOIL WITH A  BULK DENSITY OF 1.58
     AND A SATURATED WATER CONTENT OF 0.4
  4.  WHERE is THE LEADING  EDGE  OF THE  INFLOWING WATER
     IN CASES C AND D  ABOVE?
I

   I

-------
             EVAPOTRANSPIRATION  IN  CMLS
1. WATER is REMOVED FROM ROOT ZONE ONLY.
2. WATER is REMOVED FROM EACH HORIZON IN
   PROPORTION TO "AVAILABLE WATER11 IN THE
   ROOT ZONE.
3. NO WATER MOVES UPWARD FROM BELOW THE ROOT
   ZONE.

-------
        AVAILABLE WATER IN ONE LAYER OF SOIL
THE AVAILABLE WATER IS TAKEN TO BE THE AMOUNT OF
WATER STORED IN THE LAYER ABOVE THE WATER CONTENT
CORRESPONDING TO THE "PERMANENT WILTING POINT11 OF
THE SOIL.
AW = (8 - 6pWP) * T

   WHERE
     0 IS THE AVERAGE WATER CONTENT IN THE LAYER
     0pwp IS THE PERMANENT WILTING POINT
     T IS THE THICKNESS OF THE LAYER.

-------
  CHANGE IN WATER CONTENT DUE TO EVAPOTRANSPIRATION
                   (ONE LAYER ONLY)

IF THE LAYER CONTAINS ENOUGH WATER TO MEET THE
EVAPOTRANSPIRATION DEMAND, THAT AMOUNT OF WATER IS
REMOVED FROM THE LAYER AND THE AVERAGE WATER CONTENT
IS ADJUSTED ACCORDINGLY.  IF THE LAYER DOES NOT
CONTAIN SUFFICIENT WATER, THE AVERAGE WATER CONTENT
IS SET EQUAL TO THE "PERMANENT WILTING POINT11 OF THE
SOIL.

IF AW < ET THEN 6 = 9PWP

IF AW > ET THEN 8=0'- ET/T

   WHERE
     0  IS AVERAGE WATER CONTENT AFTER CORRECTING
        FOR EVAPOTRANSPIRATION,
     0' IS AVERAGE WATER CONTENT BEFORE CORRECTING
        FOR EVAPOTRANSPIRATION,
     0pwp IS THE PERMANENT WILTING POINT
     T IS THE THICKNESS OF THE LAYER.
     AW IS THE AVAILABLE WATER IN THE LAYER
     ET IS THE AMOUNT OF WATER LOST TO
        EVAPOTRANSPIRATION.

-------
CHANGE IN WATER CONTENT DUE TO EVAPOTRANSPIRATION
           (MULTIPLE LAYERS)

STEPS:
 1. CALCULATE THE AVAILABLE WATER FOR EACH LAYER
    IN ROOT ZONE.
 2. CALCULATE THE TOTAL AVAILABLE WATER.
 3. REMOVE WATER FROM EACH LAYER IN PROPORTION TO
    THE AMOUNT OF WATER AVAILABLE IN THAT LAYER
    BUT DO NOT LET THE WATER CONTENT DECREASE
    BELOW THE PERMANENT WILTING POINT.

FOR LAYER J THEN

9(J) = 9'(j) - [ET*AW(j)]/AWTOTAI_*T(j)

   WHERE
     9(j) IS THE AVERAGE WATER CONTENT OF LAYER J
          AFTER ADJUSTING FOR ET,
     6'(j) IS THE AVERAGE WATER CONTENT OF LAYER J
           BEFORE ADJUSTING FOR ET,
     AW(j) IS THE AVAILABLE WATER IN LAYER J,
     AWTQTAL IS THE AVAILABLE WATER IN THE
           ROOT ZONE,
     T(j) IS THE THICKNESS OF LAYER J.

-------
                 INFILTRATION IN CMLS

ASSUMPTIONS:
  1. WATER INFILTRATES AS PISTON DISPLACEMENT,
  2. WATER REDISTRIBUTES TO "FIELD CAPACITY"
     INSTANTANEOUSLY.

-------
            WATER PASSING SPECIFIED DEPTH

ASSUMPTION: INFILTRATING WATER ACCUMULATES IN
            UPPERMOST LAYER UNTIL THE AVERAGE
            WATER CONTENT OF THAT LAYER REACHES
            FIELD CAPACITY. EXCESS WATER, IF
            ANY, FLOWS INTO NEXT LAYER.

COMPUTATIONAL CONVENTION:

   SOIL LAYERS CAN BE DIVIDED BY BOUNDARIES
   BETWEEN THE NATURAL SOIL HORIZONS OR ANY
   OTHER DEPTH OF INTEREST. IN THIS MODEL,
   ONE LAYER ALWAYS ENDS AT THE ROOT ZONE DEPTH
   AND ONE LAYER ENDS AT THE DEPTH OF THE
   CHEMICAL.

PROCEDURE:

   SINCE THE MODEL ASSUMES NO UPWARD FLOW AND
                               \
   SINCE THE SOIL IS ASSUMED TO BE AT "FIELD
   CAPACITY" WHEN THE SIMULATION BEGINS, ANY
   WATER PASSING THE ROOTING DEPTH MOVES
                                         i
   DOWNWARD PAST ANY GREATER DEPTH.

-------
1. THEREFORE, IF DEPTH, D > ROOTING DEPTH, THE
   AMOUNT OF WATER PASSING THIS DEPTH IS THE
   AMOUNT PASSING THE ROOT ZONE DEPTH.

2. FOR DEPTH < ROOT ZONE DEPTH DO THE FOLLOWING:

      FOR EACH SOIL LAYER ABOVE THE DEPTH OF
      INTEREST,

      A. CALCULATE THE AMOUNT OF WATER NEEDED TO
         REPLENISH THE LAYER TO FIELD CAPACITY.

      B. IF THE INFILTRATING AMOUNT IS LESS THAN
         THIS AMOUNT,
         A. SET THE WATER CONTENT 6(j) OF THE
            LAYER TO ,

               6(J) = 6'(J) + I(J)/T(J)

            WHERE
               9'(j) IS THE WATER CONTENT OF THE
                     LAYER BEFORE INFILTRATION,
               I(j)  IS THE WATER INFILTRATING
                     LAYER J,
               T(j) IS THE THICKNESS OF THE LAYER
          B. SINCE ALL THE WATER IS USED UP, SET
             I(j+l) EQUAL TO ZERO.

-------
C. IF THE AMOUNT INFILTRATING IS GREATER
   THAN THE AMOUNT NEEDED TO REPLENISH
   THE LAYER,
   A. SET THE WATER CONTENT 9(j) OF THE
      LAYER TO 6FC(j).
   B. CALCULATE THE AMOUNT PASSING THE
      BOTTOM OF THE LAYER. THIS IS THE
      AMOUNT INFILTRATING THE NEXT LAYER,
      I (J+1) .

-------
              DEPTH OF CHEMICAL IN CMLS

PROCESS: FOR EACH DAY:

  1. ADJUST THE WATER CONTENT IN THE SOIL FOR
    THE DAILY EVAPOTRANSPIRATION AND INFILTRATION.

  2. CALCULATE THE AMOUNT OF WATER, Q, PASSING THE
     DEPTH OF THE CHEMICAL.

  3. IF Q > 0, THE NEW DEPTH D IS GIVEN BY

           D = D' + Q/(R*8FC)

     WHERE
        D' IS THE OLD DEPTH OF THE CHEMICAL,
        R  IS THE RETARDATION FACTOR FOR THE
           CHEMICAL IN THIS SOIL, AND
        6FC IS THE WATER CONTENT AT FIELD
           CAPACITY.

  4. IF Q = 0, THE NEW DEPTH D EQUALS THE OLD
               DEPTH D'.

-------
                 RETARDATION FACTOR
ASSUMPTION: THE ADSORPTION OF THE CHEMICAL ON THE
            SOIL IS DESCRIBED BY LINEAR,
            REVERSIBLE, EQUILIBRIUM MODEL.

THE RETARDATION FACTOR R IS GIVEN BY
     R = 1 +
WHERE
      IS THE SOIL BULK DENSITY,
   KD IS THE LINEAR SORPTION COEFFICIENT,
   0FC IS THE WATER CONTENT AT FIELD CAPACITY.

NOTE: IN CMLS, KQ MAY BE ENTERED DIRECTLY FOR
      EACH SOIL AND CHEMICAL OR IT CAN BE OBTAINED

 USING THE RELATION

   K  = K   * OC
WHERE
   KQC IS THE PARTITION COEFFICIENT NORMALIZED
       BY THE ORGANIC CARBON CONTENT OF THE
       SOIL AND
   OC  IS THE ORGANIC CARBON CONTENT OF THE SOIL
       HORIZON.

-------
            DEGRADATION OF CHEMICAL IN CMLS

ASSUMPTIONS:
   1. FIRST ORDER DEGRADATION.
   2. HALF-LIFE is CONSTANT WITHIN ONE SOIL
      HORIZON.
   3. HALF- LIFE is CONSTANT OVER TIME.
   4. FOR DEGRADATION PURPOSES, ALL OF CHEMICAL IS
      ASSUMED TO BE AT THE DEPTH CALCULATED.

CALCULATION:

   THE FRACTION OF THE CHEMICAL REMAINING IN THE
   SOIL PROFILE AT THE END OF DAY I IS GIVEN BY
          = F(I-1)*EXP{-LN(2)/HALF_LIFE(D(I))}
                                FOR I = 1,2, ...
     F(0) = 1

   WHERE
     F(l) IS THE FRACTION REMAINING,
     D(l) IS THE DEPTH OF THE CHEMICAL ON DAY I,
     HALF_LIFE(D(l)) IS THE DEGRADATION HALF- LIFE
                   OF THE CHEMICAL AT DEPTH  D(l).

-------
           RETARDATION FACTOR IN CHEMRANK

CONCEPT:  To RANK THE POTENTIAL FOR CHEMICALS TO
         LEACH PAST A SPECIFIED SOIL DEPTH BASED
         ON THE TIME REQUIRED FOR THE CHEMICALS TO
         MOVE THAT DISTANCE IN THE SOIL.  CHEMICALS
         WHICH MOVE FASTER ARE MORE APT TO REACH
         GROUND WATER.

CALCULATION:
   THE DISTANCE TRAVELED BY A CHEMICAL PER UNIT OF
   WATER APPLIED IS INVERSELY PROPORTIONAL TO THE
   RETARDATION FACTOR FOR THE CHEMICAL. THUS,
   CHEMICALS WITH SMALL RETARDATION FACTORS MOVE
   MORE RAPIDLY THAN CHEMICALS WITH LARGER VALUES.

-------
         RETARDATION FACTOR FOR UNIFORM SOIL


FOR A UNIFORM SOIL, THE RETARDATION FACTOR, RF,
IS GIVEN BY

  RF = 1 + {^KD + (f -eFC)KH}/8FC

WHERE
     IS THE SOIL BULK DENSITY,
      IS THE PARTITION COEFFICIENT,
   f IS THE POROSITY OF THE SOIL,
   0FC IS THE SOIL WATER CONTENT AT
       "FIELD CAPACITY", AND
   KH IS THE DIMENSIONLESS HENRY'S CONSTANT
       FOR THE CHEMICAL.

-------
         RETARDATION FACTOR FOR LAYERED SOIL

PROBLEM: To REACH THE GROUND WATER, A CHEMICAL MAY
         NEED TO PASS THROUGH SEVERAL LAYERS. HOW
         IS RF TO BE CALCULATED FOR SUCH A SOIL?

SOLUTION:

   IN A UNIFORM SYSTEM, THE TIME, T, REQUIRED FOR
   A CHEMICAL TO REACH A SPECIFIED DEPTH D IS

     T = D * RF * 0FC /Q

   WHERE
     RF IS THE RETARDATION FACTOR FOR THE CHEMICAL,
     9FC IS THE WATER CONTENT AT FIELD CAPACITY,
     AND
     Q IS THE FLOW RATE IN THE SOIL.

   THUS, RF = T * Q / (D * 6FC).

-------
STEPS:
  1. CALCULATE RF AND T FOR EACH LAYER ABOVE
     THE CRITICAL DEPTH D ASSUMING THE FLOW
     RATE IN THE SOIL IS UNIFORM ACROSS ALL
     DEPTHS.

  2. CALCULATE THE TOTAL TIME, TTOTAL,
     REQUIRED TO MOVE TO DEPTH D BY FINDING THE
     SUM OF THE TIMES FOR EACH LAYER.

  3. CALCULATE RF FOR THE LAYERED SYSTEM USING

          = TTOTAL * Q /
     WHERE
       6FC IS THE DEPTH-WEIGHTED AVERAGE
               WATER CONTENT AT FIELD CAPACITY.

-------
           ATTENUATION FACTOR IN CHEMRANK

CONCEPT: To RANK THE POTENTIAL FOR CHEMICALS
         TO LEACH PAST A SPECIFIED SOIL DEPTH
         BASED ON THE RELATIVE AMOUNT OF CHEMICAL
         PASSING THAT DEPTH.  THE GREATER THE MASS
         OF CHEMICAL MOVED BEYOND THE CRITICAL
         DEPTH, THE GREATER THE POTENTIAL TO
         ADVERSELY AFFECT GROUND WATER.

-------
CALCULATION:
   THE RELATIVE AMOUNT OF A CHEMICAL PASSING
   THROUGH A SOIL IN TIME T IS GIVEN BY

     Mi/Mo = EXP(-K*T)

   WHERE
        IS THE AMOUNT OF CHEMICAL ENTERING THE
         TOP OF THE SOIL AT TIME ZERO,
        IS THE AMOUNT OF CHEMICAL LEAVING THE
         LAYER OF THE SOIL AT TIME T, AND
     K   IS THE DEGRADATION CONSTANT FOR THE
         CHEMICAL IN THE SOIL.
                            (K = LN(2)/HALF-LIFE)

   THUS THE ATTENUATION FACTOR, AF, FOR THE LAYER
   IS GIVEN BY

     AF = EXP(-K*T)

-------
   IF N SOIL LAYERS EXIST ABOVE THE CRITICAL DEPTH,
THEN
     MN/MO =

   AND

     AF = EXp(-Ki*Ti-K2*T2-...-KN*TN)

   WHERE
     MN IS THE AMOUNT OF CHEMICAL PASSING THE
        CRITICAL DEPTH,
     MQ IS THE AMOUNT OF CHEMICAL AT THE SOIL
        SURFACE,
     Kj; IS THE DEGRADATION CONSTANT FOR LAYER
            I, AND
     Tj IS THE TIME REQUIRED TO MOVE THROUGH
            LAYER i (DETERMINED AS IN RF FACTOR)

-------
ir/EPA
            United States
            Environmental Protection
            Agency
            Robert S Kerr Environmental
            Research Laboratory
            Ada OK 74820
EPA/600-8-88-001
January 1988
            Research and Development
Interactive
Simulation  of the
Fate of Hazardous
Chemicals During
Land Treatment of
Oily Wastes:
           RITZ User's Guide

-------
                                                 EPA/600/8-88/001
                                                 January 1988
        INTERACTIVE  S IMULATION

                        OF  THE

   FATE  OF  HAZARDOUS  CHEMICALS

                        DURING

LAND  TREATMENT  OF  OILY  WASTES :

             RITZ  USER'S  GUIDE
                             by
               D.L.  Nofziger,  J.R. Williams,
                   Department of Agronomy
                  Oklahoma State  University
                 Stillwater,  Oklahoma  74078

                     and  Thomas E.  Short
                          CR-812808
                       Project Officer

                       Thomas E.  Short
           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

-------
                                  DISCLAIMER

The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under cooperative agreement No.
CR-812808 to the National Center for Ground Water Research.   It has been
subjected to the Agency's peer and administrative review, and it has been
approved for publication as an EPA document.  Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.

-------
                                   ABSTRACT

An interactive software system was developed to enable decision makers to
simulate the movement and fate of hazardous chemicals during land treatment of
oily wastes. The mathematical model known as the Regulatory and Investigative
Treatment Zone Model or RITZ was developed and published earlier by
Short(1985). The model incorporates the influence of oil in the sludge, water
movement, volatilization, and degradation upon the transport and fate of a
hazardous chemical. This manual describes the conceptual framework and
assumptions used by Short (1985) in developing the model. It then explains the
micro-computer hardware and software requirements, the input parameters for
the model, and the graphical and tabular outputs which can be selected.
Illustrations of the use of the software are also included. The computational
equations developed by Short (1985) are presented for completeness but are not
derived.
                                     ,111

-------
                                   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 of
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 and the saturated zones of the subsurface environment;
(b) define the processes to be used in characterizing the soil and subsurface
environment 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 and limitations of using natural
processes, indigenous to the soil and subsurface environment, for the
protection of this resource.

This user's guide serves the purpose of instructing the user to the execution
of a software package based on the Regulatory and Investigative Treatment Zone
(RITZ) model.  The guide should allow easy access to information critical to
the development of an understanding of the transport and fate of hazardous
chemicals applied during land treatment.
                                       Clinton W. Hall
                                       Director
                                       Robert S. Kerr Environmental
                                         Research Laboratory
                                      av

-------
                          TABLE OF CONTENTS




Introduction 	  1




Basic Concepts, Assumptions, and Limitations 	  1




Software Overview  	  4




Hardware and Software Requirements 	 .  	  6




Operating Conventions	  6




Getting Started  	  9




Example of Software Use	- .  .   11




 - Introduction  	   11




 - Parameter Entry 	   11




 - Output Selection  	   16




 - Output Examples 	   18




File Structure	43




References Cited 	   44




Appendix




 - Mathematical Basis of Model 	   46




 - Input Parameter Estimation  	   56




 - Parameter Averaging 	   58




Index	60
                                   v

-------
                                 INTRODUCTION

The Regulatory and Investigative Treatment Zone Model, RITZ, (Short, 1985) was
developed to help decision makers systematically estimate the movement and
fate of hazardous chemicals during land treatment of oily wastes. The model
considers the downward movement of the pollutant with the soil solution,
volatilization and loss to the atmosphere, and degradation. The model
incorporates the influence of oil upon the transport and fate of the
pollutant. This RITZ model forms the basis of this interactive software
system. The software enables users to conveniently enter the required soil,
chemical, environmental, and management parameters and checks the validity of
these entries. The user may then select graphical and tabular outputs of the
quantities of interest.

This manual describes the basic concepts of RITZ and lists the inherent
assumptions. The manual, also, describes the use of the interactive software
and the hardware and software requirements for it. Illustrative examples of
the software are presented. The appendix includes a list of the mathematical
equations used in the software.

                 BASIC CONCEPTS, ASSUMPTIONS, AND LIMITATIONS

A  land treatment site is illustrated in Figure 1. The treatment site consists
of two soil layers called the plow zone and the treatment zone. The sludge
(waste material) containing oil and pollutant is applied to the plow zone. It
is thoroughly mixed with the soil in that layer. As time passes the pollutant
and oil are degraded. Some of the pollutant is carried down through the soil
with percolating water. Some of the pollutant is volatilized and moves  into
the air above the treatment site.
 The  following  assumptions were made  by  Short(1985)  in developing this model.

   1.  Waste material  is  uniformly mixed in  the plow zone.

   2.  The  oil  in  the  waste  material  is  immobile.  It never  leaves the plow
      zone. Only  the  pollutant moves with the soil  water.

   3.  The  soil properties are uniform from  the soil surface  to  the  bottom  of
      the  treatment zone. This assumption will rarely,  if  ever, be  met  in  the
      field. The  user can estimate the  impact of  non-uniform soils  by
      comparing results  for several  simulations covering the range  of soil
      properties  present at the  site.
                                                *
   A.  The  flux of water  is  uniform throughout the treatment  site and
      throughout  time. This assumption  will rarely  be met  in ,the field.

   5.  Hydrodynamic dispersion is insignificant and  can  be  neglected. This
      assumption  gives rise to sharp leading and  trailing  edges in  the
      pollutant slug.  These sharp fronts will not exist in soils. As a  result,

-------
          Land Treatment Site
    Rainfall
                            Pollutant
                           Vapor Losses
               Evaporation
      Sludge Applied to Plow Zone
            Volatilization

             Degradation

               Leaching
                                      - Soil Surface
                                      - Plow Zone Depth
                                      - Treatment Zone Depth
              Pollutant
           Leaching Losses
  Figure 1.  Diagram of land treatment site.
    the pollutant will likely reach any depth in the treatment zone before
    the time predicted and it will remain at that depth longer than
    predicted by the model.

6.  Linear isotherms describe the partitioning of the pollutant between the
    liquid, soil, vapor,  and oil phases.  Local equilibrium between phases is
    assumed.

7.  First order degradation of the pollutant and oil are assumed.
    Degradation constants do not change with soil depth or time. This
    assumption ignores changes in biological activity with soil depth.   It
  •  also ignores the influence of loading rate, temperature,  and the quality
    of the environment for microorganisms upon the degradation rate.

8.  The pollutant partitions between the soil, oil, water, and soil vapor
    and does not partition to the remaining fractions of the sludge.

9.  The sludge does not measurably change the properties of the soil water
    or the soil so the pore liquid behaves as water.

-------
 10.   The water content of the soil is related to the hydraulic conductivity
      as described by Clapp and Hornberger (1978). That is,

              k/ks = (e/es)2b+3

      where k is the hydraulic conductivity at a volumetric  water content of
      8, ks is the saturated hydraulic conductivity or the conductivity of the
      soil at the saturated water content, 9S, and b is the  Clapp and
      Hornberger constant for the soil.

Field validation of the model is in progress. The user is cautioned to
consider the assumptions in the model and to apply the model only where
appropriate. The writers are aware the assumptions are only  simplistic
approximations to the continuum of nature. Many of the assumptions were made
to either simplify the mathematical solution or because there was insufficient
experimental data to permit more realistic descriptions of the system.

The model presents results for the specific parameters entered without any
measure of uncertainty in the calculated values. The user is encouraged to
compare results for a series of simulations using parameters in the expected
ranges for the site to obtain an estimate of this uncertainty. For example, if
the site contains two soil layers, the user may want to run  the simulation
twice, once for the soil properties of each layer.

-------
                              SOFTWARE OVERVIEW

The software can be divided functionally into the following three parts.

1.  Parameter Entry
                   This part of the program  enables the user to define  the
                   land treatment system to  be modeled. This includes
                   specification of the soil parameters, properties of  the
                   pollutant and oil, and environmental and management
                   parameters. These user inputs can be made by means of a
                   data entry editor which allows the user to move the  cursor
                   around  the screen to enter or modify parameters in any
                   sequence. These inputs may be saved in disk files for use
                   at a later time. The values entered are verified as  much
                   as possible as they are entered. When the user has
                   finished entering the parameters, a final check is made to
                   determine if the set of parameters is consistent as  a
                   whole.

2.  Output Selection
                   This part of the program  enables the user to specify the
                   desired graphs and tables. The user may also specify the
                   desired output device. Tabular outputs from the model may
                   be directed to the screen, printer, or a text file.  These
                   entries are also made by  means of the data entry editor.

                   Graphical outputs available in this software include the
                   following:

                     1.  A circle graph of mass balance indicating the
                         portions of the pollutant leached, volatilized, and
                         degraded.

                     2.  A line graph of the pollutant volatilized as a
                         function of time.

                     3.  A line graph of the pollutant leached below the
                         treatment zone as a function of time.

                     A.  A line graph of the position of the top and bottom
                         of the pollutant as a function of time.

                     5.  A line graph of the concentration of pollutant as a
                         function of time for selected depths.

                     6.  A line graph of the concentration of pollutant as a
                         function of depth at selected times.

                     7.  Bar graphs of the concentration of pollutant in
                         different phases as functions of time and depth.

-------
                   Tabular outputs include

                     1.  Input soil, pollutant, oil, environmental, and
                         operational parameters.

                     2.  Calculated parameters relating to the treatment
                         system.

                     3.  The amount of pollutant volatilized, leached, and
                         degraded and the computational mass balance error.

                     A.  The quantity of pollutant volatilized as a function
                         of time.

                     5.  The quantity of pollutant leached below the
                         treatment zone as a function of time.

                     6.  The position of the top and bottom of the pollutant
                         as a function of time.

                     7.  The concentration of pollutant in different phases
                         as a function of time at selected depths.

                     8.  The concentration of pollutant in different phases
                         as a function of depth at selected times.

3. Computations/Display
                   This  part of  the software performs the specified
                   calculation and displays the desired results.

When the software is executed,  an introductory screen is displayed followed by
the parameter entry screens. When the user has finished entering the
parameters and the entries are verified, the output selection screen is
selected. When the desired outputs have been specified, the computations and
outputs are displayed. When all of the outputs have been displayed the system
returns to the output selection screen. This provides the user with the
opportunity to obtain additional outputs for the same input parameters. If no
additional outputs are desired, the user may return to the parameter entry
screen by pressing the  key. Each time the user returns to the data entry
editor, the values selected most recently are displayed. Thus, only the
parameters to be changed need to be entered. Thus if a series of pollutants
are to be simulated for one treatment site, the soil, environmental, and
management parameters need to be entered only once. Repeated simulations can
be made easily by simply modifying the properties of the pollutant.

Illustrations of the operation of the software follow a description of the
operating conventions.

-------
                      HARDWARE AND SOFTWARE REQUIREMENTS

This software requires an IBM^ PC, XT, AT, or a compatible computer with at
least 256K bytes of random access memory, one floppy disk drive, and an 8087
or 80287 math coprocessor. An IBM color/graphics board and a compatible
monitor are required to fully utilize the software with graphics. A monochrome
card and monitor can be used for tabular output only. A printer is useful but
not essential. If the printer is compatible with the IBM graphics printer,
copies of the graphics may be printed.

The operating system must be PC-DOS or MS-DOS version 2.0 or later. The
GRAPHICS.COM file from your DOS diskette must be executed before executing
this software to obtain copies of the graphics on the printer.
                             OPERATING CONVENTIONS

The following conventions are used throughout this software.

  1.  Program Interruption; The user may interrupt the program at any time the
      system is asking for an input by pressing the  key. Control in the
      program reverts to the previous data entry screen. If the  key is
      pressed from within the parameter entry option, the program is
      terminated and control is returned to the disk operating system.

  2.  Special Keys; Cursor control keys and function keys are used in the data
      entry editor. The keys and their functions are listed below.

              This key is used to move up one line in the editor. If
                        the cursor is already on the first entry on the
                        screen, the cursor moves to the last entry on the
                        screen.

            This key is used to move down one line in the editor.
                        If'the cursor is already in the last entry on the
                        screen, the cursor moves to the first entry.

           This key is used to move the cursor one character to
                        the right. If the cursor is at the right end of the
                        entry on the line, this key does nothing.

            This key is used to move the cursor one character to
                        the left. If the cursor is under the left character in
                        the entry, this key does nothing.

                   This key moves the cursor from its present position to
                        the beginning of the last entry on the screen.
2. IBM is a registered trademark of International Business Machines, Inc.

-------









This key moves the cursor from its present position to
the beginning of the first entry on the screen.

The parameter entry process requires three screens.
One screen is for soil properties, one for pollutant
and oil properties, and one for environmental and
operational parameters. In this case, the  key
moves to the next screen in thee series. For example,
if the screen for soil properties is displayed,
pressing this key will display the pollutant and oil
properties screen. Pressing it again will display the
operational and environmental screen.

This key is used to move to the previous screen when
entering the land treatment site parameters.

This key is used to obtain brief help messages
relating to the parameter being entered.

This key is used to enter and calculate a weighted
average value of a soil parameter from values for
different soil depths. See the Appendix for details.

If the parameters entered into this model at one time
have been saved in a file, those values can be input
to the system from the file. The  key enables the
user to specify the name of the input file. If the
file exists, its values are input and displayed by the
editor. If the file is not found, the values in the
editor remain unchanged. The user may view the
directory of a disk by pressing  when the file
name is requested.

Parameters entered into the system can be saved in
disk files for use at another time. Pressing the 
key enables the user to specify the name of an output
file. After the file is specified, the present soil,
chemical and environmental parameters are written to
disk. Pressing  when the output file is requested
enables the user to view a disk directory.

This key is used to terminate data entry on a
particular screen and to proceed to the next phase of
the program.

This key is used to interrupt the present process and
to return to the previous data entry screen.

This key is used to terminate entry of a particular
parameter. Any characters to the right of the  cursor
are truncated when the  key is pressed.

-------
       This key is used to delete one character to the left
                  of the cursor.  If the cursor is at the beginning of
                  the entry,  nothing is deleted.

          This key is used to delete the character at the
                  present cursor location.

File Names; File names may be any legal MS-DOS file name. File
extensions may be used to facilitate organization of files.
                           *
Unknown Parameters; When entering parameters into the editor, the user
may signify that a value is unknown by entering only a period or decimal
point. Entering a period for an input parameter defining the land
treatment site will result in further prompting for the entry. In many
cases, the additional prompt will provide additional information about
the required parameter. It may also provide a method of estimating the
parameter from other parameters which may be known.

Specifying No Data; When tables or graphs of concentration as functions
of time or depth are selected as outputs, the user has opportunity to
specify 15 times or depths of interest. If fewer times or depths are
desired, "no data1 can be specified for the remaining entries. No data
is specified by entering a period or decimal point instead of a number.

Copy Graphics On Printer; When graphs are displayed on the screen, they
can be printed on an IBM graphics printer or a compatible machine by
pressing the 

key or the and keys. The

key results in smaller copies of the screen. The GRAPHICS.COM must be executed before RITZ if copies of graphs will be made.


-------
                                GETTING STARTED

Making a Working Copy; The software is distributed on a single diskette. The
first step is to make a working copy of the software. The original copy should
then be placed in a safe place. The following steps can be followed to make a
working copy.
Fixed Disk Systems
                      1.   Make a new directory for the RITZ model using  the
                          MKDIR command of DOS.  For example:
                          MKDIR \RITZ 

                      2.   Copy the contents of the distribution diskettevto
                          the new directory using the COPY command of  DOS. If
                          the distribution diskette is in disk drive A,  enter
                          the following command:
                          COPY A;*.* \RITZ /V 
Floppy Disk Systems
                      1.   FORMAT a new floppy diskette with the /S option.  To
                          do this place your DOS diskette in drive A and a  new
                          diskette in drive B.  Then enter
                          FORMAT B;/S 

                      2.   If you have a color/graphics card, copy the
                          GRAPHICS.COM program from the DOS diskette to the
                          working diskette using the COPY command. To do this
                          enter
                          COPY A;GRAPHICS.COM B; /V 

                      3.   Copy the contents of the distribution diskette to
                          the new diskette using the COPY command. This can be
                          done by removing the DOS diskette in drive A and
                          replacing it with the distribution diskette and
                          entering the command
                          COPY A:*.* B; /V 

Details  on  the use  of  the COPY,  FORMAT,  and MKDIR commands are given in your
DOS manual.

The software  is  distributed to run on a  system with a color graphics card.  If
your  computer has this card,  your working copy is now complete.  If your
computer does not have this card, you will need to execute the configuration
program  included on the  diskette. To configure the software for a monochrome
system

   1.   In a  floppy disk system,  place the working diskette in the default disk
       drive.  In  a fixed  disk system,  use the CD command to make the directory
       containing the RITZ software the default directory.

-------
  2.  Execute program CONFIG by entering

             CONFIG

      The program will ask you to specify the type of monitor. Specify the
      monitor matching that in your system. The program will modify the RITZ
      software for your system. The software on the working diskette should
      then be ready for use.

Executing the Program; To execute the program,
        ~ 	     '• '                 «
  1.  Place the floppy diskette in the default disk drive (or define the
      directory containing the software to be the default directory).

  2.  Enter GRAPHICS to install the memory resident software for
      printing graphics screens.

  3.  Enter RITZ to execute the model.

You may find it more convenient to make a batch file to execute steps 2 and 3
as one command. This file would contain the following lines:
       GRAPHICS
       RITZ
                                      10

-------
                           EXAMPLE OF SOFTWARE USE
                                   RITZ
              REGULATORY AND INVESTIGATIVE TREATMENT ZONE MODEL
   This software is designed to estimate the movement and fate of chemicals
   applied as oily wastes in land treatment sites.  The user is required to
   enter the properties of the chemicals and oil in the waste material, the
   soil properties of the treatment site, the management practices used,
   and the relevant parameters describing the environment at the site.   v
   Outputs of the model include the quantity of the pollutant degraded,
   leached, and volatilized, the concentration of pollutant in the different
   phases at different times and depths, and the quantity of pollutant
   volatilized and leached as a function of time. Outputs may be displayed
   in graphical and tabular forms.
   This software was developed by D.L. Nofziger and J.R. Williams, Oklahoma
   State University, Stillwater, Oklahoma.  The software is based on a
   mathematical model of the treatment zone developed by Thomas E. Short,
   R.S. Kerr Environmental Research Laboratory, Ada, Oklahoma.

   Press any key to continue:
Screen 1. Purpose of the program.
Introduction: The first screen which appears when the software is run is
displayed in Screen 1.  This gives the user the purpose of the software and the
individuals responsible for it.

Parameter Entry; This part of the software enables the user to define the
properties of the treatment site and chemicals. The data entry editor is used
for this purpose. (See OPERATING CONVENTIONS for details on the use of the
editor.) Three screens are used for these inputs. The  and  keys
can be used to move from one screen to another. Values shown in this manual
are for illustration only.

Screen 2 is the screen used for defining the soil properties of the treatment
site. Since the model assumes the soil at the treatment site is uniform with
depth, only one value is entered for each property. If the soil is not uniform
with depth, the user may obtain an average from known values at different
depths by pressing the  key. The averaging scheme used is described in the
Appendix. If the site is not uniform from one position to another, the user
may obtain a spatial average for use in this model or the model may be run
several times for different smaller sites. The parameters to be entered on
this screen are

  1.  Identification.code: This is simply a string of characters which serve
      to identify this set of data for the user's reference. It is displayed
      with outputs from the program.
                                      11

-------
     Identification code
     Soil name
     Fraction organic carbon
     Bulk density,  kg/m3
     Saturated water content,  m3/m3
     Sat. hydraulic conductivity,  m/day
     Clapp and Hornberger constant
       Soil Properties
                     Site //I
                     Tipton Sandy Loam
                     0.0050
                     1500
                     0.410
                     5.0000E-001
                     4.9000
              
              
              
             
             
    , 
Display help for entries
Average across depths
Input parameters from data file
Save parameters in output data file
Proceed - all entries made
Abort program
Edit other entry screens
Screen 2. Data entry screen for soil properties.
  2.  Soil name: This again serves to identify the soil at the treatment site.


  3.  Fraction organic carbon (foc): This is the ratio of the mass of organic
      carbon in the soil to the mass of soil solids.


  4.  Bulk density (p): This is the ratio of the mass of soil solids to the

      total volume of the soil. That is, it is the ratio of the mass of solids

      to the volume of solids,  liquids, and gases in the soil.


  5.  Saturated water content (8S): This is the ratio of the volume of water
      in"the soil to the total volume of the soil when the soil pores are

      filled with water.


  6.  Saturated hydraulic conductivity (ks): This is the hydraulic

      conductivity of the soil when all of the soil pores are filled with

      water. It is the constant of proportionality between the flux density

      and the gradient in potential in Darcy's law.


  7.  Clapp and Hornberger constant (b): This is the constant in the equation

      of Clapp and Hornberger (1978) relating the relative saturation of the

      soil to the relative conductivity of the soil.  That is


             e/es = (k/ks)2b+3


      where k is the .hydraulic conductivity of the soil at a volumetric water

      content 8 and ks is the saturated hydraulic conductivity at the

      saturated water content,  9g.
                                      12

-------
                     Oil and Pollutant Properties
     Name of pollutant in sludge
     CAS number
     Concentration of pollutant in sludge, g/kg
     Organic carbon partition coefficient, m3/kg
     Oil-water partition coefficient
     Henry's law constant
     Diffusion coef. of pollutant in air, m2/day
     Half-life of pollutant, days
     Concentration of oil in sludge, g/kg
     Density of oil, kg/m3
     Half-life of oil, days
                                             Pollutant^l
                                             123-4567
                                              .OOOOE+000
                                              .2000E-002
                                              .OOOOE+001
                                              .5000E-005
                                              .3000E-001
                                              .OOOOE+001
                                              .5000E+002
                                              .OOOOE+003
                                             4.5000E+001
              
              
             
             
    , 
               Display help for entries
               Input parameters from data file
               Save parameters in output data file
               Proceed - all entries made
               Abort program
               Edit other entry screens
Screen 3. Data entry screen for pollutant and oil properties.
Screen 3 is used to enter the properties of the pollutant and the oil in the

waste material. These entries are described below.
  1.  Name of the pollutant in sludge: This is the name of the pollutant whose

      properties are entered below. This name is for identifying output tables

      and graphs.


  2.  CAS number: This unique Chemical Abstracts Number may be entered to

      provide positive identification for the pollutant being modeled. This

      number is also displayed with the outputs.


  3.  Concentration of pollutant in sludge (Sp): This is the concentration of
      the pollutant in the sludge when it was applied to the soil.


  4.  Organic carbon-water partition coefficient (KQQ): This is the partition

      coefficient between the pollutant in soil and water normalized to the

      soil's organic carbon content. That is


             CS = KOCfOCcW

      where Cg and C^ are the concentrations of pollutant in the soil and
water, respectively, and
                                   is the fraction organic carbon in the soil.
                                       13

-------
  5.   Oil-water partition coefficient (KQ):  This is the partition coefficient
      for the pollutant between the oil and  water phases.
      That is

             C0 = KoCW
      where CQ and C^ are the concentrations of the pollutant in the oil and
      water phases, respectively,  and KQ is  the oil-water partition
      coefficient.

  6.   Henry's law constant (Kg): This is the constant for partitioning the
      pollutant between the vapor and water  phases. That is

             Cv = KHCW

      where Cy and C^ are the concentrations of the pollutant in the vapor and
      water phases, respectively.
  7.  Diffusion coef.  of pollutant in air (D^):  The diffusion coefficient of
      the pollutant in air is used to determine pollutant losses in the vapor
      phase.

  8.  Half-life of the pollutant (tip): This is the time required for one-half
      of the original amount of pollutant to be transformed to some other
      product.  It is based on the assumption that the transformation follows
      first-order or pseudo first-order kinetics.

  9.  Concentration of oil in the sludge (SQ): This is the concentration of
      oil in the sludge at the time of application.

 10.  Density of oil (po): This is the density of the oil in the sludge. That
      is, it is the mass of oil per unit volume of oil.

 11.  Half -life of oil (tio): This is the time required for one-half of the
      original amount of oil in the sludge to be biologically degraded. It is
      based on the assumption that the transformation of the oil in the sludge
      will follow first-order kinetics.

Screen A is used to enter or edit data relating to the operation of the
treatment site and the environment at the site.  The parameters needed include
the following:

  1.  Sludge application rate (SAR): This is the mass of sludge or waste
      material -applied per hectare of land area.

  2.  Plow zone depth (pzd): This is the depth to which the sludge or waste
      material is incorporated. See Figure 1 for more information.

  3.  Treatment zone depth (tzd): This is the depth of the bottom of the soil
      considered to be part of the treatment zone. Chemical movement below
      this depth is lost from the system and is considered as leached.

  A.  Recharge rate (V
-------
                Operational and Environmental Factors
     Sludge application rate,  kg/ha
     Plow zone depth,  m
     Treatment zone depth,  m
     Recharge rate, m/day
     Evaporation rate, m/day
     Temperature, degrees C
     Relative humidity
     Diffusion coef. of water vapor in air, m2/day
                                1.5000E+005
                                0.150
                                1.500
                                0.0060
                                0.0025
                                25.0
                                0.500
                                2.0000E+000
              
              
             
             
    , 
Display help for entries
Input parameters from data file
Save parameters in output data file
Proceed - all entries made
Abort program
Edit other entry screens
Screen 4. Data entry screen for operational and environmental factors.



  5.  Evaporation rate (E): This is the average flux density of water

      evaporating from the soil.


  6.  Air temperature (T): This is the average air temperature at the site.


  7.  Relative humidity (RH): This is the average relative humidity at the

      site expressed as a fraction (rather than a percent).


  8.  Diffusion coef. of water vapor in air (Dw): This diffusion coefficient

      of water vapor in air is used to estimate the vapor losses of the

      pollutant.
The keyboard will be the primary method of entering parameters into the model.

However, the software enables the user to save manually entered values in data

files for use at a later time. This is done from within the data entry editor

by means of the  and  function keys as explained in the section on

OPERATING CONVENTIONS. When saving data, the system will request the name of

the output file from the user. It will then write the current values of the

input parameters in that file. When reading parameters from a file, the system

will prompt the user for the name of the input file. The data will then be

read and the editing screens updated to those values. When naming input and

output files, the user is advised to develop a system of names and extensions

which will facilitate identification of the file contents. When a file name is

requested, the user may press the  key to view a directory of files.
                                      15

-------
                             Output Options
    Graphs:
      Mass balance
      Pollutant volatilized vs. time
      Pollutant leached vs. time
      Position of pollutant vs. time
      Concentration vs. time at selected depths
      Concentration vs. depth at selected times
      Concentration bar graphs
    Tables:
      Input parameters             %
      Calculated parameters
      Mass balance
      Pollutant volatilized vs. time
      Pollutant leached vs. time
      Position of pollutant vs. time
      Concentration vs. time at selected depths
      Concentration vs. depth at selected times
      Output device for tables
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y

                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  Y
                                  SCREEN
        
        
Display help for entries
Proceed - all entries made
Abort option and return to parameter entry screen
Screen 5. Screen for selection of desired outputs from model.
Output Selection: This portion of the software enables the user to select the

types of outputs desired and the desired output device. This selection process
begins with Screen 5. If any concentration outputs are selected, one or two

additional screens are required to select the depths and times of interest.

The use of the three screens are illustrated in this section.

Screen 5 illustrates the selection of outputs from the model. For each option,
the user enters Y if that option is desired or N if it is not desired. In this

example, all the entries are Y to generate all the possible types of output.

The entries on Screen 5 are as follows:

  1.  Graphs:


        a.  Mass balance: This option displays a pie chart of the relative
            amount of the pollutant degraded, leached, and volatilized.


        b.  Pollutant volatilized vs. time: This option displays a graph of
            the flux density of pollutant removed from the treatment site in

            the vapor phase as a function of time.


        c.  Pollutant leached vs. time: This option displays a graph of the

            flux density of pollutant leached from the treatment zone as a
            function of time.


        d.  Position of pollutant vs. time: This option displays a graph of
            the location of the top and bottom of the pollutant as a function

            of time.
                                      16

-------
      e.   Concentration vs.  time at selected depths:  This option displays  a
          graph of the concentration of pollutant as  a function of time at
          one or more depths selected by the user using Screen 6.  Graphs of
          the total concentration of pollutant and concentrations in water,
          soil, vapor, and oil phases are displayed sequentially.  For each
          phase, the software displays a depth and draws the line for that
          depth. It then waits 'for the user to press  a key.  If that key is
          not , 

, or the system will display the line for the next depth selected. If is pressed, the remaining depths for this phase are not drawn and the system proceeds to draw the graph for the next phase. If

is pressed, the screen is printed on the printer. If is pressed, the system returns to Screen 5. f. Concentration vs. depth at selected times: This option displays a graph of the concentration of pollutant as a function of depth for one or more times selected by the user using Screen 7. This option operates in the same manner as the concentration vs. time graphs described above. g. Concentration bar graphs: This option presents a series of bars representing the treatment zone. Within each bar the concentration of pollutant in one phase at a particular time is displayed qualitatively using one of six patterns. The bars are redrawn for different times selected by the user (Screen 7). In this way the user can see the change in depth and concentration of the pollutant with changes in time. Different bars on the screen represent the total concentration of pollutant, pollutant concentration in water, soil, vapor, and oil, and the oil content. 2. Tables: a. Input parameters: This table displays the parameters entered by the user to define the current scenario. b. Calculated parameters: This table contains selected chemical and physical parameters calculated from the input parameters. c. Mass balance: This table lists the absolute and relative amounts of pollutant degraded, volatilized, and leached along with the mass balance error. d. Pollutant volatilized vs. time: This is a table of the flux density of pollutant leaving the treatment site in the vapor phase as a function of time. e. Pollutant leached vs. time: This is a table of the flux density of pollutant leached from the treatment zone as a function of time. f. Position of pollutant vs. time: This table displays the location of the top and bottom of the pollutant slug at different times. 17


-------



















Dept s of I terest
Depth 1
Depth 2
Depth 3
Depth 4
Depth 5
Depth 6
Depth 7
Depth 8
Depth 9
Depth 10
Depth 11
Depth 12
Depth 13
Depth 14
Depth 15
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
meters
0.00
0.05
0.10
0.15
0.25
0.50
0.75
1.00
1.25
1.50





Display help for entries
Proceed - all entries made
Abort option and return to output option screen
Screen 6. Selection of depths of interest for concentration tables and graphs.
        g.  Concentration vs. time at selected depths: This table contains the
            total pollutant concentration, the concentration of pollutant in
            water, soil, vapor, and oil, and the oil content at user selected
            times and depths. These tables are structured so that
            concentrations for all times at one depth occur on one page.

        h.  Concentration vs. depth at selected times: This table is similar
            to that described above. It differs in that the output is
            organized so that concentrations for all depths and one time occur
            on one page.

        i.  Output device for tables: Tabular output can be displayed on the
            screen or printer. It may also be saved in disk files for later
            use. This line enable the user to specify the desired device.
            Entries in this line may be SCREEN, PRINTER, or a legal file name.

If one or more of the concentration options is desired the depths or times
desired are entered using Screens 6 and 7, respectively. In each case the user
enters the depths or times of interest. A period indicates 'no data" or no
value.
Outputs of Model; The following pages illustrate the outputs from the RITZ
model for the inputs shown in screens 2, 3, and 4 and the outputs selected in
screens 5, 6, and 7. Graphical outputs were generated by pressing the 

key. 18


-------



















Times of I terest
Time 1
Time 2
Time 3
Time A
Time 5
Time 6
Time 7
Time 8
Time 9
Time 10
Time 11
Time 12
Time 13
Time 14
Time 15
days
days
days
days
days
days
days
days
days
days
days
days
days
days
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00







: Display help for entries
: Proceed - all entries made
: Abort option and return to output option screen
Screen 7. Selection of times of interest for concentration table and graphs,
                                      19

-------
Table 1.   Table of input parameters describing land treatment site.
                          INPUT DATA - SOIL PROPERTIES

              Fraction organic carbon                  :  0.0050
              Bulk density (kg/m3)                     :  1500.0
              Saturated water content (m3/m3)          :  0.4100
              Saturated hydraulic conductivity (m/day) :  5.0000E-001
              Clapp and Hornberger constant            :  4.9000
                   INPUT DATA - OIL AND POLLUTANT PROPERTIES

        Concentration of pollutant in the sludge (g/kg)
        Organic carbon partition constant (m3/kg)
        Oil-water partition coefficient
        Henry's law constant
        Diffusion coefficient of pollutant in air (m2/day)
        Half-life of the pollutant (day)
        Concentration of oil in the sludge (g/kg)
        Density of the oil (kg/m3)
        Half-life of the oil (day)
        Diffusion coefficient of water vapor in air (m2/day)
          1.
          2.
          5.
          5,
          4.
          3,
          2.
          1.
          4,
OOOOE+000
2000E-002
OOOOE+001
5000E-005
3000E-001
OOOOE+001
5000E+002
OOOOE+003
5000E+001
          2.OOOOE+000
               INPUT DATA - OPERATIONAL AND ENVIRONMENTAL FACTORS
                  Sludge application rate (kg/ha)
                  Plow zone depth (m)
                  Treatment zone depth (m)
                  Recharge rate (m/day)
                  Evaporation Rate (m/day)
                  Air Temperature (deg C)
                  Relative humidity
1.5000E+005
0.150
1.500
0.0060
0.0025
25.0
0.500
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
                 RITZ
                                      20

-------
Table 2.   Table of calculated parameters for site described in Table 1.
                             CALCULATED PARAMETERS

        Ratio of the density of water vapor to liquid
        Boundary layer thickness (m)
        Soil-water partition coefficient (m3/kg)
        Degradation rate constant of pollutant (I/day)
        Degradation rate constant of oil (I/day)
        Water content of soil (m3/m3)
        Soil pore water velocity (m/day)
        Initial oil content in the plow zone
        Initial pollutant content in the plow zone (g/m3)
        Air content of the soil (m3/m3)
        Effective diffusion coefficient of
                          pollutant vapor in soil (m2/day)
        Initial pollutant loading (g/m2)
2.3E-005
4.6E-003
 .1E-004
 .3E-002
1.5E-002
2.9E-001
 .1E-002
 .5E-002
 , OE+002
1.
2.
2.
2.
1.
9.5E-002

9.9E-OOA
1.5E+001
                       BASIC INFORMATION ABOUT THE SYSTEM

  Maximum residence of the pollutant in the plow zone (days)      : 35
  Maximum residence of the pollutant in the treatment zone (days) : 138
  Treatment zone breakthrough time (days)                         : 102
  Retardation factor in the lower treatment zone                  : 1.6E+000
  Velocity of the pollutant in the lower treatment zone (m/day)   : 1.3E-002
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number:  123-4567
       RITZ
                                       21

-------
Table 3.   Mass balance table summarizing the amount of pollutant  degraded,
           volatilized, and leached as well as the computational error.
  Amount loaded

  Amount degraded
  Amount volatilized
  Amount leached
  Computational error
  MASS BALANCE

Mass of Pollutant
       g/m2
     1.5E+S01

     l.AE+001
     7.7E-003
     9.4E-001
     4.8E-009
                                             Relative Amount
l.OE+002
9.
5.
6.
  AE+001
  1E-002
  3E+000
3.2E-008
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
                                               RITZ
                    MASS  BALANCE
                                          Le AC lie dl
                                          Vo 1 At i 1 i
                                 i on-a.1  Eif»if*oif»:   O . OOOOK
Figure 2.   Mass balance graph summarizing  information  in  Table  3.
                                      22

-------
Table 4.   The flux density  of  pollutant lost to the atmosphere as a function
           of time.

Time Flux
days g/m2-day
0.00 3.0E-001
1.55 5.7E-004
3.09 2.8E-004
4.60 1.8E-004
6.09 1.3E-004
7.55 1.1E-004
8.99 8.6E-005
10.41 7.3E-005
11.80 6.2E-005
13.18 5.5E-005
14.54 4.8E-005
15.87 4.3E-005
17.19 3.9E-005
18.49 3.5E-005
Identification Code:
VAPOR FLUX
Time
days
19.78
21.04
22.29
23.53
24.75
25.95
27.14
28.32
29.48
30.63
31.77
32.89
34.01
35.11
Site //I
VERSUS TIME
Flux
g/m2-day
3.2E-005
3.0E-005
2.7E-005
2.5E-005
2.4E-005
2.2E-005
2.1E-005
1.9E-005
1.8E-005
1.7E-005
1.6E-005
1.5E-005
1.4E-005
1.4E-005

Soil Name: Tipton Sandy Loam
Compound Name: Pollutant//!
CAS Number: 123-4567
Uapor

g l.OE-002
a
I \
p i.OE-004 N^
F ^-^
1 ^^v
u
X l.OE-006
t . OF-008 	 1 	
Flux (g/»2-da«)


-^-^
versus Tine


^— -_

Time Flux
days g/m2-day
42.42 7.0E-006
49.74 4.3E-006
57.05 2.8E-006
64.37 1.9E-006
71.69 -1.4E-006
79.00 l.OE-006
86.32 7.6E-007
93.63 5.8E-007
100.95 4.4E-007
108.27 3.4E-007
115.58 2.6E-007
122.90 2.1E-007
130.21 1.6E-007
137.53 1.3E-007

RITZ
Identification:
Sitettl
Soil Name:
Tipton Sandy Loan
Pollutant Nam:
Pollutant*!
CAS Number:
123-4567

             0     25     50     75    100     125    150
                            Tim (days)
 Figure 3.    Graph of flux density of pollutant  in  vapor phase as a function of
             time.
                                       23

-------
Table 5.   The  flux, density of pollutant leached below the  treatment zone as a
           function of time.
LEACHATE FLUX VERSUS TIME
Time Flux
days g/m2-day
102.42 3.3E-002
103.28 3.3E-002
104.14 3.2E-002
104.99 3.2E-002
105.85 3.2E-002
106.71 3.1E-002
107.56 3.1E-002
108.42 3.1E-002
109.27 3.0E-002
110.13 3.0E-002
110.99 3.0E-002
111.84 2.9E-002
112.70 2.9E-002
113.56 2.9E-002
Identification Code: Site
Time
days
114v41
115.27
116.12
116.98
117.84
118.69
119.55
120.41
121.26
122.12
122.97
123.83
124.69
125.54
//I
Flux
g/m2-day
2.9E-002
2.8E-002
2.8E-002
2.8E-002
2.7E-002
2.7E-002
2.7E-002
2.6E-002
2.6E-002
2.6E-002
2.6E-002
2.5E-002
2.5E-002
2.5E-002

Time
days
126.40
127.26
128.11
128.97
129.82
130.68
131.54
132.39
133.25
134.11
134.96
135.82
136.67
137.53

Flux
g/m2-day
2.5E-002
2.4E-002
2.4E-002
2.4E-002
2.3E-002
2.3E-002
2.3E-002
2.3E-002
2.2E-002
2.2E-002
2.2E-002
2.2E-002
2.1E-002
2.1E-002

Soil Name: Tipton Sandy Loam
Compound Name: Pollutant//!
CAS Number: 123-4567







RITZ
                  Leachate Flux
                           versus Tine
J. . V£T WX

L
e l.OE+000
a
c
h
a
t l.OE-001
e
F
1
« l.OE-002
X
i.OE-003

f











— • 	 , 	
— 	 b_


— — I 	 1 	 1 	 1— 	 1 	 P 	 1 —




Identification:
Sitettl
Soil Nam:
Tipton Sandy Loan

Pollutant Nam:
Pallutantttl

CAS NuNl*i>:
123-4567





100   105   110   115   120   125
                Tim (days)
                                          130   135   140
Figure 4.
 Graph of flux density of pollutant leached below the treatment
 zone as a function  of time.
                                        24

-------
Table 6.    The location of  the top and bottom of the pollutant  as  a function

            of time.
                 DEPTH OF  BOTTOM AND TOP OF  POLLUTANT VERSUS  TIME
Time
days
0.00
4.15
8.13
11.94
15.61
19.14
22.54
25.83
29.02
32.11
35.11
Top
m
0.00
0.01
0.03
0.04
0.06
0.07
0.09
0.10
0.12
0.14
0.15
Bottom
m
0.15
0.20
0.26
0.31
0.36
0.40
0.45
0.49
0.53
0.57
0.61
Time
days
35.11
45.35
55.59
65.83
76.08
86.32
96.56
106.80
117.05
127.29
137.53
Top
m
0.15
0.29
0.42
0.56
0.69
0.83
0.96
1.09
1.23
1..36
1.50
Bottom
m
0.61
0.75
0.88
1.02
I.'IS
1.29
1.42
> 1.50
> 1.50
> 1.50
> 1.50
  Identification Code:  Site //I

  Soil Name: Tipton  Sandy Loam

  Compound Name: Pollutant//!

  CAS Number: 123-4567
                                               RITZ
                        Slag Position versus Tine
          1.50
                    25
50     75     100
   Tim (days)
                                                 125
                                                           Identification:
                                                            Sitettl

                                                           Soil Nam:
                                                            Tipton Sands Loan

                                                           Pollutant Nam:
                                                            Pollutant*!

                                                           CflS Hunker:
                                                            123-4567
150
 Figure 5,   Location of the top  and bottom of  the pollutant as  a  function of

             time.
                                         25

-------
Table 7.   Concentration of pollutant in different phases and oil  content  as  a
           function of time at selected depths.
Depth = 0
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
.000 meters
Total
Pollutant
g/m3
l.OE+002
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
5.9E+001
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
»
Soil
g/kg
6.5E-003
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000

Vapor
g/m3
3.2E-003
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
0 . OE+000

Oil
g/m3
2.9E+003
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
0 . OE+000
Oil
Content
m3/m3
2.5E-002
2.1E-002
1.8E-002
1.6E-002
1.4E-002
1.2E-002
7.9E-003
5.4E-003
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Depth - 0.050 meters
                             CONCENTRATION PROFILE
                                     Pollutant in
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
lotaj.
Pollutant
g/m3
1 . OE+002
7.9E+001
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
Water
g/m3
5.9E+001
5.2E+001
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
Soil
g/kg
6.5E-003
5.7E-003
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
Vapor
g/m3
3.2E-003
2.9E-003
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
Oil
g/m3
2.9E+003
2.6E+003
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
U1J.
Content
m3/m3
2.5E-002
2.1E-002
1.8E-002
1.6E-002
1.4E-002
1.2E-002
7.9E-003
5.4E-003
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                      26

-------
Table 7.   Continued.
Depth = 0
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
.100 meters
Total
Pollutant
g/m3
1 . OE+002
7.9E+001
6.3E+001
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
5.9E+001
5.2E+001
4.6E+001
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0. OE+000

Soil
g/kg
6.5E-003
5.7E-003
5.0E-003
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000

Vapor
g/m3
3.2E-003
2.9E-003
2.5E-003
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000

Oil
g/m3
2.9E+003
2.6E+003
2.3E+003
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
Oil
Content
m3/m3
2.5E-002
2.1E-002
1.8E-002
1.6E-002
1.4E-002
1.2E-002
7.9E-003
5.4E-003
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Depth = 0.150 meters
                             CONCENTRATION PROFILE
                                     Pollutant  in
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
lotaj.
Pollutant
g/m3
1 . OE+002
7.9E+001
6.3E+001
5.0E+001
0 . OE+000
0. OE+000
0 . OE+000
0 . OE+000
Water
g/m3
5.9E+001
5.2E+001
4.6E+001
4.0E+001
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
Soil
g/kg
6.5E-003
5.7E-003
5.0E-003
4.4E-003
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
Vapor
g/m3
3.2E-003
2.9E-003
2.5E-003
2.2E-003
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
Oil
g/m3
2.9E+003
2.6E+003
2.3E+003
2.0E+003
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
Ull
Content
m3/m3
2.5E-002
2.1E-002
1.8E-002
1.6E-002
1.4E-002
1.2E-002
7.9E-003
5.4E-003
  Identification Code:  Site  //I
  Soil Name: Tipton  Sandy  Loam
  Compound Name: Pollutants/1
  CAS Number:  123-4567
RITZ
                                       27

-------
Table 7.   Continued.
Depth = 0
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
.250 meters
Total
Pollutant
g/m3
0 . OE+000
2.2E+001
1.9E+001
1.7E+001
1.5E+001
0. OE+000
0 . OE+000
0. OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
0. OE+000
4.8E+001
4.2E+001
3.7E+001
3.3E+001
0 . OE+000
0 . OE+000
0 . OE+000

Soil
g/kg
0 . OE+000
5.3E-003
4.7E-003
4.1E-003
3.6E-003
0 . OE+000
0. OE+000
0. OE+000

Vapor
g/m3
0 . OE+000
2.6E-003
2.3E-003
2.1E-003
1.8E-003
0. OE+000
0. OE+000
0. OE+000

. Oil
g/m3
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
Oil
Content
m3/m3
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
0. OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Depth = 0.500 meters
                             CONCENTRATION PROFILE
                                     Pollutant  in
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
louaj.
Pollutant
g/m3
0. OE+000
0. OE+000
0 . OE+000
1.4E+001
1.2E+001
1.1E+001
0. OE+000
0. OE+000
Water
g/m3
0 . OE+000
0. OE+000
0. OE+000
3.0E+001
2.7E+001
2.4E+001
0 . OE+000
0. OE+000
Soil
g/kg
0. OE+000
0. OE+000
0 . OE+000
3.4E-003
3.0E-003
2.6E-003
0. OE+000
0. OE+000
Vapor
g/m3
0. OE+000
0. OE+000
0 . OE+000
1.7E-003
1.5E-003
1.3E-003
0. OE+000
0. OE+000
Oil
g/m3
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
0 . OE+000
VJ1J.
Content
m3/m3
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                      28

-------
Table 7.   Continued.
Depth = 0
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
.750 meters
Total
Pollutant
g/m3
0 . OE+000
O.OE+000
0. OE+000
0 . OE+000
0 . OE+000
8.8E+000
6 . AE+000
0 . OE+000
CONCENTRATION' PROFILE
Pollutant in

Water
g/m3
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
1.9E+001
1.4E+001
O.OE+000

Soil
g/kg
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
2.1E-003
1.6E-003
O.OE+000

Vapor
g/m3
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
-1.1E-003
7.8E-004
O.OE+000

Oil
g/m3
O.OE+000 .
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
-O.OE+000
0 . OE+000
Oil
Content
m3/m3
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Depth = 1.000 meters
                             CONCENTRATION PROFILE
                                     Pollutant in
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
lULcU.
Pollutant
g/m3
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
5.3E+000
0 . OE+000
Water
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
1.2E+001
0 . OE+000
Soil
g/kg
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
1.3E-003
O.OE+000
Vapor
g/m3
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
6.4E-004
O.OE+000
Oil
g/m3
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
U1JL
Content
m3/m3
O.OE+000
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant#l
  CAS Number: 123-4567   ,
RITZ
                                      29

-------
Table 7.   Continued.
Depth = 1
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
.250 meters
Total
Pollutant
g/m3
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
3 . 2E+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
7 . OE+000

Soil
g/kg
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0 . OE+000
7.7E-004

Vapor
g/m3
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
3.8E-004

Oil
g/m3
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
0. OE+000
Oil
Content
m3/m3
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
0. OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Depth = 1.500 meters
                             CONCENTRATION PROFILE
                                     Pollutant  in
Time
days
0.00
10.00
20.00
30.00
40.00
50.00
75.00
100.00
lotax
Pollutant
g/m3
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
Water
g/m3
O.OE+'OOO
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0 . OE+000
Soil
g/kg
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
Vapor
g/m3
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
Oil
g/m3
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0 . OE+000
U1J.
Content
m3/m3
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
0 . OE+000
0. OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                      30

-------
                Concentration  of Pollutant (s/w3) us Tine
l.OE+003
C l.OE+001
0
n
c
e l.OE-001
n
t
p
a l.OE-003
t
i
0
" l.OE-005
t AF-AA9
Total


























,
























.











                    25     50     75     100
                              Tine (days)
                                                         Identification:
                                                          Siteftl

                                                         Soil Nawe:
                                                          Tipton Sandy Loam

                                                         Pollutant Nam:
                                                          Pollutant*.!

                                                         CAS Number:
                                                          123-4567
       150
Figure 6.   Concentration of total pollutant as a  function of  time for  depths

             of  0.1, 0.5,  1.0, and 1.5 meters.
                Concentration of Pollutant (g/i*3) vs Tine
A. V1TVVO
C l.OE+001
0
n
c
e l.OE-001
n
t
K»
a l.OE-003
i
0
" l.OE-005
1 OF-007
: 	




















Water












Identif icatian:
Sitettl
Soil Nai«e:
Tipton Sandy Loan
Pollutant Nane:
Pollutant*!
CAS NuMber:
123-4567




                    25      50     75     100
                             Tine (days)
125
150
Figure 7.   Concentration of pollutant in water as  a  function  of time  for

             depths of 0.1,  0.5,  1.0, and 1.5 meters.
                                          31

-------
                Concentration of Pollutant  (g/kg) us Tine
J. , UA-UIU
C l.OE-003
0
n
e l.OE-005
n
t
r
» l.OE-007
i
o
n l.OE-009
] .OF-Oll










., Soil









1 —




























Identification:
SitetU
Soil Nam:
Tipton Sandy Loan
Pollutant Nane:
Pollutanttti
CAS NUM!I«I>:
123-4567





                   25     50     75     100
                             Tine (dags)
125
150
Figure  8.    Concentration of  pollutant  in soil as  a function of time  for

             depths  of 0.1,  0.5,  1.0, and 1.5 meters.
                Concentration of Pollutant (g/*?) us Time

C l.OE-003
o
n
c
e l.OE-005
n
t
p
J l.OE-007
t
i
o
" l.OE-009
1 AF-At 1


























' 	 '-











Uapor

























                    25      50      75     100
                             Tine  (days)
125
                                                         Identification:
                                                          SiteM.

                                                         Soil Nam:
                                                          Tipten Sandy Loan

                                                         Pollutant Nam:
                                                          Pollutant*!
150
Figure 9.   Concentration  of pollutant  in vapor  as a function of time for

             depths of 0.1,  0.5, 1.0,  and 1.5 meters.
                                          32

-------
       l.OE+004
                 Concentration of Pollutant (g/*3) us Tine
     C l.OE+002
     o
     n

     e l.OE+000
     n
     t
     p
     * l.OE-002

     i
     o
     n l.OE-004
       l.OE-006
                                            Oil
                    25     SO     75     100
                              Tine (days)
125
         Identification:
          Sitettl

         Soil  Nam:
          Tipton Sandy Loan

         Pollutant Nam:
          Pollutant»l

         CAS Nunber:
          123-4567
150
Figure  10.    Concentration of pollutant in oil as a function of,time for  depth

              of 0.1  meters. Curves for 0.5,  1.0,  and 1.5 meter depths  are not

              visible since the  concentration at these depths is zero.
                       Oil Content (1*3/1*3) us Tine
j. . vn-vvj.
C l.OE-003
0
n
e l.OE-005
n
t
p
» l.OE-007
i
0
n l.OE-009
l.OE-011

*""* ~











0 25 50 75 100 125 15
Identification:
Si tell
Soil Nam:
Tipton Sandy Loan

Pollutant Nam:
Pollutant*!

CAS NuMber:
123-4567





0
Tim (days)
Figure 11.   Oil  content as a  function of  time for depths of 0.1 meters.  Oil

              content curves for 0.5, 1.0,  and 1.5 meter  depths are not shown

              since the oil content is zero below the plow zone.
                                         33

-------
Table 8.   Concentration of pollutant in various phases and oil content  as  a
           function of depth at selected times.
Time = 0.
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
00 days
Total
Pollutant
g/m3
1 . OE+002
1. OE+002
1 . OE+002
1 . OE+002
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
5.9E+001
5.9E+001
5.9E+001
5.9E+001
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
>
Soil
g/kg
6.5E-003
6.5E-003
6.5E-003
6.5E-003
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000

Vapor
g/m3
3.2E-003
3.2E-003
3.2E-003
3.2E-003
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
0 . OE+000

Oil
g/m3
2.9E+003
2.9E+003
2.9E+003
2.9E+003
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
Oil
Content
m3/m3
2.5E-002
2.5E-002
2.5E-002
2.5E-002
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Time = 10.00 days
                             CONCENTRATION PROFILE
                                     Pollutant  in
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
luuajL
Pollutant
g/m3
O.OE+000
7.9E+001
7.9E+001
7.9E+001
2.2E+001
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
Water
g/m3
O.OE+000
5.2E+001
5.2E+001
5.2E+001
4.8E+001
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
Soil
g/kg
0 . OE+000
5.7E-003
5.7E-003
5.7E-003
5.3E-003
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
Vapor
g/m3
O.OE+000
2.9E-003
2.9E-003
2.9E-003
2.6E-003
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
Oil
g/m3
O.OE+000
2.6E+003
2.6E+003
2.6E+003
0 . OE+000
O.OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
U1J.
Content
m3/m3
2.1E-002
2.1E-002
2.1E-002
2.1E-002
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                      34

-------
Table 8.   Continued.
Time = 20

Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
.00 days
rri-. 4. _ 1
LOuai
Pollutant
g/m3
0 . OE+000
0 . OE+000
6.3E+001
6.3E+001
1.9E+001
0. OE+000
- 0 . OE+000
0 . OE+000
0 . OE+000
0. OE+000
CONCENTRATION
PROFILE

Pollutant in
Water
g/m3
0. OE+000
0 . OE+000
4.6E+001
4.6E+001
4.2E+001
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
Soil
g/kg
0 . OE+000
0 . OE+000
5.0E-003
5.0E-003
4.7E-003
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
Vapor
g/m3
0. OE+000
0. OE+000
2.5E-003
2.5E-003
2.3E-003
0. OE+000
0. OE+000
0. OE+000
0 . OE+000
0. OE+000
Oil
g/m3
0 . OE+000
0 . OE+000
2.3E+003
2.3E+003
0. OE+000
0 . OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000

f\l T
Ull
Content •
m3/m3 ,<
1.8E-002
1.8E-002
1.8E-002
1.8E-002
0 . OE+000
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Time = 30.00 days
                             CONCENTRATION  PROFILE
                                      Pollutant  in
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
lotai
Pollutant
g/m3
0 . OE+000
0 . OE+000
0. OE+000
5.0E+001
1.7E+001
1.4E+001
0 . OE+000
0. OE+000
0. OE+000
0 . OE+000
Water
g/m3
0. OE+000
0. OE+000
0. OE+000
4.0E+001
3.7E+001
3.0E+001
0. OE+000
0 . OE+000
0 . OE+000
0 . OE+000
Soil
8/kg
0. OE+000
0. OE+000
0. OE+000
4.4E-003
4.1E-003
3.4E-003
0. OE+000
0. OE+000
0. OE+000
0. OE+000
Vapor
g/m3
0 . OE+000
0 . OE+000
0. OE+000
2.2E-003
2.1E-003
1.7E-003
0. OE+000
0. OE+000
0. OE+000
0. OE+000
Oil
g/m3
0. OE+000
0. OE+000
0. OE+000
2.0E+003
0 . OE+000
0. OE+000
0 . OE+000
0 . OE+000
0. OE+000
0. OE+000
Ull
Content
m3/m3
1.6E-002
1.6E-002
1.6E-002
1.6E-002
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
0. OE+000
  Identification Code:  Site  //I
  Soil Name: Tipton  Sandy  Loam
  Compound Name: Pollutant//!
  CAS Number:  123-4567
RITZ
                                       35

-------
Table 8.   Continued.
Time = 40
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
.00 days
Total
Pollutant
g/m3
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
1.5E+001
1.2E+001
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
3.3E+001
2.7E+001
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000

Soil
* g/kg
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
3.6E-003
3.0E-003
O.OE+000
0 . OE+000
O.OE+000
O.OE+000

Vapor
g/m3
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
1.8E-003
1.5E-003
0 . OE+000
O.OE+000
O.OE+000
O.OE+000

Oil
g/m3
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
Oil
Content
m3/m3
1.4E-002
1.4E-002
1.4E-002
1.4E-002
O.OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Time = 50.00 days
                             CONCENTRATION  PROFILE
                                     Pollutant  in
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
lotaj.
Pollutant
g/m3
0 . OE+000
O.OE+000
O.OE+000
0 . OE+000
0 . OE+000
1.1E+001
8 . 8E+000
0 . OE+000
0 . OE+000
O.OE+000
Water
g/m3
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
2.4E+001
1.9E+001
O.OE+000
O.OE+000
O.OE+000
Soil
g/kg
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
2.6E-003
2.1E-003
0 . OE+000
O.OE+000
O.OE+000
Vapor
g/m3
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
1.3E-003
1.1E-003
O.OE+000
O.OE+000
0 . OE+000
Oil
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
U1J.
Content
m3/m3
1.2E-002
1.2E-002
1.2E-002
1.2E-002
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
  Identification Code: Site //I
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                       36

-------
Table 8.   Continued.
Time = 75
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
.00 days
Total
Pollutant
g/m3
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
6.4E+000
5 . 3E+000
O.OE+000
O.OE+000
CONCENTRATION PROFILE
Pollutant in

Water
g/m3
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
1.4E+001
1.2E+001
0 . OE+000
0 . OE+000

Soil
g/kg
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
1.6E-003
1.3E-003
O.OE+000
O.OE+000

Vapor
g/m3
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
7.8E-004
6.4E-004
0 . OE+000
O.OE+000

Oil
g/m3
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
Oil
Content
m3/m3
7.9E-003
7..9E-003
7.9E-003
7.9E-003
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
0 . OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
Time = 100.00 days
                             CONCENTRATION PROFILE
                                     Pollutant in
Depth
m
0.000
0.050
0.100
0.150
0.250
0.500
0.750
1.000
1.250
1.500
louaj.
Pollutant
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
3.2E+000
O.OE+000
Water
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
7 . OE+000
O.OE+000
Soil
g/kg
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
O.OE+000
7.7E-004
0 . OE+000
Vapor
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
3.8E«-004
0 . OE+000
Oil
g/m3
0 . OE+000
0 . OE+000
0 . OE+000
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
O.OE+000
O.OE+000
O.OE+000
U1X
Content
m3/m3
5.4E-003
5.4E-003
5.4E-003
5.4E-003
O.OE+000
0 . OE+000
O.OE+000
0 . OE+000
0 . OE+000
0 . OE+000
  Identification Code: Site #1
  Soil Name: Tipton Sandy Loam
  Compound Name: Pollutant//!
  CAS Number: 123-4567
RITZ
                                      37

-------
                 Concentration of Pollutant 
* l.OE-003
i
0
n l.OE-005
i (or-<\M
~H











%












Total






Identification:
Sitettl
Soil Name:
Tipton Sandy Loam
Pollutant Nane:
Pollutanttti
CAS Nuhber:
123-4567




             0.00    0.25    0.50   0.75   1.00    1.25    1.50
                                Depth 
Figure 12.    Concentration of  total pollutant  as a function of depth  for times

               of 10, 50,  and  100 days.
                 Concentration of Pollutant  (sr/i»3) us Depth
J.. V&TVVO
C l.OE+001
0
n
c
• l.OE-001
n
p
« l.OE-003
i
o
" l.OE-005
l.OE-007



















,


















Hater









                                                            Identification:
                                                            Sitettl

                                                            Soil Name:
                                                            Tipton Sandy Loam

                                                            Pollutant Name:
                                                            Pollutanttti

                                                            CAS Number:
                                                            123-4567
             0.00   0.25   0.50    0.75    1.00    1.25    1.50
                                Depth (N)
Figure 13.    Concentration of  pollutant in water as  a function of depth for

               times of  10, 50,  and 100  days.
                                           38

-------
                 Concentration of Pollutant  (sr/kg) ws Depth

C l.OE-003'
o
n
c
e l.OE-005
n
t
» l.OE-007
t
i
o
» i.OE-009
i AF-AI i




























Soil




















































                                                            Identification:
                                                            Site«l

                                                            Soil Na«e:
                                                            Tipton Sandy Loan

                                                            Pollutant Nam:
                                                            PollutantHl

                                                            CAS NuMber:
                                                            123-4567
             0.00   0.25    0.50    0.75    1.00    1.25    1.50
                                Depth (N)
Figure 14.    Concentration of  pollutant in soil as a  function  of depth  for

               times of  10, 50,  and 100  days.
                 Concentration of  Pollutant (g/i«3) us  Depth
i. Vt-WVJ.
C l.OE-003
o
n
• l.OE-005
n
t
r
a l.OE-007
t
i
0
" i.OE-009
1 AC A1 1


























Uapor






Identification:
Sitettl
Soil Name:
Tipton Sandy Loan
Pollutant Name:
Pollutant*!
CAS Hunker:
123-4567




             0.00    0.25    0.50    0.75    1.00    1.25    1.50
                                Depth (N)
 Figure  15.    Concentration of pollutant  in vapor  as a function of  depth for

               times of 10,  50, and 100 days.
                                           39

-------
                Concentration of Pollutant (g/i»3) us Depth
i . vAi-wn
C l.OE+002
0
n
» l.OE+000
n
t
a l.OE-002
i
0
•> l.OE-004
l.OE-006
0.








—







' Oil '
*






00 0.25 0.50 0.75 1.00 1.25 1.
Identification:
Sitettl
Soil Nane:
Tipton Sandy Loan
Pollutant Nane:
Pollutantttl
CAS NuMbep;
123-4567



50
Depth (M)
Figure 16.   Concentration of pollutant in oil  as  a function of depth for  10
             days  after application. Note the concentration decreases to zero
             at the  plow zone depth. The concentration was zero at 50, and 100
             days.
                     Oil Content  d*3/i«3) vs Depth
1. VJC.-WJ.
C l.OE-003
0
n
e l.OE-005
n
p
« l.OE-007
i
o
n l.OE-009
l.OE-011
0.
















00 0.25 0.50 0.75 1.00 1.25 1.
Identification:
Sitettl
Soil Name:
Tipton Sandy Loan
Pollutant Name:
, Pollutantttl
CAS Nunbep:
123-4567



50
Depth (M)
Figure 17.   Oil content as a function of depth for times of 10, 50, and 100
             days.  The oil does not move downward but the oil content
             decreases due to degradation.
                                       AO

-------
        Tine:   10.00 days
       Depth (H).
              'Total
    Soil Nane: Tipton  Sandy Loan
   CflS Hunker: 125-4567
    .  „	f Pollu
—Concentration o
 Mater       Soil
llutant '-•
   Uapor
                               Oil
 Oil
Content
n
          0.00
          0.25
          0.50
          0.75
          1.00
          1.25
          1.50
          Pattern description for total concentration of pollutant (g/n3).
            H.OE+001   >1.0E+000   M.OE-001   >1.0E-002   >1.0E-003   >=O.OE+000
Figure 18.    Concentration  bar graphs representing  the pollutant  and oil  in
               the treatment  zone at  a time  of 10 days. The  concentrations
               represented by the patterns in each  phase can be displayed by
               pressing the  key.
         Tine:   30.00 days
     c&'Ucsi Il5^5n67SandB Loa"
Concentration of Pollutant	—
                                       Soil
                     Uapor
                                     Oil
                           Oil
                          Content
                          n
           1.50
            Pattern description for total concentration of pollutant  (g/n3).
              >1.0E+001   H.OE+000  >1.0E-001   >1.0E-002   >1.0E-003   >=O.OE+000
Figure 19.    Concentration bar graphs for 30 days.
                                           Al

-------
        Tine:   50.00 days
                     Soil  Mane: Tipton Sandy Loan
                    CAS Number: 123-4567
              -Concentration  of Pollutant 	
 Total        Hater       Soil         Uapor



^1      ^1      ^^\      ^\
                                                                 Oil
                                                                 Oil
                                                                Content
           0.25
           0.50
           0.75
           1.00
           1.25
           1.50
           Pattern description  for concentration of pollutant in water (g/n3).
             >5.9E+000   >5.9E-001   >5.9E-002   >5.9E-003   >5.9E-004   >=OTOE+C

Figure 20.     Concentration bar graphs  for  50 days.
         Tine:  100.00 days
           0.25
           0.50
           0.75
           1.00
           1.25
           1.50
                         Soil  NaMe: Tipton Sandy Loan
                                     3-
                                   CAS Number: 123-4567
                              -Concentration of Pollutant
                            Hater
                             Soil
                                     Uapor
Oil
  Oil
Content
n
Pattern description for concentration  of pollutant in water (g/«3)
  >5.9E+000
          >5.9E-001   >5.9E-C
                                                >5.9E-003   >5.9E-004   >=O.OE«000
Figure  21.    Concentration bar graphs  for  100 days.
                                              42

-------
                                FILE STRUCTURE

Disk files are used in this software for two purposes. The first is to store
input parameters entered at one time for use at another time. The second is
for storing output tables for later printing or display or for use in other
documents.

The input parameter files are made up of a single record of binary
information. The record is composed of parameters in the sequence listed in
screens 2, 3, and A. All numeric entries are stored as floating point values.
All alphanumeric entries are stored as strings.

Tabular data stored in files are written as text in ASCII characters.

-------
                             REFERENCES  CITED

 1.   Clapp,  Roger  B.  and George M.  Hornberger.  1978.  Empirical equations  for
     some soil hydraulic properties. Water Resources  Research 14:  601-604.

 2.   Jury, W.A., W.F.  Spencer,  and  W.J. Farmer.  1983.  Behavior assessment
     model for trace  organics  in soil:Model description.  J.  Environ.  Qual.
     12:558-564.
                             >
 3.   Laskowski, D.A.,  C.A.I. Goring, P.J.  McCall,  and R.L.  Swann.  1982.
     Terrestrial environment.  In Environmental Risk Analysis for Chemicals,
     R.A. Conway (Ed.).  Van Nostrand-Reinhold Co.,  NY.  pp 198-240.

 4.   Karickhoff, Samuel  W.  1981.  Semi-empirical estimation of sorption of
     hydrophobic pollutants on  natural sediments and  soils.  Chemosphere
     10:833-846.

 5.   Karickhoff, S.W., D.S. Brown,  and T.A. Scott.  1979.  Sorption of
     hydrophobic pollutants on  natural sediments and  soils.  Water Research
     13:241-248.

 6.   Millington, J.R.  and J.M.  Quirk.  1961. Permeability of porous solids.
     Trans Faraday Soc.  57:1200-1207.

 7.   Ralston,  Anthony. 1965. A  First Course In Numerical Analysis,  McGraw-
     Hill Book Co., New  York,  pp 121-129.

 8.   Short,  Thomas E.  1985. Movement of contaminants  from oily wastes during
     land treatment.  Proceedings of Conference on Environmental and Public
     Health  Effects of Soils Contaminated with Petroleum Products,  Amherst,
     MA.

 9.   Swartzendruber,  Dale.  1960.  Water flow through a soil profile as
     affected  by the  least permeable layer. J.  of Geophysical Research
     65:4037-4042.

10.   Verschuren, K. 1983. Handbook  of  Environmental Data on Organic
     Chemicals. Van Nostrand Reinhold  Co., New York,  New York., 1310  pp.
                                     44

-------
APPENDIX
 45

-------
                          MATHEMATICAL BASIS OF RITZ

This section summarizes the mathematical equations used in this version  of  the
RITZ software. They are presented for your information only. No attempt  is
made here to explain the mathematical derivations of these equations.  See
Short(1985) for those derivations.

Total Pollutant Concentration; The total concentration of the pollutant
Crri(x,t) at position x and time t is given by
 •L                              ^
   Cij(x,t) = 0                          for x < top of pollutant  slug

   Crji(x,t) = C-poexp(-(jpt)               for top of pollutant slug <  x  <  pzd

   CT(x,t) = CToRexp(-Mpt)/(R + RTexp(-no(t-(x-pzd)/Vp)))

                                        for pzd < x < bottom of pollutant  slug

        ,t) = 0                          for x > bottom of pollutant  slug
where
   C-jo is the initial pollutant concentration,
   Mp  is the degradation constant for the pollutant
           (Up = ln(2)/tip),
   Mo  is the degradation constant for the oil
           (MO = ln(2)/40),
   R  is the retardation factor for the pollutant  in  the treatment zone,
   R-p is the contribution of oil to the retardation,
   Vp is the velocity of the pollutant in the lower treatment zone,  and
   pzd is the depth of the plow zone .

Cr|i(x,t) is the mass of pollutant per unit volume of soil.  In this software
these units are grams of pollutant per cubic meter of soil.  The initial
pollutant concentration is given by

   CTo = SAR • Sp  /

where
   SAR is the sludge application rate and
   Sp is the concentration of pollutant in the  sludge.
                                      46

-------
The retardation factor, R, is given by

   R = 1 + (PKD + (9S - 9)KH) / 6

where
   p  is the bulk density of the soil,
   9  is the water content of the soil on a volume basis,
   9S is the saturated water content of the soil on a volume basis,
   Kj) is the partition coefficient for pollutant in the soil, and
   Kg is the dimensionless value of Henry's Law constant, (
The partition coefficient is given by Kp = K0Qf0Q where KQQ is the organic
carbon partition coefficient and f0£ is the fractional organic carbon content
of the soil.

The parameter R-^ is given by

   RlJ = ^oCKQ - Kjj) / 9

where
   $o is the initial oil content or the volume fraction occupied by oil, and
   KQ is the partition coefficient for oil.

The volumetric water content of the soil, 9, is given by

   6 = 8s[Vd/k8]1/(2bf3)

where
   V^ is the recharge rate,
   ks is the saturated conductivity of the soil, and
   b is the Clapp and Hornberger constant for the soil.

The velocity of the pollutant in the lower treatment zone, Vp, is given by

   Vp = Va / R

where Va = V^/9 is the aqueous or pore water velocity.
                                      47

-------
Pollutant Concentration in Water; The concentration of pollutant in water,
C^(x,t) at position x and time t is given by

   Cy(x,t) = 0                          for x < top of pollutant slug

   Cw(x,t) = CT(x,t) / 9(R + RTexp(-^i0t))

                                        for top of pollutant slug < x < pzd

   Cw(x,t) = CT(x,t) / R6          *    for pzd < x < bottom of pollutant slug

       ,t) = 0                          for x > bottom of pollutant slug
where all the symbols are those defined for the total pollutant concentration.
C^(x,t) is the mass of pollutant in water per unit volume of water. In this
software these units are grams of pollutant per cubic meter of water.

Concentration of Pollutant in Soil: The concentration of the pollutant in the
soil phase Cg(x,t) at position x and time t is given by

   Cs(x,t) = KDCw(x,t)

where
   Kp is the soil: water partition coefficient for the pollutant and
   C^(x,t) is the concentration of pollutant in water.

Cg(x,t) is the mass of pollutant in water per unit mass of soil solids. In
this software these units are grams of pollutant per kilogram of soil.

Concentration of Pollutant in Vapor; The concentration of the pollutant in the
vapor phase Cy(x,t) at position x and time t is given by

   Cv(x,t) = KHCw(x,t)

where
   KJJ is the dimensionless (Henry's Law) vapor:water partition coefficient and
   C^j(x,t) is the concentration of pollutant in water.

CyCx.t) is the mass of pollutant per unit volume of vapor. In this software
these units are grams of pollutant per cubic meter of vapor.
                                      48

-------
Concentration of Pollutant in Oil; The concentration of pollutant in the oil
phase C0(x,t) at position x and time t is given by
   C0(x,t) = K^Cx.t)                            for x < pzd

   C0(x,t) = 0                                    for x > pzd

where
   Ko is the dimensionless oiliwater partition , coefficient for the pollutant,
   pzd is the depth of the plow zone, and
   C^(x,t) is the concentration of the pollutant in water.

CQ(x,t) is the mass of pollutant per unit volume of oil. In this software
these units are grams of pollutant per cubic meter of oil.

Oil Content: The oil content $(t) in the plow zone at time t is the volume of
oil per unit volume of soil and is given by
        = $0exp(-M0t)

where
   $0 is the initial oil content in the plow zone and
   Mo is the degradation constant for oil.

The oil content is uniform throughout the plow zone at each instant of time.
The oil content is zero below the plow zone at all times. The initial oil
content <&o is given by

   
-------
                           RT/R)exp[Moxtop/Vp - F(xtop)] - RT/R}

                                        for 0 < xtop < pzd
                            (xtop - Pzd)/Vp - G(xtop)
                                        for pzd < xtop < tzd
where
   F(xton) = ^aV^lnU + xt?D/g),
                         {top
   top-* = ^oavp lnU + xtop/B-'.
G(xto_) = aV:1ln[(g + xtoJ/(g + pzd)],
   g = DS5/DA + a,
   a = KHDs/Va8,
   DA is the diffusion coefficient of the pollutant vapor in air,
   Dg is the diffusion coefficient of the pollutant vapor in the soil,
   6 is the thickness of the stagnant boundary layer above the soil, and
   tzd is the depth of the treatment zone.

Although the equations above hold for all depths, numerical overflow occurs  in
the first equation when Moxtop/^p •"•s verv large. In this case, an approximate
form of the equation is used which is

   ttop(xtop) =  Mo^MoXtop/Vp - F(xt0p) + ln(1 + RT/R)}
                                        for 0 < xtop < pzd.

The diffusion coefficient of the pollutant in the soil, Dg is given by

   DS = DAn10/3/ei

where n is the initial air content of the soil (Millington and Quirk, 1961).

The thickness of the stagnant boundary layer (Jury et al., 1983) is given  by

   6 = DwPwvCl - RH)/2EpWL

where
   Ify is the diffusion coefficient of water vapor in air,
   RH is the relative humidity of the air,
   E is the evaporation rate,
       is the density of water vapor, and
       is the density of liquid water.
The ratio of the density of water vapor to the density of liquid water  is
given by (Short, 1985)

   PWV/PWL = ao + aiT + a2T2 + a3T3
where
   T is the temperature in degrees Celsius,
   a0 = 4.60843696E-06,
   ax = A.0710817E-07,
   a2 = 3.02943E-09, and
   a3 = 3.9405E-10.
                                      50

-------
Time at Which the Bottom of the Pollutant Slug Reaches a Specified Depth; The
bottom of the pollutant slug is located at the plow zone depth at time zero.
It moves downward through the treatment zone as time increases. The time at
which the bottom of the slug reaches a position xt,ottom is given by

                    = °                           for xbottom * Pzd

                    = ^bottom " Pzd)/vp

                                                  for xbottom > Pzd
where
   pzd is the depth of the bottom of the plow zone and
   Vp is the velocity of the pollutant in the lower treatment zone.

Flux of Pollutant Vapor for a Specified Position of the Top of the Pollutant
Slug and the Corresponding Time; The flux of pollutant vapor, J(t(xtOD)),
moving upward out of the treatment zone at the time t is given by

   J(t) = aVpCToexp(-Mpt)/{(g - a + xtop)[l + (RT/R)exp(-Mot) ] }

                                                  for 0 < xtop < pzd

   J(t) = aVpCToexp(-npt)/{(g - a + xtop)[l + (RT/R)exp(-noAt)] }  ,

                                                  for pzd < xtop < tzd

where
   t = ttop(xtOp) as defined previously and
Total Amount of Pollutant Lost as Vapor; The amount of pollutant lost in the
vapor form can be obtained by integrating the vapor flux over the time in
which the pollutant is in the plow zone and the treatment zone. That is
         r1
   Mv - Jo J(t)dt
where t is the time at which the top of the pollutant slug reaches the bottom
of the treatment zone. It is computationally more efficient to change variable
of integration and integrate over distance. This integral then becomes
        rpzd                     tzd
   My = JQ   J(t(x))(dt/dx)dx + J ^ J(t(x))(dt/dx)dx

The integrands in the above equation are

   II = aClo^P^pttopW^B + x)                   for 0 < x < pzd (term  1)

   I2 = aCToexp(-Mpttop(x)/{(g + x)(l + (RT/R)exp(-MoAt))}

                                                   for pzd < x < tzd (term  2)

where At = ttop(x) - (x - pzd)/Vp. The integration is carried out numerically
using Romberg integration. Convergence is assumed when the difference between


                                      51

-------
consecutive approximations to the integral is less than l.OE-06 percent of the
pollutant applied.

Total Amount of Pollutant Leached Below the Treatment Zone; The amount of
pollutant leached below the treatment zone, M^, is obtained by integrating the
product of the recharge rate and the pollutant concentration in water at the
treatment zone depth. That is
        ft
   MT * L V 9C (tzd.t)dt
    L   J0  a  W
where
   V&9 = the recharge rate and
   C^(tzd,t) is the concentration of pollutant in water defined previously.

This integration is also performed numerically using Romberg integration
(Ralston, 1965). Convergence is assumed when the difference between
consecutive approximations to the integral is less than l.OE-06 percent of the
pollutant applied.

Total Amount of Pollutant Degraded in the Treatment Zone: The amount of the
pollutant degraded, MJJ, within the entire treatment zone is equal to the sum
of the amounts degraded in the plow zone and in the treatment zone. That is
         pzd
=
  J
                        (x,t   (x))(dt/dx)dx
         U  p          T    top
         tb
+ I   M Acc(t)dt
  J0
  -
+
  J
            p
         tzd"
             u Acc(t)(dt/dx)dx
         pzd  p
where tb = tto_(pzd) is the time at which the top of the slug reaches the
depth of the plow zone and Acc(t) is the mass of pollutant accumulated in the
lower treatment zone. The first integral represents the degradation within the
plow zone. The second integral represents the degradation in the lower
treatment zone before the top of the slug reaches the lower treatment zone.
The third integral represents the degradation in the lower treatment zone
after the slug is entirely in that zone. These integrals are evaluated by the
Romberg integration with the same convergence criteria as for volatilization
and leaching.
                                      52

-------
The accumulation of pollutant in the lower treatment zone, Acc(t), is given by

   Acc(t) = CToexp(-npt){(xbottom - xtop) - VpM-1ln(H(xbottom)/H(xtop))}

where

   H(x) = 1 + (RT/R)exp(-M0(xb - x)/Vp)

and

   xb = pzd + Vpt.

Mass Balance Error; Pollutant applied to the soil must be volatilized,
leached, or degraded by the time the top the slug reaches the treatment zone
depth. Each of these three components are evaluated above. If the
computational techniques are accurate, the sum of these should be equal to the
amount of pollutant applied. The mass balance computational error is given by

   Error = MT - Mv - ML - MD

where Mf is the mass of pollutant applied to the plow zone. The other symbols
were defined previously.
                                      53

-------
Table 9. List of symbols with meaning and units as used in this section.


   b        Clapp and Hornberger constant, dimensionless
   Crp       total concentration of pollutant in all phases, g/rn^
   Cy       concentration of pollutant in water, g/mj
   Cg       concentration of pollutant in soil, g/kg
   Cy       concentration of pollutant in vapor, g/m
   CQ       concentration of pollutant in oil, g/m^
   GIJO      total concentration of pollutant at time zero, g/m^
   D^       diffusion coefficient of pollutant in air, m^/day
   Dg       diffusion coefficient of pollutant vapor in soil, m^/day
   Dy;       diffusion coefficient of water vapor in air, m^/day
   E        evaporation rate, m/day
   fgc      fractional organic carbon content of soil
   J        flux of pollutant vapor, g/m^-day
   k        unsaturated hydraulic conductivity, m/day
   ks       saturated hydraulic conductivity of soil, in/day
   KD       soilrwater partition coefficient of pollutant, nrVkg
   Kg       vapor:water partition coefficient of pollutant
            or the dimensionless Henry's law constant, dimensionless
   KQ       oil:water partition coefficient of pollutant, dimensionless
            organic-carbon:water partition coefficient, nrVkg
            total amount of pollutant degraded, g/m
   ML       total amount of pollutant leached below treatment zone, g/m^
   My       total amount of pollutant lost in vapor form, g/m^
   pzd      plow zone depth, m
   R        retardation factor for pollutant (ignoring oil), dimensionless
   RP       contribution of oil to retardation of pollutant, dimensionless
   RH       relative humidity, dimensionless
   SAR      sludge application rate, kg/ha
   So       initial concentration of oil in the sludge, g/kg
   Sp       initial concentration of pollutant in the sludge, g/kg
   T        temperature,- °C
   t        time, days
   tip      degradation half-life of the pollutant, days
   ti0      degradation half-life of the oil, days
   tzd      treatment zone depth, m
   V^       recharge rate, m/day
   Va       pore water velocity, m/day
   V        velocity of the pollutant in the lower treatment zone, m/day
   x        distance from the soil surface, m
   p        bulk density of soil, kg/m^
   Po       density of oil, kg/m^
            density of water vapor, kg/rn^
            density of liquid water, kg/^
   6        water content on a volume basis, itr/nr
   6_       saturated water content on a volume basis, nr/mr
                                      54

-------
Table 9. Continued.
            oil content (volume fraction of oil) at time t, nrVn
   $0       initial oil content (volume fraction of oil), nrVnr
   Mp       degradation constant of pollutant, days"
   Ho       degradation constant of oil, days"^-
   6        thickness of stagnant boundary layer, m
   H        initial air content of soil, m-Vm^
                                      55

-------
                          INPUT PARAMETER ESTIMATION

The user of this software must provide soil, chemical, and environmental
parameters to define the land treatment site. The parameters may be obtained
experimentally for the site, based on published values such as those in
Verschuren (1983), or estimated from related parameters. The software includes
a few built in estimators for certain required parameters. These are intended
for use in situations in which the required parameter is unknown. They should
be used with caution. In this secteion, the approximations available for each
parameter are described briefly. Table 10 contains a list of the numerical
parameters with their units and symbols used in the previous section.

Fractional organic carbon content
                    If this  is not known  but  the  organic matter  content  of  the
                    soil  is  known, this is approximately equal to the product
                    of 0.4 and the fractional organic matter  content.

Saturated water content
                    This  can be  estimated from  the bulk density,  p,  and
                    particle density,  ps, of  the  soil using the  equation
                     9S = 1  - p/ps. The particle  density for  most mineral
                    soils is between  2600 and 2700 kg/mj.  If  the particle
                    density  is not known  a value  of  2650 kg/m3 is usually a
                    good  estimate.

Clapp and Hornberger constant
                    If this  parameter is  not  known,  it can be estimated  using
                    the values presented  by  Clapp and Hornberger for different
                    soil  textures. This table will be displayed  on  the screen
                    if the help  key is pressed.

Organic carbon partition coefficient
                    If this  parameter is  not  known,  it can be estimated
                    (Karickhoff,  1981) from  the water  solubility, S (g/m3),
                    the molecular weight, MW  (g/mole), and the melting point,
                    MP (°C)  of the pollutant. If
                    x = -0.921og(S/(55556-MW) -4.404), then the  organic  carbon
                    partition coefficient, KQQ, is approximately

                         KQC ~ 10X                     if melting point < 25°C
                            „ 10x -  O.OKMP  -  25)

                                                      if melting,point > 25°C.

                    If these pollutant properties are not  known,  KQQ can be
                    estimated from the octonal-water partition coefficient,
                    KQW,  using the relation  of  Karickhoff  et  al.  (1979)
                                      56

-------
Table 10.   Input parameters required by the RITZ model.
Input Parameter
Fractional organic carbon content
Bulk density
Saturated water content of soil
Saturated hydraulic conductivity
Clapp and Hornberger constant
Concentration of pollutant in sludge
Organic carbon partition coefficient
Oil-water partition coefficient
Henry's law constant
Diffusion coefficient of pollutant in air
Half -life of pollutant
Concentration of oil in sludge
Density of oil
Half-life of oil
Sludge application rate
Plow zone depth
Treatment zone depth
Recharge rate
Evaporation rate
Air temperature
Relative humidity
Diffusion coefficient of water vapor in air
Units
--
kg/m3
m3/m3
m/day
--
g/kg
m3/kg
--
--
m^/day
days
g/kg
kg/m3
days
kg/ha
m
m
m/day
m/day
degrees C
--
m^/day
Symbol
foc
P
9s
ks
b
Sp

K0
%
DA
4?

PO
t50
SAR
pzd
tzd
Vd
E
T,
RH
Dy
Oil-water partition coefficient
                    If this is not known,  it can be approximated by the
                    octonal water partition coefficient for the  pollutant.
Henry's law constant
                    If the dimensionless  Henry's law constant  is  not  known,  it
                    can be determined from the value of the constant  in units
                    of atm-m3/mole by dividing the dimensioned value  by 0.024.

                    If the dimensioned constant is not known,  the
                    dimensionless value of Kg can be estimated according to
                    Laskowski et al.  (1982) from the water solubility,
                    molecular weight, and vapor pressure of the pollutant
                    using the relation

                         KH ~ VP-MW / (760-S)

                    where S is the solubility of the pollutant (g/m3),  MW is
                    the molecular weight  (g/mole), and VP is the vapor
                    pressure (mm of Hg).
                                       57

-------
                              PARAMETER AVERAGING

The soil parameters in this model are assumed to be uniform throughout the
treatment site. This will not be true in general. The software includes an
option to calculate a weighted average value for soil properties known for
different layers in the soil. This section outlines the averaging schemes
employed. The software enables the user to enter values of d^ and V.^ for each
layer. It then calculates the average and places it in the data entry screen.
                                *.
Depth Weighted Average; The average value calculated for all parameters except
the saturated hydraulic conductivity is the depth weighted average of the
values for each layer. Consider a site in which the depth of the soil layer  i
is d^ for i = 1, 2, 3, ..., N and dg = 0 and d^ is equal to the treatment zone
depth. If the parameter of interest has a value V^ for i = 1, 2, 3,  ..., N,
then the depth-weighted average V is given by

   V = E^ 1 w-V-
   v   ^i=l wivi

where w^ = (d.^ - d^_^)/djj for i = 1, 2, 3, .. ., N.

Average Saturated Hydraulic Conductivity; If d^ contains the depths  of each
layer of soil as explained above for depth weighted averages and if  k^
contains the corresponding saturated hydraulic conductivities for each layer,
the equivalent conductivity, kg, for the layered soil (Swartzendruber, 1960)
is given by             '

   ks = dN /

Screen 8 illustrates the use of the averaging feature built in to the
software. In this case, the  key was pressed when the user was being
prompted for the fraction organic carbon content. The treatment zone was made
up of 5 layers so the user chose to use this averaging scheme to compute the
average value for the site. In this case, each line includes an entry for the
depth of the layer and the fraction organic carbon content for the layer. The
two numbers must be separated by a comma or a blank space. When the   key
is pressed, the average value is calculated and placed in the appropriate line
on Screen 2. The user can then continue entering data there.

NOTE; THE AVERAGE IS CALCULATED TO THE MAXIMUM DEPTH ENTERED. THIS MAXIMUM
DEPTH SHOULD CORRESPOND TO THE DEPTH OF THE TREATMENT ZONE.
                                      58

-------
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.


Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,
Depth,


m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
,
,
»
,
>
,
,
,
,
,
,
,
»
,
»
and
and
and
and
and
and
and
and
and
and
and
and
and
and
and
Averaging
fraction organic
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
fraction
Display help
Proceed - all


Abort
option
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
Screen
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
carbon
0
0
0
1
1








.10 0
.15 0
.30 0
.05 0
.50 0








.02
.007
.005
.002
.001








for entries
entries made
and return to parameter
entry screen
Screen 8. Screen for depth wei
                                      59

-------
                            Index
A.
abort 6
assumptions 1
Clapp and Hornberger
     constant 3
computer 6
concentration bar graphs

concentration graphs 17
configuration
   graphics card 9
   monochrome card 9
cursor keys 6

D
data entry editor 5, 6,

degradation 2, 14
depth-weighted averages
     * > 11
directory 15
disk directory 15
disk files 4, 7, 43
dispersion 1

E
editor 5, 6, 11
Esc 5,  6
execution 10
file names 8
file output 18
file structure 43
flux of water 1
function keys 6 -

G
graphs 4, 8
   printouts 8
1C
keys
   Backspace 8
   cursor 6
   Delete 8
   down arrow 6
   End 6
   Enter 7
   Esc 7
   Fl 7
   F10 7
   F2 7
   F7 7, 15
   F8 7, 15
   function 6
   Home 6
   left arrow 6
   PgDn 7
   PgUp 7
   right arrow 6
   up arrow 6

L
land treatment site 2

M
mass balance 16
model assumptions 1
no data 8, 18
non-uniform soils 11

O
oil 1
oil properties 13
operating system 6
output device 18
output options 4, 16
outputs
   graphs 4
   tables 5
half life 14
hardware 6
Henry's Law constant 14
hydraulic conductivity
     function 3
input parameter 56
installation 9
   fixed disk 9
   floppy disk 9
   graphics card 9
   monochrome card 9
parameter entry 4, 11
   file 15
   keyboard 15
parameter estimation 56
partition coefficient
   oil 13
   organic carbon 13
   vapor 14
plow zone 1
plow zone depth 14
                             60

-------
pollutant concentration
   oil 13, 48
   soil 13, 48
   total 46
   vapor 14, 48
   water 13, 14, 47
pollutant properties 13
printer 6
printer graphics 8
printer output 18
program execution 10
program interruption 6

R
recharge rate 14
software operation 5
soil properties 1, 11

T
tables 5
text files 18
treatment zone 1
treatment zone depth 14

U
units 56
unknown parameters 8, 56

W
waste application 1
site description 1
                             61

-------
  DAYS
SATURATED
   ZONE
  SOLUTE
TRANSPORT
MODELING
   MOC
                                 CO

-------
           OBJECTIVES
Highlight assumptions, limitations, and
uncertainty in groundwater modeling.

Introduce processes that control fate
and transport of contaminants in the
saturated zone.

Discuss management/regulatory con-
cerns in model application.

Provide an overview on the groundwater
modeling procedure.

Identify data type and quality require-
ments to satisfy the objectives of
modeling.

Provide insight into reviewing a model
application.

Expose EPA personnel to a widely used
solute transport code (MOC).

Provide a basis and foundation (theory)
of MOC.

Provide hands on experience with the
use of MOC.
                              Cjeoi
                                  Irans, inc

-------
                SCHEDULE
8:00    Introduction
        -  overview of course
        -  objectives
        -  schedule
        -  class notes
        -  role of modeling

8:15    Modeling Overview
        -  problem statement
        -  processes that control fate and transport
        -  equations
        -  uncertainty
        -  management/regulatory concerns

9:15    Solute Transport Process
        —  source
        -  velocity variations
        -  dispersion
        -  heterogeneity

10:00  Break

10:15  Modeling Procedure/Numerical
        Methods
        —  governing equations
        -  boundary conditions
        -  initial conditions
        -  finite-difference methods
        -  gridding
        -  general application procedure
                                                rans, inc.
                                                GKOUNDWATDI SPECIALISTS

-------
       SCHEDULE (Continued)

11:15  MOCCode
       -  features
       -  data requirements
       -  application
       -  solution procedure
       -  mass balance
       -  output

12:00  Lunch

1:00   Rocky Mountain Arsenal Case Study

1:30   Data Input Instructions For MOC

2:00   Special Problems
       -  point sources
       -  point sinks
       -  reaction terms
       -  subgrid option

2:15   Example Problem
       -  preprocessors
       -  introduction to problem
       -  computer facilities
       -  hands on use

4:00   Discussion
       -  example problem results
       -  special problems
       -  questions
       -  summary of course
5:00   Adjourn
                                      GeoT
                                           lr«
 trans, inc.
GxouNDwmx SPECIALISTS

-------
       CLASS NOTES
1.      Equation Derivation
2.      Model Documentation (MOC)
3.      Program Updates
4.      Bibliography
5.      Perspective on Modeling
6.      Case History
7.      Selected Slides
8.      Class Problem
9.      Class Problem (Results)
                               Geol
                                   Iransjnc.
                                   CJOOKDWATDl SPECIALISTS

-------
                                BP-F-1-28
   MODEL MISUSE


MOST MODELING MISTAKES OCCUR
IN MODEL APPLICATION; COMMON
WAYS TO MISUSE MODELS INCLUDE:

(1) OVERKILL-MAKING THE MODEL
   MORE COMPLEX THAN DATA
   AND/OR GOALS REQUIRE;
(2) IMPROPER CONCEPTUALIZATION -
   BASING THE MODEL ON A POOR
   OR INCOMPLETE SITE
   CHARACTERIZATION;
(3) IMPROPER MODEL SELECTION -
   SELECTING A MODEL WHEN NOT
   FULLY UNDERSTANDING ITS
   LIMITATIONS;
(4) IMPROPER BOUNDARY CONDITION
   AND/OR EQUATION PARAMETER
   SPECIFICATION:
(5) APPLICATION OF A GENERIC
   MODEL TO A SPECIFIC SITE:
(6) INAPPROPRIATE PREDICTION -
   FORECASTING UNDER VASTLY
   DIFFERENT CONDITIONS THAN
   THOSE USED FOR HISTORY
   MATCHING;

-------
                                 BP-F-1-27
MODEL MISUSE (CONT.)
 (7) MISINTERPRETATION - POOR
    HYDROLOGIC INTERPRETATION
    OF COMPUTED RESULTS;
 (8) MISUSE OF NUMERICAL
    APPROXIMATION (IMPORTANCE
    OF MASS BALANCE);
 (9) UNDETECTED CODING ERROR
    (IMPORTANCE OF MODEL
    VERIFICATION).

-------
                                                        Bf-F-a-13
             QUALITY ASSURANCE
  TECHNICAL COMPETENCY
   — APPROACH
   — SYSTEM CONCEPTUALIZATION
   — INTERPRETATION OF RESULTS

  PROGRAM USE
  PROGRAM MODIFICATIONS
COMMON TO ANY
TECHNICAL PROJECT
SPECIAL CONCERN
WHEN MODELS
ARE USED
LEVEL AND EFFORT OF QUALITY ASSURANCE WILL DEPEND ON
PROJECT NEEDS

-------
                                                       BP-F-«-i4
    PROGRAM USE - QA CONCERNS
NUMERICAL ACCURACY
—  MODEL SELECTION
—  GRID SIZE
—  TIME STEP SIZE
MODELING ASSUMPTIONS
—  BOUNDARIES AND CONDITIONS
—  ZONATION
—  CALIBRATION CRITERIA

DATA PREPARATION
-  REDUCTION FROM FIELD DATA
—  KEYPUNCHING
INVOLVE DECISIONS
- DOCUMENT REASONS
- MAKE COMPARISONS
- CHECK
REQUIRES ACCURACY
- CHECK
- MAINTAIN RECORDS

-------
                                           BP-F-6-15
   PROGRAM MODIFICATIONS
FOR MANY APPLICATIONS, IT IS NECESSARY TO
MODIFY EXISTING PROGRAMS THAT HAVE BEEN
DOCUMENTED AND TESTED

COSMETIC-CHANGE INPUT OR OUTPUT FORMAT

MAJOR-CHANGE PHYSICAL DESCRIPTION OR
NUMERICAL METHOD

-------
                                            BP-F-6-18
    ALL PROGRAM MODIFICATIONS
SHOULD BE DOCUMENTED AND TESTED

 DOCUMENTATION SHOULD INCLUDE
   • REASON FOR CHANGE
   • DESCRIPTION OF CHANGE
   • NEW LISTING OF PART OF PROGAM CHANGED

 TESTING SHOULD
   • BE CONSISTENT WITH CHANGES MADE
     (I.E., MAJOR CHANGES REQUIRE EXTENSIVE TESTING
     WITH KNOWN SOLUTIONS)
   • HAVE DOCUMENTION OF TEST
      -  DISCUSSION OF TEST PROBLEM SELECTION
      -  GRAPHICAL COMPARISON
      -  COMPLETE OUTPUT

-------
                                          BP-F-6-11
          SCHEDULING
MODELING EFFORTS SHOULD BE WELL-
COORDINATED WITH OTHER PROJECT TASKS

MAXIMUM TECHNICAL FEEDBACK TO AVOID DELAYS

-------
                                                                         BP-F-6-12
MAJOR PROJECT TASKS
MODELING APPLICATION
        PROJECT
        PLANNING
          DATA
        COLLECTION
          DESIGN
          DATA
        COLLECTION
       AND ANALYSIS
          QUALITATIVE
       CHARACTERIZATION
              REPORT
              WRITING
         MODEL NEED
        AND SELECTION
            SYSTEM
       CONCEPTUALIZATION
        SENSITIVITY ANALYSIS
        MODEL CALIBRATION
                                           QUANTITATIVE PREDICTION

-------
                                               BP-F-a-4
              COMMENTS

THE EFFECTIVE USE OF MODELS USUALLY REQUIRES A
TEAM EFFORT AMONG TECHNICAL PEOPLE HAVING
DIFFERENT SKILLS

COMMUNICATION SKILLS ARE NECESSARY FOR THOSE
RESPONSIBLE FOR MODEL DEVELOPMENT AND
APPLICATION

SKILL LEVELS HIGHER THAN 5 REQUIRE FORMAL
TRAINING, DEDICATED SELF INSTRUCTION, OR
EXPERIENCE WORKING WITH WELL-TRAINED PEOPLE

SKILL LEVELS OF 5  OR LESS CAN BE ATTAINED FROM SELF
INSTRUCTION, SHORT COURSES, OR EXPERIENCE

-------
^$fess>3
j^S^o
-V^CSssC
XO^S^->Q ,xV/ Q
^§§§|oS§
  Figure 8.2
  Statistical distribution of flow paths around local heterogeneities
  leads to dispersion. The process is shown here at a microscopic
  scale where pore space surrounds gravel-sized grains. (From R. A.
  Freeze and J. A. Cherry, Groundwater,©\919, p. 384. Reprinted by
  permission of Prentice-Hall, Inc., Englewood Cliffs, N. J.)

-------
                                              C(jt,0)/N
                                                        Longitudinal concentration
                                                                 profile
                 Uniform
                flow field
       Initial
    tracer slug
                                            Ellipse of variance
                                            due to dispersion
                                                                         C(0,y)
Lateral concentration
      profile
Figure 8.1
Dispersion of an instantaneous point source in a uniform flow field. The longitudinal
and lateral distributions of concentration in the ellipse are shown by the superimposed
graphs.

-------
                      v, = 0.1 m day'1, / = 400 days
Figure 8.6
One-dimensional longitudinal dispersion showing effect of changing the
value of the dispersion coefficient.

-------
                    RP2465-5
                                                           EPRI
       EPRI EA-4ttO
       RP2O5-S
       Final Report
       August 1M6
                  A Review of Field-Scale Physical Solute Transport
                  Processes in Saturated and Unsaturated Porous
                  Media
                    Contractor: Ttnnwto Vallty Authority
  I O.OOO
   1,000
r-   IOO
cr
UJ
in
5
      I0
       10
                           °   A
                        A

                                                A    A
                                           TRACER CWTOM
                                           TEST   EVENTS TRACERS
                                 FRACTURED
                                 MEDIA       °
                                 POROUS      m
                                 MEDIA
                   10
                              IOO        1000      10,000    100,000
                                SCALE  (m)
figure 2-1.   Sc*le of Observation Versus Longitudinal Dispersivity for the
Saturated Zone

-------
   1.0
   0.8 ¥-
.0

N

H  0.6
X
o
LU
X

LJ

r  0.4
LU
OC
   0.2
   0.0
           1   	CHALK RIVER SITE
           I   	 MOBILE SITE
           |          (RUN 1)
                     MOBILE SITE
                      (RUN 2)
                             I

                             L	
                                                 I
                     Y/////A -
     o.o
0.2      0.4       0.6       0.8

RELATIVE CONDUCTIVITY, K/Kmax
1.0

-------
0.2
                t=3 days
           y-z  PLANE
                                    ID
                                    •z.
                                    o

                                    »-
                                    o
                                    UJ
                                             0.2
                                       t = 8.53 days
                              0    5   10 m
x-z PLANE

-------
                               ' LEVEL  2
         300   400   500  600   200
                          TIME, hrs.
                                      300   400   500
UJ
§  1-'
   0.8

   0.6
  o
    0.0
          LEVEL 3
       200   400   600
                         ^LEVEL 4-
                          200   400

                          TIME, hrs.
                                            200   300
  §
_l Z
UJ UJ
oc o
    0.6
      .
     -4
  O  0.2
  O
     0.0
        - LEVEL 6
         100   200
                                    T  I
                               LEVEL  7
                     300            200
                         TIME,.hrs.
                         FIELD DATA

                         MODEL PREDICTION
                                       300   400   500

-------
   0.20
UJ
oc
   0.16
Z
o
ft  0.12
Z
UJ
o
Z
o
o

UJ  0.08
   0.04
    0.0
                                               o   EXPERIMENT

                                             	ADJUSTED K(Z)

                                             	UNADJUSTED K(Z)
                                                   a, =400 cm
                               12       16       20

                                     TIME, days
                                                        24
32

-------
                            BP-F-2-1
GROUNDWATER
  FLOW MODEL
(MATHEMATICAL)

-------
                                                  BP-F-2-7
 DIAGRAM OF THE MAJOR COMPOMENTS OF
   THE GROUNDWATER FLOW EQUATION
   WATER
   MASS
  BALANCE
  WATER
MOMENTUM
 BALANCE
GROUNDWATER
   FLOW
  EQUATION
  DARCY'S
 EQUATION

-------
         BASIC EQUATION
                                                BP-F-2-8
rate of
mass in
                 rate of
                 mass out
rate of mass
accumulation
ax
                              w = s
                                     ah
                                     at
       CONFINED, ARTESIAN FLOW

-------
                                            ••   a
                                           - SP-F-2-9
    BASIC EQUATION (CONT.)
PARTIAL DIFFERENTIAL EQUATION



SECOND ORDER



LINEAR



DIFFUSION EQUATION



DEPENDENT VARIABLE - h



INDEPENDENT VARIABLES - x,y,t



EQUATION PARAMETERS - T, W, S

-------
                                        BP-F-2-10
   EQUATION PARAMETERS
TRANSMISSIVITY, T
A MEASURE OF AN AQUIFER'S ABILITY TO TRANSMIT
WATER THROUGH ITS ENTIRE THICKNESS

STORAGE COEFFICIENT, S
A MEASURE OF AN AQUIFER'S ABILITY TO TAKE IN AND
RELEASE WATER FROM STORAGE

SOURCE/SINK TERM, W
RECHARGE OR DISCHARGE, SUCH AS PUMPAGE

-------
INDEPENDENT VARIABLES
  DISTANCE




      • x- AND y- DIRECTIONS




      • TWO-DIMENSIONAL




  TIME




      • TRANSIENT,!

-------
                                           BP-F-2-12
    MAJOR ASSUMPTIONS


> POROUS MEDIA

 DARCY'S LAW

> SLIGHTLY COMPRESSIBLE FLUID

 SMALL VERTICAL VARIATION IN PROPERTIES
 AND HEAD

 SINGLE AQUIFER WITH AREAL, CONFINED FLOW

 LINEAR AQUIFER VERTICAL COMPRESSIBILITY

 PRINCIPAL COMPONENTS OF THE TRANSMISSIVITY
 ALIGNED WITH COORDINATE AXES

-------
                           BP-F-3-1
    SOLUTE
  TRANSPORT
    MODEL
(MATHEMATICAL)

-------
                                                    BP-F-3-3
 DIAGRAM OF THE MAJOR COMPONENTS
 OF THE SOLUTE TRANSPORT EQUATION
   MASS
 BALANCE
FOR SPECIES
  WATER
MOMENTUM
 BALANCE
  SOLUTE
TRANSPORT
 EQUATION
 DARCY'S
 EQUATION

-------
                                        BP-F-3-S
 BASIC EQUATION (CONT.)
PARTIAL DIFFERENTIAL EQUATION
SECOND ORDER
LINEAR (CONSTANT DENSITY)
CONVECTION-DIFFUSION EQUATION
DEPENDENT VARIABLE - C
INDEPENDENT VARIABLES - x, y, t
EQUATION PARAMETERS - 0, D, q, RC*

-------
                                     BP-F-3-7
INDEPENDENT VARIABLES
   DISTANCE
       • x-, y-, AND z- DIRECTIONS
   TIME
       • TRANSIENT, t

-------
                                           BP-F-3-8 "
     EQUATION PARAMETERS
• POROSITY, J0f
  A MEASURE OF INTERSTITIAL SPACE CONTAINED
  IN THE ROCK. IT IS EXPRESSED AS A RATION OF
  VOID SPACE TO THE TOTAL (GROSS) VOLUME OF
  THE ROCK

• DARCY VELOCITY, q

• DISPERSION, D

• SOURCE/SINK TERM, RC*

-------
                    node
          /*/• /•/•
123 4
b
T

-------
                             or- r-3-
+ Q3i + Q4i + Qsi = Ax, Ay, S,
                         ah,

                         lit7
               3h
    =Ax, T2, ( —)21
               3y
1* Ax,T21
              h2 - h
              - -
                Ay
               ! = — Ax2 —
            ! - hi
     at      At
               Ax2

        '•J = T "A? v"'

-------
            NODE POINT
>















^r -».
/
\
V
Ay
f
^
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
*. 	 ^
    Ax
MESH  CENTERED
                               I
              Ay
    Ax
BLOCK CENTERED

-------
                               >'->' - <.' - 12 f-
No Matter Which Numerical
Approach (F.D., F.E., or M.O.C.) is
Chosen, the End Result is a System
of Linear Algebraic Equations that
Must be Solved.
   For Steady-State—One Solution
   For Transient—Repeated Solutions

-------
Matrix Solution Important to User

   • Most Time-Consuming Part of
Computer Calculations
   • Controls In-Core Storage
Requirements
   • Optimal Choice of Method is
Problem Dependent

-------
Iterative Methods—Use

Convergence—How Many Guesses
Required? Does it Converge? When
has it Converged?

Improving Speed of Convergence—
Acceleration and Other Parameters
   • Problem Dependent
   • Must be Specified

-------
        CONVERGENCE
1  2 3 4  5  6 7 8  9 1011 12 13 14
       ITERATION NUMBER

-------
                                       X- --2
BOUNDARY CONDITIONS
Map
View
             t
          Aquifer
          Boundary
The Values of the
Required Solution of
the Problem on the
Aquifer Boundary

-------
                                             BP-F-2-2S
  BOUNDARY CONDITIONS (CONT.)

SPECIFIED VALUE
  • CONSTANT HEAD AT A LARGE LAKE,
   DISTANT BOUNDARY
SPECIFIED FLUX
  • WELL
  • IMPERMEABLE ROCK
  • GROUNDWATER DIVIDE
VALUE-DEPENDENT FLUX
  • HEAD-DEPENDENT LEAKAGE FROM
   A CONFINING BED, RIVER

-------
Constant
Head Boundary
Recharge
1 1 ill
          Confining
        1||||||| Artesian 1
          Illlll Aquifer |
                                    $v No-Flow
                                    ^Boundary
                                    . ••»-.. i -

-------
                                              BP-F-3-18
         INITIAL CONDITIONS
CONCENTRATIONS ARE SPECIFIED AT THE BEGINNING
OF THE TIME PERIOD INSIDE THE REGION OF INTEREST

-------
                                                                   BP-F-1-10
                            DETERMINE NECESSITY
                            OF NUMERICAL MODEL
                            COMPILE & INTERPRET
                              AVAILABLE DATA
                             COLLECT DATA AND
                              OBSERVE SYSTEM
                    CONCEPTUALIZATION
      PREPARE DATA FOR
        MODEL USING
    ESTIMATED PARAMETERS
      INTERPRET RESULTS
IMPROVE
COMPEPTUAL
MODEL
GOOD
COMPARISON
            RESULTS
            SATISFACTORY
                            SENSITIVITY RUNS
                          IS MORE DATA NEEDED?
                               PREDICTIVE
                               PREDICTIVE
                             SCENARIO RUNS
                            SENSITIVITY RUNS
                         HISTORY
                         MATCHING
                         (FIELD PROBLEM)
                PREPARE DATA FOR
                  MODEL USING
              ESTIMATED PARAMETERS
COMPARE RESULTS WITH
    OBSERVED DATA
                         POOR
                         COMPARISON
                                                          YES

-------
                                BP-F-1-18
MODEL APPLICATION
MODEL APPLICATION HAS THREE
MAIN PHASES:
(1) SYSTEM CONCEPTUALIZATION;
(2) HISTORY MATCHING OR MODEL
   CALIBRATION;
(3) PREDICTION.

MOST APPLICATIONS INVOLVE EACH
OF THE THREE PHASES, BUT TO
DIFFERENT DEGREES OF EFFORT.

-------
                                   BP-F-1-20
MODEL APPLICATION (CONT.)
    SYSTEM CONCEPTUALIZATION
    INVOLVES ORGANIZING AVAILABLE
    INFORMATION ON THE GROUND-
    WATER SYSTEM IN AN INTERNALLY
    CONSISTENT FRAMEWORK. THE
    CONCEPTUALIZATION PRODUCES
    FACTORS CONTROLLING THE FLOW
    SYSTEM SUCH AS STRATIGRAPHY
    AND GEOMETRY, BOUNDARY AND
    INITIAL CONDITIONS, AND
    HYDROLOGIC PARAMETERS.

    SYSTEM CONCEPTUALIZATION IS
    SUBJECTIVE AND QUANTITATIVE
    STANDARDS OF PERFORMANCE ARE
    GENERALLY NOT AVAILABLE.

-------
                                   BP-F-1-22
MODEL APPLICATION (CONT.)
    HISTORY MATCHING OR MODEL
    CALIBRATION IS USED TO REFINE
    ESTIMATES OF HYDROLOGIC
    PARAMETERS AND BOUNDARY
    CONDITIONS BY COMPARING MODEL
    RESULTS WITH OBSERVED DATA.

    THE HISTORY MATCHING PROCEDURE
    CAN BE PERFORMED BY EITHER TRIAL
    AND ERROR OR BY AUTOMATIC
    REGRESSION. FOR BOTH, SENSITIVITY
    ANALYSIS IS PART OF THE MATCHING
    PROCESS.

    FOR HISTORY MATCHING,
    PERFORMANCE CAN BE JUDGED BY
    HOW WELL COMPUTED VARIABLES
    (E.G., HYDRAULIC HEAD) COMPARE TO
    MEASURED VALUES.

-------
MODEL APPLICATION (CONTINUED)
     f  START J
f START  J
mates
Update model
„. parameter esti
1
initial parameter estimates,
model specification
-*• RUN MODEL
1
COMF
CALCUI
ad & OBSE
VALI
<
S
initial parameter
, estimates
MODEL
SPECIFICATION

*~ **• —
1 if
§ II
«' en ^*-
.ATED "°
ERVED |_
\


RUN MODEL
\

CALCULATE
CRITERIA
yes
JES
good

f STOP )
X_ _/ trial and error



converg

ARE RESULTS
ACCEPTABLE
\
yes


I
1
1
1
1
1
1
1
ence? J
no

t STOP J
y J automatic
   PROCEDURES FOR MODEL CALIBRATION USING
   TRIAL AND ERROR AND AUTOMATIC HISTORY
   MATCHING APPROACHES

-------
                                   BP-F-1-25
MODEL APPLICATION (CONT.)
    PREDICTION IS USUALLY THE FINAL
    AND SHORTEST PORTION OF A MODEL
    APPLICATION. PREDICTIONS ARE
    BASED ON MODEL RESULTS USING
    THE BEST ESTIMATE OF SYSTEM
    PARAMETERS OBTAINED BY HISTORY
    MATCHING AND FORECASTING THE
    EFFECTS OF A SYSTEM CHANGE.

    AS WITH SYSTEM CONCEPTUAL-
    IZATION, PREDICTION IS SUBJECTIVE
    AND QUANTITATIVE STANDARDS OF
    PERFORMANCE ARE GENERALLY NOT
    AVAILABLE.

-------
    FEATURES OF METHOD
  OF CHARACTERISTICS CODE
•   2 Dimensional (areal)
   Flow and solute transport
   Steady state and transient
   Finite-difference for flow
   Method of characteristics for
   transport
   Rectangular block centered FD grid
   Variety of boundary conditions
   Adapted for decay, adsorption
   Thoroughly documented (1978)
   Extensively tested and used
   Adapted for PC use
   Preprocessor
                             GeoT"
                                Irans, inc.
                                CTOumimuu SPECIALISTS

-------
         DATA REQUIREMENTS
       •  Grid data
       •  Time data
       •  Execution parameters
       •  Output options
       •  Porosity
       •  Storage coefficient
       •  Dispersivity
       •  Transmissivity*
       •  Saturated thickness*
       •  Diffuse recharge*
       •  Flux boundary data
       •  Leakance*
       •  Initial heads*
       •  Initial concentrations*
*arrays (data sets)
                                  GeoT
                                     In
rans,mc.
iKXKDWxm mcuum

-------
      OUTPUT
Echo of input data
Computed head distribution
Flow mass balance
X and Y velocities
Stability criteria
Computed concentration
distribution
Chemical mass balance
Observation well data
                       GeoT
                           In
 rans,mc.
QMXMDVXm BfCIALBTS

-------
                             Summary of MOC  Updates
05/16/79      ijiprove accuracy (source/sink representation)



08/26/81      variable length pumping period



10/12/83      double precision; convenient printout; output routine calls



06/10/85      increase number of particles per node (16)



07/26/85      improve calculating dispersive concentration changes



08/02/85      first order decay; linear equilibrium adsorption



08/08/85      smaller transport subgrid



08/12/85      improve output readability



07/02/86      improve particle-velocity calculations



10/20/86      minimize zero divide check errors



03/02/87      mass balance calculations in double precision



03/05/87      add SIP solution routine



05/15/87      ijiprove efficiency



01/29/88      ijiprove operational aspects

-------
0  CUMULATIVE MASS BALANCE -- (IN FT**3)
    RECHARGE AND INJECTION
    PUMPAGE AND E-T WITHDRAWAL
    CUMULATIVE  NET  PUMPAGE
    WATER RELEASE FROM STORAGE
    LEAKAGE INTO AQUIFER
    LEAKAGE OUT OF AQUIFER
    CUMULATIVE  NET  LEAKAGE
        MASS BALANCE RESIDUAL
        ERROR  (AS PERCENT)
-.63115E+09
 .37869E+09
-.25246E+09
 .OOOOOE+00
 .54938E+09
-.80194E+09
-.25257E+09
-.10783E+06
-.91339E-02
0  RATE MASS BALANCE -- (IN C.F.S.)
    LEAKAGE INTO AQUIFER
    LEAKAGE OUT OF AQUIFER
    NET LEAKAGE    (QNET)
    RECHARGE AND INJECTION
    PUMPAGE AND E-T WITHDRAWAL =
    NET WITHDRAWAL   (TPUM)
 .87043E+00
-.12706E+01
-.40017E+00
-.10000E+01
 .60000E+00
-.40000E+00

-------
CHEMICAL MASS BALANCE

       MASS IN BOUNDARIES     =   .54938E+10
       MASS OUT BOUNDARIES    =  -.84234E+10
       MASS PUMPED IN         =   .63115E+12
       MASS PUMPED OUT        =  -.29071E+11
       MASS LOST BY DECAY     =   .OOOOOE+00
       MASS ADSORBED ON SOLIDS=   .OOOOOE+00
       INITIAL MASS ADSORBED  =   .OOOOOE+00
       INFLOW MINUS OUTFLOW   =   .59915E+12
       INITIAL MASS DISSOLVED =   .26880E+11
       PRESENT MASS DISSOLVED =   .62728E+12
       CHANGE MASS DISSOLVED  =   .60040E+12
       CHANGE TOTL.MASS STORED=   .60040E+12
     COMPARE RESIDUAL WITH NET FLUX AND MASS ACCUMULATION:
       MASS BALANCE RESIDUAL  =  -.12501E+10
       ERROR  (AS PERCENT)    =  -.19636E+00

-------
                                       •      •
                                          B
Figure  9.—Parts of  finite-difference  grids  showing
  the initial geometry of particle  distribution for  the
  specification  of four  (A), five (B),  eight  (C), and
  nine (D) particles per cell.

-------
            EXPLANATION
    •  Initial location of particle
    O  New location of particle
   —^ Flow  line and direction of flow
   	Computed path  of particle

Figure 1.—Part  of hypothetical finite-
  difference grid showing  relation of
  flow field  to  movement of points.

-------
   ./-I./
                                     /'1.7-1
      •
      o
           EXPLANATION
Node o( finite-difference grid
Location of particle p
Xor Y component of velocity

Area of influence for interpolating velocity
  in X  direction at particle p

Area of influence for interpolating velocity
  in Y  direction at particle p
Figure  2.—Part  of  hypothetical  finite-difference  grid
  showing  areas over (which bilinear  interpolation  is
  used to compute  the  velocity  at  a point.  Note that
  each area  of  influence Is equal  to  one-half of the
  nr«n of a mil

-------
                        or 7/wE-SW/»
           -f
                  or
       TINJT
FLOW-
           * 77MX}


         J fcr //*«/  //

-------
C START ^J
1


     READ GEOLOGIC.
      HYDROLOGIC.&
       CHEMICAL
         INPUT
         DATA
          I
   GENERATE UNIFORM
     DISTRIBUTION OF
    TRACER PARTICLES
  COMPUTE HYDRAULIC
    GRADIENTS FOR
     ONE TIME  STEP
  COMPUTE  DISPERSION
 EQUATION COEFFICIENTS
       COMPUTE
    GROUND-WATER
      VELOCITIES
    DETERMINE LENGTH
    OF TIME INCREMENT
      FOR EXPLICIT
     CALCULATIONS
          I
     MOVE PARTICLES
          I
 GENERATE NEW PARTICLES
     OR REMOVE OLD
      PARTICLES AT
APPROPRIATE BOUNDARIES
          1
    COMPUTE AVERAGE
 CONCENTRATION IN EACH
  FINITE-DIFFERENCE CELL
           I
   COMPUTE EXPLICITLY
      THE CHEMICAL
   CONCENTRATION AT
         NODES
          i
ADJUST CONCENTRATION
   OF EACH PARTICLE
          i
       COMPUTE
    MASS BALANCE
        END OF
      TIME STEP?
      SUMMARIZE AND
      PRINT RESULTS
        END OF
       PUMPING
        PERIOD
                     YES ^ £ND OF
                           SIMULATION?
Figure 8.—Simplified flow chart illustrating the major steps  In the  calculation
                            procedure.

-------
  o
  A
         EXPLANATION
Node of finite-difference grid
Previous location of particle p
Computed new location of particle p
Corrected new location of particle p
Flow line and direction of flow
Computed path of flow

Zero transmissivity  (or no-flow boundary)
Figure 4.—Possible movement ol particles near
     an impermeable (no-flow)  boundary.

-------
                              AT START CX MIT
                             PARTICLE MOVEMENT
                                                                                       DID f AITICIC
                                                                                    O«ICINAU IN THAT
                                                                                       SOURCE CEU '
                                                         It SOUtCf CEU
                                                      OCA HO AlOMC EOCf
                                                         of
                                                            FJ770
   OCIMMIXC I -T
COOROINAHS (X MOOf
  •Him pAnnciE it
    toc*uorM0.SIIION
                                                                                                               At OLD rAMlCLl IN
                                                                                                                   NEW CELL
                                                                                                                       f J420-M40
                                IS OLD
                             OCATiON IN A
                            SOURCE CELL OR
                             A SlN« CUL '
                                J480-JS7
                                                          MAI PARtlCLC
                                                          CMANCCO CELL
                                                            IOCAIION7
                                                                                                              CMAIE NEW PAHTlClf
                                                                                                                        f IS«0
          I AND T
COO*OlKATlS OF CELL
Al NEW LOCATION Of
  PARTICLE rioM.ll!0
                            COMPUTi OISTANCC
                             MOVtO IN I AND »
                                OIKtCIIONJ
                                                      1ACE NEW PARTICLE
                                                     AT ORIGINAL LOCATION
                                                        Of OLD PARTICLI
      S Nf
    LOCAIION
   IN A PUMTINC
OR CONSTANT -MCAO
                            COUfUTt OlilANCt
                                    TlUVtllO
                             MTONO MXJNOAIIT
                                                                                     MJUOv: PARTICLE
                                                                                       FROM CRIO
                                                                                             FM4O-J710
                          RELOCATE PARTICLE INTO
                           AQUIFER (T REFLECTION
                             ACROSS IOUNOART
IES

CALL iUMOUTINE
CNCON TO COMPUTE
NEW CONCENTRATION!
F1JM


STO*f OMEHVATION WELL
MI" F4000-40.0


    Flgur* 10.—Generalized How chert of lubroutlrw MOVE. Numbere indicate line numbers where the operation Is executed.

-------
                              Summary of M3C Updates
05/16/79      inprove accuracy (source/sink representation)



08/26/81      variable length pumping period



10/12/83      double precision; convenient printout; output routine calls



06/10/85      increase number of particles per node (16)



07/26/85      improve calculating dispersive concentration changes



08/02/85      first order decay; linear equilibrium adsorption



08/08/85      smaller transport subgrid



08/12/85      improve output readability



07/02/86      improve particle velocity calculations



10/20/86      minimize zero divide check errors



03/02/87      mass balance calculations in double precision



03/05/87      add SIP solution routine



05/15/87      improve efficiency



01/29/88      iinprove operational aspects

-------
         LIST OF SUBROUTINES FOR SOLUTE-TRANSPORT MODEL


  NAME                                 PURPOSE

MAIN    	  Control Execution

PARLOD  	  Data Input and Initialization

ITERAT	  Compute Head Distribution
-*
GENPT   	  Generate or Reposition Particles

VELO    	  Compute Hydraulic Gradients, Velocities,
                  Dispersion Equation Coefficients, and
                  Time Increment for Stable Solution to
                  Transport Equation

MOVE    	  Move Particles

CNCON   	  Compute Change in Chemical  Concentrations
                  and Compute Mass Balance for Transport
                  Model

OUTPT   	  Print  Head Distribution and Compute Mass
                  Balance for Flow Model

CHMOT   	   Print  Concentrations, Chemical Mass Balance,
                  and Observation Well Data

-------
     SECTION 1
EQUATION DERIVATION

-------
  DERIVATION OF EQUATIONS DESCRIBING
  SOLUTE TRANSPORT IN GROUND WATER

111
  U. S. GEOLOGICAL SURVEY
  Water-Resources Investigations 77-19

-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
3. Recipient's Accession No.
 I. Title and Subtitle

  DERIVATION OF EQUATIONS DESCRIBING SOLUTE  TRANSPORT  IN
  GROUND WATER
                                                 5. Report D*te
                                                   April 1977
7. Author(s)
  Leonard  F.  Konikow  and David  B.  Grove
                                                 8. Performing Organization Rept.
                                                   No-  USGS/WRI-77-19
9. Performing Organization Name and Address   U.S. Geological Survey
                                    Water Resources Division
                                    Mail Stop  413, Box  25046
                                    Denver Federal Center
  	Denver,  Colorado 80225
                                                 10. Project/Task/Work Unit No.
                                                 II. Contract/Grant No.
 2. Sponsoring Organization Name and Address
  Same as  9 above
                                                 13. Type of Report & Period
                                                    Covered
                                                           Final
                                                                      U.
IS. Supplementary Notes
 16. Abstracts
       A  general equation describing the  three-dimensional transport and  dispersion  of
  a reacting solute  in flowing ground water is derived from the principle of conserva-
  tion of mass.  The derivation presented in this  report is more detailed but less
  rigorous than derivations published previously.   The general  solute-transport equation
  relates concentration changes to hydrodynamic dispersion, convective  transport, fluid
  sources and sinks, and chemical reactions.  Because both dispersion and convective
  transport depend on the velocity of ground-water flow, the  solute-transport equation
  must be solved in  conjunction with the  ground-water flow equation.
17. Key Words and Document Analysis.  17o. Descriptors
  Groundwater, Dispersion,  Computer models, Path  of pollutants
 I7b. Identifiers/Open-Ended Terms
   Solute-transport models
 17c. COSATI Field.'Group
 18. Availability Statement
  No restriction on distribution
                                      19. Security Class (This
                                         Report)
                                           UNCLASSIFIED
                                                           20. Security Class (This
                                                              Pace
                                                                TJNCLASSIFIED
           21. No. of Pages
                35
                                                            22. Price
 FORM NTis-39 i«ev. 10-73)  ENDORSED BY ANSI AND UNESCO.
                                                     THIS FORM MAY BE REPRODUCED
                                                                                 USCOMM-DC S2«8'PT4

-------
DERIVATION OF EQUATIONS DESCRIBING SOLUTE TRANSPORT




IN GROUND WATER




By Leonard F. Konikow and David B. Grove









U.S. GEOLOGICAL SURVEY






Water-Resources Investigations 77-19
                              April 1977
                           Revised Jan.  1984

-------
                UNITED STATES DEPARTMENT OF THE INTERIOR

                       CECIL D.  ANDRUS,  Secretary

                           GEOLOGICAL SURVEY

                       V.  E. McKelvey, Director
For additional information .write to:

U.S. Geological Survey
Water Resources Division
Mail Stop 413, Box 25046
Denver Federal Center
Denver, Colorado  80225

-------
                                CONTENTS
                                                                       Page

Abstract	   1
Introduction	•	   2
Ground-water flow	   3
     General flow equation	   3
     Equations of state	   4
     Flow velocity	   5
     Simplifying assumptions	   6
          Homogeneous fluid properties	   6
          Two-dimensional areal flow	   7
          Alignment of coordinate axes	   9
Solute-transport equation	  10
     Derivation of general transport equation	  10
     Two-dimensional areal solute transport	  16
     Dispersion coefficient	  24
Summary	  28
References cited	  29
                              ILLUSTRATIONS
                                                                       Page

Figure 1.  Representative elementary volume (REV) of aquifer	  11
       2.  Representative volume of aquifer having variable
             saturated thickness	*•	•  17
                                    iii

-------
    DERIVATION OF EQUATIONS DESCRIBING SOLUTE TRANSPORT IN GROUND WATER
                By Leonard F. Konikow and David B.  Grove
                                ABSTRACT

     A general equation describing the three-dimensional transport and
dispersion of a reacting solute in flowing ground water is derived from
the principle of conservation of mass.  The derivation presented in this
report is more detailed but less rigorous than derivations published
previously.  The general solute-transport equation relates concentration
changes to hydrodynamic dispersion, convective transport, fluid sources
and sinks, and chemical reactions.  Because both dispersion and convective
transport depend on the velocity of ground-water flow, the solute-transport
equation must be solved in conjunction with the ground-water flow equation.

-------
                               INTRODUCTION

     In recent years there has been an increased awareness of problems of
ground-water contamination.  Reliable predictions of contaminant movement
can only be made If we understand and can quantitatively describe the
physical and chemical processes that control solute transport in flowing
ground water.  Several reports have been published recently that develop
and present solute-transport equations to compute the concentration of a
dissolved chemical species in ground water as a function of space and
time.  Examples of these reports include Reddell and Sunada (1970), Bear
(1972), and Bredehoeft and Finder (1973).  The three main processes
affecting solute transport, and consequently chemical concentrations, are
convective transport, hydrodynamic dispersion (Including diffusion and
mechanical dispersion), and chemical reactions.  Because convective trans-
port and hydrodynamic dispersion depend on the velocity of ground-water
flow, the solute-transport equation must be considered in conjunction with
the ground-water flow equation.

     Aquifers generally have heterogeneous properties and complex boundary
conditions.  Therefore, the solution of the partial differential equations
that describe the solute-transport processes generally require the use of
a deterministic, distributed parameter, digital simulation model.  Among
the reports that describe or present numerical models to solve the solute-
transport equations are Reddell and Sunada (1970), Bredehoeft and Finder
(1973), Finder (1973), Ahlstrom and Baca (1974), Gupta and others  (1975),
Grove (1976), and Lantz and others (1976).  Furthermore, several documented
case histories show that where adequate hydrogeologic data are available,
solute-transport models can be used to compute the rates and directions of
spreading of contaminants from known or projected sources.  Examples of
model applications to field problems include Konikow and Bredehoeft  (1974),
Robertson (1974), Robson (1974), Konikow (1976), and Segol and Finder
(1976).

     These models use either finite-difference methods, finite-element
methods, or the method of characteristics.  The selection of the "best"
numerical method depends largely on the nature of the specific field
problem, but also depends to some extent on the mathematical background
of the analyst.  Although solute-transport models are best utilized when
the analyst is thoroughly familiar both with the equations and with the
numerical algorithm, the increasing availability of documented and published
programs affords the opportunity for the use of a model by persons with
only minimal familiarity with both.

     The basic purpose of this report is to derive a general form of the
solute-transport equation from general principles in a more detailed,
step-by-step, but less rigorous manner than has been done in previously
published literature.  The report is Intended to serve as an introduction
to quantitative modeling of solute-transport processes in ground water.
It will also show how the general solute-transport equation can be modified
or simplified for application to a variety of different types of field
problems.  Because of the interrelation between the flow equation and the
solute-transport equation, the former will also be presented in some

-------
detail although not  specifically derived.   It Is assumed that  the mathemat-
_ical background of the reader includes at  least a familiarity  with partial
differential equations.


                            GROUND-WATER  FLOW

                          General flow equation

     A quantitative  description of ground-water flow is a prerequisite to
accurately representing solute transport in aquifers.   A general form of
the equation describing the transient flow of a compressible fluid in a
nonhomogeneous anisotropic aquifer may be  derived by combining Darcy's Law
with the continuity  equation.  By following the developments of Cooper
(1966) and of Bredehoeft and Finder (1973), the general flow equation may
be written in cartesian tensor notation as:
^h* fe+pg^)J-pa"
                                                  3P
                                            8   dm.
                                       *-  Z  ^r + 'v*           <»
                                                               2
where k. .  is the  intrinsic permeability  (a second-order tensor), L ;
       •z-j
      p   is the  fluid density, ML~ ;
is the dynamic viscosity, ML  T  ;


                                            -2
                                  -1  -2
      P   is the  fluid pressure,  ML T
      g   is the  gravitational acceleration constant, LT
      z*  is the  elevation of the reference point above a standard
            datum, L;
      W*  = W*(x,y,z,t) is the volume  flux per unit volume (positive

            sign  for outflow and negative for inflow), T  ;
                                                    « Q
      P*   is the density of the source/sink fluid, ML~ ;
      a    is the vertical compressibility coefficient of the medium,

             LM"1!2;
      p    is the fluid density at a reference pressure, temperature, and

             and  concentration, ML  ;
      t    is the effective porosity (dimensionless);
                                                           -1  2
      6    is the compressibility coefficient of the fluid,  LM  T ;

-------
                                                  3
      VQ   is the reference volume of the fluid, L ;

      m.   is the mass of species i in the reference volume v , M;
       7>                                                     0
      e    is the number of species, (dimensionless);
      x.   are the cartesian coordinates, L; and
       ts
      t    is time, T.

     The summation convention of Cartesian tensor analysis is implied in
equation 1.  That is, each term is summed over the range of its subscripts.
Bredehoeft and Finder (1973) note that the derivation of equation 1 is
based on the following assumptions:

     1.  The porous medium may only deform vertically.
     2.  Isothermal conditions prevail.
     3.  The volume of individual grains remains constant during deforma-
           tion of the medium.
     4.  Fluid density is a linear function of pressure and concentration,
           as indicated by the following relationship:
                                         v~
                                          0
                                             -1 -2
where P   is the reference fluid pressure, ML  T  ; and
      m.  is the mass of species i in the reference volume v  at the
            reference pressure, M.

     5.  The permeability is independent of pressure, temperature, and
           concentration.
     6.  There is no change in volume caused by mixing fluids of different
           ionic concentrations.
     7.  The proportionality constants a and 3 are independent of pressure
           and concentration.
     8.  Hydraulic head gradients are the only significant driving
           mechanism.
     9.  The vertical velocity of grains is negligible.
                           Equations of state

     The density and viscosity of ground water are both related to its
temperature, pressure, and chemical content.  Because isothermal condi-
tions have been assumed, temperature changes need not be considered.

     Equation 2 expresses the dependence of density on both the pressure
and the mass concentrations of all species.  Equation 2 may be rewritten

-------
in terms of the concentration of a single chemical species of Interest
as:

                    P - PQ + P0B(P-P0) + Y(C-Co)                        (3)

where C  is the mass concentration per unit volume of solution for the

            solute species of interest, ML  ;
      C  is the concentration of the solute at the reference pressure
       0                       -3
            and temperature, ML  ; and
      Y  is the constant of proportionality between concentration and
            fluid density (dimensionless).

     If the relationship indicated by equation 3 is substituted for
equation 2, then equation 1 may be rewritten as:
                                         9P        3P
                                    + ey     + W*p*                     (4)
     Viscosity may be similarly expressed as a linear function of con-
centration by the following:

                        y - U0 + A(C - CQ)                              (5)


where y  is the dynamic viscosity of the fluid at the reference pressure,
                                             -1 -1
           temperature, and concentration, ML  T  ;  and
      A  is the constant of proportionality between concentration and
                       2-1
           viscosity, L T  .

                               Flow velocity

     The seepage velocity, or average interstitial velocity,  of ground-
water flow may be computed as:


                                v*-r

where V. is the seepage velocity in the direction of x.,  LT  ;  and
       1,                                              T*
      q. is the specific discharge, or specific flux, in  the  direction
       ^            _i
           of x., LT  .
               I*

     The specific discharge may be computed directly from Darcy's Law,
which is written as:

-------

                          Simplifying assumptions

     The general flow equation written as equation 4 can be simplified
considerably if certain conditions can be satisfied.  Several of these are
described next.

Homogeneous fluid properties

     When changes in concentrations of dissolved chemical species are
relatively small, the fluid density and viscosity remain essentially
constant.  This assumption of homogeneous fluid properties both simplifies
the flow equation and allows it to be solved independently from the solute
transport equation.  If density is independent of concentration, then
the third term on the right side of equation 4 can be eliminated from the
flow equation.

     We may aim next to express equation 4 in terms of hydraulic head
rather than pressure.  Following the development of Hubbert  (1940), we may
define the hydraulic head as:
where h  is the hydraulic head, L; and
                                    -1 -2
      P  is atmospheric pressure, ML  T  .

     If we differentiate equation 8 with respect to x., for constant
density we obtain:


                           3h  m 3z*     1_  3P
                           3x. " 3x.     pg  3x.
                             Tf     If           Tr


                       3P
Solving equation 9 for -r— yields:
                             3P       /3h     3z* \
                             ax. "  P8l 3x. "  3x. I                       (10)
                              i      \  i     i /

-------
     We may similarly differentiate equation 8 with respect to time to
obtain:
                                                                      ui;
                           at    at    pg  at
Because -r— = 0,  we can express equation 11 as:
        ot •

                              3P       3h
     If we next substitute  the  relations indicated by equations 10 and  12
into equation 4,  and then divide both sides of the equation by the constant
density, we obtain:
     Equation 13 may be further  reduced if we consider that
where K.. is the hydraulic  conductivity  tensor, LT  , and that
       '
                            8  -  g(pa +  p  eg)                          (15)
                             s           o


where S  is the specific storage,  L
       S

     By substituting equations  14  and 15 into equation 13, we obtain:



                               %)  -.'.£*«•


Two-dimensional areal flow

     In many ground-water studies  it can be reasonably assumed that ground-
water flow is areally two-dimensional.   This allows the three-dimensional
flow equation to be reduced to  the case  of two-dimensional areal flow, for
which several additional simplifications are possible.  The advantages of
reducing the dimensionality of  the equations include less stringent data
requirements, smaller computer  storage requirements, and shorter computer
execution times to achieve a numerical solution.

-------
     An expression similar to equation 16 may be derived for the two-
dimensional areal flow of a homogeneous fluid and written as:
                      _**  f If  H — »-"- 1 • C V\ Y_- J
                     3x. l ij  3x. I    a  3t
                       * \  "    J/

where b is the saturated thickness of the aquifer, L, and it is assumed
that the hydraulic conductivity, specific storage, and hydraulic head
represent vertically integrated mean values (Cooley, 1974).

     The transmissivity of the aquifer may be defined as:

                             1.. = K.. b                                (18)

                                  2 -1
where T.. is the transmissivity, L T  .
       tj

     Similarly, the storage coefficient of the aquifer may be defined as:

                              S = SB b                                  (19)

where S is the storage coefficient  (dimensionless).

     After substituting the relationships indicated by equations 18 and 19
into equation 17, we obtain:


                      •sr- (t;* ^r\ " s IT + w                        (20)
where W = W(x,y,t) » W*b is the volume flux per unit area, LT~  .

     Although fluid sources and sinks may vary in space and time, they have .
been lumped into one term  (W) in the previous development.  There are
several possible ways to compute W.  If we consider only sources and sinks
such as (1) direct withdrawal or recharge, such as pumpage from a well,
well injection, or evapotranspiration, and (2) steady-state leakage into or
out of the aquifer through a confining layer, streambed, or lake bed, then
for the case of two-dimensional horizontal flow, the source/sink terms may
be specifically expressed  as:
                                        K
                  W(x,y,t) - Q(x,y,t) - -* (H  - h)                      (21)
                                        Ul    0

where Q  is the rate of withdrawal  (positive sign) or recharge  (negative

           sign), I/T1;
      K  is the vertical hydraulic  conductivity of the confining layer,
                                     -1
           streambed, or lake bed,  LT   ;
                                        8

-------
      m  is the thickness of the confining layer, streambed,  or lake
           bed, L; and
      H  is the hydraulic head in the source bed, stream,  or  lake,  L.
       s

Alignment of coordinate axes

     The cross-product terms of the permeability tensor drop  out when
 the coordinate axes are aligned with the principal axes of the tensor
(Bredehoeft, 1969); that is, k.. = 0 when i t j.  Therefore,  the only
                              tj
permeability terms with possible nonzero values are k  , k  ,  and k
r          •>                                         xx   yy        zz
This assumption simplifies the general flow equation, which can now be
rewritten to include all permeability terms as:
     Darcy's Law may be written similarly for the three flow directions
as:

                            k

                     qx = "i  \ 3x •  H6 3x /                        (23a)


                            kyy /3P
                     qy ~ ~  v  \*y ' "B 3y /                        (23b)

                            k
                             zz
                     qz = '  p  V 3z '  "B 3z  /                        (23c)
                                 /3P^    3z*
                                 (-37 + P8 -3
     For the case of two-dimensional areal flow, if the coordinate axes
are aligned with the principal directions of the transmissivity tensor,
equation 20 may be written as:
                    T       +  -T       =s.
                3x   xx Zx    3y   yy ty    & 3t * w

-------
                        SOLUTE-TRANSPORT EQUATION

                Derivation of general transport equation

     An equation describing the three-dimensional transport and dispersion
of a reacting dissolved chemical in flowing ground water will be derived
from the principle of conservation of mass (continuity condition).  The
derivation presented here is based on the developments of Reddell and
Sunada (1970), Bear (1972), and Bredehoeft and Finder (1973).

     The principle of conservation of mass requires that the net mass of
solute entering or leaving a specified volume of aquifer during a given
time interval must equal the accumulation or loss of mass stored in that
volume during the interval.  This may be expressed in a verbal equation as:

     (Rate of Solute Accumulation)

              = (Rate of Solute Inflow) - (Rate of Solute Outflow)

                + (Rate of Chemical Production by Reactions)            (25)
     This relationship may then be expressed mathematically by considering
all fluxes into and out of a representative elementary volume (REV), as
described by Bear (1972, p. 19).  The REV shown in figure 1 is centered
at coordinates (x, y, z) and has dimensions (Ax, Ay, Az).

     First we will determine fluxes in the x-direction.  We know that at
the center of the REV the mass flux of solute in the x-direction across the
y-z plane (face IJKL) is equal to (CV * eAyAz), where C is the concentration

of the solute (ML  ), V * is the instantaneous mass velocity of the solute

(LT~ ), and Ay and Az are the dimensions (L) of face IJKL.  Note that the
term (eAyAz) simply represents the total effective cross-sectional area
through which flow is occurring.

     We need expressions for the fluxes across the outer faces ABCD and
EFGH.  The difference between the mass flux of solute across face ABCD and
face IJKL equals the rate of change of mass flux in the x-direction times
the distance between these two faces.  The rate of change of the mass flux
                                    3
of solute in the x-dlrection equals -5— (CV * eAyAz), and the distance from
                                    OX    X
the center to face ABCD is (-Ax/2).  Thus, the mass flux through face ABCD
equals


                     CV * eAyAz - |- (CV * eAyAz )-^
                       X          oX    X         f-

Similarly, the distance from the center to face EFGH is (+Ax/2) and the
mass flux through face EFGH is given by
                                      10

-------
Figure 1.—Representative elementary volume  (REV) of aquifer.
                              11

-------
                          eAyAz +    (CV* eAyAz
     The net mass flux of the solute entering or leaving the REV in the
x-direction equals the difference (input-output) between the two previous
terms:


          (Net Mass Flux)  = fcv * eAyAz - |- (CV * eAyAz )|^
                         x   L  x          «x    x        i


                                   * eAyAz +
                                                                       (26)
This equation can also be derived by writing a Taylor series for the mass
flux term about the point (x, y, z) (Reddell and Sunada, 1970, p. 39).
Note that if the value of the derivative in the x-direction (equation 26)
is negative, then the mass flux of solute entering through face ABCD is
greater than the mass leaving through face EFGH.  If all other fluxes
balance, there will then be a positive accumulation of solute mass in the
REV.  Conversely, if the derivative is positive, there will be a decrease
over time of solute mass stored in the REV.

     It can similarly be shown that for the y-direction:


                  (Net Mass Flux)  - - J-  (CV * eAxAz)Ay              (27a)


and for the z-direction:


                  (Net Mass Flux)  - - j-  (CV * eAxAy)Az              (27b)
                                 z     dz    z


     Solute may also enter or leave the REV as a-flux through a source or
sink (W).  This may be expressed mathematically as:
                  (Source/Sink Mass Flux)REy - C'W*AxAyAz                (28)


where C' is  the concentration of  the  solute in  the source or  sink

           fluid, ML~3.
                                     12

-------
     If a sink (withdrawal) is considered positive in sign and a source
(recharge or injection) is considered negative in sign, then when all other
fluxes balance, a positive W must be balanced by a decrease over time of
                                        3C
solute mass stored in the REV (negative -r-), and vice versa.  Note that
for a sink C' is equivalent to the concentration in the aquifer at the
location of the sink.

     A particular solute may also be added to or removed from solution
within the REV by the effects of chemical reactions.  Examples of such
reactions include radioactive decay, ion exchange, and adsorption.  The
amount of solute that is produced (that is, added to or removed from
solution) within the REV is equal to the rate of production of the solute
times the volume of solution and may be expressed as :

                                                    8
               (Solute Mass Produced )REV - eAxAyAz
where R« is the rate of production of the solute in reaction k of 8
            different reactions  (positive for addition of solute and

            negative for removal), ML~ T~ .

     The conservation of mass for a given solute may be expressed in a
continuity equation by combining the terms in equations 26, 27a, 27b, 28,
and 29, resulting in:
          j- (CeAxAyAz) - -  ^  (CVx*  eAyAz)Ax
                         -  -  (CV  *  eAxAz)Ay - -- (CV  *  eAxAy)Az
                           «y    y            dz   z
                         - C'WAxAyAz +  EAxAyAz      R.                  (30)
                                                fc-1  *


     If we assume that changes over  time in porosity of  the  aquifer  are
not significant and Ax, Ay, and Az are constants,  then their derivatives
equal zero.  Using an indicial notation  to represent directions  in
which x.., x,, and x. correspond to the x-, y-, and z-dlrections  respec-

tively, equation 30 can be rewritten as:


         eAXlAx2Ax3 || - - AXlAx2Ax3 -gjy (CV^* e)  -  C'WAx^x^
                                       7c

                                       8
                         + eAx1Ax2Ax3  £ *fc                            (31)



                                     13

-------
Dividing both sides of equation 31 by (Ax-Ax-Ax-) to remove the common
factors results in:
                                                                       (32)
     The instantaneous mass flux of the solute is given as CV>*.  As shown
by Bear (1972, p. 101), this flux can be separated into two parts:
                                          o
                            CV.* = CV. + CV.                           (33)
      o
where V . is the deviation of the mass average velocity of the solute from
       *£"
            the average insterstitial velocity of the fluid

            (V. = V.* - V.), LT'1.
              Is    i*     V

The term CV. represents the convective flux of solute carried by the
           1^
average fluid motion through the REVv  Neglecting diffusion, the term
 o
CV. represents the dispersive flux resulting from velocity fluctuations.
  'Z'

     Bear (1972, p. 101) also shows that the dispersive flux can be
approximated by:

                             CV. = - D.. I0-                            (34)
                               ^      ^3 3x.
                                           3

where D . . is the coefficient of mechanical dispersion (a second-order
        3              2-1
             tensor), L T  .

Equation 34 indicates that the dispersive flux is directly proportional
to the concentration gradient and occurs in a direction from higher
concentrations towards lower concentrations.  The form of equation 34 is
analogous to Pick's Law describing diffusive flux, as described by Bear
(1972, p. 78).  In considering flowing ground water, diffusive fluxes are
assumed to be negligible in comparison to dispersive fluxes.  If diffusion
is negligible, the coefficient of mechanical dispersion is equivalent to
the coefficient of hydrodynamic dispersion described by Bear (1972,
p. 606).

     Coupling between forces of one type and fluxes of another type are
discussed by Bear  (1972, p. 85-90).  By following the development of
Bredehoeft and Finder (1973), and by assuming that the only significant
driving mechanism  is the gradient of hydraulic head and that Darcy's Law
is fully valid, we eliminate the necessity of considering coupled
processes.
                                     14

-------
     As a result of  substituting  equation 34  into  equation 33,  the  instan-
taneous mass  flux may be  expressed  in  terms of  the dispersion coefficient
•as:

                        CV. *  =  CV .  - D. .  -r-^—
                                           J

     Equation 35 can next be  substituted  into equation 32  to  yield:


            ** ^1       II   f  i           
-------
                 Two-dimensional areal solute transport

     A solute-transport equation for problems Involving two-dimensional
areal flow may be derived in a manner analogous to the previous derivation
of a general three-dimensional equation by assuming that vertical varia-
tions in head and concentration are negligible.  Consider the total volume
of aquifer under a representative square area, as shown in figure 2.
Note that the vertical dimension is represented by the saturated thick-
ness, b, and that b may vary within the representative area.   In this
case the mass flux through face ABCD is approximately equal to
<
                          ***> - !; (
-------
Figure 2.—Representative volume of aquifer having variable saturated
                               thickness.
                                   17

-------
Similarly, the mass flux through face EFGH may be given by

                                              Ax
                               Ix" (°V £Ayb)
                      p  £Ayb •  a^ | «•„  (-"/wi  2
     The net mass flux of the solute entering or leaving the representative
volume thus equals the difference between the two previous terms:
               (Net Mass Flux>x - - |^ (w^* eAyb\Ax                    (43)


     It can also be shown In the same manner that for the y-directlon:
               (Net Mass Flux)  = - y- { CV * eAxbJAy                    (44)


     In the case of two-dimensional areal flow, It Is assumed that any
flux across the upper or lower faces of the representative volume is
included in the source/sink term.  Solute entering or leaving the represen-
tative volume through a source or sink may be expressed mathematically
as:

                 (Source/ Sink Mass Flux) - C'WAxAy                      (45)
     Solute added to or removed from solution within the representative
volume by the effects of chemical reactions may be expressed as:


                                                 8
                (Solute Mass Produced) « eAxAyb Y) R,,                   (46)
                                                fc-1  *


     The conservation of mass for a given solute may be expressed in a
continuity equation by combining the terms in equations 43, 44, 45, and
46, resulting in:
               J£ (CeAxAyb)  - - |j (cVx* eAyb) Ax
                                                   e
                                 C'WAxAy +  eAxAyb  £ Rfc                 ^7^
                                                  fe-1
                                     18

-------
     Tf E, Ax,  and Ay are constants, equation 47 may be rewritten as:


                       • - eAxAy \- ( CV * b]
                                {?(*/*)
                                           8
                                    EAxAyb 2*  •**    •                
-------
     The left side of equation 49 may be expanded as:



                               1 C -57 + b !£                             (52)
     Substituting equation 52 into equation 51 and solving for dC/dt
results in:
                  i£-i-i-/hn   3C \   1 J_ /hrv \
                  3t   b 3x^ ( b°tj 3Xj. j " b 3x^ ( bCVt )
                         C 3b   C^    f,
                         b 3t "  eb + A  fe                            (53)
     If the saturated thickness is constant in space, the spatial deriv-
atives of b are equal to zero.  Under these conditions, equation 53 can
be further simplified to:
                    at
                           CH/.C ab    .,                             f  .
                            eb   b at +  1* ^                          (54)
     The two-dimensional solute-transport equation can be reduced further
if changes in saturated thickness over time are negligible and if the
solute is not affected by chemical reactions.  Under these conditions
equation 54 may be simplified to:
                                                    C'W
                                      20

-------
     The difference between equation 53 and equation 54 can be presented
more explicitly by expanding the first two terms on the right side of
equation 53.  After combining terms, equation 53 may then be rewritten
as:
             at
                               Cl T1   ^  ^L.
                             	W   C  ob  ,  ^-« _                          /CAN
                             ~!b " b "3T+  ^ ^                         (56)
     Thus,'the difference between equations 53 and  54 is equal  to:
In other words, the error in computed concentrations caused by  assuming
that the saturated thickness is uniform, when  it actually varies  in  space,
is inversely proportional to the saturated  thickness and directly propor-
tional to the divergence of the saturated thickness.   If the  rate of
change in saturated thickness is small compared to  the total  saturated
thickness, the simpler equation 54 can be used as a reasonable  approxi-
mation to equation 53.
      If  we consider c  as a spatial variable, then equation 48 could
 instead  be divided by  (AxAy).   Following the steps that led to equation
 51  then  results in:
                                       8
                           C'W  +  eb 2  R.                              (57)
                                      fe-1   ^
                                   21

-------
     The equation of continuity for  water may be written as:
                     •air (pebV  +  pw  •  °                            (58)
                       ^»
For homogeneous fluids,  density (p)  is  constant.  We may therefore divide
through by p to obtain:
                  +   3  (eby }   +  w  =  o                               (59)
            31       3x •      t-
     We may rewrite equation 57  by expanding  the  term on the left side,
expanding the second term on the right  side,  and  rearranging to obtain:
                3C   .   „  3(eb)
                                                  C'M
                          8
                    - .eb T Rfc  -  0                                     (60)
'fi
   After adding the term CW to both sides of equation  60  and  after
   further factoring and rearranging of terms,  we obtain:
                               22

-------
                                     8C
                                            +  c'w
                                  .]   -  cw  -  «£
          |9(cW.)
^  +  -gf-  +  w'•  "  w  "  rt£ R*   "  °
 From equation 59 it is apparent that the fifth term  on  the  left
 side equals zero.   Rearranging terms and dividing through  by  cb
 results in the following general solute-transport equation:
    3C      1   3   /_     3C
    It  =  lb"5
                                           \
                                            )
                                         H
 If porosity is constant in space, equation 62 reduces to:
                               W(C-C')
                               — -                               (63)
It is interesting to note that when W represents withdrawal only,  then
C' = C and the third term on the right side of equation 63 becomes
equal to zero.  Therefore, withdrawals from the aquifer produce
concentration changes only indirectly through the effects of the
withdrawals on the velocity field, rather than by any direct effect  on
the mass or concentration of solutes.

     If vertical variations of head or concentration  are  significant,
then the two-dimensional equations previously derived would  not  precisely
describe areal solute transport.   Cooley (written commun. , 1976) shows
that when a two-dimensional solute-transport equation is  derived with a
more rigorous vertical integration of the three-dimensional  equation,
the third dimension is not actually eliminated in converting to  areal
coordinates but instead is transformed to boundary conditions.
                                    23

-------
                         Dispersion coefficient

     The solution of the solute- transport equation requires consideration
of the dispersion coefficient.  Because of its tensorial properties, its
consideration may not appear to be straightforward.  Hence, we will next
consider this coefficient in more detail.

     Bear (1972, p. 580-581) states that hydrodynamic dispersion is the
macroscopic outcome of the actual movements of individual tracer particles
through the pores and that it includes two processes.  One process is
mechanical dispersion, which depends upon both the flow of the fluid and
the nature of the pore system through which the flow takes place.  The
second process is molecular and ionic diffusion, which because it depends
on time, is more significant at low flow velocities.  Bear (1972) further
states that the separation between the two processes is artificial.  In
developing our model we assume for flowing ground-water systems that the
definable contribution of molecular and ionic diffusion to hydrodynamic
dispersion is negligible.

     The relationship between the dispersion coefficient, the fluid flow,
and the nature of the pore system is given in tensor notation by Scheidegger
(1961, p. 3275) as:

                                      V  V
where o...   is the dispersivity of the porous medium (a fourth-order
       ifjmrl
               tensor), L;
      V  and V  are the components of the flow velocity of the fluid in
       fit      rl                                         »
               the m and n directions, respectively, LT  ; and

      |v|   is the magnitude of the velocity vector, LT~ .

     Scheidegger (1961, p. 3275) states that the dispersivity tensor
possesses 81 components, but that even in the case of an anisotropic medium,
symmetry properties reduce the number of components to 36.  Both Scheidegger
(1961) and Bear (1972) show that the dispersivity of an isotropic porous
medium can be defined by two constants.  These are the longitudinal dis-
persivity of the medium, a, , and the transverse dispersivity of the medium

a .  For an isotropic porous medium the components of the dispersivity

tensor in three dimensions (i,j <" 1,2,3) are:


                                     a                                 (68a)
                                                                       (68b)

-------
                   a.... » a....  • a....  »  a....  e  0
                                  I (aL -  V                          (68d)
     The components of the dispersion coefficient  for  three-dimensional
flow may be stated explicitly by expanding equation  67 for  a range of
three on i and J.   After eliminating terms with coefficients that equal
zero (shown by equation 68c), we obtain:
                           v v           v V           V V
                            11            99            V
                           V V           V V           V V
                            11            99           ^
                oo = aoon  ~±-A  + ao90o ~£-£  + a->->i* -*-
                22    2211  |y|      2222 jyj      2233 jy|
                           V V           VV           V V
                -.,...           ,,,.           ,,,
                33    3311           3322          3333
                                   a. --  +-:-            (69c)
                           |v|           |y|            |y|
                                 V V           V V
               D., = D,. = a191, -^  +a1,,1-?-i                    (69d)
                12    21    1212 |v|      1221 |V.
                                 V V           V V
                ._ = D,. -a.,., -^-  + a, „,  -^                    (69e)
                13    31    1313 |vj      1331 jv|
                                 V V           V V
               D,, = D., = a,,,, -^  + a     -±                      (69f)
                23    32    2323
     If we substitute the identities presented in equations 68a and b
into equations 69a, b, and c, we obtain directly:
                        V1V1       V2V2       V3V3
                        -^-^  +0.-^-^  +aT-^^                     (70a)
                                 T          T
                                   25

-------
                            V V        V V        V V
                            TT  + *'  T7  + *T TT                 <70b>
                             |v|      •   Ivl      *  |vl
                            V V        V V       V V
                             11        22        33
                                  -X -    -, -
Next, by substituting the identities given by equation  68d into equations
69d, e, and f, we see that:
                                       - V
                                             V V
                                       - V
                       D23 " °32 * <*L - V
     Scheidegger (1961) and Bachmat and Bear (1964)  also  show that for a
Cartesian coordinate system x« in which one of the axes,  say  x.,  coincides

with the direction of the average velocity, then V..  =  |v|  and \^  = 0.

Substituting these relations into equations 70 and 71  we  obtain:
                                 DL                                    (72.)
where D. and D  are respectively the longitudinal and transverse
                                   2-1
         dispersion coefficients, L T
                                    26
                                                                       (72b)
                   D12 ' D21 " D13 - D31 C D23 " D32 " °               (72c)

-------
     Solving equations 72a and b for a, and a., results in:
                                a  - -f-                               (73a)
                                 L    V
                                     DT
                                a  - -T
     Introducing equations 73a and b into equations  70 and  71 produces:


                          (V.)2        (V)2        (V.)2

                 "n • °         + °T -7  + "T -fl
                           (v,)2        
                                       V V
                 D.  - D«. -  (D. - D_) -i-3-                            (7Ae)
                  13    31     L    T   jv|2
                 D23 " D32 '  (DL * DT)    l                            (7Af )
                                        v
                                           2
     Equation 74 defines the local  transformation of  D..  from orthogonal
                                                       «v
axes, in which x, is parallel to V., to global  cartesian  axes.   Note that

while D.., D.., and D__ must have positive values,  it is  possible for the

cross-product terms (equations 74d, e, and f) to have negative values.
                                     27

-------
For the case of two-dimensional flow, all components and terms in equa-
tion 74 that have a subscript 3 are eliminated.

     The magnitude of the velocity,  |v|, is defined as:
            2 + v 2 + v 2
            L    V2  + V3
                                                                       (75)
     In summary, the components of the dispersion tensor that must be
evaluated for three-dimensional flow in an isotropic porous medium are:
Dll D12 D13
D21 D22 D23
D31 D32 D33
                                        D   D   D
                                         xx  xy  xz
                                        D   D   D
                                         yx  yy  yz
                                         zx  zy  zz
                                                 (76)
                                 SUMMARY

     A general equation describing the three-dimensional transport and
dispersion of a reacting dissolved chemical in flowing ground water was
derived nonrigorously from the principle of conservation of mass.  The
general equation relates concentration changes to hydrodynamic dispersion,
convective transport, fluid sources and sinks, and chemical reactions.

     Concentration changes caused by dispersion are assumed to be a
function of both the dispersion coefficient and the concentration gradient.
The dispersion coefficient is a second-order tensor and is related to
the dispersivity of the porous medium and to the flow velocity of the
ground water.  If solute concentrations are affected by chemical reactions',
specific mathematical expressions describing the rates of reactions must
be incorporated into the general solute-transport equation.

     Because both the dispersion coefficient and convective transport
depend on the flow velocity, the solution of the solute-transport equation
requires the definition of the velocity field, which in the general case
requires that the flow and solute-transport equations be solved simul-
taneously.  However, the solution of these equations can be considerably
simplified if conditions of homogeneous fluid properties and (or) two-
dimensional flow can be validly assumed.
                                     28

-------
                            REFERENCES CITED

Ahlstrom, S. W., and Baca, R. G., 1974, Transport model user's manual:
     Battelle Pacific Northwest Laboratories rept. BNWL-1716, 25 p.

Bachmat, Y., and Bear, J., 1964, The general equations of hydrodynamic
     dispersion in homogeneous, isotropic, porous mediums:  Jour. Geophys.
     Research, v. 69, no. 12, p. 2561-2567.

Bear, J., 1972, Dynamics of fluids in porous media:  New York, Am. Elsevier
     Publishing Co., 764 p.

Bredehoeft, J. D., 1969, Finite difference approximations to the equations
     of ground-water flow:  Water Resources Research, v. 5, no. 2,
     p. 531-534.

Bredehoeft, J. D., and Pinder, G. F., 1973, Mass transport in flowing
     groundwater:  Water Resources Research, v. 9, no. 1, p. 194-210.

Cooley, R. L., 1974, Finite element solutions for the equations of
     ground-water flow:  Hydrology and Water Resources Pub. no. 18,
     Desert Research Inst., Univ. Nevada, 134 p.

Cooper, H. H., Jr., 1966, The equation of groundwater flow in fixed and
     deforming coordinates:  Jour. Geophys. Research, v. 71, no. 20,
     p. 4785-4790.

Grove, D. B., 1976, The use of Galerkin finite element methods to solve
     mass transport equations:  unpub. Ph. D. thesis, Colorado School of
     Mines, Golden, Colo., 152 p.

Gupta, S. K., Tanji, K. K., and Luthin, J. N., 1975, A three-dimensional
     finite element ground water model:  California Water Resources Center,
     contr. no. 152, 119 p.

Hubbert, M. K., 1940, The theory of ground-water motion:  Jour. Geology,
     v. 48, no. 8, pt. 1, p. 785-944.

Konikow, L. F., 1976, Modeling solute transport in groundwater:  Internat.
     Conf. on Environmental Sensing and Assessment, Las Vegas, Nev., Proc.,
     art. 20-3.

Konikow, L. F., and Bredehoeft, J. D., 1974, Modeling flow and chemical
     quality changes in an irrigated stream-aquifer system:  Water Resources
     Research, v. 10, no. 3, p. 546-562.

Lantz, R. B., Pahwa, S. B., and Grove, D. B., 1976, Development of a
     subsurface waste disposal simulation model [abs]:  EOS, Am. Geophys.
     Union Trans., v. 57, no. 4, p. 249.

Pinder, G. F., 1973, A Galerkin-finite element simulation of ground-water
     contamination  on Long Island, New York:  Water Resources Research,
     v. 9, no. 6, p. 1657-1669.

                                    29

-------
Reddell, D. L., and Sunada, D. K., 1970, Numerical simulation of dispersion
     in groundwater aquifers:  Colorado State Univ. Hydrology Paper 41,
     79 p.

Robertson, J. B., 1974, Digital modeling of radioactive and chemical waste
     transport in the Snake River Plain aquifer at the National Reactor
     Testing Station, Idaho:  U.S. Geol. Survey open-file rept., 41 p.

Robson, S. G., 1974, Feasibility of digital water-quality modeling illus-
     trated by application at Barstow, California:  U.S. Geol. Survey
     Water-Resources Inv. 46-73, 66 p.

Scheidegger, A. E., 1961, General theory  of dispersion in porous media:
     Jour. Geophys. Research, v. 66, no. 10, p. 3273-3278.

Segol, G., and Finder, G. F., 1976, Transient simulation of saltwater
     intrusion in southeastern Florida:  Water Resources Research, v. 12,
     no. 1, p. 65-70.
                                     30

-------
      SECTION 2






MODEL DOCUMENTATION

-------
   Techniques of Water-Resources Investisations
      oF the United States Geological Survey
                  Chapter C2

COMPUTER MODEL OF  TWO-DIMENSIONAL
   SOLUTE TRANSPORT AND DISPERSION
            IN GROUND WATER
            By L F. Konikow and J. D. Bredehoeft
                    Book 7
        AUTOMATED DATA PROCESSING AND COMPUTATIONS

-------
                DEPARTMENT OF THE INTERIOR


                  WILLIAM P. CLARK, Secretary



                   U.S. GEOLOGICAL  SURVEY


                     Dallas L. Peck, Director
                    First printing  1978
                    Second printing 1984
     UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1978
For sale by the Books and Open-File Reports Section, U.S. Geological Survey,
              Federal Center, Box 25425, Denver, CO 80225

-------
                             PREFACE

    The series of manuals on techniques describes procedures for plan-
ning and executing specialized  work  in  water-resources  investigations.
The material is grouped under major  headings called books and further
subdivided into sections and chapters; section C of Book 7 is on computer
programs.
    This  chapter presents a  digital computer  model for  calculating
changes in the concentration of a dissolved chemical species in flowing
ground  water.  The  computer program represents a  basic and general
model that may have to be modified by the user for efficient application to
his specific field problem. Although this model will produce reliable  cal-
culations for a wide variety of field problems, the user is cautioned that in
some cases the accuracy and efficiency of the model can be affected  sig-
nificantly by his selection of values for certain user-specified options.
                                                                        m

-------
                                          CONTENTS
Abstract		-	
Introduction _			
Theoretical background 	
        Flow equation	
        Transport equation	
        Dispersion  coefficient  	
        Review of assumptions	
Numerical methods 	
    Flow equation	
    Transport equation	
        Method of characteristics	
        Particle tracking	
        Finite-difference approximations	
        Stability criteria 	
        Boundary and initial conditions	
        Mass balance	
        Special problems 	
Computer program	
    General program features	
    Program segments	
        MAIN	
        Subroutine PARLOD  _	
        Subroutine ITERAT	
        Subroutine GENPT 			
        Subroutine  VELO  	
        Subroutine MOVE			
Pure
  1
  1
  2
  2
  3
  3
  4
  4
  4
  5
  6
  6
  7
 11
 13
 14
 15
 19
 20
 21
 21
 22
 22
 22
 23
 23
                                            Page
Computer program—Continued
    Program segments—Continued
        Subroutine CNCON			    26
        Subroutine OUTPT	    26
        Subroutine CHMOT	.	    25
Evaluation of model	    25
    Comparison with analytical solutions —    26
    Mass balance  tests 	    28
        Test problem 1—spreading of a tracer
          slug	    28
        Test problem 2—effects of  wells	    31
        Test  problem  3—effects   of  user
          options  	    32
    Possible program modifications	    34
        Coordinate system and boundary
          conditions	    36
        Basic equations 	    36
        Input and output	    36
Conclusions  			    87
References cited „		    87
Attachment I, Fortran IV program listing	    41
Attachment II, Definition of selected program
  variables	    74
Attachment III, Data input formats	    76
Attachment IV, Input data for test problem 3    79
Attachment V, Selected output for test
  problem 3	    80
                                            FIGURES
                                                                                                Page
 1. Part of hypothetical finite-difference grid showing relation of flow field to movement of points...      7
 2. Part of hypothetical finite-difference grid showing areas over which bilinear interpolation is used
        to compute the velocity at a point	      7
 8. Representative change in breakthrough curve from time level fc—1 to k			     11
 4. Possible movement of particles near an impermeable (no-flow) boundary	     15
 B. Replacement of points in source cells adjacent to a no-flow boundary			     16
 6. Replacement of points in source cells not adjacent  to a no-flow boundary for negligible regional
        flow  (a) and for relatively strong regional flow (b)  	     17
 7. Relation between possible initial locations of points and indices of adjacent nodes	     19
 8. Simplified flow chart illustrating the major steps in the calculation procedure		     21
 9. Parts of finite-difference  grids showing the initial geometry of particle distribution for the specifi-
        cation of four (a), five (b), eight (c), and nine (d) particles per cell	     28
10. Generalized flow chart of subroutine MOVE  	     24
11. Comparison  between analytical  and  numerical  solutions  for  dispersion  in  one-dimensional,
        steady-state flow	     26

-------
VI                                         CONTENTS


                                   FIGURES—Continued
                                                                                             P.«e
12.  Comparison between analytical and numerical solutions for dispersion in plane radial steady-
        state flow	    28
13.  Grid, boundary conditions, and flow field for test problem 1			    29
14.  Mass-balance errors for test problem 1	    30
15.  Grid, boundary conditions, and flow field for test problem 2	    30
16.  Mass-balance errors for test problem 2	    31
17.  Grid, boundary conditions, and flow  field for test problem 3	    32
18.  Effect of NPTPND on mass-balance error for test problem 3; CELDIS=0.50 in all cases	    33
19.  Effect of CELDIS on mass-balance error for test problem 3; NPTPND=9 in all cases	    34
                                           TABLES
                                                                                             Page
 1.  List of subroutines for solute-transport model	    20
 2.  Model parameters for test problem 1 	    29
 3.  Model parameters for test problems 2 and 3	    31
 4.  Effect of NPTPND on accuracy, precision, and efficiency of solution to test problem 3	    33
 5.  Effect of CELDIS on accuracy, precision, and efficiency of solution to test problem 3	    33

-------
    COMPUTER  MODEL OF  TWO-DIMENSIONAL  SOLUTE TRANSPORT
                      AND DISPERSION  IN  GROUND  WATER
                            By L F. Konikow and J. D. Bredehoeft
                  Abstract

  This report presents a model that simulates solute
transport in flowing ground  water. The  model  is
both general and flexible in that it can ba applied
to a wide range of problem types. It is applicable
to  one- or two-dimensional  problems  involving
steady-state or transient flow. The model  computes
changes in  concentration over time  caused by the
processes of  convective transport,  hydrodynamic
dispersion,  and mixing  (or  dilution)  from fluid
sources. The model assumes that the solute is non-
reactive and that gradients of fluid density, viscos-
ity, and temperature do not affect the velocity dis-
tribution. However, the aquifer  may  be hetero-
geneous and (or)  anisotropic.
  The model couples  the ground-water flow equa-
tion with the solute-transport equation. The digital
computer program uses an alternating-direction im-
plicit procedure to solve a finite-difference approxi-
mation to  the  ground-water  flow equation, and  it
uses  the method  of  characteristics  to solve the
solute-transport equation. The latter uses a particle-
tracking procedure to represent convective transport
and a two-step explicit  procedure to solve a finite-
difference equation  that describes  the effects of hy-
drodynamic dispersion, fluid sources and sinks, and
divergence of velocity. This  explicit procedure has
several stability criteria, but the consequent time-
stop limitations are automatically determined by the
program.
  The report includes a listing of the computer pro-
gram, which is written  in FORTRAN IV and con-
tains about 2,000 lines.  The  model is  based on  a
rectangular, block-centered, finite-difference grid. It
allows the specification  of any number of injection
or withdrawal wells and of spatially varying diffuse
recharge or discharge,  saturated thickness, trans-
missivity, boundary conditions, and initial heads and
concentrations. The program also permits  the desig-
nation of up to five nodes as observation points, for
which a summary  table of head  and  concentration
versus time is printed at the end of the calculations.
The data input formats for the model  require three
data cards  and from seven to nine data sets to de-
scribe  the  aquifer  properties,  boundaries,  and
stresses.
  The accuracy of the model was evaluated for two
idealized  problems for  which  analytical  solutions
could  be  obtained. In the case of  one-dimensional
flow  the  agreement was nearly  exact,  but in  the
case  of plane radial flow a small  amount of  nu-
merical dispersion occurred. An analysis of several
test problems indicates that the  error in  the mass
balance will be generally less than  10 percent. The
test problems demonstrated that the accuracy and
precision  of  the  numerical solution is sensitive to
the initial number of particles placed in  each  cell
and to the size of the time increment, as determined
by the stability  criteria. Mass balance errors  are
commonly the greatest during the first several time
increments, but tend to  decrease and stabilize with
time.

             Introduction

   This  report  describes and   documents a
computer model  for  calculating transient
changes in  the concentration  of a nonreac-
tive  solute  in  flowing ground water.  The
computer program solves  two simultaneous
partial  differential equations.  One equation
is the ground-water flow equation, which de-
scribes the head distribution in the aquifer.
The second is  the solute-transport equation,
which describes the chemical  concentration
in the system.  By coupling the flow equation
with the solute-transport equation, the model
can be applied to both steady-state and tran-
sient flow problems.
   The purpose of the simulation model is to
compute tiie   concentration of a  dissolved
chemical species in an aquifer at  any speci-
fied  place  and time.  Changes  in chemical
concentration   occur   within  a   dynamic
ground-water  system primarily due to four

-------
                 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
distinct processes:  (1) convective transport,
in which  dissolved chemicals are  moving
with the flowing ground water; (2) hydro-
dynamic dispersion, in which molecular and
ionic diffusion and small-scale variations in
the velocity of flow through the porous media
cause the  paths of dissolved molecules and
ions to diverge or  spread from the average
direction of  ground-water  flow;  (3)  fluid
sources, where water of one composition is
introduced into water of a different composi-
tion;  and  (4) reactions,  in  which  some
amount of a particular dissolved chemical
species may be added to or removed from the
ground water due  to chemical and physical
reactions in the water or between the water
and the solid aquifer  materials.  The model
presented in this report assumes (1) that no
reactions occur that affect the concentration
of the species of interest, and (2) that gra-
dients  of  fluid density,  viscosity, and  tem-
perature do not affect the velocity distribu-
tion.
    This  model  can be applied to a wide
variety of field problems. However, the user
should first become aware of the assumptions
and  limitations  inherent in  the  model, as
described  in  this report. The computer pro-
gram presented in this report is offered as a
basic working tool that  may  have to be
modified by the user for efficient application
to specific field  problems.  The program is
written in FORTRAN IV and is compatible
with  most high-speed computers. The data
requirements, input  format specifications,
program options, and output formats are all
structured in a general manner that should
be readily adaptable to many field problems.
   This report includes a detailed description
of the numerical  method used to solve the
solute-transport equation. The reader is as-
sumed to have (or can obtain elsewhere) a
moderate familiarity  with finite-difference
methods and ground-water flow models.

    Theoretical  Background

              Flow equation
   By following the derivation of Finder and
Bredehoeft (1968), the equation describing
the transient two-dimensional areal flow  of
a homogeneous compressible fluid through a
nonhomogeneous anisotropic aquifer can  be
written in Cartesian tensor notation as
                     W
                3*
(i)
where
     Tit          is the transmissivity ten-
                   sor, Lf/T\
     h           is the hydraulic head, L;
     S           is the  storage coefficient,
                   (dimensionless);
     t            is the time, T;
     W= W(x,y,t) is the volume flux per unit
                   area (positive sign for
                   outflow  and  negative
                   for inflow), L/T; and
     x( and Xj     are the Cartesian coordi-
                   nates, L.
If we only consider fluxes of (1) direct with-
drawal or  recharge, such as well pumpage,
well  injection, or evapotranspiration,  and
 (2)  steady leakage into or out  of the aquifer
through a confining  layer, streambed, or
lakebed, then W(x,y,t)  may  be expressed
as

     W(x,y,t)=*Q (x.y.t) ~(H.- h)     (2)
                        //»
where
     Q   is the  rate  of  withdrawal  (posi-
           tive sign)  or  recharge  (negative
           sign),L/r;
     K,  is the vertical hydraulic conductiv-
           ity of the confining layer, stream-
           bed, or lakebed, L/T;
     m  is the  thickness  of  the  confining
           layer, streambed, or lakebed, L;
           and
     H,  is the hydraulic head in the source
           bed, stream, or lake, L.
   Lohman (1972) shows that an expression
 for  the  average seepage velocity of ground
 water can  be derived  from  Darcy's law. This
 expression can be written  in  Cartesian ten-
 sor  notation as
                        3h
                                      (3)

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                                    3
where
     Vt  is the seepage velocity in the direc-
          tion of xlt L/T;
     KH is the hydraulic conductivity tensor,
          L/T; and
     c   is the effective porosity of the aqui-
          fer,  (dimensionless).

          Transport equation
  The equation used to describe the two-di-
mensional areal transport and dispersion of
a given nonreactive dissolved chemical  spe-
cies in flowing ground water was derived by
Reddell and  Sunada  (1970),  Bear  (1972),
Bredehoeft and Finder (1973), and Konikow
and  Grove  (1977). The  equation  may be
written as

                                    C'W

                             t,/=l,2  (4)
where
     C  is the concentration of the dissolved
          chemical species, M/L3;
     DtJ is the coefficient of  hydrodynamic
          dispersion   (a  second-order  ten-
          sor), L»/T;
     b   is the  saturated thickness  of the
          aquifer, L; and
     C'  is the concentration of the dissolved
          chemical in a source or sink fluid,
  •Qt
  The first term on the right side of equa-
tion  4 represents the change in  concentra-
tion  due to hydrodynamic  dispersion.  The
second term describes the effects of convec-
tive  transport, while the third term repre-
sents a fluid source  or sink.

         Dispersion coefficient
  Bear (1972, p. 580-681) states that hydro-
dynamic dispersion is the macroscopic  out-
come of the actual  movements of individual
tracer particles through the pores and  that
it includes  two processes.  One  process  is
mechanical dispersion, which depends  upon
both the flow of the fluid and the nature  of
the pore system through which the flow takes
place. The second process is molecular  and
ionic diffusion, which because it depends on
time, is more  significant at  low flow veloci-
ties. Bear (1972)  further  states that the
separation between the two processes is arti-
ficial. In developing our model we assume for
flowing ground-water  systems  that the de-
finable contribution  of molecular and ionic
diffusion  to  hydrodynamic dispersion  is
negligible.
  The  dispersion coefficient may be related
to the  velocity of ground-water flow and  to
the nature of the aquifer using Scheidegger's
 (1961) equation:
                      VmV(.
              **\i= CUfwm |»ri             * •* '
                       In
where
    «
-------
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                       w^     <«>

  Note  that while DZI  and !>„„ must have
positive values, it is possible for the cross-
product terms (eq  10) to  have negative
values if V, and V, have opposite signs.

         Review of assumptions
  A number of assumptions have been made
in the development  of the previous equa-
tions. Following is a list of the main assump-
tions that must be carefully evaluated before
applying the model to a field problem.
1. Darcy's law is valid and  hydraulic-head
     gradients are the only significant driv-
     ing mechanism for fluid flow.
2. The porosity and hydraulic conductivity
     of the aquifer are constant with time,
     and porosity is uniform  in space.
3. Gradients of fluid density, viscosity, and
     temperature do  not affect the velocity
     distribution.
4. No chemical reactions occur that affect
     the  concentration of the  solute, the
     fluid properties, or the aquifer  proper-
     ties.
5. Ionic and molecular diffusion are negli-
     gible contributors to the total disper-
     sive flux.
6. Vertical variations in head and  concen-
     tration are negligible.
7. The aquifer is homogeneous and isotropic
     with respect to the coefficients of longi-
     tudinal and transverse dispersivity.
  The nature of a specific field problem may
be such that not all of these underlying as-
sumptions  are completely valid. The degree
to which field conditions deviate from these
assumptions will affect the applicability and
reliability of the model for that problem. If
the deviation from a particular assumption
is significant, the governing equations will
have to be modified to account for  the ap-
propriate processes or factors.

       Numerical Methods

   Because aquifers have variable  properties
 and complex boundary conditions,  exact ana-
                           lytical solutions to the  partial differential
                           equations of flow  (eq 1) and  solute trans-
                           port  (eq 4)  cannot  be  obtained  directly.
                           Therefore, approximate  numerical  methods
                           must be employed.
                             The numerical methods require  that the
                           area of interest be subdivided by a grid into
                           a number of smaller  subareas. The model
                           developed here utilizes a rectangular,  uni-
                           formly spaced, block-centered,  finite-differ-
                           ence grid, in which nodes are defined at the
                           centers of the rectangular cells.

                                         Plow equation
                             Finder and Bredehoeft (1968) show that
                           if the coordinate axes are  alined with the
                           principal directions of  the transmissivity
                           tensor, equation 1 may be approximated by
                           the following implicit  finite-difference equa-
                           tion:
                                      h*-ij,k — ht,i.k I

                               "-"•"I   (AX)»   J
                                                                 (ID
                            where
                                i,j,k      are  indices in the x,  y, and
                                            time  dimensions,   respec-
                                            tively;
                                Az,Ay,At  are  increments in the x, y,
                                            and  time dimensions,  re-
                                            spectively; and
                                qw        is the volumetric rate of with-
                                            drawal or recharge at the
                                            (i,j) node, L*/T.
                            Note that k represents the  new time level
                            and fc-1 represents the previous time level.
                            To  avoid confusion  between  tensor  sub-

-------
                   MODEL OP SOLUTE TRANSPORT IN GROUND WATER
scripts and nodal indices, the latter are sep-
arated by commas.
  The finite-difference equation (eq 11)  is
solved numerically for each node in the grid
using an  iterative alternating-direction im-
plicit (ADI)  procedure.  The derivation and
solution of the finite-difference equation and
the use of the iterative ADI procedure have
been  previously discussed in  detail in the
literature. Some of the more relevant refer-
ences include Finder and Bredehoeft (1968),
Prickett  and Lonnquist  (1971),  and  Tres-
cott, Finder, and Larson  (1976).
  After the head distribution has been com-
puted for a given time step, the velocity of
ground-water flow is computed at each node
using an  explicit  finite-difference  form of
equation 3. For example, the velocity in the
x direction at node  (i,j)  would be computed
as
  The velocity in the x direction can also be
computed on  the boundary  between  node
(i,j) and node (t+1,;)  using the following
equation :
                                     (13)

where the  hydraulic  conductivity on the
boundary is computed as the harmonic mean
of the hydraulic conductivities  at the two
adjacent nodes.
  Expressions similar to equations 12 and 13
are used to compute the velocities in the y
direction at (t,/) and (t,/+Vfc) respectively.
Note  that equation 13, which computes the
head  difference over a distance Ax, is  more
accurate than equation 12, which computes
the head difference over 2Ax.


           Transport equation

         Method of characteristics
  The method of characteristics is used in
this model to solve the solute-transport equa-
tion.  This  method was developed  to  solve
hyperbolic  differential equations.  If solute
transport is dominated by convective trans-
port, as is common in many field problems,
then equation 4 may closely approximate a
hyperbolic partial differential equation and
be highly  compatible with  the method of
characteristics.  Although it is difficult to
present a rigorous mathematical proof for
this numerical  scheme, it has been success-
fully applied to a variety of field problems.
The development of this technique for prob-
lems of flow through porous media has been
presented by Carder, Peaceman, and Pozzi
(1964), Finder and Cooper  (1970),  Reddell
and  Sunada  (1970), and Bredehoeft  and
Finder  (1973).  Carder,  Peaceman,  and
Pozzi (1964) state that this technique does
not introduce numerical dispersion  (artifi-
cial dispersion resulting from the numerical
calculation process). They and Reddell and
Sunada (1970)  also compared  solutions ob-
tained  using the method of characteristics
with those  derived  by analytical methods
and found good agreement for  the cases in-
vestigated. Applications  of  the method to
field  problems   have  been  documented  by
Bredehoeft  and  Finder  (1978),  Konikow
and Bredehoeft (1974),  Robertson  (1974),
Robson (1974),  and Konikow (1977).
  The approach taken by the method of char-
acteristics is not to solve equation 4 directly,
but rather to solve an equivalent system of
ordinary differential equations. Konikow and
Grove  (1977, eq 61) show that by consider-
ing saturated thickness as a variable and by
expanding the  convective transport term,
equation 4 may  be rewritten as
                                                                 -V.
                                                        W(C  - C')

                                                           tb
                                  •   (14)
                                            Equation 14 is the form of the solute-trans-
                                            port equation that is solved in the computer
                                            program presented in this report. For con-
                                            venience we may also write equation 14 as
                                                                                  (15)

-------
                  TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
where
            F -
                                      (16)
  Next consider representative fluid  par-
ticles that are convected with flowing ground
water. Note that changes with time  in prop-
erties of the fluid, such as concentration, may
be described either  for  fixed points within
a stationary coordinate system as successive
fluid particles pass the reference points, or
for reference fluid particles  as they move
along their respective paths past fixed points
in space. Aris (1962, p. 78) states that "as-
sociated with these two descriptions are two
derivatives  with  respect  to  time."  Thus
"dC/fit is the rate of change of concentration
as observed  from a fixed  point,   whereas
dC/dt is the rate of change as observed when
moving with the fluid particle.  Aris (1962)
calls the latter the material derivative.
  The material derivative  of concentration
may be defined as
                       dx

           _                  dy
         dt   "ftt  "Qx dt   "Qy dt
Note the correspondence of the second and
third terms on the right side of equation 15
with the second and third terms on the right
side of equation 17. The latter includes the
material derivatives of position, which are
defined by velocity.  Thus  for  the x and y
components,  respectively,  of  position and
velocity we have
 and
                  dt

                  dy
                 ~dt
                                      (18)
                                      (19)
  If we next substitute  the  right sides of
equations 15, 18, and 19 for the correspond-
ing terms in equation 17, we obtain

         —=— — (bDi^-)+F.     (20)
          dt   b 3x1     fiXf
  The solutions  of the system of equations
comprising equations 18-20 may be given as
                                      (21)
and are called the characteristic curves of
equation 15.
  Given solutions to equations 18-20, a solu-
tion to the partial differential equation  (eq
15) may be obtained by following the char-
acteristic  curves.  This  is accomplished  nu-
merically  by  introducing a set of moving
points (or reference particles)  that can be
traced within the stationary coordinates of
the finite-difference grid. Garder, Peaceman,
and Pozzi  (1964,  p. 27) state,  "Each point
corresponds to one characteristic curve,  and
values of  x, y, and C are obtained as func-
tions of t for each characteristic." Each point
has a concentration and position associated
with it and is moved through the flow field
in proportion  to the flow velocity at its loca-
tion. Intuitively, the method may be visual-
ized  as tracing a  number of fluid particles
through a flow field and observing changes
in chemical concentration in the fluid par-
ticles as they move.

              Particle tracking

   The first step in the  method of character-
istics involves placing a number of trace-
able particles or  points in each cell of the
finite-difference grid to form a  set of points
that are distributed in a geometrically  uni-
form pattern throughout the area of inter-
est. It was found  that  placing from four to
nine points per cell provided satisfactory re-
sults  for  most two-dimensional problems.
The location or position of each particle is
specified by its x- and  y- coordinates in the
finite-difference grid. The initial concentra-
tion assigned  to each point  is the initial  con-
centration associated with the node of the
cell containing the point.
   For each time step every point is moved a
distance proportional to the length of the
time  increment and the velocity at the loca-
tion of the point.  (See fig. 1.) The new posi-
tion of a point is thus computed with the fol-
lowing finite-difference forms  of equations
18 and 19:
                                                                                    (22)
                                              and

-------
                    MODEL OF  SOLUTE TRANSPORT IN GROUND WATER
L
I
                  EXPLANATION
           •  Initial location of particle
           O  New location of particle
          —^- Flow line and direction of flow
          	Computed path of particle

        Figure 1.—Part of hypothetical finite-
          difference grid  showing relation of
          flow field to movement of  points.
                                        (23)
where
     p          is  the  index  number for
                   point identification ; and
     Bxf and &yf are the distances moved in
                   the x and y directions, re-
                   spectively.
  The  x and y  velocities  at the  position
of  any particular  point  p,  indicated as
V•»<,.»,]» for time k are calculated through
bilinear interpolation over the area  of half
of a cell using the x and y velocities com-
puted at adjacent nodes and cell boundaries.
For example, figure  2 illustrates that the
velocity in the x direction of point p, located
in the southeast quadrant of cell (i,j) , would
be computed using bilinear interpolation be-
tween the x velocities computed with  equa-
tions 12 and IS at (i,j), (t,; + l), (t + VW),
and (t-r-V6 ,;'+!). Similarly, the velocity in
the  y direction of point p would be based on
the  y velocities computed at (tj),  (t +!,/),
.jf The time index is distinguished
with  an asterisk here because  this  tempo-
rarily assigned  average  concentration rep-
resents the new  time level only with  respect
to convective transport. The moving points
simulate convective transport because the
concentration at each node of the grid will
change with  each  time  step as  different
points having different concentrations enter
and leave the area of that cell.

      Finite-difference approximations
  The total change in concentration in an
aquifer may be computed by solving equa-
tions 18-20. Equations 18 and 19, which are
related to changes in concentration caused

-------
8
      TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
by convective transport  alone,  are  solved
by the  movement of  points  as described
previously. The changes  in concentration
caused by hydrodynamic dispersion, fluid
sources, divergence of velocity, and changes
in saturated thickness  are calculated  using
an explicit finite-difference approximation to
equation 20, which can be expressed as
  f~ 1  3      3C     ~\
 fl -- (60V— ) +F  .
  L b 3*<     3z,     J
                                    (24)
  Note that a  solution  to  equation  20 re-
quires the computation of the change in con-
centration at the tracer  particles. However,
primarily because of the difficulty in comput-
ing the concentration gradient at a large
number of moving points, the change in con-
centration  represented  by  equation 20  is
solved at each node of the grid rather  than
directly at the location  of  each point. The
material  derivative of concentration on any
characteristic curve (or  for any tracer par-
ticle) is then related to the change in con-
centration for a node during one time  step,
which  was computed with the solution  to
equation 24.
  The right side of equation 24 can be con-
sidered as the  sum of two separate terms,
as follows :
        ACijU = ( ACU,*) ! + ( AC,,M) „    (25)
where
          k) i  is the change in concentration
                caused   by hydrodynamic
                dispersion,  and is  defined
                as
and
                                     (26)
     (ACu.k)n is the  change in concentra-
                tion  resulting from an ex-
                ternal  fluid  source  and
                changes in saturated thick-
                ness, and from equation 16
                is defined as
.   (27)
    = A/
W(C  -  C')

   €/)
                                   First we will examine the change in con-
                                 centration due to dispersion, partly follow-
                                 ing the development of Reddell and Sunada
                                  (1970). The right side of equation 26 can be
                                 expanded according to the summation con-
                                 vention of tensor notation to obtain

                                          .$ *.{wje+wj£.}
                                           T  -
                                                3»
                                                                      (28)
                                   A finite-difference approximation for the
                                 derivative in the z direction at  (i,j) may be
                                 written as
                                 2-(M)j£
                                  3z      Qz
                              )
                                                    AZ
                                                       AZ
                                                                      (29)
                                  In the following expansion of equation 29
                                  it  is implied that concentrations  (C)  are
                                  known from the previous (fc-1) time level;
                                  hence, equation 29 is an explicit finite-differ-
                                  ence equation. The spatial derivatives of con-
                                  centration at (i+Vz,j) may be approximated
                                  by

                                                 +HJ
                                  and
                                                           AZ
                                                        C       —C
                                                          H-H./+1
                                                                      (30)
                 vu         2Ay
                                    (31)
Because concentrations are defined  only  at
nodes,  we must express  the right  side  of
equation 31 in terms of concentrations  at
nodes. Assuming that the concentration at a

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
cell boundary is approximately equal to the
average (arithmetic mean) of the concentra-
tions at adjacent nodes, we have
                       +C
        <+K.H->
and

       H-KJ-i
                                     (32)
                                     (33)
Substitution  of equations  32 and  33 into
equation 31 results in:
                                                                                (34)
                                              Similarly, the spatial derivatives of con-
                                           centration at (i-l/2,j) are
                                            (—\
                                            \dx)i-
                                                                Ax
                                                                                 (35)
                                             and
                                                                                (36)
                                            After substituting equations 30, 34, 35, and
                                            36 into equation 29, we have
                    bD
                             (AX)'

                           J) '  U+l
      (AX)'

-C«-,-Cl
                       wrH-HJP  <-
                                       +
                                       4AXAV
                                                                                  (37)
   A finite-difference approximation for  the
 derivative in the y direction in equation 28
                                            may be developed for node (i,/) in an analo-
                                            gous manner to equation 37 to produce
 "dv


                                                          AJ/
                         ^
                                             (Atf)«
                                        -C, .. -C.
                                         ~  «-U-t~ <-U'
                                                                                  (38)
                                4AXAV

  Equation 28 may then be solved explicitly
by substituting the relationships expressed
                                             by equations 87 and 38 for the terms within
                                             brackets on the right side of equation 28.

-------
10
TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
  Next we will examine the change in con-
centration  denoted  by equation 27. Substi-
tuting explicit finite-difference  approxima-
tions for tiie terms in equation 27, we have
                                     (39)
  Equations 28, 37, 38, and 39 together pro-
vide a solution to equation 24, which in turn
allows us to solve equation 20 and complete
the definition of the characteristic curves of
equation 15.
  Because the processes of convective trans-
port, hydrpdynamic dispersion,  and mixing
are  occurring  continuously and simultane-
ously, equations 18,  19, and 20 should  be
solved simultaneously.  However,  equations
18 and 19 are  solved by particle movement
based on implicitly computed heads  while
equation 20 is solved  explicitly with respect
to concentrations. Because the change in con-
centration at a source node due to mixing is
proportional to the difference in concentra-
tion between the node and the  source fluid
 (see eq 27), the accuracy  of estimating the
concentration at the node  during a time  in-
crement  will clearly  affect the  computed
change. Similarly, because the change in con-
centration due to dispersion is proportional
to the concentration gradient at a point, the
accuracy of estimating  the concentration
                            gradient will clearly affect the accuracy of
                            the numerical results. As the  position of a
                            front or breakthrough curve advances with
                            time, say from the k-1 to k time level, the
                            concentration gradient at any fixed reference
                            point and the concentration  differences at
                            sources are continuosly  changing. The con-
                            sequent  limitations imposed by estimating
                            nodal concentrations in a strict explicit man-
                            ner can  be minimized by using a two-step
                            explicit  procedure in which equation 24  is
                            solved at each node by giving equal weight
                            to  concentration  gradients computed  from
                            the concentrations at the previous time level
                            (k-1)  and to concentration gradients com-
                            puted from concentrations at time level (k*),
                            which represents the convected position of
                            the front at the new time level (k)  prior to
                            adjustments of concentration for dispersion
                            and mixing. Figure 3 illustrates the sequence
                            of calculations to solve equations 18-20 over
                            a given time increment.  First the concentra-
                            tion  gradients  at the  previous time level
                            (k-1) are determined  at each node. Then
                            the front is convected to a new position for
                            time level k* based on  the velocity of flow
                            and length of the time  increment. Next the
                            concentration gradients at each node are re-
                            computed for the new position of the front.
                            The concentration distribution for  the new
                            frontal position is then adjusted at each node
                            in two  steps: first based on  concentration
                            gradients at  k-1 and second  based on con-
                            centration gradients at  k*.
                              The   finite-difference  approximation  to
                            equation 24 may thus be  expressed as
             0.5
                 b


              0.5 A/
         dxi
                           a.v,
                                (bD:
                      ac
                         /,M
                         *   '-
                                                                 (40)
 in which the appropriate finite-difference ap-
 proximations for the terms within brackets
 are indicated by equations 37, 38, and 39.
                               The new nodal concentrations at the end
                             of time increment k are computed as
                                                         »        (41)

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                        11
                                              0.6
                                       RELATIVE DISTANCE
                                                                                      1.0
  Figure 3.—Representative change in breakthrough curve from time level It—1 to k. Note that concentration
                  changes are exaggerated to help Illustrate the sequence of calculations.
where £«./,»• is the average of the concentra-
tions of all points in cell (ij) after equations
22 and 28 were solved for all points for time
step k, and ACM-t is  the change in concentra-
tion caused  by  hydrodynamic dispersion,
sources, and sinks, as calculated in equation
40.
  Because the concentrations of points in a
cell vary about the concentration of the node,
the change in concentration computed at a
node using equation  40 cannot be applied
directly in all cases to the concentrations of
the points. If the change in concentration at
the node  (AC1J(»)  is positive, the increase is
simply  added  to  the point concentrations.
But if the concentration change is negative,
it is applied to points in that cell as a per-
centage  decrease  in concentration at  each
point that is equal to the percentage decrease
at the node. This technique preserves a mass
balance within each cell, but when a decrease
in concentration is computed for a node, it
will  also  prevent a possible but erroneous
computation of negative  concentrations  at
those points that had a concentration less
than that at the node.

              Stability criteria
   The  explicit  numerical solution of  the
solute-transport  equation  has a number of
stability  criteria  associated with it. These
may require that the time step used to solve
the flow equation be subdivided into a num-
ber of smaller time increments to accurately
solve the solute-transport equation.
   First, Reddell and  Sunada  (1970, p.  62)
show that  for  an  explicit  finite-difference
solution of equation 26 to  be stable,

-------
12
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
           (A*)'    (Ay)'   2
Solving equation 42 for At, we see that

                           0.5
              Min
          (over grid)
                      (AZ)
                                     (42)
                                    . (43)
  Because the solution to equation 26 is ac-
tually written as a set of N equations for N
nodes, the maximum permissible time incre-
ment is the smallest At computed for any in-
dividual node in the entire grid. The smallest
At will then occur at the node having  the
largest value of
               (A*)' (Ay)*
  Next consider the effects of mixing ground
water of one concentration with injected or
recharged  water of a  different  concentra-
tion, as represented by the source terms in
equation 39. The change in concentration in
a source node  cannot exceed  the difference
between the source concentration (C' . ) and
                                  »>/
the concentration in the aquifer  (C
-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                                    13
                                     (55)
where y  is the fraction of the grid dimen-
sions that particles will be allowed to move
(0
and
Because these criteria are governed by the
maximum velocities in the system, and since
the computed velocity of a  tracer particle
will  always  be  less  than or equal to the
maximum velocity computed at a node or cell
boundary, we have to check only the latter.
Substituting the grid velocities and solving
equations 56 and 57 for At results in
and
                   (V.)t
                                     (58)
                                     (59)
   If the time  step  used to solve  the  flow
equation exceeds the smallest of the time
limits determined by equations 43, 49, 58, or
69, then the time step will be subdivided into
the appropriate number of  smaller time in-
crements  required  for  solving the solute-
transport equation.

      Boundary and initial conditions

   Obtaining a solution to the  equations that
describe ground-water flow and solute trans-
port requires the specification of boundary
and initial conditions for the  domain of the
problem. Specifications for  solving the flow
equation must  be compatible  with  the solu-
tion of the solute-transport equation. Several
different types  of boundary conditions  can
be incorporated into  the  solute-transport
model. Two general types are incorporated
in this model;  these are constant-flux  and
constant-head conditions. These can be used
to represent the  real  boundaries  of an
aquifer as well as  to  represent  artificial
boundaries for the  model.  The  use of the
latter can  help to minimize  data  require-
ments and the areal extent of the  modeled
part of the aquifer.
  A constant-flux  boundary can be used to
represent  aquifer  underflow,  well  with-
drawals, or well injection.  A finite flux is
designated by specifying the flux rate as a well
discharge or injection rate for the appro-
priate nodes. A no-flow boundary is a spe-
cial  case of a  constant-flux boundary. The
numerical procedure used in this model re-
quires  that the  area  of interest  be sur-
rounded by a  no-flow boundary. Thus the
model will automatically  specify the outer
rows and columns of the finite-difference grid
as no-flow boundaries. No-flow boundaries
can also be located elsewhere  in the grid to
simulate  natural   limits  or  barriers  to
ground-water flow. No-flow boundaries are
designated by  setting the transmissivity
equal to zero at appropriate nodes, thereby
precluding the flow of water or dissolved
chemicals across the boundaries of the cell
containing that node.
  A constant-head boundary  in the model
can represent parts of  the aquifer where the
head will not change with time, such as re-
charge  boundaries or  areas beyond the  in-
fluence  of hydraulic stresses.  In this model
constant-head boundaries are simulated by
adjusting the leakage term (the last term on
the right side of equation 11) at the appro-
priate nodes. This is accomplished by setting
the  leakance coefficient  (K./m)  to a suffi-
ciently high value (such as 1.0 «-') to allow
the head in the aquifer at a node to be im-
plicitly  computed  as a value  that  is essen-
tially equal to the value of H,, which in this
case would be specified as the desired con-
stant-head altitude. The  resulting  rate of
leakage into or out of the designated con-
stant-head cell would equal the flux required
to maintain the head  in the aquifer at the
specified constant-head altitude.
  If a constant-flux or constant-head bound-
ary represents a fluid source, then the chemi-
cal concentration  in the source fluid (C')
must also be specified. If the boundary rep-
resents a fluid sink, then the concentration
of the produced fluid  will equal the concen-
                                                                                               m

-------
14
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
tration in the aquifer at the location of the
sink.
  Because solute transport directly depends
upon hydraulic and concentration gradients,
the head and concentration in the aquifer at
the start of  the simulation period must be
specified. The initial conditions can be deter-
mined from field data and (or)  from previ-
ous simulations. It is important  to note that
the simulation  results may be  sensitive to
variations or errors in the initial conditions.
In  discussing  computed  heads,  Trescott,
Finder, and Larson (1976, p. 30) state:
  If initial conditions are specified so that transient
flow is occurring in the system  at the start of the
simulation, it should be recognized that water levels
will change during the simulation, not only in re-
sponse to the new pumping stress, but also due to
the initial conditions. This  may or may not be the
intent of the user.

               Mass balance
   Mass  balance calculations are performed
after specified time increments to help check
the numerical accuracy and precision of the
solution. The principle of conservation  of
mass requires  that the  cumulative  sums of
mass inflows and outflows (or net flux) must
equal the accumulation of mass (or change
in mass stored). The difference between the
net flux  and the mass  accumulation  is the
mass residual  (Rn) and is one measure of
the numerical  accuracy of the  solution. Al-
though a small residual does not prove that
 the numerical solution  is accurate, a large
 error in the mass balance is undesirable and
 may indicate the presence of  a  significant
 error in the numerical solution.
   The  model uses  two methods to  estimate
 the error in the mass balance. Both are based
 on the magnitude of the mass  residual, Rm>
 which is computed from
               fl.=AM,-M,          (60)
 where
      AM. is the change in mass stored in the
             aquifer, M; and
      Mf  is the net mass flux, M.
    The  two mass  terms,  AM.  and Mf, are
 evaluated using the following equations:
                                  AM.
                                                               (61a)
                           where 
-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                        15
             Special problems

  There are a number of special problems
associated  with the  use  of the method of
characteristics to  solve the solute-transport
equation. Some of these problems are asso-
ciated  with the movement  and tracking of
particles, while other problems are related to
the computational transition between the
concentrations of particles within a cell and
the average concentration at that node. We
will next describe  the more significant prob-
lems and the procedures used  to minimize
errors that might result from them.
  One possible problem is related to no-flow
boundaries.  Neither  water  nor  dissolved
chemicals can be allowed to cross a no-flow
boundary. However, under certain conditions
it might be  possible for  the  interpolated
velocity at the location of a particle near a
no-flow boundary to be such that the particle
will be convected across the boundary during
one time increment. Figure 4 illustrates such
a possible  situation,  which arises from the
deviation between the  curvilinear flow line
and the linearly projected  particle path. If
a  particle  is  convected across  a  no-flow
boundary,  then it is relocated within the
aquifer by reflection across the boundary, as
also shown in figure 4. This correction thus
will tend to relocate the particle closer to the
true flow line.
  Fluid sources and sinks also require special
treatment.  Because they tend  to represent,
singularities in the velocity field, the use of
a central difference formulation  (eq  12) to
compute the velocity at a node  may indicate
zero or very small velocities at the nodes.
Therefore,  the velocity components  at a
source  or  sink node cannot be used for in-
terpolation of the velocity at a point within
or adjacent to that cell. To help maintain
radial flow to or from a sink or source, re-
spectively,  the velocities  computed on  the
boundaries of source or sink  cells are as-
signed  to that node. The appropriate bound-
ary velocities are  determined on the basis of
the quadrant of interest. This can be illus-
trated  by  referring  again  to figure 2. If  a
point is located in the southeast quadrant of
cell (f,j), the x velocity at node  (t,/)  would
     o
     A
       EXPLANATION
Node of finite-difference grid
Previous location of particle p
Computed new location of particle p
Corrected new location of particle p
Flow line and direction of flow
Computed path of flow

Zero transmissivity (or no-flow boundary)
   Figure 4.—Possible movement of particles near
       an impermeable (no-flow) boundary.
be set equal to V.(H_H j>(and the y velocity to
^»(U+W)'  Corresponding  adjustments are
made for points in other quadrants, so that
the magnitude and  direction of velocity  at
the node are not fixed  for a given time in-
crement, but depend on the relative location
of the point of interest within the cell. A
similar approximation is made when a point
of interest is located in a  cell adjacent to a
source or sink. Thus, if the same point, p, in
figure  2 were  located in an  unstressed cell
but the adjacent cell (t+l,j)  represented a
source  or  sink,  then  the  y  velocity  at
node  (t'+l,;)  would be  approximated  by
V,(i+l i+V4) in order  to estimate the y velocity
at point p. A  corresponding approximation
for the x velocity at node (t,; + l) would be
made using ^t(i+^J+D^  a source  or sink
were located at (t,/+1).
  The maintenance  of a reasonably uniform
and continuous spacing of points requires
special treatment in areas  where sources and
sinks dominate the flow field. Points will con-
tinually move out of a  cell that represents a
source, but few or none will move in to  re-

-------
16
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
place them and thereby maintain a continuous
stream of points.  Thus, whenever  a point
that originated in a source cell moves out of
that source cell, a new point is introduced
into the source cell to replace it. Placement
of new points in a source cell is compatible
with and analogous to the generation of fluid
and solute mass at the source.
  The procedure  used to replace points in
source  cells that  are adjacent  to  no-flow
boundaries is illustrated in figure 5. Here a
steady, uniformly spaced stream of points is
maintained by generating a new point at the
same relative position in the source cell as
the new position  in the adjacent cell of the
point that left the source cell. As an example,
point 7 was convected from  cell (t-1,;) to
cell (i,j). So the replacement point (22) was
placed at a location within cell (t-1,;) that
is identical to the  new location of point 7
within cell  (t,/).
   The procedure used to  replace points in
source cells that  lie within the aquifer and
not adjacent  to a no-flow boundary is  illus-
trated in figure 6. Here a  steady, uniformly
spaced stream of particles is maintained by
generating a new point in the source cell at
the original  location of the point that left
the source cell.  When a relatively strong
                             source is imposed on a relatively  weak  re-
                             gional flow field, as illustrated in figure  6a,
                             then radial flow will be maintained in the
                             area of the source, and all initial and replace-
                             ment  points will move  symmetrically away
                             from node  (t,/). For example, after point 7
                             moves from cell (t,;) to (t+1, ;-l), the re-
                             placement point (18) is positioned at time k
                             in cell ({,;')  at the same location as the ini-
                             tial position  of  point  7. Although the  re-
                             placement procedure illustrated earlier by
                             figure 5 would work just as well for the case
                             illustrated in figure 6a,  it would not be satis-
                             factory for the  situation presented in figure
                             6b, which illustrates the imposition of a rela-
                             tively  weak  source in  a relatively strong
                             regional flow field. In this  case the velocity
                             distribution within  the source cell does  not
                             possess radial  symmetry, and the velocity
                             within the upgradient part of the source cell
                             is much lower than the velocity within  the
                             downgradient part of  the source cell.  Re-
                             placement of points at original locations in
                             source cells, as  illustrated in figure 6b, will
                             maintain a steady stream of points leaving
                             the source cell in proportion to the velocity
                             field. However, the use of the procedure illus-
                             trated in figure 5 for the case presented in
                             figure 6b would result in the accumulation of
                    lima *-1
                                                                     tim *
                          -3 —-4
                           1J    14
                                                              *»
                                                                         °2    °3
                                                                         °7    °8
                                                                                    "19
                                         EXPLANATION
                                   •  NodaoKinita-dlttaranc* 0/id
                                   •p Initial location of panic to •
                                   Of Naw location o* panicla a
                                   Aa Location of raplacamant partlela •
                                 n
                   Conatant-haad aewca


                   Zaro oanamitaivity (or no-flow bondary)
               Hgur* 5.—Replacement of points in source cells adjacent to a no-flow boundary.

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                         17
                          tiim *-1
                        \  Vl /
                        \    4
                                     / M./-1
                                  12"
                                                     10°
                                                 •Pis
           lima*

         °2   V/-1 °3
                                                              /
                                                                       °M
                                                           time *
                              i
                       £A
                                                                03
                      012
                                            (b)
                                        EXPLANATION

                                   •  Nodi of finite -diffaranca ftrid
                                   •p Initial location of particla p
                                   Op Naw location of paftiela p
                                   Ap Location of laplacamant panlela p
                                     Fluid aourca
                Figure 6.—Replacement  of  points  In source  cells  not  adjacent  to  a
                  no-flow  boundary for  negligible  regional flow (a)  and for  relatively
                  strong regional flow (b).
points in the low-velocity area of the source
cell (i,/), with few points being replaced into
the  high-velocity  area,  where  convective
transport is the greatest.
  Although we normally expect points to be
convected out  of source cells, figure 6b also
demonstrates the possibility that points may
sometimes enter a source cell. This can also
occur when two  or more source cells of dif-
ferent strengths are adjacent to each other.
An erroneous multiplication of points might
then result if  points that did not originate
in a particular source cell are replaced when
they in turn are convected out of that source
cell. Therefore, points leaving a  source cell
are replaced only if they had originated in
that source cell.
  Hydraulic sinks also require some  special
treatment. Points will continually move into
a cell representing a strong sink,  but  few or
none will move out. To avoid the resultant
crowding and  stagnation of tracer  points,
any point moving into a sink  cell  is removed
from the flow field after the calculations for
that time  increment have  been  completed.
The numerical removal of points which enter

-------
18
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
sink cells is analogous to the withdrawal of
fluid and solute mass through the hydraulic
sink. The combination of creating new points
at sources and destroying old points at sinks
will tend to maintain the total  number of
points in the flow field at a nearly constant
value.
   Both the flow model  and  the transport
model assume  that sources and sinks act
over  the entire cell  area  surrounding  a
source or sink node. Thus, in effect,  heads
and concentrations  computed at source or
sink nodes represent average values over the
area of the cell. Part of the total concentra-
tion change computed at  a source node repre-
sents mixing between the source  water at
one concentration and the ground water at a
different concentration  (eq  39). It can be
shown  from the relationship between the
source concentration (C'..k ) and the aquifer
concentration  (Cw.*_,), as  indicated  by
equation 44, that the following constraints
generally must be met in a source cell :
 and
                for
 If  it is  assumed that the sources act over
 the area of the source cell and that there is
 complete vertical mixing, then  these  same
 constraints should also apply to all points
 within the cell. Because of the possible devia-
 tion  of  the  concentrations of  individual
 points within a source cell from the average
 concentration, the change  in  concentration
 computed at  a source  node (AC(i>.k) should
 not be applied directly to each of the points
 in  the cell. Rather, at  the end  of each time
 increment the concentration of each point in
 a source cell is updated by setting it  equal
 to  the final  nodal concentration.  Although
 this may  introduce a small amount of  nu-
 merical dispersion  by  eliminating possible
 concentration variations within the area of a
 source cell, it prevents the adjustment of the
 concentration at any point in the source cell
 to a value that would violate the constraints
 indicated by  equation 65.
    In  areas of divergent flow there may be a
 problem because some cells can become void
                            of points where pathlines become spaced
                            widely apart. This would result in a calcula-
                            tion  of zero change in concentration at a
                            node due to convective transport,  although
                            the nodal concentration would  still be  ad-
                            justed for changes caused by hydrodynamic
                            dispersion  (eq 28). Also, some numerical
                            dispersion is generated at nodes in and  ad-
                            jacent to the cells into which the convective
                            transport of solute was underestimated  be-
                            cause of the resulting error in the concentra-
                            tion gradient. This might not cause a serious
                            problem if only a few cells in a large grid
                            became void or if the voiding were transitory
                            (that is, if upgradient points were convected
                            into void cells during later or  subsequent
                            time   increments).  Figure  6a   illustrates
                            radial flow, which represents the most severe
                            case of divergent flow. Here it  can be seen
                            that when  four points per cell  are used to
                            simulate convective transport, then  in  the
                            numerical procedure four  of the  eight sur-
                            rounding cells would erroneously not receive
                            any  solute  by convection from the adjacent
                            source. If  eight  points per cell were used
                            initially, then at a distance of two rows or
                            columns from the source only 8 of 16 cells
                            would  be on pathlines originating  in  the
                            source cell. So, while  increasing the  initial
                            number of points per cell would help, it is
                            obvious that for purely radial flow, an  im-
                            practically  large initial number of  points
                            per cell would be required to be certain that
                            at least one particle pathline passes from the
                            source through every cell in the grid.
                               The problem of cells becoming void of par-
                            ticles can be minimized by limiting the num-
                            ber of void cells to a small percentage of the
                            total  number  of cells that  represent  the
                            aquifer. If the limit is exceeded, the numeri-
                            cal solution to the solute-transport equation
                            is terminated at the end of that time incre-
                            ment  and the "final" concentrations at that
                            time are saved.  Next the problem is reini-
                            tialized at  the time of termination by re-
                            generating  the  initial particle distribution
                            throughout the grid and assigning the "final"
                            concentrations at the time of termination as
                             new "initial" concentrations  for nodes  and
                             particles. The solution to the solute-transport

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                                    19
equation is then simply continued in time
from this new set of "initial" conditions until
the total simulation period has elapsed. This
procedure preserves the mass balance within
each cell but also introduces a small amount
of numerical dispersion by eliminating vari-
ations  in   concentration within  individual
cells.
  To help  minimize the amount of numeri-
cal dispersion resulting from the  regenera-
tion of points, the program also includes an
optimization routine that attempts to main-
tain an approximation of the previous con-
centration  gradient within a cell.  The opti-
mization routine aims to meet the  following
constraints:
    C*   is the concentration of  the  nth
           point in cell ({,/), M/L8;
    Nf   is the total  number of points ini-
           tially placed in a cell; and
    Ci,m  is the concentration at node (l,m),
           which represents a cell  adjacent
           to (t,?)  and on a line that starts
           at ({,/)  and extends through the
           coordinates of the point (n) of
           interest, as illustrated in figure
           7, M/L».

Note  that equation  66a  simply  indicates
that a mass balance must be preserved  in a
cell regardless of the range in variation of
point  concentrations within the cell. Equa-
tions  66b and c indicate that the concentra-
tion of any point must lie between Cu and
the concentration  at  the node adjacent to
particle n. The coordinates of the  adjacent
node would take on values of l=i or Z=t±l
and m=; or m=;±l. For example, figure 7
shows that for point 2, the coordinates (l,m)
would equal ({,;'-!), while for point 3, (l,m)
would equal   (t+l,/-l). The optimization
                                                               .'•/-1
                                                            •4 *u -6
                                                            •e "7 -a
                                                          EXPLANATION
                                                   •   Node of finite -difference grid
                                                   •„ Location of particle n

                                                 Figure 7.—Relation  between possible  Ini-
                                                   tial locations of points and indices of ad-
                                                   jacent nodes.


                                             routine is written so  that if equations 66a-c
                                             cannot be satisfied simultaneously for node
                                             (i,j) within two iterations, then to avoid fur-
                                             ther computational delay all C* are simply
                                             set equal to Ctil.
                                                    Computer  Program

                                               The computer program serves as a means
                                             of translating the numerical algorithm into
                                             machine  executable  instructions.  The pur-
                                             pose of this chapter is to describe the overall
                                             structure of the program and to  present a
                                             detailed  description of  its  key  elements,
                                             thereby providing a link between the numeri-
                                             cal methods and the computer code. We hope
                                             that this link will make it easier for the
                                             model user to understand and, if necessary,
                                             modify  the  program.  The  FORTRAN  IV
                                             source program developed for this model is
                                             listed in attachment I and  includes almost
                                             2,000 lines. For reference purposes columns
                                             73-80 of each line contain a label that is
                                             numbered  sequentially  within  each   sub-
                                             routine. The  definition of selected variables
                                             used in the program is presented  in attach-
                                             ment II; this  glossary therefore also  serves
                                             as a key for relating the program variables

-------
20
TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
to their corresponding mathematical terms.
The computer program is compatible with
many scientific computers; it has been suc-
cessfully run on Honeywell, IBM, DEC, and
CDC computers.


       General program features

  The program is segmented  into a  main
routine and eight subroutines. The name and
primary purpose of each segment are  listed
in Table 1. Each  program segment will be
described in more detail in later sections of
this chapter.

Table 1.—List of subroutines for solute-transport model
    Name
                        ParpoM
MAIN	Control execution.
PARLOD ..Data input and initialization.
ITERAT ...Compute head distribution.
GENPT	Generate or reposition particles.
VELO	Compute hydraulic gradients, velocities,
            dispersion equation coefficients, and
            time increment for stable solution to
            transport equation.
HOVE	Move particles.
CNCON	Compute change in chemical concentra-
            tions and compute mass balance for
            transport model.
OUTPT	Print head distribution and compute
            mass balance for flow model.
CHMOT —Print concentrations,  chemical  mass
            balance, and observation well data.
   The major steps in the calculation pro-
 cedures are summarized in figure  8, which
 presents a simplified flow chart of  the over-
 fill structure of the computer program. The
 flow chart illustrates that the tracer particles
 may  have to  be moved more  than once to
 complete  a given time step. In other words,
 the time step used to implicitly solve the flow
 equation  may have to be  subdivided into a
 number of smaller  time increments for the
 explicit   solution   of  the solute-transport
 equation. The maximum time increments al-
 lowable for the explicit calculations are com-
 puted automatically by the model.  Thus, the
 model  user cannot specify  an erroneously
 large increment or  an inefficiently small in-
                           crement for  solving  the  solute-transport
                           equation. For transient flow problems, some
                           discretion is still required in the specifica-
                           tion of the initial time step and of the time-
                           step multiplier, as  discussed  by Trescott,
                           Finder, and Larson (1976, p. 38-40).
                              The general program presented here  is
                           written to allow a grid having up to 20 rows
                           and 20 columns. Because the numerical pro-
                           cedure requires that the outer rows and col-
                           umns  represent no-flow  boundaries, the
                           aquifer itself is then  limited  to maximum
                           dimensions of 18 rows and 18 columns. If a
                           problem requires a larger grid, then the ap-
                           propriate arrays must be redimensioned ac-
                           cordingly.   These  arrays  are   contained
                           in COMMON statements PRMK, HEDA,
                           HEDB, CHMA, CHMC, and DIFUS,  and in
                           DIMENSION  statements  on lines   C170,
                           G200, H140, and 1160.
                              The program allows the specification  of
                           one pumping well  per node. The wells can
                           represent injection  (recharge) or withdrawal
                            (discharge).  If more  than one  well  exists
                           within the area of a cell, then the flux spe-
                           cified for that node should represent the net
                           rate of injection or withdrawal of all wells
                           in that cell. The model assumes that stresses
                           are constant with time during each pumping
                           period (NPMP).  But the total  number  of
                           wells, as well as their locations, flux rates,
                           and source concentrations, may be changed
                           for successive pumping  periods. The pro-
                           gram  also allows the  specification of obser-
                           vation wells at as many as five nodes in the
                            grid.  For nodes that  are designated as ob-
                            servation wells, at the end of the simulation
                            period or after every 50 time increments the
                            model will print a summary table of the head
                            and concentration  at  the previous time  in-
                            crements.
                              The program also includes a node identi-
                            fication array (NODEID), which allows cer-
                            tain nodes or zones  to be  identified  by a
                            unique code number.  This feature can save
                            much time in the preparation of input data
                            by easily equating each code number with a
                            desired boundary condition, flux, or source
                            concentration.

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                      21
                     START
              \ READ GEOLOGIC.
               \ HYOROLOGIC.&
               \   CHEMICAL
                \     INPUT
                     DATA
j
GENERATE UNIFORM
CtlCTDlDl ITl/"*M f\C
I/IS IKIbU I ION Or
TRACER PARTICLES

COMPUTE DISPERSION
EQUATION COEFFICIENTS


DETERMINE LENGTH
OF TIME INCREMENT
FOR EXPLICIT
CALCULATIONS
*
MOX/P PARTlPI FQ

j
ftFKlFRATF MF\I/ PARTlPl FC
OR REMOVE OLD
DADTIPI CC AT
rMn 1 lULfca A I
APPROPRIATE BOUNDARIES


COMPUTE AVERAGE
CONCENTRATION IN EACH
FINITE-DIFFERENCE CELL
|
COMPUTE EXPLICITLY
THE CHEMICAL
NODES

|
\SUMMARIZE AND /
DQlfUT DCCIIITC /
rnlN 1 KcbULTS /
i

1 Ji
( STOP J

COMPUTE HYDRAULIC
»b fltl A FMCKJTC CfiO
ONE TIME STEP
|
— COMPUTE
* GROUND-WATER
t/Cl rtfMTICG
VcLOUITIcS







|
ADJUST CONCENTRATION
OF EACH PARTICLE


COMPUTE
MASS BALANCE
t'
^



.
^


JL
S ,x 	 EN° O^X NO






















to



iO



































































            Plgur* 8.—Simplified flow chart Illustrating the major steps In the calculation
                                      procedure.
           Program segments
                  MAIN
  The primary purpose of the MAIN routine
is to control the overall execution  sequence
of the program. Subroutines for input, ex-
ecution, and output are called from MAIN
and the elapsed time simulated is compared
with  the  desired  total simulation  period.
Also,  lines A500-A580 serve to store  (or

-------
22
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
record)" observation well data for transient
flow problems.

           Subroutine PARLOD

  All input data are read through subroutine
PARLOD. These  data define the properties,
boundaries,  initial conditions, and stresses
for the aquifer, as well as spatial grid and
time-step factors. The values of many vari-
ables are also initialized here. After the data
are read, some preliminary calculations are
made,  such as  (1)  determining time incre-
ments for the flow model (lines B780-B890),
(2)  computing the  harmonic mean Jrans-
missivities in the  x and y directions (B1670-
B1800),  (3)  adjusting transmissivity for
anisotropy  (B1810-B1820), (4)  computing
iteration  parameters   (B1840-B1910  and
B2880-B2980),and (5) checking for possible
inconsistencies among  the  input  data
(B3140-B3290).  A printout is also provided
of all input data so that the data may be re-
checked and each run identified.

            Subroutine ITERAT

   This subroutine  solves  a finite-difference
approximation of the flow equation (eq 11)
using  an iterative ADI procedure. The ma-
trix generated by  the  finite-difference ap-
proximation  is solved  using  the Thomas
algorithm,  as described by von Rosenberg
 (1964, p. 113). Row calculations are made in
lines C270-C610,  and column calculations are
made  in lines  C630-C970. The calculations
are assumed to have converged on a solution
if the maximum  difference at all nodes be-
tween heads computed along rows and heads
computed along columns is less than the spec-
ified tolerance. Convergence is checked on
lines C940-C950. Note that here (for ex-
ample, lines C380,  C700,  C930, and C1150)
and in other subroutines the thickness array
 (THCK) is used to check whether a node is
 in the aquifer.
   It should also  be noted here that the flow
model, as written, assumes that  the trans-
 missivity of the aquifer is independent of the
 head  (or saturated thickness) and remains
 constant with time. If this assumption is not
                           appropriate to the particular aquifer system
                           being modeled, then the solution algorithm
                           presented in this subroutine should be modi-
                           fied  accordingly. For example, flow models
                           published by Prickett and Lonnquist (1971,
                           p. 43-45) and Trescott, Pinder, and Larson
                           (1976) include such a modification.
                              All parameters involved in the calculation
                           of heads are defined as double precision vari-
                           ables and all calculations  involving  these
                           parameters are  performed in double pre-
                           cision. The number of double precision vari-
                           ables and operations can  be  reduced sig-
                           nificantly if the program is to be executed on
                           a  high-precision scientific computer, thereby
                           improving the efficiency of the model by re-
                           ducing computer storage  requirements and
                           execution time.
                              The iterative ADI procedure used to solve
                           the  finite-difference  equations  is not neces-
                           sarily the best possible solution technique for
                           all problems. For example, it may be difficult
                           to obtain a solution using the iterative ADI
                           procedure for cases of steady-state flow when
                           internal nodes in  the grid have zero trans-
                           missivity and for  cases in which the trans-
                           missivity is highly anisotropic. In such  cases,
                           a strongly implicit  procedure, such as the
                           one documented  by  Trescott, Pinder, and
                           Larson (1976), should be substituted for the
                           solution algorithm contained in subroutine
                           ITERAT.

                                        Subroutine GENPT
                              The   primary  purpose   of  subroutine
                           GENPT  is  to generate  a uniform initial
                           distribution of tracer particles throughout
                           the  finite-difference grid. This is done  either
                           at the start of a simulation period or  at an
                           intermediate time when too many cells have
                           become void of particles.  In the latter case,
                           the program attempts  to preserve an  ap-
                            proximation of the previous  concentration
                           gradient within  each  cell  (lines  D1420-
                            D2040).
                              The placement of particles is accomplished
                            in lines  D510-D1410. The program allows
                            the placement of  either four, five, eight, or
                            nine particles per cell. Of course each option
                            will result in a slightly different geometry

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                       23
 Figure 9.—Parts of finite-difference grids showing
  the Initial geometry of particle distribution for the
  specification of four (A), five (8), eight (C), and
  nine (0) particles per cell.

and density of points, as illustrated by figure
9. The most regular or uniform patterns are
produced when four or nine particles per cell
are specified. If a different number of par-
ticles per cell or a different placement geom-
etry are desired,  this subroutine could be
modified accordingly.
  As  particles  are  moved  or  convected
through the grid during the calculation pro-
cedure, there is a need to remove particles at
fluid  sinks  and create  particles  at fluid
sources. A buffer array  (called LIMBO)  is
created  on lines D430-D480 that  contains
particles that can be added later to the grid
at sources and that also contains space to
store particles removed at sinks or discharge
boundaries.

             Subroutine VELO
  Subroutine VELO accomplishes three  ob-
jectives. First, it computes the flow velocities
at nodes and  on cell boundaries  by solving
equations having the form of equations 12
and 13. The velocities are computed on lines
E420-E680. Second, the  dispersion equation
coefficients are calculated. These  coefficients
represent terms factored out of equations 37
and 88, as follows:
DISP(IX,IY,1) = (bD..),<+HJ)/(A*)*  (67a)
DISP(IX,IY,2) = (bDn) ,tt^,/(*V)*  (67b)
                                                                                (67c)
DISP(IX,IY,8)-
DISP(IX,IY,4).
Note  that  each dispersion coefficient  (/)„,
£>,,, !?„, D,z) is computed on cell boundaries
using the  relationships expressed in equa-
tions  8-10. Therefore,  the equation  coeffi-
cients computed by equation 67 are stored
as forward values from the indicated node in
the DISP array. Third, this subroutine com-
putes (on  lines E1050-E1240  and E1800-
E1930)  the  minimum  number of particle
moves (NMOV) required to solve the trans-
port equation for the given time step so that
the maximum time increment for the trans-
port equation solution will not exceed any of
the criteria indicated by equations 43,  49,
58, and 59.

             Subroutine MOVE

   Although this subroutine has only one main
function, which is to move the tracer par-
ticles in accordance with equations 22 and
23, it is the longest and perhaps the most
complex segment of the program. The com-
plexities are mainly introduced by the treat-
ment of particles at the  various  types of
boundary  conditions. To help  illustrate  the
calculation procedure followed within sub-
routine MOVE, a flow chart is presented in
figure 10. The numbers in the flow chart in-
dicate the  corresponding lines  in subroutine
MOVE  where the  indicated  operation  is
executed.
   If a node represents a fluid source or sink,
then particles must be respectively created or
destroyed  in these  cells.  If  the  value  of
pumpage  (REG)  at a  node  dees not equal
zero, then the node is assumed to represent
either a fluid source  (for REC<0) or a fluid
sink  (for  REC>0). Recharge or discharge
can also be represented by the  RECH array.
But it is assumed that this  type of flux is
sufficiently diffuse so that it does not induce
areas or points of strongly divergent or con-
vergent flow and therefore particles need not
be created or destroyed at these nodes. Note
that here and in other subroutines the pres-
ence of a constant-head boundary is  tested
by checking the  value of leakance (VPRM)

-------
                           COMPUTE ELAPSED TIME
                              AT START Of NEXT
                             PARTICLE MOVEMENT
                           	F210
                              START NEXT MOVE
                                           F2I
1
SELECT NEI

r PARTICLE L
F 3101
          YES
   DETERMINE «-»
COORDINATES OF NODE
  WHERl PARTICLE IS
    LOCATED FM0.4TO
                             DETERMINE IN WHICH
                           QUADRANT OF CELL THE
                            PARTICLE IS LOCATED
                           	F MO-1740
      IS
   U ADMAN
LOCATED IN OK
ADJACENT TO A
  SOURCE OR
    SINK?
SET VELOCITY AT
SOURCE /SINK NODE
•VELOCITY ON
ADJACENT CELL
BOUNDARY


USE BILINEAR
INTERPOLATION TO
COMPUTE X AND V
VELOCITY OF PARTICLE
F 1770 -2030
COMPUTE X AND Y
COORDINATES OF CELL
AT NEW LOCATION OF
PARTICLE F20M-2110


COMPUTE DISTANCE
MOVED IN > AND Y
DIRECTIONS
F 2040 -2050
                             COMPUTE DISTANCE
                             PARTICLE TRAVELED
                              KYOND BOUNDARY
                            	F2150-2320
                           RELOCATE PARTICLE INTO
                            AQUIFER BY REFLECTION
                             ACROSS BOUNDARY
                            	F2210-23 BO
                                                                                     DID PARTICLE
                                                                                  ORIGINATE IN THAT
                                                                                     SOURCE CELL ?
                                                          If SOURCE CELL
                                                           TED ALONG EDGE
                                                          OF AQUIFER 7
                                                             F2720
                                                               7
 SUM NUMBER OF
 PARTICLES AND
CONCENTRATIONS
   IN CELL AT
 NEW LOCATION
        F2420-2430
CREATE NEW PARTICLE
          F2S80-2690
SOURCE OH SINK 7

   2480-2620
                                                                                                                PLACE NEW PARTICLE IN
                                                                                                                  OLD CELL AT SAME
                                                                                                                  RELATIVE POSITION
                                                                                                                 AS OLD PARTICLE IN
                                                                                                                      NEW CELL
                                                                                                                          F3420-3460
                                                                                        ISOLD
                                                                                     OCATION IN A
                                                                                   SOURCE CELL OR
                                                                                     A SINK CELL >
                                                                                       2480-2
                                                           HAS PARTICLE
                                                           CHANGED CELL
                                                             LOCATION?
                                                                                                                CREATE NEW PARTICLE
                                                                                                                	   F2580-2680
                                                                                                                 PLACE NEW PARTICLE
                                                                                                                AT ORIGINAL LOCATION
                                                                                                                  OF OLD PARTICLE
                                                                                                                          F 2770-3380
                                                                                        S N
                                                                                      LOCATION
                                                                                    IN A PUMPING
                                                                                  OR CONSTANT-HEAD
                                                                                        SINK?
                                                                                        580-36
                                                                                   REMOVE PARTICLE
                                                                                      FROM GRID
                                                                                           F3640-1710
                                                                                      CALL SUBROUTINE
                                                                                     CNCON TO COMPUTE
                                                                                    NEW CONCENTRATIONS
                                                                                                  F39SO
                                                                                                            STORE OBSERVATION WELL
                                                                                                            DATA FOR STEADY -FLOW
                                                                                                                 CASES
                                         3
                                         o
                                         a
                                                                                                                                                          M
                                                                                                                                                          CO
                                                                                                W
                                                                                                w
                                                                                                09
                                                                                                O
                                                                                                CS
                                                                                                8
                                                                                                w
                                                                                                CO
                                                                                                                                                      co
                                                                                                                                                      H
                                                                                                                                                      o
                                                                                                                                                      z
                                                                                                                                                      CO
    Figure 10.—Generalized flow chart of subroutine MOVE. Numbers indicate line numbers where the operation Is executed.

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                      25
at each node. If VPRM exceeds 0.09, it is as-
sumed that the node represents a constant-
head boundary condition and is treated as a
fluid source or  sink accordingly. At  a con-
stant-head node the difference in head be-
tween the aquifer and the source bed is used
to determine whether  the node represents a
fluid  source  or sink  (for  example, lines
F2500-F2520).

           Subroutine CNCON
  This subroutine computes the change in
concentration at each  node and at each par-
ticle for the given time increment. Equation
39, which denotes the change  in concentra-
tion resulting from sources, divergence of
velocity, and changes in saturated thickness,
is solved on lines G350-G610.  On the G520
the value of the storage coefficient is checked
to determine whether  the aquifer is confined
or unconfined. It assumes that if S<0.005,
then the aquifer is confined and
If S^O.005, the model assumes  that
 •= 'dh/'dt. If this criterion is  not appropriate
to  a particular aquifer  system, then line
G520 should be  modified accordingly. The
change  in  concentration caused by  hydro-
dynamic dispersion   is computed  on lines
G640-G770 as indicated by equations 37 and
38.
   The nodal changes in concentration caused
by convective  transport are  computed on
lines G850-G940. The number of cells that
 are void of particles  at the new time level
are also counted in this set of statements on
 lines G880-G910, and then compared  with
 the critical number of void cells (NZCRIT)
 to  determine  if  particles should be regen-
 erated at  initial positions before the next
 time level is started (lines G960-G1020).
   The new (time level k) concentrations at
 nodes are computed on the basis of the previ-
 ous concentration  at time  fc-1 and  the
 change during fc-1 to k. The  adjustment at
 nodes is accomplished on lines G1060-G1180,
 while the  concentration  of particles is ad-
 justed on lines G1210-G1S60.
   A mass balance for the solute  is next com-
 puted (lines  G1400-G1730) at the end of
 each time increment.  In computing the mass
of solute withdrawn or leaking out of the
aquifer at fluid sinks, the concentration at
the sink node is assumed to equal the nodal
concentration computed at time level fc-1.

            Subroutine OUTPT
  This subroutine prints the results of the
flow model calculations. When  invoked, the
subroutine prints  (1) the new hydraulic
head matrix (lines H190-H260), (2) a nu-
meric map of head values  (H300-H890), and
(3)  a drawdown map (H510-H710). This
subroutine also computes a mass balance for
the flow model and  estimates  its  accuracy
(H420-H820). A mass balance is performed
both for cumulative volumes since the initial
time and for flow rates during the present
time  step. The  mass balance results are
printed on lines H840-H930.

            Subroutine CHMOT
   This subroutine prints (1)  maps of con-
centration (lines 1250-1380), (2) change in
concentration from initial conditions (1440-
1580), and (3) the results of the cumulative
mass balance for the solute  (1670-1860).
The accuracy of the chemical mass balance is
estimated on lines 1610-1660 using  equations
62 and 64.  The former  is  not computed if
there was no change  in the total mass of
solute stored in the aquifer. The latter is not
computed if the initial concentrations  were
zero everywhere. Lines 1890-11140 serve to
print the head  and concentration data  re-
corded at observation  wells. These data are
 recorded after each time step for a transient
flow problem and  after each particle move-
ment for a steady-state flow  problem. The
 data are  printed after every 50 time incre-
 ments and at the  end of the simulation
 period.
       Evaluation  of  Model

  Comparison with analytical solutions

   The accuracy of the numerical solution to
 the solute-transport equation can be evalu-

-------
26
                 TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
ated in part by analyzing relatively simple
problems for which analytical solutions are
available and then comparing the numerical
calculations with the  analytical  solution.
Figure 11 presents such a comparison  for a
problem of one-dimensional steady-state flow
through  a  homogeneous  isotropic porous
medium. The analytical solution is obtained
with  the  following  equation presented  by
Bear  (1972, p. 627) :
      C(x,t)-C0
        C.-C,
                                     (68)
where
    erfc
           is the complimentary error func-
             tion, and
    q-.iV  is the specific discharge, LT~l.
Bear  (1972, p. 627)  shows that equation 68
is subject to the following initial conditions:
               -oo
-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                  27
equation 68.  In the analysis of one-dimen-
sional  test problems, it was assumed that
porosity equals 0.35, velocity equals 3.0x10-'
ft/a (9.1 x 10-" m/s), and time equals 10.0
days.
  As shown  in figure  11, comparisons be-
tween the analytical and numerical solutions
were made for two  different values of dis-
persivity. For the  higher dispersion there
was essentially an exact agreement between
the two curves. In the case of low dispersion,
there is a very small difference at some nodes
between the concentrations computed analyt-
ically and those computed numerically. This
difference is caused primarily by the error in
computing the concentration at a node as the
arithmetic average of the concentrations  of
all  particles located  in that cell. This is not
considered to  be a serious problem since this
error is not cumulative. Also note in the case
of low dispersion that the grid spacing (10
ft or  3.05 m)  was coarse relative  to the
width  of the breakthrough  curve between
concentrations of 0.05  and 0.95.  Neverthe-
less, the numerical  model still  accurately
computed the shape and position of the front.
  In  computing  the  numerical solutions
shown in figure 11 the program was executed
using  nine  particles  per cell  and with
CELDIS = 0.50 (y in equations  54-55). The
10-day simulation  required  52 time incre-
ments and used about 40 seconds of  cpu  on
a Honeywell 60/68 computer.
  An analytical solution is also available for
the problem of plane radial flow in which a
well continuously injects a tracer at constant
rate qK and constant concentration C0. Bear
 (1972, p. 638)  indicates that the following
equation is appropriate for this problem (al-
though there  are some  limitations discussed
by  Bear):
            C  1 ^ (r'/2-Gt )
          	=-prfr{—       }       (69)
          Co  2    lV4/3«1P )
 where

     G

     r
       ••Vr;
is the radial distance from
  the center of  the well,
  L; and
    r= (ZGt) *  is the average radius of the
                  body of  injected water,
                  L.
   Again, the general computer program had
to be somewhat modified  to permit a  suit-
able comparison  to  be made between the
analytical solution and the numerical model.
One change involved the direct calculation of
velocity at any point based on its  distance
from the well using the following equation:

                                     (70)

The other significant change was  made in
subroutine GENPT to allow  the initial place-
ment of 16 particles per cell, rather than the
present maximum of 9. In the analysis of
test problems for radial flow, it was assumed
that porosity equals 0.35, the injection rate
 (gw) equals 1.0 ftVs (0.028 mVs), saturated
thickness equals 10.0 ft (3.05 m), and longi-
tudinal dispersivity equals 10.0 ft (3.05 m).
  The application of the method of character-
istics,  which was  written  for two-dimen-
sional Cartesian  coordinates, to  a problem
involving radially symmetric divergent flow
represents a severe test of the model. Never-
theless, it can be  seen in figure 12 that there
is good agreement between the analytical and
numerical solutions  after  both  relatively
short and long times. However, the presence
of some numerical dispersion is evident, par-
ticularly for the longer time. The numerical
dispersion is introduced in  part during  the
regeneration of particles after the number of
cells void of particles has exceeded the criti-
cal number.  The  geometry of initial particle
placement minimized this problem in cells
that lay in  the same row or column of  the
grid as the injection well. The circles in  fig-
 ure 12, which indicate concentration values
 computed at these nodes, agree closely with
the analytical solution. The greatest errors
 occur at nodes on radii  from  the injection
 well that are neither parallel to nor 45° from
 the main axes of the grid.  These results in-
 dicate that this Cartesian coordinate  model
 is  not best suited for application  to purely
 radial flow  problems. However, if radially
 divergent flow is limited  to areas of several

-------
 28
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
  1.0
  0.9
  0.8
  0.7
P 0.4
  0.2
  0.1
  0.0
                                                     EXPLANATION
                                                — Analytical solution
                                                  Numerical solution:
                                                • Nodra on radii pinll*) too/Id
                                                • Nods* on radii 45°to grid
                                                A NodM on intennodlata radii
              100
                         200
                                   300         400         600

                                      RADIAL DISTANCE. IN FEET
                                                (00
                                                           700
                                                                      800
Figure 12.—Comparison between analytical and numerical solutions for dispersion  in plane radial steady-state flow.
  rows and columns  within a more uniform
  regional flow field, the model will accurately
  compute concentration distributions. To ap-
  ply the method of characteristics to a prob-
  lem of plane radial  flow, it would be best to
  rewrite the program in a system of radial
  coordinates, which  should improve the ac-
  curacy for those problems to the same order
  shown in  figure 11 for the analysis of one-
  dimensional flow.
             Mass balance tests

    The accuracy and precision of the numeri-
  cal solution can  also be partly evaluated by
  computing the magnitude of the error in the
  mass  balance. The mass  balance error will
  depend on the nature of the problem and will
  vary from one time step to the next. During
  the development of the program, the model
  was  applied to a variety  of hypothetical
  solute-transport problems to  assure its flexi-
                             bility, transferability, and accuracy under a
                             wide range of conditions. To illustrate the
                             range in mass balance errors that might be
                             expected and some of the factors that affect
                             it, several of these problems are presented
                             here.

                              Test problem 1—spreading of a tracer slug
                               The first test described here was designed
                             to evaluate the accuracy of simulating the
                             processes of convective transport and  disper-
                             sion  independent of the effects of chemical
                             sources. Thus, a slug of tracer was initially
                             placed in four cells of a grid whose boundary
                             conditions generated a steady-state flow field
                             that was moderately divergent in some places
                             and moderately convergent in  other  places,
                             as illustrated in figure 13. The aquifer was
                             assumed to be homogeneous and isotropic.
                             Because flow was assumed to  be in  steady
                             state, the storage coefficient was set equal to
                             0.0. The parameters used to define problem

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                29
                                                                    EXPLANATION

                                                            HP No-flow boundary

                                                            ||:i:Pr| Constant- head boundary

                                                            [ + I Initial concentration (Co) equals
                                                            I*	±1  100; elaewhera Co=0

                                                            • 90— Computed potentiometric altitude.
                                                                 Contour interval 2.0 foet
                                                                 (0.61 meter)
                                                               A/= 900 feet (274 meters)

                                                               A/r 900 feet (274 meterg)
                 Figure 13.—Grid, boundary conditions, and flow field for test problem 1.
1 are listed in table 2.  The slug of known
mass was then  allowed  to  spread  down-
gradient for a period of 2.0 years.

    Table 2.—Model parameters for test problem 1
     Aquifer propertie*
   Numerical parameter*
   £=0.005 ft/a
       (1.6x10'm/s)
    ft =20.0 ft
       (6.1 m)
   5=0.0
    i=0.80
 .,/.*=O.SO
      Ax =900 ft
          (274 m)
      Ay =900 ft
          (274 m)
 CELDIS=0.49
NPTPND=9
  The model  first computed a steady-state
head  distribution,  shown  in  figure 13, and
velocity  field. The model  required 12 time
increments (or particle movements) to simu-
late a 2.0-year period. The model was run to
simulate conditions of no dispersion (a/,=0.0
ft)  as well as moderate dispersion (at=100
ft or 30.6 m). The mass balance error com-
puted using equation  64 is shown in figure
14 for  both conditions. In  these tests the
error averages  1.9 percent and is always
within a range of  ±8 percent. Much of the
error is related to the calculation  of nodal
concentrations based on the arithmetic mean
of particle concentrations in each cell. When
a particle moves across a  cell boundary, its
area of influence shifts entirely from the first
node to the second. Thus,  depending on the
local density of points and  local concentra-
tion gradients, the use of an arithmetic mean
to compute nodal  concentrations may give
too much weight to some particles and too
little weight to others. The use of a weighted
mean,  in which  the weighting  factor is  a
function of the distance between a node and
a particle, reduced  the error to some degree.
But the improvement in precision was small
compared with the  increase in computational
requirements, so this algorithm  was not in-
cluded in the general  program. Because the
error caused by using an arithmetic mean is
not cumulative, it is not considered  a serious

-------
30
 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS


	1	1	1	
                                                                            K        A    :
                                                                           '   N    '      \  '
                                                      1.0

                                                TIME. IN YEARS

                              Figure 14.—Mass balance errors for test problem 1.
                                                        1.5
                                                                                              N  -I
2.0
                                                                                  EXPLANATION

                                                                              No-flow boundary
                                                                              Constant-head boundary
                                                                          G   Injection well

                                                                          ©   Withdrawal well

                                                                        —90- Computed potentiometric altitude.
                                                                               Contour interval 2.0 feet
                                                                               (0.61 meter)
                                                                            A X= 900 feet (274 meter*)

                                                                            £/= 900 feet (274 meters)
                      Figure 15.—Grid, boundary conditions, and flow field for test problem 2.

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                               31
problem. Furthermore, figure 14 shows that
the error decreases for a higher dispersivity
because  dispersion  smooths  out   sharp
fronts and  minimizes strong concentration
gradients.

      Test problem 2—effects of wells
  The second problem was designed to eval-
uate the application  of the  model to prob-
lems in which  the flow field is  strongly in-
fluenced by wells. The grid  and boundary
conditions  used to define this problem are
illustrated  in figure  15. The problem con-
sists  of one injection  well  and  one with-
drawal well, whose effects are superimposed
on  a  regional  flow field controlled by two
constant-head  boundaries. The  parameters
for problem 2 are defined  in table  3. The
aquifer was also assumed to be homogeneous
and isotropic. The model simulated a period
of 2.4 years and assumed steady-state flow.
  The  model required  18 time increments
 (or particle movements)  to  simulate a 2.4-
year period of solute transport. Problem  2
was also evaluated for  conditions of no dis-
persion (o£, = 0.0 ft) as well as moderate dis-
persion (aL=100 ft or 30.5 m). The mass bal-
ance error was computed using equation 62
and is shown in figure 16 for  both conditions.
The average of the 36 values shown in figure
16  is -0.06 percent; the error always falls
within the  range  of. ± 8 percent. It can be
     10.0
        Table 3.—Model parameters for test problems 2 and 3
            Aquifer properties
              •nd stresses
                                Numerical parameters
           K =0.005 ft/s
              (1.5x10' m/8)
           6=20.0 ft
              (6.1 m)
           S=0.0
           e=0.30
        or/oi=0.30
           C'=100.0
           c.=o.o
           g.=1.0 ftVs
              (0.028 mVs)
      Ax =900 ft
          (274 m)
      Ay =900 ft
          (274 m)
 CELDIS=0.50
NPTPND=9
        seen that in this case the errors are essen-
        tially  coincident  for  almost 1  year,  after
        which the error  appears to be dependent on
        the magnitude of dispersion. However, the
        model output showed that when ot = 100  ft
        (30.5  m),  the  leading  edge of  the  break-
        through curve (or chemical front)  reaches
        the constant-head sink just prior to 1.0 year.
        When at = 0.0 ft, the  leading  edge  of the
        breakthrough curve still had not entered the
        constant-head sink after 2.4 years. Because
        the two curves  in figure  16  are essentially
        coincident prior  to 1.0 year, it thus appears
        that the divergence of the two curves is not
        caused directly by the difference in disper-
        sivity. Rather, it is related to the difference
        in arrival times at the hydraulic sinks and is
        a direct effect of the manner in which con-
                      0.6
1.0             1.6
   TIME, IN YEARS
  2.0
2.6
                          Figure 16.—Mass balance errors for test problem 2.

-------
32
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
centrations are computed at sink nodes and
(or) the method of estimating the mass of
solute  removed from the  aquifer  at  sink
nodes during each time increment.

   Test problem 3—effects of user options
  In addition to the input options that con-
trol  the form  or  frequency of the output,
there  are  two  execution parameters  that
must be specified  by the user and  influence
the accuracy, precision,  and efficiency  (or
computational cost) of the solution to a par-
ticular problem. These execution parameters
are the initial number of particles  per node
(NPTPND)  and  the maximum fraction of
the grid dimensions that particles are  al-
lowed  to move  (y in  equations 54-55  or
CELDIS in  the program). The third test
problem was designed to allow an evaluation
of both of these parameters. As illustrated
                            in figure 17, this problem  consists of one
                            withdrawal  well  located in  a regional  flow
                            field that is  controlled by two constant-head
                            boundaries.  The  contamination  sources are
                            three  central  nodes along  the upgradient
                            constant-head  boundary. The  model param-
                            eters for test problem 3 are the  same as for
                            test problem 2, as listed in table  3. However,
                            for test problem 3 solutions were obtained
                            using a range in values for CELDIS and
                            NPTPND.
                              The solution to this problem was found to
                            be sensitive  to the density of tracer particles
                            used in the simulation. Figure 18 shows how
                            the error in the mass balance  varied  with
                            time for cases of NPTPND  equal  to 4,  5, 8,
                            and 9. Table 4 lists the  execution time and
                            the mean and standard deviation of the mass
                            balance error for each case. These data clear-
                            ly indicate that the accuracy and precision
                                                                    EXPLANATION

                                                                No •flow boundary

                                                           {Illll Constant -head boundary

a                                                                Constant*head boundary
                                                                 and contaminant source

                                                             $  Withdrawal well

                                                          —90.0— Computed potentlometrlc altitude.
                                                                 Contour interval 2.5 feet
                                                                 (0.76 meter)
                                                              & *= 900 feet (274 meters)
                                                              A/= 900 feet (274 meters)
                  Figure 17.—Grid, boundary conditions, and flow field for test problem 3.

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                        33
    20.0
                                                                 EXPLANATION
                                                                •    • NPTPND* 4
                                                                O— —O NPTPNDs 5
                                                                6	A NPTPND'8
                                                                O	O NPTPND* 9
   -10.1
                                      1.0              1.6

                                         TIME. IN YEARS
                       2.0
2.6
  Figure 1B.—Effect of NPTPND on mass balance error for test problem 3; CELDIS=0.50 in all cases.
Table 4.—Effect of NPTPND on accuracy, precision,
     and efficiency of solution to test problem 3
Han balance error
(percent)
NPTPND
4 	
5 	
8 	
9 	

epu-ceeonda '
	 12.8
	 14.0
	 17.9
	 19.2

Memn
1.49
.90
.48
.26
Standard
deviation
6.83
2.29
1.53
.69
 'The program wa> executed on a Honeywell 60/68 computer;
CELDIS = 0.60.
of the solution are  directly proportional  to
particle density,  while the efficiency  of the
solution is inversely related to NPTPND.  In
other words, a better solution will cost more.
It is important to note that the oscillations
or scatter shown in figure 18  decrease with
time and that there is essentially no  differ-
ence among the solutions and  among the
mass balance errors for times greater than
about 1.5 years.
  Next the effect of  CELDIS (or y)  was
evaluated  for test  problem  3  by  setting
NPTPND=9 and running the model with
several possible values of CELDIS. Figure
19 shows how the error in the mass balance
varied with time for cases of CELDIS equal
to 0.25, 0.50, 0.75, and 1.00. Table 5 lists the

Table 5.—Effect of  CELDIS  on  accuracy,  precision.
     and efficiency of solution to test problem 3
Mass balance error
(percent)
CELDIS
0.26 	
.60 	
,75 	
1.00 ... .

epu-eeeonda '
	 84.6
	 19.2
	 14.4
	 12.1

Mean
1.50
.26
.66
.25
Standard
deviation
2.99
.69
.69
1.48
  1 The program wai executed on a Honeywell (0/68 computer;
NPTPND=».
execution time and the mean and standard
deviation of the mass balance error for each
case.  These data indicate that the  relation-
ship between CELDIS and the mass balance
error is not as simple and straightforward
as for NPTPND. It is apparent that the re-
sults for 0.50, 0.75, and 1.00 are similar, and
of these, the results for CELDIS=0.50 ap-

-------
34
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
   16.0
                                                                      CELDIStO.25
                                                                 O	-o CELDIS xO.50
                                                                 ft	A CELDIS:0.75
                                                                 D— -
-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                      85
 Coordinate system and boundary conditions
  After the finite-difference grid is designed,
the first program modification that should be
made is to modify the array dimensions for
the specific grid used. This will  permit the
most efficient use of computer storage. The
array sizes should be set equal to NX, NY,
and  NPMAX, which are specified on  Input
Card 2. The maximum number of particles,
NPMAX, may be computed from the follow-
ing equation:
NPMAX==(NX-2) (NY-2) (NPTPND)
            + (N.)  (NPTPND) + 250   (71)
where
     N,   is the number of nodes  that repre-
          sent fluid sources, either at wells
          or at constant-head cells.
The values of NX and NY should be substi-
tuted for the  20-by-20  arrays contained
in COMMON statements  PRMK, HEDA,
HEDB, CHMA, CHMC, and DIFUS, and in
DIMENSION statements  on lines   C170,
G200, H140, and 1160. The value  of NPMAX
should replace 3200 in the PART array in
all the CHMA COMMON statements.
   Although this program is designed for ap-
plication to two-dimensional areal flow prob-
lems, it can  be applied  directly to two-di-
mensional cross sections. In this case  the z-
coordinate would replace the ^-coordinate.
Then the user would  have to  assume and
specify unit  width  (THCK  array) for  Ay
and  substitute hydraulic  conductivity for
transmissivity in data set 3 of  attachment'
 III. If the problem involves  transient flow,
then specific storage (S,) should be  substi-
 tuted for the storage coefficient.  Also, if  re-
 charge or discharge is to be specified through
 the  RECH array (data set 5), values  should
 be divided by the thickness of the layer (Az)
 to reduce the  dimensionality of the  stress
 rate to  (T~l) rather than (LT-1) as indi-
 cated in the  documentation. In applying the
 cross-sectional model to a field problem it is
 important that conditions meet the inherent
 assumption  that there exist no significant
 components of flow into or out of the plane1
 of  the  section.  Because  this  assumption
 would probably be  impossible to meet in the
vicinity of a pumping well, the  use of the
REC array  (data set 2) should  usually be
limited to representing special or known-flux
boundary conditions.
  The program can also be applied directly
and simply to one-dimensional problems. In
this case one of the dimensions (NX or NY)
should be reduced to a value of 3, of which
the outer two are used to represent the no-
flow boundaries around the one-dimensional
row or column.
  The  most complex type of change would
involve  rewriting the program for applica-
tion to other than a two-dimensional rectan-
gular grid. One possibility includes problems
of flow to or from  wells  in  which radial
symmetry can be assumed. This would allow
variables to be expressed in terms of r-z co-
ordinates. Another possibility is to simulate
three-dimensional flow in x-y-z coordinates.
A three-dimensional  finite-difference  flow
model is available (Trescott, 1975) and would
be compatible with the method-of-character-
istics solution to  the solute-transport equa-
tion.
  It is  sometimes convenient to separately
associate certain parts of the grid or certain
boundary conditions with corresponding field
conditions or hydrologic  units. The analysis
of flow patterns and  water-quality  changes
may then be aided by performing separate
mass balances (or budgets)  for each char-
acteristic type of  node.  The  nodal types
or  zones  can be conveniently identified
through the NODEID array. Then the mass
balance routines in subroutines CNCON and
 (or)  OUTPT would have to be  modified to
tally fluxes separately for each NODEID; for
an example,  see Konikow (1977). Similarly,
if a coupled stream-aquifer system  is being
considered, a separate  subroutine  may  be
added to route streamflow downstream and
progressively  account   for  ground-water
gains and losses and for tributary inflow or
diversions. An example of such a modifica-
tion is discussed by Konikow and Bredehoeft
 (1974).
   For certain types of problems it may be
 desirable to be able to specify  a  constant-
concentration boundary condition. The pro-

-------
36
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
gram could be modified to allow this by using
a predetermined value or range in values of
NODEID to identify this type of boundary.
Then a statement could be added between
lines G1090 and G1100  to reset  the concen-
tration at the node equal to the constant con-
centration where this condition  is specified.
The value of the constant concentration can
be stored in the CNRECH array. Note that
the mass balance calculation as  presently
written will not account for the mass of
solute added or removed at a constant-con-
centration boundary.

              Basic equations
   The basic equations that are solved by this
model were derived under a number of limit-
ing assumptions. Some  of these  assumptions
can be overcome through modifications of the
basic equations and corresponding changes
in the program.
   The program assumes that molecular dif-
fusion is negligible. But if it is  necessary to
consider the process of molecular diffusion in
a particular problem, the coefficient  of hy-
drodynamic dispersion (Dtl) can be redefined
as the sum of the coefficient of mechanical
dispersion, which is defined by the right side
 of equation 5,  and a  coefficient of molecular
 diffusion. The consequent program modifica-
 tion would have  to  be made only in sub-
 routine VELO (lines E1280-E1680).
   The solute-transport  equation can also  be
 modified to include the effects of first-order
 chemical reactions, as  was done by Robert-
 son (1974). The reaction term  could  be in-
 cluded in the right side of equation 39. The
 corresponding program modification  would
 be required in subroutine CNCON.
   In certain problems  the range in concen-
 trations may be so great that the dependence
 of fluid properties, such as density and vis-
 cosity, on the  concentration  may have to be
 considered because of the dependence of fluid
 flow on  variations in fluid properties. In this
 case the flow equation  (eq 1) would have to
 be  rewritten  in terms  of  fluid  pressure,
 rather than hydraulic head, such as equation
 15 of Bredehoeft and Finder (1973, p. 197).
 Then the program can  be modified to iterate
                           between the solutions to the flow and solute-
                           transport equations if the  change in fluid
                           properties at any node exceeds some criterion
                           during one time increment.
                              The flow equation can also be modified for
                           application to unconfined aquifers in which
                           the saturated thickness is a direct function
                           of water-table elevation.  This would  require
                           the inclusion of steps in subroutine ITERAT
                           to correct the transmissivity for changes in
                           saturated thickness.  Such a feature is in-
                           cluded in the  two-dimensional  flow model
                           documented by Trescott,  Finder, and Larson
                            (1976).

                                         Input and output

                              The input and output formats have been
                            designed  for flexibility of  use and  general
                            compatibility with the analysis of a variety
                            of types of flow problems. If any of  the for-
                            mats  are not suitable for  use with  a par-
                            ticular problem, they should be modified ac-
                            cordingly. All input formats are described
                            in attachment  III  and  contained  in sub-
                            routine PARLOD in the program.
                              It has  been  assumed that several aquifer
                            parameters  are  constant and  uniform  in
                            space, such  as storage  coefficient, effective
                            porosity, and dispersivity. If any of these are
                            known to vary in space, they should be re-
                            defined as  two-dimensional  arrays.  Then
                            statements to allow these arrays to  be read
                            into the  program  should be added  to sub-
                            routine PARLOD. Similarly, values of leak-
                            ance and source concentrations  (CNRECH)
                            are only read in data set 7, where values can
                            be associated only with a limited number of
                            unique node identification codes. If the varia-
                            tions  of  these parameters  are known on  a
                            more detailed scale, then they too can be read
                            as additional data sets by adding appropriate
                            statements to  subroutine PARLOD. For ex-
                            ample, a typical sequence of statements for
                            reading one data set is represented  by lines
                            B2650-B2750, where the initial water-table
                            elevations  (data  set 8)  are read. This se-
                            quence of statements can then be replicated
                            for reading in a different data set and in-
                            serted into subroutine PARLOD.

-------
                   MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                         37
  A labeled listing of the input data deck for
test problem 3 is provided in attachment IV.
This example illustrates the  use of the data
input formats specified in attachment III and
shows that only  a few  data cards are re-
quired by the model to simulate a relatively
simple problem. This example will also allow
the user to verify that his program deck and
computer yield essentially the same results
as  obtained by the  documented  program.
Thus, selected parts of the  output for test
problem 3 are included in attachment V. Not
all  of the printed  output from test problem
3 has been  duplicated in attachment III. In-
stead, it contains only a sufficient selection to
illustrate the type and form of output pro-
vided by the model, as well  as to allow the
user to compare his calculated values of cri-
tical parameters, such as head, velocity, and
concentration, with the values computed by
the documented model.
             Conclusions

   The  model  presented in this report can
simulate the two-dimensional transport and
dispersion of  a nonreactive  solute in either
steady-state or transient ground-water flow.
The program  is  general and flexible in that
it can  be  readily and directly  applied  to  a
wide range of types of problems, as defined
by aquifer properties, boundary conditions,
and stresses. However, some program modi-
fications may  be required  for application to
specialized problems or conditions  not in-
cluded  in the general model.
   The  accuracy of the numerical results can
be evaluated by comparison with analytical
solutions only  for relatively simple and ideal-
ized problems; in these cases there was good
agreement between the numerical and analy-
tical results. Mass balance tests also help to
evaluate the accuracy  and precision of the
model results.  The error in the  mass balance
is generally less than 10 percent. The range
in mass balance  errors  is commonly  the
greatest  during  the first few time incre-
ments, but tends to decrease  and stabilize
with time. For some problems  the  accuracy
and precision of the numerical results may
be sensitive to the initial number of particles
placed in each cell and to the size of the time
increments, as  determined  by the stability
criteria  for the  solute-transport equation.
The results of several numerical experiments
suggest  that  the accuracy and precision of
the results are essentially independent of the
magnitude of the dispersion coefficient, and
comparable accuracies are attained for high,
low, or zero dispersivities.
         References  Cited

Aris,  Rutherford, 1962,  Vectors, tensors, and the
    basic equations  of fluid mechanics: Englewood
    Cliffs, N. J., Prentice-Hall, 286 p.
Bear, Jacob, 1972,  Dynamics  of fluids  in porous
    media: New York, Am. Elsevier Publishing Co.,
    764 p.
Bredehoeft, J.  D., and Finder, G.  F., 1973,  Mass
    transport in flowing ground water: Water Re-
    sources Research, v. 9, no. 1, p. 194-210.
Carder, A. 0.,  Peaceman, D. W., and Pozzi, A. L.,
    Jr., 1964, Numerical calculation of multidimen-
    sional miscible  displacement by the  method of
    characteristics:  Soc. Petroleum  Engineers Jour.,
    v. 4, no. 1,  p. 26-36.
Konikow, L.  P., 1977, Modeling chloride movement
    in the alluvial  aquifer at  the Rocky Mountain
    Arsenal, Colorado: U.S. Geol. Survey Water-
    Supply Paper 2044, 43 p.
Konikow, L. F., and Bredehoeft, J.  D.,  1974, Model-
    ing flow and chemical  quality changes in an
    irrigated  stream-aquifer  system:  Water Re-
    sources Research, v. 10, no. 3,  p. 646-662.
Konikow, L. F., and Grove, D. B.,  1977,  Derivation
    of equations describing  solute  transport in
    ground water:  U.S. Geol. Survey  Water-Re-
    sources Investigatons 77-19, 30 p.
Lohman, S. W., 1972, Ground-water hydraulics: U.S.
    Geol. Survey Prof. Paper 708,  70 p.
Finder, G. F., and Bredehoeft, J. D., 1968, Applica-
    tion of the digital computer for aquifer evalua-
    tion: Water Resources Research,  v. 4, no. 6,
    p. 1069-1093.
Finder, G. F.,  and  Cooper, H. H.,  Jr., 1970, A nu-
    merical technique for calculating the transient
    position  of the saltwater front:  Water  Re-
    sources Research, v. 6, no. 8, p. 876-882.
Prickett. T. A., and Lonnquist, C. G., 1971, Selected
    digital computer techniques for groundwater re-
    source evaluation: Illinois Water  Survey  Bull.
    65, 62 p.

-------
38
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
Reddell, D. L., and Sunada, D. K., 1970, Numerical
    simulation of dispersion in groundwater  aqui-
    fers:  Colorado State  Univ.  Hydrology  Paper
    41, 79 p.
Robertson,  J. B., 1974, Digital modeling of radio-
    active  and  chemical waste  transport in the
    Snake  River Plain aquifer at the National Re-
    actor Testing Station, Idaho: U.S. Geol. Survey
    Open-File Rept. IDO-22054, 41  p.
Robson, S. G., 1974,  Feasibility  of digital water-
    quality modeling  illustrated  by application  at
    Barstow, California: U.S. Geol. Survey Water-
    Resources Investigations 46-73, 66 p.
Scheidegger, A.  E., 1961, General theory of disper-
    sion in porous media: Jour. Geophys. Research,
    v. 66, no. 10, p. 3273-3278.
                                Trescott,  P.  C., 1975, Documentation of finite-dif-
                                    ference model  for simulation  of three-dimen-
                                    sional ground-water  flow:  U.S. Geol. Survey
                                    Open-File Rept. 75-438, 32 p.

                                Trescott,  P.  C., Finder, G. F.,  and Larson, S. P.,
                                    1976, Finite-difference model for aquifer simu-
                                    lation in two dimensions with results of numeri-
                                    cal experiments:  U.S.  Geol.  Survey  Techniques
                                    of Water-Resources Investigations, Book 7, Chap.
                                    Cl, 116 p.

                                von  Rosenberg, D. U., 1969,  Methods for  the  nu-
                                    merical  solution of  partial differential equa-
                                    tions: New York, Am. Elsevier Publishing  Co.,
                                    128 p.

-------
COMPUTER PROGRAM AND RELATED DATA

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                      41
C
C
C
C
C
C
C
C
C
C
C
C
C
C
                              Attachment  I
                  FORTRAN  IV  Program  Listing
      ***»******•••*•**<
                                              i***********************
       SOLUTE TRANSPORT  AND  DISPERSION  IN A POROUS MEDIUM
        NUMERICAL  SOLUTION 	  METHOD OF CHARACTERISTICS
        PROGRAMMED BY J.  D.  BREDEHOEFT  AND L. F. KONIKOW
  110
 •A**************************************************************
 DOUBLE PRECISION DMIN1/DEXP/DLOG/DABS
 REAL •8TMRX,VPRM/HI,HR,HC/HK,HT/REC»RECH,TIM,AOPT,TITLE
 REAL *8XDEL/YDEL/S/AREA,SUMT,RHO*PARAM,TEST/TOL»PINT,HMIN,PYR
 REAL *8TINT«ALPHA1*ANITP
 COMMON /PRMI/ NT1H/NPMP/NPNT/N1TP/N/NX/NY/NP/NREC/1NT/NNX/NNY/NUMO
1BS»NMOV,IMOV»NPMAX/ITMAX,NZCRIT/IPRNT/NPTPND,NPNTMV*NPNTVL»NPNTD»N
2PNCHV/NPDELC
 COMMON /PRMK/ NODEID(20/20)/NPCELL(20/20)/L1NBO(500)/IXOBS(5)/IYOB
1S(5)
 COMMON /HEDA/ THCK<20/20>/PERM(20,20)/TMWL<5/50)/TMOBS(50)/ANFCTR
 COMMON /HEOB/ TMRXC20/20/2)/VPRM<20/20>/HK20/20>,HR<20/20>/HC<20/
120>/HK(20/20)/WT(20/20>,REC(20/20)/RECH(20/20)/TIM(100)/AOPT(20)/T
2iTLEdO)»XDEL*YDEL/S/AREA*SUMT,RHO*PARAM,TEST»TOL/PINT*HMIN*PYR
 COMMON /CHMA/ PART(3/3200),CONC(20/20)/TMCN<5*50)/VX(20/20)/VY(20*
120)/CONINT(20/20)/CNRECH(20/20)/POROS/SUMTCH/BETA/T1HV/STORM/STORM
2I/CMSIN/CMSOUT,fLMlN/FLMOT,SUMIO/CELDIS/DLTRAT/CSTORM
 A**************************************************************
  	LOAD DATA —
 INT«0
 CALL PARLOO
 CALL 6ENPT

 	START COMPUTATIONS	
    	COMPUTE ONE  PUMPING  PERIOD- —
    150 INT«1/NPMP
    (INT.GT.1) CALL  PARLOD
    	COMPUTE ONE  TIME  STEP	
    130 N«1/NTIM
 IPRNT«0
    	LOAD NEW DELTA T	
 TINT'SUMT-PYR*(INT-1>
 TDEL«DMINKTIM(N),PYR-TINT)
 SUMT»SUMT+TDEL
 TIM(N)«TDEL
 REMN«MOD(N/NPNT)
 A******************••**••*»•••A**************************** ***
-------
 42             TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program listing—Continued


C     	OUTPUT ROUTINES	
  120 IF (REMN.Efl.O.O.OR.N.EO.NTIM.OR.MOD(N,50).EQ.O) CALL  CHMOT
      IF (SUMT.GE.(PYR*INT)>  60 TO 140
  130 CONTINUE

C     	SUMMARY OUTPUT	
  UO CONTINUE
      IPRNT«1
      CALL CHMOT
  150 CONTINUE
      STOP
                                               r **»**»*»•**••••***•*»•
      •it************************************************************
      END
      SUBROUTINE PARLOO
      DOUBLE PRECISION OMINl,DEXP,DLOG,DA8S
590
600
610
620
630
640
6SO
660
670
680
690
700
710
720
730-
 10
 20
      REAL *8TMRX,VPRM,HI,HR,HC,HK,WT,REC,RECH,TIM,AOPT,TITLE              B   30
      REAL *8XOEL,YDEL,S,AREA,SUMT,RHO,PARAH,TEST,TOL,PINT,HMIN,PYR        B   40
      REAL *8FCTR,TIHX,TINIT,PIES,YNS,XNS,RAT,HMX,HMY                      8   SO
      REAL *8TINT,ALPHA1,ANITP                                             B   60
      COMMON /PRMI/ NTIM,NPMP,NPNT,NITP,N,NX,NY,NP,NREC,INT,NNX,NNY,NUMO   B   70
     1BS,NMOV,IMOV,NPMAX,1TMAX,NKRIT,IPRNT,NPTPND»NPNTMV,NPNTVL,NPNTD,N   B   80
     2PNCHV/NPOELC                                                         B   90
      COMMON /PRMK/ NODE ID(20,20),NPCEIL(20,20),L1MBO(500),IXOBS(5),IYOB   B  100
     1S<5)                                                                 B  110
      COMMON /HEOA/ THCK(20,20),PERM(20,20),TMWL(5,50),TMOBS(50),ANFCTR    B  120
      COMMON /HEOB/ TMRX(20,20,2>,VPRM(20,20),H1(20,20),HR(20,20),HC(20,   B  130
     120),HK(20,20),WT(20,20),REC(20,20),RECH(20,20),TIM(100),AOPT(20),T   B  UO
     2ITLE(10),XDEL,YDEL,S,AREA,SUMT,RHO,PARAM,TEST,TOL,PINT,HMIN,PVR      B  150
      COMMON /CHMA/ PART(3,3200),CONC(20,20),TMCN(5,50),VX(20,20),VY(20,   B  160
     120>*CONINT(20,20)»CNRECH(20,20)»POROS,SUMTCH,BETA,TIMV,STORM*STORM   B  170
     2l*CMSIN,CMSOUT,FLMIN,FLMOT»SUMIO,CELDIS*DLTRAT»CSTORM                B  180
      COMMON /BALM/ TOTLQ                                                  B  190
      COMMON /XINV/ OXINV,DYINV,ARINV,PORINV                               B  200
      COMMON /CHMC/ SUMC(20,20),VXBOY(20,20),VYBOY(20,20)                  B  210

      IF (1NT.GT.1) GO TO 10                                               B  230
      WRITE (6,750)                                                        B  240
      READ (5*720) TITLE                                                   8  250
      WRITE (6,730) TITLE                                                  B  260

C     	INITIALIZE TEST AND CONTROL VARIABLES	                          B  280
      STORMI«0.0                                                           B  290
      TEST-0.0                                                             B  300
      TOTLO-0.0                                                            B  310
      SUMT-0.0                                                             B  320
      SUMTCH-0.0                                                           B  330
      INT»0                                                                B  340
      IPRNT-0                                                              B  350
      NCA-0                                                                B  360
      N»0                                                                  B  370
      1MOV«0                                                               B  380
      NMOV-0                                                               B  390

C     	LOAD CONTROL PARAMETERS	                                        B  410
      READ (5,7*0) NTIM,NPMP,NX,NY,NPMAX,NPNT,N1TP,NUMOBS,ITMAX,NREC,NPT   B  420
     1PND,NCOOES,NPNTMV,NPNTVL,NPNTO,NPOELC,NPNCHV                         B  430
      READ (5/800) PINT,TOL,POROS,BETA,S,T1)1X,T1N1T,XDEL,YOEL,DLTRAT,CEL   B  440
     1DIS,ANFCTR                                                           B  450
      PTR>PINT*86400.0*365.25                                              B  460
      NNX«NX-1                                                             B  470

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                             43
                        FORTRAN IV program listing—Continued
      NNY«NY-1
      NP-NPMAX
      DX1NV"1.0/XDEL
      DYINV-1.0/YDEL
      ARINV=OXINV*DYINV
      PORINV'1.0/POROS
C     	PRINT CONTROL PARAMETERS	
      WRITE (6/760)
      WRITE (6/770) NX/NY/XDEL/YDEL
      WRITE (6/780) NTIM/NPMP/PINT/TIMX/TINIT
      WRITE (6/790) S/POROS/BETA/DLTRAT/ANFCTR
      WRITE (6/870) NITP/TOL/ITMAX/CELDIS/NPMAX/NPTPND
      IF (NPTPND.LT.4.0R.NPTPND.GT.9.0R.NPTPND.E0.6.0R.NPTPND.E«.7> WRIT
     1E (6/880)
      WRITE (6/890) NPNT/NPNTMV/NPNTVL/NPNTD/NUMOBS/NREC/NCODES/NPNCHV/N
     1POELC
      IF (NPNTMV.EQ.O) NPNTMV999
      GO TO 20

C     	READ DATA TO REVISE TIME STEPS  AND  STRESSES  FOR SUBSEQUENT
C        PUMPING PERIODS	
   10 READ (5/1060) ICHK
      IF (ICHK.LE.O) RETURN
      READ (5/1070) NT IM/NPNT/NITP/1TMAX/NREC/NPNTMV/NPNTVL/NPNTD/NPDELC
     1/NPNCHV/PINT/TIMX/TINIT
      WRITE (6/1080) INT
      WRITE (6/1C90) NTIM/NPNT/NITP/ITMAX/NREC/NPNTMV/NPNTVL/NPNTD/NPDEL
     1C/NPNCHV/PINT/TIMX/TINIT
      *•***••<
      	LIST TIME INCREMENTS	
   20 DO 30 J«1/100
      TIM(J>«0.0
   30 CONTINUE
      TIM(1)«TINIT
      IF (S.EO.0.0) GO TO 50
      DO 40 K'2/MT1M
   40 TIM(K)«T1MX*TIM(K-1>
      WRITE (6/470)
      WRITE (6/490) TIM
      GO TO 60
   50 TIM(1)»PYR
      WRITE (6/480) TIMd)
c
c
      ••••»»»*****»***»*i
                                       .**•***••,
   60
      	INITIALIZE
      IF (INT.GT.1)
      DO 70 IY-1/NY
      00 70 IX«1/NX
      VPRM(IX/IY)'0.0
      PERH(IX/IY)»0.0
      THCK(IX/IY)»0.0
      RECH(IX/1Y)»0.0
      CNRECH(IX/IY)*0
      REC(IX/IY)«0.0
      NODEIO(IX/IV)«0
      TMRX(IX/IY/1>«0
      THRX(IX/I Y/2)"0
      HI(1X/IY)«0.0
      HR(IX,1Y)«0.0
      HC(IX/IY)«C.C
      HK(IX/IY)«0.0
      WT(IX/IY)«0.0
      VX(IX/IY)«0.0
MATRICES	
GO TO 100
B 480
B 490
B SOO
B 510
B 520
B 530
B 540
B 550
B 560
B 570
B 580
B 590
B 600
B 610
B 620
B 630
6 640
B 650
B 660
B 670
B 680
B 690
B 700
B 710
B 720
B 730
B 740
B 750
B 760
B 770
B 780
B 790
B 800
B 810
B 820
B 830
B 840
B 850
B 860
B 870
B 880
B 890
B 900
6 910
B 920
B 930
B 940
B 950
B 960
8 970
B 980
B 990
B1000
81010
B1020
B1030
B1040
B1050
B1060
B1070
B1080
B1090

-------
 44
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
   70
C
c
   80
   90
  100
  110
  120
  130
  140

  ISO
  160
C
C
                                    i A****************
  170
                   FORTRAN IV program listing—Continued

VYdx* I v)»0.0
VXBDY(IX,IY)=0.0
VYBDY(IX*IY)»0.0
CONC(IX*IY)'0.0
CONINT(IX*IY)*0.0
SUMC(IX*IY)«0.0
CONTINUE
*************************************i
	READ OBSERVATION WELL LOCATIONS	
IF (NUMOBS.LE.O) GO TO  100
WRITE (6*900)
DO 80 J»1*KUMOBS
READ (5*700) IX*1V
WRITE (6*810) J*IX*IY
IXOBS(J)-IX
IV08S(J)»IV
DO 90 I*1*NUMOBS
DO 90 J«1*50
TMWL(I*J)»C.O
TMCN(I*J)>0.0
******<
	READ PUKPAGE DATA —  (X-Y COORDINATES AND  RATE  IN  CFS)	
   	SIGNS : WITHDRAWAL *  POS.;  INJECTION » NEG.	
   	IF  INJ. WELL* ALSO READ CONCENTRATION OF  INJECTED  WATER	
IF (NREC.LE.O) GO  TO 120
WRITE (6*910)
DO 110 I*1*NREC
READ (5*710) IX*IV*FCTR*CNREC
IF (FCTR.LT.0.0) CNRECH(IX*IY)«CNREC
REC(IX*IY)»FCTR
WRITE (6*820) IX*IV*REC(IX*IY)*CNRECH(IX*IY)

IF (INT.GT.1) RETURN
AREA'XDEL'YOEL
WRITE (6*690) AREA
WRITE (6*600)
WRITE (6*610) XDEL
WRITE (6*610) VDEL
****•*•**!
-—READ TRANSMISSIVITY  IN  FT**2/SEC INTO VPRM
   	FCTR  « TRANSMISSIVITY MULTIPLIER  	>
WRITE (6*530)
READ (5*550) INPUT*FCTR
DO 160 IY«1*NY
IF (INPUT.£0.1) READ (5*560) (VPRM(IX*IY)*IX«1* NX>
00 ISO IX"1*NX
IF (INPUT.NE.1) GO  TO 130
VPRM(IX*IY)*VPRM(IX*IY)*FCTR
GO TO 140
VPRM(IX*IY)»FCTR
IF (IX.EQ.1.0R.IX.EO.NX) VPRM(I X*IY)«0.0
IF (IV.EQ.1.0R.IV.EO.NY) VPRM(IX* IV)*0.0
CONTINUE
WRITE (6*520) (VPR«UX,1Y)*IX«1*NX)

	SET UP COEFFICIENT MATRIX 	  BLOCK-CENTERED  GRID	
	AVERAGE  TRANSMISSIVITY  	 HARMONIC  MEAN	
IF (ANFCTR.NE.0.0)  GO TO 170
WRITE (6*1050)
ANFCTR«1.0
PIES-3.1415927*3.1415927/2.0
YNS»NY*NY
                                                     ARRAY	
                                                     FT«*2/SEC	
 B1100
 B1110
 B1120
 81130
 81140
 81150
 81160
 B1170
 B1180
 81190
 81200
 81210
 B1220
 81230
 B1240
 81250
 81260
 81270
 81280
 81290
 B1300
 81310
 81320
 B1330
 81340
 B1350
 B1360
 81370
 B1380
 B1390
• B1400
 81410
 B1420
 81430
 B1440
 B1450
 B1460
 81470
 81480
 81490
 81500
 81510
 81520
 B1530
 81540
 81550
 81560
 B1570
 81580
 81590
 81600
 81610
 81620
 81630
 81640
 81650
 81660
 B1670
 61680
 B1690
 B1700
 B1710

-------
               MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                        45
                    .EO.0.0)  60  TO  180
                    ,0*VPRM(IX/lY)*VPRM(IX»1/I Y)/(VPRM(IX/IY)*XDEL+VPRM
180
190
200
210
220
230
240
                   FORTRAN IV program listing—Continued

 XNS«NX*NX
 HMIN«2.0
 00 180 IY-2/NNY
 DO 180 IX«2/NNX
 IF (VPRMUX/IY) ,
 TMRX(IX/1Y/1)«2,
1(IX41/IV)*XDEL>
 THRX(IX/IV,2>«2.0*VPRM(IX/IY)«VPRM(IX/IY*1)/(VPRM(IX/IY>*YDEL*VPRM
1(1X/IY*1)«YDEL)
    	ADJUST COEFFICIENT  FOR ANISOTROPY	
 TMRX(IX/1V/2)>TMRX(IX/IY/2)*ANFCTR
    	COMPUTE MINIMUM ITERATION PARAMETER  (HMIN)	
 IF (TMRX(IX/IY/1).EQ.O.O> 60 TO 180
 IF (TMRXdX/IY/2).EO.0.0) 60 TO 180
 RAT=TMRX(IX/IY/1>*YDEL/(TMRX(IX/IY/2)*XDEL>
 HMX«PIES/(XNS«(1.0*RAT))
 HMY»PIES/(YNS«(1.0*(1.0/RAT>))
 IF (HMX.LT.HM1N) HMIN'HMX
 IF (HMV.LT.HMIN) HM1N«HMY
 CONTINUE
 • •**»»«**«**»**»»**»*»*»**»*»**«****«t»***«*»*«»i
 	READ AQUIFER THICKNESS	
 WRITE (6/510)
 READ (5/550) INPUT,FCTR
 DO 210 IY«1/NY
 IF (INPUT.EQ.1) READ (5/540) ( THCK( I X/I Y>/IX«1,NX>
 DO 200 IX'1/NX
 IF (INPUT.NE.1) 60 TO 190
 THCK(IX/IY)BTHCK(IX/IY)*FCTR
 60 TO 200
 IF (VPRMUX/IY).NE.0.0) THCK(IX/IY)«FCTR
 CONTINUE
 WRITE (6/500) (THCK(1X/IY)/IX«1/NX)

 	READ DIFFUSE RECHARGE  AND DISCHARGE	
 WRITE (6/830)
 READ (5/550) INPUT/FCTR
 DO 240 IY-1/NY
 IF (INPUT.EQ.1) READ (5/560)
 DO 230 IX»1/NX
 IF (INPUT.NE.1) 60 TO 220
 RECH(IX/IV)*RECH(IX/IY)*FCTR
 60 TO 220
 IF (THCK(IX/IY).NE.O.O) RECH(I X,IY)«FCTR
 CONTINUE
 WRITE (6/840) (RECH(IX/IY)/IX'1/NX)

 	COMPUTE PERMEABILITY FROM TRANSMISSIVITY	
 	COUNT NO. OF CELLS IN  AQUIFER	
 	SET N2CRIT « 2X OF THE NO.  OF  CELLS IN  THE  AQUIFER	
 DO 250 IX-1/NX
 DO 250 IY«1/NY
 IF (THCK(IX/IY).EQ.O.O)
250
                                  (RECH(IX,IY)/IX»1 /NX)
                             60  TO  250
260
 PERM(IX/IV)-VPRM(IX/IV)/THCK(IX/IY>
 NCA«NCA*1
 VPRM(IX/IY)»0.0

 AAO«NCA*AREA
 NZCRIT«(NCA*25)/SO
 URITE (6/620)
 DO 260 IY-1/NY
 WRITE (6/650) (PERM(IX/I Y)/IX«1/NX)
B1720
B1730
B1740
81750
B1760
B1770
B1780
B1790
B1800
B1810
B1820
81830
B1840
B1850
81860
81870
B1880
81890
B1900
B1910
B1920
B1930
B1940
B1950
B1960
B1970
81980
B1990
82000
82010
B2020
82030
62040
B2050
82060
82070
82080
82090
B2100
82110
82120
82130
82140
82150
82160
B2170
B2180
82190
82200
82210
B2220
82230
82240
82250
82260
B2270
62280
82290
B2300
82310
82320
B2330

-------
 46              TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program listing—Continued

      WRITE  (6*630) NCA*AAQ*NZCRIT                                         82340

C     	READ NODE IDENTIFICATION CARDS — -                                 B2360
C        -—SET VERT. PERM.* SOURCE CONC.* AND DIFFUSE RECHARGE — -         B2370
C           	SPECIFY CODES TO FIT TOUR NEEDS	                          B2380
      WRITE  (6*570)                                                        82390
      READ (5/550) INPUT,FCTR                                              82400
      00 280 IY»1,NY                                                       82610
      IF (INPUT.EQ.1) READ (5*640) (NODEID(IX*IY)* IX*1*NX)                 82420
      DO 270 IX«1*NX                                                       82430
  270 IF (INPUT.NE.1.AND.THCK(I X*IV).NE.0.0) NODE ID(IX/IY)•FCTR            82440
  280 WRITE  (6*580) (NODE 1D(1X*IY)*IX«1*NX)                                82450
      WRITE  (6*920) NCODES                                                 82460
      IF (NCODES.IE.0) 60 TO 310                                           82470
      WRITE  (6*930)                                                        82480
      DO 300 IJ-1/NCODES                                                   82490
      READ (5*850) 1COD E * FCTR1 *FCU2*FC TR3*OVERRD                          82500
      DO 290 IX*1*NX                                                       82510
      DO 290 IY«1*NY                                                       82520
      IF (NOOEIO(IX*IY).NE.ICODE) 60 TO 290                                82530
      VPRM(IX*IY)*FCTR1                                                    82540
      CNRECH(IX*IY)»FCTR2                                                  82550
      IF (OVERRD.NE.O) RECH( I X* I Y)«FCTR3                                   82560
  290 CONTINUE                                                             82570
      WRITE  (6*860) ICODE*FCTR1*FCTR2                                      82580
  300 IF (OVERRD.NE.O) WRITE (6*1100) FCTR3                                82590
  310 WRITE  (6*590)                                                        82600
      DO 320 IY»1/NY                                                       82610
  320 WRITE  (6*520) (VPRMIX*IY)*IX«1*NX)                                  82620

C     	READ WATER-TABLE ELEVATION	                                     82640
      WRITE  (6*670)                                                        82650
      READ (5*550) INPUT*FCTR                                              82660
      DO 350 IY«1*NY                                                       82670
      IF (INPUT.EQ.1) READ (5*660) IHT(IX*I V)* IX*1*NX)                     82680
      DO 340 IX«1,NX                                                       82690
      IF (INPUT.(HE.1) 60 TO 330                                            82700
      WT(lX*IV)*bT(IX*IY)*FCTR                                             82710
      GO TO  340                                                            82720
  330 IF (THCK(IX*IV).NE.O.O> UT(IX*IV)>FCTR                               82730
  340 CONTINUE                                                             B2740
  350 WRITE  (6*680) (WT(IX*IY>*IX«1*NX)                                    82750
C     A***************************************************************     B 2 760
C     	SET INITIAL HEADS	                                              82770
      00 360 1X«1,NX                                                       82780
      DO 360 1Y»1*NY                                                       82790
      HI(IX*IY)«WTHI(IX*IY)                                                  82830
C                                                                          B2840
      CALL OUTPT                                                           82850
C     •*•»•••*•••»*»•**••»»•*•*•••»•*••••»••***•••••••••*•»•»»**•»•••      82860
C     	COMPUTE ITERATION PARAMETERS	                                   82870
      DO 370 10*1*20                                                       82880
      AOPT(1D)«0.0                                                         82890
  370 CONTINUE                                                             82900
      ANITP-NITP-1                                                         82910
      ALPHA1«DEXP(OLOG(1.0/HNIN)/ANITP>                                    B2920
      AOPT(1)»HMIN                                                         82930
      DO 380 IP*2*NITP                                                     82940
  380 AOPT(IP)«AOPT(IP-1)*ALPHA1                                           82950

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER

                        FORTRAN IV program listing—Continued
                                                                       47
            (6/450)
            (6/460) AOPT
C
c
                            I **•****•**•*****••»**••******»•***•*****••
  390
  400
  410
  420
WRITE
WRITE
* * i
	READ INITIAL CONCENTRATIONS AND COMPUTE  INITIAL  MASS  STORED—-
READ (5/550) INPUT/FCTR
DO 420 IY-1/NY
IF (INPUT.EG.1) READ  (5/660)  (CONC(IX,IY)/IX«1/NX)
DO 410 IX-1/NX
IF (INPUT.NE.1) GO TO 390
CONC(1X/IY)«CONC(IX/IY)*FCTR
60 TO 400
IF (THCK(IX/IY).NE.O.O) CONC(IX/IV)»FCTR
CON I NT (IX/IY)«CONC(IX/IY)
STORMI«STORK1»CON1NT(IX/IY)*THCK(IX/IY)*AREA*POROS
CONTINUE
C
C
C
C
c
c
    	CHECK DATA SETS
    DO 440 1X-1/NX
    DO 440 IY«1/NY
    IF (THCK(IX/IY).GT
    IF (TMRX(IX/IY/1).
    IF (TMRXdX/IY/2).
    IF (NODEID(IX/IY).
    IF (WT(IX/IY).NE.O
    IF (RECH(IX/IY).NE
    IF (REC(IX/IV).NE.
    IF (PERMdX/IY) .GT
    IF (NODEID(IX/IV).
    IF (WT(1X/IY).NE.O
    IF (RECH(IX/IY).NE
    IF (REC(IX/IY).NE.
    IF (THCK(IX/IY).GT
440 CONTINUE
    ••****<
    RETURN
    ••***•I
  430
                   FOR  INTERNAL CONSISTENCY	
                  .0.0) GO TO 430
                  GT.0.0) WRITE  (6/940)  IX/IY
                  GT.0.0) WRITE  (6/950)  IX/IY
                  GT.O) WRITE (6/960)  IX/IY
                  .0) WRITE  (6/970)  IX/IY
                  .0.0) WRITE (6/980)  IX/IY
                  0.0) WRITE (6/990)  IX/IY
                  .0.0) GO TO 440
                  GT.0.0) WRITE  (6/1000)  IX/IY
                  .0) WRITE  (6/1010)  IX/IY
                  .0.0) WRITE (6/1020)  IX/IY
                  0.0) WRITE (6/1030)  IX/IY
                  .0.0) WRITE (6/1040)  IX/IY
                                                                 i *****
                                                     t*****************
  450 FORMAT (1H1/20HITERATION PARAMETERS)
  460 FORMAT (3H   /1G20.6)
  470 FORMAT (1H1/27HTIME  INTERVALS  (IN  SECONDS))
  480 FORMAT (1H1/15X/17HSTEADY-STATE  FLOW//5X/57HTIME
     1 FOR SOLUTE-TRANSPORT SIMULATION • /G12.5)
  490 FORMAT (3H   /10G12.5)
  500 FORMAT (3H   /20FS.1)
  S10 FORMAT (1H1/22HAQUIFER THICKNESS
  S20 FORMAT (3H   /20F5.2)
  S30 FORMAT (1H1/30HTRANSMISSIV1TY  MAP
  S40 FORMAT (20G3.0)
  5SO FORMAT (11/610.0)
  560 FORMAT (2064.1)
  570 FORMAT (1H1/23HNODE  IDENTIFICATION
  S80 FORMAT (1H /20I5)
  590 FORMAT (1H1/4SHVERTICAL PERMEABILITY/THICKNESS
  600 FORMAT (1HO/10X/12HX-Y SPACING:)
  610 FORMAT (1H /12X/1061 2. 5 ).
  620 FORMAT (1H1/24HPERMEABILTV MAP  (FT/SEC))
  630 FORMAT (1HO/////10X/44HNO. OF  FINITE-DIFFERENCE
     1 /I4//10X/28HAREA OF AQUIFER IN MODEL
     20X/47HNZCRIT   (MAX. NO. OF CELLS  THAT
                                                  INTERVAL (IN SEC)
                                  (FT))

                                   (FT«FT/SEC»
                                   MAP//)
                                                (FT/(FT*SEC)>)
                                                 CELLS IN AQUIFER «
                                       •  /612.5/10H   SQ. FT.////1
                                       CAN  BE  VOID OF/20X/56HPARTI
B2960
62970
B2980
62990
63000
63010
B3020
63030
B3040
B3050
63060
63070
63080
63090
63100
63110
63120
83130
83140
83150
63160
63170
63180
63190
63200
63210
63220
63230
63240
83250
63260
63270
63280
83290
B3300
63310
83320
83330
63340
63350
B3360
63370
83380
83390
83400
83410
B3420
B3430
63440
B34SO
B3460
63470
B3480
83490
83500
B3510
83520
B3S30
B3S40
B3550
B3560
83570

-------
48
               TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
                                                         • I*/)
                                     CELL  •  *612.4)
                      FORTRAN IV program listing—Continued

   3CLES;  IF EXCEEDED* PARTICLES  ARE  REGENERATED)    <
640 FORMAT (2011)
650 FORMAT (3H   *20F5.3)
660 FORMAT (2064.0)
670 FORMAT (1H1*11HWATER TABLE)
680 FORMAT (1H *20F5.0)
690 FORMAT (1HO»10X*19HAREA  OF ONE
700 FORMAT (212)
710 FORMAT (212*268.2)
720 FORMAT (10A8)
730 FORMAT (1HO*10A8)
740 FORMAT (1714)
7SO FORMAT (1H1/77HU.S.G.S.  METHOD-OF-CHARACTER1STICS MODEL FOR SOLUTE
   1 TRANSPORT IN GROUND WATER)
760 FORMAT (1H0*21 X,21 HI N P U T     DATA)
770 FORMAT (1HO*23X*1 6H6RI0  DESCRIPTORS//13X*30HNX     (NUMBER OF COLUM
   1NS)  •  *I4/13X*28HNY     (NUMBER OF ROWS)      »*16/13X*29HXDEL  (X
   2-OISTANCE IN FEET)  * *F7.1/13X*29HYDEL   (V-DISTANCE IN FEET) • *F7
   3.1)
780 FORMAT (1HO*23X*1 6HTINE   PARAMETERS//! 3X»40HNTIM    (MAX. NO. OF TI
   1ME STEPS)       • *I6/13X*40HNPMP   CNO.  OF  PUMPING PERIODS)
   2 * *I6/13X*39HPINT   (PUMPING  PERIOD  IN  YEARS)       «*F10.2/13X,39
   3HTIMX   (TIME INCREMENT  MULTIPLIER)    »*F10.2/13X*39HTINIT  (INIT
   4IAL TIME STEP IN SEC.)    **F8.0)
790 FORMAT (1HO*14X*34HHYDROL06IC  AND  CHEMICAL PARAMETERS//!3X,1HS*7X*
   129H(STORA6E COEFFICIENT)        »*5X*F9.6/13X»28HPOROS    (EFFECTIVE
   2 POROSITY)*8X,3H*   *F8.2/13X*39HBETA     (CHARACTERISTIC LEN6TH)
   3  •  *F7.1/13X*31HDLTRAT (RATIO OF TRANSVERSE  TO/21 X*30HLONGITUO I
   4NAL DISPERSIVITY)   « *F9.2/13X*39HANFCTR   (RATIO  OF T-YY TO T-XX)
   5    •  *F12.6)
800.FORMAT (1265.0)
810 FORMAT (1H *16X*I 2*5X*I2*4X*I 2)
820 FORMAT (1H *7X*214*3X*F7.2*5X*F7.1)
830 FORMAT (1H1,39HDIFFUSE RECHARGE AND DISCHARGE  (FT/SEC))
840 FORMAT (1H *1P10E10.2>
850 FORMAT (12*3610.2*12)
860 FORMAT (1HO*7X*I2*7X*E10.3*SX*F9.2)
870 FORMAT (1HO*21X*20HEXECUTION  PARAMETERS//13X*39HNITP    (NO. OF ITE
   1RATION PARAMETERS)  « * 14/13X*39HTOL    (CONVERGENCE CRITERIA - ADI
   2P) • *F9.4/13X*39HITMAX   (MAX.NO.OF ITERATIONS  -  ADIP) • *14/13X*3
   34HCELDIS (MAX.CELL  DISTANCE PER MOVE/24X*28HOF  PARTICLES - H.O.C.)
   4    • *F8.3/13X*30HNPMAX (MAX. NO. OF PARTICLES)*7X*2H> *I4/12X*3
   S2H NPTPND (NO. PARTICLES PER  NOOE)*6X*3H*   *I4)
880 FORMAT (1HO*SX*47H*** WARNING  •**   NPTPND  MUST  EQUAL  4*5*8* OR 9.)
890 FORMAT (1HO*23X*15HPR06RAM OPTIONS//1 3X*30HNPNT    (TINE STEP INTER
   1VAL FOR/21X*18HCOMPLETE  PR1NTOUT)*7X*3H»   *I4/1 3X*31HNPNTMV (MOVE
   2INTERVAL FOR CHEM./21X*28HCONCENTRAT10N  PRINTOUT)  •   *I4/13X*29HN
   3PNTVL (PRINT OPTION-VELOCITV/21X*24HO"NO;  1-FIRST TIME STEP;/21X*1
   47H2*ALL TIME STEPS)*8X*3H«  * I 4/13X*31HNPNTD   (PRINT  OPT ION-D ISP.C
   50EF./21X,24HO«NO; 1-FIRST TIME  STEP;/21X*17H2-ALL TIME STEPS)*8X*3
   6H«  *I4/13X*32HNUMOBS (NO. OF  OBSERVATION  WELLS/21X*28HFOR HVDR06R
   7APH PRINTOUT) •  * 14/13X*35HNREC    (NO.  OF PUMPING  WELLS)    • *I5
   8/13X*24HNCOOES (FOR NODE IDENT.)*9X*2H*  * 15/13X*25HNPNCHV (PUNCH V
   9ELOCITIES)*8X*2H« * 15/13X*36HNPDELC (PRINT OPT.-CONC.  CHANGE)  •  *
   SI4)
900 FORMAT (1HO*10X*29HLOCATION OF  OBSERVATION WELLS//17x»3HNO.*5X,1HX
   1*5X*1HY/I
910 FORMAT (1HO*10X*28HLOCATION   OF  PUMPING   WELLS//11X*28HX   Y    RA
   1TEUN CFS)   CONC./>
920 FORMAT (1HO»5X*37HNO. OF NODE  IDENT.  CODES SPECIFIED  • *I2)
930 FORMAT (1HO*10X*41HTHE FOLLOWIN6 ASSIGNMENTS  HAVE BEEN MAOE:/5X*51
   1HCOOE NO.     LEAKANCE     SOURCE  CONC.      RECHARGE)
83580
B3S90
B3600
63610
B3620
B3630
B3640
B3650
83660
83670
B3680
B3690
B3700
83710
83720
83730
B3740
83750
83760
83770
83780
83790
B3800
B3810
B3820
83830
B3840
B3850
B3860
83870
B3880
B3890
83900
B3910
B3920
B3930
B3940
83950
B3960
B3970
B3980
B3990
B4000
B4010
84020
B4030
84040
84050
84060
B4070
B4080
84090
84100
84110
B4120
B4130
B4140
B4150
84160
84170
84180
84190

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                             49
                        FORTRAN IV program littmg—Continued

  940 FORMAT <1H ,5X,61H*«* WARNING ••*   THCK.EQ.0.0  AND  TMRX(X).GT.0.0
     1  AT  NODE IX «,I4,6H, IV «,I4)
  950 FORMAT (1H /5X/61H*«* WARNING **•   TNCK.EQ.0.0  AND  TMRX(V).6T.O.0
     1  AT  NODE IX «,I4,6H, IY »»14)
  960 FORMAT (1H /5X/61H*** WARNING *•*    THCK.EQ.0.0 AND  NODE ID .G~T .0.0
     1  AT  NODE IX «,I4,6H, IV «,I4>
  970 FORMAT <1H ,5X,56H««« WARNING ***   THCK.EQ.0.0  AND  WT.NE.0.0  AT  N
     10DE  IX «/I*,6H/ IV «/!*>
  980 FORMAT (1H ,5X,58H**« WARNING «••   THCK.EQ.0.0  AND  RECH.NE.0.0 AT
     1  NODE IX «,I4,6H, IV «,I4>
  990 FORMAT (1H ,5X,58H**« WARNING *•*
     1  NODE IX »,I4,6H, IV «,14>
 1000 FORMAT (1H ,5X»61H*** WARNING **•
     1  AT  NODE IX «,I4,6H, IV «,I4>
 1010 FORMAT (1H ,5X,56H*** WARNING *•*   PERM.EQ.0.0  AND  WT.NE.0.0  AT  N
     10DE  IX »,I4,6H* IV «/!*)
 1020 FORMAT (1H »5X,58H*** WARNING • •'•   PERM.EQ.0.0  AND  RECH.NE.0.0 AT
     1  NODE IX >,14,6H, IV «,I4)
 1030 FORMAT (1H /5X,58H««* WARNING ***    PERM.EQ.0.0 AND  REC.NE.0.0 AT
     1  NODE IX »,I4,6H* IV »/!*>
 1040 FORMAT (1H »5x,58H,*«* WARNING •••   PERM.EQ.0.0  AND  THCK.GT.0.0 AT
     1  NODE IX «,14,6H, IV «,I4)
 1050 FORMAT (1HO,5X,45H**« WARNING **•   ANFCTR
     1X,34HDEFAULT ACTION: RESET ANFCTR • 1.0)
 1060 FORMAT (ID
 1070 FORMAT (1014,365.0)
 1080 FORMAT (1H1,5X,25HSTART PUMPING PERIOD NO.
     1G TIME STEP* PUMPAGE* AND PRINT PARAMETERS
 1090 FORMAT (1HC,14X,9HNTIM   > •14/15X,9HNPNT
     1I4/15X»9HITMAX  * ,I4/15X/9HNREC   • ,I 4/15X,9HNPNTHV  «  ,I4/15X,9H
     2NPNTVL • ,I4/15X,9HNPNTD  » 114/1SX»9HNPDELC * •I4/15X»9HNPNCHV •
     3»14/15X,9HPINT   « 'F10.3/15X/9HTIMX   • ,f10.3/15X,9MT1NIT   • ,F1
     40.5/5
 1100 FORMAT (1H ,46X,E10.5)
      END
      SUBROUTINE HERAT
      DOUBLE PRECISION DMIN1»DEXP,DLOG,DABS
      REAL *8TMRX,VPRM,H1/HR,HC»HK,WT»REC»RECH/TIM,AOPT,TITLE
      REAL »8XDEL»YDEL/S/AREA,SUMT,RHO,PARAM,TEST,TOL,PtNT/HMIN,PYR
      REAL *8B/G/U/AxC/E/f»DR/DC/T8AR/TMK*COEF/BLH/BRK/CHK*QL#BRH
      COMMON /PRMI/ NTIM/NPMP/NPNT/N1TP,N,NX,NY,NP,NREC»1NT/NNX/NNY/NUMO
     1BS.NMOV/IMCV/NPMAX,ITMAX/NZCRIT,IPRNT,NPTPND/NPNTMV/NPNTVL»NPNTD/N
     2PNCHV/NPDELC
      COMMON /PRCIK/ NODE I 0 ( 20/?0)/NPC ELL (20*20)/L IM80 ( 500) / IXOB S ( 5 ) / IYOB
     1S(5)
      COMMON /HE DA/ THCK(20/20>»PERH(20*20)»TNWL(5,50)/TMOBS(50),ANFCTR
      COMMON /HEDB/ TMRX(20,20,2),VPRM(20,20),HI(20,20),HR(20,20),HC(20,
     120),HK(20,20),WT(20,20),R£C(20,20),R£CH(20,20),TIM(100),AOPT(20>,T
     2ITLE(10),XDEL,Yi>EL,S,AREA,SUMT,RHO,PARAM,TEST,TOL,PlNT,HMIN,PYR
      COMMON /BALM/ TOTLQ
      COMMON /XINV/ DXINV,DYINV,ARINV,PORINV
      DIMENSION b(20), 8(20), G(20>
C     **•*•*<
      KOUNT'O
      	COMPUTE ROW AND COLUMN	
         	CALL NEW ITERATION PARAMETER	
   10 REMN«MOD(KOUNT,NITP)
      IF (REMN.EQ.O) NTH»0
      NTH«NTH*1
      PARAM«AOPT(NTH)
                                           THCK.EQ.0.0  AND  REC.NE.0.0 AT

                                           PERM.EQ.0.0  AND  NODE ID.GT.0.0
                                                 WAS  SPECIFIED  AS  0.0/23
                                                 ,I2//2X,75HTHE  FOLLOWIN
                                                 HAVE  BEEN  REDEFINED:/)
                                                  •  ,14/15X,9HN1TP   • ,
C
C
C
C
B4200
B4210
B4220
B4230
B4240
B4250
84260
84270
84280
B4290
B4300
B4310
B4320
B4330
84340
B4350
84360
84370
84580
B4390
84400
B4410
B4420
64430
84440
84450
B4460
B4470
B4480
B4490
B4SOO
84510
B4520
B4530
84540-
                                                     ,**•»•***•**•»•*•
      	ROW COMPUTATIONS	
   10
   20
   30
   40
   50
   60
   70
   80
   90
C 100
C 110
C 120
C 130
C 140
C 150
C 160
C 170
C 180
C 190
C 200
C 210
C 220
C 230
C 240
C 250
C 260
C 270

-------
 50              TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program toting—Continued

      TEST«0.0                                                             C  280
      RHO»S/TIM(N)                                                         C  290
      BRK»-RHO                                                             C  300
      00 50 IY«1,NY                                                        C  310
      00 20 M«1,NX                                                         C  320
      W.EQ.O.O> 60 TO 30                                     C  380
      COEF*VPRM(IX/IY>                                                     C  390
      QL«-COEF*WTUX,IV)                                                   C  400
      A«TMRX*DY1NV                                                C  440
      TBAR*A«C*E«F                                                         C  450
      THK«T8AR*PARAM                                                       C  460
      BLH — A-C-RHO-COEF-TMK                                                C  470
      BRHȣ*F-TMK                                                          C  480
      OR«BRH*HC(IX/IY)+BRK*HK(IX,IY)-E*HC(IX,lr-1>-F*HC                                                  C  510
      B(IX)-C/U(IX)                                                        C  520
      G(IX)«(OR-A*G(IX-1))/W(IX)                                           C  530
   30 CONTINUE                                                             C  540
C                                                                          C  550
C        	BACK SUBSTITUTION	                                           C  560
      00 40 J»2,hX                                                         C  570
      IJ-J-1                                                               C  580
      IS«NX-IJ                                                             C  590
   40 HR-e*HR(IS + 1*IY>                                    C  600
   50 CONTINUE                                                             C  610
W            W^WWWWWWWWWWwWWWKWWW                 W                       ^  Q ^ \J
C     	COLUMN COMPUTATIONS	                                            C  630
      DO 90 IX«1,NX                                                        C  640
      00 60 M«1,NY                                                         C  650
      U(M)«0.0                                                             C  660
      8(N)>0.0                                                             C  670
   60 G(M)«0.0                                                             C  680
      DO 70 IY»1/NY                                                        C  690
      IF (THCK(IX^IV).EO.O.O) GO TO 70                                     C  700
      COEF'VPRMUXsIV)                                                     C  710
      QL--COEF*WT(1X,IY>                                                   C  720
      A»TNRX{ IX, IY-1»2)«OYINV                                              C  730
      C«TMRXTBAR»PARAM                                                       C  780
      BLH«-A-C-RHO-COEF-TMK                                                C  790
      8RH«E+F-TNK                                                          C  800
      OC«BRH«HR( 1X/IY) + BRK«HK(IX,IY)-E«HR(IX-1,1Y)-F*HR(IX»1/IY) + OL»RECH   C  810
     1(IX/IY)*REC(IX,1Y)«ARINV                                             C  820
      H
-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                            51
                        FORTRAN IV program listing—Continued
   80
   90
  100
  110
    IJ-J-1
    IB«NY-IJ
    H'c(ix,iB)«G(iB>-B(ie)*Hc(ix/iB*i)
    IF  (THCK(IX,IB).EO.O.O>  GO TO  80
    CHK»DABS(HC(1X/IB)-HR(IX/IB>>
    IF  (CHK.GT.TOL)  TEST«1.0
    CONTINUE
    CONTINUE
    • * **
    KOUNT*KOUNT*1
    IF  (TEST.EO.0.0) GO  TO  120
    IF  (KOUNT.GE.ITMAX)  GO  TO 100
    GO  TO  10
    ••••*•**•••*•**•*****»»•»*•*•<
    	TERMINATE PROGRAM —  ITMAX
    WRITE  (6/160)
    DO  110  IX«1/NX
    DO  110  IY«1/NY
    HK(IX/IY)«HC(IX/IY>
    CALL OUTPT
    STOP
  120
C
c
  130
                                                                 r *••*
                                     •ft********************************
                                     EXCEEDED	
      *****•<
                        r**********************************************
    	SET NEW  HEAD  (HK)	
    DO 130 IY«1»NY
    DO 130 1X-1/NX
    IF (THCK(IX/1Y).EO.O.O>  GO  TO  130
    HR(IX/IY)aHK(lX/IY)
    HKUX/I Y)»HC(IX/I Y)

    	COMPUTE  LEAKAGE  FOR MASS BALANCE	
    IF (VPRMUX/m.EO.0.0)  GO  TO  130
    DELQ«VPRM(1X/IY)*AREA*(WT(IX/IY)-HK(IX/IY))
    TOTLO«TOTLQ*DELO*TIM(N)
    CONTINUE
      WRITE
      WRITE
          (6/140)
          (6/150)
N
KOUNT
      *******************i
      RETURN
      ****•*****!
                                                                 i *****
                              i*********<
                                                       MAX.  NO.  ITERATION
140 FORMAT (1HC//3X/4HN a  /1I4)
ISO FORMAT (.IN /2X/23HNUMBER  OF  ITERATIONS  «  /1I4)
160 FORMAT (1HO/SX/64H***   EXECUTION  TERMINATED  --
   1S EXCEEDED   ***/26X/21HFINAL  OUTPUT  FOLLOWS:)
    END
    SUBROUTINE GENPT
    REAL *8THRX/VPRH/HI,HR/HC/HK/UT/REC/RECH,TIH/AOPT/TITLE
    REAL *8XDEL/YDEL/S/AREA/SUMT/RHO/PARAM/TEST/TOL/PINT/HMIN/PYR
    COMMON /PRMI/ NTIM/NPMP/NPNT/NITP/N/NX/NY/NP/NREC/INT/NNX/NNY/NUMO
   1BS/NMOV/IMCV/NPMAX/ITMAX/NZCRIT/IPRNT/NPTPND/NPNTMV/NPNTVL/NPNTD/N
   2PNCHV/NPDELC
           /PRftK/ NODEID<20/20)/NPCELL(20/20)/LIM80(500)/IXOBS(5),IrOB
      COMMON
     1S(5)
      COMMON
      COMMON
           /HE DA/ THCK(?0/20>/PERM(20/20)/TMWL(5/50)/TMOBS(50)/ANFCTR
           /HEDB/ TMRX(20/20/2)/VPRM(20/20)/HI(20/20)/HR(20/20)/HC(20/
     120),HK(20/20)/WT(20«20>/REC(20/20)/RECH(20,20)/TIM(100)/AOPT(20)*T
     2ITLE(10)/XDEL/YDEL/S/AREA/SUMT/RHO/PARAM/TEST/TOL/PINT/HMIN/PYR
      COMMON /CHMA/ PART(3/3200)/CONC(20/20)/TMCN(5/50)/WX(20/20)/VY(20/
     120)/CONINT (20/20)/CNRECH(20/20)/POROS/SUMTCH/BETA/TIMV/STORM/STORM
                                                       C
                                                       C
                                                       C
                                                       C
                                                       c
                                                       c
                                                       c
                                                       c
                                                       c
                                                       c
  900
  910
  920
  930
  940
  950
  960
  970
  980
  990
C1000
C1010
C1020
C1030
C1040
C10SO
C1060
C1070
C1080
C1090
C1100
C1110
C1120
C1130
C1140
C1150
C1160
C1170
C1180
C1190
C1200
C1210
C1220
C1230
C1240
C1250
C1260
C1270
C1280
C1290
C1300
C1310
C1320
C1330
C1340
C1350
C1360
C1370-
                                                          10
                                                          20
                                                          30
                                                          40
                                                          50
                                                          60
                                                          70
                                                          80
                                                          90
                                                         100
                                                         110
                                                         120
                                                         130
                                                         140

-------
52              TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS

                       FORTRAN IV program luting—Continued

      2I«CMSIN,CMSOUT«FLMIN»FLMOT,SUMIO*CEL01S*OLTRAT«CSTORM                0  150
       DIMENSION RP(8)/ RN<8>,  IPT(8)                                       0  160

       F1*0.30                                                              0  180
       F2«1.0/3.0                                                           0  190
       IF (NPTPND.E0.4) F1-0.25                                             0  200
       IF (NPTPNO.EQ.9) F1>1.0/3.0                                          0  210
       IF (NPTPND.E0.8) F2»0.25                                             0  220
       NCHK»NPTPND                                                          0  230
       IF (NPTPNO.EQ.5.0R.NPTPND.E0.9) NCHK>NPTPND-1                        D  240
       IF (TEST.GT.98.) GO TO 10                                            D  250

 C     	INITIALIZE VALUES	                                              0  270
       STORM»0.0                                                            0  280
       CMSIN«0.0                                                            D  290
       CMSOUT'0.0                                                           0  300
       FLMIN»0.0                  -                                         0  310
       FLMOT*0.0                                                            D  320
       SUMIO-0.0                                                            D  330
 C     •*»••*»•»***•••**»**»«»«»»*»***«»*»******»*********«*«•*•*»•«»*      D  340
    10 00 20 10-1*3                                                         D  350
       00 20 IN-1/NPMAX                                                     D  360
    20 PART(ID»IN)«0.0                                                      0  370
       DO 30 IA«1,8                                                         D  380
       RP(IA)«0.0                                                           0  390
       RN(IA)»0.0                                                           D  400
    30 1PT(IA)«0                                                            D  410
 C     	SET UP LIMBO ARRAY	                                             D  420
       00 40 IN»1,500                                                       0  430
    40 LIMBO(IN)»0.0                                                        0  440
       INO-1                                                                 D  450
       DO 50 IL»1,500/2                                                     D  460
       LIMBO( ID-1NO                                                        0  470
    SO 1NO-INO+1                                                            0  480

 C     ---INSERT PARTICLES	                                               D  500
       00 410 IX-1/NX                                                       0  510
       00 410 IY-1/NY                                                       0  520
       IF (THCK(1X/IV).EQ.O.O)  GO TO  410                                    D  530
       KR-0                                                                 0  540
       TEST2-0.0                                                            0  550
       METH-1                                                               D  560
       NPCELL(IX,IY>«0                                                      0  570
       C1«CONC.EG.O.O.OR.THCK(IX + 1,IY-1).EQ.O.O.OR.THCK(IX-1 /   D  620
      1IY»1).£0.0.0.OR.THCK(IX-1/IY-1).£8.0.0)  TEST2«1.0                    0630
       IF «THCKUX/IY»1) ,EO.O.O.OR-.THCMIX,1Y-1 ).EQ.O.O.OR.THCK.EO.O.O>.AND.NPTPND.GT.5)  TEST2-1.0          0  650
       CNOOE"C1*<1.0-F1)                                                    0  660
       IF (TEST.LT.98.0.OR.TEST2.6T.0.0) GO TO  70                           D  670
       SUMC»CONC*CONC*CONC
-------
           MODEL OF SOLUTE TRANSPORT IN GROUND WATER

                  FORTRAN IV program lifting—Continued
                                                                           53
 80
 90
100
110
120

130
140
150
160
EVET«(-1.0)**IT
DO UO IS«1/2
EVES«<-1.0)**IS
PART(1 /INO)»IX*F1*EVET
PART(2/IND)»IY+F1*EVES
PART (2 /I NO) «-PART (2, J NO)
PART(3/1ND)«C1
IF (TEST. LT. 98.0. OR. TEST2.GT. 0.0) 60 TO 130
IXO«IX*EVET
IYO«IY*EVES
KR«KR+1
IPT(KR)>INO
IF (METH.EO.Z) CO TO 80
PART<3/INO)«CNODE»CONC 
IF )  100/110/120
RP(KR)«CONC»C1-PART<3/INO)
GO TO 130
RP(KR)»0.0
RN(KR)»0.0
60 TO 130
RP(KR)»C1-PART<3/INO)
RN(KR)«CONC(IXO/IYO)-PART<3/INO)
1ND-1NO+1
CONTINUE

IF (NPTPNO.EO. 5. OR. NPTPNO.ee. 9) 60 TO 150
GO TO 160
   --- PUT ONE PARTICLE AT CENTER OF CELL ---
PART(1,INO) — IX
PART<2/INO)»-IY
PART<3/1NO)«C1
INO«IND+1
   --- PLACE NORTH/ SOUTH/ EAST/ AND WEST PARTICLES ---
IF (NPTPNO.LT.8) GO TO 290
CNOOE"C1«(1.0-F2)
00 280 1T»1,2
EVET»<-1.0)**IT
PART (1/INO)>IX«F2*E VET
PART(2/INO)«-IV
PART(3/IND)'C1
IF (TEST. LT. 98.0. OR. TEST2.6T. 0.0) 60 TO 220
IXO»IX*EVET
170
180
190
200
210

220
IPT(KR)«INO
IF (HETH.EQ.2) 60 TO 170
PART(3/IND)*CNODE*CONCUXD/IY)*F2
60 TO 180
PART(3/IND)*2.0*C1*CONC(IXD/IY)/ 190/200/210
RP(KR)»CONC(IXO/IY)-PART(3/INO)
RN(KR)«C1-PART(3/IND)
60 TO 220
RP(KR)«0.0
RN(KR)«0.0
60 TO 220
RP(KR)-C1-PART(3/IND)
RN(KR)«CONC(IXO/IY)-PART(3/IND)
1ND-INO»1
PART(1/INO)»IX
  770
  780
  790
  800
  810
  820
  830
  840
  850
  860
  870
  880
  890
  900
  910
  920
  930
  940
  9SO
  960
  970
  980
  990
01000
01010
01020
01030
01040
010SO
01060
01070
01080
01090
01100
01110
01120
01130
01140
01150
01160
01170
01180
01190
01200
01210
01220
01230
01240
01250
01260
01270
01280
01290
01300
01310
01320
01330
01340
013SO
01360
01370
01380

-------
54    -          TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program luting—Continued

      PART(2/INO)«IY+F2*EVET                                               D1390
      PART(2/IND)*-PART(2/IND>                                             OUOO
      PART(3/INO)«C1                                                       01410
      IF (TEST.LT.98.0.OR.TEST?.6T.0.0) GO TO  380                          OUZO
      IVO«IY*EVET                                                          OU30
      KR»KR+1                                                              01440
      IPT(KR)«INO                                                          01450
      IF (METH.E0.2) 60 TO 230                                             01460
      PART(3/IND)«CNODE*CONC(IX,IYO)*F2                                    01470
      GO TO 240                                                            01480
  230 PART(3/IND)«2.0*C1*CONC(IX/1YD>/(C1+CONC(IX/IYO)>                    01490
  240 IF (C1-CONC(IX/IYD)) 250*260/270                                     01500
  250 RP(KR)«CONC(1X/IYD)-PART(3/INO>                                      01510
      RN(KR)»C1-PART(3/INO)                                                01520
      GO TO 280                                                            01530
  260 RP(KR>*0.0                                                           01540
      RN((CR)«0.0                 -                                         01550
      GO TO 280                                                            01560
  270 RP(KR>«C1-PART(3/IND)                                                01570
      RN(KR)»CONC(IX/IYO)-PART(3/IND)                                      01580
  280 IND«IND*1                                                            01590
C                                                                           01600
  290 IF (TEST.LT.98.0.OR.TEST2.GT.0.0) GO TO  410                          01610
      SUMPT=0.0                                                            01620
C         —COMPUTE CONC. GRADIENT WITHIN CELL	                          01630
      00 300 KPT«1/NCHK                                                    01640
      IK«IPT(KPT>                                                          01650
  300 SUMPT»PART(3/IK)+SUMPT                                               01660
      CBAR*SUMPT/NCHK                                                      01670
C         	CHECK MASS BALANCE WITHIN CELL AND  ADJUST  PT.  CONCS.	       01680
      SUMPT-0.0                                                            01690
      IF (CBAR-C1) 310/410/330                                             01700
  310 CRCT«1 .0-UBAR/C1)                                                   01710
      IF (METH.EQ.1) CRCT«CBAR/C1                                          01720
      DO 320 KPT*1/NCHK                                                    01730
      IK«IPT(KPT)                                                          01740
      PART(3/IK)«PART(3/IK>*RP(KPT)*CRCT                                   01750
  320 SUMPT«SUMPT*PART(3/IK>                                               01760
      CBARN'SUMPT/NCHK                                                     01770
      GO TO 350                                                            01780
  330 CRCT«1,0-(C1/CBAR)                                                   01790
      IF (METH.EQ.1) CRCT=C1/CBAR                                          01800
      00 340 KPT-1/NCHK                                                    01810
      IK»IPT(KPT)                                                          D1820
      PART(3/IK>*PART(3/1K)+RN(KPT)*CRCT                                   01830
  340 SUMPT«SUMPT*PART(3/IK)                                               01840
      CBARN'SUMPT/NCHK                                                     01850
  350 IF (CBARN.Ea.C1) GO TO  410                                          01860
C         	CORRECT FOR OVERCOMPENSATION	                                D1870
      CRCT-C1/CBARN                                                        01880
      00 380 KPT'1/NCHK                                                    01890
      IK'IPT(KPT)                                                          01900
      PART(3/IK)«PART(3/IK)«CRCT                                          01910
C        	CHECK  CONSTRAINTS	                                          01920
      IF (PART(3/IK)-C1)  360/380/370                                       01930
  360 CL1M«C1-RP(KPT)+RN(KPT)                                              01940
      IF (PART(3/IK).LT.CLIM> GO TO 390                                    D1950
      GO TO 380                                                            01960
  370 CLIM«C1*RP(KPT)-RN(KPT)                                              01970
      IF (PART(3/IK).GT.CLIM) GO TO 390                                    01980
  380 CONTINUE                                                             01990
      GO TO 410                                                            02000

-------
MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                            55
       FORTRAN IV program listing— Continued
390 TEST2-1.0
    DO (00 KPT«1,NCHK
    IK«IPT(ICPT>
400 PART(3*1IO»C1
410 CONTINUE
    NP«INO
    IF (INT.EO.O) CALL
       r * * * 1
    RETURN
       CHMOT
                  I**********************************
   END
   SUBROUTINE VELO
   DOUBLE PRECISION  DM IN1/DEXP/DLOG/DABS
   REAL •8TMRX/VPRM/HI/HR/HC/HK/WT/REC/RECHVTIM/AOPT/TITLE
   REAL *8XDEL/YDEL/S,AREA,SUMT,RHO,PARAM/TEST/TOL/PINT/HMIN/PVR
   REAL *8RATE/SLEAK/D1V
   COMMON /PRMI/ NTIM/NPMP/NPNT/NITP/N/NX/NY/NP/NREC/INT/NNX/NNY/NUMO
  1BS/NMOV/IMCV/NPMAX/ITMAX/NZCRIT/IPRNT/NPTPND/NPNTMV/NPNTVL/NPNTD/N
  2PNCHV,NPDELC
   COMMON /PRMK/ NODE ID(20,20)/NPCELL(20/20)»LIMBO(500),IXOBS(5),IYOB
  1S(S>
   COMMON /HEDA/ THCK(20,20),PERM(20,20)/TMWL(5/50>/TMOBS(50)/ANFCTR
   COMMON /HEDB/ TMRX(20/20/2)/VPRM(20/20)/HI(20/20),HR(20/20)/HC(20/
  120)/HK(20/20)/UT(20/20)/REC(20/20)/RECH(20,20)/TIM(100)/AOPT(20)/T
  2ITLE(10),XDEL/YDEL/S,AREA,SUMT/RHO,PARAM,TEST/TOL/PINT/HMIN/PVR
   COMMON /X1NV/ DXINV/DYINV,ARINV/PORINV
   COMMON /CHMA/ PART(3,3200),CONC(20,20)/TMCN(5/50)/VX(20/20),VY(20/
  120)/CONINT(20/20)/CNRECH(20/20)/POROS/SUMTCH/BETA,TIMV/STORM/STORM
  21/CMS IN/CMSOUT/FLMIN/FLMOT/SUMIO/CELDIS/DLTRAT/CSTORM
   COMMON /CHKC/ SUMC(20/20>/VXBDY(20/20)/VYBDY(20/20)
   COMMON /DIFUS/ DISP (20/20/4)

   	COMPUTE VELOCITIES  AND  STORE	
   VMAX = 1 .OE-10
   VMAY*1.OE-10
   VMXBO=1.OE-10
   VMYBD»1.OE-10
   TMV«TIM(N)
   LIM«0

   DO 20 IX*1/NX
   DO 20 IY«1/NY
   DO 10 1Z-1/4
10 OISP(IX/IY/H)«0.0

   IF (THCKUX,IV).EQ.0.0) GO  TO  20
   RATE«REC(IX/IY)/AREA
   SLEAK«(HK(IX/IY)-WT(IX/1Y))*VPRM(IX/IY)
   D1V=RATE*SIEAK*RECH(IX/IY>

      	VELOCITIES  AT NODES	
         	X-DIRECTION	
   GRDX«(HK(1X-1,IY)-HK(IX+1,IY))*DXINV*0.50
   IF (THCK(IX-1/IY).EQ.0.0)  GRDX = (HK(IX/IY)-HK(I X»1/IY))«DXIN V
   IF (THCKUX + 1/IY).EQ.0.0)  GRDX«(HK(IX-1/IY)-HK(I X/IY))•DXINV
   IF (THCK(IX-1/IY).EQ.0.0.AND.THCK(IX»1/IY).EQ.0.0) 6RDX«0.0
   VX(IX/IY)«PERM(IX,IY)*GROX«PORINV
   ABVX«ABS(VX(IX/IY))
   IF (ABVX.GT.VMAX)  VMAX'ABVX
         	Y-DIRECTION	
   GRDY«(HK(IX/IY-1)-HK(1X,IY+1))*DYINV«0.50
   IF (THCK(IX/IY-1).EQ.0.0)  GROY«(HK(IX,IY)-HK(IX/IY*1))*DYINV
02010
D2020
D2030
D2040
D20SO
D2060
D2070
D2080
D2090
02100
02110-
                                                            10
                                                            20
                                                            30
                                                            40
                                                            50
                                                            60
                                                            70
                                                            80
                                                            90
                                                           100
                                                           110
                                                           120
                                                           130
                                                           140
                                                           150
                                                           160
                                                           170
                                                           180
                                                           190
                                                           200
                                                         E 210
                                                         E 220
                                                           230
                                                           ?40
                                                           250
                                                           260
                                                           270
                                                           280
                                                           290
                                                           300
                                                           310
                                                           320
                                                           330
                                                           340
                                                           350
                                                           360
                                                           370
                                                           380
                                                           390
                                                           400
                                                           410
                                                           420
                                                           430
                                                           440
                                                           450
                                                           460
                                                           470
                                                           480
                                                           490
                                                           SOO
                                                           S10
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E
                                                                         E

-------
56
          TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
C
C
   20
C
C
                  FORTRAN IV program listing—Continued

IF (THCK(IX/IV«1).EQ.O.O) 6ROY»(HK( I X/I Y-1 )-HK ( I X/I Y» *DY 1NV
IF (THCK(IX/IY-1).EQ.O.O.AND.THCK(IX/IV+1).EQ.O.O) GRDY-0.0
VY(IX/IV)*PERM(IX/IV)*6ROV*PORINV*ANFCTR
ABVY«ABS(VY(IX/IY)>
IF (ABVY.6T.VMAY) VMAY-ABVY

   — VELOCITIES AT CELL BOUNDARIES — -
GRDX"(HK(IX/IY)-HK(IX»1, IV»*OXINV
PERNXa2.0*PERM(IX/IY>*PERM(IX*1/IY)/(PERM(IX/IV)+PERM(IX«1/IV)>
VXBDY(IX/IY)*PERMX*GRDX*PORINV
GRDY«(HK(IX/IY)-HK(IX/IV«1))*DVINV
PERMY>2.0*PERM(IX/IY)*PERM(IX/ir+1)/(PERM(IX/IY)«PERM(IX/IV+1)>
VYBOY(IX/IY)«PERMY*GRDY*PORINV*ANFCTR
A8VX«ABS(VX80Y(IX/IY)>
ABVY*ABS(VYBDY(IX/IY))
IF (ABVX.GT.VMXBD) VHXBO=ABVX
IF (ABVY.GT.VMYBD) VMYBD«ABVY

IF (OIV.GE.0.0) GO TO  20
TDIV»(POROS*THCK(IX/1Y))/DABS(DIV)
IF (TOIV.LT.TMV) TMV«TOIV
CONTINUE
******
	PRINT VELOCITIES—-
IF (NPNTVL.EO.O) GO TO  80
              ,2) GO
              ,1 .AND
                          TO 30
                          .N.EQ.1)
GO TO 30
                    (VX(1X/IY)/IX«1/NX)
                    (VXBDYUX/IV)/IX«1/NX)
                    (VYUX/I Y)/1X«1,NX)
                    (VVBDY(IX/IV)/IX«1/NX)
C
C
    IF  (NPNTVL.EO.
    IF  (NPNTVL.EQ.
    GO  TO 80
 30 WRITE (6/220)
    WRITE (6/330)
    DO  40 IY»1/NY
 40 WRITE (6/350)
    WRITE (6*340)
    00  50 IY«1/NY
 50 WRITE (6/350)
    WRITE (6/360)
    WRITE (6/330)
    DO  60 IY-1/NY
 60 WRITE (6/350)
    WRITE (6/340)
    DO  70 IY-1/NY
 70 WRITE (6/3SO)
    	PUNCH VELOCITIES	
 80 IF  (NPNCHV.EQ.O) GO TO 110
    IF  (NPNCHV.EQ.2) GO TO 90
    IF  (NPNCHV.EQ.1.AND.N.EQ.1) 60 TO 90
    GO  TO 110
 90 WRITE (7/510) NX/NV/XDEL/VDEL/VNAX/VNAV
    DO  100 IY-1/NY
    WRITE (7/520) (VX(IX/IY)/IX»1,NX)
100 WRITE (7/S20) (VY(1X/IY)/IX-1/NX)
    A***************************************
    	COMPUTE NEXT TIHE STEP	
110 WRITE (6/390)
    WRITE (6/400) VHAX/VHAY
    WRITE (6/410) VMXBD/VNVBD
    TOELX»CELDIS*XDEL/VMAX
    TDELY-CEtOIS*YDEL/VMAY
    TDELXB»CELOIS«XOEL/VHX8D
    TDELYB*CELDIS*YOEL/VHYBD
    TIMV«AHIN1(TDELX/TOELV/TDELXB/TOELVB>
    WRITE (6/310) TMV/TIHV
                                                     >A****************
E 520
E 530
E 540
E 550
E 560
E 570
E 580
E 590
E 600
E 610
E 620
E 630
E 640
E 650
E 660
E 670
E 680
E 690
E 700
E 710
E 720
E 730
E 740
E 750
E 760
E 770
E 780
E 790
E 800
E 810
E 820
E 830
E 840
E 850
E 860
E 870
E 880
E 890
E 900
E 910
E 920
E 930
E 940
E 950
E 960
E 970
E 980
E 990
El 000
£1010
E1020
El 030
E1040
E1050
E1060
E1070
E1080
E1090
E1100
E1110
£1120
E1130

-------
                  MODEL OP SOLUTE TRANSPORT IN GROUND WATER              57

                        FORTRAN IV program listing—Continued

      IF (TMV.LT.TIMV) GO  TO 120                                           E1140
      LIH—1                                                               E1150
      60 TO 130                                                            E1160
  120 TIHV-TNV                                                             E1170
      LIM«1                                                      ~         E1180
  130 NTIMV«TIf1(N)/TIMV                                                    £1190
      NMOV»NTIMV«1                                                         E1200
      WRITE (6/420) TIHV,NTI«V»NHOV                                        £1210
      TIKV«TIM(N)/NNOV                                                     E1220
      WRITE (6/370) TIH(N)                                                 £1230
      WRITE (6/380) TINV                                                   £1240
C                                                                          E12SO
      IF (BETA.CO.0.0) 60  TO 200                                           £1260

C     	COMPUTE DISPERSION COEFFICIENTS	                                E1280
      ALPHA-BETA                                                           £1290
      AINGMLPHA                                                           E1300
      TRAN>DLTRAT*ALPHA                                                    £1310
      XX2-XOEL*XDEl                                                        E1320
      YY2»YDEL*YDEl                                                        E1330
      XY2«4.0«XOEL<>YDEL                                                    E1340
      00 1SO IX»2,NNX                                                      E1350
      00 150 IY-2/NNY                                                      E1360
      IF (THCK(IX/IY).EQ.O.O) GO TO 150                                    E1370
      VXE»VXBDY(IX,IY)                                                     E1360
      VYS»VYBOV(IX*IV)                                                     £1390
      IF (THCK(IX»1/IY).EO.O.O) 60 TO 140                                  EUOO
C        	FORWARD COEFFICIENTS: X-DIRECTION	                           £1410
      WYE»(VY8DY(IX/IY-1)*VYBDV(IX + 1.IY-1)+VYS*VYBDY(1X*1/I V))/4.0         E1420
      VXE2*VXE*VXE                                                         £1*30
      VVE2»VYE*VYE                                                         E1440
      VMGE»SORT(VXE2+VY£2)                                                 E14SO
      IF (VHGE.LT.1.0E-20) GO TO 140                                       £1460
      DALN»AING*VNGE                                                       £1470
      DTRN*TRAN*VMGE                                                       £1480
      VMGE2=VMGE«VMGE                                                      £1490
C           	XX COEFFICIENT	                                           E1500
      DISP(IX,IV,1)»(DALN*VXE2«DTRN*VYE2)/(VMGE2*XX2)                      £1510
C           	XY COEFFICIENT-—                                           E1520
      DISP(IX,IY»3)«(DALN-DTRN)*VXE*VYE/(VMGE2«XY2)                        El 530
C        	FORWARD COEFFICIENTS: Y-DIRECT10N	                           E1540
  140 IF (THCK(IX,IV«1).EQ.O.O) 60 TO 150                                  £1550
      VXS*(VXBDY(IX-1sIV)«VXE«VXBDY(lX-1«IY«1)+VXBDV(IX/IY*1))/400         E1560
      VYS2*VYS*VYS                                                         E1570
      VXS2>VXS*VXS                                                         E1580
      VMGS«SORT(VXS2»VYS2)                                                 £1590
      IF (VMGS.LT.1.0E-20) GO TO 150                                       £1600
      DALN«ALNG*VNGS                                                       £1610
      DTRN'TRAN*VMGS                                                       E1620
      VMGS2«VMGS*VMGS                                                      £1630
C           	YY COEFFICIENT — -                                           E1640
      DISP(IX/1Y/2)«(DALN*VYS2+DTRN*VXS2)/(VMGS2*YV2)                      E1650
C           	YX COEFFICIENT	                                           E1660
      DISP(IX,IV,4)c(DALN-DTRN)*VXS*VYS/(VMGS2*XY2)                        £1670
  150 CONTINUE                                                             E1680

C     	ADJUST CROSS-PRODUCT TERNS FOR ZERO THICKNESS	                  £1700
      00 160 IX»2»NNX                                                      £1710
      00 160 IY«2*NNY                                                      E1720
      IF (THCK(IX«IY«1>.EQ.O.O.OR.THCK(IX+1,IY+1).EQ.O.O.OR.THCK(IX/IY-1   E1730
     1).EQ.O.O.OR.THCK(IX + 1,1Y-1).EO.O.O) DISP(IX/1Y/3)«0.0                E1740
      IF (THCK(IX+1,IV).Ee.O.O.OR.THCK(IX+1,IV«1).EQ.O.O.OR.THCK(IX-1,IV   E17SO

-------
 58
              TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
                        FORTRAN IV program listing—Continued

     1).Ea.O.O.OR.THCK(IX-1*IY*1).EQ.O.O)  DISP(IX*IY*4)»0.0
  160 CONTINUE
C     *********»***»*****»*************************»*l
C     	CHECK FOR STABILITY OF EXPLICIT METHOD	
      TIMDIS'0.0
      00 170 IX»2»NNX
      00 170 IY«2*NNY
      TOCO«DISP(IX*IV*1)+DISP(IX*IY*2)
  170 IF (TDC0.6T.TIMDIS) TIMOIS-TDCO
      TIMOC«0.5/TIMDIS
      WRITE (6*4*0) T1MOC
      NT1MD«TIM(N)/TIMDC
      NOISP«NTIMD+1
      IF (NDISP.LE.NMOV) 60 TO 180
      NMOV»NOISP
      TIMV«TIM(N)/NMOV
      LIM«0
  180 WRITE (6*430) TIMV*NTIMD*NHOV
C     ********««***************************»*i
C     	ADJUST OISP. EQUATION COEFFICIENTS FOR  SATURATED  THICKNESS	
      00 190 IX»2*NNX
      DO 190 IY«2»NNY
      BAVX«0.5»(THCK(IX*IY)*THCK(IX*1*1Y))
      BAVY«0.5*(THCK(IX,IY)+THCK(IX*IV*1))
      DISP(IX*IY*1)«DISP(IX»IY*1)*BAVX
      D1SP(IX*IY*2)«OISP(1X*IY*2)*BAVY
      DISP(IX*IY*3)«DISP(IX*IY*3)*BAVX
      OISP(IX*IY*4)»DISP(IX*IY,4)*BAVY
      *******************i
190

200
210

220

230


240
                                               r***********************
  2SO


  260


  270


  280


  290
    IF  (LIM)  210/220*230
    WRITE  (6/530)
    GO  TO  240
    WRITE  (6/540)
    60  TO  240
    WRITE  (6*550)
             ****
    --- PRINT  DISPERSION EQUATION COEFFICIENTS
    IF  (NPNTD.EQ.O)  60 TO 300
    IF  (NPNTO.EQ.2)  GO TO 250
    IF  (NPNTD.EQ.1.AND.N.E0.1) 60 TO 250
    GO  TO
    WRITE
                  (OISP(IX*IY*1)*IX«1*NX)
          300
          (6*450)
    WRITE (6*460)
    00 260  IY«1*NY
    WRITE (6*500)
    WRITE (6*470)
    00 270  IY"1*NY
    WRITE (6*5CO) (OISP(IX*IY*2)*IX*1*NX)
    WRITE (6*480)
    DO 280  IY-1/NY
    WRITE (6*500) (OISP(IX*IY*3)*IX*1*NX)
    WRITE (6*490)
    00 290  IY'1*NY
    WRITE (6*500) (OISP(IX*IY*4)*IX«1*NX)
    ****** i
C
C
C
C
  300 RETURN
      *••***<
310 FORMAT (1H *19H TMV (MAX. INJ.) • *612.S/20H
   112.5)
                                                     TIMV (CELDIS)
                                                                   • *6
E1760
61770
E1780
E1790
E1800
E1810
E1820
E1830
E1840
E1850
E1860
E1870
E1880
E1890
E1900
E1910
E1920
E1930
E1940
E1950
E1960
E1970
E1980
E1990
E2000
E2010
E2020
E2030
E2040
E2050
E2060
E2070
E2080
E2090
£2100
E2110
E2120
E2130
E2140
E2150
E2160
E2170
E2180
E2190
E2200
E 2-210
E2220
E2230
E2240
E2250
E2260
E2270
E2280
E2290
E2300
E2310
E2320
E2330
E2340
E2350
E2360
E2370

-------
                  MODEL OP SOLUTE TRANSPORT IN GROUND WATER
                                                                59
  320 FORMAT
  330 FORMAT
  340 FORMAT
  350 FORMAT
  360 FORMAT
  370 FORMAT
  380 FORMAT
  390 FORMAT
  400 FORMAT
  410 FORMAT
  420 FORMAT
  430 FORMAT
  440 FORMAT
  450 FORMAT
     1MGRID
  460 FORMAT
  470 FORMAT
  480 FORMAT
  490 FORMAT
  SOO FORMAT
  510 FORMAT
  520 FORMAT
  530 FORMAT
  540 FORMAT
  550 FORMAT
     1ECT10N
           FORTRAN IV program listing—Continued

(1H1/12HX VELOCITIES)
(1H ,25X,8HAT NODES/)
(1HO/25X/1JHON BOUNDARIES/)
(1H /10G12.3)
<1H1,12HY VELOCITIES)
(3H   ,11HTIM   » ,1612.5)
(3H   ,11HTIMEVELO « ,1612.5)
<1H1,10X,29HSTAB1L1TY CRITERIA — M.O.C.//)
<1HC,8H VMAX « ,1PE9.2,5X,7HVMAY • /1PE9.2)
(1H ,8H VMXBD" ,1PE9.2,5X,7HVMYBD« ,1PE9.2)
(1HO/8H TIMV « ,1PE9.2,5X,8HNTIMV • ,\5»5X/7HNMOV *  ,!5/>
<1HO/8H TIMV * »1PE9.2,5X,8HNTIMD • /15,5X/7MNMOV «  »I5)
(3H   /11HTIMEDISP • ,1E12.5>
<1H1,32HDISPERSION EQUATION COEFFICIENTS,10X,25H«<0-IJ>«
(2I4,2F10.1,2F10.7)
(8F10.7)
<1HO,10X,42HTHE LIMITING
<1HC»10X»40HTHE LIMITING
<1HO,10X,58HTHE LIMITING
RATE)
                         STABILITY
                         STABILITY
                         STABILITY
                                           CRITERION
                                           CRITERION
                                           CRITERION
IS
IS
IS
CELDIS)
BETA)
MAXIMUM
INJ
   10
c
c
 END
 SUBROUTINE MOVE
 REAL *8TMRX,VPRM,H1*HR,HC,HK,WT,REC,RECH,T1M*AOPT,T1TLE
 REAL *8XOEL,YDEL,S,AREA,SUMT,RHO,PARAM,TEST,TOL,PINT,HMIN,PYR
 COMMON /PRMI/ NTIM,NPMP,NPNT,NITP,N,NX,NY,NP,NREC,INT,NNX,NNY,NUMO
1BS/NMOV,IMOV,NPMAX,ITMAX/NZCRIT»IPRNT,NPTPND,NPNTHV,NPNTVL,NPNTO,N
2PNCHV,NPDELC
 COMMON /PRMK/ NODEID(20,20),NPCELL(20,20),LIMBO(500),IXOBS(5),IVOB
1S(5>
 COMMON /HEDA/ THCK < 20, 20) ,P-ERM (20,20) ,TMWL( 5,50) ,TMOBS(50)# ANF CTR
 COMMON /HEOB/ TMRX(20,20,2),VPRM(20,20),HI(20,20>,HR(20,20),MC(20,
120),HK(20,20),UT(20,20),REC(20,20),RECH(20,20),TIM(100),AOPT(20),T
2ITLE(10),XDEL,YOEL,S,AREA,SUMT,RHO,PARAM,TEST,TOL,PINT,HMIN,PYR
 COMMON /XIMV/ DXINV,OYINV,ARINV,POR1NV
 COMMON /CHMA/ PART(3,3200),CONC(20,20),TMCN(5,50),VX(20,20),VY(20,
120),CONINT(20,20),CNR£CH(20,20),POROS,SUMTCH,BETA,TIMV,STORM,STORM
2I,CMSIN,CMSOUT,FLMIN,FLMOT,SU«IO,CEH>IS,DLTR»T,CSTORK
 COMMON /CHMC/ SUMC(20,20 ) ,VXBOY(20,20),VYBDY(20,20)
 DIMENSION XNEW(A), YNEW(A), DIST(4)

 WRITE  (6,680) NMOV
 SUMTCHBSUMT-TIM(N)
 FlsO.249
 IF (NPTPND.E0.5) F1«0.299
 IF (NPTPND.EQ.9) F1-0.333
 CONST1«TIMV*DXINV
 CONST2'TIMV«DYINV
 	MOVE PARTICLES 'NMOV TIMES	
 DO 650 IMOV«1,NHOV
 NPTM«NP
    	MOVE EACH PARTICLE	
 DO 590 IN«1,NP
 IF (PART(1,1N).EQ.O.O) GO TO 590
 KFLAG'O
      i *i
       	COMPUTE OLD LOCATION	
E2380
E2390
E2400
E2410
E2420
E2430
E2440
E2450
E2460
E2470
E2480
E2490
E2500
E2510
E2520
E2530
E2540
E2550
E2560
E2570
E2580
E2590
E2600
E2610
E2620
E2630
E2640-
                                                                 10
                                                                 20
                                                                 30
                                                                 40
                                                                 50
                                                                 60
                                                                 70
                                                                 80
                                                                 90
                                                                100
                                                                110
                                                                120
                                                                130
                                                                140
                                                                150
                                                                160
                                                                170
                                                                180
                                                                190
                                                                200
                                                                210
                                                                220
                                                                230
                                                                240
                                                                250
                                                                260
                                                                270
                                                                280
                                                                290
                                                                300
                                                                310
                                                                320
                                                                330
                                                                340
                                                                350

-------
60
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                        FORTRAN IV program listing—Continued

      JFLAG«1
      IF  GO TO 30
      IFLAG»-1
      PART(2»IN)»-PART<2»IN)
   30 VOLD*PART(2*IN)
      IY«YOLD+0.5
      IF (THCK(1X,IV).EQ.O.O) 60 TO 560
C     ••••*••****•*»**•*•*»»*•«**»»*»***»•»«*****•••*•*•••»•»**••*»*•
C           	COMPUTE NEW LOCATION AND LOCATE CLOSEST NODE	
C           	LOCATE NORTHWEST CORNER- —
      1VX»XOLD
      IVY-TOLD
      IXE*IVX*1
      IVS»IVY*1

C        	LOCATE QUADRANT, VEL. AT 4 CORNERS/ CHECK FOR BOUNDARIES—
      CELDX»XOLD-IX
      CELDY»VOLD-IY
      IF (CELDX.EQ.O.O.ANO.CELDY.EQ.O.O)  60 TO 280
      IF (CELDX.6E.O.O.OR.CELOY.6E.O.O) 60 TO 70
C              	PT. IN NH QUADRANT	
      VXNW«VXBDY
      VXSU»VXBDY(IVX«IVS)
      VXSE»VX(IXE*1VS)
      VYNU*VYBDY(IVX,IVV)
      VVNE"VYBDV(IXE»IVY)
      VVSW»VV(IVX»IYS)
      VYSE«VY(1XE,IYS)
      IF (THCKdVX,IVY).£0.0.0) 60 TO SO
      IF (RECdXE/IVY).EQ.0.0.AND.VPRHUXE,IVY).LT.0.09) 60  TO  40
      VXNE-VXNW
   40 IF (REC(IVX«IVS).EQ.0.0.AND.VPR«(IVX,IYS).LT.0.09) 60  TO  SO
      VYSU«VYNW
   SO IF  60 TO 270
      VYSE-VYNE
      GO TO 270
C
   70 IF (CELOX.LE.O.O.OM.CELOY.6E.O.O) 60 TO 130
C              	PT. IN NE  QUADRANT —
   80 VXNU»VX(IVX»IVY)
      VXNE«VXBDY(IVX,IVY)
      VXSU«VX(IVX,IYS)
      VXSE-VXBOY(IVX,IYS)
      VYNH«VYBDY(IVX,IVY)
      VYNE«VYBDY(IXE,IVY)
      VYSW«VYUVX,IY$)
      VYS6«VY(IXE,IYS)
      IF (CELDJT.EQ.O.O) 60  TO 120
      IF (THCK(1XE,IVV).EQ.O.O)  60 TO 100
      IF (RECUVX,IVY).EQ.0.0.AND.VPRMUVX,IVY).LT.0.09) 60  TO  90
      VXNU'VXNE
   90 IF (REC(IXE»IYS).EO.0.0.AND.VPR«(IXE/IYS).LT.0.09) 60  TO  100
                                                            360
                                                            370
                                                            380
                                                            390
                                                            400
                                                            410
                                                            420
                                                            430
                                                            440
                                                            4SO
                                                            460
                                                            470
                                                            480
                                                            490
                                                            SOO
                                                            510
                                                            S20
                                                            530
                                                            540
                                                            550
                                                            560
                                                            570
                                                            580
                                                            590
                                                            600
                                                            610
                                                            620
                                                            630
                                                            640
                                                            650
                                                            660
                                                            670
                                                            680
                                                            690
                                                            700
                                                            710
                                                            720
                                                            730
                                                            740
                                                            750
                                                            760
                                                            770
                                                            780
                                                            790
                                                            800
                                                            810
                                                            820
                                                            830
                                                            840
                                                            850
                                                            860
                                                            870
                                                            880
                                                            890
                                                            900
                                                            910
                                                            920
                                                            930
                                                            940
                                                            950
                                                            960
                                                            970

-------
                MODEL OF SOLUTE TRANSPORT IN GROUND WATER              61

                       FORTRAN IV program lifting—Continued

    VYSE-VYNE                                                            F  980
100 IF  (REC(IVX,IYS>.EQ.0.0.AND.VPRM(IVX,IVS).LT.0.09)  60  TO 270        F  990
    IF  (THCK(IXE*IVS).EQ.O.O)  GO TO  110                                  F1000
    VXSU-VXSE                                                   -        F1010
110 IF  (THCKC1VX,IVY).£0.0.0)  60 TO  270                                  F1020
    VYSU*VYNU                                                            F1030
    60  TO 270                                                            F1040
120 IF    60 TO  270                                  F1060
    VYSW'VVNU                                                            F1070
    60  TO 270                                                            F1080
                                                                         F1090
130 IF  (CELOY.LE.O.O.OR.CELDX.GE.O.O)  GO  TO  190                          F1100
             	PT. IN  SW  QUADRANT	       ~                            F1110
140 VXNW«VXBDY(IVX,IVY)                                                  F1120
    VXNE«VX.Efi.0.0.AND.VPRM.LT.0.09)  60  TO 160        F1240
    VXSE'VXSW                                                            F1250
160 IF  (REC(1XE,IVY).EO.0.0.AND.VPRMCIXE,IVY).LT.0.09)  60  TO 270        F1260
    IF    60 TO  170                                  F1270
    VXNE*VXNU                                                            F1280
170 IF  (THCK(IXE»IYS>.EQ.O.O)  60 TO  270                                  F1290
    VYNE'VYSE                                                            F1300
    GO  TO 270                                                            F1310
180 IF  (RECUXE/1VY).EO.0.0.AND.VPRMUXE,IVY).LE.0.09)  GO  TO 270        F1320
    IF  .Ee.O.O.AND.VPRMUVX,IVY).lT.0.09>  60  TO 270        F1540
    IF  (THCK(IXE«IVV).EQ.O.O)  60 TO  230                                  F15SO
    VXNW-VXNE                                                            F1560
230 If  (THCK(IVXtirS).EO.O.O)  60 TO  270                                  F1570
    VYNU-VYSU                                                            F1580
    60  TO 270                                                            F1S90

-------
 62              TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program toting—Continued

  240 IF (RECUVX,IVY).EO.0.0.AND.VPRMUVX,IVY) .LE.0.09)  GO  TO  270        F1600
      IF (THCK(IXE/IVY).EO.O.O) 60 TO 270                                  F1610
      VXNW'VXNE                                                            F1620
      60 TO 270                                                            F1630
  2SO IF (RECUVX/IVY).EO.0.0.AND.VPRMIVX,IVY).LE.0.09)  GO  TO  270        F1640
      IF (THCK(IVX,IYS>.EO.0.0) 60 TO 270                                  F16SO
      VYNW-VYSW                                                            F1660
      60 TO 270                                                            F1670
C                                                                          F1680
  260 IF (CELDX.EQ.0.0.AND.CELOV.LT.0.0) 60 TO  80                          F1690
      IF (CELDX.LT.0.0.AND.CELDY.EO.0.0) GO TO  140                         M700
      IF (CELDX.6T.0.0.AND.CELDY.EQ.0.0) 60 TO  200                         F1710
      IF (CELOX.EQ.0.0.AND.CELDY.GT.0.0) 60 TO  200                         F1720
      WRITE (6*690) IN/IXsIV                                               F1730
  270 CONTINUE                                                             F1740

C        	BILINEAR INTERPOLATION	                                      F1760
      CELXDsXOLO-IVX                                                       F1770
      CELOXH«AMOO.EQ.0.0)  VXN«VXNW*VXNE     F1840
      VXS«VXSW*(1.0-CELOX)*VXSE«CELDX                                      F1850
      IF (THCK(IVX,IYS).EO.0.0.OR.THCK( IXE,IVS).EQ.0.0)  VXS«VXSW*VXSE     F1860
      XVEL«VXN*(1.0-CELOY)*VXS*CELDY                                       F1870
      IF (THCKUVX/IVY). EQ.0.0. AND. THCK(IXE/IVY). EQ.0.0)  XVEL«VXS          F1880
      IF (THCKUVX/1YS).EQ.0.0.AND.THCKUXE/IYS).EQ.0.0)  XVEL-VXN          F1890
C           	Y VELOCITY	                                               F1900
      CELDYH«AMOD                                                       F2020
      YVEL«VY(IX,IY)                                                       F2030
  290 DISTX«XVEL*CONST1                                                    F2040
      DISTY»YVEL*CONST2                                                    F2050

C        	BOUNDARY CONDITIONS	                                         F2070
      TEMPX-XOLO+OISTX                                                     F2080
      T£MPY»YOLD*DISTY                                                     F2090
      INX«TEHPX*0.5                                                        F2100
      INY*TE«PY*0.5                                                        F2110
      IF (THCK(INX/INY).GT.O.O>  GO TO 330                                  F2120
(     A***************************************************************     F2130
C           	X BOUNDARY	                                               F2140
      IF (THCK(INX,IY).EO.0.0) GO TO  300                                  F2150
      PART(1/IN)«TEMPX                                                     F2160
      GO TO 310                                                            F2170
  300 BEYON-TERPX-1X                                                       F2180
      IF (BEVON.LT.0.0) BEYON«BEYON+0.5                                    F2190
      IF (8EVON.GT.O.O) BEYON«6EYON-0.5                                    F2200
      PARTd#IN>«TEMPX-2.0*BEYOM                                          F2210

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                             68
  310
  320
  330

  340
                        FORTRAN IV program listing—Continued

      INX«PART(1,1N)*0.5
      TEMPX=PART(1/IN)
      **************************************************<
            	Y BOUNDARY	
      IF  60 TO  330
      PART(2»IN)«TEMPY
      GO TO 340
      *********<
      BEYONsTEMPY-IY
      IF (BEYON.LT.0.0) BEYON«BEYON*0.5
      IF (BEYON.GT.0.0) BEYON«BEYON-0.5
      PART(2,IN)»TEMPY-2.0«BEYON
      INY«PART(2*IN>+0.5
      TEMPY«PART<2/IN>
      GO TO 340
      PARTd ,IN)«TEMPX
      PART(2»IN)«TEMPY
      CONTINUE
      ****** *********************************** 1
         	SUM CONCENTRATIONS AND COUNT  PARTICLES	
      SUNC*1 .
C
c
      ****** 1
                                               I ****** 1
                                                     AT  OLD  LOCATION—>-
         	CHECK FOR CHANGE  IN  CELL LOCATION	
      IF UX.EQ.INX.AND.IY.EO.INY) GO TO 580
C           	CHECK FOR CONST.-HEAD BDY. OR SOURCE
      IF (REC(IX/IY).LT.O.O)  GO  TO 350
      IF (REC(1X/IY).GT.O.O)  GO  TO 360
      IF (VPRH< IX,1Y).LT.0.09) GO TO 540
      IF .LT.HK(IX,IY>> GO TO 360
      GO TO 540

C           	CREATE NEW PARTICLES AT BOUNDARIES	
  350 IF (IFLAG.GT.O) GO TO 550
      KFLAG-1
  360 DO 370 IL«1,500
      IF (LIMBO(IL).EO.O) GO  TO  370
      IP'LIMBOdL)
      IF (IP.LT.IN) GO TO 380
  370 CONTINUE

C           	GENERATE NEW PARTICLE	
      IF (NPTH.EO.NPMAX) GO TO 600
      NPTM«NPTH+1
      IP-NPTM
      GO TO 390
  380 LIKBOUD-O
C
  390 IF (KFLAG.EQ.O) GO TO 520
      IF .EQ.O.O.OR.THCK.EQ.O.O.OR.THCK GO TO 520
      IF .EO.O.O.OR.THCK.EQ.O.O) GO TO 520
C              	IF CENTER SOURCE	
      IF (JFLAG.LT.O) GO TO 500
      JJ«4
      AN-TENPY-YOLD
      AD'TEMPX-XOLD
      DISTMV»SQRT((AD*AD)*(AN*AN))
      IF (AD.EO.0.0) GO TO 410
      SLOPE-AN/AD
F2220
F2230
F2240
F2250
F2260
F2270
F2280
F2290
F2300
F2310
F2320
F2330
F234Q
F2350
F2360
F2370
F2380
F2390
F2400
F2410
F2420
F2430
F2440
F2450
F2460
F2470
F2480
F2490
F2500
F2510
F2520
F2530
F2540
F2550
F2560
F2570
F2580
F2590
F2600
F2610
F2620
F2630
F2640
F26SO
F2660
F2670
F2680
F2690
F2700
F2710
F2720
F2730
F2740
F2750
F2760
F2770
F2780
F2790
F2800
F2810
F2820
F2830

-------
64
              TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                          NEW  COORDINATES  AND  VERIFY	
                              420
                      FORTRAN IV program listing—Continued

    BI«VOLD-SLOPE*XOLO
    XC1-IX-F1
    XC2»IX*F1
    YC1-IY-F1
    YC2»IY*F1
             	COMPUTE
    DO 400 IK»1»4
    YNEW(1K)«0.0
    XNEU(IK)«0.0
400 DIST(IK)«0.0
    VNEUd )» (SLOPE*XC1>*BI
    XNEVd)*XC1
    VNEU(2>*(SLOPE»XC2>«8I
    XNEW(2)»XC2
    IF (SLOPE.EO.0.0> 60 TO
    YNEW<3)«YC1
    XNEU(3)*(VC1-BI)/SLOPE
    YNEU(4)«YC2
    XNEW<4)«/SLOPE
    60 TO 430
410 YNEW(1)*1Y-F1
    XNEW(1)=XOLD
    YNEW<2)«1Y-»F1
    XNEW<2)»XOLD
420 JJ«2
430 DO 440 II«1,JJ
440 DIST**2«(VNEW-TE»1PY)**2>*1.00001
    IACOO
    DISTCK-2.0
    DO 460 IG»1,JJ
    IF (DIST(1G).GE.D1STHV.AND.D1STUG>.IT.DISTCK>  GO TO 450
    60 TO 460
    IXC«XNEW*0.50
    IYC«YNEWdG>+0.50
    IF (IXC.NE.IX.OR.IYC.NE.IY)  GO  TO  460
    I»CC«I6
    DISTCK-OISTdG)
    CONTINUE
    IF (IACC.LT.1.0R.IACC.6T.4)  GO  TO  S10
       (XNEW(IACC).EQ.XC1.0R.XNEU(IACC).EO.XC2)  60  TO 470
       (YNEU  60  TO 480
       TO 510
                            VNEU(IACC)«YC1
                            YNEW(IACC)«YC2
C
C
450




460




470


480

490


500


510




520
      IF
      IF
      GO
      IF
      IF
      GO
      IF
      IF
   (YNEW(IACC).LT.VCI)
   (VNEU(IACC).6T.YC2)
   TO 490
   (XNEUdACO.CT.XC1)
   (XNEW(IACC).GT.XC2)
PARTd ,IP)«XNEW(IACC)
PART(2»1P>«YNEW(1ACC)
GO TO 530
PART(1,IP)«-IX
PART(2»IP)»IY
60 TO 530
PARTd »IP)«XOLD
PART(2,IP)-YOLO
60 TO 530
        	IF EDGE SOURCE OR
            — X POSITION	
OLX-INX-IX
PARTd,IP)*TEMPX-DLX
            	V POSITION	
DLY-INY-lt
                            XNEW(IACC)»XC1
                            XNEW(IACC)"XC2
                                   SINK	
F2840
F28SO
F2860
F2870
F2880
F2890
F2900
F2910
F2920
F2930
F2940
F29SO
F2960
F2970
F2980
F2990
F3000
F3010
F3020
F3030
F3040
F3050
F3060
F3070
F3080
F3090
F3100
F3110
F3120
F3130
F3140
F3150
F3160
F3170
F3180
F3190
F3200
F3210
F3220
F3230
F3240
F3250
F3260
F3270
F3280
F3290
F3300
F3310
F3320
F3330
F3340
F3350
F3360
F3370
F3380
F3390
F3400
F3410
F3420
F3430
F3440
F3450

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER               65

                        FORTRAN IV program listing—Continued

      PART(2*1P)»TEMPY-DLY                                                F3460
      IF (KFLA6.6T.O) 60 TO 530                                           F3470
C              	IF SINK	                                              F3480
      SUMC(IX*IY)>SUMC(IX,1V)«CONC(IX,IV)                        _        F3490
      NPCELL (IX*1Y»NPCELL(IX*IY>*1                                       F3500
C                                                                         F3510
  530 PART(2»IP)«-PART(2»IP)                                              F3520
      PART<3*IP>»CONCUX*IY>                                              F3S30
      IF (REC(IX*IV).EQ.O.O) 60 TO 540                                    F3540
C     •*«»•*»••***••*•»»»•••*»***»*»*******•••*••••*•*»»••««*••••»••••    F3550
C           	CHECK FOR DISCHARGE BOUNDARY AT NEW LOCATION	            F3S60
  540 IfLAG-1.0                                                           F3570
  550 IF .LT.HKC INX*1 NY)> 60 TO 56  F3580
     10                                                                   F3S90
      IF (REC(INX'INY).GT.O.O) 60 TO 560                                  F3600
      60 TO 590                                                           F3610

C           	PUT PT. IN LIMBO	                                        F3630
  560 PARTU »IN>«0.0                                                      F3640
      PART(2,IN)»0.0                                                      F3650
      PART(3*IN>«0.0                                                      F3660
      DO 570 1D»1*500                                                     F3670
      IF (LIMBO(1D).6T.O) 60 TO 570                                       F3680
      LIMBO(ID)*IN                                                        F3690
      60 TO 590                                                           F3700
  570 CONTINUE                                                            F3710
C                                                                         F3720
  580 IF (IFLAG.LT.O) PART<2*IN)"-TEMPY                                   F3730
      IF (JFLAG.LT.O) PART(1,IN)»-TEMPX                                   F3740
  590 CONTINUE                                                            F37SO
C     	END OF LOOP	                                                   F3760
C     «•••»•*•••»•••»*•»*•••**»»»»«**»*•*»***»•»*»••»•»*••»••****•«•*•    F 3770
      60 TO 620                                                           F3780
C        	RESTART MOVE IF PT. LIMIT EXCEEDED	                         F3790
  600 WRITE (6*700) IMOV*IN                                               F3800
      TEST*100.0                                                          F3810
      CALL 6ENPT                                                          F3820
      DO 610 IX»1*NX                                                      F3830
      DO 610 IY«1,NY                                                      F3840
      SUMC(IX*IY)«0.0                                                     F38SO
  610 NPCELL (IX*m«0                                                     F3860
      TEST'0.0                                                            F3870
      60 TO 10                                                            F3880

  620 SUMTCH*SUMTCH«TIHV                                                  F3900
C        	ADJUST NUMBER OF PARTICLES	                                 F3910
      NP'NPTM                                                             F3920
      WRITE (6*670) NP*IHOV                                               F3930

      CALL CNCON                                                          F3950

C        	STORE DBS. WELL DATA FOR STEADY FLOW PROBLEMS	              F3970
      IF (S.6T.O.O) 60 TO 640                                             F3980
      IF (NUMOBS.LE.O) 60 TO 640                                          F3990
      J«MOD(IMOV*50)                                                      F4000
      IF (J.EQ.O) J-50                                                    F4010
      TMOBS(J)*SUNTCH                                                     F4020
      00 630 I-1*NUMOBS                                                   F4030
      TMWL(I»J)>HK(IXOBS(I)*IVOBS(I)>                                     F4040
  630 TNCN(I*J)'CONC(IXOBS(I)*IYOBS(I))                                   F4050
C        	PRINT CHEMICAL OUTPUT	                                       F4060
  640 IF (IMOV.6E.NMOV) 60 TO 660                                         F4070

-------
66              TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

                        FORTRAN IV program listing—Continued

  650 IF .EQ.O.OR.MOO(IMOV,50).EO.O>  CALL  CHMOT           F4080

  660 RETURN                                                               F4100
C     **•***»*•*•«*******•*»**»**•****»*»•*»»»»»**•**•*»»***»••*•»«*•*     F4110
C                                                                          F4120
C                                                                          F4130
C                                                                          F4140
  670 FORMAT <1HC/2X,2HNP/7X/2H* ,8X, 14,10X,11H1MOV      • /8X,!4)          F4150
  680 FORMAT (1HO»10X,61HNO. OF PARTICLE MOVES REQUIRED  TO  COMPLETE  THIS   F4160
     1 TIME STEP  « ,I4//>                                                 F4170
  690 FORMAT (1HO/5X/53H**• WARNING • •*     QUADRANT NOT LOCATED  FOR  PT.   F4180
     1 NO.  *I5,11H , IN CELL »2IO                                         F4190
  700 FORMAT (1HO/5X/17H ***   NOTE   ***,10X,23HNPTM.EQ.NPMAX  	  1MOV«   F4200
     1,I4,5X,8HPT. NO.=,I4,5X,10HCALL GENPT/)                              F4210
      END                                                                  F4220-
      SUBROUTINE CNCON                                                     G   10
      REAL  *8TMRX,VPRM,HI,HR,HC/Hr>WT,REC»RECH,TIM,AOPT,TITLE              G   20
      REAL  *8XDEL,YDEL»S,AREA,SUMT,RHO,PARAM,TEST,TOL,PINT,HMIN/PYR        G   30
      REAL  *8FLW                                                           G   40
      COMMON /PRMI/ NTIM,NPMP,NPNT,NITP,N,NX,NY,NP,NREC,INT,NNX,NNY,NUMO   G   SO
     1BS,NMOV,IMOV,NPMAX,ITMAX,N2CR1T,IPRNT,NPTPND,NPNTMV,NPNTVL,NPNTD/N   G   60
     2PNCHV,NPDELC                                                         G   70
      COMMON /PRMK/ NODE I 0(20,20),NPCELL(20,20),LIMBO<500),IXOBS(5),IY06   G   80
     1S(5)                                                                  G   90
      COMMON /HEDA/ THCK(20,20)/PERM(20,20),TMWL(5,50),TMOBS(50),ANFCTR    G  100
      COMMON /HEDB/ TMR X(20,20/2),VPRM(20,20)/HI(20,20),HR(20,20),HC(20,   G  110
     120)/HK(20/20)/WT(20,20),REC(20,20),RECH(20,20),TIH(100),AOPT(20)/T   G  120
     2ITLE(10),XOEL,YDEL,S,AREA/SUMT/RHO,PARAM,TEST,TOL,PINT,HMIN,PYR     G  130
      COMMON /XIKV/ OXINV/OYINV/ARINV,PORINV                               G  140
      COMMON /CHHA/ PART(3/3200)/CONC(20,20),THCN(5/50)/VX(20,20),VY(20,   G  150
     120),CONINT(20,20),CNRECH(20,20),POROS,SUMTCH,BETA,T1MV/STORM,STORM   G  160
     2I/CMSIN/CMSOUT/FLMIN/FLHOT/SUMIO/CELDIS/OLTRAT/CSTORM                G  170
      COMMON /DIFUS/ DISP(20,20/4)                                         G  180
      COMMON /CHMC/ SUMC(20/20)/VXBDY(20/20)/VYBDY(20/20)                  G  190
      DIMENSION CNCNCC20/20)/ CNOLD(20,20)                                 G  200

      ITEST'O                                                              G  220
      DO 10 IXM/NX                                                        G  230
      DO 10 IY=1,NY                                                        G  240
      CNOLO( IX,IY)«CONC(IX/IY)                                             G  250
   10 CNCNC(1X*IY)*0.0                                                     G  260
      APC«0.0                                                              G  270
      N2ERO«0                                                              G  280
      TVA*AREA*TIMV                                                        G  290
      ARPOR«AREA*POROS                                                     G  300
C     •••»*••••••*»•**»*•***•••»••••**»•••**••••**•****••••••********     G  310
C     	CONC. CHANGE fOR 0.5*TIHV DUE TO:                                 G  320
C           RECHARGE* PUMPING, LEAKAGE, DIVERGENCE OF  VELOCITY...          G  330
      CONST»0.5*TIMV                                                       G  340
   20 DO 60 IX=1/NX                                                        G  350
      DO 60 IY»1,NY                                                        G  360
      IF (THCK(IX,IY>.EQ.O.O) GO TO 60                                     G  370
      EQFCT1«CONST/THCK(IX,IY)                                             G  380
      EQFCT2«EQFCT1/POROS                                                  G  390
      C1«CONC(1X,IY)                                                       G  400
      CLKCN*0.0                                                            G  410
      SLEAK'(HK(IX,IY)-UT(IX,IY»*VPRM(IX,IY)                              G  420
      IF (SLEAK..LT.O.O) C LKCN = CNREC H ( I X ,1 Y)                                G  430
      IF (SLEAK.GT.0.0) CLKCN-C1                                           G  440
      CNREC-C1                                                             G  450
      RATE'RECd X/I Y)*ARINV                                                6  460
      IF (RATE.LT.0.0) CNREC«CNRECH(I X,IY)                                 G  470

-------
              MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                           67
                     FORTRAN IV program litting— Continued

   DIV«RATE+SLEAK«RECH(IX/IV)
   IF (S.EQ.0.0) 60 TO 30
   DERH«-HR(IX/IY)>/TIM(N)
   DIVsDIV*S*DERH
   IF (S.LT. 0.005) GO TO 30
   ...NOTE: ABOVE STATEMENT  ASSUMES  THAT S«0.005  SEPARATES  CONFINED
            FROM UNCONFINEO  CONDITIONS; THIS CRITERION SHOULD BE
            CHANGED IF FIELD CONDITIONS ARE DIFFERENT.
   DELC«EaFCT2*(C1«(DIV-POROS*OERH)-RATE*CNREC-SLEAK*CLKCN-RECH(IX/IV
  1>*CNRECH(IX/IY»
   GO TO 40
30 DELC«E«FCT2*(C1*DIV-RATE*CNREC-SLEAK*CLKCN-RECH(IX/IY)*CNRECH(IX/I
 40
   CNCNC(IX/1Y)«CNCNC(1X/IY)*DELC
   --- CONC. CHANGE DUE  TO  DISPERSION  FOR  0.5*TIMV ---
      -—DISPERSION WITH TENSOR COEFFICIENTS"-
   IF (BETA. EQ. 0.0) GO  TO  50
   Xl«DISP(IX/IY/1)*(CONC(IX«1/IY)-C1)
   X2«DISP(IX-1/IY/1)*(CONC(IX-1/IY)-C1)
   Y1*DISP
   XX1«D1SP-CONC(IX-1/IY)-C
  10NCUX-1/1Y-1))
50 CNCNC(IX/IY)«CNCNC(IX/IY)*EOFCT1*
   IF (APC.GT.0.0) GO  TO  80
   IF (REC(IX/IY).NE.O.O.OR.VPRM(1X/IY).GT.0.09)  GO TO 90
   NZERO-NZERO+1
   GO TO  90
   CONC(IX,IY)«SUMC(IX,IY)/APC
   CONTINUE
      --- CHECK NUMBER  OF  CELLS VOID OF  PTS. ---
   IF (N2ERO.GT.O) URITE  (6/290)  NZEROsIMOV
   IF (NZERO.LE.NZCRIT) GO TO  20
   TEST«99.0
   URITE  (6/3CO)
   WRITE  (6/320)
   DO 100 IY«1/NY
   WRITE  (6/330) (NPCELL(IX/IY)/IX-1/NX)
   GO TO  20

   --- CHANGE CONCENTRATIONS  AT NODES ---
   DO 130 IX-1/NX
   DO 130 IY*1/NY
   IF (THCKUX/m.EQ.O.O) GO  TO 120
   CONC(IX/IV)'CONC(IX/IV)+CNCNC(IX/IY)

                                                                         6
                                                                         6
                                                                         G
                                                                         G
  480
  490
  SOO
  510
G 520
G 530
G 540
G 550
G 560
G 570
G 580
G 590
G 600
G 610
G 620
G 630
G 640
G 650
6 660
G 670
G 680
G 690
G 700
G 710
6 720
G 730
G 740
G 750
G 760
G 770
G 780
G 790
6 800
G 810
G 820
  830
  840
  850
G 860
G 870
G 880
6 890
G 900
6 910
6 920
G 930
6 940
G 950
G 960
  970
  980
  990
G1000
G1010
61020
61030
61040
61050
61060
61070
61080
G1090
6
6
6
                                                                         6
                                                                         6
                                                                         6

-------
 68
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                        FORTRAN IV program lifting—Continued
  120

  130
C
c
  no

  ISO
  160

  170
  180
c
c
  190

  200
  210
  220

  230
C
C
  240
  250
  260
  270
                                               r••*»*****•*»»»•«**•***•
NPCELL(IX*IY)*0
SUMC(IX*IY)«0.0
IF (CONC(IX*IY).LE.O.O) 60 TO 130
CNCPCT-CNCNC(IX*IY)/CONC(IX*IY)
SUMC(IX*IV)>CNCPCT
60 TO 130
IF (CONC(IX,IY).GT.O.O) WRITE (6*310) I X*IY*CONC(IX*IV)
CONC(IX/IY)«0.0
CONTINUE
**•»*•*•»!
	CHAN6E CONCENTRATION OF PARTICLES	
00 180 IN«1*NP
IF (PART(1*IN).EO.O.O) GO TO 180
1NX«A8S(PART(1*IN)>*0.5
INY*A8S(PART(2*IN))*0.5
	UPDATE CONC. OF PTS. IN SINK/SOURCE CELLS	
IF (REC(INX*INV).NE.O.O) GO TO 140
IF (VPRM(INX*INY).LE.0.09) 60 TO 150
PART(3*IN)*CONCUNX*INV)
60 TO 180
IF (CNCNC(1NX*INY).LT.O.O) 60 TO 170
PART(3*IN)*PART(3*IN>+CNCNC(INX*INY>
60 TO 180
IF (CONC(INX*INY).LE.O.O) 60 TO 160
IF (SUMC(INX*INY).LT.-1.0) 60 TO 160
PART(3*IN)*PART(3*IN)«PART(3*IN)*SUMC(INX*INV)
CONTINUE
WRITE (6*280) TIM(N)*T1MV*SUMTCH
•»•******»*••••**•**»**»»*»»•••*••**»«**»*•*•*•••»»•**••*•*•••••
	COMPUTE MASS BALANCE FOR SOLUTE	
CSTORM«0.0
STORM'0.0
00 270 1X«1/NX
00 270 IY«1/NY
IF (THCK(IX*IY).EO.O.O) GO TO 270
SUMC(IX*IV)*0.0
   	COMPUTE MASS OF  SOLUTE IN STORAGE	
STORM«STORH + CONe) 200*210*190
CMSOUT»CMSOUT+REC(IX*IY)*CNOLO(IX,IY)*TIMV
GO TO 210
CMSIN'CMSIN*REC(IX*IY)*CNRECH(IX*IV)*TIMV
IF (RECH(IX*IY)) 230*240*220
CMSOUT>CMSOUT+RECH(IX*IV)*CNOLD(IX*IY)*TVA
GO TO 240
CHSIN»CMSIN«RECH(1X*IY)*CNRECH(IX*IV)*TVA
**»•••••*•»»»»«»»•**•••**»•**•••**•***•*•*•.
   —-ACCOUNT FOR BOUNDARY FLOW	
IF (VPRM(IX*IY).EO.O.O) GO TO 270
FLW»VPRH(IX*IV)*(WT(IX*IV)-HK(IX*IV»
IF (FLW.6T.O.O) 60 TO  250
IF (FLU.LT.0.0) GO TO  260
60 TO 270
      	MASS IN BOUNDARY DURING TIME STEP	
FLNIN«FLMIK«FLW*CNRECH(IX*IY)*TVA
GO TO 270
      	MASS OUT DURING TIME STEP	
FLNOT»FLMOT+FLW*CNOLD(IX*IV)*TVA
CONTINUE
   !•••••<
   	COMPUTE CHANGE IN MASS OF SOLUTE STORED	
                                        I DISCHARGED	
61100
61110
G1120
61130
61140
61150
61160
61170
61180
61190
61200
61210
G1220
61230
61240
G1250
61 260
61270
G1280
61290
G1300
61310
61320
61 330
61340
61350
61 360
61370
61380
61390
61400
61410
61420
61430
G1440
G1450
61460
61470
61480
G1490
61 500
G1510
61520
61530
G1S40
G1550
G1S60
61570
G1S80
61590
61600
61610
G1620
61630
61640
61650
61660
G1670
61680
61690
G1700
61710

-------
                 MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                        FORTRAN IV program listing—Continued
      CSTORN-STORM-STORMI
      SUMIO»FLM1N»FLMOT-CMSIN-CMSOUT
C
c
C
c
      ** I
                                             I ••••••*•I
      	REGENERATE PARTICLES IF 'NZCRIT* EXCEEDED-
      IF (TEST.6T.98.0) CALL GENPT
      TEST-0.0
      ******i
      RETURN
                                • *1612.5»10X*11HTIMV
                                                          i •***««*
                                                     * *1G12.5*10X*
                                                     CALL  GENPT    ••*/>
                                                   AT  NODE  •  *2I4*4X,7HC
280 FORMAT (3H   ,11HTIM(N>
   19HSUMTCH « *G12.5)
290 FORMAT (1HO»5X*40HNUMBER OF CELLS WITH ZERO  PARTICLES   •   *I4*5X*9
   1HIMOV  «  *I4/>
300 FORMAT (1HO*5X*44H***   NZCRIT EXCEEDED
310 FORMAT (1H *5X* 37H***CONC.GT.O. AND*. THCK .E«. 0
   10NC « *G10.4,4H •**)
320 FORMAT (1HO*2X*6HNPCELL/>
330 FORMAT (1H *4X*20I3)
    END
    SUBROUTINE OUTPT
    REAL *8TMRX,VPRM,HI*HR,HC*HK,WT*REC»RECH*TIH»AOPT*TITLE
    REAL *8XDEL*YDEL*S*AREA*SUMT*RHO*PARAM*TEST*TOL*PINT*HMIN,PVR
    COMMON /PRMI/ NTIM*NPMP*NPNT*N1TP*N*NX*NY*NP*NREC*INT*NNX*NNY*NUMO
   1BS*NMOV*IMOV*NPMAX»ITMAX*NZCRIT*IPRNT*NPTPND*NPNTMV*NPNTVL»NPNTO*N
   2PNCHV/NPDELC
           /PRMK/ NODEID(20*20)*NPCELL(20*20),LINBO(500)*IXOBS(5)»IYOB
      COMMON
     1S(S)
      COMMON
      COMMON
        /HEDA/ THCK(20*20)/PERM(20*20>*TMHL<5*50)*TMOBS<50)»ANFCTR
        /HEDB/ TMRX(20/20*2)»VPRM(20/20)»HI(20/20),HR(20/20)/HC(20<.
C


C
120),HK(20/20)/WT(20,20)/REC(20,20)/RECH(20,20)/TIM(100)*AOPT(20)/T
2ITLE(10)»XDEL*YDEL*S»AREA»SUMT»RHO»PARAM,TEST»TOL»PINT»HMIN»PtR
 COMMON /BALM/ TOTLO
 DIMENSION IHC20)
     »**•*<
 TIMD'SUMT/86400.
 TIMYmSUMT/(86<.00.0*365.25)
 	PRINT HEAD VALUES	
 WRITE  (6*120)








C
c










WRITE (6*130) N
WRITE (6*UO) SUMT
WRITE (6*150) TIMO
WRITE (6*160) TIMY
WRITE (6*170)
DO 10 1Y«1*NY
10 WRITE (6*180) (HK(IX*I
IF (N.EO.O) GO TO 110






Y),IX«1,NX)

•A*********************************************
- — PRINT HEAD MAP 	
WRITE (6*120)
WRITE (6*130) N
WRITE (6*140) SUMT
WRITE (6*150) TIND
WRITE (6*160) TIMY
WRITE (6*170)
00 30 IY«1,NY
DO 20 IX«1*NX
20 IH(IX)«HK(IX,IY)+0.5
30 WRITE (6*190) (IH(ID)*










ID»1*NX)
                                         I****************************
G1720
G1730
61740
G17SO
G1760
G1770
61780
61790
61800
61810
61820
61830
61840
61850
61860
61870
61880
61890
61900
61910
61920
61930-
H  10
   20
   30
   40
   50
   60
   70
   80
   90
H 100
H 110
H 120
H 130
H 140
H 150
H 160
H 170
H 180
H 190
.H 200
H 210
H 220
H 230
H 240
H 250
H 260
H 270
H 280
H 290
H 300
H 310
H 320
H 330
H 340
H 350
H 360
H 370
H 380
H 390
H 400

-------
70
               TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                       FORTRAN IV program listing—Continued
  40
  50
  60
  70
  80
  90
 100
	COMPUTE  WATER  BALANCE  AND  DRAWDOWN	
QSTR=0.0
PUMPsQ.O
TPUM«0.0
QINsO.O
OOUT=0.0
QNET*0.0
DELO«0.0

PCTERR=0.0
WRITE (6/290)

DO 80 IY»1/NY
DO 70 IX*1/NX
IHUX)=0.0
IF (THCK( IX/IY).EQ.0.0) GO  TO  70
TPUM=REC(IX/IY)+RECH(IX/IY)«AREA+TPUM
IF (VPRM(IX/IY).EQ.0.0) GO  TO  60
DELQ=VPRM(IX/IY)»AREA*(WT(1X/IY)-HK(1X/IY))
IF (DtLQ.GT.0.0)  GO  TO  40
QOUTsQOUT+DELQ
GO TO 50
QIN=QIN*DELQ
ONET«ONET»DELO
DDRWsHI(IX/lY)-HKUX/lY)
IH(IX>»DDRW+0.5
OSTR=OSTR*DDRW*AREA*S
CONTINUE
   	PRINT DRAWDOWN HAP	
WRITE (6/300)  (1H(IX)/IX«1/NX)
CONTINUE
PUMP«TPUM*SUMT
DELSs-aSTR/SUMT
ERRMB=PUMP-TOILQ-QSIR
DEN=PUMP+TOTLU
IF (ABS(DEN).EO.ABS(ERRMB)> JCK=1
IF (DEN.EQ.O.U) GO  TO 100
IF (JCK.E0.1) GO  TO 90
PCTERR=ERRH6*20G.O/DEN
GO TO 100
IF (QIN.EQ.0.0) GO  TO 100
PCTERRs100.0*QNET/bIN
   	PRINT MASS  BALANCE DATA  FOR  FLOW  MODEL	
WRITE (6/240)
              PUMP
              QSTR
              TOTLU
              E RRME
              WRITE  (6/280) PCTERR
              Q1N/QOUT/QNET
              TPUM
              DELS
              WRITE  (6/280) PCTERR
     WRITE
     WRITE
     WRITE
     WRITE
          (6/2SO)
          (6/230)
          (6/260)
          (6/270)
    IF (JCK.EQ.O)
    WRITE (6/200)
    WRITE (6/210)
    WRITE (6/220)
    IF (JCK.EQ.1)
 110
    RETURN
    ••••**i
                    i ******* t
                                                   i*****************
120 FORMAT
130 FORMAT
140 FORMAT
            (1H1/23HHEAD DISTRIBUTION - ROW)
            (1X/23HNUMBEK OF TIME STEPS • /1IS)
            (8X/16HTIME(SECONOS) » /1G12.S)
H 410
H 420
H 430
H 440
H 4SO
H 460
H 470
H 480
H 490
H 500
H 510
H 520
H 530
H 540
H SSO
H 560
H 570
H 580
H 590
H 600
H 610
H 620
H 630
H 640
H 650
H 660
H 670
H 680
H 690
H 700
H 710
K 720
H 730
H 740
h 750
H 760
H 770
H 780
H 790
H 8UO
H 810
H 820
H 830
H 840
H 850
H 860
H 870
H 880
H 890
H 900
H 910
H 920
H 930
H 940
M 950
H 960
H 970
H 980
H 990
H1000
H1010
H1020

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                        71
                        FORTRAN IV program listing—Continued
  150
  160
  170
  180
  190
  200

  210
  220
  230
  240
  250
  260
  270
  280
  290
  300
 FORMAT  (8X,16HTIME    = /1E12.5)
 FORMAT  <8X,1oHTIME(YEARS)   = /1E12.5)
 FOKMAT  <1H )
 FORMAT  (1HO/10F12.7/1CF12.7)
 FORMAT  (1HO/20I4)
 FORMAT  <1HG,2X,33HRATE MASS BALANCE --  (IN
1 /G12.5/10X/8HQOUT *  /G12.5/10X/8HQNET «
 FORMAT  (1H /17X/8HTPUM *  /G12.5)
        (1H /17X/8HDELS *  /G12.S/)
        (4X/29HWATER RELEASE FROM STORAGE
C.F.S.) //10X/8HQIN
/G12.5/)
 FORMAT
 FORMAT (4X/29HWATER RELEASE FROM  STORAGE *  /1E12.5)
 FORMAT (1HO,2X,23HCUMULATIVE MASS BALANCE//)
 FORMAT (4X/29HCUMULATIVE  NET  PUMPAGE   «  /1E12.5)
 FORMAT (4X/29HCUMULATIVE  NET  LEAK.AGE   •  /1E12.5)
 FORMAT (1HO/7X/25HMASS BALANCE RESIDUAL  *  /G12.5)
 FORMAT (1H /7X/25HERROR   (AS PERCENT)    *  /G12.5/)
 FORMAT (1H1/8HDRAWDOWN)
 FORMAT OH   /20I5)
 END
 SUBROUTINE CHMOT
 REAL *8TMRX/VPRMrHI/HR/HC/HK/WT/REC,RECH,TIM,AOPT/TITLE
 REAL •8XDEL/YDEL/S/AREA/SUMT/RHO/PARAM/TEST/TOL/PINT/HMIN/PYR
 COMMON /PRMI/ NTIM/NPMP/NPNT/NITP/N/NX/NY/NP/NREC/INT/NNX/NNY/NUMO
1BS/NMOV/IMCV/NPMAX/1TMAX/NZCRIT/1PRNT/NPTPND/NPNTMV/NPNTVL/NPNTD/N
2PNCHV/NPDELC
 COMMON /PRMK/ NODE ID(20/20)/NPCELL(20/20)/LIMBO(500>/IXOBS(5)/IYOB
1S(5)
 COMMON /HE DA/ THCK(20/20 ) /PERM(20/20)/TMWL(5/50)/TMOBS(50)/ANFCTR
 COMMON /HE OB/ THRX(20,20/2)/VPRM(20/20)/HI{20/20)/HR(20/20)/HC(20/
120)/HK(20/20)/WT(20/20)/REC(20/20)/RECH(20/20)/TIM(100)/AOPT(20)/T
2ITLE(10)/XDEL/YDEL/S/AREA/SUMT/RHO/PARAM/TEST/TOL/PINT/HMIN/PYR
 COMMON /CHMA/ PART(3/3200)/CONC(20/20)/TMCN(5/50)/VX<20/20)/VY(20/
120)/CONlNT(20/20)/CNRECH(20/20)/POROS/SUMTCH/BETA/TIMV/STORM/STORM
2I/CMSIN/CMSOUT/FLMIN/FLMOT/SUMIO/CELDIS/DLTRAT/CSTORM
 DIMENSION IC(20)
                  I
 TMFY«86400. 0*365.25
 TMYR»SUMT/TNFY
 TCHD-SUMTCH/86400.0
 TCHYR«SOMTCH/TMFY
 IF (IPRNT.GT.O) GO TO 100

 	PRINT CONCENTRATIONS	
 WRITE (6,160)
 WRITE (6/170) N
 IF (N.GT.O) WRITE  (6/180) TIM(N)
       (6/190) SUMT
               SUMTCH
               TCHD
               TMYR
               TCHYR
               I MOV
   10
   20
C
C
 WRITE
 WRITE
 WRITE
 WRITE
 WRITE
 WRITE
 WRITE
 DO 20
 DO 10
 IC(IX)»CONC(IX/IY)*0.5
 WRITE (6/240) (IC(IX)/IX*1/NX)

 IF (N.EO.O) GO TO 150
 IF (NPDELC.EO.O) 60 TO 50
 	PRINT CHANGES IN CONCENTRATION	
 WRITE  (6/230)
            (6/450)
            (6/200)
            (6/210)
            (6/460)
            (6/380)
            (6/220)
            IY-1/NY
            IX-1/NX
                                       i »********•»***'
H1030
HlOtO
H1050
H1060
H1070
H1080
H1090
H1100
H1110
H1120
H1130
H1140
H1150
H1160
H1170
H1180
H1190
H1200-
                            10
                            20
                            30
                            40
                            50
                            60
                            70
                            80
                            90
                            100
                            110
                            120
                            130
                            140
                            150
                            160
                            170
                            160
                            190
                            200
                            210
                            220
                            230
                            240
                            250
                            260
                            270
                            280
                            290
                            300
                            310
                            320
                            330
                            340
                            350
                            360
                            370
                            380
                            390
                            400
                            410
                            420
                            430
                            440

-------
 72
              TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
                        FORTRAN IV program toting—Continued
C
C
C
C
(6*170)
(6,180)
(6/190)
(6/450)
(6,200)
(6/210)
(6/460)
(6/380)
(6/220)
IY»1/NY
IX«1/NX
N
TIM(N)
SUMT
SUMTCH
TCHO
TMYR
TCHYR
I MOV
    WRITE
    WRITE
    WRITE
    WRITE
    WRITE
    WRITE
    WRITE
    WRITE
    WRITE
    00 40
    00 30
    CN6>CONC(IX/IY)-CONINT(IX/IY)
 30 IC(IX)*CN6
 40 WRITE  (6/240) (1C( I X)/IX*1/NX)
            >*»**•
    	PRINT MASS BALANCE DATA FOR SOLUTE	
 50 RESID«SU«iO-CSTORM
    IF (SUMIO.EO.0.0) GO TO 60
    RESID-SUM10-CSTORM
    ERR1-RESI 0*200.0/(SUMIO+CSTORM)
 60 IF (STORMI.EO.0.0) 60 TO 70
    ERR3--100.0*RES10/(STORM1-SUMJO)
 70 WRITE  (6/220)
    WRITE  (6/2SO)
    WRITE  (6/220)
    WRITE  (6/260) FLMIN
    WRITE  (6/270) FLMOT
    RECIN--CMSIN
    RECOUT«-CMSOUT
    WRITE  (6/290) RECIN
    WRITE  (6/280) RECOUT
    WRITE  (6/300) SUMIO
    WRITE  (6/310) STORMI
    WRITE  (6/320) STORM
    WRITE  (6/330) CSTORM
    IF (SUMIO.EQ.0.0) 60 TO 80
    WRITE  (6/340)
    WRITE  (6/3SO) RESIO
    WRITE  (6/360) ERR1
 80 IF (STORMI.EO.0.0) 60 TO 90
    WRITE  (6/370)
    WRITE  (6/360) ERR3
    A****************************************
    	PRINT HYOR06RAPHS AFTER 50 STEPS
 90 IF (MODUMOV,50).EO.O.AND.S.£0.0.0)
    IF (MOD(N/50).EQ.O.AND.S.6T.O.O) 60
    60 TO  1SO
100 WRITE  (6/390) TITLE
    IF (NUMOBS.LE.O) 60 TO 150
    WRITE  (6/400) INT
    IF (S.6T.O.O) WRITE (6/410)
    IF (S.EO.0.0) WRITE (6/420)
       	TABULATE HVOR06RAPH OATA	
    M02«0
    IF (S.6T.O.O) 60 TO 110
    NTO-NMOV
    IF (NNOV.6T.50) NTO*MOD(IMOV/SO)
    60 TO  124)
110 NTO-NTIM
    IF (NTIN.6T.SO) NTO»MOD(N/50>
120 IF (NTO.EO.O) MTO-50
    00 140 J»1/NUMOBS
     ****••**••***»***•»**•
OR END OF SIMULATION	
60 TO 100
TO 100
  450
  460
  470
  480
  490
  500
  510
  520
  530
  540
  550
  560
  570
  580
  590
  600
  610
  620
  630
  640
  650
  660
  670
  680
  690
  700
  710
  720
  730
  740
  750
  760
  770
  780
  790
  800
  810
  820
  830
  840
  850
  860
  870
  880
  890
  900
  910
  920
  930
  940
  950
  960
  970
  980
  990
11000
11010
11020
11030
11040
11050
11060

-------
                  MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                           73
                        FORTRAN IV program listing—Continued
  130
  no

  150
    TMYR-0.0
    WRITE (6*430) J»IXOBS(J>,IYOBS•**•*
C
c
C
c
    RETURN
    ****** I
  160
  170
  180
  190
  200
  210
  220
  230
  240
  250
  260
  270
  280
  290
  300
  310
  320
  330
  340
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
    FORMAT
   1TION:)
350 FORMAT
360 FORMAT
370 FORMAT
                                               >*»****(
(1H1/13HCONCENTRATION/)
(1X/23HNUMBER OF TIME STEPS > »115)
(8X/16HDELTA T       • /1G12.5)
(8X,16HTIME(SECONDS) • ,1612.5)
(3X,21HCHEM.TIME(OAYS>    • /1E12.5)
(8X,16HT1ME(YEARS)   « /1E12.S)
(1H )
(1H1/23HCHANGE IN CONCENTRATION/)
(1HO/20I5)
(1H /21HCHEMICAL MASS BALANCE)
(8X/25HMASS IN BOUNDARIES       /1E12.5)
(8X/25HMASS OUT BOUNDARIES      /1E12.5)
(8X/25HMASS PUMPED OUT          /1E12.5)
(8X/25HMASS PUMPED IN           /1E12.5)
(8X/25HINFLOW MINUS OUTFLOW     /1E12.5)
(8X/25H1NITIAL MASS STORED      /1E12.S)
(8X/25HPRESENT MASS STORED      /1E12.5)
(8X/25HCHANGE MASS STORED       /U12.5)
(1H /5X/S3HCOMPARE RESIDUAL WITH NET FLUX
AND MASS ACCUMULA
             (8X/25HMASS BALANCE RESIDUAL  • /1E12.5)
             (8X/25HERROR  (AS PERCENT)    • /1E12.5)
             (1H /5X/55HCOMPARE INITIAL MASS STORED  WITH
                                            CHANGE  IN  MASS
                               COMPLETED • ,115)
   ISTORED:)
380 FORMAT (1X/23H NO. MOVES
390 FORMAT (1H1/10A8//)
400 FORMAT (1HO/5X/65HTIME  VERSUS  HEAD  AND  CONCENTRATION AT SELECTED 0
   1BSERVATION POINTS//1Sx/19HPUMPING PERIOD  NO.  »I4////)
410 FORMAT (1HO/16X,19HTRANSIENT   SOLUTION////)
420 FORMAT (1HO/15X/21HSTEAOY-STATE  SOLUTION////)
430 FORMAT (1HO/20X,22HOBS.WELL NO.     X    Y,17X/1HN,6X,40HHEAD (FT)
   1   CONC.(MG/L)    TIME  (YEARS)//24X/I3/9X,\2/3X/IIII)
440 FORMAT (1H /58X,I 2/6X/F7.V8X/F7.1/8X,F7. 2)
450 FORMAT (1H /2X/21HCHEM.TIME(SECONDS)  •  /E12.S)
460 FORMAT (1H /2X/21HCHEM.TIME(YEARS)    *  /E12.5)
    END
11070
11080
11090
11100
11110
11120
11130
11140
11150
11160
11170
11180
11190
11200
11210
11220
11230
11240
11250
11260
11270
11280
11290
11300
11310
11320
11330
11340
11350
11360
11370
11380
11390
11400
11410
11420
11430
11440
11450
11460
11470
11480
11490
11500
11510
11520
11530
11540
11550
11560-

-------
74
TECHNIQUES OF WATER-RESOURCES  INVESTIGATIONS
                                  Attachment  II
              Definition  of Selected  Program  Variables
AAQ       area of aquifer in model
ALNG      BETA
ANFCTR   anisotropy factor (ratio of T,, to Tu)
AOPT      iteration parameters
AREA      area of one cell in finite-difference grid
BETA      longitudinal dispersivity of porous
             medium
CELDIS    maximum distance across one cell that
             a particle is permitted to move in
             one step (as fraction of width of
             cell)
CLKCN    concentration of leakage through con-
             fining layer or streambed
CMSIN    mass of solute recharged into aquifer
CMSOUT   mass of solute discharged from aquifer
CNCNC    change in concentration due to disper-
             sion and sources
CNCPCT   change in concentration as percentage
             of concentration at node
CNOLD    concentration at node at end of pre-
             vious time increment
CNREC    concentration of well withdrawal or
             injection
CNRECH   concentration in fluid source
CONG      concentration in aquifer at node
CONINT   concentration in aquifer at start of
             simulation
Cl        CONG at node (IX.IY)
DALN     longitudinal dispersion coefficient
DDRW    drawdown
DELQ     volumetric rate of leakage across a
             confining layer or streambed
DELS      rate of change in ground-water storage
DERH     change in head with respect to time
DISP      dispersion equation coefficients
DISTX    distance particle moves in ^-direction
             during time  increment
DISTY    distance particle moves in y-direction
             during time  increment
DLTRAT   ratio of transverse to longitudinal
             dispersivity
DTRN     transverse dispersion coefficient
FCTR      multiplication or conversion factor
FLMIN    solute  mass entering modeled area
             during time  step
FLMOT    solute  mass leaving modeled area
             during time  step
GRDX     hydraulic gradient in ^-direction
GRDY     hydraulic gradient in y-direction
HC        head from column computation
HI        initiaLhead in aquifer
HK        computed head at end of time step
HMIN     minimum iteration parameter
                             HR        head from row computation in sub-
                                          routine ITERAT; elsewhere HR
                                          represents head from previous time
                                          step
                             IMOV     particle movement step number
                             INT       pumping period number
                             IPRNT    print control index for hydrographs
                             ITMAX    maximum permitted number of
                                          iterations
                             IXOBS    x-coordinate of observation point
                             IYOBS    y-coordinate of observation point
                             KOUNT   iteration number for ADIP
                             LIMBO    array for temporary storage  of
                                          particles
                             N         time step number
                             NCA      number of aquifer nodes in model
                             NCODES  number of node identification codes
                             NITP     number of iteration parameters
                             NMOV    number of particle movements (or time
                                          increments) required to complete
                                          time step
                             NODEID  node identification code
                             NP        total number of active particles in grid
                             NPCELL  number of particles in a cell  during
                                          time increment
                             NPMAX   maximum number of available particles
                             NPHP    number of pumping periods or simu-
                                          lation periods
                             NPNT    number of time steps between printouts
                             NPTPND  initial number of particles per node
                             NREC    number of pumping wells
                             NTIM     number of time steps
                             NUMOBS  number of observation wells
                             NX        number of nodes in x-direction
                             NY        number of nodes in y-direction
                             NZCRIT   maximum number of cells that can be
                                          void of particles
                             NZERO   number of cells that are void of
                                          particles at the  end of a time
                                          increment  .
                             PARAM    iteration parameter for current
                                          iteration
                             PART     1. x-coordinate of particle;  2. y-coordi-
                                          nate of particle;  3. concentration of
                                          particle. Also note that the signs of
                                          coordinates are used as flags to store
                                          information on original location of
                                          particle.
                             PERM     hydraulic conductivity (in LT1)
                             PINT     pumping period in years
                              POROS    effective porosity
                              PUMP    cumulative net pumpage
                              PYR       total duration of pumping period
                                           (in seconds)
                              QNET     net water flux (in LT ')

-------
                      MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                             75
                          Definition of teleeted program variables—Continued
QSTR      cumulative change in volume of water
              in storage
REG        point source or sink; negative for in-
              jection,  positive  for withdrawal
              (in L'T')
RECH      diffuse recharge or discharge; negative
              for recharge, positive for discharge
              (in LT-1)
RN         range in concentration between regen-
              erated particle and adjacent node
              having lower concentration
RP         range in concentration between regen-
              erated particle and adjacent node
              having higher concentration
S           storage coefficient (or specific yield)
SLEAK     rate of leakage through confining
              layer or streambed
STORM     change in  total  solute mass in storage
              (by summation)
STORMI    initial mass of  solute in storage
SUMC      summation of concentrations  of all
              particles in a cell
SUMIO     change in total solute mass in storage
              (from inflows—outflows)
SUMT      total elapsed time (in seconds)
SUMTCH   cumulative elapsed time during
              particle  moves (in seconds)
THCK      saturated  thickness of aquifer
TIM        length  of specific  time step
              (in seconds)
TIMD      elapsed time in days
TIMY      elapsed  time in years
TIMV      length of time increment for particle
              movement (in seconds)
TIMX      time step multiplier for transient flow
              problems
TINIT      size of initial time step for transient
              flow problems (in seconds)
TITLE      problem description
TMCN      computed concentrations at observation
              points
TMOBS     elapsed times for observation point
              records
TMRX      transmissivity  coefficients  (harmonic
              means on cell boundaries;  forward
              values are stored)
TMWL      computed heads at observation points
TOL        convergence criteria (ADIP)
TOTLQ     cumulative net leakage through con-
              fining layer or streambed
TRAN      transverse dispersivity of porous
              medium
VMAX      maximum value of VX
VMAY      maximum value of VY
VMGE      magnitude of velocity vector
VMXBD    maximum value of VXBDY
VMYBD    maximum value of VYBDY
VPRM      initially used to read transmissivfty
              values at nodes; then after line
              B2270, VPRM equals leakance factor
              for confining layer or streambed
              (vertical  hydraulic  conductivity/
              thickness). If VPRM^0.09, then the
              program assumes that the node is a
              constant-head boundary and is flag-
              ged for subsequent special treat-
              ment in calculating convective trans-
              port.
VX         velocity in x-direction at a node
VXBDY    velocity in x-direction on a boundary
              between nodes
VY         velocity in y-direction at a node
VYBDY    velocity in {/-direction on a boundary
              between nodes
WT         initial water-table  or potentiometric
              elevation, or constant head in
              stream or source bed
XDEL      grid spacing in x-direction
XOLD      x-coordinate of particle at end of pre-
              vious time increment
XVEL      velocity of particle in x-direction
YDEL      grid spacing in y-direction
YOLD      ^-coordinate of particle at end of pre-
              vious time increment
YVEL      velocity of particle in y-direction

-------
76
TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
                                Attachment III
                             Data  Input  Formats
Card Column
1 1-80
2 1-4

6- 8


9-12

13-16

17-20

21-24


26-28

29-32


33-36

Format
10A8
14

14


14

14

14

14


14

14


14

Variable
TITLE
NTIM

NPMP


NX —illy' &
Btffl&L
NY y<&++

NPMAX

NPNT


NITP

NUMOBS


ITMAX

Definition
Description of problem
Maximum number of time steps in «
pumping period (limit=100)*.
Number of pumping periods. Note
that if NPMP>1, then data set
jj * /i 10 must be completed.
1b($j~~5WD- Number of nodes in x direction
^ 	 (limit=20)*.
(J Number of nodes in y direction
(limit=20)*.
Maximum number of particles
(limit=3200) *. (See eq 71.)
Time-step interval for printing
hydraulic and chemical output
data.
Number of iteration parameters
(usually 4^NITPs^7).
Number of observation points to be
specified in a following data set
(limit=6)*.
Maximum allowable number of it-
erations in ADIP (usually 100
        37-40


        41-44

        45-48


        49-62



        63-66



        67-60



        61-64


        66-68
       14


       14

       14


       14



       14



        14



        14


        14
NREC


NPTPND

NCODES


NPNTMV



NPNTVL



NPNTD



NPDELC


NPNCHV
Number of pumping or injection
  wells to be specified in a following
  data set
Initial number of particles per node
  (optiems=4, 6, 8, 9).
Number of node identification codes
  to be specified in a following data
  set (limit=10)*.
Particle movement interval (IMOV)
  for printing chemical output data.
  (Specify 0 to print only at end of
  time steps.)
Option for printing computed veloci-
  ties (0=do not print; l=print for
  first time step; 2=print for all
  time steps).
Option for printing computed dis-
  persion equation coefficients  (op-
  tion definition same as for
  NPNTVL).
Option for printing computed
  changes in concentration  (0=do
  not print; l=print).
Option to punch velocity data  (op-
  tion definition same as for
  NPNTVL). When specified,  pro-
  gram  will punch  on  unit 7  the
  velocities at nodes.
   Set footnote* at rad of UbU.

-------
                  MODEL OP SOLUTE TRANSPORT IN GROUND WATER
                                                                 77
                                Data input formate—Continued
Card Column
8 1-6
6-10

11-15
16-20

21-26

26-30


81-55

36-40

41-46

46-60

61-66

66-60
Date Number
•et of eardi
Format
G5.0
G6.0

G6.0
G6.0

G6.0

G6.0


G6.0

G6.0

G6.0

G5.0

G6.0

G5.0

Format
Variable
PINT
TOL

POROS
BETA

S

TIMX


TINIT

XDEL

YDEL

DLTRAT

CELDIS

ANFCTR

Variable
Definition
Pumping period in years.
Convergence criteria in ADIP
(usually TOL^O.Ol).
Effective porosity.
Characteristic length, in feet
(=longitudinal dispersivity).
Storage coefficient (set 5=0 for
steady flow problems).
Time increment multiplier for trans-
ient flow problems. TIMX is dis-
regarded if 5=0 .
Size of initial time step in seconds.
TINIT is disregarded if 5=0.
Width of finite-difference cell in
x direction, in feet.
Width in finite-difference cell In
y direction, in feet.
Ratio of transverse to longitudinal
dispersivity.
Maximum cell distance per particle
move (value between 0 and 1.0).
Ratio of T,, to T...

Definition
 1    Value of NUMOBS   212
        (limit=5)*
 2    Value of NREC
      a. 1
      b. Value of NY
          (limit=20)*
      a. 1
      b. Value of NY
           (limit=20)*
      a. 1
      b. Value of NY
           (limit=20)*
      a. 1
      b. Value of NY
           (limit=20)»
212,
2G8.2
          IXOBS, IYOBS
IX, IY, REG, CNRECH
II, GlO.O  INPUT, FCTR
20G4.1    VPRM
II, GlO.O  INPUT, FCTR
20G3.0    THCK

II, GlO.O  INPUT, FCTR
20G4.1    RECH

II, GlO.O  INPUT, FCTR
2011      NODEID
x and y coordinates of observation
  points. This data set is eliminated
  if NUMOBS is specified as =0.
x and y coordinates of pumping (+)
  or injection  (—)  wells, rate  in
  ft'/s, and if an injection well, the
  concentration  of injected water.
  This data set is eliminated if
  NREC=0.
Parameter card t for transmissivity.
Array for temporary storage of
  transmissivity data, in ff/s. For
  an  anisotropic aquifer, read in
  values of T,, and the program will
  adjust for anisotropy by multi-
  plying Tn by ANFCTR.
Parameter cardt for THCK.
Saturated  thickness of aquifer, in
  feet.
Parameter  cardt for RECH.
Diffuse recharge  (—) or discharge
  (+),inft/s.
Parameter  cardt for NODEID.
Node identification matrix  (used to
  define constant-head nodes or
  other boundary conditions and
  stresses).
8e* footnotes at end of table.

-------
78
        TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                                  Data input formats—Continued
 Data
  get
Number
of cards
                            Format
                                             Variable
                                                                              Definition
       Value of NCODES
         (limit=10)»
               I2.3G10.2, ICODE, FCTR1,
                 12         FCTR2, FCTR3,
                            OVERRD
 10
       a. 1
       b. Value of NY
            (limit=20)*

       a. 1
       b. Value of NY
            (limit=20)*
               II, G10.0   INPUT, FCTR
               20G4.0     WT
               II, G10.0
               20G4.0
INPUT, FCTR
CONG
       a. 1
                II
ICHK
        b. 1                10I4.3G5.0  NTIM, NPNT,  NITP,
                                        ITMAX, NREC,
                                        NPNTMV, NPNTVL,
                                        NPNTD, NPDELC,
                                        NPNCHV, PINT, T1MX,
                                        TINIT

        c. Value of NREC   212, 2G8.2  IX,  IY,  REC, CNRECH
Instructions for using NODEID
  array. When NODEID=ICODE,
  program sets Ieakance=FCTRl,
  CNRECH=FCTR2, and if
  OVERRD is nonzero,  RECH
  =FCTR3. Set OVERRD=0 to
  preserve values of RECH specified
  in data set 5.
Parameter cardt for WT.
Initial water-table or potentiometric
  elevation, or constant head in
  stream or source bed, in feet.
Parameter cardt for CONG.
Initial concentration in aquifer.

This data set allows time step param-
  eters, print options, and pump-
  age data to be revised for  each
  pumping period of the simulation.
  Data set 10 is only used if NPMP
  >1. The sequence of cards  in data
  set 10  must be  repeated (NPMP
  —1) times (that is, data set 10
  is required for each pumping
  period  after the first).
Parameter to check whether any re-
  visions are desired. Set ICHK=1
  if data are to be revised, and then
  complete data set  lOb  and c. Set
  ICHK=0 if data are not to be re-
  vised for the next pumping period,
  and skip rest of data set 10.
Thirteen parameters to  be  revised
  for next pumping period; the
  parameters were previously  de-
  nned in the description of  data
  cards 2  and 3.  Only include this
  card if ICHK=1 in previous part
  a.
Revision  of previously defined  data
  set 2. Include part c only  if
  ICHK=1 in previous  part a and
  if NREC>0 in previous part b.
  • These HmiU can be modified if necessary by changing the corresponding  array dimensions in the COMMON statements of  the
program.
  f The  parameter  card must be the first card of the Indicated data sets. It is used to specify whether the parameter is constant
and uniform, and can be defined by one value, or whether it varies in space and must be defined at each node. If INPUT = 0,  the
data set has a constant value, which la defined by FCTR. If IN PUT =1, the data set la read from cards as described by part b.
Then  FCTR is a multiplication factor for the values read In the data set

-------
                MODEL OF SOLUTE TRANSPORT IN GROUND WATER
                                                                 79
                            Attachment  IV
                  Input  Data  for  Test  Problem 3
  Card I
  Card 2
  Card3

Data Set 1 •

Data Set 2
Data Set 3
Data Set 4
Data Set 5
Data Set 6
TEST PROBLEM NO. 3 (STEADY  FLOW,  1 WELL/ CONSTANT-HEAD BOUNDARIES)
   1   1    9  103200    1    7    2  100   1   9   2  10   1   0   0
  2.5.0001  0.30 100.   0.0   0.0  0.0 900. 900.  0.3 0.50  1.0
 5 4
 5 7
 A 7   1.0
0  0.1
0  20.0
0  0.0
1  1.0
OOOOOOOOO
022111220
OOOOOOOOO
coooooooo
ooooooooo
ooooooooo
ooooooooo
ooooooooo
022222220
OOOOOOOOO
Data Set 7 •
Data Set 8
 1    1.0
1   1.0
                     100. U
                         ii.O
        0.0100.100.100.100.100.100.100. 0.0
        0.0  75. 75. 75. 75. 75. 75. 75. 0.0


Data Set9  0  0.0

-------
80
TECHNIQUES OP WATER-RESOURCES INVESTIGATIONS
                                Attachment  V
                 Selected  Output  for Test  Problem  3
      U.S.&.S. METHOD-OF-CHARACTERIST1CS MODEL FOR SOLUTE  TRANSPORT IN GROUND WATER

      TEST  PROBLEM NO. 3 (STEADY  FLOW,  1 WELL. CONSTANT-HEAD  BOUNDARIES)

                          INPUT     DATA

                            GRID DESCRIPTORS

                 NX    (NUMBER OF  COLUMNS)  «     9
                 NY    (NUMBER OF  ROWS)     a    10
                 XDEL  (X-OISTANCE  IN  FEET) *   900.0
                 YDEL  (Y-OISTANCE  IN  FEET) =   900.0

                            TIME  PARAMETERS
                 NTIM   (MAX.  NO.  OF  TIME STEPS)
                 NPMP   (NO.  OF  PUMPING PERIODS)
                 PINT   (PUMPING PERIOD IN YEARS)
                 TIMX   (TIME  INCREMENT MULTIPLIER)
                 TINIT  (INITIAL TIKE STEP IN SEC.)

                   HYDROLOGIC  AND  CHEMICAL PARAMETERS

                 S       (STOWAGE  COEFFICIENT)       '
                 POROS   (EFFECTIVE POROSITY)        =
                 BETA    (CHARACTERISTIC LENGTH)      •
                 DLTRAT  (KATIO  OF TRANSVERSE TO
                         LONGITUDINAL 01SPERSIVITY)   «
                 ANFC1K  (RATIO  OF T-YY TO T-XX)      •

                          EXECUTION PARAMETERS

                 NITP   (NO.  OF  ITERATION PARAMETERS)
                 TOL    (CONVERGENCE  CRITERIA - ADIP)
                 ITMAX  (MAX.NO.OF ITERATIONS - ADIP)
                 CELDIS (MAX.CELL  DISTANCE PER M3VE
                            OF PARTICLES - M.O.C.)
                 NPMAX  (MAX.  NO.  OF  PARTICLES)
                 NPTPND (NO.  PARTICLES PER NODE)
                            PROGRAM  OPTIONS

                  NPNT   (TIME STEP INTERVAL FOR
                         COMPLETE  PRINTOUT)
                  NPNTKV (MOVE INTERVAL  FOR CHEM.
                         CONCENTRATION  PRINTOUT)
                  NPNTVL (PRINT OPTION-VELOCITY
                         0«NO; 1«F1RST  TIME STEP.'
                         2 = ALL T-IME  STEPS)
                  NPNTD  (PRINT OPTION-DJSP.COEF.
                         0«NO; 1»F1RST  TIME STEP.'
                         2»ALL TIME  STEPS)
                  NUMObS (NO. OF OBSERVATION WELLS
                         FOR HYDROGRAPH PRINTOUT)
                  NREC   (NO. OF PUMPING WELLS)
                  NCODtS (FOR NODE  IOENT.)
                  NPNCHV (PUNCH VELOCITIES)
                  NPDELC (PRINT OPT.-CONC. CHANGE)
                                           1
                                           1
                                           2.50
                                           0.00
                                           0.
                                         0.000000
                                         0.3U
                                        100.0

                                         0.30
                                         1.000000
                                         7
                                         0.0001
                                       100

                                         o.sou
                                      3200
                                         9
                                      1

                                     10
                                      2
                                      1
                                      2
                                      0
                                      0

-------
         MODEL OF SOLUTE TRANSPORT IN GROUND WATER


               Selected output for test problem 8—Continued


           STEADY-STATE  FLOW

TIME INTERVAL (IN  SEC)  FOR  SO'LUTE-TRANSPORT SIHULATION «  0.768944*06

      LOCATION OF  OBSERVATION WELLS

            NO.      X      Y
81
                    5
                    5
      LOCATION  OF   PUhPING  WELLS

      X   Y   RATEdN  CFS)   CONC.

      4   7      1.00         0.0

      AREA OF ONE  CELL =   0.81004*06

      X-Y SPACING:
          900.00
          90C.OO
TRANSnISSIVI
0.00 0.00
0.00 0.10
0.00 0.10
0.00 0.10
0.00 0.10
0.00 0.10
0.00 0.10
0.00 0.10
0.00 0.10
0.00 O.OU
TY MAP
0.00
0.10
0.10
0.11
0.10
0.10
0.10
0.10
0.10
o.cc
(FT«FT/SEC)
0.00 0.00 0.00
0.10 0.1C 0.10
0.10 0.10 0.10
0.10 0.10 0.10
(J.10 0.10 0.10
0.10 0.10 0.10
0.10 0.10 0.10
0.10 0.10 0.10
0.10 0.10 0.10
0.00 0.00 0.00
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
00
10
10
10
10
10
10
10
10
00
0
0
0
0
0
0
0
0
0
0
.00
.10
.10
.10
.10
.10
.10
.10
.10
.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
AQUIFER
0
0
0
0
0
0
0
0
0
0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
THICKNESS
0.0
20.0
2U.O
20.0
20. U
20.0
20.0
20.0
20.0
0.0
0.0
20. C
20.0
20.0
20.0
20. U
20.0
20. C
20.0
0.0
(FT)
tl.C
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
0.0

0
20
20
20
20
20
20
20
20
0

.U
.0
.0
.0
.0
.0
.0
.0
.0
.0

0
20
20
20
20
20
20
20
20
0

.0
.0
.0
.0
.0
.0
.0
.0
.0
.0

0.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
0.0

0.0
20.0
20.0
20.0
20.0
20. U
20.0
20.0
20.0
0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0

-------
82
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

        Selected output for teat problem S—Continued
DIFFUSE RECHARGE AND
O.OOd+OU
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOa+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
DISCHARGE
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
(fT/SEC)
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00

O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.DOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00

O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+UO
O.OOd+00

O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00

O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00
O.OOd+00

3.30d+00
0.30d»00
3.30d+00
O.DOd+00
O.DDd+00
0.33d+00
0.30d+00
0.30d+00
0.30d+00
3.334+00
PERMEABILTY NAP (FT/SEC)
  0.0000.0000.0000.0000.OCOO.0000.0000.0000.000
  0.0000.0050.0050.0050.0050.0050.0050.0050.000
  0.0000.OOSO.OOSO.00 SO.OOSO.OOSO.OOSO.OOSO.000
  0.0000.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.000
  0.0000.OOSO.OCSO.OOSO.OOSO.OOSO.OOSO.OOSO.000
  0.0000.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.000
  0.0000.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.OOSO.000
  0.0000.OOSO.OOSO.00 SO.OOSO.OOSO.OOSO.DOS 0.000
  0.0000.U050.0050.0050.0050.0050.0050.0050.000
  0.0000.0000.0000.0000.UOOO.0000.0000.0000.000
         NO.  OF FINITE-DIFFERENCE  CELLS  IN  AQUIFER  *    56

         AREA OF AQUIFER IN MODEL   •   O.*5360«+08    SO.  FT.
         NZCRIT
                  (MAX.  NO. OF  CELLS  THAT  CAN  BE  VOID  OF
                   PARTICLES;  IF  EXCEEDED,  PARTICLES  ARE  REGENERATED)

-------
  MODEL OF SOLUTE TRANSPORT IN GROUND WATER

        Selected output for tett problem S—Continued

NODE 1DENT1FICAT10N HAP
                                   83
0
0
0
0
0
0
0
0
0
0
0
2
u
0
0
0
0
0
2
0
0
I
0
0
0
0
0
0
2
0
0
1
0
0
0
0
0
0
2
0
0
1
0
0
0
0
0
0
2
0
0
1
0
0
0
0
0
0
2
0
0
2
0
0
0
0
0
0
2
0
0
2
0
0
0
0
0
0
2
0
0
0
u
0
0
0
0
0
0
0
     NO. OF NODE IDENT.  COOES SPECIFIED *   2

          THE FOLLOWING  ASSIGNMENTS HAVE BEEN MADE:
    CODE NO.     LEAKANCE     SOURCE  CONC.      RECHARGE
                 0.100e + U

                 0.100e*01
  0.00

100.00
TICAI
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0
1
0
0
0
u
0
0
1
0
PERMEABILITY/TH1
.00 0.00 0.00 0.
.00 1.00 1.00 1.
.00 0.00 0.00 0.
.00 0.00 0.00 0.
.00 C.CC 0.00 0.
.00 0.00 0.00 0.
.00 0.00 0.00 0.
.00 0.00 0.00 0.
.00 1.00 1.00 1.
.00 0.00 0.00 0.
CKNESS
00 0.00
00 1.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 1.00
00 0.00
(FT/ (FT«SEC> )
0.00 0.00 0.00
1.00 1.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
0.00 0.00 0.00
1.00 1.00 0.00
0.00 0.00 0.00
 WATER  TABLE
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
100.
0.
0.
0.
0.
0.
0.
75.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.

-------
84
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

        Selected output for test problem 3—Continued
ITERATION PARAMETERS
0.2*67*00-01
O.A57299d-01
0.8*75390-01
.157080
.291125
.539560
1.00000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
o.coooco
0.000000
o.ooooco
o.oorrooo
0.000000
CONCENTRATION
NUMBER OF TIME STEPS
I IME< SECONDS)
CHEM.T1M£< SECONDS)
CHEM. TIME ( DAYS)
TIME< YEARS)
CHEM.T1M£( YEARS)
NO. MOVES C
0
0
Q
0
0
0
U
0
0
0
0
0
0
0
0
0
0
0
0
0
0
U. 00000
O.OOOOOe+00
O.OOOOOe+00
C.OOOOOe+00
O.OOOOOetOO
OMPLETEO
0
U
0
0
0
0
0
0
0
0
0
0
0
0
0
0
U
0
U
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0

0
0
0
0
0
0
0
0
0
0

0
0
Q
0
0
0
0
0
0
0
                       N •    1
                       NUMBER OF  ITERATIONS *   20

-------
                                      aeieetea output /or teit problem 3—Continued
HEAD DISTRIBUTION - ROU
NUMBER OF  TIME STEPS *      1
       TIHE(SECONDS) *   0.788944*08
       TIME    *   0.91313**03
       TIME(TEARS)   •   0.25000«*01
0.0000000 0.0000000
0.0000000 99.9999995
0.0000000 95.9387858
0.0000000 91.8816815
0.0000000 87.8530674
0.0000000 83.9382225
0.0000000 80.3627221
0.0000000 77.5265176
0.0000000 75.0000003
0.0000000 0.0000000
HEAD DISTRIBUTION - ROU
NUMBER OF TIME STEPS «
TIME(SECONOS) « 0
TIMECOATS) • 0
TIME(TEARS) • 0

000000
0 100 100 100 100 100
0 96 96 96 96 96
0 92 91 92 92 92
0 88 88 88 88 88
0 84 84 83 84 84
0 80 80 77 80 81
0 78 77 77 77 78
0 75 75 75 75 75
000000
0.0000000 0.0000000 0.0000000 0.0000000 0.0000033
99.9999995 99.9999995 99.9999995 99.9999995 99,9999995
95.9346978 95.9468712 95.9958792 96.0611455 96.1171357
91.8531641 91.8569301 91.9755221 92.1315893 92.2591335
87.7393101 87.6521342 87.9176617 88.2305223 88.4600393
83.5988909 83.0946482 83.8124811 84.4128118 84.774712?
79.6233998 77.3151005 79*8248158 80.8335448 81.2863911
77.2168501 76.7175099 77.3381095 77.8101323 78.0688953
75.0000003 75.0000002 75.0000003 75.0000003 75.0000034
0.0000000 0.0000000 0.0000000 0.0000000 0.0000033
1
.788944*08 »
.91313**03
.25000e*01

000
100 100 0
96 96 0
92 91 d
88 89 0
85 IS 0
81 81 0
78 78 0
75 75 0
000
0.0000000 0.0000000
99.9999995 0.0000000
96.1482887 0.0000000
92.3277521 0.0000000
88.5758019 0.0000000
84.9396259 0.0000000
81.4683757 0.0000000
78.1790838 0.0000000
75.0000004 0.0000000
0.0000000 0.0000000



i







O
M
f
o
axmos >•
H
»
VNSPORT IN G!
*v
O
55
O
S3
M




g.


-------
                                   Selected output for test problem 3—Continued
                                                                                                                   oo
DRAWDOWN
0
0
0
0
U
0
0
0
0
0

0
0
-95
-91
-87
-83
-79
-77
0
0

0
0
-95
-91
-87
-83
-79
-76
0
C

0
0
-95
-91
-87
-82
-76
-76
0
0

0
0
-95
-91
-87
-83
-79
-76
0
0

0
0
-95
-91
-87
-83
-80
-77
0
0

0
0
-95
-91
-87
-84
-80
-77
0
0

U
0
-95
-91
-88
-84
-80
-77
0
0

0
0
0
0
0
0
0
0
0
0
  CUMULATIVE  MASS  BALANCE

   CUMULATIVE  NET  PUMPAGE    «   0.78894e + li8
   WATER RELEASE FROM STORAGE  «   O.OOOOOe+00
   CUMULATIVE  NET  LEAKAGE    '   0.78895e»08

       MASS  BALANCE RESIDUAL   *   -767.00
       ERROR   (AS  PERCENT)     *  -0.97219e-03
                                                                                                                   H
                                                                                                                   W
                                                                                                                   O
&
a
m
CO
  RATE MASS  BALANCE — (IN  C.F.S.)
          OIN  «    2.7857
          OOUT *   -1.7857
          ONET
X VELOCITIES
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
   0.000
n
m 1 • Uw V
TPUM *
DELS *

O.OOU
0.935e-14
0.757e-07
0.528e-06
0.211e-05
0.628e-OS
0.137e-04
0.573e-05
0.708e-12
0.000

0.000
0.935e-14
0.757e-07
0.52de-06
0.211e-05
0.628e-05
0.137e-04
0.573«-05
0.708e-12
0.000
U
1.0000
0.00000
AT NODES

0.000
-0.924e-14
-0.749e-07
0.229e-06
0.186e-05
0.781e-05
0.282e-04
0.749e-05
0.925e-12
0.000
ON BOUNDARIES
0.000
-0.278e-13
-0.225e-06
-0.697e-07
0.161e-05
0.934*-05
0.427e-04
0.925e-05
0.1 14e-1 1
C.OOO


0.000
-0.6V9e-13
-0.566e-06
-0.113e-05
-0.165e-05
-0.198e-05
-0.1b6e-05
-0.112e-05
-0.139e-12
0.000

O.OCO
-0.112e-12
-0.908e-06
-0.220e-05
-0.492e-05
-0.133e-04
-0.465e-04
-0.115e-04
-0.142e-1 1
0.000


0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
0.

0.
-0.
-0.
-0.
-0.
-0 .
-0.
-0.
-0.
U.


000
131e-12
106e-05
254e-05
536e-05
122e-04
326e-04
101e-04
125e-11
000

000
149e-12
121«-05
289e-05
579e-05
1 1 1e-04
187e-04
874e-05
108e-11
000


0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
0.

0.
-0.
-0.


000
139e-12
112e-05
263e-05
502e-05
891e-05
135e-04
677e-05
835e-12
000

000
128e-12
104e-05
-0.236e-05
-0.
-0.
-0.
-0.
-0.
0.
42Se-OS
670e-OS
839e-OS
479e-05
592e-12
000


0.000
-0.996e-13
-0.807e-04
-0.182e-0>
-0.320e-35
-0.488«-35
-0.588e-35
-0.3*2e-35
-O.*22e-12
0.000

0.000
-0.712e-U
-0.577e-3S
-0.127e-3S
-0.214e-35
-0.305e-3>
-0.337e-35
-0.204e-35
-0.252e-ia
0.000


0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
0.

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.


000
712e-13
577e-06
127e-05
214e-05
J05«-05
337e-05
204e-05
252e-12
000

000
000
000
000
000
000
000
000
000
000


0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000

0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
RESOURCE
03
_,
z

to
H
o
>•
H
o
to












-------
                                    Selected output for teat problem S—Continued
T VELOCITIES
                          AT  NODES
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000


0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000





X «
BD»
(MAX
0.000 0.000
0.752*-04 0.753e-04
0.752e-04 0.7S4e-04
0.749e-04 0.7S9e-04
0.736e-04 0.764C-04
0.694e-04 0.751e-04
0.594e-04 O.S91e-04
0.497e-04 0.42Bv-04
0.468e-04 0.411e-04
0.000 0.000
ON BOUNDARIES

0.000 0.000
0.752e-04 0.753e-04
0.751e-04 0.756e-U4
0.746e-04 0.762e-04
0.725«-04 0.767e-04
0.662e-04 U.736e-04
0.525e-04 0.446e-04
0.468e-04 0.411e-04
0.000 0.000
0.000 0.000



STABILITY CRITERIA 	 M.O.C.

3.26e-C5 VHAY « 9.57e-05
4.65e-05 VMYBD= 1.07e-04
. INJ.) - 0.11955e«08
0.000
0.751e-04
0.754e-04
0.768e-04
0.811e-04
0.957e-04
0.590e-04
0.214e-04
0.318e-04
0.000


0.000
0.75U-04
0.757e-04
0.779e-04
0.844e-04
0.107e-03
0.111e-04
0.318e-04
0.000
O.OUO








0.000
0.742e-04
0.743e-04
0.748e-04
0.756e-04
0.749e-04
0. 599e-04
0.447e-0t
0.433e-04
0.000


0.000
0.742e-04
0.745e-04
0.751e-04
0.760e-04
0.738e-04
0.461e-04
0.433e-04
0.000
0.000








0.000
0.729«-04
0.729e-04
0.725e-04
0.715e-04
0.68Se-04
0.611e-04
O.S40«-04
O.S20e-04
0.000


0.000
0.729e-04
0.728«-04
0.722«-04
0.707e-04
0.663e-04
O.S60e-04
O.S20e-04
0.000
0.000








0.000
0.719e-3i
0.717e-3t
0.709*-34
0.693e-34
0.664e-3'<
0.621e-04
O.S82e-3'.
0.568e-0'«
0.000


0.000
0.719e-04
0.714e-3i
0.704e-0i
0.682e-3l
0.646e-3«
0.596e-34
O.S68e-34
0.000
0.000








0.000
0.713e-04
0.710e-04
0.701e-04
0.684e-04
0.6SBe-04
0.626e-04
O.S99e-04
0. 589e-04
0.000


0.000
0.713e-04
0.708e-04
0.695e-04
0.673e-04
0.643e-04
0.609e-04
O.S89e-04
0.000
0.000








0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000


0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000








iv (CELOIS) * 0.42045e»07













gg
o
o
f
o
•^
en
o
s
•-3

H
>
1^^
01
3
\J
H
^H
2!
0
o
o
<
^
H
w
90
TIHV «  4.20e»06
                      NTIHV  =
                          18
 T1H (N)   •   0.788944+08
 TIMEVELO  «   0.41523e»07
 TIMEOISP  =   0.30143«+08
TIMV
4.15e+06
                      NTIMD
                                         NflOV
                                         NMOV  *
                                                    19
                                                    19
         THE  LIMITING STABILITY CRITERION  IS  CELDIS

         NO.  OF PARTICLE MOVES REQUIRED TO  COMPLETE THIS  TIME  STEP
                                                                    19

-------
88
TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS
                          Selected output for teat problem 3—Continued
                CONCENTRATION
NunotN gr lint site's i
DELTA T 0.788944*08
T1ME( SECONDS) 0.78894d«08
CHEW. TIBE( SECONDS) 0.78894e+08
CHEM.TIME(OAYS) 0.91313e*03
TIME(YEARS) 0.2SOOOe+01
CHEN.TIHE(YEARS) 0.25000«+01
NO.
0
0
0
0
0
0
0
0
0
0
MOVES COMPLETED 19
0 U 0 0
0 2 98 100
0 4 96 100
0 7 92 99
0 9 89 96
1 10 81 89
1 8 56 73
0 2 20 3i
0 0 1 5
0000
0
98
96
93
88
80
46
19
3
0
000
200
400
700
900
10 1 0
8 1 0
300
000
000
CHEMICAL MASS BALANCE









MASS IN BOUNDARIES
MASS OUT BOUNDARIES
MASS PUMPED IN
MASS PUMPED OUT
INFLOW MINUS OUTFLOW
INITIAL MASS STORED
PRESENT MASS STORED
CHANGE MASS STORED
COMPARE RESIDUAL WITH








NET
MASS BALANCE RESIDUAL

ERROR (AS PERCENT)

0.946*2e»13
-0.13340e+OB
O.OOOOOe+03
-0.96281e+09
0.84881«+10
O.OOOOOe+00
0.84631e»10
0.84631e»10
FLUX AND MASS
0.24910e«08
0.29390e«00

-------
                                  StleeUd output for test problem S—Continued

TEST PROBLEM NO. 3 (STEADY  FLOW,  1  HE|_L»  CONSTANT-HEAD BOUNDARIES)
     TIME VERSUS HEAD AND  CONCENTRATION  AT  SELECTED OBSERVATION POINTS*

              PUMPING PERIOD  NO.     1
                                                                                                                o
                                                                                                                o
                                                                                                                W
                                                                                                                r
               STEADY-STATE  SOLUTION
OBS.WELL NO.    X

     1          5
                                                            0
                                                            1
                                                            2
                                                            3
                                                            4
                                                            5
                                                            6
                                                            7
                                                            8
                                                            9
                                                           10
                                                           11
                                                           12
                                                           13
                                                           14
                                                           15
                                                           16
                                                           1 7
                                                           18
                                                           19
                                                                   HEAD  (FT)
                                                   0.0
                                                  92.0
                                                  92.C
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92.0
                                                  92. 0
                                                  92.0
                                                  92.0
                                                                                 C3MC.MG/L)
 3.0
 0.0
 0.2
 1.2
 2.9
15.5
33.0
53.1
64.6
72.9
79.8
35.4
89. 4
92.2
94.3
95.8
97.0
97.8
98.4
98.7
                                                                                                TIME  (TEARS)
                                                                                           09
                                                                                           O
                                                                                           r
                                                                                           H
                                                                                           W
                                                                                                               2!
                                                                                                               09
                                                                                                               •B
                                                                                                               O
0.00
0.13
0.26
0.39
0.53
0.66
0.79
0.92
1 .05
1 .18
1.32
1.45
1 .58
1.71
1.84
1.97
2.11
2.24
2.37
2.50
H
5g
O
§
e
0
^
H
W
»






                                                                                                               00
                                                                                                               co

-------
                                 Selected output for test problem 3—Continued
         OBS.UELL  NO.

              2
o
o
*
2
a
I
I
«
i
 o
 i
 2
 3
 4
 5
 6
 7
 B
 9
10
11
12
13
14
15
16
1?
18
19
        HEAD  (FT)
 D.O
79.8
79.8
79.8
79.8
79.8
79.8
79.8
79.8
79.8
79.8
79.6
79.8
79.8
79.8
79.8
79.8
79.8
79.8
79.8
                     COVC.(
H
W
JO
»
K
co
O
G
»
O
w
CO
NM
2!
M
co
H
i-^
s
>
Hj
s
s:
CO

-------
    SECTION 3



PROGRAM UPDATES

-------
                            NOTE  ON  COMPUTER PROGRAM UPDATE
                                                              January 29, 1988

Reference t  'Computer model  of two-dimensional solute  transport  and dispersion in
ground water."  by L. F. Konikow and  J. D.  Bredehoeft  (1978); U.S.  Geological Survey
Techniques of Water-Resources Inv., Book 7, Chapter C2.

     This update  includes Modifications to iiprove operational  aspects of the
code.  Two additional input  data  checks are performed to verify zero thickness
on  the  grid  boundaries  and  consistency  between   input  thicknesses  and
transnissivlties.   The  location  of the cell with the  Maxima  velocity will be
printed on output if CELDIS is the liilting tiie step size  criterion. A minor
error  in  MOVE related  to velocities near  boundaries has been  corrected.  The
INTEGER*2 specification statement has been  changed  to confon to ANSI Standard
Fortran 77.   No changes in  input  foriats are  required.

     The modifications  can be impleaented  by deleting the  following lines of
the codei
         B3160
        SC  85
         C  86
         D  35
        SE  85
         E  86
        SE 671
        SE 681
         E1055R
E1060
E1072R
E1073R
E1074R
E1077R
E1078R
E1081R
E1090
E1100
E1120
E1125
E2451R
E2452R
E2461R
E2470
F  35
F 250
F 260
F 730
F 750
F 960
F 980
F1230
F1250
F1510
F1530
 F2041R
 F2051R
 G1412R
SG1537R
SH  65
 H  66
and by  inserting the following statements in their proper sequential locations
 (as indicated  by the line numbers in coluins 73-80):
       *                REVISED JANUARY 1988
       IF  (THCK(IX.IY).GT.O.O) THEN
       IF (IX.EQ.l.OR.IY.EQ.l.OR.IX.EQ.NX.OR.IY.EQ.NY)
      1     WRITE (6,935) IX,IY
       GO TO 430
       END IF
       IF  (IX.NE.l) THEN
       IF (TMRX(IX-1,IY,1).GT.O.O) WRITE (6,940) IX,IY
       END IF
       IF  (IY.NE.1) THEN
       IF (TMRX(IX,IY-1,2).GT.O.O) WRITE (6,950) IX,IY
       END IF
   935 FORMAT (1H ,5X.54H*** WARNING ***   THCK.NE.0.0 ON BOUNDARY
      IE IX -,I4,6H, IY -.14)
       INTEGER PTID
                                              *     A  68A
                                                   B3161
                                                   B3162
                                                   B3163
                                                   B3164
                                                   B3165
                                                   B3182
                                                   B3183
                                                   B3184
                                                   B3185
                                                   B3186
                                                   B3187
                                           AT NOD  B4195
                                                   B4196
                                                   D  36

-------
    MAXVXI-0
    MAXVXJ-0
    MAXVYI-0
    MAXVYJ-0
    IF (JCK.EO.O) THEN
     IF (ABVX.GT.VMXBD) THEN
       VMXBD-ABVX
       MAXVXI-IX
       MAXVXJ-IY
     END IF
     IF (ABVY.GT.VMYBD) THEN
       VMYBD-ABVY
       MAXVYI-IX
       MAXVYJ-IY
     END IF
    END IF
    WRITE (6,394) VMXBD.VMYBD
115 TDELXB-CELDIS*XDEL/VMXBD
    ITCD-0
    IF (TDELYB.LT.TDELXB) ITCD-1
    TIMV-AMIN1(TDELXB,TDELYB)
    IF (AMAX1(VMXBD,VMYBD).LE.1.0E-10) WRITE(6,570)
    IF (ITCD.GT.O) THEN
       MJ-MAXVYJ-H
       WRITE (6,534) MAXVYI,MAXVYJ,MAXVYI,MJ
     ELSE
       MI-MAXVXI+1
       WRITE (6,535) MAXVXI.MAXVXJ.MI.MAXVXJ
    END IF
394 FORMAT  (1HO,5X,46HMAXIMUM EFFECTIVE SOLUTE VELOCITIES.  X-VEL - ,
   1  1PE9.2.5X.8HY-VEL - .1PE9.2)
410 FORMAT  (1H0.5X.35HMAXIMUM FLUID VELOCITIES.  X-VEL - .1PE9.2.5X,
   1  8HY-VEL -  .1PE9.2)
                ,4X.52HMAX. Y-VEL. IS CONSTRAINT AND OCCURS BETWEEN NOD
                       AND (,I2,1H,,I2,1H))
535 FORMAT  (1H  .4X.52HMAX. X-VEL. IS CONSTRAINT AND OCCURS BETWEEN NOD
   1ES  (,I2,1H.,I2,7H) AND (.I2.1H.,I2,1H))
    INTEGER PTID
    CONST1-TIMV*DXINV/RF
    CONST2-TIMV*DYINV/RF
    IF  (THCK(IXE+1,IVY).NE.O.O) VXNE-VXNW
    IF  (THCK(IVX.IYS-H).NE.O.O) VYSW-VYNW
    IF  (THCK(IVX-l.IVY).NE.O.O) VXNW-VXNE
    IF  (THCK(IXE.IYS-H).NE.O.O) VYSE-VYNE
    IF  (THCK(IVX.IVY-l).NE.O.O) VYNW-VYSW
    IF  (THCK(IXE-H.IYS).NE.O.O) VXSE-VXSW
    IF  (THCK(IXE.IVY-l).NE.O.O) VYNE-VYSE
    IF  (THCK(IVX-l.IYS).NE.O.O) VXSW-VXSE
290 DISTX-XVEL*CONST1
    DISTY-YVEL*CONST2
220 CMSOUT-CMSOUT+RECH(JX,JY)*TVA2*(CNOLD(IX,IY)+C1)
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
                                                                        E
534 FORMAT (1H
   1ES (,I2,1H.,I2,7H)
   286
   287
   288
   289
SE 645
 E 672
   673
   674
   675
   676
   682
   683
   684
   685
   686
SE 687
 E1079R
 E1082R
 Ellll
 E1112
 E1121
 E1126
 E2062
 E2063
 E2064
 E2065
 E2066
 E2067
 E2068
 E2453R
 E2454R
 E2471
 E2472
 E2602
 E2603
 E2604
 E2605
                                                                        F
                                                                        F
                                                                        F
                                                                        F
                                                                        F
                                                                        F
                                                                        F
    36
   251R
   261R
   731
   751
   961
   981
 F1231
 F1251
 F1511
 F1531
 F2040
 F2050
SG1538R

-------
                          NOTE ON COMPUTER PROGRAM  UPDATE
                                                                   May  15,  1987
Reference»   'Computer  model  of  two-dimensional solute  transport  and  dispersion  in
ground  water." by  L.  F.  Konikow  and J. D. Bredehoeft (1978); U.S. Geological  Survey
Techniques of Water-Resources 7/iv.,  Book 7, Chapter C2.

      The  following   modifications  were  made  to  improve  model   efficiency,
especially  for  one-dimensional  problems.   The   way  in  which  particles  are
removed and replaced has  been changed so that the total  number of  particles  is
minimized.   The  program has  also been modified to use  only one row of particles
for  one-dimensional  problems.  For example,  if the user specifies 9 particles
(NPTPND) for a one-dimensional  problem  (NX  or NY-3), then the program will use
3  particles  aligned  in the  direction  of flow.   The  final change  is  that the
computation of the exponential decay term has been moved outside  of the DO  loops
in MOVE and CNCON.  No changes in  input formats are required.

      This  update  also notes that all  REAL*8  statements should  be changed to
the  ANSI Standard DOUBLE  PRECISION statement.  These changes are not included  in
the   listing below,  and   are only  required  for  compilers that  do  not  accept
the  REAL*8 statement.

      The modifications can be  Implemented  by deleting the  following  lines  of
the  code:
          D 220
          D 460 -  D 480
          D1050
          D2060
          F 222
          F 267R
          F2415R
          F2580 -  F2620
 F2640
 F2680
 F2690
SF2867R
SF3477
 F3770
 F3780
 G 343R
 G 345R
 G 431R
 G 472R
   476R
   489R
 G1444R
SG1694E
SG1694F
G
G
SG1695C
SG1695D
SG1696E
SG1696F
SG1697C
SG1697D
and  by inserting  the following statements  in their proper sequential  locations
(as  indicated by  the line numbers in columns 73-80)«
              •            REVISED MAY 1987 BY D.J. COODE              *    A  67
              IF((NX.E0.3.0R.NY.E0.3).AND.NPTPND.NE.l) WRITE(6,883)             B 614A
           883 FORMAT (1H0.5X,56H*«« ONE-DIMENSIONAL *•• WILL USE ONLY 1 ROU OF P  B4013A
             1ARTICLES/13X.35HUSE 2 PARTICLES FOR NPTPND - 4 OR 5/13X.35HUSE 3 P  B4013B
             2ARTICLES FOR NPTPND - 8 OR 9/13X.31HUSE 4 PARTICLES FOR NPTPND - 1  B4013C
             36)                                                      B4013D
              IONED-0                                                  D 172
              IF(NX.EQ.3.0R.NY.E0.3) IONED-1                               D 173
              IF (NPTPND.EO.5.AND.IONED.EQ.1) Fl-0.25                        D 202
              IF (NPTPND.EQ.8.AND.IONED.NE.1) F2-0.25                        D 22^
              IF (NPTPND.EQ.8.AND.IONED.E0.1) Fl-1.0/3.0                      D 222
              J.F(IONED.EQ.1.AND.T.T.EQ.1.AND.IS.EQ.2) CO TO 140                 D 782
              IF(IONED.E0.1.AND.IT.E0.2.AND.IS.EQ.l) CO TO 140                 D 783
              IF(IONED.EQ.1.AND.IS.EQ.2) PTIDdNDM                         D 834A
              IF(IONED.EQ.1.AND.ITT.EQ.1.AND.ISS.E0.2) CO TO 138               D1019A
                                            -1-

-------
      IF(IONED.E0.1.AND.ITT.EQ.2.AND.ISS.E0.1)  GO TO 138
      IF(IONED.EQ.l) THEN
       IF(rr.E0.1.AND.ISS.E0.2) PTIDdNDM
       IFdT.EQ.2) THEM
        IF(ISS.EO.l) PTID(IND)-13
        IFdSS.EQ.2) PTIO(INDM6
       END IF
      END IF
      IF ((NPTPND.E0.5.AND.IONED.NE.D.OR.NPTPND.EQ.9) GO TO 150
      IF (NPTPND.EQ.8.ANO.IONED.EQ.1) GO TO ISO
      IFdONED.EO.i) GO TO 290
      KP-IND-1
      DOUBLE PRECISION DCYFa.DCYT.DCfT2
      IONED-0
      IF(NX.E0.3.0R.NY.E0.3) IONED-1
      IF (NPTPND.EQ.5.AND.IONED.EQ.1) Fl-0.25
      IF (NPTPND.EQ.8.AND.IONED.NE.1) F2-0.25
      IF (NPTPND.E0.8.AND.IONED.E0.1) F1-F2
      DCYT-l.DO
      DCYT2-1.DO
      IF(DECAY.NE.O.O) THEN
       DCYT-DEXP(-DaFa)
       DGfT2-DEXP (-DCYFa*0. 5DO)
      END IF
      PART(3.IN)-PART(3.IN)«DCYT
  360 CONTINUE
C           	GENERATE NEW TEMPORARY PARTICLE	
  398 SUMC(JX.JY)-SUMC(JX.JY)+CONC(JX.JY)»DCYT2
      PART(3.IP)-CONC(JX.JY)«DCYT
c     *••**••*•**••*••***•*•**•*••*•••*•*•*********•*•*•**•***********
         	INSERT TEMPORARY PARTICLES INTO LIMBO LOCATIONS	
      IF(NPTM.EQ.NP) GO TO 620
      IN-NPTM
      00 595 IL-1,500
      IP-LIMBO (ID
      IF(IP.EQ.O) GO TO 595
      PART(l.IP)-PARTd.IN)
      PART(1.IN)-0.0
      PART(2.IP)-PART(2.IN)
      PART(2.IN)-0.0
      PART(3,IP)-PART(3.IN)
      PART(3.IN}-0.0
      PTID(IP)-PTIDdN)
      PTID(IN)-0
      LIMBO dL>*0
      IN-IN-1
      IF(IN.LE.NP) GO TO 596
  595 CONTINUE
  596 NPTM-IN
      GO TO 620
      DOUBLE PRECISION DCYFCT,DCYT.DCYT2
      DCYT-1.00
      DCYT2-1.DO
      IF(DECAY.NE.O.O) THEN
       DCYT-OEXP(-DCYFCT)
       DCYT2-DEXP (-OCYFCT'O.500)
      END IF
      GO TO 70
      CUCN-CNRECH (IX. IY) «DCYT2
      CNREC-CNRECH(iX.IY)»DCYT2
      CNREC2-CNRECH(IX.IY )*DCYT2
      IF (NPCELL(IX.IY).LE.O) C1-CNOLD(IX,IY)«DCYT2
      DELOa-CNOLD (IX, IY )-CNOLD (IX, IY )«DCYT
      IF (FLU.GT.0.0) FLMIN-FLMIN+FLU*YT*CNOLD(1.JY)«DCYT
      IF (FLU.LT.0.0) FLMOT-FLMOT+FLU«YT*CNOLD(1,JY)«DCYT
      IF (FLU.GT.0.0) FLMIN-FLMIN+FLU*YT«CNOLD(NHX.JY)*DCYT
  272 IF (FLU.LT.0.0) FLMOT-FLMOT+FLU*YT*CNOLD(NMX.JY)*DCYT
      IF (FLU.GT.0.0) FLMIN-FLMIN+FLU«XT*CNOLD(JX,1)»DCYT
      IF (FLU.LT.0.0) FLMOT-FLMOT+FLU«XT*CNOLD(JX.1)*DCYT
      IF (FLU.GT.0.0) FLMIN-FLMIN-t-FLU*XT*CNOLD(JX.NMY)*DCYT
  274 IF (aU.LT.0.0) FLMOT-FLMOT+FLU*XT*CNOLD(JX.NMY)*OCYT
D1019B
D1034A
D1034B
D1034C
D1034D
D1034E
D1034F
D1034G
D1051
D1052
D1132
D2061
F
F
F
F
F
F
F
F
F
F
    33
   211A
 F 211B
 F 217
   222A
   222B
   268A
   268B
   268C
   268D
   268E
   268F
 F2416R
 F2581
 F2641
SF2868R
SF3478
 F3762
 F3763
 F3764
 F3765
 F3766
 F3767
 F3768
 F3769
 F3771
 F3772
 F3773
 F3774
 F3775
 F3776
 F3777
 F3778
 F3779
 F3781
 F3782
 F3783
 F3784
 F3785
 G  42
   346A
   346B
   346C
   346D
 G 346E
 G 346F
 G 347R
 G 431A
 G 472A
 G 476A
 G 489A
 G1444A
SG1694G
SG1694H
SG1695E
SG1695F
SG1696G
SG1696H
SG1697E
SG1697F
G
G
G
G
                                      -2-

-------
                         NOTE ON COMPUTER PROGRAM  OPTION

                                                                 MARCH 5,  1987
Reference»  "Computer model of two-dimensional solute transport and dispersion in ground
water," by L. F.  Konikow and J. D. Bredehoeft (1978); U,S.  Geological Survey Techniques
of Water-Resources Inv., Book  7, Chapter C2.

     The following modifications will  convert  the numerical  solution technique
for the flow equation from the  iterative  Alternating  Direction Implicit Procedure
in the original code to a  Strongly  Implicit Procedure.   In  those few situations
when  iterative ADI has difficulty converging,  SIP  is  usually successful and
relatively efficient.

     To use the model  with the  SIP  routine, the user  should specify NITP - 10
in input card  2. No  other  changes  in the  input formats  are  needed.

     To convert the  program,  the following  lines of code should be deleted?

           B 185                  B3960-B3971           H 126
           B 615                  B4015
           B2910-B2981           C 10-C1370

and the following  lines Inserted in their proper sequential  locations:


   C     *        S.I.P. ROUTINE ADAPTED  BY RICHARD HEALY — 1979       *    A  53
   C     *         S.I.P.  - REVISED SEPTEMBER  1982  &  FEB.  1987        *    A  54
         REAL  *8HMAX                                                         B  63
         COMMON /BALM/ TOTLQ.TOTLQI.TPIN.TPOUT.HMAX                          B 192
         HMAX-1.0                                                             B 453
         IF (NITP.NE.10) WRITE  (6,886)                                       B 616
        1RATION PARAMETERS) - .I4/13X,39HTOL     (CONVERGENCE CRITERIA - SIP  B3961
        2)  -  .E9.2/13X.39HITMAX   (MAX.NO.OF ITERATIONS - SIP)  -,I4/13X,3  B3972
     886 FORMAT (1H0.5X,38H***  WARNING ***  NITP SHOULD EQUAL 10 )            B4016
         SUBROUTINE  ITERAT                                                   C  10
         DOUBLE PRECISION  DMIN1,DEXP,DLOG,DABS                               C  20
   C     	C  30
   C     SOLUTION BY THE STRONGLY IMPLICIT  PROCEDURE                          C  40
         C_	,_...._	_____.._	^-r-^_	    f  Kfl
                                                             — ——™»»    ^  ^y
   C                                                                          C  60
         REAL  *8TMRX,VPRM.HI,HR.HC.HK.WT.REC.RECH.TIM.AOPT.TITLE             C  70
         REAL  *8DEL,ETA,V,XI,RHOP,TEMP,TEST3                                  C  80
         REAL  *8XDEL,YDEL,S,AREA,SUMT.RHO.PARAM,TEST.TOL,PINT,HMIN.PYR       C  90
         REAL  *8B,G,W,A,C,E,F,DR,DC.TBAR,TMK,COEF,BLH,BRK,CHK,QL,BRH,HMAX    C 100
         REAL  *8DXINV,DYINV,ARINV.PORINV                                      C 110
         COMMON /PRMJ/ NTIM,NPMP,NPNT.NITP.N,NX,NY,NP,NREC,INT,NNX.NNY,NUMO  C 120
        IBS,NMOV,IMOV,NPMAX,ITMAX,NZCRIT,IPRNT,NPTPND,NPNTMV,NPNTVL,NPNTD,N  C 130
        2PNCHV.NPDELC.ICHK                                                   C 140
         COMMON /PRMC/ NODEID(40,40),NPCELL(20,20),NPOLD(20,20),LIMBO(500),  C 150
                                       -1-

-------
c
c
c
    1IXOBS(5),IYOBS(5)
     COMMON /HEDA/ THCK(40,40).PERM(40,40),TMUL(5,50),TMOBS(50).ANFCTR
     COMMON /HEDB/ TMRX(40,40,2),VPRM(40,40),HI(40,40),HR(40,40),HC(40,
    140),HK(40,40),UT(40,40),REC(40,40),RECH(40.40),TIM(100),AOPT(20),T
    2ITLE(10),XDEL,YDEL,S.AREA.SUMT.RHO.PARAM,TEST.TOL.PINT.HMIN.PYR
     COMMON /BALM/ TOTLQ.TOTLQI.TPIN.TPOUT.HMAX
     COMMON /XINV/ DXINV.DYINV.ARINV.PORINV
     DIMENSION DEL(40,40),ETA(40,40),V(40,40),XI(40,40),
    1IORDER(21),RHOP(40),TEMP(40),TEST3(201)
     DATA IORDER/1,2,3,4,5,1,2.3,4,5,11*!/

      COMPUTE AND PRINT ITERATION PARAMETERS

     PQIN-0.0
     PQOUT-0.0
     KOUNT—1
     DO 1000 I-l.NX
     DO 1000 J-l.NY
1000 HRd.J)-HK(I.J)
     IF(INT.NE.l) GO TO 40

      	INITIALIZE ORDER OF ITERATION PARAMETERS

     IN01-NX-1
     JN01-NY-1
     I2-IN01-1
     J2-JN01-1
     L2-NITP/2
     PL2-L2-1
                                                                          C
                                                                          C
                                                                          C
  160
  170
  180
                                                                          C 190
C
C
c
c
c
c
c
c
c
      COMPUTE MAXIMUM PARAMETER FOR PROBLEM

     DX-(1./NX)**2
     DY-(1./NY)**2
     W-1-AMIN1(2*DX/(1+ANFCTR*DX/DY),2*DY/(1+DY/(ANFCTR*DX)))

      	 COMPUTE PARAMETERS IN GEOMETRIC SEQUENCE 	

     PJ—1.
     DO 20 I-1.L2
     PJ-PJ+1
 20  TEMP(I)-1.-(1.-W)**(PJ/PL2)

      	 ORDER SEQUENCE OF PARAMETERS 	

     DO 30 J-l.NITP
 30  RHOP(J)-TEMP(IORDER(J))
     WRITE(6,1002) HMAX,NITP,(RHOP(J),J-1,NITP)
1002 FORMAT(IX,6HBETA- ,F4.2,/,1X,I3,23H ITERATION  PARAMETERS.,6(/1X,
    16E15.6))

      INITIALIZE DATA FOR A NEW ITERATION
C
C
C
C
C
C
C
C
C
C
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
200
210
220
230
240
250
260
270
280
290
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
670
680
                                    -2-

-------





]



c
c
c













c
c
c

c
c
c
c
c
c
c


c
c

c
c
c






c
c
c
40 KOUNT-KOUNT+1
IF(KOUNT.LE.ITMAX) GO TO 50
WRITE (6, 160)
CALL OUTPT
WRITE (6 ,1003) (TEST3(I),I-1,KOUNT)
.003 FORMAT ( IX, 39HMAXIMUM HEAD CHANGE FOR EACH ITERATION:,
+20(/,1X,10(F12.5)))
STOP
50 IF(MOD(KOUNT,NITP))60,60,70

INITIALIZE DATA FOR A NEW ITERATION

60 NTH-0
70 NTH-NTH+1
W-RHOP(NTH)
TEST3(KOUNT+1)-0
TEST-0
DO 80 I-l.NX
DO 80 J-l.NY
DEL(I,J)-0
ETA(I,J)-0
V(I.JM)
80 XI(I,J)-0
BIGI-0
RHO-S/TIH(N)

CHOOSE SIP NORMAL OR REVERSE ALGORITHM

IF(MOD(KOUNT,2)) 100,230,100


	 ORDER EQUATIONS WITH ROW 1 FIRST- 3X3 EXAMPLE i
123
456
789

100 DO 210 J-2.JN01
DO 210 I-2.IN01

	 SKIP COMPUTATIONS IF NODE IS OUTSIDE AQUIFER BOUNDARY 	
IF(THCK(I,J).EQ.O.) GO TO 210

	 COMPUTE COEFFICIENTS 	

D-TMRX(I-1,J,1)/XDEL
F-TMRX(I,J,1)/XDEL
B-TMRX(I.J-1.2)/YDEL
H-TMRX(I.J,2)/YDEL
CH-DEL(I,J-1)*B/(1.+W*DELU.J-1))
GH-ETA (1-1 , J )*D/ (1 . +W*ETA (1-1 , J ) )

	 SIP 'NORMAL1 ALGORITHM 	
	 FOWARD SUBSTITUTE, COMPUTING INTERMEDIATE VECTOR V 	
C 690
C 700
C 710
C 720
C 730
C 740
C 750
• C 760
C 770
C 780
C 790
C 800
C 810
C 820
C 830
C 840
C 850
C 860
C 870
C 880
C 890
C 900
C 910
C 920
C 930
C 940
C 950
C 960
C 970
C 980
C 990
C1000
C1010
C1020
C1030
C1040
C1050
C1060
C1070
C1080
C1090
C1100
C1110
C1120
C1130
C1140
C1150
C1160
C1170
C1180
C1190
C1200
C1210
-3-

-------
C                                                                         C1220
      E~ B-D-F-H-RHO-VPRMU.J)                                            C1230
      BH-B-U*CH                                                           C1240
      DH-D-H*GH                                                           C1250
      EH-E+WMCH+GH)                                                      C1260
      FH-F-W*CH                                                           C1270
      HH-H-W*GH                                                           C1280
      ALFA-BH                                                             C1290
      BETA-DH    .                                                         C1300
      GAMA-EH-ALFA*ETA(I,J-l)-BETA*DEL(1-1,J)                             C1310
      DEL(I,J)-FH/GAMA                                                    C1320
      ETA(I,J)-HH/GAMA                                                    C1330
      QL~VPRM(I,J)*WT(I,J)                                               C1340
      RES—D*HK (1-1. J )-F*HK (1+1, J )-H*HK (I, J-H )-B*HK (I, J-l)-               Cl 350
     1E*HK(I,J)-RHO*HR(I,J)+QL+RECH(I.J)+REC(I,J)*ARINV                   C1360
      V(I,J)-(HMAX*RES-ALFA*V(I,J-l)-BETA*V(1-1,J))/GAMA                  Cl370
  210 CONTINUE                                                            C1380
C                                                                         C1390
C      	BACK SUBSTITUTE FOR VECTOR XI 	                               C1400
C                                                                         C1410
      DO 220 J-1.J2                                                       C1420
      J3-NY-J                                                             C1430
      DO 220 1-1,12                                                       C1440
      I3-NX-I                                                             C1450
      IF(THCK(I3,J3).EQ.O.) GO TO 220                                     C1460
      XI(I3,J3)-V(I3fJ3)-DEL(I3,J3)*XI(I3-H,J3)-                          C1470
     1ETA(I3.J3)*XI(I3,J3+1)                                              C1480
C                                                                         C1490
C      	 COMPARE MAGNITUDE OF CHANGE WITH CLOSURE CRITERION	          C1500
C                                                                         C1510
      TCHK-DABS(XI(I3,J3))                                                C1520
      IF(TCHK.GT.BIGI)BIGI-TCHK                                           C1530
      HK(I3,J3)-HK(I3,J3)+XI(I3,J3)                                       C1540
  220 CONTINUE                                                            C1550
  221 IF(BIGI.GT.TOL)TEST-1                                               C1560
      TEST3(KOUNT+1)-BIGI                                                 Cl570
      IFCTEST.EQ.l.) GO TO 40                                             C1580
      DO 130 IY-1.NY                                                      C1590
      DO 130 IX-l.NX                                                      C1600
C     	CUMULATE PUMPAGE AND RECHARGE FOR MASS BALANCE	                C1610
      IF  (REC(IX.IY).GT.O.O) GO TO 32                                     C1620
      PQIN-PQIN+REC(IX,IY)                                                C1630
      GO TO 34                                                            C1640
   32 PQOUT-PQOUT+REC(IX,IY)                                              C1650
   34 IF  (RECH(IX.IY).GT.O.O) GO TO 36                                    C1660
      PQIN-PQIN+RECH(IX,IY)*AREA                                          C1670
      GO TO 38                                                            C1680
   36 PQOUT-PQOUT+RECH(IX,IY)*AREA                                        C1690
C     	COMPUTE LEAKAGE FOR MASS BALANCE	                              C1700
   38 IF  (VPRM(IX.IY).EQ.O.O) GO TO 130                                   C1710
      DELQ-VPRM(IX,IY)*AREA*(WT(IX,IY)-HK(IX,IY))                         C1720
      IF  (DELO.GT.0.0) GO TO 125                                          C1730
      TOTLQ-TOTLQ+DELQ*TIM(N)                                             C1740
                                    -4-

-------

-125
130






C
c
C
C
C
C
c
230


C
C

C
C




c
c
c
c

















340
1030
C
C
GO TO 130
TOTLQI-TOTLQI+DELQ*TIM (N )
CONTINUE
TPIN-PQIN*TIM(N)+TPIN
TPOUT-PQOUT*TIM (N )+TPOUT
WRITE (6 ,140) N
KOUNT-KOUNT+1
WRITE(6,150) KOUNT, (TEST3(I),I-1,KOUNT)
RETURN


	 ORDER EQUATIONS WITH THE LAST ROW FIRST- 3X3 EXAMPLE.
789
456
123

DO 340 JJ-1.J2
J-NY-JJ
DO 340 I-2.IN01

	 SKIP COMPUTATIONS IF NODE IS OUTSIDE OF AQUIFER BOUNDARY 	
IF (THCK(I.J).EQ.O.) GO TO 340

	 COMPUTE COEFFICIENTS 	
D-TMRX(I-1,J,1)/XDEL
F-TMRX(I,J,1)/XDEL
B-TMRXU,J-1,2)/YDEL
H-TMRX(I,J,2)/YDEL

	 SIP 'REVERSE1 ALGORITHM 	
	 FOUARD SUBSTITUTE, COMPUTING INTERMEDIATE VECTOR V 	

E—B-D-F-H-RHO-VPRM (I , J )
CH-DEL(I.J+1)*H/(1.+V*DEL(I,J+1))
GH-ETA(I-1,J)*D/(1.+U*ETA(I-1,J))
BH-H-W-CH
DH-D-W*GH
EH-E+WMCH+GH)
FH-F-U*CH
HH-B-U*GH
ALFA-BH
BETA-DH
GAMA-EH-ALFA*ETA (I . J+l )-BETA*DEL (1-1 . J )
DEL(I,J)-FH/GAMA
ETA(I,J)-HH/GAMA
QL— VPRM(I,J)*WT(I,J)
RES— D*HK (1-1 , J )-F*HK (1+1 , J )-H*HK (I , J+l )-B*HK (I , J-l )-
1E*HK (I , J )-RHO*HR (I , J HQL+RECH (I , J )+REC (I , J ) *ARINV
V (I , J )- (HMAX*RES-ALFA*V (I , J+l )-BETA*V (1-1 , J ) )/GAMA
CONTINUE
FORMAT(1X,8F13.2)

	 BACK SUBSTITUTE FOR VECTOR XI 	
C1750
C1760
C1770
C1780
C1790
C1800
C1810
C1820
C1830
C1840
C1850
C1860
C1870
C1880
C1890
C1900
C1910
C1920
C1930
C1940
C1950
C1960
C1970
C1980
C1990
C2000
C2010
C2020
C2030
C2040
C2050
C2060
C2070
C2080
C2090
C2100
C2110
C2120
C2130
C2140
C2150
C2160
C2170
C2180
C2190
C2200
C2210
C2220
C2230
C2240
C2250
C2260
C2270
-5-

-------
      DO 350 J-2.JN01                                                     C2280
      DO 350 13-1,12                                                      C2290
      I-NX-I3                                                             C2300
      IF(THCK(I.J).EQ.O.) GO TO 350                                       C2310
      XIU,J)-V(I,J)-DEL(I.J)*XI(I-H,J)-ETA(I,J)*XI(I,J-1)                C2320
C                                                                         C2330
C      	COMPARE MAGNITUDE OF CHANGE WITH CLOSURE CRITERION	         C2340
C                                                                         C2350
      TCHK-DABS(XHI.J))                                                  C2360
      IF(TCHK.GT.BIGI) BIGI-TCHK                                          C2370
      HK(I,J)-HK(I,J)+XI(X.i)                                             C2380
  350 CONTINUE                                                            C2390
      GO TO 221                   "                                        C2400
  140 FORMAT(1HO//3X.4HN - ,114)                                          C2410
  150 FORMAT(IX,22HNUMBER OF ITERATIONS- ,I3,/,1X,                         C2420
     139HMAXIMUM HEAD CHANGE FOR EACH ITERATION.,20(/,IX,10F12.5))        C2430
  160 FORMAT(1H0.5X, 53H*** EXECUTION TERMINATED — MAX # ITERATIONS EXC  C2440
     1EEDED/26X.21HFINAL OUTPUT FOLLOWS:)                                 C2450
C     RETURN                                                              C2460
      END                                                                 C2470-
      REAL *8HMAX                                                         H  34
      COMMON /BALM/ TOTLQ.TOTLQI.TPIN.TPOUT.HMAX                          H 127
                                    -6-

-------
                          NOTE ON COMPUTER PROGRAM UPDATE

                                                           MARCH 2. 1987

Reference i  'Computer model of two-dimensional solute transport and dispersion in
ground water," by L. F. Konikow and J. D. Bredehoeft (1978); U.S. Geological Survey
Techniques of Water-Resources Inv., Book 7, Chapter C2.

     The following minor  modifications were iade to perform all flow mass-
balance calculations in double precision (real*8),  to place two time-step
parameters  in a new  common block,  and to Improve the output formats.  These
changes do  not affect the accuracy of the solution, but may result in more
accurate calculation of the flow mass balance for some problems.  The time-
step parameter change is  required only for some FORTRAN compilers, and then
is needed only for transient flow problems with multiple pumping periods.  No
changes in  input  formats  are required.

     The modifications can be Implemented by deleting the following lines of
the code»

A 420     B 860      B2980     B3970      D 440     F3570     11030     11040

and by inserting  the following statements in their proper sequential locations
(as indicated by  the line numbers in  columns 73-80)>
        *                REVISED MARCH 1987 BY D.J. GOODE               *    A  64
        REMN-1.0                                                             A 338
        IF  (NPNT.GT.O) REMN-MOD(N.NPNT)                     .                A 421
        REAL *8TOTLQ,TOTLQI,TPIN,TPOUT                                      B  62
        COMMON /HEDD/ TINIT.TIMX                                            B 157
        WRITE  (6,490)  (TIM(K),K-l,NTIM)                                      B 861
        WRITE  (6,460)  (AOPT(IP),IP-1,NITP)                                  B2981
       2P) - .E9.2/13X.39HITMAX  (MAX.NO.OF ITERATIONS - ADIP) - ,I4/13X,3  B3971
        REAL *8TOTLd,TOTLOi,TPIN,TPOUT,PQIN,PQOUT,DELQ                      C  52
     40 LIMBO(IN )-0       '                                                   D 441
    540 IFLAG-1          '                                                    F3571
        REAL *8TOTLQ,TOTLQI,TPIN,TPOUT,PQIN.PQOUT.DELQ,OOUT.QIN,QNET,DDRW   H  32
        REAL *8QSTR,TPUM,PUMP,TOTLQN.SRCS,SINKS,ERRMB.DENOM.PCTERR          H  33
    110 NTO-N                                                                11031
        IF  (N.GT.50) NTO-MOD(N,50)                                          11041

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                    OCTOBER 20, 1986
  Reference:  "Computer model of two-dimensional solute transport and in ground water,"
by L. F. Konikow and J. D. Bredehoeft (1978).  U.S. Ceol. Survey Techniques of
Water-Resources  Inv., Book 7, Chapter C2.
     Following are  several modifications that  will  prevent zero-divide  checks
that  nay   otherwise  occur  In  certain   cases.   The  nodifications  can  be
Implemented  by  deleting the  following lines of the code:

                             E 412      E 491

and by inserting the following  statements in their proper sequential  locations
(as indicated by the line  numbers in columns 73-80)t

    PERHX-0.0                                                             E  414
    PERMD-PERN(IX.IY)+PERM(IX-1.IY)                                      E  415
    IF (PERMD.GT.0.0) PERMX-2.0*PERM(IX.IY)*PERM(IX-1.IY)/PERMD          E  416
    PERMY-0.0                                                             E  494
    PERMD-PERM2(IX.IY)+PERM2(IX.IY-1)                                    E  495
    IF (PERMD.GT.0.0) PERMY-2.*PERM2(IX,IY)*PERM2(IX.IY-1)/PERMD         E  496

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                        JULY  2.  1986

Refer-encei   'Computer model  of two-dimensional solute  transport  and dispersion  in
ground water." by  L.  F.  Konikow and J. D.  Bredehoeft  (1978).   U.S.   Ceol. Survey
Techniques of Water-Resources Inv., Book 7, Chapter C2.

     The  following modifications were made  primarily to yield  more  efficient
and  more accurate  calculations of particle velocities,  particularly in the
vicinity of permeability  contrasts.   No changes  in  input  formats are  required.
The  modifications can be  implemented by deleting  the  following lines of the
code>
            SB 210         E  451          E 531        E 630
             E 190         E  460          E 540        E 640
             E 431         E  511          E 600        SF 170
             E 441         E  521          E 610        SG 190

and by inserting  the following statements in their  proper sequential
locations  (as Indicated by the-line numbers  in columns  73-80):

C     *               REVISED JULY 1986                               *
      COMMON /CHMC/  SUMC(20,20),VXBDY(40,40),VYBDY(40,40),PERM2(40.40)
      DK-0.0
      PERM2(IX,IY)-0.0
C     COMPUTE HARMONIC MEAN PERMEABILITY
      DO  445 IX-2.NNX
      DO  445 IY-2.NNY
      IF  (THCK(IX.IY).EQ.O.O) GO TO 445
      PNODE-PERM(IX.IY)
      PERM(IX,IY)-2.0*PNODE*PERM(IX+1,IY)/(PNODE+PERM(IX-H,IY))
      PERM2(IX,IY)-2.0*PNODE*PERM(IX,IY+1)/(PNODE+PERM(IX,IY+1))
  445 CONTINUE
      COMMON /CHMC/  SUMC(20,20),VXBDY(40,40),VYBDY(40,40),PERM2(40,40)
      PERMX-2.0*PERM(IX,IY)*PERM(IX-1,IY)/(PERM(IX.IYHPERM(IX-1,IY))
      IF  (THCK(IX-l.IY).NE.O.O) GO TO 13
      PERMX-PERM(IX.IY)
      DHX-HK(IX,IY)-HK(IX+1,IY)
   13 IF  (THCK(IX-H.IY).NE.O.O) GO TO 14
      PERMX-PERM(IX-l.IY)
      DHX-HK(IX-1,IY)-HK(IX,IY)
   14 IF  (THCK(IX-1,IY).EQ.O.O.AND.THCK(IX+1,IY).EQ.O.O) DHX-0.0
      VX(IX,IY)-PERMX*GRDX*PORINV
      PERMY-2.*PERM2(IX,IY)*PERM2(IX,IY-1)/(PERM2(IX,IY)+PERM2(IX,IY-1))
      IF  (THCKttX,IY-1).NE.0.0) GO TO 15
      PERMY-PERM2(IX,IY)
      DHY-(HK(IX,IY)-HK(IX,IY+D)
   15 IF  (THCKUX,IY+1).NE.0.0) GO TO 16
      PERMY-PERM2(IX,IY-1)
      DHY-(HK(IX,IY-1)-HK(IX,IY))
   16 IF  (THCKdX,IY-1).EQ.0.0.AND.THCKUX,IY+1).EQ.0.0) DHY-0.0
      VY(IX,IY )-PERMY*GRDY*PORINV*ANFCTR
      VXBDY(IX,IY )-PERM(IX,IY)*GRDX*PORINV
      VYBDYUX, IY)-PERM2 (IX, IY)*GRDY*PORINV*ANFCTR
      COMMON /CHMC/  SUMC(20,20),VXBDY(40,40),VYBDY(40,40),PERM2(40,40)
      COMMON /CHMC/  SUMC(20,20),VXBDY(40,40),VYBDY(40,40),PERM2(40,40)
 A  63
SB 211
 B 297R
 B 965
 B3291
 B3292
 B3293
 B3294
 B3295
 B3296
 B3297
 B3298
SE 191
   412
   423
   424
   425
   426
   427
   428
   452
   461
   491
   503
   504
   505
   506
   507
   508
   509
   541
   611
   641
SF 171
SG 191
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E

-------
              United  States Department of the Interior

                               GEOLOGICAL SURVEY
                                 RESTON, VA. 22092

In Reply Refer To:                                    January 15,  1986
WGS-Mail Stop 431
Memorandum

To:  Users of "Computer Model of Two-Dimensional Solute Transport and
     Dispersion in  Ground Water," by L.  F. Konikow and J. D.  Bredehoeft
     (1978)

The attached updates to this model have  been developed during the last two
years;  their verification has just been  completed.  Overall,  they serve to
improve the accuracy, efficiency, and usability of the model.  Revisions can
be implemented by following the instructions in the updates.   Otherwise,
copies  of the updated computer program on tape are available  from:

               U.S. Geological Survey
               WATSTORE Program Office
               437  National Center
               Reston, VA  22092
                                       Leonard F. Konikow
                                       Research Hydrologist
Attachments

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                              August  12,  1985

Reference:  "Computer model of two-dimensional solute transport and dispersion in ground
water" by L. F. Konikow and J. D. Bredehoeft (1978):  U.S. Geological Survey
Techniques of Water-Resources Investigations, Book  7, Chapter C2.

     Following are several general modifications to the subject computer
program that  are  designed primarily to improve the  readability of the output.
These minor changes will not affect the calculations.  They also will allow
the user to specify NPTPND  =  1;  this will reduce significantly  the computer
time required to  solve a given problem, but  may adversely affect the numerical
accuracy of the solution to the solute-transport equation.  Hence,  this option
is recommended only for prelininary or test  runs in which high numerical
accuracy of the solution to the transport equation  is not needed, such  as  when
the objective of  the run is to check  input data,  boundary conditions, or
overall problem formulation.

     The modifications can be implemented by deleting the following lines  of
the code:
               A 450
               A 610
               B 611
B1630
B2330
B2620
B4011
E2410
E2530
E2540
E2550
E2560
E2570
G 770
and by inserting the following statements  in  their proper sequential locations
(as indicated by the line numbers in columns  73-80):
       IF (TDEL.EO.(PYR-TINT))  IPCK=1
       IF (REMN.EQ.O.O.OR.N.EO.NTIM.OR.IPCK.E0.1)  CALL OUTPT
   120 IF (REMN.EO.O.O.OR.N.EO.NTIM.OR.MOO(N/50).EO.O.O».IPCK.EQ.1)
      1 CALL CMMOT
      1.AND.NPTPNO.NE.16.AND.NPTPND.NE.1)  WRITE (6/880)
       IF (NPTPNO.EQ.1) WRITE  (6/882)
       WRITE (6/755)
       WRITE (6/755)
       WRITE (6/755)
       WRITE (6/755)
   160 WRITE (6/840) ( VPRMUX/I Y) / IX»1/NX)
       WRITE (6/755)
       WRITE (6/755)
       WRITE (6/755)
   260 WRITE (6/840) (PERM(IX/IY)/IX=1/NX)
       WRITE (6/755)
   320 WRITE (6/840) (VPRM(IX/IY)/IX = 1/NX)
       WRITE (6/755)
   880 FORMAT (1HO/5X/47H*** WARNING  ***  NPTPND  MUST  =  1/4/5/8/9/ OR 16)
   882 FORMAT (1HO/5X/53H*** CAUTION  ***  USE  OF  NPTPND=1 MAY CAUSE LOSS 0
      1F ACCURACY)
       IF (NPTPNO.E0.1) 60 TO  410
   350 FORMAT (1H /1P10E12.3)
   460 FORMAT (1MO/35X/14HXX COEFFICIENT/)
   470 FORMAT (1MO/35X/14HYY COEFFICIENT/)
   480 FORMAT (1HO/35X/14HXY COEFFICIENT/)
   490 FORMAT (1MO/35X/14HYX COEFFICIENT/)
   500 FORMAT (1M /1P10E11.2)
       CNCNC(IX/IY)=CNCNC(IX/IY)+EQFCT1*(X1+X2+Y1»Y2+XX1-XX2+YY1-VY2)
                                                  336
                                                  406
                                                  451
                                                  611
                                                  612
                                                  612
                                                  614
                                                61195
                                                81347
                                                3U35
                                                B1515
                                                B1631
                                                B1945
                                                B2075
                                                B2315
                                                B2331
                                                B2615
                                                32621
                                                62655
                                                B4011A
                                                B4012A
                                                B4012B
                                                01115
                                                E2411
                                                E2531
                                                E2541
                                                E2551
                                                E2561
                                                E2571
                                                G 771

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                             August 8,  1985

Reference;  "Computer model of two-dimensional solute transport and dispersion in
ground water" by L. F. Konikow  and  J. D. Bredehoeft  (1978):  U.S. Geological  Survey
Techniques of Water-Resources Investigations, Book 7, Chapter C2.

     The following modifications  will allow the  user to  solve the  transport
equation within a smaller  grid than  is used to solve the flow equation.   This
can yield a great savings  in  computer calculation time and  storage
requirements for problems  in  which the hydraulic gradients  within  the  area  of
interest for transport are influenced by hydraulic stresses and  (or) boundary
conditions outside of the  area in which solute transport is occurring.

     The approach is to define a  primary finite-difference  grid  for the flow
model using the standard input data  formats of the computer code.   A smaller,
secondary subgrid for transport is then defined  within the  coordinates of the
primary grid.  The revised input  formats are  structured  so  that  implementation
of the changes will  not necessitate  the modification of  any existing data
sets; that is, identical results  will be obtained with the  modified code as
with the unmodified  program for problems which do not  apply a transport
subgrid.

     To utilize this new option,  the user should specify "NX"  as a negative
value in field 3 of  data card 2.  When the program reads such a  negative
value, it will then  reset  NX = -AW  and read  a new data  set, immediately
following data card  2, which  contains the following four integer values  in
free format (that is, values  are  separated by commas or  spaces):

          MX,MY     The X" and y-coordinates  within  the  primary grid  of the
                     upper-left node  of the transport subgrid, and
          MMX,MMY  The grid  coordinates of the  lower-right node of the
                     transport subgrid.

For example, if the  primary grid  is  20 by 20  (that is, NX = 20 and NY = 20),
and if you want the  subgrid for transport to  be  10 by  10, with the  upper left
node of the subgrid  corresponding to node (3,4)  of the primary grid, then your
new data would specify MX «  3, MY = 4, MMX -  12, and  MMY =  13.

     Note that the "window" for the transport subgrid can overlap  all  or any
part of the primary  grid,  but can not extend  beyond it.  Also note  that  unless
the subgrid overlaps, the  first or last row or column  of the primary grid,
which are no-flow boundaries, then all nodes  of  the subgrid can  be  "active"

-------
nodes.  Finally, the subgrid should be located so that there will not be a
significant amount of convective (advective) transport across the subgrid
boundary, or else the accuracy of the solution will be adversely affected.

     The boundaries of the transport subgrid are assumed to represent a
constant concentration-gradient condition.  Thus, the revised model has been
programmed to allow no dispersive flux across the outer boundary of the
subgrid and to set the concentration of any convective (advective) flux across
the boundary equal to the concentration at the node just inside the boundary,
regardless of whether the flux is into or out of the subgrid at that node.

     The following instructions for implementing this option assume that all
"updates" released or dated prior to the date of this "update note" have
already been incorporated.  The modifications can be implemented by deleting
the following lines in the code:

     A 570          D 650          E1960          F3840          G1330
     B 990          D 800          £1970          F4050          G1340
     B1130          D 810          E2000          G 231          G1350
     B1140          D1022          E2010          G 241          G1421
     B1150          D1023          £2020          G 351          G1431
     B1380          D1075          £2032          G 361          G1440
     B1390          D1090          £2190          G 370          G1443R
     B1400          D1170          £2200          G 381R         G1445R
     B2550          D1180          £2220          G 420          G1490
     B3020          D1380          £2230          G 460          G1499R
     B3030          D1390          £2250          G 474R         G1520
     B3040          £ 480          £2260          G 477R         G1530
     B3080          E 560          £2280          G 480          G1536R
     B3092R         E 670          E2290          G 595          G1560
     B3420          E 680          F2420          G 851          G1585
     B3500          £1350          F2430          G 861          G1600
     B3550          £1360          F2500          G 870          H 140
     B3840          £1510          F2866R         G 900          H1070
     B3850          £1530          F2875          G1010          H1190
     D 511          £1650          F2885          G1020          I 350
     D 521          £1670          F3530          G1061          I 360
     D 530          E1710          F3580          G1071          I 380
     D 600          E1720          F3635          G1080          I 540
     D 610          E1740          F3662          G1160          I 550
     D 620          E1760          F3664          G1280          I 580
     D 630          £1810          F3666          G1300          11090
     D 640          £1820          F3830          G1310

and inserting the following statements in their proper sequential locations,
as indicated by the line numbers in columns 73-80 (for identification
purposes, all inserts for this update are labeled with an "S" in column 74):

-------
c
t
* REV. JULY-DEC. 1985 TO ALLOW SECONDARY SUBGRID FOR TRANSPORT
COMMON /MEDC/ MXxMYxMMX/MMYxNMXxNMYxMCHK
IX=IXOBSCI)-MX+1
If  (IXOBS(I).LT.MX.OR.IXOBS(I).GT.MMX) GO TO 110
IY=IYOBS(I)-MY+1
IF  (IYOBS(I).LT.MY.OR.IYOBS(I).GT,MMY) GO TO 110
TMCN(IxJ)=CONC(lXxIY)
COMMON /MEDC/ MXxMYxMMXxMMYxNMXxNMYxMCHK
MCHK=0
NCA2=0
NMX-NX
NMY=NY
MMX=NX-1
MMY«NY-1
   --- READ UPPER LEFT AXD LOWER RIGHT NOOAL COORDS. OF
      TRANSPORT SUBGRID* IN FREE FORMAT/ IF NX.LT.O ---
IF (NX.GT.O) GO TO 5
NX=-NX
MCHK=1
READ (5**) MXxMYxMMXxMMY
NMX=MMX-MX+1
NMY = MMY-MY*1      ••                  :
CONTINUE
WRITE (6x755)                  '
IF (MCHK.GT.O) WRITE (6x775) NMXxNMYxMXxMYxMMXxMMY
WRITE (6x888)
00 75 !Y«1xNMY
DO 75 IX=1xNMX
CNRECM(lXxIY)«0.0
CONC(IX/IY)=0.0
CONINT(IX,IY)=0.0
SUMCCIX/.IY)=0.0
NPCELL(IX/IY)»0
CONTINUE
JX=IX-MX+1
JY=IY-MY*1
IF (JX.LT.1.0R.JY.LT.1.0R.JX.GT.NMX.OR.JY.GT.NMY) GO TO 105
IF (FCTR.LT.0.0) CNRECM( JXx JY)=CNREC
REC(IX,IY)«FCTR
WRITE (6x820) IX/I YxRECCIX/I Y) xCNREC
IF (MCHK.EO.O) GO TO 250
IF (IX.LT.MX.OR.IX.GT.MMX)  GO TO 250
IF (IY.LT.MY.OR.IY.GT.MMY)  GO TO 250
   75
  105
  110
      IF  (MCHK.GT.O)  NZCRIT*(NCA2+25)/50
      IF  (NZCRIT.EQ.O)  NZCRIT=1
      IF  (MCHK.EQ.O)  GO TO  265
      AA02«NCA2*AREA
      WRITE (6x633)
      WRITE (6x635) NCAxAAQ
      WRITE (6x634)
      WRITE (6x630) NCA2xAAQ2xNZCRIT
      GO  TO 267
  265  CONTINUE
  267  CONTINUE
      JX=IX-HX+1
SA
SA
SA
SA
SA
SA
SA
SA
SB
SB
So
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
SB
    61
    62
   205
   562
   563
   564
   565
   571
   155
   353
   354
   355
   356
   432
   433
   434
   435
   436
   437
   438A
   438B
   438C
   438D
   439A
   4396
   439H
   545
   565
   617
SB1161
SB1162
SB1163
SB1164
SB1165
SB1166
SB1167
SB1168
SB1372
SB1373
SB1374
SB1382
SB1392
SB1402
SB2262
SB2263
S32264
SB2265
SB2301
SB2305
SB2332
SB2333
SB2334
SB2335
SB2336
SB2337
SB2338
SB2339
SB2345
SB2544

-------
      JY=IY-MY+1                                                         SB2545
      IF (JX.LT.1.0R.jY.LT.1.0R.JX.GT.NHX.OR«JY.GT.NMY) 60 TO 285        SB2546
      CNRECH(JX,JY)«FCTR2                                                SB2551
  285 CONTINUE                                                           SB2555
C        	IF USING SMALLER SUBGRIO FOR TRANSPORT/ READ INITIAL         SB3005
C           CONCENTRATION ARRAY FOR SUBGRIO NODES ONLY—-                SB3006
      00 420 IY»1,NMY                                                    SB3021
      JY»IY+MY-1                                                         SB3025
      IF (INPUT.E0.1) READ (5/660) (CONC(IX,IY),IX«1,NMX)                SB3031
      00 410 IX=1,NMX                                                    SB3041
      JX=IX+MX-1                                                         SB3045
  390 IF (THCK(JX/JY).NE.O.O) CONC(IX,IY)*FCTR                           SB3081
      CFCTR»CONINT(IX,IY)*THCK(JX,jr)*AREA                               S33093R
  500 FORMAT (3H   /25F5.1)                                              SB3421
  580 FORMAT (1M ,4013)                                                  SB3501
  630 FORMAT (1HO/05X/05X/44HNO. OF FINITE-DIFFERENCE CELLS IN AQUIFER - SB3551
  633 FORMAT (1HO,////2X,26HFLOW MODEL (PRIMARY GRID):/)                 SB3582
  634 FORMAT (1HO,///2X,18HTRANSPORT SUBGRID:/)                          SB3583
  635 FORMAT (1HO,05X,05X,44MNO. OF FINITE-DIFFERENCE CELLS IN AQUIFER * SB3564
     1 ,I4//10X,28HAREA OF AQUIFER IN MODEL  " ,G12.5,10H   SQ. FT./)    SB3585
  755 FORMAT (1H )                                                       SB3715
  775 FORMAT C1HO,18X,31HSECONOARY SUBGRIO FOR TRANSPORT//16X,30HNMX   ( SB3763
     1NUMBER OF COLUMNS)  *  ,I4/16X,30NNMY   (NUMBER OF ROWS)     =  /I SB3764
     24//16X/38MCROSS-REF. TO PRIMARY GRID      IX  IY/46X/8H 	/18 SB3765
     3X/28HFIRST NODE (UPPER LEFT) AT: /2I4/18X/28HLAST NODE (LOWER RICH SB3766
     4T) AT: ,214)                                                       SB3767
     2 POROSITY),8X,3H»  ,F9.3/13X,39HBETA    (LONGITUDINAL OISPERSIVITV SB3841
     3)  *  ,F7.1/13X,31HDLTRAT  (RATIO OP TRANSVERSE TO/21X,30HLONGITUDI SB3851
  888 FORMAT (1H1)                                                       SB4017
      COMMON /HEDC/ MX,MY,MMX,MMY,NMX,NMY,MCM£                           SO 125
C        	TRACK PARTICLE LOCATIONS IN COORDINATES OF PRIMARY GRID	   SO 505
      00 410 IX«1,NMX                                                    SO 512
      JX»IX*MX-1                                                          SD 515
      DO 410 IY*1,NMY                                                    SO 522
      JY=IY*MY-1                                                          SO 525
      IF (THCK(JX/JY).EQ.O.O) GO TO 410   .                               SO 531
      IF (IX.EQ.1.0R.IX.EQ.NMX.OR.IY.EQ.1.0R.XY.EQ.NMY)  TEST2«1.0        SO 595
      IF (VPRM(JX,JY).GT.0.09)  TEST2-1.0                                 SD 601
      IF (RECUX,JY).NE.O.O)  TEST2-1.0                                   SO 611
      IF (THCK(JX+1sJV+1).EQ.O.O.OR.THCK(JX+1sJY-1>.EQ.O.O.OR.THCK(JX-1, SD 621
     1JY+1).EQ.O.O.OR.THCK(JX-1,JY-1).EQ.O.O)  TEST2=1.0                   SD 631
      IF ((THCK(JX/JY*1).EO.O.O.OR.TMCK(JX,JY-1).EQ.O.O.OR.TMCK(JX+1/JY)  SO 641
     1.EQ.O.O.OR.THCK(JX-1,JY).EQ.O.O).ANO.NPTPND.GT.5)  TEST2=1.0        SO 651
      PART(1,IND)sJX+F1*EVET                                              SO 801
      PART(2,IND)=JY*F1«EVES                                              SD 811
      PART(1,INO)=(JX+F1«EVET)+F2*EVET2                                  SD1022A
      PART(2,INO)*(JY*F1*EVES)*F2»EVES2                                  S01023A
  150 PART(1,IND)«JX                                                     SD1076
      PART(2,INO)»-JY                                                    S01091
      PART(1,INO)*JX+F2*EVET                                              SD1171
      PART(2,IND)«-JY                                                    SD1181
      PARTC1,IND)=JX                                                     SD1381
      PART(2,INO)ejY+F2*EVET                                              SD1391
      COMMON /HEOC/ MX,MY,MMX,MMY,NMX,NMY,MCMK.                           SE 145
      IF (IX.GT.NMX.OR.IV.GT.NMV)  GO  TO 12                                SE 315
   12 JCK«0                                                              SE 343
      IF (IX.LT.(MX-1).OR.IY.LT.(MY-1).OR.IX.GT.(MMX+1).OR.IY.GT.(MMY*1)  SE 344
     1)  JCK=1                                                             SE 345
      IF (A8VX.GT.VMAX.ANO.JCK.EQ.O)  VMAX*ABVX                           SE 481
      IF (ABVY.GT.VMAY.AND.JCK.EQ.O)  VMAY-ABVY                           SE 561

-------
     IF  (A6VX.GT.VMXBO.ANO.JCK.EO.O)  VMXBO«ABVX
     IF  (ABVY.GT.VMYBO.AND.JCK.EQ.O)  VMYBD«ABVY
     IF  (JCK.GT.O)  GO  TO  20
     DO  150  IX=2/MMX
     00  150  IY»2/MMY
     JX»IX-MX*1
     JY»IY-MY+1
     IF  (JX.LT.1.0R.JY.LT.1) GO TO 150
     IF  ((IX-M).GT.MMX) GO TO 140
     DISP(JX/JY/1)=(DALN*VXE2+OTRN*VYE2)/(VMGE2*XX2)
     IF  ((IY-1).LT.MY.OR.(IY+1).GT.MMY) GO TO 140
     DISP(JX/JY/3)=(DALN-DTRN>*VXE*VYE/(VMGE2*XY2)
     IF  ((IV-M).GT.MMY) GO TO 150
     DI$P(JX/JY/2)*(DALN*VY$2+DTRN*VXS2)/(VMGS2*YY2)
     IF  ((IX-1).LT.MX.OR.(IX+1).GT.MMX) GO TO 150
     DISP(JX/JY/4)=(DAUN-DTRN)*VXS*VYS/(VMGS2*XY2)
     00  160  IX«2/MMX
     00  160  IV=2/MMY
     JX=IX-MX+1
     JY=IY-MY+1
     IF  (JX.LT.1.0R.JY.LT.1) GO TO 160
   1>.EQ.O.O.OR.THCK(IX+1,IV-1).EQ.O.O> DISP( JX/ JY/3>=0.0
   1).EQ.O.O.OR.THCK(IX-1/IY*1).EQ.O.O) DISP( JX/ JY/4)«0.0
     00  170  IX=1/NMX
     00  170  IY*1/NMY
     00  190  IX=2/MMX
     DO  190  IY=2/MMY
    IF (JX.LT.1.0R.JY.LT.1) GO TO 190
    DISP(JX,JY,1)«DISP(JX,JY,1>*BAVX
    OISP(JX,JY,2)»DISP(JX*JY/2)*BAVY
    DISP(JX/JY/3)=DISP(JX/JY/3)*BAVX
    00 260 IY=1/NMY
260 WRITE (6x500) (OISPCIXx I Y,1 ) xIX'
    DO 270 IY=1/NMY
270 WRITE (6/500) (DISPUX/I Y/2>/IX=1 ,NMX>
    DO 280 IY=1/NMY
280 WRITE (6/500) (DISP(IX/IY/3)/IXc1 /NMX)
    DO 290 IY=1/NMY
290 WRITE (6/500) (DISP(IX/I Y/4)/IX«1/NMX)
    COMMON /HEOC/ MX/MY/MMX/MMY/NMX/NMY/MCHK
    NLOC'O
    JXsIX-MX+1
    JY-IY-MY+1
    JNX«INX-MX-»-1
    JNY«INY-MY+1
    IF (MCHK.EQ.O) GO TO 342
    IF (JNX.LT.1.0R.JNX.GT.NMX.OR.JNV.LT.1.0R.JNY.GT.NMY) NLOC=1
    IF (NLOC.GT.O) GO TO 345
342 CONTINUE
    SUMC(JNX/JNY)aSUMC(JNX/JNY)-fPART(3/IN)
    NPCELL(JNX/JNY)=NPCELL(JNX/JNY)+1
345 CONTINUE
    IBD=0
    IF (VPRM(IX/IY).LT.0.09) GO TO 348
348 IF (MCHK.EQ.O) GO TO 540
    IF (JX.EQ.1.AND.VXBDY(IX-1/IY).GT.O.O.ANO.JNX.GT.JX) IBD*1
    IF (JX.EQ.NMX.AND.VXBOY(IX/IY).LT.O.O.AND.JNX.LT.JX) IBD=1
 SE  671
 SE  681
 SE  7C5
 SE1351
 SE1361
 SE1375
 SE1376
 SE1377
 SE1405
 SE1511
 SE1525
 SE1531
 SE1555
 SE1651
 SE1665
 SE1671
 SE1711
 SE1721
 SE1725
 SE1726
 SE1727
 SE1741
 SE1761
 SE1811
 SE1821
 SE1961
 SE1971
 SE1973A
 SE1973B
 SE1973C
 SE2001
 SE2011
 SE2021
 SE2033
 SE2191
 SE2201
 SE2221
 SE2231
 SE2251
 SE2261
 SE2281
SE2291
 SF 125
SF 335
SF 415
SF 475
SF2395
SF2396
SF2397
SF2398
SF2405
SF2407
SF2421
SF2431
SF2445
SF2478
SF2501
SF2522
SF2523
SF2524

-------
    IF CJY.EQ.1.ANO.VYBDYCIX,IY-1).GT.O.O.AND.JNY.GT.JY) 180=1         SF2525
    IF (JY.EO.NMY.AND.VYBDY NPOLDUX/JY)=NPOLD(JX/JY)-1                 SF2886
525 DLXsINX-IX                                                         SF3425
    PARTC1,IP)=TEMPX-DtX                                               SF3435
    OLY=INY-IY                                                         SF3455
    PART(2/IP)*TEMPV-Ol.Y                                               SF3465
    PART(2/IP)=-PART(2/IP>                                             SF3475
    PART(3*IP)*CONC(JX,JY>*EXP(-OCYFCT>                                SF3477
    SUMCUX,JY)»SUMC                                 SF3479
    NPCELt(JX/JY)=NPCELl(JXsJY)+1                                      SF3481
    GO TO 540                                                          SF3483
    PART(3*IP)=CONC(JX,JY>                                             SF3531
550 IF (NLOC.GT.O) GO TO 565                                           SF3575
    IF (VPRM(INX/INY).GT.0.09.AND.WT(INX/INY).LT.MK(INX,INY)) GO TO 56 SF3581
560 IF (NPOLD(JNX/JNY).LE.O) GO TO 590                                 SF3636
565 CONTINUE                                                           SF3638
    IF (NLOC.GT.O) GO TO 56?                                           SF3661
    SUMC(JNX,JNY)=SUMC(JNX,JNY)-CONC(JNXxJNY)                          SF3663
    NPCELL(JNXxJNY)aNPCELLUNX,JNY>-1                                  SF3665
    NPOLD(JNX,JNY)*NPOLD                                               SF4051
630 CONTINUE                                                           SF4053
    COMMON /HEDC/ MX/MY*MMX^MMY*NMX/NMY/MCMK                           SG 135
    00 10 IX»1xNMX                                                     SG 232
    DO 10 IV*1/NMY                                                     SG 242
 20 00 60 IX=1/NMX               .                                      SG 352
    00 60 IY=1/NMY                                                     SG 362
    JXsIx*HX-1                                                         SG 366
    JY=IY*MY-1                                                         SG 367
    IF (THCK(JXsJY).EQ.O.O)  GO TO 60                                   SG 371
    EQFCT1eRFFCT/THCK(JX,JY)  .                                         SG 382R
    SLEAKc
-------
   X1=DISP(IX/IY/1)*(CAVG(IX+1/IY)-C2)
   If (ZX.LE.1) GO TO 42
   X2=DISP(IX-1/IY/1)*(CAVG(IX-1/IY)-C2)
   IF (IY.GE.NMY) GO TO 43
   Y1*OISP(IX/IY/2)*(CAVG(IX/IY+1)-C2)
   IF (IY.LE.1) GO TO 46
   Y2=DISP(IX/IY-1/2)*(CAVG(IX/IY-1)-C2)
   IF .GT.0.09) GO TO 90
    00 100 IV*1sNMV
100 WRITE (6/330) (NPCELL ( IX,IY),IX«1 ,NMX)
110 DO 130 IX«1/NMX
    00 130 IY«1/NMY
    JX*IX+MX-1
    JY=IY+MY-1
    IF (THCK(JX/JY).EQ.O.O) GO TO 120
120 IF (CONC(IXsIY).GT.O.O) WRITE (6/310) JX/ JY/CONC
-------
    FLW=VPRM(JX/JY>*(WT(JX/JY)-HK(JX/JY))
          	SUBGRIO BOUNDARIES	
    IF (MCHK.EQ.O) 60 TO 275
    YT=YDEL*TIMV
    XT=XOEL*TIMV
    00 272 IY=MY/MMY
    IX«MX
    JY=IY-MY+1
    FLW=TMRX(IX-1/IY/1)*
    IF (FLU.GT.0.0) FLMIN»FLHIN+FLW*XT*CNOLO(JX,NMY)*EXP(-DCYFCT)
274 IF (FLW.LT.0.0) FLMOT«FLMOT+FLW«XT*CNOLDUX,NMY)*EXP(-OCYFCT)
275 CONTINUE
190 FORMAT (1HO,30I4)
300 FORMAT (3M   ^2515)
    COMMON /HEOC/ MX/MY,MMX/MMYxNMX,NMY/MCHK
    DO 20 IY«1xNMY
    00 10 IX=1,NMX
 20 WRITE (6/240) (IC(IX)*IX*1sNMX)
    00 40 IY = UNMY
    DO 30 IX»1,NMX
 40 WRITE (6/240) (1C(IX)/IX=1,NNX)
    JX=IXOBS(J)
    JY=IYOBS(J)
    C1INT=0.0
    IX»JX-MX*1
    IY=JY-MY*1
    IF (JX.LT.MX.OR.JY.LT.MV.OR.JX.GT.MMX.OR.JY.GT.MMY) GO TO 125
    C1INT*CONINT(IXsIY>
    GO TO 127
125 WRITE (6/435)
127 WRITE (6/440) MOZ/WT(JX,JY)/C1INT/
435 FORMAT (1M /3X/45H** NOTE **  THIS OBS. WELL IS LOCATED OUTSIDE/16
   1X/24HOF THE TRANSPORT SUBGRIO)
 SG1601
 SG1692A
 SG1692B
 SG1692C
 SG1692D
 SG1694A
 SG1694B
 SG1694C
 SG1694D
 SG1694E
 SG1694F
 SG1695A
 SG1695B
 SG1695C
 SG16950
 SG1696A
 SG1696B
 SG1696C
 SG1696D
 SG1696E
 SG1696F
 SG1697A
 SG1697B
 SG1697C
 SG16970
 SG1696A
 SH1071
 SH1191
 SI 125
   351
   361
   381
   541
   551
   581
 SI1074
 SI1076
 SI1082
 SI1083
 SI1084
 SI1085
 SI1086
 SI1087
 SI1088
SI1091
SI1525
 SI1526
SI
SI
SI
SI
SI
SI

-------
     To optimize computer storage  (or memory) requirements while using this
option, it may be desirable to redimension the arrays used by the flow model
to the size  of the primary grid and the arrays used by the transport model to
the size of  the subgrid.  For example,  if it were desired to  use a primary
grid of (40,40) but keep the transport subgrid within the original dimensions
of (20,20),  then the following changes would be required in the  common and
dimension statements of the program:



     COMMON  /PRMC/ NODEID<40,40>,NPCELL(20,20),NPOLD(20,20),LIMBO<500),  SA  145
     COMMON  /MEDA/ THCKUO,40) ,PERM(40,40) ,TMWL(5,50) ,TMOBS ( 50)/ANFCTR   SA  170
     COMMON  /MED6/ TMRX(40,40,2),VPRM(40,40),HI(40,40),HR(4C,40),HC(40,  SA  180
    140),MK(40,40),WTC40,40),REC<40,40>,RECH(40,4C»,TIM(100>,AOPT(20>,   SA  190
     COMMON  /CHMA/ PART(3,6400),CONC(20,20),TMCN(5/50),VX(40,40),VY(40,  SA  211
    140),CONINT(20,20),CNRECH<20,20),POROS,SUMTCH,BETA,TIMV,STORM,STORM  SA  220
     COMMON  /PRMC/ NOOEIO(40/40),NPCELLC20/20)/NPOLD(20/20)/LIMBO(500)/  SB   95
     COMMON  /MEOA/ THU(40/40)/PERM(40/40)/TMWLC5/50)/TMOBS(50)/ANFCTR   SB  120
     COMMON  /HEOB/ TMRX(40/40/21/VPRM(40/40)/HI(40/40)/HR(40/40)/HC(40/  SB  130
    140),HK(40,40),UT(40,40),REC(40,40),RECM<40,40),TIM(100),AOPT(20),   SB  140
     COMMON  /CHMA/ PART(3/6400)/CONC(20/20)/TMCN(5/50)/VX(40/40)/VY(40/  SB  161
    140)/CONINT(20/20),CNRECH(20/20)/POROS/SUNTCH/BETA,TIMV/STORM/STORM  SB  170
     COMMON  /CHMC/ SUMC(20/20)/VXBDY(40/40)/VYBDY(40/40)                 SB  210
     COMMON  /PRMC/ NOOEIO(40/40)/NPCELL(20/20)/NPOLO(20/20)/LIMBO(500)/  SC   85
     COMMON  /HEDA/ THCKUO/40)/PERM(40/40)/TMWL(5/SO)/TMOBS(50)/ANFCTR   SC  110
     COMMON  /HEOB/ TMRX(40/40/2)/VPRM(40/40)/HI(40/40)/HR(40,40)/HC(40/  SC  120
    140)/HK(40/40),HT(40,40)/REC(40/40)/RECH(40/40)/TIM(100),AOPT(20)/   SC  130
     DIMENSION H(40)/  6(40),  G(40)                                       SC  171
     COMMON  /PRMC/ NODEIDUO,40),NPCELL(20,20)/NPOLD(20,20)/LIMBO(500),  SO   65
     COMMON  /HEOA/ THCK(40/40)/PERM(40/40)/TMHL(5/50)/TMOBS(50)/ANFCTR   SO   90
     COMMON  /HEDB/ TMRX(40,40,2),VPRM(40,40),Hl(40,40),HR(40,40),MC(40x  SD  100
    140),HK(40,40),HT(40,40),REC(40/40),RECH(40,40),TIM(100)/AOPT(20)/   SO  110
     COMMON  /CHMA/ PART(3,6400),CONC(20,20),TMCN(5,50),VX(40,40),VY(40,  SO  131
    140>/CONINT(20,20),CN«ECH(20,20),POROS,SUMTCH,BETA,TIMV/STORM/STORM  SO  140
     COMMON  /PRMC/ NOOEIDUO/40),NPCELL(20,20),NPOLD(20,20),LIMBO(500)/  SE   85
     COMMON  /HEOA/ TMCK(40,40),PERM(40,40),TMHL(5/50),TMOBS(50)/ANFCTR   SE  110
     COMMON  /HEOB/ TMRX(40/40/2),VPRM(40,40),HI(40,40),MR(40,40),HC(40,  SE  120
    140),HK(40,40),WT(40,40),REC(40,40),RECH(40,40),TIM(100),AOPT(20),   SE  130
     COMMON  /CHMA/ PART(3,6400),CONC(20,20),TMCN(5,50),VX(40,40),VY(40,  SE  161
    140),CONINT(20,20)/CNRECH(20,20),POROS,SUMTCH,BETA,TIMV,STORM,STORM  SE  170
     COMMON  /CHMC/ SUMC(20,20),VXBDY(40,40),VYBDY(40,40)                 SE  190
     COMMON  /PRMC/ NODEIO(40,40) ,NPCELK20,20) ,NPOLD(20/20)/LIMBO (500) ,  SF   65
     COMMON  /HEOA/ THCK(40,40),PERM(40,40),TMWL<5,50),TMOBS(50),ANFCTR   SF   90
     COMMON  /HEDB/ TMRX(40,40,2),VPRM(40,40),HI(40,40),MR(40,40),HC(40,  SF  100
    140),HK(40,40),WT(40,40),REC(40,40),R£CH(40,40),TIM(100),AOPT(20),   SF  110
     COMMON  /CHMA/ PART(3,6400),CONC(20,20),TMCN(5,50),VX(40,40),VY(40,  SF  141
   140)/CONINT(20/20)/CNRECH(20/20)/POROS/SUMTCH/BETA,TIMV,STORM/STORM  SF  150
     COMMON  /CHMC/ SUMC(20/20)/VXBDY(40,40),VY60Y(40,40)                 SF  170
     COMMON /PRMC/ NODE 10(40,40),NPCELL(20,20),NPOLO(20,20),LIMBO(500),  SG   75
     COMMON  /HEOA/ THCK(40,40),PERM(40,40),TMHL(5,50)/TMOBS(50),ANFCTR   SG  100
     COMMON /HEOB/ TMRX(40,40,2)/VPRM(40,40),HI(40,40),HR(40/40),HC(40,  SG  110
   140)/HK(40/40),WT(40,40),REC(40,40),RECM(40,40),TIM(100)/AOPT(20),   SG  120
    COMMON /CHMA/ PART(3,6400),CONC(20,20),TMCN(5,SO),VX(40,40),VY(40/  SG  151
   140),CONINT(20,20),CNRECM(20,20),POROS,SUMTCH,BETA,TIMV,STORM,STORM  SG  160
     COMMON  /CHMC/ SUMC(20,20)/VXBOY(40/40),VY6DY(40,40)                 SG  190
    COMMON  /PRMC/ NOOEID(40,40),NPCELL(20,20),NPOLD(20,20),LIM80(500),  SH   65
     COMMON  /HEDA/ THUUO,40) ,PERM (40,40) ,TMHL ( 5,50)/TMOBS (50) , ANFCTR   SH   90
     COMMON  /HEDB/ TMRX(40,40,2),VPRM(40,40),HI(40,40),HR(40,40),HC(40,  SH  100
   140),HK(40,40),HT(40,40),REC(40,40),RECH(40,40),TIM(100),AOPT(20),   SH  110
     DIMENSION IM(40)        '                                            SH  141
     COMMON  /PRMC/ NODEIO(40,40),NPCELL(20,20)/NPOLD(20,20),LIMBO(500)/  SI   65
     COMMON /HEDA/ THCK(40,40),PERM(40,40),TM«L(5,50),TMOBS(50),ANFCTR   SI   90
     COMMON  /HEOB/ TMRX(40,40,2)/VPRM(40/40)/HI(40,40>,MR(40,40),HC(40,  SI  100
   140),HK(40,40),WT(40,40),REC(40,40),RECH(40,40),TIM(100),AOPT(20),   SI  110
     COMMON /CHMA/ PART(3,6400),CONC(20,20),TMCN(5,50),VX(4C,40),VY(40,  SI  131
   140),CONINT(20,20),CNRECH(20,20),POROS,SUMTCH,BETA,TIMV,STORM,STORM  SI  140

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                           August 2,  1985

Reference ;  "Computer model of two-dimensional solute transport and dispersion in
ground water" by L.  F. Konikow and J. D. Bredehoeft  (1978):  U.S. Geological Survey
Techniques of Water-Resources Investigations, Book 7. Chapter C2.

     The following modifications to  the  subject  computer code will allow the
model to simulate a first-order irreversible-rate reaction and  (or)
equilibrium-controlled sorption-desorption  for a linear  isotherm.

     An example of a first-order irreversible-rate reaction  is  radioactive
decay.  The rate constant,  x,_is defined as

                            X -./« 2/tM

where tj$ is the half -life of the solute.  The decay is applied  directly to
the tracer particles (rather than  at the nodes of the  finite-difference grid)
with the following exponential function:

                            k     k-1
                          Cp  = Cp    exp(-XA/)


where Cp is the solute concentration of  the tracer particle,
      k is the index in the time dimension; and
      A/ is the time increment.
This exponential formulation has no  numerical stability  restrictions.  However,
if the half -life is much smaller than the time step (A/) for solving the
transport equation, then some numerical  accuracy will  be lost.

     The equilibrium-controlled sorption-desorption is expressed  in terms of a
retardation factor (RA, as:
where Kd  is the distribution coefficient,  L3/M;

      Pb is the bulk density of the solid, M/L3; and
      e is the porosity.

     The revised input data formats will allow either one or both of these
types of reactions to be implemented by specifying a value of 1 for a new
variable, NREACT,  in columns 69-72  of input card 2.   If NREACT - 1, then
you must also insert a new card (or line) immediately following input card 3
that specifies values in the following order for DK (Kd), RHOB  (Pb), and
THALF  (tj$)  (in seconds), using a free format  (that is,  the three values are
separated by spaces or commas, but blanks are not read as zeros).  This
modification to the input data will not require any modifications to existing
data sets for nonreactive solutes (that is, identical results will be obtained

-------
with and without these  program changes).

      These modifications were evaluated by comparing the numerical solutions
with analytical solutions (van Genuchten  and Alves, 1982) for  various
combinations of parameters.   As can be seen in figure l, the agreement is
excellent for all cases.
    NO DECAY AND NO SORPTION
                                               DECAY WITH NO SORPTION
0.1-

0.8-

0.7-




OJ-

0.4-


0.3




O.H
    0   10   20  30   40  SO   10  70
                  DISTANCE (CM)


    SORPTION WITH NO DECAY
                                   90  no
                               - Analytical
                                 Numerical
                               ^S days
                 40  90   tO  70   80  tO   tOO
                 DISTANCE (CM)
                                                                          	Analytical
                                                                           •  Numerical
                                                                          ti/j= 4.5 days
                                                  «  20   jo  «o   so  w  TO
                                                            DISTANCE (CM)

                                              DECAY WITH SORPTION
                                                                            BO
                                                                                to  100
                                                      20
                                                        10  40  SO  tO   70
                                                            DISTANCE (CM)
                                                                            BO  tO
                                                                                   100
Figure 1.—  Comparison of numerical  and analytical solutions for transport
     with reactions in one-dimensional,  steady-state flow  (Ax = 2 cm.
     V = 25  cm/day, and  
-------
      This set of modifications also revises the structure of  subroutine
 CNCON.  Previously, equation 40 was solved twice — using concentrations at
 times A:-l and k* over  0.5 & each and summing the results.  Now, following
 the modifications of Tracy  (1982), the concentrations at times  k-1 and k* are
 first averaged and equation 40 then solved once over A/.  This will yield a
 reduction in execution time on the order of 5 percent.

      The program modifications can be implemented by deleting the following
 statements:
      A  60
      B 430
      B3100
      B3690
      E 710
      E1080
      E1830
      E2360
      E2460
      F2040
G
G
F2050
F2865
G 200
  220
  320
G 330
G 340
G 380
G 400
G 410
G 430
G 440
G 470
  475
  620
  650
  660
  670
  680
  690
G 700
G 710
G 720
G 730
G 740
G 750
G 760
  .800
  810
                             G
                             G
G1498
G1502
G1535
G1678
I 610
I 660
11360
11370
11380
                             G1470
 and by inserting the  following Fortran statements in the proper sequential
 location:
C     *  REV. MAY-AUG. 1985 BY  L.  KONIKOW  AND M.  PERSON TO INCLUDE:  *
C     *       DECAY AND EQUILIBRIUM  SORPTION-DESORPTION REACTIONS    *
C     *                                                               *
      COMMON /CHMR/ RF/DK/RHOB/THALF,DECAY/ADSORB,SORBI/DMASS1/CSTM2
      SORBI=0
      RMOB=0.0
      DMASS1-0.0
      DECAY=0.0
      THALF=0.0
     1PND/NCODES/NPNTMV/NPNTVL/NPNTD/NPOELC/NPNCHV/NREACT
         	READ REACTION TERMS  IN  FREE  FORMAT	
      IF (NREACT.EQ.1) READ  (5/*) DK/RHOS/THALF
      RF=1.0+(DK*RHOB/POROS)
      IF (THALF.GT.0.0) DECAY=ALOG(2.0)/THALF
      WRITE (6/895) DK/RHOB/RF/THALF,DECAY
      CFCTR2=DK*RHOB
      CFCTR=CONINT(IX/IY)«THCK(IX/IY)*AREA
      SORBI=SORBI+CFCTR*CFCTR2
  410 STORMI=STORMI*CFCTR*POROS
  740 FORMAT (1814)
  895 FORMAT (1HO/23X/14HREACTION TERMS//13X/37HDK      (DISTRIBUTION CO
     1EFFICIENT) s /E12.5/13X/37HRHOB     (BULK DENSITY OF SOLIDS)   = /E
     212.5/13X/37HRF      (RETARDATION FACTOR)       = /E12.5/13X/37HTMA
     3LF   (HALF LIFE OF DECAY/IN SEC)=  /E12.5/13X/37MDECAY   (DECAY CON
     4STANTsLN 2/THALF)= ,E12.5)
      COMMON /CHMR/ RF/DK./RHOB/THALF/DECAY/ADSORB/SORBI/DMASS1/CSTM2
      TDIV=(POROS*THCK(IX/IY)*RF)/DABS(DIV)
      WRITE (6/392)
      IF (RF.LE.1.0) GO TO 115
      WRITE (6/394)
      VMAX«VMAX/RF
                                                       A   59R
                                                       A   60R'
                                                       A   69R
                                                       B  182R
                                                       B  295R
                                                       B  296R
                                                       B  396R.
                                                       B  397R
                                                       B  398R
                                                       B  399R
                                                       B  431R
                                                       B  454R
                                                       B  455R
                                                       B  532R
                                                       B  534R
                                                       B  635R
                                                       B3015R
                                                       B3092R
                                                       B3094R
                                                       B3096R
                                                       B3691R
                                                       B4121R
                                                       B4122R
                                                       B4123R
                                                       B4124R
                                                       B4125R
                                                       E  205R
                                                       E  711R
                                                       E1055R
                                                       E1071R
                                                       E1072R
                                                       
-------
TIMV (CELOIS)
    VMAY=VMAY/RF
    VMXBD=VMXBD/RF
    VMYBD=VMYBO/RF
    WRITE (6x400) VMAX/VMAY
    WRITE (6/410) VMXBD/VMYBD
115 TDELX=CELDIS*XDEL/VMAX
    TDCO*(OISP(IX/IY/1)+OISP(lX/IY/2))/RF
310 FORMAT (1HO/19H TMV (MAX. INJ.) = /G12.5/20H
392 FORMAT (1MO/5X/16MFLUID VELOCITIES)
394 FORMAT (1MO/5X/27MEFFECTIVE SOLUTE VELOCITIES)
400 FORMAT (1H /8H VMAX = /1PE9.2t5X/7HVMAY = /1PE9.2)
    COMMON /CHMR/ RF/DK/RHOB/THALF/DECAY/ADSORB/SORB I/OMASS1/CSTM2
    IF (THALF.GT.0.0.AND.THALF.LT.TIMV) WRITE (6/685)
    DCYFCT=TIMV*DECAY
    DCY2=DCYFCT*0.5
290 DISTX=XVEL*CONST1/RF
    DISTY=YVEL*CONST2/RF
          	DECAY PARTICLES	
    PART(3/IN)=PART(3/IN)«EXP(-DCYFCT)
398 SUMC(IX,IY)=SUMC(IX/IY)+CONC(lXxIY)*EXP(-DCY2)
685 FORMAT (1HO/5X/51H**« CAUTION ***  DECAY HALF-LIFE IS LESS THAN TI
   1MV/V23X/24HACCURACY MAY BE AFFECTED/23X/34H(REDUCE TIMV BY DECREAS
   2ING CELOIS))
    COMMON /CHMR/ RF/DK/RHOB/THALF/DECAY/AOSORB/SORBI/OMASSI/CSTMZ
    DIMENSION CNCNC(20/20)/CNOLD(20/20)/CAVG(20/20)
    CAVG(IX/IY)=CONC(IX/IY)
    TVA2=TVA«0.5
    TMCMK=TIMV»10.0
    SRCDCY=0.0
    RFPOR=RF*POROS
    RFPORA=RFPOR*AREA
    	CONC.  CHANGE DUE TO:
          MIXING  AT SOURCE  CELLS...
          ...WITH DECAY OF  RECHARGE  DURING TIME INCREMENT
    CONST=TIMV
    DCYFCT=TIMV*DECAY
    DCYHLF=DCYFCT*0.5
    RFFCT=CONST/RF
    GO TO 70
    EQFCT1=RFFCT/THCK(IX/IY)
    C1=CAVG(IX/IY)
    C2 = C1
    CLKCN=C1
    IF (SLEAK.GE.0.0) GO TO 25
    CLKCN=CNRECH(IX/IY)*EXP(-OCYHLF)
    SRCOCY=SRCDCY*(CNRECH(IXxIY)-CLKCN)*SLEAR«TVA
 25 CONTINUE
    IF (RATE.GE.0.0)  GO TO  27
    CNREC=CNRECH(IX/IY)*EXP(-DCYHLF)
    SRCDCY«SRCDCY+(CNRECH(IX/IY)-CNREO*RATE*TVA
 27 IF (RECH(IX/IY).GE.O.O) GO TO  29
    CNREC2=CNRECH(IX/IY)*EXP(-DCYHLF)
    SRCDCY=SRCDCY+(CNRECH(IX/IY)-CNREC2)*RECH(IX/IV)*TVA
 29 CONTINUE
    IF (THALF.GT.TMCHK.OR.THALF.EQ.O.O)  GO TO 37
    IF (OIV.GE.0.0) GO  TO 37
    IF (CNOLD(IX/IY).LE.O.O.OR.CONC(IX/IY).LE.O.O)  GO  TO  37
E1074R
E1075R
E1076R
E1077R
E1078R
E1081R
E1831R
E2361R
E2451R
E2452R
E2461R
F 172R
F 205R
F 265R
F 267R
F2041R
F2051R
F24UR
F2415R
F2866R
F4174R
F4175R
F4176R
G 195R
G 201R
G 255R
G 295R
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
                      G
  296R
  297R
  305R
  306R
  321R
  322R
  323R
  341R
  342R
  343R
  344R
  345R
  381R
  4Q1R
  403R
  411R
  425R
  431R
  432R
  435R
  471R
  472R
  473R
  474R
  A76R
  477R
  473R
  482R
  484R
  486R

-------
 37
220
    C1=EXP((ALOG(CNOLD(IX,IY))+ALOG(CONC(IX/IY)))*0.5)
    IF  (NPCELLdX/IYKLE.O) C1=CNOLD(IX/I Y) *EXP (-OC YHLF)
    CONTINUE
    --- CONC. CHANGE DUE TO DISPERSION FOR TIMV ---
    X1=DISP(IX,IY/1)**CCAVG(IX/IY+1)+CAVG(IX-1/IY-M)-CAVG(IX/IY-1)-C
    AVG(IX-1/IY-1)>
    YY1=DISP(IX/IY,4)*CCAVG(IX+1/IY)+CAVG(IX+1/IY+1)-CAVG(IX-1/IY)-CAV
    G(IX-1/IY+1)>
    YY2=DISP(IX,IY-1,4)*)
                                 ... AND DECAY ---
    ADSORB=0.0
    RFDCY=RFPOR*DECAY*TVA*0.50
    C1=CONC
    C1B=C1*THCK
    DELDCY=CNOLD(IX,IY>-CNOLD(IX,IY)*(EXP<-DCYFCT))
    DMASS1=DMASS1-DELDCY*THCK(IX/IY)*RFPORA
    STORM=STORM+C1B*ARPOR
       --- COMPUTE MASS ADSORBED ---
    ADSORB = C1B*DMRHOB*AREA+AOSORB
    CMSOUT=CMSOUT+RECCIX,IY)«TIMV*«1.0-HTFCT>*CNOLO(IX,IY)+WTFCT«C1)
    CMSOUT=CMSOUT+RECM
    FLMOT=FLMOT+FLW*TVA*((1.0-WTFCT)*CNOLD(IX/IY)+WTFCT*C1)
       --- COMPUTE MASS LOST BY DECAY ---
    CSTM2=CSTORM*ADSORB-SORBI
    DMASS1=DMASS1*SRCDCY
    COMMON /CHMR/ RF/DK/RH06/THALFxDECAY/ADSORB/S6RBI/OMASS1/CSTH2
 50 RESID=SUMIO-CSTM2+DMASS1
    IF (SUMIO.GT. (STORMI + SOR8D) GO TO 70
    £RR3=-100.0*(RES1D>/(STORMI+SORBI-SUMIO)
    WRITE (6*295) DMASS1
    WRITE (6/296) ADSORB
    WRITE (6/298) SOR6I
    WRITE (6/332) CSTM2
    IF (SUMIN.NE. 0.0. AND. SUMIO.GT. ( STORMI+SORBI ) )  GO TO 90
295 FORMAT (8X/25MMASS LOST BY DECAY     = /1E12.5)
296 FORMAT (8X/25MMASS ADSORBED ON SOLIDS= /1E12.5)
298 FORMAT (8X/25HINITIAL MASS ADSORBED  = /1E12.5)
310 FORMAT (8X/25MINITIAL MASS DISSOLVED = /1E12.5)
320 FORMAT (8X/25MPRESENT MASS DISSOLVED = /1E12.5)
330 FORMAT (8X/25MCMANGE MASS DISSOLVED  = /1E12.5)
332 FORMAT (8X/25MCHANGE TOTL.MASS STORED= /1E12.5)
G
G
G
G
G
G
G
G
G
G
G
                                                                        G
                                                                        G
                                                                        G
                                                                        G
                                                                        G
                                                                        G
   468R
   489R
   592R
   621R
   651R
   661R
   671R
   681R
   691R
   701R
   711R
   721R
   731R
   741R
   751R
   761R
   8A1R
   935R
GU11R
GU12R
GU42R
GU43R
G1444R
G14A5R
GU71R
G1472R
G1473R
G1499R
G1536R
G1679R
G1715R
G1725R
G1726R
: 155R.
I 611R
I 637R
  661R
  751 R
  752R
  753R
  795R
  845R
I1341R
I1342R
I1343R
11361R
I1371R
I1381R
I13E2R

-------
                                  REFERENCES

Tracy, J. V., 1982, Users guide and documentation for adsorption and decay
     modifications to the USGS solute transport model:  U.S. Nuclear
     Regulatory Commission, Report NUREG/CR-2502, 138 p.

Van Genuchten, M. Th., and Alves, W. J., 1982, Analytical solutions of the
     one-dimensional convective-dispersive solute transport equation:  U.S.
     Dept. of Agriculture, Technical Bulletin 1661, 151 p.

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                              July 31, 1985

Reference i  "Computer model of two-dimensional solute transport and dispersion  in
ground-water," by L. F.  Konikow and J. D. Bredehoeft (1978):   U.S. Geological Survey
Techniques of Water-Resources Investigations, Book 7, Chapter C2.


The solute-transport  equations presented  in the cited report include two
terms that should have canceled out during the derivation and expansion of the
basic governing equation.  Although the corresponding changes to the code were
documented previously in the updates of August 26,  1981, the correct equations
are reproduced below  for clarification.   (The code  changes reduce slightly the
cpu time.)
dC

dt
             b    dXj
                             dc
a*,.
        W(C - C')
                                                                       (14)
           W(C  -  C')

              (.b
                                                                (16)
                             W(C - C')
                                                                (27)
                              CiJ,k-l
                                                                (39)
     AC:
               0.5  At
                    ±-{U>,    "'1;
                                  '7
                                      dx.i
0.5 At
b
9 ,.n 9C(**) > t
, ^D/y fl > f
9.v. • 9x .-
wfc, » - r1 )
rT ( V- JL 4r *«• /
€
                                                                      (40)

-------
                       BOTE 01 COMPUTER PROGRAM UPDATE
                                                            July 26, 1985


  Reference: "Computer model of tvo-dlmenslonaJ to Jute transport Mod disper-
  sion In ground vater" by 1. F. Konikov and J. D. Bredehoeft (1978): U.S.
  Geological Survey Techniques of Vater-Resources Investigations,  Book 7,
  Chapter C2.
    THE FOLLOWING MODIFICATIONS WILL CORRECT A PROGRAMMING ERROR  IN  THE
ROUTINE TO CALCULATE  CHANGES IN CONCENTRATION CAUSED BY DISPERSION.
FOR MOST PROBLEMS/  THE  DIFFERENCE IN THE RESULTS MILL BE NEGLIGIBLE.
HOWEVER/ IF THE PROBLEM BEING SOLVED IS ONE IN WHICH DISPERSIVE FLUXES
ARE VERY STRONG RELATIVE TO CONVECTIVE (ADVECTIVE) FLUXES/ THEN THE
ORIGINAL CODE COULD HAVE GENERATED NEGATIVE MASS-BALANCE ERRORS OF UP
TO SEVERAL PERCENT.

    THE MODIFICATIONS CAN BE IMPLEMENTED BY DELETING:   G1120
                                                         G1130
                                                         G1UO

AND INSERTING THE FOLLOWING STATEMENTS IN THEIR PROPER SEQUENTIAL
LOCATIONS (AS INDICATED BY THE LINE NUMBERS IN COLUMNS 73-60):

    CNCPCTsQ.O                                                           G1084G
    IF (CONC(IX/IY).GT.O.O) CNCPCT«CNCNC(IX/IY)/CONC(IX/IY>             G10BSG
    IF (CNCPCT.LT.0.0)  SUMC(IX/IY)=CNCPCT                               G1125G

-------
                          NOTE ON COMPUTER PROGRAM UPDATE
                                                                June 10,  1985

     Reference: "Computer model of ttro-dimenaional solute  transport and diapei—
     si on In ground vater" by L. F. Konikov and J.  D. Bredehoeft (1978):  U.S.
     Geological Survey Techniques of Water-Resources Investigations,   Book 7,
     Chapter C2.

     An additional program modification has been made since the previous  update
     of October 12, 1983.   The  following changes will allow the user  to specify
     16 particles per node.   This option can lead to increased numerical  accuracy
     in some cases.
         1.  Delete  the following Fortran statements from the source  code:
              A 210
              B 160
              B 600
              B 610
              B4010
              D 130
              D 155
D 160
D 380
D 510
D 520
D 832
D 840
D 880
D 900
D 920
D 940
0 950
D 970
D 980
D1000
D1010
D1020
D1030
D1040
E 160
F 140
F 175
F2905
G 150
G 230
G 240
G 350
G 360
G 850
G 860
G1060
G1070
G1420
G1430
I 130
         2.  Insert  the following Fortran statements in the proper  sequential
    location:                  |. .,(
    *  REV. JUNE-AUG.  1984  BY W. SANFQRD TO ALLOW  16  PTS.  PER NODE *    A  59
    COMMON /CHMA/  PART(3/6400),CONC(20/20),TMCN(5/50)/VX(20,20),VY(20/  A 211
    COMMON /CHMA/  P ART (3/6400»COMC (20* 20) /TMCN (5/ 50) ,VX < 2C/20), V V (20,  B 161
    If (NPTPND.NE.4.AND.NPTPND.NE.5.AND.NPTPND.NE.S.AND.NPTPND.NE.9.    B 601
   1    .ANO.NPTPNO.NE.16)  WRITE(6/880)                                  B 611
880 FORMAT (1HO,5X,47H*»*  WARNING *** NPTPNO MUST  EQUAL 4*5/8/9,OR 16)  B4011
    COMMON /CHMA/  PART(3'6400)/CONC(20/20)/TMCN(5/50>/VX(20/20)/Vr(20s  D 131
    COMMON /CHMP/  PTID(6400)                                              0 156
    DIMENSION RPT(16>/RNT(16)/RP(16)/RN(16)/IPT(16)                      0 161
    IF (NPTPND.EQ.16)  F1=0.25                                            0 223
    IF (NPTPNO.EO.16)  F2=0.125                                           D 224
    00 30 IA=1/16                                                         0 381
    RPT(IA)=0.0                                                           D 401
    RNT(IA)=0.0                                                           0 402
    DO 410 IX»2,NNX                                                       D 511
    DO 410 IY=2/NNY                                                       0 521
    KR2=0                                                                 0 541
    RR=KR+1                                                               D 793
    IF (NPTPNO.EG.16)  CO TO 72                                           0 795
    IPT(KR)=IND                                                           0 833
    GO TO 76                                                              0 835
 72 IF (TEST.LT.98.0.OR.TEST2.GT.0.0) GO TO 135                          D 841
 76 IF (TEST.LT.98.0.OR.TEST2.GT.0.0) GO TO 139                          0 845
    PARTC=CNODE+CONC(IXD/IYD)*F1                                         0 901
 80 PARTC=2.0*C1*CONC(IXO/IYO)/(C1*CONC(IXO/IYO))                        0 921
100 RPT(KR)=CONC(IXD,IYD)-PARTC                                          0 941

-------
    RNT(KR)=C1-PARTC
110 RPJUR)=0.0
    RNT(KR)=0.0
120 RPT(KR)=C1-PARTC
    RNT(KR)=CONC(IXO/IYO)-PARTC
130 IF (NPTPNO.E0.16) GO TO 135
    PART(3/IND)=PARTC
    RP(KR)=RPT(KR)
    RNUR) = RNT(KR)
    GO TO 139
135 00 138 ITT = 1,2
    EVET2=(-1.0)*«ITT
    00 138 ISS '= 1,2
    EVES2=<-1.0)**ISS
    PART(1,IND)=(IX+F1*EVET)+F2*EVET2
    PART(2,IND)=(IY+F1*EVES)+F2*EVES2
    PART(2,IND)=-PART(2,IND)
    KR2=KR2+1
    IF (TEST.LT.98.0.0R.TEST2-.GT.O.O)
    PART(3,IND) = PARTC
    RP(KR2) = RPT(KR)
    RNUR2) - RNT(KR)
    IPTUR2) = INO
    GO TO 137
136 PART(3,IND) = C1
137 PTIO/CONC<20/20>,TMCN<5,50>'VX<20/20>/VY(20/
              PART (3* 6400) *CONC( 20/20) 'TMCN(5s50>'VX(20/20>/VY(20/
              PTIDC6400)
              16) F1=0.25
IF (NPTPND.EQ. 16) F2=0.125
IF (NPTPNO.EC.16) GO TO 400
GO TO (401, 411/421/431/441,451,461/471, 481), ITEM
GO TO (482/4 83/4 84 / 485 / 486/487, 488/489/490/491,4 92/493, 494/49 5 /
       496, 497), ITEM
GO TO 441
GO TO 530
PART(1,IP)=IX-F1-F2
PART(2/IP)=IY-F1-F2
PTIO(IP)=1
GO TO 530
PART(1/IP)=IX-F1-F2
PART(2/IP)=IY-F1*F2
PTID(IP)=2
GO TO 530
    PART(2/IP)=IY-F1-F2
    (>TJD(IP)=3
 0 951
 D 971
 0 981
 D10C1
 01011
 D1012
 01013
 D1014
 D1015
 01016
 01017
 01018
 01019
 01021
 01022
 01023
 01024
 D1C25
 01026
 D1027
 D1028
 01029
 01031
 01032
 01033
 01034
 D1035
 01036
 01037
 01036
 01039
 01045
 E 161
 F 141
 F 176
 F 223
 F 224
 F29C1
 F2906
 F2921
 F2922
 F2924
 F3266
 F3267
 F3268
 F32t9
 F3271
 F3272
 F3273
 F3274
 F3275
F3276
F3277
F3278

-------
485
486
AS?
488
   GO TO 530
   PARTC1/IP)=IX-F1+F2
   PARTC2/1P)=IY-F14F2
   PTID(IP>=4
   GO TO 530
   PART(1/1P)=IX-F1-F2
490
491
492
 493
    PTID(IP>=5
    GO TO 530
    PAKT(1/1P)
    PART<2,IP)
    PTID(IP)=6
    GO TO 530
    PA?T(1/IP)
    PART(2/IP)
    PTIOCP)=7
    GO TO 530
    PARTU/IP)
    PART(2/IP>
    PTID(IP>=8
    GO TO 530
    PART(1/IP)
    PART(2/IP>
    PTIO(IP)=9
    GO TO 530
    PART(1/IP)
    PART(2/IP)
    PTID(IP>=1
    GO TO 530
    PART<1/IP)
 495
 496
497
             =IX-M-F2
             =IY+F1+F2
             =IX-F1+F2
             =IY+F1-F2
             =IX-F1+F2
             =IY+F1+F2
             =IX+F1-F2
             =IY-F1-F2
             =IX+F1-F2
             =IY-F1+F2
             0
             =IX+F1+F2
     PT10(IP)=1
     GO  TO  530
     PART(1/IP)
     PART(2/IP)
     PTID(IP)=1
     GO  TO  530
     PART(1rIP)
     PART(2/IP)
     PTID(IP>=1
     GO  TO  530
     PART(1,IP>
     PART(2/IP>
     PTIO(IP>=1
     GO  TO  530
     PART(1/IP)
     PART(2/IP)
     PTID(IP)=1
     GO  TO  530
     PART (1,IP)
              1
              =1X*F1*F2
              =IY-F1*F2
              2

              =lX-»-F1-F2
              =IY+F1-F2
              3

              =IX+F1-F2
              *IY+F1+F2
              4

              =IX+r1*F2
              =IY+F1-F2
              5

              =IX*F1**2
     PTIDCIP)=16
     GO TO 530
     COMMON /CUM A/
                  PART(3/64GO)/CCNC(2C'/2C)/TMCN(5/50)/VX(2C<20)/VY(2G/
F32S1
F3282
F32£3
F3284
F3285
F3286
F3267
F3283
F3289
F329.1
F3292
F32.93
F3294
F3295
F3296
F3297
F329S
F3299
F3301
F33C2
F3305
F33C4
F3305
F3306
F3307
F3308
F33C9
F3311
F3312
F3313
F33U
F3315
F3316
F3317
F3318
F3319
F3321
P3322
F3323
F3324
F3325
F332C
P3327
F332;
F332?
F3331
F3332
F3323
F3334
F3335
F 5 3 3 6
F3337
G  151

-------
    DO 10 IX=2/NNX                                                      G 231
    DO 10 IY=2/NNY                                                      G 241
 20 DO 60 IX=2/NNX                                                      G 351
    DO 60 IY=2/NNY                                                      G 361
 70 DO 90 IX-2/NNX                                                      G 851
    DC 90 IY=2/NNY      .                                                G 861
110 DO 130 IX=2/NNX                                                     G1061
    DO 130 IY=2/NNY                                                     G1071
    00 270 IX=2/NNX                                                     G1421
    DO 270 IY = 2/NNY                                                     GU31
    COMMON /CMMA/ PART(3/6400)/CONC<20/20)/TMCN(5/50)/VX(20/20)/VY(20*  I 131

-------
                             NOTE ON COMPUTER PROGRAM UPDATE

                                                              October 12, 1983

     Reference;    "Computer model of two-dimensional solute transport and dispersion
     in ground-water," by L. F.  Konikow and J. D.  Bredehoeft (1978):  U.S.
     Geological  Survey Techniques of Water-Resources Investigations, Book 7,
     Chapter C2.


     Several additional program  modifications have been made since the  previous
     update of August 26, 1981. These can be implemented  as follows:

          1.   The following changes assure that all time-step and  flow calculations
          are perfomed entirely  in double precision.  These changes will only make
          a difference on certain computers.

               a)   Delete the following  Fortran statements from the source code:

                    A 115
                    B  55

               b)   Insert the following  Fortran statements in the proper sequential
               location:
 REAL *8TMSUM/ANTIM/TDEL                                                   A 116
 REAL *8TMSUM/ANTIM/TDEL                                                   B  56
 REAL *80XINV/OYINV/ARINV/PORINV                                          3  57
 REAL *80XINV/DYINV/ARINV/PORINV                                          C  45
 REAL *80XINV/OYINV/ARINV/PORINV                                          E  35
 REAL *80XINV/DYINV/ARINV/PORINV                                          F  33
 REAL *80XINV/DYINV/ARINV/PORINV                                          G  35
           2.   The  following format changes are made to provide more convenient
           printouts  for most situations.

               a)   Delete:

                    B1630         B2620         B3600
                    B2330         B3440

               b)   Insert:


160 WRITE (6x840)  (VPRM(IX/IY),1X = 1/NX)                                     B1631
260 WRITE (6/540)  (PERMCIX,IY)/IX = 1/NX)                                     52331
320 WRITE (6/840)  (VPRM(IX/IY)/IX=1/NX)                                     326?1

-------
            3.   The following changes will assure that all output routines are called
            at the end of the last time step of a pumping period.

                a)   Delete:

                     A  450
                     A  610

                b)   Insert:
     IPC.ISQ
     IF  (TOEL.EO.(PyR-TlNT))  IPCK=1
     IF  (REMN.EQ.O.O.OR.N.EG.NTXM.C8.IPCK.EQ.1)  CALL  OuTPT
120  IF  
     IF  (NPNCHV.EQ.O)
     SNOFILE(7)
 155 CONTINUE
   GO TO 155
                                         A  702
                                         A  703
                                         A  704
                                         A  705
            5.   The following changes were made primarily to improve the interpolated
            estimates of velocities at tracer particles located in wells adjacent to no-
            flow boundaries and to assure a more consistent and uniform regeneration
            of particles at  fluid  sources.
                 a)   Delete:
               420
               430
               440
               450
               500
               510
               520
               530
     F 780
     F 800
     F1010
     F1030
     F1070
     F1280
     F1300
     F1340
     F1560
     F1580
     F1620
F1660
F1840
F1860
F1880
F1890
F1920
F1930
F1940
F1950
F1960
F1980
F1990
F2715
F2725
F2735
F2745
F2755
F2765
F2775
F2785
F2795
F2805
F2815
F2825
F2835
F3390
F3395
F3410
F3420
F3430
F3440
F3450
F3460

-------
              b)   Insert:
     *                REVISED  OCTOBER  1983
     DHXsHK(lX-1/lY)-HK(IX+1/IY)
     IF  
-------
                 5b) Inserts (continued)
1275
1277

1273
1279
1280
1282

1283
1 284
1235
1287

1283
1239

1290
         GO  TO 1277
         GO  TO 1278
         GO  TO 1279
IF CTHCK(IVX,IVY).EQ.O.O>
IF CTHdUlXE/IYS>.EQ.O.O>
IF (THCIUIXE/IVY>.EQ.O.O>
GO TO 1290
VXNW*VXSW
IF (THCK
IF (THCK(IVX/IVY).EQ.O.O)
IF 
GO TO 1290
VXSE*VXNE
IF (THCIUIVX/IVYKGT.O.C) GO TO 1284
VYNHSVYNE
VXSWsVXNW
VYSWSVYSE
                                          F1744A
         GO  TO 1282
         GO  TO 1283
         GO  TO 1284
,EQ
.EQ
,EQ
                      ,0)
                      ,0)
                      ,0)
GO
GO
GO
TO
TO
TO
•GT.0.0)  GO  TO  1289
GO TO 1290
IF (THCIUIVX/IYS),
IF (THCIUIXE/IVY>,
IF (THCHCIXE/IYS)
GO TO 1290
VXSU-VXNW
IF (THCK(IXE/IVY)
VYNE«VYNW
VYSEsVYSW
VXSE=VXNE
CONTINUE
CELYQsCELDYH*2.0
VYWsVYNW»(1.0-CELYO)*VYSW*CELYO
VYEsVYNE-(1.0-CELYD)*VYSE*CELYO
1287
1288
1289
F1744C
F17440
F1744E
F1744F
F1744G
F1744H
F1744I
F1744J
F1746A
F17468
F1746C
F17460
F1740E
F1746F
F1746G
F1746M
F1746I
F1746J
F1743A
F17488
F1748C
F17480
F1743E
F1748F
F1743G
F1748H
F1748I
F1749A
F1921
F1931
F1951

-------
        Reference:
                          NOTE ON COMPUTER PROGRAM UPDATE

                                                              August 26, 1981
        "Computer  model of  two-dimensional solute  transport and
        dispersion in ground  water," by  L. F. Konikow and J. D.
        Bredehoeft (1978):  U.S.  Geological Survey  Techniques of
        Water-Resources Investigations,  Book 7, Chapter C2.


        Several additional program modifications have been made since the previous
        updates of March 26 and December 4,  1980.  One  improvement included  in
        these changes is a correction to allow for variable-length pumping periods and
        (or) changing pumping rates during simulations  involving transient flow and
        the use of data set 10.  All of the  changes since March 26, 1980  can be
        implemented as follows:

        1.   Delete the following Fortran statements from the source code:

        A 140         B 880         E2050         G  530         H 125
        A 380         B3790         F  60         G  540         H 436
        A 470         C  80         F 180         G  550         H 438
        A 620         C 145         F4080         G  560         H 715
        B  90         C1200         G  70         G  570         H 720
        B 185         D  60         G 490         G  580         H 725
        B 460         E  80         G 500         G  590         H 735
        B 640         E 270         G 510         G  600         I  60
        B 700         E 720         G 520         H  60         11160                i
                                                              11530

        2.   Insert the following Fortran statements in the proper sequential
            location:
C      *                 REVISED DECEMBER 1980                             *    A  56
C      *                 REVISED AUGUST 1981                               *    A  57
       REAL *3TMSUM/A,MT1M          '                                            A 115
     2PNCHV/NPDELC/1CHK                                                        A 141
       TMSUMsQ.O                                                                A 265
       IF  (INT.GT.1) TMSUMsTMSUM+PYR                                          A 325
       TINT=SUMT-TMSUM                                                          A 381
       IF  (S.EQ.O.O.ANO.ICHK.EQ.O.AND.(N.GT.1.0R.INT.GT.1) )  GO  TO 101       A 435
  101  CALL MOVE                                                                A 471
       IF  (SUMT.GE.(PYrt+TMSUM))  GO TO 140                                     A 621
       REAL *8TMSUM/ANTIM                                                      a  55
     2PNCHV/NPDELC/ICHK                                                        B  91
       COMMON /BALM/ TOTLQ/TOTLQI/TPIN/TPOUT                                  S 186
       TPINsQ.O                                                                 B 317
       TPOUT=0.0                                                                B 318
       ICHKsQ                                                                   B 395
       IF  (NITP.LE.O)  WRITE (6/385)                                           B 615
       IF  (ZCHK.LE.O)  WRITE (6/1110) INT                                      B 695
       IF  (ICHK.LE.O)  GO TO 20                                                 B 7U'i
       PYRsPINT*S6400.0*365.25                                                 B aJ5
       IF  (NPNTMV.EU.O) NPNTMV=999                                            B 8j
       IF  (TINIT.GT.PYR)  WRITE  (6/475)                                         B 3i
    50  ANTIM=NTIM                                                               B 882
       00 55 Ksl/NTIM                                                           B 884
    55  TIM«)»PYR/ANTIM                                                        B 886
       IF  (INT.GT.1.AND.ICHK.LE.O) RETURN                                     B1345

-------
                        NOTE ON COMPUTER PROGRAM UPDATE
                                                          March 26,  1980

     Reference;   "Computer model of tuo-dimensional solute transport and
     dispersion in ground-water," by L.  F. Konikow and J. D. Bredehoeft
     (1978):  U.S. Geological Survey Techniques of Voter-Resources Inuesti-
     gotions, Book 7t Chapter C2.


     Several  additional program modifications have been made since the
     previous update of May 16, 1979.  These can be implemented as follows:

          1.  Delete the following Fortran statements from the source
             code:
                   E1980
                   E1990
                   E2030
                   F 480

          2.   Insert the following Fortran statements in the proper
              sequential location:


IF  (THC<( IX,IY).EQ.O.O>  GO TO  199                                         E1972
3AVX32.0*THC«IX,IY>*THCK(IX+1/IY)/(THCK(IX,IY)+THCK        E1976
OISP(IX,IY,4>»DISP
-------
IS LONGER TH
      IF  (NX.GT.3.AND.NY.GT.3)  60 TO 185
      HMX-PIES/XNS
      HMY»PIES/YNS
      HMIN*DMIN1(HMIN/HMX/HMY)
  185  CONTINUE
  475  FORMAT  (1 HO/5X/65H***  WARNING ***   INITIAL TIME STEP
    1AN  PUMPING  PERIOO/25X/34H***ADJUST EITHER TINIT OR
    2  «  /I6/13X/39HPINT    (PUMPING PERIOD IN YEARS)      a/F11.3/13X/39
  885  FORMAT  (1HO/5X/38H***  WARNING ***  NITP MUST BE POSITIVE)
 1110  FORMAT  (1H1/5X/25HSTART PUMPING PERIOD NO. /I2//2X/23HNO  PARAMETER
    1S REDEFINED/)
    2PNCHV/NPOELC/ICHK
      COMMON  /BALM/  TOTLQ/TOTLQI/TPIN/TPOUT
      PQIN»0.0
      PQOUTaQ.O
;      	CUMULATE PUMPAGE AND RECHARGE FOR MASS BALANCE	
      IF  (REC(IX/IY).GT.O.O) GO TO 32
      PQINaPQIN+RECUX/IY)
      GO  TO 34
   32  PQOUTsPQOUT+RECUX/IY)
   34  IF  (R£CH(IX/IY).GT.O.O) GO TO 36
      PQINsPQIN + RECHUX/IY)*AREA
      GO  TO 38
   36  PQOUTapQOUT+RECH(lX/IY)*AREA
 -  38  IF  (VPRM(IX/IY).EQ.O.O) GO TO 130
      TPINsPQIN*TIM(N)+TPlN
      TPOUT«PQOUT*TIM(N)+TPOUT
     2PNCHV/NPDELC/ICHK
     2PNCHV/NPDELC/ICHK
      TMV»TIM(N)*1.0E5
      MAXX'O
      MAXYaQ
      IF  (TDIV.GE.TMV) GO TO 20
      TMV»TOIV   _                                                 _  _
      MAXXaJX
      MAXYaiy
       IF (AMAX1(VHAX,VM.AY/VMX8D/VMYBD).LE.1.0E-10)  WRITE(6/570)
  200  IF (NMOV.EQ.1) GO  TO  235
       IF (LIM) 210/220/230
       WRITE  (6/560) MAXX/tlAXY
       GO TO  240
  235  WRITE  (6/580)
  560  FORMAT  (1H /15X/35HHAX.  INJECTION  OCCURS  AT  CELL  IX a ,13,7H  IY a

  570  FORMAT  (1 HO/5X/47H*** WARNING  •*•   DECREASE  CRITERIA IN c 230-260)
  530  FORMAT  (1HQ,10X/63H*TIME  INCREMENT FOR SOLUTE TRANSPORT EQUALS TIM
      1E  STEP  FOR FLOW*)
      2PNCHV/NPDELC/ICHK
       IF (HOO(IMOV/50).E3.0)  I?RNT=1
       IF (M00( IMOV/NPNTMV).EQ.O)  IPRNT*-1
  650  IF (IPRNT.NE.O)  CALL  CHMOT
      2PNCHV/NPOELC/ICH<
       CMREC2=C1
       IF (RECH(IX/.IY) .LT.0.3)  CP4REC 2« CNRECH (I X , I Y }
       DELCsEQFCT2«(C1«OIV-aATE*CNR£C-SLcAiC«CLICC.-J-RECH(IX,IY)*C.NREC2)
    COMMON /BALM/ TJTLU/TOTL«H *TP IN,TPOUT
    TPUMapQlN*PuOUT
    .PUHPafP IN + TPOJT
   2PNCHV/NPDELC/ICHK
    IF (S.Ea.Q.O.AND.N.LT.NTIM.AND.INT.GT.C) GO TO  100
150 IPRNTsQ
    RETURN
440 FORMAT CH /5dX/!2/6x/F7.1/8X/F7.1/8X/F7.3)
                                                                        81912
                                                                        B1913
                                                                        81914
                                                                        81915
                                                                        81916
                                                                        83384
                                                                        83385
                                                                        33791
                                                                        84015
                                                                        84532
                                                                        84533
                                                                        C  81
                                                                        C 146
                                                                        C 192
                                                                        C 193
                                                                        C1181
                                                                        C1182
                                                                        C1183
                                                                        C1184
                                                                        C1185
                                                                        C1186
                                                                        C1187
                                                                        C1188
                                                                        C1189
                                                                        C1201
                                                                        C1232
                                                                        C1233
                  61
                  81
                 275
                 284
                 285
                 722
                 724
                 725
                 726
               £1125
               E2052
               E2054
               E2102
               £2104
               £2106
               E2635
               £2636
               E2637
               £2633
               £2639
               F   61
               F4J65
               F4082
               F4QS5
               G   71
               ti  455
               G  475
               G  595
               H   61
               H  126
               H  71 6
               H  721
               I   61
               I  905
               11155
               11165
               1*531

-------
Table 1 -- List of Fortran statements to be deleted from original
           source deck (line identification numbers in columns  73-80)
A 150
fi 100
B 190
B3900
C 90
C 150
D 70
D 350
D 870
D1080
E 90
F 70
F 220
F 360
F2710
- A 160
- B 110


- C 100

- D 80
- D 360


- E 100
- F 80
- F 240
- F 390
- F3380
F3400
F3470 - F3500
F3540
F3630 - F3640
F3740
G 80 - G 90
G 640
G1100
G1500
G1540
G1590
G1630
G1660
G1680
H 70 - H 80
H 130
H 490
H 570
H 730
H 870
H 890
H 920
H1060
H1080
H1130
I 70
I 620
I 800




- H 580
- H 820

- H 900
- H 930

- HL110

- I 80
- I 640



 Table 2  --  (follows)  --  Listing of new or replacement Fortran state-
            ments to be inserted into program at sequential location
            indicated  by  line identification number in columns 73-80.
            For example,  statement A 55 should be inserted between
            lines A 50 and A 60.

-------
                    NOTE ON OCMPUTER PROGRAM UPDATE
                                                       May 16, 1979
Reference:  "Computer model of tuo-dunensional solute transport and
dispersion in ground water," by L. F. Konikow and J. D. Bredehoeft
(1978):  U.S. Geological Survey Techniques of Water-Resources Inves-
tigations, Book 7t Chapter C2.

     A number of refinements and modifications have been made to the
computer program that is documented in the above report since it was
published.  These changes improve the accuracy of the numerical solu-
tion for many problems, but slightly increase the execution time and
core storage requirements.  The degree of improvement in accuracy is
problem dependent.  For some problems the changes will have no sig-
nificant effect on the numerical solution, but for other problems
the improved accuracy will more than offset the increased cost.
Thus, in general it is recommended that these changes be incor-
porated into the program.

     Most of the program changes affect the treatment of tracer
particles at nodes that represent fluid sources and sinks.  Other
changes improve the methods for calculating the mass balance errors
for the fluid and for the solute.  The program changes will be
invisible to the model user in the sense that the input data require-
ments and formats are not affected.

     Implementation of these changes will require the deletion of
146 statements from the original program listing (Attachment I in
the report), and the insertion of 179 new statements.  The state-
ments to be deleted are listed sequentially in Table 1 of this
update note.  The new statements to be inserted are listed sequen-
tially in Table 2.  All statements have unique identification numbers
in columns 73-80; the sequential order of these line identification
numbers must be maintained in making the program changes.

-------
    IF (NPOLDUX/IY) .GT.Q) NPOLD
-------
                               Table 2
    »                     REVISED  APRIL  1979                        *
    COMMON /PRMC/ NODE I0(20/20)/NPCELL(20/20)/NPOLO(20/20)/LIMBO(500)/
   1 IX03S(5),IYJBS(5)
    INTEGER OVERriD
    COMMON /PRMC/ NOOE10(20,20),NPCELL(20,20),NPOLO(20,20),LIMdO(500),
   1 !XOJS(5),IYOiS(5)
    COMMON /dALM/ TOTLQ,TOTLiJI
    TOTLQI=0.0
520 FORMAT (1H ,7X,2I4,3X,F9.4,3X,F8.2)
    COMMON /PRKC/ NODE 10(20,20),NPCELL(20,20),NPOLO(20,20),LIMBO(500),
   1 IX03S(5),IY08S(5)
    COMMON /3ALM/ TQTLQ,TOTLQI
    IF (OELQ.GT.0.0) GO TO 125
    GO TO 130
125 TOTLai"TOTLQI+DELD*TIH(N)
    INTEGER *2 PTIO
    COMMON /PRKC/ NOOE10(20,20),NPCELL(20,20),NPOLO(20,20),LIM80(500),
   1lXOdS(5),IY03S(5)
    COMMON /CHMP/ PTIO(3200)
 10 00 20 IN*1,NPMAX
    PTIO(IN)*0
    00 20 I0s1,3
    NPOLO(IX,IY)sNPTPNO
150
    PTID(INO)=KR
    PART(1,IND)sIX
    PTIO(INO).*S
    IF (EVET.LT
    IF (EVET.GT
    IF (EVET.LT.O)
    IF (EVET.GT.O)
    COMMON /PRMC/
                O)
                O)
                PriD(INO)=6
                PTIO(INO)=8
                PTIO(INO)s7
                PTIO(I NO)=9
               NOOEI0(20/20)/NPCELL(20/20)/NPOLD(20/20)/LIMBO(500)/
390
1 IX03S(5) /I YOBS( 5)
 INTEGER *2 PTIO
 COMMON /PRMC/ NODEID(20/20)/NPCELL
1 IX03S(5) /IY08S(5;
 COMMON /CHMP/ PTIO(3200)
 F1«0.30
 F2«1. 0/3.0
 IF (NPTPNO.EQ.4) FlaQ.25
 IF (NPTPNO.EQ.9) F1*F2
 IF (NPTPNO.Ea.8) F2aQ.25
 IF (KFLAG.EQ.O) GO TO 398
 IF (THCK (IX+1,IY).EQ. 0.0. AND. (VPRM
1).LT.O.O.OR.THCK(IX,IY-1 ) . EQ. 0. 0) .
2C(IX/IY-H).LT.O.O.OR.THCK(1X^IY-H)
 IF (THCK(IX-1/IY) .EQ. 0.0. AND. (VPRM
1).LT.O.O.OR.THCK(IX/IY-1) . EQ. 0. 0) .
2C(lX«IY*1).lT.O.O.OR.THCKCIXsIY+1>
 IF (THCK (IX, I Y- 1 ). EQ. 0.0. AND. (VPRM
1 ) ,LT.O.O.OR.THCK(IX-1/IY).EQ.O.O).
2C(IX+1,IY>.LT.O.O.OR.THCK(IX+1,IY)
 IF (THCK(IX,IY+1).EQ.O.O.AND.(VPRM
1 ) .LT.O.O.Oft.THCK(ZX-1«ZY) . EQ. 0. 0) .
                                       (20/20)/NPOLD(20/20)/LIMBO(500)/
                                       (IX/IY-1)
                                       AND. ( VPRM
                                       .EQ.O.O) )
                                       ( I X , I T-1 )
                                       AND. ( VPRM
                                       .EQ.O.O)>
                                       (IX-1,IY)
                                       ANO.(VPRM
                                       .EQ.O.O))
                                       (IX-1,IY)
                                       ANO. ( VPRM
                                                .GE. 0.09
                                                ( IX, I Y+1
                                                 GO  TO 5
                                                .GE.0.09
                                                ( IX/ I Y +1
                                                 GO  TO 5
                                                . GE.0.09
                                                
-------
    FCT2»NPCELL(IX/IY)
    IF (FCT2.GT.0.0) WTFCT3FCT1/FCT2
    FLMOTsFLMOT+FLW*TVA*((1.0-WTFCT)*CNOLD(IX/If)+WTFCT *CONC ( IX/I Y»
265 NPOLO(IX/IY)*NPCELL(IX/IY)
    fy,PCELL(IX/IY)*0
    COMMON /PRMC/ NODE I 0(20/20)/NPCELL(20/20)/NPOLD(20/20)/LIMBO(500)/
   1IX03S(5)/IY08S(5)
    COMMON /3ALM/ TOTLQ/TOTLGI
    PQINsO.O
    P30UTsO.O
    TPIN-0.0
    TPOUTaQ.O
    IF (REC(IX/IY).GT.0.0) GO TO 32
    PQINapQIN + RECUX/IY)
    GO TO 34
 32 PQOUTapQQUT+REC(IX/IY)
 34 IF (RECH(IX/IY).GT.0.0) GO TO 36
    PQINapQlN+RECH(IX/IY)*AREA-
    GO TO 38
 36 PQOUTsPQOUT+RECH(IX/IY)*AREA
 33 IF (VPRMUX/IY) .EQ.0.0) GO TO 60
    TPUMaTPUM+PQIN+PQOUT
    TPINSPQIN*SUMT
    TPOUTapQQUT*SUMT
    TOTLQNaTOTLQ*TOTLQI
    SRCS*QSTR-TPIN+TOTLQI
    SINSINKS)*O.S
    IF (DENOM.EJ.0.0) GO  TO 100
    PCTERR=EfiRM8*100.0/DENOM
    WRITE (6/211) TPIN
                  TPOUT
                  TOTLQI
                  TOTLQ
                  TOTLQN
                  PCTERR
130
201
202
203
204
210
211
212
240
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
WRITE
FORMAT
FORMAT
FORMAT
FORMAT
FORMAT
FORMAT
FORMAT
FORMAT
FORMAT
COMMON
          (6/212)
          (6/202)
          (6/203)
          (6/260)
          (6/280)
          (6/201)
          (6/202)
          (6/203)
          (6/204)
          (6/211)
          (6/212)
                  QIN
                  QOUT
                  QNET
                  PQIN
                  PQOUT
            (1HO/10F12.7)
            (1HO/2X/33HRATE  MASS BALANCE — (IN C.F.S.)
            (4X/29HLEA
-------
  SECTION 4






BIBLIOGRAPHY

-------
SELECTED BIBLIOGRAPHY ON SOLUTE-TRANSPORT

        PROCESSES IN GROUND WATER
           Leonard  F. Konlkow
         U. S. Geological Survey
            Reston,  VA  22092
           revised:   January  1988

-------
                                 CONTENTS
Topic                                                            Page
General	3
Theory and Development of Solute-Transport Equation	      5
Macrodlsperslon and Stochastic Approaches	      6
Diffusion and Dispersion	      9
Accounting for Reactions	     12
Flow of Immiscible Fluids and Multiphase Transport	     15
Transport in Fractures	17
Analytical Solutions	18
Numerical Methods and Models	20
Parameter Determination and Tracers	     24
Analysis of Field Problems	     27
Aquifer Reclamation and Management Aspects	     30
Freshwater-Saltwater Relationships 	     31

-------

-------
Bear, Jacob, 1979, Hydraulics of groundwater:  McGraw-Hill, New York,
     567 p.

Bear, J.,  and Verruijt,  A.,  1987,  Modeling groundwater flow and pollution:
     D. Reidel Publishing Company, Dordrecht, Holland, 414 p.

Bird, R. B., Stewart, V. E., and Lightfoot, E. N., 1966, Transport
     phenomenal  John Wiley and Sons, New York, 780 p.

Domenlco,  P. A., 1977, Transport phenomena in chemical rate processes
     In sediments:  Ann. Rev. Earth Planet. Sci., v. 5, p. 287-317.

Freeze, R. A., and Cherry, J. A., 1979, Groundwatert  Prentice-Hall, Inc.,
     Englewood Cliffs, N.J., 604 p.

Glllham, R. U., and Cherry, J. A., 1982, Contaiinant migration In
     saturated unconsolidated geologic deposits!  Geological Society of
     America, Special Paper 189, p. 31-61.

Greenkorn, R. A., 1983, Flow phenomena in porous medlai  Marcel Dekker,
     Inc., New York, 550 p.

Grlsak, G. E., and Jackson, R. E., 1978, An appraisal of the hydro-
     geological processes involved in shallow subsurface radioactive
     waste management in Canadian terrain:  Inland Waters Directorate,
     Ottawa, Canada, Scientific Series No. 84, 194 p.

IAHR-ISSS, 1972, Fundamentals of transport phenomena in porous media:
     Proc. Second Symposium, Univ. of Guelph, Ontario, Aug. 7-11,
     1972, 797 p.

National Research Council, 1984, Groundwater contamination:  Studies  in
    Geophysics, National Academy Press, Washington, D. C., 179 p.

Nelson, R. W., 1978, Evaluating the environmental consequences of
     groundwater contamination, 1.  An overview of contaminant arrival
     distributions as general evaluation requirements:  Water Resources
     Research, v. 14, no. 3, p. 409-415.   [Also see parts 2-4, same
     issue, p. 416-450.]

Rellly, T. E., Franke, 0. L., Buxton, H. T., and Bennett, G. D., 1987, A
     conceptual framework for ground-water solute-transport studies with
     emphasis on physical mechanisms of solute Movement:  U.S. Geol. Survey
     Water-Resources Inv. Report 87-4191, 44 p.

Schwartz,  F. W., 1975, On radioactive waste management:  An analysis
     of the parameters controlling subsurface contaminant transport:
     Jour. Hyd., v. 27, p. 51-71.

-------
General, continued


Vrba, J., and Romijn,  E..  [eds.l,  1986,  Impact of agricultural activities
     on ground watert   International  Assoc.  of Hydrogeolegists, International
     Contributions to Hydrogeology, v. 5, 332 p.

Yaron, B., Dagan, G.,  and Goldshiid, J., eds., 1984, Pollutants in porous
     •edia — The unsaturated zone between soil surface and groundwateri
     Ecological Studies 47, Springer-Verlag,  Berlin, Heidelberg, New York,
     Tokyo,  296 p.

-------
              THEORY AND DEVELOPMENT OF SOLUTE-TRANSPORT EQUATION
Bachnat, Y., and Bear. J., 1964. The general equations of hydrodynamic
     dispersion in homogeneous, isotroplc, porous tediums:  Jour.
     Geophys. Research, v. 69, no. 12, p. 2561-2567.

Biake, T. R., and Garg, S. K.. 1976, On the species transport equation
     for flow in porous Media»  Water Resources Research, v. 12, no. 4.
     p. 748-750.

Bredehoeft, J. D., and Pinder, G. F., 1973, Mass transport in flowing
     ground watert  Water Resources Research, v. 9, no. 1, p. 194-209.

Dagan,  Gedeon,  1987,  Theory  of  solute  transport by  groundwaten  Ann.  Rev.
     Fluid Mech., v.  19,  p.  183-215.

Domenico, P. A., and Palciauskas, V. V., 1979, The volume-averaged
     •ass-transport equation  for chemical diagenetic Modelst  Jour.
     Hydrology, v. 43, p. 427-438.

Gray, W. G., 1982, Derivation of vertically averaged equations describing
     multiphase flow in porous mediai  Water Resources Research, v. 18,
     no. 6, p. 1705-1712.

Gray, W. G., and O'Neill, K., 1977, Comment on "On the species transport
     equation for flow In porous media" by Thomas R. Blake and Sabodh K.
     Gargi  Water Resources Research, v. 13, no. 3, p. 695-696.

Hatton, T. A., and Llghtfoot, E. N., 1984, Dispersion of trace solutes  in
     flowing groundwaten  Water Resources Research, v. 20, no. 9, p. 1253-
     1259.

Konikow, L. F., and Grove, D. B., 1977,  Derivation of equations describing
     solute transport in ground water*  U.S. Geol. Survey Water-Resources
     Inv. 77-19, 30 p.

Parker, J. C., and Genuchten, M. Th. van, 1984, Flux-averaged and volume-
     averaged concentrations  in continuum approaches to solute transport!
     Water Resources Research, v. 20. no. 7, p. 866-872.  [Also see Comment
     by G. Dagan and E. Bresler, and Reply by authors*  Water Resources
     Research, v. 21, no. 8, p. 1299-1302.]

Sposito, G., and Barry, D. A., 1987, On the Dagan model of solute transport
     in groundwatert   Foundational aspectsi  Water Resources Research, v. 23,
     no. 10. p. 1867-1875.

Whitaker, S., 1967, Diffusion and dispersion in porous media«  Am.
     Inst. Chen. Eng. Jour., v. 13, p. 420-427.

-------
                  HACRODISPERSION AND STOCHASTIC APPROACHES


Black,  T. C., and Freyberg, D. L., 1987, Stochastic modeling of
     vertically averaged concentration uncertainty In a perfectly stratified
     aquifer> Water Resources Research, v. 23, no. 6, p. 997-1004.

Bresler, E., and Dagan, G., 1981, Convective and pore scale dispersive
     solute transport in unsaturated heterogeneous fieldst  Water Resources
     Research, v. 17, no. 6. p. 1683-1693.

Gala, M. A., and Greenkorn. R. A., 1986, Velocity effects on dispersion in
     porous Media with a single heterogeneity!  Water Resources Research,
     v. 22, no. 6, p. 919-926.

Chu, Shu-Yuan, and Sposito, G.. 1980, A derivation of the macroscopic
     solute transport equation for honogeneous, saturated, porous media:
     Water Resources Research, v. 16, no. 3, p. 542-546.

Dagan,   Gedeon,   1986,   Statistical  theory   of   grounduater  flow  and
     transport* Pore to  laboratory,  laboratory to formation,  and formation to
     regional scalet Water Resources Research, v. 22, no. 9, p. 120S-134S.

Davis,  A. D., 1986, Deterministic modeling of  a dispersion  in heterogeneous
     permeable medlat  Ground Water, v. 24, no. 5, p. 609-615.

Gelhar,  L.  W.,  1986,  Stochastic  subsurface hydrology  from  theory to
     applications!  Water Resources Research,  v.  22, no. 9, p. 135S-145S.

Gelhar,  L. W.,  and Axness,  C.  L..  1983.  Three-dimensional  stochastic
     analysis  of  macrodispersion  in  aquiferst  Water  Resources  Research, v.
     19,  no. 1,  p. 161-180.    [Also,  see Comment  by Cushman  and  Reply by
     authors  in v. 19, no. 6, p. 1641-1644.]

Gelhar, L. W., Gutjahr, A. L., and Naff, R. L., 1979, Stochastic analysis
     of macrodispersion in a stratified aquifer:  Water Resources Research,
     v. 15. no. 6, p. 1387-1397.

Gillban, R. W.. Sudicky, E. A., Cherry, J. A., and Frind, E. 0., 1984,
     An advection-diffusion concept for solute transport in heterogeneous
     unconso11dated geological deposltst  Water Resources Research,
     v. 20, no. 3, p. 369-378.   [Also, see Comment by Guven, Holz, and
     Melville and Reply by Sudicky and Gillham, v. 22, no. 1,  p. 89-94, Jan.
     1986.]

Gupta,  V. K., and Bhattacharya, R. N., 1986, Solute dispersion in multi-
     dimensional periodic saturated porous media:  Water Resources Research,
     v. 22, no. 2, p. 156-164.

Guven,  0., Holz, F. J., and Melville, J. G., 1984, An analysis of dispersion
     in a stratified aquifer: Water Resources  Research, v. 20, no. 10, p.
     1337-1354.

-------
Macrodisperslon and Stochastic Approaches, continued


Giiven.O., and  Molz, F.  J.,  1986,  Deterministic and  stochastic  analyses  of
     dispersion in an unbounded stratified porous Medium*  Water Resources
     Research,  v. 22,  no. 11, p. 1565-1574.

Matheron, G., and De Marsily, G., 1980, Is transport in porous media
     always diffusive?  A counterexamples  Water Resources Research,
     v. 16,  no. 5, p.  901-917.

Hercado, A., 1984, A note on micro and macrodispersions  Ground Water, v. 22,
    no. 6, p. 790-791.

Neuman, S. P.,  Winter, C. L., and Newman, C. M.,  1987, Stochastic theory of
    field-scale Fickian dispersion in anlsotropic porous media:  Water
    Resources Research, v. 23, no. 3. p. 453-466.

Pickens, J. F., and Grisak, G. E., 1981, Scale-dependent dispersion in a
     stratified granular aquifer* Water Resources Research, v. 17, no. 4,
     p. 1191-1211.

Pickens, J. F., and Grisak, G. E., 1981, Modeling of scale-dependent
     dispersion in hydrogeologic systems*  Water Resources Research,
     v. 17, no. 6, p.  1701-1711.

Rao, P. V., Rortier, K. M., and Rao, P. S. C., 1981, A stochastic
     approach for describing convective-dispersive  solute'transport In
     saturated porous mediat  Water Resources Research, v. 17, no. 4,
     p. 963-968.

Refsgaard, A.,  1986, Laboratory experiments on solute transport  in
     non-homogeneous porous media*  Nordic Hydrology, v. 17. no. 4/5, p.
     305-314.

Schwartz, F. W., 1977, Macroscopic dispersion in porous media*  The
     controlling factors*  Water Resources Research, v. 13, no. 4,
     p. 743-752.

Silliman, S.  E.,  and  Simpson,  E. S.,  1987,  Laboratory evidence  of the
     scale effect  In  dispersion of solutes in porous  media*   Water Resources
     Research,  v. 23,  no. 8, p. 1667-1673.

Smith, L., and Schwartz, F. W., 1980, Mass transport, 1. A stochastic
     analysis of macroscopic dispersion*  Water Resources Research,
     v. 16,  no. 2, p.  303-313.

Smith, L., and Schwartz, F. W., 1981, Mass transport, 2.  Analysis of
     uncertainty in prediction*  Water Resources Research, v. 17, no. 2,
     p. 351-369.

-------
Macrodlspersion and Stochastic Approaches, continued
Sposito, G., Jury, U. A., and Gupta, V. K., 1986, Fundamental problems in
     the stochastic convection-dispersion model of solute transport in
     aquifers and field soils:  Water Resources Research, v. 22, no. 1,
     p. 77-88.

Tang, D. H., and Finder, G. F., 1979, Analysis of mass transport with
     uncertain physical parameterst  Water Resources Research, v. 15,
     no. 5, p. 1147-1155.

Tang, D. H., Schwartz. F. U., and Smith, L., 1982, Stochastic modeling
     of mass transport in a random velocity field:  Water Resources
     Research, v. 18, no. 2, p. 231-244. [Also see Commentary by Dagan
     and Reply by authors in v. 19, no. 4, p. 1049-1054.]

Ufflnk, G. J. H., 1983, A random walk method for the simulation of
     macrodispersion  in a stratified aquifer:  Relation of Groundwater
     Quantity and Quality (Proceedings of the Hamburg Symposium),  IAHS Publ
     no. 146. p. 103-114.
                                       8

-------
                           DIFFUSION AND DISPERSION
Bachmat, Y., 1967, On the similitude of dispersion phenomena In
     homogeneous and Isotropic porous mediumi  Water Resources Research,
     v. 3, no. 4, p. 1079-1083.

Bachmat, Y., and Bear, J., 1983, The dispersive flux in transport
     phenomena in porous medlai  Adv. Water Resources, v. 6, p. 169-174.

Baker, L. E., 1977, Effects of dispersion and dead-end pore volume in
     •Iscible floodingi  Soc. Petrol. Eng. Jour., v.  7, no. 3, p. 219-227.

Banks, R. B., and Jerasate, S., 1962, Dispersion in unsteady porous
     •edia flow.  ASCE, Jour. Hyd. Div., v. 88, no. HY3, p. 1-21.

Bear, Jacob, 1961, On the tensor form of dispersion in porous media:
     Jour. Geophys. Research, v. 66, no. 4, p. 1185-1198.

Biggar, J. W., and Nielsen, D. R., 1960, Diffusion effects  in miscible
     displacement occurring in saturated and unsaturated porous materials:
     Jour. Geophys. Research, v. 65, no. 9, p. 2887-2896.

Blackuell, R. J., 1962, Laboratory studies of miscroscopic dispersion
     phenomenal  Soc. Petrol. Eng. Jour., v. 2, no. 1, p. 1-8.

Brenner, H., 1961, The diffusion model of longitudinal mixing  in
     beds of finite length.  Numerical values: Chemical Engineering
     Science, v. 17, p. 229-243.

Bresler, E., and Dagan, G., 1979, Solute dispersion in unsaturated
     heterogeneous soil at field scale:  II.  Applications:  Soil Sci.
     Soc. Am. Jour., v. 43, p. 467-472.

Coats, K. H., and Smith, B. D., 1964, Dead-end pore volume and
     dispersion  in porous media:  Soc. Petrol. Eng. Jour., v. 4, no. 1,
     p. 73-84.   [Also see discussions in v. 4, no. 3, p. 282-284.]

Dagan, G., and Bresler, E., 1979, Solute dispersion in unsaturated
     heterogeneous soil at field scale:  I.  Theorys  Soil Scl. Soc.
     Am. Jour., v. 43, p. 461-467.

de Josselin de Jong, G., 1958, Longitudinal and transverse diffusion
     in granular deposits:  Am. Geophys. Union Trans., v. 39, no. 1,
     p. 67-74.

Feenstra, S., Cherry, J. A., Sudicky, E. A., and Haq,  Z., 1984, Matrix
     diffusion effects on contaminant migration from an injection well  in
     fractured sandstone:  Ground Water, v. 22, no. 3, p. 307-316 [also, see
     Discussion  by G.  R.  Walter  and  Reply by  authors,   v.  22,  no.  6,  p.
     786-787].

-------
Diffusion and Dispersion, continued

Fried, J.  J.,  and  Combarnous,  M.  A.,  1971, Dispersion  in porous  media,  in
     Advances in hydrosciencet  New York, Academic Press, v. 7. p. 170-282.

Goltz, H. N., and Roberts, P. V., 1986, Three-dimensional solutions for solute
     transport in an infinite medium with mobile and immobile zonest Water
     Resources Research, v. 22,  no. 7, p. 1139-1148.

Green, T., 1984, Scales for double-diffusive fingering in porous media: Water
     Resources Research, v. 20,  no. 9, p. 1225-1229.

Harleman, D. R. F., Helhor, P. F., and Ruier, R. R., Jr.. 1963,
     Dispersion-permeability correlation in porous mediat  ASCE, Jour.
     Hyd. Div., v. 89,  no. HY2,  p. 67-85.

Harleman, D. R. F., and Rumer, R. R., Jr., 1963, Longitudinal and
     lateral dispersion in an isotroplc porous mediums  Fluid Mechanics
     Jour., v. 16, pt.  3, p. 385-394.

Hoopes, J. A., and Harleman, D.  R. F., 1967, Wastewater recharge and
     dispersion in porous media»  ASCE, Jour. Hyd. Div.. v. 93,
     no. HY5, p. 51-71.

Hunt, B. W., 1973, Dispersion in nonuniform seepage»  ASCE, Jour. Hyd.
     Div., v. 99, no. HY2, p. 293-299.

Ogata, A., 1961, Transverse diffusion in saturated  isotroplc granular
     media«  U.S. Geol. Survey Prof. Paper 411-B, 8 p.

	1964a, The spread of a dye stream in an isotropic granular medium>
     U.S. Geol. Survey Prof. Paper 411-G, 11 p.

	1964b, Mathematics of dispersion with linear adsorption isotherm:
     U.S. Geol. Survey Prof. Paper 411-H, 9 p.

	1970, Theory of dispersion in a granular medium:   U.S. Geol.
     Survey Prof. Paper 411-1, 34 p.

Perkins, T. K., and Johnson, 0. C., 1963, A review of diffusion and
     dispersion in porous media:  Soc. Petrol. Eng. Jour., v. 3,
     no. 1, p. 70-84.

Pozzi, A. L., and Blackwell, R. J., 1963, Design of laboratory models
     for study of miscible displacement:  Soc. Petrol. Eng. Jour.,
     v. 3, no. 1, p. 28-40.

Saffman, P. G., 1959, A theory of dispersion in a porous medium: Jour.
     Fluid Mech., v. 6, p. 321-349.

	1960, Dispersion due to molecular diffusion and macroscopic
     mixing in flow through a network of capillaries:  Jour. Fluid Mech.,
     v. 7, p. 194-208.
                                      10

-------
Diffusion and Dispersion, continued
Scheidegger, A. E., 1961, General theory of dispersion In porous media:
     Jour. Geophys. Research, v. 66, no. 10, p. 3273-3278.

SI 11 loan, S. E., Konikow, L. F., and Voss, C. I., 1987, Laboratory
     Investigation of longitudinal dispersion In anlsotroplc porous media»
     Water Resources Research, v. 23. no. 11, p. 2145-2151.

Simpson, E. S., 1962, Transverse dispersion In liquid flow through
     porous ledla»  U.S. Geol. Survey Prof. Paper 411-C, p. C1-C30.

Skibitzke, H. E., 1964, Extending Darcy's concept of ground-water notion:
     U.S. Geol. Survey Prof. Paper 411-F, 6 p.

Skibitzke, H. E., and Robinson, G. H., 1963, Dispersion of ground water
     flowing through heterogeneous Materialsi  U.S. Geol. Survey Prof.
     Paper 386-B, 3 p.

Smith,  I. M., Farraday, R. V., and O'Connor, B. A., 1973, Raylelgh-Ritz
     and Galerkin-finite elements for diffusion-convection problems:
     Water Resources Research, v. 9, no. 3, p. 593-606.

Taylor, Geoffrey, 1953, Dispersion of soluble Batter in solvent flowing
     slowly through a tube:  Royal Soc. London Proc.. Ser. A, v. 219,
     p. 187-203.

Warren, J. E., and Sklba, F. F., 1964, Macroscopic dispersions Soc.
     Petrol. Eng. Jour., v. 4, no. 3, p. 215-230.
                                      11

-------
                           ACCOUNTING FOR REACTIONS
Amundsen, N. R., 1950, Mathematics of adsorption in beds, lit Jour.
     Phys. Coiloid Chen., v. 54, p. 812-820.

Bahr, J. H., and Rubin, J., 1987, Direct conparlson of kinetic and iocai
     equilibrium formulations for solute transport affected by surface
     reactions: Water Resources Research, v. 23, no. 3, p. 438-452.

Banks, R. B., and All, I., 1964, Dispersion and adsorption in porous
     media flowi  ASCE Jour. Hyd. Dlv., v. 90, no. HY5, p. 13-31.

Borden, R. C., and Bedlent, P. B., 1986, Transport of dissolved hydrocarbons
     influenced by oxygen-1 lilted bI©degradation — 1. Theoretical develop-
     ment!  Water Resources Research, v. 22, no. 13, p. 1973-1982.  [Also
     see Part 2. Field application, same issue, p. 1983-1990.]

Cameron, D. R., and Klute. A., 1977, Convective-dispersive solute
     transport with a combined equilibrium and kinetic adsorption model<
     Water Resources Research, v. 13, no. 1, p. 183-188.

Cederberg, G. A., Street, R. L., and Leckle, J. 0., 1985, A groundwater mass
     transport and equilibrium chemistry model for multlcomponent systems:
     Water Resources Research, v. 21, no. 8, p. 1095-1104.

Charbeneau, R. J., 1981, Grounduater contaminant transport with adsorption
   •  and ion exchange chemistry:  Method of characteristics for the case
     without dispersiom  Water Resources Research, v. 17. no. 3, p. 705-
     713.

Dria, M. A., Bryant, S. L., Schechter, R. S.. and Lake, L. V., 1987,
     Interacting precipitation/dissolution waves:  The movement of inorganic
     contaminants in groundwateri  Water Resources Research, v. 23, no. 11,
     p. 2076-2090.

Fenske, P. R., 1979, Time-dependent sorption on geological materials:
     Jour. Hydrology, v. 43, p. 415-425.

Grove, D. B., and Stollenwerk, K. G., 1985, Modeling the rate-controlled
     sorption of hexavalent chromium!  Water Resources Research, v. 21,
     no. 11, p. 1703-1709.

Grove, D. B., and Stollenwerk, K. G., 1987, Chemical reactions simulated
     by ground-water-quality modelsi  Water Resources Bulletin, v. 23, no. 4,
     p. 601-615.

Gupta, S. P., and Greenkorn. R. A., 1973. Dispersion during flow in
     porous media with bilinear adsorption!  Water Resources Research,
     v. 9, no. 5, p. 1357-1368.
                                      12

-------
Accounting for Reactions, continued
Gupta, S. P., 1974, Determination of dispersion and nonlinear adsorption
     parameters for flow in porous media«  Water Resources Research,
     v. 10, no. 4, p. 839-846.

Hlgglns, G. H., 1972, Sorption in flow through porous media:  In
     Fundamentals of transport phenomena in porous mediat  Elsevier
     Publishing Co., p. 384-392.

Jennings, A. A., 1987, Critical chemical reaction rates for
     multicomponent grounduater contamination models«  Water Resources
     Research, v. 23, no. 9, p. 1775-1784.

Jennings, A. A., Kirkner, D. J., and Theis, T. L., 1982, Multicomponent
     equilibrium chemistry  in~groundwater quality modelss  Water Resources
     Research, v. 18, no. 4, p. 1089-1096.

Lai, S. H., and Jurinak, J. J., 1972, Cation adsorption In one-dimensional
     flow through sol 1st  A numerical solutions  Water Resources
     Research, v. 8, no. 1, p. 99-107.

Lapldus, L., and Amundson, N. R., 1952, Mathematics of adsorption  in
     beds, VIi  Jour. Phys. Chem., v. 56, p. 984-988.

HI Her, C. W., and Benson, L. V., 1983, Simulation of solute transport
     in a chemically reactive heterogeneous systems  Model development
     and applications  Water Resources Research, v. 19, no. 2, p.  381-391.

Palclauskas, V. V., and Domenlco, P. A., 1976, Solution chemistry, mass
     transfer, and the approach to chemical equilibrium in porous
     carbonate rocks and sediments.  Geological Society of America
     Bull., v. 87, p. 207-214.

Reardon, E. J., 1981, Kd's - Can they be used to describe reversible ion
     sorption reactions in contaminant migration?  Ground Water, v. 19,
     no. 3, p. 279-286.

Rubin, Jacob, 1983, Transport of reacting solutes in porous media:
     Relation between mathematical nature of problem formulation
     and chemical nature of reactions:  Water Resources Research,
     v. 19, no. 5, 1231-1252.

Rubin, Jacob, and James, R. V., 1973, Dispersion-affected transport of
     reacting solutes in saturated porous media:  Galerkin method applied
     to equilibrium-controlled exchange in unidirectional steady water
     flows  Water Resources Research, v. 9, no. 5, p. 1332-1356.

Satter, A., Shum, Y. M., Adams, W. T., and Davis, L. A., 1980, Chemical
     transport in porous media with dispersion and rate-controlled
     adsorption:  Soc. Petrol. Eng. Jour., p. 129-138.
                                      13

-------
Accounting for Reactions, continued
Valocchi, A. J., 1984, Describing the transport of ion-exchanging con-
     tan inants using an effective Kd approachi  Water Resources
     Research, v. 20, no. 4, p. 499-503.

Valocchi, A. J., 1985. Validity of the local equilibrium assumption for
     •odeling sorbing solute transport through homogeneous soils:  Water
     Resources Research, v. 21, no. 6, p. 808-820.

Valocchi, A. J., 1986, Effect of radial  flow on deviations  from local
     equilibrium during sorblng solute transport through homogeneous
     sol 1st  Water Resources Research, v. 22, no. 12, p. 1693-1701.

Valocchi, A. J., Street, R. U., and Roberts, P. V., 1981, Transport of
     ion-exchanging solutes in grounduateri  Chromatographic theory
     and field simulation!  Water Resources Research, v. 17, no. 5,
     p. 1517-1527.

Willis, C., and Rubin, J., 1987, Transport of reacting solutes subject to a
     moving dissolution boundaryt  Numerical methods and solutions:  Water
     Resources Research, v. 23, no. 8, p. 1561-1574.
                                      14

-------
              FLOW OF IMMISCIBLE FLUIDS AND MULTIPHASE TRANSPORT
Abriola, L. M.( and Finder, G. F., 1985, A multiphase approach to the
     •odeling of porous media contanination by organic compounds, 1.
     Equation development!  Water Resources Research, v. 21, no. 1,
     p. 11-18  [Also see Part 2, Numerical Simulation:  same issue,
     p. 19-26.]

Allen, H. B., Ill, 1985, Numerical modelling of multiphase flow in porous
     mediai  Adv. Water Resources, v. 8, p. 162-187.

Baehr, A. L., 1987, Selective transport of hydrocarbons in the
     unsaturated zone due to aqueous and vapor phase partitioning:  Water
     Resources Research, v. 23, no. 10, p. 1926-1938.

Baehr, A. L., -and Corapcioglu, M. Y., 1987, A compositional multiphase model
     for groundwater contamination by petroleum products,  2.  Numerical
     solution: Water Resources Research, v. 23, no. 1, p.  201-213.  [Also see
     Part 1. Theoretical considerations, same issue, p. 191-200.]

Crittenden, J. C., Hutzler, N. J., and Geyer, D. G., 1986, Transport of
     organic compounds with saturated groundwater flow:  model development
     and parameter sensitivity:  Water Resources Research, v.  22, no. 3,
     p. 271-284.

Cushman, J. H., 1984, On unifying the concepts of scale, Instrumentation,
     and stochastlcs in the development of multiphase transport theory:
     Water Resources Research, v. 20, no. 11, p. 1668-1676.

Eckberg, D. K., and Sunada, D. K., 1984, Nonsteady three-phase immiscible
     fluid distribution in porous media: Water Resources Research, v. 20,
     no. 12, p. 1891-1897.

Faust, C. R., 1985, Transport of immiscible fluids within and below the
     unsaturated zone:  A numerical model:  Water Resources Research, v. 21,
     no. 4, p. 587-596.

Gupta, S. P., and Greenkorn, R. A., 1974, An experimental  study of
     immiscible displacement with an unfavorable mobility ratio in porous
     media*  Water Resources Research, v. 10, no. 2, p. 371-374.

Hubbert, M. K., 1953, Entrapment of petroleum under hydrodynamic
     conditionsi  Bull. American Assoc. Petroleum Geol., v. 137, no. 8,
     p. 1954-2026.

Kuppusamy, T., Sheng, J., Parker, J. C., and Lenhard, R. J., 1987,
     Finite-element analysis of multiphase Immiscible flow through soils:
     Water Resources Research, v. 23, no. 4. p. 625-631.

Letkeman, J. P., and Ridings, R. L., 1970, A numerical coning model:
     Soc. Petrol. Eng. Jour., v. 10, no. 4, p. 418-424.
                                      15

-------
Flow of Immiscible Fluids and Multiphase Transport, continued


Liu, P. L-F., Cheng, A. H-D., Liggett, J. A., and Lee, J. H., 1981,
     Boundary integral equation solutions to Moving interface between
     two fluids in porous mediai  Water Resources Research, v. 17, no. 5,
     p. 1445-1452.

Osborne, H., and Sykes, J., 1986, Numerical nodeling of immiscible organic
     transport at the Hyde Park Landfillt  Water Resources Research, v. 22,
     no. 1, p. 25-33.

Pinder, G. F., and Abriola, L. H., 1986, On the simulation of nonaqueous phase
     organic compounds in the subsurfacet  Water Resources Research, v. 22,
     no. 9, p. 109S-119S.

Sudicky, E. A., and Frlnd, E. 0., 1982, Contaminant transport in frac-
     tured porous mediat  Analytical solutions for a system of parallel
     fracturesi  Water Resources Research, v. 18, no. 6, p. 1634-1642.

Varnon, J. E., and Greenkorn, R. A., 1970, Nonuniqueness of steady
     state fingering solutions in porous media>  Water Resources Research,
     v. 6, no. 5, p. 1411-1414.
                                      16

-------
                            TRANSPORT IN FRACTURES
Bibby, R., 1981, Mass transport of solutes in dual-porosity media:
     Water Resources Research, v. 17, no. 4, p. 1075-1081.

Grisak, G. E., and Pickens, J. F., 1980, Solute transport through fractured
     media, 1.  The effect of matrix diffusion!  Uater Resources Research,
     v. 16, no. 4, p. 719-730.

Neretnieks, I., 1983, A note on fracture flow dispersion mechanisms in
     the groundi  Water Resources Research, v. 19, no. 2, p. 364-370.

Novakowski, K. S., Evans, G. V., Lever, D. A., and Raven, K. G., 1985, A
     field example of measuring hydrodynamic dispersion in a single
     fracture»  Water Resources Research, v. 21, no. 8, p. 1165-1174.

Schwartz, F. W., Smith, L., and Crowe, A. S., 1983, A stochastic analysis
     of macroscopic dispersion in fractured media«  Water Resources Research,
     v. 19, no. 5, p. 1253-1265.

Smith, L., and Schwartz, F. W., 1984, An analysis of the influence of
     fracture geometry on mass transport in  fractured media:  Water
     Resources Research, v. 20, no. 9, p. 1241-1252.

Tang, D. H., Frlnd, E. 0., and Sudlcky, E. A., 1981, Contaminant transport
     in fractured porous mediai  Analytical solution for a single fracture»
     Water Resources Research, v. 17, no. 3, p. 555-564.
                                      17

-------
                             ANALYTICAL SOLUTIONS

Al-Nlami, A. N. S., and Rushton, K. R., 1979, Dispersion  in stratified
     porous mediat  Analytical solutions!  Water Resources Research,
     v. 15, no. 5, p. 1044-1048.

Eldor, H., and Oagan, G., 1972, Solutions of hydrodynamic dispersion
     in porous mediat  Water Resources Research, v. 8, no. 5,
     p. 1316-1331.

Goltz, M. N., and Roberts, P. V., 1987, Using the method of moments to
     analyze three-dimensional diffusion-limited solute transport from
     temporal and spatial perspectives!  Water Resources Research, v. 23, no.
     8, p. 1575-1585.

Gureghian, A. B., and Jansen, G., 1985, One-dimensional analytical solutions
     for the migration of a three-member radionucllde decay chain in a
     multilayered geologic medium*  Water Resources Research, v. 21, no. 5,
     p. 733-742.

Holly, D. E., and Fenske, P. R., 1968, Transport of dissolved chemical
     contaminants in ground-water systemsi  in  Nevada Test Site,
     Geological Society of America Memoir 110,  p. 171-183.

Hsieh, P. A., 1986, A new formula for the analytical solution of the radial
     dispersion problemi  Water Resources Research, v. 22, no. 11,
     p. 1597-1605.

Huyakorn, P. S., Ungs, M. J.,  Mulkey,  L.  A.,  and Sudicky, E. A., 1987,  A
     three-dimensional  analytical  method  for  predicting  leachate migration!
     Ground Water, v. 25, no. 5, p. 588-598.

Lenau, C. W., 1972, Dispersion from recharge welli  ASCE, Eng. Hech.
     Div. Proc. Pap., v. 98, no. EM 2, p. 331-344.

	1973, Contamination of discharge well from recharge welli  ASCE,
     Hyd. Div. Proc. Pap. no. 9958, v. 99, no. HY8, p. 1247-1263.

Lindstrom, F. T., 1976, Pulsed dispersion of trace chemical concentrations
     in a saturated sorbing porous medlumi  Water Resources Research,
     v. 12. no. 2, p. 229-238.

Marino, M. A., 1974, Models of dispersion in a granular medium! Jour.
     Hydrology, v. 23, p. 313-318.

	1974, Distribution of contaminants in porous media flowi   Water
     Resources Research, v. 10, no. 5, p. 1013-1018.

      L978, Flow against dispersion in nonadsorbing porous media!
     Jour. Hydrology, v. 37, p. 149-158.
                                      18

-------
Analytical Solutions, continued
	1978, Flow against dispersion In adsorbing porous media:   Jour.
     Hydrology, v. 38, p. 197-205.

Moench, A. F., and Ogata, A., 1981, A numerical inversion of the Laplace
     transform solution to radial dispersion in a porous medium:  Water
     Resources Research, v. 17, no. 1, p. 250-252.

Ogata, A., 1976, Two-dimensional steady-state dispersion in a saturated
     porous tediumi  U.S. Geol. Survey Jour, of Research, v. 4, no.  3,
     p. 277-284.

Ogata, A., and Banks, R. B., 1961, A solution of the differential equation
     of longitudinal dispersion in porous media:  U.S. Geol. Survey
     Prof. Paper 411-A, p. A1-A7.

Parlange, J. Y., and Starr, J. L., 1977, Comment on "Analytical
     solution of the equation for transport of reactive solutes through
     soils" by H. M. Selim and R. S. Hansel1<  Uater Resources Research,
     v. 13, no. 3, p. 701.

Selim, H. H., and Hansel 1, R. S., 1976, Analytical solution of the
     equation of transport of reactive solutes through soils:  Uater
     Resources Research, v. 12, no. 3, p. 528-532.

Shamir, U. Y., and Harleman, D. R. F., 1967, Dispersion in layered
     porous media:  ASCE, Jour. Hyd. Div., v. 93, no. HY5, p. 237-260.

Sudicky, E. A., and Frind, E. 0., 1984, Contaminant transport in fractured
     porous mediat  analytical solution for a two-member decay chain
     in a single fracture: Uater Resources Research, v. 20, no. 7, p. 1021-
     1029.

van Genuchten, H. Th., and Alves, W. J., 1982, Analytical solutions
     of the one-dimensional convective-dlspersive solute transport
     equations  U.S. Department of Agriculture, Technical Bull. No.
     1661, 151 p.

Tang, D. H., and Babu, D. K., 1979, Analytical solution of a velocity
     dependent dispersion problem:  Uater Resources Research, v. 15,
     no. 6, p. 1471-1478.
                                      19

-------
                         NUMERICAL METHODS AND MODELS
Ah1strom,  S. V., and Baca, R. G., 1974, Transport model user's manual:
     Battelle Pacific Northwest Laboratories rept. BNUL-1716, 25 p.

Burnett, R. D., and Frind, E. 0., 1987, Simulation of contaminant
     transport in three dimensions, 2, Dimensionality effects:  Water
     Resources Research, v. 23, no. 4, p. 695-705.

Cheng, R.  T., and Hodge, D. S., 1976, Finite-element method in modeling
     geologic transport processes!  Mathematical Geology, v. 8, no. 1,
     p. 43-56.

Chatwal, S. S., Cox, R. L., Green, D. U., and Ghandi, B., 1973,
     Experimental and mathematical modeling of liquid-liquid miscible
     displacement in porous mediat  Water Resources Research, v. 9,
     no. 5, p. 1369-1377.

Domenico,  P. A., and Robbins, G. A., 1984, A dispersion scale effect  in
     model calibrations and field tracer experiments!  Jour. Hydrology,
     v. 70, p. 123-132.

Frind, E.  0., and Hatanga, G. B., 1985, The dual formulation of flow  for
     contaminant transport modeling, 1. Review of theory and accuracy
     aspectsi  Water Resources Research, v. 21, no. 2, p. 159-169.

Garder, A. 0., Peaceman, D. W., and Pozzi, A. L., Jr., 1964, Numerical
     calculation of multidimensional miscible displacement by the method
     of characteristics!  Soc. Petrol. Eng. Jour., v. 4, no. 1, p. 26-36.

Gray, W. G., and Pinder, G. F., 1976, An analysis of the numerical solution
     of the transport equation:  Water Resources Research, v. 12, no. 3,
     p. 547-555.

Grove, D.  B., 1977, The use of Galerkin finite-element methods to solve
     mass-transport equations:  U.S. Geol. Survey Water-Resources Inv. 77-49,
     55 p.

Grove, D.  B., and Stollenwerk, K. G., 1984, Computer model of one-dimensional
     equilibrium controlled sorptlon processes:  U.S. Geol. Survey Water-
     Resources Inv. 84-4059, 58 p.

Guymon, G. L., Scott, V. H., and Herrman, L. R., 1970, A general
     numerical solution of the two-dimensional diffusion-convection
     equation by the finite-element method!  Water Resources Research,
     v. 6, no. 6, p. 1611-1617.

Huyakorn,  P.S., Andersen. P. F., Mercer, J. W., and White, H. 0., Jr., 1987,
     Saltwater intrusion in aquifers: Development and  testing  of a
     three-dimensional finite element model: Water Resources Research, v. 2,
     no. 2, p. 293-312.
                                      20

-------
Numerical Methods and Models, continued
Huyakorn, P. S., Jones, B. G., and Andersen, P. P., 1986, Finite element
     algorithms for simulating three-dimensional groundwater flow flow and
     solute transport in multilayer systems*  Water Resources Research, v. 22,
     no. 3, p. 361-374.

Huyakorn, P. S., Lester, B. H., and Mercer, J. W., 1983, An efficient finite
     element technique for modeling transport In fractured porous media,
     1.  Single species transport!  Water Resources Research, v. 19,
     no. 3, p. 841-854.

INTERA Environmental Consultants, Inc., 1979, Revision of the documentation
     for a model for calculating effects of liquid waste disposal in deep
     saline aquiferst  U.S. Geol. Survey Water-Resources Inv. 79-96, 73 p.

INTERCOM? Resource Development and Engineering Inc., 1976, A model for
     calculating effects of liquid waste disposal in deep saline aquifers:
     U.S. Geol. Survey Water-Resources Inv. 76-61.

Javandel, I., Doughty, C., and Tsang, C. P., 1984, Groundwater transport:
     Handbook of mathematical models:  American Geophysical Union, Water
     Resources Monograph 10, 228 p.

Khaleel, R., and Reddell, D. L., 1986, MX solutions of convective-
     dispersion problems:  Ground Water, v. 24, no. 6, p. 798-807.

Konlkow, L. P., and Bredehoeft, J. D., 1978, Computer model of two
     dimensional solute transport and dispersion  in ground water:  U.S.
     Geol. Survey Techniques of Water-Resources Inv., Book 7. Chap. C2,
     90 p.

Lantz, R. B.. 1970, Rigorous calculation of miscible displacement using
     immiscible reservoir simulators:  Soc. Petrol. Eng. Jour., v. 10,
     no. 2, p. 192-203.

Mercer, J. W., Larson, S. P., and Paust, C. R., 1980, Finite-difference
     model to simulate the areal flow of saltwater and freshwater
     separated by an  interface:  U.S. Geol. Survey Open-File Report
     80-407, 88 p.

Nalluswami, M., Longenbaugh, R. A., and Sunada, D. K., 1972, Finite
     element method for the hydrodynamic dispersion equation with
     •ixed partial derivatives:  Water Resources Research, v. 8, no. 3,
     p. 1247-1250.

Naymik,  T.  G., 1987,  Mathematical  modeling  of solute transport  in the
     subsurface:  Critical Reviews  in  Environmental  Control,  v.  17,  no. 3, p.
     229-251.
                                      21

-------
Numerical Methods and Models, continued
Peaceman, D. U., and Rachford, H. H., Jr., 1962, Numerical calculation of
     multidimensional nisclble displacement:  Soc. Petrol. Eng. Jour., v. 2,
     no. 4, p. 327-339.

Plckens, J. F., Gillhan, R. W., and Cameron, D. R., 1979, Finite-element
     analysis of the transport of water and solutes in tile-drained soils:
     Jour. Hydrology, v. 40, no. 2, p. 243-264.

Pickens, J. F., and Lennox, U. C., 1976, Numerical simulation of waste
     movement in steady groundwater flow systems«  Water Resources
     Research, v. 12, no. 2, p. 171-180.

Pinder, G. F., and Cooper, H. H., Jr., 1970, A numerical technique
     for calculating the transient position of the saltwater front:  Water
     Resources Research, v. 6, no. 3, p. 875-882.

Pinder, G. F., and Shapiro, A., 1979, A new collocation method for the
     solution of the convection-dominated transport equation:  Water
     Resources Research, v. 15, no. 5, p. 1177-1182.

Price, H. S., Cavendish, J. C., and Varga, R. S., 1968, Numerical methods
     of higher-order accuracy for diffusion-convection equations:  Soc.
     Petrol. Eng. Jour., v. 8, p. 293-303.

Prickett, T. A., Nayilk, T. G., and Lonnquist, C. G., 1981. A "Random-
     Walk" solute transport model for selected groundwater quality
     evaluations:  Illinois State Water Survey, Bull. 65, 103 p.

Reeves, H., and Cranwell, R. M., 1981, User's manual for the Sandia
     waste-isolation flow and transport model  (SWIFT) Release 4.81:
     Sandia National Lab., NUREG/CR-2324, U.S. Nuclear Regulatory
     Commission, 145 p.

Reeves, H.. Ward, D. S., Davis, P. A., and Bonano, E. J., 1986, SWIFT II
     self-teaching curriculum:  Illustrative problems for the Sandia
     waste-isolation flow and transport model for fractured media:  Sandia
     National Laboratories, NUREG/CR-3925, U.S. "Nuclear Regulatory Commission,
     96 p.

Reddell, 0. L., and Sunada, D. K., 1970, Numerical simulation of dispersion
     in groundwater aquifers:  Colorado State Univ. Hydrology Paper 41,
     79 p.

Settarl, A., Price, H. S., and Dupont, T., 1977, Development and appli-
     cation of variational methods for simulation of misclble displace-
     ment  in porous media:  Soc. Petrol. Eng. Jour., v. 17, no. 3, p. 228-246.

Segol, G., Pinder G. F., and Grey, W. G., 1975, A Galerkin-finite element
     technique for calculating the transient position of the saltwater
     front:  Water Resources Research, v. 11, no. 2, p. 343-347.
                                      22

-------
Numerical Methods and Models, continued
Shamir, U. Y., and Harleman, D. R. F., 1967, Numerical solutions for
     dispersion in porous mediai  Water Resources Research, v. 3, no. 2,
     p. 557-581.

Tagamets, T., and Sternberg, Y. M., 1974, A predictor-corrector method
     for solving the convection-dispersion equation for adsorption in
     porous mediai  Water Resources Research, v. 10, no. 5, p. 1003-1011.

van Genuchten, M. T., Finder, G. F., and Frind, E. 0., 1977, Simulation of
     two-dimensional contaminant transport with Isoparametric hermitian
     finite elements!  Water Resources Research, v. 13, no. 2, p. 451-458.

Voss, C. I.. 1984. SUTRA — Saturated Unsaturated Transport—
     A finite-element simulation model for saturated-unsaturated
     fluid-density-dependent ground-water flow with energy transport
     or chemically-reactive single-species solute transport!  U.S. Geol.
     Survey Water-Resources Invest. Report 84-4369.
                                      23

-------
                      PARAMETER DETERMINATION AND TRACERS
Davis, S. N., Thompson, G. H., Bentley, H. U., and Stiles, G.,
     1980, Ground-water tracers - A short reviews  Ground Water,
     v. 18, no. 1, p. 14-23.

Caspar, £., and Oncescu, H., 1972, Radioactive tracers In hydrology:
     Elsevler Publishing Co., New York, 342 p.

Grove, 0. B., and Beeten, U. A., 1971, Porosity and dispersion constant
     calculations for a fractured carbonate aquifer using the two-well
     tracer Methods  Water Resources Research, v. 7, no. 1, p. 128-134.

Grove, D. B., Beeten, W. A., and Sower, F. B., 1970, Fluid travel time
     between a recharging and discharging well pair in an aquifer having
     a uniform regional flow fields  Water Resources Research, v. 6,
     no. 5, p. 1404-1410.

Gupta, S. P., and Greenkorn, R. A., 1974, Determination of dispersion
     and nonlinear adsorption parameters for flow in porous medias  Water
     Resources Research, v. 10, no. 4, p. 839-846.

Guvanasen, V., and Guvanasen, V. H., 1987, An approximate seminalytical
     solution for tracer injection tests in a confined aquifer with a radially
     converging flow field and finite volume of tracer and chase fluid:  Water
     Resources Research, v. 23, no. 8, p. 1607-1619.

Guven. 0., Falta, R. W., Holz, F. J., and Melville, J. G., 1985, Analysis
     and interpretation of single-well tracer tests in stratified aquifers:
     Water Resources Research, v. 21, no. 5, p. 676-684.

Hoehn, E., and Santschi, P. H., 1987, Interpretation of tracer displacement
     during  infiltration  of river  water  to groundwatert   Water  Resources
     Research, v. 23, no. 4, p. 633-640.

Huyakorn, P. S., Andersen, P. F., Molz, F. J., Guven,  0., and Melville, J. G.,
     1986, Simulations of two-well tracer tests in stratified aquifers at the
     Chalk River and the Mobile sites:  Water Resources Research, v. 22, no.
     7, p. 1016-1030.

Knopman, D. W., and Voss, C. I., 1987, Behavior of sensitivities in the
     one-dimensional advection-dispersion equations Implications for parameter
     estimation and sampling designs  Water Resources Research, v. 23, no. 2,
     p. 253-272.

Kreft, A., and Zuber, A., 1979, Determination of aquifer parameters by a
     two-well pulsed method using radioactive tracers—Commentss Jour.
     Hydrology, v. 41, no. 2, p. 171-176.
                                      24

-------
Parameter Detemination and Tracers, continued

Mackay, D. M., Freyberg, D. L., Roberts, P. V., and Cherry, J. A., 1986,
     A natural gradient experiment on solute transport In a sand aquifer —
     1. Approach and overview of plume movement:  Water Resources Research,
     v. 22, no. 13, p. 2017-2029.   [Also see Parts 2-5, same Issue.]

Holz, F. J., Giiven, 0., Melville, J. G., Crocker, R. D., and Hatteson, K. T.,
     1986, Performance, analysis, and simulation of a two-well tracer test at
     the Mobile sltei  Water Resources Research, v. 22, no. 7, p. 1031-1037.

Molz, F. J., Melville, J. G., Gliven, 0., Crocker, R. D., and Matteson, K.
     T., 1985, Design and performance of single-well tracer tests at the
     Mobile sitei Water Resources Research, v. 21, no. 10, p. 1497-1502.

Murty, V. V. N., and Scott, Vr H.,  1977. Determination of transport model
     parameters In groundwater aquifersi  Water Resources Research, v. 13, no.
     6, p. 941-947.

Naymik, T. G., and Sievers, M. E.,  1985, Characterization of dye tracer
     plumes: In situ field experiments!  Ground Water, v. 23, no. 6,
     p. 746-752.

Ogata, Akio, 1963, Effect of the injection scheme on the spread of
     tracers In ground-water reservoirs:  U.S. Geol. Survey Prof. Paper 475-B,
     p. B199-B202.

Pickens, J. F., Jackson, R. E., Inch, K. J., and Merritt, W. F., 1981,
     Measurement of distribution coefficients using a radial injection
     dual-tracer test:  Water Resources Research, v. 17, no. 3, p. 529- 544.

Rainwater, K. A.. Wise, W. R., and Charbeneau, R. J., 1987, Parameter
     estimation through groundwater tracer tests>  Water Resources Research,
     v. 23, no. 10, p. 1901-1910.

Sauty, J. P., 1980. An analysis of hydrodispersive transfer in aquifers:
     Water Resources Research, v. 16, no. 1, p. 145-158.

Schwarzenbach, R. P., and Westall, J., 1981. Transport of nonpolar
     organic compounds from surface water to groundwater.  Laboratory sorption
     studies* Environmental Science & Technology, v. 15, no. 11,
     p. 1360-1367.

Smart, P. L.. and Laidlaw, I. M. S., 1977, An evaluation of some
     fluorescent dyes for water tracing:  Water Resources Research, v. 13, no.
     1. p. 15-33.

Strecker. E. W., and Chu, Wen-sen,  1986, Parameter identification of a
     ground-water contaminant transport model:  Ground Water, v. 24,
     no. 1, p. 56-62.
                                      25

-------
Parameter Detent Ination and Tracers, continued
Sudicky, E. A., Gillhan, R. V., and Frind, E. 0., 1985, Experimental
     investigation of soiute transport in stratified porous media—1. The
     nonreactive cases  Water Resources Research, v. 21, no. 7, p. 1035-
     1041.

Umarl,  A., Uiliis, R., and Liu, P. L. F., 1979, Identification of aquifer
     dispersivitles in two-dimensional transient groundwater contaminant
     transports An optimization approachi  Water Resources Research, v. 15,
     no. 4, p. 815-831.

Wagner, B. J., and Corelick, S. H., 1986, A statistical methodology for
     estimating transport parameterss  Theory and applications to one-
     dimensional advective-disperslve systemss  Water Resources Research,
     v. 22. no. 8, p. 1303-1315.

Webster, D. S., Proctor, J. F., and Marine, I. W., 1970, Two-well
     tracer test in fractured crystalline rock:  U.S. Geol. Survey Water
     Supply Paper 1544-1, 22 p.

Wiebenga, W. A., Ellis, W. R., Seatonberry, B. W., and Andrew, J. T. G.,
     1967, Radioisotopes as groundwater tracerst  Jour. Geophys. Research, v.
     72, no. 16, p. 4081-4091.

Wood, W. W., 1981, A geochemical method of determining dispersivity in
     regional ground-water systemst  Jour. Hydrology, v. 54, p. 209-224.

Wood, W. W., and Ehrlich, G. G., 1978, Use of Baker's yeast to trace
     mlcrobial movements in ground waters  Ground Water, v. 16, no. 6,
     p. 398-403.
                                      26

-------
                          ANALYSIS OF FIELD PROBLEMS
Barcelona, M. J., and Naynik, T. G., 1984, Dynamics of a fertilizer
     contaminant plume in grounduaters  Reprinted from Environmental Science
     & Technology, v. 18, p. 257.

Bennett, G. D., Mundorff, M. J., and Hussaln, S. A., 1968, Electric
     analog studies of brine coning beneath fresh-water wells in the Punjab
     Region, West Pakistani  U.S. Geol. Survey Water-Supply Paper 1068-J, 31
     P-

Boomer, P. M., and Schechter, R. S., 1979, Mathematical modeling of
     in-situ uranium leaching:  Soc. Petrol. Eng. Jour., p. 393-400.

Brown, D. L., and Silvey, W. D., 1977, Artificial recharge to a fresh
     water-sensitive brackish-water sand aquifer, Norfolk, Virginia: U.S.
     Geol. Survey Prof. Paper 939, 53 p.

Freeberg, K. M., Bedient, P. B., Connor, J. A., 1987, Modeling of TCE
     contamination and recovery in a shallow sand aquifers  Ground Water,
     v. 25, no. 1, p. 70-80.

Goltz, M. N., and Roberts, P. V., 1986, Interpreting organic solute
     transport data from a field experiment using physical nonequillbrium
     •odelsi  Journal of Contaminant Hydrology, v. 1. no. 1/2, p. 77-93.

Grove, D. B., and Wood, W. W., 1979, Prediction and field verification of
     subsurface-water quality changes during artificial recharge, Lubbock,
     Texasi Ground Water, v. 17, no. 3, p. 250-257.

Helweg, 0. J., and Labadie, J. W., 1977, Linked models for managing river
     basin salt balance:  Water Resources Research, v. 13, no. 2, p. 329-336.

Jackson, R. E.. and Inch, K. J., 1980, Hydrogeochemical processes
     affecting the migration of radlonuclldes  in a fluvial sand aquifer at the
     Chalk River Nuclear Laboratories:  Environment Canada, Inland Waters
     Directorate, NHRI Paper No. 7, 58 p.

Jackson, R. E., and others, 1985, Contaminant hydrogeology of toxic organic
     chemicals at a disposal site, Gloucester, Ontario:  1.  Chemical
     concepts and site assessment:  Environment Canada, Inland Waters
     Directorate, NHRI Paper No. 23, 114 p.

Kipp, K. L., Jr., Stollenwerk, K. G., and Grove, D. B., 1986, Groundwater
     transport of strontium 90 in a glacial outwash environment:  Water
     Resources Research, v. 22, no. 4, p. 519-530.

Konikow, L. F., 1977, Modeling chloride movement in the alluvial aquifer
     at the Rocky Mountain Arsenal, Colorado:  U.S. Geol. Survey Water Supply
     Paper 2044, 43 p.
                                      27

-------
Analysis of Field Probleis, continued
Konikow, L. F., and Bredehoeft, J. D., 1974, Modeling flow and chemical
     quality changes in an irrigated stream-aquifer system:  Water Resources
     Research, v. 10, no. 3.  p. 546-562.

Konikow, L. F., and Person, H. A., 1985, Assessment of long-term salinity
     changes in an irrigated stream-aquifer system:  Water Resources Research,
     v. 21. no. 11, p. 1611-1624.

LeBlanc, D. R., 1984, Sewage plume in a sand and gravel aquifer, Cape Cod,
     Massachusettsi  U.S. Geol. Survey Water-Supply Paper 2218, 28 p.

Lewis, B. D., and Goldstein.  F. J.. 1982, Evaluation of a predictive
     ground-water solute-transport model at the Idaho National Engineering
     Laboratory, Idahot  U.S. Geol. Survey Water-Resources Investigations
     82-25. 71 p.

MacFarlane, D. S., Cherry, J. A., Egboka, B. C. E., Sudicky, E. A., Dance, J.
     T., Nicholson, R. V., Greenhouse, J. P., and others, 1983, Migration of
     contaminants in groundwater at a landfill:  A case study  (in 7 parts:  1.
     Groundwater flow and plume delineation^ 2. Groundwater monitoring
     devices* 3. Tritium as an indicator of dispersion and recharge; 4. A
     natural-gradient dispersion test) 5. Cation migration in the dispersion
     testi  6. Hydrogeochemistryt and 7. DC, VLF, and Inductive resistivity
     surveys):  Jour. Hydrology, v. 63, p. 1-197.

Mattraw, H. C., Jr.,  and Franks, B. J. [eds.], 1986, Movement and fate of
     creosote waste in ground water, Pensacola, Florida:  U.S. Geological
     Survey Toxic Waste-Ground-Water Contamination Program:  U.S. Geological
     Survey Water-Supply Paper 2285, 63 p.

Mercado, A., 1976, Nitrate and chloride pollution of aquifers:  A regional
     study with the aid of a single-cell model:  Water Resources Research, v.
     12, no. 4, p. 731-747.

Mundorff, M. J., Carrigan. P. H.. Jr., Steele, T. 0., and Randall, A. D.,
     1976,  Hydrologlc evaluation of salinity control and reclamation projects
     in the Indus Plain, Pakistan—A Summary:  U.S. Geol. Survey Water-Supply
     Paper 1608-0, 59 p.

Pankow, J.  F., Johnson, R. L., Hewetson, J. P., and Cherry, J. A., 1986,
     An evaluation of contaminant migration patterns at two waste disposal
     sites on fractured porous media in terms of the equivalent porous
     medium  (EPM) model:  Journal of Contaminant Hydrology, v. 1, no. 1/2,
     p. 65-76.

Pinder, G.  F., 1973,  A Galerkin-finite element simulation of ground-water
     contamination on Long Island, New York:  Water Resources  Research, v. 9,
     no. 6, p. 1657-1669.
                                      28

-------
Analysis of Field Problems, continued

Robertson, J. B., 1974, Digital nodeling of radioactive and chemical waste
     transport in the Snake River Plain aquifer at the National Reactor
     Testing Station, Idaho*  U.S. Geol. Survey Open-File Rept. IDO-22054, 41
     P-

	1977, Numerical modeling of subsurface radioactive solute transport from
     waste-seepage ponds at the Idaho National Engineering Laboratory! U.S.
     Geol. Survey Open-File Rept. 76-717, 68 p.

Robson, S. G., 1974, Feasibility of digital water-quality modeling illustrated
     by application at Barstow, California!  U.S. Geol. Survey Water-Resources
     Inv. 46-73, 66 p.

Robson, S. G., and Saulnier, G. J., Jr., 1980, Hydrogeochemistry and simulated
     solute transport, Plceance Basin, Northwestern Colorado: U.S. Geol.
     Survey Open-File Rept. 80-72, 89 p.

Segol, G., and Finder, G. F., 1976, Transient simulation of saltwater
     Intrusion in southeastern Florida:  Water Resources Research, v. 12, no.
     1, p. 65-70.

Souza, W. R., and Voss, C. I., 1987, Analysis of an anlsotropic coastal
     aquifer system using variable-density flow and solute transport
     simulation: Journal of Hydrology, v. 92, p. 17-41.

Thurman, E.  M.,  Barber, L.  B.,  Jr., and  LeBlanc,  D. R.,  1986,  Movement and
     fate  of  detergents   in  groundwatert   A  field   studyt    Journal  of
     Contaminant Hydrology, v. 1, no. 1/2,  p. 143-161.

Valocchi, A. J., Roberts, P. V., Parks, G.  A., and Street, R. L., 1981,
     Simulation of the transport of ion-exchanging solutes using laboratory
     determined chemical parameter values:   Ground Water, v. 19, no. 6, p.
     600-607.
                                      29

-------
                  AQUIFER RECLAMATION AND MANAGEMENT ASPECTS

Andersen, P. F., Faust, C. R., and Mercer, J. W., 1984, Analysis of conceptual
     designs   for   remedial   measures   at   Liparl   Landfill,   New   Jersey:
     Ground Water,  v.  22,  no.  2,  p.  176-190.

Gorelick, S. M., Evans, B., and Reason, I., 1983, Identifying sources of
     grounduater pollution:  An optimization approach: Water Resources
     Research, v. 19, no. 3, p. 779-790.

CorelIck, S. M., Reason, I., and Cottle, R. W., 1979, Management model of
     groundwater system with a transient pollutant source»  Water Resources
     Research, v. 15, no. 5, p. 1243-1249.

Corelick, S. M., Voss, C. I., and others, 1984, Aquifer reclamation design:
     The use of contaminant transport simulation combined with nonlinear
     programming:  Water Resources Research, v. 20, no. 4, p. 415-427.

Javandel, I., and Tsang, Chin-Fu, 1986, Capture-zone type curves:  A tool
     for aquifer cleanup:  Ground Water, v. 24, no. 5, p. 616-625.

Kaunas, J. R., and Halines, Y. Y., 1985, Risk management of groundwater
     contamination in a multiobjective  framework:  Water Resources Research,
     v. 21, no. 11, p. 1721-1730.

Massmann, J., and Freeze, R. A., 1987, Groundwater contamination from waste
     management sites: The Interaction between risk-based engineering design
     and regulatory policy, 1, Methodology:  Water Resources Research, v.  23,
     no. 2, p. 351-367.   [Also see Part 2, Results: same issue, p. 368-380.)

Herritt, M.  L.,  1986,  Recovering fresh water stored in saline limestone
     aquifers:  Ground Water, v. 24, no. 4, p. 516-529.

Sharefkin, M., Shechter, M., and Kneese, A., 1984, Impacts, costs, and
     techniques for mitigation of contaminated groundwater:  A review: Water
     Resources Research, v. 20, no. 12. p. 1771-1783.  [Also see
     Comment by R. S. Main, v. 22, no. 3, p. 429-430.]

Wagner, B. J., and Corelick, S. M., 1987, Optimal groundwater quality
     management under parameter uncertainty:  Water Resources Research, v. 23,
     no. 7, p. 1162-1174.

Ward, D. S., Buss, D. R., Mercer, J. W., and Hughes, S. S., 1987, Evaluation
     of a groundwater corrective action at the Chem-Dyne hazardous waste site
     using a telescopic mesh refinement modeling approach:  Water Resources
     Research, v. 23, no. 4, p. 603-617.

Willis, R., 1979, A planning model for the management of groundwater quality:
     Water Resources Research, v. 15, no. 6, p. 1305-1312.
                                      30

-------
       SECTION 5






PERSPECTIVE ON MODELING

-------
 Proceedings of the NATO Advanced Research Workshop on
 Advances in Analytical and Numerical Groundwaler Flow and Quality Modelling
 Lisbon. Portugal
 June 2-6.1987

 Library of Congress Cataloging in Publication Data
 NATO Advanced Workshop on Advances in Analytical and Numerical Groundwater Flow and
 Quality Modelling (1987: Lisbon, Portugal)
    Groundwater flow and quality modelling / edited by E. Custodio, A. Gurgui. J. P. Lobo
 Ferreira.
    p.    cm. — (NATO ASI series. Series C, Mathematical and physical sciences; vol. 224)
    "Proceedings of the NATO Advanced Research Workshop on Advances in Analytical and
 Numerical Groundwater Flow and Quality Modelling, Lisbon, Portugal, June 2-6,1987"—T.p.
 verso.
    "Published in cooperation with NATO Scientific Affairs Division."
    Includes index.
    ISBN 90-277-2655-8
    1.  Groundwater flow—Mathematical models—Congresses.  2.  Water, Underground
 —Quality—Mathematical   models—Congresses.   I.  Custodio,   Emilio.  II. Gurgui.  A.
 (Antonio). 1953-   . III.  Ferreira. J. P. Lobo (Joao Paulo Lobo). 1952-
 IV.  North Atlantic Treaty Organization. Scientific Affairs Division.  V.  Title.  VI.  Series:
 NATO ASI series. Series C, Mathematical and physical sciences; no. 224.
 TC176.N386  1987
 628.ru — dc19                                                   87-30968
                                                                     CIP
 Groundwater Flow  and
 Quality  Modelling
 edited by

 E. Custodio
 Polytechnic University of Catalonia (ETSICCP/DIT) and
 International Course on Groundwater. Barcelona. Spain

 A. Gurgui
 Polytechnic University of Catalonia (ETSIIB/DHT) and
 International Course on Groundwater. Barcelona. Spain
and
J. P. Lobo Ferreira
Civil Engineering National Laboratory.
Lisbon. Portugal
Published by D. Reidel Publishing Company
P O. Box 17.3300 AA Dordrecht. Holland

Sold and distributed in the U.S.A. and Canada
by Kluwer Academic Publishers.
101 Philip Drive. Norwell. MA 02061. U.S.A.

In all other countries, sold and distributed
by Kluwer Academic Publishers Group.
PO Box 322.3300 AH Dordrecht. Holland

D. Reidel Publishing Company is a member of the Kluwer Academic Publishers Group
All Rights Reserved
* 1988 by D. Reidel Publishing Company, Dordrecht. Holland.
No part ol the material protected by this copyright notice may be reproduced or utilized
in any lorm or by any means, electronic or mechanical, including photocopying, recording
or by any information storage and retrieval system, without written permission (rom the
copyright owner.

Printed in The Netherlands
D. Reidel Publishing Company

Dordrecht / Boston / Lancaster / Tokyo

Published in cooperalion wilh NATO Scientific Aflairs Division

-------
 PRESENT LIMITATIONS AND PERSPECTIVES ON  MODELING
 POLLUTION PROBLEMS IN  AQUIFERS
            Leonard  F. Konikow
            U.S. Geological Survey
            431 National Center
            12201 Sunrise Valley  Drive
            Reston, VA 22092  U.S.A.
 ABSTRACT.  In recent years there  has  been an explosive increase in the
 DSC of deterministic, distributcd-parameter, ground-water simulation models
 for analyzing contaminant transport in ground-water systems.   Predictive
 errors arising strictly from  inaccuracies in the equation-solving  algorithm
 • are usually much smaller than the predictive errors associated  with or
 produced by:   (1) theoretical misconceptions or overidealizations about the
 system that are incorporated into the  model, (2) uncertainty and error in
 the specification of system  properties, boundary conditions, and initial
 conditions, and (3) uncertainty in future stresses.   The next major level  of
 improvement  in ground-water simulation  models will not arise  from
 improved numerical procedures; rather, a greater investment must  be made
 towards  obtaining more accurate descriptions of aquifer properties and
 their variability.  It is especially critical for transport models  that
 variability in the permeability and  porosity fields  be defined  as accurately
 and precisely as possible.  The tradeoff between accuracy and  cost  will
 always be a difficult one to resolve, but will always have to  be  made for
 both  model development and data collection.


 I.  INTRODUCTION
<.
i      In  recent  years  there  has been an explosive  increase in the  use  of
; deterministic, distributed-parameter,  ground-water simulation models  for
1 analyzing contaminant transport  in  ground-water systems and for  predicting
i lystem responses to  changes in stresses.  Because many people  who are
" using ground-water  models or relying  on their  results are  not  fully  aware
 °f  the assumptions and idealizations that have been  incorporated  into them,
 'here is  a danger that some  may  infer that the accuracy of such  a
 Prediction is  somehow coincident  with  the  numerical accuracy of  the
 Wathematical  solution to the governing equations.   This assumption is
 obviously invalid; the  numerical  errors arising  strictly from inaccuracies  in
 'he equation-solving algorithm are usually much smaller than the  predictive
 
-------
644

conditions, and  initial conditions, and  (3) uncertainty in future stresses.
This false sense of  accuracy  and unjustified confidence may arise from  the
quantitative  nature  of numerical  simulation, and the uninitiated may  tend
to confuse quantitativeness and precision  with reliability and accuracy.

     The  purpose  of  this  paper  is  to  review  some   of  the practical
limitations   on   modeling   complex   ground-water   pollution   problems.
Theoretical  shortcomings  and   data   constraints  are   summarized  and
contrasted, and  a  case history is  presented  in which the predictive accuracy
of a solute-transport model is analyzed.
2.   GROUND-WATER  FLOW

      A general form of  the equation  describing the transient flow  of a
slightly compressible fluid in a nonhomogeneous anisotropic aquifer may be
derived by combining Darcy's Law with the continuity equation (for
detailed derivations see Bear, 1979; or Freeze and  Cherry,  1979).  A
general ground-water flow equation may be  written in Cartesian tensor
notation  as
                   a      ai-       3h
                              - S,— + W*                         (1)
where  K-,: is  the hydraulic conductivity tensor, LT~'; Ss is  the  specific
storage, L"1; h is  the hydraulic  head, L; W* = W*(x,y,z,t) is  the  volume
flux per unit volume (positive sign for outflow and negative for inflow),

T"'; x; are  the Cartesian coordinates.  L; and  t  is time, T.   The
summation convention of Cartesian tensor  analysis is implied in equation I.
That is, each term is summed over the range of its  subscripts.

     The dependent  variable in eq.  1  is the hydraulic head.  However, in
cases  where  fluid properties,  such as density or viscosity, vary  significantly
in time or space because of changes in pressure, temperature, or chemical
composition, the  fluid is nonhomogeneous, and the relations  among  water
levels, heads, pressures,  and fluid velocities are less  straightforward.
Calculations  of flow rates and  directions then require pressure, density,
and elevation data, instead of just head measurements.  Davies (1987)  shows
that using the concept of equivalent  freshwater head as a basis for
analyzing and modeling areal ground-water flow in  a variable-density
system can  lead to significant errors in predicted  flow  directions and
velocity magnitudes.

     In some ground-water studies it can  be  reasonably assumed that
ground-water flow is two-dimensional.  This allows the three-dimensional
flow equation to be  reduced  to  the  case of two-dimensional  areal or
cross-sectional  flow,  for which  several additional simplifications are
possible.   The advantages of  reducing the  dimensionality of  the equations
include less stringent data requirements and simpler, more efficient
                                                                                                                                                                    645
                                                                                        mathematical or numerical solutions.   The disadvantage  is the risk  of
                                                                                      '  losing meaningful details  of the system  and inducing compensating errors
                                                                                        during the model  calibration procedure.   If  there are minor components of
                                                                                        flow into or out of the plane of the  two-dimensional model, they may
                                                                                        have a much greater impact on  the solute concentration field  than on the
                                                                                      .  head distribution.

                                                                                             Because contaminant transport in ground  water  is  strongly dependent
                                                                                        oo ground-water  flow,  it  is  often  feasible to use a ground-water flow
                                                                                        model to analyze  directions  of flow and transport,  as well  as  travel times.
                                                                                        An  illustrative  example is the analysis at the Love Canal area,  Niagara
                                                                                        Falls, New York,  described by Mercer and  others (1983). Faced with
                                                                                        inadequate and uncertain  data describing the system,  they used  Monte
                                                                                        Carlo simulation  and uncertainty analysis to estimate a  range  of travel
                                                                                      •  times (and their associated probabilities)  from the contaminant source  area
                                                                                        to the Niagara  River.
                                                                                        3.  SOLUTE  TRANSPORT IN GROUND WATER

                                                                                             The  purpose of a model  that simulates solute transport in ground
                                                                                        water is to compute the concentration of a dissolved  chemical  species  in an
                                                                                        aquifer at any specified place and time.   Because  convective transport and
                                                                                        hydrodynamic dispersion depend on the  velocity of ground-water  flow, the
                                                                                        mathematical simulation model  must  solve at  least  two simultaneous  partial
                                                                                        differential equations.  One is the equation of  flow,  from which
                                                                                        ground-water velocities are obtained,  and the  second is  the  solute-transport
                                                                                        equation, describing the chemical concentration  in  ground water.

                                                                                             The  theory  behind the equation describing solute transport has been
                                                                                        well  documented  in the literature (see, for example. Bear,  1979).  Changes
                                                                                        in  chemical concentration  occur  within  a dynamic  ground-water system
                                                                                        primarily due to  four distinct  processes:   (1) convective (or advective)
                                                                                        transport,  in  which dissolved chemicals  are moving with the flowing
                                                                                        ground water; (2)  hydrodynamic dispersion, in  which  molecular and  ionic
                                                                                        diffusion and small-scale variations in the velocity of flow through  the
                                                                                        Porous media cause the paths of dissolved molecules and  ions  to  diverge or
                                                                                        spread from  the average direction  of ground-water flow; (3) fluid sources,
                                                                                        where water  of one composition is introduced  into and mixed  with water
                                                                                        °f  a  different composition; and  (4) reactions, in which  some amount  of a
                                                                                        Particular  dissolved chemical species  may be added to or  removed from the
                                                                                        ground water due to chemical,  biological and physical reactions in the
                                                                                        *ater or  between  the  water and the solid aquifer  materials. There are
                                                                                        "gnificant practical problems in quantifying the role  of each  of  these four
                                                                                        Processes  in  the  field.

                                                                                             A generalized form of the solute-transport equation  is presented by
                                                                                            e (1976), in which terms are incorporated to represent  chemical
                                                                                        factions and solute concentrations both  in the  pore fluid and  on the solid
                                                                                        '"'face, as follows:

-------
646
where CHEM equals

                                                                   (2)
      —     for linear  equilibrium  controlled ion-exchange reactions,
     33t
             for s chemical rate-controlled reactions, or
 k-,  R"
-\(eC + pbC) for  decay, and where C  is concentration  of the solute,

ML"3; Vj  is the seepage velocity  in the direction x;, LT"1;  DJJ  is the
coefficient of hydrodynamic dispersion (a  second-order  tensor),  L T"1;  C

is concentration of the solute  in  the  source or sink fluid, ML"  ; pj,  is bulk

density  of the  solid,  ML"3; «  is  the effective porosity; C is concentration
of the species adsorbed  on the solid  (mass of solute/mass of sediment); R^

is the rate of production  of the  solute in  reaction k, ML"3T" ;  and  X  is

the decay constant (equal  to In 2/half life), T"'.

      In the conventional  formulation of equation  2, the dispersion
coefficient itself is a function  both  of the intrinsic  properties of  the
aquifer (such as  heterogeneities in hydraulic conductivity) and  of  the  fluid
flow.  This relationship was expressed by  Scheidegger (1961) as:
                    '  °ijmn
                                                                    (3)
 where  ajjmn is  the  dispersivity or  characteristic length of the porous
 medium (a fourth-order tensor),  L;  Vm and  Vn are the components of the

 flow velocity  of the fluid  in the m and  n directions,  respectively, LT"  ;

 and  IVl is the  magnitude  of the velocity vector, LT1.   Both Scheidegger
 (1961)  and Bear (1979) show that the  dispersivity of an isotropic porous

 medium can be  defined by  two  constants.  These are the longitudinal
 dispersivity of the medium, OL, and the  transverse  dispersivity of the

 medium, «]-.  Most  applications  of  transport models to  ground-water

 contamination problems that have been documented to date have been

 based  on  this conventional formulation.
                                                                                                                                                                      647
        The consideration of solute transport  in  a  porous medium  that  is
  gnisotropic  would  require the estimation of more than  two parameters. For
  example, Moranville and others (1977) and  Greenkorn (1983)  indicate that
  for the case of transversely  isotropic media, the dispersion tensor can be
  characterized by six scalar  invariants. In practice, it  is rare  that field
  values for even the two constants o^ and otj  can be defined uniquely.
  Thus, it appears impractical  to  be able to measure  or  define as many as
  5jx dispersivity constants  in  the field.  So,  although anisotropy in hydraulic
  conductivity (a second-order tensor)  is recognized and  accounted  for  in
  ground-water flow  simulation,  it  is commonly  assumed  out of convenience
  that the same  system is isotropic with respect  to dispersion.

        The error that can be  introduced  by  neglecting material anisotropy is
  illustrated  in field  data where  o-r has been shown  to be sensitive to
  direction.    An example is  the  study of a contaminant plume at Barstow,
  California  (Robson,  1974; 1978). Robson  applied  two-dimensional
  solute-transport models in both  area!  and cross-sectional planes.   To achieve
  a  best fit to the  field data, he had  to  reduce the value of ctj in the
  cross-sectional  model  by a factor of 100 from the value of 60 ft used in
  the area) model (OL - 200 ft in  both planes).   As  flow is predominantly
  horizontal  in this study area, Robson (1978) explains the change  in
  dispersivity  values as follows:   'In the areal-oriented model DL  and  D-r
  ire essentially measures of  mixing along aquifer bedding  planes,  as  is DL
  in the profile  model, whereas Of in  the profile model  is  primarily  a
  measure of  mixing  across bedding planes.*

        If  single values of ct^  or  crj- are used in predicting  solute transport
  when the flow direction is  not  always parallel to one  of the principal
  directions of anisotropy, then dispersive  fluxes will be  either overestimated
  or underestimated  for various parts of the  flow system (depending on
  whether the values of at  and  a-f are characteristic  of  dispersive transport
  in the horizontal  or vertical direction).   This can lead  to  significant  errors
  in predicted concentrations.

        There are many aspects of solute transport and dispersion in
  anisotropic  porous  media that are still poorly  understood.  Fattah  and
  Hoopes  (1985)  state "the tensor  nature of the  dispersion coefficient in
  inisotropic  porous  media . .  . has not been  established." Celhar and Axness
  0983) and Neuman and others  (1987) present  general three-dimensional
  stochastic analyses of macrodispersion (or field-scale dispersion) in
  anisotropic media.   Gelhar and .Axness conclude  that the macrodispersivity
  coefficient  (A;:) is  a  second-rank symmetric tensor;  because the
  off-diagonal  terms are nonzero,  there  is an  offset between the mean flow
  direction and the  principal axis associated  with the largest principal value
  of AJJ when the flow  is not parallel to  the principal axes of  the hydraulic
  conductivity tensor.  On the  other hand, Neuman and others  conclude that
  "large  Peclet numbers, the  dispersivity  tensor reduces  to  a single
  Principal component parallel  to  the mean velocity, regardless  of the
  orientation of the velocity  vector; at small and intermediate Peclet numbers
  '"ere  is an  offset toward  the axis of highest spatial correlation, a direction
•  Opposite  from that inferred  by  Gelhar and  Axness.

-------
648

      The seepage  velocity is obviously a critical parameter  to define  for  a
field problem  involving solute  transport,  as  it directly  influences both
convective transport and the dispersive flux.  An  expression  for  the
average seepage  velocity of ground  water can be derived from Darcy's law
and  can be written in Cartesian  tensor notation as:
           V:
                      3h
(4)
The velocity distribution can  be  computed  on the basis  of observed or
calculated hydraulic gradients.  But the relation  between velocity  and
dispersion is at least  partly dependent  on the relationship between the scale
of observations of concentration  and the scale  of definition of  velocity.  In
general,  for a  field problem at a scale that incorporates local and regional
aquifer heterogeneities, the  smaller or finer the scale  at  which  the velocity
is defined,  the smaller  will be the apparent magnitude of the dispersion
coefficient.

      Although OL is  generally deemed to be an  intrinsic property of the
aquifer,  it  is found  in  practice to be dependent  on  and  proportional to the
scale of  the measurement.   Celhar (1986) shows that most reported values
of 
-------
650

Well documented and efficient multiphase models applicable to
contamination of ground water by  immiscible organic chemicals are not yet
generally available.   One of  the very  few documented field applications  is
Osborne  and Sykes (1986).  Jn summary,  regardless of how accurately »e
can solve the commonly accepted governing equation, cq. 2, that equation
itself is  not  necessarily a  definitive and  sufficient description  of the
processes controlling  contaminant transport at the scale  of  most field
problems. Hence, any predictions arising from the solution  of  that equation
should be appropriately qualified.
4.   NUMERICAL METHODS

     Analytical solutions are available  to solve the solute-transport
equation (see, for example. Bear, 1979;  Javandel  and others,  1984; and  van
Genuchten and Alves, 1982). However, obtaining  the exact  analytical
solution  to the partial differential equation requires that the properties and
boundaries of the flow system  be  highly and perhaps unrealistically
idealized.  For simulating most field  problems, the  mathematical benefits
of obtaining an exact analytical solution are probably  outweighed by  the
errors  introduced  by the  simplifying  approximations of  the  complex field
environment  that  are required to apply the analytical approach.

     The solute-transport equation is in general  more difficult  to solve
numerically than  the ground-water flow equation, largely  because certain
mathematical  properties of  the  transport equation vary depending upon
which  terms in the  equation are dominant in a particular  situation.  When
solute  transport is dominated by conveclive transport, as is  common in
many field  problems, then  cq.  2 approximates a  hyperbolic  type of
equation (simitar  to equations describing the propagation of  a  wave or  of
a shock  front).   But if a system is dominated  by dispersive  fluxes,  such as
might  occur  where fluid  velocities are  relatively low and aquifer
dispersivities  are  relatively high, then eq. 2  becomes more  parabolic in
nature (similar to the ground-water flow  equation).  The numerical  methods
that work best for parabolic partial differential  equations  are  not best  for
solving hyperbolic equations, and vice versa.  Thus, no one numerical
method or simulation  model will be ideal for the entire spectrum of
ground-water transport problems likely  to be encountered  in the field.
Further  compounding  this difficulty  is  the  fact that the ground-water  flow
velocity  within a given multidimensional  flow  field will normally vary
greatly,  from near zero in  low permeability  zones  or near  stagnation
points, to several feet or meters per  day  in  high  permeability  areas or  near
stress points.  Thus, for a  given single ground-water system, the
mathematical characteristic of the transport  process may vary between
hyperbolic and parabolic, so that no  one  model may even be best over  the
entire  domain of a  single problem.

     Three types of numerical methods are  commonly  used to  solve the
solute-transport equation:  finite-difference  methods, finite-element methods,
and the  method of characteristics.  Each  has some  advantages,
disadvantages, and special limitations for applications to field  problems.   A
                                                                                                                                                                    651
  comprehensive review of solute-transport models  is presented by  Anderson
  (|979).     The model survey of van der Heijde and  others  (1985) reviews a
  total  of 84 numerical, mass-transport  models.  Hamilton and others  (1985)
  compare the application  of three  different transport  models to a single
  field  problem; they conclude that the Peclet  number  is a  critical criterion
  lo evaluate.

        The  method of characteristics was originally  developed to solve
  hyperbolic equations.  If solute  transport is dominated by convective
 • transport,  as is common  in  many  field problems, then equation 2 may
  closely  approximate a hyperbolic equation and be highly compatible with
  the method of characteristics.  Documented  models based on variants  of
 ,' this approach include Konikow and Bredehoeft (1978) and  Prickett  and
:j others (1981).  Finite-difference and finite-element methods  can accurately
> and efficiently  solve  the transport equation,  particularly  when  dispersive
  transport is large compared to convective  transport. However, problems of
  numerical  dispersion and oscillations may  induce significant errors  for
 . some  problems. Examples of recently documented three-dimensional,
  transient, finite-difference  models that simultaneously solves the  fluid
  pressure, energy-transport, and  solute-transport equations for
.. nonhomogeneous  miscible fluids include Kipp (1987)  and  Reeves  and others
'. (1986).   A two-dimensional  finite-element transport model  is documented by
  Voss (1984).
                                   I
        Because none of the standard  numerical methods are ideal for a  wide
  range of transport problems, there  is currently much  research on mixed or
  adaptive methods that aim to minimize  numerical errors and combine  the
  best features of  alternative  standard numerical approaches.   Examples
  include Carrera and  Mellon! (1987), Ewing,  Russell, and  Wheeler (1983).
  Fujinawa (1986), and  Neuman  (1984).
  5.  MODEL DESIGN  AND CALIBRATION

       In  the development of a  deterministic  ground-water model for a
  specific area and purpose,  we must  select an appropriate level  of
  complexity (or, rather, simplicity). We are inclined  to believe that  finer
  resolution in a model will  yield greater accuracy.   However, there  also
  exists the practical constraint  that even when appropriate  data  are
  *vailable, a finely discretized  three-dimensional  numerical  transport model
  ""ay be too large or  too expensive  to run on available  computers.  The
  Election of the appropriate model and appropriate level of  complexity  will
  remain subjective and dependent on  the judgement and experience  of the
  •nalysts, the objectives of the study, and level  of  prior information on the
  *ystem of interest. The  trade-off between accuracy and cost will always be
  * difficult one to  resolve,  but  will always have to be made. In any case,
  Managers and other users of model  results must be made aware that these
  •'ade-offs and  judgements have been made and  may affect  the reliability
  or the model.

       In general,  it is more  difficult  to  calibrate solute-transport model  of

-------
652

an aquifer than it is to calibrate a ground-water  flow model.  Fewer
parameters need to be defined to compute the  head distribution  with  a
flow model  than are required  to compute concentration changes with  a
solute-transport  model.   Because  the ground-water  seepage velocity  is
determined from the head distribution, and because both convectivc
transport and  hydrodynamic dispersion are functions of the seepage
velocity,  a  model  of ground-water  flow in an aquifer  is often calibrated
before  a  solute-transport model is developed.  In fact,  in a  field
environment perhaps the single most important key  to  understanding a
solute-transport  problem  is the development of  an accurate definition  (or
model)  of the flow system.  This is particularly relevant to transport  in
fractured rocks, where simulation is commonly based on porous-media
concepts.   Although  the  potential  field can often be simulated, the required
velocity field  may be greatly in error.

      Major  questions in the application of a ground-water model concern
the  model's  ability to represent the processes that  are controlling responses
in the  system  of interest and the reliability of the  predictions.   First  and
foremost, it  must  be demonstrated  that the model  accurately  solves the
governing equations  (this is rarely  a  problem for the solution to  the flow
equation, but  may sometimes be  significant  for the  solute-transport
equation).  Some errors may be introduced by inappropriate approximations
inherent  in the  assumed governing equations because of inadequacies of the
conceptual model.   However, in most model applications to field  problems.
the  dominant  cause of errors in  model output is  the presence of  errors or
uncertainty in the input data,  which reflect our inability to  accurately and
quantitatively  describe the aquifer properties, stresses, and  boundaries.
Concerning both concepts and  parameters, Watson's  (1969) cautioning
statement is relevant:  'Just because we do or must  describe  the  world in a
given way does  not  mean  that the world  is really that way.*

      To  demonstrate that  a deterministic  ground-water simulation model  is
realistic,  it is  usual  to compare field observations  of aquifer responses
(such as changes in  water  levels  or potcntiometric heads  for  ground-water
flow problems and concentration for  transport problems) to corresponding
values obtained  from the model.   The objective of  this calibration
procedure is to minimize differences  between the observed  data and
computed values.  Usually, the model  is considered calibrated  when it
reproduces historical data within some  acceptable level of accuracy.
Although  a poor match  provides  evidence  of errors  in  the concepts (or
hypotheses) underlying the simulation  model,  a  good match in itself does
not  prove the  validity or adequacy of the  model.

      Matalas and  Maddock (1976) argue that model  calibration is
synonymous  with parameter estimation.  The  calibration of a deterministic
ground-water model  is often accomplished through a  trial-and-error
adjustment of  the  model's  input data (aquifer properties, sources  and sinks,
and  boundary  and initial conditions)  to modify the  model's output. Because
a large  number of interrelated factors affect the output, this  may  become
a highly  subjective procedure.  Advances in parameter  identification
procedures, such as described by Cooley (1982). Knopman and Voss (1987),
                                                                            653
pleuman (1980),  Umari  and others (1979), Wagner and  Gorelick  (1986), and
yen (1986), help to  eliminate some of the subjectivity inherent  in model
calibration. However, the hydrologic experience and  judgement of the
giodcler continues to be a major  factor  in calibrating  a  model  both
,ccurately  and efficiently.  The modeler  should  be familiar  with the
specific field  area being studied  to know that both  the  data base and the
guinerical model adequately  represent prevailing field  conditions. The
modeler must  also recognize that  the  uncertainty in  the  specification of
sources, sinks, and boundary  and  initial conditions should be evaluated
during the calibration procedure  in the  same manner as the uncertainty in
aquifer properties.   Failure to recognize the uncertainty inherent both in
the input  data and  in  the calibration data  may  lead to  Tine-tuning" of  the
model  through artificially precise  parameter adjustments strictly to  improve
the match  between observed and  calculated variables. This may  only  serve
to falsely increase the  confidence in  the model  without  producing an
equivalent  (or any)  increase in its predictive accuracy.

     Figure I  illustrates in  a general  manner the role of models in
providing input to the  analysis of ground-water contamination  problems.
The value  of  the modeling approach  is  its  capability to integrate
site-specific data with  equations describing  the relevant  processes as a basts
for predicting changes  or responses in ground-water  quality.  A major
difference  between evaluating existing contaminated  sites and new or
planned sites  is that for the  former,  if  the contaminant source  can  be
reasonably well  defined, the  history of contamination  itself  can,  in  effect,
serve  as a  surrogate  long-term tracer  test that provides critical  information
on velocity and dispersion at a regional scale.   However, it  is common
when  a contamination  problem  is  recognized, that the  locations, timing, and
strengths of the contaminant sources are for  the most part unknown.  At
new sites,  historical  data are commonly  not  available to provide  a basis
for model  calibration and to serve as  a control  on the accuracy of
predictions. As indicated in  Fig.  1, there should be  allowances for feedback
from the stage of interpreting model  output both  to the data collection
and analysis phase  and  to the conceptualization  and mathematical
definition of  the relevant governing processes.


6.  EXAMPLE OF MODEL APPLICATION  TO FIELD  PROBLEM

     The  Idaho National Engineering  Laboratory (INEL) is  located on 890
X). mi. of semiarid land  in the eastern  Snake  River  Plain of southeast
(
-------
654
                                                                                                                                                                    655
referred (or additional  details.

      As described in those two reports, the eastern  Snake River  Plain is a
large structural and  topographic basin  about  200 miles  long  and  50 to 70
miles wide.  It is  underlain  by 2,000  to 10,000 feel  of  thin  basaltic lava
flows, rhyolite deposits, and  interbedded alluvial and lacustrine sediments.
These formations contain  a vast amount of ground water and comprise the
major aquifer  in  Idaho, which is  known as the  Snake River  Plain aquifer.
Ground-water flow is generally to the  southwest  at  relatively high
velocities (5 to 20 feet  per day, or 1.5 to  6 meters per day), according to
the reports on this area.  The principal water-bearing zones occur in  the
basalts, the permeability fabric of which is highly heterogeneous,
anisotropic, and  complicated  by secondary permeability  features, such as
fractures, cavities, and lava  tubes.

      Because of the concern  about the ground-water contamination
resulting  from the waste discharge, in  1973 Robertson developed  a digital
solute-transport model to simulate  the underlying aquifer system  to help
analyze ground-water flow and contaminant transport at the site.   The
numerical  model was based on the method  of characteristics. He  first
calibrated  a flow model  for  a 2,600  sq. mi. (6,600 sq. km.)  area,  and  then
calibrated  the  transport model for a  smaller  part of that area in which
contamination  was of concern.  The  calibration  of the  transport  model was
based on  a 20-year history of contamination,  as documented by samples
from about 45 wells near and downgradient  from the known point sources
of contamination. These data showed that  chloride and  tritium  had spread
over  a  15 sq. mi. (39 sq. km.) area and migrated as far as  5 mi. (8  km.)
downgradient from discharge points.  The distribution of waste chloride
observed  in November 1972 is shown  in fig.  2. Robertson notes that the
degree of  observed lateral  dispersion in the plumes is particularly large.

      Robertson used the calibrated transport  model  to predict future
concentrations  of chloride, tritium, and strontium-90 for the years I9SO
and  2000 under a  variety of  alternative possible  future stresses.   The
scenario that came closest  to  what actually occurred for the chlorides
included assumptions that disposal continues at 1973 rates and the Big Lost
River recharges the aquifer in odd numbered  years.   The projections
indicated  that  by 1980 the leading edges of both the chloride (see fig. 3)
and  the tritium plumes  would  be  at or near  the  INF.L southern boundary.

      Lewis and Goldstein (1982) report  that eight wells  were drilled
during the summer of  1980 near  the  southern boundary to help fill data
gaps  and  to monitor contaminants  in ground  water flowing  across the
INEL boundary.   They also used the data  from  the eight wells  to help
evaluate the accuracy of Robertson's  predictive model.  The  distribution  of
waste chloride  observed  in  October 1980 is  shown in  figure 4.  A
comparison  of  fig. 4 with fig.  2 indicates that the leading edge of  the
chloride plume had advanced  2M to 3  miles (4 to 5 km)  during  that  8 year
period, and that  the  highest concentrations increased from around  85  mg/L
to around  100  mg/L.
         A comparison of figs. 3 and  4 indicate  that although the observed
   and  predicted  plumes show general  agreement in the direction, extent, and
   magnitude of contamination, there  exist some  apparently significant
   differences in  detail.   The observed  plume is  broader and  exhibits more
   lateral spreading than  was predicted, and  has  not spread as far south  and
   as close  to the  INEL  boundary  as  was predicted. Also, the predicted
   secondary  plume north of  the Big  Lost  River, emanating from the Test
   Reactor  Area,  was essentially  not detected  in  the  field at  that time.

         In  view of these differences  between the predicted and  observed
   concentrations  it is reasonable to ask why  the errors occurred, whether the
   errors are  significant  in  relation to the overall problem,  and  whether the
   model predictions had  any value. Lewis and Goldstein  (1982)  present  a
   number of factors which  they felt contributed to  the discrepancy.  These
   reasons can be  summarized as:  (1) less dilution from recharge during
   1977-80 because of  below-normal river flow; (2) chloride disposal  rates at
   the ICPP facility  were increased during the several years preceding  1980;
   (3) the model  grid may have  been too coarse; (4)  the model calibration
 •  selected  inaccurate hydraulic and transport parameters; (5) vertical
   components of flow and  transport  may  be  significant in the aquifer but
   can  not  be evaluated  with the two-dimensional areal model; (6) there may
   be too few wells  to accurately map  the actual plumes, and some existing
   wells may not be constructed  properly to  yield representative  measurements;
   and  (7)  the numerical  method introduces some errors (however. Grove's
  • 1977 analysis of this  same system  used  finite-difference  and  finite-element
   methods, and comparisons  of  numerical  results offer no  basis  for
   concluding that the numerical solution algorithm used  by Robertson  was in
   itself a significant source  of  the predictive errors).

         Although these factors can be  expanded upon,  and additional  factors
   added, it is extremely difficult  to  assess the contribution of any single
   factor to the total error.   One approach that  would  help would be to
   recalibrate the  earlier  model using the now extended historical record and
   use  it to test some of these hypotheses. Other factors could only be  tested
   if new models are developed  that  incorporate additional  or more complex
   concepts, such  as  density  differences and  three-dimensienal  flow. Such a
   recalibration and  model revision should lead to  a model  that  has  greater
   predictive  power and  reliability.  Whether  the errors in this case were
   significant in  relation  to  the  overall problem  can  be best (or  perhaps, only)
   answered by those who sponsored the model study  in light of (1) what
   they  expected, (2)  what actions  were taken or not  taken  because of these
   Predictions, and (3) what  predictive  alternatives were available.  It is clear,
   however, that  the  model  predictions  represented  one  hypothesis of future
 -  contaminant spreading that could be tested in the field,  and  indeed  the
 -'•.  1980 test drilling was  designed  to  a  large extent to test  that  very
 .'  Prediction.  The process of collecting data  is  most  efficient when  guided
 :  by an objective of hypothesis testing.  So perhaps for the  INEL site, a
•'.".  major value of the model  so far has been  to  help  optimize the data
 1  collection  and  monitoring process;  that  is, the  predictive model  offers a
 .  means to  help  decide  how frequently  and  where  water samples  should be
 •?  collected to track the plume.

-------
   656

   7.  CONCLUSIONS

        To make the most reliable prediction  for  a  given ground-water
   problem, all relevant information should be considered and evaluated in
   order to arrive at  the best estimate  of  the  future  behavior of the system
   Deterministic simulation models can  help one  accomplish this quantitatively
   by providing a format- to  integrate and  synthesize  all of that available
   information in a  manner consistent with theories describing  the governing
   processes.   Our present understanding of the many processes affecting
   ground water is sufficiently adequate to allow us,  in  theory,  to forecast
   the behavior of a ground-water system.   In  practice we are severely
   limited by  the inadequacy of data available to  describe aquifer properties
   and historical  stresses  and  responses  and by an  inability  to predict future
  stresses.   Overall, extreme  caution is  required  in making, presenting, and
  accepting predictions of  future ground-water behavior.  Because the
  confidence  in estimates of  future  stresses decreases with  predictive time,
  and because historically observed system  behavior  may  not  reflect the
  relative dominance  or strengths of  different governing  processes under a
  new set  of  stresses, forecasts will  have greater  uncertainty  with increasing
  predictive time.

        The theoretical elegance and power of  the purely deterministic
  approach must  be tempered by the  practical  compromises  imposed  by
  considering  the complex real world of the heterogeneous geological  media
  that contain the ground-water resources of the  world.   This compromise
  comes in the form of melding, in  one way  or  another,  statistical  and
  deterministic approaches.  These ways may range from  a  straightforward
  sensitivity analysis using  a  deterministic model  10 development and solution
  of  stochastic differential  equations  that directly  incorporate  the variance in
  aquifer parameters.

       An  advantage  of deterministic models  is that  they represent processes
  and  thus  have cause and  effect relationships built  into  them.    But
  careful attention must be paid to the accuracy with which  future "causes"
  (stresses)  can be predicted (or  estimated),  because that can  be the  major
 source of error  in  the predictions of  future "effects" (system responses).
 This asset of representing processes  provides  the  basis  for predicting or
 extrapolating responses beyond the  range of  previously  observed stresses
 and  responses.   However, concepts inherent in a given model (for  example,
 two-dimensional flow and vertically averaged parameter  values, or assumed
 geometry  and boundary  conditions) may be adequate over  the observed
 range of stresses, but prove  to be  oversimplified  or  invalid approximations
 under  a new and previously inexperienced type or magnitude of stresses.  It
 should  be recognized that  when model parameters  have  been  adjusted to
 obtain  a  'best fit"  to historical data, a bias may  be  induced  towards
 extrapolating existing trends when predicting  future  conditions.

      Between the two general types  of ground-water problems -- flow  and
 transport — the latter is clearly the more difficult to handle  and predict.
 Flow problems can  often be  simplified through the principle of
superposition, so  that changes in head  resulting from changes in stress  can
                                                                               657
 •' be computed directly. The  flow problem is merely a subset of  the transport
   problem, in which  the flow must  be described  in  terms  of actual  heads
  ' |pd hydraulic gradients,  not just their changes.   Also, the conventional
   jolute-transport equation  provides  a less definitive description of the
   processes and factors  affecting solute  concentration than  does the  flow
   equation for hydraulic head; the former is more difficult and  costly to
   so|ve accurately.  Because the response times for propagation of fluid
 • pressure changes  is  much shorter than  for  the migration  of solutes,  the
   spatial  scale for  variance  in concentration  is much smaller than the scale
   for variance in  head.  Thus, the response  times for  transport  phenomena
   ire proportionately  greater, and it  is more difficult  to calibrate and
   evaluate transport models  with field data  than it  is  for flow  models.

        The  next major  level of improvement in ground-water simulation
 T models will probably  not arise from improved numerical  procedures; rather,
   a greater investment must  be  made  in  obtaining more  accurate descriptions
   of aquifer  properties  and  their variability. Better  definitions  are needed of
 -.  the geometry, boundary conditions, heterogeneities of the system being
 <  analyzed, and field  data  to calibrate and  verify models.   It is especially
 '
-------
 658

 Dagan, G., 1987, Theory or  solute transport by  groundwater:   Ann, Rev
      Fluid Mech., v. 19,  p.  183-215.

 Davies,  P. B.,  1987,  Modeling  area),  variable-density,  ground-water  f\0
      using equivalent freshwater head - analysis of potentially          *
      significant errors: ill Proc.  Solving  Ground  Water Problems with
      Models, v. 2, Natl. Water Well Assoc., Dublin, OH, p. 888-903.

 Davis,  A. D.,  1986, Deterministic modeling  of  dispersion  in  heterogeneous
      permeable media:   Ground  Water,  v, 24, no. 5, p. 609-615.

 Domenico,  P.  A.,  and  Robbins,  G.  S.,   1984, A  dispersion scale  effect  in
      model  calibrations and field tracer  experiments:   Jour.  Hydrology
      v.  70, p.  123-132.

 Ewing,  R. E.,  Russell,  T.  F.,  and  Wheeler,  M.  F.,  1983,  Simulation  of
      rniscible   displacement  using  mixed  methods  and  a  modified
      method   of  characteristics:  SPE   12241,  Proc.  7th  SPE  Symp. On
      Reservoir Simulation, p. 71-81.

 Fattah,  Q.  N.,  and   Hoopes,  J. A.,  1985,  Dispersion  in   anisotropic,
      homogeneous,  porous media:   Jour, of  Hyd. Eng., ASCE,  v.  111  no'
      5,  p. 810-827.

 Faust, C.  R..  1985,  Transport of  immiscible  fluids  within and  below the
      unsaturated zone:   A  numerical model:   Water  Resources  Research
      v.  21, no. 4, p. 587-596.

 Freeze, R.  A.,  and  Cherry,  J. A.,  1979, Groundwater:  Prentice-Hall,  Inc.
      Englewood Cliffs,  NJ, 604  p.

 Ftjinawa,   K...   1986,   A   'characteristic'   finite   element  scheme  for
      convective-dispersive  transport  with  non-equilibrium  reaction:    Int.
      Jour. Numer. Methods  Eng.,  v. 23,  p. 1161-1178.

Glrabcdian,  S.  P., and  LeBlanc, D. R., 1987, Results of  spatial  moments
      analysis for a  natural-gradient tracer  test  in  sand  and  gravel,  Cape
      Cod, Mass.:  [abs.J  Eos, v. 68, no. 16, p. 322-323.

Gtlhar,  L.  W.,  1986,  Stochastic  subsurface  hydrology   from  theory  to
      applications: Water  Resources Research,  v. 22, no. 9, p. I35S-145S.

Gelhar, L. W.,  Gutjahr, A. L., and  Naff,  R. L., 1979,  Stochastic  analysis of
      macrodispersion in  a stratified  aquifer:  Water  Resources  Research,  v.
      IS, no. 6, p. 1387-1397.

Gelhar,  L. W., and Axness, C.  L.,  1983,  Three-dimensional  stochastic
      analysis  of  macrodispersion in  aquifers:  Water  Resources  Research.
      v. 19, no.  I, p. 161-180.

Grcenkorn, R.  A., 1983,  Flow  phenomena in  porous media:   Marcel  Dekker,
                                                                               659
        Inc.,  New York, 550 p.

   Grove.  D.  B.,   1976,  Ion  exchange  reactions  important  in  groundwater
        quality  models:   in^ Advances in  Groundwater  Hydrology, American
        Water Res. Assoc., p. 144-152.

   Grove,  D.  B., 1977, The use  of  Galerkin  finite-element  methods  to  solve
        mass-transport  equations:  U.S.  Geol.  Survey  Water  Resources Inv.
        77-49, 55  p.

   Hamilton,  D.  A.. Wiggert, D. C., and Wright, S. J.,  1985,  Field  comparison
        of three mass transport  models:  Jour. Hydraulic Eng.,  ASCE,  v.  Ill,
        no. I. p. 1-11.

>  Javandel, I., Doughty, D., and  Tsang,  C-F.,  1984, Groundwater  transport:
•2*.       Handbook  of  mathematical  models:   Am. Geophys.  Union, Water Res.
V  '     Monograph 10, 228 p.

i: Kipp. K-  L-  Jr- l987.  HST3D:   A computer code  for  simulation  of heat
        and  solute transport in  three-dimensional ground-water flow  systems:
        U.S.  Geol. Survey  Water-Res.  Inv. Report 86-4095, 517 p.

   Knopman,  D.  S, and Voss, C. I., 1987, Behavior of  sensitivities in the  one-
        dimensional advection-dispersion equation:  Implications for parameter
        estimation  and sampling  design:   Water  Resources Research, v. 23,  no.
        2, p.  253-272.

   Konikow,  L. F.,  1981, Role of numerical simulation in analysis of ground-
        water quality  problems:   Science of  the  Total  Environment, v. 21,
        p. 299-312.

   Konikow,  L.   F.,   and   Bredehoeft,  J.  D.,  1978,  Computer  model  of
        two-dimensional solute  transport  and   dispersion  in  ground   water:
        U.S.  Geol.  Survey  Techniques of Water-Resources Inv., Book  7, Chap.
        C2, 90 p.

   Lewis,  B.   D.,  and  Goldstein, F.  J.,  1982,  Evaluation  of  a  predictive
        ground-water   solute-transport   model    at    the    Idaho   National
        Engineering  Laboratory,  Idaho:  U.S.  Geol.   Survey   Water-Resources
        Investigation  82-25,  71 p.

•{,  Lewis,  F.  M.,  Voss, C.  I.,  and  Rubin,  J,  1987,  Solute  transport  with
j£       equilibrium aqueous complexation   and  either  sorption   or  ion
Jg       exchange: simulation methodology  and applications:  Jour. Hydrology,
f       v. 90, p. 81-115.

        ly, D. M, Ball, W. P.,  and Durant,  M. G., 1986, Variability of  aquifer
        sorption properties  in   a  field  experiment  on  groundwater  transport
        of   organic   solutes:    methods   and   preliminary   results:    Jour.
        Contaminant Hyd, v.  1, p. 119-132.

-------
660

Mackay, D.  M-,  Freyberg,  D.  L., Roberts, P. V.,  and Cherry, J. A.,  1986,  4
      natural  gradient  experiment on  solute transport in a  sand aquifer:  I
      Approach  and  overview  or plume movement:  Water Resources  Res
      v. 22. no.  13. p. 2017-2029.

Matalas, N.  C.,  and  Maddock, T,  III,  1976, Hydrologic semantics:   Water
      Resources  Research,  v. 12, no. I, p. 123.

Matheron,  G.,  and  De Marsily. G.,   1980,  Is  transport  in  porous  media
      always diffusive?  A  counterexample:  Water  Resources  Research,  v
      16, no. 5,  p. 901-917.

Mercer,  J.  W.. Silka, L. R., and Faust, C. R., 1983, Modeling ground-water
      flow at Love Canal, New York:   Jour. Env. Eng.,  ASCE,  v.  109, no.
      4,  p. 924-942.

Moll, F. J.,  Giiven, O.,  Melville, J. G.,  and Keely,  J. F.,  1987, Performance
     and analysis of aquifer tracer  tests with implications  for  contaminant
      transport modeling -- a project summary:   Ground  Water,  v. 25, no. 3,
      p. 337-341.

Moranville,  ML B., Kessler, D.  P., and  Grcenkorn,  R.  A.,  1977,  Directional
      dispersion  coefficients in anisotropic porous media:   Ind. Eng. Chem..
      Fundam., v. 16,  no.  3, p. 327-332.

Neuman, S.  P.,  1980, A statistical approach  to the inverse problem of
       aquifer hydrology, 3, Improved solution  method and added
       perspective: Water Resources  Research, v. 16,  no.  2, p. 331-346.

Neuman, S.  P.,  1982, Statistical characterization  of aquifer  heterogeneities:
      an overview:  iii Narasimhan, T. N.,  ed..  Recent trends  in
      hydrogeology,  Geol.  Soc.  Am.  Spec. Paper  189,  p.  81-102.

Neuman, S.  P.,  1984, Adaptive Eulerian-Lagrangian finite-element method
      for  advection-dispersion:   Int. Jour. Numer. Methods  Eng., v. 20,  p.
      321-337.

Neuman, S.  P., Winter,  C.  L., and Newman,  C. M, 1987, Stochastic theory
      of field-scale  Fickian dispersion in anisotropic porous media:  Water
      Resources Research, v.  23, no. 3, p. 453-466.

Osborne, M., and Sykes, J., 1986, Numerical modeling of immiscible  organic
      transport at the Hyde Park Landfill:  Water Resources Research, v.
      22, no. I,  p.  25-33.

Finder,  G.  F., and Abriola, L.  M., 1986, On the  simulation  of  nonaqueous
      phase organic compounds in the  subsurface:  Water Resources
      Research, v. 22. no. 9. p.  109S-119S.

Prickett, T.  A.,  Naymik, T. G.,  and  Lonnquist, C.  G., 1981, A
     "Random-Walk"  solute transport model for selected  groundwater quality
                                                                                                                                                                   661
    evaluations:   Illinois State Water Survey,  Bull.  65,  103  p.

Reeves. M., Ward,  D. S., Johns, N. D., and Cranwell, R. M., 1986, Theory
     and implementation for SWIFT II,  the  Sandia  waste-isolation flow
     and transport model  for  fractured  media, release 4.84:  Sandia Natl.
     Labs., NUREG/CR-3328,  SAND83-1159, Albuquerque, NM, 189 p.

gobertson, J.  B., 1974, Digital  modeling of radioactive and  chemical waste
     transport  in  the Snake River  Plain  aquifer at the National  Reactor
     Testing  Station, Idaho: U.S. Geol. Survey Open-File Report  IDO-220S4,
     41 p.

Robson, S. G.,  1974, Feasibility of digital water-quality modeling  illustrated
     by application at Barstow, California:  U.S. Geological Survey
     Water-Resources Investigations 46-73, 66  p.

Robson, S. G.,  1978, Application of digital profile  modeling techniques to
     ground-water solute transport  at Barstow, California: U.S. Geological
     Survey  Water-Supply Paper  2050, 28 p.

Rubin, J.,  1983, Transport of reacting solutes  in porous media: Relation
     between mathematical  nature  of problem  formulation  and chemical
     nature  of reactions:   Water Resources Research, v. 19, no. 5, p.
      1231-1252.

Scheidegger, A. E., 1961, General theory  of dispersion  in  porous  media:
    Jour.  Geophys.  Research,  v. 66, no.  10, p. 3273-3278.

Smith,  L.,  and  Schwartz, F.  W.,  1980, Mass transport,  1, A stochastic
     analysis of macroscopic dispersion:   Water Resources Research, v. 16,
     no. 2, p. 303-313.                                              	

Umari, A.,  Willis,  R., and  Liu, P. L.-F., 1979, Identification of aquifer
     dispersivities in two-dimensional  transient groundwater contaminant
     transport: an optimization approach:  Water Resources Research,  v.
      15, no.  4, p. 815-831.

v»n Genuchten, M. T., and Alves, W. J.,  1982,  Analytical  solutions of  the
     one-  dimensional convective-dispersive solute-transport  equation: U.S.
     Dept.  Agriculture, Tech.  Bull.  1661,  151  p.

v»» der Heijde, P. K. M.,  Bachmat, Y.,  Bredehoeft,  J. D.,  Andrews,  B.,
     Holtz, D., and  Sebastian, S., 1985,  Groundwater management:   The use
     of numerical models:   Am. Geophys. Union,  Water Resources
     Monograph 5, 2nd ed., 180 p.

V<|W, C. I.,  1984, SUTRA -- Saturated Unsaturated Transport ~ A
     finite-element simulation  model for  saturated-unsaturated
     fluid-density-dependent ground-water flow with energy transport  or
     chemically-reactive single-species solute  transport: U.S.  Geol. Survey
     Water-Resources Invest. Report 84-4369, 409 p.

-------
662

Wagner, B. J., and  Gorelick, S. M,  1986. A statistical methodology  for
      estimating transport  parameters:  Theory and applications to
      one-dimensional advective-dispersive systems:  Water Resources
      Research, v. 22, no.  8, p. 1303-13IS.
                                                                                                      663
Watson, R.  A., 1969,  Explanation and prediction  in  geology
     Geology,  v. 77,  p. 488-494.
Jour, of
Yeh, W. W-G.,  1986, Review of parameter identification procedures in
     groundwater hydrology:   The inverse problem:  Water Resources
     Research, v. 22, no. 1, p. 95-108.
— — — —



V^. fflfl
we *

-jMtooaRpPu

»*A«ACA.MC
aoujwcco

CTCNRND

CfiTONflNC
|
XXlSOfG
NTfWMmr
t
JL
. —
CUJBRKIDNf

KMtuanet
ramspom

CONCOTW.M:
OCOVdVMG
nfSC%.O«A*
PROCESSES
—



XXLS
<«c »

                                                                                       Figure 1.  Schematic overview  of the role of simulation models in
                                                                                       evaluating ground-water contamination  problems (from Konikow,  1981).
                                                                                       F'8ure 2.   Map of  ICPP-TRA  vicinity showing  observed distribution of
                                                                                             chloride  in the Snake River Plain aquifer water  in 1972 (from
                                                                                       K°bertson,  1974).

-------
664
Figure  3.   Model-projected  distribution of  waste chloride  in  the Snake
River Plain aquifer for 1980. ICPP-TRA vicinity, assuming disposal
continues at  1973 rates and the Big Lost River recharges the aquifer in
odd numbered  years (from  Robertson.  1974).
Figure  4.  Distribution of waste chloride in  the  Snake River  Plain  aquifer-
ICPP-TRA vicinity,  October 1980 (from  Lewis  and Goldstein,  1982).

-------
  SECTION 6






CASE HISTORY

-------
Modeling Chloride Movement
in the Alluvial Aquifer at
the Rocky Mountain Arsenal,
Colorado
By LEONARD F. KONIKOW
GEOLOGICAL SURVEY WATER-SUPPLY PAPER 2044

-------
 UNITED STATES DEPARTMENT OF THE INTERIOR

              JAMES G. WATT, Secretary


               GEOLOGICAL SURVEY

                Dallas L. Peck, Director
                First printing  1977
               Second  printing  1983
                                                                                                                   CONTENTS
                                                                      Pa*
Abstract 	      1
Introduction	      1
    Selection of study area  	      2
    Procedure of investigation 	      4
    Acknowledgment*	      4
Simulation model 	      4
    Background 	•.	      4
    Flow equation 	      6
    Transport equation	      6
    Dispersion coefficient	      7
    Numerical methods ...,	      7
    Boundary conditions	      9
Description of study area	     10
    History of contamination 	     10
    Contamination pattern	     12
    Hydrogeology	     12
Application of simulation model	     16
    Finite-difference grid	     15
    Data requirements	     15
        Aquifer properties	     15
        Aquifer stresses	     18
    Calibration of flow model	.'	     20
    Calibration of solute-transport model	     23
    Predictive capability 	     32
    Application to water-management problems	     36
Summary and conclusions  	     40
References cited 	     42
                                                                                                                ILLUSTRATIONS
For (ale by the Distribution Branch, U.S. Geological Survey,
     604 South Pickett Street, Alexandria, VA 22304
FIOURB  1-4.  Maps showing:
                1. Location of study area	     3
                2. Major hydrologic features	    11
                3. Observed chloride concentration, 1966	    13
                4. General water-table configuration in the alluvial aquifer
                     in and  adjacent to the Rocky  Mountain Arsenal,
                     1966-71 	    14
           5.  Finite-difference grid used to model the study area	    16
           6.  Graph showing change  in standard error of estimate for suc-
                cessive simulation tests	    19
                                                                    m

-------
IV
                                CONTENTS
                                                                      P»ge
FIGURE     7. Graph  showing relation between the assumed rate  of net
                recharge in irrigated areas and the mean difference of ob-
                served and computed water levels	    20
        8-18. Maps showing:
                 8. Computed chloride concentration, 1966	    26
                 9. Observed chloride concentration, January 1961  	    26
                10. Computed chloride concentration at the start of 1961  ..    27
                11. Observed chloride concentration, January-May 1969  ..    28
                12. Computed chloride concentration at the start of 1969  ..    29
                13. Observed chloride concentration, May 1972 	    30
                14. Computed chloride concentration at the start of 1972  ..    31
                15. Chloride concentration predicted for 1980, assuming that
                     pond C is filled with fresh water during 1972-80	    32
                16. Chloride concentration predicted for 1980, assuming that
                     recharge from pond C is minimal during 1961-80 ...    34
                17. Computed drawdown caused by maintaining two con-
                     stant-head sinks along the northern boundary of the
                     Rocky Mountain Arsenal  	    36
                18. Chloride concentration predicted for 1980, assuming that
                     artificial  recharge  from Pond  C  is  coupled  with
                     drainage through two hydraulic sinks	    38
          19. Generalized  cross  section from vicinity of source of artificial
                recharge through hydraulic sink (represented as a well) ...    39
                                TABLES
TABU      1.  Summary of main data requirements for numerical model	
           2.  Generalized history of disposal pond operations at the Rocky
                Mountain Arsenal, 1943-72 	
           3.  Elements of hydrologic budget computed by ground-water flow
                model	
                                                                       17
                                                                       21
                                                                       23
                                                                                                                           CONTENTS
                                                                                                                   CONVERSION FACTORS
                                                                                              English units used in this report may be converted to metric unit* by the following
                                                                                            conversion factors:
                                                                                                     namatrt
                                                                                                     English unite
                                                                                            Feet (ft)  ............................................
                                                                                            Feet per year (ft/yr)  --------------------
                                                                                            Feet per day (ft/d) ..........................
                                                                                            Feet per second per foot (tft/sl/ft)
                                                                                            Square feet (ft1)  	
                                                                                            Feet squared per day (ft'/d)

                                                                                            Cubic feet per second (ft»/s)
 Multiply by
4.047 x 10~3
 .3048
 .3048
 .3048
1.0

 .0929
 .0929

2.832 x 10-*
                                                                                            Cubic feet  per  second per mile  1.760 x 10~*
                                                                                              (IftVsl/mi).
                                                                                            Miles (mi)  	  1.609
                                                                                            Square miles (mi1)  	  2.690
         IboMofa
         Metric unite
Square kilometers (km1).
Meters (m).
Meters per year (m/yt).
Meters per day (m/d).
Meters per second per meter
  Um/B]/m).
Square meters (m1).
Meters squared per day
  (m'/d).
Cubic  meters  per second
  (mVs).
Cubic meters per second per
  kilometer ((mVs]/km).
Kilometers (km).
Square kilometers (km1).

-------
  MODELING CHLORIDE MOVEMENT IN THE
       ALLUVIAL AQUIFER AT THE ROCKY
        MOUNTAIN ARSENAL, COLORADO

                     By LEONARD F. KONIKOW
                             ABSTRACT
  A solute-transport model that can be used to predict the movement of dissolved
chemicals in flowing ground water was applied to a problem of ground-water con-
tamination at the Rocky Mountain Arsenal, near Denver, Colo. The model couples a
finite-difference solution to the ground-water flow equation with the method-of-charac-
teristics solution to the solute-transport equation.
  From 1943 to 1956 liquid industrial wastes containing high chloride concentrations
were disposed into unlined ponds at the Arsenal. Wastes seeped out of the unlined dis-
posal ponds and spread for many square miles in the underlying shallow alluvial
aquifer. Since 1956 disposal has been into an asphalt-lined reservoir, which contributed
to a decline  in ground-water contamination by 1972. The simulation model quan-
titatively integrated the effects of the major factors that controlled changes in chloride
concentrations and accurately reproduced the 30-year history, of chloride ground-water
contamination.
  Analysis of the simulation results indicates that the geologic framework of the area
markedly restricted the  transport and dispersion of dissolved chemicals in the
alluvium. Dilution, from irrigation recharge and seepage from unlined canals, was an
important factor in reducing the level of chloride concentrations downgradient from
the Arsenal. Similarly, recharge of uncontaminated water from the unlined ponds since
1966 has helped to dilute and flush the contaminated ground water.

                         INTRODUCTION

  The contamination  of a ground-water resource is a serious problem
that can have long-term economic and physical consequences  and
might not be easily  remedied. Although the prevention of ground-
water contamination provides the most satisfactory result  (Wood,
1972), the capability to predict the movement of dissolved chemicals
in flowing ground water is also needed in order to (1) plan and design
projects to minimize ground-water contamination, (2) estimate spatial
and temporal variations of chemical concentrations,  (3) estimate the
traveltime of a contaminant from its source to a ground-water sink (a
discharge point, such  as a stream, spring, or well), (4) help design an
effective and efficient monitoring system, and (5) help evaluate the

-------
 2             ROCKY MOUNTAIN ARSENAL. COLORADO

physical and economic feasibility of alternative reclamation plans for
removing contaminants from an aquifer and (or) preventing the con-
taminants from spreading.
  Reliable predictions of contaminant movement can be made only if
we  understand the processes  controlling convective transport, hy-
drodynamic dispersion, and chemical  reactions that affect the dis-
solved chemicals in ground water, and if these processes can be ac-
curately represented in a systematic model. For a model to be usable
in a variety of hydrogeologic situations, the modeling technique must
be accurate, functional, and transferable. Because aquifers generally
have  heterogeneous properties and complex boundary conditions,
quantitative predictions would appear to require the use of a  deter-
ministic, distributed parameter, digital simulation model.
  This study is part of the US. Geological Survey's Subsurface Waste
Program, the objective of which is to appraise the impact of waste dis-
posal on the Nation's water resources. The main objective of this study
was to demonstrate the applicability of the method-of-characteristics
model to a problem of conservative (nonreacting) contaminant  move-
ment through an alluvial aquifer. By studying a field problem in which
the effects of reactions are negligible, the effects of other processes
that affect solute transport may be isolated and described more ac-
curately. This study should serve as a basis for investigating more
complex systems whose chemical reactions are significant and in-
teract with the other processes. The purposes of this report are (1) to
briefly describe the  general simulation model and (2) to demonstrate
its application to a complex field problem.
  Because convective transport and hydrodynamic dispersion depend
on  the velocity of ground-water flow, the  mathematical simulation
model must solve two simultaneous partial differential equations. One
is the equation of flow, from' which ground-water velocities are ob-
tained, and the second is the solute-transport equation, describing the
chemical concentration in the ground water. Three general classes of
numerical methods  have been used to solve these partial differential
equations: finite-difference methods, finite-element methods, and the
method of characteristics. Each method has some advantages,  disad-
vantages, and special limitations for application to field problems.
                     SELECTION OF STUDY AREA
  The field area selected for this study is in and adjacent to the Rocky
Mountain Arsenal, near Denver, Colo. (See fig. 1.) A 30-year history of
ground-water contamination in this area  is related to the disposal of
liquid industrial wastes into ponds (Petri, 1961; Walker, 1961; Walton,
1961). The Rocky Mountain Arsenal area is well suited for this study
           MODELING CHLORIDE MOVEMENT, ALLUVIAL AQt|
                                          _ ADAMS CO	
                                          ARAPAHOE CO
                                                                                                              10 KILOMETERS
                   PlOUHB 1.—Location of study area.

because (l) the geology and hydrology of the area are well known, (2)
adequate, though limited, water-quality data are available to calibrate
the mathematical model, and (3) the history of liquid waste-disposal
operations  at the  Arsenal can  be  approximately reconstructed.
Furthermore, the waste water has a very high chloride concentration,
which can serve  as a conservative tracer.

-------
ROCKY MOUNTAIN ARSENAL. COLORADO
                  PROCEDURE OF INVESTIGATION

  This investigation was conducted in three distinct phases. During
the first phase, all available data were collected, interpreted, and
analyzed to produce  accurate, comprehensive, and quantitative
descriptions for the alluvial aquifer of its (1) geologic properties and
boundaries, (2) hydraulic properties, boundaries, and stresses, and (3)
chemical sources arid distributions over space and time. Many of the
geologic and hydraulic interpretations were presented by Konikow
(1975). Most chemical data are presented in this report.
  During the second phase of the investigation, a steady-state flow
model was developed to estimate recharge rates to the aquifer and to
compute ground-water flow velocities. In the third phase of the in-
vestigation, the solute-transport model was calibrated to reproduce
the observed history of  ground-water contamination at the Rocky
Mountain Arsenal. Much of the output from the flow model was used
as input to the solute-transport model.

                       ACKNOWLEDGMENTS

  John D. Bredehoeft, U.S. Geological Survey, and George F. Finder,
formerly with  the Survey and now at Princeton University, jointly
developed the original version of the solute-transport model used in
this study. J.D. Bredehoeft was also instrumental both in the further
development of this model and in the selection of the area for this
study. Their work is gratefully acknowledged. Many data were sup-
plied by the Rocky Mountain  Arsenal, the U.S. Army Corps of
Engineers, and the* Colorado Department of Health, and their assist-
ance also is appreciated.

                     SIMULATION MODEL
                         BACKGROUND
  The purpose of the simulation model is to compute  the concentra-
tion of a dissolved chemical species in an aquifer at any specified place
and time. Changes in chemical concentration occur within a dynamic
ground-water system primarily due to four distinct processes:

1. Convective transport, in which dissolved chemicals are moving with
    the flowing ground water.
2. Hydrodynamic dispersion, in which molecular and  ionic diffusion
    and small-scale variations in the velocity of flow through the
    porous media cause the paths of dissolved molecules and ions to
    diverge or spread from the average direction of ground-water
    now.
3. Mixingjor dilution), in which water of one composition is introduced
            r of a different composition.
MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                              4. Reactions, in which some amount of a particular dissolved chemical
                                                                  species may be added to or removed from the ground water due to
                                                                  chemical  and physical reactions in the water or between  the
                                                                  water and the solid aquifer materials.
                                                              The model presented in this report assumes that no reactions occur
                                                              that affect the concentration of the species of interest and that the
                                                              density and viscosity of the water are constant and independent of the
                                                              concentration. Robertson  (1974) expanded the model to include the
                                                              effects of radioactive decay and ion exchange with a linear adsorption
                                                              isotherm.
                                                                The modeling technique used in this study couples an implicit finite-
                                                              difference procedure to solve the flow equation and the method of
                                                              characteristics to  solve  the  solute-transport equation.  The  ap-
                                                              plicability of  this (or  any other) type of model to complex  field
                                                              problems can  only be demonstrated  by first testing it for a variety of
                                                              field conditions in which observed records of contaminant movement
                                                              can be compared with concentration changes computed by the model.
                                                              In this manner, the accuracy, limitations, and efficiency of the method
                                                              can be shown for a wide range of problems. Also, calibrating the model
                                                              in an area for which historical data are available will provide insight
                                                              into the use of the model in areas where few or no data are available.

                                                                                      FLOW EQUATION

                                                                By following the derivation of Finder and Bredehoeft (1968), the
                                                              equation describing  the transient two-dimensional  flow of a
                                                              homogeneous  compressible  fluid through a nonhomogeneous  an-
                                                              isotropic aquifer may be written in cartesian tensor notation as:
                                                             where
                                                                  TJ: is the transmisflivity tensor, l?IT\
                                                                  h  is the hydraulic head, L;
                                                                  S  is the storage coefficient, L°;
                                                                  t   is the time, 7;
                                                                  W is the volume flux per unit area, LIT; and
                                                                 x, y are cartesian coordinates.

                                                             If we only consider fluxes of (1) direct withdrawal or recharge, such as
                                                             well pumpage, well injection, or evapotranspiration, and (2) steady
                                                             leakage  into or  out  of the aquifer through a confining  layer,
                                                             streambed, or lake bed, then W(xy,0 may be expressed as:
                                                                                                    t - h),
                                                   (2)

-------
6
              ROCKY MOUNTAIN ARSENAL, COLORADO
where
    Q is the rate of withdrawal (positive sign) or recharge (negative
         sign), LIT;
    Kg is the vertical hydraulic conductivity of the confining layer,
         streambed, or lake bed, LIT;
    m is the thickness of the confining layer, streambed, or lake bed,
         L; and
    Ha is the hydraulic head in the source bed, stream, or lake,  L.

  Lohman  (1972) showed that an expression for the average seepage
velocity of  ground water can be derived from Darcy's Law. This ex-
pression  can be written  in cartesian tensor notation as:

                                   lh                       (3)
 where
     Vjf is the seepage velocity in the direction of *,-, LIT;
     Ky is the hydraulic conductivity tensor, LIT; and
     n  is the effective porosity of the aquifer, L°.

                      TRANSPORT EQUATION

   The equation used to describe the two-dimensional transport and
 dispersion of a given dissolved chemical species in flowing ground
 water was derived by Reddell and Sunada (1970),  Bear (1972), and
 Bredehoeft and Finder (1973) and may be written as:
  8C
  d*  "

 where
     C
 8C.
'**
                                    nb
                                                    i'j-1,2,  (4)
        is the concentration of the dissolved chemical species, Mil?;
     Dy is the dispersion tensor, I?IT;
     b  is the saturated thickness of the aquifer, L;
     C' is the concentration of the dissolved chemical in a source or
          sink fluid, MIL?; and
     Rfr is the rate of production of the chemical species in reaction k
          of s different reactions, M/L3T.
   The first term on the right side of equation 4 represents the change
 in concentration due to hydrodynamic dispersion and is assumed to be
 proportional to the concentration gradient. The second term describes
 the effects of convective transport, and the third term represents  a
 fluid source or sink. The fourth term, which describes chemical reac-
                                                                        MODEUNG CHLORIDE MOVEMENT, ALLUVIAL AQ


                                                               tions, must be written explicitly for all reactions affecting the chemi-
                                                               cal species of interest. This term may be eliminated from equation 4
                                                               for the case of a conservative (nonreactive) species.

                                                                                     DISPERSION COEFFICIENT

                                                                  The dispersion coefficient may be related to the velocity of ground-
                                                               water flow and to the nature of the aquifer using Scheidegger's (1961)
                                                               equation:
                                                                                                              vv
                                                                                                                                           (5)
                                                                              where
                                                                                  a(jmn w tne dispersivity of the aquifer, L;
                                                                                  Vm and Vn are components of velocity in the m and n directions,
                                                                                     LIT; and
                                                                                  | vl is the magnitude of the velocity, LIT.

                                                                                Scheidegger (1961) further showed that, for an isotropic aquifer, the
                                                                              dispersivity tensor can be defined in terms of two constants. These are
                                                                              the longitudinal and transverse dispersivities of the aquifer (a t and a2,
                                                                              respectively). These are related to tv j longitudinal and transverse dis-
                                                                              persion coefficients by
                                                                              and
                                                                                                                                           (6)
                        UP-a, | V|.                          (7)

  After expanding equation 5, substituting Scheidegger's identities,
and eliminating terms with coefficients that equal zero, the compo-
nents of the dispersion coefficient for two-dimensional flow in an
isotropic aquifer may be stated explicitly as:

                                                             (8)
                                                                                                                                          (9)


                                                                                                                                         (10)
                                                                                                   NUMERICAL METHODS
                                                                               Because aquifers have variable properties and complex boundary
                                                                             conditions, exact solutions to the partial differential equations of flow

-------
8
   ROCKY MOUNTAIN ARSENAL, COLORADO
(eq 1) and solute transport (eq 4) cannot be obtained directly.
Therefore, an approximate numerical method must be employed.
  Finder and Bredehoeft (1968) showed that if the coordinate axes
are aligned with the principal directions of the transmissivity tensor,
equation 1 may be approximated by the following implicit finite-
difference equation:
F
L
                 (Ajc)2
                                      [
                                       f
                                       L
                                             (A,y)2
                                  A*
 where
      i , j , k are indices in the x, y, and time dimensions, respectively;

  Ax, Ay, At are increments in the x, y, and time dimensions, respec-
                 tively; and

         ?u> is the volumetric rate of withdrawal or recharge at the
                 (iJ)node,Ls/T.

    The numerical solution of the finite-difference equation requires
  that the area of interest be subdivided into small rectangular cells,
  which constitute a finite-difference grid. The finite-difference equa-
  tion is solved numerically, using an iterative alternating-direction im-
  plicit procedure  described by Finder (1970) and Prickett and Lonn-
  quist (1971).
    After the head distribution has been computed for a given time step,
  the velojto of ground-water flow is computed at each node, using an
                                                                                       MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                                                                                                 9
                                                                   explicit finite-difference form of equation 3. For example, the velocity
                                                                   in the x direction at node (i,j) would be computed as:
                                                                                                                              (12)
                                                                                                      2Ax
 A similar expression is used to compute the velocity in the y direction.
   The method of characteristics presented by Carder, Peaceman,
 and Pozzi (1964) is used to solve the solute-transport equation (eq 4).
 The development and application of this technique in ground-water
 problems has been presented by Finder and Cooper (1970), Reddell
 and Sunada  (1970), Bredehoeft and  Finder (1973), Konikow  and
 Bredehoeft  (1974), Robertson  (1974), and Robson  (1974).   The
 method actually solves a system of ordinary differential equations
 that is  equivalent to  the partial differential equation  (eq 4)  that
 describes solute transport.
  The numerical solution is achieved by introducing a set of moving
 points that can be traced with reference to the stationary coordinates
 of the finite-difference grid. Each point has a concentration associated
 with it and is moved through the flow  field in proportion to the flow
 velocity  at its location. The moving points  simulate convective
 transport because the concentration  at each node of the finite-
 difference grid changes as different points enter and leave its area of
influence. Then, the additional change in concentration due to disper-
sion and to fluid sources is computed by solving an explicit finite-
difference equation. In this study, four points were initially distributed
in each cell of the grid.
                                                                                       BOUNDARY CONDITIONS

                                                                    Several different types of boundary conditions can be represented in
                                                                  the simulation model. These include:
                                                                  1. No-flow boundary  By specifying a transmissivity equal to zero at a
                                                                       given node, no flow can occur across the boundary of that cell of
                                                                       the finite-difference grid. The  numerical method used  in this
                                                                       model also requires that the outer rows and columns of the finite-
                                                                       difference grid have zero transmissivities.
                                                                  2. Constant-head boundary: Where the head in the aquifer will not
                                                                       change with time, a constant-head condition is maintained by
                                                                       specifying a very high value of leakance (1.0 [ft/sl/ft or [m/s]/m),
                                                                       which is the ratio of the vertical hydraulic conductivity to the
                                                                       thickness of the confining layer, streambed, or lake bed. The rate
                                                                       of leakage is then a function of the difference between the head

-------
10
ROCKY MOUNTAIN ARSENAL, COLORADO
     of the aquifer and the head in the source bed, stream, or lake and
     is computed implicitly by the model.
 3. Constant flux: A constant rate of withdrawal or recharge may be
     specified for any node in the model.
  At any boundary that acts as a source of water to the aquifer, the
chemical concentration of the source must also be defined.

                 DESCRIPTION OF STUDY AREA

                    HISTORY OF CONTAMINATION
  The Rocky Mountain  Arsenal has been operating since 1942, pri-
 marily manufacturing and processing chemical warfare products and
 pesticides. These operations have produced liquid wastes that contain
 complex organic  and  inorganic chemicals,  including a  charac-
 teristically high chloride concentration that apparently ranged up to
 about 5,000 mg/1 (milligrams per liter).
  The liquid wastes were disposed into several unlined ponds (fig. 2),
 resulting in the contamination of the underlying alluvial aquifer. On
 the basis of available records, it is assumed that contamination first
 occurred at the beginning of 1943. From 1943 to 1956 the primary dis-
 posal was into pond A. Alternate and overflow discharges were col-
 lected in ponds B, C, D, and E.
  Much of the area north of the Arsenal is irrigated, both with surface
 water  diverted from one of  the irrigation canals, which are also
 unlined, and  with  ground  water  pumped from irrigation wells.
 Damage to crops irrigated with shallow ground water was observed in
 1951,  1952, and 1953  (Walton, 1961).  Severe crop damage was
 reported during 1954, a year when the annual precipitation was about
 one-half the normal amount, and ground-water use was heavier than
 normal (Petri, 1961).
  Several investigations have been conducted since 1954 to determine
 both the cause of the problem and how to prevent further damages.
 Petri and Smith (1956) showed that an area of contaminated ground
 water of several square miles existed north and northwest of the dis-
 posal ponds. These data clearly indicated that the liquid wastes seeped
 out of the unlined disposal ponds, infiltrated the underlying alluvial
 aquifer, and migrated downgradient toward the South Platte River. To
 prevent additional contaminants from entering the aquifer, a 100-
 acre (0.405 km2) evaporation pond (Reservoir F) was constructed in
 1956, with an asphalt lining to hold all subsequent liquid wastes
 (Engineering News-Record, Nov. 22,1956).
  In 1973 and 1974 there were new (and controversial) claims of crop
 and livestock damages allegedly caused by ground water that was con-
 taminated at the Arsenal (The Denver Post, Jan. 22, 1973; May 12,
                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL AQ
                   11
                                                                                                         1O4-5O'
                                                                   39-50' -
                                                                               L	ROCKY MOUNTAIN AKSENALl BOUNDARY	

                                                                                          O       l       2 MILES
                                                                                          h
                                                                                          0     1    2 KILOMETERS
                                                                                            EXPLANATION
                                                                                    Irrigated «r««

                                                                                    Irrigation w*ll
Unlimd rnarvolr

Lln«d rcnrvolr
                                                                FIGURE 2.— Major hydrologic features. Letters indicate disposal-pond designation*
                                                                                     assigned by the VS. Army.
                                                                1974; May 23, 1974). Recent data collected by the Colorado Depart-
                                                                ment  of Health  (Shukle,  1975) have  shown  that  DIMP
                                                                (Diisopropylmethylphosphonate), a nerve-gas byproduct about which
                                                                relatively little is known, has been detected at a concentration of 0.57
                                                                ppb (parts per billion) in a well located approximately 8 miles  (12.9
                                                                km) downgradient from the disposal ponds and 1 mile (1.6 km) upgra-
                                                                dient from 2 municipal water-supply wells of the City of Brighton. A
                                                                DIMP concentration of 48 ppm (parts per million), which is nearly

-------
12
ROCKY MOUNTAIN ARSENAL, COLORADO
100,000 times higher, was measured in a ground-water sample col-
lected near the disposal ponds. Other contaminants detected in wells
or springs in  the area include DCPD (Dicyclopentadiene), endrin,
aldrin, and dieldrin.
                    CONTAMINATION PATTERN

  Since 1955 more than  100 observation wells and test holes have
been constructed to monitor changes in water quality and water levels
in the  alluvial aquifer. The areal extent of contamination has been
mapped on the basis of chloride concentrations in wells, which ranged
from normal background concentrations of about  40 to 150 mg/1 to
about 5,000 mg/1 in contaminated ground water near pond A.
  Data collected during 1955-56 indicate that one main plume of con-
taminated water extended beyond the northwestern boundary of the
Arsenal and that a small secondary plume extended beyond the north-
ern boundary. (See fig. 3.) However, the velocity distribution computed
from the water-table map available at that time  (Petri and Smith,
1956) could not, in detail, account for the observed pattern of spread-
ing from the sources  of contamination. Because contaminant
transport depends upon flow, the prediction of concentration changes
requires the availability of accurate, comprehensive, and quantitative
descriptions for the aquifer  of its hydraulic properties, boundaries,
and stresses.
                         HYDROGEOLOGY

  The  records of about 200  observation wells, test holes, irrigation
wells, and domestic wells were compiled, analyzed, and sometimes
reinterpreted  to describe the hydrogeologic characteristics of the
alluvial aquifer in and  adjacent to the Rocky  Mountain Arsenal.
Konikow (1975) presented four maps that show the configuration of
the bedrock surface, generalized water-table configuration, saturated
thickness of alluvium, and transmissivity of the aquifer. These maps
show that the alluvium forms a complex, nonuniform, sloping, discon-
tinuous, and heterogeneous aquifer system.
  A map showing the general water-table configuration for 1955-71
is presented in figure 4. The assumptions and limitations of figure 4
were discussed in more detail  by Konikow (1975).  Perhaps the
greatest change from previously available maps is the definition of
areas in which the alluvium either is absent or is unsaturated most of
the time. These areas form internal barriers that significantly affect
ground-water flow patterns within the aquifer.  The contamination
patten shown in figure 3 clearly indicates that the migration of dis-
solved^Wpride in this aquifer was  also significantly constrained by
the acaHB' boundaries.
MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
13
                                                                                                         104'5O'
                                                                    39-50' h-
                                                                               L	ROCKY MOUNTAIN ARSENALl BOUNDARY BaBB	|
                                                                                          0       1       2 MILES
                                                                                          I	1—H	'
                                                                                          0    1     2 KILOMETERS
                                                                                            EXPLANATION
                                                                             •    Data point (Sept. 1955-March 1966)
                                                                         —300 -  Line of equal chloride concentration (in milligram* per liter).
                                                                                   Interval variable

                                                                          LA.-.;«! Area in which alluvium I* absent or unaaturated

                                                                             FIGURE 3.— Observed chloride concentration, 1956.

                                                                  The general direction of ground-water movement is from regions of
                                                                higher water-table  altitudes to those of lower water-table altitudes
                                                                and is  approximately perpendicular to the water-table contours.
                                                                Deviations from the general flow pattern inferred from water-table
                                                                contours may occur in some areas because of local variations in
                                                                aquifer  properties, recharge,  or discharge. The nonorthogonality at
                                                                places between water-table contours and aquifer boundaries indicates
                                                                that the approximate limit of the saturated alluvium does not consist-
                                                                ently represent a no-flow boundary, but that, at son^Hfoces, there

-------
14
ROCKY MOUNTAIN ARSENAL, COLORADO
                                        1CW50'
                          0    1    2 KILOMETERS
                            EXPLANATION
           ,00— WATER-TABLE CONTOUR - Shows approximate altitude of
                  water table, 1955-71. Contour interval 10 feet 13 meters).
                  Datum is mean sea level
                Area in which alluvium ii absent or unsaturated
 FlOURB 4.— General water-table configuration in the alluvial aquifer in and adjacent
                 to the Rocky Mountain Arsenal, 1955-71.

 may be significant flow across this line. Such a condition can readily
 occur in areas where the bedrock possesses significant porosity and
 hydraulic conductivity, or where recharge from irrigation, unlined
 canals, or other sources is concentrated. Because the hydraulic con-
 ductivity of the bedrock underlying the alluvium is generally much
 lower than that of the  alluvium, ground-water  flow through the
 bedrock was assumed to be  negligible for the purposes of this in-
 vestigation.
                                                                                        MODELING CHLORIDE MOVEMENT, ALLUVIAL A
15
                                                                    The position of the boundary that separates the alluvial aquifer
                                                                  from the areas in which the alluvium is either absent or unsaturated
                                                                  may actually change with time as the water table rises or falls in
                                                                  response to changes in recharge and discharge, although the bound-
                                                                  ary was assumed to remain stationary for the model study. The effect
                                                                  of the changing boundary was most evident in the vicinity of pond A.
                                                                  A map of the water-table configuration during the period when pond A
                                                                  was full (Konikow,  1976) shows that during this time, there was
                                                                  ground-water flow from pond A to the east and northeast into  the
                                                                  alluvial channel underlying the valley of First Creek, in addition to the
                                                                  northwestward flow indicated in figure 4.

                                                                              APPLICATION OF SIMULATION MODEL

                                                                                       FINITE-DIFFERENCE GRID

                                                                    The limits of the modeled area were selected to include the entire
                                                                  area having chloride concentrations over 200  mg/1 and the areas
                                                                  downgradient to which the contaminants would likely spread, and to
                                                                  closely coincide with natural boundaries and divides in the ground-
                                                                  water flow system. The model includes an area of approximately 34
                                                                  mi2 (88 km2).
                                                                    The modeled area was  subdivided into a finite-difference grid of
                                                                  uniformly spaced squares. (See fig. 5.) The grid contains 25 columns
                                                                  (i) and 38 rows 0)- Because of the irregular boundaries and discon-
                                                                  tinuities of the alluvial aquifer, only 516 of the total 950 nodes in the
                                                                  grid were actually used to compute heads (or water-table altitudes) in
                                                                  the aquifer. Each cell of the grid is 1,000 feet (305 m) on each side. By
                                                                  convention, nodes are located at the centers of the cells of the grid. All
                                                                  aquifer properties and stresses must be defined at all nodes of the grid.

                                                                                       DATA REQUIREMENTS

                                                                    Many factors influence the flow of ground water and its dissolved
                                                                  chemicals through the alluvial aquifer near the Rocky Mountain Ar-
                                                                  senal. To compute  changes in chloride concentration, all parameters
                                                                  and coefficients incorporated into equations 1 and 4 must be defined.
                                                                  Thus, many input data are required for the model, and the accuracy of
                                                                  these data will affect the reliability of the computed results. The main
                                                                  input data  requirements  for  modeling chloride movement  in this
                                                                  alluvial aquifer are summarized in  table 1.

                                                                                        AQUIFER PROPERTIES
                                                                    The transmissivity of an aquifer reflects the rate at which ground
                                                                  water will  flow through the aquifer under a unit hydraulic gradient
                                                                  (Lohman and others, 1972). Konikow  (1975)  showed  that  the

-------
16
ROCKY MOUNTAIN ARSENAL, COLORADO
                                       104-50-
             39-55'
 39-50-
              [	ROCKY MOUNTAIN ARSENM.IBOUNPARY __i	\	I

                         n        1        2 MILES
                         0    1     2 KILOMETERS
                           EXPLANATION
              Zero-tran»mi«ivitv cell

              Conitant-head call
         A   Ditposal-pond call. Letter correspond* to designation in figura 2
          I    Irrigation-racharga call
          L   Canal-leakage cell
      	Boundary ot araas in which alluvium it absent or unsaturated

          FIGURE 6.— finite-difference grid used to model the study area.

 transmissivity of the alluvial aquifer in this study area ranges from 0
 to over 20,000 fWd (over 1,800 nWd), and that the saturated thickness
 is gene^fe less than 60 feet (18 m). The highest transmissivities,
                                                                                            MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                                                                                                    17
                                                                              TABLE 1. — Summary of main data requirement* for numerical
                                                                                                   model
                                                                                                   Aquifer pnpertiei
                                                                                                                           Aquifer i
                                                                                            Transmissivity
                                                                                            Storage coefficient
                                                                                            Saturated thickness
                                                                                            Effective porosity
                                                                                            Diaperaivity
                                                                                            Boundaries
                                                                                            Initial chloride
                                                                                              concentration
                                                                                                     Ground-water withdrawals
                                                                                                     Irrigation recharge1
                                                                                                     Canal leakage1
                                                                                                     Disposal-pond leakage1
                                                                                              {Quantity ind qnOHy mutt ba Mined.-
                                                                    greatest saturated  thicknesses, and lowest  hydraulic  gradients
                                                                    generally occur near the South Platte River in the northwestern part
                                                                    of the modeled area. The finite-difference grid was superimposed on
                                                                    the maps of transmissivity and saturated  thickness presented by
                                                                    Konikow (1975), and corresponding values were determined for each
                                                                    node of the grid.
                                                                      The storage coefficient of the aquifer is an approximate measure of
                                                                    the relation between changes in the amount of water stored in the
                                                                    aquifer  and changes in head. Because no changes in head with time
                                                                    occur in steady-state flow, a value for this parameter is needed only
                                                                    for an analysis of transient (time-dependent) flow, which was not con-
                                                                    sidered  in this study.
                                                                      Values of effective porosity and dispersivity of the aquifer must be
                                                                    known to solve the solute-transport equation. Because no field data
                                                                    are available to describe these parameters in this study area, values
                                                                    were selected by using a trial-and-error adjustment within a range of
                                                                    values determined for  similar aquifers in other areas.
                                                                      No-flow and constant-head boundaries used in this model are indi-
                                                                    cated in figure 5. Constant-head boundaries were specified where it
                                                                    was believed that either underflow into or out of the modeled area or
                                                                    recharge was sufficient to maintain a nearly constant  water-table
                                                                    altitude at that point  in the aquifer. Altitudes  assigned to the con-
                                                                    stant-head cells  were determined  by  superimposing  the  finite-
                                                                    difference grid (fig. 5)  on the water-table map (fig. 4).
                                                                      No data were available to describe the chloride concentrations in
                                                                    the aquifer when the Arsenal began its operations. Because more re-
                                                                    cent measurements indicated that the normal background concentra-
                                                                    tion may be as low as 40 mg/1, an initial chloride concentration of 40
                                                                    mg/1 was assumed to have existed uniformly throughout the aquifer in
                                                                    1942.

-------
18
ROCKY MOUNTAIN ARSENAL, COLORADO
                      AQUIFER STRESSES
  No direct measurements of long-term aquifer stresses were availa-
ble. Hence, these factors were estimated, primarily using a mass-
balance analysis of the observed flow field.
  The  areas that had probably been irrigated during most of the
period  from 1943 to  1972 were  mapped from aerial photographs.
These irrigated areas are shown in figure 2. In the model, irrigation
was assumed to occur at 111 nodes of the finite-difference grid, which
represents an area of  1.11x10" ft2 (1.03X107 m2).
  The net rate of recharge from irrigation and precipitation on irrig-
ated areas was estimated through a trial-and-error analysis, in which
the simulation model was used to compute the water-table configura-
tion for various assumed recharge  rates.  Initial estimates of net
recharge  were used  in  a  preliminary calibration of  the model.
Transmissivity values  and boundary conditions in the model were ad-
justed between successive simulations with an objective of minimizing
the differences between observed and computed water-table altitudes
in the irrigated area. The standard error of estimate (or scatter) is a
statistical measure similar to the standard deviation (Croxton, 1953,
p. 119). It is used here to indicate the extent of deviations  between
computed and observed heads. Figure 6 shows that the standard error
of estimate generally  decreased as successive simulation tests were
made.  After about seven tests, additional adjustments  produced only
small improvements in the fit between the observed and computed
water tables.
   A final estimate of the net recharge rate in irrigated areas was
made using the set of parameters developed for the final test of figure
6. Figure 7 shows that the mean of the differences between observed
and computed heads at all nodes in the irrigated area is minimized
 (equal to zero) when a net recharge rate of approximately 1.54 ft/yr
 (0.47 m/yr) is assumed. Also, irrigation recharge was assumed to have
a chloride concentration of 100 mg/1.
   The recharge rate due to leakage from unlined canals was similarly
estimated to  be approximately 2.37 ft/yr (0.72 m/yr),  which  is
 equivalent to 0.40 [ftVsl/mi (0.0070 [mVsl/km). The standard error of
 estimate in this case was about 1.3 feet (0.40 m). Canal leakage was
 assumed to have a chloride concentration of 40 mg/1.
   Changes in the chemical concentration of ground water in irrigated
areas  are partly caused by the mixing (or dilution) of ground water
having one concentration with recharged water having a  different
concentration. Because the magnitude of this change is a function of
the gross recharge, rather than of the net recharge, an estimate of the
 gross recharge must  be made. Hurr, Schneider, and  Minges (1975)
 presented data indicating that the average rate of application of ir-
 rigation water in the  South Platte River valley is about 4.2  ft/yr (1.3
                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL A)
19
                                                                                                   10  11  12  13  14  15 1e
                                                                                    SIMULATION TEST NUMBER
                                                                  FIGURE 6.—Change in standard error of estimate for succedvenmulationtecU.

                                                               m/yr). Hurr, Schneider, and Minges (1975) also stated that 45 to 50
                                                               percent of the applied irrigation water IB recharged to the aquifer.
                                                               Thus, the gross recharge to the aquifer in irrigated parts of the study
                                                               area was assumed to equal 1.9 ft/yr (0.58 m/yr).
                                                                 In the study area irrigation water is  derived both from surface
                                                               water, diverted through canals and ditches, and from ground water,
                                                               pumped from  irrigation wells. The difference between the gross
                                                               recharge and the net recharge, 0.35 ft/yr (0.11 m/yr), was assumed to
                                                               equal the total ground-water withdrawal rate through wells.
                                                                 It was estimated from data presented by Schneider (1962) and Me-
                                                               Conaghy, Chase, Boettcher, and Major (1964) that 62 irrigation wells
                                                               operated in the study area during 1955-71, the period represented by
                                                               the water-table map (fig. 4). Only a small number of wells were drilled
                                                               after 1965 (Hurr and others, 1975, p. 5), so the estimate based on data
                                                               up to 1964 is probably an accurate approximation. By multiplying the
                                                               total ground-water withdrawal rate by the irrigated  area and then
                                                               dividing by  the number  of irrigation wells, the average sustained
                                                               pumping rate per well is computed to be 0.02 ft3/s (5.7xlO~4 ms/s).

-------
20
ROCKY MOUNTAIN ARSENAL, COLORADO
    O.4O
                   NET RECHARGE. IN METERS PER YEAR
                   0.4S
                                                0.56
             1.4
                      1.B       1.6       1.7

                   NET RECHARGE. IN FEET PER YEAR
                                                1.8
                                                         1.9
                                                           0.0 2
FIGURE 7.— Relation between the assumed rate of net recharge in irrigated areas and
         the mean difference of observed and computed water levels.
  Leakage from the unlined disposal ponds at the Arsenal represents
both a significant source of recharge to the aquifer and the primary
source of ground-water contamination in the area. Because no records
were available to describe the variations in discharge of liquid wastes
to the 6 unlined ponds, the general history of their operation was
reconstructed primarily from an analysis of aerial photographs, which
were available in 20 sets with varying degrees of coverage during
1948-71. The summary in table 2 shows that four characteristic sub-
periods were identified during which the leakage rates and concentra-
tions were assumed to remain constant for modeling purposes.

                  CALIBRATION OF FLOW MODEL
  The flow model  computes the head distribution  (water-table
altitudes) in the aquifer on the basis  of the specified aquifer proper-
ties, boundaries, and hydraulic stresses. Because the ground-water
seepage velocity is  determined from  the head  distribution, and
because both convective transport and hydrodynamic dispersion are
functions of the seepage velocity, an accurate model of ground-water
flow is a prerequisite to developing an adequate and reliable solute-
transport model. In general, the flow model was calibrated by compar-
ing observed water-table  altitudes with corresponding computations
of the model.
  Insufficient field data were available to accurately calibrate a tran-
sient-flo^kdel. However, the use of the  disposal ponds varied over
MODELING CHLORIDE MOVEMENT. ALLUVIAL AQUIFER
21
                                                                  TABLE 2. — Generalized history of disposal pond operations at the Rocky Mountain
                                                                                         Arsenal, 1943- 72
                                                                                          IN.A. out applicable)
Y«n
1943-66


1967-60


1961-67


1968-72



A**nf»
A — Full 	
BAB — Full 	
C — 1/2 Full 	
A — Empty 	
B.D.E — Empty 	
C — Full 	
A — Empty 	
BAE — 1/3 Full 	
C — 1/3 Full 	
A — Empty 	
B.D.E — Empty 	
C — Full 	

Compnud
fcnkM.

0.16
.18
.64
.0
.0
1.08
.0
.06
.36
.0
.0
1.08

AMvnrad chloride
eonecntntioD

4,000
3,000
3,000
N.A.
N.A.
1,000
N.A.
600
500
N.A.
N.A.
160

                                                                 time and induced the only significant transient changes noted in the
                                                                 area. Several water-level measurements in observation wells at the
                                                                 Arsenal showed that the water table fluctuated locally by up to 20 feet
                                                                 (6 m), mainly in response to filling and emptying of the unlined ponds.
                                                                 Therefore, the hydraulic history of the aquifer was approximated by
                                                                 simulating four separate steady-flow periods, based on the generalized
                                                                 history of disposal pond operations shown in table 2.
                                                                  The first period simulated was 1968-72, when it was assumed that
                                                                 pond C was full and that all other unlined ponds were empty. Con-
                                                                 stant-head boundary conditions were applied at the 5 nodes corre-
                                                                 sponding with pond C, and the rate of leakage from pond C was com-
                                                                 puted implicitly by the model to be about 1.08 fWs (0.031 ms/s). A com-
                                                                 parison of the heads computed for 1968—72 with the observed water-
                                                                 table configuration for 1955-71 shows good agreement in most of the
                                                                 modeled area. The computed heads were within 2.5 feet (0.75 m) of the
                                                                 observed heads at more than 84 percent of the nodes. The greatest
                                                                 residuals (difference between the observed and computed heads at a
                                                                 node) were between 7.5 and 9.5 feet (2.3 and 2.9 m). Residuals in this
                                                                 range only occurred at less than than 1.5 percent of the nodes, and
                                                                 only at nodes near the disposal ponds, where the greatest variations in
                                                                 observed water-table  altitudes had been measured.  It  must  be
                                                                 emphasized that the general water-table configuration presented in
                                                                 figure 4 represents a composite of water-level measurements made
                                                                 during 1955-71 and is not necessarily an accurate representation of
                                                                 the water-table configuration  at any specific time durij^^Bat period.

-------
22
ROCKY MOUNTAIN ARSENAL, COLORADO
  The observed water-table configuration indicates that a source of
recharge to the alluvial aquifer occurs in  an area  located approx-
imately one-half mile (0.80 km) northeast  of the center of pond C,
which corresponds to the node at (»- 10,> = 25). This recharge might
represent leakage from an unlined canal (Sand Creek Lateral) or the
concentrated discharge of seepage through the unsaturated alluvium
to the east and south. The model analysis indicated  that an average
flux of about 0.10 ftVs (0.0028 mVs) would be required to maintain the
observed hydraulic gradient in this area. This average flux was thus
assumed to have existed at this node from 1943-72. Because this
recharge would probably be uncontaminated, it was assumed to have a
chloride concentration of 80 mg/1.
  Similarly, the water-table map presented by Konikow (1975) indi-
cates that significant recharge may occur in or near the industrial
area located south of pond A and north of the fresh water reservoirs.
Thus, constant-head boundary conditions were applied to the three
nodes located at (i = 5,j = 32-34). The model computed that a com-
bined total of about 0.09 ftVs (0.003 m3/s) of recharge would occur
there during 1968-72. The source of this recharge could be infiltrated
surface runoff from paved areas in the industrial complex. Because
the chloride concentration of some ground-water samples taken in
this area were  slightly  above normal  background levels, it was
assumed that any recharge from this area would have a chloride con-
centration of 200 mg/1.
  The second period simulated  was 1943-56, when  pond C was
assumed to leak at 50 percent of the rate computed  for 1968-72. All
other unlined ponds were assumed to be full during 1943-56 and were
represented as constant-head boundaries in the model. Except for the
changes at the disposal ponds, all other parameters and boundary con-
ditions in the model were identical with the 1968-72 simulation. The
head distribution for 1957-60 was assumed to be the same as during
1968-72 because of the apparent similar use of the disposal ponds.
Therefore, the 1957-60 period did not require a separate flow simula-
tion.
  The third period simulated was 1961-67, when ponds B, C, D, and E
were all assumed to leak at one-third of the rates computed for the
periods when each was full. As in the second simulation period, all
other parameters and boundary conditions in the model were assumed
to be unchanged.
  The flow model calculated a mass balance for each simulation run to
check the numerical accuracy of the solution. As part of these calcula-
tions, the net flux contributed by each separate hydrologic component
of the model was also computed and itemized to form  a hydrologic
budget for the aquifer in the modeled area. The hydrologic budget is
valuable because it provides a measure of the relative importance of
                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL AQBHER
                                                                                                                           23
                                                               each element to the total budget. The hydrologic budgets for the final
                                                               calibrations of the four steady-state flow models are presented in table
                                                               3. The date in table 3 indicate that the major sources of ground-water
                                                               inflow are (1) infiltration from irrigated fields, (2) underflow through
                                                               the aquifer into the study area, (3) seepage losses from the unlined ir-
                                                               rigation canals, and (4) infiltration from the unlined disposal ponds.
                                                               The major ground-water outflow occurs as (1) seepage into the South
                                                               Platte River,  (2) withdrawals from irrigation wells, and (3) underflow
                                                               through the aquifer out of the study area. The computed total flux
                                                               through the aquifer in the study area averages about 14 ft3/s (0.40
                                                               nWs). However, most of this is flowing through the part of the aquifer
                                                               north and west of the Arsenal boundary that receives most of the
                                                               recharge and has the highest transmissivity.

                                                                  TABLE 3. — Elementf of hydrologic budget computed by ground-water flow model
                                                                                                        Computed flux1 OMM
                                                                                                1943-M
                                                                                                        1967-60
                                                                                                                1961-67
                                                                                                                       1868-78
Well discharge 	
Irrigation recharge 	
Canal leakage 	
Pond A 	
Pond B . 	
Pond C 	
Pond D 	
Pond E 	

Industrial area 	
Underflow across:
Southwest boundary 	
Southeast boundary 	
Northeast boundary 	
South Platte River 	
First Creek 	
Sand Creek Lateral 	
Total flux:
Recharge 	
Discharge 	
	 -1.264
	 6.648
	 1.606
	 165
	 022
	 642
	 108
	 060
	 076
	 .019

	 4.713
	 096
	 -.487
	 -12.361
	 -.041
	 100

	 14.133
	 -14.133
-1.264
6.648
1.606
.0
.0
1.083
.0
.0
.081
.094

4.473
.089
-.496
-12.290
-.122
.097

14.171
-14.171
-1.264
6.648
1.606
.0
.007
.361
.036
.017
.086
.147

4.822
.109
-.478
-12.214
.014
.100

13.953
-13.963
-1.264
6.648
1.606
.0
.0
1.083
.0
.0
.081
.094

4.473
.089
-.498
-12.290
-.122
.097

14.171
-14.171
                                                                 'A ixwttY. nlu. in Ou* UN. fawlata* nduri* to th. xprffo in UM nodiM «••; • taftttn nh»
                                                               chart* fan tin aquifer.
                                                                         CALIBRATION OF SOLUTE-TRANSPORT MODEL

                                                                 The solute-transport model applied to the Rocky Mountain Arsenal
                                                               area was designed to compute changes in the chloride concentration
                                                               in the alluvial aquifer during 1943-72. Heads and fluxes computed by
                                                               the flow model were used as input to the transport model. A different
                                                               velocity field was computed for each steady-state flow period outlined
                                                               in table 2.

-------
24
ROCKY MOUNTAIN ARSENAL, COLORADO
  The solute-transport model was calibrated mainly on the basis of
the chloride concentration pattern that was observed in 1956 (fig. 3).
Field measurements of the effective porosity and dispersivity of the
aquifer were not available, so a range of realistic values were tested in
a sensitivity analysis. The computed concentrations were most sensi-
tive to variations in the value of effective porosity and least sensitive
to the transverse dispersivity. A comparison of observed and com-
puted chloride concentration patterns indicated  that an effective
porosity of 30 percent and longitudinal and transverse dispersivities of
100 feet (30 m) were best.
  After appropriate concentrations were assigned to all sources, and
an  initial  background concentration of 40 mg/1 was assigned to all
nodes in the aquifer to represent conditions at the end of 1942, the
transport  model was run for a 14-year simulation period (1943-56).
The model computed a chloride  concentration pattern (fig. 8) that
agreed closely with the observed pattern (fig. 3). The small difference
in the directions of the axes of the main plumes between the observed
and computed  data is probably due mainly to errors in the computed
flow field, rather than to errors in the transport model.
  Since 1956, all disposal has been into the asphalt-lined Reservoir F,
thereby eliminating the major source of contamination. However, that
alone could not eliminate the contamination problem because large
volumes of contaminants were already present in the  aquifer. In
January 1961 sufficient data were again available to contour the pat-
tern of contamination (fig. 9). Although this is more than 4 years after
the source of  contamination  had been apparently eliminated, only
minor changes can be observed in the overall contamination pattern.
These changes include a small downgradient spreading of dissolved
contaminants  and a significant decrease in chloride concentrations
near the center of the contaminated zone. At this time the downgra-
dient spreading was most noticeable near the northeastern limit of
the contaminated zone, where a third distinct plume had formed. Dur-
ing 1957-72 water in the unlined disposal ponds was derived pri-
marily from local surface runoff and canal diversions,  which had
relatively low chloride concentrations. Thus, much of the observed im-
provement in water quality near the center of the contaminated zone
from 1957 to 1961 was probably the result of dilution by recharge from
the former disposal  ponds and from the Sand Creek Lateral.
  The solute-transport model was next used to simulate the period
1957-60, using the  chloride concentrations computed for the end of
1956 as initial  conditions. The chloride concentrations thus computed
for  the end of 1960 (or the start of 1961) are illustrated in figure 10,
which can be compared with the observed pattern for January, 1961
(fig. 9)^jy; computed concentrations show the same general changes
from jthat occurred in the observed chloride pattern. However,
                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER      25
                                                                                                        104*50'
                                                                  39'5O' -
                                                                      104-55'
                                                                                 	KOCKY MOUNTAIN AKSEN All BOUND
                                                                                         0       l       2 MILES
                                                                                         I	r—H	'
                                                                                             1    2 KILOMETERS
                                                                                          EXPLANATION
                                                                        - MO - Line of equal chloride concentration (in milligram t per liter).
                                                                                Interval variable
                                                                        LiSi'^fciJJ Area in which alluvium is absent or untaturated

                                                                          FIGURE 8.— Computed chloride concentration, 1966.

                                                               the model results indicated a more direct discharge toward the South
                                                               Platte River than was observed, and the model did not indicate any
                                                               spreading to the northeast to form  a third plume. Some of this ob-
                                                               served spreading may have been caused by transient changes in the
                                                               flow field that could not be reproduced with the steady-state flow
                                                               model.
                                                                 Available data suggest that recharge of the aquifer was relatively
                                                               low from 1961 to about  1968. Nevertheless, data collected in early
                                                               1969, the next time for which field data were available, d^adicate the
                                                               occurrence of a further significant decrease in the ovei^^pze of the

-------
26
ROCKY MOUNTAIN ARSENAL, COLORADO
   39-50' -
                L	ROCKY MOUNTAIN
                                       ARSENALI BOUNDARY
                                             ..J
                                   i
                                           2 MILES
                           0    I     2 KILOMETERS
                             EXPLANATION
                 Data point (Jan. 1961)

              ~~~ Lin* of equal chloride concentration (in milligrams par litar).
                    Interval variable
                I Area in'which alluvium i* absent or uniaturated

             ?    Position of contour uncertain

         FIGURE  9.— Observed chloride concentration, January 1961.
affected area. (See fig. 11.) Apparently, as  the contaminated water
continued to migrate downgradient, its chloride concentration was
diminished  by  dispersion and dilution.  Also,  the  concentrations
decreased even  more near the former disposal ponds. Chloride con-
centrations greater than 1,000 mg/1 were now limited to only a few iso-
lated areas.
  The observed data from 1961 and 1969 also indicate that some con-
taminants were  present in the aquifer near the freshwater reservoirs,
MODELING CHLORIDE MOVEMENT, ALLUVIAL AQ
   27
                                                                                                                          104*50-
                                                                                   39'SO' -
       |	ROCKY MOUNTAIN AKSENAll BOUND,

                  0       I       2 MILES
_l
                                                                                            0    1    2 KILOMETERS
                                                                                               EXPLANATION
                                                                               	Line of equal chloride concentration (In milligrams par liter).
                                                                                    Interval variable

                                                                                   Area In which alluvium It absent or unsaturated
                                                                         FIGURE 10.— Computed chloride concentration at the §Urt of 1961.
                                                                 adjacent to the industrial area. It is unlikely that these contaminants
                                                                 were derived from the disposal ponds and, thus, were not predicted by
                                                                 the model. The chloride concentration pattern computed for the end of
                                                                 1968 (or start of 1969), using the chloride concentrations computed
                                                                 for the end of 1960 as initial conditions, is presented in figure 12. For
                                                                 the most part, the solute-transport model has also reproduced this
                                                                 period of record, from 1961 to 1969, fairly well.
                                                                   From about 1968 or 1969 through 1974, pond C was apparently
                                                                 again maintained full most of the time by diverting water from the
                                                                 freshwater reservoirs to the south. Available data showed that by

-------
28
ROCKY MOUNTAIN ARSENAL, COLORADO
                                          1O4-5CV
   39'50' -
[	ROCKY MOUNTAIN ARS
                                MOUNTAIN AgSENAll BOUNDARY
                           0       I        2 MILES
                           0    1     2 KILOMETERS
                             EXPLANATION
             •    Data point (Jan.-May 1969)
          	300	 Lin* of equal chloride concentration (in milligramt per liter).
                    Interval variable
          Liv.ijiiJtiJ Ar*8 in wnich elluvium '• abtent or untaturated
             T    Petition of contour uncertain
       FIGURE  11.— Observed chloride concentration, January-May 1969.

 1972 the areal extent of contamination, as indicated by chloride con-
 centration, had significantly diminished (fig. 13), and concentrations
 above 1,000 mg/1 were now limited to just two small parts of the main
 zone of contamination. Because both are areas of relatively low hy-
 draulic conductivity, it appears that low flow velocities have retarded
 the movement of the contaminated ground water out of or through
          areas. Chloride concentrations were almost at normal back-
              in the middle of the affected area. This largely reflected
                                                                           MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
29
                                                                                                                             1W50'
                                                                                     39.50- -
                                                                                              ROCKY MOUNTAIN AR.SEMAL.I BOUN
                                                                                              0    1    2 KILOMETERS
                                                                                                EXPLANATION
                                                                                 	 Line of equal chloride concentration (in milligramt per liter).
                                                                                        Interval variable
                                                                                     Area in which alluvium it absent or untaturated
                                                                           FIGURE 12.—Computed chloride concentration at the ftart of 1969.
                                                                   the infiltration of fresh water from pond C, which had the effect of
                                                                   diluting and flushing the contaminated ground water.
                                                                     The pattern of contamination computed for the start of 1972, by
                                                                   using the chloride concentrations computed for the end of 1968 as ini-
                                                                   tial conditions, is presented in figure 14. The computed pattern agrees
                                                                   fairly well with the observed pattern (fig. 15), although the former
                                                                   shows somewhat longer plumes. After a 30-year simulation, the model
                                                                   has identified (1) the two areas where  high chloride concentrations
                                                                   were still present, (2) the reduction in size and strentfhpf the main

-------
30
ROCKY MOUNTAIN ARSENAL, COLORADO
                                           104'50'
    39-50' -
                 [	ROCKY MOUNTAIN ARSENALl BOUNDARY

                                            2 MILES
                                    1
             0     1    2 KILOMETERS
               EXPLANATION
    Data point (May 19721
o	Lin* of equal chloride concentration (in milligrams per liter).
      Interval variable
                • I  Area in which alluvium it absent or unsaturated
             -•••-**
              1     Position of contour uncertain
            FIGURE 13.— Observed chloride concentration, May 1972.
 plume, and (3) the significant reduction in chloride concentration in
 the middle of the contaminated zone.
   The simulation model computes the velocity of ground-water flow at
 each node of the  model,  but these data could not be independently
 verified with field data. The computed velocities ranged from less than
 1.0 ft/d (0.3  m/d) to over 20 ft/d  (6.1 m/d). Because the computed
 velocity varies greatly within the modeled area, depending on the
 transmissivity and hydraulic gradient, one value of velocity cannot be
MODELING CHLORIDE MOVEMENT, ALLUVIAL AP,
                                                                                                                                                   31
                                                                                                                              1O4-50'
                                                                                      39-5O' -
                                                                                           104
                                                                                               ROCKY MOUNTAIN AUSENAll BOUND
                                                                                               O       1        2 MILES
                                                                                               h
                                                    ..J
                                                                                                              0    1     2 KILOMETERS
                                                                                                                EXPLANATION
                                                                                                 	 Line of equal chloride concentration (In milligrams per liter).
                                                                                                       Interval variable

                                                                                                    Area In which alluvium Is absent or unsaturated
                                                                            FIOUHE 14.—Computed chloride concentration at thecUrt of 1972.
                                                                    used to estimate the average time of travel of dissolved chemicals be-
                                                                    tween two points on a flow line. For problems where this type of infor-
                                                                    mation is desired, the computer program could be easily modified to
                                                                    provide these data.
                                                                      Mass balance calculations were performed during the calibration
                                                                    procedure to check the numerical accuracy of the model simulations.
                                                                    Errors in the mass  balance were always less than 1 percent for the
                                                                    flow  model, but averaged about 14 percent  for the solute-transport
                                                                    simulations. The latter is somewhat higher than desirable and indi-

-------
32
ROCKY MOUNTAIN ARSENAL. COLORADO
   39.50- _
               [	[tOCKY MOUNTAIN ARSENAtlBOU
                                           J
                                         2 MILES
                          0    1    2 KILOMETERS
                            EXPLANATION
         	aoo	Lin* of equal chloride concentration (|n milligrams per liter).
                  Interval variable
          EgO|| Area In which alluvium ii absent or umatureted

 FiOUBl 16.—Chloride concentration predicted for 1980, aiauming that pond C U filled
                    with fresh water during 1972-80.

 cates the need for further refinements in the numerical procedure
 used to solve the transport equation.

                      PREDICTIVE CAPABILITY

   Once a model has been adequately calibrated, it can be used to pre-
 dict or analyze the effects of either future or past changes in stresses
 or boundary conditions. The Rocky Mountain Arsenal model, which
 was calibrated over a 30-year historical period, appeared to be reliable
 enough to be used for limited  predictive purposes. The predictive
                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                                                                                            33
capability of this model can help to (1) isolate the effects of past
measures, (2) evaluate the causes of present and recurring problems,
and (3) predict future concentrations under a variety of assumptions.
Several simulation tests were made, both to demonstrate the potential
uses of this type of model and to better understand the nature of the
ground-water contamination problem at the Arsenal.
  Because of the relative importance of leakage from pond C (table 3),
a simulation test was made to evaluate its possible effect on future
chloride concentrations in the aquifer. One simulation run was made
to predict the chloride concentration in 1980 if pond C were kept full of
fresh water during 1972-80. These results (fig. 15) indicate that if the
artificial recharge due to leakage from pond C were to continue, then
in 1980 there would remain only one area of significant contamina-
tion, which would be confined entirely within the Arsenal boundaries.
There would also still be one small area north  of the Arsenal that
would contain chloride concentrations between 200 and 500 mg/1.
  The importance of the effects of artificial recharge from pond C can
be illustrated by computing what the chloride concentration would be
in 1980 if pond C were not kept full, and then comparing this pattern
with the one in figure 15. Therefore, the  chloride concentrations in
1980 were recomputed after assuming that the minimal recharge rate
from pond C that was estimated for 1961-67 (table 3) had continued
during 1968-80. These results are presented in figure 16 and indicate
that if the artificial  recharge from pond  C would not occur during
1961-80, then in 1980, which is about  25 years after the sources of
ground-water contamination were supposedly eliminated, there would
still be two relatively large areas of contaminated ground water re-
maining.
  A comparison of the results presented  in figure 16 with those shown
in figure 15 indicates that the effects of artificial recharge from  pond
C had significantly increased the rate of water-quality recovery in the
aquifer. In addition to the effects of dilution, the recharge created a
mound on the water  table, which increased the hydraulic gradients,
and consequently, the  flow velocities. In  effect, it "pushed" or
"flushed" the contaminated ground water out of the area faster than
would have occurred naturally. The apparent difference in the mass of
chlorides present in the aquifer between  figures 15 and 16 is caused by
two factors. First, in figure 15 a  greater percentage  of the  total
chloride  is contained in areas having a concentration less than 200
mg/1. Second, during the simulation period upon which figure  15 is
based, a greater mass of chlorides has discharged from the aquifer in
the modeled area, mostly to the South  Platte River and as ground-
water underflow across the model boundaries.
  By comparing figures 15 and 16, it can also be inferred that it would
probably take at least many decades for this contaminated aquifer to

-------
34
ROCKY MOUNTAIN ARSENAL, COLORADO
                                        104-501
   39.50- -
               [	ROCKY MOUNTAIN ARSENALl BOUND

                          0        l       2 MILES
                                           J
                                         I
                          0    I    2 KILOMETERS
                            EXPLANATION
         	900	Lin* of equal ctilorid* concentration (in milligrami per liter).
                  Interval variaole
          ^^^^1 Area in which •lluvlum it abMnt or untaturated

 Fwtmi 16.— Chloride concentration predicted for 1980, assuming that recharge from
                   pond C is piinimul during 1961-80.

 naturally recover its original water-quality characteristics. But it can
 also be inferred that appropriate water-management policies can help
 to reduce this restoration  time to  the order of years, rather than
 decades.
  More than  one-half of the total ground-water flux through the
 modeled area is derived  from recharge  from irrigation applications
 and canal leakage. (See table 3.) A simulation test was designed to
 demonstrate that these sources of recharge have an important effect
 on the chemical concentrations in the aquifer. A simulation run was
         MODELING CHLORIDE MOVEMENT, ALLUVIAL Ar.
made for 1943-56, assuming  zero  dilution from these  recharge
sources. The chloride concentrations computed for 1956 under this
assumption were then compared with the computed pattern shown in
figure 8. If there were no dilution from recharge, in 1956 the area con-
taminated with chloride concentrations greater than 1,000 mg/1 would
have been slightly larger than the area within the 500 mg/1 isochlor
shown in figure 8, and the higher concentrations would have spread to
the north  much earlier. Therefore, dilution from irrigation recharge
and seepage from unlined canals were both important factors in
reducing the level of chloride concentrations downgradient from the
Arsenal.

          APPLICATION TO WATER-MANAGEMENT PROBLEMS
  Changes in water use or water management in an area can signifi-
cantly affect both the flow and chemical quality of ground water.
Because a wide variety of alternative decisions or policies are possible
regarding water planning,  management, and quality  control,  it is
difficult to determine the optimum set of alternatives that will mini-
mize detrimental effects and maximize beneficial effects. Thus, an ac-
curate solute-transport model can be a valuable planning tool because
it can help to analyze the relative sensitivity of the aquifer system to
different management alternatives and demonstrate the impact of
specific practices on the chemical quality of ground water.
  To demonstrate this use of the model, a hypothetical change in
ground-water  management  at the  Rocky  Mountain  Arsenal  was
simulated. This illustrative example evaluates a proposal (Konikow,
1974) to maintain hydraulic sinks along those parts of the northern
boundary  of the Arsenal under which the plumes of contaminated
ground water were moving. Construction of the hydraulic sinks could
physically involve either drilling a line of wells for pumping or ex-
cavating a trench or ditch below the water table for drainage. The
main purpose of the hydraulic sink would be to intercept and remove
contaminated  ground water  from the aquifer  before it  migrates
downgradient from the Arsenal. This proposal is evaluated here only
to demonstrate the general value of the model and is intended neither
to endorse this particular plan nor to suggest that any changes in
water management should necessarily be implemented at the Rocky
Mountain  Arsenal.
  The hydraulic sinks could greatly modify the heads  and  flow
velocities in their vicinities. Therefore, these hydraulic changes would
have to be computed before the effects on solute transport could be
predicted.  For this example problem a simplified scheme was used to
generate hydraulic sinks in the flow model that was calibrated for the
period 1968-72. Two sinks, A and B, were represented by  imposing

-------
36
ROCKY MOUNTAIN ARSENAL, COLORADO
 39-5O- -
                        ROCKY MOUNTAIN ARSENALJBOUNDARY

                        O       1        2 MILES
                        O    1    2 KILOMETERS
                          EXPLANATION
               Line of conitant water-level decline, in feet. Interval variable

               Area in which alluvium is absent or unsaturated

               Hydraulic sink
                 Constant-head altitude at A-5.085 ft
                 Constant-head altitude at B-5.124 ft

 FIGURE 17.— Computed drawdown  caused by maintaining two constant-head rinks
          along the northern boundary of the Rocky Mountain Arsenal.
 constant-head boundary conditions at appropriate nodes in the finite-
 difference grid  of the model. Sink A was along the northwestern
 boundary and was simulated as a constant head of 5,085 feet (1,550 m)
 at nh (i = 5-9, j - 17). Sink B was along the northern boundary
MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                                                                                                            37
                                                                 and was simulated as a constant head of 6,124 feet (1,562 m) at nodes
                                                                 (14,19), (16,20), and (16,21).
                                                                  The effect of these two constant-head sinks on steady-state ground-
                                                                 water flow  is illustrated in figure 17, which depicts the computed
                                                                 drawdown (water-level  decline) relative to the heads computed for
                                                                 1968-72. The water-level declines extend a  greater distance to the
                                                                 south and southeast of the sinks than to the north and northwest
                                                                 because of the much lower availability of recharge to the south and
                                                                 southeast. The model computed that the discharge of sink A would be
                                                                 about 0.98 ft'/s  (0.028 mVs), whereas discharge of sink B would be
                                                                 about 0.15 ftVs (0.0042 mVs). The greater computed discharge of sink
                                                                 A reflects its location in an area of higher transmissivity and its closer
                                                                 proximity to the outer constant-head boundary of the model. The ex-
                                                                 tent and magnitude of drawdown shown in figure 17 indicates that hy-
                                                                 draulic gradients, and resulting flow velocities and solute transport,
                                                                 would be  significantly affected if the hydraulic sinks were actually
                                                                 present.          i
                                                                  The two hydraulic sinks were assumed to have begun operating in
                                                                 1968, and pond C was assumed to have remained full after 1968. The
                                                                 chloride concentration pattern then computed for 1980 is shown in
                                                                 figure 18, which represents the combined effects of artificial recharge
                                                                 from pond C and artificial drainage to the two hydraulic sinks. This
                                                                 pattern can be compared with the one shown in figure 15, which just
                                                                 represents  the effect of artificial recharge from pond  C. The  most
                                                                 noticeable difference is that figure 18 shows the persistence of higher
                                                                 chloride concentrations in larger areas downgradient (north) of the
                                                                 hydraulic sinks.
                                                                  Although it may seem anomalous to produce less water-quality im-
                                                                 provement with a source-sink combination than with a source alone,
                                                                 this occurrence is not unreasonable and can be explained with the aid
                                                                 of a schematic cross-section through the source of artificial recharge
                                                                 and the hydraulic sink. (See fig. 19.) Figure 19 shows both the original
                                                                 water table with recharge from the unlined pond and the new steady-
                                                                 state water-table position and directions of flow that were established
                                                                 after pumping began. The change in hydraulic gradients caused by the
                                                                 sink  indicates that the velocity of ground-water flow would increase
                                                                 on the upgradient side of the  sink. On the downgradient side of the
                                                                 sink, there would be a reversal of the hydraulic gradient within a
                                                                 small area near the sink. However, just beyond the area of reversal is
                                                                 an area where the new gradients would be  much less than before,
                                                                 creating a  zone of near stagnation in the aquifer. Because of ex-
                                                                 tremely low flow velocities in this zone, any contaminants that were
                                                                 present in this area at the time the sink was constructed would re-
                                                                 main in the area much longer than if no sink had b^& constructed.

-------
38
ROCKY MOUNTAIN ARSENAL, COLORADO
                                          104'StV
    39'50' -
                           ROCKY MOUNTAIN ARStNALJ BOUNDARY

                                   I        2 MILES
                                             .J
                          '0
                           I-
                           0    I     2 KILOMETERS
                               EXPLANATION
                   Lino of equal chloride concentration (In milligrams per liter).
                     Interval variable

                   Area In which alluvium it absent or unsaturated

                   Hydraulic link
 FIGURE 18.— Chloride concentration predicted for 1980, assuming that artificial
    recharge from pond C ia coupled with drainage through two hydraulic sinks.

   The construction of the hydraulic sinks would slightly increase the
 rate of water-quality improvement between 1968 and 1980 in the area
 between the source and the sinks. In fact, the results of the simulation
 run using sinks indicate that the chloride concentrations in the area
 between the source and the sinks would virtually reach an equilibrium
 pattern after about 10 years.
   Comparison of figures 15 and 18 indicate that water management
 using a source-sink combination as postulated here would produce less
                                                                                          MODELING CHLORIDE MOVEMENT. ALLUVIAL AQUIFER
                                                                                                                                 39
                                                                                     SOUTH
                                                                                   6200-
                                                                                   6050
                                                                                                                                             164O
                                                                                                            O      1 KILOMETER

                                                                                                            EXPLANATION
                                                                                              Saturated alluvium
                                                                                              Bedrock
                                                                                                                     Original water level
                                                                                                	Water level after pumping
                                                                                                 •*-   Direction of flow after pumping
FIGURE  19.— Generalized cross-section from vicinity of source of artificial recharge
               through hydraulic sink (represented aa a well).
desirable results overall than would a simple artificial recharge source
alone. The primary reason for this appears to be that the sinks were
placed near the middle of the contaminant  plume. In the study area
this type of scheme would be most effective if the hydraulic sinks were
located at or just beyond the maximum extent of the plume. Similarly,
if an artificial recharge source is to be used for flushing and dilution, it
should be placed just upgradient of the contaminated zone, if possible.
In general, designs would depend on the circumstances of each specific
problem and should be evaluated individually.
  At the Rocky Mountain Arsenal, the hydraulic sinks would provide
a secondary benefit that is not apparent in  the map of chloride con-
centration. Drainage due to the hydraulic sinks would produce water-
table declines that might prevent soil salinity problems caused by high
water  tables in contaminated areas. This  same effect would also
reduce possible discharge of contaminated ground water to springs or
streams, such as First Creek.
  Removing pollutants from a contaminated aquifer may seem to be
an almost impossible task. While this may be true for some or even
most contaminated aquifers, others may be highly amenable to one or
more plans for artificial reclamation that could significantly acceler-
ate the  rate of water-quality improvement in  the  aquifer. The
feasibility of any such  reclamation plan would be strongly dependent
on the hydraulic properties of the aquifer and on the type and source
of contamination. An accurate model of the aquifer is a prerequisite
for reliably predicting changes in concentration and thereby estimat-
ing the benefits that might be derived from a given plan.

-------
40
              ROCKY MOUNTAIN ARSENAL, COLORADO
                  SUMMARY AND CONCLUSIONS
  The movement of dissolved chemicals in flowing ground water can be
simulated with a computer model that couples a finite-difference solu-
tion to the ground-water flow equation with the method-of-charac-
teristics solution to the solute-transport equation. The usefulness of the
model was illustrated by its application  to a problem of ground-water
contamination in the Rocky Mountain Arsenal area, Colo., where the
model integrated the effects of several factors that controlled changes
in chloride concentrations and successfully reproduced the record of
chloride contamination observed during a 30-year period.
  The method of characteristics offers one workable solution tech-
nique to the solute-transport equation. Although every ground-water
contamination problem is in many  ways unique, the processes  con-
trolling solute transport are the same.  Thus, this method-of-charac-
teristics model is generally applicable  to a wide variety of ground-
water contamination problems.
  The predictive accuracy of the model is most limited by the  ade-
quacy of the input data. The results of applying the model to the
ground-water contamination problem at the Rocky Mountain Arsenal
indicate that where adequate hydrogeologtc data are available, the
model can be used to predict the rates and directions of spreading of
conservative (nonreactive)  contaminants from known or projected
sources.
  The predictive capability of the model can be helpful in designing a
monitoring network. By indicating the most probable and least proba-
ble areas of future contamination and the rate of spreading, optimal
locations and sampling frequencies for observation wells can be deter-
mined. The model can also indicate areas where contaminated ground
water might seep into surface water.
  In some cases of aquifer contamination it may be both physically
and economically feasible to institute a reclamation, program to im-
prove or control the quality of ground water. Because a large variety of
water-management plans can be proposed for any one problem, an ac-
curate model of flow and solute transport in the aquifer could be an in-
valuable tool for planning an efficient and effective program.
  Conclusions of this study that pertain specifically to the Arsenal can
be summarized as follows:
1. Ground water in the alluvial  aquifer  in and adjacent to the Rocky
     Mountain Arsenal flows predominantly to the north  and north-
     west. The alluvium generally has the greatest transmissivity,
     saturated thickness, and rate of recharge in the area between the
     South Platte  River and the north  and northwest boundaries of
     the^pcky Mountain Arsenal. Within the boundaries of the Ar-
     s^^pbhe main source of ground-water recharge since 1943 has
     amHrentlv been infiltration from the unlined disooRal oonda.
        MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
41
2. The pattern of chloride contamination was principally controlled by
    the rates and directions of ground-water flow. Thus, variations in
    the hydraulic conductivity (or transmissivity) and boundary con-
    ditions of the aquifer directly influence solute transport. The ob-
    served contamination pattern clearly indicates that the spread-
    ing of contaminants was significantly restricted or constrained
    by the areas in which the aquifer is either absent or unsaturated.
    Model analyses support the hypothesis that dilution, both from ir-
    rigation recharge and from seepage from unlined canals, was an
    important factor in reducing the level of chloride concentrations
    in the contaminated ground water that flowed past the bound-
    aries of the Arsenal.
3. Most of the contaminants that seeped into the aquifer were proba-
    bly derived from the overflow ponds  (ponds B, C, D, and E).
    Because the original primary disposal pond (pond A) was located
    in an area where the aquifer had a relatively low hydraulic con-
    ductivity, only small quantities of contaminated ground water
    could seep through the aquifer downgradient from pond A.
4. In 1972, approximately 16 years after the main sources of ground-
    water contamination had been eliminated, a large area of the
    alluvial aquifer still contained chloride concentrations that were
    significantly above normal background levels. But the magnitude
    and extent of contamination, as measured by the chloride con-
    centration, had significantly diminished in comparison with ob-
    servations during 1956 and 1961.  By  1980 high chloride con-
    centrations (that is, greater than 200 mg/1) will probably occur in
    only two comparatively small areas. One is north of the Arsenal,
    near First Creek, and the other is near and downgradient from
    the site of pond A.
6. The diluting and flushing effects of freshwater recharge from pond
    C contributed to a significant reduction in the concentrations and
    total quantity of chlorides present in the aquifer.
6. If the Rocky Mountain Arsenal or a similar industrial plant were
    first beginning operations today, this model could be used with
    reliable aquifer descriptions to predict the magnitude and extent
    of ground-water contamination that could be expected to result
    from  the disposal of nonreactive liquid  industrial wastes into
    unlined ponds. Perhaps more importantly, the model could have
    been  used either to determine where the disposal ponds should
    have been located to minimize the extent of contamination or to
    demonstrate, in the first place, that this particular method was
    inappropriate for liquid waste disposal in this particular environ-
    ment and that an alternative method was nee
7. The stringent date requirements for applying the
    mnAol r»ftinf«»H mit rfofiHf>nrtf>R in Hntj» extfltinor at the start of this"

-------
42
ROCKY MOUNTAIN ARSENAL, COLORADO
     investigation. The subsequent analysis and reinterpretation  of
     existing hydrogeologic data led to a revised conceptual model that
     accounted  for the effects  on  ground-water  flow  and solute
     transport of the areas in which the alluvium either is absent or is
     unsaturated most of the time.
8. The model for chloride  movement cannot predict the  behavior  of
     nonconservative (reactive) chemical species. But any comprehen-
     sive study or management plan of ground-water quality at the
     Arsenal would need to include any such reacting species with ap-
     propriate modifications to the model.

                          REFERENCES CITED

Bear, Jacob, 1972, Dynamics of fluids in porous media: Am. Hsevier Publishing Co.,
    New York, 764 p.
Bredehoeft, J. D., and Finder, G. F., 1973, Mass transport in flowing groundwater:
    Water Resources  Research, v. 9, no. 1, p. 194-210.
Croxton, F. E., 1953, Elementary statistics with  applications in medicine and the
    biological sciences: New York, Dover Publishers, Inc., 376 p.
Engineering News-Record, 1956, Asphaltic membrane is used to leakproof a lake: v.
    157, no. 21, p. 40-41.
Carder, A. O., Peaceman, D. W., and Pozzi, A. L., Jr., 1964, Numerical calculation of
    multidimensional miscible displacement by the method of characteristics: Soc.
    Petrol. Eng. Jour., v. 4, no. 1, p. 26-36.
Hurr, R. T., Schneider, P. A.. Jr., and Minges, D. R.. 1975, Hydrology of the South Platte
    River valley, northeastern  Colorado: Colorado Water Conserv.  Board  Water-
    Resources Circ. 28, 24 p.
Konikow, L. F., 1974,  Reclamation of a contaminated aquifer [abs.J, in  Abstracts with
    programs: Geol.Soc. America, v. 6. no. 7, p. 830-831.
	1975, Hydrogeologic maps of the alluvial aquifer in and adjacent to the Rocky
    Mountain  Arsenal, Colorado: U.S. Geol. Survey Open-File Rept. 74-342.
      . 1976, Modeling solute transport in ground water, in Internal. Conf. on Environ-
    mental Sensing and Assessment: Las Vegas, Nev., art. 20-3.
Konikow, L. F., and  Bredehoeft, J. D., 1974, Modeling flow and chemical quality
    changes in an irrigated stream-aquifer system: Water Resources Research, v. 10,
    no. 3, p. 546-562.
Lohman, S. W., 1972, Ground-water hydraulics: VS. Geol. Survey Prof. Paper 708,70 p.
Lohman, S. W., and others, 1972, Definitions of selected ground-water terms—Revisions
    and conceptual refinements: U.S. Geol. Survey Water-Supply Paper 1988,21 p.
McConaghy, J. A., Chase, G. H., Boettcher, A. J., and Major, T. J., 1964, Hydrogeologic
    data of the Denver Basin, Colorado: Colorado Water Conserv. Board Basic-Data
    Rept. 15, 224 p.
Petri, L. R., 1961, The movement of saline ground water in the vicinity of Derby, Col-
    orado: in Ground Water Contamination Symposium: Robert A. Taft Sanitary Eng.
    Center Tech. Rept. W61-5, p. 119-121.
Petri, L. R., and Smith, R. O., 1956, Investigation of the quality of ground water in the
    vicinity of Derby,  Colorado: VS. Geol. Survey open-file report, 77 p.
Pinder, G. P., 1970, A digital model for aquifer evaluation: VS. Geol. Survey Tech-
    niques Water-Resource Inv., book 7, chap. Cl, 18 p.
Pinder, G. F., and Bredehoeft, J. D., 1968, Application of the digital computer for
    aquifer evaluation: Water Resources Research, v. 4, no. 5, p. 1069-1093.
                                                                                                      MODELING CHLORIDE MOVEMENT, ALLUVIAL AQUIFER
                                                                                                                                                   43
                                                                          Pinder, G. F., and Cooper, H. H., Jr., 1970, A numerical technique for calculating the
                                                                              transient position of the saltwater front: Water Resources Research, v. 6, no. 3, p.
                                                                              876-882.
                                                                          Prickett, T. A., and Lonnquist, C. G., 1971, Selected digital computer techniques for
                                                                              groundwater resource evaluation: Illinois State Water Survey Bull. 65, 62 p.
                                                                          Reddell, D. L., and Sunada, D. K., 1970, Numerical simulation of dispersion in ground-
                                                                              water aquifers: Colorado State Univ. Hydrology Paper 41, 79 p.
                                                                          Robertson, J. B., 1974, Digital modeling of radioactive and chemical waste transport in
                                                                              the Snake River Plain aquifer at the National Reactor Testing Station, Idaho: VS.
                                                                              Geol. Survey Open-File Rept. IDO-22054, 41 p.
                                                                          Robson, S. G., 1974, Feasibility of digital water-quality modeling illustrated by applica-
                                                                              tion at Barstow, California: VS. Geol. Survey Water-Resources Inv. 46-73, 66 p.
                                                                          Scheidegger, A. E., 1961, General theory of dispersion in porous media: Jour. Geophys.
                                                                              Res.,  v. 66. no. 10, p. 3273-3278.
                                                                          Schneider, P. A., Jr., 1962, Records of logs and selected wells and teat holes, and chemi-
                                                                              cal analyses of ground water in the South Platte River basin in western Adams and
                                                                              southwestern Weld Counties, Colorado: Colorado Water Conserv. Board Basic-Data
                                                                              Rept. 9, 84 p.
                                                                          Shukle, R. J., 1975, 1974-76 groundwater study of the Rocky Mountain Arsenal  and
                                                                              some surrounding area: Colo. Dent, of Health, Denver, Colo., 20 p.
                                                                          Walker, T. R., 1961, Ground-water contamination in the Rocky Mountain Arsenal area,
                                                                              Denver, Colorado: Geol. Soc. America Bull., v.  72, p. 489-494.
                                                                          Walton, Graham, 1961, Public health aspects of the contamination of ground water in
                                                                              the vicinity of Derby,  Colorado: in Ground Water Contamination Symposium:
                                                                              Robert A. Taft Sanitary Eng. Center Tech. Rept. W61-5, p. 121-125.
                                                                          Wood, L. A., 1972, Groundwater degradation — Causes and cures:  Proc. 14th Water
                                                                              Quality Conf., Urbana, Dl., p. 19-26.

-------
          • •       ' *•' \ '»« " f .' i — -.'.,•..' i,"* •* "v.'VJ -i*1^'-". -^-V."v^. ,-^'i . *!.-..• -"-^v''-^i^"ri^jy^v/.j^sftt^i.5";'?\c'*"^'.''"T»-' j-'s't .^ii>i5i' 0.-"J'":'"ij*\,i %'. i  •"-'
                   .... -;.•'. v' .> Ji" . . ;• _ •.*.••' • V|S«?t - : *> •_;- ••*".  •  : *•?" .- •- .-v- «jf* . *- Jf^V-**24'*-* «;v '-t,V**'5,"4. VN-.-;^--.* - •• .^v?S'5ji?."i^rJ ;*,•,-** •*> "i;,-.: i-» - ' • »


         STUDIES IN HlbPHYSICS 1   S??^:?;;
                                                             ;v.--fe-";-
                  .'••."•   # V" •v^v-;.- "- ,*w*:-. "•*' • * •-  •>.   -  ,• •  •' '. -     .  •  -   'f'-*"    v- -••.- • .-. '


                  '^'^•^•^i^^


                  '     ''' '                                         •••••••-•-•
                                                                                           •-•

,. :v-,:.:.^- .-.
 ^ '  " .    ^V   ™  -  _^ .^-. ^_ _^^  — ,^ — _^ _— .^^ —  ^ .^  —  _^ —      " > •  •  •  . -J'*" •  '           •  • •. ' '-^  . r- ^ •  "*.
         Geophysks Study Com
                .; :-.;W.:,J
-------
                             Groundwater  Contamination
                             and  Aquifer  Reclamation   at  the
                             Rocky   Mountain  Arsenal,
                             Colorado
 6
                              l.KONARD I-.  KONIKOW
                              U.S. Geological Survry

                              DOl'CLAS \V. THOMPSON
                              I'.S. Army Corfi.v of Eiifii
                              ABSTRACT
pin
                              Croimdvvatcr contamination .it tin- Kix-ky Mountain Arsenal. Colorado. is related In the dispos.il ol lu
                              industrial wastes and (u industrial leaks and spills thai lia\r occurred during llir 10-vr lnslnr\ nl opri.ilioi,
                              of the Arsenal.  From 1913 to 195ft llic liquid wastes were discharged into milmrcl ponclv which leMilled MI
                              contamination of part of the underlying alluvial aquifer. Since l!)5(i.  disposal lias liren accomplished l>\
                              discharge into an .ispli.ilt-lincd reservoir. which significantly reduced the volume ol contaminants ciitcrim:
                              the aquifer.  In the mid-1970s toxie organic elieinieals were deleeted outside ol the Arsenal in llie alliivi.il
                              U(|iiilei The Colorado Department of Health issued three orders, \\lijeh ealli-d for ilia h.ilt to nn.intlioi i/< i!
                              dischariii's. (2)  cleanup, and (3) uionndwaler nioiiitiiriim Snlisei|iiently  a management eomniitmeiil ».<••
                              made to mitigate tile prolilem. A pilot tJmiimlu.ilcr containnienl and trealinc'nt system was ninslriM h oiindaiy. (2i Irealiim the ».iler with an uctiv.iteil . .n!"n,
                              priKrss, and (3) injecting the treated water on the dowiitfradient side of the harrier tliiiiuiih several reili.iiX'
                              wells. Because ol the success ol the pilot operation, it is IMMIIK expanded at present to intercept most ol lln
                              contaminated underflow crossing the entire north iKiimdary. llow<'ver. homnlary interception alone eannoi
                              achii'M ai|iiiler restoratiiin at the Arsenal. It is anticipated that the overall final program will also li;i\e to
                              include elements of source containment and isolation, source elimination, process modification to reduce tin
                              volume of wastes generated, and development of alternative waste-disposal procedures th.il are nonpolliilinu
                              A variety ol alternatives have IHVM proposed and are currently hcinu evaluated to determine the most leasllili
                              for implementation. The research plannini;. and design studies that are neeessarx  to achieve the reclamation
                              goal at (lie Arsenal illustrate that an ellcclive a<|iiifei restoration program is dillicnlt to drsiun and c\|X'iivi\<
                              to iniplemenl.



l\'TR()ntK'TIO\                                         I'iisi-s xvlicrc a Sfrious H'-omulxvater contamination priililein i'X-
                                                           ists. (lie single most iiii|Xiilant rcinedi.il aclinn tliul can lit' taken
The I'ontaniination of a grovnulwatei resiMirif is u serious pro!)-      is to eliminate the source of contamination Hut  even then.
lem that  cat) have  loiin-terin cconoinn and physical ciinse-      coiitaininatits already in the a<|iiiln xvill continue to  miuial.
(piences Ix'causc  in most cases the problem  is  neither easily      and spread unless some action is taken to innnoliih/c. ueu-
nor cjnickly remedied. Wood (1972) concluded, "The most sal-      trali/e, or IXMIIOVC them. Hence, there is often a need In clean
isfaclory cure lor grouiulwaler pollution i. preventiini.'  In many      up or restore eontaminatc'd aquifers.

                              93

-------
94
    l.r.ONUM) K  KONIkOVA 1111(1 IMU CI.AS  \\  THOMPSON
  Tin' "rcstorabilit)  nl a contaminated ai|HJlcr is dependent
on thehydrogcologicandgeochcmical properties ol the allccled
aquifer and on tin* chemical aiul physical properties ol  the
contaminant. Restoration ol a contaminated aqnitcr is neither
technically nor economically IcasiMc  in many  cases.  Factors
frequently hindering restoration include (I) tlie slow diffusive
nature ol  gronndwatcr How. (2)  (lie dillicnlty of defining scc-
ondary permeability effects. (.'}) tlie generally low  oxygen eon-
tent and lack of biologic reactivity in grouudvvatcr. (4) (lie re-
tention ol some chemicals in the aquilcr because they tend to
he siirlx'd hy minerals  in the rocks making up the aquifer. (5)
the lack ol transfcrability ol some restoration techniques from
one  site to another, and Ifi) the lack ol knowledge almut  the
source ol the contamination.
  Effective aquifer restoration programs, il technically feasihle.
are Ixilli dillicnlt to design and expensive' to iniplement. Never-
theless. in res|)onse  to pulilic or governmental demands for
positive action in clearly documented cases where gronudwatcr
contamination threatens pnl>lic  health, aqniler cleanup pro-
"rams are being re<|uired anil instituted more frequently.  Some
programs are being financed and operated hy the  federal gov-
ernment. Examples include the Kocky Mountain Arsenal. Col-
orado,  where irrigation and domestic water-supply wells in
adjacent areas have Keen contaminated from industrial wastes
stored  at  tin- Arsenal,  and also \Vurtsmith Air Force  Base.
Michigan, where toxic  organic solvents used in aircraft  main-
tenance have entered and spread through the underlying aqui-
fer. Other programs may he implemented I >ccanse  of violations
of federal  regulations.  Kor example, a recent Justice Depart-
ment suit was filed  in North Carolina under  the  imminent
ha/urd provision of the Resource Conservation and Hccovcry
Act. the suit asks that the defendants". . . permanently restore
the aqniler to a condition commensurate with sale  human use"
\Hazartluns  \\'ttntc .Yens 2(2'. Jan. 21. 19SO.  p.   12).  As an
example of an aquifer restoration program being initiated  he-
cause of state regulations, a chemical company  in northern
Michigan has come to an agreement with the state  of Michigan
to remove the contaminants from the  soil and gronndwater at
their former dump site; the projected cost is $12 million to S15
million (The Wall Stn-el Journal. Sept. 25, UWI.  p. 48).
  General management options  lor restoring water quality in
aquifers currently available include the following: (1) eliminate
the source ol contamination but allow restoration to proceed
only through  natural Hushing,  dilution, and  gcochcmical or
biological  reactions;  (2) accelerate removal ol contaminants
through withdrawal wells,  drains, or  trenches: (3) accelerate
Hushing with artificial recharge; (4) install "impermeable" bar-
riers to contain a contaminated area; (5) induce in situ chemical
or biologic reactions that would neutralise or immohili/.e  (he
contaminant; and ((i) excavate .mil remove the contaminated
part of the aquifer. The selection of the best  approach for a
paiticular situation requires the ability to predict changes in
How ant) chemical concentration in the aquiler lor each possible
management alternative. This iu turn  requires both adequate
field data to describe the aquifer systems and the development
ol accurate simulation  models to define the grouudwater How
system, pollutant-transport mechanisms, and  nature and rate
ol chemical or biological reactions.
                                            COLORADO
 —i
                              5        10 MUSS

                                10 KILOMETERS
 KKil'hT* () I  IjK-.ition ol sliuly area
  This  chapter  locuscs  on the  groundwatcr  contamination
problem at the Kocky Mountain Arsenal, which is located near
Denver, Colorado (sec Figure (>. 1). This area is well suited for
serving as a case study to illustrate data requirements, inves-
tigative approaches, anil management options  related to the
reclamation of contaminated aquifers because (I) the 40-yr his-
tory ol groundvvatcr contamination  is  relatively  well docu-
mented  in the scientific and engineering literature; (2) the
geology and hydrology of the area are  fairly well known; (.'})
adequate, though limited, water-quality data are available to
calibrate numerical simulation models; (4) the  locations and
strengths ol contaminant sources can be approximately recon-
structed; (5) a management commitment has been  made to
aquifer reclamation; ant) (ft) construction, operation, and eval-
uation ol a pilot  reclamation system at the Arsenal have been
completed.
DF.SCHII'TION OF  SITDY  AUK A

History of Contamination

The Hockv  Mountain Arsenal has been operating since  1942.
primarily manufacturing and processing chemical warfare prod-

-------
Coiitiiiiiiiiution ami Aquifer Rtrlinnatiun
lifts and pesticides. These operations have produced liquid
wastes that contain coniplev organic and inorganic chemicals,
including a characteristically  high chloride concentration that
apparently ranged up to about 5(XK) mg/L.
  The liquid wastes were discharged to several iinlined ponds
(Figure 6.2). resulting in the contamination ol the underlying
alluvial aquifer. On the basis of available records, it is assumed
that contamination first occurred  at  the  beginning of  19-43.
From  1943  to  1956 the  primary disposal was into pond A.
Alternate and overflow discharges were collected in ponds B,
C. D. and E.
  Much of the area north of the Arsenal is irrigated. Imth with
surface water diverted from one of the irrigation canals, which
are also iinlined. and with gromulwatcr pumped from irrigation
well:..  Some damage  to crops irrigated with shallow ground-
water was observed in 1951. 1952,  and 1953 (Walton. 1961).
Severe crop damage was reported during  195-4,  a vear when
the annual precipitation, was about one hall the normal amount
and groundwater use  was heavier than normal (Fetri, 1%1).
  Several investigations have been conducted since 195-4 to
determine Itoth the cause of the problem and how to prevent
further damage. Petri and Smith (1956) showed that an area ol
contaminated gronndwater of several square miles existed  north
and northwest of the disposal ponds.  These data clearly indicate
                                      IW50-
r~
i
I 	 J
0
t-
0

—mm .1,
- _ *
• Irrigation
1 '"*•«., ,««-,... v
1
1 KIM'KV MIM'NTAIN AJ^SKNAI. HIH'MiABV 	
1 1 HatS
1 t «ILO*IC1Cn
EXPLANATION
!••>' ^^» unllrwd rMCrvoir
«Mll ^^ Lined rlMrvoir
J
J




KKil'Ht* 6.2  Major Imlrulogic features: Inters iiuiicati' dis|x>s.il-|XHnl
ilrsimutions assigned In lli<- I'.S. Army iKimikovv. 19771.
that the lii|iiid wastes seeped out ol (lie iinlined disposal ponds.
infiltrated the underlying alluvial aquilcr. and migrated down-
gradient toward the South I'latle Hiver. To prevent addition,il
contaminants from entering the aqniler. a KKI-acre (0.045-knr'1
eva|xiration pond (reservoir F) was constructed with an asphalt
lining in 1956 to hold all subsequent liquid wastes. Although
the liner eventually tailed,  even  if the lining were to have
remained totally  impervious, this  new  disposal pond  in itself
would not eliminate the contamination problem because large
amounts ol contaminants were alrcadv  present in and slow|\
migrating through (lie aquifer.
   From about 196S or 1969 through about 197-1. pond (.' was
maintained lull most ol  the time by diverting water trom (lie
freshwater reservoirs to the south. This resulted in the inlil-
tration ol about I ft'/sec (0.03 in'/sce)  ol  freshwater into the
alluvial aquifer. This artificial recharge bad  the effect of diluting
and Hushing the  contaminated groundwater away from pond
(.' (aster than would have occurred otherwise. By 1972 tin1 are.il
extent and magnitude ol contamination, as indicated bv chlo-
ride concentration, had significantly diminished, (.'hloride con-
centrations were  then above 1000 mg'L in only two relatively
small parts of the contaminated area and were almost at normal
background levels in the middle ol the aflectcd area  limme-
diately downgradient from pond C).
   In 1973 and 197-4 there were new claims oi crop and livestock
damages allegedly caused by groundwater that was contami-
nated at the Arsenal. Data collected by the Colorado Depart-
ment of Health  (Shukle. 1975) show that  dnsnpropylinethvl-
phosphonate(DIMP). a nerve-gas by-product, has been detected
at a concentration of 0.57 part per billion (ppb) in a well located
approximately 8 mi (12.9 km) downgradient from the disposal
ponds and I mi (1.6 km) upgradient trom two municipal water-
supply wells of the city of Brighton. A DIMP concentration  of
48 parts per million (ppin). which is nearly lOO.(XK) times higher.
was measured in a groundwater sample collected near the dis-
|H>sal ponds. Other contaminants detected in wells or  springs
in the area include dicyclopcntadicnc (IX.'PI)I, cndrin. aldriu.
dieldrin. and several organo-sullur compounds.
  The detection of these chemicals, which  were manufactured
or used at the Arsenal,  in areas o(T the Arsenal property led
the Colorado Department of Health to issue cease and desist,
cleanup, and monitoring orders in April  1975 to the  Rocky
Mountain Arsenal and Shell Chemical  Company, which was
leasing industrial facilities on the site. The Cease and Desist
Order called for a halt to unaiithori/.cd discharges of contam-
inants into surface water and  groundwater just north of the
Arsenal. The cleanup order applied to all sources of DIM P and
DCPD located at the facilities. The third order called for a
groundwater monitoring program,  the results of which would
be rc|x>rtcd to the State Health Department on a regular basis.
Consequently, a program that included groundwater monitor-
ing and studies to determine a means to intercept contaminants
flowing across the north boundary ol the  Arsenal was estab-
lished by the U.S. Army.
  As a result of continued monitoring, additional contaminants
have been identified in  the groundw.ilcr at the Arsenal. The
most widespread of those found are Nemagon (dibromochlo-
ropropanc) and various industrial solvents. Nemagon contain-

-------
9fi
    I.K.ONAHI) K.  kONIKOXX  tliul DOl'CI.AS XV.  THOMPSON
in.ilinii li.is been identified as probably resulting from Arsenal-
relaled activities, xvhcrcas the industrial solvents identified arc
not unique to Arsenal activities. F.xtrcmely low concentrations
of Xcir.agon  (< 2 ppb) have been found in wells located im-
mediately.,west of the Arsenal Ixmndary. Otlier organic con-
taminants associated with pesticide manufacturing have been
found in wells located in a centrally located manufacturing plant
area known as the South Plants area. These contaminants prob-
ably entered the aquifer from accidental spills and leaks and
appear to be migrating from this  area very slowly.

Hxdrogeology

The records ol  several hundred observation wells,  test holes,
irrigation wells, and domestic wells were compiled and anal-
y/cd to describe the hydrogeologic characteristics of the alluvial
aquitcr in and adjacent to the llocky Mountain Arsenal. Kon-
ikoxv (1975) presented lour maps that  show the configuration
ol bedrock surlace. generalized water-table confirmation, sat-
urated thickness of alluvium, and  transmissivity ol the aquifer.
These maps show that the alluvium forms a complex, sloping.
discontinuous, and heterogeneous aquifer system.         -~
  A map  showing the general water-table configuration for
1955-1971 is presented in Figure (i.3. The assumptions and
limitations of Figure (i.'} are discussed  in more detail by kon-
ikow (1975). The areas in which the alluvium  either is absent
or is unsaturated most ol the time form internal  barriers (hat
significantly affect groundwater flow  patterns within the aquifer
and. hence, significantly influence solute trans|x>rt.
  The general  direction of groundwater movement is from
regions of higher water-table altitudes to those ol  lower water-
table altitudes and  is approximately  perpendicular to the
water-table contours. Deviations from the general flow pattern
inferred from water-table contours may occur in some areas
because of local variations in aquifer properties,  recharge, or
discharge. The nonorthogonality at places between water-table
contours and aquifer lM>nndaries indicates that the approximate
limit of the saturated alluvium does not consistently represent
a no-llow boundary  but that, at  some places, there may In-
significant flow across  this line. Such a condition can readily
occur in areas where the bedrock  |x>sscsscs  significant porosity
and hydraulic conductivity or where recharge  from irrigation,
unlined canals, or other sources is concentrated.  Recatisc the
hydraulic conductivity ol the bedrock underlying  the alluvium
is generally much lower than that of the alluvium, groundwater
lloxv and contaminant  transport  through the bedrock are as-
sumed to be secondary considerations compared with llow and
transport i" the alluvial aquifer. Croundwatcr withdrawals in
the area are predominantly  from wells tapping the alluvial
aquifer.

Contamination  Pattern

Since 1955 several  hundred observation wells and lest holes
have been constructed to monitor changes in water qualilv and
xvaler levels in the alluvial aquifer. The areal extent of contam-
ination has been mapped on the basis ol concentration ol chlo-
ride. 131 NIP. and other inorganic and organic compounds in
                         EXPLANATION
       nix>— WATER-TABLE CONTOUR - Showt aporo»tmata almuda o<
               watar tabla. 1955-71. Contour interval 10 faat 13 matan).
               Datum it maan aaa lava)
            Ar*« in which alluvium it ftbMnt or unt*turar«d
HOIHI. (i..°)  (tt'iieral watrr-talilccniiliunralion in llie alluvial aquifer
ill and adjacent to the Hooky Mountain Arsenal. 1055-1971 ikonikou.
1977)
wells. Chloride concentrations ranged from normal background
concentrations of about 40 to 150 ing/I, to about 5(XH) mg/L in
contaminated groundwater  near pond A. Chloride data col-
lected during 1955-19.% indicate that one main plume ol con-
taminated water extended beyond the northwestern Ixnmdary
of thi' Arsenal  and that a small secondary  plume extended
beyond the northern Ixnmdary (see  Figure (>.-!). The contam-
ination pattern  shown  in Figure (>.4 clearly  indicates that the
migration of contaminants in  (his aquilcr is  also signilieantK
constrained by  the aquifer l>ouiularics.
  Because chloride generally  behaves as a conservative (that
is, nonreactive) solute in groundwater. it  is often assumed that
chlorides can IK- used to indicate the maximum extent ol con-
tamination from a source  that contains chloride. But  this as-
sumption  is  not always reasonable because  chloride is also a
common natural constituent in gromidwater.  At  the  Kocky
Mountain Arsenal the extent of contamination as indicated by
chloride concentration reflects a dilution ratio of about 3.1:1
from  the contaminant  source  to the definable downgradient
limit of contamination. However, the extent  ol contamination
as indicated by some of the organic coui|X>unds. such as DIM P
(sec Itobsou,  I9SI), is much greater  because  they have a zero
background concentration and can be delected to trace con-

-------
('.ontdinintition intd Aifiiijt'i'
                &*.  7-
                 --'    erf'
                        •i     i    t •iLOxiim
                         EXPLANATION
          •    O«l point (S«pt 19S5-M»cti 19561
       	300	Lin* o* aqual cMond* conc«nlr«iion {in milhgrami p«r lil«rl.
                 (n,«t..l v«n«bl«

       K  \\ Af«» in which alluviuni it abMnl or unialurat«d

KKil'HK (i.-t  Olisi'iAed ililoridc iinifrntratiiin  in H)">i> (ki>nikc adsorlxxl. Oilier diHercnces among
shapes and locations of  plumes of different contaminants arise
liecause they entered the aquifer at significantly different times
and (or) locations within the Arsenal. For example, the Ncni-
agon  plume  occurs west  of  the chloride plume because the
source of the Ncmagon was not from  the  disposal ponds hul
apparently from a spill  that occurred west  of the ponds.
  Contaminants have also liccn delected in  several  shallow
bedrock wells in or near the Arsenal. However, at present there
are inadequate data to  define the areal extent, depth of pen-
etration, or'rate of 'spreading of contaminants  in the bedrock.
APPLICATION OF  SIMl'LATION  MODF.LS
The reliable assessment ol hazards or risks arising from ground-
water contamination problems and the design of efficient and
elleclive techniques to mitigate them require the capability to
predict the  Ix'li.ivior of chemical  contaminants in  (lowing
grouudwater. Reliable and (|iiautitativc predictions of contam-
inant movement can only be made il the processes controlling
convcctive tr.ins|x>il, hydrodxnainic dispersion; and cliemical.
physical, anil biological reactions thai  affect solute concentra-
tions in the ground arc nuderstiHKl. 'fhese prm-esses. in turn.
must be expressed in precise mathematical equations having
defined parameters. The theorx and developnienl ol the equa-
tions describing Krouudwatcr flow and solute liansport lia\e
been well  documented  in  the literature. Perhaps the  most
important  technical advancement  iiT the anal)sis of pomid-
walcr contamination problems during  the past ID yr has been
the development ol deterministic numerical simulation nuxlek
that efficiently solve the noverninU flow and transport equations
for the properties and boundaries ol a specific field situation
Although many of the processes that  ailed waste  movement
are individually well understood, their complex interactions in
a heterogeneous environment may not be  understood well
enough for the net outcome to be reliably predicted. Thus, the
analysis of pomidwater contamination problems ean be erealK
aided by the application of deterministic numerical simulation
models that solve the equations describing groundwater llou
and  solute  trans|X)rt.
  Tin1 solute-transport model described by konikow and Hrc-
dehoelt (I97S) was used to simulate the movement of chloride
through the alluvial aquifer at the  Arsenal in an ellurl lo  re-
produce the 30-yr (19-13-1972) history of contamination, lo help
test  hy|M)llieses concerning governing processes and  parame-
ters  to develop an improved conceptual model of the problem.
to aid in  setting priorities lor the collection of additional data.
and  lo evaluate |Missible management alternatives  (Konikow.
1977). The model included an area of approximately 3-4 mi-' (SS
km-'). The stringent data requirements for applying the' solnte-
transport model pointed out deficiencies in the data base avail-
able at the  start of  the slud\. Specifically, it was found dial the
velocity distribution determined from the water-table config-
uration mapped in 1956 (see Pctri and Smith. 195ft) was in part
inconsistent with the observed pattern of contaminant spread-
ing.  The subsequent quantitative analysis and  reinterpret.ition
of available hydrogeologic data, based partK on feedback from
the numerical  simulation model, led to a revised conceptual
model of the aquifer properties and bonnduries that  incorpo-
rated the strong influence of the internal  harriers within the
alluvial aquifer
  The solute-transport model ol konikow (1977) was calibrated
mainly on the basis of (he chloride concentration pattern that
was observed in 195fi (Figure 6.4). Computed chloride patterns
agreed closely with observed patterns, which during the 30-yr
history were available only for  19.%.  I9KI.  l%9,  and 1972.
The  calibrated model was then used to analyze the ellccts  of
future and  past changes in stresses and boundary conditions.
For  example, comparative analyses illustrated that it would
probably take at least many decades for this contaminated aqui-
fer to recover naturally its original water-quality characteristics.
But  it was  also inferred that  appropriate  water-management
policies for aquifer reclamation can help to reduce  this resto-
ration time to the order of years,  rather than decades, for the
relatively mobile contaminants, konikow (197-4) also noted that
(lie simulation results showed that a reclamation scheme using
a network  of interceptor wells would aid in  containing and
removing the contaminated grouudwaler.

-------
                                                                   i.i.oNvui) K  kOMKovv•  antl i)or<;i..\s vv THOMPSON
   Hobsou (19M) developed and calibrated a solute-transport
model lor 1)1 Ml1 to help evaluate (1) (lie mechanisms mid pa-
rameters controlling 131MF migration, (2) future DIMF con-
centrations in nearby municipal water supply wells, and (3) the
effectiveness of various gronndwatcr harrier configurations de-
signed to halt otl-Arsenal movement of contaminated ground-
water. Tin- model included an area of ahont 90 mi- (230 kin-)
and assumed that DIMF is conservative. Using the calibrated
model. Rohsnn was ahle to reconstruct the historical movement
of .DIMF in the aquifer lietween 1952 and 1975. to estimate
DIMF concentrations in the1 South Flatte River resulting from
discharge of contaminated groundwatcr. and to predict future
DIMF concentrations under a variety of assumed mana.uenient
alternatives.
  To evaluate more fully the range of engineering approaches
or alternatives that would he feasible for construction along the
north honndary of the Arsenal, Warner (1979) modeled a smaller
part of the aquifer (2.5 mr or ft. 4 kmj) in that area in  much
finer detail.  He predicted the impact on DIMF concentration
of implementing a variety of interception schemes that incor-
porated variants of a  basic plan that  included elements of
groundvvater withdrawal, a barrier, ami  reinjection of treated
water.  Among other findings.  Warner (1979) showed that a
proper!)  operated hydraulic harrier, consisting of a line of
pumping wells,  would he just as effective as a bentonite barrier
in stopping the movement of DIM F-contaminatcd gronndwatcr
across the northern  boundary of the Arsenal.
  It is recogni/.ed that oilier organic contaminants of concern
may hi- sorbed or altered by chemical and biological reactions
as they move through the aquifer. The movement of a solute
that is sorbed will be retarded  relative to the  movement of a
conservative solute. This  is beneficial in the sense that in a
given  time a contaminant that is sorbed will not migrate as far
as a conservative contaminant. However, the soiption process
could pose a significant obstacle to aquifer reclamation Ix-cansc
even after the contaminant  source- has been  eliminated,  the
sorbed organics could later desorb and continue to migrate
through the aquifer, perhaps still (losing a ha/urcl alter all con-
servative contaminants have been llu.shcd out of the aquifer.
Sorption  processes can and have been incorporated into solute-
transport mtxlels  (see  C.rove.  197(il. and this allows a  more
realistic evaluation to be made of their behavior and response
to mi|X>se(l  aquifer  reclamation stresses. Although this then
presents  no great conceptual difficulty,  in practice- it is  quite
difficult to determine the coefficients that describe the rates of
reactions and exchange capacity of the aquifer material for each
individual contaminant.
  An overall systems-management model is currently in final
development under the sponsorship of  the U.S.  Army. This
computer model is expected to provide a valuable management
and decision-making tool to aid  in evaluating aquifer recla-
mation alternatives at the Rocky Mountain Arsenal. The model
will be composed of numerous modules, including (1) ground-
vvater flow.  (2) solute transport. (3) groundvvatcr interception
and control. (4) surface-water control. (5) groundvvater and sur-
face-water treatment, (h'l cost  estimation, and (7) re|xirl and
graphics output. The model will be evaluated and verified using
the Kockv Mountain Arsenal as a test case because of the abun-
 dance of historical data there. After verification, selected al-
 ternatives for contamination control and elimination at the Ar-
 senal will lie modeled with a goal of predicting long-term system
 responses and costs. If successful, this model will be applied
 to Installation Restoration  programs under way at other loca-
 tions.
AOUIFKR  KKSTORATION  FKOCRAM

Response to Cease and Desist Orders

As  a  result of the (Vase and Desist Orders, an  Installation
Restoration program was established at the  Rocky Mountain
Arsenal under the direction of the Frogram Manager for Chem-
ical Demilitari/ation and Installation Restoration. Aberdeen
Froving ('.round. Maryland This office  was later rcorgani/ed
into the U.S. Army Toxic and  Ha/ardons Materials' Agency
(USATHAMA), which  currently directs the  Installation  Res-
toration program at  the Arsenal. The- main  objective of this
program is to limit  the migration of contaminants from the
Arsenal to the degree required by applicable  federal and  state
regulations. The program is primarily concerned with contam-
ination problems resulting from historical activities on the Ar-
senal  as opposed to ongoing operations.
  The installation Restoration program consists ol three major
parts  or subprograms that include regional groundwater mon-
itoring, contaminant migration control, and elimination ol con-
taminant sources. This  program had been nig.mi/.cd to allow a
phased approach in developing and implementing contaminant
control systems,  thereby accelerating the reduction of potential
environmental hazards. More than $25 million has been ex-
pended to date  in the  Installation  Restoration program, ex-
cluding the costs associated with construction of the control
systems.
  A comprehensive groundvvater monitoring program was de-
veloped based on historical contaminant distribution infor-
mation and initiated late in 1975. It included sample collection
from Ixilh oil-site and adjacent off-site wells.  This monitoring
program has  been continually updated since that time to in-
clude additional wells and analytical parameters as required.
Currently, it involves  the  collection and analysis ol samples
from  91) to KM)  wells on a quarterly basis. The  information
generated from the monitoring program is used to define the
distribution and track the migration of known contaminants.
identify new contaminants, develop design criteria for contam-
ination control and treatment systems,  and evaluate  the op-
eration of  existing systems.
  The subprogram concerning contaminant migration control
at the Arsenal boundaries was initiated  in late 1975 with the
goal of rapidly eliminating the migration of contaminants off
the Arsenal's grounds.  Boundary control was  the only viable1
option because of the  already wide distribution  of contami-
nants, the long travel times associated with contaminant mi-
gration from  the sources to the boundaries,  and the lack  of
precise definition of all source areas.  Filot and lull-scale Ixxmd-
ary  control systems have been implemented  at the northern
Arsenal boundary, and plans have been  developed to expand

-------
C.iniliiinindtinn tinil A^III/CC Kcclaination
                                                       99
the treatment system along llic northwestern Arsenal bound-
ary. Those systems will lie discussed in more  detail later in
this chapter.
  Planning lor the control and elimination of contaminant sources
evolved several years later as additional  data became available
on specific source areas. The goal is to control or eliminate the
contaminant sources on the Arsenal grounds and thereby elim-
inate the need lor boundary control in the future. Studies have
been undertaken to aid further identification and definition of
contaminant  sources, to  develop feasible source  control and
elimination alternatives, and to develop  control and treatment
systems. A summary of the strategy and progress of this sub-
program is given at the end of this chapter.
Contaminant Migration Control at Arsenal Boundaries

Hccansc the contamination that resulted in the issuance of the
< lease  and Desist Order  was detected in surface water and
groundwater immediately north of the Arsenal, the primary
locus of the Installation Ucstoration program during 197(i and
1977 was the northern Arsenal boundary. A dike was con-
structed to stop the miration oil the Arsenal of contaminated
surface water. Studies were initiated to determine a  feasible
alternative for stopping the flow of contaminated groundwater
oil the  Arsenal without significantly altering the normal ground-
watcr flow pattern in  the area. The concept selected involved
interception of the grouudwatcr a  short distance south of  the
northern Arsenal boundary, treatment of the water to  remove
the contaminants, and reinjcction of the treated water at  the
boundary.
  Two method!) were proposed for intercepting the  flow of
groundwater. The first method involved the use of  a hydraulic
barrier, one or two lines of closely spaced pumping wells that
would  provide for dewatcring of the aquifer along or between
the lines.  The permeability in the area is  sufficiently high lor
this concept to have worked,  but the gradient is shallow and
concern was expressed  over the potential lor excessive recy-
cling of water from the reinjection wells back to the  withdrawal
wells. As a result of this concern and  to provide an additional
safety factor, a second method was  selected that involved  the
use of a slurry cntoll wall  to lorm an impermeable barrier
between, the withdrawal and reinjection wells.
Treatment Process

l«ite in 1975 a lalx>ratorv study was initiated to evaluate various
methods for removing organic compounds from representative
groundvvrtter samples from the area. Treatment  processes in-
vestigated include granular activated-carbon adsorption, pow-
dered activated-carbon  adsorption, chemical oxidation  using
ultraviolet (I'V) light and ozone, and an ionic exchange resins.
Key chemical parameters for analysis included Dl M P and D( 'PI).
Extensive lalxiratory studies were conducted using standard
isotherm tests for evaluating the carbons and resins and using
hatch reactor tests for evaluating the I'V/ozone  process. The
anionic exchange resins were dropped from further consider-
ation because of low cllicicncy and high cost. A vries ol licit!
studies was initiated on the carbon adsorption and I'V/o/one
oxidation processes to permit further evaluation.
  Powdered activated-carbon adsorption tests incorporating a
polymeric coagulant were conducted usjnu a standard  Arinv
Krdlator water-treatment unit (chemical addition, mixing, up-
llowclarilicationHSwcder. 1977). fGranular activated-carbon ad-
sorption tests were conducted  usinj; a dynamic -flow, multi-
column system (Sweder, 1977).  l'V/o/onc oxidation tests were
conducted using a continuous-flow, mechanically mixed reactor
(Hunts et ill.. I97K). Granular activatcd-carlxm was found to
be more efficient (110 mg ol carbon/I, of water) in removing
the contaminants than was the powdered activated-carbon (20 ft (305 in).  Ik-cause little operational information
was available on groundwater contamination control  systems
similar to the one  proposed, the Army decided to install  a
limited pilot containment system in the area of high-contam-
inant concentrations and evaluate the possibility of extending
the treatment system across the entire affected part of the
northern )>oundary.
  The North Honndary  Pilot System AMPS) was constructed
and placed in operation  in July  I97S  It included the following
live subsystems: a barrier, dewatering wells, reinjection wells.
treatment plant, and monitoring  wells. A schematic diagram
of the system is provided in Kigure h'.5.
  The barrier was constructed by filling a 3-ft-wide,  1500-lt-
long trench, averaging 25 ft  in depth, with a mixture of soil
and bentouite clay.  The barrier was anchored approximately 2
ft into the bedrock all along the alignment.
  The dewatering wells were installed south  (npgradient) of
the barrier approximately 225 ft apart on a straight line parallel
to the barrier.  There were  six S-inch-diametcr  wells placed
within 30-inch-diameter gravel-packed holes.  Kach well was
screened throughout the entire saturated portion of tlir alluvi.il
aquifer.  A submersible  pump and flow control system  were
installed at each well site. Water  from the wells was pumped
through an underground manifold to a  single sump at the treat-
ment plant.

-------
 KM)
    LEONAIUJ  K.  KONIKOW IIH(I l)()l;CI.AS  \V  THOMPSON
  The injection  wells were installed north ulowngradicnt) of
the harrier, approximately HX) It apart on a straight line be-
tween the harrier and the northern Arsenal honndary. There
were twelve 18-inch-dianioter wells, which were installed in
36-inch-diameter gravel-packed holes. The recharge wells were
screened to a point al>ove the water tahle. Treated water was
continuously injected into  the recharge wells l»y gravity flow
through an  underground manifold  system.  Sensors  and (low
control valves were installed in the wells to prevent overflow
or surface discharge in  the event that a well experienced an
excessively high huildup of hydraulic head hecause of clogging
of well screens or other factors.
  The treatment plant subsystem was designed to treat 10.(XX)
gallons of water per  hour.  It consisted of two  mixed-media
pressure filters, each 4 It in diameter, and two adsorl>er vessels
IIH columns), eaeh 10 fl in diameter and II It high,  designed
to contain .lUiil 20.(MM) Hi ('I I (HI kg) of granular activated e.n
lion  Water Irom the collection sump -was pumped through the
filters in parallel to remove suspended material,  then through
the carbon adsorbers, and  finally to the injection wells. Only
one carbon a'dsorber was in operation at any  one time. \Vhen
the 1)1 NIP concentration approached 50 pph. the carbon was
replaced.  During  1978-1981.  replacement was  required ap-
proximate!)  once every  9 mouths. The exhausted carbon was
transported  offsite (or regeneration by a commercial vendor.
Carbon usage rates ranged trom 1(X) to 150 mg of carbon/I, of
water. The treatment system  was designed to be largely au-
tomatic and simple to operate by incorporating automatic back-
washing of the  filters and sensors  for control of pumps and
valves.
  Ten monitoring wells were installed both  upgradicnt and
downgradient ot  the pilot containment system. They were cased
with small-diameter PYC  pipe  and screened in the alluvial
               PILOT TREATMENT
                  PLANT
KK.l'HK (v.'i  S> lie-main   Typical gas rliroiiiutograpliv7iiia.ss spcctromvtry scan ol
 north boundary pilot treatment system iitllnrnt.
aquifer. Water levels  and chemical qualitv  wen1 monitored
periodically to provide information on the ctlcctivencss ol the
operation  of the system.
  The cost of the barrier and the wells as constructed in 197S
was $450,(XX). The facility for housing  the treatment  system
cost approximately $4(MXX). The treatment equipment was ob-
tained under a lease/service contract agreement with  a com-
mercial vendor with an initial cost of approximately  SKX).(XX)
and a yearly fee ranging from S 1:15.000  to SI50.000.
  The NBPS operated successfully  for a period of approxi-
mately 3 yr. For example, during fiscal year 1979, downtime
was less than I |XTecnt ot operating time.  The granular acti-
vated carbon cflcctivcly removed the organic contaminants from
the groundwater, generally to a level of less than 10 pph. as
illustrated by a comparison ol typical gas chroinatography/mass
spectrometry analyses of the influent (Figure  (i.(i) and effluent
(Figure 6.7) of the treatment system. The flow of groundwater
downgradient  from the  NHPS was essentially unchanged
(D'Appolonia Consulting Engineers. Inc.. 1979).  Preliminary.
data indicate that the concentration of orgauics in the ground-
water downgradieut from  the pilot system has diminished sig-
nificantly.

Expanded Containment System

As a result of the successful operation of the pilot containment
system, construction of the expanded containment system was
begun in early 1981. The expanded system consists ol a (i8(X)-
ft barrier ranging from 25 to 50  ft deep, 54 withdrawal  wells.
and 38 reinjection (or  recharge) wells. The expanded barrier
effectively intercepts all the contaminated groundwater flowing
across the northern Arsenal boundary in the alluvial aquifer.
The  expanded treatment  system is designed to treat  3ft.IKK)
gallons (!3(i.(XX) I.) of water per hour. The adsorlx-rs used in
the pilot operation have been replaced with three pnlsed-bed
adsorbers  designed to contain  30.(XX) II) (H.ofX) kg) ol earlxin

-------
 Contamination mid Aytiijer Reclamation
                                                      101
                                    If
                                      JV.A.
                                             5    J
                                             I    1
       1    2   3  «   ',  6   7   «   »   10   II  I?  13 14  16  16
                          RETENTON TME

FKil'HK (i.7  Typical gas clmiinatograpliy/mass s|K'clronielry scan of
noitli Ixiiindaiy pilot treatment svslein cllliient.
ruch. Tlit1 new adsorbers should he much more efficient than
the old ones because the anticipated carbon usage rate is only
25 to 30 nig of carlwn/L of water. The mixed-media filters have
been replaced with cartridge filters, which are easier to main-
tain. The whole system is highly automated and will require
only intermittent monitoring by a single operator.  The esti-
mated cost for the expanded system is approximately $6 million.
The expanded system became operational in 1983.

Other Contaminant Migration Control Systems

Concepts have Ix-en developed for two additional  Ixiundury
contaminant migration control systems located along the north-
western  Arsenal boundary  (Figure b'.8). One  system will lie
located at the southern end of that  Ixmndary and  the other
midway along that Ixnmdary. Both systems  have been devel-
oped primarily to control (he migration of low concentrations
of Nemagon across the  boundary. Both systems will br similar
in si/e to the NBPS and will incorporate granular activatcd-
carlxm treatment of the gronndwater. The system to be located
on the southern end of the boundary (Irondalc System) was
constructed under the  direction of Shell Chemical Company
and  incorporates a hydraulic barrier for interception  of the
groundwater, along with the injection wells. It became oper-
ational in 19H3.  The other system, to be constructed by the
Army,  will  incorporate a slurry cutoff wall, withdrawal wells.
and reinjeclion wells, similar to the pilot system. It is scheduled
to l>e operational in 1985.

Planning for Control and Elimination of Contaminant
Sources

Contaminant migration control at the boundaries of the Hocky
Mountain Arsenal  was  initiated to stop or severely limit the
migration ol contaminants of) the Arsenal grounds as soon as
possible.  Owing to the si/.e ol the Arsenal  and the  extent of
 the source areas,  the boundary control systems could IK- re-
 quired to operate for an  indefinite' period of time. The only
 way to limit this  requirement and  the  associated cost  is to
 control or eliminate the contaminant sources. Therefore a stud)'
 was initiated in 1980 to identity and assess existing and inno-
 vative control or elimination alternatives that are capable of
 bringing the Arsenal into compliance with all applicable federal
 and state environmental laws and-rcgulations. Another stud)
 objective was to develop  preliminary cost data and technical
 data for use in a subsequent detailed evaluation and comparison
 of alternatives. A study team made up of 12 government and
 independent scientists and engineers was established to con-
 duct and manage the study. A review ol historical operations.
 past study reports, and data from ongoing studies was made t<«
 identify, where (xissible, potential sources of contaminant mi-
 gration problems.
  The next phase of the  study involved the development .-a!
 control strategies.  Guidelines and  criteria for developmental)
 the strategies were required because of the complexity of and
 relationships between the contaminant sources and migration
 characteristics. In addition, some degree ol commonality ol
 structure or organi/ation among the strategies was needed to
 enable a comparison and ranking ol the alternatives to IK- de-
 veloped As a result, a hierarchical approach and structure lor
 generation and classification of control strategies  were devel-
 oped incorporating five levels of detail ranging from cowept
 to unit operation (Kocky Mountain Arsenal Contamination Con-
 trol Study Team, unpublished report, August 1981). Each team
 member individually developed a  number of strategies using
 the hierarchical approach and determined the problem defi-
 nition and technical data-base deficiencies associated with *-ach
 scheme. The schemes were then submitted  to the group as a
 whole lor integration and  evaluation.
  Screening criteria were developed to aid in evaluating and
 comparing the alternative schemes. The  goal was to pitwlmv

    EXPLANATION
  «Dtwit*lnQ Ml!
 ffl  LKKM) TrMtnwil
                 NORTHWEST BOUNDARY
                 SYSTEM (PraooMdl
     IRONDALf SYSTEM
     (9w« Ownc« Co
KKJUHK (> H  Ideation ol existing anil pro|x>s
-------
 102
                                                                   I.KONAHI) K  KONIKOU  and Dot c:i.\S \V. TIIOMI'sUN
 a set of criteria that could IK- applied at the various hierarchical
 levels, thereby enabling a general screening of the .schemes
 rather than a detailed evaluation ol'each one. The major criteria
 selected li>r use are as follows:

   1. Availability of technology,
   2. Amount of additional data required,
   3. Cost and time needed to  fill data gaps,
   4. Life cycle costs—capital and O&M,
   5. Compatibility between systems,
   6. Degree of risk—environmental and technological,
   7. Compliance with regulatory requirements.

   The individual schemes developed by the study team mem-
 bers were integrated,  evaluated, and screened by the  study-
 group as a whole. This work resulted in the presentation of 14
 alternative schemes that were recommended for detailed eval-
 uation by the Contamination Control Study Team. The schemes
 incorporate various aspects of the technologies listed in Table
 6.1. The schemes address only the known contaminant sources
 at the  Arsenal and therefore may have to IK- expanded if ad-
 ditional sources are identified in the future.
   In addition  to the  development of the  alternative schemes,
 the study team identified a number of data gaps concerning
 both problem definition and technology development that must
 be filled before final selection  of a control or elimination al-
 ternative can  be made.  Studies have been included  in  the
 overall Installation Restoration program to fill these data gaps.
 They include additional hydrogeologic definition of certain areas
 on the Arsenal, surface-water hydrology definition, technology
 development for water treatment, and technology development
 tor disposal of contaminated soil and residue. As the data from
 these additional studies become available, the study team will
 further evaluate and revise the alternatives as required with
 the goal  of selecting one alternative for implementation.
  The  implementation of the selected alternative will l>e con-
 ducted using a phased approach. As soon as a particular part
TABLE B.I  Contaminant  Source Control and  t'liminatiou
Technologies

C.roimdwatt'r Intercept inn
  Hydraulic harrier
  Slum trench
  Drvvalrrint; trench I I'rcncli ilr.iin)
Water Treatment
  Adsorption (carlx>n and resin)
  Chemical addition/coagulation/prccipitation
  Filtration
  Membrane separation
  Chemical oxidation
  Activated iludue
  Volatile stripping
  Ion exchange
Contaminated Soil and Residue Treatment
  Incineration             ~
  fixation/stabilization
  In situ forced leaching
  l-'vc.ivatum and  dis|»>sal
 of the alternative is defined and design criteria are developed.
 construction will IK- initiated. For example, the elimination of
 Basin F will probably In- one of the first major actions initiated
 Ix-cause it is known  to leak and because the extent and nature
 of the contamination associated with this area of the  Arsenal
 have been better defined than  elsewhere.  Tin- control and
 elimination of known contaminant sources at the Kocky Moun-
 tain Arsenal are currently expected to involve a 5-yr construc-
 tion program that is scheduled to start  in  19S5. A final cost
 estimate for the construction program has not been developed,
 but preliminary estimates range from $50 million to SHX) mil-
 lion.
 SUMMARY  AND CONCLUSIONS

 Removing pollutants from a contaminated aquifer may seem
 to be an almost impossible task. While this may be true  for
 some contaminated aquifers, others may be amenable to one
 or more plans for artificial reclamation that could significantly
 accelerate the rate of water-quality improvement in the aquifer.
 The feasibility of any such reclamation plan would be strongly
 dependent on  the  hydraulic and chemical  properties of the
 aquifer, on the type and source of contamination, and on the
 duration and area! extent of contamination.  Because a variety
 of reclamation plans can be proposed lor any one problem,  an
 accurate model ol How and contaminant transport in the aquifer
 is an invaluable tool for  planning  an efficient  and effective
 program.
  The control and elimination of contaminant migration and
 contaminant sources at the Rocky Mountain Arsenal represent
 a large, complex, and costly undertaking (over S25 million has
 been spent in the Installation Restoration program). An exten-
 sive well-monitoring program has been required to define the
 extent of the conta'mination and the relationships between the
 sources and contaminant  migration patterns. Control of con-
 taminant migration at the Arsenal Itoundarics has proved fea-
 sible using a  system  involving groundwater  interception.
 treatment, and rcinjection. The system was operated success-
 fully without adversely affecting  the  How and distribution  of
 groundwater dowugradieut from the treatment system, and it
 has resulted in a significant decrease in the concentration  of
organic contaminants in groundwater downgradient from tin-
 pilot system.
  Although Unindary-control systems can be used successfully
to stop or restrict the migration of contaminants off the Arsenal's
grounds, they cannot solve the problem of continued contam-
 inant migration from the source areas to the environment. The
overall solution thus involves the  control or elimination of the
contamination at the sources. A program has  been successfully
initiated at the Rocky Mountain Arsenal to develop and assess
source control and elimination strategies. Through additional
data collection and feasibility studies,  a single strategy will be
selected and  implemented using a phased  construction ap-
proach.  The ultimate goal of these activities is to bring the
Arsenal into compliance with all  applicable  federal and state
environmental laws  and regulations.
  The  great  difficulty  and expense  involve d  in mitigating

-------
('iintainiitatiini tiiul Aquifer Kirltinnitiiin

Uronndwatcr contamination prolilcms do not  lessen (lie need
In do so; they do illustrate (lie long-term hcuclits ol planning
and designing waste-disposal activities to prevent or nn'niini/.e
future contamination liauirds.
ACKNOWLEDGMENTS
The Installation Restoration -program  at  the Rocky Mountain
Arsenal (RMA) is being funded and directed In the U.S. Army
Toxic  and Hazardous  Materials  Agency.  Aberdeen Proving
(Ground. Maryland, in cooperation with  the Rocky Mountain
Arsenal. Denver, Colorado  Tlic authors wish  to thank the
personnel from these organizations lor their support. Special
thanks are extended to Carl lx>\cn. ("hid. Process Develop-
ment and Evaluation Division.  RMA.  and Donald Mailer, Ru-
lu'l.iiid I lager. Inc., Tucson. Ari/.ona. lor providingoperational
and cost data on the RMA contaminant control svslenis.
REFERENCES
Buhls. R. K.. 1'. <;. Malonr. anil I). U. 'Ilininpsim (197S).  Evaluation
  of ullraxiolct/ozoiie treatment of Hockx Mountain Arsenal gromul-
  xx-aler. C.S. Army Eniiinivr Watei-u-uij\ t'.v»criiiii'iil Station. Tech-
  nical Report  V-76-/, 7S pp.
DAppolonia Consultmn Engineers. Inc.  (1979). Evaluation of north
  houndary pilot ciintahniifiit system.  KMA. Denver. Colorado. I'roj-
  eel Number  RM79-3S9. Wl pp.
drove,  D. B.  (I97(i'. Ion exchange reactions in Kroumlwatcr quality
  miKli.'K. in Ailranrex in GnHiiiihrntrr Hijtlnilnnij.  Am.  Water !<<•-
  sour. Assot'.. pp. 144-152.
Konikoxv. L. F. (197-1).  licflamatioii ol a ct)ntaminati'(l ai|iiilfr. (•<•(*/.
  Stir. Am.  Almtr. Profiraiii.\ 6.  S30-H31.
konikovv, L. K. (1975).  llydrotieologic maps ol'tlie alluvial aquifer in
                                                             103
  and adjaei'iit to the Kix-ky Monnluin Aiscnal. <.'olniad'i. I'.S. C.cnl
  Sun . t)iM'>i-Hlc Ki-fi. 7/-W2
Konikow, L. F. (1977). Modeling clilornlc nuni-iiicnl in tlir allnxi.il
  ai|iiiliT at llic Kix'kx Mountuin Arsenal. (!olni.«ln. C..S (.in/  Sun
  \\uti-i -Siii>iil  2(H-I. 4.) pp.
Konikou. I,. K.. ami J. I), llreclrliiicll |I97S). Coinputor miMlel ol IVMI-
  ilimc'iisional solute  trans|x»l and dis|MTsion in ground w.ilci, C.S
  (•I'til. SUIT.  '/i>c/iNJf/iurrr\ lin .. Huvk  ~. (°/n//i
  C2. 9t) pp.
Hetri. I..  U  (ISKil). II if tiioxcmciit ol -..ilinc uroniul water in I lie viiiinU
  ol Derhy, Ooloradn. in  (•nmiir/ \\utfi f'tintiimiiwtiini .Si/mjxi.viiiin.
  HoU-rt A. Taft Sanitarx  KIIK. Center Teili  Hep  Whl-5. pp. II')-
  121
I'ctn. I.. K.. and K. (). Smitli (l'JV)i. Imestication ol  the i|U,ilit\  <>l
  )!roiin(l watri  in tlie vicinity of  l)i-rl>\. (.'olor.nlo.  C.S. (ii'iil. Sun .
  O/trii-fi/r HI'II.. 77 pp.
Kolison.  S. ('.  |I9SI). (loinpiiler simulation ol moxvmetit ol 1)1 Ml'-
  contaminated Kroiimlualcr near tlic HiK'kx  Mountain AISCIM!. (.Hl-
  orad(K  'I. F. /immir and (.*. (). Hi^s. fd** . in />r('ifirr//>i/f/f/ nml
  C.iininiltKili i Ciinliiiiiiiiinil 7Vuii.v/inr(. Anifiicui S\iiim. Holier! A  lalt Sanitary Kim  (it-liter Tecli.
  Rep W6I-5. pp.  121-125.
Warner.  ]. W. (1979). Digital-transport model stiulx of diJMipropyl-
  mi'thylphosphoiiate  (DIMP) croinul-xx'ater cuiitainiuation at the  Koiky
  Mountain Arsenal.  Colorado. C.S. (.'<•«/. Siirr. ();«'ii-/;i/i- HCJJ. '•)-
  676. 39 pp.
U'ood. L. A. (1972). (>roundwater degradation — i-auso and cures,  in
  Prucvetlinns l-ltli Water Quality Coiif.. I'rli.in.i. III.,  pp. 19-25.

-------
   SECTION 7






SELECTED SLIDES

-------
    £-W
       - '  J If •
    C  -
    o
3.
     q. ^ - ^    A  - -
 r.
         . (V ^ V
         : ^ " °*i}
3-0

-------

-------
at
  il
  dt
l
•^^•^

b
*

d*
*-(}>&.*£
V/ N H-
 r' /x
^c\


-------
  I   (
LfiTTI CE-
                   &*.)!>

-------

I

-------
              FOR.  MO-C-
   f r to/umm  X  6  rows)
                  >• x
 IX =/
                              t
                             AX
                            ,.,_A
Jo \jvt> fifty

-------
              or
  -f
          or
       N
TINTT

 \
M
i   \TQ/U
          FOHFIJDW
      il
/i
H
                 >*•
     !i
        .
              f  wi+k 3
SUMTCH -*

-------
Q
  ~ 1ST W-'-
= w,
w    = w
                     ntt
    -eb

-------
















WELL






node,J



                        L/
         INFLOW = QBc,=zf(Hs -hi)
Schematic:  representation of constant-head boundary condition
< F :i. om Vas?;;,  .1.984 f )

-------
(\ffre


-------
   00
   c
            RELATIVE SORBED CONCENTRATION
  »- o
  M- >-•
  a B>
  m oa
—^t» M


MS

?i?
1TSC *
1 3 M.
 "t M, 0
2 »°°
" oc:
-» 2U

T-S
3 «• »-
£=?
9-03
•/* B M
to  M


•*  «,
    O
    M
        3)
        m
        m

        g

        CO
        CO
        O
        m
        o

        o
        O
        z
        o
        m
        O
        Z

-------
f
or


-------
/ftt
             - AC

-------
                 REACTION TERMS:
             INPUT DATA INSTRUCTIONS
1. SET NREACT = 1 IN COL. 72 ON INPUT CARD 2.

2. SPECIFY VALUES FOR DK, RHOB. AND THALF IN FREE
    FORMAT ON NEW LINE IMMEDIATELY FOLLOWING
    DATA CARD 3.

-------
            b
co
lt/%
      -

-------
\*
     USE OF
           MX
             HMX
      I
    I
    7
    /C
 A7
s«A
          6 7
AT/Vvyv
      VV^A4l^-/
                          Primary
     Primary
      A/X ^

-------
     USE OF SECONDARY SUBGRID FOR TRANSPORT:
             INPUT DATA INSTRUCTIONS
1. SPECIFY NX AS A NEGATIVE VALUE IN FIELD 3 OF
    DATA CARD 2.

2. SPECIFY THE UPPER LEFT AND LOWER RIGHT CORNERS
    OF THE SUBGRID (MX, MY, MMX. MMY) IN FREE FORMAT
    IN A NEW DATA CARD IMMEDIATELY FOLLOWING DATA
    CARD 2.

-------
  SECTION 8






CLASS PROBLEM

-------

-------
(M  .

tO  '

CO  -

O  i

(0  c  '
 u     "•
 Z     e
    0^"  ik
    7  -

-------
: SAMPLE PROBLEM TOR lCTTnV.ilT.iUH JJP 2-3UHEHSIOK&L SOLDTE-TRANSPORT MODEL BASED
, on METHOD OF CflMBm^MjifTgyTCS i

 This exercise is designed to illustrate the application of the  model  to  a
 realistic problem, using a variety of boundary conditions.  The boundaries,
 properties, and stresses are approximations of the long-term contamination
 problem at the Rocky Mountain Arsenal, Colorado, U.S.A.   (Reference:   Konikow,
 1977, -Modeling chloride aoveaent in. the alluvial aquifer at the Rocky Mountain
 Arsenal, Colorado:  U.S. €eol. Survey Water-Supply Paper 2044,  43  p.).   To
 minimize computational costs for the class exercise,  the problem has  been
 simplified to use a relatively coarse -grid and uniform aquifer  properties.

 !The class will be divided into groups .of two.   The basic data file is designed
 to simulate the .problem for a 2(Hyear period using NPTPND - 9, CELDIS =
 0.50, € = 0.20, and «L =MDO ft.  Each group will run several variants of the
 basic problem that requires a modification of the basic data file, as
 follows:

    - i.   JfPTPNJ* = J.
      2.   NPTPND - 4
    '3.  ''NPTPND'-* 5
      -4,   tfPTPND - 8
                            as a nrrngfranfr-hcad condition*
      6.   Longitudinal dispersivity =   0.0 feet
      7.   IrangitudiTial dispersivtty -  20.0 feet
      ^«.   Twngitxtatnal flispersivity -"--500; Q -feet
      9.   Transverse dispersivity   =   0.0 feet
     10.   Transverse dispersivity   = 100.0 feet
     11.   CELDIS" 0.25
     J2
     13.   Porosity = 0.30
     It.  'Use fi x 3 sulngrifl .for
     16.      — 10 years
    _!&.   1^ = J^D-^nd ^ =— LO
     19.   10 years of contaainati«n followed by 10 years of flushing with
          •                      (same Q, C'  =
                      lal 4"ff^T

 The results of all runs wlH be compared and discussed in class so that all
 participants can see tl» Tesults of runs Trade by other groups.  Each group
 "will ~plt>t "tare uuuuuiiliatipn -veiBnB tine for observation well 3 for their runs.
 ComparisCTW yill •« i i^nrtygtg t^ sensitivity of precision, accuracy, and
 efficiency of ±fte srrtTrt-jrm-*T> -pByj^jpns of the selected parameters.

-------
       CLASS PROBLEM — Input data  for base conditions
ROCKY HT  ARSENAL -- APPROXIMATE SOLUTION
   1    1   10  132000   1    7   1 100   4
                                            1
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0
0
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0
0
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0
0
0
0
0
0
0
0
0
0
0
  20.  .001 0.20 100.  0.0   0.0  0.02000.2000.  0.3  0.5   1.0
 6 e
0310  0.2
0510  0.2
0710  0.2
0605 -1.0      1000.
1  0.01
  0
  0
  0
  0
  0
  0
  0
  0
  0
  0
1  40.0
  0
  0
  0
  0
  0
  0
  0
  0
0  0.0
1  1.0

0011111100
0000000000
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
                   0
                   0
                   0
                   0
                   0
                   0
                   0
                   0
                   0
                   0
                   0
0111111110
  1   0.1
 1   1.0
    10.0


250 250 250 250 250 250
      20  25  30  35  40  45  50  55

0  10.0

-------
                                                              <*r
                                  ritdi

                                                                                                                                "1
                                                                                                                                 -i
                                                                                                                                 j

                                                                       /•'••.

C too-
I
^
h—
   ¥»t>


                                                                                                   :.::
o 100
                                                                             — r
   IX)

   /to
                                      -.-:/':-
                                      -y:-~
                                      /••;:'••
                                                                   ::j
                                                                                                            •-.


7     f    f
                                                                   /O
                                                                   //    Av
                                                                                                                    -JT": :r:;T
•--y—-

-------
                               v7~-iHe   or  Ob*-
  JDOO
^

C too
o 300-
   0~
                                      I
                                     I
t   (0    //
                                                             »3
                                    J-t»ii
ll
ff  /;   TO

-------
  JflOO
0/AHe^i
                                                                ;

                                                                'pi
   (00
<
*
0 30CK


  /IC
               2-     5
      f    L    7     V    Ct
     J     b    '     *    7


     fl     or-   y~'"lt    ('
n   if    /;    10

-------
                                  PARTICLE PATHS
                                                                          j mis
                                                                         10
                                                       20
(note:  map  is  mirror  image;  east & west are reversed from original map and grid)

-------
      CLASS PROBLEM — / = 20 years
  Contours of concentrations of nodes
Contours of concentrations of particles

-------
      SECTION 9






CLASS PROBLEM - RESULTS

-------
                                                                                               MOC
                           -
  JD00
>*'ii
   Yw
     I .....
         -DtitripfcOv^

         fktttk  resu.lrs :
                                              'sttoAitiin^'
                                          i *t
3

o
  /to
            rrj:
              j i •

                 ih^
                           -::J
'".'I'
     :rE,u-i'
                                                 ':SL
                                                       i::-:|.


                                                                            in
                                           O.I1.


                                           OJd^70
•for conjT.-( i
                                      7    f   f    M    /i   IT-   &
                                                                                   -I
                                                                                    ..
                                  Or-
                                                                                               /;

-------
                                        >ni(MI M > M 10 I MH

                                        >1H IOIM««»OIM»>
-------
- • . . .

v.  T~iH-e   xr  Ob*-
 nofa. t>, 
-------
•fhttcJL-"retijTS. :  Vary
                                                                   If   If   ZO

-------

-------
ccr  Obs- frW/  3   ^o
-------
                                        imow to * » 10 » •*«

                                        1IH IOINAMOH»HIM»iOGtlJWVm *C«MtID
                                             AMOTCmiHIMOGtlJWVttf ACCIXtID                                                 S


                                              v±    Tine,     or    Obi-  \J*II    3       ftiofo.    4,
:           t ::::::::   :  :::::::       ::::::  :  •  s    :;::•:::    ::     ::::::::
• HsISImSSiSSSSEiS! SI  ••  •!•••••       ••••••  •  I  •    -• .-^ t • •    ••     ••••'!• *

: SfUiiiMhissiiM^ "  »  »•»::       "•"•"•  •  •  -    ••  i::: !    :      •::::
• •l*li*!«l!*l»"i*!""" ••  *•  •••••••       ••••••  •  •  •    •• •{• • • i    ••     I!!!!!! I
• •••••••••••••••••••B ••  ••  •••••••       ••••••  •  •  •    •'. ••• • • •    ••     SISSSSI S
• •••••{••••••••IiiaiB ••  ••  •••!•••       ••••••  •  •  •    '.* ••• • • •    ••     ••••••• •
                                                 • '.  ••••••••••§•••••••••[••••
                                                 - •  ••••••••••••••••••••••••i
                                                 • •  •••••••••••••••••••••••••

                                                 • t  •••••••••••••••••••••••••
                                                 • •  •••••••••••••••••••••••••
                                                 • •  ••••••••••^•••••{••••••!
                                                 • •  ••••••••i«iiuiii|iiaiiii
b
f
/'
                                                                                                                       /JT

-------
                                                         miOM *> • ja to i MCH

                                                         MM IOIMAnOWIHIIMtftOG«miVt*iifl91i**iI«]i
                                                                                                                                                    •                     • ••••. ••••••••••••••••••••
                                                                                                                                                    ••                    • ••••^•••••••••••{••••••l
                                                                                                                                                    SS                    • •••••i.MiMiill"**"Il"3
                                                                                                                                                    S!                    "5"1*5*v•!•••••••ilit""li"
                                                                                                                                                    ••                    • ••••••^••••••••iiiliMii
                                                                                                                                                    ••                    ••••••••> •••••••••••••••••

                                                                                                                                                    •                     ::::::::ri::;i:::::::i:>:

                                                                                                                                                     :                    :::i:::.:r::!i::::!]:::::
                                                                                                                                                     •                    ••••••••••.•••••••••••••••
                                                                                                                                                     •                    ••••••••••»•••••••••••••••
                                                                                                                                                     •                    ••••••••••• •••••••••{••••i
                                                                                                                                                     *                    •••••BBB|iii•••••••••••!••
                                                                                                                                                     •                    •••••••itiik*••••••!•!••••
                                                                                                                                                     •   •                •••••••••••• <••••••••••••• •
                                                                                                                                                     •   •                ••»•••••••••>•!!•••••*•••• •
                                                                                                                                                     1 • •    f           ••••••••••••!«•••••••••••• •
                                                                                                                                                     • • ••   • •           •••••••••••••  •••••••••••• •
i. noves con'ierta •
3 .1 n « A
0
0
0
0
0
0
0
0
0
0
0
0
10
10
10
10
10
12
12
t2
19
10
13
0
10
Id
10
to
It
It
17
11
11
11
10
3
10
10
14
112
III
411
0
0
21
142
2>
0
10
11
20
It
21
It

III
m
S10
21
0
20090
10
10
It
17
11
tt
41
22*

ItO
111
21
l«»92
19
to
t>
21
»l
0
0
t9)
111
101
11

to
10
11
12
IM
921
>10
201

tt
10

10
10
10
11
tr
21
It
2t
It
to
to

0
9
0
0
0
0
0
0
0
0
0
                                                                                                                                                                                   ••  •  ••••• t «••
                                                                                                                                                                                   •i  •  ••••• •. ••
                                                       •f

                                                       A/
                                                                                                     10
/y
t       j

-------
                    CLASS PROBLEM —  Input data for Run 5
(*indicates  lines modified or inserted in comparison to base conditions)
            SOCKY  MT i»SEN4L -- flPPROXIMSTE  SOLUTION — CONSTANT-HEAD SOURCE
1 1 13 132000 1
20. .001 3.20 100. 0.0





0
6 5
3 5
6 8
3 3
V 5
310 0.2






0510 0.2
0710 0.2
1











0.31
3 1 . 1 .
0 1 . 1 . 1
•3 1 . 1 . '
Q 1 . 1 .
0 1 . 1 .
0 1. 1 .
C 1. 1.
0 1. 1.
0 1. 1.
3 1. 1.
C 1. 1.

. 1. 1 .
1 . 1 .
. 1. 1.
1 . 1 .
. 1. 1 .
. 1. 1 .
1. 1.
1. 1 .
. 1. 1 .
. 1. 1 .
. 1. 1 .
                                   7   5 100   3   9
                                   0.0  0.02000.2000.
                              2100
                              0.3  0.5  1.0
           1  40.0
0 1
0 1
0 1
0 1
3 1
0 1
0 1
0 1
0 1
0 1
0 1
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 0
1 1
1 1
1 1
11110
11113
111110
111110
11110
10110
10113
11110
11110
11110
111110
           0  0.0
           1  1.0

           0011111 IOC
            0000020000
            0111111110
                0.1
                0.1
               1 .0
10.0
1030.
                    250 250 250 250 250 250
                              2*7
                 20  25  30  35  40  45  50  55

            0  10.0

-------
                   CLASS PROBLEM ~ Input data for Run 14
^indicates lines modified or inserted in comparison to base conditions)
ROCKY MT ARSE
1 1 -10
3/3/3/11
20. .001 C.
o 3
3 5
6 a
8 8
9 5
0310 0.2
0510 0.2
0710 0.2
0605 -1.0
1 0.01
n 1 . 1 .










1











0
1
0 1
0 1
0 1.
0 1
0 1
0 1.
0 1.
0 1
0 1.
0 1
40.0
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0 1 1
0.0
1.0
1 .
1 .
1.
1 .
1 .
1.
1 .
1 .
1 .
1 .

1
1
1
1
1
1
0
0
1
1
1


NAL — APPRCXIMAT
132000 1 7
20 100. 0.0 0.0
1000.
1 . 1 1 - i -
1 . 1
1 . 1
1. 1
1 . 1
1 . 1
0 1
0 1
1 . 1
1. 1
1 . 1

1 1
1 1
1 1
1
1
1
1
1
1
1
1


. 1 .
. 1 .
. 1.
. 1 .
. 1 .
. 1.
. 1 .
. 1 .
. 1.
1 .

1 1
1 1
1 1
1 1
1 1
0 1
0 1
1 1
1 1
1 1
1 1


1 . 1
1
1. 1
1 1
0 1
0
1. 1
1
1. 1
1 .

1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0


                                      5100   4   9   1   1   0   0   0   0

                                       0.02000.2000.  0.3  0.5  1.0
            0011111100
            0000000000
            0111111110

             1   0.1
            1   1.0
10.0
                   250 250 250 250 250 250
                 20  25  30  35  40  45  50  55

            0  10.0

-------
                    CLASS PROBLEM — Input data for Run 18
('indicates lines modified or inserted in comparison to  base  conditions)
            ROCKY «T ARSEM4L — APPROXIMATE  SOLUTION    •* RETARDATION AND DECAY •*
              1   1  10  132COO   1   7   5103   49  12   00001
              20. .001  0.20 100.  3.0  0.0  O.C2003.2000.  0.3  0.5  1.0
              0.1030CEOO  2.00000500  3.15576E03
             t> 3
             3 5
             4 S
             3 6
             9 5
            031C  0.:
            0510  0.2
            0710  0.2
            06C5 -1.0    1000.
            1  0.01

                                              0
                                              C
                                              0
                                              0
                                              0
                                              0
                                              0
                                              0
                                              0
                                              0
                                              0
0
o
3
D
j
0
0
0
0
c
0
1 . 1 . 1.
1 . 1 . 1.
1 . 1 . 1.
1 . 1 . 1.
1 . 1 . 1 .
1 . 1 . 1.
1. 1. 0
1 . 1 . 0
1 . 1 . 1 .
1 . 1 . 1 .
1 . 1. 1.
1. 1. 1.
1. 1. 1.
1. 1 . 1.
1. 1 . 1.
1 . 1 . 1 .
1. 1. 0
1. 1. 0
1. 1 . 1.
1 . 1. 1.
1 . 1 . 1 .
1. 1 . 1.
1 . 1.
1. 1.
1 . 1.
1. 1.
1. 1.
1 . 1.
1. 1.
1. 1.
1. 1.
1 . 1 .
1 . 1.


1











C
1
0
c
0
i.
0
0
0
0
0
0
0
0
0
0
3
0
1
1 .
1
1 .
0.0
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
.0
.0
1 .
1 .
1.

1
1
1
1
1
1
0
0
1
1
1


1 . 1 . 1.
1 . 1 . 1 .
1. 1. 1 .

1111
1111'
1111
1111
1111
1101
1101
1111
1111
1111
1111


.
.
.

0
0
0
0
A
0
0
0
0
0
0


            0011111100
            0000000000
            0111111110

             1   0.1
            1    1.0
10.0
                    250  250 25C 250  250 250
                 20  25   30  35  40  45  50  55

            C   10.0

-------
                    CLASS  PROBLEM ~ Input data for Run  19
(*indicates  lines modified or inserted in comparison  to base conditions)
           ROCKY
             1
            10.  .
           6 5
           3 5
           6 o
           3 &
           ? 5
          3310   0
          0510   0
          071C   0
          0605  -1
          1  0.01

            0
            0
            0
            0
            0
            0
            0
            0
            0
            0
            0
          1  40.0
MT ARSENAL — 10-YR CONTAMINATION +  10-YR FLUSHING
2  10  132000   1   7   5  100   4   9   1   1   0   0
001  0.23 100.  0.0  0.0  C.02000.2000.  0.3 0.5  1.0
.2
.2
.2
.0    1000.
1.
1 .
1.
1.
1.
1.
1.
1 .
1
1.
i .
1. 1.
1. 1.
1 . 1.
1. 1.
1. 1.
1. 1.
1. 0
1 . 0
1 . 1 .
1. 1.
1. i.
,
.
.
B
m
f
f

\ .
i.
i.
. 1. 1.
. 1. 1.
. 1. 1.
. 1. 1.
. 1. 1.
. 0 1.
. 0 1.
. 1 . 1 .
1 . 1 . 1 .
1 . 1. 1.
. 1. 1.
1.
1 .
1.
1.
1.
1.
1.
1 .
1
1.
1.











0
1
0
0
0
0
0
u
0
0
0
0
0
0.
1.
1
1
1
1
1
1
1
1
1
1
1
0
0
0011111
1
1
1
1
1
1
1
1
1
1
1


100
1
1
1
1
1
1
0
0
1
1
1



1
1
1
1
1
1
1
1
1
1
1



1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
1 C 1
1 0 1
1 1 1
1 1 1
1 1 1
1 1 1



1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 0
1 o



0000000000
          0111111110
           1   0.1
          1   1.0
      10.0


  250 250 250 250 250 250
                20  25  30  35  40  45  50  55

           0  10.0
           1
             1   1   7 100   4   5   0   0   0
           0310  0.2
           0510  0.2
           0710  0.2
           0605 -1.0      10.
                                10.  0.0  0.0

-------