United States
                   Environmental Protection
                   Agency
Robert S. Kerr Environmental
Research Laboratory
Ada OK 74820
                   Research and Development
EPA/600/S2-86/062 Sept. 1986
&EPA         Project Summary
                   Performance  and  Analysis  of
                   Aquifer Tracer Tests with
                   Implications for  Contaminant
                   Transport  Modeling

                   Fred J. Molz, Oktay Giiven, Joel G. Melville, and Joseph F. Keely
                    The scale-dependence of dispersivity
                   values used in contaminant transport
                   models to estimate the spreading of
                   contaminant plumes by hydrodynamic
                   dispersion processes was investigated
                   and found to be an artifact of conven-
                   tional modeling approaches (esp., verti-
                   cally  averaged parameters in two-
                   dimensional plume simulations). The
                   work  reported here shows that varia-
                   tions  in hydraulic conductivity with
                   depth result in significant variations in
                   ground-water flow and contaminant
                   transport velocities;  it is the resulting
                   velocity variations that, if vertically av-
                   eraged,  give rise to apparent scale-
                   dependency of dispersion (e.g., in-
                   creased dispersion with increasing
                   travel distance). Special depth-selective
                   observation well designs  are recom-
                   mended by the authors for use in tracer
                   tests,  so that detailed estimates of the
                   variations in hydraulic conductivity,
                   and flow and transport velocities can be
                   obtained. Innovative modeling tech-
                   niques, that take advantage of the de-
                   tailed information obtainable from such
                   tests (by emphasizing advective trans-
                   port, as  opposed to dispersive trans-
                   port),  have been developed by the au-
                   thors. These modeling techniques are
                   shown to have an element of true pre-
                   dictive ability, being able to closely sim-
                   ulate actual results with little or no cal-
                   ibration.
                    This Project Summary  was  devel-
                   oped by  EPA's Robert S. Kerr Environ-
                   mental Research Laboratory, Ada, OK,
                   to announce key findings of the re-
                   search project that is fully documented
in a separated EPA publication of the
same title (see ordering information at
back).

Introduction
  Due to worsening national environ-
mental problems, hydrologists are be-
ing asked to identify, assess or even an-
ticipate situations  involving ground
water contamination. Many of the U.S.
Environmental Protection Agency's reg-
ulatory activities relate to prevention or
remediation of such situations. In both
regulatory and assessment activities,
increasing use is being made of com-
plex mathematical  models that are
solved with the aid  of the digital com-
puter. Some of the principal areas
where mathematical models can be
used to assist in the management of
EPA's ground water protection pro-
grams are:
  (1)  appraising the physical extent,
    and chemical and biological qual-
     ity, of ground-water reservoirs
    (e.g., for planning purposes),
  (2) assessing the potential impact of
    domestic, agricultural, and indus-
    trial practices (e.g., for permit is-
    suance, EIS's, etc.),
  (3) evaluating the probable outcome
    of remedial actions at hazardous
    waste sites,  and  of aquifer
    restoration techniques generally,
 (4) providing exposure estimates and
    risk assessments for health-
    effects studies, and
 (5) policy formulation (e.g., banning
    decisions, performance stand-
    ards).

-------
These activities can be broadly catego-
rized as being either site-specific  or
generic modeling  efforts, and can  be
further subdivided into applications to
point-source or nonpoint-source prob-
lems. The success of these efforts de-
pends  on the accuracy  and efficiency
with which the natural process con-
trolling the behavior of  ground water,
and the chemical and biological species
it transports, are simulated.
  Models are collections of partial dif-
ferential equations that contain a num-
ber of parameters which represent aqui-
fer physical  properties and  must  be
measured in  the field. Of the various
parameters involved, the hydraulic con-
ductivity distribution  is of major impor-
tance.  Other parameters such as those
relating to sorption, hydrodynamic dis-
persion, and chemical/biological trans-
formation are  important also, but
hydraulic conductivity is more funda-
mental because, combined with the hy-
draulic gradient and  porosity, it relates
to where the water is moving and how
fast. Therefore, this  communication is
devoted mainly to the conceptualization
and measurement of hydraulic conduc-
tivity distributions and the relationship
of such  measurements to dispersion
(spreading) of contaminants in aquifers.
Discussion
  For the  most part, contemporary
modeling technology  is built around
two-dimensional models having physi-
cal properties, such as hydraulic con-
ductivity, that are  averaged over the
vertical thickness of the aquifer. In such
a formulation, the longitudinal disper-
sivity is forced to be the major aquifer
property related to contaminant spread-
ing. This is not due  to any fundamental
theoretical limitation. The major limita-
tion is that dependable and economical
field approaches for measuring
vertically-variable hydraulic conductiv-
ity distributions are  not available. In the
absence of such data, one has no choice
in a modeling sense but to use some
type of vertically-averaged advection-
dispersion approach built  around full
aquifer longitudinal dispersivities.
   In order to begin to overcome this
limitation,  a series of single-well (Fig-
ure 1) and two-well (Figure 2)  tracer
tests were performed at a field site near
Mobile, Alabama. A major  objective of
this communication is to describe these
tracer tests and  discuss some practical
               Injection
                        J (t)
   Withdrawal

  Q = QOUT
I
                   Upper
    / /Sf S/SS/ / //// ///
  Confining Layer
                                         Injection-
                                         Withdrawal
                                         Well
                                          Observation
                                          Well
                                          With
                                          Multilevel
                                          Samplers
                           Lower Confining Layer

Figure 1.    Vertical cross-sectional diagram showing single-well test geometry.
implications of the results with regard
to modeling of contaminant dispersion
in aquifers. The tests utilized multilevel
sampling  wells which had to be  de-
signed and installed carefully.
  The authors describe the design and
construction of a multilevel sampling
well system for use with chemical trac-
ers in a variety of confined and uncon-
fined aquifers. The actual sampling sys-
tem is not perfected and should be
viewed as a prototype. However, it ap-
peared to work in a satisfactory manner
at the Mobile site. As shown in Figure 3,
the screened portions of  these multi-
level observation wells  are composed
of three-foot long slotted sections alter-
nating with seven-foot long solid sec-
tions.
  As also shown in Figure  3, a two-inch
diameter  PVC  insert was constructed
with slotted and solid portions that
matched with those of the observation
well screen. The insert was designed to
hold any wires, tubing, or  instrumenta-
tion that ultimately would be placed in
an observation  well.  Composed of
threaded ten-foot long sections, the in-
serts extended all the way to the land
surface. In order to isolate the various
sampling  zones, inserts were fitted ex-
ternally with cylindrical annular inflat-
able packers.
  After the required probes, tubing and
wires were placed within the inserts, the
sampling sections were isolated inter-
nally with silicone rubber plugs. The
complete insert was constructed on the
surface, then placed in the well, using a
crane,  positioned  and  the packers  in-
flated.  After installation, each isolated
sampling zone appeared  as shown in
Figure 3. A conductivity probe was
placed near the zone center, and two
lengths of vacuum tubing connected
the sampling zone to the  surface. This
tubing could be  used  with  peristaltic
pumps to mix the contents of the sam-
pling zone and to obtain ground water
samples for data on the arrival of tracers
used in the experiments to simulate
contaminant movement.
  In the recent past, some hydrologists
advocated the use of single-well or two-
well tracer dispersion tests as a means
of measuring full-aquifer longitudinal
dispersivity.  However,  analyses of
single- and two-well tests of the Mobile
site  and at the Borden site  in Canada
(both with stratified aquifers) indicated
that if this is done, the resulting number
will  have little physical meaning. In the

-------
           Injection Well
             (Source)
             Withdrawal Well
                 (Sink)
                                                             Multi-Level
                                                             Observation
                                                             Well
                                Plan View
    x/xxxxxx
    xxXXXXXXXXXxxx
                        Vertical Section in x-z Plane
Figure 2.   Two-well test geometry in a stratified aquifer.
case of single-well tests, the full aquifer
breakthrough curves measured in ob-
servation wells are  determined mainly
by the hydraulic conductivity profile in
the  region between  the injection-
withdrawal well and an  observation
well if the travel distance  between the
injection-withdrawal well  and the ob-
servation well  is typical of most test
geometries. Thus,  information about
the hydraulic conductivity profile is nec-
essary for  meaningful  test interpreta-
tions. The relative concentration versus
time  data  recorded at the  injection-
withdrawal well itself  is  primarily  a
measure of the combined  local and
semi-local  dispersion  that has taken
place during the experiment. The ef-
fects of such dispersion depend in part
on the hydraulic conductivity distribu-
tion  in the  aquifer,  and in part on the
size of the experiment. As the size of the
experiment increases,  the  effects of
local vertical dispersion will become
large compared to the effects of local
horizontal (radial) dispersion.
  The two-well simulations  of  experi-
ments conducted at the Mobile  site
show that the concentration versus time
breakthrough curve measured at the
withdrawal well would be very sensitive
to variations of the hydraulic conductiv-
ity in the vertical. Without the use of the
kind of multilevel observation wells
used in the test, little useful information
about the hydraulic or dispersive char-
acteristics of the aquifer (e.g.,  aquifer
stratification or values of local  disper-
sivities) would be  obtained.  Factors
such as the length of the injection pe-
riod, the use of recirculation, and the
physical size of the experiment all have
strong effects  on the  breakthrough
curve measured at the withdrawal well
(Figure 4) making the interpretation of
field results  difficult, especially with
conventional modeling approaches
(Figure 5). This can be addressed more
satisfactorily if aquifer  stratification
(Figure  6) is measured and properly
taken into account (Figure 7).


Conclusions
  Based on the above observations and
the large values for full-aquifer disper-
sivities that consistently result from cal-
ibrated  areal  ground water transport
models, the authors believe that the fol-
lowing working conclusions are war-
ranted:
   I. Local longitudinal hydrodynamic
     dispersion plays a relatively unim-
     portant role in the transport of
     contaminants  in aquifers. Differ-
     ential advection (shear flow) in the
     horizontal direction is much more
     important.
   II. The concept of full-aquifer dis-
     persivity commonly used in
     vertically-averaged (areal) models
     will  not be applicable over  dis-
     tances of interest in most contam-
     ination  problems. If one has  no
     choice but to apply a full-aquifer
     dispersion concept, the resulting
     dispersivity will not represent a
     physical property of the aquifer.
     instead, it will be an  ill-defined
     quantity that will depend on the
     size and type of experiment used
     for its supposed measurement.
  III. Because of conclusion II, it makes
     no sense to perform tracer tests
     aimed  at measuring full-aquifer
     dispersivity. If an areal model is
     used, the modeler will end up ad-
     justing the dispersivity  during the
     calibration process anyway, inde-
     pendent of the measured value.
  IV. When tracer tests are performed,
     they should be aimed at determin-
     ing the hydraulic conductivity dis-
     tribution. The theoretical and ex-
     perimental work presented in this
     report indicate that the variation
     of horizontal hydraulic  conductiv-
     ity with respect to vertical position
     is a key aquifer property related to
     the spreading of contaminants.
  V. Two- and three-dimensional mod-
     eling approaches should  be  uti-
     lized which emphasize variable
     advection rates in the  horizontal
     direction and hydrodynamic dis-
     persion in the transverse direc-
     tions, along with sorption and
     microbial/chemical degradation.

-------
                                          2" PVC Removable Insert
                  Aquifer
     70ft.
    (21 m)
\

t

I

TT
M
I!
m
is
' t
3'
T
J_

(0.9 1 mf
(2.1 m)
                                 Vacuum ^
                                 Tubing
                                       Plug
Figure 3.    Diagram of a completed multilevel sampling well. This and similar systems were
           used at the Mobile site.
      25.0

      22.5

      20.0

      17.S

      15.0

      12.5

      10.0

       7.5

       5.0

       2.5
I
c
.g
I
o
O
            40
                  120    200   280    360   440    520   600    680   760
          0    SO     160    240    320   400   480    560    640    720

                                       Time (hrs)

Figure 4.    Measured tracer concentration versus time in the withdrawal well during the
           two-well test.
the authors are suggesting, therefore, is
that the time may have arrived to begin
changing  from a homogeneous to a
vertically-stratified concept when deal-
ing with contaminant transport, realiz-
ing fully that such an approach will be
interim in nature and  not totally correct.
Performance and simulation of several
single- and double-well tracer tests sug-
gests that the  stratified approach is
much more compatible with valid phys-
ical concepts,  and  at  least in some
cases, results in a mathematical model
that  has a degree of true predictive
ability. Nevertheless, real-world appli-
cations will undoubtedly  require cali-
bration, which in the approach recom-
mended here would involve varying the
hydraulic  conductivity  distribution
rather than the longitudinal dispersivity.
The benefit is that when calibrating with
an estimated hydraulic conductivity dis-
tribution, one is dealing with the physi-
cal property that probably dominates
the dispersion process,  rather than
dealing with a fitting parameter that has
little,  if any, physical  relationship to the
problem.
  The change from  a  vertically-
homogeneous  to a vertically-stratified
approach will not be easy from a field
measurement viewpoint,  nor will it be
inexpensive. One obvious implication
of this study is that until better field
characterization tools  are made  rou-
tinely available, any type of ground
water contamination  analysis and recla-
mation plan will be difficult, expensive
and possibly unable  to meet all of the
desired objectives in a reasonable time
frame.
  VI. In order to  handle the  more
     advection-dominated flow sys-
     tems described in conclusion V,
     one will have to utilize or develop
     numerical  algorithms that are
     more accurate than those utilized
     in the  standard dispersion-
     dominated models.
                                    Summary
                                      Much of contemporary modeling
                                    technology related to contaminant
                                    transport may be viewed as an attempt
                                    to apply vertically homogeneous aqui-
                                    fer concepts to real aquifers. Real
                                    aquifers are not homogeneous, but they
                                    are not perfectly stratified either. What

-------
    501-
I
o
    45
    40
    35
       30
%   25
I
       20
    15
    JO
                             O
                             O
                            O
                                 8
                                 Q  O
                                O

                                    O
          40  / 120     200    280    360     440    520    600    680    760
Figure 5.
             80      160    240     320    400    480     560     640    720
                                        Time (hrs)

         Calculated tracer concentration versus time in the withdrawal well based on an
         assumed homogeneous, isotropic aquifer with no local dispersion (circles} shown
         together with the results of the present two-well test (full line).

-------
            38


            40


            42


            44


            46


      -§.    48

      f
      <§    50


            52


            54


            56


            58
                   • 1    .2    .3    .4   .5    .6    .7    .8    .9    1.
Figure 6.    Normalized hydraulic conductivity distribution inferred from travel times measured
            during the two-well test.



^
1
c
.0
1
c
8
c
5




25.0
22.5
20.0

17.5
15.0

12.5

10.0

7.5
5.0

2.5
n rt

>%m
/>§£>
Ot \ ^
•' \°°
I V^.-o
,' \ O
! V o o
I \ ^ ^/\
— . f\- i i \
,' '^' ' o\

o / •/N" '
-------
      Fred J. Molz, Oktay Guven, and Joel G. Melville are with Auburn University,
        Auburn, AL 36849; and Joseph F. Keelyfformerly the EPA Project Officer, see
        below for  present contact) is with Oregon Graduate Center in Beaverton,
        Oregon. The complete report, entitled "Performance and Analysis of Aquifer
        Tests with Implications for Contaminant Transport Modeling," fOrder No. PB
        86-219 086/A S; Cost: $11.95, subject to change) will be available only from:
             National Technical Information Service
             5285 Port Royal Road
             Springfield, VA 22161
             Telephone: 703-487-4650
      Inquiries should be directed to:
             Robert S. Kerr Environmental Research Laboratory
             U.S. Environmental Protection Agency
             Ada, OK 74820
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S2-86/062
                                                        2863
                                «•


-------