A study of the effectiveness of electrical resistivity surveys for detecting plumes of contaminated ground water at three waste-water impoundment sites in the Georgia Coastal Plain [1986]

A STUDY OF THE EFFECTIVENESS OF ELECTRICAL RESISTIVITY SURVEYS FOR DETECTING PLUMES OF CONTAMINATED GROUND WATER AT THREE WASTE-WATER IMPOUNDMENT SITES IN THE GEORGIA COASTAL PLAIN BY John C. Donahue
OPEN FILE REPORT 86-5
DEPARTMENT OF NATURAL RESOURCES J. Leonard Ledbetter, Commissioner
ENVIRONMENTAL PROTECTION DIVISION Harold F. Reheis, Assistant Director
GEORGIA GEOLOGIC SURVEY William H. McLemore, State Geologist
J
ATLANTA 1986

The preparation of this report was financed in part through grant no. G004353-86-l from the U. S. Environmental Protection Agency under the provisions of Section 106 of the Federal Water Pollution Control Act of 1972 as amended.
I
I
-j

TABLE OF CONTENTS

INTRODUCTION . ...... . ........... . ............... . ........... . .... . 1

::r~~.::::::::::::::::::::::::::::::::::

3 8

NATURE OF TH.E WASTE . . . . I I I 10
SCHOOL SEWER POND SITE, WILKINSON COUNTY, GEORGIA . . .............. 12

TEXTILE MILL WASTE-POND COMPLEX, JEFFERSON COUNTY, GEORGIA ....... 29

MUNICIPAL SEWER POND SITE, DOOLY COUNTY, GEORGIA . .. .............. 42

CONCLUSIONS . . . . . . . . . . 51 I I I I I I I I I

ACKNOWLEDGEMENTS . 55 I I I

REFERENCES .......... ....... 56 ~

I I I I I I I I

APPENDIX I: CALCULATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

APPENDIX II: ADDITIONAL DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

i
. J

LIST OF ILLUSTRATIONS

FIGURE

PAGE

1 Locations of Surface Impoundments Examined for This Investigation .... , .... , .................... , . . . . . . . . . . 2

2 Schematic Diagram of the Electrode Arrays Used for BOSS System Measurements ,,,,,,,, , , , ,, . .. . .. ..... ... .... 9

3 Detail Map of the Wilkinson County School Pond Site ....................................................... 13

. 4 Log-Log Plot of Apparent Resistivity Data for Sounding B, School Sewer Pond Site ........................ 16

- 5 _.Log-Log Plot of Apparent Resistivity Data for Sounding C, School Sewer Pond Site ........................ 17
6 Cummulative Plot of Apparent Resistivity Data for Sounding C, School Sewer Pond Site . . . . . . . . . . . . . 18

7 Log-Log Plot of Apparent Resistivity Data for Sounding D, School Sewer Pond Site ....................... 19

8 Log-Log Plot of Apparent Resistivity Data for Sounding E, School Sewer Pond Site , . , . , , .............. .. , . 20

9 Log-Log Plot of Apparent Resistivity Data for Sounding F, School Sewer Pond Site . . . . . . . . . . . . . . . . . . . . 21

10 Comparison of Depth and Resistivity Predictions for Soundings B, C, D, and F, School Sewer Pond
Site ....................................................... 23

11 Contour Plot of Resistivity Profile Data, School
Sewer Pond Site ................................... , . , . . . . . . 26

12 Panel Diagram of Resistivity Profile Data Around the School Sewer Pond Site ................................. 27

13 Detail Map of the Textile Mill Waste-Pond Site, Jefferson County, Georgia ................................ 30

14 Log-Log Plot of Apparent Resistivity Data for Sounding A, Textile Mill Waste-Pond Site ................ 32

15 Log-Log Plot of Apparent Resistivity Data for Sounding B, Textile Mill Waste-Pond Site . . . . . . . . . . . . . . . . . 33

.. J
ii

FIGURE

PAGE

16 Cumulative Plot of Apparent Resistivity Data for Sounding B, Textile Mill Waste-Pond Site ................. 34

17 Log-Log Plot of Apparent Resistivity Data for Sounding D, Textile Mill Waste-Pond Site ............... 35

18 Cumulative Plot of Apparent Resistivity Data for Sounding D, Textile Mill Waste-Pond Site ................... 36

19 Comparison of Depth and Resistivity Predictions for Soundings A, B, and D, Textile Mill Waste-
Pond Site .......... , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

20 Detail Map of the Municipal Sewer Pond Site, Doo ly County, Georgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

21 Log-Log Plot of Apparent Resistivity Data for Sounding D, Municipal Sewer Pond Site . . . . . . . . . . . . . . . . 46

22 Cumulative Plot of Apparent Resistivity Data for Sounding D, Municipal Sewer Pond Site .................. 47

23 Log-Log Plot of Apparent Resistivity Data for Sounding E, Municipal Sewer Pond Site ..................... 48

I
1
.I
iii
J

INTRODUCTION
This study, continuing investigations begun in 1985 (Donahue and Meehan, 1986), examines the effectiveness of electrical resistivity surveys in detecting plumes of contaminated ground water emanating from waste-water impoundments.
Resistivity surveys have been used in similar studies before (Warner, 1969; Stollar and Roux, 1975; Benson et al., 1982) and have been found useful for detecting plumes if the contaminants are electrolytes. Three waste impoundments were selected for examination: a sewer pond serving a school in Wilkinson County, Georgia; a sewer pond serving a small municipality in Dooly County, Georgia; and a pond complex treating wastes from a textile mill in Jefferson County, Georgia. The locations of the impoundment sites are shown in Figure 1.
The criteria used to select ponds for the present study were similar to those used in the previous study. The impoundments would be situated in the outcrop belts of Cretaceous and Tertiary clastic sediments in the Georgia Coastal Plain, which contain the recharge areas of certain major aquifers. The selected impoundments would also have received high health-hazard and contamination-potential ratings in the Surface Impoundment Assessment Program (Carver, 1982) (these rankings are rough and preliminary). The impoundments would be situated in areas subject to a minimum of cultural interference and would be sufficiently small and free of brush to permit timely surveying. The impoundments would also be expected to treat waste water with a higher electrolyte content than local ground water.
J
1
. j

Figure 1. Locations of Surface Impoundments Examined for This Investigation. 1) Textile Mill Waste-Pond Complex, Jefferson County; 2) School Sewer Pond, Wilkinson County; 3) Municipal Sewer Pond, Dooly County.

METHODS
Resistivity methods determine the ability of the ground to resist the flow of electricity. The ability of the ground to resist electric currents depends on the amount of, and the composition of, a) the solid earth materials, b) the ground water, and c) the non-conducting fluids (gases are generally the only non-conducting fluids present in earth materials in the Georgia Coastal Plain). Increased electrolyte content in the ground water results in decreased ability of the ground water and the ground --to resist electric currents. Ground water of low electrolyte content that has locally become contaminated with electrolyte-rich waste water is therefore a suitable target for resistivity methods.
Resistivity methods generally involve injecting a measured flow of electricity into the ground, via a system of electrodes, and measur1ng potential differences with other electrodes placed suitably nearby. The current measurements (amperes) and the potential measurements (voltages) are then used, along with some factor related to the electrode arrangement (typically expressed in feet or meters), to compute resistivities (generally expressed as ohm-feet or ohm-meters).
A resistivity, as calculated above, represents the true resistivity of a homogeneous electrically isotropic material. The ground however, is typically a heterogeneous material and may be partly or wholly anisotropic. A resistivity, as determined above, then is related only in some complicated fashion to the true resistivities of the various parts of the ground and is, therefore, called an apparent resistivity.
I
I j
3
J

The present study used three types of electrode arrangements: the Wenner array, the BOSS System*, and the Schlumberger array. The Wenner array uses four equally spaced electrodes placed along a line at the surface of the ground. The two outer electrodes inject the current and the two inner electrodes are. used for measuring potential differences. The BOSS System is explained more fully in the equipment section. The Schlumberger array consists of four electrodes placed along a line at ground surface, with the two outer electrodes for injecting current and the two inner electrodes for measuring potentials. The distance between the two inner electrodes is typically one-fifth or less of the distance between the outer electrodes. The term 'a-spacing' is used to refer to equal distances between electrodes in the Wenner array or between the five electrodes used for a single measurement series for the BOSS System. The a-spacing is part of the electrode arrangement factor used in calculating resistivities for these two arrangements. In the Schlumberger array, the equivalent electrode arrangement parameters are: 2L, the distance between the current electrodes; 21, the distance between the potential electrodes; and x, the distance between the point midway between the potential electrodes and the point midway between the current electrodes. Calculations for deriving apparent resistivities from data for each of the three types of electrode arrangements are given in Appendix I.
Electric sounding is the resistivity technique used to determine how the apparent resistivity varies with depth. With the BOSS System and the Wenner array, the technique involves increasing the a-spacing about a fixed center point on the electrode line.
J
*Trademark of Bison Instruments, Inc., Minneapolis, Mn.
J 4

The increase in the separation of the current electrodes causes more current to travel deeper (and wider) through the ground. The potential electrodes, however, maintain their same relative position with respect to the current electrodes. For isotropic homogeneous materials, if the current is held constant, the potential differences registered at the potential electrodes decreases proportionately as the electrode array is expanded (that is, as the electrode arrangement factor increases) and the resistivity remains constant (at whatever value is characteristic for the material involved). However, in a heterogeneous material (for instance, a layered material), if the current is held constant, the potential decreases are not proportional to the electrode separation, as the separation increases, and changes in the measured resistivity are registered. Sounding data may be interpreted by various graphic techniques or by computer analysis. The soundings performed during the present study employed the cumulative-type Wenner array, the BOSS System, and the Schlumberger array.
In the cumulative-type Wenner sounding, the a-spacing is increased 1n equal steps (Telford et al., 1976). The method should work well where: a) layering is the only significant heterogeneity, b) layers are not too thick, and c) the length of the equal steps is much smaller than the layer thicknesses. The data can then be used to develop a cumulative plot. A point on a cumulative plot 1s the sum of the apparent resistivities up to and including the apparent resistivity for a particular a-spacing plotted versus that particular a-spacing (an arithmetically scaled graph is used). Interpretation is accomplished by drawing straight-line segments that pass through or close to as
5

many points as possible. Intersections of the straight-line segments are then noted and the values on the a-spacing ax1s that correspond to the intersections represent the depths to interfaces (layer boundaries).
BOSS System data and cumulative-type (and conventional) Wenner sounding data may be used to construct log-log plots of apparent resistivity versus a-spacing for comparison to master curves. Information provided with the master curves allows the depths to interfaces and the resistivities of various layers to be estimated. The table of master curves available for the present study allows curve-matching for soundin-gs intercepting up to four layers (Mooney and Wetzel, 1956). BOSS and Wenner data may also be interpreted by computer inversion using a program by Zohdy and Bisdorf (1975). The program, which is valid for soundings intercepting up to 50 layers, calculates theoretical curves and depths to and resistivities of layers from sounding data.
Schlumberger sounding data can also be rendered onto log-log apparent resistivity-versus-L (current electrode separation) plots or subjected to computer analysis. During the course of the present study, no table for Schlumberger data could be located. A computer program for Schlumberger curve-matching exists but has not been entered into the computer available to this facility.
Electric profiling is a resistivity technique used to determine how apparent resistivity varies laterally. The current study used Wenner profiles, although one Schlumberger profile was attempted and abandoned after difficulty arose with obtaining stable potential readings. In a Wenner profile, the a-spacing is held fixed while the
6
J

array is moved about the area of interest, typically along a network of survey lines. In a Schlumberger profile the current electrodes are placed relatively far apart and the potential electrodes, with 21 fixed, are moved along the line between the current electrodes. If necessary, the current electrodes can be repos~tioned (using the same 2L), and the process of moving the current electrodes (maintaining the same 21) can be repeated.
In the present study, the soundings were used to attempt to locate a horizon of interest and the profiles were used to examine its electrical behavior laterally. Resistivity methods are treated in considerable detail in Van Nostrand and Cook (1966) and in Telford et al. (1976).
7

EQUIPMENT
The resistivity apparatus used in this investigation consists of a Bison Model 2390 square-wave low-frequency alternating-current transmitter and receiver outfit with a Model 2365 BOSS System cable and switchbox set for soundings. The outfit includes metal stake electrodes and separate cables for Wenner and other electrode arrays. Short lengths of cable for perfo~ing profiles at a-spacings of 8 meters or less were procured to cut down on unnecessary weight. The data output consists of digital displays of electrical potential in millivolts (mV) and of current in milliamperes (ma). Details for setting up the equipment and making measurements are explained in the operator's manuals (Bison Instruments, 1979, 1983).
In the BOSS System, two multiconductor cables are extended along a line from each side of a center and are connected to the switchbox. Electrodes are placed at the center (0 meters) and at current outlets along each cable at distances of 1/2, 1, 2, 4, 8, 16, 32, 64, 128, and 256 meters from the center. These connections permit measurements to be made at a-spacings of 1/2, 1, 2, 4, 8, 16, 32, 64 and 128 meters. Five electrodes, one a-spacing apart, are used in each measurement series: two current electrodes, two potential electrodes, and one electrode remaining idle (not activated). The five electrode arrays that are used in each measurement series are shown schematically in Figure 2. A full sounding using each array at each a-spacing out to and including 128 meters yields a total of 45 separate measurements.
8

c

p

idle

p

J 1

1

-1 ~ a-spacing

center

c

p

idle

c

r

c
1 A-array
p C-array

c

p

p

c

idle

o,-array

--.--1----,1 idle

c

p

p

c

.------..1------r-~

C

C

idle

P

P

~,....-----,- 1--lr--------r-1-----,t

8-array

C -- current electrode P - potential electrode

Figure 2. Schematic Diagram of the Electrode Arrays Used for BOSS

System Measurements. In a BOSS System measurement series at a par-

J

ticular a-spacing, a resistivity measurement is made using each of the

l

arrays shown above, which overlap by two

Note that the a-spacings.

o1

and

o2

arrays

are

Wenner

arrays

j

9

NATURE OF THE WASTE Two of the ponds investigated during this study treat domestictype sewage with little or no industrial waste content. This type of waste water contains certain substances -- nitrate, phosphate, chloride, carbon dioxide, ammonia, and various metals -- which exist as or can form ions in water. The introduction of sewage or treated sewage into ground water of sufficiently low electrolyte content could be expected to lower the resistivity of the ground water. Warner (1969) has documented several cases in which resistivity surveys revealed zones of low resistivity associated with discharge from septic tanks
and cesspools. c. Boswell (personal communication) of the Water
Protection Branch of the Georgia Environmental Protection Division has commented that nitrate is generally the only ionic or ion-forming contaminant subject to being leached out of sewage ponds. Other ions or ion-forming contaminants are either decomposed, precipitated, or sorbed.
The textile mill operations include the cleaning and dyeing of wool, synthetic, and wool/synthetic blend fabrics. Two waste-water streams, a high-level one and a low-level one, are produced. The highlevel stream contai-ns domestic-type sewage and waste water from the dyeing and cleaning operations and is subjected to treatment. Certain contaminants in the treatment effluent are subject to limits: sulfide, chromium, phenolic compounds, and suspended solids. Because of low concentration or lack of electrolytic behavior, none of these substances is expected to be capable of causing any detectable change 1n the electrical properties of local ground water. Chemical oxygen demand, biological oxygen demand, and pH are also subject to limits (pH
10

limited to range of 6-9). A large difference in pH between the ground water and any leachate might give rise to a detectable change in resistivity. Among the substances in the process wastes not subject to limits are sodium and sulfate, which are present incident to wool cleaning and are both ionic in solution. Sodium and sulfate, along with substances that might cause extremes of pH, however, are subject to decomposition, sorption, or precipitation.
The low-level waste stream consists of boiler blowdown, spent dilute saline solution from the recharge of water softeners, storm runoff, and discharge from several small creeks. This stream is channelled into the diversion canal (see Fig. 13) and discharged to the river without treatment. Electrolytes are present in this stream, although at very low concentrations.

_j

11

1

SCHOOL SEWER POND SITE, WILKINSON COUNTY, GEORGIA
SITE DESCRIPTION Figure 3 shows a detail map of the impoundment area. In this
area, according to the State Geologic Map (Pickering et al., 1976), undifferentiated Cretaceous and Tertiary sediments underlie the valley floors, the Twiggs Clay underlies valley walls; and the Irwinton Sand caps the ridges. A private well sunk in a nearby town encountered interbedded sands and clays (mostly kaolinitic) at the same horizon as the impoundment. Herrick (1961) ascribed these sands and clays to the Tuscaloosa Formation, but, they probably belong to the Huber Formation (Buie, 1978). Several hundred meters east of the pond, several investigators observed lumps of hard, hackly kaolin exposed by grading and ditching along a road. Such kaolin is typical of the Huber Formation. Records of test borings incident to the construction of the pond and lithologic logs of nearby municipal wells are not known to exist.
Ground-water flow in the impoundment area is estimated to have a major component toward the Town Branch of Bear Camp Creek. The water table in the valley of the branch is shallow particularly on the west side of the impoundment as is evidenced by crawfish burrows and by standing water in holes left by rotted tree stumps.
North of Town Branch, the topography slopes gently toward the branch with grades typically about 3%. Maximum relief is about 20 feet. Short steep slopes exist along the road grade. South of the flood plain of the branch, the topography slopes more steeply, with grades up to 13%.
12
J

A
-- - -
N
c ~
~'i;
~ ~
&~ ~f
e
~
~
-.;::

.. center, sounding A ..,

WOODED

pond
o utlet

profile 1

.... ""
A
E

road
E

~
, 0
~ ~

.". '
8 0

0

50m

...J

lmm & 2m 1: 2000

power line
X X )(
chain-link fence
====
14twer line

""o,.-+.,,.
traverse line: crou ticks data points on profile lineL Pacings In meters Indicated next to each tick round ticka ends of aounding lines lozenges sounding centera

Figure 3. Detail Map of the Wilkinson County School Pond Site
1
13

j

1

Except for cleared areas along the road in front of the pond and within the fence around the pond, the vegetative cover consists of

1

fairly clean woods. The undergrowth had been burned out in the area

west of the pond. Although at the time of the survey a dro~ght had

been 1n progress for about a month, deciduous plants were in the pro-

cess of sprouting leaves. The rootward stumps of freshly severed

muscadines (Vitis rotundifolia), a type of deciduous vine, of approxi-

mately one centimeter in diameter were delivering streams of fluid at a

steady drip, an unusually high rate. Deciduous plants at this phase of

their annual cycle may pose a source of interference for resistivity

measurements.

1

Most of the culture is concentrated in the northern part of the

study area. A gravel road runs along the north side of the impoundment

site. A power line serving floodlights inside the pond compound paral-

leLs the road part way. A galvanized steel culvert crosses under the

road northeast of the pond. A terra cotta sewer line (with one iron

manhole cover) which feeds the pond is buried beneath the road. A

chain-link fence surrounds the pond. Except for a cleared land-survey

line with iron pins at several locations, the south side of the area is

cu'lture-free. Several residences SO to 100 meters northeast of the

pond are not connected to any sewer line and presumably dispose of do-

mestic wastes with septic tanks. A patch of dampness extended from the

culvert up the road ditch to a point behind the residences. A public

water supply well is located approximately 200 meters east of the

impoundment.

14
_j

The pond treats domestic-type sewage from an elementary/junior high school complex. The school is permitted to discharge 21,000 gallons (88,000 liters) of treated waste water per day into Town Branch of Bearcamp Creek. Little waste is treated during the summer when school is not in session. The pond, built in 1967, is of the facultative type and of earth construction, with a depth of approximately 6 feet (1.8 meters) and an area of approximately 4 acres (1.6 ha.). Some of the earth used in construction originated off-site.
MEASUREMENT Five BOSS System soundings (soundings A, B, D, E, and F) and a
cumulative-type Wenner sounding (sounding C) were conducted at this site. The Wenner sounding used the same line and center as BOSS sounding B. Data from the Wenner sounding were used to construct a cumulative plot at the site (see Fig. 6). Log-log plots were constructed for all soundings and were compared with master curves from the table of Mooney and Wetzel (1956) (See Figs. 4, 5, 7, 8, and 9). Data from soundings B, C, D, and F were additionally subjected to computer analysis using the program of Zohdy and Bisdorf (1975).
The log-log plot for BOSS sounding A could not be matched with any of the curves in the master curve table, possibly because the sewer line interfered with the measurements. BOSS sounding E was matched with a master curve in the table, although the match may not be strictly valid because the sounding crossed a metal culvert (see Fig. 6). The depths to and resistivities of various layers predicted by each of the above-mentioned data treatments are summarized in Figure 10.
15

- Sch of s Pond, ~lkin" n Coun~ GA
80 ~ Sc undl ~ 8, I glot pot 2H ~ a4 08/8

iQ ..u...a '-' vv / i;; ,{a v ._,
I

,......1...3,.9J

1\
\ :\.~0 ..,
\ "'
1\
~~
'

40~

~

.I..
i

Ic:
.i

!
sf"

t
;;
i "

c
.i
l

~
~c~0 [I

. .........
-)(
e ~ ':'!
-.I - ... ~
! ci : ':': .i a;; .I ~~
~ .... ~
,.. -3
.. .:
. - ~l5 ..!: i"
.s ! .5

-.. -e e e
o~.-
.. .. .. .. . - N 1'1 ...
. . . -e:~-=
.5 .i .5
~~~d
:!:'
_ ...... "'
~ ,.. >- ,..
JIJIJIJI

a e
..I
.,.!
.c:
l
- ....
:;:
!
.5
.,..a.!.. )
N

-i/,.
!!
,:.
:ii

l;t,.415

IPIN tnt '"ittivity in ohmmlltrs

'-'

2000

1000 800

800 500 400

300

200

100

- Figure 4. Log-Log Plot of Apparent Resistivity Data for Sounding B,
School Sewer Pond Site. The heavy line traces the matching master curve.

J

16

,
j
1
1

Schoo St ""' and, ~ilk in ~n Cou ty, GA Wonn r S undl g C, IOOIog plat
P.1 2Hr., 8/86

I

J

..

I

l


'1:
.s

-

!i!

! i
E

a.
-3 )

8

c

i!

'i

J

1

j
~- . , ....... "':p

.-;
,;
:i
j

IPP ronr nrirtfvity In ohm-mllt11 >II

2000

1000 BOO

600 500 400

300

200

100

Figure 5. Log-Log Plot of Apparent.Resistivity Data for Sounding C, School Sewer Pond Site. The heavy line traces the matching master curve.

17

.. 1600

~

Oi

.e.E
.t::.

. E

~

:~

14000

..~..
.~..

c

12000

.~..~....
:;
E
.:I

School Stwer Pond, Wilkinson County, GA Wennw Sounding C, cumul1tive plat 2Hz., 04108/86

10000

8000

6000

4000

2000

0

5

10

15

20

25

Figure 6. Cumulative Plot of Apparent Resistivity Data for Sounding C,
School Sewer Pond Site. The numbers on the a-spacing scale corresponding to the circled points are depths to interfaces.

18

rw" SchpolS ~nd, ~ilkimo County GA
BO S S< ~ ndin 0, 1< lOt pi I

!
I

2H . 04 09/8

5
.r

I...

_a 321 .....

' 1\~

\

I

1:1;
:1;

" .I
I i

.iuae
a

i11 .5 8.

2 1(0)

~

1\ -,.......... ,....

v

v ....... ~ .....2'~ I I! E

a~..;

~t .~..

.. FJ

X...
E

...

- ......
~ gg ill
..5 .!i .5

~]61

~
I

~

~

i
i

. g e

.ta

:;
- j

.~
!!"'

~
i
.5

~~~..~..~..
N,..._N,.._
=- N ,_, ...
.;:I ;:a;.j_

-
..~..
'i
..-5.
a
-3 )

~\ -,.

~

..~;;

t ~~;i

IIIP ~nt r11i1ti'lity in ohmmotm

2000

1000 800

600 soo 400

300"""'

200

>I:
100

Figure 7. Log-Log Plot of Apparent Resistivity Data for Sounding D, School Sewer Pond Site. The heavy line traces the matching master curve.

19

Sch ol S ,..., ond, lvilkino County GA 80 S So ~nd i~ E, lc -lot pi I 2H , 04 09/8
I""'
;'i
X"'
\

... 8 ...
!
c:
.9
"a" .
-1 )

...
If

I.

~
~

1r
1/

...,f

~L-------------~--L-~~---L---L----L---~!I~307 PPront resistivity in ohmmotoa '~-

2000

1000 800

600 '500 400

300

200

100

Figure 8. Log-Log Plot of Apparent Resistivity Data for Sounding E, School Sewer Pond Site. The heavy-line traces the matching master curve.

20
.J

Sci onl $ w~r 'ond, ~r l,konl n Coun1 GA 80 S S u11du a F, I 9 log p ot 2H , 04 1018

~ ~ :1 ~
..5..
0:
ii
i
N M

I

- N ~.
: !; ~

.!!.!!.!!.!!

. -- --- -----i

wo ront rn~tivity in ohmmtltrs

--L-----'----L~--L__L__l ___~--~---~------ -----------~ )I:

2000

11100 100

100 500 oo

300

200

100

Figure 9. Log-Log Plot of Apparent R~sistivity Data for Sounding F, School Sewer Pond Site. The heavy line traces the matching master curve.

21

COMPUTER TREATMENT

CUMULA nvE PLOT

CURVE TABLE MATCH

l

~--~-om Q~m

Om Um

t----+--1.83m---------------- t---+--1.6m

t----1---- 3.2m

~ 102U1m
~

CQ

I

~
~
~

14.12m

I
46.40m

~--r---Om
1----+---2.0 1m

~

~
~

113ID

c..J
~
5 ~
CJ)

738

12.22m 15.99m

..
20.0m
'-J-

Figure 10, (See next page.)

J

22

CCWPUTER TREATMENT

224

581

337

154

~

~

56

Q

(!)

~

~

~

243

m m
5.08 n-

~-+--11.49m

CUMULAnVE PLOT

CURVE TABLE MATCH

250nm-

Om

750!lm-

O.Sm

1.7m

- 25nm-

3.4m

75Unm-

V\_

Om

~

0.99m

~

1.94m

1.1..

3.16m

~

~

6.07m

Om 130Dnm-
1.7m
130nm-

390onm-

4.3m 5.1m

1300!lm-

~

Figure 10. Comparison of Depth and Resistivity Predictions for Soundings B, C, D, and F, School Sewer Pond Site. The predictions are determined by computer analysis (Zohdy and Bisdorf, 1974), by referral to a table of master curves (Mooney and Wetzel, 1956), and by cumulative plots. Lines join interface predictions that corroborate one another.
J
23

j

Resistivity predictions made us~ng the computer and master curve

I

table treatments and depth predictions made using the various data

treatments agree approximately with one another in many instances (see

1

Fig. 10). However, many instances of disagreement exist which arise

from several sources.

J One source is that the master curve table is valid for a maximum

)

of four layers, whereas, the computer can fit curves for up to a max~

mum of 50 layers. Thus ~n cases such as those for soundings D and F,

the computer can assign as many layers and resistivities as is needed

for a close match. Then the combination of layer thicknesses, depths,
J
and resistivities predicted by the computer may be very different from

I

the combination predicted using the table of master curves even though

the master curve and the computer-generated curve (and the field curve)

appear very nearly congruent.

Another source for disagreement is that the computer cannot recog-

nize anything about the quality of the data. In certain cases; the 2A

extrapolation from computing BOSS data can be of low quality and enter-

ing that extrapolation as a data point for the computer treatment can

1

lead to somewhat improper matches and resistivity/depth predictions.

T. Schmi~t of the Georgia Geologic Survey (personal communication) sug-

gests doing one computer treatment using all the extrapolations and

interpolations resulting from a BOSS calculation and then doing another

computer treatment using only the results of the D1/D 2 calculation (see Appendix I).

A problem arose with the computer analysis of the sounding C data

in which data at small a-spacings is notably more erratic than that at

large a-spacings. The lower quality data at the small a-spacings low-

ered the overall quality of the computer-generated curve of depth/

resistivity predictions.

24

.J

The cumulative sounding technique has certain previously mentioned

1

limits concerning the thicknesses of layers and the s1ze of the step-

wise increases 1n a-spacing. Selecting a proper size for the a-spacing

1

steps has been a recurring problem in that the one-meter or two-meter

steps used for this study are too large to resolve thin, usually near-

J

surface layers.

I

Wenner profiles were conducted using a-spacings of 2 meters and 8

meters to test the layers beneath the 2-meter and 5.5-meter interfaces

1

predicted from the cumulative plot for sounding C. The 2-meter profile

proved to be subject to excessive interference, probably from vegeta-

J

tion. A good example of such interference was encountered near the in-

I

tersection of profile lines 5 and 6 (see Fig.9). Near the intersection

are two data points (one on each profile line) about two meters apart.

J

An apparent resistivity of 1139 ohm-meters was registered on one of the

points and an apparent resistivity of 762 ohm-meters was registered on

I

the other. No such difficulties were encountered with the 8-meter pro-

file. Figures 10 and 11 show, respectively, a contour plot of and a
J fence diagram of apparent resistivity data obtained from the 8-meter

J

profile.

I

DISCUSSION

The very high apparent resistivities registered by the 8-meter

profile in the bluffs south of Town Branch are considered to reflect

the greater depth to the water table in that part of the area. The

contour plot and the fence diagram also show a pronounced resistivity

j

low east of the sewer pond. The low apparent resistivities registered

j

in the vicninty of the intersection of profile lines 1 and 2 may, in

.I

25

J

'------.. 290 - - - - - - - - - - - - - - - - - - - - -
I
I
1

0
L

50m

I

1mm 2m

1:2000

-290-
topographic contour contour intvel 10 feet
-9oo-
r"ltivltv contour In ohm-meters contour Interval Irregular

.Q .
traver line:
croa ticks data pOints on profile line, with reaitt lvltlea in ohm-meters Indicated, Pacing i1 8 meters

Other cultural and physiographic features are n Indicated in Figure 10.

Figure 11. Pond Site, Contour Plot of Resistivity Profile Data, School Sewer

26

..--

-

-

AB ~m
t: 8000
'\. 'i..~"
~..<. :)
Profile 1 A
.N......

E'
Profile 4

Vertical ocale ahows apperent reistlvlty logarithmically In ohm-meters
Horizontal acale ahowa dlotenceo arithmetically In metera
...............
............
............
............ ............ D

- - Profile 1 (continued) lOOOD-rr--------r-------r------~------~r-------~

~t. ~ ~,o'~

1561

97

------

N78E

-

Horizontal Scale
1mm. - 2m. 1:2000

0

50m

I

I

I

I

I

l

Figure 12. Panel Diagram of Resistivity Profile Data Around the School Sewer Pond Site. The profiles were run using a Wenner array with an 8-meter a-spacing (frequency 2Hz.). The inset (upper left) shows a plan view. The traverse has been separated at points B and D to provide a better view.

i,
j

part, reflect interference from the culvert; however, the rest of the

l

area east of the pond is apparently culture-free. The apparent

resistivities east of the pond are lower than those west of the pond,

1

where the water table is obviously very shallow, and indicate that the

ground east of the pond is anomalous. A check of Figure 7 shows that

the apparent resistivities measured for sounding D are generally lower

than those measured for soundings B, C, E, and F (see Figs. 4, 5, 8,

and 9), further indicating the anomalous condition of the ground east

I

of the pond. The depth and resistivity predictions from both the com-

puter treatment and the table curve match for data from sounding D in-

dicated a zone of very low resistivity sandwiched between two zones of

much higher resistivity, a sounding pattern expected of a plume of con-

taminated ground water. A thin lens of montmorillonitic clay might

also cause such a sounding pattern but would be out of place in the

Huber Fo~ation. The proximity of the postulated plume to the pond

might indicate that the pond is the source of the plume, however, the

anomaly is situated across the expected direction of ground-water flow

away from the pond. The anomaly does lie in the expected direction of

ground-water flow away from the residences immediately northeast of the

study area which lack sewer hookups. Such a location suggests that

septic-tank discharge may be partly responsible for the anomaly.

28

TEXTILE MILL WASTE-POND COMPLEX, JEFFERSON COUNTY, GEORGIA
SITE DESCRIPTION Figure 12 shows a detail map of the study area. According to the
State Geologic Map (Pickering et al., 1976), the Twiggs Clay underlies the ground at lower elevations and the Irwinton Sand underlies the ground at higher elevations. The Twiggs Clay is typically montmorillonitic (Huddlestun and Hetrick, 1979). The estimated direction of ground-water flow is, overall, toward the Ogeechee River (south, just off the bottom of figure 12) and, locally, toward the small valleys such as the one occupied by the pond complex. The floor of Pond 3 (now abandoned) is level. Natural slopes with up to 15% grades occur in the high ground around the pond complex, although local man-made slopes such as the bank of the diversion canal may be steeper. Maximum relief 1n the area is about 35 feet (10.7 meters). Most of the area is cleared and planted with grass, although a strip of fairly clean woods is present on the west side of the diversion canal and a strip of swampy, brushy woods separates Pond 3 from the Ogeechee River.
Culture occurs throughout the area. A waste treatment plant and a laboratory are located on the hill east of Pond 2. The floor of Pond 3 is underlain by a drainage-pipe network. A power line passes by the northern end of profile line 4. Barb-wire fences mark the property boundaries in the western part of the area. The line that carries treated waste water to the river is located east of Pond 3. Unsurfaced roads, earth-work dikes, and a diversion canal are also present.
The mill prepares, dyes, and finishes wool, blended wool/synthetlc, and synthetic fabrics. The waste-wa~er treatment facilities
29
J

N
tr*"r \ln8: round ticks '" endS ot soundin9 11nes
cross tlcl<S '" data poil\tS of\ proille \\n8, proi\18S u ""'8"n8r ., ., ,.itt\
3 "'et., aspaein9 tor 11nes 2 af\d 3, 8 !1'et8r asPaein9 ior \\n8 4
\oz"9es"' tou"dln9 c8nttlfS
--1.00---
toP grep\"'iC contour. co"0tour int81"1al 10 ieet
\

240- ,.

Figure 13. Detail MaP of the !extile Mill Waste-fond Sit Jefferson

I

\

30

originally consisted of a stabilizing pond (Pond 1), an aeration pond

]

(Pond 2), and a polishing pond (Pond 3). The mill had been permitted

1

to discharge up to 1.5 million gallons (6.3 million liters) per day of

I

treated waste water from the waste-pond complex into the Ogeechee

River. Production capacity at the mill was later expanded and a waste

treatment plant was installed. The authorized discharge was raised to

4.0 million gallons (16.9 million liters) per day. Pond 3 was abandon-

ed and Ponds 1 and 2 were converted, respectively, for storage and

equalization uses. Pond 2 occupies approximately 4 acres (1.6 ha.) and

I

is impounded with concrete-skirted earth dikes. Pond 3 was impounded

with unfaced earth dikes and occupied about 7 acres (2.8 ha.). The

types of waste water have been discussed in the introduction.

MEASUREMENT The electrical sounding techniques employed at this site consisted
of two cumulative-type Wenner soundings (soundings B and D), a BOSS System sounding (sounding A), and a Schlumberger sounding (sounding C). The investigators were unable to obtain a table of master curves for Schlumberger soundings. A computer program exists for treating Schlumberger sounding data but is not yet entered into the computer available to the Geologic Survey. Data from the BOSS and Wenner soundings were subjected to computer treatment using the program of Zohdy and Bisdorf (1975). Curves prepared from these data were also compared with master curves due to Mooney and Wetzel (1956) (see Figs. 13, 14, 16). The data from the Wenner soundings were also rendered onto cumulative plots (see Figs. 15, 17). The array for sounding 0 was expanded in one-meter steps for smaller a-spacings in order to save time. Two sets of sums, one for the one-meter expansion and one for the two-meter expansion,
31

IX ill MIll W 111 Po d, JtH rson Co" rtv. GA BOS ~So ndin A, lo -log pi t
Hz. 03/ 04/8!

c
..i
.5

. 5

~

!t

1

t-o
~._, 31.2

I
/

......

/ .
;:;....(

li ...

!

1

,L ;i.

- ,;
e s

;.e..
5

r ...'8

c:
0

:I

0:

Si l

v~~

~
0 ~

I
!...
~. .._, 73.!
vl7
~_/

..

~
)(
e -,.. ...
.:: ~
-s!a !

~
e , i.!! .! .ac:

= .l:

~ ~

:!

u
i

'! .!! .5

.E.,_E..E,

-_:.,;,...; ......

e e e

-=~.:!

.Ie

I -~

.sI

- -- ~NNe...N~se:i - ..... ,., ....
.I;.I;]:5':I;

...
j
I
.i
, - l!
-5 ,.
...

1-.=.s..

~

= :!!
j ,..!.! .::;

lpp tnt resistivity in ohm-mttm
~

200

100 90 80 70 60 50

40

30

20

10

Figure 14. Log-Log Plot of Apparent Resistivity Data for Sounding A, Textile Mill Waste-Pond Site, The heavy line traces the matching master curve.
32

....

Te ilo ill V. m P nd, Jol 1r1an Co ntv, GA

5 "'

1

We nor aunt n9 B, log-log plat 2H ., 0' 04/B

.i
.5

'.ic..

.N..

~e:fI'
:;;<E 51 .1
J. s
c
rrv..
~ 8! , /
v
I
I

$.

c: .5!
.!s.. 1s"

II;
:s.e.;. ."c.,
8.

<!). 28 (!)~

"...'. ")('

~ ~

-~

':'!
~

!a t! ~

'! ;~

0

~
'Q.
.lJ

:"i
! ~

?:: :~

; :
. 5

....e e e

N

.N..;

-. N ...
. ;j
i j
.i .5 .i

aaa~ "N' N.... N"'
-- -.. - N ... ....
... .:0.. .II.;.
~ ~ .!! .!!

...
e
i
.5
- ..":..
-3 ,..
....
N

IPPI nt millivily in ohm-motors

200

100 90 BO 70 60 50

40

30

20

10

Figure 15. Log-Log Plot of Apparent Resistivity Data for Sounding B, Textile Mill Waste-Pond Site. The heavy line traces the matching master curve.
33

j Textile Mill Wasta Pond, Jlffnon County, GA Wennlf' Sounding 8, cumulati't'l plat 2Hz.. 031 (].418 6
1Cial 1051 1032
so

0

20

30

Figure 16. Cumulative Plot of Apparent Resistivity Data for Sounding B, Textile Mill Waste-Pond Site. The numbers on the a-spacing scale corresponding to the circled points are depths to interfaces.
34

l
I

-,-------------r-r-,--r--r--.----,----,-------.-----------e~ ~

ljTw1n1 ~iltlt

~~illolunWd ~1Q11

P< 0,

~d, Jlf og-lo1

'IOn lol

Co

nly,

GA

~

SH , 03 ~6/B

-~

"ii
!

~~------------+-4--+--~-+---+--~~----~------4-------------~ ~

'.;..
E

.g
z

,. 8
c:
0

.!.,I
0

c.;.. i

:!
~
8 0'"

.,~ 102g

~ j ~~

1000 800

600 500 400

300

. -,... C"''.
~> > >
~ ~ ~

opp en! ristivity in ohmmtlln

~

200

100

Figure 17. Log-Log Plot of Apparen~ Resistivity Data for Sounding D, Textile Mill Waste-Pond Site. The heavy line traces the matching master curve.
35

1

1

11.000

1

I

J 8000
I
7000

sooo

5000

4000

3000

2000

1000

Textile Mill Wut Pond, Jrffenon County, GA Wenner Sounding 0, cumulatiw plot 5Hz.. D3/0&/Sii
11328

a

5

10

15

2D

25

Figure 18. Cumulative Plot of Apparent Resistivity Data for Sounding D, Textile Mill Waste-Pond Site. The numbers on the a-spacing scale corresponding to the circled points are depths to interfaces.

36

cr:NPUTER TREATMENT

CUMULA nVE PLOT

CURVE TABLE MATCH

l

m-;::::=:;:::::::::- Om
O.JOm 1.02m

1.74m

l

2.61m

Om 120!'lm-
l.Jm
4anm-

5.1m

120flm-

1---+---7.60m

7.6m

12flm-
vv

5Gnm

---.....--om

...
y

m

.,_-+--0.66m
- 1.21m 2

51lf2m

~
~
i
CQ

----- 1---~-7.70m

8 .5m

~
Q
~ 2

13.5m

> <,.

t---~-24.72m-----.-+lJ--t--20.0m
zsn.m

Figure 19. (See next page.)
37

---..---om
1---+--- 1.2m

cavi'I.JTCR TREATMENT

CUMULAnVE PLOT

CURVE TABLE MATCH

1

B63!2m 2596.0.m

Om 0.74m

Om
P.----

820.Qrn-

Om O.Bm

2.93m

2640!2m-

981Dm
~
~
~ 306nm
Q
~
~ nnm
!J)

5.35m 8.73m
~
13.70m

213nm-

41m S.Om

5.5m-

82nm-

-'\...
12.0m

LrV

Figure 19. Comparison of Depth and Resistivity Predictions for Soundings A, B, and D, Textile Mill Waste-Pond Site. The predictions are determined by computer analysis (Zohdy and Bisdorf, 1975), by referral to master curve tables (Mooney and Wetzel, 1956), and by cumulative plots. Lines join interface predictions that corroborate one another.
38

were computed for the cumulative plot.

I

The resistivities, layer thicknesses, and depths to interfaces

predicted by the various data treatments are shown in Figure 19. As

was the case with the school sewer pond site, depth and resistivity

predictions made by the various data treatments agree approximately and

are so indicated in Figure 19. Also, as with the case of the school

sewer pond site, instances of disagreement among the predictions arose.

The disagreements resulted from the same sources. The cumulative plots

for soundings B and D failed to predict or failed to place accurately

several thin near-surface interfaces (see Fig. 19). For soundings A

and B, the data points and points on the computer-generated curve di-

verge from the four layer master curve (from the curve table) at large

a-spacings, because of the inability of the master curve to accommodate

more than four layers. Interestingly, although the data from sounding

D are of lower quality at small a-spacings, points on the computer-

generated curve do not depart appreciably from either the data points

or from the four-layer master curve (this situation contrasts with that

of sounding Cat the school pond site) (See Figs. 5 and 17). The cause

of the uncorroborated interface prediction at 13.5 meters for the cumu-

lative plot for sounding B remains uncertain.

Wenner profiles were run at a 3-meter a-spacing along two lines

(profiles 2 and 3) on the floor of pond 3. Except for the east end of

line 3, the apparent resistivities were low (typically 40 to 60 ohm-

meters) and the horizontal profile nearly flat. A Schlumberger profile

line was attempted on the floor of pond 3 but was abandoned because the

potential readings were erratic (profile line 1). The current elec-

trode line extended nearly the width of pond 3 and perhaps the trans-

mitter lacked the power to transmit through such a distance. Another

39

Wenner profile line (line 4) with an 8-meter a-spacing was run west of

1

the diversion canal. Low apparent resistivities were registered along

this line near a small creek that flows into the diversion canal (see

Appendix II). A contour plot of apparent resistivities could not be

constructed because the apparent resistivity pattern on the floor of

Pond 3 was flat and because the a-spacing used on the floor of the pond

was different from that used west of the canal.

DISCUSSION The apparent resistivities registered for the floor of Pond 3 are
much lower than those registered for the surrounding high ground. A review of the resistivities predicted for various layers by soundings shows that a zone of low resistivity commences about one meter beneath the floor of the pond and continues to considerable depth (see Fig.l9). Although a plume of leachate possibly could account in part for the low predicted resistivities and for the low apparent resistivities registered by both soundings and profiles, an identifiable plume should give a sounding pattern of a zone of low resistivity sandwiched between zones of high resistivity. Such a pattern is not evident for any of the soundings taken on the floor of Pond 3, and, the persistence of the low resistivity zone past a depth of 25 meters seems uncharacteristic of a plume. The presence of a thick section of Twiggs Clay, which underlies valley bottoms in the area, would better explain the vertical persistence of the low resistivity zone. The Twiggs Clay is characteristically montmorillonitic, and, the high exchangeable-ion content of montmorillonites would account for the low resistivity. If a plume existed at this site, it might well be impossible to identify. Moreover, if the impoundment at this site were to be situated directly in
40

the Twiggs Clay, the clay would probably act as an effective natural liner and prevent the formation of plumes.
According to the computerized prediction, sounding D intercepted a low resistivity zone at a depth of about 8 meters. This distance is approximately the height of the sounding center above the floor of Pond 3 and possibly represents the depth to the top of the Twiggs Clay in that part of the study area.
41

1
MUNICIPAL SEWER POND SITE, DOOLY COUNTY, GEORGIA
I

SITE DESCRIPTION

1

Figure 20 shows a detail map of the impoundment site. According

to the State Geologic Map (Pickering et al., 1976), residuum from the

Ocala Limestone underlies the high ground and clastic sediments of the

undifferentiated Claiborne Group underlie the valleys. A road cut near

l

Turkey Creek, a few kilometers northeast of the study area, exposed medium sands and clayey medium sands, in places gravelly. P. Huddlestun

of the Georgia Geologic Survey (personal communication) notes that

Claiborne sediments of northwest Dooly County are uniform and very

fine-grained and resemble loess. The Claiborne sediments of this area

outcrop only rarely and contain no gravel. The sediments exposed 1n

the road cut are stream terrace deposits. The Ocala residuum is well

exposed in a series of railroad cuts northwest of the study area. The

residuum appears as masses of cherty shell hash and concretionary chert

set in an earthy brownish-red matrix. Pieces of similar-appearing

chert occur as float at the impoundment site.

Local ground-water flow is estimated to have major components

bearing toward Turkey Creek and toward the small stream immediately

southwest of the impoundment.

Topography varies across the study area. The flat swampy terrain

of the Turkey Creek flood plain occupies the southern and eastern por-

tions of the study area. The bluff sides forming the northwestern

boundary of the Turkey Creek valley occupy the central and western

parts of the area. Natural slopes range up to 10% in grade and man-

made slopes such as the cut accommodatin~ the northwest corner of the

42
.J

.i
1
~
l

I

~
:~...

'

:~: (~hool y~
sounding~ C and 0

I I I
I
!~/
"~ OQ~Q
~0

/

I

I

I

=s
~

I ;..;8.:_:

(I)

I cQ

I ~ ::....

po~

I

~ ~

I (I)

I

N SWAMPY WOODS

0
1mm 2m , :2000

50m

A

I'

I 0

tr.var line:

cron ticks date pelntl on profile llne1, specing 8 meten

round tick ends of toundlng llne1

lozengH 10u ndlng centers

-zoo-
topographic contour contour lntervel 10 fftt
---1000-rnlltlvlty contour In ohm-meterL p.c:lng 8m, contour interval lrreguler

)( )( )( chain-link fence
cover boundery
city limit

Figure 20. Detail Map of the Municipal Sewer Pond Site, Dooly County, Georgia.
43

impoundment may be steeper. Level to very gently sloping bluff tops occupy the northern portion of the study area in the vicinity of the school yard. The total relief is about 25 feet (7.6 meters),
The area south and east of the impoundment is covered with fairly clean woods, although a strip extending five to ten meters away from the fence on the north, west, and south sides of the pond is kept free of trees and brush. Except for a narrow strip of woods separating the impoundment yard from a school yard, the north and west sides of the impoundment area are mostly open.
Except for the chain-link fence surrounding the pond, the culture is situated in the western part of the area. In the northwestern part of the area, a chain-link fence separates a kindergarten ground from the remainder of the school yard. The sewage inlet line enters the pond in the western part of the area. A pig yard adjoins the impoundment area on the west.
The pig yard contained more than a dozen animals 1n a fenced dirtfloored enclosure with dilapidated rat-infested sheds for shelter. The animals apparently had no access to range, and no other sort of sanitary provisions were obvious. Apparently, the animal wastes were slmply allowed to accumulate on the ground. Out of regard for the nearby school and municipality, the County had declared the pig yard to be a health hazard and was in the process of bulldozing it when the survey began. The survey party did not attempt to enter the pig yard area to take measurements.
The pond, built in 1968, is of earth construction with a concrete skirt covering the interior face of the dikes and with a concrete bar part way across the middle. The pond L8 facultative, with an area of
44

j

approximately 5 acres (2 ha.), and treats domestic wastes for a small

1

municipality. The authorized discharge of treated waste water 1s

40,000 gallons (169,000 liters) per day into Turkey Creek.
1

1

MEASUREMENT

Four BOSS System soundings and one cumulative-type Wenner sounding

were attempted. An extremely resistive layer present beneath the

ground at the higher (topographic) elevations gave rise to electrical

potential differences that were too large for the instrument to meas-

J

ure, particularly at small a-spacings. Thus only one BOSS System

sounding (sounding E) and the Wenner sounding (sounding D) generated

sufficient data for plotting acceptable field curves.

The log-log plot of the data from the Wenner sounding predicts a

interface at 8.6 meters and a surface layer with a resistivity of 8200

ohm-meters (see Fig. 21). The field curve had to be compared with a

two-layered master curve because the lack of data at smaller a-spacings

would not permit more layers to be identified. The cumulative plot of

the Wenner sounding data predicts interfaces at about 5 meters, 10

meters, and 15 meters beneath the surface (see Fig. 22). The interface

identified at 5 meters is probably spurious because too few data points

exist at small a-spacings to permit valid straight line segments to be

constructed on the plot. Data from soundings A, B, and C are g1ven in

Appendix II. The field curve for sounding E was matched with a four-

layer curve (curve family X234; resistivity ratio 1:10:3:1; interface

depth ratio 1;2;6; from Mooney and Wetzel, 1956) (see Fig. 21). Inter-

faces were predicted at depths of 1.25 meters, 2.5 meters, and 7.5 me-

ters, with the following resistivities:~layer 1-91.5 ohm-meters, layer

2-915 ohm-meters, layer 3-275 ohm-meters, layer 4-91.5 ohm- meters.

45

:

1

,..un cioo Sow Pond Oooly County, A ~In 11 S und l g 0, I g4og p 01

i
i

_!;

Hz 03/ 8/86

..~f

......

~---------------r-+--~-r--4---4----4------~~~---+--------------~

-r--------~r+-+-+-4--+-~~--+-----~--------~-

10000

5000 4000

3000

2000

1000

Figure 21. Log-Log Plot of Apparent Resistivity Data for Sounding D, Municipal Sewer Pond Site. The heavy line traces the matching master
curve.

46

1

MuniciiJII Sewer Pond, Oooly County, GA

1

Wenner Sounding 0, cumulative plot

2Hz., 03/18186
1
I

= 28!180
~ =
1

28568

e~
=Cl

.s 239153

l

."0
~ c: ~

;)o.
~

6000 4000 2000

a

20

25

Figure 22. Cumulative Plot of Apparent Resistivity Data for Sounding D, Municipal Sewer Pond Site. The numbers on the a-spacing scale corresponding to the circled points are depths to interfaces.
47

: -r--------~--------------r-;--r-.~-r---r----.-----,--------M.~

Municipal Sewer Pond, Oool Co nty, GA

~

B.OSS Sounding E, lo.91n1hmo pi 1

.j:

2Hz., 03119/86

.5

I

!

-t--------t------;-}-'E~----~+-~~---+---+----+-----~------~ ~

-t-------Itt-).~."~------------~+-~-4~-+---+----+-----~------~ ~

,.,

~

.:: !

,...??1

, -!
1:

)

\.~
\
~,

300

200

..

~ \.i ,;. :i ~ .e 100 "ilo ao 10

........

).(..

. ! e
~
~

",:.';
,;.; ~

='1:
I...
!!
'5
~

!
I!

1: ~
~

? :i

~

-~

'i
-~

...e
'"1

E
"'

..E..

-... ....

-:: :: ::

i ~ II

~ ;

.i .i .5

..

c!!
"1
- . "'

c!! ;";'
...

~
."....'.. ...

c!!
"1
"'

. ,..
!!

,..
!!

!=,!..

,..
!!

-

IPP rani r11i11i1 ly in ohmmun >~:~

so 50 40

30

20

Figure 23. Log-Log Plot of Apparent Resistivity Data for Sounding E, Municipal Sewer Pond Site. The h~avy line traces the matching master curve.

48

Seven profile lines (two were interrupted and offset) were run

1

using a Wenner array with an a-spacing of 8 meters (see Fig. 20). The

8 meter a-spacing was selected because it was expected to test the 10

meter interface identified by the Wenner cumulative plot and because it

was the smallest or next smallest a-spacing associated with recordable

data on BOSS System soundings A, B, and C. Six of the profile lines

were run with the array collinear with the direction of advance. For

profile line 6, however, the array was set up parallel to the chain-

link fence north of the pond and was advanced broadside away from the

fence. Readings were taken at distances of 0.5 meter, 1 meter, 2

meters, 4 meters, and 8 meters from the fence. The purpose of the line

was to determine if the fence interfered with measurements (see Appen-

dix II).

DISCUSSION The 10 meter interface identified by the cumulative Wenner plot
and the 8.6 meter interface identified by matching the log-log plot of sounding D are considered to be equivalent. The interface is roughly consistent with the expected contact between the Ocala residuum and the Clairborne sediments. The Ocala residuum with its high content of chert 1s likely identical with the 8200 ohm-meter upper layer encountered by sounding D and is the likely cause of the very high resistivities registered by the profile lines and the other soundings in the northern part of the area. Employing a Lee electrode array may be necessary to cope with highly resistive ground.
The generally lower resistivities encountered by sounding E and by profile lines in the southern part of the area reflect the flooded and waterlogged condition of the ground in the flood plain and the absence of Ocala residuum.
49

Experience on earlier surveys indicated that chain-link fences had

I

little effect on resistivity measurements. The experience with profile

line 6 showed that such fences could provide a source of interference,

although the fence in this case did not interfere seriously beyond

about 4 meters. The distance through serious interference extends

probably depends heavily on the condition of the ground. The weather

had been rainy for several days before this line was run.

The profile portion of the survey revealed an area of very low

apparent resistivity west and southwest of the pond (see Fig. 20). The

apparent resistivities registered here were much lower than those reg-

istered in the water-logged and partly submerged Turkey Creek flood

plain. These differences 1n resistivity indicate that the area west

and southwest of the pond 1s anomalous. The area is downhill from the

p1g yard and is located in the expected direction of ground-water flow

away from both the pig yard and the pond. Although a plume emanating

from the pond could contribute to the anomaly, electrolytes leached

from or washed off from the pig yard probably contribute heavily to the

anomalously low resistivity in this area.

J

50

j
j

CONCLUSIONS

1

The surveys conducted in the course of the present study found low

l

resistivity anomalies at three waste-pond sites. In the case of the school sewer pond site, the anomaly is believed due to septic tank dis-

charge. At the municipal sewer pond site, the anomaly is probably due

substantially to seepage and run-off from a nearby pig yard. In the

case of the textile mill site, the anomaly coincides with an abandoned

waste pond occupying a valley underlain by a montmorillonitic clay unit

(the Twiggs Clay).

The surveys at these three sites employed both BOSS System and

cumulative-type Wenner soundings although a Schlumberger sounding was

attempted at the textile mill waste-pond site. Data from both the BOSS

and the cumulative-type Wenner soundings were rendered onto log-log

plots for comparison with a table of master curves. Once a matching or

near-matching curve was located, information as to depths to and resis-

tivities of various layers was developed. The curve table used could

treat soundings intercepting up to four layers. Data from Wenner and

selected BOSS soundings were also subjected to computer analysis with a

program that could treat soundings intercepting up to SO layers. Data

from the Wenner soundings were further rendered onto cumulative resis-

tivity plots, which give information as to depths of various layers.

Schlumberger sounding data may be referred to tables of master curves

or may be analyzed with a computer. No table, however, was available,

nor, was the appropriate computer program entered into the computer

available to the Georgia Geologic Survey.

The cumulative resistivity plots were found to be insensitive to thin (usually near-surface) layers unles~ the stepwise increases in

51

J

a-spac1ng were kept (in some cases, unrealistically) small. Data

1

treatment by matching log-log plots of data with master curves is

limited by the number of layers for which the master curves are calcu-

lated and plotted. The present study used a master curve table due to

I

Mooney and Wetzel (1956), which 1s valid for soundings intercepting up

to four layers. The use of either cumulative plots or computer analy-

S1S can overcome the limitation with the table of master curves as to

the number of layers. If ground conditions are appropriate, the cumu-

lative plot can detect an indefinite number of layers. The computer

1

analysis carried out for the present study used a program (Zohdy and

Bisdorf, 1975) which is valid for a great deal more than four layers

I

(up to SO layers). Erratic data can compromise the validity of both a

computer analysis and a curve match using the master curve table. Com-

puter analysis is especially vulnerable because the machine cannot dis-

tinguish between lower quality data and higher quality data which may

arise during the same sounding. With the tabular curve match, the in-

vestigator can, in many cases, recognize low quality data and ignore

it. The cumulative-type Wenner sounding is time consuming and tedious

to conduct and will not give resistivity predictions. However, the

data 'and cumulative plot can be worked up promptly and with a minimum

of equipment. The computer and master curve table treatments require

access to curve tables, which may not be readily available, or to a

computer terminal, which is not field-portable.

Schlumberger soundings can be evaluated with either a curve table

or with computer analysis. The limitations discussed above for compu-

ter and curve table treatment for Wenner and BOSS soundings also apply

to Schlumberger data. Although non-symmetric Schlumberger arrays are

52

J

conventionally used in profiles rather than in soundings, the calcula-

I

tions needed for reducing raw data from a non-symmetrical Schlumberger

array are cumbersome and require an electronic calculator to accomplish

J

1n a timely fashion.

1

Despite the problems, the sounding data treatments are valuable

aids for interpreting resistivity surveys. In the course of the pres-

J

ent study, the treatments aided in identifying unusual ground condi-

tions (the presence of a thick montmorillonitic clay layer) at the tex-

)

tile mill site and in the preliminary identification and location of

plumes of electrolytes 1n ground water at the other two sites. If data
J could be telephoned in to a terminal operator and depth/resistivity

J

predictions formulated on twice-daily or more frequent cycles, the com-

puter data treatment would be by far the most useful of the treatment

I

methods.

I

Profiles attempted during the study employed both Wenner and

Schlumberger arrays. Perhaps because the instrumentation lacked the

J

necessary power to transmit through large distances, the single Schlum-

berger profile attempted during the study was beset with erratic read-

J

ings and was abandoned. Wenner profiles were employed at all three

)

sites and aided in revealing anomalous ground and indicating its

extent.

The experience at the school sewer pond site suggests that Wenner

profiles using small a-spacings are subject to excessive interference

from vegetation. Profiles using small a-spacings are probably best

limited to bare or grass-covered areas such as the floor of Pond 3 at

the textile mill site. Deciduous trees and vines are thought to pose

the major source of interference due to ~egetation. The interference
J
is probably at a max1mum during the spr1ng of the year and could be

53

j

lessened by conducting surveys during the fall and winter, when these

l

plants are dormant.

Chain-link fences are a common feature at waste-pond sites. Al-

1

though experience with previous surveys tended to show that such fences

1

have little effect on resistivity measurements, the experience at the

municipal sewer pond site indicated that, under certain circumstances,

I

such fences can pose a source of interference. It is probably advis-

able to keep a survey line at least one a-spacing away from a chain-

1

link fence or to run a few profile stations at successively greater

distances away from the fence to determine if it presents a problem.

54 I
.J

j
ACKNOWLEDGEMENTS
1

The author wishes to thank the owners of the various impoundments
1 for their assistance in allowing access to the impoundment sites. The

}

author also wishes to thank Newton Anderson of the Dixie Kaolin Co. for

allowing survey lines to cross his land behind the school sewer pond.

I

55

REFERENCES

l

Benson, R.C., Glaccum, R.A., and Noel, M,R., 1982, Geophysical

Techniques for Sensing Buried Wastes and Waste Migration, USEPA

contract #68-30-3050, J.J. van Ee, project officer, Advanced

Monitoring Systems Division, USEPA, Las Vegas, NV 89114.

Bison Instruments, Inc., 1979, Bison Instruments Signal Enhancement
Earth Resistivity System Model 2390 Instruction Manual, Bison Instruments, Inc., Minneapolis, MN 55416.

Bison Instruments, Inc., 1983. Bison Instruments Offset Sounding System Model 2365 Instruction Manual, Bison Instruments, Inc., Minneapolis, MN 55416.

Buie, B.F., 1978, The Huber Formation of Eastern Central Georgia, Short Contributions to the Geology of Georgia, Georgia Geologic Survey
Bulletin 93, p. 1-7.

Carver, R.E., 1982, Surface Impoundment Assessment Final Report, report to the State Geologist of Georgia.

Donahue, J. and Meehan, D.L., 1986, Resistivity Studies Around Three Waste-Water Impoundment Sites in the Georgia Coastal Plain, Georgia Geologic Survey Open File Report 86-4, 27p.

Herrick, S.M., 1961, Well Logs of the Coastal Plain of Georgia, Georgia Geologic Survey Bulletin 70, 462p.

Huddlestun, P.F., and Hetrick, J.H., 1979, Stratigraphy of the Barnwell Group of Georgia, Georgia Geologic Survey Open File Report 80-1,
89p.

Mooney, H.M., and Wetzel, W.W., 1956, The Potentials About a Point Electrode and Apparent Resistivity Curves for a Two-, Three-, and Four-Layer Earth, University of Minnesota Press, Minneapolis. Mn, 146p., with separately bound master curve table, 243 leaves.

Pickering, S.M. and staff, 1976, Geologic Map of Georgia, Georgia Geologic Survey, Atlanta, Georgia.

Stollar, R.L., and Roux, P., 1975, Earth Resistivity Surveys-- A Method for Defining Ground-Water Contamination, Ground Water, vol.
13, no. 2, p.l45-150.

Telford, W.M., Geldart, L.P., Sheriff, R.E., and Keys, D.A., 1976, Applied Geophysics, Cambridge University Press, Cambridge, London,
New York, New Rochelle, Melbourne, Sydney, 860p.

Van Nostrand, R.G., and Cook, K.L., 1966, Interpretation of Resistivity Data, U.S. Geological Survey Professional Paper 499, 310p.

Warner, D.L., 1969, Preliminary Field Studies Using Earth Resistivity Measurements for Delineating Zones of Contaminated Ground Water, Ground Water, vol. 7, no. 1, p. 9-16.

Zohdy, A.A.R., and Bisdorf, R.J., 1975, Computer Programs for the Forward Calculation and Automatic Inversion of Wenner Sounding
Curves, U.S. Department of Commerce National Technical Information Service Publication PB-247 265, 47p~

56

APPENDIX I: CALCULATIONS
Apparent resistivities were calculated for Wenner arrays using the relation:

R* :a 21ra V/I,

where: R* is the apparent resistivity in ohm-meters,

Vis the potential in millivolts,

I is the current in milliamps,

a is the a-spacing (separation between electrodes)

in meters.

In the BOSS System, the apparent resistivity at an a-spacing "a"

can be calculated by finding the apparent resistivities for arrays

o1 and o2 (Wenner-left and Wenner-right arrays, see Fig. 2) from the
above formula and then averaging them (Bison Instruments, 1983).

Resistances at '3a (three times an a-spacing) can be computed from the

relation:

R3A '"' ~(~ + ~ ) + Ra - Ra

2A

4A

2A

A

where:

R3A is the resistance (not resistivity) at 3a;
~ is a resistance found by computing V/I 2A
for o1 and for o2 at 2a (twice a certain
a-spacing)(V and I are potential and current,

as above) and then averaging the results;

~ is a resistance computed at 4a (four times
4A
a certain a-spacing) in the same manner as F-lLl2A,

57

J

~ is a resistance found by computing V/I

1

2A for array B at 2a;

~ is a resistance found by computing V/I

1

A
for array B at a.

J

Resistances at 2a (twice an a-spacing) can be calculated from the

relation:

I

R2A a 2(RC - ~ )

A A

I

where

RC is a resistance (V/I) calculated for array C
A

at a;

I

~A is a resistance calculated by averaging V/I

1

for arrays o1 and o2 at a.

Either resistance R3A or R2A can be converted to an apparent

I

resistivity by appropriate substitutions into the Wenner resis tivi ty

relation:

I

R* .. 211' (3a) R3A

I

or

I

R* :a 211' (2a) R2A. Apparent resistivities were calculated for Schlumberger arrays

using the relation (Telford et al., 1976):

R* ,. ~ (L2 - x2)2 (y)

2l L2 + x2

I

where: R* is the apparent resistivity, 21 is the distance between the potential electrodes, L is half the distance between the current
. electrodes,
X is the distance between the point midway

58

J

between the current electrodes and the point

1

midway between the potential electrodes, -

I

V is the potential in millivolts,

I is the current in milliamps.

J

If this array is deployed symmetrically (i.e., the current electrode

spread and the potential electrode spread have the same midpoint),

I

then x = 0 and the relation simplifies to:

~ L2 (V) R*

1

U I

I

1
I I I I

59

j
j

1

I

APPENDIX II: ADDITION DATA

SCHOOL, WILKINSON COUNTY, GEORGIA

Profiles (2-meter a-spacing, Wenner array, 2Hz., apparent resistivities

f

in ohm-meters) (see Fig. 3)

Profile 1 (in order, west to east)
1334 1293 off scale 1925 756 605
97

Profile 2 (in order, north to south)
525 565
23

Profile 3 (in order, north to south)
1349 1070 278

Profile 4 (in order, west to east)
1113 445
779

Profile 5 (in order, west to east)
1390 241 1139

Profile 6 (in order, west to east)
762 1172

60

J TEXTILE MILL, JEFFERSON COUNTY, GEORGIA Schlumberger Sounding (Sounding c. 2Hz . apparent resistivities

in ohm-meters) (see Fig. 13)

1

21

2L

apparent

21

2L

apparent

(meters) (meters) resistivity (meters) (meters) resistivity

.I

1

5

106.6

3

15

53.9

6

89.0

18

50.8

I

7

77.7

8

69.8

21

48.0

24

43.9

10

65.3

27

41.6

I 1

2

10

12

14

16

18

20

30

63.0 59.6 56.5 54.1 52.1 50.9

4

20

24

28

32

36

40

38.6
49.9 44.4 40.8 37.8 35.4 32.7

I

I

Profiles (Wenner array, 5Hz., apparent resistivities in ohm-meters)

(see Fig. 13)

Profile 2 (3-meter a-spacing) (in order. east to west)

1)66.9 2)62.2 3)69.7 4)58.6 5)65.6 6)60.5 7)57.5 8)57.3 9)57.7 10)56.9 11)55.4 12)60.3

13)62.9 14)63.4 15)69.1 16)63 . 9 17)59.5 18)60.1 19)47.5 20)50.3 21)42.4 22)40.7 23)39.0 24) 39.6

25) 41.3 26)42.4 27)47.7 28)43.0 29)49.7 30)47.1 31)48.6 32)47.3 33)50.1 34)47.3 35)49.2 36)48.6

37)53.7 38)57.5 39)57.3 40)62.2 41)52.0 42)52.9 43)55.2 44)56.0 45)52.4 46)57.5 47)46.7 48)46.2

49)55.8 50)45.4 51)47.7 52)45.8 53)46.5 54)45.0 55)47.1 56)50.7 57)48.0 58)54.8 59)48.6 60)56.5 61)63.1

61

J

TEXTILE MILL, JEFFERSON COUNTY, GEORGIA

1

Profiles (coqtinued) (Wenner array, 5Hz., apparent resistivities in

J

ohm-meters) (see Fig. 13)

Profile 3 (3-meter a-spacing)

I

(in order, east to west)

1)78.5

11) 62.7

21)57.3

31)54.8

41) 115

2)58.4

12) 65.7

22)49.4

32)52.9

42) 116

I

3) 66.5 4)49.7

13) 61.2 14)67.8

23)49.0 24)48.8

33)48.0 34)63.9

43)132 44)134

5)61.4

15) 65.8

25)50.1

35)50.9

45)155

I

6)45.2 7) 60.1

16)68.6 17) 65.2

26)46.0 27)49.0

36)62.2 37)61.2

46) 180 47)206

8)53.1

18) 65. 6

28)46.9

38) 67.6

48)247

I

9) 66.7 10) 62.9

19) 62.4 20) 56.7

29)50.1 30)51.1

39)75.9 40)87 .2

49)319 50) 658

51)791

I

52)1034

I

Profile 4 (8-meter a-spacing) (in order, north to south)

I

1) 687 2)804

3)837

l

4)924 5)919 6)892

7)879

I

8)839 9)670

10)630 11)443 12)352 13)275 14)185 15) 117 16)152 17) 98.3 18)116

19)72.6 20) 51.2 21)79.6 22)82.9 23)53.6 24) 41.7 25)35.0 26)35.3 27) 65.7

1
l

62
J

MUNICIPALITY, DOOLY COUNTY, GEORGIA

1

BOSS Soundings (2Hz., resistivities in ohm-meters, a-spacings in meters)

J

l

Apparent Resistivities:

t

a-spacing

Sounding A

Sounding B

Sounding c

0.5

OS

OS

OS

1

OS

OS

OS

1.5

*

*

*

I

2

OS

OS

OS

3

*

*

*

4

3279

OS

OS

l

6 8

*
2125

*
4187

*
5615

12

*

*

*

16

712

888

2241

24

*

*

*

32

279

985

420

64

*

*

*

*-interpolated or extrapolated ~OSS resistivity, not calculated os-out of range of the instrument

Profiles (Wenner array, 2Hz., apparent resistivities in ohm-meters)

(see Fig. 20)

Profile 1 (8-meter a-spacing) (in order, west to east)

1)103 2)136 3)144 4)264 5)232 6)295 7)266 8)268

9)250 10)261 11) 221 12)_229 13)252 14)240 15)244 16)233

17)200 18)243 19)186 20)206 21)208 22)202 23)260

63

MUNICIPALITY, DOOLY COUNTY, GEORGIA

1

Profiles (continued) (Wenner array, 2Hz., apparent resistivities in

I

ohm-meters) (see Fig. 20)

Profile 2 (8-meter a-spacing)

(in order, west to east)

1)2592 2)3062 3)2187 4)3165 5)2437 6)3246 7)2899 8)2818 9)2193

10)1786 11) 1261 12)1206 13)1814 14) 1633 15) 1917 16)1389 17) 1356 18)786

19)630 20)545 21)482 22)507 23)460 24)522 25)462

Profile 3 (8-meter a-spacing) (in order, west to east)

1)5275 2)3185 3)3627 4)2422 5)4085 6)3949

7)4097 8)3929 9)4896 10) 3592 11) 3780 12)3215

13)2381 14)2542 15)2130 16)1884 17)1924

Profile 4 (8-meter a-spacing) (in order, west to east)

1) 198 2)285 3)307 4)382 5)335 6)359

7)289 8)331
9)305 10)305 11) 308 12)563

13)332 14)555 15) 311 16)379 17)266

64

MUNICIPALITY, DOOLY COUNTY, GEORGIA

l

Profiles (continued) (Wenner array, 2Hz., apparent resistivities in

ohm-meters) (see Fig. 20)

Profile 5 (8-meter a-spacing) (in order, north to south)

1)666
2) 613
3)691 4)367

5)460 6) 130 7)125 8)87.2

9)90.7 10)76.9 11)86.9 12)82.9

Profile 6 (8-meter a-spacing, broadside traverse, distances in meters)

distance from fence

apparent resistivity

0.5

153

1.0

286

2.0

585

4.0

1522

8.0

1964

Profile 7 (8-meter a-spacing) (in order, north to south)

1)1067 2) 784 3) 781 4)502

5)465 6)244 7)268 8)200

9) 167 10)155

65