Geophysical Prospocting34, t197, 1986.

GEOTHERMAL PROSPECTING BY GEOELECTRIC

SOUNDINGS IN NE GREECE*

C. THANASSOULAS*{., G.-A. TSELENTIS***

and N' KOLIOS**

Assrnlcr
Tnnuessoulls, C., TsEreNrrs,G.-A. and Kortos, N. 1986,Geothermal Prospectingby Geo-
electricSoundingsin NE Greece,GeophysicalProspecting34' 83-97'
In total j'7 direct current resistivity soundingswere carried out during a geothermal
exploration survey of the Genisea,NE Greece,geothermalfield. The data revealeda high
eleitrical conductivity zone at the center of the investigatedarea and suggestedthat an
anomalousheat sourcelay beneaththe study area.
This was confirmed by subsequentdrilling data. Temperaturemeasurements, from 1l
boreholes,were used for the construction of isotherms that correlated very closely with the
geoelectricdata.

INrRooucrtoN
Greece,in its privileged geodynamicsetting of collision area betweenthe African
and Eurasianplates,hasparticularlyattractivegeothermalprospects.
The active distensivetectonicsof the Macedonia region (NE Greece),have
causedrift valley structuresand associatedsedimentarydepositsin the Tertiary
post-orogenicbasinswith thermalanomaliesin most of them.The Tertiary Xanthi-
Komotini a narrow valley,80 km long and 20 km wide, with Lake Vistonisat its
center (figs 1, 2a and 2b). It has been consideredas one of the promising areasin
Greecefor explorationof low enthalpygeothermalcnergyresources.
To get a tetter insight into the subsurfacestructureand thermal conditionsof
the 'KJssani' geothermalarea, located on the southwestmargin of the Xanthi-
Komotini valley,a geophysicalsurveyof dc resistivityand temperaturemeasure-
mentswascarriedout during August-September 1983.

* paperreadat the 47thEAEG meeting,Budapest, Feb-

June1985,last materialreceived
ruary1986.
Street,Athens608,Greece'
** Instituteof GeologyandMineralExploration,70 Messoghion
*** Universityof Athens,Department of Goethermyand Geophysics, Panepistimiopolis,
Greece.
Ilissia,Athens15701,
83
84 C . T H A N A S S O U L A S ,G . - A . T S E L E N T I S A N D N . K O L I O S

NEoGENE
__ffi
PALEOGENE

Fig. 1. Location map of the investigatedarea. Inset maps give (a) generalgeologicalsetting of
the Xanthi-Komotini valley, (b) detailed geology of the investigated area and (c) its tectonic
structure as obtained from seismicreflection studies.

Gporoctc Ssrrluc
The investigatedareais locatedin the westernpart of the TertiaryXanthi-Komotini
basin.The pre-Tertiaryformationsat the northern and southernedgesof the basin
consistof metamorphicrocks,mainly gneiss,shales,and amphibolites.
Someigneousrocks are encounteredin the form of granodiorites(for example,
the Xanthi Granite of Oligoceneage),while volcanic rocks (mainly andesites),are
found towards the north and south parts of the basin as veins and small cylinders
within sedimentsof Paleogene age.

:;!{:ir,ij
 GEOTHERMAL PROSPECTING 85

Fig. 2. (a) Production of hot water at borehole 83 and (b) 84 showing the depositionof salts
on the tube.

The basin is coveredby the following sediments:

1 . Paleogeneformations-limestones,conglomeratesand interbeddedlayersof clay

and sandstone;
2. Neogeneformations-mainly the lake sedimentsof the Kessaniarea and
3. Quaternary sediments-the alluvial sedimentsof the northern part of the basin
consistingof alternatingclays,fine sandsand coarseto gravellysands.
86 C . T H A N A S S O U L A S ,G . . A . T S E L E N T I S A N D N . K O L I O S

The major known faults in the area have an E-W and NE-SW direction.
Seismicreflection(fig. lc) studiesrevealedthat the basin has an aeymmetricstruc-
ture, deepeningtowards the north wherethe sedimentsare thickest.

GeocgEtrtlcAL AND Tnrnunr INvrsrlcertoNs

In the region there are many hot springs, some well known from the past for their
curative properties. Most of these springs are located on alluvial sediments, 2 m
above sea level, in a marsh region of about I km wide, with seasonalvariations of
their dischargeand temperature due to the nearby sea(fig. 3).
The exit points of the springs can be located easily from the small cones of
calcium salts deposited. Within a small area around the main spring, we located
more than 100 exit points of hot water, 18 of which were sampled for chemical
analysis(table 1a).
To study the chemical composition and temperature of the deep waters, we also
sampled some shallow wells used for irrigation purposes (table lb) and some deep
exploratoryboreholes(table 1c).
Due to the special geological conditions of the area (thicknessof sediments,near
sea),we did not use Na-K or Ca-Na-K geothermometers but the SiO, geother-
mometer, that gave an initial water temperature of the order of 100"C.
In the north the geochemical results give a ratio Na/Cl > I while in the
southwards direction this ratio becomes less than one (Na/Cl < 1). This might be
becausepermeablesedimentsdominate the south part of the region and saline water
intrudes from the nearby sea.The boundary between these two regions is between

Fig. 3. Locationofshallowwellq hot springsand exploratoryboreholes.

 :foon@droh
---OA-OOO- -! 1
NN9Q
- - n- $
- -r 6Nr -d6o- o- h
O - -d 6r o c )
oociodct-ictdd J J ,i d i o o o d o --.: o o d : d -j -.,:A-:AAi

U)
H = = E qh E-N NE
.l q ?q q*
--,No-, N.I
= EE
r, qO =
q - Adq6 E^iqcisriEri Ari E
h r - ' - e+ , - r i q i rdi n n 6 n n 6 6 h
h h ry.:r
EqeEqqs
d ri r ^i 6 J
$c-6
$oo
:Nngo-

o 38=o8=3993 3RR9c:3R3,g3e:!993 6C'$O\On

99-Q-:Q
AAAA: :::: : +i A ^l; j ^ ; ^ A ^ a : : : : : -;

xxxx=xx=xY
AAAAAA A::
- - 5 i - 5 6 ; ;j i6;ii ni6; X;;; !oo
+ + + + ci ri -t.i + + -r -i + + + di -; ++.++oo
gRXA=:!lllYlq x==x=---==-xx-x-=x
- - -
! -

ttt -
z --,^:^;^A^::A^::x:::

!f,O\tOrS-rn-x r -
SS-N\O.o-r--l - 6 -!a- - O c tl9O -o a]
O - o6 - dil).1
_.| d lar N c) - h o\ rh\onryc)
r-!hh
ch oooJ-idJd-jd ri.+ ri r; ri ri r; rj r; + + ri r; + rj ri r; r; .++++dd

y?!loQoo<)qo6 gQQOQOOQOOOOOOOOOO 6hc)c)c)h

r, a - -, - o s -, r, vtG:vtv: q a a - i n - h n h 6 _
i 6 6 ci no c> I
:\ q
-; o -j r; + -i 6i o -i -i h 6 q ri rirj r vj ri !d ri;;
sv:f,!f,$+t\fr+*$$:f*+r++$ "{;;;
ij Un
<f$+.+'
n.i.i

4
o '- ?9 dQ O rf, t+ - *
1
I I Q F e
- - n . i - - - -c- i d5 ' \o
c
b +\o; -s +d 6 65VVEi;
\ U - t r+
= tn" 1h* ch c -\ * ri9 nd !f, ri a Y?'i yl r; ri r; ri -i qj 4 4 r; ri 6 r; ri ri d -;;
6rd6lddNdNNNdNNNNdNd + +-
-atNN'
a

383933:3o3
AAAdAA ::
-Yl Y 9 - -?t !-i =9 =q -q- 9- -9:9xyk---=v=v ?9 9 ==Y-'Y^,cg
9urr-O
O ssvvssssvvvsvvvssv ssooocj

qEqEEqqflFfr ==-=x-xY===-==Y-YY
----YY-:YY-e
oohhoh
nodd6
NN-NC)dN
ssssEgsgggss35$ss ++-i-id+

U 3g3g:::ri$ 999 :-t v9 . 1!.. 1 9 v l g

1 O @<> O \o $ Rf O :t r a
. . 1a a . l +o+or\o
oo-j'f,+-.:+cj-j,i n "1 cl R.! c] a..l t$oo
6i6ii'-.:oo

Nde'OOO.l
Q
n iNi F- l -ic<
.r )
- -i " r1 qN - -,
= q 3 3 it 3\ c j ( j .3c j 63 <gt 3
'o.o
g g i8r ,go s rg r
r,aj
:- j a 3 6 R s - j
<t(i i,c i

nq!:O--<f -\Od r-oNO-r(

t cntr ;;3DB
lieY-,?"?iHqC e6{roe.1r:.!rNO*S33giGg
Csq hr l $r !- fqoE- N6 nS6f
s ? F!i839^9SFfl;KS39R
lhl h n1 ah :nl h 6! "h! .h! n ni :h{ .6! n n -ni rni n c i n n N N n N c i l d
8 3nrxs
F { - _ e n
irrqino

. : i

o> |

\
H3 U eEFSF3gHF
run-ni .. ==3=?A===E=E3ee=3A
333333333333333333
EE3=sE
FSRFF-

nEfrE3RRs33 EeEE=AEi=EEEEEEsEE EERRS

rrrrrrrrrr ***srs iretsssrrrrrr d.dil.d;i

pQ hn h \O q rnnn h nh qr
U t Nt N -; q
N

N - N iHi - d
c9s ooi . t t ? . " ?q c 1 - i r 6 - i - t
<hh--S.+--itq.1iiii6ri*n
q qJ o n:: ij *.o + o ^r
iossoo
J
!)
6<; e c r+ n r - o O - -6_= o. J6 - n -o6 -*- - h
* w * , +r + + cl .t V n.O r
>z rN -- + 6+ +e - --
lf
N o S E
Fl
88 C . T H A N A S S O U L A S .G . . 4 . T S E L E N T I S A N D N . K O L I O S

+oc

..+oc

profileat (a)wells,(b)boreholes.
Fig.4. Temperature

the villagesof Potamia and N. Kessani and coincideswith a major fault unit
(fig. 1b).
In addition to chemicalsampling,we measuredtemperaturesat 11 shallow
irrigation wells and l1 deep explorationboreholes.Temperatureswere measured
with a platinum (Pt) probe forming the branch of a Wheatstonetype resistance
bridge.
Locationsof the wellsand boreholesstudiedare shown in fig. 3. Temperature--
depth profilesfor all measuredwells and boreholesare shown in fig. 4a and 4b,
respectively.All profilesexhibit nonlinearbehaviour.Note also that most of them -
extrapolateto about 16"C.
The variation of the geothermalgradientwith depth is shownin fig.5; one can
deducethat the geothermalgradientsencounteredare greaterthan the meanearth
geothermalgradient(0.3"C/10m). The oscillationof the graphsreflectsthe frequent
interchangeof permeable, impermeibleformationsin the area.This sort of stratig-
 GEOTHERMAL PROSPECTING 89

to**T-
ollopo **T-
:f-
opo *-r
o10

I
I
trl
fso t-
I
t-1'
''l
\-1
z
I
lr
tl
t-
lo ls
t
r
lo

hoo-
tl
r
|
ll I
*:r
tr@rrcotoo10 o10010
*q*
o10
to -rL :* /- -
I L
-17
B
-le
10 11
7.,,
t5o
I t|-
m h*t +m
o*-*|,--
'c'o
"l
rg ( i---56-ffio tl I
lO,
1150 : r- 50i I
"^ I, lm
?l

Fig. 5. Measured geothermal gradients.

raphy prohibits upward heat flow becauseof the low thermal conductivities of the
strata encountered and the constraints that impermeable formations impose upon
the developmentof convectivecurrents.

GsopHysrcAL INVESTTcATToNs
The use of electrical resistivity techniquesin geothermal exploration is based on the
fact that the resistivity of water at high temperaturesdecreasessignificantly.
During the present investigation, 77 Schlumberger resistivity soundings were
performed. Current electrodeswere generally separated by up to 1000 m (ABl2 :
500 m), while a small number of soundings were performed with AB up to 2000 m
and one sounding with AB up to 4000 m in an attempt to define the deep limestone
bedrock.
The geoelectricsoundings are grouped into 14 geoelectric traverses (fig. 6, Tl-
T14), of which T1-Tl1 are east-west and T12-T14, north-south. From fig.6 we
judge that the resistivity soundings cover the area under investigation satisfactorily,
with a mean distance between soundings of about 300 m and a separation between
traversesof about 400 m.

Rrsurrs
Spatial changeof apparent resistiuity
The amount of data so recordedcan provide an insight into the three-dimensional
distributionof geoelectric
parametersin the area.Figures7a-7d show the apparent
 r,)
tr
t t
T 1-V7 Ye 7z ?a in in in #t;
F/w w1Lr1
KoUTSor.lT2-V is F6p< i,. pu;q[rz
T3- F i6 F i4rs \
fa* ff91_ iO 3.9 s
?.3_ ia*T4 \

'"*,.,
-itflff*\
rlo-sj
i., _trrerr_*f*r^
' iI
'"
tn-?9- ?7-2-PF_ 'p,Fr*Ttt
'+
d d "' o----i*-
FF

Fig. 6. Location ofthe geoelectricsoundings.

AB/2.50
A

99
\\
".r
)
.0
,t2
I
I

II

A9/2"25O
c
pominrmum,/_1po in .flm, AB in m
O
for differentcurrentelectrodeseparations.
Fig. 7. Spatialvariationof apparentresistivities
 GEOTHERMAL PROSPECTING 9l

E
o
o,

-t'
46810
,f/m
Fig. 8. Salinityagainstapparentresistivity.(Resistivitydata takenfrom fig. 6.)

resistivity contours over the area with ABl2 eqt:al to 50, 100, 250 and 350 m,
respectively.Judging from these diagrams we seethat:
l. for AB/2 equal to 50, 100 and 250 m the apparent resistivity becomes minimal
(p^ < 5 Om) towards the southeast of the area. This pattern probably reflects
saline water entrance into the upper formations facilitated by existing faults;
2. towards the southwest of the area the curves have roughly a NE-SW direction,
suggestingthe existenceof a lateral geoelectric discontinuity probably due to a
fault in the bedrock.

Apparent resistivity values from these diagrams were compared with salinity
values calculated from the geochemical data of the existing springs in the area
(table 1). The results are plotted in fig. 8 from which it can be seen that resistivity
and salinity exhibit an inverse linear correlation, as one would expect.

Vertical uariation of apparent resistiuity

The interpretation of the apparent resistivity curves was performed in two stages.
During the first stage we used 2 and 3 layer model curves; with the auxiliary point
method we found an approximate geoelectric model. Next, we used a computer
program based on Ghosh's method to improve the above model until the RMS
difference between observed and calculated data was less than 3oh. The geoelectric
model thus reflectsa possiblegeoelectricsection.
The interpretation of the resistivity field curves is shown diagrammatically for
traversesT1-T14 (fig. 9). As seen from these diagrams, the following formations can
be identified:
1. Surface formations covering all the investigated area with large lateral variations
in their electrical resistivity and with little thickness. These formations corre-
spond to soft deposits (alluvium and sandy clay) of the Quaternary period. The
electrical resistivity of these formations lies between 11 Qm and 90 Om depend-
ing upon the relative amount of concentrated sand and clay.
2. The next geoelectricformation is almost uniform on all traverses,it was identi-
fied, with electrical resistivity values lying btiiween 1.8 Om and ll Qm, with the
 N
fri.,.;'i'','.
"lf;11i;
a

E
a

o
o
$ffi
fi - "f;g1"9
Ia
T
'lf Fl l : '
It .,
'r': ..:r
ll ,

6 Ji:i l.- "

{_{_{ru
&R

u f--{---}q
x
ai;

o
N
a

o
a

ts
o
a

@
q

E
a

R
o

F
a

N
F

p
@

p
I I
\
ts
ts
 t
N

N
q
I

i
o
D
6

I
o

o
o

3
' K H -j--------r'ud--{------u
*----+ A
H z
q z/gv

@
a

N
a

z
o I'
I

e
@
I
I
N

o t
@
@

o
q

oa
9!
ol u a
@ l-----l------L.u -F-------6-----6-- -o
F--!--*---Au uxx-6 o
"Rg "KHF nibN
 Lriri

I
a.l

E
o

F
bIJ
N a
N

tn
6
0)

v l

-l

@
-
61
E 8 oo
ooc
o
HX
q=
90
o N
q
o["Et
t,'il
E E

o
a iiv

D vl
@ x+

l o
o
o:
g.:
I O;
OE

Nii
x N
e-H*-'---fi1-'
aav z-. z/gv z :bo- ( c))
 GEOTHERMAL PROSPECTING 95

Table 2. Souruling locations

of geothermalinterest.

Traverse Soundine

T7 s34,S33
T8 S8
T9 S1
Tl0 s 1 3S
, 16
T11 s27,S28,S29

means of the smallest and largest value, on all traverses, 3.4 Qm and 8.2 Qm,
respectively.
From the results of boreholes 81-11 we see that the upper part of this
formation corresponds to the upper sandstone series of the Oligocene-Eocene
formation. The formation is several hundred meters thick in the north of the
investigatedarea, while in the south it does not exceed 100 m.
3. The last and deepest formation, identified geoelectrically, has a high electrical
resistivitygenerallybetween 13.6Qm and 44 Qm. The apparent resistivitycurves
'lows'
show that in some placesthe formation resistivitybecomesvery low. The
lying in the east of the investigated area might reflect sea water intrusions, while
the rest are located in the main area of geothermal interest and are listed in
table 2.

For geoelectric traverse Tll, this low resistivity region reaches the surface at
sounding locations S28 and S29, which coincides with the location of the hot water
springs of Genisea. From the same section we see that the conductive zone dips
westwards reaching its greatestdepth below sounding location S27.
As the interpretation diagram shows, the above formation has resistivities lying
between 4.7 Qm and 5 Qm for the region between soundings S27 and S28, while in
the immediately adjacent areas the resistivity is higher. This suggeststhat hot min-
eralized water flows towards the surfaceand lowers the formation's resistivity.
The same picture is found at traverseTl0 at locations S13, 16, 17 and 25, while
in section T9 the conductive zone is very concentrated and is located just below
sounding Sl at a depth of about 125 m. At the same depth, borehole 83 found
water with a temperature of 78'C.
In traverseT8, this conductive zone is expanded,and has a higher resistivity.
Similarly, expansionis observed in section T7 towards the east with even higher
resistivity. It must be noted that the shape of the resistivity curves is affected by sea
water intrusion in this area.

Spatial uariation of the bedrock

The results of the geoelectric interpretations with information about the possible
depth to the bedrock are compiled in fig. 10, qhich shows the spatial variation of
the depth to geoelectricbedrock for the whole of the investigated area.
96 C . T H A N A S S O U L A S ,G . - A . T S E L E N T I S A N D N . K O L I O S

*'%
"o-

N
aO\
t d-/

- Regionot mox. oeepQn

Depth to bedrock in m from gr. surfoce.

Fig. 10. Map showing the depth to bedrock.

This diagram shows that the bedrock becomes deeper northwards, and that a
zone with a depth greater than 400 m crosses the investigated area from west to
east.
It must be noted that all the boreholes that produced hot water were located in
the south of the investigated region, where the bedrock is shallow. In the northern
part of the area all the drilled boreholes indicated high temperatures but the dis-
charge was unsatisfactory,probably becausethe boreholesdid not reach bedrock.

Sperrnr VlRrlrroN oF TEMPERATURE

The great number of wells in the investigated area allow the assessmentof the areal
variation of temperature and its comparison with variations of resistivity as deduced
from the geoelectricsurveys.
From temperature measurementsin each well, we calculated the formation tem-
perature corresponding to a depth of 350 m (fig. I 1).
We see that the curves of equal temperature are elongated in the NE-SW
direction. The good correlation between temperature and resistivity data is obvious.
For example, the high temperature region (T > 50"C) coincides with the low resis-
tivity region (p"<75 Am). A comparison of figs 9 and 10 shows that the high
temperature region corresponds to the area where the depth to the geoelectric
bedrock is greatest.

CoNcrusroNs
The geoelectric investigations showed that there are three geoelectric formations
with electrical resistivities,from top to bottom, between l1-90 Qm, 1.8-11 Qm, and
1444 Qm.
 G E O T H E R M A LP R O S P E C T I N G 97
The spatialvariationof the temperatureof the formationwe+foundto correlate
very closelywith electricalresistivityvariations,with the resistivitylows following
very closelythe high-temperaturearea.
A numberof boreholesrevealedsomeshallowwater horizonsof limited extent
(figs2a and 2b).
The high temperaturesare characteristicallyfound where the bedrock dips
towardsthe centerof the areain an E-W direction.
This study illustratesthe possibilityof localizingsmall geothermalfeaturesby a
detailedand carefullydesignedgeoelectricsurvey.