The structure of the ophiolitic belt in Albania inferred ... · In this work the existing ground...

231

ANNALS OF GEOPHYSICS, VOL. 48, N. 2, April 2005

Key words Total Magnetic Field – ophioltic belt –2.5D modelling – Albania

1. Introduction

The existing studies of the potential fieldsin Albania are mainly restricted to an analysis

of the ground gravimetric data rather than thegeomagnetic data (Lubonja et al., 1977; Arapi,1982; Bushati, 1988; Frasheri et al., 1991;Veizaj, 1995). Frasheri et al. (1995) present re-sults from some potential field interpretationsand their relation to the regional geologicproblems. Veizaj (1995) presents a model ofthe crust based on some gravimetric profilesacross Albania also using the interpretation ofthe Bouguer anomaly map of Albania and thefree air data of Adriatic Sea. Bushati (1988)compiled the Bouguer anomaly map of Alba-nia and related the gravity high at the Midritazone to the moderately serpentinized mantle

The structure of the ophiolitic belt in Albania inferred

from geomagnetic anomalies

Ilir Kane (1), Alexandros Stampolidis (1), N. Grigoris Tsokas (1), Salvatore Bushati (2),Despoina Kondopoulou (1) and Panagiotis Tsourlos (1)

(1) Department of Geophysics, School of Geology, Aristotle University of Thessaloniki, Greece(2) Centre of Geophysical and Geochemical Exploration (CGGE), Tirana, Albania

AbstractThe ground magnetic measurements in Albania were used for the compilation of the Total Magnetic Field (TMF)anomaly map of Albania. The magnetic data were processed and interpreted in order to study the structure of theophiolitic belts of Albania. The ophiolites of Albania are placed at the Mirdita zone and are divided into two par-allel alignments which are called the eastern and western ophiolitic belts. They are associated with strong poten-tial field anomalies and their characteristics are considered crucial for a better understanding of the tectonic set-tings of Albania. The ground TMF data used in this study were acquired over various campaigns (1990-1994) andcover most of Albania’s territory. The data were compiled to a map after reduction to the epoch 1990.4. Thestrongest magnetic anomalies in Albania appear along the known ophiolitic belts which trend NE-SW to the northand NW-SE to the south. Several processing steps were applied to the unified and gridded data in order to obtaininformation on the distribution of the magnetic sources. The magnetic sources were subsequently modeled usinga 2.5D inversion technique. The magnetic properties of the ophiolites determined from laboratory measurementson rock samples, while their lateral extent was calculated from the processing of the magnetic data and used asconstraints to the inversion procedure. The bottom of the ophiolitic belts is considered to be predominated byharzburgites. They exhibit lower magnetization than other rocks of the complex, i.e. gabbros, basalt. In fact gab-bros are associated with the observed high frequency magnetic anomalies. The modeling results indicate that thethickness of the ophiolites is reduced from east to west. The eastern ophiolitic belt has a maximum thickness ofapproximately 12 km at its northern section (Kukesi and Lura massifs). The magnetic sources appear with rela-tively small thickness at the western ophiolitic belt. Boundaries of the eastern ophiolites are characterized by ver-tical contacts.

Mailing address: Dr. Alexandros Stampolidis, Depart-ment of Geophysics, School of Geology, Aristotle Univer-sity of Thessaloniki, P.O. Box 352-1, Thessaloniki, GR-54124 Greece; e-mail: [email protected]

232

Ilir Kane, Alexandros Stampolidis, N. Grigoris Tsokas, Salvatore Bushati, Despoina Kondopoulou and Panagiotis Tsourlos

rocks there. Bushati et al. (1996) suggestedthat the ophiolites have thicknesses varyingfrom 10-14 km at the NE part of the ophioliticbelt to 6 km at the south-eastern Bulqiza Mas-

sif. The western ophiolites are typically only2-3 km thick.

The measurements of the geomagnetic fieldin Albania consist mostly of regional datasets

1 2

Fig. 1. Simplified geologic map of Albania with the sample sites at the ophiolitic rocks. The tectonic zones are:Sazani, Ionian, Kruja, Krasta-Cukali, Alps, Mirdita, Korabi, Gashi, depression of Mirdita zone and Peri-Adriat-ic depression.

Fig. 2. A simplified geologic map of Mirdita zone. I – western belt; II – eastern belt; III – volcanic rocks; IV– gabbros; V – Cretaceous limestones; VI – Tithonian breccia; VII – Paleocene Quaternary sedimentary rocks;VIII – Heterogeneous ophiolitic mélange; IX – Upper Triassic-Lower Jurassic limestones. The codes in the cy-cles taken from table I and denote the sites where the rock samples were extracted for laboratory measure-ments. The numbers in polygons denote the ophiolitic massifs. 1 – Tropoja; 2 – Kukesi; 3 – Lura; 4 – Bulqiza;5 – Shebeniku-Pogradec; 6 – Krrabi; 7 – Terbuni; 8 – Gomsiqe; 9 – Skenderbeu; 10 – Shpati; 11 – Vallamara;12 – Devolli; 13 – Voskopoja; 14 – Morava. The location of the eight profiles, where 2.5D modeling was at-tempted, is shown as well (i.e. Pr1-Pr8).

Tabl

eI.

Mag

netic

pro

pert

ies

of t

he r

ock

sam

ples

col

lect

ed f

rom

the

out

crop

s of

oph

iolit

ic r

ocks

. The

site

s ar

e sh

own

in f

ig. 2

in

cycl

es w

ith t

heir

code

nam

es. M

easu

rem

ents

of

the

susc

eptib

ility

wer

e ca

rrie

d ou

t in

the

labo

rato

ry u

sing

a B

artin

gton

MS2

sus

cept

ibili

ty m

eter

and

the

mea

sure

men

tsof

the

mag

netiz

atio

n w

ere

carr

ied

out

usin

g a

MIN

ISPI

Nm

agne

tom

eter

. No.

is

the

num

ber

of s

ampl

es f

or e

ach

site

. MN

RM

is N

atur

al R

eman

ent

Mag

-ne

tizat

ion,

κis

mag

netic

sus

cept

ibili

ty,M

IND

is in

duce

d m

agne

tizat

ion

(mag

netic

fiel

d in

tens

ity w

as c

alcu

late

d by

the

IGR

F m

odel

),Q

is th

e K

oeni

gs-

berg

er r

atio

(M

NR

M/M

IND).

MT

OT

is th

e su

m o

f M

NR

Mpl

us M

IND

(sup

pose

d to

be

para

llel)

.

Roc

k ty

peN

o.Si

teC

ode

Lon

g.L

at.

Mag

netic

Sus

cept

ibili

tyD

NR

MI N

RM

MN

RM

M

IND

QM

TO

T

κ(1

0− 5un

its S

I)

10−

3A

m− 1

10−

3A

m− 1

10−

3A

m− 1

Min

Max

Ave

r.M

inM

axA

ver.

5B

ulqi

zaB

l20

.16

41.5

664

295

076

422

9−3

70.

4244

921

.225

.70.

8246

.98

767

193

421.

0712

525

.71.

0051

.44

Gjin

arG

j20

.24

41.0

092

015

9011

7034

943

2.7

12.2

7.5

36.5

0.20

44.0

Har

zbur

gite

910

6010

7−2

92.

526

100.

2746

.54

Lur

a 1

L20

.17

41.7

1100

7540

2200

--

49

880

.73

0.10

88.7

5L

ura

217

9014

529

411

862

.40.

1270

.45

Terb

unT

r19

.90

41.9

824

3037

0026

0021

8−4

611

5635

95.5

0.36

130.

5

5K

rast

aK

r20

.21

41.4

322

6028

0025

0019

3−2

9.7

540

1100

900

91.5

9.

899

1.5

Dun

ite4

Kal

imas

hK

l20

.22

42.0

539

057

0040

0-

--

--

--

-4

2000

--

--

-

226

014

37.

77

7.3

0.95

14.2

92

Boz

haj

Bo

20.5

840

.54

170

3580

045

016

2−3

0.9

560

3016

.41.

8249

.41

Gab

bro

350

0020

343

7018

1.5

0.38

251.

53

3500

032

947

470

1234

.20.

3817

04.1

9B

ulsh

arB

u20

.02

41.8

928

863

845

0-

--

--

--

-5

530

268

−23

411

.85.

519

.50.

2825

.0

Dio

rite

7M

ash/

kor

M93

043

0010

00-

--

--

--

-

3V

ithku

qV

20.5

340

.46

510

1540

070

033

2.2

7.5

1.5

320

225

.50.

0827

.53

1520

020

3.5

45.5

200

546.

30.

3674

6.3

3St

rato

ber

St20

.62

40.5

540

680

600

312

−14

0.15

4.8

2.1

21.9

0.09

24

Bas

alt

652

211

3578

029

425

.40.

3214

.87.

60.

3529

.55

Pesh

qesh

Ps20

.03

41.8

519

750

4360

034

300

120

32.3

88.7

1374

120

1415

0.08

1535

3K

urbn

esh

Ku

20.1

41.7

425

0030

200

2200

059

7518

394

169

791

0.21

960

229

500

95−5

024

30.

3010

346

Rra

sa2

Rr

20.0

641

.95

1055

015

100

1260

028

5−2

.460

135

115

472

0.24

587.

6

Ker

atop

hyre

6R

rasa

1R

r20

.06

41.9

518

0063

0027

800

342

13.5

330

1190

800

1484

0.53

2284

5Pr

ecaj

Pr20

.17

42.1

817

7019

050

1800

031

3−1

1.5

1975

3700

2000

5629

0.75

7629

Dia

base

4B

raka

Br

19.9

41.9

479

8011

400

1100

027

110

1965

3100

2800

3648

0.76

6448

233

The structure of the ophiolitic belt in Albania inferred from geomagnetic anomalies

234


measured over limited areas for mineral explo-ration purposes. Further, measurements werecarried out to map regions that include the ul-tramafic massifs (Angelopoulos and Bushati,1993). The total magnetic field and the verticalcomponent were measured on some profilesacross Albania (Geology of Albania, 1982).However, they comprised a small number andthey were rather sparse to give an idea on thedistribution of the geomagnetic sources in Al-bania and no modelling was attempted.

The main subsurface magnetic sources inAlbania are the ophiolites which are mainly lo-cated at the Mirdita zone (fig. 1). The ophioliticrocks of the Mirdita zone generate very distin-guishable magnetic anomalies compared tothose of the surrounding sedimentary rock for-mations. Thus, the study of the magnetic field isconsidered a useful tool for the structural inves-tigation of the ophiolites in Albania.

In this work the existing ground geomagnet-ic data were compiled to produce a map of themagnetic field of Albania. Different filters andtransformations were applied to the gridded da-ta in order to obtain information about the dis-tribution of the magnetic sources. Also, inter-pretation methods such as «terracing» and 2.5Dinversion were employed to reconstruct themagnetic sources along selected profiles per-pendicular to the ophiolitic belts. The resultscompared with other geophysical studies(Bushati, 1988; Bushati et al., 1996) and exist-ing geological information, hopefully con-tribute to the study of ophiolites in Albania.

2. Geological setting

The Albanides represent the assemblage ofgeological structures that lie in the territory ofAlbania. They are located between the Dinar-ides in the north and the Hellenides in the south.Together they form the Dinaric branch of theMediterranean Alpine belt. The Albanides aredivided into Internal (east) and External (west)zones and between them extends the Midritazone. The Internal Albanides are characterizedby the continental formations of the Korabi-(Pelagonian in Greece) zone, while the externalzones by a continuous sedimentation and defor-

mation from Triassic to Paleogene times. TheExternal Albanides are the Krasta-Cukali andthe Kruja zones, corresponding to the Pindosand Gavrovo nappes in Greece respectively, andthe Ionian and Sazani zones (fig. 1), (Aubouin,1959; Geology of Albania, 1982). The CentralMidrita zone comprises mainly ophiolites asearly mentioned.

The Midrita zone ophiolites (fig. 2) form abroad, westward-convex arc, ca. 250 km longand up to 70 km wide, with an outcrop area ofapproximately 4000 km2 (7000 km2 if the sedi-mentary cover is excluded). They belong to theSSE trending Dinaric mountainous system(Nowack, 1929) and link Dinaric ophioliteswith Hellenic ophiolites (Aubouin and Ndojaj,1964). The Midrita zone represents an oceaniclithosphere segment one of the most completeand coherent sequences among the Mediter-ranean ophiolites and is constituted by NNW-SSE trending sub-parallel belts.

In the Mediterranean Alpine belts, a westernlherzolite ophiolite belt, comprising the Midritaophiolite, and an eastern harzburgite belt (Var-dar) were distinguished in the early literature(Nicolas and Jackson, 1972). Based on petro-logical variability, geochemical and metallo-genic studies, a western-type lherzolitic beltand an eastern-type harzburgitic belt have beendistinguished in the Midrita ophiolite itself(Ndojaj, 1963; Dede et al., 1966; Tashko, 1976;Shallo et al., 1985, 1989; Shallo, 1992; Becca-luva et al., 1994; Robertson and Shallo, 2000)(fig. 2). The eastern-type ophiolitic belt in-cludes the Tropoja, Kukesi, Lura, Bulqiza,Shebeniku-Pogradec massifs, while the mainmassifs of the western-type belt are the Krrabi,Terbuni, Gomsiqe, Kashnjet, Skenderbeu, Ku-turumani, Shpati, Devolli and Morava massifs.Summary logs of the western and eastern-typeophiolites of Midrita zone are presented in fig.3 (Robertson and Shallo, 2000).

The western-type ophiolites (Triassic-Juras-sic) – This belt is relatively thin (ca. 2-3 km)and consists of high-Ti tholeiites within vol-cano-sedimentary series in the western periph-eral parts and a volcanic sequence in the west-ern part, associated with lherzolite mantle suiteand minor harzburgite and dunite tectonite and

235


ultramafic and mafic cumulates (Shallo et al.,1985; Beccaluva et al., 1994; Hoxha, 2001).They are disrupted into a series of large thrustsheets and substantial detached blocks that aremainly eastward-dipping, to steeply inclined,folded and cut by steep faults. Ophiolitic vol-canics and sedimentary cover units are pre-served in kilometre-scale synclines in NorthernAlbania (Robertson and Shallo, 2000).

The eastern-type ophiolites (Jurassic) – Thisbelt is characterized by low-Ti tholeiites ofbasalt-andesite and andesite-dacite (rhyolite) se-ries underlain by sheeted dyke complex, andquartzdiorite-plagiogranite and gabbros (ISPGJ-IGJN, 1982; Shallo et al. 1985; Beccaluva et al.,1994; Hoxha, 2001). Ultramafic mantle se-quences are represented by clinopyroxene-poorharzburgite, overlain by harzburgite-dunite andboth interbedded with and overlain by thicktransitional dunite, rare wehrlite-lherzolite and

cumulate pyroxenite. The eastern-type ophio-lites reveal a major thrust sheet, 6-8 km thick,segmented into several large massifs. The exis-tence of a stack of thrust sheets in the eastern-type ophiolite indicates that they were internal-ly disrupted. Triassic limestone blocks along thefront of the biggest thrust and sub-alkali volcan-ites (not belonging to the ophiolite series) in thecenter of the ophiolites indicate the presence ofexotics emplaced during the Late Jurassic-EarlyCretaceous ophiolite obduction, and are possi-bly derived from seamounts resting on the de-scending Triassic oceanic slab (Hoxha, 2001).

Detailed structural mapping reveals that thedeep mantle section in both massifs washarzburgitic and that the major differences be-tween the two ophiolitic belts are restricted to theuppermost mantle and lower crustal section(Nicolas and Boudier, 1999). These are typically«ophiolitic» in the eastern massifs, being com-posed of a thick dunitic transition zone rich inbasaltic impregnations and chromite depositsoverlain by a lower crust of layered gabbros. Incontrast, in western massifs the uppermost man-tle is composed of highly strained to myloniticlherzolites which originate from more depletedharzburgites by impregnation and tectonic dis-persion of melt during deformation occurring at1000°-800°C. Layered gabbros are locally ab-sent, and the crust can be reduced to diabasedikes or sills and extrusives. The contrast be-tween the eastern and western massifs is as-cribed to successive episodes of magmatic andamagmatic spreading in a slow spreading envi-ronment (Nicolas et al., 1999). Furthermore, anundisturbed lateral transition exists between thewestern and eastern-type ophiolites, as seen inthe western part of the Bulqiza and Shebenikumassifs (Manika, 1994; Manika et al., 1997; Be-bien et al., 1998). In some areas this transitionzone is complicated by eastward-dipping thrusts,backthrusts and sub-vertical faults, which led tothe contact previously being interpreted as a re-gional thrust (Beccaluva et al., 1994). Robertsonand Shallo (2000) suggested that the western andeastern-type ophiolite formed together and wereemplaced as a single thrust sheet. Gradually theophiolites moved westwards over the continentalcrust, overthrusted the Krasta-Cukali zone. Theuppermost part of the western and eastern vol-

Fig. 3. Summary logs of the western and easternophiolites, Midrita zone (Robertson and Shallo,2000).

236


canic ophiolite is overlain by a remarkable dep-osition, a «block-in-matrix-type» mélange set inan argillite matrix, considered to be Late Juras-sic-Early Cretaceous in age (ISPGJ-IGJN 1982;Shallo 1984; Shallo et al., 1985; Hoxha, 2001).

The nature of the contact between the Kras-ta-Cukali zone and the Mirdita zone is variablein different areas, ranging from an eastward-dipping thrust (e.g., Miraka, central region;Voskopoja, southern region), to a high-anglefault (e.g., Hajmeli, northern region). By con-trast, the contact of the eastern-type ophiolitewith the Korabi zone to the east varies from ahigh to a moderate-angle west-dipping reversefault zone (fig. 11a,b), to a westward-dippinglow-angle thrust (e.g., Bulqiza ophiolitic mas-sif, central region), or an overturned fault (e.g.,Selishte, Kolesjani, Luma R., northern region).The majority of these fault contacts were activeduring the Early Tertiary. Taking account ofsub-surface evidence (Frasheri et al., 1996), thewestern ophiolite contact is regionally a rela-tively gently inclined thrust (< 45°), whereasthe eastern contact is steeper (> 50°).

3. Data acquisition and reduction

TMF intensities were measured over vari-ous campaigns during the period 1990-1994 bythe Magnetometry Department of the CGGE,Tirana. A total of 25460 magnetic stations weremeasured using a MP-2 proton precession mag-netometer (resolution 0.1 nT), covering most ofthe Albanian territory which is about 28000km2 (Angelopoulos and Bushati, 1993) (fig. 4).The coordinates and the elevation of the sta-tions were extracted from the topographic mapwith scale 1:25000. The accuracy of the meas-urements was ± 25 nT for the Internal Alban-ides and ± 10 nT for the External Albanides(Angelopoulos and Bushati, 1993).

The distribution of magnetic stations is de-picted in fig. 4 and was not regular due to therugged topography of the country. At the regionof Kukesi (10 × 12 km) for example, samplingwas quite dense with measurements obtainedevery 100 m (a total of 8350 points in this re-gion). Conversely, the Bulqiza Massif was notcovered with a dense grid: the profiles in this

zone were 2-3 km apart and the magnetic sta-tions were every 500 m.

At the ophiolitc massifs of Lura, Shebenikuand Gjinar the measurements were also ob-tained on profiles positioned 2 km apart andstationed every 500 m. Over the Puka Massifthe measurement grid was every 500 m. TheExternal Albanides (western part of the coun-try) were covered by a sparse grid, with stationsevery 500-1000 m, while some empty corri-dors, 5-10 km wide, exist at the southern part ofthe country. Due to the very rough topography

Fig. 4. Location of the ground magnetic stations.

237


Fig. 5. The residual total magnetic field anomalymap of Albania.

in the area of the Albanian Alps (20 × 15 km) nodata exist at this region.

Griding was performed using a minimumcurvature algorithm and with constant gridspacing equal to 500 m. The measured datawere corrected and reduced for the 1990.4epoch utilizing the IGRFGRID program (Phillips,1997). The Definite Geomagnetic ReferenceField (DGRF) was calculated and then removedfrom the grided data. Furthermore, a planar re-gional trend was removed by the use of a leastsquares algorithm. The planar trend representslong wavelength components rising from to-pography, regional features and magnetic corefields not adequately be removed by the DGRF

model used. The resulted residual total fieldanomaly map, on which the data enhancementtechniques and interpretations were applied, isshown in fig. 5.

4. Rock magnetic properties

Some rock samples were collected from themost representative outcrops of the ophioliticrocks in order to study their magnetic proper-ties. These values were used as reference for themodelling of the magnetic sources.

Crustal magnetic anomalies reflect the spa-tial variation of the total magnetization of therocks. To understand the origin of this variabil-ity it is necessary to know the magnetic proper-ties of the rocks.

Total magnetization T is the sum of two vec-tors: remanent magnetization ∆F (Natural Rema-nent Magnetization, NRM) and induced magneti-zation F. That is, n/T F F F B= + = + l∆ ∆ ,where κ is the magnetic susceptibility of the rock,µ is the magnetic permeability and B is theEarth’s magnetic induction. In ophiolitic rocks theNRM is mostly TRM (Thermal Remanent Mag-netization) and CRM (Chemical Remanent Mag-netization). TRM is usually primarily acquiredwhen rooks cool down below their Curie temper-ature and CRM can be either primary or second-ary and is acquired when the rocks are subjectedto chemical transformation, recrystalization etc.There is evidence of strong chemical transforma-tion of the ophiolitic rocks in Albania (Shallo etal., 1989) that lead to the existence of the CRM inthese rocks (Frasheri, 1989).

Measurements of the remanent magnetiza-tion and the magnetic susceptibility were per-formed on the ophiolitic rock samples extractedfrom different ophiolitic outcrops in Albania(fig. 2). Magnetic susceptibility and magnetiza-tion have been measured on 84 rock samplescollected from 10 different sampling sites withharzburgites, dunites, gabbros, basalts and dior-ites outcrops (fig. 2).

The results of the magnetic susceptibilityand NRM measurements, which were carriedout on those samples, are shown in table I. Theresults agree with the work of Frasheri (1989)who measured the induced and remanent mag-

238


netization of some ultrabasic rocks calculatingalso the Q (Koenigsberger ratio) on 700 sam-ples in the sense that the magnetism of ultraba-sic rocks varies in a broad band of values con-ditioned by the presence of ferromagnetic min-erals. The ultrabasic rocks appear to be ferro-magnetic with broadly varying values of mag-netic susceptibility and NRM, both in ampli-tude and direction.

The harzburgites of the Bulqiza Massif havethe lowest magnetic susceptibility (average700×10−5 units SI), whereas in the Lura Massifthey exhibit higher values of susceptibility(2200×10−5 units SI, table I). When these rocksare serpentinized, i.e. undergoing a processwhich causes the creation of secondary magnet-ic phases, the magnetic properties are muchstronger. In this case, magnetic susceptibilityreaches very high values (e.g., 7500×10−5 SI).

Dunites have a magnetic susceptibilityhigher than the non-serpentinized harzburgites.Four dunitic samples from the site of Kalimashexhibit high magnetic susceptibility, around2000×10−5 units SI (table I).

Gabbros seem to be weakly magnetized. Themagnetic susceptibilities of the Gabbro samplesat Bulshar are lower than those of the non-ser-pentinized harzburgites. However, the growth ofthe amount of magnetite in gabbros enhancestheir magnetic susceptibility and they can be-come strongly magnetic. The magnetic suscepti-bility, as in the Gabbro samples from the site ofBozhaj, changes in a very broad limit, e.g., from200-400 to 5000×10−5 and the maximum valuesare of the order of 35000×10−5 units SI.

The basalts are comparable to non-serpen-tinized harzburgites. But in some samples veryhigh values of κ were found e.g., 1500×10−5

units SI, the average value being in the order of600-700×10−5 units SI.

The susceptibility measurements indicate ahigh degree of variability in magnetic propertieswithin the same rock type and between differentrock types. This variability may be caused bydifferent magnetic phases with varying concen-tration. Magnetic minerals in plutonic rocksmainly occur as intergrowth and/or as inclusionsin mafic minerals like olivine and pyroxene.

The inclination and declination of the rema-nent magnetization of the samples was also

found to be very scattered and thus we were notable to extract any reliable conclusion about thedirection of the remanent magnetization vector.Note that Frasheri (1989) observed similar be-haviour.

The Koenigsberger ratio for the most of oursamples was less then one (table I). Exceptionsare shown by samples taken from Kalimash(Q=9.8) and some samples taken from Bozhaj(Q=1.82). Also Frasheri (1989) mentioned thatthe Koenigsberger ratio was less than one for57% of the samples from dunites and 47% fromhatzburgites where the ratios were 0.5 and 0.6,respectively. As a result, for the transformationsof the magnetic field and modelling of magnet-ic sources we considered that the magnetizationof the ophiolites was of induced type.

Frasheri et al. (1991) suggested that the sed-iments which surround the ophiolitic belts, andbelong to the adjacent tectonic zones (Korabiand Krasta-Cukali), are characterized by verylow magnetic susceptibility values. Their valuesare of the order of 5-10×10−5 SI and they rarelyreach values as high as 100×10−5 SI. Compari-son of the magnetic properties of the rocks inAlbania suggest that, as expected, intrusiverocks are more magnetic than effusive rocksand sediments and therefore they are responsi-ble for the large magnetic anomalies in the area.The different kinds of intrusive rocks have mag-netic properties that vary from practically non-magnetic to very highly magnetic materials.Thus, the values to be used for the modellingpurposes must be selected taking into accountthis fact.

5. Data processing

The map of the residual total magnetic fieldanomaly of Albania is shown in fig 5. The totalfield magnetic anomalies associated with ophi-olitic rocks are clearly identified by their posi-tive high amplitude values. However, within theophiolite belts the anomaly pattern is complexdue to the superposition of many high-ampli-tude anomalies of different wavelengths. This isexpected, since we have magnetic sources withdifferent dimensions, magnetization and burialdepth.

239


The most remarkable features in this mapare the strong anomalies associated with almostevery ophiolithic outcrop and the weak anom-alies that surround the ophiolithic belts margins.The stronger anomalies are correlated with theKukesi, Lura, Puka and Bulqiza massifs withmaxima up to 800 nT. The Puka Massif in par-ticular, is characterized by high frequencyanomalies. At the Shebeniku and Gjinar massifsthe anomalies are of smaller magnitude at about400 nT (fig. 5). The northern part of the Tropo-ja Massif and the Central Mirdita zone over thevolcano sedimentary series are characterized bymagnetic anomalies with low magnitude. Overthe sedimentary basins of the Quaternary de-pression of the Mirdita zone, which are the Bur-reli to the north and the Librazhd-Moker to thesouth, anomalies of moderate magnitude up to150 nT with dominant low frequencies occur.This provides evidence for the continuation ofthe ophiolites under these basins.

The magnetic anomalies correlated with theophiolitic belts of the Mirdita zone are separat-ed in a northern and a southern part by a corri-dor, which coincides with the Shengjergji fly-sch corridor (fig. 2). This tectonic unit played asignificant role in the geology of Albanides.Geophysical data (Frasheri et al., 1995) showthat the Shengjergji flysch corridor, representsa structural carbonatic-terrigeneous withoutophiolites under the flysch deposits. The dis-ruption of the magnetic anomalies in this areasuggests that the ophiolitic belts of the Alban-ides are tectonically split into two parts. Final-ly in the adjacent (to the Mirdita zone) Korabiand Krasta-Cukali zones no significant magnet-ic anomalies are present.

In order to obtain some information aboutthe location and nature of the magnetic sourcesthe data were subjected to various transforma-tions. Low pass filtering, upward continuation,directional derivatives, reduction to the poleand pseudogravity transformations were ap-plied in order to enhance the data. Finally, the«terracing» operator (Cordell and McCafferty,1989) was applied to delineate the horizontalboundaries of the magnetic sources.

The ophiolites comprise a 30-70 km widebelt, thus, the longest wavelength anomaliesgenerated from these sources should have a

wavelength larger than 30 km. A low pass filterwith cut off wavelength of 25 km was consid-ered appropriate to isolate the anomalies causedby the relatively deep structures of the ophi-olitic belt.

The low pass filtered map is shown in fig. 6.The anomalies which dominate the filtered mapare centered over the outcrops of the ophioliticmassifs of Kukesi, Lura, Bulqiza, Puka to thenorth and Shebeniku and Gjinar to the south.The strongest anomaly is the one over the mas-sif near the town of Kukesi. The anomaly at thePuka Massif remains pronounced despite filter-ing. To the south there are some isolated low

Fig. 6. The total magnetic field anomaly filtered byremoving all wavelengths smaller then 25 km.

240


magnitude anomalies centered over the ophi-olitic massifs. Overall, the anomalies to thenorth of the country are more extended and ofhigher magnitude than those to the south.

The upward continuation process favoursthe long wavelength anomalies (Henderson andZietz, 1949), caused by deep structures. Figure7 depicts the total magnetic field continued up-ward to the altitude of 2500 m above groundlevel. The ground surface, where the measure-ments were taken, is considered flat for this op-eration. The strongest anomalies of the upwardcontinued field are observed in the northernpart especially over the eastern belt.

Positive gravity anomalies are located oversurplus mass concentrations but the same is notnecessary true for magnetic anomalies when themagnetization and the ambient field are not bothdirected vertically. In general the magneticanomalies are dipolar and asymmetrical in shape.This complexity can be eliminated by using re-duction to the pole and pseudogravity transfor-mations (Baranov, 1957; Arkani-Hamed, 1988).

The reduction to the pole operation trans-forms the magnetic anomaly to the anomalythat would be measured at the north magneticpole, where the ambient field is directed verti-cally down. The reduced to the north magnetic

Fig. 7. Upward continuation of the total magneticfield anomaly to 2500 m.

Fig. 8. Reduced to the pole total magnetic fieldanomaly of Albania.

241


pole field of Albania is shown in fig. 8. Theanomalies at the center of the ophiolitic mas-sifs have much larger magnitude comparedwith the small anomalies surrounding the ophi-olitic belts.

Cordell and McCafferty (1989) described atechnique to produce a kind of equivalentsource from measured gravity or magnetic da-ta. This is an iterative technique that transformsthe magnetic data into a field comprised of uni-form domains separated by abrupt boundaries.Thus, density or magnetization within theequivalent layer should remain constant exceptacross these abrupt boundaries. The result pro-duces pseudo-geological boundaries since itderives from the processing of a physical prop-erty. The so-called «terracing» operator is notequated to downward continuation since itdoes not admit the presence of a transition zonebetween the boundaries of the distinguishabledomains. The operation is based on the sign ofthe curvature of the potential field at any par-ticular point. The value of each point of a datagrid is revised upwards or downwards based onthe algebraic sign of the curvature (i.e. the sec-ond vertical derivative) at that point. Repeatedadjustments are made until the terraced effectis achieved. The resulting «terrace» does notdirectly reflect density or magnetization butcan be scaled in order to do so. This schemewas applied to the pseudogravity field of Alba-nia and the result is shown in fig. 9. The resultindicates that the ophiolites are mainly con-fined at depth to the area in which they out-crop. The only exception is seen in the LuraMassif where there is a small extension of the«terraced» boundary east of the geologic out-crop. The «terraced» boundaries at the easternside of the belt are very sharp in comparisonwith the western ones. This is justified by geo-logical evidence that suggest that the edges ofthe ophiolitic massifs at the eastern side are al-most vertical.

The second vertical derivative (Evjen, 1936;Elkins, 1951) of a potential field measured on ahorizontal surface is obtained by multiplying thespectrum of the anomaly by the square of thewavenumber. This procedure helps to better re-solve and accentuate shallow sources since thehigher wavenumber anomalies are magnified.

The steepest horizontal gradient of pseudo-gravity or reduced to the pole magnetic anom-aly caused by a tabular body tends to overlaythe edges of the causative body. Actually, thesteepest gradient will be located directly overthe edge of the body if the edge is vertical andfar removed from all other edges or sources.We calculated the gradient by applying simplefinite-difference formulas to the magneticmeasurements by first reducing them to thenorth magnetic pole and the results are pre-sented in fig. 10. The maxima of the horizontalgradient are calculated and plotted by blackdots on fig. 10. The interpolation of these max-

Fig. 9. The result of «terracing» applied to the totalfield magnetic anomaly of Albania after the removalof all wavelengths smaller than 20 km.

242


ima, showing the steepest gradient values,helps to delineate the edges of the magneticsources. A range of maxima is well correlatedwith the eastern edge of the ophiolites (fig. 10),where the greatest values of the maxima coin-cide with their outcrops. This result indicatesagain that the eastern edge is nearly vertical.

Finally, taking into account the informationprovided by the above described techniques,

modelling was attempted along selected pro-files (fig. 2), perpendicular to the ophioliticbelts. A 2.5D forward method was applied to fitthe synthetic anomaly caused by a certain mod-el to the anomaly observed along the profiles.This procedure provides the magnetic anomalyalong a profile assuming that the anomalousbodies are finite prisms of polygonal cross-sec-tion and uniform magnetization. The prismswere assumed to have their y axis 20 km long ateither side of each profile. The program SAKI(Webring, 1985) was used for the inversion ofthe observed magnetic data. It uses semi-auto-matic Marquardt inversion of gravity and mag-netic profiles.

The modelled ophiolites were grouped atcertain bodies differing in their magnetic sus-ceptibility values. The initial susceptibility val-ues used in the modelling procedure were takenfrom table I. The horizontal boundaries of thebodies, which were taken from previous pro-cessing steps, were used as constrains in the in-version scheme. Depths to the top of ophioliteswere calculated by means of spectrum analyses,while the ophiolites outcrop were used as con-straints. Depths to the bottom of magneticsources, shapes and final susceptibility valueswere calculated by means of inversion.

In our models we assume that the bottompart of the ultramafics in both ophiolitic belts ispredominated by harzburgites. Harzburgitesconstitute also the upper part of the ultramaficsat the eastern-type ophiolite, while lherzolitesconstitute also the upper part of the ultramaficsat the western-type ophiolite, both character-ized by relatively low susceptibility values,ranging from 200 to 1700×10−5 SI. The upperpart of the ultramafic sequence of the western-type ophiolite is predominated by plagioclasedunite and plagioclase lherzolite and horn-blendites, whereas that of the eastern-type isthicker and mainly composed of transitionaldunites and layered chromites. Crustal cumu-late sequence in the eastern-type ophiolite ispredominated by gabbros, which exhibit highmagnetic susceptibility values, while in thewestern-type is troctolitic. Taking all the aboveinto account we split the ophiolites of Midritazone into a number of simple modelling bodieswith differing magnetic susceptibility, in order

Fig. 10. The second horizontal derivative of the mag-netic anomaly of Albania. The derivative is calculatedafter reducing the anomaly to the north magnetic pole.The maxima of the gradient are denoted by black dots.

243


to simulate the magnetic anomalies. Thesemodelling bodies do not correspond to singleophiolitic formations, but to ophiolitic blockswhich include the ultramafic and crustal ophi-olitic sequences. Their susceptibilities are rep-resentative for the whole sequence.

Profile Pr1 (fig. 2) passes through western-type ophiolite. Two modeling bodies with sus-ceptibilities that are slight different (fig. 11a)were used in this model. The thickness of lowmagnetic susceptibility Neocene molassic de-posits, which cover the ophiolites range from1.0 to 1.5 km. The thickness of ophiolites at this

part of Albania (fig. 2) does not exceed 3 km. Inprofile Pr2 (fig.2) the modeling bodies (fig.11b) are similar to those of Pr1, their thicknessdoes not exceed 4.5 km and the sedimentarycover is less than 1 km thick. The modelingbody 1 in Pr3 (fig. 11c), which correspond toDevolli Massif, appears detached from the mainophiolitic body 2, and has relatively low sus-ceptibility. Body 1 in Pr4 (fig. 11d) correspondsto western-type ophiolite, while body 2 to east-ern-type. They have similar magnetic propertiesand their thickness is about 4 km. The thicknessof the sedimentary cover in this profile is about

Fig. 11a-d. 2.5D modeling of the magnetic sources along the profiles Pr1 (a), Pr2 (b), Pr3 (c) and Pr4 (d) (seefig. 2 for the location of profiles).

a b

c d

244


2 km. In profile Pr5 (fig. 12a) modelling bodies1 and 2 correspond to dunite and harzburgiterich bodies. Dunite exhibits three times highersusceptibility values than harzburgite. Thethickness of the ophiolites is less than 4 km.

At the northern part of Albania the ophio-lites become thicker (10-12 km). We introducesome modeling bodies on top of the main ophi-olite blocks (correspond to gabbroitic, basalticand peridotic bodies) in order to compensatethe high frequency part of the magnetic field(fig. 12b-d, profiles 6-7-8), which is character-ized by higher magnetic properties. Their depth

extent is limited (1.5-3 km). Those bodies in-fluence the abrupt fluctuation of the observedfield. The overall thickness of the ophiolites isincreasing, reaching up to 12 km at the north-eastern part where the Kukesi and Lura massifscrop out, while the western part remains rela-tively thin.

The nearly vertical boundaries of the easternmassifs used in the models, after the interpreta-tion of the horizontal gradient as discussedabove, appear to produce a very good fit withthe observed data. This suggests the existence ofnearly vertical faults. The thickness is reducing

a b

c d

Fig. 12a-d. 2.5D modeling of the magnetic sources along the profiles Pr5 (a), Pr6 (b), Pr7 (c) and Pr8 (d) (seefig. 2 for the location of profiles).

245


from west to east. This is in line with the over-thrust structure of the ophiolites at the westernpart, over the sediments of the Krasta-Cukalizone. In general, the thickness of the eastern beltbecomes smaller from north to south. However,in the southern part, the thicknesses of the east-ern and western belts are the same.

6. Discussion and conclusions

We have compiled the total field magneticmap of Albania from ground measurements.The magnetic datasets used do not cover the en-tire Albanian region and have variable density,thus we had to interpolate and regrid the data.

Processing and interpretation techniqueswere applied to enhance the information fromthe magnetic data and to produce structural andgeological models of the ophiolitic belts in Al-bania, which are characterized by strong mag-netic anomalies, compared to the surroundingformations. From our study and from previousgeological and geophysical work in this area,we conclude that the ophiolitic belts in the Al-banides are tectonically split into two parts, anorth and a south part, from Shengjergji corri-dor under which there are no ophiolitic bodies.The compiled magnetic map (fig. 5) supportsthe previous geological studies which suggestthat there is no continuation of the ophiolitesunder the sediments of the Shengjergji corridor(absence of strong magnetic anomalies). Gab-bros exhibits the highest magnetic susceptibili-ty values. It is two to three times higher thanthe susceptibility of the harzburgites. The sedi-mentary cover has thickness that is about 1000-1500 m. The eastern ophiolitic belt has thegreater thickness at the northern part. It is about12 km near the Kukesi and Lura massifs. Themagnetic sources have smaller thicknesses inthe western belt at the northern part. Near theGomsiqe Massif the thickness does not exceed3 km. There is an abrupt change in thickness inthe northern part from the western to the east-ern belts. From south to north the thickness ofthe eastern belt increases. Steep eastern bound-aries of the ophiolitic bodies appears at themagnetic models correspond to nearly verticalfaults.

Acknowledgements

This work was carried out as part of theSeis-Albania project (No. 972342) in theframework of the NATO «Science for Peace»program. We want to thank Prof. Anastasia Ki-ratzi and Prof. Betim Muco for the inclusion ofthis work in the «Science for Peace» programand the employees of CGGE (Centre of Geo-physical and Geochemical Exploration) inTirana, Albania for their help. Prof. AlfredFrasheri, Prof. Minella Shallo and Dr. SimoSpassov for their help and useful suggestions.

REFERENCES

ANGELOPOULOS, A. and S. BUSHATI (1993): Unification ofgravity, magnetic, radioactivity of Albania, Greece andother Balkan’s, Final Report CGGE, Tirana, Albania.

ARAPI, S. (1982): The study of the relationship of the grav-ity field distribution in the Ionic, Kruja and Sazani out-er tectonic zones in the context of the geologic-geo-physical researches of the perspective structures for oiland gas, MSc. Thesis, Geophysical Enterprise, Fier, Al-bania.

ARKANI-HAMED, J. (1988): Differential reduction to pole ofregional magnetic anomalies, Geophysics, 53, 1592-6000.

AUBOUIN, J. (1959): Contribution a l’etude geologique de laGrece septentrionale. Les confins de l’Epire et de laThessalie, Ann. Geol. d. Pays Helen., 10, 1-525.

AUBOUIN, J. and I. NDOJAJ (1964): Regard sur la géologie del’Albanie et sa place dans la géologie des Dinarides,Bull. Soc. Géol. Fr., Ser. 7, VI (6), 593-625.

BARANOV, V. (1957): A new method for interpretation ofaeromagnetic maps, pseudogravimetric anomalies,Geophysics, 22, 359-83.

BEBIEN, J., M. SHALLO, K. MANIKA and D. GEGA (1998):The Shebenik Massif (Albania): a link between MORand SSZ-type ophiolites?, Ofioliti, 23, 7-15.

BUSHATI, S. (1988): Regional study of gravity field scatter-ing in the inner Albanides, for the tectonic and miner-alogenic regioning, MSc. Thesis, Geophysical Enter-prise, Tirana, Albania.

CORDELL, L. and A.E. MCCAFFERTY (1989) A terracing op-erator for physical property mapping with Potentialfield data, Geophysics, 54, 621-34.

DEDE, S., R. SHEHU and M. SHALLO (1966): Magmatizmiintruziv me lhgiperi dhe lidhya e mineralizimit me te,Permbledhje. Stud., 3, 1-15.

ELKINS, T.A. (1951): The second derivative method of grav-ity interpretation, Geophysics, 16, 29-50.

EVJEN, H.M. (1936): The place of the vertical gradient ingravitational interpretations, Geophysics, 1, 127-136.

FRASHERI, A. (1989): Physical properties of chrome iron oreand ultrabasic rocks in the Albanides, in LeobenerHefte zur Angewandten Geophysik, 2, 100-125.

246


FRASHERI, A., V. LUBONJA, P. NISHANI, S. BUSHATI, A. HY-SENI and V. LECI (1991): Les donnes geophysiques surles relation entre les zones tectoniques des Albanides aterre et sur le plateau continental de la mer Adriatique,in Colloque sur la Geologie de l’Albanie (Paris, Soc.Geol. Fr.).

FRASHERI, A., P. NISHANI, S. BUSHATI and A. HYSENI (1995):Geophysical study of the Albanides, Boll. Geofis. Teor.Appl., 37 (146), 159-178.

GEOLOGY OF ALBANIA (1982): Notes of the new geologicmap of Albania 1:200000, I.S.P.G. J. Tirana, p. 342.

HENDERSON, R.G. and I. ZIETZ (1949): The upward contin-uation of the anomalies in total magnetic intensityfields, Geophysics, 14, 517-534.

HOXHA, L. (2001): The Jurassic-Cretaceous orogenic eventand its effects in exploration of sulfide ores, Albanianophiolites, Albania, Eclogae Geol. Helv., 94 (2001)339-350.

ISPGJ-IGJN (1982): Geology of Albania, Tirana, Albania(published in 1990, pp. 492, in Albanian).

LUBONJA, L., A. FRASHERI and A. QIRINXHI (1977): The grav-ity and magnetic surveys usage for regional geologicalstudies in Albania, Summary of Studies, 7, 49-62.

MANIKA, K. (1994): Petrologie du massif ophiolitique deShebenik (Albanie), Ph.D. Thesis, Université Paris-Sud,pp. 297 (unpublished).

MANIKA, K., M. SHALLO and D. GEGA (1997): The plutonicsequence of Shebenik ophiolitic complex, Albania: evi-dence for dual magmatism, Ofioliti, 22, 93-99.

NDOJAJ, I.G. (1963): Mbi disa probleme te magmatizmit nevendin tone, Bil. Shkiperise Nat., 2, 1-13.

NICOLAS, A. and F. BOUDIER (1999): Slow spreading accre-tion and mantle denudation in the Mirdita ophiolite (Al-bania), J. Geophys. Res., 104, 15,155-15,167.

NICOLAS, A. and E.D. JACKSON (1972): Répartition en deuxprovinces des peridotites des chaines alpines longeant laméditerranée: implications géotectoniques, Bull. SuisseMiner. Petrol., 53, 385-401.

NOWACK, E. (1929): Geologische Ubersicht von Albanien,Salzburg.

PHILLIPS, J. (1997): Potential-field geophysical software forthe PC version 2.2, U.S. Geol. Surv. Open-File Rep. 97-725.

ROBERTSON, A. and M. SHALLO (2000): Mesozoic-Tertiary tec-tonic evolution of Albania in its regional Eastern Mediter-ranean context, Tectonophysics, 316 (3), 197-254.

SHALLO, M. (1992): Geological evolution of the Albanianophiolites and their platform periphery, Geol. Rund-sch., 81 (3), 681-694.

SHALLO, M., DH. KOTE, A. VRANAJ and I. PREMTI (1985):Magmatismi ofiolitile i shqiperise, Ist. SP. Geol.,Tirana, Albania.

SHALLO, M., DH. KOTE, A. VRANAJ and I. PREMTI (1989):Some petrological features of the ophiolite of Alban-ides, Bul. Shk. Gjeol., 2, 9-27.

TASHKO, A. (1966): Disa dallime gjeokimike ge verehenbrenda shkembinjeve ultrabazike te vendit tone,Permbledhje. Stud., 4, 101-118.

VEIZAJ, V. (1995): The relationship between Albanid oroge-ny and Apulian platform according to the gravity data,in First Symposium of Oil and Gas, Fier, Albania.

WEBRING, M. (1985): SAKI, a Fortran program for general-ized linear inversion of gravity and magnetic profiles,U.S. Geol. Surv., Open-File Rep. 85-122.

(received February 10, 2004;accepted November 11, 2004)

The structure of the ophiolitic belt in Albania inferred ... · In this work the existing ground...

Documents

Transcript of The structure of the ophiolitic belt in Albania inferred ... · In this work the existing ground...