GeolCarp_Vol57_No1_57

GEOLOGICA CARPATHICA, FEBRUARY 2006, 57, 1, 57—65

www.geologicacarpathica.sk

GPS study (1996—2002) of active deformation along the

Periadriatic fault system in northeastern Slovenia:

tectonic model

MARKO VRABEC

, POLONA PAVLOVČIČ PREŠEREN

and BOJAN STOPAR

University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Geology, Aškerčeva 12, SI-1000 Ljubljana,

Slovenia; mvrabec@ntfgeo.uni-lj.si

University of Ljubljana, Faculty of Civil and Geodetic Engineering, Department of Geodesy, Jamova 2, SI-1000 Ljubljana, Slovenia;

polona.pavlovcic@fgg.uni-lj.si; bojan.stopar@fgg.uni-lj.si

(Manuscript received December 6, 2004; accepted in revised form June 16, 2005)

Abstract: We present GPS-derived 6 year (1996—2002) displacements of 9 sites in northeastern Slovenia, spanning
across the faults of the Periadriatic fault (PAF) system. Site velocities relative to the stable Eurasian plate, while close to
or within the uncertainty limits, indicate predominately N- to NNE-directed movements in the range from 0.5 to 2 mm/yr,
which is consistent with the previously published continuous and episodic GPS observations from the region. Our results
support the recent idea about the ongoing eastward extrusion of the Eastern Alpine domain and confirm that the PAF
system could represent the dextral southern boundary of the extruding unit. However, the deformation in the Slovenian
part of the PAF system is not limited to a single strike-slip zone, but is accommodated within the larger area. Measurable
dextral displacements in the range of ~ 1 mm/yr exist on the Sava and Labot (Lavanttal) faults of the PAF system. No
dextral displacements were observed along the eastern continuation of the Sava fault, which suggests southward displace-
ment transfer and possibly absorption of deformation in the transpressive Sava Folds foldbelt situated south of the fault.
Ongoing thrusting of the Northern Karavanke unit north of the PAF implies active transpression along the main PAF
zone, whereas the region between the PAF and the Sava fault is apparently deformed transtensionally.

Key words: Slovenia, Periadriatic fault, GPS geodesy, neotectonics, transpression, transtension, extrusion.

Introduction

The Periadriatic fault (PAF) is a major post-collisional
structural feature of the Alpine Orogen. Along its entire
length, the PAF system exhibits complex geometrical and
kinematic relationships (e.g. Schmid et al. 1989), some of
which remain controversial (e.g. Viola et al. 2001). The
easternmost outcropping segment of the PAF is located ap-
proximately on the Austrian-Slovenian border (Figs. 1, 2).
Displaced paleogeographical markers and kinematic recon-
structions indicate that it accomodated at least 100 km of
dextral motion (Kázmér & Kovács 1985; Frisch et al.
1998; Fodor et al. 1998), which mostly happened in the
Miocene during eastward extrusion of the Eastern Alps
out of the Adria-Europe collision zone (Ratschbacher et
al. 1991). The dextral PAF acted as a southern border of
the extruding wedge, whereas the sinistral northern
boundary was located along the Northern Calcareous Alps
(Ratschbacher et al. 1991; Frisch et al. 1998). The main
phase of dextral movements on the easternmost PAF seg-
ment ended at the beginning of the Pliocene (Fodor et al.
1998), when extrusion was stopped due to the termination
of subduction in the Carpathians, providing until then a
free boundary for eastward escape (Horváth & Cloething
1996). This event was reflected by the onset of inversion
of extensional structures of the Pannonian Basin (Fodor et
al. 1999) and by transpressional deformation in most of
central and northern Slovenia (Fodor et al. 1998; Márton

et al. 2002). In the territory of eastern Slovenia and north-
eastern Croatia, this inversion episode is documented to
have lasted at least until the end of the Pliocene (Tomljen-
ović & Csontos 2001), and could have continued into
Quaternary times (e.g. Placer 1999).

It has recently been realized that extrusional tectonics

in the Eastern Alpine-Pannonian domain might be active
today. For example, Quaternary to Recent reactivations of
strike-slip zones were documented in the Pannonian Basin
(e.g. Lőrincz et al. 2002). Some of those deeply-seated in-
tra-basinal fault zones are believed to have acted as a con-
tinuation of the PAF during the Miocene extrusion
processes (Fodor et al. 1998). But more importantly, a
GPS-based analysis of intraplate deformation in Central
Europe (Grenerczy et al. 2000; Grenerczy 2002) suggests
that the territory of the Eastern Alps is being actively dis-
placed eastward at the rate of ~ 1.3 mm/yr, as the N- to
NW-moving Adriatic microplate collides with Europe
(Fig. 1). Due to the sparse network of GPS sites available
to those studies, the exact position and character of the
southern boundary of the extruding Alpine-North Pannon-
ian block remained unclear. Is it the PAF, as assumed by
Grenerczy (2002), or is the deformation distributed in a
wider belt? For example, the distribution of seismic activi-
ty in Slovenia, in terms of both magnitude and frequency
of events (Poljak et al. 2000), suggests that the tectonical-
ly most active area is located in the Dinarides, closer to
the rigid Adria. Could then a significant part of the Adria-

VRABEC, PAVLOVČIČ PREŠEREN and STOPAR

Europe convergence be absorbed in the PAF zone, or
close to it, as implied by wide-aperture GPS data? To ad-
dress these questions, we present the analysis of GPS-de-
rived 6 year (1996—2002) displacements of 9 stations in
northeastern Slovenia, spanning across the Sava, Šoštanj,
Periadriatic and Labot faults of the Periadriatic fault system.

Structure and evolution of the PAF system in

Slovenia

The Periadriatic fault is a major, WNW-ESE- to E-W

trending structural and topographical boundary, running
along the Slovenian-Austrian border (Figs. 2, 3). The fault
zone is up to several km wide and is organized into elon-
gated fault-parallel shear lenses and strike-slip duplexes,
which contain highly-deformed Paleozoic, Mesozoic and
Tertiary rocks (Fig. 3). The fabric of the synkinematic to-
nalite intrusion of Oligocene age emplaced along the fault
zone suggests initial coaxial N-S shortening (von Gosen
1989), but all subsequent brittle deformation is dextral to
reverse-dextral (von Gosen 1989; Polinski & Eisbacher
1992; Fodor et al. 1998). Displaced paleogeographical
markers of Paleozoic—Mesozoic (Kázmér & Kovács 1985)
and Tertiary (Fodor et al. 1998) age imply 300—500 km of
dextral separation along the PAF. However, a realistic esti-

Fig. 1. Active deformation in the eastern part of the Adria-Europe
collision zone, inferred from wide-aperture GPS data (Grenerczy
2002). Velocities shown are relative to the stable Eurasia reference
frame. The Eastern Alpine—North Pannonian unit (shaded) is moving
eastward with respect to both the Adriatic microplate and Europe,
and the Periadriatic fault system is the presumed southern boundary
of the extruding unit. White squares with dots are the GPS stations
used in the analysis of Grenerczy (2002); several additional stations
outside the area covered by this map were also used. Permanent IGS
network station GRAZ, shown also in Fig. 4, is marked for reference.
White dots – GPS network analysed in this study.

Fig. 2. Simplified tectonic map of the easternmost outcropping part of the Periadriatic fault system, Slovenia and Austria (see Fig. 1 for
location). Tertiary-Quaternary basins are shown in lighter colour. White dots – positions of GPS stations analysed in this study. IF – Idri-
ja fault, HF – Hochstuhl fault, LF – Labot (Lavanttal) fault, NKFS – Northern Karavanke flower structure, PAF – Periadriatic
fault, SF – Sava fault, SKSZ – Southern Karavanke shear zone, ŠF – Šoštanj fault, VB – Velenje Basin, ŽF – Žužemberk fault.

GPS STUDY OF THE DEFORMATION ALONG THE PERIADRIATIC FAULT IN SLOVENIA

mate of the actual dextral slip on the PAF is around
100 km (e.g. the kinematic reconstruction of Frisch et al.
1998). The rest of the present-day separation was pro-
duced by extreme extensional stretching of the extruding
northern block. The easternmost outcrop of the Periadriat-
ic fault zone is overlain by 17 Ma old Miocene sediments
(Fig. 3), indicating that major strike-slip movements on
the main fault terminated by then. Displacement was then
transferred southward to the adjacent Southern Kara-
vanke shear zone

(Fig. 3). Disrupted and tilted Neogene

sediments inside the zone imply that the deformation last-
ed until after mid-Miocene, but ceased by Pliocene as the
Pliocene sediments covering the faults in the zone are un-
affected (Fodor et al. 1998). The post-mid-Miocene slip
alone might be significant, since the stratigraphic succes-
sions of Eocene to mid-Miocene sediments differ marked-
ly across the shear zone (Jelen et al. 1992).

A major change in regional tectonics occurred at

around the Miocene-Pliocene transition, when termina-
tion of subduction in the Carpathians blocked further
eastward extrusion of the Eastern Alpine domain (Hor-
váth & Cloething 1996). Perhaps triggered by this event,
the Adria microplate might have started rotating in a
counterclockwise sense at about the same time (Márton
et al. 2002, 2003). The changed boundary conditions
have profoundly affected the architecture and organiza-

tion of the Slovenian PAF system, which is not a straight
strike-slip corridor in its present configuration, but com-
prises of two major parallel to subparallel faults with sev-
eral tens of kilometers of known displacement, the PAF
and the Sava fault, and of younger oblique faults with
smaller displacements, which either branch off the main
PAF zone or dextrally displace it (Fig. 2).

Both the main PAF fault branch and the Southern Kara-

vanke shear zone are dextrally offset for 10—14 km by the
NW—SE trending Labot (Lavanttal) fault (Figs. 2, 3). The
majority of dextral displacement is believed to have oc-
curred in Pliocene—Quaternary (Kázmér et al. 1996); dextral
activity clearly postdates the mid-Miocene—Pliocene time
bracket of the latest activity of the Southern Karavanke
shear zone which the Labot fault cross-cuts (Fodor et al.
1998). Šoštanj fault, which forks off the PAF zone and joins
with the Labot fault (Figs. 2, 3) might have acted as a by-
pass fault, transferring dextral deformation from the western
PAF branch eastward. Pliocene activity of the Šoštanj fault
is evidenced by formation of the transtensional Velenje Ba-
sin situated at the fault (Fig. 3). The basin contains an up to
1000 m thick succession of Pliocene-Quaternary sediments,
which thickens considerably towards the Šoštanj fault
(Brezigar 1986). Well-documented post-depositional dex-
tral deformation of the basin fill in the fault zone (Vrabec
1999), as well as a prominent geomorphological expression

Fig. 3. Detailed structural map of the Periadriatic fault and associated structures in northeastern Slovenia (see Fig. 2 for location), com-
piled and extended from structural interpretation maps of Fodor et al. (1998) which were based on Buser (1978), Mioč & Žnidarčič
(1977, 1983), and Premru (1983). Fault kinematics (where shown) are geological time-scale kinematics inferred from map relationships
and fault-slip data. Big white dots show locations of GPS stations. SKSZ – Southern Karavanke shear zone.

VRABEC, PAVLOVČIČ PREŠEREN and STOPAR

of the fault trace and possible offsetting of the fluvial net-
work (Fodor et al. 1998), prove continuation of fault activi-
ty into Quaternary and possibly Recent times.

Most of the post-Miocene deformation in the PAF sys-

tem was probably taken by the Sava fault, a regional-scale
structure subparallel to the Periadriatic fault, which ex-
tends from the eastern Italy across northern Slovenia and
eventually merges with the Šoštanj fault and Labot fault
in the east (Figs. 2, 3). Dextral kinematics of the fault are
obvious from the architecture of the fault zone, fault-slip
data (Fodor et al. 1998; Vrabec 2001), and from separation
of a distinct Oligocene volcanogenic formation, which in-
dicates 30—60 km of dextral displacement (Hinterlechner-
Ravnik & Pleničar 1967; Placer 1996a,b). Deformed
Neogene sediments inside the fault zone only roughly
constrain timing of activity to post-mid-Miocene times
(12 Ma), but the association of the fault with Pliocene-
Quaternary basins, as well as indications for deformation
of Quaternary sediments along the fault zone (Vrabec
2001), impressive topographic expression of the fault, and
occasional seismicity, indicate Pliocene-Quaternary and
probably recent activity.

Additionally, part of the deformation was absorbed in

the lenticular area between the Sava and Periadriatic
faults (Fig. 2). Structural analysis, fault-slip data and
paleomagnetic declination measurements in Tertiary
rocks demonstrate complex internal deformation within
this “shear lens”, involving clockwise domino-block
rotations (Fodor et al. 1998). Transpressional deforma-
tion and high topography prevail in the western part of
the lens, whereas the eastern part was probably de-
formed transtensionaly, as suggested by the occurrence
of small fault-bounded Pliocene-Quaternary basins
(Fig. 3) and by fault-slip data (Fodor et al. 1998). North
of the Periadriatic fault, the Northern Karavanke Mts
form a dextral transpressive flower structure, which is
thrusted over the Pliocene-Quaternary sediments of the
foreland Klagenfurt Basin for several kilometers (Fig. 2;
Polinski & Eisbacher 1991; Nemes et al. 1997). This
implies that some strike-slip deformation along the
main PAF branch persisted well into the Quaternary.
Thrust structures of the Northern Karavanke unit extend
into the study area (Fig. 3).

Part of the displacement from the Sava fault might also

have been transferred southward during the Quaternary
(Fodor et al. 1998; Vrabec 2001; Vrabec & Fodor 2006).
The rectangular Gorenjska Basin of Quaternary age could
have formed due to extension in a releasing step between
the Sava fault and the Žužemberk fault system (Figs. 2, 4).
East of the Gorenjska Basin, the E-W trending foldbelt
named the Sava Folds (Figs. 2, 4) formed in Pliocene—Qua-
ternary times (Placer 1999; Tomljenović & Csontos 2001).
The structure of the Sava Folds region suggests it could
have formed by transpressive shortening, perhaps absorbing
a part of the Sava fault dextral movement by deformation
partitioning (Vrabec 2001). Geomorphic indicators, like
high relief, elevated Pliocene-Quaternary gravel terraces,
and an antecedent fluvial network (Placer 1999) imply
young, possibly active uplift of the Sava Folds region.

The Slovenian part of the PAF system displays only

modest seismicity, at least compared to the areas south of
its southern boundary, the Sava fault (e.g. Poljak et al.
2000). Many of the instrumentally recorded earthquakes
are distributed along or close to the faults of the PAF sys-
tem. Events are generally too weak to allow reliable deter-
mination of focal mechanisms, but the few published
solutions favour nearly horizontal dextral strike-slip mo-
tions on the Sava and Labot faults (Reinecker & Lenhardt
1999; Poljak et al. 2000).

GPS data acquisition and processing

Network setup

A network of 9 GPS stations was established in 1995—96

with the primary goal of providing a high-precision ex-
ternal coordinate frame for monitoring mining-induced
subsidence in the Velenje coal mine area, but also with
potential usefulness for geodynamic applications in
mind, as the network traverses all the major faults of the
PAF system: Sava fault, Šoštanj fault, Periadriatic fault
and Labot fault (Fig. 3). The stations are marked either
by metal pins driven into bedrock (sites KMNK, PONI,
LUCE, URGO and MRZL), or are established on ~ 1 m
tall concrete pillars seated in bedrock (sites JERI, SKOR,
LUBE and VEKO). Stations JERI, SKOR and LUBE are
located close to the Velenje Basin, where underground
mining of the Pliocene coal seam is causing surface sub-
sidence and horizontal deformation of up to several cm/yr.
Decade-long observations of the Velenje Basin local
GPS network (not presented here) indicate that those 3 sta-
tions, seated in competent pre-Pliocene bedrock, are not
influenced.

Measurement campaigns

The stations were occupied in 3 campaigns, performed

on 10—12 July 1996, 1—3 September 1999, and 3—5 Sep-
tember 2002. In each campaign, all stations were occu-
pied simultaneously. Measurements were done in two
24-hour sessions with a sampling interval of 15 s (1996)
or 30 s (1999 and 2002). Data were collected by dual-fre-
quency receivers Trimble 4000SSE and 4000Ssi with
Trimble Compact L1/L2 (with ground plane) and Trim-
ble 4000SST external geodetic GPS antennas. To ensure
maximum consistency of observations, the same receiv-
ers and antennas were used at each network station in all
three campaigns. To minimize periodic influences, the
campaigns were performed at the same time of the year.

Data processing

GPS observations were processed using the Bernese GPS

Software, version 4.2 (Hugentobler et al. 2001) in ITRF2000
reference frame. International GPS Service (IGS) precise orbits
were used and five IGS reference stations, Wettzell, Zimmer-
wald, Graz, Medicina and Penc, which have known precise

GPS STUDY OF THE DEFORMATION ALONG THE PERIADRIATIC FAULT IN SLOVENIA

coordinates and velocities in ITRF2000, were used to fix the
network and the reference frame.

GPS observations for the IGS reference stations were

obtained from Scripps Orbit and Permanent Array Center
at the University of California, San Diego, USA (ftp://
garner.ucsd.edu/pub/). Information about the receivers
and antenna types used at the IGS stations was obtained
from the IGS Central Bureu (http://igscb.jpl.nasa.gov/
network/list.html). Coordinates of the IGS stations and
their velocities in ITRF2000 reference frame were taken
from the ITRF web site (ftp://lareg.ensg.ign.fr/pub/itrf/
itrf2000/). Precise ephemeris of the GPS satellites and
Earth rotation parameters were obtained from the GPS-In-
formations-und-Beobachtungssystem (GIBS) at the
Bundesamt für Geodaesie und Kartographie, Germany
(http://gibs.leipzig.ifag.de/). Those parameters, available
only in ITRF94 for 1996 and in ITRF97 for 1999, were
transformed to a common ITRF2000 reference frame with
the computer program TRNFSP3N of J. Kouba, obtained
at National Geodetic Survey, USA (http://www.ngs.noaa.gov/
gps-toolbox/trnfsp3.htm). Phase center corrections were
accounted for by using IGS phase center calibrations.
Station coordinates were then recalculated to the epochs
of the individual campaigns (1996.53, 1999.67 and
2002.67).

GPS phase observations were processed in a differen-

tial mode, that is on the basis of the carrier-phase differ-
ences. For each day of observation 13 independent
baseline vectors were computed. After the construction
of single differences, double differences of the iono-
sphere-free linear carrier-phase combination L3 were pro-
cessed. For short baselines, the L1&L2 ambiguities were
solved using a SIGMA-dependent strategy, and a QIF
(Quasi-Ionosphere-Free) algorithm was used for long
baselines. We applied elevation dependent weights of
observations (model cosz). Processing was performed
with the Saastamoinen troposphere model, with one site
troposphere parameter for two hours of observations.

The initial results of the GPS data processing were daily

solutions for baseline vectors, and daily positions of all
network points. For consistency check, point coordinates
computed for each day of observation were transformed
with the Helmert transformation. Normal equations for dai-
ly solutions of all 3 campaigns were then combined to
make the final solution, with IGS stations fixed in the
ITRF2000 reference frame. The results are coordinates of
network points in ITRF2000 for each campaign epoch.

Site velocities and velocity error estimations

Velocity components of network stations were estimat-

ed from computed positions at each campaign epoch, fol-
lowing the strategy of Bernese GPS Software (Hugentobler
et al. 2001, p. 287). For apriori velocities of our network
sites we used the velocities computed according to the
NUVEL-1A NNR model (McCarthy 1996), where veloci-
ties of the IGS reference stations in ITRF2000 were held
fixed. For campaign measurements, we estimated the long-
term station velocities by assuming constant displacement
rates between the campaigns. Velocity components ac-
cording to ITRF2000 with their respective formal errors
are listed in Table 1.

The formal errors of site velocities determined from re-

peated GPS campaigns are generally estimated on the as-
sumption that measurement noise is uncorrelated in time
( » white « noise). However, if time correlated noise ( » co-
loured« noise) is present, the true velocity noise is un-
derestimated with the scale factor of 2 to 11 (Mao et al.
1997; Dixon et al. 2000). Since our three campaigns were
performed in nearly the same period of the year (1996.53,
1999.67 and 2002.67), we assumed that no seasonal sig-
nal in computed velocities exists, but we nevertheless in-
flated our estimated formal errors of velocity vectors
with a pessimistic scale factor of 10.

Computed horizontal velocities of network points

were then transformed from ITRF2000 to the stable Eur-

Table 1: Velocities computed for the 1996—2002 period, with RMS (root mean square error, 1

σ). Initial processing results in ITRF2000

with formal RMS are given in the left half of the table. Velocities relative to stable Eurasia with scaled-up RMS (accounting for refer-
ence frame transformation and possible presence of coloured noise) are presented in the right half.

GPS STUDY OF THE DEFORMATION ALONG THE PERIADRIATIC FAULT IN SLOVENIA

The station VEKO from our network and the IGS per-

manent station GRAZ, both belonging to the Eastern Al-
pine tectonic unit, show an obvious, statistically
significant difference in velocity compared to the other
stations of our network (Fig. 4). The high-strain bound-
ary between the two domains matches the position of the
Labot fault. The relative velocities of stations across the
fault suggest ~ 1 mm/yr of dextral movement, which is
consistent with the slip sense postulated from focal
mechanisms of earthquakes (Reinecker & Lenhardt
1999). While similar in magnitude, the diverging veloci-
ty vectors of sites VEKO and GRAZ do not suggest that
the Eastern Alpine unit behaves like a coherent, east-
ward-moving block. This is not surprising since the kine-
matic model of Miocene extrusion, based on palinspastic
restoration (Frisch et al. 1998), demands considerable in-
ternal deformation of the extruding unit, which was
achieved predominatly by translation and rotation of in-
ternal fault-bounded blocks. Miocene counterclockwise
rotations of such blocks were also demonstrated by pale-
omagnetic data (Márton et al. 2000). Diverging site ve-
locities, as well as the occurrence of earthquakes along
the NNW—SSE trending faults north of the PAF and only
minor seismicity in the areas between them (Reinecker &
Lenhardt 1999), could suggest that a similar domino-
block tectonic mechanism is active today. A denser net-
work of GPS stations within the Eastern Alpine unit will
be needed to investigate this possibility.

Another high-strain zone, likely corresponding to the

Sava fault, can be inferred in the southern part of the study
area (Fig. 4). The relative velocities of stations KMNK and
LUCE indicate ~ 1 .2 mm/yr of dextral displacement along
the fault. However, the area south of the Sava fault appar-
ently does not behave as a rigid block, since the stations
KMNK and MRZL diverge significantly. The station
KMNK is located at the margin of the Quaternary Gorenjs-
ka Basin, where N-S trending normal faults were observed,
thus the movement of site KMNK away from MRZL could
indicate active extension in the Gorenjska Basin. The site
MRZL, located in the Sava Folds, does not show statisti-
cally significant movement with respect to Eurasia, and no
fault-parallel movement relative to stations PONI and
JERI north of the Sava fault. One possible explanation
could be slip absorbtion by active shortening and uplift of
the Sava Folds area, producing predominantly vertical
movement of the MRZL site. Due to the inherently large
uncertainty of determining vertical movements with GPS
(e.g. Hugentobler et al. 2001), we could not check this hy-
pothesis at the present stage of our study, since the mea-
sured vertical movement rates are still way below
statistical significance.

The station URGO, located in the Northern Karavanke

Mts north of the PAF (Fig. 4), moves northward more than
1 mm/yr with respect to stable Eurasia. This could reflect
ongoing thrusting of the Northern Karavanke range onto
its foreland. The Quaternary age of transpressive thrusting
is clearly documented west of the study area in the
Klagenfurt Basin (e.g. Polinski & Eisbacher 1992). For the
Northern Karavanke Peca thrust, located close to the

URGO station, a geologically young age of thrusting was
also inferred from significantly tilted deposits of uncon-
solidated laminated clay, which fill karstic channels with-
in the Mesozoic beds of the thrusted unit (Placer 1996b).

The PAF segment between the stations URGO, LUCE,

JERI, LUBE and PONI does not seem to accomodate any
dextral slip (Fig. 4). However, only one of those stations
(URGO) is located north of the PAF, and if indeed situat-
ed on an active thrust of possibly transpressive character,
it could be following a complex displacement trajectory,
which could easily obscure mm-scale dextral movements
relative to the stations south of the PAF. At least one ad-
ditional station positioned in a stable foreland of the
Northern Karavanke thrust system would be needed to
address these concerns.

The stations LUCE, JERI and PONI, located within the

Sava-PAF shear lens, exhibit very uniform NNW-ward
movement of ~ 1 .4 mm/yr with respect to stable Eurasia.
Movement of those sites relative to the stations KMNK
and MRZL south of the Sava fault indicates active tran-
stension inside the eastern part of the shear lens. The
transtensional deformation is in agreement with structur-
al and geomorphic indicators (Fig. 3; Fodor et al. 1998),
presented in the introductory chapter. An additional sta-
tion from this area, SKOR, is not moving with respect to
stable Eurasia and is possibly an outlier due to unknown
non-tectonic influences.

Active deformation in the Šoštanj fault area is at

present not well understood. The sites LUBE, JERI and
PONI imply significant fault-perpendicular shortening of
~

1 .5 mm/yr, and only a very small (insignificant at the

present precision level) component of dextral shear
(Fig. 4). A detailed surface and subsurface structural
study of the Šoštanj fault zone, covering the area be-
tween stations SKOR and JERI (Vrabec et al. 1999), did
not reveal any indications for fault-perpendicular short-
ening. The station LUBE north of the fault does not
move with respect to stable Eurasia, but similarly to sta-
tion MRZL it might undergo vertical uplift, not present-
ly detectable with GPS because of the lack of vertical
precision. Significant local relief and high altitude (rela-
tive to surroundings) of the entire Southern Karavanke
shear zone region might indicate uplift of this area. Per-
haps, as speculated for the PAF and Northern Karavanke
thrust, the deformation in the Šoštanj fault—Southern
Karavanke shear zone area is transpressive and the dex-
tral component is too small to be detectable geodetically
within the 6-year timespan. But alternatively, like the
similarly stationary site SKOR, the LUBE station could
be an outlier too, especially since neither of those sites is
monumented directly into the bedrock, but are instead
positioned on concrete pillars.

Conclusions

While the results of our GPS data analysis after the first

6 years of observation are still close to or within the
uncertainty limits, they demonstrate that measurable dis-

VRABEC, PAVLOVČIČ PREŠEREN and STOPAR

placements exist in the Slovenian PAF system. The de-
rived velocity field and inferred modes of ongoing defor-
mation are for the most part consistent with geological
knowledge of the area and with the published earthquake
focal mechanisms. We summarize the inferred displace-
ment rates, modes of deformation, and possible mecha-
nisms of displacement transfer in Fig. 5.

Our results confirm that the active dextral deformation

accross the PAF system could account for the difference
in kinematics between the Adria microplate and the ex-
truding Eastern Alpine—North Pannonian unit, deter-
mined from the wide-aperture GPS data (Grenerczy
2002). However, deformation in the PAF system of north-
eastern Slovenia is complex and obviously not tied to a
single strike-slip corridor. We could resolve ~ 1 .2 mm/yr
of dextral movement on the Sava fault which parallels
the PAF, but no displacement on the fault was detected
in the eastern part of the study area, which implies trans-
fer of deformation to other structures. In our preliminary
analysis the movements of stations south of the Sava
fault are seemingly consistent with the established model
for Pliocene-Quaternary displacement transfer southward
via the pull-apart Gorenjska Basin, and/or absorbtion of
dextral slip by transpressive shortening and uplift of the
Sava Folds south of the restraining bend of the Sava fault
(Fodor et al. 1998; Vrabec 2001; Vrabec & Fodor 2006).
No dextral displacement was detected along the PAF
zone, but there is indication for active ~ 1 mm/yr north-
ward propagation of the Northern Karavanke thrust sys-
tem situated just north of the PAF. This could indicate
transpressive deformation with an at present undetect-

able dextral component along the PAF and its branching
fault, the Šoštanj fault, where also significant fault-per-
pendicular shortening is implied. The region between the
Sava fault and the PAF apparently undergoes transten-
sion, which is also indicated by structural and geomor-
phic features of that area. A detectable dextral slip of
~

1 mm/yr was found along the Labot fault. We speculate

that this slip could reflect the domino-block deformation
mechanism inside the extruding Eastern Alpine unit.

To better constrain site velocities in terms of reduced

uncertainty, we plan to continue re-occupying the net-
work in 2-year intervals. Additionally, the recent public
concern about increased seismicity in the Velenje Basin
stimulated expansion of the network with 9 more stations
that were stabilized and measured for the first time in
Summer 2003. The new stations were primarily selected
to monitor the Šoštanj fault, but also reach over the PAF
and the Labot fault. The expanded network will hopeful-
ly provide better insight into the kinematics of the area
that is currently least understood. On the other hand, a
deeper understanding of the kinematics of regional de-
formation and testing the validity of our findings would
require a wider network of reasonably spaced GPS sites,
covering the length of the PAF system and entering the
surrounding areas. We initiated such a project in 2003 in
the framework of the PIVO (Periadriatic fault—Istria Ve-
locity Observations) experiment, which included 36 sta-
tions distributed in the territory of Slovenia and northern
Croatia. While the data analysis is still in progress, the
first preliminary results (Weber et al. 2006) agree with
conclusions about the kinematics of the PAF system pre-
sented in this paper.

Finally, it is hoped that other workers will provide data

which would further test and elaborate the ongoing ex-
trusion model. One area of particular interest is the sinis-
tral northern boundary of the extruding wedge. The other
major challenge would be studying internal deformation
of the extruding unit and investigating how extrusion is
compensated in front of the extruding wedge.

Acknowledgments:

The research was financially sup-

ported by the Velenje Coal Mine Company. In particular,
we wish to thank D. Potočnik and M. Koželj for support
and participation in measurement campaigns. We also
gratefully acknowledge the useful advice and stimula-
tion given by John Weber when we started tectonic inter-
pretation of our GPS data.

References

Altamimi Z., Sillard P. & Boucher C. 2002: ITRF2000: A new re-

lease of the International Terrestrial Reference Frame for earth
science applications. J. Geophys. Res. 107, B10, 2214—2232.

Brezigar A. 1986: Coal seam of the Velenje coal mine. Geologija

28, 29, 319—336 (in Slovenian with English abstract).

Buser S. 1978: Geological map of SFR Yugoslavia 1 : 100,000,

sheet Celje. Federal Geol. Surv., Belgrade.

Dixon. T.H., Miller M.M., Farina F., Wang H. & Johnson D. 2000:

Present day motion of the Sierra Nevada block and some tec-

Fig. 5. Fault slip rates and possible modes of ongoing deforma-
tion, inferred from our GPS data. See text for discussion. Tectonic
model based on Vrabec & Fodor (in press). Depicted Adria move-
ment and net extrusion rate according to Grenerczy (2002).

GPS STUDY OF THE DEFORMATION ALONG THE PERIADRIATIC FAULT IN SLOVENIA

tonic implications for the basin and range province: North
American Cordillera. Tectonics 19, 1—24.

Fodor L., Jelen B., Márton E., Skaberne D., Čar J. & Vrabec M.

1998: Miocene-Pliocene tectonic evolution of the Slovenian
Periadriatic Line and surrounding area – implication for Al-
pine-Carpathian extrusion models. Tectonics 17, 690—709.

Fodor L., Csontos L., Bada G., Györfi I. & Benkovics L. 1999:

Tertiary tectonic evolution of the Pannonian basin system and
neighbouring orogens: a new synthesis of paleostress data. In:
Durand B., Jolivet L., Horváth F. & Séranne M. (Eds.): The
Mediterranean Basins: Tertiary extension within the Alpine
Orogen. Geol. Soc. Spec. Publ. 156, 295—334.

Frisch W., Kuhlemann J., Dunkl I. & Brügel A. 1998: Palinspastic re-

construction and topographic evolution of the Eastern Alps dur-
ing late Tertiary tectonic extrusion. Tectonophysics 297, 1—15.

von Gosen W. 1989: Fabric developments and the evolution of the

Periadriatic Lineament in Southeast Austria. Geol. Mag. 126,
55—71.

Grenerczy Gy. 2002: Tectonic processes in the Eurasian-African

plate boundary zone revealed by space geodesy. In: Stein S. &
Freymueller J.T. (Eds.): Plate boundary zones. AGU Geody-
namics Ser. Vol. 30, 67—86.

Grenerczy Gy., Kenyeres A. & Fejes I. 2000: Present crustal move-

ment and strain distribution in Central Europe inferred from
GPS measurements. J. Geophys. Res. 105, B9, 21835—21846.

Hinterlechner-Ravnik A. & Pleničar M. 1967: The andesite of

Smrekovec and its tuff. Geologija 10, 219—237 (in Slovenian).

Horváth F. & Cloething S. 1996: Stress-induced late-stage subsid-

ence anomalies in the Pannonian Basin. Tectonophysics 266,
287—300.

Hugentobler U., Schaer S. & Fridez P. 2001: Bernese GPS Software.

Version 4.2. Astronomical Institute of the University of Berne.

Jelen B., Aničić B., Brezigar A., Buser S., Cimerman F., Drobne K.,

Monostori M., Kedves M., Pavšič J. & Skaberne D. 1992:
Model of positional relationship for upper Paleogene and Mi-
ocene strata in Slovenia. In: Montanari A., Coccioni R. & Odin
G.S. (Eds.): International Geological Conference on Miocene
Epoch. Abstracts and field trips, Ancona, Italy, 71—72.

Kázmér M. & Kovács S. 1985: Permian-Paleogene paleogeography

along the Eastern part of the Insubric-Periadriatic Lineament
system: Evidence for continental escape of the Bakony-Drau-
zug unit. Acta Geol. Hung. 28, 71—84.

Kázmér M., Fodor L., Józsa S., Jelen B., Herlec U. & Kuhlemann J.

1996: Late Miocene paleogeography of Slovenia and the
Southern Alps: a palinspastic approach. In: Amann G., Han-
dler R., Kurz W. & Steyrer H.P. (Eds.): 6

Symposium on

Tektonik, Strukturgeologie, Kristalingeologie. Extended ab-
stracts, Salzburg, Austria, 212—214.

Lőrincz K.D., Horváth F. & Detzky G. 2002: Neotectonics and its rela-

tion to the Mid-Hungarian Mobile Belt. In: Cloething S.A.P.L.,
Horváth F., Bada G. & Lankreijer A.C. (Eds.): Neotectonics and
surface processes: The Pannonian basin and Alpine/Carpathian
system. EGU Stephan Mueller Spec. Publ. 3, 247—266.

Mao A., Harrison G.A. & Dixon T.H. 1997: Noise in GPS coordi-

nate time series. J. Geophys. Res. 104, B2, 2797—2816.

Márton E., Kuhlemann J., Frisch W. & Dunkl I. 2000: Miocene ro-

tations in the Eastern Alps – palaeomagnetic results from in-
tramontane basin sediments. Tectonophysics 323, 163—182.

Márton E., Fodor L., Jelen B., Márton P., Rifelj H. & Kevrić R.

2002: Miocene to Quaternary deformation in NE Slovenia:
complex paleomagnetic and structural study. J. Geodynamics
34, 627—651.

Márton E., Ćosović V., Drobne K. & Moro A. 2003: Palaeomag-

netic evidence for Tertiary counterclockwise rotation of Adria.
Tectonophysics 377, 143—156.

McCarthy D. 1996: IERS conventions (1996). IERS Technical Note

21, Observatoire de Paris, Paris.

Mioč P. & Žnidarčič M. 1977: Geological map of SFR Yugoslavia

1:100,000, sheet Slovenj Gradec. Federal Geol. Surv., Belgrade.

Mioč P. & Žnidarčič M. 1983: Geological map of SFR Yugoslavia

1:100.000, sheet Ravne. Federal Geol. Surv., Belgrade.

Nemes F., Neubauer F., Cloething S. & Genser J. 1997: The

Klagenfurt basin in the Eastern Alps: an intra-orogenic decou-
pled flexural basin? Tectonophysics 282, 189—203.

Placer L. 1996a: Displacement along the Sava fault. Geologija 39,

283—287.

Placer L. 1996b: Peca thrust at the Periadriatic lineament. Geologija

39, 289—302.

Placer L. 1999: Structural meaning of the Sava folds. Geologija 41,

191—221.

Polinski R.K. & Eisbacher G.H. 1992: Deformation partitioning

during polyphase oblique convergence in the Karawanken
Mountains, southeastern Alps. J. Struct. Geol. 14, 1203—1213.

Poljak M., Živčić M. & Zupančič P. 2000: The seismotectonic char-

acteristics of Slovenia. Pure Appl. Geophysics 157, 37—55.

Premru U. 1983: Geological map of SFR Yugoslavia 1 : 100,000,

sheet Ljubljana. Federal Geol. Surv., Belgrade.

Reinecker J. & Lenhardt W.A. 1999: Present-day stress field and

deformation in eastern Austria. Int. J. Earth Sci. 88, 532—550.

Ratschbacher L., Frisch W., Linzer H.G. & Merle O. 1991: Lateral

extrusion in the Eastern Alps. Part 2. Structural analysis. Tec-
tonics 10, 257—271.

Schmid S.M., Aebli H.R., Heller F. & Zingg A. 1989: The role of

the Periadriatic line in the tectonic evolution of the Alps. In:
Coward M.P., Dietrich D. & Park R.G. (Eds.): Alpine Tecton-
ics. Geol. Soc. Spec. Publ. 45, 153—171.

Tomljenović B. & Csontos L. 2001: Neogene-Quaternary structures

in the border zone between Alps, Dinarides and Pannonian ba-
sin (Hrvatsko zagorje and Karlovac basin, Croatia). Int. J.
Earth Sci. 90, 560—578.

Viola G., Mancktelow N. & Seward D. 2001: Late Oligocene-Neo-

gene evolution of Europe-Adria collision: New structural and
geochronological evidence from the Giudicarie fault system
(Italian Eastern Alps). Tectonics 20, 999—1020.

Vrabec M. 1999: Style of postsedimentary deformation in the Plio-

Quaternary Velenje basin, Slovenia. Neu. Jb. Geol. Paläont.
Mh. 8, 449—463.

Vrabec M. 2001: Structural analysis of the Sava fault zone between

Trstenik and Stahovica (in Slovenian with English abstract).
Unpublished PhD thesis, University of Ljubljana, 1—94.

Vrabec M. & Fodor L. 2006: Late Cenozoic tectonics of Slovenia:

Structural styles at the northeastern corner of the Adriatic mi-
croplate. In: Pinter N., Grenerczy Gy., Weber J., Stein S. &
Medak D. (Eds.): The Adria microplate: GPS geodesy, tec-
tonics, and hazards. Springer, 151—168.

Vrabec M., Čar J. & Veber I. 1999: Kinematics of Šoštanj fault in

the Velenje basin area: Insights from subsurface data and pa-
leostress analysis. RMZ Materiali in Geookolje 46, 623—634.

Weber J., Vrabec M., Stopar B., Pavlovčič Prešeren P. & Dixon T.

2006: The PIVO-2003 experiment: A GPS study of Istria
peninsula and Adria microplate motion, and active tectonics
in Slovenia. In: Pinter N., Grenerczy Gy., Weber J., Stein S.
& Medak D. (Eds.): The Adria microplate: GPS geodesy, tec-
tonics, and hazards. Springer, 305—320.