GeolCarp_Vol66_No5_427

www.geologicacarpathica.com

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

, OCTOBER 2015, 66, 5, 427—438 doi: 10.1515/geoca-2015-0035

Evidence of a plate-wide tectonic pressure pulse provided by

extensometric monitoring in the Balkan Mountains (Bulgaria)

MILOŠ BRIESTENSKÝ

, MATT D. ROWBERRY

, JOSEF STEMBERK

, PETAR STEFANOV

JOZEF VOZÁR

, STANKA ŠEBELA

, PUBOMÍR PETRO

, PAVEL BELLA

, PUDOVÍT GAAL

and CHOLPONBEK ORMUKOV

Institute of Rock Structure and Mechanics, Academy of Sciences of the Czech Republic, v.v.i., V Holešovičkách 94/41, 182 09, Praha 8,

Czech Republic; briestensky@irsm.cas.cz; rowberry@irsm.cas.cz;

stemberk@irsm.cas.cz

National Institute of Geophysics, Geodesy, and Geography, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., Bl. 3, 1113 Sofia,

Bulgaria; psgeo@abv.bg

Earth Science Institute, Slovak Academy of Sciences, Dúbravská cesta 9, P.O. Box 106, 840 05 Bratislava 45, Slovak Republic;

jozef.vozar@savba.sk

Karst Research Institute ZRC SAZU, Titov trg 2, SI-6230 Postojna, Slovenia; sebela@zrc-sazu.si

State Geological Institute of Dionýz Štúr, Jesenského 8, 040 01 Košice, Slovak Republic; lubomir.petro@geology.sk

Slovak Caves Administration, Hodžova 11, 031 01 Liptovský Mikuláš, Slovak Republic; bella@ssj.sk; gaal@ssj.sk

Central Asian Institute for Applied Geosciences, Timur Frunze 73, 720027 Bishkek, Kyrgyz Republic; ch.ormukov@caiag.kg

(Manuscript received October 29, 2014; accepted in revised form June 23, 2015)

Abstract: The EU-TecNet monitoring network uses customized three-dimensional extensometers to record transient de-
formations across individual faults. This paper presents the first results from two newly established monitoring points in
the Balkan Mountains in Bulgaria. The data from Saeva Dupka, recorded across an EEN-WWS striking fault, show sinis-
tral strike-slip along the fault and subsidence of the southern block. Much of the subsidence occurred around the time of the
distal M

= 5.6 Pernik Earthquake. An important transient deformation event, which began in autumn 2012, was reflected

by significant compression and following extension, across the monitored fault. The data from Bacho Kiro, recorded
across a NE—SW striking fault, show sinistral strike-slip along the fault and subsidence of the north-western block. The
same important deformation event was reflected by changes in the strike-slip, dip-slip, and horizontal opening/closing
trends. These results have been compared to data from other monitoring points in the Western Carpathians, External
Dinarides, and Tian Shan. Many of the sites show evidence of simultaneous displacement anomalies and this observation
is interpreted as a reflection of the plate-wide propagation of a tectonic pressure pulse towards the end of 2012.

Key words: Eurasian Plate, Balkan Peninsula, active tectonics research, aseismic transient deformations, slow-slip
phenomena, pressure pulse, EU-TecNet.

Introduction

Until the turn of century, the geoscientific community tended
to accept the long standing assumption that stress accumula-
tion along faults was released by either continuous aseismic
sliding or earthquakes resulting from the instantaneous fail-
ure of locked faults (Peng & Gomberg 2010). This assump-
tion came to be challenged around twenty years ago after it
was  discovered  that  slow-slip  phenomena  were  far  more
common in the vicinity of tectonic plate boundary faults than
previously  thought.  Evidence  came  from  a  number  of  dis-
tinct  regions  including  southern  Japan  (Hirose  et  al.  1999)
and Cascadia (Dragert et al. 2001). It has come to fundamen-
tally change our understanding of the ways in which tectonic
stresses  arising  from  plate  motions  are  accommodated  by
slip  on  faults  (Gomberg  2010).  Nonetheless,  slow-slip  phe-
nomena  are  not  unique  to  the  depths  of  subduction  zone
plate  interfaces.  They  occur  on  faults  in  many  settings  and
appear to span a continuum so that, as a result, it is no longer
possible to characterize fault slip modes in aseismic or seis-
mic  terms  (Peng  &  Gomberg  2010).  These  phenomena  are
manifest  as  geodetically  observed  aseismic  transient  defor-

mations. While geoscientists most commonly use seismic
and geodetic monitoring as the basis with which to recognize
slow-slip phenomena, less abundant strainmeters and tiltme-
ters also measure aseismic transient deformations, and with
much greater resolution than GPS (Agnew 2009).

The use of extensometers as a means with which to recog-

nize  aseismic  transient  deformations  is  best  exemplified  by
the  fault  displacement  monitoring  network  EU-TecNet
(Stemberk et al. 2010; Košvák et al. 2011). It was established
at the turn of the century in order to accurately quantify the
displacements that occur across specific faults over decadal
timescales as it is simply not possible to obtain this informa-
tion from geodetic measurements such as GPS (Košvák et al.
2002; Briestenský et al. 2014a). The network now comprises
nearly one hundred and fifty monitoring points spread across
mainland Europe along with a smaller number of distal sites in
the Arctic, Africa, Asia, North America, and South America.
In this paper, the background section outlines a number of its
most  important  findings,  while  the  methods  section  illus-
trates both the procedure used to select suitable sites and the
technical parameters of the installed extensometers. However,
the main aim of the paper is to describe and present the first

428

BRIESTENSKÝ, ROWBERRY, STEMBERK, STEFANOV, VOZÁR, ŠEBELA, PETRO, BELLA, GAAL and ORMUKOV

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

two  years  of  results  from  two  of  the  most  recently  estab-
lished  monitoring  points.  These  are  located  in  the  caves  of
Saeva  Dupka  and  Bacho  Kiro  in  the  Balkan  Mountains  of
Bulgaria (Fig. 1). The presented results are then compared to
data  obtained  over  the  same  period  from  monitoring  points
in  the  Western  Carpathians,  the  External  Dinarides,  and  in
the Kyrgyz Ala-Too Range.

Background

The EU-TecNet monitoring network deploys permanently

installed extensometers that are able to measure relative fault
displacements in three dimensions along with the horizontal
and vertical angular rotations of the opposing fault sidewalls
(Marti  et  al.  2013).  These  extensometers  are  detailed  later
but the fact that they are able to record three dimensional dis-
placements is hugely advantageous given that movement be-
tween fault planes is generally characterized by horizontal or
vertical  slip  (Košvák  2006).  The  monitoring  points  are  usu-
ally located underground in settings such as mines or caves
as this helps to ensure that the recorded slow-slip phenomena
are the result of endogenic processes. Monitoring points on
the surface are more likely to be influenced by exogenic pro-
cesses including gravitationally induced slope deformations,
extreme rainfall events, prolonged periods of rainfall, or sea-
sonal temperature changes. The area around each new site is
mapped  to  ascertain  whether  the  obtained  results  could  be
influenced  by  deep-seated  slope  deformations  and,  if  so,
the  data  are  then  interpreted  appropriately  (Briestenský  et
al.  2011a).  No  evidence,  irrespective  of  whether  prolonged

periods of rainfall or extreme rainfall events are considered,
has ever been found to suggest that precipitation influences
the  data  from  underground  monitoring  points.  Furthermore,
seasonal  temperature  changes  may  influence  the  opening/
closing component across the fault, but this influence dimin-
ishes rapidly with depth (Gosar et al. 2009): the seasonal am-
plitude has been found to decrease from around 1 mm at the
surface  to  less  than  0.05 mm  at  depths  of  more  than  10 m
(Briestenský et al. 2010).

The displacements at any given monitoring point are typi-

cally  characterized  by  long  periods  of  geodynamic  stability
during which none of the three displacements move or dur-
ing which one or more of them is characterized by progres-
sive creep. These periods of stability are then interrupted by
shorter periods of anomalous or increased geodynamic activ-
ity during which one or more of the displacement components
will  be  affected  by,  for  example,  a  conspicuous  reversal  in
the progressive creep trend; a sudden enduring displacement;
or a series of oscillatory displacements. These shorter periods
of  geodynamic  activity  normally  last  for  no  more  than  six
months  before  resumption  of  the  progressive  creep.  How-
ever,  in  terms  of  aseismic  transient  deformations,  the  most
important finding from the network is represented by the rec-
ognition  of  pressure  pulses  (Stemberk  et  al.  2010).  This  is
the term given to a period of time in which a number of scat-
tered  monitoring  points  are  affected  by  anomalous  or  in-
creased  geodynamic  activity.  It  is  known  that  slow-slip
phenomena  are  especially  sensitive  to  stress  perturbations
(Peng & Gomberg 2010). Pressure pulses are therefore par-
ticularly  significant  from  a  tectonic  perspective  as  they  are
thought to reflect the widespread redistribution of stress and

Fig. 1.  The  location  of  the  recently  established
fault  displacement  monitoring  points  at  Saeva
Dupka  Cave  (N 43°2’48.67”,  E 24°11’9.17”)
and

Bacho

Kiro

Cave

(N 42°56’49.80”,

E 25°25’47.82”) in the Balkan Mountains of Bul-
garia.  The  cities  of  Pleven,  Plovdiv,  Ruse,  Sofia
and Shumen together with the Balkan, Pirin, Rila,
and  Rhodope  Mountain  Ranges  are  also  marked.
The  inset  presents  the  location  of  the  study  area
with respect to the rest of the Balkan Peninsula.

429

PLATE TECTONIC PRESSURE PULSE PROVIDED BY EXTENSOMETRY, BALKAN Mts (BULGARIA)

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

strain within the shallow crust (Košvák et al. 2007). Further-
more, the changes that characterize the pulse will vary from
one monitoring point to the next, depending on the location
of the source of the pulse in relation to the fault geometry
and orientation of the stress axes.

It is very difficult to constrain the relationship between

anomalous or increased geodynamic activity and significant
local or regional seismic events while sudden displacements
have been noted both before (Košvák et al. 1992; Briestenský
et al. 2007) and after earthquakes (Košvák et al. 1998; Dobrev
& Košvák 2000). Irrespective of whether the sudden displace-
ments are preseismic or postseismic, they are not thought to
be instantaneous mechanical movements generated by a spe-
cific earthquake event, and both phenomena are interpreted
as reflections of the redistribution of stress and strain in the
crust. Furthermore, recent investigations have shown that the
progressive creep recorded at a number of sites in Central
Europe changed around the same time as the M

= 9.3 Indian

Ocean Earthquake on 26 December 2004 (Stemberk et al.
2010) and the M

= 9.0 Tohoku Earthquake on 11 March

2011 (Briestenský et al. 2014b). The latter study found that
these changes also correlated with natural gas concentration
anomalies, thereby supporting the results of earlier research
(Briestenský et al. 2011b).

Sites

Saeva Dupka

The first study site, Saeva Dupka, 3 km to the south of the

village  of  Brestnitsa  in  Lovech  Province,  is  situated  on  the
southern  slope  of  the  small  Brestnitsa  Polje.  The  cave  has
developed  in  the  massive  light  Tithonian  limestones  of  the
Brestnitsa  Formation.  This  formation  has  been  subjected  to
intensive karstification and faults in the vicinity of the cave
are often expressed morphologically on the surface (Shanov
& Kostov 2015). The entrance is found at 510 m a.s.l., its to-
tal  length  is  205 m,  and  its  denivelation  is  40 m.  Inside  the
cave, considered to be one of the most beautifully decorated
in  the  country,  one  large  gallery  separates  five  chambers.
The  inner  temperature  fluctuates  between  7—11 °C  and  its
relative humidity ranges from 90—98 %. This cave is of par-
ticular seismotectonic significance as it hosts a number of in-
dicative features such as cave breccia, fallen stalactites, and
dislocated or cracked stalactones. The cave breccia has been
interpreted  as  a  result  of  a  catastrophic  paleo-earthquake
while  the  fallen  stalactites  have  been  interpreted  as  reflec-
tions of reactivation of the fault to the south of the cave dur-
ing the Pleistocene (Shanov & Kostov 2015).

Bacho Kiro

The second site, Bacho Kiro, 5 km to the west of the town

of Dryanovo in Gabrovo Province, is situated in the valley of
the  Andaka  River.  The  cave  has  developed  mainly  in  the
Barremian  bioclastic  and  organogenic  limestones  of  the
Emen  Formation  and  the  Barremian  to  Aptian  sandstones,
marls,  siltstones,  limestones  of  the  Bulgarene  Formation.

These  formations  have  again  been  subjected  to  intensive
karstification,  with  at  least  thirty-five  caves  known  in  the
area  around  Dryanovo  Monastery,  the  longest  of  which  is
Andaka Cave at 5000 m (Suchkov & Sinnyovsky 2010). The
entrance to the cave is found at 335 m a.s.l., its total length is
3600 m, and its denivelation is 65 m. Inside the cave, a com-
plex labyrinth of corridors, spread over four levels, connects
twelve  chambers.  The  inner  temperature  is  stable  at  13 °C
and its relative humidity is constant at 95%. It hosts a num-
ber  of  karst  features  including  stalagmites,  stalactites,  and
flowstones.  This  cave  is  of  particular  archeological  signifi-
cance as it has been found to have hosted one of the earliest
known  Aurignacian  burials  while  further  excavations  have
revealed  some  of  the  earliest  human  remains  found  in  Bul-
garia (Beron et al. 2006).

Methods

In 2012, two potential underground observatories for fault

displacement  monitoring  were  surveyed  in  detail,  at  Saeva
Dupka  and  Bacho  Kiro  in  the  central  part  of  the  Balkan
Mountains.  The  applied  methodology  follows  that  previ-
ously  outlined  by,  for  example,  Briestenský  et  al.  (2014a).
Each reconnaissance survey was conducted with the aim of
identifying significant tectonic structures which have played,
or continue to play, an important role in the development of
the specific cave system. These surveys were supplemented
by further investigations in the vicinity of the significant tec-
tonic structures, focusing specifically on the analyses of slick-
ensides and damaged speleothems. Damaged speleothems are
particularly significant as these commonly reflect recent tec-
tonic  movements  (Gilli  2005)  although,  of  course,  it  is  im-
portant to consider all the mechanisms that may lead to this
phenomenon  (Becker  el  al.  2006).  Moreover,  any  evidence
of new speleothem growth was also recorded, as tectonic ac-
tivity helps mineralized water to penetrate along faults prior
to  the  precipitation  of  its  calcium  carbonate.  Suitable  faults
for displacement monitoring were identified at both sites, the
results of which provide evidence for the active tectonic re-
gime.  The  selected  monitoring  point  at  Saeva  Dupka  lies
across a fault striking EEN-WWS with a 170/80 (dip direc-
tion°/dip°)  while  the  selected  monitoring  point  at  Bacho
Kiro  lies  across  a  fault  striking  NE-SW  with  a  310/70.  In
both instances, the strike orientation represents a fundamen-
tal control on the development of the specific cave system.

The faults were each instrumented with a precise extenso-

meter  called  a  TM-71  (Košvák  1969).  These  extensometers
have  been  designed  to  take  advantage  of  the  moiré  pheno-
menon of optical inference as generated by concentric spirals
and  parallel  lines  (Nishijima  &  Oster  1964).  Moiré  patterns
appear when two sets of identical overlapping periodic mark-
ings are not perfectly aligned: such misalignment generates a
series of macroscopic interference fringes. When two overlap-
ping sets of concentric spirals are poorly aligned, a family of
hyperbolic interference fringes is generated, in which the total
number of fringes reflects both the distance between the in-
dividual rings and the center to center distance between the
individual  spirals.  The  common  principle  axis  indicates  the

430

BRIESTENSKÝ, ROWBERRY, STEMBERK, STEFANOV, VOZÁR, ŠEBELA, PETRO, BELLA, GAAL and ORMUKOV

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

direction  of  the  displacement  (Fig. 2A,B).  In  contrast,  when
two  overlapping  sets  of  parallel  lines  are  poorly  aligned,  a
family  of  parallel  interference  fringes  is  generated,  in  which
the total number of fringes reflects both the distance between
each  individual  line  and  the  total  amount  of  rotation
(Fig. 2C,D). These moiré patterns can be used to measure ex-
tremely small movements due to the fact that a comparatively
small  relative  movement  between  the  overlapping  periodic
markings  results  in  comparatively  large  changes  in  the  ob-
served moiré patterns: the displacement is magnified (Oster &

Nishijima 1963). Therefore, if the number of hyperbolic inter-
ference fringes is known, the amount of displacement can be
quantified through a simple trigonometric transformation.
Likewise, the same is true for angular rotations, if the number
of parallel interference fringes is known (Košvák 1991).

The TM-71 uses the deterministic nature of moiré patterns

in  order  to  measure  movements  across  a  range  of  different
discontinuities  (Klimeš  et  al.  2012).  Although  the  extenso-
meter appears to be a single unit (Fig. 3), this impression is
misleading:  it  actually  comprises  two  distinct  constituent
components, which are able to move independently, one held
in place by the left arm and one held in place by the right arm.
Each of these components incorporates two glass plates, one
orientated  vertically  and  one  orientated  horizontally,  into
which one spiral grid and two sets of parallel lines have been
etched.  The  extensometer  superimposes  these  four  glass
plates, two on each constituent component, to the extent that
when  the  etched  patterns  align  perfectly  the  overlapping
plates  do  not  generate  any  moiré  patterns.  However,  if  the
glass plates move, a series of interference fringes is generated,
which can then be used to quantify the amount of movement.
Moreover, the use of horizontal and vertical glass plates, com-
bined with the configuration of the spiral grid and two sets of
parallel lines, enable us to record these movements in three-di-
mensions. The precision of the instrument is governed by the
number of lines etched into the glass plates: relative displace-
ments can be measured in three Cartesian coordinates (x, y, z)
with a precision of better than  ± 0.007 mm while horizontal and
vertical angular rotations (gxy and gxz) can be measured with
a precision of better than  ± 0.00016 rad (Marti et al. 2013).

Results

Saeva Dupka

The monitored fault is associated with the majority of

speleothem  damage  recorded  in  the  cave  while  its  strike  is
mirrored  by  the  orientation  of  the  main  passage  (Fig. 4).
Data from this monitoring point have now been recorded for
more than two years and these demonstrate oblique displace-
ment along the fault (Fig. 5). The vertical displacement com-
ponent  shows  that  the  southern  block  subsided  by  about

Fig. 2. An illustration of the moiré phenomenon of optical inference
as  generated  by  concentric  spirals  (A  and  B)  and  parallel  lines
(C and D). A – Two overlapping concentric spirals aligned precisely
on top of one another; B – Two overlapping concentric spirals dis-
placed  with  respect  to  one  another.  The  displacement  generates  a
family of hyperbolic interference fringes (see text for further details);
C – Two overlapping sets of parallel lines aligned precisely on top
of  one  another;  D  –  Two  overlapping  sets  of  parallel  lines  rotated
with respect to one another. The rotation generates a family of paral-
lel interference fringes (see text for further details).

Fig. 3. A permanently installed extensometer used to measure aseis-
mic  transient  deformations  across  a  fault.  These  instruments  take
advantage of the moiré phenomenon to record displacement and ro-
tation in three-dimensions. They comprise two distinct components,
one held in place by the left arm and one held in place by the right,
which  move  independently  of  one  another:  the  number  of  macro-
scopic  fringes  changes  as  a  result  of  these  transient  deformations.
The inset presents an example of the macroscopic fringes shown on
one of the combined indicators (see text for further details).

Fig. 4. A plan of the Saeva Dupka study site showing the position of
the most significant fault and location of the extensometric moni-
toring point. The map adapted after Beron et al. 2006.

431

PLATE TECTONIC PRESSURE PULSE PROVIDED BY EXTENSOMETRY, BALKAN Mts (BULGARIA)

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

0.075 mm with respect to the northern block while the hori-
zontal dextral strike-slip displacement component shows that
it has also moved to the east by about 0.05 mm. Much of the
vertical displacement occurred around the time of the distal
M

= 5.6 Pernik Earthquake. This event, the epicenter of

which was located 24 km west of Sofia, took place at
00.00 UTC on 22 May 2012 and was followed by a series of
large aftershocks which culminated in a M

= 4.6 event on

14 July 2012. It is not possible to know whether the recorded
vertical displacement was preseismic or postseismic due to
the monthly monitoring interval, the large number of seis-
mic events, and the considerable length of time over which
they occurred. The opening/closing component shows that the
fault has an overall tendency to open, indicating extension,
in total by about 0.05 mm per year. However, this trend was

interrupted towards the end of 2012, when the opening/closing
component recorded a sudden pulse of compression and ex-
tension prior to resumption of the progressive long-term trend
(Fig. 5, Phases A and B). The magnitude of this sudden pulse
of compression and extension suggests that the movements
were endogenic even though the event was not accompanied
by significant strike-slip or dip-slip. The combination of sub-
sidence of the southern block and opening across the fault pro-
vides evidence of a generally extensional tectonic regime.

Bacho Kiro

The observed fault is associated with conspicuous speleo-

them damage while its strike is mirrored by a number of others
in the cave, along which many of the passages have devel-

Fig. 5. The displacements registered along the EEN-WWS striking fault in Saeva Dupka and along the NE-SW striking fault in Bacho Kiro.
The structural block models characterize the fault displacements that began at both sites towards the end of 2012 and which reflect signifi-
cant aseismic transient deformations (red arrow – strike-slip, blue arrow – dip slip, black arrow – fault opening/closing). The schematic
model in middle right shows that the regional stress-field (white arrows) was orientated approximately N-S during the compressional defor-
mation phase.

432

BRIESTENSKÝ, ROWBERRY, STEMBERK, STEFANOV, VOZÁR, ŠEBELA, PETRO, BELLA, GAAL and ORMUKOV

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

oped,  whose  slickensides  indicate  previous  dextral  strike-
slip  (Fig. 6).  In  terms  of  the  general  trends,  the  first  two
years  of  data  recorded  at  this  monitoring  point  demonstrate
oblique  displacement  along  the  fault,  with  the  vertical  dis-
placement  component  showing  that  the  northwestern  block
has  subsided  by  about  0.05 mm  with  respect  to  the  south-
eastern block and the horizontal sinistral strike-slip displace-
ment  component  showing  that  the  northwestern  block  has
also moved towards the southwest by about 0.1 mm (Fig. 5).
In  detail,  however,  these  general  trends  are  more  compli-
cated, especially the period from towards the end of 2012 to
the  middle  of  2013.  This  has  been  divided  into  four  stages
(Fig. 5, Phases A—D): Phase A is characterized by relative up-
lift  of  the  northwestern  block,  sinistral  strike-slip  displace-
ment,  and  no  opening/closing  across  the  fault;  Phase B  is
characterized by relative uplift of the south-eastern block, neg-
ligible sinistral strike-slip displacement, and sudden compres-
sion (fault closing) across the fault; Phase C is characterized
by relative subsidence of the south-eastern block, negligible
sinistral strike-slip displacement, and sudden extension (fault
opening)  across  the  fault;  and  Phase D  is  characterized  by
relative  subsidence  of  the  north-western  block,  negligible
dextral strike-slip movement of the north-western block, and
no opening/closing across the fault. The sudden pulse of com-
pression  and  extension  recorded  in  the  middle  of  2013
(Fig. 5,  Phases B  and  C)  is  not  typical  of  the  seasonal  open-
ing/closing  trends  recorded  at  surface  monitoring  points
while the size of the displacements, about 0.05 mm, suggests
that they were caused by endogenic processes.

Discussion

The slow-slip phenomena are manifest as aseismic transient

deformations which can be identified as a result of seismic and
geodetic monitoring or through the use of strainmeters and
tiltmeters. An example of the latter is provided by the moni-
toring network EU-TecNet. This network was established on
the basis that faults are sensitive to perturbations in the re-
gional stress field and, therefore, it should be possible to iden-

tify  aseismic  transient  deformations  using  customized  exten-
someters  installed  across  individual  fault  structures  in  the
shallow  crust.  In  contrast  to  many  other  studies  made  using
strainmeters and tiltmeters, this monitoring network is exten-
sive,  with  nearly  one  hundred  and  fifty  monitoring  points
spread  across  mainland  Europe.  This  enables  us  to  analyse
aseismic  transient  deformations  with  great  precision  on  the
continental  scale  and  it  is  argued  that  our  method  represents
the  most  appropriate  approach  to  characterizing  the  progres-
sive deformation of the shallow crust. Data from the network
have  demonstrated  that  deformation  of  the  shallow  crust  re-
flects  the  process  of  stress  redistribution  within  the  mantle-
lithosphere  while  also  showing  that  it  is  unusual  for
conspicuous periods of stress redistribution, here termed tec-
tonic pressure pulses, to be associated with seismic events.

The recorded data indicate that the Bulgarian region was

affected by a tectonic pressure pulse towards the end of 2012
and the direction of maximum horizontal stress at that time
was  orientated  orthogonally  to  the  monitored  fault.  In  con-
trast,  at  Bacho  Kiro,  the  onset  of  the  compressional  phase
was marked by a notable acceleration in the sinistral strike-
slip trend while closure across the fault only occurred at the
end  of  this  compressional  phase.  Therefore,  to  account  for
closing  across  an  EEN-WWS  fault  at  Saeva  Dupka  and  ac-
celeration in the sinistral strike-slip trend across an NE-SW
fault  at  Bacho  Kiro,  the  direction  of  maximum  horizontal
stress,  1

σ, was orientated approximately SSE—NNW (or, at

least, in the NW or SE quadrant). This is an important obser-
vation given that the direction of regional extension is gener-
ally thought to be N—S in the area of Bulgaria (e.g. Papazachos
et al. 1998; Burchfiel et al. 2000; Kotzev et al. 2008). The pre-
sented results suggest that the general pattern of N—E exten-
sion can be interrupted periodically by short-lived episodes of
compression with approximately the same direction.

In order to establish whether the proposed pressure pulse

was a regional phenomenon, the data obtained from a num-
ber  of  other  fault  displacement  monitoring  sites  have  been
investigated:  these  sites  are  located  in  the  Western
Carpathians, the External Dinarides, and in the Kyrgyz Ala-
Too  Range  on  the  northern  edge  of  the  Tian  Shan.  The
extensometric network in the Western Carpathians of Slovakia
comprises  more  than  forty  monitoring  points.  Of  these,
thirty-two, at twenty-seven different sites, had been installed
across  significant  faults  at  the  time  of  the  pulse  in  2012
(Fig. 7).  Seventeen  of  these  points  were  characterized  by
anomalous  displacements  around  the  time  of  the  proposed
tectonic pressure pulse towards the end of 2012. This repre-
sents a significant proportion given that, as mentioned previ-
ously,  the  physical  characteristics  of  a  specific  fault  and  its
orientation  in  relation  to  the  source  of  the  disturbance  pro-
vide  the  fundamental  controls  as  to  whether  or  not  it  will
react  to  a  tectonic  event.  The  anomalies  recorded  at  each
monitoring point are summarized in Table 1. All of the sev-
enteen affected monitoring points record changes in at least
one  of  the  two  slip  components  while  only  three  record
opening/closing across the fault. The different types of iden-
tified anomaly include initiation of strike-slip displacements,
conspicuous reversals in the progressive creep trend, and os-
cillatory displacements (Figs. 8, 9, 10). It has not been pos-

Fig. 6. A plan of the Bacho Kiro study site showing the position of
the most significant faults with previous strike-slip displacements
and location of the extensometric monitoring point. The map adapted
after Beron et al. 2006.

435

PLATE TECTONIC PRESSURE PULSE PROVIDED BY EXTENSOMETRY, BALKAN Mts (BULGARIA)

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

Fig. 9. The structural block models characterize the strike-slip displacement component during the significant tectonic pressure pulse to-
wards the end of 2012. A – initiation of strike-slip displacements at sites which were not previously characterized by horizontal slip,
B – reversal in the sense of displacement that changes the progressive long-term trends, C – large oscillatory displacements which do not
generally change the progressive long-term trend.

Fig. 10. The dip-slip displacements registered across faults in Na Landrovci Cave and Plavecká Cave in the Western Carpathians of Slovakia.
The pressure pulse is expressed as a reversal in the sense of displacement at Na Landrovci Cave and large oscillatory displacement at
Plavecká Cave.

The extensometer is installed across a fault striking E-W with
a 354/78. This site presents evidence for the proposed pressure
pulse (Fig. 11). The onset of the compressional phase is re-
flected by both the strike-slip and dip-slip components. Dur-
ing this phase, the southern block was uplifted by about
0.3 mm with respect to the northern block and the horizontal

strike-slip displacement component shows about 0.1 mm of
sinistral movement, demonstrating oblique thrusting along the
fault as a result of significant pressure possibly from the N-S
or NNE-SSW direction (SE quadrant) at the time of the pulse
in 2012. Such orientation is in good accordance with the gen-
eral N-S striking maximum horizontal stress axis in Slovenia

436

BRIESTENSKÝ, ROWBERRY, STEMBERK, STEFANOV, VOZÁR, ŠEBELA, PETRO, BELLA, GAAL and ORMUKOV

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

Fig. 11 The displacements registered along an E-W striking fault in Vrh Svetih Treh Kraljev in the Southern Limestone Alps of Slovenia
(N 46°0’39.81”, E 14°10’23.11”). The structural block models characterize the fault movements recorded during the compressional and
relaxation phases of the significant tectonic pressure pulse (blue arrow – dip slip, red arrow – strike-slip).

Fig. 12. The displacements registered along a N-S striking fault in Ala Archa in the Kyrgyz Ala-Too Range of Kyrgyzstan
(N 42°38’14.87”, E 74°29’41.05”). The structural block models characterize the fault movements recorded during the compressional and
relaxation phases of the significant tectonic pressure pulse (A and B) (blue arrow – dip slip, black arrow – fault opening/closing).

(Bada et al. 2007). The onset of the relaxation phase is re-
flected by the dip-slip component. During this phase, the
southern block subsided by about 0.2 mm with respect to the
northern block, prior to resumption of the progressive hori-
zontal and vertical trends.

The subterranean monitoring point of Ala Archa in

Kyrgyzstan  is  located  in  the  Kyrgyz  Ala-Too  Range  on  the
northern  edge  of  the  Tian  Shan.  The  extensometer  is  in-
stalled across a fault striking N-S with a 270/20. This distal
site  also  appears  to  present  evidence  for  the  proposed  pres-
sure (Fig. 12). The short compressional phase is reflected by
about 0.05 mm of closing of the fault while the onset of the
relaxation phase is reflected by a significant subsidence and
opening across the fault. During this phase, the fault opened
by about 0.4 mm while the western block subsided by about
0.1 mm with respect to the eastern block.

Conclusions

The EU-TecNet monitoring network was established on

the  basis  that  faults  are  sensitive  to  perturbations  in  the  re-
gional  stress  field  and,  therefore,  it  should  be  possible  to
identify  aseismic  transient  deformations  using  customized
extensometers  installed  across  individual  fault  structures  in
the  shallow  crust.  This  paper  presents  the  first  results  from
two  newly  established  monitoring  points  in  the  Balkan
Mountains in Bulgaria. The data from Saeva Dupka, recorded

across an EEN-WWS striking fault, show 0.05 mm of sinis-
tral strike-slip along the fault and 0.075 mm of subsidence of
the southern block. Much of the subsidence occurred around
the time of the distal M

= 5.6 Pernik Earthquake on 22 May

2012. Furthermore, an important aseismic transient deforma-
tion, which began in autumn 2012, is reflected by significant
fault  closing  and  opening,  or  compression  and  extension,
across  the  monitored  fault.  The  data  from  Bacho  Kiro,  re-
corded across a NE-SW striking fault, show 0.1 mm of sinis-
tral strike-slip along the fault and 0.05 mm of subsidence of
the north-western block. The same important transient defor-
mation is reflected by changes in the strike-slip, dip-slip, and
opening/closing trends and this event has been divided into
four distinct phases on the basis of the recorded data. To ac-
count for closing across an EEN-WWS fault at Saeva Dupka
and  acceleration  in  the  sinistral  strike-slip  trend  across  a
NE-SW fault at Bacho Kiro, the direction of maximum hori-
zontal  stress,  1

σ, must have been broadly N-S during the

aseismic transient deformation towards the end of 2012. This
contrasts markedly the generally accepted picture of N-S ex-
tension  in  this  part  of  the  Eurasian  Plate  (e.g.  Kotzev  et  al.
2008; Olaiz et al. 2009). It suggests that extension can be in-
terrupted  periodically  by  short-lived  episodes  of  crustal
stress  inversion.  These  results  have  been  compared  to  data
from monitoring points in the Western Carpathians, External
Dinarides, and Tian Shan. The sites all provide evidence of
simultaneous displacement anomalies and this observation is
interpreted  as  a  reflection  of  the  propagation  of  a  tectonic

437

PLATE TECTONIC PRESSURE PULSE PROVIDED BY EXTENSOMETRY, BALKAN Mts (BULGARIA)

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

pressure pulse towards the end of 2012. These results high-
light the need for protracted monitoring, across large areas,
in order to better understand the aseismic transient deforma-
tions that characterize slow-slip phenomena.

Acknowledgments:  The  authors  would  especially  like  to
thank Peter Zvonár and Pubomír Sliva for their assistance at
Driny Cave and Sedmička Cave, respectively. This study has
been conducted thanks to the support of the long-term con-
ceptual development research organization RVO: 67985891.
The EU-TecNet fault displacement monitoring network was
established  within  the  framework  of  COST  Action  “3-D
Monitoring  of  Active  Tectonic  Structures”.  This  paper  is
published  within  the  framework  of  CzechGeo-EPOS  “Dis-
tributed system of permanent observatory measurements and
temporary monitoring of geophysical fields in the Czech Re-
public”  (MŠMT  Project:  LM2010008).  The  authors  also
wish  to  acknowledge  financial  support  provided  by  the
Czech Ministry of Education, Youth, and Sports (COST OC
625.10  and  LM2010008);  the  Czech  Science  Foundation
(GA205/05/2770,  GA205/06/1828,  and  GA205/09/2024);
the Grant Agency of the Academy of Sciences of the Czech
Republic (IAA300120801); the Grant Agency of the Slovak
Academy of Sciences (VEGA 1/0141/15); the Bulgarian Sci-
ence  Fund  (DO  02.260/18.12.2008);  the  Karst  Research
Programme  (P6-0119);  and  EPOS  (European  Plate  Observ-
ing Systems) Project (FP7-Infrastructure 262229).

References

Agnew D.C. 2009: Instrumental, theoretical, temporal, and statisti-

cal challenges in the search for transient deformations. Eos 90
(suppl.), G32A-01.

Bada G., Horváth F., Dövényi P., Szafián P., Windhoffer G. &

Cloetingh S. 2007: Present-day stress field and tectonic inver-
sion of the Pannonian basin. Global and Planetary Change 58,
165—180.

Becker A., Davenport C.A., Eichenberger U., Gilli E., Jeannin P.-Y.

& Lacave C. 2006: Speleoseismology: a critical perspective. J.
Seismology 10, 371—388.

Beron P., Daaliev T. & Jalov A. 2006: Caves and speleology in Bul-

garia. Pensoft, Sofia, 1—507.

Briestenský M., Stemberk J. & Petro L. 2007: Displacements regis-

tered around March 13, 2006 Vrbové earthquake M = 3.2 (West-
ern Carpathians). Geol. Carpathica 58, 487—493.

Briestenský M., Stemberk J. & Rowberry M.D. 2014a: The use of

damaged speleothems and in situ fault displacement monitor-
ing to characterise active tectonic structures: an example from
Západní Cave, Czech Republic. Acta Carsologica 43, 129—138.

Briestenský M., Thinová L., Praksová R., Stemberk J., Rowberry

M.D. & Knejflová Z. 2014b: Radon, carbon dioxide, and fault
displacements in central Europe related to the Tohoku Earth-
quake. Radiat. Prot. Dosim. 160, 78—82.

Briestenský M., Stemberk J., Michalík J., Bella P. & Rowberry

M.D. 2011a: The use of a karstic cave system in a study of ac-
tive tectonics: fault movements recorded at Driny Cave, Malé
Karpaty Mts (Slovakia). J. Cave Karst Stud. 73, 114—123.

Briestenský M., Thinová L., Stemberk J. & Rowberry M.D. 2011b:

The use of caves as observatories for recent geodynamic activity
and radon gas concentrations in the Western Carpathians and
Bohemian Massif. Radiat. Prot. Dosim. 145, 166—172.

Briestenský M., Košvák B., Stemberk J., Petro L., Vozár J. &

Fojtíková L. 2010: Active tectonic fault microdisplacement
analyses: a comparison of results from surface and underground
monitoring in western Slovakia. Acta Geodyn. Geomater. 7,
387—397.

Burchfiel C.B., Nakov R., Tzankov T. & Royden L.H. 2000: Ceno-

zoic extension in Bulgaria and Northern Greece: the northern
part of the Aegean extensional regime. In: Bozkurt E., Win-
chester J.A. & Piper J.D.A. (Eds.): Tectonics and magmatism
in Turkey and the surrounding area. Geol. Soc. London, Spec.
Publ. 173, 325—352.

Dobrev N.D. & Košvák B. 2000: Monitoring tectonic movements in

the Simitli Graben, SW Bulgaria. Eng. Geol. 57, 179—192.

Dragert G., Wang K. & James T.S. 2001: A silent slip event on the

deeper Cascadia subduction interface. Science 292, 1525—1528.

Gilli E. 2005: Review on the use of natural cave speleothems as

palaeoseismic or neotectonics indicators. C.R. Geosci. 337,
1208—1215.

Gomberg J. 2010: Slow-slip phenomena in Cascadia from 2007 and

beyond: a review. Geol. Soc. Amer. Bull. 122, 963—978.

Gosar A., Šebela S., Košvák B. & Stemberk J. 2009: Surface versus

underground measurements of active tectonic displacements
detected with TM71 extensometers in western Slovenia. Acta
Carsologica 38, 213—226.

Hirose H., Hirahara K., Kimata F., Fujii N. & Miyazaki S. 1999: A

slow thrust slip event following the two 1996 Hyuganada Earth-
quakes beneath the Bungo Channel, southwest Japan. Geophys.
Res. Lett. 26, 3237—3240.

Klimeš J., Rowberry M.D., Blahůt J., Briestenský M., Hartvich F.,

Košvák B., Rybář J., Stemberk J. & Štěpančíková P. 2012: The
monitoring of slow-moving landslides and assessment of stabil-
isation measures using an optical—mechanical crack gauge.
Landslides 9, 407—415.

Košvák B. 1969: A new device for in situ movement detection and

measurement. Exp. Mech. 9, 374—379.

Košvák B. 1991: Combined indicator using moiré technique. In: 3rd

International Symposium on Field Measurements in Geome-
chanics (Oslo), 9—11 September 1991. A.A. Balkema, Rotterdam,
53—60.

Košvák B. 2006: Deformation effects in rock massifs and their long-

term monitoring. Q. J. Eng. Geol. Hydrogeol. 39, 249—258.

Košvák B., Vilímek V. & Zapata M.L. 2002: Registration of micro-

displacement at a Cordillera Blanca fault scarp. Acta Mont.
Ser. A 19, 61—74.

Košvák B., Dobrev N.D., Zika P. & Ivanov P. 1998: Joint monitor-

ing on a rock face bearing an historical bas-relief. Q.J. Eng.
Geol. Hydrogeol. 31, 37—45.

Košvák B., Mrlina J., Stemberk J. & Chán B. 2011: Tectonic move-

ments monitored in the Bohemian Massif. J. Geodyn. 52, 34—44.

Košvák B., Nikonov A.A., Pereděrin V.I., Sidorin A.J. & Enman S.V.

1992: Monitoring of microdisplacements along ruptures at Garm
Geodynamic Test Site. Izv., Phys. Solid Earth 28, 761—775.

Košvák B., Cacoń S., Dobrev N.D., Avramova-Tačeva E., Fecker

E., Kopecký J., Petro L., Schweitzer R. & Nikonov A.A. 2007:
Observations of tectonic microdisplacements in Europe in rela-
tion to the Iran 1997 and Turkey 1999 earthquakes. Izv., Phys.
Solid Earth 43, 503—516.

Kotzev V., King R.W., Burchfiel C., Todosov A., Nurce B. &

Nakov R. 2008: Crustal motion and strain accumulation in the
South Balkan region inffered from GPS measurements. Earth-
quake monitoring and Seismic Hazard Mitigation in Balkan
Countries. NATO Science Series: IV: Earth and Environmental
Sciences 81, 19—43.

Marti X., Rowberry M.D. & Blahůt J. 2013: A MATLAB® code for

counting the moiré interference fringes recorded by the optical-
mechanical crack gauge TM-71. Comput. Geosci. 52, 164—167.

438

BRIESTENSKÝ, ROWBERRY, STEMBERK, STEFANOV, VOZÁR, ŠEBELA, PETRO, BELLA, GAAL and ORMUKOV

GEOL

EOL

EOLOGICA CARPA

OGICA CARPA

OGICA CARPATHICA

THICA

THICA, 2015, 66, 5, 427—438

Nishijima Y. & Oster G. 1964: Moiré patterns: their application to

refractive index and refractive index gradient measurements. J.
Opt. Soc. Amer. 54, 1—5.

Olaiz A.J., Muñoz-Martín A., De Vicente G., Vegas R. & Cloetingh

S. 2009: European continuous active tectonic strain-stress map.
Tectonophysics 474, 33—40.

Oster G. & Nishijima Y. 1963: Moiré patterns. Sci. Amer. 208, 54—63.
Papazachos B.C., Papadimitriou E.E., Kiratzi A.A., Papazachos C.B.

& Louvari E.K. 1998: Fault plane solutions in the Aegean Sea
and the surrounding area and their tectonic implication. Boll.
Geofis. Teor. Appl. 39, 199—218.

Peng Z. & Gomberg J. 2010: An integrated perspective of the con-

tinuum between earthquakes and slow-slip phenomena. Nat.
Geosci. 3, 599—607.

Shanov S. & Kostov K. 2015: Dynamic tectonics and karst.

Springer—Verlag, Berlin, Heidelberg, 1—123.

Stemberk J., Košvák B. & Cacoń S. 2010: A tectonic pressure pulse

and increased geodynamic activity recorded from the long-term
monitoring of faults in Europe. Tectonophysics 487, 1—12.

Suchkov D. & Sinnyovsky D. 2010: The canyon of Dryanovo River,

Gabrovo District. Ann. Univ. Min. Geol. “St. Ivan Rilski” 53,
Part 1, Geol. & Geophys., 119—124.