This article was downloaded by: [Chulalongkorn University]

On: 07 January 2015, At: 02:30

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Geomatics, Natural Hazards and Risk

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/tgnh20

Monitoring vertical displacements by

precise levelling: a case study along the
Tuzla Fault, Izmir, Turkey
a a
A. Sabuncu & H. Ozener
a
Geodesy Department, Bogazici University Kandilli Observatory
and Earthquake Research Institute (KOERI), Cengelkoy, Istanbul
34684, Turkey
Published online: 03 Jul 2013.

Click for updates

To cite this article: A. Sabuncu & H. Ozener (2014) Monitoring vertical displacements by precise
levelling: a case study along the Tuzla Fault, Izmir, Turkey, Geomatics, Natural Hazards and Risk,
5:4, 320-333, DOI: 10.1080/19475705.2013.810179

To link to this article: http://dx.doi.org/10.1080/19475705.2013.810179

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
 Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015
 Geomatics, Natural Hazards and Risk, 2014
Vol. 5, No. 4, 320–333, http://dx.doi.org/10.1080/19475705.2013.810179

Monitoring vertical displacements by precise levelling: a case study

along the Tuzla Fault, Izmir, Turkey

A. SABUNCU* and H. OZENER

Geodesy Department, Bogazici University Kandilli Observatory and Earthquake
Research Institute (KOERI), Cengelkoy, Istanbul 34684, Turkey

(Received 31 January 2013; in final form 28 May 2013)

Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

The Aegean region and its surrounding area of Western Turkey is one of the most
seismically active and rapidly deforming regions in the world. The study area,
Izmir, is located between 26
150 –28
200 E longitude and 37
450 –39
150 N
latitude in the Aegean region of Western Anatolia. The Tuzla Fault passes
through Izmir, which is the third largest city in Turkey. In this study, we approach
the problem of estimating and investigating the vertical displacements along the
Tuzla Fault using a network of high precision levelling line. We established a
levelling route with eight benchmarks, along the fault line. Six precise levelling
campaigns were performed between 2009 and 2012, and the collected data in the
surveying campaigns were processed by using the least-squares adjustment
method and global testing in order to evaluate the vertical displacement. The
results of the precise levelling indicated that the vertical displacements in the study
area are not as significant in this period as were expected. However, compared
with previous studies conducted in the same area, this study is different not only
from the technique applied, but also it is carried out for the first time.

1. Introduction and kinematics of the study area

The Earth is a natural laboratory for geoscientists to study various topics such as
seismicity, tectonics and earthquakes. Devastating natural disasters often lead to sig-
nificant economic and life losses throughout the world. Therefore, to understand,
evaluate and manage these problems, systematic research and development should
be carried out. The Aegean region, along with the surrounding coastal areas of
Greece and Western Turkey, is one of the most seismically active and rapidly
deforming regions in the world. The intense seismic activity around the Aegean Sea
and its region, including large parts of Greece and Western Anatolia, has been the
most significant geodynamic phenomenon in the Eastern Mediterranean region in
the past century (Ambraseys & Finkel 1991 ). The study area is situated in the middle
of the well-known African Plate in the south, the Eurasian Plate in the north, the
Anatolian Plate in the east and the Hellenic Arc in the west (figure 1) (Ergun & Oral
2000; Kocyigit 2000; Utku 2000; Yilmaz 2000; Taymaz 2001). Moreover, the Aegean
region is under the control of two main motions with N–S extensional tectonics. The
first motion has a westward escape with a 20–25 mm/yr rate of the Anatolia plate,
bounded by the North Anatolia and East Anatolia Fault zones with the intersecting
Karliova depression of East Anatolia. The second motion is the N–S extension of

*Corresponding author. Email: asli.turgutalp@boun.edu.tr

Ó 2013 Taylor & Francis

 Monitoring Vertical Displacements by Precise Levelling 321
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Figure 1. The interactions of major plates around the Aegean region and Western Anatolia
(modified and data from Reilinger et al. 2006).

the Western Anatolian and Aegean plates with a rate of about 30–60 mm/yr. A
group of E–W-trending grabens, which are bounded by E–W-trending normal fault
zones that extend about 100–150 km, have been developing at the end of these
motions (Yılmaz 2000). The western part of Turkey, the Aegean Sea, Greece and its
adjacent areas, and part of the north-eastern Mediterranean have experienced both
major earthquakes and the effects of the active part of the Alpine-Himalayan Oro-
genic Belt system (Mc-Kenzie 1972, 1978; Jackson et al. 1982; Mercier et al. 1989;
Armijo et al. 1996). In the literature, multi-disciplinary research related to interac-
tions throughout the Arabia–Africa and Eurasian plates has been performed for sev-
eral time periods. This research indicates that the region is mainly under pure shear
stress from an internally deforming counterclockwise rotation of the Anatolian plate
relative to the Eurasian plate (Reilinger et al. 2006). A number of different studies in
the area have been performed in order to understand the kinematics of the Aegean
region. R
ozsa et al. (2005) used repeated levelling observations in order to determine
vertical
movements and tectonic activity in the upper Rhine graben which revealed that the
slow tectonic environment had a mean movement 0.25 mm/yr up to 60 years.
Grzempowski et al. (2009) monitored the subsidence at the stations that are located
in Poland-Silesia, which is ascribed to compaction of sediments. Gimenez et al.
(2009) studied about the repeated observations of levelling along the coastline of the
Eastern Betic Cordilere over the past 27 years, and the results show that the vertical
movement is nearly 0.2 mm/yr. In addition, levelling was repeatedly observed in
 322 A. Sabuncu and H. Ozener

order to assess the vertical movements caused by the magma injection and human
development in eastern California near Caldera (Howle et al. 2003). D’Anastasio
et al. (2006) studied the levelling line that was used in the Appennies in order to
reveal short-term vertical movements and Schlatter et al. (2005) have studied the ver-
tical movements in the vicinity of Basel, Switzerland, but their investigations have
not shown any dramatic vertical movements over the past 30 years. Spampinato
et al. (2013) analysed the vertical displacements in Eastern Sicily and Southern
Calabria in Italy by using precise levelling technique. The result indicates that corre-
lated instrumental and geological data make it possible to understand and assess the
active tectonic structures which are in charge of the vertical displacements. The maxi-
mum subsidence rates up to 30 cm/yr were monitored with interferometric synthetic-
aperture radar (InSAR) and precise levelling data in Northern Iran (Motagh et al.
2007). In addition, in these studies, of the kinematics of the study area, the active tec-
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

tonics and geological data are correlated to identify the vertical displacements and
movements.

1.1. Seismic activity and major faults in the study area

The study area, Izmir, which is the third largest city in Turkey, with a population of
4 million, is located on the mid-Aegean coast of Western Anatolia (figure 2) (see the
Turkish Statistical Institute web page). The tectonic framework of the study area
indicates that dense earthquake activities have occurred frequently in this region
throughout history (figure 3), affecting not only Izmir,_ but also Çesme, Urla,

Figure 2. The precise levelling benchmarks in the study area. The upper right of the map
shows Turkey and the study area location.
 Monitoring Vertical Displacements by Precise Levelling 323
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Figure 3. The seismicity of Izmir and Aegean Sea (Mw  3.5) NEMC-KOERI (1900–2012).
The three blue squares denote the precise levelling benchmarks in the study area.

Do ganbey, Karaburun and surrounding cities and towns. A comprehensive study
based on the recorded data of the Turkish General Directorate of Mineral Research
and Exploration (GDMRE) on the active faults and seismicity of Izmir _ and its vicin-
_
ity identified 13 active or possibly active faults in the central part of Izmir and the
nearby towns (Emre et al. 2005) including the Izmir, _ Guzelhisar, Gulbahce, Mene-
men, Seferihisar, Yeni Foca, Bornova, Gumuldur, Gediz Graben, Dagkizilca, Man-
isa, Kemalpasa and Tuzla Faults (figure 4).
The Tuzla Fault, which is about 42-km long on the ground, is situated mainly
_
between Izmir Bay in the north and Kuşadasi Bay in the south, with a NE–SW linea-
ment trending (Emre & Barka 2000; Ocakoglu et al. 2004, 2005; Uzel and Sozbilir
2008). Scientific studies including bathymetric and seismic data have indicated that
the Tuzla Fault enters the Aegean Sea from a SW direction and extends for 50-km
long. In the scientific literature, the Tuzla Fault has various names such as the Cumao-
vasi Fault, the Cumali Reverse Fault and the Orhanli Fault Zone (Saroglu et al. 1987,
1992; Esder et al. 1988; Yılmaz 2000; Genc et al. 2001; Uzel and Sozbilir 2008, Bayrak
& Bayrak 2012 ). The Tuzla Fault forms the western margin of the Cumaovasi Basin
that runs through Gaziemir and Doganbey town, which can be divided into three seg-
ments: Çatalca, Orhanli and Cumali. The 15-km long Çatalca segment is the northern
part of the Tuzla Fault with N35E lineament trending. Moreover, according to the
Quaternary geomorphological data, the Çatalca segment is a right-lateral strike-slip
fault. The central segment of the Tuzla Fault is Orhanli, which is about 16-km long
with N50E lineament trending. The southern part of the Tuzla Fault is the Cumali seg-
ment, which begins from the Cumali Thermal Spa and crosses through the Doganbey
Cape. It is about 15-km long on the ground and continues under the sea for a total
 324 A. Sabuncu and H. Ozener
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Figure 4. The active and possibly active faults in Izmir (modified from Emre et al. 2005).

length of more than 25 km (Ocakoglu et al. 2005). The epicentre of the 6 November
1992 (Mw ¼ 6.0) earthquake was 38.07 N latitude and 36.90 E longitude, and it
caused damage to 100 buildings which was the largest earthquake on the Tuzla Fault
in recorded history (Ilhan et al. 2004; Radius 1997). The focal mechanism solutions of
the 1992 earthquake on the Tuzla Fault indicate that this fault is a right-lateral strike-
slip fault (T€urkelli et al. 1995). Though the morphology at the Doganbey promontory
is seen left lateral, the focal mechanism solution indicates that the Tuzla Fault charac-
ter is right lateral (Tan & Taymaz 2001). The morphological and structural features of
the Tuzla Fault indicate that its early left-lateral offsets were later overprinted by
right-lateral offsets. Moreover, several hot springs occur in the central part of the
fault, which indicates that the hot springs are associated with active faults in the area
(Uzel et al. 2010). In addition, geological observations reveal a right-lateral offset of
200–700 m at young riverbeds of the Holocene age along the Tuzla Fault, and the lat-
est earthquake (Mw ¼ 6.0) indicates that the focal mechanism solution is right lateral
(Emre & Barka 2000; Ocakoglu et al. 2004).

2. Data acquisition and processing

A comprehensive compilation of geodynamic studies of the crustal movements
showed that a determination and comparison of geometric or physical measurements
and observations should be carried out with the same stations at different epochs.
The smaller the time interval and the smaller the amount of movement, the more
accurate the measuring method has to be in order to determine significant changes
(Schlatter et al. 2005). Analyses of geophysical observations allow comparisons
 Monitoring Vertical Displacements by Precise Levelling 325

dating back millions of years. The Geodesy Department of Kandilli Observatory and
Earthquake Research Institute initiated geodetic research in the study area in 2006
(Halicioglu 2007; Halicioglu & Ozener 2008). Reconnaissance was performed in
the study area by taking into account different parameters such as the distance to the
fault, and the rock types for the establishment of the microgeodetic network. The
microgeodetic network comprises 16 stations with planned density sites that are situ-
ated along the fault line and splayed homogenously throughout the city. The obser-
vations were performed along the Tuzla Fault and in its vicinity by GPS and precise
levelling techniques during the period 2009–2012 (Ozener et al. 2012). In this study,
we focused on the precise geometric levelling technique and determination of the ver-
tical displacements in Izmir on the Tuzla Fault and its vicinity.
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

2.1. Precise levelling method

Field studies, analysis of the surveying and interpretation of the results cover a long
period in order for geoscientists to understand vertical movements. These vertical dis-
placements can be monitored by different measurement techniques such as sea-level
observation, geological and seismic observation, GPS observation and levelling data.
The geometric precise levelling technique is the most accurate and precise among
them (Kowalczyk et al. 2012). Furthermore, the oldest, simplest and most accurate
methods for determining the height differences between successive points are geomet-
ric precise levelling. In addition, precise levelling is more accurate than GPS observa-
tions in evaluating and interpreting the vertical displacement in the study area.
We established a levelling route along the fault line from the selected station to
detect the vertical displacements in detail. A further critical prerequisite was that all
benchmarks needed have good monuments on the ground and are stable. With regard
to monumentation, the benchmarks were established with stainless steel pegs by
using drill and epoxy in the bedrock so that they would not be affected by surface
movements, because otherwise displacements might represent monumentation dis-
placements instead of vertical displacements. The levelling route consisted of eight
benchmarks, including three main stations and five auxiliary stations and all

Table 1. Coordinates of levelling benchmarks in UTM coordinate system, benchmark Ids

and approximate distance between benchmarks.

Station Station ID Latitude (UTM) Longitude (UTM) Distance (m)

Kaplica KPLC 491869.37 4215258.73 950

Çeşme CESM 491919.77 4214351.74
1150
Nivelman1 NIV1 491950.69 4213506.58
1200
Huzur Sitesi HZUR 491264.48 4213331.09
1100
Nivelman2 NIV2 490489.49 4212639.72
1000
Nivelman4 NIV4 489443.91 4212440.96
1000
Nivelman3 NIV3 489119.98 4213524.62
1100
Doganbey DBEY 488742.04 4214369.27
 326 A. Sabuncu and H. Ozener
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Figure 5. The levelling route in the study area with all benchmarks.

benchmarks were placed approximately 1000 m apart. Table 1 shows all station UTM
coordinates, the four-digit station IDs and the distances between the benchmarks.
The main benchmarks were Kaplica (KPLC), Huzur Sitesi (HZUR) and Doganbey
(DBEY), and the study area and the precise levelling stations are shown in figure 2.
The levelling line shown in figure 5 was about 7500 m, and the route was measured in
double- run mode. The distances between KPLC and HZUR and between HZUR
and DBEY were 2800 m and 4700 m, respectively (Sabuncu 2010). Six precise level-
ling surveys were carried out in the study area during the period 2009–2012.
In addition, significant procedures applied during the study included the following.

 Equal number of set-ups for forward and backward measurements.

 Maximum allowed sight length of 50 m.
 Use of invar rod and rod correction with metal base.
 Instrument calibration before and after each survey.

First and second precise levelling surveys were conducted on 2009 and 2010 using
digital-level Topcon DL-101C with invar staff, while in 2011 and 2012, a new precise
levelling instrument, Trimble DiNi, was used instead. The precision of Topcon DL-
101C and Trimble DiNi were 0.4 mm/km and 0.3 mm/km with invar staff,
respectively.

3. Data-processing strategy
The following strategies were applied in order to analyse the observations. The height
differences were calculated by fixing KPLC point’s height at 100.00 m for every mea-
surement epoch. In this case, we did not need all points’ orthometric heights because
the aim is to determine and assess the vertical displacements by fixing one main
station’s heights at 100.000 m. The summary set of the levelling results for 2009,
2010, 2011 and 2012, and the height differences are shown in table 2. In order to
 Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Table 2. The summary set of levelling.

Heights (m) Height Differences (m)

2011 2011 July— 2012

2011 2011 2012 2012 2010–2011 Februaryruary— 2012 February–
Stations 2009 2010 February July February June 2009–2010 February 2011 July February 2012 June

KPLC 100.00000 100.00000 100.00000 100.00000 100.00000 100.00000 0.00000 0.00000 0.00000 0.00000 0.00000
HZUR 108.26038 108.25375 108.25051 108.25252 108.25071 108.25318 0.00663 0.00324 0.00201 0.00181 0.00247
DBEY 198.79536 198.78789 198.78375 198.78447 198.78861 198.78989 0.00747 0.00414 0.00072 0.00414 0.00128
Monitoring Vertical Displacements by Precise Levelling
327
 328 A. Sabuncu and H. Ozener

Table 3. The height differences between HZUR and DBEY

benchmarks.
Survey periods Height differences
of HZUR–DBEY (m)

2009 August 90.535

2010 May 90.534
2011 February 90.533
2011 July 90.531
2012 February 90.538
2012 June 90.537

determine and evaluate the vertical displacements in the network, we first examined
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

the height differences between the HZUR and the DBEY benchmarks (table 3). The
results indicated that there was no significant vertical displacement in the network
from 2009 to 2012 (figure 6).
In a geodetic network, deformation analysis is usually conducted in three steps.
These steps are adjusted by the least-squares method, a global test and geometrical
examination of the deformation differences between the two campaigns (Niemeier
et al. 1982; Chrzanowski et al. 1991; Erol 2008). In the first step, the measurements
are obtained from different campaigns at times t1 and t2, and are adjusted with free-
adjustment methods. Approximate coordinates should be taken to be identical for
two different campaign adjustments. In addition, outliers and systematic errors are
determined and eliminated in this step. In the next step, the global-congruency-test
method is performed in order to detect the stable points in the network between the
intervals Dt ¼ t2  t1. During the global-congruency test, the measurements are

Figure 6. The height difference of HZUR–DBEY benchmarks for every year of the study
period, respectively.
 Monitoring Vertical Displacements by Precise Levelling 329

obtained from different campaigns at t1 and t2, and are adjusted by the combined-
free-adjustment method. The free-adjustment calculations of networks were carried
out one by one before the combined free adjustment is performed for both measure-
ments. In the last step, deformation and localization determination methods are
applied if displacements or shape-shifting occurred in the network as a result of the
global-congruency test (Erol 2008).
To determine the displacement vectors of geodetic network points between cam-
paigns, coordinate unknowns should be calculated as follows:

x Sxx ¼ s^ 20 Q‘‘ :
‘ þ v ¼ A^ ð1Þ

The coordinate unknown differences should be tested as a zero value or not. Then
the H0 null hypothesis is established as follows:
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

H0 : Eðx1 Þ ¼ Eðx2 Þ;
H0 ¼ x2  x1 ; ð2Þ
H0 ¼ d ¼ x2  x1 ¼ 0:

Concerning the test data with the null hypothesis, Vi, S values are calculated for each
epoch by using the following formulation:

V1 ¼ v T1 P1 v 1 s1 ¼ V1 =f1 ;
V2 ¼ v T2 P2 v 2 s2 ¼ V2 =f2 ; ð3Þ
Vc ¼ v Tc Pc v c sc ¼ Vc =fc :

The degrees of freedom, for the first and second epochs of free adjustments, are
denoted by f1 and f2, and the degree of freedom for combined free adjustment is
denoted by fc; TC, the test value, is calculated for the global test.
The test value is calculated as follows:

V0 ¼ V1 þ V2 ;
f0 ¼ f1 þ f2 ;
ð4Þ
r ¼ fC  f0 ;
TC ¼ ððVC  V0 Þ=rÞ=ðV0 =f0 Þ:

After the calculation, the TC test value is collated with a Fischer distribution that is
denoted by the (F) value. If Tc > Fr;f0 ;1a , the network has been deformed from t1 to
t2 and the null hypothesis is rejected. In this case, the next stage is the localizing of
the deformation. In order to figure out which points have important and logical
movement at the Dt time interval, this step should be calculated for each point sepa-
rately as follows:

d ¼ x2  x1 ; s2o ¼ Vo =fo ;
T ¼ d T Qtdd d=r s2o ; ð5Þ
ðQd Þ ¼ ðQx1 Þ þ ðQx2 Þ:
 330 A. Sabuncu and H. Ozener

The TH test value is compared with threshold value, which is obtained from a Fischer
distribution with r, f0 and s ¼ 1  a ¼ 0:95 parameters. Also, the TH test value
should be calculated for each point in the network except for the stable points; TH is
denoted as follows:

TH ¼dT Q  1dd d=r S02 ; ð6Þ

where dT is the transpose of d and d is the height difference for the two campaigns,
and
d ¼ H2  H1 ; ð7Þ

S02 ¼ ðV1 TP1 V1 þ V2 TP2 V2 Þ=f1 þ f2 ð8Þ

Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

where S02 is the a posteriori variance value that is obtained from the adjustment for
the two campaigns. Then, if TH > Fr;fo ;1a , it indicates that the point heights are
changed significantly. The global test was repeated until there was no deformation at
any of the benchmarks.
In this study, all procedures mentioned above were applied to all obtained levelling
data sets in a successive manner. The adjustment was carried out by a global test for
three-year measurements. Initially, the global test was performed with KPLC–
HZUR–DBEY benchmarks; then Qdd and variance and covariance matrices were
computed. In the second step, the aim was to determine localization. In the localiza-
tion process, which was carried out for the three benchmarks, calculation should
continue until there is no deformation in the network. The first localization results
indicate that there is a deformation in the network and that the HZUR station has
the maximum test value, which determines the deformation. The HZUR station was
subtracted from the localization process in order to eliminate the deformation; then
the former procedure and calculations were repeated again for the KPLC and
DBEY stations in order to continue localization. In the last step of the localization
process, there was no deformation in the network and so the deformation value was
denoted by dt. The deformation value for the KPLC station was 0.0 mm; for the
HZUR station, it was 7.2 mm; and for the DBEY station, it was 2.5 mm during the
period 2009–2010 (Sabuncu 2010). In addition, all of the procedures were applied for
the successive year measurements, and there was no deformation in the network at
the end of the least square adjustment.

4. Results and discussion

Geodynamic studies have indicated that the Aegean region, Western Anatolia and
the surrounding area need to be monitored continuously with different scientific
techniques in order to understand and evaluate the geodynamic phenomena. Several
multi-disciplinary studies have already been carried out to investigate the geody-
namic settings, seismicity and kinematics in the study area, but most of these studies
have been on a smaller scale. Monitoring the risk zones of high seismic activity and
population concentration using different geodetic techniques is the initial step in the
process of assessing and mitigating the seismic hazard. Different types of these stud-
ies provide the necessary information for prioritizing the planning effort for safety
from earthquake hazard. An appropriate geodetic technique should be chosen with
 Monitoring Vertical Displacements by Precise Levelling 331

consideration of the deformation type and the deforming area proximity, and it
should also take into consideration the urban area and suitable processing
techniques.
In addition, our precise levelling surveys were conducted six times from 2009
through 2012 in the study area and all six survey campaigns indicated that the verti-
cal displacements in the study area are not as significant as we expected. From the
figure 6, it is evident that the height differences between the HZUR and DBEY sta-
tions are not significant. The maximum height differences between these stations
were 7 mm from July 2011 to February 2012. Also, least square adjustment indicates
that there is no deformation in the network except for the 2009–2010 survey. There-
fore, compared with previous research studies conducted in the study area, this study
is different because of the technique that is applied for the first time. The geodetic
observations can provide information for only a short-time window geologically.
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

The time interval between the measurements must be large enough so that dramatic
vertical changes can be observed. The greater the time interval between observations,
the more accurate and precise the measurements of vertical movements will be.
Correlated geodetic and geological data make the results possible to clarify the
seismicity, kinematics of the structure and what is in charge of vertical displacements.
In order to make reliable assessments of vertical displacement, precise levelling
surveys should be performed periodically over the long run in the study area.

Acknowledgements
The authors would like to thank the Geodesy Department of the Kandilli Observa-
tory and Earthquake Research Institute and the project’s members for their support.
The authors gratefully acknowledge the cooperation and contribution of Dr. Mus-
tafa Acar, who helps in the process. We also thank officials and local people in the
region for their help. The maps were drawn using GMT 4.5 software (Wessel and
Smith, 2001). This study was mainly supported by TUBITAK-CAYDAG under
grant No. 108Y295 and Bogazici University Scientific Research Projects (BAP)
under grant No. 5056.

References
Armijo R, Meyer B, King GCP, Rigo A, Papanastassiou D. 1996. Quarternary evolution of
the Corinth rift and its implications for the later Cenozoic evolution of the Aegean.
Geophys J Int. 126:11–53, doi:10.1111/j.1365-246X.1996.tb05264.x
Ambraseys NN, Finkel CF. 1991. Long term seismicity of Istanbul_ and the marmara sea
region. Terra Nova. 3:527–539.
Bayrak Y, Bayrak E. 2012. An evaluation of earthquake hazard potential for different regions
in Western Anatolia using the historical and instrumental earthquake data. Pure Appl
Geophys. 169:1859–1873. doi:10.1007/s00024-011-0439-3.
Chrzanowski A, Chen YQ, Secord JM, Chrzanowski AS. 1991. Problems and solutions in the
integrated monitoring and analysis of dam deformations. CISM Journal. 45(4):547–
560.
D’Anastasio E, De Martini PM, Selvaggi G, Pantosti D, Marchioni A, Maseroli R. 2006.
Short-term vertical velocity field inthe Apennines (Italy) revealed by geodetic levelling
data. Tectonophysics. 418:219–234.
Emre O, Barka A. 2000. Active faults between Gediz graben and Aegean Sea (Izmir region).
Proceedings of International Symposia on Seismicity of Western Anatolia; Turkey.
 332 A. Sabuncu and H. Ozener

Emre O, Ozalp S, Dogaz A, Ozaksoy V, Yildirim C, Goktas F. 2005. The report on faults of
Izmir and its vicinity and their earthquake potentials. General Directorate of Mineral
Research and Exploration Report No. 10754; Ankara (Turkey): GDMRE.
Ergun M, Oral EZ. 2000. General tectonic elements of the Eastern Mediterranean and implica-
tions. Proceedings of International Symposia on Seismicity of Western Anatolia,
Izmir, Turkey.
Erol S. 2008. Determination of deformations with GPS and levelling measurements [Doctoral
Thesis]. Istanbul, (Turkey): Istanbul Technical University.
Esder T, Caglav F, Pekatan R, Yakabag A. 1988. The feasibility report for the area and the
geothermal wellhole of Cumali-Tuzla (Seferihisar-Izmir) (in Turkish). GDMRE
Report No. 8146; Ankara (Turkey): GDMRE.
Genc C, Altunkaynak S, Karacik Z, Yazman M, Yilmaz Y. 2001. The Cubuklu graben, south
of Izmir: tectonic significance in the Neogene geological evolution of the Western Ana-
tolia. Geodinamica Acta. 14:45–55.
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

Gimenez J, Boruq MJ, Gil AJ, Alfaro P, Estevez A, Surincah E. 2009. Comparison of long-
term and short-term uplift rates along an active blind reverse fault zone (Bajo Segura,
SE Spain). Stud Geophys Geod. 53:81–98.
Grzempowski P, Badura J, Cacon S, Przybylski B. 2009. Recent vertical movements in the
Wraclaw section of the middle Odra fault zone. Acta Geodyan Geomater. 6(3):339–
349.
Halicioglu K. 2007. Network design and optimization for deformation monitoring on Tuzla
fault-Izmir and its vicinity [M.Sc. Thesis]. Istanbul, (Turkey): Bogazici University
Kandilli Observatory and Research Institute.
Halicioglu K, Ozener H. 2008. Geodetic network design and optimization on the active Tuzla
fault (Izmir, Turkey) for disaster management. Sensors. 8:4742–4757. doi:10.3390/
s8084742.
Howle JF, Langbein JO, Farrar CD, Stuart K, Wilkinson SK. 2003. Deformation near the
Casa Diablo geothermal well celd and related processes Long Valley calderai Eastern
California, 1993–2000. J Volvanol Geoth Res. 127:365–390.
Ilhan T, Utku M, Ozyal N, Utku Z. 2004. Earthquake risk of izmir region. Izmir (Turkey):
Dokuz Eylul University Marine Science Institute Press.
Jackson JA, Gagnepain J, Houseman G, King GCP, Papadimitriou P, Soufleris C, Virieux J.
1982. Seismicity, normal faulting and the Geomorhological development of the Gulf
of Corinth (Greece): the Corinth earthquakes of February and March 1981. Earth and
Pla Sci Lett. 57:377–397; doi:10.1016/012-821X(82)90158-3.
Kocyigit A. 2000. Seismicity of Southwestern Turkey. Proceedings of International Symposia
on Seismicity of Western Anatolia, Izmir, Turkey.
Kowalczyk K, Bednarczyk M, Kowalczyk A. 2012. Relational database of four precise level-
ling campaigns in Poland. 8th Int Conf Environ Eng. 1–3: 1356–1361.
Mc-Kenzie DP. 1972. Active tectonics of the Mediterranean region. Geophy J Res. 30:109–
185. doi:10.1111/j.1365-246X.1972.tb02351.x
Mc Kenzie DP. 1978. Active tectonics of Alphine-Himalayan belt: the Aegean region and sur-
rounding regions. Geophy J R Ast Soc. 55:217–254. doi:10.1111/j.1365-246X.1978.
tb04759.x.
Mercier J, Sorel D, Vergely P, Simeakis K. 1989. Extensional tectonic regimes in the Aegean
basins during the Cenozoic. Basin Res. 2:49–71, doi:10.1111/j.1365-2117.1989.
tb00026.x.
Motagh M, Djamour Y, Walter TR, Wetzel HU, Zschau J, Arabi S. 2007. Land subsidence
in Mashhad valley, northeast Iran: results from InSAR, levelling and GPS. Geophys
J Int. 168(2):518–526.
Niemeier W, Teskey WF, Lyall RG. 1982. Precision, reliability and sensitivity aspects of an
open-pit monitoring network. Aust J Geod Photogramm Surv. 37:1–27.
 Monitoring Vertical Displacements by Precise Levelling 333

Ocakoglu N, Demirbag E, Kuşçu I. 2004. Neotectonic structures in the area offshore of

Alaçatı, Doganbey and Kuşadası (Western Turkey): evidence of strike-slip faulting in
the Aegean extensional province. Tectonophysic. 391:67–83. doi:10.1016/j.tecto.
2004.07.008.
Ocakoglu N, Demirbag E, Kuşçu I. 2005. Neotectonic structures in Izmir Gulf and surround-
ing regions (Western Turkey): evidences of strike-slip faulting with compression in the
Aegean extensional regime. Marine Geology. 219:155–171. doi:10.1016/j.margeo.
2005.06.004.
Ozener H, Dogru A, Acar M. 2012. Determination of the displacement along the Tuzla Fault
(Aegean region–Turkey): preliminary results from GPS and precise levelling techni-
ques. J Geod. doi:10.1016/j.jog.2012.06.001
RADIUS Project Group. Earthquake hazard and damage scenario. RADIUS Projet Report
_
of Izmir. Izmir _
(Turkey): Metropolitan Municipality of Izmir Publication; 2001.
R
ozsa S, et al. 2005. Determination of displacements in the upper Rhine graben Area from
Downloaded by [Chulalongkorn University] at 02:30 07 January 2015

GPS and leveling data. Int J Earth Sci. 94(4):538–549

Reilinger R, McClusky S, Vernant P, Lawrence S, Ergintav S, Cakmak R, Ozener H, Kadirov
F, Guliev I, Stepanyan R, et al. 2006. GPS constraints on continental deformation in
the Africa-Arabia-Eurasia continental collision zone and implications for the dynam-
ics of plate interactions. J Geophy Res. 111:B05411. doi:10.1029/2005jb004051.
Sabuncu A. 2010. Investigation of crustal movement along Tuzla fault-Izmir [M.Sc. Thesis].
Istanbul, (Turkey): Bogazici University Kandilli Observatory and Research Institute.
Saroglu F, Emre O, Boray A. 1987. The seismicity and the active faults of Turkey. GDMRE,
Report No: 8174; Ankara (Turkey): GDMRE. (in Turkish).
Saroglu F, Emre O, Kuscu I. 1992. Turkish active faults map. Ankara (Turkey): General
Directorate of Mineral Research and Exploration.
Schlatter A, Schneider D, Geiger A, Kahle HG. 2005. Recent vertical mpvements from precise
levelling in the vicinity of the city of Basel, Switzerland. Int J Earth Sci. 94:507–514.
doi:10.1007/s00531-004-0449-9.
Spampinato CR, Braitenberg C, Monaco C, Scicchitano G. 2013. Analysis of vertical move-
ments in eastern Sicily and southern Calabria (Italy) through geodetic leveling data.
J Geod. 66:1–12. doi:10.1016/j.jog.2012.12.002.
Tan O, Taymaz T. 2001. Source parameters of November 6, 1992 Do ganbey (Izmir) earth-
quake (Mw ¼ 6.0) obtained from inversion of teleseismic body-waveforms. 4th Int
Turkish Geology Symposium; Adana (Turkey): Çukurova University. pp. 171.
Taymaz T. 2001. Active tectonics of the North and Central Aegean Sea. Proceedings of Sym-
posia on Seismotectonics of the North-Western Anatolia-Aegean and Recent Turkish
Earthquakes; Istanbul (Turkey): Faculty Of Mines, Istanbul Technical University.
Turkish Statistical Institute web page; [cited 2013 Jan 21]. Available from: http://www.turkstat.
gov.tr/PreHaberBultenleri.do?id=13425
T€
urkelli N, Kalafat D, ve G€ undogdu O. 1995, 6 Kasım 1992 Izmir (Do ganbeyli) depremi saha
g€ozlemleri ve odak mekanizma ç€ oz€um€u. Jeofizik. 9(10):343–348 (in Turkish).
Utku M. 2000. Position of Western Anatolia in Turkey’s seismicity. Proceedings of Interna-
tional Symposia on Seismicity of Western Anatolia, Izmir, Turkey.
Uzel B, Sozbilir H, Ozkaymak Ç. 2010. Neotectonic evolution of an actively growing superim-
posed basin in western anatolia: the inner bay of Izmir. Turkish J Earth Sci 21,4:439–471.
Uzel B, Sozbilir H. 2008. A first record of a strike-slip basin in western anatolia and its tectonic
implication: the cumaovası basin. Turkish J Earth Sci. 17:559–591.
Yilmaz Y. 2000. Active tectonics of Aegean region. Proceedings of International Symposia on
Seismicity of Western Anatolia, Izmir, Turkey.
Wessel P, Smith WHF. 2001. The generic mapping tools (GMT) version 4.5 technical reference
& cookbook; Honululu, (Hawaii): SOEST/NOAA.