Preliminary report on the Mw:6.1 Cephalonia earthquake of 26th Jan 2014 (in English)

MINISTRY OF INFRASTRUCTURE, TRANSPORTATION AND NETWORKS

EARTHQUAKE PLANNING AND PROTECTION ORGANIZATION
INSTITUTE OF ENGINEERING SEISMOLOGY & EARTHQUAKE ENGINEERING (ITSAK)

THE EARTHQUAKE OF 26/1/2014 (M6.1)
IN CEPHALONIA (GREECE): STRONG GROUND
MOTION, SOIL BEHAVIOUR AND RESPONSE OF
STRUCTURES
(PRELIMINARY REPORT)

THESSALONIKI
FEBRUARY 2014



THE EARTHQUAKE OF 26/1/2014 (M6.1)
IN CEPHALONIA (GREECE): STRONG GROUND
MOTION, SOIL BEHAVIOUR AND RESPONSE OF
STRUCTURES
(PRELIMINARY REPORT)

This Report is redacted by the researchers of EPPO-ITSAK(in alphabetic order):
Karakostas Ch., Civil Engineer, Researcher A
Lekidis V., Civil Engineer, Researcher A
Makra K., Dr. Civil Engineer, Researcher B
Margaris B., Dr. Seismologist, Research A
Morfidis K., Civil Engineer, Researcher C
Papaioannou Ch., Dr. Seismologist, Research A
Rovithis M., Dr. Civil Engineer, Researcher D
Salonikios T., Civil Engineer, Researcher B
Savvaidis A., Dr. Geophysicist, Researcher B
Theodoulidis N., Dr. Seismologist, Researcher A

Acknowledgements
K. Konstantinidou, MSc IT and the staff of the Laboratory, S. Zacharopoulos Civil
Engineer, A. Marinos, N. Adam technicians, contribute to effective operation of the
EPPO-ITSAK strong motion network and assure its data transfer to the central
computer facilities in Thessaloniki.
Digital data of the mainshock are open to the public at:
http://www.itsak.gr/news/news/65 .


1. STRONG GROUND MOTION
1.1. Introduction
In January 26, 2014, 13:55 GMT (15:55 local time) a strong earthquake with magnitude
M6.1 (HVR) occurred at the southwestern coasts of the Cephalonia island, about 9km
southwest of the Lixouri town. According to the Seismological Center of the Aristotle
University of Thessaloniki, it was a shallow crustal event with epicenter 38.161N,
20.340E and depth 10km. (http://geophysics.geo.auth.gr/ss/station_index_en.html). At
18:45GMT (20:45 local time) a strong aftershock with magnitude M5.5 (HVR)
followed the mainshock. From the focal mechanisms of both eartquakes it is deduced
that they are related to the Cephalonia Transform Fault (CTF) (Scordilis et al. 1985).
This is a dextral strike-slipe fault with a thrust component (Papazachos and Papazachou
1997, 2003).

Figure 1.1 Epicenter of the 26/1/2014 mainshock (M6.1, red star), aftershock (M5.5,
pink star) and aftershock distribution (M≥4.0) during the first two days of the seismic
sequence in Cephalonia (source: Geophysical Laboratory, Aristotle Univ. Thessaloniki).
Accelerographs of EPPO-ITSAK in yellow squares and seismographs in pink triangles.
Focal mechanisms are also shown.


The seismic sequence of Cephalonia is related to the strike slip fault (Cephalonia
Transform Fault: CTF) (Scordilis et al., 1985). In Fig. 1.1 the epicenter of the
mainshock (13:55GMT) is shown by red star while for the aftershock (18:45 GMT) by
pink star. Focal mechanisms determined by the Columbia University
(www.globalcmt.org) are also given.
From the aftershock distribution during the first two days following the mainshock a
fault length around 18km appears that corresponds to a moment magnitude M6.1. The
aftershock distribution of about 48 hours after the mainshock is shown in Fig. 1.1.
Due to the moderate magnitude of the earthquake, the ground shaking was felt on the
Cephalonia, on the islands of Ithaki, Lefkas and Zakynthos, as well as on areas of
western Greece and Peloponnesos. According to the EMSC the ground motion was also
felt in large part of continental Greece and in south Italy and Albania (Fig. 1.2).

Figure 1.2. Map of observed macroseismic intensities of the 26/1/2014 mainshock in
Cephalonia (EMSC, 2014).
On 27th of January scientific and technical staff of EPPO-ITSAK arrived on the
Cephalonia island in order to install additional accelerographs and seismographs on the
island and record effects of strong ground shaking on natural and built environment.


During the last three years the EPPO-ITSAK has installed throughout Greek a dense
network of continuous recording accelerographs. Their accelerometers are broadband,
of high resolution (24bits), with absolute GPS time. Recordings of this network are
transferred in real time at the central computing unit of the EPPO-ITSAK in
Thessaloniki. Consequently, strong ground motion parameters of the mainshock
recorded at the Cephalonia and Ithaki islands (peak ground acceleration, velocity,
displacement and spectral values) were provided in short time in the form of a
preliminary report on the web site of EPPO-ITSAK (www.itsak.gr). In addition, in less
than 10 minutes (almost real time) after the earthquake origin time, the preliminary
shakemaps
were
generated
and
were
available
to
the
public
(http://portal.ingeoclouds.eu/sitools/client-user/Shakemaps/project-index.html) .
1.2. Network of strong ground motion
The permanent strong motion stations on the Cephalonia & Ithaki Regional Unity
(digital instruments CMG-5TD-EAM) were installed in the town of Argostoli (ARG2:
building of Regional Authority) and in the village of Vasilikades (VSK1: building for
Sitizens’ Service Center) and in the village of Vathi (ITC1: building of Municipality
Technical Department). A temporary network of three accelerographs was deployed in
27 and 28 of January, 2014, mainly within the area strongly affected by the mainshock
around the town of Lixouri. More specifically, three accelerographs were installed; (i)
in the townhall of Lixouri (LXR1), (ii) in the old school of the village Chavriata(CHV1)
and (iii) in a private house of the Ag. Thekli village (AGT1). In addition, two
seismographs one in the Fiskardo village (FSK1) and another south to Argostoli
(VVA1) were also installed (Fig. 1.3). All these instruments are of continuous mode and
from their data analyses significant information for the seismic source properties of the
mainshock and its aftershock, ground motion prediction in the near field and influence
of site conditions on ground motion will result. Data from the stations ARG2, VSK1,
CHV1, LXR1 and AGT1 are transferred in real time to the seismological station of
Thessaloniki to improve the hypocenter and magnitude accuracy of the aftershock
activity for the national seismographic network.
1.3. Mainshock Strong Motion Recordings in Cephalonia
Ground motion of the mainshock was recorded by the permanent accelerograph network
on the islands of Cephalonia and Ithaki as well as throughout Greece. In near real time,
in about 10 minutes after the origin time, preliminary shakemaps were produced and
uploaded on the web (Fig. 1.4). These maps include distribution of instrumental
intensity, peak ground acceleration, velocity and spectral acceleration values for natural
periods T=0.3, 1.0, 3.0 sec.



Figure 1.3. Permanent (ARG2, VSK1, ITC1) and temporary (CHV1, LXR1, AGT1)
accelerograph stations and temporary stations of seismographs (VVA, FSK1), installed
by EPPO-ITSAK on the Cephalonia and Ithaki islands.

Figure 1.4. Shakemaps for the 26/1/2014(M6.1) earthquake in Cephalonia.


High values of spectral acceleration (>1000cm/s/s) at the station ARG2 are observed in
the range of low natural periods, T<0.3sec (Fig. 1.5). However, relatively high spectral
accelerations (>500 cm/s/s) for periods up to 0.7sec are apparent in the N-S component.
At the site of Vasilikades (VSK1), spectral values are up to four times lower than those
of ARG2 but the spectral shape is different relatively enriched for periods T>0.5sec
(Fig. 1.6). Strong ground motion bracketed duration, for ground acceleration >0.1g,
was about 9sec in both horizontal components and 6sec in vertical component. In Figs.
1.7 and 1.8 acceleration, velocity and displacement time histories and their
corresponding pseudo-velocity and acceleration response spectra (D=0.05) are also
provided for the aftershock of 26/1/2014, 18:44GMT (M5.5). Peak ground acceleration
at the station of Argostoli was 116cm/s/s while in Vasilikades 25/cm/s/s. For the
accelerograms processing the software ViewWare was used (Kashima, 2005).
In Table 1.1, recorded peak ground acceleration, velocity and displacement for the
events of 26/1/2014 are given, along with corresponding epicentral distance from the
recording stations of EPPO-ITSAK. Peak ground values were observed in horizontal
components. For the same event, in the town of Lixouri, the accelerograph of the
Geodynamic Institute of NOA, recorded a peak ground acceleration of 0.53g (Kalogeras
I., personal communication).
Table 1.1. Ground motion parameters for the 26/1/2014 mainshock observed at
Argostoli (ARG2) and Vasilikades (VSK1).

STATION

Epicentral
Distance
(km)

Peak Ground
Acceleration
(cm/s/s)

Peak
Ground
Velocity
(cm/s)

Peak Ground
Displacement
(cm)

Argostoli (ARG2)

13

383.4

20.5

3.7

Vasilikades(VSK1)

34

95.0

8.4

2.1



Figure 1.5. Acceleration, velocity and displacement time histories recorded at Argostoli
(ARG2) station and their corresponding pseudo –velocity and acceleration response
spectra for the mainshockof 26/1/2014, 13:55GMT (M6.1) and damping D=0.05.



Figure 1.6. Acceleration, velocity and displacement time histories recorded at
Vasilikades (VSK1) station and their corresponding pseudovelocity and acceleration
response spectra for the mainshock of 26/1/2014, 13:55GMT (M6.1) and damping
D=0.05.



Figure 1.7. Acceleration, velocity and displacement time histories recorded at Argostoli
(ARG2) station and their corresponding pseudo–velocity and acceleration response
spectra for the aftershock of 26/1/2014, 18:44GMT (M5.5) and damping D=0.05.



Figure 1.8. Acceleration, velocity and displacement time histories recorded at
Vasilikades (VSK1) station and their corresponding pseudo–velocity and acceleration
response spectra for the aftershock of 26/1/2014, 18:44GMT(M5.5) and damping
D=0.05.


According to information compiled by Papazachos and Papazachou (1997, 2003) since
mid of 15th century AD the causative fault where the mainshock of 26/1/2014 occurred,
produced events whose maximum magnitude reached M7.2.
The most recent large event on the causative fault occurred in January 17, 1983 with
magnitude M7.0. Despite its large magnitude this event caused a macroseismic intensity
IMM=VI (Bulletin of Geodynamic Inst., NOA) while a peak ground acceleration of
0.17g was recorded at an epicentral distance of 35km in Argostoli (Theodoulidis et al.,
2004).
In Fig. 1.9, comparison of acceleration response spectra recorded at Argostoli of the
26/1/2014 (M6.1) earthquake with that of 17/1/1983 (M7.0), is presented. For periods
less than 1.2sec, ground motion of the recent event is much stronger (two to three times)
than that of 1983, which is probably due to shorter hypocenter-to-station distance.

Figure 1.9. Comparison of horizontal components acceleration response spectra of the
17/1/1983 (M7.0) and 26/1/1983 (M6.1), Cephalonia earthquakes.


2. GEOTECHNICAL ISSUES AND SEISMIC RESPONSE OF LIFELINES
2.1. GEOLOGICAL, GEOTECHNICAL AND GEOPHYSICAL
CHARACTERISTICS OF THE BROADER ARGOSTOLI AREA
2.1.1 ARG2 strong motion site
The broader area of Argostoli town is characterized geologically by pliocenic
sediments. From a geotectonic point of view, the above sediments are part of the Paxos
zone and are composed of sandstones, conglomerates, limestones, marly limestones and
marls. Above those Pliocenic sediments, marine silty clay soils form the offshore zone
along the present coastline of Argostoli town. The above soil formations which are rich
in fosils and organics are characterized by low stiffness and high plasticity index
whereas their thickness is at 30m approximately. The geological map of the broader
area of Argostoli is given in Figure 2.1.
An accelerometric station (marked as CH in Figure 2.2) had been installed in the
telecommunications (OTE) building in the town of Argostoli. Since 2012, a new 24bit
accelerometric station (code name: ARG2) is installed in a two-story building that
houses the Prefecture of Ionian Islands (Figure 2.1) located in a close distance
(approximately 200m) from the old CH station. ARG2 station is currently operated by
ITSAK as part of the Greek National Accelerometric Network.

Figure 2.1. Geological map of the broader Argostoli area (Part of IGME map, scale
1:50.000) indicating the location of ARG2 accelerometric station.



Figure 2.2. ARG2 station installed at the Perfecture building of Argostoli as part of the
Greek National Accelerometric Network operated by ITSAK and former CH station
location where geotechnical survey was performed.
The geotechnical and geophysical characteristics of the subsoil close to CH station were
derived by means of in-situ tests (NSPT, Cross-Hole) performed in 1996 within a
collaborative research project (AUTH, ITSAK and EEPO 1996).
Figure 2.3 summarizes the obtained results in terms of soil classification, NSPT blows
count, P and S waves propagation velocity. The latter leads to a mean Vs30 shear wave
propagation velocity at 420m/sec corresponding to soil class Β (stiff preconsolidated
clay of thickness larger than 70m) according to Greek Seismic Code ΕΑΚ2003 and EC8
(Table 2.1). Given the close distance and the comparable geological setting between
ARG2 and CH stations the above geotechnical profile may be considered representative
also for the subsoil conditions characterizing ARG2 station.

Figure 2.3. Subsoil characteristics at CH station: Soil statigraphy and classification and
variation of NSPT blows count, Vs and Vp velocities with depth (AUTH – ITSAK –
EPPO 1996)


Response spectra corresponding to NS and EW component of the M6.1 Cephalonia
earthquake (26th Jan. 2014, 13:55 GMT) as recorded from ARG2 accelerometric station
are compared in Figure 2.4 with the design elastic spectrum corresponding to soil type
B according to EAK2003 and EC8, respectively. It is observed that the recorded
spectral accelerations are higher than the code-defined values of EAK2003 between 0.1
and 0.3sec. On the other hand, EC8 design spectrum predicts higher spectral values
closer to the recorded ones due to a soil amplification coefficient that is adopted by EC8
contrary to EAK2003. For the particular soil type (i.e. B) the above coefficient is equal
to 1.20 which multiplies the peak ground acceleration (i.e. 0.36g for zone III) in rock
conditions.
Similar comparisons for the vertical component of seismic motion are plotted in Figure
2.5. In this case, ΕΑΚ2003 design spectrum predicts lower spectral amplitudes in the
high frequency range and significantly higher values in the low frequency range with
respect to the recorded response. On the contrary, EC8 design spectrum envelops the
recorded motion in terms of both amplitude and frequency content.
Table 2.1. Eurocode EC8 – Description of soil class Β
Soil
type

Description of stratigraphic
profile

B

Deposits of very dense sand,
gravel, or very stiff clay, at least
several tens of meters in
thickness, characterized by a
gradual increase of mechanical
properties with depth

Vs,30
(m/s)

Parameters
NSPT
(blows/30cm)

360-800

>50

cu (KPa)

>250



Figure 2.4. Comparison of EAK2003 and EC8 elastic design spectra with response
spectra of M6.1 earthquake (26/01/2014, 13:55 GMT) recorded from ARG2
accelerometric station: N-S and E-W components.

Figure 2.5. Comparison of EAK2003 and EC8 elastic design spectra with response spectra of
M6.1 earthquake (26/01/2014, 13:55 GMT) recorded from ARG2 accelerometric station:
Vetical (U-D) component.

MINISTRY OF INFRASTRUCTURE, TRANSPORATION AND NETWORKS


2.1.2. The bridge DeBosset
Within the multidisciplinary project “Restoration and strengthening of the DeBosset bridge in
Argostoli Cephalonia’ (Pitilakis and associates 2006, Rovithis and Pitilakis 2011), a set of
geotechnical field tests were performed in properly selected locations along the bridge axis
(Figure 2.6). The above field survey included sampling boreholes, SPT tests, Cross-Hole
tests and microtremor array measurements complemented by laboratory tests on soil
specimens to provide a detailed knowledge on natural, mechanical and dynamic properties of
the foundation soil at the bridge site. Synthesis of the above geotechnical data allowed a
detailed 2D cross-section of the bridge foundation soil in terms of soil stratigraphy and shear
wave propagation velocity variation with depth (Figure 2.7). Of particular interest is the
presence of a soft surface silty-clay layer (Vs=140-170m/sec) whereas at the depth of
approximately 35m the stiff sandstone layer (Vs=1000m/sec) may be considered as the
seismic bedrock.
Based on the soil profile shown in Figure 2.7, the mean Vs30 shear wave propagation velocity
is equal to 340m/sec. Thus, the particular subsoil conditions close to the shoreline of
Argostoli town are classified as soil type D (soft clays of high plasticity index IP > 50 and
thickness higher than 10m) and soil type C according to EAK2003 and EC8, respectively
(Table 2.2).

Figure 2.6. Geotechnical and geophysical field tests in DeBosset bridge (Pitilakis and
associates, 2006; Rovithis & Pitilakis, 2011)



Figure 2.7. 2D cross-section of the soil profile at the bridge site (Pitilakis and associates,
2006; Rovithis & Pitilakis, 2011)
Table 2.2. Eurocode EC8 – Description of soil type C
Soil
type

profile

C

Deep deposits of dense or
medium-dense sand, gravel or
stiff clay with thickness from
several tens to many hundreds of
meters.

2.2.

Vs,30
(m/s)
180-360

Parameters
NSPT
(blows/30cm)
15 - 50

cu (KPa)

70-250

GEOLOGY OF THE BROADER VASILIKIADES AREA

2.2.1. VSK1 accelerometric station site
Vasilikiades village (where VSK1 accelerometric station is installed) is located at the northern
part of the Cephalonia Island. The geological setting consists of thin-bedded pelagic
limestones of the upper Cretaceous era. From a geotectonic point of view, the above
geological formations are part of the the Paxos zone. The geologic map of the area is given in
Figure 2.8. The exact location of VSK1 accelerometric station is also shown. In the absence
of other geotechnical – geophysical information, the VKS1 site may be classified as soil type
A according to Greek Seismic Code EAK2003 (rocky or semi-rocky formations sufficiently
extended in space and depth and conditioned by the fact that they do not present intense
weathering) and Eurocode 8 (Table 2.3)



The response spectrum of the horizontal ground motion (EW and NS components) recorded
by VSK1 accelerometric station during the earthquake on January 26, 2014 (13:55 GMT) is
significantly lower than the elastic design spectrum defined by EAK2003 or Eurocode 8 for
soil type A (Figures 2.9 and 2.10).

Figure 2.8. Geological map of northern part of Cephalonia Island (Part of IGME map, scale
1:50.000) indicating the location of VSK1 accelerometric station.
Table 2.3. Eurocode 8 – Description of Soil type A
Soil
type

profile

A

Rock or other rock-like
geological formation, including at
most 5m of weaker material at
the surface

Vs,30
(m/s)
> 800

Parameters
NSPT
(blows/30cm)
-

cu (KPa)
-



Figure 2.9. Comparison of EAK2003 and EC8 elastic design spectra with response spectra
of M6.1 earthquake (26/01/2014, 13:55 GMT) recorded from VSK1 accelerometric
station: N-S and E-W components.

Figure 2.10. Comparison of EAK2003 and EC8 elastic design spectra with response spectra
of M6.1 earthquake (26/01/2014, 13:55 GMT) recorded from VKS1 accelerometric station:
U-D component.
2.3 Geotechnical failures
On 28th of January 2014 a group of researchers from ITSAK visited the affected area of
Cephalonia Island to investigate earthquake-induced geotechnical failures.



As a general comment, it may be stated that main M6.1 earthquake event of 26/01/2014
imposed extensive geotechnical failures. These were observed mainly in the Western part of
the Island (peninsula of Paliki) and may be grouped in the following categories:
•
•
•
•

Slope failures due to local landslides occurred in soil or rock formations.
Failures on stone retaining walls and soil settlement
Extensive cracks on the road network
Failures of quay walls and port marines

Figure 2.11 depicts the geographical distribution of the geotechnical failures recorded mainly
along the perimeter of Argostoli bay. It is worth mentioning that most of the geotechnical
failures were gathered along a North-South axis almost parallel to the fault rupture as defined
by the distribution of the aftershocks epicenters (Figure 1.1).

Figure 2.11. Geographical distribution of geotechnical failures imprinted by ITSAK
researchers during their in situ visit on January 28, 2014.
2.3.1 Landslides and rock sliding effects
A large number of local landslides were observed in several locations along the road network
of Paliki peninsula in the Western part of Cephalonia island. Most of the landslide phenomena
occurred in cracked limestone formations (Photo 2.1). Large limestone segments of volume
larger than 2m3 (Photo 2.2) were detached and collapsed due to the large natural slope in



conjunction to the almost vertical tensile cracks (Photo 2.1). Local slope failures were also
observed in rock type formations (consolidated red conglomerates and breccia) covering
partially the road (Photo 2.3). In this case, it is assumed that the landslide mechanism was
mobilized due to the combined action of vertical and horizontal seismic motion.
2.3.2 Failures on stone retaining walls and soil settlement
Repeatable seismic induced failures on stone retaining walls were recorded. Of particular
interest is the case of a stone wall along the road to Atheras village, that was damaged in two
successive locations having an estimated failure length at 10m and 20m respectively (Photo
2.4). A direct effect of the above failure was the subsequent land sliding of the retained soil
causing the closure of the road network. A similar type of failure was recorded at a stone
retaining wall of a physical slope upon which a single-storey house was founded (Photo 2.5).
It should be mentioned that heavy rainfall contributed to the evolution of landslide
phenomena that were observed on site at the particular time period.
Particularly extensive cracks were observed at the stone retaining wall supporting the
foundation soil of a church located in the Chavriata village (Photo 2.6). Although, the above
retaining wall didn’t collapse it is estimated that the evident disruption of the stone parts
should have minimized the lateral strength of the wall leading to a particularly low safety
factor against future seismic loading.
The above failures may be attributed to significant lateral earth pressures imposed on the old
stone walls during seismic shaking. Of course, possible structural deficiencies and poor
condition of joint mortars should also have contributed to the observed damage.
Foundation soil settlement in the range of 3 to 5cm was observed at a 2-storey building
located in Ag. Dimitrios village (Photo 2.7).

Photo 2.1. Landsides of cracked limestone formations



Photo 2.2. Detachment of large limestone segments

(α)
Photo 2.3. Landslides of rock formations

(β)

Photo 2.4. Extensive failure of stone retaining wall along the road network to Atheras Village



Photo 2.5. Retaining stone wall failure supporting the foundation soil of a single-story
structure

Εικόνα 2.6. Chavriata village: Extensive cracks on a three-level retaining stone wall
supporting the foundation soil of a church

Photo 2.7. Foundation soil settlement observed in a 2-storey building.



2.3.3 Cracks along the road network
Extensive cracks in the major part of the road network mainly within the Paliki peninsula
were observed as direct result of the geotechnical failures described above combined with the
old construction age of the road network . Significant problems with cracks of up to 20m long
and up to 50cm wide were recorded at the road network joining Argostoli with Lixouri,
Vilatoria, Kardakata and Zola villages (Photo 2.8)
2.3.4 Seismic response of port facilities at Argostoli and Lixouri.
Extensive failures in the form of longitudinal cracks of jetties and displacement or/and
rotation of quay walls were recorded at Lixouri port and (to a lesser extent) at Argostoli port.
At the Lixouri port, the earthquake-induced horizontal displacement of the concrete quay
walls was measured in the range of 5cm to 15cm followed by a rocking deformation towards
the waterfront thus demonstrating a 3-dimensional pattern of their seismic response (Photo
2.9a and 2.9b). The above response mechanism in combination with possible liquefaction
phenomena (although obvious signs of extensive soil liquefaction were not observed) and
consequent lateral spreading of the liquefied soil led to extensive cracks in several parts of the
port’s jetties parallel to the coastline (Photo 2.9c and 2.9d).
At the Argostoli port, the observed failures were less extensive with small displacement and
rocking deformation of the quay walls , as shown in Photos 2.10a and 2.10b. In the above
type of failure, the material discontinuity between concrete quay walls and the debris
embankment should be taken into account. For this case, liquefaction phenomena were less
extensive due to the clayey composition of the soil formations close to the DeBosset bridge
(Pitilakis et al., 2006). Finally, it is worth noticing that despite the extensive damage, both
ports remained functional.



(a)

(b)

(c)

(d)

(e)
(f)
Photo 2.8. Cracks on the road network: (a) Argostoli – Lixouri (b) Argostoli - Kardakata (cd) Lixouri - Vilatoria (e) Ag. Dimitrios – Livadi (f) Argostoli – Zola



(a)

(b)

(c)
(d)
Photo 2.9. Lixouri port: (a-b) Lateral transposition and turn of seawalls (c-d) Long cracks at
the port’s jetties parallel to the seashore



(a)
(b)
Photo 2.10. Argostoli port: (a-b) Seawall detachement along the seashore close to DeBosset
bridge.
2.4. Preliminary Conclusions on Geotechnical Engineering Aspects
From a geotechnical point of view, the main earthquake-induced geotechnical failures may be
summarized as follows:
• The road network was significantly affected by the earthquake triggered landslides at
slopes composed mainly of limestone formations
• A large number of stone retaining walls failures were observed that may be attributed to
significant lateral earth pressures during seismic shaking imposed on the old stone walls.
• Significant displacements and rocking motion were recorded on the quay walls of Lixouri
port and (to a lesser extent) Argostoli port combined with extensive cracks on the jetties
parallel to the shoreline that should be associated with liquefaction and lateral spreading
phenomena.



3. RESPONSE OF STRUCTURES
3.1. Type of structures
The types of structures that are located at the stricken areas were mainly constructed after the
1953 Cephalonia earthquake. These are divided in four major categories in relation to the type
of the load bearing system. In the wider meisoseismal area where structures were strongly
affected by the 26/01/2014 earthquake the structural systems are grouped as follows:
• One to two storey masonry buildings: These buildings are further subdivided according
location criteria. There are masonry buildings that were constructed by clay or stone or
concrete bricks and by higher quality mortar, which are mainly located at Argostoli and
Lixouri. Also there are one storey masonry buildings that are located at small villages.
These buildings have walls that are composed by roughly treated stones and low-strength
clay mortar . These buildings are of secondary use serving as barns or stables and are not
numerous, since most were destroyed during the catastrophic earthquake that took place in
the island in 1953.
• Reinforced concrete buildings: These buildings are located at the whole island and were
constructed mainly after the 1953 earthquake, till nowadays. These buildings are of one to
four storeys, they have well reinforced concrete frames, as well as shear walls that were
qualified as such under Eurocode standards . Also there are many reinforced concrete
buildings that were constructed after 1953 and are characterized as monuments. The most
important towns on the island are Argostoli and Lixouri and the main building stock there
belong to the present category.
• Monumental and other cultural heritage masonry buildings: These buildings serve mainly
as churches or schools (at country villages) and have one or two storeys. They were
constructed by adopting traditional antiseismic techniques and in most cases their
construction was financed by benefactors.
• Other buildings: During the in situ inspections some wooden buildings were found, as well
as some stone and reinforced concrete bridges.
3.2 Type of damage on structures at the stricken area
The earthquake of 26 January 2014 at Cephalonia cause limited damage to the structures of
the island. In less than 24 hours after the main event, five EPPO-ITSAK researchers arrived at
the island. Among others, extensive inspections were performed and the damage on buildings
and other structures was recorded. Next day after the earthquake there was not significant
available information about the range of the damage throughout the meisoseismal area, so
was decided to inspect buildings in the two main towns of Argostoli and Lixouri on foot and
asking the citizens to show the damaged buildings. Two days after the earthquake important
information was available about the situation at the villages, so was decided to inspect these



areas. It was deduced that the damaged structures were mainly located at the Paliki peninsula,
which is the part of Cephalonia where Lixouri is located. Inspections were performed at the
villages of Agios Dimitrios, Livadi, Vilatoria, Agia Thekla, Kalata, Monopolata, Kaminarata,
Mandourata, Favata, Havdata, Havriata, Vouni and Manzavinata.
Bellow, follows the description of damage that were observed on the load bearing systems of
buildings according to the aforementioned categories. Also photos are given from the most
important types of damage.
• One to two storey masonry buildings: These buildings were not constructed according to
any seismic code and are located mostly at the small villages. Diagonal cracks and/or partial
collapses where observed for buildings of this category. These are low importance
buildings, of secondary use or occasionally habitable and inadequately maintained. In some
cases, cracks that were formed due to previous earthquakes and which were inadequately
repaired with different type of mortar opened further. Also some buildings were totally
abandoned and had partially collapsed. The number of these buildings was significantly
limited and it is not possible to define a specific area where many damaged buildings of this
type are concentrated.
• Reinforced concrete buildings: The majority of the existing building stock in Cephalonia
pertains to this group. The most common type of damage was the detachment of the infill
walls from the surrounding concrete beam – column frames. This type of damage was
observed in numerous cases, and was a reason for concern of the population since it was the
main type of observed damage to structures. After detailed inspections no cracks were
observed at the reinforced concrete elements of almost every building that was checked.
Also, in most cases no diagonal cracks were observed at the infill walls. This fact indicates
that the infill walls were well constructed. Also it was observed that typically, along the
height of the infill wall two concrete wall ties were constructed. Due to the good
construction practices that were applied to the infill walls, the reinforced concrete frames
were supported and the developed inter-storey drift was significantly limited. This way it is
possible to deduce that the absence of diagonal cracks at the infill walls is related with the
absence of cracks at the reinforced concrete elements. Of course, some exceptions were
observed. In some reinforced concrete (R/C) buildings significant damage to structural
elements or/and in infill walls were observed. These structures, with significant damage at
the structural elements and infill walls are mainly located along the road at the north of
Lixouri (towards Agios Dimitrios and Livadi). The most serious damage was observed in
the village of Livadi, where the second floor of a three storey building totally collapsed,
coming to rest on the ground floor. However, in most cases it was found that concrete
buildings were reinforced with adequate number of steel bars and stirrups and no failures
due to the inelastic elongation of steel bars were observed. Observed failures were mainly
due to concrete crashing and diagonal tension. Also there is a subcategory of public and
museum R/C buildings that were constructed after the 1953 earthquake. These buildings
have local failures at their structural element that can be attributed to the reduced strength



due to the reduced durability and the intensity of the earthquake. Through corrosion of the
reinforcement due to carbonation of concrete, the strength of the R/C structural elements is
significantly reduced and the most common type of damage is the loss of concrete cover.
These buildings were built half a century ago and were not properly maintained to preserve
their strength capacity, which was significantly reduced by the pass of time.
• Monumental and other cultural heritage masonry buildings: The temples and other cultural
heritage buildings were one or two storey masonry structures. Some of these buildings had
no damage or light damage. Many of them had cracks at the exterior walls (no detailed
inspection was performed). More specifically, the temples churches at the south of the
Paliki peninsula had suffered significant damage. In some cases strengthening techniques
were applied to these buildings in the past and local failures at the top of the temples were
observed. The bell towers were not constructed in touch with the temples in order to avoid
impact phenomena of the bell tower to the main structure. In masonry buildings that were
constructed at the villages through funding by benefactors damage at the exterior walls were
observed.
• Other buildings: During the in situ inspections some wooden buildings as well as some
stone and reinforced concrete bridges were found. No damage to stone or reinforced
concrete bridges were observed during the inspections. Also, a wooden building that was
examined had no any damage. Additional remarks: During the inspections the detachment
and displacement of many tiles from the roof of many buildings was observed mainly at
Argostoli and Lixouri and less at the other villages. This is explained by the fact that
wooden roofs were constructed without any wooden board panel under the tiles. In contrast,
the tiles were resting on wooden rafters placed at 20 to 30cm distance. In addition to the
repair cost, this detachment and displacement of tiles had the consequence of allowing the
entrance and flow of the rainwater on furniture, electric devices and inside the electric
wires, causing short circuits. Also in many temples ceiling decorations under the wooden
roof were constructed. By the detachment and displacement of the roof tiles, these
decorations were exposed to rain water and should be repainted or repaired or possibly
reconstructed. During the main seismic event many household belongings, furniture,
commercial goods and museum contents fell at the floor and some of them were broken.
For the museum contents the damage happens through the detachment at the interfaces
where the exhibits previously bond.
3.3. Preliminary Conclusions on Seismic Response of Structures
•

In general, recorded intensity of the 26/1/2014 (M6.1) earthquake (acceleration,
energy, response spectra) does not correspond to the observed damage of the buildings
on the island. Certainly, there is damage, mainly on stone masonry buildings and on
buildings that were designed by older codes. No cases were observed in which
hysteretic damping was developed through the flexural cracking of concrete structural



elements and through a high level of inelastic deformation of steel bars. This is also
justified by the fact, that there were not observed cracks at the R/C structural
elements, with few exceptions.
• It is noted that existing buildings possess a substantial amount of strength reserves
(depending mainly on their redundancy and on the over-strength of individual
structural members), as well as possible additional energy dissipation mechanisms,
which contribute to a significant increase of their behavior factor. Experience gathered
from this and previous seismic events suggests that the seismic protection of Greek
urban areas relies also on several alternative factors (such as infill walls, regular
configuration of the structural system, proper material and workmanship quality, etc.).
Due to high level of recorded accelerations, soil-structure interaction phenomena
probably developed with positive effects on structural response. The aforementioned
remarks show that there are additional mechanisms activated for the dissipation of the
imposed seismic energy when a strong earthquake occurs. In this way, the response of
structures subjected to strong ground motion can be improved.
• Through the accumulated experience from past earthquakes as well as from the
present one, it is concluded that seismic protection, not only in Cephalonia but also in
other seismic regions, is additionally improved by other parameters such as the correct
arrangement of the load carrying structural elements, the use of shear walls, the well
constructed infill walls and the use of high quality construction details and materials.
Additionally, the high construction quality applied at the most of the buildings in
Cephalonia, together with the long lasting experience of local construction personnel
to aseismic construction contributes to positive response of the built environment.
• Additionally to the structural damage, the earthquake generated secondary damage to
commercial wares and household contents. Also, due to damage on tile-roofs it is
possible for rain water to enter the house, causing damage to furniture, electric devices
and electric networks. In these cases it is important to repair tile-roofs immediately
and if this is not possible, to cover temporarily the openings on roof. In masonry wall
buildings special and immediate measures should be applied after an earthquake, since
rain water may deteriorate the mortar between bricks as well as may cause damage to
paintings, museum contents etc.
• Finally, it is suggested that buildings constructed according to the 1959 Greek Seismic
Code or before, should be carefully checked, evaluated and when necessary
strengthened according to modern techniques. In addition, school buildings designed
according to older codes of 1959 or 1985(same coefficients as 1959), should be
checked and strengthened according to new Greek code for building strengthening.



Photos 3.1, 3.2. Two neighbor masonry buildings at the villages. The building with strong
R/C ties at top had no damage.



Photos 3.3, 3.4. Partial collapses of masonry buildings of secondary use.



Photos 3.5, 3.6. Multistorey buildings with detachment of infill walls from the R/C frames.
No damage to the structural elements were observed.



Photos 3.7, 3.8. Cracks between structural element and infill walls of the elderly people
nursary (up) and the hospital (bottom) at Argostoli. These buildings were temporarily
abandoned.



Photos 3.9, 3.10. Damage at the Labour houses at Lixouri.



Photos 3.11, 3.12. Damage at a two storey building in Agios Dimitrios.



Photos 3.13, 3.14. Three storey building severely damaged in Livadi. Total collapse of
second floor coming to rest on the ground floor.



Photos 3.15, 3.16. Two storey building with damage, at the Northern road from Lixouri.



Photos 3.17, 3.18. Flexural cracks at the ends of R/C columns at the Northern road from
Lixouri.



Photos 3.19, 3.20. Two temples at the southern area close to Lixouri.



Photos 3.21, 3.22. Temple at the southern area close to Lixouri.



Photos 3.23, 3.24. Cultural heritage masonry building and temple, with damage, at the center
of Paliki peninsula.

Photo 3.25. Damage at glass panes.

Photo 3.26. Damage on tile - roofs.

Photo 3.27. Damage of commercial wares.

Photo 3.28. Damage of commercial wares.

References
Anagnostopoulos S.A., Rinaldis, D., Lekidis, V.A., Margaris,V.N. and Theodulidis,
N. [1987] “The Kalamata, Greece earthquake of September 13, 1986,” Earthquake
Spectra Vol. 3, No 2, 365-402.
AUTH-ITSAK-EPPO (1996). Final report on Study of the effect of local soil
conditions, geomorphology and soil structure interaction on the recordings of the
national accelerographic network.
Dercourt et al. (1980). Grèce. Introduction à la géologie générale. 26 Cong. Géol. Int.
Paris. Excursions 160c-162c, Livret Guide, G12, 159p, Paris.
EAK2003, Greek Seismic Code EPPO, Athens, 2003.
EC-8, EUROCODE No 8, “Structures in seismic regions”, Commission of the
European Communities, 1989.
Hoek E., Carranza Torres C.T., and Corkum B. (2002) “Hoek – Brown failure
criterion - 2002 edition”, Proc. North American Rock Mechanics Society meeting in
Toronto in July 2002.
Kashima T. (2005): ViewWave, Building Research Institute, Tsukuba, Japan.
Lekidis, V.A. and Manos, G.H. [1993] “Observations and classification of buildings
damaged by a strong earthquake,” Proc. of International Seminar: "Post-earthquake
emergency damage and usability assessment of buildings", Athens, Greece, 22-24.
Lekidis, V.A., Theodulidis, N.P., Margaris, V.N. and Papastamatiou, D.J. [1992]
“Observations and lessons learned from recent earthquakes in Greece,” Proc. of the
10WCEE, 1, Madrid, pp. 21-26.
Lekidis,V.A. and Anagnostopoulos, S.A. [1992] “Analysis of the prefecture building
in Kalamata, for the recorded accelerograms of the September 1986 earthquake”
Proc. of 1st Greek Congress on Earthquake Engineering and Engineering
Seismology, Vol II, Greece, 331-346. (in Greek).
OMOE (2003), «Geologic and Geotechnical Research and Studies», Vol. 11,
YPECHODE, GGDE, Specific Committee for elaboration of Transeuropean Network
Issues.
OSMEO (Anath. A3, 2001) «Guidelines for Reduaction of Road Works Studies»,
EOAE.
Papazachos B. and C. Papazachou (1997). The earthquakes of Greece, Ziti Publ. Co.,
304.
Papazachos B. And C. Papazachou (2003) The Earthquakes in Greece, Ziti Publ. Co.,
286 (in greek).
Papazachos B., Mountrakis D., Papazachos C., Tranos M., Karakaisis G. And
Savvaidis A. (2001). Faults caused known strong earthquakes in Greece and
surounding area from the 5th century BC till today, Proc. 2nd Hellenic Conf. On
Earthquake Engineering and Engineering Seismology, Nov. 28-30, 2001,
Thessaloniki, Vol. A, 17-26.
Pitilakis K. et al. (2006) «Rehabilitation and Reinforcement Study of DeBosset bridge
at Argostoli, Cephalonia », Ministry of Culture.

RocLab (V1.001, 2002). http://www.rocscience.com .
Rovithis, E. & K. Pitilakis (2011). Seismic performance and rehabilitation of old
stone bridges in earthquake-prone areas: the case of debosset bridge in Greece. Proc.
IBSBI 2011, October 13-15, 2011, Athens, Greece.
Scordilis, E. M., G. F. Karakaisis, B. G. Karakostas, D. G. Panagiotopoulos, P. E.
Comninakis, and B. C. Papazachos (1985). Evidence for transform faulting in the
Ionian Sea: the Cephalonia Island earthquake sequence, Pure Appl. Geophys., 123,
388–397.
Theodoulidis N., Kalogeras I., Papazachos C., Karastathis V., Margaris V.,
Papaioannou Ch. and Skarlatoudis A., HEAD v1.0: A unified HEllenic Accelerogram
Database, Seism. Res. Letters, 75, 1, 36-45, 2004.

Preliminary report on the Mw:6.1 Cephalonia earthquake of 26th Jan 2014 (in English)

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (19)

Similar a Preliminary report on the Mw:6.1 Cephalonia earthquake of 26th Jan 2014 (in English)

Similar a Preliminary report on the Mw:6.1 Cephalonia earthquake of 26th Jan 2014 (in English) (20)

Más de Institute of Engineering Seismology and Earthquake Engineering

Más de Institute of Engineering Seismology and Earthquake Engineering (15)

Último

Último (20)

Preliminary report on the Mw:6.1 Cephalonia earthquake of 26th Jan 2014 (in English)