vol. pp. 2, ,**1 ,*3 ,,- seismic velocity structure along ... · pdf filezincan; (mu) mudurnu...

�

�

�

� � �

� � � � � �

�

�

� �

�

�

209

Bull. Earthq. Res. Inst.

Univ. Tokyo

Vol. pp.

* e-mail : mksalah @hotmail.com

east of the Anatolian block (creating mountain ranges

(McKenzie, ; Sengor, ), and a subduction zone

earthquakes (M ) with focal mechanisms that show

Earthquake Research Institute, University of Tokyo, Tokyo , Japan

Faculty of Engineering and Architecture, Suleyman Demirel University, Isparta, Turkey

General Directorate of Disaster A airs, Earthquake Research Department, Ankara, Turkey

The three-dimensional P and S wave velocity structures of the crust of the central and western

segments of the North Anatolian Fault Zone are determined by applying a tomography method to

arrival times data generated by local earthquakes that occurred beneath the study area. From the

obtained P and S wave velocity models, we further calculate Poisson’s ratio for a more reliable in-

terpretation of the imaged seismic anomalies. With the exception of high-velocity anomalies de-

tected at a depth of km, prominent low-velocity zones are clearly visible along most parts of the

studied segment down to a depth of km. High Poisson’s ratio anomalies are widely distributed

in most parts of the studied region. Seismic activity is more intense in the high-velocity and high

Poisson’s ratio zones, although it sometimes occurs in low-velocity areas. Large crustal earthquakes

occur in areas characterized by a highly heterogeneous velocity structure, but with distinctly high

Poisson’s ratio anomalies. These results indicate that the active tectonics of this area are reflected

in the seismic velocity structure, and that large crustal earthquakes occur generally in highly het-

erogeneous zones revealed by seismic tomography. These inferences are of great significance for

understanding earthquake-generating processes in areas having intense seismic and tectonic activi-

ties such as the Anatolian plate.

Seismic tomography, Poisson’s ratio, P wave velocity, S wave velocity, North Anatolian

Fault Zone

southwestern boundary of the Anatolian block (Fig.

The North Anatolian Fault Zone (NAFZ) system ). Due to the intense tectonic activity in this region

(Fig. ), about -km long, delineates the northern ( Aktug and Kilicoglu, ), the NAFZ has been

boundary of the Anatolian Plate, and is characterized the locus of large seismic events in past centuries,

by a right-lateral strike slip motion (McKenzie, ; which presents a major source of risk for the popula-

Barka and Gulen, ). In this complex tectonic en- tion (Provost ). These large earthquakes

vironment, the collision of the Arabian, African, and testify that the present-day high strain rate is accom-

Eurasian plates leads to a compressive regime to the modated seismically (Pucci ), and place this

fault among the most active strike-slip faults world-

such as the Zagros and the Caucasus), a right-lateral wide (Ambraseys, ; Ambraseys and Finkel, ;

strike-slip fault zone to the north (which accommo- Ambraseys, ).

dates the westward escape of the Anatolian block Seismicity along the active segments of the

NAFZ is characterized by frequent moderate to large

associated with back arc spreading in the Aegean

Sea (McKenzie, ; Barka and Gulen, ) to the essentially pure right-lateral strike-slip solutions

Mohamed K. Salah *, S. Sahin and M. Kaplan

e.g.,

et al.,

et al.,

Abstract

Key words :

. Introduction

Seismic Velocity Structure along the Western Segment

of the North Anatolian Fault Zone Imaged by Seismic

Tomography

+

,

-

+ , -

2, ,**1 ,*3 ,,-

03

+31, +313

1

++- **-,

#

2

,/

+

+ +/** ,**0

+31,

+322 ,**-

,**0

+31* +33/

,**,

+31, +322

+

�

� ��

�

�

�

� �

�

210

th

Mohamed K. Salah, S. Sahin and M. Kaplan

Fig. . Simplified tectonic map of Turkey, Arabian, African, and Eurasian plates (Provost

). The black arrows show the directions of relative plate velocity, and the

thick black line shows the location of the right-lateral strike-slip NAFZ. Epicenters of

August Golcuk earthquake and November Duzce earthquake are shown

by star and cross, respectively. Abbreviations are : (K) Karliova triple junction ; (Er) Er-

zincan ; (Mu) Mudurnu Valley ; (IzF) Izmit fault ; (GF) Ganos fault.

( Canitez and Uçer, ; McKenzie, ; Zanchi years, which endangers the city of Istanbul, where

and Anglier, ). The earthquakes are the over million people live ( Rockwell ;

most recent to occur on the NAFZ in the century. Papazachos ; Gurer ). Improving

The first of these earthquakes (Golcuk earthquake) our knowledge of the seismic velocity structure of

occurred on August, striking the Izmit region, the crust in which these large catastrophic events

west of the Marmara Sea. This Mw . (USGS) earth- take place will contribute to a better understanding

quake has a focal mechanism (Harvard CMT) that is of the overall tectonics of this region and the forces

consistent with a right-lateral movement along an triggering these earthquakes.

E-W strike-slip fault. Maximum surface dextral of- Although there have been extensive studies of

fsets exceeded m, and surface rupture extended the tectonics and near-surface geology of this unique

along the fault for a total rupture length of km region, research on deep crustal structure and earth-

(Barka, ; Barka , ). Three months quake activity has been hampered by the sparse

later, on November (Muller ; Utkucu coverage of seismic stations. Recently, Akyol

), the Mw . Duzce earthquake occurred. ( ) obtained a -D P-wave crustal velocity model

The focal mechanism solution shows almost a pure, for western Anatolia, which is characterized by crus-

dextral strike-slip movement on an E-W nodal plane, tal velocities that are significantly lower than aver-

dipping to to the north, with a rake between age continental values. It shows four layers down to

and (USGS, Harvard CMT (Tibi )). a depth of km (the assumed Moho depth), with a

This earthquake produced right-lateral surface rup- velocity of . km/s at depths between and km

tures over a total length of km and a maximum and a velocity of . km/s for the lowermost crust

dextral o set of m (Akyuz , ). As a (depth range km). The velocity at the shallow-

result of these two large events, , people were est layer (depth km), however, is poorly constrain-

killed and over , people were injured and/or ed due to the lack of head waves and refracted waves

lost their homes and property. The extent of the at the surface in the data set they used. The low ve-

damage was mainly due to the dense population of locities might be associated with high crustal tem-

the region. A very severe (Mw ) earthquake is ex- peratures, a high degree of fracture, or the presence

pected in the Marmara sea region within the next of fluids at a high pore pressure in the crust (Akyol

et al.,

e.g.,

e.g., et al.,

et al., et al.,

et al.,

et al., et et al.

al.,

et al.,

et al.,

et

+

,**-

+1 +333 +, +333

» +301 +31,

+33- +333 +* ,**+

,* ,**, ,**-

+1

1 .

/

++*

+333 ,*** ,**,

+, ,**-

,**- 1 + ,**0 +

/. 0.

+11 +01 ,**+ ,3

0 ,/ +/ ,+

.* 0 .-

# / ,*** ,**, ,+ ,3

,* *** -

+** ***

1

-*

�

� � ��

�

�

� �

��

�

211

w

w

Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography

Fig. . (a) The epicenteral distribution of the

events used in this study that occurred along the

northwestern side of the NAFZ. Star is the epi-

center of August Golcuk earthquake and

cross is the epicenter of November Duzce

earthquake. (b) The distribution of large earth- Fig. . Locations of seismic stations used in the

quakes that occurred along the NAFZ (M . ) present study that belong to Sabonet and Turknet

since . Thick and thin lines denote the North seismic networks. Thick and thin lines denote the

Anatolian Fault Zone (NAFZ) and the Eskisehir North Anatolian Fault Zone (NAFZ) and the Es-

Fault Zone (EFZ), respectively. kisehir Fault Zone (EFZ), respectively.

November , Duzce earthquake (M . ). These

). In this study we investigate three-dimen- and September , and aftershocks of the

sional velocity and Poisson’s ratio structures of the

crust of this area, and try to correlate them with events were recorded by Sakarya-Bolu Micro-

other geophysical observations that will strengthen earthquake Recording Network (Sabonet) seismic

our current knowledge of the region, thus helping to stations and Turkish National telemetric earth-

mitigate geoseismic hazards. quakes network (Turknet) permanent station belong-

ing to the Turkish General Directorate of Disaster

A airs (Fig. ). The remaining events are from

In this study we used events that occurred background seismicity in the area, and occurred be-

near or to the north of the western segment of NAFZ tween and ; and were also recorded by the

(Fig. a). We used aftershocks of the August , above-mentioned seismic networks. These events

Golcuk earthquake (M . ) between August are located between latitudes . . N and longi-

tudes E with focal depths down to about

km (Fig. a). The uneven distributions of both the

seismic stations and the seismic events in the study

area impose limitations on the resolution scale of the

anomalies obtained, as is explained in the following

sections. The error in the hypocentral locations does

not exceed . km for all events. Earthquakes tend to

cluster around the two main shocks, but we also

selected additional events around these clusters to

get a more uniform distribution of hypocenters in

the study area. The selected events generated

P and S arrivals recorded by the seis-

mic stations shown in Fig. . The accuracy of arrival

times is estimated to be less than . s for P wave

data and somewhat larger ( . s) for the S wave

data. We examined all the residuals stepwise with

respect to the assumed initial velocity model, and

removed data with residuals beyond the limit s.

To study the relation between the nucleation

al.,

. Data

, -1--

+1 +333

+, +333

-

/ *

+32*

+, +333 1 +

,**0 +/ +333 +*13

+.

+

# - +0*-

-1--

+332 ,**-

, +*/+ +1

+333 1 . +1 -3 / .+ /

,2 -, -/

,

, /

+333

-1--

,*1*, +.+.2 +/

-

* +/

* ,/

+

,

�

��

�

�

�

�

�

�

212

rection and km in the depth direction. Straight


Fig. . -D configuration of the grid net adopted for

the present study in horizontal (a), and depth (b)

directions. Grid spacing is . in the horizontal di-

lines in (a) show the location of vertical cross-

sections in Figs. .

Zhao

and Kanamori, ; Zhao , , ;

usu-

ally nonunique, because any given velocity anom-

terms of tectonic and geodynamic implications. Com-

zones of large crustal earthquakes (M ) and the

seismic velocity and Poisson’s ratio anomalies ob-

tained, we also collected data on events that oc-

curred in the study area starting from January

up to December , from the earthquake catalog of

the National Earthquake Information Center (NEIC),

U.S. Geological Survey (Fig. b). A close inspection

of the distribution of the epicenters of these large

events suggests that they occur mainly due to move-

ments along the NAFZ as explained in the previous

section.

The seismic tomography method was first devel-

oped by Aki and Lee ( ). Subsequently, many re-

searchers have successfully improved this technique

and applied it to various regions around the world

( Hirahara, ; Thurber, ; Spakman and

Nolet, ; Zhou and Clayton, , Zhao ,

; Nakajima ; Mishra ). In

this study, we use the tomographic method of Zhao

( ), which has been applied to many regions

having di erent tectonic circumstances (

Ser-

rano , a, b ; Kayal ). The

method uses an e cient -D ray tracing scheme to

compute travel times and ray paths. We adopted a time inversions, we use the following relation :

grid spacing of . in the horizontal direction andVp Vs

km in the depth direction (Fig. ). Velocities at

grid nodes are taken as unknown parameters, and to compute the Poisson’s ratio ( ).

the velocity at any point in the model is calculated The selection of the initially used velocity model

by linearly interpolating the velocities at the eight is an important step in any tomographic inversion

grid nodes surrounding that point. For more details because it generally a ects the amplitude and di-

about the method, see Zhao ( , ). stribution of the velocity anomalies obtained. We

The interpretation of tomographic images is adopted a crustal velocity model that is slightly

di erent from the model determined by Akyol

aly can be attributed to either a thermal or a chemi- ( ) for western Anatolia as our initial P wave

cal variation. For this reason, it is generally useful to velocity model. Their model shows four layers down

consider some other physical parameters for a more to a depth of km (the assumed Moho depth), with a

reliable interpretation of the anomalies obtained in velocity of . km/s at depths between and km

and a velocity of . km/s for the lowermost crust

pared to the seismic velocity itself, the Poisson’s ratio (depth range km). The velocity at the shal-

(or Vp/Vs ratio) is a better indicator of the content of lowest layer (depth km), however, is poorly con-

fluids and/or magma (Zhao and Negishi, ; Kayal strained due to the lack of head waves and refracted

; Takei, ; Salah and Zhao, ; Naka- waves at the surface in the data set they used. The

jima , ) or serpentinization (Kamiya model we used in our study, on the other hand, is

and Kobayashi, ; Christensen, ). Therefore, even simpler. We assumed velocities of , . , . , . ,

after Vp and Vs models are calculated from travel and . km/s at depths of , , , , and km,

et al.,

e.g.,

et al.,

et al., et al.,

et al.

e.g.,

et al., et al.,

et al.

et al.

et al.,

et al.,

. Method

,

0 +/

. -

* ,

+- +0+33/ +330 +331 ,**+

/

,0

+32*

,**/

,

+310

+311 +32-

+322 +33* +33,

+33. ,**+ ,**-

+33,

#

+332 ,**, ,**,

$ -

* , 0, + + ,

+/ .

#

+33, +33.

#

,**0

,3

0 ,/ +/ ,+

0 .-

,+ ,3

-

+332

,**, ,**, ,**-

,**+ ,**/

,*** +330 / / 3 0 * 0 //

1 1/ / +* +/ ,/ .*

-

s s

s

� �

�

� �

�

� �

�

� �

213


Fig. . Initial P wave (solid line) and S wave (dashed

line) velocity models adopted for the present study.

and vertical rays passing at this depth.

The final inversion results were obtained after

two iterations. The P and S wave root-mean-square

(RMS) travel time residuals calculated after these

iterations were . and . s, respectively. In the

following paragraphs we discuss the results of Vp,

Vs, and Poisson’s ratio only in areas having a reliable

resolution.

Figures and show velocity perturbations

relative to the initial velocity models of the P and S

waves at depths of , , , and km, respectively,

respectively (Fig. ). Because there are almost no together with the distribution of seismicity around

studies of S wave velocity structure in this area, we the studied depth slice. The locations of the epicen-

used the relation : Vs Vp/ . , to derive the initial ters of the two large earthquakes are also

S wave velocity model. We checked a number of shown. Fig. , on the other hand, shows perturba-

slightly di erent initial velocity models, but finally tions of Poisson’s ratio at the same depths. To study

selected the above-mentioned one because we found the relation between the nucleation zones of large

that it gives the minimum RMS travel time residual. earthquakes (M ) and the imaged anomalies, we

also plot their epicentral distribution superimposed

on the velocity and Poisson’s ratio structures at and

Before describing the main features of the re- km depths (Figs. , and ) because the focal

sults, we show the results of a checkerboard resolu- depths of most of these events are around km. In

tion test (CRT), (Inoue ; Zhao , the following paragraphs, we describe the main fea-

) to demonstrate the reliability of the tomo- tures of our results.

graphic images obtained. The values of assumed At the shallowest layer ( km depth), the velocity

checkerboard-type perturbations assigned to grid structure is generally heterogeneous, having strong

nodes are ; the image of which is straightfor- lateral variations amounting to . A low P and S

ward and easy to remember. Synthetic arrival times wave velocity anomaly is detected at the central

are then calculated for the checkerboard model. portion of the study area (between longitudes and

Numbers of stations and events with their exact . E) with some portions having a high P wave

locations in the synthetic data are taken to be the velocity. This low-velocity zone has a high Poisson’s

same as those in the real data set. ratio, and is characterized by intense seismic activity

The CRT results for the P and S wave velocity (Figs. , , and ). This heterogeneous structure

structures at four crustal depth slices are shown in continues down to a depth of km, but is dominated

Figs. and , respectively. The resolution is particu- by higher than average velocities. The amplitude of

larly good where many ray paths (event-station the high S wave velocity anomaly is much lower

pairs) are traversing. At a depth of km, there is than that of the P wave, and sometimes the S wave

good resolution in the areas where the seismic sta- velocity is lower than the average at the margins of

tions are located, and most of the input synthetic the well-resolved zone. This is also reflected in the

anomalies are well recovered. The surrounding parts high Poisson’s ratio anomalies seen in most parts.

have relatively poor resolution, especially where no Seismic activity is relatively intense around this

events are located. This is due to insu cient ray depth slice, and occurs mostly in the zone of average

paths criss-crossing there. For deeper layers (depths to high velocities and average to high Poisson’s ratio

, and km), the resolution is reliable only near the with minor exceptions.

station locations and the seismically active regions At depths of and km, prominent low P and

(Figs. and ). Some parts of the study area have a S wave velocities are widely seen in most parts of the

reasonable resolution at a depth of km because of study area (well-resolved zone). Poisson’s ratio is

su cient lengths and directions of both horizontal higher than average at a depth of km, but is gener-

et al., et al.,

. Resolution and Results

/

* ,1 * -*

2 3

, 2 +/ ,/

/

+ 1- +333

+*

#

/

2

+/ ++ +,

+*

+33* +33,

+33.

,

- 0

-*

-+ /

2 3 +*

2

0 1

,

$

2 +/

+/ ,/

0 1

,/

$ +/

.

� �214


Fig. . Results of a checkerboard resolution test for P wave velocity structures at four

crustal depths. The grid spacing is km. Filled and open circles show high and low ve-

locities, respectively. Perturbation scale is shown at the lower right.

Fig. . The same as Fig. , but for S wave velocity.

0

,*

1 0

� �215


Fig. . P wave velocity perturbations at depths of , , , and km, respectively. White circles denote

seismicity in the depth range shown between brackets below each map. Dark and light gray colors

denote high and low velocities, respectively. The perturbation scale is shown to the right. Thick and

thin lines denote the North Anatolian Fault Zone (NAFZ) and the Eskisehir Fault Zone (EFZ), re-

spectively. Star and cross denote the epicenters of the August Golcuk and the November

Duzce earthquakes, respectively.

Fig. . The same as Fig. , but for S wave velocity.

2 , 2 +/ ,/

+1 +333 +, +333

3 2

� �

��

�

216


Fig. . Distribution of Poisson’s ratio perturbations at four depth slices. Light and dark gray colors de-

note high and low Poisson’s ratios, respectively. Other details are similar to those of Fig. .

scribed before are also illustrated along vertical cross-

sections AA’, BB’, CC’, and DD’ (Figs. ), which

ally low at a depth of km with the exception of the Golcuk earthquake, while cross-section DD’ pas-

zone near the epicentral area of the August, ses through the hypocenter of the November,

Golcuk earthquake. Seismic activity is also concen- Duzce earthquake. The velocity and Poisson’s ratio

trated along the zones of low to average velocity/ structures are very heterogeneous at the top km

average to high Poisson’s ratio. depths, but at deeper levels, low-velocity/high Pois-

In Figs. and , we plot the seismic velocity son’s ratio anomalies dominate. The hypocenters of

and Poisson’s ratio structures at depths of and the two large events are located in distinctive

km along with the epicentral distribution of the large zones characterized by average P wave velocity, low

crustal earthquakes. It is clear that the majority of S wave velocity, and high Poisson’s ratio. The back-

the large events occur in the highly heterogeneous ground seismic activity and the majority of the large

zones dominated by high velocities at a depth of earthquakes occur also in zones dominated by aver-

km, which is underlained by lower velocity zones at age P wave velocity, low S wave velocity, and high

a depth of km. The amplitude of the low S wave Poisson’s ratio. We discuss the implications of these

velocity zone, however, is higher, which is reflected observations in the following paragraphs.

in the prominent higher than average Poisson’s ra-

tios at the two depth slices. These results imply that

fluids might be involved in triggering large events in

this tectonically active region, as is explained later. The NAFZ is an active right-lateral system,

The velocity and Poisson’s ratio structures de- which is bound to the north by the westward extrud-

ing Anatolian block (McKenzie, ; Sengor, ;

run Barka, ; Saroglu ). Since the Middle/

either in the E-W or N-S directions (see Fig. for the Late Miocene ( Ma) the NAFZ has accumulated a

location of cross-sections). Cross-sections AA’ and geologic displacement on the order of km

CC’ pass through the hypocenter of the August, (Seymen, ; Sengor, ; Barka, ; Barka and

et al.,

. Discussion

. Large earthquakes along the NAFZ

+*

2

+- +0

,/ +333

+1 +333 +, +333

+*

++ +,

2 +/ +333

2

+/

+31, +313

+33, +33,

. +- /

2/ +,*

+1 +31/ +313 +32+

/

/ +

�

� �

�

�

217


Fig. . The same as Fig. , but at a depth of km.Fig. . P wave velocity (a), S wave velocity (b), and

Poisson’s ratio (c) perturbations at a depth of km

together with the epicenteral distribution of large

earthquakes (see text for details). Other details are

similar to those of Fig. .

two main strands, the Duzce and the Mudurnu fault

segments, where GPS data indicate that the former

accommodates up to mm/yr (Ayhan ).

Farther west, the NAFZ splays again into three ma-

Hancock, ; Westaway, ; Armijo ; jor strands and the northernmost one is consider-

Hubert-Ferrari ). This displacement trans- ed to accommodate most of the present-day strain

lates into a long-term and short-term geologic slip (Barka and Kadinski-Cade, ; Straub ).

rate of . . cm/yr (Tokay, ; Seymen, ; The preceding paragraphs show a high rate of

Barka and Hancock, ) and . cm/yr (Hubert- seismic activity along the NAFZ, which is also clear

Ferrari ), respectively. Conversely, GPS from the distribution of the seismic events used in

networks have measured present-day strain rates in this study (Fig. ), most of which are aftershocks of

the northern part of the Anatolian block that reach the two large earthquakes (see section ). The

cm/yr (Reilinger , ; Straub highly heterogeneous seismic velocity and Poisson’s

; McClusky ; Kahle , ). ratio crustal structures obtained in this study are

To the east of the town of Bolu, the NAFZ is formed consistent with this unstable seismo-tectonic set-

by a main single trace, but to the west it splays into ting. The majority of large crustal earthquakes oc-

et al.,

et al.,

et al.,

et al.,

et al.,

et al., et al.,

et al., et al.,

+, ++ +/++

2

2

+* +333

+32. +33. +333

,**,

+322 +331

* / * 2 +31- +31/

+32. + 2

,**,

,

+333 ,

, - +331 ,***

+331 ,*** +333 ,***

� �218


Fig. . Vertical cross-sections of P wave velocity, S Fig. . The same as Fig. , but along line BB’.

wave velocity, and Poisson’s ratio structures along

line AA’ (see Fig. for the location of the cross-

sections). Low velocities and high Poisson’s ratios

are shown by light gray, whereas high velocities

and low Poisson’s ratios are shown by dark gray.

Star denotes the hypocenter of the August

Golcuk earthquake and big circles denote large

earthquakes. Crosses show the distribution of seis-

micity in a -km-wide zone around the profile.

The perturbation scale is shown to the right of

each panel.

and are certainly of value for identifying and charac-

terizing faults with potential for surface rupturing

earthquakes.

The velocity models obtained in this study for

western Anatolia are dominated by low velocities

with the exception of a high-velocity anomaly at a

depth of km and by high Poisson’s ratios (except at

cur close to zones characterized by average P wave km depth). The low Poisson’s ratios at a depth of

velocity, low S wave velocity, and high Poisson’s ra- km, which appear around the seismically active

tios (Figs. , and ). In summary, a comprehensive zone, have low or even no resolution, hence we con-

analysis of tectonic landforms and associated struc- sider them to be unreliable features. Akyol

tures, combining the results of seismic tomography ( ) obtained lower crustal velocities in western

and other geophysical observations, are valuable Anatolia and attributed them to high temperatures,

tools for defining the strain distribution pattern and fluids, and high pore pressure, or the presence of

its evolution in the near surface. These data are the partial melt. Their results also reflect near-surface

basis for understanding how surface deformation geological complexities and that crustal velocities

builds up. Consequently, they provide a picture of are significantly slower in western Anatolia, increas-

the characteristics of the principal slip zone at depth, ing from . km/s at a depth of km to only .

et al.

. Previous geophysical observations and ob-

tained velocity and Poisson’s ratio models

+- +. +-

.

+1 +333

+*

2

,/

,/

++ +,

,**0

0 ,/ ,+ 0 .-

/ ,

� �

��

� �

�

�

219


Fig. . The same as Fig. , but along line CC’. Fig. . The same as Fig. , but along line DD’. The

big cross denotes the hypocenter of the Novem-

ber Duzce earthquake.

km/s at the Moho. The velocities obtained at the

base of the crust are within the bounds of average

crustal velocities at high temperatures (Christensen gions are interpreted to be hot and unstable mantle

and Moony, ). These low velocities might indi- lid zones, whereas the very low velocity zones are

cate that fluid-filled faults and fractures in this seis- interpreted to be regions with no mantle lid. Al-

mically active region permeate the crust ( Al- though we have no resolution at depths greater than

Shukri and Mitchell, ; Mitchell ). It km, these observations, however, support the exis-

was also found that the average velocity for tence of the low velocities and high Poisson’s ratios

the entire Aegean region is approximately . km/s anomalies we detected at depths greater than km

(Panagiotopoulos and Papazachos, ), which is (Figs. ). Seismic activity is concentrated mostly

lower than the worldwide average continental upper in the top km depths, which might suggest the

mantle velocity of . km/s (Moony and Braile, presence of seismically active low-angle breakaway

). faults within the upper crust of western Anatolia

Recently, Al-Lazki ( ) detected broad- (Sengor, ).

scale ( km) zones of low ( km/s) velocity Measurements of coda by Akinci ( )

anomalies underlying the Anatolian plate and the in western Anatolia for the frequency range . Hz

Anatolian plateau, and even smaller-scale (( show a strong frequency dependence, which agrees

km), very low ( . km/s) velocity zones beneath with the assumption that tectonically active areas

the Isparta Angle, central Turkey and the northern show a high attenuation due to the complex struc-

Aegean Sea region. The broad-scale low-velocity re- ture of the region. Using a electrical resistivity sur-

n

n

n

n c

n

P

e.g.,

et al.,

P

P

et al.

P Q et al.

P

+/ +- +0 +-

+,

+333

+33/

+322 +331 ,/

1 3 +/

+32/ +- +0

+*

2 +

+323

,**. +321

/** 2 +33.

+ / +*

,**

1 2

�

�

�

� �

��

�

�

_

220


ter (NEIC), (http : //neic.usgs.gov/neis/epic/epic rect.

vey in northwest Anatolia, Caglar ( ) detected a study area are dominated by low velocity anomalies.

km thick surface layer of low resistivity character- Poisson’s ratios are generally higher than average

izing the sedimentary sequences of this depth range, with minor patches of low Poisson’s ratio anomalies.

and deduced that the tectonic structure is compara- Seismic activity is concentrated near the zones char-

tively complex. Below a depth of about km geoelec- acterized by low to average velocities and moderate

tric models show a more resistive structure underly- to high Poisson’s ratio. The results of a checkerboard

ing these sediments. The resistive structure is corre- resolution test show that most of the currently ob-

lated with Precambrian crystalline rocks and gneiss tained structures are reliable features. They are also

schist basement. This change from low to high re- in general agreement with many other geophysical

sistivity at a depth of about km is consistent with observations, although the scales of such studies

the change from low (at km depth) to high (at km should be considered. All these anomalies are associ-

depth) velocities we observed in our study. High ated with the complex and active tectonic setting of

average heat flow with geothermal activity ( Ilk- the region, the circulation of hot geothermal fluids,

isik, ; Pfister ; Gemici and Tarcan, and the presence of sediment fillings at the surfacial

), high rate of seismicity (Bozkurt, ), inten- layers.

sive faulting and extension-related Neogene and The various pieces of evidence mentioned in the

Quaternary volcanism ( Paton, ; Innocenti previous sections suggest that the generation of a

) are also among the main characteristics of large crustal earthquake is closely related to the

the region. surrounding tectonic environment such as rifting/

Sari and Salk ( ) found that the most pro- spreading or the relative transform motion between

nounced structural and morphological features in adjacent plates and the physical/chemical properties

western Turkey are graben-like structures created of crustal materials such as magmas and fluids. Com-

by E-W normal faulting. The grabens are filled with plex physical and chemical reactions may take place

recent sediments that give rise to relatively negative in the source zone of a future earthquake, causing

gravity anomalies. Caglar and Isseven ( ) de- heterogeneities in the material property and stress

tected three electrically conductive ( m) zones field, which may be detected by seismic tomography

at depths of km. The origins of these zones are and other geophysical methods. These results indi-

explained by the circulation of hydrothermal fluids cate that large earthquakes do not occur anywhere,

with low resistivity values ( . . m) and by the but only in anomalous areas that may be detected

e ects of a strong hydrothermal alteration in the with geophysical methods. Higher resolution seis-

rocks. They also detected a low magnetic anomaly mic imaging of this area combined with other geo-

with a low intensity of about in a central area logical, geochemical, and geophysical investigations

in northwest Anatolia, and ascribed it to hydrother- would certainly deepen our understanding of the

mal demagnetisation of the rocks due to geothermal ongoing tectonic activity and the earthquake-gen-

activity. erating processes, and would also contribute to the

mitigation of seismic hazards.

In this study we collected a number of arrival

times of P and S waves from the aftershocks of the We thank M. Uyeshima and two anonymous re-

two large earthquakes, as well as local earth- viewers for their constructive comments and discus-

quakes, and used them to map the -D crustal struc- sions, and the National Earthquake Information Cen-

ture of P wave, S wave, and Poisson’s ratio beneath

northwest Anatolia. The velocity structure is very html) for facilitating the availability of earthquake

heterogeneous at shallow layers of the crust ( and catalogs and seismological data on the web. All fig-

km depths) with amplitudes of . It is dominated ures in this paper were made using GMT (Generic

mainly by low velocities at a depth of km and high Mapping Tools) software written by Wessel and

velocities at a depth of km. At the deepest layers of Smith ( ).

the crust (depths and km) most parts of the

e.g.,

et al.,

e.g., et

al.,

. Conclusions

Acknowledgements

,**+

/

/

/

, 2

+33/ +332

,**, ,**+

+33,

,**/

,**0

,**.

. 0

/ +/

* ,2 * /,

#

/*

+333

-

, 2

0

,

2 +332

+/ ,/

0

W

W

g

�

�

�

�

�

��

�

��

�

� �

��

��

�

�

�

�

�

�

Anatolia, Geophys. J. Int., . /j. - X. . .

Dikbas, B. Yerli, R. Armijo, B. Meyer, J.B. de Chabalier,

Aki, K. and W. Lee, , Determination of the three- T. Rockwell, J.R. Dolan, R. Hartleb, T. Dawson, S.

dimensional velocity anomalies under a seismic array Christo erson, A. Tucker, T. Furnal, R. Langridge, H.

using first P arrival times from local earthquakes, .A Stenner, W. Lettis, J. Bachhuber and W. Page, , The

homogeneous initial model, , surface rupture and slip distribution of the August

. Izmit earthquake (M . ), north Anatolian fault,

Akinci, A., A.G. Taktak and S. Ergintav, , Attenuation , .

of coda waves in western Anatolia, Barka, A.A., H.S. Akyuz, H.A. Cohen and F. Watchorn, ,

, . Tectonic evolution of the Niksar and Tasova-Erbaa

Aktug, B. and A. Kiliçoglu, , Recent crustal deformation pull-apart basins, north Anatolian fault zone : their sig-

of Izmir, western Anatolia and surrounding regions as nificance for the motion of the Anatolian block,

deduced from repeated GPS measurements and strain , .

field, , . Barka, A.A. and L. Gulen, , New constraints on age and

Akyol, N., L. Zhu, B. J. Mitchell, H. Sozbilir and K. Kekovaly, total o set of the north Anatolian fault zone : Implica-

, Crustal structure and local seismicity in western tions for tectonics of the eastern Mediterranean Region,

In : Melih Tokay Symposium, Spec. Publ. Middle-east

Akyuz, H.S., A.A. Barka, E. Altunel, R.D. Hartleb and G. Techn. Univ., Ankara, pp. .

Sunal, , Field observations and slip distribution of Barka, A.A. and P.L. Hancock, , Neotectonic deforma-

the November , Duzce earthquake (M . ), Bolu- tion patterns in the convex-northwards arc of the

Turkey, in : the Izmit and Duzce earthquakes : Pre- North Anatolian fault, in The Geological Evolution of

liminary results, by A.A. Barka (Ed.), pp. , the Eastern Mediterranean, edited by J.G. Dixon and A.

Istanbul Tech. Univ., Istanbul. H.F. Robertson, , .

Akyuz, H.S., R.D. Hartleb, A.A. Barka, E. Altunel, G. Sunal, B. Barka, A.A. and K. Kadinski-Cade, , Strike-slip fault

Meyer and R. Armijo, , Surface rupture and slip geometry in Turkey and its influence on earthquake

distribution of the November Duzce earthquake activity, , .

(M . ), north Anatolian fault, Bolu, Turkey, Bozkurt, E., , Neotectonics of Turkey a synthesis,

, . , .

Al-Lazki, A.I., E. Sandvol, D. Seber, M. Barazangi, N. Turkelli Çaglar, I., , Electrical resistivity structure of the north-

and R. Mohamad, , tomographic imaging of western Anatolia and its tectonic implications for the

mantle lid velocity and anisotropy at the junction of Sakarya and Bornova zones,

the Arabian, Eurasian and African plates, , .

, . Çaglar, I. and T. Isseven, , Two-dimensional geoelectri-

Al-Shukri, H. J. and B. J. Mitchell, , Reduced seismic cal structure of the Goynuk geothermal area, northwest

velocities in the source zone of new Madrid earth- Anatolia, Turkey, ,

quakes, , . .

Ambraseys, N.N., , Some characteristic features of the Canitez, N. and B. Uçer, , Computer determinat

north Anatolian fault zone, , .

Ambraseys, N.N., , The seismic activity of the Marmara

sea region over the last years,

, .

Ambraseys, N.N. and C.F. Finkel, , The seismicity of

Turkey and adjacent areas : a historical review,

. Muhittin Salih Eren, Istanbul, pp.

Armijo, R., B. Meyer, A. Hubert and A.A. Barka, , West-

ward propagation of the North Anatolian fault into the

northern Aegean : Timing and kinematics, ,

.

Ayhan, M.A., C. Demir, A. Kilicoglu, I. Sanli and S.M. Naki-

boglu, , Crustal motion around the western seg-

ment of the north Anatolian fault zone : Geodetic meas-

urements and geophysical interpretation, paper pre-

sented at IUGG , Int. Union of Geodesy and Geophys.,

Birmingham, U.K.

Barka, A.A., , Seismo-tectonic aspects of the North Ana-

tolian fault zone, Ph.D. thesis, pp., Univ. of Bristol,

Bristol, U.K.

Barka, A.A., , The North Anatolian fault zone, Ann,

, .

Barka, A.A., , The August Izmit Earthquake,

, .

Barka, A.A., H.S. Akyuz, E. Altunel, G. Sunal, Z. Çakir, A.

J. Geophys. Res.,

Bull. Seismol. Soc. Am.,Phys. Earth Planet.

Inter.,

Tec-tonophysics,

J. Geodyn.,

et al.Geol. Soc. Spec. Publ.,

Tectonophysics,Bull. Seis- Geo-

mol. Soc. Am., dyn. Acta,

PnPhys. Earth Planet. Inter.,

Geophys. J.Int.,

J. Volcanol. Geother. Res.,Bull. Seismol. Soc. Am.,

Tectonophysics,

Bull. Seismol. Soc.Am.,

Geology,

Tecton.,

Science,

References

+* ++++ +-0/ ,.0 ,**0 *-*/-

+310

#

+ ,**,

.-2+ +1

.-33 +333 1 .

+33. .- 0*

,***

+// +0/

,**0

,.- ,0.

.1+ .2. +322

#

,**0

-3 0/

,*** +32.

+, +333 1 +

+333

0- 1*

10- 11-

+322

,**,

+, +333 00- 02.

1 + ,**+

0+ 00 - -*

,**+

,**.

3/ ++*

+*,. +*.* ,**.

+322

+2-

+.3+ +/*3 +31

+31* » +301

+.- +0/

,**,

,***

+ +2

+33/

+/**

+2** ,.*

+333

,01 ,1*

+333

33

+32+

--/

+33,

+0. +3/

+333 +1 +333

+2/2 +2/3

2+

3,

21

-,,

.+

+1

1

3, +.

+,/

+/2

+-.

12

3

3,

,1

0

,2/

� �

�

�

�

�

��

�

�

221

using silica temperature estimates from thermal springs,

Christensen, N. and W. Mooney, , Seismic velocity struc-


ions for

the fault plane solutions in and near Anatolia,

, .

Christensen, N., , Poisson’s ratio and crustal seismology,

, .

ture and composition of the continental crust : a global

view, , .

Gemici, U. and G. Tarcan, , Hydrogeochemistry of the

Simav geothermal field, western Anatolia, Turkey,

, .

Gurer, O.F., N. Kaymakçi, S. Çakir and M. Ozburan, ,

Neotectonics of the southeast Marmara region, NW

Anatolia, Turkey, , .

Harvard, , CMT catalogue. http : //www.seismolo

gy.harvard.edu/

Hirahara, K., , A large-scale three-dimensional seismic

structure under the Japan islands and the Sea of Japan,

, .

Hubert-Ferrari, A., R. Armijo, G. King, B. Meyer and A.A.

Barka, , Morphology, displacement, and slip rates

along the northern Anatolian Fault, Turkey,

(B ), , doi : . / JB .

Ilkisik, O.M., , Regional heat flow in western Anatolia

, .

Tectono-physics,

J. Geophys. Res.,

J. Geophys. Res.,

J.Volcanol. Geother. Res.,

J. Asian Earth Sci.,

J. Phys. Earth,

J. Geophys.Res.

Tectonophysics,

+33/

,-/ ,..

+330

-+-3 -+/0

310+ 3122

» ,**,

,+/ ,--

» » ,**-

+*.+ +*/+

+310 ,**,

+311

-3- .+1

,**,

+* ,,-/ +* +*,3 ,**+ ***-3-

+33/

+1/ +2.

.

+*+

+**

++0

,+

,/

+*1

,..

�

��

�

�

��

�

�

�

�

�

�

�

�

�

�

�

�

_

nucleation, , . / GL .

connaissance of the Marmara Sea region (NW Turkey) :

Innocenti, F., S. Agostini, G.D. Vincenzo, C. Doglioni, P. Papazachos, , Precursory accelerating seismic crus-

Manetti, M.Y. Savasçin and S. Tonarini, , Neogene tal deformation in the northwestern Anatolia fault

and Quaternary volcanism in western Anatolia : Mag- zone, , .

ma sources and geodynamic evolution, Paton, S., , Active normal faulting, drainage patterns

, . and sedimentation in southwestern Turkey,

Inoue, H., Y. Fukao, K. Tanabe and Y. Ogata, , Whole , .

mantle P wave travel time tomography, Pfister, M., L. Rybach and S. Simsek, , Geothermal re-

, .

Kahle, H.G., M. Cocard, Y. Peter, A. Geiger, R. Reilinger, S.C. surface heat flow density in an area of active continen-

McClusky, R. King, A.A. Barka and G. Veis, , The tal extension, , .

GPS strain rate field in the Aegean Sea and western Provost, A.-S., J. Chery and R. Hassani, , -D mechani-

Anatolia, ( ), . cal modeling of the GPS velocity field along the North

Kahle, H.G., M. Cocard, Y. Peter, A. Geiger, R. Reilinger, A.A. Anatolian Fault, , .

Barka and G. Veis, , GPS-derived strain rate field Pucci, S., N. Palyvos, C. Zabci, D. Pantosti and M. Barchi,

within the boundary zones of the Eurasian, African, , Coseismic ruptures and tectonic landforms along

and Arabian plates, (B ), the Duzce segment of the north Anatolian fault zone

. (Ms . , November ), , doi : .

Kamiya, S. and Y. Kobayashi, , Seismological evidence / JB .

for the existence of serpentinized wedge mantle, Reilinger, R.E., S.C. McClusky, M.B. Oral, W. King and M.N.

, . Toksoz, , Global positioning system measurements

Kayal, J.R., D. Zhao, O.P. Mishra, R. De and O.P. Singh, , of present-day crustal movements in the Arabian-

The Bhuj earthquake : Tomographic evidence for Africa-Eurasia plate collision zone,

fluids at the hypocenter and its implications for rupture (B ), .

Reilinger, R.E., M.N. Toksoz, S.C. McClusky and A.A. Barka,

McClusky, S.C., , Global positioning system con- , Izmit, Turkey Earthquake was no surprise,

straints on plate kinematics and dynamics in the east- , .

ern Mediterranean and Caucasus, Rockwell, T., A.A. Barka, T. Dawson, S. Akyuz and K. Tho-

(B ), . rup, , Paleoseismology of the Gazikoy-Saros seg-

Mckenzie, D.P., , Active tectonics of the Mediterranean ment of the north Anatolian fault, northwestern Tur-

region, , . key : Comparison of the historical and paleoseismic re-

Mishra, O.P., D. Zhao, N. Umino and A. Hasegawa, , cords, implications of regional seismic hazard, and mod-

Tomography of northeast Japan forearc and its impli- els of earthquake recurre

cations for interpolate seismic coupling,

( ), , doi : . / GL .

Mitchell, J.B., Y. Pan, J. Xie and L. Cong, , coda

variation across Eurasia and its relation to crustal evo-

lution, , .

Mooney, W.D. and L.W. Braile, , The seismic structure

of the continental crust and upper mantle of north

America, In : The geology of north America : an over-

view, by A. Bally, P. Palmer (ed.), Geol. Soc. Am., Boul-

der, CO, pp. .

Muller, J.R, A. Aydin and F. Maerten, , Investigating the

transition between the Mudurnu Valley and

Izmit earthquakes along the north Anatolian fault with

static stress changes, , .

Nakajima, J., T. Matsuzawa, A. Hasegawa and D. Zhao, ,

Three-dimensional structure of Vp, Vs, and Vp/Vs be-

neath northeastern Japan : implications for arc magma-

tism and fluids, , .

Nakajima, J., Y. Takei and A. Hasegawa, , Quantitative

analysis of the inclined low-velocity zone in the mantle

wedge of northeastern Japan : a systematic change of

melt-filled pore shapes with depth and its implications

for melt migration, , .

National Earthquake Information Center (NEIC), , http :

//neic.usgs.gov/neis/epic/epic rect.html.

Panagiotopulos, D.G. and B.C. Papazachos, , Travel

times of Pn waves in the Aegean and surrounding area,

, .

Papazachos, B.C., A.S. Savvaidis, G.F. Karakaisis and C.B.

Geophys. Res. Lett.,

Tectonophysics,Marine Geology,

J. Geol.Soc. Lond.,

Phys. EarthPlanet. Inter.,

Tectonophysics,

Geophys. Res. Lett.,Earth Planet. Sci. Lett.,

J. Geophys. Res.,J. Geophys. Res.,

Geo-phys. Res. Lett.,

J. Geophys. Res.,

et al.,GSA Today,

J. Geophys. Res.,

Geophys. J. Roy Astron. Soc.,

Geophys. Res.Lett.,

Lg Q

J. Geophys. Res.,

Geophys. J. Int.,

J. Geophys. Res.,

Earth Planet. Sci. Lett.,

Geophys. J. R. Astron. Soc.,

+* +*,3 ,**, *+/+11

,**,

,**/

,+1 ,-*

+33,

-31 .,+

+33* +*-+ +*..

+332

,3. -,2

+333 11 23

,**- -

+0 ,/+- ,/+0

-0+ -11

,***

,**0

- ,--/-

,--1* 1 + +333 +*

,*** +*,3 ,**. **-/12

2+3 2,, +331

,**,

,**+

/ 332- 3333

,*** ,*** +333

+ 0

- /03/ /1,* ,**+

+31,

+*3 +2/

,**-

+0 +2/* +* +*,3 ,**- *+11-0

+331

,,101 ,,113

+323

-3 /,

,**-

+301 +333

.1+ .2,

,**+

,+2.- ,+2/1

,**/

/3 1*

,**0

+32/

+0/ +10

,3

-.1

,,+

+.3

/3

,3+

,0

,*3

+*/

+++

,1

+*,

+*

+*/

-*

-*

+*,

+/.

+*0

,-.

2*

� �

�

�

�

�

�

222


nce, , .

Salah, M.K. and D. Zhao, , -D seismic structure of Kii

Peninsula in southwest Japan : evidence for slab dehy-

dration in the forearc, , .

Sari, C. and M. Salk, , Sediment thicknesses of the

western Anatolia graben structures determined by D

and D analysis using gravity data,

, .

Saroglu, F., O. Emre and I. Kusçu, , Active fault map of

Turkey, map, Gen. Dir. of Miner. Res. and Explor., An-

kara.

Sengor, A.M.C., , The north Anatolian transform fault :

its age, o set and tectonic significance,

, .

Sengor, A.M.C. , Cross-faults and di erential stretching

of hanging walls in regions of low-angle normal fault-

ing : examples from western Turkey,

, .

Serrano, I., F. Bohoyo, J. Galindo-Zaldivar, J. Morales and D.

Zhao, a, Geophysical signatures of a basic-body

rock placed in the upper crust of the external zones of

the Betic Cordillera (Southern Spain),

( ), doi : . / GL .

Serrano, I., J. Morales, D. Zhao, F. Torcal and F. Vidal, ,

P-wave tomographic images of the central Betics Albo-

ran sea (South Spain) using local earthquakes : contri-

bution for a continental collision,

( ), - , . / GL .

Serrano, I., D. Zhao and J. Morale, b, -D crustal struc-

ture of the extensional Granada basin in the conver-

J. Seismol.,

Tectonophysics,

J. Asian Earth Sci.,

J. Geol. Soc.London,

Spec. Publ. Geol.Soc. London,

Geophys. Res.Lett.,

Geophys. Res. Lett.,

.-- ..2

,**- -

+3+ ,+-

,**0

,

-

-3 .2

» +33,

+313

#

,03 ,2,

+321 #

/1/ /23

,**,

++ +* +*,3 ,**+ *+-.21

+332

,+ .*-+ .*-. +* +*,3 +332 3***,+

,**, -

/

-0.

,0

+-0

,2

,3

,/

�

� �

�

�

�

��

��

��

�

� �

�

�

��

223

east and eastern Mediterranean, (B ),


gent boundary between the Eurasia and African plates, , .

, . Westaway, R., , Present-day kinematics of the middle

Seymen, I., , Tectonic characteristics of the North Ana-

tolian Fault Zone in the Kelkit valley, Ph.D. thesis, .

pp., Istanbul Tech. Univ., Istanbul. Zanchi, A. and J. Angelier, , Seismotectonics of western

Spakman, W. and G. Nolet, , Imaging algorithms, accu- Anatolia : regional stress orientation from geophysical

racy and resolution in delay time tomography, In : and geological data, , .

Vlaar, N. J., (Eds.), Mathematical Geophysics, D. Zhao, D., A. Hasegawa and S. Horiuchi, , Tomographic

Reidel, Norwell, MA, pp. . imaging of P and S wave velocity structure beneath

Straub, C., H.G. Kahle and C. Schindler, , GPS and geo- northeastern Japan, , .

logic estimates of the tectonic activity in the Marmara Zhao, D., A. Hasegawa and H. Kanamori, , Deep struc-

Sea region, NW Anatolia, (B ), ture of Japan subduction zone as derived from local,

. regional and teleseismic events, ,

Takei, Y., , E ect or pore geometry on Vp/Vs : from .

equilibrium geometry to crack, , . Zhao, D. and H. Kanamori, , The Northridge earth-

/ JB quake : -D crustal structure in the rupture zone and its

Thurber, C.H., , Earthquakes locations and three- relation to the aftershock locations and mechanisms,

dimensional crustal structure in the Coyote Lake area, , .

Central California, , . Zhao, D., H. Kanamori and H. Negishi, , Tomography of

Tibi, R., G. Bock, Y. Xia, M. Baumbach, H. Grosser, C. Milk- source area of the Kobe earthquake : evidence for

ereit, S. Karakisa, S. Zunbul, R. Kind and J. Zschau, , fluids at the hypocenter?, , .

Rupture process of the August Izmit and No- Zhao, D. and H. Negishi, , The Kobe earthquake :

vember , Duzce (Turkey) earthquakes, seismic image of the source zone and its implications

, F F . for the rupture nucleation, ,

Tokay, M., , Geological observation on the north Anato- .

lian fault zone between Gerede and Ilgaz, paper pre- Zhao, D., K. Wang, G.C. Rogers and S.M. Peacock, , To-

sented at Symposium on the North Anatolian Fault mographic image of low P velocity anomalies above

Zone and Earthquake Belt, Miner. Res. and Explor. Inst. slab in northern Cascadia subduction zone,

of Turkey, Ankara. , .

USGS- United States Geological Survey, Implication for Zhao D., Y. Xu, D.A. Wiens, L.M. Dormon, J. Hildebrend and

earthquake risk reduction in the United States from the J. Webb, , Depth extent of the Lau backarc spread-

Kocaeli. Turkey earthquake of August . , US Geo- ing center and its relationship to subduction processes,

logical Survey Circular : . , .

Utkucu, M., S.S. Nalbant, J. McCloskey, S. Steacy and O. Zhou, H. and R.W. Clayton, , P and S wave travel time

Alptekin, , Slip distribution and stress changes as- inversions for subducting slab under the island arcs of

sociated with the November , Duzce (Turkey) northwest Pacific, , .

earthquake (Mw . ), , . (Received April , )

Wessel, P. and W.H.F. Smith, , New improved version of (Accepted June , )

Generic Mapping Tools released,

J. Geophys. Res.,

phys. Union,Tectonophysics,

Tectonophysics,et al.

J. Geophys. Res.,

J. Geophys. Res.,J. Geophys. Res.,

J. Geophys. Res.,

Geophys. Res. Lett.,J. Geophys. Res.,

Science,

Geophys.J. Int., J. Geophys. Res.,

Earth Plan-ets Space,

Science,

J. Geophys. Res.,Geophys. J. Int.,

EOS Trans. Am. Geo-

0

/13

0+ 13 +33.

+31/

+3, +,*1+ +,*3*

+33-

+322

,/3 ,1.

+33,

+// +21

+331 +33*3 +33,2

+33.

+,

,1/21 ,10*+

,**, # ,,-+- ,,-,3

+* +33/ +33.

+*,3 ,**+ ***/,, -

+32-

10- 100

2,,0 2,-0 +330

+33/

,**+ +23+ +23.

+333 +1 +332 +33/

+, +333

+ 1 3301

+31- 3320

,**+

,2/ ,3-

+331

+1 +333

++3- +333 ,/. ,/1

+33*

,**-

+333 +, 02,3 02/+

1 + ,,3 ,.+ ,. ,**1

+332 ,1 ,**1

33

13

-..

,,,

31

+*,

33

+*1

,,

22

,1.

+.. +*-

/-

,12

3/

+/-

vol. pp. 2, ,**1 ,*3 ,,- seismic velocity structure along ... · pdf filezincan; (mu) mudurnu...

Documents