Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Seismic communication wikipedia , lookup
Mantle plume wikipedia , lookup
Shear wave splitting wikipedia , lookup
Seismic inversion wikipedia , lookup
Reflection seismology wikipedia , lookup
Magnetotellurics wikipedia , lookup
Earthquake engineering wikipedia , lookup
Seismometer wikipedia , lookup
Bull. Earthq. Res. Inst. Univ. Tokyo Vol. 2, ,**1 pp. ,*3ῌ,,- Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography Mohamed K. Salah+*, S. Sahin, and M. Kaplan- + Earthquake Research Institute, University of Tokyo, Tokyo ++-ῌ**-,, Japan Faculty of Engineering and Architecture, Suleyman Demirel University, Isparta, Turkey - General Directorate of Disaster A#airs, Earthquake Research Department, Ankara, Turkey , Abstract 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 interpretation of the imaged seismic anomalies. With the exception of high-velocity anomalies detected at a depth of 2 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 heterogeneous zones revealed by seismic tomography. These inferences are of great significance for understanding earthquake-generating processes in areas having intense seismic and tectonic activities such as the Anatolian plate. Key words : Seismic tomography, Poisson’s ratio, P wave velocity, S wave velocity, North Anatolian Fault Zone +. Introduction 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 (e.g., Aktug and Kilicoglu, ,**0), 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, +31, ; which presents a major source of risk for the popula- Barka and Gülen, +322). In this complex tectonic en- tion (Provost et al., ,**-). 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 et al., ,**0), and place this east of the Anatolian block (creating mountain ranges fault among the most active strike-slip faults world- such as the Zagros and the Caucasus), a right-lateral wide (Ambraseys, +31* ; Ambraseys and Finkel, +33/ ; 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 (McKenzie, +31, ; Sengör, +313), and a subduction zone NAFZ is characterized by frequent moderate to large associated with back arc spreading in the Aegean earthquakes (M1) with focal mechanisms that show Sea (McKenzie, +31, ; Barka and Gülen, +322) to the essentially pure right-lateral strike-slip solutions * e-mail : [email protected] 209 Mohamed K. Salah, S. Sahin and M. Kaplan Fig. +. Simplified tectonic map of Turkey, Arabian, African, and Eurasian plates (Provost et al., ,**-). 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 +1 August +333 Gölcük earthquake and +, November +333 Düzce earthquake are shown by star and cross, respectively. Abbreviations are : (K) Karliova triple junction ; (Er) Erzincan ; (Mu) Mudurnu Valley ; (IzF) Izmit fault ; (GF) Ganos fault. (e.g., Canitez and » Uçer, +301 ; McKenzie, +31, ; Zanchi years, which endangers the city of Istanbul, where and Anglier, +33-). The +333 earthquakes are the over +* million people live (e.g., Rockwell et al., ,**+ ; most recent to occur on the NAFZ in the ,*th century. Papazachos et al., ,**, ; Gürer et al., ,**-). Improving The first of these earthquakes (Gölcük earthquake) our knowledge of the seismic velocity structure of occurred on +1 August, striking the Izmit region, the crust in which these large catastrophic events west of the Marmara Sea. This Mw 1.. (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, +333 ; Barka et al., ,***, ,**,). Three months quake activity has been hampered by the sparse later, on +, November (Muller et al., ,**- ; Utkucu et coverage of seismic stations. Recently, Akyol et al. al., ,**-), the Mw 1.+ Düzce earthquake occurred. (,**0) 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 0. to the north, with a rake between age continental values. It shows four layers down to +11 and +01 (USGS, Harvard CMT (Tibi et al., ,**+)). a depth of ,3 km (the assumed Moho depth), with a This earthquake produced right-lateral surface rup- velocity of 0.,/ km/s at depths between +/ and ,+ km tures over a total length of .* km and a maximum and a velocity of 0..- km/s for the lowermost crust dextral o#set of / m (Akyüz et al., ,***, ,**,). As a (depth range ,+ῌ,3 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 (Mw1) 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 210 Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography al., ,**0). In this study we investigate three-dimen- and September +/, +333 and +*13 aftershocks of the sional velocity and Poisson’s ratio structures of the November +,, +333 Duzce earthquake (Mw1.+). These 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 belonging to the Turkish General Directorate of Disaster ,. Data A#airs (Fig. -). The remaining +0*- events are from In this study we used -1-- events that occurred background seismicity in the area, and occurred be- near or to the north of the western segment of NAFZ tween +332 and ,**- ; and were also recorded by the (Fig. ,a). We used +*/+ aftershocks of the August +1, above-mentioned seismic networks. +333 Golcuk earthquake (Mw1..) between August +1 are located between latitudes -3./ῌ.+./N and longi- These events tudes ,2ῌ-,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 +333 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 -1-- events generated ,*1*, P and +.+.2 S arrivals recorded by the +/ seismic 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 Fig. ,. (a) The epicenteral distribution of the -1-events used in this study that occurred along the northwestern side of the NAFZ. Star is the epicenter of +1 August +333 Gölcük earthquake and cross is the epicenter of +, November +333 Düzce earthquake. (b) The distribution of large earthquakes that occurred along the NAFZ (Mῌ/.*) since +32*. Thick and thin lines denote the North Anatolian Fault Zone (NAFZ) and the Eskisehir Fault Zone (EFZ), respectively. Fig. -. Locations of seismic stations used in the present study that belong to Sabonet and Turknet seismic networks. Thick and thin lines denote the North Anatolian Fault Zone (NAFZ) and the Eskisehir Fault Zone (EFZ), respectively. 211 Mohamed K. Salah, S. Sahin and M. Kaplan zones of large crustal earthquakes (Mῌ/) and the seismic velocity and Poisson’s ratio anomalies obtained, we also collected data on ,0 events that occurred in the study area starting from January +32* 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 movements along the NAFZ as explained in the previous section. -. Method The seismic tomography method was first devel- oped by Aki and Lee (+310). Subsequently, many researchers have successfully improved this technique and applied it to various regions around the world (e.g., Hirahara, +311 ; Thurber, +32- ; Spakman and Nolet, +322 ; Zhou and Clayton, +33*, Zhao et al., +33,, +33. ; Nakajima et al., ,**+ ; Mishra et al., ,**-). In 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 direction and 0ῌ+/ km in the depth direction. Straight lines in (a) show the location of vertical crosssections in Figs. +-ῌ+0. this study, we use the tomographic method of Zhao et al. (+33,), which has been applied to many regions having di#erent tectonic circumstances (e.g., Zhao and Kanamori, +33/ ; Zhao et al., +330, +331, ,**+ ; Serrano et al., +332, ,**,a, b ; Kayal et al., ,**,). 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 and 0ῌ VpῌVs,,+sῌ+, s +/ km in the depth direction (Fig. .). Velocities at grid nodes are taken as unknown parameters, and to compute the Poisson’s ratio (s). 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 et al. (+33,, +33.). stribution of the velocity anomalies obtained. We The interpretation of tomographic images is usu- adopted a crustal velocity model that is slightly ally nonunique, because any given velocity anom- di#erent from the model determined by Akyol et al. aly can be attributed to either a thermal or a chemi- (,**0) 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 ,3 km (the assumed Moho depth), with a reliable interpretation of the anomalies obtained in velocity of 0.,/ km/s at depths between +/ and ,+ km terms of tectonic and geodynamic implications. Com- and a velocity of 0..- km/s for the lowermost crust pared to the seismic velocity itself, the Poisson’s ratio (depth range ,+ῌ,3 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, +332 ; Kayal strained due to the lack of head waves and refracted et al., ,**, ; Takei, ,**, ; Salah and Zhao, ,**- ; Naka- waves at the surface in the data set they used. The jima et al., ,**+, ,**/) or serpentinization (Kamiya model we used in our study, on the other hand, is and Kobayashi, ,*** ; Christensen, +330). Therefore, even simpler. We assumed velocities of /, /.3, 0.*, 0.//, after Vp and Vs models are calculated from travel and 1.1/ km/s at depths of /, +*, +/, ,/, and .* km, 212 Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography 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 *.,1 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. Fig. /. Initial P wave (solid line) and S wave (dashed line) velocity models adopted for the present study. Figures 2 and 3 show velocity perturbations relative to the initial velocity models of the P and S waves at depths of ,, 2, +/, 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 : VsVp/+.1-, to derive the initial ters of the two large +333 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 (Mῌ/) and the imaged anomalies, we also plot their epicentral distribution superimposed .. Resolution and Results on the velocity and Poisson’s ratio structures at 2 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 et al., +33* ; Zhao et al., +33,, the following paragraphs, we describe the main fea- +33.) 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 0῍. 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 structures at four crustal depth slices are shown in (Figs. 2, 3, and +*). This heterogeneous structure continues down to a depth of 2 km, but is dominated Figs. 0 and 1, 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 2, 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. 0 and 1). 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). su$cient lengths and directions of both horizontal higher than average at a depth of +/ km, but is gener- 213 Poisson’s ratio is Mohamed K. Salah, S. Sahin and M. Kaplan Fig. 0. 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 velocities, respectively. Perturbation scale is shown at the lower right. Fig. 1. The same as Fig. 0, but for S wave velocity. ῎ 214 ῎ Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography Fig. 2. P wave velocity perturbations at depths of ,, 2, +/, 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), respectively. Star and cross denote the epicenters of the +1 August +333 Gölcük and the +, November +333 Düzce earthquakes, respectively. Fig. 3. The same as Fig. 2, but for S wave velocity. ῎ 215 ῎ Mohamed K. Salah, S. Sahin and M. Kaplan Fig. +*. Distribution of Poisson’s ratio perturbations at four depth slices. Light and dark gray colors denote high and low Poisson’s ratios, respectively. Other details are similar to those of Fig. 2. ally low at a depth of ,/ km with the exception of the +333 Golcuk earthquake, while cross-section DD’ pas- zone near the epicentral area of the +1 August, +333 ses through the hypocenter of the +, November, +333 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 2 and +/ the two large +333 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 2 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 ratios at the two depth slices. These results imply that /. Discussion /. + Large earthquakes along the NAFZ 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- scribed before are also illustrated along vertical cross- ing Anatolian block (McKenzie, +31, ; Sengör, +313 ; sections AA’, BB’, CC’, and DD’ (Figs. +-ῌ+0), which run Barka, +33, ; Saroglu et al., +33,). Since the Middle/ either in the E-W or N-S directions (see Fig. . for the Late Miocene (+-ῌ/ Ma) the NAFZ has accumulated a location of cross-sections). Cross-sections AA’ and geologic displacement on the order of 2/ῌ+,* km CC’ pass through the hypocenter of the +1 August, (Seymen, +31/ ; Sengör, +313 ; Barka, +32+ ; Barka and ῎ 216 ῎ Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography Fig. +,. Fig. ++. P wave velocity (a), S wave velocity (b), and Poisson’s ratio (c) perturbations at a depth of 2 km together with the epicenteral distribution of large earthquakes (see text for details). Other details are similar to those of Fig. 2. The same as Fig. ++, but at a depth of +/ km. two main strands, the Duzce and the Mudurnu fault segments, where GPS data indicate that the former accommodates up to +* mm/yr (Ayhan et al., +333). Farther west, the NAFZ splays again into three ma- Hancock, +32. ; Westaway, +33. ; Armijo et al., +333 ; jor strands and the northernmost one is consider- Hubert-Ferrari et al., ,**,). 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, +322 ; Straub et al., +331). rate of *./ῌ*.2 cm/yr (Tokay, +31- ; Seymen, +31/ ; The preceding paragraphs show a high rate of Barka and Hancock, +32.) and +.2 cm/yr (Hubert- seismic activity along the NAFZ, which is also clear Ferrari et al., ,**,), respectively. from the distribution of the seismic events used in Conversely, GPS 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 +333 earthquakes (see section ,). The ῏,ῌ- cm/yr (Reilinger et al., +331, ,*** ; Straub et al., highly heterogeneous seismic velocity and Poisson’s +331 ; McClusky et al., ,*** ; Kahle et al., +333, ,***). 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- ῎ 217 ῎ Mohamed K. Salah, S. Sahin and M. Kaplan Fig. +-. Vertical cross-sections of P wave velocity, S wave velocity, and Poisson’s ratio structures along line AA’ (see Fig. . for the location of the crosssections). 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 +1 August +333 Golcuk earthquake and big circles denote large earthquakes. Crosses show the distribution of seismicity in a +*-km-wide zone around the profile. The perturbation scale is shown to the right of each panel. Fig. +.. The same as Fig. +-, but along line BB’. and are certainly of value for identifying and characterizing faults with potential for surface rupturing earthquakes. /. , Previous geophysical observations and obtained velocity and Poisson’s ratio models 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 2 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. tures, combining the results of seismic tomography (,**0) obtained lower crustal velocities in western Akyol et al. 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 0.,/ km/s at a depth of ,+ km to only 0..- ῎ 218 ῎ Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography Fig. +/. The same as Fig. +-, but along line CC’. Fig. +0. The same as Fig. +-, but along line DD’. The big cross denotes the hypocenter of the +, November +333 Düzce 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, +33/). These low velocities might indi- lid zones, whereas the very low Pn 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 (e.g., Al- though we have no resolution at depths greater than Shukri and Mitchell, +322 ; Mitchell et al., +331). It ,/ km, these observations, however, support the exis- was also found that the average Pn velocity for tence of the low velocities and high Poisson’s ratios the entire Aegean region is approximately 1.3 km/s anomalies we detected at depths greater than +/ km (Panagiotopoulos and Papazachos, +32/), which is (Figs. +-ῌ+0). Seismic activity is concentrated mostly lower than the worldwide average continental upper in the top +* km depths, which might suggest the mantle Pn velocity of 2.+ km/s (Moony and Braile, presence of seismically active low-angle breakaway +323). faults within the upper crust of western Anatolia Recently, Al-Lazki et al. (,**.) detected broad- (Sengör, +321). scale (/** km) zones of low (2 km/s) Pn velocity Measurements of coda Qc by Akinci et al. (+33.) anomalies underlying the Anatolian plate and the in western Anatolia for the frequency range +./ῌ+* Hz Anatolian plateau, and even smaller-scale ((,** show a strong frequency dependence, which agrees km), very low (1.2 km/s) Pn 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- 219 Mohamed K. Salah, S. Sahin and M. Kaplan 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 2 km should be considered. All these anomalies are associ- depth) velocities we observed in our study. ated with the complex and active tectonic setting of High average heat flow with geothermal activity (e.g., Ilk- the region, the circulation of hot geothermal fluids, isik, +33/ ; Pfister et al., +332 ; 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 (e.g., Paton, +33, ; Innocenti et previous sections suggest that the generation of a al., ,**/) 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 (,**0) 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 (.ῌ0 W m) zones field, which may be detected by seismic tomography at depths of /ῌ+/ 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 (*.,2ῌ*./, W 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 /* g 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. 0. Conclusions In this study we collected a number of arrival Acknowledgements times of P and S waves from the aftershocks of the We thank M. Uyeshima and two anonymous re- two large +333 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 ter (NEIC), (http : //neic.usgs.gov/neis/epic/epic_rect. northwest Anatolia. The velocity structure is very html) for facilitating the availability of earthquake heterogeneous at shallow layers of the crust (, and 2 catalogs and seismological data on the web. All fig- km depths) with amplitudes of 0ῌ. 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 2 km. At the deepest layers of Smith (+332). the crust (depths +/ and ,/ km) most parts of the 220 Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography References Aki, K. and W. Lee, +310, Determination of the threedimensional velocity anomalies under a seismic array using first P arrival times from local earthquakes, +.A homogeneous initial model, J. Geophys. Res., 2+, .-2+ῌ .-33. Akinci, A., A.G. Taktak and S. Ergintav, +33., Attenuation of coda waves in western Anatolia, Phys. Earth Planet. Inter., 21, +//ῌ+0/. Aktug, B. and A. Kiliçoglu, ,**0, Recent crustal deformation of Izmir, western Anatolia and surrounding regions as deduced from repeated GPS measurements and strain field, J. Geodyn., .+, .1+ῌ.2.. Akyol, N., L. Zhu, B. J. Mitchell, H. Sözbilir and K. Kekovalý, ,**0, Crustal structure and local seismicity in western Anatolia, Geophys. J. Int., +*.++++/j.+-0/-,.0X.,**0.*-*/-. Akyüz, H.S., A.A. Barka, E. Altunel, R.D. Hartleb and G. Sunal, ,***, Field observations and slip distribution of the November +,, +333 Düzce earthquake (M῏1.+), BoluTurkey, in : the +333 Izmit and Düzce earthquakes : Preliminary results, by A.A. Barka et al. (Ed.), pp. 0-ῌ1*, Istanbul Tech. Univ., Istanbul. Akyüz, H.S., R.D. Hartleb, A.A. Barka, E. Altunel, G. Sunal, B. Meyer and R. Armijo, ,**,, Surface rupture and slip distribution of the +, November +333 Düzce earthquake (M 1.+), north Anatolian fault, Bolu, Turkey, Bull. Seismol. Soc. Am., 3,, 0+ῌ00. Al-Lazki, A.I., E. Sandvol, D. Seber, M. Barazangi, N. Turkelli and R. Mohamad, ,**., Pn tomographic imaging of mantle lid velocity and anisotropy at the junction of the Arabian, Eurasian and African plates, Geophys. J. Int., +/2, +*,.ῌ+*.*. Al-Shukri, H. J. and B. J. Mitchell, +322, Reduced seismic velocities in the source zone of new Madrid earthquakes, Bull. Seismol. Soc. Am., 12, +.3+ῌ+/*3. Ambraseys, N.N., +31*, Some characteristic features of the north Anatolian fault zone, Tectonophysics, 3, +.-ῌ+0/. Ambraseys, N.N., ,**,, The seismic activity of the Marmara sea region over the last ,*** years, Bull. Seismol. Soc. Am., 3,, +ῌ+2. Ambraseys, N.N. and C.F. Finkel, +33/, The seismicity of Turkey and adjacent areas : a historical review, +/**ῌ +2**. Muhittin Salih Eren, Istanbul, ,.* pp. Armijo, R., B. Meyer, A. Hubert and A.A. Barka, +333, Westward propagation of the North Anatolian fault into the northern Aegean : Timing and kinematics, Geology, ,1, ,01ῌ,1*. Ayhan, M.A., C. Demir, A. Kilicoglu, I. Sanli and S.M. Nakiboglu, +333, Crustal motion around the western segment of the north Anatolian fault zone : Geodetic measurements and geophysical interpretation, paper presented at IUGG33, Int. Union of Geodesy and Geophys., Birmingham, U.K. Barka, A.A., +32+, Seismo-tectonic aspects of the North Anatolian fault zone, Ph.D. thesis, --/ pp., Univ. of Bristol, Bristol, U.K. Barka, A.A., +33,, The North Anatolian fault zone, Ann, Tecton., 0, +0.ῌ+3/. Barka, A.A., +333, The +1 August +333 Izmit Earthquake, Science, ,2/, +2/2ῌ+2/3. Barka, A.A., H.S. Akyüz, E. Altunel, G. Sunal, Z. Çakir, A. Dikbas, B. Yerli, R. Armijo, B. Meyer, J.B. de Chabalier, T. Rockwell, J.R. Dolan, R. Hartleb, T. Dawson, S. Christo#erson, A. Tucker, T. Furnal, R. Langridge, H. Stenner, W. Lettis, J. Bachhuber and W. Page, ,**,, The surface rupture and slip distribution of the +1 August +333 Izmit earthquake (M 1..), north Anatolian fault, Bull. Seismol. Soc. Am., 3,, .-ῌ0*. Barka, A.A., H.S. Akyüz, H.A. Cohen and F. Watchorn, ,***, Tectonic evolution of the Niksar and Tasova-Erbaa pull-apart basins, north Anatolian fault zone : their significance for the motion of the Anatolian block, Tectonophysics, -,,, ,.-ῌ,0.. Barka, A.A. and L. Gülen, +322, New constraints on age and total o#set of the north Anatolian fault zone : Implications for tectonics of the eastern Mediterranean Region, In : Melih Tokay Symposium, Spec. Publ. Middle-east Techn. Univ., Ankara, pp. -3ῌ0/. Barka, A.A. and P.L. Hancock, +32., Neotectonic deformation patterns in the convex-northwards arc of the North Anatolian fault, in The Geological Evolution of the Eastern Mediterranean, edited by J.G. Dixon and A. H.F. Robertson, Geol. Soc. Spec. Publ., +1, 10-ῌ11-. Barka, A.A. and K. Kadinski-Cade, +322, Strike-slip fault geometry in Turkey and its influence on earthquake activity, Tectonophysics, 1, 00-ῌ02.. Bozkurt, E., ,**+, Neotectonics of Turkey῎a synthesis, Geodyn. Acta, +., -ῌ-*. Çaglar, I., ,**+, Electrical resistivity structure of the northwestern Anatolia and its tectonic implications for the Sakarya and Bornova zones, Phys. Earth Planet. Inter., +,/, 3/ῌ++*. Çaglar, I. and T. Isseven, ,**., Two-dimensional geoelectrical structure of the Göynük geothermal area, northwest Anatolia, Turkey, J. Volcanol. Geother. Res., +-., +2-ῌ +31. Canitez, N. and B. » Uçer, +301, Computer determinations for the fault plane solutions in and near Anatolia, Tectonophysics, ., ,-/ῌ,... Christensen, N., +330, Poisson’s ratio and crustal seismology, J. Geophys. Res., +*+, -+-3ῌ-+/0. Christensen, N. and W. Mooney, +33/, Seismic velocity structure and composition of the continental crust : a global view, J. Geophys. Res., +**, 310+ῌ3122. Gemici, » U. and G. Tarcan, ,**,, Hydrogeochemistry of the Simav geothermal field, western Anatolia, Turkey, J. Volcanol. Geother. Res., ++0, ,+/ῌ,--. Gürer, » O.F., N. Kaymakçi, S. Çakir and M. » Ozburan, ,**-, Neotectonics of the southeast Marmara region, NW Anatolia, Turkey, J. Asian Earth Sci., ,+, +*.+ῌ+*/+. Harvard, +310ῌ,**,, CMT catalogue. http : //www.seismolo gy.harvard.edu/ Hirahara, K., +311, A large-scale three-dimensional seismic structure under the Japan islands and the Sea of Japan, J. Phys. Earth, ,/, -3-ῌ.+1. Hubert-Ferrari, A., R. Armijo, G. King, B. Meyer and A.A. Barka, ,**,, Morphology, displacement, and slip rates along the northern Anatolian Fault, Turkey, J. Geophys. Res. +*1 (B+*), ,,-/, doi : +*.+*,3/,**+JB***-3-. Ilkisik, O.M., +33/, Regional heat flow in western Anatolia using silica temperature estimates from thermal springs, Tectonophysics, ,.., +1/ῌ+2.. ῎ 221 ῎ Mohamed K. Salah, S. Sahin and M. Kaplan Innocenti, F., S. Agostini, G.D. Vincenzo, C. Doglioni, P. Manetti, M.Y. Savasçin and S. Tonarini, ,**/, Neogene and Quaternary volcanism in western Anatolia : Magma sources and geodynamic evolution, Marine Geology, ,,+, -31ῌ.,+. Inoue, H., Y. Fukao, K. Tanabe and Y. Ogata, +33*, Whole mantle P wave travel time tomography, Phys. Earth Planet. Inter., /3, ,3.ῌ-,2. Kahle, H.G., M. Cocard, Y. Peter, A. Geiger, R. Reilinger, S.C. McClusky, R. King, A.A. Barka and G. Veis, +333, The GPS strain rate field in the Aegean Sea and western Anatolia, Geophys. Res. Lett., ,0 (+0), ,/+-ῌ,/+0. Kahle, H.G., M. Cocard, Y. Peter, A. Geiger, R. Reilinger, A.A. Barka and G. Veis, ,***, GPS-derived strain rate field within the boundary zones of the Eurasian, African, and Arabian plates, J. Geophys. Res., +*/ (B-), ,--/-ῌ ,--1*. Kamiya, S. and Y. Kobayashi, ,***, Seismological evidence for the existence of serpentinized wedge mantle, Geophys. Res. Lett., ,1, 2+3ῌ2,,. Kayal, J.R., D. Zhao, O.P. Mishra, R. De and O.P. Singh, ,**,, The ,**+ Bhuj earthquake : Tomographic evidence for fluids at the hypocenter and its implications for rupture nucleation, Geophys. Res. Lett., ,3, +*.+*,3/ ,**,GL*+/+11. McClusky, S.C., et al., ,***, Global positioning system constraints on plate kinematics and dynamics in the eastern Mediterranean and Caucasus, J. Geophys. Res., +*/ (B-), /03/ῌ/1,*. Mckenzie, D.P., +31,, Active tectonics of the Mediterranean region, Geophys. J. Roy Astron. Soc., -*, +*3ῌ+2/. Mishra, O.P., D. Zhao, N. Umino and A. Hasegawa, ,**-, Tomography of northeast Japan forearc and its implications for interpolate seismic coupling, Geophys. Res. Lett., -* (+0), +2/*, doi : +*.+*,3/,**-GL*+11-0. Mitchell, J.B., Y. Pan, J. Xie and L. Cong, +331, Lg coda Q variation across Eurasia and its relation to crustal evolution, J. Geophys. Res., +*,, ,,101ῌ,,113. Mooney, W.D. and L.W. Braile, +323, The seismic structure of the continental crust and upper mantle of north America, In : The geology of north America : an overview, by A. Bally, P. Palmer (ed.), Geol. Soc. Am., Boulder, CO, pp. -3ῌ/,. Muller, J.R, A. Aydin and F. Maerten, ,**-, Investigating the transition between the +301 Mudurnu Valley and +333 Izmit earthquakes along the north Anatolian fault with static stress changes, Geophys. J. Int., +/., .1+ῌ.2,. Nakajima, J., T. Matsuzawa, A. Hasegawa and D. Zhao, ,**+, Three-dimensional structure of Vp, Vs, and Vp/Vs beneath northeastern Japan : implications for arc magmatism and fluids, J. Geophys. Res., +*0, ,+2.-ῌ,+2/1. 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, Earth Planet. Sci. Lett., ,-., /3ῌ1*. National Earthquake Information Center (NEIC), ,**0, http : //neic.usgs.gov/neis/epic/epic_rect.html. Panagiotopulos, D.G. and B.C. Papazachos, +32/, Travel times of Pn waves in the Aegean and surrounding area, Geophys. J. R. Astron. Soc., 2*, +0/ῌ+10. Papazachos, B.C., A.S. Savvaidis, G.F. Karakaisis and C.B. Papazachos, ,**,, Precursory accelerating seismic crustal deformation in the northwestern Anatolia fault zone, Tectonophysics, -.1, ,+1ῌ,-*. Paton, S., +33,, Active normal faulting, drainage patterns and sedimentation in southwestern Turkey, J. Geol. Soc. Lond., +.3, +*-+ῌ+*... Pfister, M., L. Rybach and S. Simsek, +332, Geothermal reconnaissance of the Marmara Sea region (NW Turkey) : surface heat flow density in an area of active continental extension, Tectonophysics, ,3+, 11ῌ23. Provost, A.-S., J. Chéry and R. Hassani, ,**-, --D mechanical modeling of the GPS velocity field along the North Anatolian Fault, Earth Planet. Sci. Lett., ,*3, -0+ῌ-11. Pucci, S., N. Palyvos, C. Zabci, D. Pantosti and M. Barchi, ,**0, Coseismic ruptures and tectonic landforms along the Duzce segment of the north Anatolian fault zone (Ms 1.+, November +333), J. Geophys. Res., +++, doi : +*. +*,3/,**.JB**-/12. Reilinger, R.E., S.C. McClusky, M.B. Oral, W. King and M.N. Toksöz, +331, Global positioning system measurements of present-day crustal movements in the ArabianAfrica-Eurasia plate collision zone, J. Geophys. Res., +*, (B/), 332-ῌ3333. Reilinger, R.E., M.N. Toksöz, S.C. McClusky and A.A. Barka, ,***, +333 Izmit, Turkey Earthquake was no surprise, GSA Today, +*, +ῌ0. Rockwell, T., A.A. Barka, T. Dawson, S. Akyüz and K. Thorup, ,**+, Paleoseismology of the Gazikoy-Saros segment of the north Anatolian fault, northwestern Turkey : Comparison of the historical and paleoseismic records, implications of regional seismic hazard, and models of earthquake recurrence, J. Seismol., /, .--ῌ..2. Salah, M.K. and D. Zhao, ,**-, --D seismic structure of Kii Peninsula in southwest Japan : evidence for slab dehydration in the forearc, Tectonophysics, -0., +3+ῌ,+-. Sari, C. and M. Salk, ,**0, Sediment thicknesses of the western Anatolia graben structures determined by ,D and -D analysis using gravity data, J. Asian Earth Sci., ,0, -3ῌ.2. Saroglu, F., » O. Emre and I. Kusçu, +33,, Active fault map of Turkey, map, Gen. Dir. of Miner. Res. and Explor., Ankara. Sengör, A.M.C., +313, The north Anatolian transform fault : its age, o#set and tectonic significance, J. Geol. Soc. London, +-0, ,03ῌ,2,. Sengör, A.M.C. +321, Cross-faults and di#erential stretching of hanging walls in regions of low-angle normal faulting : examples from western Turkey, Spec. Publ. Geol. Soc. London, ,2, /1/ῌ/23. 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), Geophys. Res. Lett., ,3 (++), doi : +*.+*,3/,**+GL*+-.21. Serrano, I., J. Morales, D. Zhao, F. Torcal and F. Vidal, +332, P-wave tomographic images of the central Betics Alboran sea (South Spain) using local earthquakes : contribution for a continental collision, Geophys. Res. Lett., ,/ (,+), .*-+-.*-., +*.+*,3/+332GL3***,+. Serrano, I., D. Zhao and J. Morale, ,**,b, --D crustal structure of the extensional Granada basin in the conver- ῎ 222 ῎ Seismic Velocity Structure along the Western Segment of the North Anatolian Fault Zone Imaged by Seismic Tomography gent boundary between the Eurasia and African plates, Tectonophysics, -.., 0+ῌ13. Seymen, I., +31/, Tectonic characteristics of the North Anatolian Fault Zone in the Kelkit valley, Ph.D. thesis, +3, pp., Istanbul Tech. Univ., Istanbul. Spakman, W. and G. Nolet, +322, Imaging algorithms, accuracy and resolution in delay time tomography, In : Vlaar, N. J., et al. (Eds.), Mathematical Geophysics, D. Reidel, Norwell, MA, pp. +//ῌ+21. Straub, C., H.G. Kahle and C. Schindler, +331, GPS and geologic estimates of the tectonic activity in the Marmara Sea region, NW Anatolia, J. Geophys. Res., +*, (B+,), ,1/21ῌ,10*+. Takei, Y., ,**,, E#ect or pore geometry on Vp/Vs : from equilibrium geometry to crack, J. Geophys. Res., +*1, +*. +*,3/,**+JB***/,, Thurber, C.H., +32-, Earthquakes locations and threedimensional crustal structure in the Coyote Lake area, Central California, J. Geophys. Res., 22, 2,,0ῌ2,-0. Tibi, R., G. Bock, Y. Xia, M. Baumbach, H. Grosser, C. Milkereit, S. Karakisa, S. Zünbül, R. Kind and J. Zschau, ,**+, Rupture process of the +333 August +1 Izmit and November +,, +333 Düzce (Turkey) earthquakes, Geophys. J. Int., +.., F+ῌF1. Tokay, M., +31-, Geological observation on the north Anatolian fault zone between Gerede and Ilgaz, paper presented at Symposium on the North Anatolian Fault Zone and Earthquake Belt, Miner. Res. and Explor. Inst. of Turkey, Ankara. USGS- United States Geological Survey, Implication for earthquake risk reduction in the United States from the Kocaeli. Turkey earthquake of August +1.+333, US Geological Survey Circular ++3- : +333. Utkucu, M., S.S. Nalbant, J. McCloskey, S. Steacy and O. Alptekin, ,**-, Slip distribution and stress changes associated with the +333 November +,, Duzce (Turkey) earthquake (Mw῏1.+), Geophys. J. Int., +/-, ,,3ῌ,.+. Wessel, P. and W.H.F. Smith, +332, New improved version of Generic Mapping Tools released, EOS Trans. Am. Geo- phys. Union, 13, /13. Westaway, R., +33., Present-day kinematics of the middle east and eastern Mediterranean, J. Geophys. Res., 33 (B0), +,*1+ῌ+,*3*. Zanchi, A. and J. Angelier, +33-, Seismotectonics of western Anatolia : regional stress orientation from geophysical and geological data, Tectonophysics, ,,,, ,/3ῌ,1.. Zhao, D., A. Hasegawa and S. Horiuchi, +33,, Tomographic imaging of P and S wave velocity structure beneath northeastern Japan, J. Geophys. Res., 31, +33*3ῌ+33,2. Zhao, D., A. Hasegawa and H. Kanamori, +33., Deep structure of Japan subduction zone as derived from local, regional and teleseismic events, J. Geophys. Res., 33, ,,-+-ῌ,,-,3. Zhao, D. and H. Kanamori, +33/, The +33. Northridge earthquake : --D crustal structure in the rupture zone and its relation to the aftershock locations and mechanisms, Geophys. Res. Lett., ,,, 10-ῌ100. Zhao, D., H. Kanamori and H. Negishi, +330, Tomography of source area of the +33/ Kobe earthquake : evidence for fluids at the hypocenter?, Science, ,1., +23+ῌ+23.. Zhao, D. and H. Negishi, +332, The +33/ Kobe earthquake : seismic image of the source zone and its implications for the rupture nucleation, J. Geophys. Res., +*-, 3301ῌ 3320. Zhao, D., K. Wang, G.C. Rogers and S.M. Peacock, ,**+, Tomographic image of low P velocity anomalies above slab in northern Cascadia subduction zone, Earth Planets Space, /-, ,2/ῌ,3-. Zhao D., Y. Xu, D.A. Wiens, L.M. Dormon, J. Hildebrend and J. Webb, +331, Depth extent of the Lau backarc spreading center and its relationship to subduction processes, Science, ,12, ,/.ῌ,/1. Zhou, H. and R.W. Clayton, +33*, P and S wave travel time inversions for subducting slab under the island arcs of northwest Pacific, J. Geophys. Res., 3/, 02,3ῌ02/+. (Received April ,., ,**1) (Accepted June ,1, ,**1) ῎ 223 ῎