- Email: [email protected]
Crustal thickness variations in the Eastern Mediterranean and southern Aegean region
Marine and Petroleum Geology 77 (2016) 190e197
Contents lists available at ScienceDirect
Marine and Petroleum Geology journal homepage: www.elsevier.com/locate/marpetgeo
Research paper
Crustal thickness variations in the Eastern Mediterranean and southern Aegean region Funda Bilim a, Attila Aydemir b, *, Abdullah Ates c a
Engineering Faculty, Geophysical Engineering Department, Cumhuriyet University, 05480 Sivas, Turkey Turkish Petroleum Corp., Sogutozu Mah., 2180. Cad., No:10, 06530 Sogutozu, Ankara, Turkey c Faculty of Engineering, Department of Geophysical Eng., Ankara University, 06100 Besevler, Ankara, Turkey b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 October 2015 Received in revised form 11 June 2016 Accepted 15 June 2016 Available online 18 June 2016
In this paper, regional analog gravity anomaly map obtained from the General Directorate of Mineral Research and Exploration (MTA) was digitized and used for the calculation of the crustal thickness (Moho depth) variations in the Eastern Mediterranean and the southern part of the Aegean Region. In the gravity anomaly map, there are mainly EeW trending apparent gravity anomalies represented by the contours up to 150 mGal. They are generally parallel to the shorelines of Africa, Turkey and Crete. Crustal thickness variations were calculated from the gravity anomalies, using an empirical equation in this study. Obtained thicknesses (Moho depths) were mapped and correlated with the previous investigations and seismological findings. According to the estimations, crustal thicknesses are about 25 e30 km along the coastal regions and more than 30 km on the onshore part of Turkey increasing up to 42 km through the eastern Anatolia. However, there are thin crustal zones around 17 km in the offshore Egypt, to the NW part of Cyprus and about 19 km to the north of Crete. They may be related with the main tectonic trends in this region except the circular thinning to the south of Kas (southwestern part of Turkey). In order to determine the locations and boundaries of prominent tectonic elements, Analytic Signal (AS) and maxspots maps of the gravity anomalies were also prepared in this study. All produced maps are generally consistent to each other and the boundaries of main tectonic units were apparently illustrated in the maxspots map from the horizontal gradient of Bouguer anomalies. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Eastern mediterranean Aegean sea Western anatolia Crustal thickness Analytic signal Maxspots map
1. Introduction Tectonic framework of the Eastern Mediterranean is dominated by the subduction of the African Plate beneath the Eurasian Plate (Fig. 1). In addition, the Anatolian Plate has been converging through the Aegean Sea and Greece with the counter-clockwise rotational tectonic escapement (30e40 mm/y; Le Pichon et al., 1995). This movement is controlled by two strike-slip fault systems (Fig. 1); the North Anatolian Fault-NAF and the East Anatolian Fault-EAF (Ketin, 1948; McKenzie, 1972; Le Pichon and Angelier, 1979; Sengor and Yilmaz, 1981). The African Plate subduction beneath the Anatolian Block triggered N-S directional extension in the western Anatolia (Sengor and Yilmaz, 1981) and crustal thickness was found about 34 km to the east and 25 km to the further west of the onshore Aegean region (Ates et al., 2012) under the
* Corresponding author. E-mail addresses: [email protected], [email protected] (A. Aydemir). http://dx.doi.org/10.1016/j.marpetgeo.2016.06.012 0264-8172/© 2016 Elsevier Ltd. All rights reserved.
influence of extension. Tectonics of the Aegean zone can be summarized by its rapid extension characteristics that allows to consider this region as one of the world’s most rapidly extending crustal thinning zone (the extension rate was given as 14 ± 5 mm/yr by Reilinger et al., 1997; McClusky et al., 2000). Tectonic model of the Eastern Mediterranean can be explained by the subduction of the African Plate dipping north beneath the Aegean Sea (Eurasian Plate) through the Hellenic and Cyprian arcs from west and to the east, respectively (McKenzie, 1978; Le Pichon and Angelier, 1979; Le Pichon, 1982). The most active part of the subduction is observed around the southern Aegean or Hellenic trench. Tomographic results indicated a high-velocity anomaly dipping north, down to 600 km beneath the Hellenic arc and it was interpreted as the subducting lithosphere of the African Plate (Wortel et al., 1990). Crustal thicknesses in the study area were calculated previously for very limited zones, mainly based on the seismological studies. However, there is no investigation to cover entire region for the Eastern Mediterranean, whether based on seismological
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
191
Fig. 1. The major tectonic elements in and around Turkey, and Eastern Mediterranean region (modified from Bozkurt and Mittwede, 2005). The large arrows show relative motions of the Anatolian Block and convergent motions of the African Plate and Arabian Platform. NAFZ: North Anatolian Fault Zone, DSFZ: Dead Sea Fault Zone, NEAFZ: North-East Anatolian Fault Zone, EAFZ: East Anatolian Fault Zone, IASZ: Izmir-Ankara-Erzincan Suture Zone.
observations or estimations from the gravity data. Zelt et al. (2005) performed a 2D inversion of refraction and reflection traveltimes along an axial seismic profile through Gulf of Corinth, Greece and found the Moho depth at 29 km to the east and 39 km to the west of the Corinth strait. Crustal thickness to the east of the Corinth is consistent with the results of Sodoudi et al. (2006) where they calculated the Moho depths using the P and S receiver functions. In their research, they found shallow Moho depths in the Aegean Sea to the north of Crete, varying between 19 and 26 km. Van der Meijde et al. (2003) also analysed receiver functions below 17 seismological stations in the Mediterranean region and North Africa. Three stations in their investigations are located in the area of our study and they found Moho depth 25 ± 1.4 km below the station KOUM to the south of Izmir, 29 ± 2.3 km below APER station to the east of Crete and 32 ± 1.1 km below the station named KEG to the south of Port Said. Di Luccio and Pasyanos (2007) used surface wave dispersion curves in the analysis and found shallower crust in the Ionian Sea, southern Italy (8e16 km) and the central part of the Eastern Mediterranean (16e24 km). Their calculations in the southern, central-western Aegean are in the range of 20e25 km and consistent with this study. Recently, ElGabry et al. (2013) investigated the crustal thickness and mantle structure of the Eastern Mediterranean using the non-linear inversion of Rayleigh wave dispersion curves extracted from the broad-band records on the profiles along the Hellenic and Cyprian arcs. In the first profile (Hellenic arc profile), they found the average thickness around 31 km and they stated that the Moho depth decreases gradually down to 21 km under the Aegean Sea, to the north of Crete. In the Cyprian arc profile, the Moho was found at a depth of 37 km in the Sinai Peninsula to the south, rapidly rising to 21 km below the Levantine Basin (to the south of Cyprus) that is known as the
Herodotus Abyssal Plain as well. Similar to the seismological investigations, crustal thickness calculations based on the gravity data or gravity modeling were performed for the limited zones or regions in the Aegean region, Turkey, Middle East and North Africa. A study covering large area was accomplished by Seber et al. (2001), attempting to develop a crustal model for the Middle East and North Africa region. In their study, they presented a 3D crustal model after they integrated and interpolated rock thicknesses and Moho depth measurements in this part of the world. The Eastern Mediterranean region was represented by 20e25 km Moho depths in general. The shallowest zone was given with the 10e15 km interval to the south of Crete and the Herodotus Abyssal Plain. Makris and Stobbe (1984) also created a crustal thickness map of the area interested, deduced from the geophysical data. They found two different shallowest zones; one of them is located to the NW of Alexandria and the second one to the NW of Cyprus about 22 km. The other shallow zone to the north of Crete is about 24 km in their map. Casten and Snopek (2006) compiled the gravity data from the land, marine and satellite sources in the area including the Hellenic subduction zone and they obtained a Bouguer anomaly map range from þ170 mGal to 10 mGal. After the 3D gravity modeling study, they found the Moho depths less than 20 km under the Cretan Sea (to the north of Crete) and it gets deeper to the south of the island (around 30 km). They proposed extremely thick sedimentary cover (up to 18 km) around the Mediterranean Ridge and the Herodotus Abyssal Plain. Tirel et al. (2004) calculated maximum thinning of the crust to the north of Crete (around 22 km) and they estimated an average thickness of 25 km to the north of Cretan Sea (the area of Cyclades islands) where the rigid block-type behaviour was suggested during the post-Miocene times. Staackmann et al. (2008) found similar results (20 km of Moho depth beneath the Cretan Sea) with Tirel
192
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
et al. (2004) and they observed a crustal thinning from west to the east. Ergun et al. (2005) modelled the gravity anomalies in twodimensions (2D) along with four different profiles to the south of Turkey and, only easternmost profile D is consistent with this study. Cowie and Kusznir (2012) mapped crustal thickness and oceanic lithosphere distribution in the Eastern Mediterranean using the gravity inversion by considering the relation of this region under the break-up of West and Central African Rift System. Their calculations were based on two different break-up ages; 225 Ma (late Triassic) and 100 Ma (mid-Cretaceous). They suggest a new rifting in the Levantine Basin where crustal thickness pattern is similar to all previous investigations, but different thickness values. All studies given above are related with the crustal thickness in the offshore areas in the Eastern Mediterranean and southern Aegean region. There are also some investigations in the onshore areas. For instance, Ates et al. (2012) indicated that crustal thickness from the shorelines to the hinterland in the Aegean and Mediterranean regions of Turkey varies from 25 to 34 km to confirm previous and above mentioned studies. Another investigation based on the gravity data was accomplished by Barazangi et al. (1993) in the Syria and they found 36e37 km Moho depth in average. In this study, the crustal thickness variation in the Eastern Mediterranean and southern Aegean region was calculated using an empirical equation on the compiled gravity data by MTA. Following the preparation of the crustal thickness map, analytic signal (AS) transformation was performed to make the anomalies more apparent with the approximate locations of causative bodies. Then, maxspots map of the horizontal gradients of the Bouguer anomalies were created to delineate the boundaries of critical tectonic elements in the study area. Finally, all results inferred from this investigation were compared with the previous studies in order to determine the differences and to expose the disputed areas that require further investigations. 2. Data and methods The gravity anomaly data in the Eastern Mediterranean and the southern Aegean region were acquired by different companies, institutes and official organizations in the Mediterranean littoral countries. All these data were initially compiled and a regional, analog Bouguer map was prepared after the free-air and Bouguer corrections by the General Directorate of Mineral Research and Exploration (MTA)-Turkey (Ozelci, 1973). In the Bouguer calculation, the density of 2.67 g cm3 was used. The analog Bouguer anomaly map of the study area was obtained from MTA. After digitizing and re-gridding, it is presented in Fig. 2, the boundaries of this map are smaller than Fig. 1 because only Eastern Mediterranean and southern Aegean region is focused in this study (the boundaries of following maps will be the same with Fig. 2). However, Fig. 1 presents larger area to indicate the extensions of significant tectonic units. In addition to the data from Ozelci (1973), Ates et al. (1999) published more recent gravity anomaly data in Turkey enabling to obtain more accurate and combined crustal thickness map particularly for the onshore part of Turkey. Unlike the previous seismological investigations, gravity anomalies are used for determination of the Moho depths in the whole region of the Eastern Mediterranean and southern Aegean. It is possible to find out different empirical equations proposed by different authors in the literature to calculate crustal thicknesses from the gravity anomalies (ie., Riad et al., 1981; Riad and El Etr, 1985; Rivero et al., 2002; Tirel et al., 2004). The equation given by Riad et al. (1981) was preferred in this investigation because the following Equation (1) was tested in Turkey by Ates et al. (2012) and, its accuracy was verified by correlating with all previous
crustal thickness investigation results in the onshore areas (Ates et al., 2012):
H ¼ 29:98 0:075Dg
(1)
H: crustal thickness (km) and D g: gravity anomaly values (mGal). The crustal thickness map was map given in Fig. 3 and this map indicates smooth and more accurate results. In order to determine the approximate locations of causative bodies, the gravity anomaly map was transformed to the Analytic Signal (AS) which is generally used for the magnetic data and this transformation is independent from the direction of magnetization (Blakely, 1996). In this study, AS was applied onto the gravity data whether they indicate positive or negative gravity anomaly. The mathematics of AS can be given as the sum of the vertical and horizontal gradients of the gravity anomaly (Equation (2)) and it is formulated in 3D by:
Aðx; yÞ ¼
vg vg vg iþ jþ k vx vy vz
(2)
Here, i, j and k are unit vectors in the x, y and z directions, respectively (Bilim and Ates, 2003, 2005). The amplitude of the AS can be given as follows;
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 2 vg vg vg þ þ jAðx; yÞj ¼ vx vy vz
(3)
The AS map of the gravity anomalies is presented in Fig. 4, as dark patches to represent the approximate locations of the causative tectonic features. The amplitude of dark patches is determined according to Equation (3). Maxspots of the horizontal gradients of the original Bouguer anomalies were calculated and locations of maximum horizontal gradients were also emplaced in Fig. 4 with red circles in different sizes. The elongations of maxspots display significant alignments with the plate boundaries and tectonic units. The horizontal gradient method has been used since the mid 1980s to locate the magnetic boundaries from the gravity or pseudogravity data (Cordell and Grauch, 1985; Blakely and Simpson, 1986). The method is based on the principle that a nearly vertical, fault-like boundary produces a gravity anomaly where the horizontal gradient is largest directly over the top of that boundary. The magnitude of the horizontal gradient is given by the Equation (4):
" #0:5 vgz ðx; yÞ 2 vgz ðx; yÞ 2 hðx; yÞ ¼ þ vx vy
(4)
and it is easily calculated using simple finite-difference relationship (Blakely, 1996). 3. Interpretation of the crustal thickness, analytic signal and maxspots maps There are several EeW trending positive anomalies with contour values up to 170 mGal in the gravity anomaly map (Fig. 2). The largest anomaly extends parallel to the African shoreline, between the Hellenic-Cyprean arcs and the Mediterranean Ridge that is called as the Herodotus Abyssal Plain (Figs. 1 and 2). The second largest one is located to the north of Crete. Southernmost of the Hellenic Arc is also represented by a large anomaly that the eastern part intersects with the largest anomaly to the south. Although the Hellenic Arc is thought to extend through northwest, there is no northwestward extension of this anomaly. The third largest positive anomaly matches with the Cyprus itself and to the north, and it
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
193
Fig. 2. The regional gravity anomaly map of the Eastern Mediterranean and southern Aegean region. Contour interval: 25 mGal.
Fig. 3. The crustal thickness map of the Eastern Mediterranean and southern Aegean region. The contour interval is 1 km.
extends through NW, into the Gulf of Antalya. The last apparent positive anomaly is a circular anomaly and it is located on the northern bend of arcs where the Hellenic and Cyprean arcs meet. Contour pattern from positive to negative contours follows the
Turkey’s boundaries and the negative contour values have increments in the negative direction through the eastern part of Turkey. The Palmyrides in Syria are represented by a large negative anomaly down to 70 mGal (Fig. 2) and this section was modelled
194
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
Fig. 4. The Analytic Signal (AS) and maxspots map of the Eastern Mediterranean and southern Aegean region. Dark patches indicate the locations of causative bodies. Circles show the edges of anomalous bodies and circle sizes are proportional to the intensity.
as the Palmyride Trough between the Aleppo Plateau to the NW and Rutbah Uplift to the SE (Barazangi et al., 1993). The crustal thickness map obtained from the gravity anomalies (Fig. 3) has almost the same pattern on the gravity anomaly map (Fig. 2). Positive gravity contours follow the shallower crustal thicknesses and on the contrary, negative contour patterns indicate a good consistency with the thicker parts of the crust in the study area. The Palmyrides in Syria are represented by 35 km crustal thickness. Thin parts of the crust in the Eastern Mediterranean are represented by 17 km contours in two different places; one of them to the NW part of Cyprus and the second one is in the Herodotus Abyssal Plain (to the NeNW of offshore, Gulf of Alexandria). In order to localize and focus on the causative bodies, the AS transformation was applied onto the gravity anomalies that cover quite large areas, because they are arised from the potential field data. In Fig. 4, the AS map of the gravity anomalies is given together with the maxspots illustrated with red circles and the causative bodies are presented as dark patches. Although AS patches are smaller than the anomalies, they are consistent to the locations of previously determined tectonic elements. The arc-shaped continuous patch to the north of Crete (from Greece to the SW end of Turkey) represents the thinner crustal anomaly in Fig. 3 and probably caused by the back-arc basin behind the magmatic uplift created by the subduction zone in the Eastern Mediterranean (Fig. 4). The positive gravity anomaly to the south of Crete is composed of a set of smaller crustal thinning patches and they are separated from the largest anomaly to the south by a crustal thickening (up to 28 km) to the SE of Crete (Fig. 3). The highest positive gravity anomaly (represented by closed contours of 170 mGal) in the far offshore of the Gulf of Alexandria is wellmatched with the thinnest crustal anomaly (17 km) in the
Eastern Mediterranean. Although it is a large gravity anomaly, its main body area in the AS map is not as large as the area that the anomaly covers. It consists of two different causative bodies; one of them is located to the SW of Cyprus and the second one is the area represented approximately with the closed contour of 18 km (Fig. 3). The thinning around the Cyprus and to the north are wellconsistent with the AS map. AS transformation is not dependent on the anomalies’ characteristics, whether they are positive or negative. Because of this, the thinning to the south of Finike and thickening in the Gulf of Antalya offshore are combined and altogether they extend from the Gulf of Antalya through the east of the onshore (to the north of Anamur). All other AS anomalous regions given above could be related with the anomalous regions in the Bouguer anomaly map (Fig. 2), but the anomalous regions in the onshore part of Turkey, starting from the Gulf of Antalya to the E and NE are not directly related with an apparent Bouguer anomaly, only the AS anomaly to the north of Adana may be related to the Adana Miocene Basin. However, the Iskenderun Basin that is the sister basin to the Adana Basin does not indicate any AS anomaly. Finally, there is another interesting inconsistency in Syria, the thickening on the Palmyrides (Fig. 3) is shifted to the W and SW in the AS map (Fig. 4). The only common characteristics for the inconsistent AS anomalies is that they are directly related with the negative Bouguer anomalies. In order to expose the boundaries of main tectonic units and causative bodies, maxspots map of the gravity anomalies was prepared (Fig. 4) and overlaid onto the AS map to make a comparison. The most obvious and longest continuous tectonic element is the Mediterranean Ridge to the south of the map. The highest positive gravity anomaly and its AS response in the far offshore of the Gulf of Alexandria is surrounded apparently by a small oval
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
shape-maxspots to the north of continuity of the Mediterranean Ridge. Although only highest positive contours of the largest anomaly give an apparent AS response, almost whole anomaly is bounded by a large oval maxspots region as well (following about 20 km closed contour in Fig. 3) and the smaller oval maxspots boundary represented the thinnest crustal area is included in this large oval region. The Cyprean Arc can be followed by the maxspots until the Hellenic Arc (eastern part of Crete). However, the Hellenic Arc is not followed from Crete to Greece with maximas. Nevertheless, the southern boundary of the back-arc basin to the north of Crete has an apparent continuous alignment from Greece to Turkey. The northern boundary of its AS response is also limited by smaller maxspots circles as well. The Bouguer anomaly around Cyprus and to the north of the island is bounded by the maxspots obviously. Similar to the Cyprean anomaly, the circular anomaly to the south of Finike is also very well-defined by the maxspots. However, the other AS anomalies in the onshore area of Turkey has no relationship with the alignments of maxspots. On the contrary to AS patches, maxspots follow the Bouguer anomaly and crustal thickness patterns. There is another inconsistency; the Palmyrides are also surrounded by the maxspots, although their AS response is shifted to the W and SW as mentioned above (Fig. 4). 4. Discussions and conclusions In this study, the crustal thickness variations inferred from the regional gravity anomaly data were calculated and, locations and boundaries of the tectonic elements and causative bodies have been determined by the AS transformation and the maxspots of the horizontal gradients from the original Bouguer anomalies. In comparison with the previous investigations, there are some differences and similarities with the crustal thicknesses obtained in this investigation. 2D inversion of traveltimes along an axial seismic profile through the Gulf of Corinth presented by Zelt et al. (2005) has some differences. They found the Moho depth at 29 km to the east and 39 km to the west of the Corinth strait, but they did not provide the reason for this difference. In this research, a shallow zone up to 19 km was found to the north of Crete and this zone extends deepening in the NWeSE direction through the Corinth strait and represented by 24e26 km contours. Extension of the shallow zone through the NW may be the main reason for the crustal thickness differences between two regions to the east and west of the strait and there are about 3 km differences probably caused by the poor quality of the deeper parts of seismic sections they used. Crustal thickness map of this study is consistent with the results of Sodoudi et al. (2006) in the Aegean Sea where they calculated the Moho depths varying between 19 and 26 km to the north of Crete. The island of Crete is surrounded from the north, east and south by the relatively shallow regions in both investigations. The results of Van der Meijde et al. (2003) are close to Moho depths calculated in this investigation. Their estimation of 25 ± 1.4 km below the station KOUM to the south of Izmir is the same with the crustal thickness map of this study. The Moho depth (29 ± 2.3 km) below the APER station to the east of Crete is consistent with our results within their error limits. Below the station KEG to the south of Port Said, they calculated almost the same crustal thickness (32 ± 1.1 km). Di Luccio and Pasyanos (2007) found significant crustal thinning towards the southern Aegean Sea (20e25 km) similar to our study. In contrast, their calculations in and around Crete has a broad variety, 26e39 km from southwest to central and northern part of the island. This variation in such a short distance needs to be reconsidered. Similarly, they calculated a thicker crust (30e34 km) beneath Cyprus and in the Levantine basin where the thickness values are in the range of 18e26 km in
195
this study. They provided a plot with error bars to compare the Moho depths derived from their investigation versus the previous researches and differences indicate a random distribution on the plot. Errors are not focused in any of the ±5 km intervals. There are 2 km differences between this study and the most recent investigation of ElGabry et al. (2013) to the north of Crete, but 21 km in average may be acceptable for the far offshore of the Gulf of Alexandria, not for the Levantine Basin (offshore Israel and Lebanon). Following their model development, Seber et al. (2001) correlated the theoretical gravity response of their model with the available Bouguer observations in this region. In the first step of their correlation, many inaccuracies have been observed in some specific zones and they introduced Airy-type isostatic compensation into the theoretical model response in order to remove or mitigate the inaccuracies. Hence, they found a reasonable fit with the observed gravity measurements. However, the grid interval of their calculation is about 25 km and not sensitive enough. Thus, they stated that lateral fluctuations in crustal and/or upper mantle structure and lateral density variations may be the reasons for the large gravity residuals. As a result of these reasons, their map in Figure 7 of their paper does not present the details in the Eastern Mediterranean and Aegean region. Large differences between the observed and their model based isostatic gravity anomalies are mainly located in the Aegean Sea and in a large zone extending from the south of Greece, crossing the south of Crete to the isle of Cyprus (Seber et al., 2001). In addition, the region with shallowest Moho depths is located to the south of Crete in Fig. 2 of their paper. However, this zone is observed to the SE of Crete in our study (Fig. 3), there is an apparent shift through the east between the Moho maps in both studies. The crustal thickness map of Makris and Stobbe (1984) is based on old gravity data, particularly in Turkey and Syria. Because of this reason, there are some differences between their map and the map in this paper. Although, thickness patterns in two investigations are similar to each other, the Bouguer anomaly and crustal thickness maps in Turkey and near offshore of the country are different, because more recent data were used in this research (Ates et al., 1999). The other main difference is observed around Palmyrides in Syria. Thickness values and patterns of thickness variations in other regions are similar, but the Moho is observed deeper (2e6 km) than this study. These differences may be caused by the use of two different formulations. Ergun et al. (2005) used a 2D approach to model gravity anomalies in the southern part of Turkey and Cyprus with assumption that changes in the strike direction of structures are gradual. Moho depths are not consistent with this study and other investigations except the profile D to the east, extending from Lebanon to the Gulf of Mersin, in their study. They used 2.0 mg/m3 density routinely for the shallowest, Pliocene-Quaternary sedimentary units and Upper Miocene evaporites dominated by halite. Deeper sediments of Miocene and older units were represented by 2.3 mg/m3 density. Sedimentary unit densities used by Ergun et al. (2005) are quite low, but the oceanic crustal density and the mantle density value used in their models are reasonable (3.0 mg/m3 and 3.38 mg/m3 respectively). The density used for the Upper Miocene and younger units is low (2.67 mg/m3 versus 2.0 mg/m3). These abnormal density values may be the reason for the inconsistencies of models along 4 profiles in their study and other investigations. Casten and Snopek (2006) prepared three different maps around the Hellenic subduction zone with gravity modeling; triangular-shaped continental crust without sedimentary cover, arc-shaped continental Moho and, arc-shaped and subducted oceanic Moho. The arc-shaped continental Moho map around the Crete island is similar to the results of this study. They found about 20 km Moho depth to the north of Crete consistent with this study. However, their map indicates 35 km Moho depth around the
196
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
eastern and western ends of the island and to the near south. On the contrary, the crustal thickness variations are increasing up to 28e29 km at most to the SE and SW of Crete then thinning back to 17e23 km (Fig. 3) through the largest Bouguer anomaly in the Eastern Mediterranean (Fig. 2) in this study. The crustal thickness variations increase up to 35 km around the Palmyrides, Syria in this investigation. Barazangi et al. (1993) found a flat Moho depth around 36e37 km that is very close to the variations in this investigation. Cowie and Kusznir (2012) presented crustal thickness and oceanic lithosphere distribution from the gravity inversion according to different break-up ages of West and Central African Rift System. They suggested another rift in the Levantine Basin extending from far offshore of the Nile delta, Egypt to the SE of Cyprus using a late Triassic (225 Ma) break-up age. They also explained the linear extension of the NE shoreline of Libya and NW shoreline of Egypt down to Alexandria as a transform fault controlled margin kinematics. They presented two cross-sections, one in the N-S direction crossing the shallowest part in the Herodotus Abyssal Plain, and the other in the NWeSE extension in the Levantine Basin (Cowie and Kusznir, 2012). The crustal thickness values are consistent with our results in the shallow zones in their cross-sections. However, deeper parts on the sections are getting steeply deeper through the shorelines (about 35e40 km) that such deep values can only be observed around Palmyrides. This inconsistency may be arised from the inversion difficulties in the transition zone between the onshore and offshore areas. There are some recent crustal thickness studies in the onshore parts of Turkey. The investigations focused or including particularly on the Aegean region were compared with this investigation, below. Tezel et al. (2007) found thickness variation about 25e40 km from western to the eastern Turkey by using surface wave dispersion analysis. In the study of Tezel et al. (2010), they only focused on the crustal thickness of western Turkey using the receiver function and the results are varying between 20 and 35 km. The most recent research based on the gravity data (Ates et al., 2012) was covering whole onshore area of Turkey. The average crustal thickness variation is in the range of 25e26 km in the onshore Aegean and Mediterranean regions and 42 km in the eastern part, near the Iranian border. These values are consistent with the contour values of crustal thickness in this research and previous investigations except Ergun et al. (2005) because they calculated 38 km crustal thickness beneath the Taurus Mountains. The locations and boundaries of critical tectonic units and causative bodies were defined by the Analytic Signal (AS) and maxspots maps of the gravity anomalies in this study. Almost all of the causative bodies and their boundaries are consistent with the produced maps. However, there are some easily recognizable changes and inconsistencies, particularly in the onshore part of Turkish Mediterranean region and around Palmyrides in Syria that are represented by negative gravity anomalies. Moreover, the AS anomalies and maxspots are not compatible with each other in these regions. This may be derived from the sparsity and low sensitivity of the gravity data in these areas. Thus, the Mediterranean onshore part of Turkey is composed of the rugged and severely tectonized Taurus Mountains where the data acquisition is quite difficult. In case of acquiring and compiling very dense and detailed gravity data in these problematic regions, more detailed results will be obtained and the research area will be subjected for the further investigations. Acknowledgments The authors thank the General Directorate of the Mineral Research and Exploration (MTA) of Turkey for the analog map of the
gravity data. The analog map was digitized by a group of undergraduate students under the supervision of Prof. Dr. Abdullah Ates at the Ankara University, Department of Geophysical Engineering. We appreciated for their enthusiasm. References Ates, A., Kearey, P., Tufan, S., 1999. New gravity and magnetic anomaly maps of Turkey. Geophys. J. Int. 136, 499e502. Ates, A., Bilim, F., Buyuksarac, A., Aydemir, A., Bektas, O., Aslan, Y., 2012. Crustal structure of Turkey from aeromagnetic, gravity and deep seismic reflection data. Surv. Geophys. 33, 869e885. Barazangi, M., Seber, D., Chaimov, T., Best, J., Litak, R., Al-Saad, D., Sawaf, T., 1993. Tectonic evolution of the northern arabian plate in western Syria. In: Boschi, et al (Eds.), Recent Evolution and Seismicity of the Mediterranean Region. Kluwer Academic Publishers, Netherlands, pp. 117e140. Bilim, F., Ates, A., 2003. Analytic signal inferred from reduced to the pole data. J. Balk. Geophys. Soc. 6, 66e74. Bilim, F., Ates, A., 2005. Analitik Sinyal yontemlerinin manyetik model verileri uzerinde karsilastirilmasi. Istanb. Univ. Muh. Fak. Yerbilim. Derg. 18, 151e162 (in Turkish with English Abstract). Blakely, R.J., 1996. Potential Theory in Gravity and Magnetic Applications. Cambridge University Press, p. 441. Blakely, R.J., Simpson, R.W., 1986. Approximating edges of source bodies from magnetic or gravity anomalies. Geophysics 51, 1494e1498. Bozkurt, E., Mittwede, S.K., 2005. Introduction: evolution of continental extensional tectonics of western Turkey. Geodin. Acta 18, 153e165. Casten, U., Snopek, K., 2006. Gravity modelling of the Hellenic subduction zone-a regional study. Tectonophysics 417, 183e200. Cordell, L., Grauch, V.J.S., 1985. Mapping basement magnetization zones from aeromagnetic data in the San Juan basin, New Mexico. In: Hinze, W.J. (Ed.), The Utility of Regional Gravity and Magnetic Anomaly Maps, pp. 181e197. Soc. Explor. Geophys. Cowie, L., Kusznir, N., 2012. Mapping crustal thickness and oceanic lithosphere distribution in the Eastern Mediterranean using gravity inversion. Pet. Geosci. 18, 373e380. Di Luccio, F., Pasyanos, M.E., 2007. Crustal and upper-mantle structure in the Eastern Mediterranean from the analysis of surface wave dispersion curves. Geophys. J. Int. 169, 1139e1152. ElGabry, M.N., Panza, G.F., Badawy, A.A., Korrat, I.M., 2013. Imaging a relic of complex tectonics: the lithosphere-asthenosphere structure in the Eastern Mediterranean. Terra Nova 25, 102e109. Ergun, M., Okay, S., Sari, C., Oral, E.Z., Ash, M., Hall, J., Miller, H., 2005. Gravity anomalies of the Cyprus Arc and their tectonic implications. Mar. Geol. 221, 349e358. Ketin, I., 1948. Uber die tektonische-mechanischen Folgerungen aus den grosssen Anatolischen erdbeben des letzten dezenniums. Geol. Rundsch. 36, 77e83. Le Pichon, X., 1982. In: Hsu, K. (Ed.), Land-locked Oceanic Basins and Continental Collision: the Eastern Mediterranean as a Case Example. Mountain Building Processes Academic Press, London, pp. 201e211. Le Pichon, X., Angelier, J., 1979. The Hellenic arc and trench system: a key to the evolution of the Eastern Mediterranean area. Tectonophysics 60, 1e42. Le Pichon, X., Chamot-Rooke, N., Lallemant, S., Noomen, R., Veis, G., 1995. Geodetic determination of the kinematics of Central Greece with respect to Europe: implications for Eastern Mediterranean Tectonics. J. Geophys. Res. 100, 12675e12690. Makris, J., Stobbe, C., 1984. Physical properties and state of the crust and upper mantle of the Eastern Mediterranean Sea deduced from geophysical data. Mar. Geol. 55, 347e363. McClusky, S., Balassanian, S., Barka, A., Demir, C., Ergintav, S., Georgiev, I., Gurkan, O., Hamburger, M., Hurst, K., Kahle, H., Kastens, K., Kekelidze, G., King, R., Kotzev, V., Lenk, O., Mahmoud, S., Mishin, A., Nadariya, M., Ouzounis, A., Paradissis, D., Peter, Y., Prilepin, M., Reilinger, R., Sanli, I., Seeger, H., Tealeb, A., Toksoz, M.N., Veis, G., 2000. Global positioning system constraints on plate kinematics and dynamics in the Eastern Mediterranean and Caucasus. J. Geophys. Res. 105, 5695e5719. McKenzie, D., 1972. Active tectonics of the mediterranean region. Geophys. J. Int. 30, 109e185. McKenzie, D.P., 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions. Geophys. J. R. Astr. Soc. 55, 217e254. Ozelci, H.F., 1973. Dogu akdeniz bolgesi gravite anomalileri. MTA Bull. 80, 54e88. Reilinger, R.E., McClusky, S.C., Oral, M.B., King, R.W., Toksoz, M.N., Barka, A.A., Kinik, I., Lenk, O., Sanli, I., 1997. Global positioning system measurements of present-day crustal movements in the arabia-africa-eurasia plate collision zone. J. Geophys. Res. 102, 9983e9999. Riad, S., El Etr, H.A., 1985. Bouguer anomalies and lithosphere-crustal thickness in Uganda. J. Geodyn. 3, 169e186. Riad, S., Refai, E., Ghalib, M., 1981. Bouguer anomalies and crustal structure in the Eastern Mediterranean. Tectonophysics 71, 253e266. Rivero, L., Pinto, V., Casas, A., 2002. Moho depth structure of the eastern part of the Pyrenean belt derived from the gravity data. J. Geodyn. 33, 315e332. Seber, D., Sandvol, E., Sandvol, C., Brindisi, C., Barazangi, M., 2001. Crustal model for the Middle East and North Africa region: implications for the isostatic
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197 compensation mechanism. Geophys. J. Int. 147, 630e638. Sengor, A.M.C., Yilmaz, Y., 1981. Tethyan evolution of Turkey: a plate tectonic approach. Tectonophysics 75, 181e241. Sodoudi, F., Kind, R., Hatzfeld, D., Priestley, K., Hanka, W., Wylegalla, K., Stavrakakis, G., Vafidis, A., Harjes, H.-P., Bohnhoff, M., 2006. Lithospheric structure of the Aegean obtained from P and S receiver functions. J. Geophys. Res. 111 (B12307), 1e23. Staackmann, M., Casten, U., Ates, A., 2008. Gravity controlled Moho depths in the aegean region. EGU general assembly. Geophys. Res. Abstr. 10. EGU2008-A08760. Tezel, T., Erduran, M., Alptekin, O., 2007. Crustal shear wave velocity structure of Turkey by surface wave dispersion analysis. Ann. Geophys. 50, 177e190. Tezel, T., Shibutani, T., Kaypak, B., 2010. Crustal structure variation in western
197
Turkey inferred from the receiver function analysis. Tectonophysics 492, 240e252. Tirel, C., Gueydan, F., Tiberi, C., Brun, J.P., 2004. Aegean crustal thickness inferred from gravity inversion. Geodynamical implications. Earth Planet. Sci. Lett. 228, 267e280. Van der Meijde, M., Van der Lee, S., Giardini, D., 2003. Crustal structure beneath broad-band seismic stations in the Mediterranean region. Geophys. J. Int. 152, 729e739. Wortel, M.J.R., Goes, S.D.B., Spakman, W., 1990. Structure and seismicity of the Aegean subduction zone. Terra Nova 2, 554e562. Zelt, B.C., Taylor, B., Sachpazi, M., Hirn, A., 2005. Crustal velocity and Moho structure beneath the Gulf of Corinth, Greece. Geophys. J. Int. 162, 257e268.
Contents lists available at ScienceDirect
Marine and Petroleum Geology journal homepage: www.elsevier.com/locate/marpetgeo
Research paper
Crustal thickness variations in the Eastern Mediterranean and southern Aegean region Funda Bilim a, Attila Aydemir b, *, Abdullah Ates c a
Engineering Faculty, Geophysical Engineering Department, Cumhuriyet University, 05480 Sivas, Turkey Turkish Petroleum Corp., Sogutozu Mah., 2180. Cad., No:10, 06530 Sogutozu, Ankara, Turkey c Faculty of Engineering, Department of Geophysical Eng., Ankara University, 06100 Besevler, Ankara, Turkey b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 October 2015 Received in revised form 11 June 2016 Accepted 15 June 2016 Available online 18 June 2016
In this paper, regional analog gravity anomaly map obtained from the General Directorate of Mineral Research and Exploration (MTA) was digitized and used for the calculation of the crustal thickness (Moho depth) variations in the Eastern Mediterranean and the southern part of the Aegean Region. In the gravity anomaly map, there are mainly EeW trending apparent gravity anomalies represented by the contours up to 150 mGal. They are generally parallel to the shorelines of Africa, Turkey and Crete. Crustal thickness variations were calculated from the gravity anomalies, using an empirical equation in this study. Obtained thicknesses (Moho depths) were mapped and correlated with the previous investigations and seismological findings. According to the estimations, crustal thicknesses are about 25 e30 km along the coastal regions and more than 30 km on the onshore part of Turkey increasing up to 42 km through the eastern Anatolia. However, there are thin crustal zones around 17 km in the offshore Egypt, to the NW part of Cyprus and about 19 km to the north of Crete. They may be related with the main tectonic trends in this region except the circular thinning to the south of Kas (southwestern part of Turkey). In order to determine the locations and boundaries of prominent tectonic elements, Analytic Signal (AS) and maxspots maps of the gravity anomalies were also prepared in this study. All produced maps are generally consistent to each other and the boundaries of main tectonic units were apparently illustrated in the maxspots map from the horizontal gradient of Bouguer anomalies. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Eastern mediterranean Aegean sea Western anatolia Crustal thickness Analytic signal Maxspots map
1. Introduction Tectonic framework of the Eastern Mediterranean is dominated by the subduction of the African Plate beneath the Eurasian Plate (Fig. 1). In addition, the Anatolian Plate has been converging through the Aegean Sea and Greece with the counter-clockwise rotational tectonic escapement (30e40 mm/y; Le Pichon et al., 1995). This movement is controlled by two strike-slip fault systems (Fig. 1); the North Anatolian Fault-NAF and the East Anatolian Fault-EAF (Ketin, 1948; McKenzie, 1972; Le Pichon and Angelier, 1979; Sengor and Yilmaz, 1981). The African Plate subduction beneath the Anatolian Block triggered N-S directional extension in the western Anatolia (Sengor and Yilmaz, 1981) and crustal thickness was found about 34 km to the east and 25 km to the further west of the onshore Aegean region (Ates et al., 2012) under the
* Corresponding author. E-mail addresses: [email protected], [email protected] (A. Aydemir). http://dx.doi.org/10.1016/j.marpetgeo.2016.06.012 0264-8172/© 2016 Elsevier Ltd. All rights reserved.
influence of extension. Tectonics of the Aegean zone can be summarized by its rapid extension characteristics that allows to consider this region as one of the world’s most rapidly extending crustal thinning zone (the extension rate was given as 14 ± 5 mm/yr by Reilinger et al., 1997; McClusky et al., 2000). Tectonic model of the Eastern Mediterranean can be explained by the subduction of the African Plate dipping north beneath the Aegean Sea (Eurasian Plate) through the Hellenic and Cyprian arcs from west and to the east, respectively (McKenzie, 1978; Le Pichon and Angelier, 1979; Le Pichon, 1982). The most active part of the subduction is observed around the southern Aegean or Hellenic trench. Tomographic results indicated a high-velocity anomaly dipping north, down to 600 km beneath the Hellenic arc and it was interpreted as the subducting lithosphere of the African Plate (Wortel et al., 1990). Crustal thicknesses in the study area were calculated previously for very limited zones, mainly based on the seismological studies. However, there is no investigation to cover entire region for the Eastern Mediterranean, whether based on seismological
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
191
Fig. 1. The major tectonic elements in and around Turkey, and Eastern Mediterranean region (modified from Bozkurt and Mittwede, 2005). The large arrows show relative motions of the Anatolian Block and convergent motions of the African Plate and Arabian Platform. NAFZ: North Anatolian Fault Zone, DSFZ: Dead Sea Fault Zone, NEAFZ: North-East Anatolian Fault Zone, EAFZ: East Anatolian Fault Zone, IASZ: Izmir-Ankara-Erzincan Suture Zone.
observations or estimations from the gravity data. Zelt et al. (2005) performed a 2D inversion of refraction and reflection traveltimes along an axial seismic profile through Gulf of Corinth, Greece and found the Moho depth at 29 km to the east and 39 km to the west of the Corinth strait. Crustal thickness to the east of the Corinth is consistent with the results of Sodoudi et al. (2006) where they calculated the Moho depths using the P and S receiver functions. In their research, they found shallow Moho depths in the Aegean Sea to the north of Crete, varying between 19 and 26 km. Van der Meijde et al. (2003) also analysed receiver functions below 17 seismological stations in the Mediterranean region and North Africa. Three stations in their investigations are located in the area of our study and they found Moho depth 25 ± 1.4 km below the station KOUM to the south of Izmir, 29 ± 2.3 km below APER station to the east of Crete and 32 ± 1.1 km below the station named KEG to the south of Port Said. Di Luccio and Pasyanos (2007) used surface wave dispersion curves in the analysis and found shallower crust in the Ionian Sea, southern Italy (8e16 km) and the central part of the Eastern Mediterranean (16e24 km). Their calculations in the southern, central-western Aegean are in the range of 20e25 km and consistent with this study. Recently, ElGabry et al. (2013) investigated the crustal thickness and mantle structure of the Eastern Mediterranean using the non-linear inversion of Rayleigh wave dispersion curves extracted from the broad-band records on the profiles along the Hellenic and Cyprian arcs. In the first profile (Hellenic arc profile), they found the average thickness around 31 km and they stated that the Moho depth decreases gradually down to 21 km under the Aegean Sea, to the north of Crete. In the Cyprian arc profile, the Moho was found at a depth of 37 km in the Sinai Peninsula to the south, rapidly rising to 21 km below the Levantine Basin (to the south of Cyprus) that is known as the
Herodotus Abyssal Plain as well. Similar to the seismological investigations, crustal thickness calculations based on the gravity data or gravity modeling were performed for the limited zones or regions in the Aegean region, Turkey, Middle East and North Africa. A study covering large area was accomplished by Seber et al. (2001), attempting to develop a crustal model for the Middle East and North Africa region. In their study, they presented a 3D crustal model after they integrated and interpolated rock thicknesses and Moho depth measurements in this part of the world. The Eastern Mediterranean region was represented by 20e25 km Moho depths in general. The shallowest zone was given with the 10e15 km interval to the south of Crete and the Herodotus Abyssal Plain. Makris and Stobbe (1984) also created a crustal thickness map of the area interested, deduced from the geophysical data. They found two different shallowest zones; one of them is located to the NW of Alexandria and the second one to the NW of Cyprus about 22 km. The other shallow zone to the north of Crete is about 24 km in their map. Casten and Snopek (2006) compiled the gravity data from the land, marine and satellite sources in the area including the Hellenic subduction zone and they obtained a Bouguer anomaly map range from þ170 mGal to 10 mGal. After the 3D gravity modeling study, they found the Moho depths less than 20 km under the Cretan Sea (to the north of Crete) and it gets deeper to the south of the island (around 30 km). They proposed extremely thick sedimentary cover (up to 18 km) around the Mediterranean Ridge and the Herodotus Abyssal Plain. Tirel et al. (2004) calculated maximum thinning of the crust to the north of Crete (around 22 km) and they estimated an average thickness of 25 km to the north of Cretan Sea (the area of Cyclades islands) where the rigid block-type behaviour was suggested during the post-Miocene times. Staackmann et al. (2008) found similar results (20 km of Moho depth beneath the Cretan Sea) with Tirel
192
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
et al. (2004) and they observed a crustal thinning from west to the east. Ergun et al. (2005) modelled the gravity anomalies in twodimensions (2D) along with four different profiles to the south of Turkey and, only easternmost profile D is consistent with this study. Cowie and Kusznir (2012) mapped crustal thickness and oceanic lithosphere distribution in the Eastern Mediterranean using the gravity inversion by considering the relation of this region under the break-up of West and Central African Rift System. Their calculations were based on two different break-up ages; 225 Ma (late Triassic) and 100 Ma (mid-Cretaceous). They suggest a new rifting in the Levantine Basin where crustal thickness pattern is similar to all previous investigations, but different thickness values. All studies given above are related with the crustal thickness in the offshore areas in the Eastern Mediterranean and southern Aegean region. There are also some investigations in the onshore areas. For instance, Ates et al. (2012) indicated that crustal thickness from the shorelines to the hinterland in the Aegean and Mediterranean regions of Turkey varies from 25 to 34 km to confirm previous and above mentioned studies. Another investigation based on the gravity data was accomplished by Barazangi et al. (1993) in the Syria and they found 36e37 km Moho depth in average. In this study, the crustal thickness variation in the Eastern Mediterranean and southern Aegean region was calculated using an empirical equation on the compiled gravity data by MTA. Following the preparation of the crustal thickness map, analytic signal (AS) transformation was performed to make the anomalies more apparent with the approximate locations of causative bodies. Then, maxspots map of the horizontal gradients of the Bouguer anomalies were created to delineate the boundaries of critical tectonic elements in the study area. Finally, all results inferred from this investigation were compared with the previous studies in order to determine the differences and to expose the disputed areas that require further investigations. 2. Data and methods The gravity anomaly data in the Eastern Mediterranean and the southern Aegean region were acquired by different companies, institutes and official organizations in the Mediterranean littoral countries. All these data were initially compiled and a regional, analog Bouguer map was prepared after the free-air and Bouguer corrections by the General Directorate of Mineral Research and Exploration (MTA)-Turkey (Ozelci, 1973). In the Bouguer calculation, the density of 2.67 g cm3 was used. The analog Bouguer anomaly map of the study area was obtained from MTA. After digitizing and re-gridding, it is presented in Fig. 2, the boundaries of this map are smaller than Fig. 1 because only Eastern Mediterranean and southern Aegean region is focused in this study (the boundaries of following maps will be the same with Fig. 2). However, Fig. 1 presents larger area to indicate the extensions of significant tectonic units. In addition to the data from Ozelci (1973), Ates et al. (1999) published more recent gravity anomaly data in Turkey enabling to obtain more accurate and combined crustal thickness map particularly for the onshore part of Turkey. Unlike the previous seismological investigations, gravity anomalies are used for determination of the Moho depths in the whole region of the Eastern Mediterranean and southern Aegean. It is possible to find out different empirical equations proposed by different authors in the literature to calculate crustal thicknesses from the gravity anomalies (ie., Riad et al., 1981; Riad and El Etr, 1985; Rivero et al., 2002; Tirel et al., 2004). The equation given by Riad et al. (1981) was preferred in this investigation because the following Equation (1) was tested in Turkey by Ates et al. (2012) and, its accuracy was verified by correlating with all previous
crustal thickness investigation results in the onshore areas (Ates et al., 2012):
H ¼ 29:98 0:075Dg
(1)
H: crustal thickness (km) and D g: gravity anomaly values (mGal). The crustal thickness map was map given in Fig. 3 and this map indicates smooth and more accurate results. In order to determine the approximate locations of causative bodies, the gravity anomaly map was transformed to the Analytic Signal (AS) which is generally used for the magnetic data and this transformation is independent from the direction of magnetization (Blakely, 1996). In this study, AS was applied onto the gravity data whether they indicate positive or negative gravity anomaly. The mathematics of AS can be given as the sum of the vertical and horizontal gradients of the gravity anomaly (Equation (2)) and it is formulated in 3D by:
Aðx; yÞ ¼
vg vg vg iþ jþ k vx vy vz
(2)
Here, i, j and k are unit vectors in the x, y and z directions, respectively (Bilim and Ates, 2003, 2005). The amplitude of the AS can be given as follows;
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 2 vg vg vg þ þ jAðx; yÞj ¼ vx vy vz
(3)
The AS map of the gravity anomalies is presented in Fig. 4, as dark patches to represent the approximate locations of the causative tectonic features. The amplitude of dark patches is determined according to Equation (3). Maxspots of the horizontal gradients of the original Bouguer anomalies were calculated and locations of maximum horizontal gradients were also emplaced in Fig. 4 with red circles in different sizes. The elongations of maxspots display significant alignments with the plate boundaries and tectonic units. The horizontal gradient method has been used since the mid 1980s to locate the magnetic boundaries from the gravity or pseudogravity data (Cordell and Grauch, 1985; Blakely and Simpson, 1986). The method is based on the principle that a nearly vertical, fault-like boundary produces a gravity anomaly where the horizontal gradient is largest directly over the top of that boundary. The magnitude of the horizontal gradient is given by the Equation (4):
" #0:5 vgz ðx; yÞ 2 vgz ðx; yÞ 2 hðx; yÞ ¼ þ vx vy
(4)
and it is easily calculated using simple finite-difference relationship (Blakely, 1996). 3. Interpretation of the crustal thickness, analytic signal and maxspots maps There are several EeW trending positive anomalies with contour values up to 170 mGal in the gravity anomaly map (Fig. 2). The largest anomaly extends parallel to the African shoreline, between the Hellenic-Cyprean arcs and the Mediterranean Ridge that is called as the Herodotus Abyssal Plain (Figs. 1 and 2). The second largest one is located to the north of Crete. Southernmost of the Hellenic Arc is also represented by a large anomaly that the eastern part intersects with the largest anomaly to the south. Although the Hellenic Arc is thought to extend through northwest, there is no northwestward extension of this anomaly. The third largest positive anomaly matches with the Cyprus itself and to the north, and it
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
193
Fig. 2. The regional gravity anomaly map of the Eastern Mediterranean and southern Aegean region. Contour interval: 25 mGal.
Fig. 3. The crustal thickness map of the Eastern Mediterranean and southern Aegean region. The contour interval is 1 km.
extends through NW, into the Gulf of Antalya. The last apparent positive anomaly is a circular anomaly and it is located on the northern bend of arcs where the Hellenic and Cyprean arcs meet. Contour pattern from positive to negative contours follows the
Turkey’s boundaries and the negative contour values have increments in the negative direction through the eastern part of Turkey. The Palmyrides in Syria are represented by a large negative anomaly down to 70 mGal (Fig. 2) and this section was modelled
194
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
Fig. 4. The Analytic Signal (AS) and maxspots map of the Eastern Mediterranean and southern Aegean region. Dark patches indicate the locations of causative bodies. Circles show the edges of anomalous bodies and circle sizes are proportional to the intensity.
as the Palmyride Trough between the Aleppo Plateau to the NW and Rutbah Uplift to the SE (Barazangi et al., 1993). The crustal thickness map obtained from the gravity anomalies (Fig. 3) has almost the same pattern on the gravity anomaly map (Fig. 2). Positive gravity contours follow the shallower crustal thicknesses and on the contrary, negative contour patterns indicate a good consistency with the thicker parts of the crust in the study area. The Palmyrides in Syria are represented by 35 km crustal thickness. Thin parts of the crust in the Eastern Mediterranean are represented by 17 km contours in two different places; one of them to the NW part of Cyprus and the second one is in the Herodotus Abyssal Plain (to the NeNW of offshore, Gulf of Alexandria). In order to localize and focus on the causative bodies, the AS transformation was applied onto the gravity anomalies that cover quite large areas, because they are arised from the potential field data. In Fig. 4, the AS map of the gravity anomalies is given together with the maxspots illustrated with red circles and the causative bodies are presented as dark patches. Although AS patches are smaller than the anomalies, they are consistent to the locations of previously determined tectonic elements. The arc-shaped continuous patch to the north of Crete (from Greece to the SW end of Turkey) represents the thinner crustal anomaly in Fig. 3 and probably caused by the back-arc basin behind the magmatic uplift created by the subduction zone in the Eastern Mediterranean (Fig. 4). The positive gravity anomaly to the south of Crete is composed of a set of smaller crustal thinning patches and they are separated from the largest anomaly to the south by a crustal thickening (up to 28 km) to the SE of Crete (Fig. 3). The highest positive gravity anomaly (represented by closed contours of 170 mGal) in the far offshore of the Gulf of Alexandria is wellmatched with the thinnest crustal anomaly (17 km) in the
Eastern Mediterranean. Although it is a large gravity anomaly, its main body area in the AS map is not as large as the area that the anomaly covers. It consists of two different causative bodies; one of them is located to the SW of Cyprus and the second one is the area represented approximately with the closed contour of 18 km (Fig. 3). The thinning around the Cyprus and to the north are wellconsistent with the AS map. AS transformation is not dependent on the anomalies’ characteristics, whether they are positive or negative. Because of this, the thinning to the south of Finike and thickening in the Gulf of Antalya offshore are combined and altogether they extend from the Gulf of Antalya through the east of the onshore (to the north of Anamur). All other AS anomalous regions given above could be related with the anomalous regions in the Bouguer anomaly map (Fig. 2), but the anomalous regions in the onshore part of Turkey, starting from the Gulf of Antalya to the E and NE are not directly related with an apparent Bouguer anomaly, only the AS anomaly to the north of Adana may be related to the Adana Miocene Basin. However, the Iskenderun Basin that is the sister basin to the Adana Basin does not indicate any AS anomaly. Finally, there is another interesting inconsistency in Syria, the thickening on the Palmyrides (Fig. 3) is shifted to the W and SW in the AS map (Fig. 4). The only common characteristics for the inconsistent AS anomalies is that they are directly related with the negative Bouguer anomalies. In order to expose the boundaries of main tectonic units and causative bodies, maxspots map of the gravity anomalies was prepared (Fig. 4) and overlaid onto the AS map to make a comparison. The most obvious and longest continuous tectonic element is the Mediterranean Ridge to the south of the map. The highest positive gravity anomaly and its AS response in the far offshore of the Gulf of Alexandria is surrounded apparently by a small oval
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
shape-maxspots to the north of continuity of the Mediterranean Ridge. Although only highest positive contours of the largest anomaly give an apparent AS response, almost whole anomaly is bounded by a large oval maxspots region as well (following about 20 km closed contour in Fig. 3) and the smaller oval maxspots boundary represented the thinnest crustal area is included in this large oval region. The Cyprean Arc can be followed by the maxspots until the Hellenic Arc (eastern part of Crete). However, the Hellenic Arc is not followed from Crete to Greece with maximas. Nevertheless, the southern boundary of the back-arc basin to the north of Crete has an apparent continuous alignment from Greece to Turkey. The northern boundary of its AS response is also limited by smaller maxspots circles as well. The Bouguer anomaly around Cyprus and to the north of the island is bounded by the maxspots obviously. Similar to the Cyprean anomaly, the circular anomaly to the south of Finike is also very well-defined by the maxspots. However, the other AS anomalies in the onshore area of Turkey has no relationship with the alignments of maxspots. On the contrary to AS patches, maxspots follow the Bouguer anomaly and crustal thickness patterns. There is another inconsistency; the Palmyrides are also surrounded by the maxspots, although their AS response is shifted to the W and SW as mentioned above (Fig. 4). 4. Discussions and conclusions In this study, the crustal thickness variations inferred from the regional gravity anomaly data were calculated and, locations and boundaries of the tectonic elements and causative bodies have been determined by the AS transformation and the maxspots of the horizontal gradients from the original Bouguer anomalies. In comparison with the previous investigations, there are some differences and similarities with the crustal thicknesses obtained in this investigation. 2D inversion of traveltimes along an axial seismic profile through the Gulf of Corinth presented by Zelt et al. (2005) has some differences. They found the Moho depth at 29 km to the east and 39 km to the west of the Corinth strait, but they did not provide the reason for this difference. In this research, a shallow zone up to 19 km was found to the north of Crete and this zone extends deepening in the NWeSE direction through the Corinth strait and represented by 24e26 km contours. Extension of the shallow zone through the NW may be the main reason for the crustal thickness differences between two regions to the east and west of the strait and there are about 3 km differences probably caused by the poor quality of the deeper parts of seismic sections they used. Crustal thickness map of this study is consistent with the results of Sodoudi et al. (2006) in the Aegean Sea where they calculated the Moho depths varying between 19 and 26 km to the north of Crete. The island of Crete is surrounded from the north, east and south by the relatively shallow regions in both investigations. The results of Van der Meijde et al. (2003) are close to Moho depths calculated in this investigation. Their estimation of 25 ± 1.4 km below the station KOUM to the south of Izmir is the same with the crustal thickness map of this study. The Moho depth (29 ± 2.3 km) below the APER station to the east of Crete is consistent with our results within their error limits. Below the station KEG to the south of Port Said, they calculated almost the same crustal thickness (32 ± 1.1 km). Di Luccio and Pasyanos (2007) found significant crustal thinning towards the southern Aegean Sea (20e25 km) similar to our study. In contrast, their calculations in and around Crete has a broad variety, 26e39 km from southwest to central and northern part of the island. This variation in such a short distance needs to be reconsidered. Similarly, they calculated a thicker crust (30e34 km) beneath Cyprus and in the Levantine basin where the thickness values are in the range of 18e26 km in
195
this study. They provided a plot with error bars to compare the Moho depths derived from their investigation versus the previous researches and differences indicate a random distribution on the plot. Errors are not focused in any of the ±5 km intervals. There are 2 km differences between this study and the most recent investigation of ElGabry et al. (2013) to the north of Crete, but 21 km in average may be acceptable for the far offshore of the Gulf of Alexandria, not for the Levantine Basin (offshore Israel and Lebanon). Following their model development, Seber et al. (2001) correlated the theoretical gravity response of their model with the available Bouguer observations in this region. In the first step of their correlation, many inaccuracies have been observed in some specific zones and they introduced Airy-type isostatic compensation into the theoretical model response in order to remove or mitigate the inaccuracies. Hence, they found a reasonable fit with the observed gravity measurements. However, the grid interval of their calculation is about 25 km and not sensitive enough. Thus, they stated that lateral fluctuations in crustal and/or upper mantle structure and lateral density variations may be the reasons for the large gravity residuals. As a result of these reasons, their map in Figure 7 of their paper does not present the details in the Eastern Mediterranean and Aegean region. Large differences between the observed and their model based isostatic gravity anomalies are mainly located in the Aegean Sea and in a large zone extending from the south of Greece, crossing the south of Crete to the isle of Cyprus (Seber et al., 2001). In addition, the region with shallowest Moho depths is located to the south of Crete in Fig. 2 of their paper. However, this zone is observed to the SE of Crete in our study (Fig. 3), there is an apparent shift through the east between the Moho maps in both studies. The crustal thickness map of Makris and Stobbe (1984) is based on old gravity data, particularly in Turkey and Syria. Because of this reason, there are some differences between their map and the map in this paper. Although, thickness patterns in two investigations are similar to each other, the Bouguer anomaly and crustal thickness maps in Turkey and near offshore of the country are different, because more recent data were used in this research (Ates et al., 1999). The other main difference is observed around Palmyrides in Syria. Thickness values and patterns of thickness variations in other regions are similar, but the Moho is observed deeper (2e6 km) than this study. These differences may be caused by the use of two different formulations. Ergun et al. (2005) used a 2D approach to model gravity anomalies in the southern part of Turkey and Cyprus with assumption that changes in the strike direction of structures are gradual. Moho depths are not consistent with this study and other investigations except the profile D to the east, extending from Lebanon to the Gulf of Mersin, in their study. They used 2.0 mg/m3 density routinely for the shallowest, Pliocene-Quaternary sedimentary units and Upper Miocene evaporites dominated by halite. Deeper sediments of Miocene and older units were represented by 2.3 mg/m3 density. Sedimentary unit densities used by Ergun et al. (2005) are quite low, but the oceanic crustal density and the mantle density value used in their models are reasonable (3.0 mg/m3 and 3.38 mg/m3 respectively). The density used for the Upper Miocene and younger units is low (2.67 mg/m3 versus 2.0 mg/m3). These abnormal density values may be the reason for the inconsistencies of models along 4 profiles in their study and other investigations. Casten and Snopek (2006) prepared three different maps around the Hellenic subduction zone with gravity modeling; triangular-shaped continental crust without sedimentary cover, arc-shaped continental Moho and, arc-shaped and subducted oceanic Moho. The arc-shaped continental Moho map around the Crete island is similar to the results of this study. They found about 20 km Moho depth to the north of Crete consistent with this study. However, their map indicates 35 km Moho depth around the
196
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197
eastern and western ends of the island and to the near south. On the contrary, the crustal thickness variations are increasing up to 28e29 km at most to the SE and SW of Crete then thinning back to 17e23 km (Fig. 3) through the largest Bouguer anomaly in the Eastern Mediterranean (Fig. 2) in this study. The crustal thickness variations increase up to 35 km around the Palmyrides, Syria in this investigation. Barazangi et al. (1993) found a flat Moho depth around 36e37 km that is very close to the variations in this investigation. Cowie and Kusznir (2012) presented crustal thickness and oceanic lithosphere distribution from the gravity inversion according to different break-up ages of West and Central African Rift System. They suggested another rift in the Levantine Basin extending from far offshore of the Nile delta, Egypt to the SE of Cyprus using a late Triassic (225 Ma) break-up age. They also explained the linear extension of the NE shoreline of Libya and NW shoreline of Egypt down to Alexandria as a transform fault controlled margin kinematics. They presented two cross-sections, one in the N-S direction crossing the shallowest part in the Herodotus Abyssal Plain, and the other in the NWeSE extension in the Levantine Basin (Cowie and Kusznir, 2012). The crustal thickness values are consistent with our results in the shallow zones in their cross-sections. However, deeper parts on the sections are getting steeply deeper through the shorelines (about 35e40 km) that such deep values can only be observed around Palmyrides. This inconsistency may be arised from the inversion difficulties in the transition zone between the onshore and offshore areas. There are some recent crustal thickness studies in the onshore parts of Turkey. The investigations focused or including particularly on the Aegean region were compared with this investigation, below. Tezel et al. (2007) found thickness variation about 25e40 km from western to the eastern Turkey by using surface wave dispersion analysis. In the study of Tezel et al. (2010), they only focused on the crustal thickness of western Turkey using the receiver function and the results are varying between 20 and 35 km. The most recent research based on the gravity data (Ates et al., 2012) was covering whole onshore area of Turkey. The average crustal thickness variation is in the range of 25e26 km in the onshore Aegean and Mediterranean regions and 42 km in the eastern part, near the Iranian border. These values are consistent with the contour values of crustal thickness in this research and previous investigations except Ergun et al. (2005) because they calculated 38 km crustal thickness beneath the Taurus Mountains. The locations and boundaries of critical tectonic units and causative bodies were defined by the Analytic Signal (AS) and maxspots maps of the gravity anomalies in this study. Almost all of the causative bodies and their boundaries are consistent with the produced maps. However, there are some easily recognizable changes and inconsistencies, particularly in the onshore part of Turkish Mediterranean region and around Palmyrides in Syria that are represented by negative gravity anomalies. Moreover, the AS anomalies and maxspots are not compatible with each other in these regions. This may be derived from the sparsity and low sensitivity of the gravity data in these areas. Thus, the Mediterranean onshore part of Turkey is composed of the rugged and severely tectonized Taurus Mountains where the data acquisition is quite difficult. In case of acquiring and compiling very dense and detailed gravity data in these problematic regions, more detailed results will be obtained and the research area will be subjected for the further investigations. Acknowledgments The authors thank the General Directorate of the Mineral Research and Exploration (MTA) of Turkey for the analog map of the
gravity data. The analog map was digitized by a group of undergraduate students under the supervision of Prof. Dr. Abdullah Ates at the Ankara University, Department of Geophysical Engineering. We appreciated for their enthusiasm. References Ates, A., Kearey, P., Tufan, S., 1999. New gravity and magnetic anomaly maps of Turkey. Geophys. J. Int. 136, 499e502. Ates, A., Bilim, F., Buyuksarac, A., Aydemir, A., Bektas, O., Aslan, Y., 2012. Crustal structure of Turkey from aeromagnetic, gravity and deep seismic reflection data. Surv. Geophys. 33, 869e885. Barazangi, M., Seber, D., Chaimov, T., Best, J., Litak, R., Al-Saad, D., Sawaf, T., 1993. Tectonic evolution of the northern arabian plate in western Syria. In: Boschi, et al (Eds.), Recent Evolution and Seismicity of the Mediterranean Region. Kluwer Academic Publishers, Netherlands, pp. 117e140. Bilim, F., Ates, A., 2003. Analytic signal inferred from reduced to the pole data. J. Balk. Geophys. Soc. 6, 66e74. Bilim, F., Ates, A., 2005. Analitik Sinyal yontemlerinin manyetik model verileri uzerinde karsilastirilmasi. Istanb. Univ. Muh. Fak. Yerbilim. Derg. 18, 151e162 (in Turkish with English Abstract). Blakely, R.J., 1996. Potential Theory in Gravity and Magnetic Applications. Cambridge University Press, p. 441. Blakely, R.J., Simpson, R.W., 1986. Approximating edges of source bodies from magnetic or gravity anomalies. Geophysics 51, 1494e1498. Bozkurt, E., Mittwede, S.K., 2005. Introduction: evolution of continental extensional tectonics of western Turkey. Geodin. Acta 18, 153e165. Casten, U., Snopek, K., 2006. Gravity modelling of the Hellenic subduction zone-a regional study. Tectonophysics 417, 183e200. Cordell, L., Grauch, V.J.S., 1985. Mapping basement magnetization zones from aeromagnetic data in the San Juan basin, New Mexico. In: Hinze, W.J. (Ed.), The Utility of Regional Gravity and Magnetic Anomaly Maps, pp. 181e197. Soc. Explor. Geophys. Cowie, L., Kusznir, N., 2012. Mapping crustal thickness and oceanic lithosphere distribution in the Eastern Mediterranean using gravity inversion. Pet. Geosci. 18, 373e380. Di Luccio, F., Pasyanos, M.E., 2007. Crustal and upper-mantle structure in the Eastern Mediterranean from the analysis of surface wave dispersion curves. Geophys. J. Int. 169, 1139e1152. ElGabry, M.N., Panza, G.F., Badawy, A.A., Korrat, I.M., 2013. Imaging a relic of complex tectonics: the lithosphere-asthenosphere structure in the Eastern Mediterranean. Terra Nova 25, 102e109. Ergun, M., Okay, S., Sari, C., Oral, E.Z., Ash, M., Hall, J., Miller, H., 2005. Gravity anomalies of the Cyprus Arc and their tectonic implications. Mar. Geol. 221, 349e358. Ketin, I., 1948. Uber die tektonische-mechanischen Folgerungen aus den grosssen Anatolischen erdbeben des letzten dezenniums. Geol. Rundsch. 36, 77e83. Le Pichon, X., 1982. In: Hsu, K. (Ed.), Land-locked Oceanic Basins and Continental Collision: the Eastern Mediterranean as a Case Example. Mountain Building Processes Academic Press, London, pp. 201e211. Le Pichon, X., Angelier, J., 1979. The Hellenic arc and trench system: a key to the evolution of the Eastern Mediterranean area. Tectonophysics 60, 1e42. Le Pichon, X., Chamot-Rooke, N., Lallemant, S., Noomen, R., Veis, G., 1995. Geodetic determination of the kinematics of Central Greece with respect to Europe: implications for Eastern Mediterranean Tectonics. J. Geophys. Res. 100, 12675e12690. Makris, J., Stobbe, C., 1984. Physical properties and state of the crust and upper mantle of the Eastern Mediterranean Sea deduced from geophysical data. Mar. Geol. 55, 347e363. McClusky, S., Balassanian, S., Barka, A., Demir, C., Ergintav, S., Georgiev, I., Gurkan, O., Hamburger, M., Hurst, K., Kahle, H., Kastens, K., Kekelidze, G., King, R., Kotzev, V., Lenk, O., Mahmoud, S., Mishin, A., Nadariya, M., Ouzounis, A., Paradissis, D., Peter, Y., Prilepin, M., Reilinger, R., Sanli, I., Seeger, H., Tealeb, A., Toksoz, M.N., Veis, G., 2000. Global positioning system constraints on plate kinematics and dynamics in the Eastern Mediterranean and Caucasus. J. Geophys. Res. 105, 5695e5719. McKenzie, D., 1972. Active tectonics of the mediterranean region. Geophys. J. Int. 30, 109e185. McKenzie, D.P., 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions. Geophys. J. R. Astr. Soc. 55, 217e254. Ozelci, H.F., 1973. Dogu akdeniz bolgesi gravite anomalileri. MTA Bull. 80, 54e88. Reilinger, R.E., McClusky, S.C., Oral, M.B., King, R.W., Toksoz, M.N., Barka, A.A., Kinik, I., Lenk, O., Sanli, I., 1997. Global positioning system measurements of present-day crustal movements in the arabia-africa-eurasia plate collision zone. J. Geophys. Res. 102, 9983e9999. Riad, S., El Etr, H.A., 1985. Bouguer anomalies and lithosphere-crustal thickness in Uganda. J. Geodyn. 3, 169e186. Riad, S., Refai, E., Ghalib, M., 1981. Bouguer anomalies and crustal structure in the Eastern Mediterranean. Tectonophysics 71, 253e266. Rivero, L., Pinto, V., Casas, A., 2002. Moho depth structure of the eastern part of the Pyrenean belt derived from the gravity data. J. Geodyn. 33, 315e332. Seber, D., Sandvol, E., Sandvol, C., Brindisi, C., Barazangi, M., 2001. Crustal model for the Middle East and North Africa region: implications for the isostatic
F. Bilim et al. / Marine and Petroleum Geology 77 (2016) 190e197 compensation mechanism. Geophys. J. Int. 147, 630e638. Sengor, A.M.C., Yilmaz, Y., 1981. Tethyan evolution of Turkey: a plate tectonic approach. Tectonophysics 75, 181e241. Sodoudi, F., Kind, R., Hatzfeld, D., Priestley, K., Hanka, W., Wylegalla, K., Stavrakakis, G., Vafidis, A., Harjes, H.-P., Bohnhoff, M., 2006. Lithospheric structure of the Aegean obtained from P and S receiver functions. J. Geophys. Res. 111 (B12307), 1e23. Staackmann, M., Casten, U., Ates, A., 2008. Gravity controlled Moho depths in the aegean region. EGU general assembly. Geophys. Res. Abstr. 10. EGU2008-A08760. Tezel, T., Erduran, M., Alptekin, O., 2007. Crustal shear wave velocity structure of Turkey by surface wave dispersion analysis. Ann. Geophys. 50, 177e190. Tezel, T., Shibutani, T., Kaypak, B., 2010. Crustal structure variation in western
197
Turkey inferred from the receiver function analysis. Tectonophysics 492, 240e252. Tirel, C., Gueydan, F., Tiberi, C., Brun, J.P., 2004. Aegean crustal thickness inferred from gravity inversion. Geodynamical implications. Earth Planet. Sci. Lett. 228, 267e280. Van der Meijde, M., Van der Lee, S., Giardini, D., 2003. Crustal structure beneath broad-band seismic stations in the Mediterranean region. Geophys. J. Int. 152, 729e739. Wortel, M.J.R., Goes, S.D.B., Spakman, W., 1990. Structure and seismicity of the Aegean subduction zone. Terra Nova 2, 554e562. Zelt, B.C., Taylor, B., Sachpazi, M., Hirn, A., 2005. Crustal velocity and Moho structure beneath the Gulf of Corinth, Greece. Geophys. J. Int. 162, 257e268.