Batı Anadolu Kara Ve Kıyı Ötesi Ml≥5.0 Depremlerine Ait Artçı Şok Dizilerinin (2005-2015) İstatistik Özellikleri

Abstract

Bu çalışmada Batı Anadolu kara ve kıyı ötesi ML≥5.0 depremlerinin artçı şok alanlarında, ana şokları izleyen 150 günlük sürede meydana gelen artçı şok dizilerinin istatistik özellikleri, KRDAE-UDİM deprem kataloğu (2005-2015) kullanılarak incelenmiştir. Artçı şokların istatistik özelliklerini incelemek için Gutenberg-Richter yasasına göre b değeri ve geliştirilmiş Omori yasasına göre p değeri ZMAP program paketi kullanılarak hesaplanmıştır. İncelenen 17 deprem için artçı şok alanında, 150 günlük sürede meydana gelen artçı şok sayılarının 10 – 2914 arasında ve hesaplanan tamamlılık magnitüdü (Mc) değerlerinin ise 1.8 – 3.2 arasında değiştiği saptanmıştır. Bu 17 ana şokun artçı şok dizileri için hesaplanan b değeri aralığı 0.69±0.002 – 2.51±0.4 olup, p değeri aralığı ise 0.33±0.008 – 2.25±0.19’dur. Bu depremlerden artçı şok sayısı göreceli olarak yeterli (432 – 2914) ve b ve p değerlerinin standart sapmaları düşük (sırasıyla 0.02 – 0.05 ve 0.03 – 0.19) olan 5 adet depremin (24 Mayıs 2014 ML=6.5 Gökçeada açıkları, Ege Denizi; 10 Haziran 2012 ML= 6.0 Ölüdeniz açıkları, Muğla; 19 Mayıs 2011 ML=5.9 Kütahya, Simav; 17 Ekim 2005 ML=5.9 İzmir Seferihisar; 3 Mayıs 2012 ML=5.1 Kütahya) istatistik özellikleri yorum aşamasında kullanılmıştır. Çalışma bölgesindeki jeotermal alanlarda meydana gelen depremlerin artçı şok dizileri için göreceli olarak daha yüksek b ve p değerleri (>1) hesaplanmıştır. 19 Mayıs 2011 ML=5.9 Kütahya, Simav depremi artçı şok dizisi için b değeri 1.77±0.05 ve p değeri 1.09±0.04 olarak, 17 Ekim 2005 ML=5.9 İzmir Seferihisar depremi artçı şok dizisi için b değeri 1.26±0.03 ve p değeri 2.25±0.19 olarak hesaplanmıştır. Ana şokun meydana geldiği fayın türüne göre ana şoklara ait artçı şok dizileri için hesaplanan b değerlerinin değişim gösterdiği saptanmıştır. Normal fay üzerinde meydana gelen 19 Mayıs 2011 ML=5.9 Kütahya, Simav depremi artçı şok dizisi için hesaplanan b değeri 1.77±0.05’dir. Doğrultu atımlı fay üzerinde meydana gelen 10 Haziran 2012 ML= 6.0 Ölüdeniz açıkları, Muğla ve 24 Mayıs 2014 ML=6.5 Gökçeada açıkları, Ege Denizi depremi artçı şok dizileri için hesaplanan b değerleri sırasıyla 0.728±0.04 ve 0.69±0.02’dir. Sonuç olarak normal fay üzerinde meydana gelen depremlerin artçı şok dizileri için hesaplanan b değerlerinin doğrultu atımlı fay üzerinde meydana gelen depremlerin artçı şok dizileri için hesaplanan b değerlerinden daha yüksek olduğu saptanmıştır.

In this work, we analyzed statistical properties of aftershock sequences of ML≥5.0 earthquakes occurred in the land and offshore of western Anatolia (36°-42°N and 26°-30°E) by using Kandilli Observatory and Earthquake Research Institute, National Earthquake Monitoring Center’s (KOERI, NEMC) earthquake catalog (2005-2015). The Anatolian plate between the right-lateral strike-slip North Anatolian Fault Zone (NAFZ) and the left-lateral strike-slip East Anatolian Fault Zone (EFZ) escapes westward from the eastern collision zone between the Arabian and Eurasian plates. The extension in western Anatolia is caused by the westward movement of the Anatolian-Aegean region in a counter clockwise direction (W-SSW) with increasing rates from 20 mm/y in central Anatolia to 30 mm/y near the Hellenic trench. The crust beneath western Anatolia is dominated by graben systems and thermal structures of widespread young alkaline volcanic rocks. The regional heat flow is about 110 mW/m2. The geothermal systems in the west Anatolia lie within or along the E-W trending graben systems. The shallowest Curie Point Depth (CPD) regions are located in the west of Kütahya (~ 7 km) and the northeast of Denizli (~ 9 km). The crust in the west Turkey is thin near the coastlines (~ 28 km) compared to the crust in the east Turkey (~42 km). Crustal earthquakes beneath western Anatolia and Aegean region occur along the strike-slip faults and the normal faults and deep earthquakes occur along the subduction zones along the Hellenic and Cyprian arcs. In general, there are three main types of earthquake sequences; Type I : the main shock-aftershocks sequence, Type II : the foreshocks-main shock-aftershocks sequence, and Tyep III : the swarm. Type I occurs in homogeneous material with a uniform external stress. A number of aftershocks with decreasing magnitude and frequency following the mainshock is observed. Type II occurs in heterogeneous material in some degree with a non-uniform external stress. A slow build up seismicity (foreshocks) is observed previous to Type I sequences. Type III occurs in extremely heterogeneous material with very concentrated external stress. They are commonly observed in volcanic or fractured areas of magma intrusions. In this study, the following three laws are applied to analyze statistical properties of the aftershock sequences: Gutenberg-Richter’s law, the modified Omori law and Bath’s law. Gutenberg-Richter’s law describes a linear relationship defined by a and b parameters to model the frequency magnitude distribution of the aftershock sequences. Parameter a is a measure of seismic activity of a specific volume. Parameter b is a measure of the size distribution of events in the same volume. The b values are used to explain the seismicity and the physical properties of the crust in the different tectonic settings. Analyzing aftershock sequences is very important in terms of giving insight about crustal structure and stress distribution after a main shock. There is a great literature on b values. The range of estimated b value of the Gutenberg-Richter relationship for the aftershock sequences is between 0.6 and 1.4 with an average of 1.0. The b value is dependent on the mechanical structure of materials and stress conditions. The b value increases with the ductility of rock. An increased material heterogeneity increases the b value. The b-value decreases when stress increases. An increase in the thermal gradient may cause an increase in the b value. In volcanic areas, high b values are observed near magma chambers and highly cracked volumes. In creeping sections of faults, high b values are found, whereas asperities exhibit a low b values. The modified Omori law describes an exponential relationship defined by k, c and p parameters to explain the decay of aftershock rates with time. Parameter k and c are related to the total number of aftershocks above given cutoff and the rate of aftershock activity during the initial part of the sequence. The parameter p, which is a measure of the decay rate of aftershocks, changes from 0.9 to 1.5 and its variability may be related to the structural heterogeneity, stress and temperature in the crust. The fast decay rate of aftershock activity is associated with high regional heat flow and the presence of less spatial heterogeneities of the fault shear strength. High temperatures accelerate the stress relaxation in the aftershock area. High p values are calculated for the areas where the larger slip occurs during the main shock. Low p values are calculated in the regions where the crustal heat flow is low and the aftershock activity decays gradually. Low p values are calculated for non-ruptured areas. Bath’s law is a statistical law describing the difference in magnitude between a main shock and its largest recorded aftershock as nearly constant 1.2. The selection of aftershock area and duration for aftershock sequences are important to analyze the statistical properties of aftershock sequences.We considered a circle to select the aftershock area. The radius of the circle is defined by an empirical formula with surface magnitude Ms of the main shock. We used aftershocks occurred in the aftershock area during 150 days following the main shock. To analyze statistical properties of aftershock sequences of ML≥5.0 earthquakes (2005-2015) occurred in the land and offshore of western Anatolia, we used ZMAP program package, which is Mat-lab based, open-source code. ZMAP includes a set of tools to help seismologists to analyze the catalog data for the catalog quality assessment, b and p value mapping, etc. This program package allows user to select quickly subsets in space, time, and magnitude, plot histograms, compute b or p values, compare the frequency-magnitude distributions of different time periods and locations, compare day time versus night time activity, compute the fractal dimension of hypocenters, create cross sections, overlay topography, compute stress-tensor inversions. Maps are computed on an interactively defined grid excluding low-seismicity areas. There are two methods in ZMAP to map seismicity: a) constant radii and b) a constant number of samples. The constant radii method produces maps with a continuous spatial resolution but varying sample sizes. Therefore, uncertainties can vary significantly in space. A constant sample size method results in more homogeneous uncertainties, but the resolution, which is inversely proportional to the density of earthquakes, will vary across the region of interest. Since the magnitudes in the selected KOERI, NEIC catalog (both Md and ML) are not given in homogenous manner, we converted the duration magnitudes Md to local magnitudes ML before using the catalog data. Then, the completeness magnitude (Mc), which is minimum magnitude above all earthquakes are reliably recorded, is determined by using the maximum curvature method. The b values are estimated by using both weighted least squares and maximum likelihood methods. The constant radius method is used to produce b and p value maps for the aftershock areas. In the interpretation step, the b and p value maps are used together with their standard deviation and goodness of fit maps. We investigated b and p values for the aftershock areas of 17 main shocks occurred during 150 days following the main shocks. The range of number of aftershocks for 17 main shocks is between 10 and 2914. The range of the completeness magnitudes of the aftershock sequences for these main shocks is between 1.8 and 3.2. The ranges of the estimated b and p values for the aftershock areas are 0.69±0.002 – 2.51±0. and 0.33±0.008 – 2.25±0.19, respectively. The average completeness magnitude is around 2.64 which is relatively high and indicates that the microearthquake activity is not equally monitored in time and space domain. The standard deviations of b and p values are low (0.02 – 0.05 and 0.03 – 0.19) for five main shocks (24 May 2014 ML=6.5 Gökçeada offshore, Aegean Sea; 10 June 2012 ML= 6.0 Ölüdeniz offshore, Muğla; 19 May 2011 ML=5.9 Kütahya, Simav; 17 October 2005 ML=5.9 İzmir Seferihisar; 3 May 2012 ML=5.1 Kütahya) having relatively high number of the aftershocks (432 – 2914). Therefore, we used mainly the results of these five main shocks in the interpretation step. We determined relatively high b and p values (>1) for the aftershock series of main shocks occurred within the geothermal areas. The b and p values are estimated as 1.77±0.05 and 1.09±0.04 for 19 May 2011 ML=5.9 Kütahya, Simav earthquake’s aftershock sequences, 1.26±0.03 and 2.25±0.19 for 17 October 2005 ML=5.9 İzmir Seferihisar earthquake’s aftershock sequences. We determined the different b values for the aftershock sequences of the main shocks occurred on the different type of the faults. We estimated the b value as 1.77±0.05 for 19 May 2011 ML=5.9 Kütahya, Simav earthquake’s aftershock series whose main shock occurred on the normal fault. We estimated the b values as 0.728±0.04 ve 0.69±0.02 for 10 June 2012 ML= 6.0 Ölüdeniz offshore, Muğla, 24 May 2014 ML=6.5 Gökçeada offshore, Aegean Sea earthquakes’ aftershock sequences whose main shocks occurred on the strike slip faults. We concluded that the estimated b values for the aftershock sequences whose main shocks occurred on the normal fault are higher than the estimated b values for the aftershock sequences whose main shocks occurred on the strike slip fault. The results of this study suggest that the strategy for the station distribution and objective of recording micro earthquake activity should be revisited to provide improved data base to investigate b and p values and the asperities in both space and time domain within the aftershock areas. In addition, the reporting location errors and magnitude types in homogenous manner in the catalogs will be important for the detailed studies in the aftershock areas.