Determination of recent tectonic activity in the central part of the North Anatolian Fault

Kuzey Anadolu Fayı‟nın orta bölümündeki güncel tektonik aktivitenin belirlenmesi

Abstract

One of the important fault systems all over the world is the North Anatolian Fault (NAF). The NAF is a long fault which stretches from Karlıova to Northern Aegean. Although a lot of studies exist, the central part of the NAF has not been studied in detail. In this study, the central part of NAF from Amasya to Kastamonu has been evaluated using geodetic tech-niques, GPS (Global Positioning System) and InSAR (Synthetic Aperture Radar Interferometry), and modeled with DEFNODE software. The geodetic techniques to determine the tectonic activity have been used since early 1900. The iner-tial methods were not sufficient to calculate the strain accumulation or slip rate of the fault. Espe-cially, the interseismic studies cannot be possible. After the space geodesy development, the earth sci-ences started to use these techniques. The most use-ful techniques of them are GPS, InSAR, VLBI and SLR. In this study, we used GPS and InSAR tech-nique because of the financial constrains, mobility and accuracy. The GPS campaigns were carried out in four years periodically at 16 force-centered sta-tions. Every year, the measurements were done in the same week. Each station was measured at least 8 hours in three days. All stations were not measured simultaneously because of the lack of equipments. So, we had continuous stations to link measurements which were carried out in two stages. The measure-ments obtained from campaigns and downloaded from internet belonging to IGS sites close to study area were processed using GAMIT/GLOBK soft-ware. The process was run in three steps. First step was made with GAMIT, the all GPS data belonging to same day had put in the same day directory and the required files for process were updated from process center like SOPAC (Scripps Orbit and Per-manent Array Center). The stations coordinates and its covariance matrix were calculated. In this step, the repeatabilities were checked to protect data against to blunder and systematic errors. Second step, the calculated data (called quasi-observation) were combined with the international global net-work data computed by center, SOPAC. Each data had different sites that the all quasi-observation should have various weights. In the third step, the combined data were transformed in the reference frame like ITRF (International Terrestrial Reference Frame) using common sites and transform parame-ters. After that, we calculated the sites velocities ac-cording to Eurasia plate in ITRF2000 reference frame using sites which have well known coordi-nates and have represented Eurasia plate very well. The GPS results are given in the text. InSAR method depends on phase information in two SAR images by calculating phase difference between each pair of corresponding image points. The new image with interference pattern of fringes due to relative phase difference is called an interferogram. Phase change in the interferogram shows differ-ences in the range distance between image acquiring platform (aircraft, satellite) and point on the ground. One complete phase cycle (2π radians) corresponds to a relative range change by half of wavelength. Main InSAR components are three, first is the im-age, at least two, second is the topography removal tools, for example, other interferogram that does not contain deformation or DEM (Digital Elevation Model), third is orbital information. In this study, we used the ERS-1 and ERS-2 (European Remote Sens-ing satellite) data obtained from ESA (European Space Agency) with track numbers 164 and 300. All the possible combinations were processed but the expected phase gradient was not calculated as useful as it was useful to determine tectonic activity. Main results of these are the atmospheric effects on the interferogram. The topography also rises to effect because of the gradient differences. Finally, GPS results were modeled by DEFNODE software developed by Prof. Dr. Robert McCaffrey. The DEFNODE calculated strain and slip rate on the fault using velocity vectors of sites, locking depth, fault geometry and continental block defini-tions. The fault geometry was obtained from field work. The fault geometry described the continental block boundaries. All modeling results calculated from the software are shown in the text. According to results obtained from GPS and InSAR, the topography and atmosphere are important for InSAR to calculate interseismic strain accumulation of the hilly region. The other important result is that the main branch of the NAF has faster slip rate than splays of the NAF in the study area.

Özet

Dünyanın en aktif yanal atımlı fay sistemlerinden birisi olan Kuzey Anadolu Fayı (KAF), doğuda Karlıova’dan başlar batıda Kuzey Ege Denizi’nde Saros Körfezi’ne kadar uzanır. Kuzey Anadolu Fay Zonu (KAFZ) içinde ağaç dalları şeklinde birçok yan kolu vardır ve bu yan kollar Anadolu’nun içlerine doğru uzanmaktadır. Tüm fay zonu sismik olarak aktif ise de, fayın aktivitesinin önemli bir bölümü ana kol üzerindedir. Kuzey Anadolu Fayı’nın orta kesiminde yan kolların keserek birbirin-den ayırdığı birçok kıtasal blok vardır. Bölgedeki nispeten en büyük yan kollar inceleme için seçil-miş ve aralarındaki hareket belirlenmeye çalışılmıştır. Fay sistemlerinin güncel metotlarla izlenme-si 1980’li yıllardan sonra uydu jeodezisinde meydana gelen önemli ilerlemeler sayesinde mümkün olmuştur. Bu bağlamda, yersel tekniklerin yerini uzay ve uydu tekniklerinin alması genel fotoğrafın görülmesinde büyük kolaylıklar sağlamıştır. Bu çalışmada, jeodezik metotlardan GPS (Küresel Ko-numlandırma Sistemi) ve InSAR (Yapay Açıklıklı Radar İnterferometri) kullanılmıştır. Çalışmanın amacı iki metot ile sonuçlar üreterek karşılaştırmak ve/veya metotların zayıf yönlerini birbiri ile tamamlamak ve böylece bölgenin depremselliğini ortaya koymaktır. GPS ile kurulan jeodezik ağ dört yıl boyunca periyodik olarak ölçülmüştür. InSAR tekniği ile bölgeyi kapsayan görüntülerden interferogramlar oluşturulmuş ve görüntü alım zamanları arasında yüzeyde meydana gelen uydu bakış yönündeki değişimler hesaplanmaya çalışılmıştır. InSAR metodu atmosfer ve bölge topograf-yasına bağlı olarak istenen sonuçlara ulaşmamıştır. GPS sonuçlarının modellenmesi ile yan kolla-rın ana kola göre nispeten yavaş olduğu belirlenmiştir.