3-D velocity structure of eastern Marmara region from local earthquake tomography

Abstract

Local earthquakes located in 40-41.5 N and 28-30.5 E geographic coordinates between the dates January,2003-October,2007 were selected to use for local earthquake tomography. Data were obtained from short period and broadband seismic stations dispersely located in the region. This study was conducted to reveal 3-D P wave velocity structure in the target area. To prepare data set, relocation procedure was done. To determine the 1-D velocity model for the study region, earthquakes which have azimuthal gap less than 1800 and recorded at least by 7 stations were selected. Obtained 1-D velocity model has 32 km depth. Reliable solutions were observed to 17 km due to the depth distributions of the earthquakes. To consolidate the robustness of the velocity model, shifting test was applied and quarry blast data were used. Results show that the 1-D velocity model represents the study region. Using reference earth model for local earthquake tomography was decided. Later on, observing more confidential parts of the velocity changes in the study region, synthetic models were produced for the checkerboard test. At the same time, control parameters were set to obtain reasonable solution after running tomography algorithm. 3-D tomographic inversion based on damped least square inversion was applied to 14329 P wave arrival times and the results of 3-D tomographic inversion were tried to correlate with geologic verifications in the region. Generally low velocities change between 5.3-5.7 km/s through vertical extension of the faults. Vertical extensions of the NAF branches are observed between 2 -15 km depths when the resolution parameter of the data set taken into account. These properties are relatively similar for extracted profiles. Location of the vertical extensions of the fault zones mainly show the same depth range similar to done previous studies in the region.