The Usage of Correlation Method for Micro-Earthquake Analysis at Salavatlı Geothermal Area, Aydın, Turkey.

Abstract:

Induced seismicity typically refers to minor earthquakes and tremors that are caused by human activity that alters the stresses and strains on the Earth's crust. Earthquakes which are smaller magnitude are called micro-earthquakes. Micro-earthquakes could be observed in areas which includes energy technologies that involve injection or withdrawal of fluids from the subsurface. Micro-earthquakes are the result from these kind of process. Geothermal areas are good example of that kind of areas. So we chose Salavatlı geothermal area , Aydın, Turkey to study. In this area, we have working production and injection wells, and these wells are active since 2000.

The main aim of this study is to find similar micro-earthquakes with correlation method and to check if they come from same source or location. So the data which was collected from 9 broad-band stations which were installed at the study area, between June 2010 and April 2013 were used for this study. 977 micro-earthquakes were located with SEISAN. 815 of them could be used for the correlation analysis due to quality of the data. GISMO correlation toolbox was used for this process. The duration of waveforms are 10 seconds because the longest micro-earthquake in this study was 8 seconds. We took lower limit as 0.9 for the correlation co-efficient. 34 event of similar waveforms were found.

When we examined location of all micro-earthquakes, we saw that they were scattered over the study area. This might be caused from wrong P or S wave pickings or inadequate crustal velocity model. So, firstly, we decided to check the initial velocity model by using VELEST software. For the VELEST processing 334 best located micro-earthquakes were selected on the base of similiar criterias. After getting minimum 1D velocity model for the study area, all micro-earthquakes were relocated again and then mapped. The location of most of them changed and got better. Then with final 1D velocity model, events of similar waveforms were mapped. An example of the similar waveforms can be seen in the figure. In general, when monthly location of micro-earthquakes were checked, the direction of the injected water from re-injection wells became clearly seen on the seismicity maps.