Geophysical study and monitoring of the recorded events nearby the Dokan Dam , NE Iraq

In this study an attempt has been made to monitor and perform analysis to the occurred earthquakes recorded by the installed seismometers in the close proximity to the Dokan dam. Three-component continuous high resolution data logger seismometers had recorded thousands of seismic events from January to December 2013. Twenty five earthquakes were utilized for further processings that covered the studied area and surroundings. The analyses include identifying the body waves from different azimuths within optimal range of frequencies from 3 to 500Hz. The waveforms were used in determining the various source parameters including location, focal depth and magnitude from their arrival times and amplitudes. Epicentral map was drawn which depicts the epicentral distances of the events from the source. Furthermore, the estimated crustal structure below the studied area reveals that the seismic activities were occurred in shallow depths (less than 10 km) within the upper part of the crust and the resulted average values for the Pand S-wave velocities were 5.97 km/sec and 3.23km/sec respectively.


1.INTRODUCTION
In this paper an investigation had been done to gather the seismic data and to sift through this data to identify potential events. Iraq is located in the northern Arabian plate including the western edge of the zagros mountain range, where the convergent tectonic boundary between the Eurasian and Arabian plates is revealed by a fold and thrust belt [1]. In general, Iraq has a rather well-documented history of seismic activity. The seismic history for this region reveals annual seismic activity of different strength. The northern part of Iraq depicts the highest seismic activity with strong diminution of earthquakes in the southern and southwestern parts of the country [2]. The seismic activity when occurred, naturally or induced, the vibrations travel outwards through the ground from the source. Each event radiates seismic waves that travel throughout Earth, and several earthquakes per day produce distant ground motions that, although too weak to be felt, are readily detected with modern instruments anywhere on the globe [3]. Also, initial review of collected data and published bulletins confirm that a large number of occurring small events (magnitude < 4) are either not being recorded or detected by distant stations, or they are not being reported by the scarce number of neighboring seismic stations in Turkey and Iran.

2.SEISMIC INSTRUMENTION
The study area is located at latitude 35° 57' ‫״6.22‬ and longitude 44° ‫״31'75‬ in the northeastern part of Iraq. Dokan hydroelectric dam Figure.(1) is equipped with the seismic monitoring system, with its own internal GPS engine that receives accurate time signals to observe and detect the seismicity over time in three directions which greatly enhanced the triaxial force-based digitization at 24bit-sps recordings of local and regional seismic activities without the risk of losing any of the data. This is used to monitor the ground vibration of the crest, and records the effects that any large scale seismic events (earthquakes) may have on the structure of the dam. Also, seismographs include triaxial geophones were installed in shallow boreholes and powered by solar panels.

3.DATA PROCESSING AND SPECTRAL ANALYSIS
Seismic signals are made up of waveforms composed of different frequencies. Continuous analog data were converted into discrete digital data which is then analyzed and processed.
More specifically, analog signals are sampled at a specific sampling frequency. The seismic waves observed in earthquake records manifest clearly non-stationary characteristics, as well as a wide frequency content [4]. These would be used to gain further information about individual event properties such as source location or other source parameters. In processing technique, the date and arrival-time (h-m-s) for each event is important. If events are triggered on noise rather than legitimate seismic data, it is possible to remove, the processing associated with this event or the events that are deemed to be noise. Noise may occur at certain frequencies, and are associated with an event which can be identified and removed from the true data signal by noise frequency filter. The most common signal processing operation is to filter the signals to enhance certain features and suppress others [5]. Arrival times are processed which identify when the p-wave and s-waves are first detected.
The P-wave and S-wave picks were fixed on the waveform Figure.(3). Filtering is used to improve the quality of the signals by removing noise to obtain optimal Signal/Noise ratio at the time of P-and S-wave picks. (t s -t p ) Where (t s -t p )= time difference taken from the earthquake record between the arrival of Pand S-waves , v p = velocity of P-waves, v s = velocity of s-waves.
Fast fourier transform is a mathematical routine used to convert seismic waveforms from the time domain to the frequency domain. Performing this maintenance on the data will help improve the performance of the arrival time pickers, and thus will obtain more accurate source location and source parameter calculations. It is important to note that events must be source located before source parameter calculations are performed [ 8].
Determining the magnitude and azimuth of the events is an integral part of processing earthquake data and is done routinely with nearly all earthquakes located, whether global or local. The magnitude of an earthquake is normally estimated by measuring the ground amplitudes record at stations. The general form of empirical equation defining magnitude is [9]: Where A is the maximum ground amplitude in micrometers of the wave used, T the wave period in sec, Q is an empirical function of epicentral distance, the distance and h is the focal depth. An example for the azimuth and magnitude determination for the event which occurred in 22 Nov 2013 is shown in Figure.(4). Information about the frequency content, amplitude of the signals and using the hyperion programs ver.14.0 helped in estimating the source location.
However,the crustal structure and velocities may differ significantly from region to region, and that the event location can be significantly improved when local travel-time curves or crustal models are available.Consequently,the travel time can be used as a function of depth10

4.RESULTS
The calculated source parameters are illustrated in Table.(1) for the selected events occurred in 2013. The value of the magnitude was varied between 1.1 to 5.5. The maximum focal depth beneath the studied area is less than 10 km.
The total recorded event numbers occurred in 2013 versus magnitude were plotted in Figure.(6-a) which reveals that the magnitude of most of the events were between 1-2.
Moreover, the estimated P-and S-waves velocities are nearly 5.97 km/sec and 3.23km/sec respectively as shown in Figure.(6-b). They are differentiated from the other events by the observation that no single earthquake in the sequence is obviously the main shock and not succeeded by series aftershocks with magnitudes ranging between 1.6 and 5.5.

5.CONCLUSIONS
In this study it can be concluded that the detected swarms of earthquakes striking the area in a relatively short period of time were possibly associated with the relative movements of the Eurasian and Arabian tectonic plates. While the large number of occurring small events

6.ACKNOWLEDGEMENTS
I would like to acknowledge the staff of dokan dam directorate which their assistance was facilitated the development of this work. Special thanks are due to the anonymous reviewers for their comments which improved the quality of this manuscript.