m Khandan Bakavoli - International Institute of Earthquake Engineering and Seismology, Tehran, Iran
e Haghshenas - International Institute of Earthquake Engineering and Seismology, Tehran, Iran
m Amel Sakhi - Faculty of Engineering, Urmia University, Urmia ,Iran

This paper presents the results of an experiment carried out in order to study topographic site effect. The hill was instrumented using 7 seismometers, which were installed across the irregularity. Two stations were located on each side of the foothill, two stations at the top and the other three at the middle part of the hill. 58 events were recorded by the aforementioned temporary velocimetric network. These events were recorded at reference station and at least three other stations. The recorded motions were analyzed using different methods like standard spectral ratio and horizontal to vertical ratio. The window length which begins with S waves is 15 seconds for all dataset. The events are then filtered and cosine tapered in order to minimize border effects due to windowing. The frequency of vibration of the hill has been derived. The amplifying potential of the ground motion has been determined by plotting amplification curves for different stations. It is shown that both methods are consistent in predicting frequencies of vibration but shows different level of amplification. The station at the hill-top shows a relatively strong amplification at frequency range of 3-4 Hz. Besides, the events as non-stationary signals were analyzed using new signal processing techniques like time-frequency distributions and wavelet analysis. The best time-frequency distributions were selected to obtain the best results. It is shown that these techniques would be helpful in recognizing frequencies of vibration that may not be revealed by standard methods.

Topography, H/V, SSR ,Signal Processing