| Summary: | 碩士 === 國立成功大學 === 地球科學系碩博士班 === 100 === To characterize the ambient noise levels in the Taiwan region and to understand its relationship with climate parameters, we used two methods to analyze the continuous waveforms recorded at Taiwan broadband seismic stations. On the basis of these results, we are able not only to evaluate the station performance and provide crucial information for site selection, but also to realize the controlling factors of noise.
First, we extract the noise amplitude of 9 frequency bands in 2-18 Hz from seismic data collected at 18 BATS stations between 2006 and 2009. Then we compare these high-frequency noise amplitudes with ambient temperature, atmospheric pressure, rainfall and wind-speed, respectively. The results show that the high-frequency noise is affected by rainfall and wind-speed. However, 2 Hz noise exhibits seasonal variation for the stations in western Taiwan, which may reflect the seasonal change of wave climate. High-frequency noises in western Taiwan, Kenting and Taitung are larger than elsewhere, implying intensive human activities in these areas. Stations that are close to river are characterized by a delay of returning to background high-frequency noise level after typhoons.
Secondly, we follow the method proposed by MaNamara and Buland (2004) to compute the probability density functions of power spectral density (PDFs) of ambient noise. The data is collected at 95 broadband stations in BATS, TAIGER, and CWBSN from July 2006 to June 2008. The result shows that the high-frequency noise (〉 1 Hz) is much higher than the global low noise model. But the long-period noise is within the global high and low noise models. The spatial variation shows that noise level in the plain area is 20-50 dB higher than the mountainous area, and the noise level along the western coast is 10-30 dB higher than the eastern coast. This implies that the excitation of the secondary microseism is more effective in the shallower Taiwan Strait than in the deep Pacific. Noise amplitude in 1-10 Hz presents significant diurnal and weekly variation. Significant annual variation that is associated with the wave climate can be seen at temporal variation of the 2-8 s noise amplitude. We also verify that the borehole stations present lower noise level by about 20 dB when compared with corresponding surface stations. It means that the ambient noise could be much reduced by installing borehole seismic stations and therefore enhance the ability of detecting smaller earthquakes.
|
|---|
