3.0 Analysis and Results Using the commercial software package Matlab [8], data was converted into the ASCII file format which enabled further processing. A baseline correction was also performed on the raw data using a Butterworth high pass filter at 0.1%. After this, data was ready to be processed with Nakamura‟s technique and The Frequency Domain Decomposition (FDD) method. 3.1 Data Analysed by Nakamura’s H/V Technique Nakamura‟s method of analysing ambient vibration measurements was first proposed in 1989 by Yutaka Nakamura [6]. Despite several recognized shortcomings of Nakamura‟s method, such as questionable ability to identify which wave types cause microtremors, recently it has gained popularity as an accurate and cost-effective way of determining the fundamental site period [1]. Nakamura proposes that the soil layer does not amplify the vertical component of ambient noise. Therefore, the period at which a plot of the ratio of horizontal motion to vertical motion (H/V) peaks should correspond directly to the fundamental period of that site [1]. Using the free software package Geopsy [7], time history plots were generated for each test as can be seen in Figure 3. From the time history plot, Geopsy‟s window selection tool was used to select “windows” of a desired amount of time to be used in further analysis. The goal of window selection was to remove the influence of flawed data in determining the peak in the H/V ratio plot. Any large spikes in time history plot were avoided, as well as segments of the test that were influenced by other noise sources, such as a lawnmowers or traffic in very close proximity. Windows were typically 40 seconds in length, as recommended by SESAME guidelines [12]. Fig. 3 Time History Plot in Geopsy 454
RkJQdWJsaXNoZXIy MTMzNzEzMQ==