| The Comprehensive Nuclear-Test-Ban Treaty (CTBT) was adopted by the General Assembly on Sep. 10, 1996. Annex 1 to the CTBT Protocol had listed the 50 seismological stations (arrays) comprising the primary network of International Monitoring System (IMS ), and Lanzhou seismic array (PS13) was in it. China is one of the State Party of CTBT. According to the project that the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) is building IMS, the Institute of Geophysics, China Seismological Bureau (IGCSB) was designated to be the artificial person and undertaker of Lanzhou array building project permitted by relevant ministry in China. After the field survey, the technique and implementation group of IGCSB chose Dajianshan where is 18km away from Lanzhou seismic station in the south-west as the array site among several backup sites, and enforced a site survey on August, 1999. Eignt suits of instruments supported by the Analysis and Prediction Center, CSB were used for the field observation, and 18 sites were deployed on two concentric circles and a beeline respectively. During the 18 days, we achieved about 3GB waveform data. The content of this paper is data processing and design of Lanzhou seismic array.According to CTBTO, IMS array comprises 9 vertical Short Peroid (SP) seismometers and a broad-band seismometer. Usually, the best practical design of IMS array adopt concentric circles deployment, i.e. to deploy 3 SP seismometers on the inner ring and five ones on the outer ring with near equidistance, and then with one broad-band seismometer at the center. The elements on the two rings should avoid being deployed on a line in the radial direction. According to this kind of design method, the most important thing is to confirm the radius of the two rings for Lanzhou seismic array on the principle of highest signal coherence combined with lowest noise coherence.By using Welch's averaged, modified periodogram method (P.D. Welch, 1967), the author calculated noise power spectrum of every site to see if the noise level of the array site accord with IMS standard (the noise level at 5 Hz must lower than -135dB), and the result suggest that the noise level accord with the IMS standard. To find out the principle of noise's variation, the root mean square of every site had been calculated and analysised, the result show that several sites had obvious daily variation which is relevant to the TV transmittion tower on the top of Dajianshan. We must analysis the coherence of noise and signal of every site pairs for the comfirmation of the two radius. The cross-correlationcurves method as a traditional method is to find out the distance with the biggest SNR to be the radius. The author tried a new coherence function method combined with the cross-correlation curves traditional method to confirm that the radius of the seismic array are 380m and 1500-1750m respectively. The result suggest that if there are enough distance combined by every site, the radius will be more accurate. The coherence method was mentioned early (Ota Kulhanek, 1973) but no one used it to confirm redius of a seismic array.According to the data processing results (the two radiuses), the author chose nine sites as a elementary array from 18 sites and calculated its array response to find that the main lobe is big and sharp while the side lobe is small, this result suggest that the resolution of this elementary array is very high, so it has the ability to locate earthquakes accurately. This paper also show a rough estimate of the gains of this array, because it needs one year's data to collect one year's noise correlation variation curves and then give a more practical detection ability of the seismic array.One result of this paper is 'seismic array site survery data processing software' exploited by the author. The software can calculate the power spectrum of noise for every site, noise coherence function for every site pair, the cross-correlation curves, the root mean square of noise for every site and the array response.This thes... |
PDF Full Text Request