零相位化和非线性叠加对提取气枪震源地震信号波形的影响研究*

(1.中国科学技术大学 地球和空间科学学院地震学与地球内部物理实验室,安徽 合肥 230026; 2.中国地震局滇西地震预报实验场办公室,云南 大理 671000; 3.南方科技大学 地球与空间科学系,广东 深圳 518055)

主动源; 气枪震源; 零相位化处理; 非线性叠加; 信号提取

Effects of Nonlinear Stacking and Zero-phase Processing on Signal Extraction from Airgun Source Waveform
YAO Jiaqi1,CAO Wenzhong1,YE Beng1,2,ZHANG Wei3

(1.Laboratory of Seismology and Physics of Earth's Interior,School of Earth and Space Sciences,University of Science and Technology of China,Hefei 230026,Anhui,China)(2.Western Yunnan Earthquake Prediction Study Area,China Earthquake Administration,Dali 671000,Yunnan,China)(3.Department of Earth and Space Sciences,Southern University of Science and Technology,Shenzhen 518055,Guangdong,China)

active source; airgun seismic source; zero-phase processing nonlinear stacking; signal extraction

备注

使用零相位化处理和非线性叠加方法提取低信噪比记录中的有效信号,通过定量评估结果波形的走时和振幅等信息的精准程度,探究此处理流程是否可提高提取气枪震源有效信号波形的准确性。对合成波形和云南宾川主动源实验中实际资料数据进行零相位化处理,选用相似性加权、时间域相位加权、时频域相位加权、改进的时频域相位加权4种非线性叠加方法提取有效信号。结果 表明,基于上述处理方法可有效提高提取波形的准确性:(1)零相位处理可以提高波形数据的分辨率,使有效信号到时点变为波形峰值点,有利于在非线性叠加后拾取出准确的到时位置;(2)零相位化数据的非线性叠加结果卷积震源子波可基本恢复地震信号波形记录,可降低非线性叠加对有效信息的损失;(3)使用时间域相位加权叠加和相似性加权叠加方法可获得较好结果,但可能会压制低信噪比小振幅信号; 基于时频域的相位加权类叠加方法对有效波形成分影响较大,但对小振幅信号保幅较好。

We investigate whether using zero-phase waveform record can reduce the influence of nonlinear stacking on the effective signals extracted from low signal-noise ratio data by evaluating the fitness of traveltime and waveform quantitatively.Then semblance weighted stacking,time-domain phase weighted stacking,time-frequency domain phase weighted stacking and improved time-frequency domain weighted stacking methods are applied to process the synthetic data and real data of air-gun experiment in Binchuan,Yunnan.The results show that:(1)zero-phase processing can significantly improve the data revolution and the onset time picking precision of seismic phase;(2)nonlinear stacking results using zero-phased data show merely effects on effective signals,and the raw records can be recovered by convolution of the stacking results with source wavelet; and(3)time-domain phase weighted stacking or semblance weighted stacking method provide better recovery of final stacking results but probably affect the signals with low signal-noise-ratio,while the results from the phased weighted stacking method based on time-frequency transform method show very large influences on effective signals but give good recovery of signals with low amplitude.