地球物理学进展 ›› 2017, Vol. 32 ›› Issue (4): 1784-1790.doi: 10.6038/pg20170450

• 应用地球物理学Ⅱ • 上一篇    下一篇

基于jittered采样的浅层三维地震数据处理及应用

李鹏, 张建清, 尹剑   

  1. 长江勘测规划设计研究院 长江地球物理探测(武汉)有限公司, 武汉 430010
  • 收稿日期:2016-10-23 修回日期:2017-05-27 出版日期:2017-08-20 发布日期:2017-08-20
  • 作者简介:李鹏,男,1983年生,湖北人,博士,研究方向为地震资料处理和解释.(E-mail:lipeng2@cjwsjy.com.cn)
  • 基金资助:

    水利部“948”课题(201409)、中国地质大学(武汉)地球内部多尺度成像湖北省重点实验室2016年度开放基金(SMIL-2016-04)和2016年湖北省博士后创新岗位“水库大坝渗漏精细探测方法技术研究”联合资助.

Shallow seismic data processing and application based on jittered sampling

LI Peng, ZHANG Jian-qing, YIN Jian   

  1. Ch angJiang Institute Of Survey, Planning, Design and Research, ChangJiang Geophysical Exploration & Testing CO., LTD(WUHAN), Wuhan 430010, China
  • Received:2016-10-23 Revised:2017-05-27 Online:2017-08-20 Published:2017-08-20

摘要:

目前,浅层三维地震勘探在国内外尚处于起步阶段,其勘探深度在100 m以内,勘探目标速度低,地震资料主频较高,根据采样定理,允许的最大道间距较中深层地震勘探要小得多,因此浅层三维地震勘探的观测系统设计和数据处理方法还有许多亟待解决的问题,本文将信号处理领域的压缩感知理论应用到浅层三维地震的观测系统设计和数据处理方法中,研究具有重要的实际意义.理论研究表明:随机采样数据比有规律的欠采样数据能更好的进行波场重建,随机欠采样数据可以将相干混叠转化为不相干噪声,从而将地震数据解释问题转变成了数据去噪问题,同时采用jittered随机采样方式可以避免完全随机采样的数据存在的数据点空间位置过于集中或过于稀疏的情况.本文采用模型分析和实际数据研究的方法,结果显示:在野外进行浅层三维地震数据采集时,如采用jittered随机欠采样方式进行检波器的布置,可以实现以较少的检波器布置较大范围的观测系统,从而大大提高了野外数据采集工作的效率,同时,这种随机的布线可以较灵活地适应野外的工作环境,在采用规则布线时,尽量希望测区足够平坦、开阔,无大尺度障碍物,而随机的布线可以适应有障碍物的地形地貌,在遇到障碍物或不能跨越的地形时,可适当增大道间距,在开阔位置减小道间距进行数据补偿.得到的结论是:这种浅层三维地震道间距无法满足奈奎斯特采样条件的情况,通过检波点的随机分布所采集的地震数据可以有效降低地震资料的混叠效应.

Abstract:

At present, shallow 3-D seismic exploration is still in its infancy, its exploration depth is less than 100 meters, the exploration target velocity is low and the frequency of seismic data is high. According to the sampling theorem, the permissible maximum trace interval is smaller than the mid-deep exploration. Therefore, there are still many problems to be solved in the geometry design and data processing of the shallow three-dimensional seismic exploration. In this paper, the theory of compressed sensing in signal processing is applied to shallow seismic geometry design and data processing. Theoretical research shows that random sampling data can better reconstruct the wavefield than undersampled data, The sampled data can transform the coherent aliasing to non-coherent noise, which transforms the seismic data interpretation problem into the data denoising problem. At the same time, the jittered random sampling method can avoid the situation that the spatial data points of the completely random sampled data are too concentrated or too sparse. Through the model and the actual data research, the results show that:in the field of shallow three-dimensional seismic data acquisition, such as the use of jittered random undersampling method for the detector layout can be achieved with fewer detectors layout of a wide range of geometry, which greatly improves the efficiency of the field work, at the same time, this random layout can be more flexible to adapt to the field environment. In the use of regular layout, as far as possible hope that the survey area is flat enough, open, no large-scale obstacles, and random layout can be adapted to the topography of obstacles. In the case of obstacles, it is appropriate to increase the trace interval, in the open position to reduce the trace interval for data compensation. This kind of shallow 3-D seismic trace interval can not satisfy the Nyquist sampling condition, and the random distribution of seismic data can effectively reduce the aliasing effect of seismic data.

中图分类号: