| 数据域特征波反射走时反演 |
| |
| 引用本文： | 冯波,王华忠,冯伟.数据域特征波反射走时反演[J].地球物理学报,2019,62(4):1471-1479. |
| |
| 作者姓名： | 冯波 王华忠 冯伟 |
| |
| 作者单位： | 1. 波现象与反演成像研究组(WPI), 同济大学海洋与地球科学学院, 上海 200092;2. 资源建设部, 中山大学图书馆, 广州 510275 |
| |
| 基金项目： | 国家自然科学基金（41574098，41704111，41774126），国家重大专项（2016ZX05024-001，2016ZX05006-002）资助. |
| |
| 摘 要： | 地震波的运动学信息(走时、斜率等)通常用于宏观速度建模.针对走时反演方法,一个基本问题是走时拾取或反射时差的估计.对于成像域反演方法,可以通过成像道集的剩余深度差近似计算反射波时差.在数据域中,反射地震观测数据是有限频带信号,如果不能准确地确定子波的起跳时间,难以精确地确定反射波的到达时间.另一方面,如果缺乏关于模型的先验信息,则很难精确测量自地下同一个反射界面的观测数据同相轴和模拟数据同相轴之间的时差.针对走时定义及时差测量问题,首先从叠前地震数据的稀疏表达出发,利用特征波场分解方法,提取反射子波并估计局部平面波的入射和出射射线参数.进一步,为了实现自动和稳定的走时拾取,用震相的包络极值对应的时间定义反射波的到达时,实现了立体数据中间的自动生成.理论上讲,利用包络极值定义的走时大于真实的反射波走时,除非观测信号具有无限带宽(即delta脉冲).然而,走时反演的目的是估计中-大尺度的背景速度结构,因此走时误差导致的速度误差仍然在可以接受的误差范围内.利用局部化传播算子及特征波聚焦成像条件将特征波数据直接投影到地下虚拟反射点,提出了一种新的反射时差估计方法.既避免了周期跳跃现象以及串层等可能性,又消除了振幅因素对时差测量的影响.最后,在上述工作基础之上,提出了一种基于特征波场分解的新型全自动反射走时反演方法(CWRTI).通过对泛函梯度的线性化近似,并用全变差正则化方法提取梯度的低波数部分,实现了背景速度迭代反演.在理论上,无需长偏移距观测数据或低频信息、对初始模型依赖性低且计算效率高,可以为后续的全波形反演提供可靠的初始速度模型.理论和实际资料的测试结果证明了本文方法的有效性.
|
| 关 键 词： | 特征波场分解 立体数据空间 反射时差测量 走时反演 反射波层析 |
| 收稿时间： | 2017-07-27 |
Characteristic wavefield decomposition based reflection traveltime inversion |
| |
| FENG Bo,WANG HuaZhong,FENG Wei.Characteristic wavefield decomposition based reflection traveltime inversion[J].Chinese Journal of Geophysics,2019,62(4):1471-1479. |
| |
| Authors: | FENG Bo WANG HuaZhong FENG Wei |
| |
| Institution: | 1. Wave Phenomena and Inversion Imaging Research Group(WPI), School of Ocean and Earth Science, Tongji University, Shanghai 200092, China;2. Collection Development Department, Sun Yat-Sen University Library, Guangzhou 510275, China |
| |
| Abstract: | In exploration seismology, macro-velocity models are often estimated using kinematic information of seismic data or migrated gathers. Traveltime inversion can be implemented in both image and data domain. Typical image domain inversion methods, such as the Ray-based tomographic migration velocity analysis or the wave-equation migration velocity analysis, require calculation of common image gathers, which is quite computationally time consuming. For data domain traveltime inversion methods, it is difficult to measure the reflection traveltime difference between observed and synthetic data, especially in the absence of prior information of subsurface structure. In this paper, we start from the sparse representation of prestack seismic data, using the characteristic wavefield decomposition (CWD) to separate reflection phases and estimate ray-parameters. Meanwhile, traveltime of each reflection phase is defined by its envelope maximum. Therefore, the stereo data space can be estimated directly from CWD data. Based on the characteristic wavefield focusing imaging conditions, a novel method for reflection traveltime residual estimation is proposed, which can eliminate the influence on traveltime residual caused by unreliable amplitude information. Finally, a method of characteristic wavefield reflection traveltime inversion (CWRTI) is developed. It neither requires low-frequency seismic data and/or long-offset acquisition nor a good initial model. In addition, CWRTI is computationally efficient, for it eliminates the need for calculating and saving the common imaging gathers and traveltime picking. Furthermore, it is a fully automatic procedure, making it quite promising for establishment of an automatic macro-velocity model. Tests on synthetic and real data demonstrate the effectiveness of CWRTI. |
| |
| Keywords: | Characteristic wavefield decomposition Stereo data space Reflection traveltime residual estimation Traveltime inversion Reflection tomography |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《地球物理学报》浏览原始摘要信息 |
| 点击此处可从《地球物理学报》下载免费的PDF全文 |
|
