基于局部斜率属性的VSP波场分离研究

基于垂直地震剖面(VSP)资料中上、下行波视速度的差异，利用地震剖面同相轴局部斜率属性参数，提出了一种分离上、下行波场的新方法.首先利用Fourier变换初步分离上、下行波场，然后利用平面波分解滤波器(Plane Wave Destruction (PWD) Filter)技术估计初始分离波场的同相轴局部斜率属性参数，在此基础上对VSP原始资料波场分离.该方法是一种时间域最小平方优化分离波场的方法，不存在其他滤波方法阈值滤波器边界的影响，减少了因镶边问题带来的假象.模拟和实际资料处理结果表明，该方法与传统方法相比，分离出的上、下行波噪声假象少，振幅保持好，更好地消除了上、下行波的相互影响.

VSP wavefield separation using local slopes attribute

Based on difference between upgoing and downgoing wavefields in vertical seismic profiling (VSP) data, a new method to separate upgoing and downgoing wavefields using local event slopes attribute is developed. In this method, we first use Fourier Transform to obtain initially separated wavefields, and then use plane wave destruction technology to estimate local event slopes of initially separated wavefields. Finally, original VSP data is separated on the basis of local event slopes. Our proposed method, which is a kind of optimization approach in time domain, doesn’t have taper effects that result from window filter band and attenuates artifacts. Results of synthetic and field data examples show that, compared with conventional method, separated wavefields using local slopes are clean and amplitudes are maintained. The interference between upgoing and downgoing wavefields can be eliminated better.