【摘要】 面波是地震勘探中广泛存在的一种规则干扰波，在资料处理中要去除面波又十分棘手。常规的滤波方法在滤除面波的同时也对与面波处于同一频带的有效信号造成了损害。本文基于在较低频率处面波的能量强于反射波、在小频率范围和小空间范围内面波能量变化缓慢的假设条件下，提出在小波变换频带分解的基础上，先用面波的视速度对面波作线性时移，使面波逐道相干，再利用KL分解或沿x方向进行小波变换的方法来提取面波，并将其从原始资料中减去。经过合成数据和实际资料的处理证明，这种方法可以十分有效地消除面波，同时不损失反射波的低频成分。更多还原

【Abstract】 Surface wave is a regular noise wave existing widely in seismic data, and its elimination is very difficult in data processing. Usual filtering methods eliminate surface wave, but damage useful signals which coexist with surface wave in the samefrequency band. On the basis of the fact that surface wave energy is stronger thanreflection energy in low frequency range, and on the assumption that surface waveenergy varies slowly in small frequency and space ranges, we advance a new methodthat includes the following steps:. On the basis of frequency-band decomposition of wavelet transform, lineartime shift of surface wave is achieved by using its apparent velocity, making surfacewave intertrace-coherent.. Surface wave is abstracted by making KL decomposition or doing wavelettransform in x-axis direction, then is subtracted from the original seismic data.Synthetic and real data processing results show that the method can desirably eliminate surface wave , without damage of low-frequency reflection contents.更多还原