Three-component seismic exploration through P-wave source and three-component geophone is an effective technique used in complicated reservoir exploration. In three-component seismic exploration data processing, one of the difficulties is static correction of converted wave. This paper analyzes propagation characteristics of non-converted and converted refracted waves, and discovers a favorable condition for the formation of converted refracted wave, i.e. the velocity of overlaying medium S wave is much lower than that of underlying medium S wave. In addition, the paper proposes the static correction method of converted wave based on PPS converted refracted wave, and processes the real three-component seismic data with better results of static correction of converted wave.
PPS converted refracted wave, converted wave, static correctionThe recent seismic exploration technology of P-wave source generating seismic wave and three-component geophone receiving P wave (P-P wave) and converted wave (P-S wave) is the most efficient method in complicated reservoir prospecting [1][2][3][4] . Considering the low and rapidly changed velocity of near-surface S wave, how to eliminate the influence of near-surface S wave on reflected S-wave propagation time (S-wave static correction) is especially important [5] .There have been three static correction methods of converted wave: (a) the calculation of S-wave static correction value from inversion of S-wave velocity based on Love wave dispersion [6] , (b) the inversion of near-surface S-wave velocity by the non-converted refracted S wave [7] , and (c) the direct derivation of the maximum stack power of S-wave static correction [5] . Since the widely used P-wave seismic source in seismic survey makes it difficult to observe Love wave and nonconverted refracted S wave, the application of the first two methods has limitations. The third method cannot fundamentally solve the problem of long wave-length S-wave static correction for its approximation to residual static correction.In 2002, Ritzwoller and Levshin [8] proposed the use of surface wave dispersion and non-converted refracted P and S waves in the inversion of surface S-wave velocity. Li [9] suggested the adoption of first break travel-time difference of Z and X component seismic record to calculate S-wave static correction value. Heelan [10] studied how to calculate amplitude intensity of refracted wave. Hall [11] explained non-converted and converted refracted waves, and briefly described the propagation kinematics characters. Most studies in recent decades on refracted wave have focused on non-converted refracted wave (PPP and SSS refracted wave) [12][13][14][15][16][17] , with little work on