Multicomponent seismic exploration opens a new world for improving seismic imaging and for extracting valuable additional information about subsurface physical characteristics. This paper anatomizes the multicomponent amplitude responses, mainly focusing on AVO responses, and amplitude relationships among P-SV converted-waves, P-waves and S-waves. First, the relationship between different rock physical parameters and P-SV wave AVO response is discussed extensively. Then, an improved approximation of P-SV wave stacked amplitude is derived and can be used for P-SV wave synthetics and S-impedance inversion from P-SV converted wave. Finally, an optimum multicomponent event registration method is suggested by quantitatively analyzing the AVO attribute relationship between P-SV wave and P-wave. In addition, a novel method of amplitude-ratio extraction from multicomponent seismics for consideration in the probability distributions of lithology and fluid is also proposed.
Based on the re-expression of the three-term AVO equation for the PP wave reflection coefficient given by Aki and Richards (1980), two improved AVO formulas involving elastic parameters and their applications are presented in this paper. Unlike traditional AVO approximations, the new expressions can be used not only to extract new attributes which are sensitive to lithology and fluids, but also to invert the elastic-parameter reflectivity volumes, such as the Lamé constant, the shear modulus, as well as bulk density simultaneously. Furthermore, multi-attributes AVO analysis and integrated oil/gas prediction is favored because there is no loss in accuracy for the new equations. The results of the rock physics analysis and applications to real data show that the new method is more stable and less ambiguous than the conventional methods, which are currently used to discriminate reservoir sands from other lithologies and between low- and full-saturation of hydrocarbons. In addition, the inversion results on synthetic data indicate that the modified approximations can provide more reliable results in layered parameters, like Pimpedance, S- impedance, and LMR (Lambda-Mu-Rho).