An integrated method for seismic velocity modeling based on collocated cokriging

Abstract

The modeling results of seismic velocity (VP and VS) largely influence the accuracy of seismic inversion and depth migration. In the process of velocity modeling, seismic attributes and sedimentary facies are very important as covariates, especially sedimentary facies, which affect geological conditions in the whole area. Because of the sparse distribution of well loggings in the early stage of petroleum exploration, integrating multiple variables to control the modeling process becomes essentially critical. The conventional cokriging method and the two-step modeling approach, which do not use sedimentary facies as quantitative data, are not able to introduce enough information to enhance the quality of a velocity model. The proposed collocated cokriging with sedimentary facies (CCK with sedimentary) method directly uses the digital model of sedimentary facies as a covariate to calculate the velocity results. CCK with sedimentary integrates three kinds of data, logging, seismic attributes, and sedimentary facies, where the logging data is the primary variable, and seismic attributes and sedimentary facies are the two covariates. Field examples indicated that the velocity models of the P-wave for CCK with sedimentary is better than that for collocated cokriging (CCK), and the interval velocity has been proven to be a better seismic attribute of the kriging method. The same is true for the VS results. The velocity models obtained by CCK with sedimentary contain less abnormal values with high continuity and stability. Therefore, CCK with sedimentary has been proven to be an effective method to calculate velocity distributions from seismic data, and its applicability for other elastic properties needs further research.