First-arrival tomography for building velocity model by jointly using seismic lines and sparse nodes

The exploration of the deep structure of petroliferous basins has gradually become a research hot spot in recent years. A key issue for seismic prospecting is how to improve the accuracy of velocity model building in deep regions. Since the conventional seismic lines have limited offset, and cannot record the far-offset first-arrival data that contain the deep structural information. As a result, the velocity model building methods using the first arrivals of seismic line data can hardly construct the velocity model of deep structures. The data recorded by seismic nodes possess wide-angle and wide-azimuth features and have been widely used in the study of deep structures in the field of global/regional seismology. However, the dense deployment of seismic nodes will significantly increase the required number of seismic nodes and field working load. By contrast, the sparse deployment of seismic nodes can hardly preserve the velocity building accuracy at a shallow depth, resulting in inaccurate velocity model building in deep regions. To cope with this problem, we propose a new frrst-arrival tomographic velocity-model-building method by jointly using seismic lines and sparse seismic nodes: We form a new observation system by combining the conventional seismic lines and seismic nodes, and seismic tomography is used to model oil field deep velocity by combining the first-arrival time of seismic nodes and conventional seismic lines. Both numerical experiments and practical application results show that the combination of seismic lines and sparse seismic nodes can form good data coverage in both shallow and deep layers, where frrst-arrival tomography has high modeling accuracy. In the case of large node spacing, the combined method of conventional seismic line and node only adds a little more work than conventional line acquisition, but its tomographic velocity modeling still has a high modeling accuracy, which is similar to the high-cost high-density node data modeling results. The method proposed in this paper can be easily applied to the regions that are still using the conventional seismic lines and can achieve high accuracy in the deep-region velocity model building with little increase in workload. Therefore, the proposed method is of great significance for the exploration of the deep regions in petroliferous basins.