Pseudo-3D Depth Reconstruction from Multi-Vintage 2D and 3D Seismic Data in The Offshore Southeast Sumatra

Abstract

The availability of legacy 2D seismic data in many mature basins offers an opportunity to generate 3D-like subsurface images without the need for expensive 3D seismic acquisition. This study proposes and demonstrates a workflow to construct a pseudo-3D seismic volume from multi-vintage 2D and 3D seismic data in Offshore Southeast Sumatra. The objective is to improve subsurface imaging and depth-structure mapping while minimizing exploration cost and time. Major challenges include variations in amplitude, frequency, phase, and time alignment across datasets acquired with different parameters, as well as determining the optimum line spacing required for reliable 3D reconstruction. The workflow begins with 2D–3D matching and amplitude balancing, followed by spatial interpolation using the 3D Sparse Radon Interpolation technique, which reconstructs coherent reflectors between 2D lines. The generated pseudo-3D volume is then merged with available 3D seismic data to create a unified volume for interpretation. Subsequent time-to-depth conversion is performed using interval velocity data, and the velocity model is iteratively updated using time-preserving tomography based on depth mis-ties between seismic horizons and well markers until convergence is achieved. The resulting depth-structure map provides improved consistency with well control and reveals subtle structural features that were not observed in the individual 2D datasets. The proposed approach demonstrates that pseudo-3D reconstruction from multi-vintage 2D seismic data can serve as a practical and cost-efficient alternative for early-stage exploration and pre-survey evaluation, especially in offshore areas where 3D seismic acquisition is economically constrained.