Post-stack Deterministic Seismic Inversion using Sparse Optimisation Utilizing Gaussian Dense Layer Simulation

Abstract

The main objective in inverting seismic data is to estimate layer characteristics instead of layer boundaries. It is about getting as much as layer boundaries as seismically permitted with correct position and amplitude. However, the number of layers identified by seismic inversion methods is relatively in small number. Meaning, inversion methods tend to thicken the layers due to band-limited of seismic data and wavelet inaccuracy. It is well-accepted that sparse-layer inversion has the ability in recovering full RC series that are permittable seismically yet it tends to produce blocky impedance. Sparse-layer inversion is also computationally extensive because of large amount of competing relectivity model. In this paper, we test sparse optimisation inversion performance using gaussian dense-layer simulation instead of sparse-layer model.