Stochastic Inversion of Post-Stacked Seismic Data of Penobscot, New Scotland

Abstract

Seismic characterization plays a fundamental role in the description of subsurface properties from acquired seismic information. The process that converts seismic information into elastic properties of the reservoir under study is known as seismic inversion. Many inversion methods are based on the deterministic method, whose resolution is affected by the limitation frequency content of the seismic. For this reason, it is proposed the implementation of stochastic inversion of post-stacked seismic data from conventional and fractal initial models that are related by their fractal and statistical properties, allowing the obtaining of deterministic inversions slightly altered from each other, all with a good fit to the initial seismic data and thus obtaining a higher frequency content in the impedance sections. The statistical information of the final inversions obtained was calculated through a routine in Matlab. The results obtained were verified with those derived from other attributes to identify relevant structural and/or stratigraphic features that would allow the characterization of the Mississauga Formation of the Penobscot Field, located in Nova Scotia, Canada. The stochastic process developed allowed the delimitation and identification of the tops of the target sands, in addition to providing probability percentages that confirm sands 2 and 4 as the packages with the best conditions for the storage of hydrocarbons.