Optimization of Reservoir Simulation and Petrophysical Characterization in 4D Seismic

Abstract 4D seismic has become a widely accepted technique to interpret changes between successive 3D seismic surveys in terms of fluid substitution and pressure depletion in a producing reservoir. However, most time-lapse studies have been mostly qualitative and based on simplified reservoir representation in order to adjust to the time constraints of today's oil market. In this paper, we present how reservoir simulation constrained by stochastic characterization and non-linear optimization can be used in an integrated series of tools to refine 4D interpretation. Because of its direct relationship with pore fluid content and properties, we use seismic impedance rather than seismic amplitude as the primary data between the various steps of our 4D interpretation loop. Non-linear inversion of the 3D seismic data sets allows a preliminary interpretation. Stochastic simulation of the lithology and porosity, constrained by these "observed" impedance volumes and by well logs, provide the static reservoir characterization for the reservoir simulator. Once simulated production matches the recorded production history, empirical or Biot/Gassmann-type petrophysical models are used to calculate the "simulated" impedance volume from the fluid saturation and pressure distribution calculated by the reservoir simulator. Non-linear optimization is used iteratively to improve first the production history match and next the agreement between "simulated" and "observed" impedance volumes over time. This optimization is performed over a limited set of poorly constrained parameters in the permeability calculation and petrophysical models. In the case study presented here, the analysis of a turbidite reservoir in the South Timbalier 295 field, our results show how stochastic characterization helps reproducing the complexity of reservoir fluid dynamics while the results of the optimization underlines the robustness of the 4D interpretation despite the large amount of unknowns. Introduction Time-lapse, or 4-D, seismic monitoring is an integrated reservoir exploitation technique based on the analysis of successive 3D seismic surveys. Differences over time in seismic attributes are directly related to changes in pore fluids and pore pressure during the drainage of a reservoir under production. The detection of areas with significant changes or with unaltered hydrocarbon-indicative attributes, can be used to determine drilling targets where hydrocarbons remain trapped after several years of production. Making sure that seismic differences are related to fluid flows is critical for a complete time-lapse seismic study. Noise associated with differences in acquisition can generate seismic differences between surveys that are not related to the reservoir drainage pattern. In this paper, we describe how reservoir simulation can be used to generate independent impedance maps to validate or constrain 4-D impedance maps obtained from the inversion of successive legacy data sets. A complete 4D analysis is an iterative loop where the original interpretation can be refined along the later steps. First, we summarize the steps preceding the reservoir simulation, including the 3D seismic processing and inversion, and the preliminary time-lapse interpretation. We then describe the elastic models, the properties of reservoir fluids and the reservoir characterization that can be used to link the impedance volumes to fluid and lithology distributions in the reservoir.