Ensemble-based conditioning of reservoir models to seismic data

Abstract

While 3D seismic has been the basis for geological model building for a long time, time-lapse seismic has primarily been used in a qualitative manner to assist in monitoring reservoir behavior. With the growing acceptance of assisted history matching methods has come an equally rising interest in incorporating 3D or time-lapse seismic data into the history matching process in a more quantitative manner. The common approach in recent studies has been to invert the seismic data to elastic or to dynamic reservoir properties, typically acoustic impedance or saturation changes. Here we consider the use of both 3D and time-lapse seismic amplitude data based on a forward modeling approach that does not require any inversion in the traditional sense. Advantages of such an approach may be better estimation and treatment of model and measurement errors, the combination of two inversion steps into one by removing the explicit inversion to state space variables, and more consistent dependence on the validity of assumptions underlying the inversion process. In this paper, we introduce this approach with the use of an assisted history matching method in mind. Two ensemble-based methods, the ensemble Kalman filter and the ensemble randomized maximum likelihood method, are used to investigate issues arising from the use of seismic amplitude data, and possible solutions are presented. Experiments with a 3D synthetic reservoir model show that additional information on the distribution of reservoir fluids, and on rock properties such as porosity and permeability, can be extracted from the seismic data. The role for localization and iterative methods are discussed in detail.

Synergizing Crosswell Seismic and Electromagnetic Techniques for Enhancing Reservoir Characterization

“…Therefore, a full-wave approach is needed (Choy et al 1980). In a more recent article, Leeuwenburgh et al (2011) history matched 4D seismic for the tracking of the fluid fronts of a 3D synthetic example, showing consistent multimodel history matching by means of the combined usage of production and seismic data. The seismic information was incorporated into the model by first inverting it for saturation levels and then using these attribute data for constraining the reservoir saturation with the EnKF.…”

“…The 4D seismic gained prominence in monitoring the changes within a reservoir, with many studies using them for reservoir history-matching purposes (Skjervheim et al 2007; Sedighi-Dehkordi and Stephen 2010; Leeuwenburgh et al 2011;Kazemi et al 2011;Trani et al 2013). Skjervheim et al (2007) successfully incorporated inverted time-lapse seismic data and production data into the EnKF for reservoir model updating, improving the permeability estimate for synthetic and real field cases of the North Sea.…”

“…The method relies, however, on the prior interpretation of the seismic data that may differ from the actual reservoir state. Leeuwenburgh et al (2011) history matched 4D seismic data by directly integrating seismic amplitude data into the process, avoiding the need for inversion and demonstrating improvements in reservoir forecasting (Gibson and Levander 1988;Yilmaz 2001). Even though convolutional methods are the standard techniques for surface seismic imaging, their ray-theory assumption is not always adequate for crosswell imaging (Sheriff and Geldart 1995;Gao et al 2012).…”

A Sparse Bayesian Imaging Technique for Efficient Recovery of Reservoir Channels With Time-Lapse Seismic Measurements

“…The EnKF is a popular Monte Carlo variant of the standard Kalman filter (KF) and has been successfully applied in a variety of history matching problems [8,39,36,40]. As the KF, the EnKF estimation process is based on forecast and analysis cycles but differs in the way the first two moments of the system state are represented by an ensemble of state vectors, approximating the KF estimate and associated error covariance matrix by the ensemble sample mean and covariance.…”

A Time Domain Update Method for Reservoir History Matching of Electromagnetic Data

“…[29], and represents the distribution of the system state via a collection of state vectors (ensemble) that approximates the covariance matrix of the system by a sample covariance matrix computed from the ensemble. Despite the fact that the EnKF updates are based on means and covariances only (i.e., second order statistics neglecting higher order moments of the joint probability density distribution of the model variables) and are computed from a finite size ensemble, impacting the quality of the history matching estimates, the EnKF has shown to work remarkably well in practice [14], [30]. Achieving good matching for a variety of different problems, the EnKF has naturally become one of the methods of choice for reservoir history matching.…”