Geostatistical Modeling with Seismic Characterization of Wara/Burgan sands, Minagish Field, West Kuwait

Abstract

The Minagish field covers an area of 90 km2, located in south-western part of Kuwait in onshore position. The studied Cretaceous Wara and Upper Burgan reservoirs, deposited in fluvio deltaic environment (clastic rocks), consist of vertically stacked sands with extensive lateral facies variation. Lower Burgan sands are more significant and blocky in nature with little variations in their properties. Reservoirs geometry, their heterogeneities and structural setting are the key issues for the development of the reservoirs. The geostatistical methodology used to simulate a high-resolution geological model representative of the reservoir heterogeneity will be described in this paper. The paper will particularly discuss the techniques used for the integration of seismic attributes to constrain the facies modeling, as well as a nested simulation workflow for a realistic representation of heterogeneities. The three dimensional structural grid was classically based on the seismic interpretation (faults and horizons) and the well correlations. Facies simulation was performed using PluriGaussian functions approach in a non-stationary frame based on vertical proportion curves (VPC) matrix. The distribution has been done separately in three zones: Wara, Upper Burgan and Lower Burgan. For Wara and Upper Burgan zones, nested simulations were used. The lithology (sand or shale) was simulated in a first step, under the constraint of proportion facies maps extracted from seismic data. In a second step, the depositional environment facies were simulated in the sand lithology. These simulations were based on two Gaussian functions to better integrate the various orientations of the different deposits, and their complex spatial relationship. For the Lower Burgan zone, because of the very low variability of the sand proportion, a stationary facies simulation was run, using a truncated Gaussian algorithm. Eventually, the paper underlines the capacity of the PluriGaussian Simulation approach to realistically mimic sedimentary bodies, and to easily incorporate seismic derived information.

Truncation map estimation based on bivariate probabilities and validation for the truncated plurigaussian model

“…Carrillat et al (2010) compared the use of TPG on a giant carbonate oilfield with application of other methods. Albertão et al (2005) and Al-Anezi et al (2013) used TPG to describe the distribution of facies for complex reservoir models while using varying category proportions computed from seismic data.…”