Geostatistical Modeling of the Third Upper and Third Lower Sands of the Burgan Formation, Greater Burgan Field, Kuwait
By
W. Scott Meddaugh1, Jamal Al-Hamoud2, R. H. Kirby3
(1) ChevronTexaco Exploration and Production Technology Co, Houston, TX (2) Kuwait Oil Company, Kuwait, Kuwait (3) ChevronTexaco Overseas Petroleum Co, San Ramon, CA
A geostatistical modeling study of the Burgan Third Upper (3U) and Third
Lower (3L) sands was used to examine reservoir geology uncertainty issues prior
to a finite difference flow simulation study. Separate models for the 3U and 3L
units were generated for two representative portions of the Greater Burgan
field. A variety of approaches including Gaussian Sequential Simulation (GSS),
Sequential Indicator Simulation (SIS), Collocated Cokriging with GSS, and
Boolean Modeling were evaluated. The following final workflow was used to model
the 3U and 3L units: (1) Multi-binary
SIS to distribute two facies (“sand” and
“shale”); (2) GSS to distribute effective porosity within each facies using
appropriate facies and geological interval-specific data; and, (3) facies-dependent
transform to add permeability to the reservoir models. The “sand” facies
represents higher quality reservoir rock (shale volume less than 40%, effective
porosity greater than 10% and permeability greater than 100 md). Sand volume
ranges from zero at the top of the 3U and 3L to over 50% at the bottom of the
units.
Multiple realizations of the data-driven models and a set of models for which the semivariogram parameters were altered to generate models with varying connectivity were evaluated using a 3D streamline-based flow simulator to establish the appropriate cases for finite-difference fluid-flow simulation. Water breakthrough times were 1300 +/- 130 days for the data-driven base case, 1100 +/- 37 days for the enhanced connectivity case, and 1450 +/- 80 days for the lower connectivity case.