Geostatistical lithofacies modeling of the upper sandstone member/Zubair formation in south Rumaila oil field, Iraq

Geological facies modeling is a crucial problem for reservoir characterization as it affects the reservoir heterogeneities and fluid flow performance prediction. The main purpose of this research is to adopt a stochastic simulation to construct 3D lithofacies models of the tidal/estuarine depositional environment of the upper sandstone member in south Rumaila oil field, located in Iraq. Based on core measurements, the upper sandstone member has three main lithofacies: sand, shaly sand, and shale. Literature review indicates that the formation is encompassed of mainly sandstone with some inter-bedded shale zones. To reconstruct the 3D lithofacies model, the sequential indicator simulation (SISIM) was adopted to build the categorical image, pixel by pixel, considering the nonparametric condition distribution. Specifically, SISIM depends on the variogram to address and model the variation between any two spatial points from the available data. Therefore, 12 different variograms were constructed given the three lithofacies in four different azimuth directions: 0 degrees, 45 degrees, 90 degrees, and 135 degrees. The resulting lithofacies models in the four selected azimuth directions have shown frequent tidal lithofacies channeling and indicate an approximate matching with the original description of the formation depositional environment of the tidal-dominated and sand-rich environment. The generated lithofacies model in 135 degrees direction has sand channels prevailing towards the southeast shoreline of the reservoir. The created lithofacies model also preserves the reservoir complexity and heterogeneity because it was created using a high-resolution gridding system with approximately two million grids. Additionally, the resulting tidal lithofacies model ensures reservoir heterogeneity as the petrophysical properties are then distributed given each lithofacies with distinct indicator variograms.