Sand Production Tendency in the World's Largest Clastic Oil Reservoir - An Evolving Experience

The mature Greater Burgan field has the largest clastic oil reservoir in the world producing from multiple units for more than 70 years. Wara and Burgan are the two main sandstone reservoirs in the field with several sub-units of varying mineralogical and mechanical properties. Continuous oil production from some low strength reservoir units has resulted in pressure depletion and the associated water encroachment has led to the initiation of sand production in few wells. This paper presents an approach to analyze rock mechanics, reservoir and production data to predict the sanding tendency in a heterogeneous rock of the world's largest clastic oil reservoir. Analytical poro-elastic geomechanical sanding evaluation approach was used to study rock failure and sand production. A number of wells (both with sand production and without sand production) were selected for geomechanical analysis based on the data availability and their respective location across the field. Various types of rock mechanical tests were performed with stringent quality control criteria to determine the mechanical characteristics of the rocks. Field calibrated geomechanical model along with reservoir and production data were used to build a calibrated sanding model for the Wara reservoir. The sanding model was then utilized to create sanding evaluation logs; perforation optimization and safe operating envelop plots for existing and future wells. As per the calibrated geomechanical model, the reservoir units comprise intervals of variable rock strengths. The sanding model calibrated from offset wells suggested that drawdown and rock strength are the most sensitive parameters in failing the rock and sand production. The wells with and without sanding had mixed response to water breakthrough and there does not seem to be any obvious pattern of sanding observed with respect to the onset of water production in the wells. It was also noticed that few offset wells did not exhibit sand production despite strength and stress conditions in these wells were favorable for early sand production. Potential explanation was found for those cases. These models helped to identify selective intervals to optimize sand-free production with limited drawdowns. However, for wells with very high drawdowns (installed with artificial lift pumps and 1000 psi drawdown pressure), none of the prolific sand interval could be stable and downhole sand control measures may be required. The combination of measured and modelled parameters have helped to understand the sanding tendency and behaviors of a highly heterogeneous reservoir. This analysis has produced guidelines on the best and worst well trajectories and optimum perforation orientations to mitigate well-life sanding risks with a rock strength sensitivity. Furthermore, data priorities have been identified in order to come up with holistic sand management strategy.