Application of Multivariate Methods to Optimize Development of Thin Oil Zones in a Mature Carbonate Reservoir

Abstract

Exploitation of thin oil zones in a mature field with complex carbonate geology under strong water drive offers many challenges. The primary objective is effective oil recovery from the thin oil zones without excessive water production. The initial development phase targeting thin remaining oil zones in a giant, mature carbonate field in Saudi Arabia has been guided by reservoir simulation results, with performance generally exceeding expectations. However, performance of individual horizontal wells has varied greatly. Multivariate statistical methods have been applied across the gamut of reservoir parameters for these wells to gain further insights into critical success factors and mechanisms. Response variables were established (producing time to reach various watercut thresholds) to gauge well performance. Principal component, factor, and multiple regression analyses were applied to independent reservoir parameters for a population of 20 horizontal wells placed in the target zone. These parameters included zone thickness, standoff from fluid contacts, vertical permeability contrast, thickness of low-permeability interval, reservoir contact, net/gross ratio, completion design, extent of fracturing, zone porosity, proximity to injectors, and trajectory orientation. Multivariate analysis conclusively demonstrated that the principal factor governing well performance in the early period (up to three years) was the vertical permeability contrast or in other words, the extent to which a permeability baffle exists between the thin low-permeability zone and the underlying thick high-permeability zone. Other parameters may contribute to well performance beyond the 30% watercut threshold and will be addressed in a future paper. The findings from this study have been translated into Best Practices for exploiting thin oil zones and have been applied in further developing the thin oil zone in the subject field.