Faulting and Fluid Flow in Carbonate Reservoirs An example from the OSUAER Field, UAE

This thesis investigates the impact of low permeability faults on fluid flow in carbonate reservoirs. Special attention is given to the situation where such faults are cut by open fractures (partially sealing faults). Microstructural and petrophysical measurements of fault rocks from outcrops in the UAE were analysed, suggesting the damage zone (DZ) to faults could act as conduits, whereas the fault core (FC) is more likely to act as a baffle. If these fault rocks were present within the OSUAER (OS), the permeability contrast between fault and host rock would suggest that they could act as baffles in the DZ or seals in the FC. In the outcrops, the FC is cut by open fractures and varies in fault rock thickness (FRT) over a short distance. If open fractures are present in the subsurface they would impact fluid flow. Simple simulation models suggest that it is important to accurately account for fault- rock thickness and permeability in reservoirs with low permeability faults that are not cut by open fractures. However, fault thickness and permeability have less significance to fluid flow in a reservoir with a partially sealing fault. The difference in oil production and similarities in bottom-hole pressure means that it is difficult to distinguish between a reservoir with a partial fault sealing and a fractured reservoir without a sealing fault. Experiments were conducted to assess how a partly sealing fault would impact oil production from a reservoir with open fractures. Samples with fractures but no sealing fault produced 86% of the oil compared to 44% in samples with partially sealing faults. Numerical modelling was used to investigate the effects of fluid flow in the samples during the core flooding experiments. Simulations showed varied oil production across regions but pressure behaviour was similar throughout the sample due to the presence of open fractures. In addition, faults act as pressure barriers and reduce oil production, so pressure equilibration in a fractured reservoir is not conclusive evidence that partially sealing faults are not present. Moreover, simulation models were able to reproduce experimental results, giving confidence in modelling an actual reservoir with open fractures and partially sealed faults. A case study was conducted to improve the historical matching of oil production from the OS field by applying realistic fault transmissibility multipliers (TMs). These studies suggest that faults are discontinuous and potentially cut by open fractures. The findings also indicated that faults act as conduits. The best history match was with TMs of 5 with a discrete fracture network. This model predicted 13% more oil production than the basic case without faults. This research shows that two types of open fractures can interact with low-permeability faults. First, less fractured faults (partially sealing faults) cannot act as pressure barriers but may reduce oil production; these are difficult to identify based on production data and are not usually documented in carbonate reservoirs. Second, fractures along faults or in their DZ may act as fluid conduits. This type of fault appears to exist in the OS field, but at the same time, it is not possible to generalise to all faults in the field.