TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractA Geological Model was built and an Uncertainty Assessment approach was used to better understand the reservoir behaviour. Conceptual models were used to constrain the number of realizations, all of which are equi-probable solutions that honour both hard and soft data.Ranking of models was done by progressively analyzing the cases/ scenarios considered. Initially, inconsistent scenarios/ cases were discarded. In other words, those that did not comply with the data distribution expected and/or the spatial distribution and consistency with the conceptual geological models are discarded. The weight of each criterion in the decision that leads to abandon a particular case or scenario varies according to the variable modeled.The final phase involved ranking the different models using body connectivity and volumetrics. The ranking was divided in sets where one variable was considered and others added progressively until all the variables were taken into account, including structure and fluid contacts uncertainty. The selected scenarios were then dynamically evaluated for well and development placement.

Carbonate reservoirs are notoriously complex and difficult to characterize, due to their inherit homogeneity. The ability to understand and model such homogeneity accurately, leads to better reservoir management and improved field development strategy in terms of well placement and optimized well patterns. Dynamic reservoir model was built from the static geological model i.e. using MDT, pressure volume temperature (PVT), routine core analysis (RCA) and special core analysis (SCAL) etc. As part of the reservoir model validation process, history matching of the model was conducted to match the observed data. After history matching was completed, tracers injection analysis and streamlines modelling was conducted in other validate the reservoir model, the well patterns and improved the full field development strategy. The full field development scheme includes water alternating gas (WAG) miscible hydrocarbon gas injection, with 5 observers covering both the flank and crestal locations. The tracer analysis included the injection of 11 different tracers in 11 different injectors, and the monthly monitoring of the producers within the well pattern. Streamline model was built simultaneously from the convention compositional model, in order to analysis and predict the different tracers break through times i.e. both observed from the field and simulated. In addition to this, streamline time of flight (TOF) analysis, the effect of tracking of different tracer components, geological and geophysical impact was evaluated, in order to improve the breakthrough time match between observed and simulated time. As a result of this analysis reservoir management improved, as the source of increasing higher GOR in specific wells were discovered. Remedial actions were recommended to help reduced the increasing high GOR in the respective wells. Also the field development strategy improved, as injector's contribution per well pattern well was quantified. As a result injection could be redistributed. In producer wells with little or no support from their respective injectors, a plan was made to ensure that such patterns could be close appropriately. This improved and maintains the voidage replacement ratio (VRR) in the field, according to reservoir guidelines. This paper describes how by using gas tracer injection and streamline modelling reservoir management and field development can be improved. In addition to the improvements in reservoir management and field development strategy, several lessons learnt and best practice were suggested from the tracer and streamline study conducted. They include but are not limited to; Different tracers can show different concentration level at breakthrough wells.Wells further away from injectors can show earlier breakthrough than well close to injectors, as it is dependent on reservoir connectivity.In analyzing model breakthrough times, the measurable tracer in the field needs to be used, as opposed to the actually first breakthrough seen in the simulator. As below such concentration cannot be measured.Operational changes in wells need to be captured properly in the simulator, in order ensure improve prediction of tracer breakthrough times.

Carbonate reservoirs are notoriously complex and difficult to characterise, due to their inherit homogeneity. The ability to understand and model such homogeneity accurately, leads to better reservoir management and improved field development strategy in terms of well placement and optimised well patterns. Dynamic reservoir model was built from the static geological model i.e. using MDT, pressure volume temperature (PVT), routine core analysis (RCA) and special core analysis (SCAL) etc. As part of the reservoir model validation process, history matching of the model was conducted to match the observed data. After history matching was completed, tracers injection analysis and streamlines modelling was conducted in other validate the reservoir model, the well patterns and improved the full field development strategy. The full field development scheme includes water alternating gas (WAG) miscible hydrocarbon gas injection, with 5 observers covering both the flank and crestal locations. The tracer analysis included the injection of 11 different tracers in 11 different injectors, and the monthly monitoring of the producers within the well pattern. Streamline model was built simultaneously from the convention compositional model, in order to analysis and predict the different tracers break through times i.e. both observed from the field and simulated. In addition to this, streamline time of flight (TOF) analysis, the effect of tracking of different tracer components, geological and geophysical impact was evaluated, in order to improve the breakthrough time match between observed and simulated time. As a result of this analysis reservoir management improved, as the source of increasing higher GOR in specific wells were discovered. Remedial actions were recommended to help reduced the increasing high GOR in the respective wells. Also the field development strategy improved, as injector's contribution per well pattern well was quantified. As a result injection could be redistributed. In producer wells with little or no support from their respective injectors, a plan was made to ensure that such patterns could be close appropriately. This improved and maintains the viodage replacement ratio (VRR) in the field, according to reservoir guidelines. This paper describes how by using gas tracer injection and streamline modelling reservoir management and field development can be improved. In addition to the improvements in reservoir management and field development strategy, several lessons learnt and best practice were suggested from the tracer and streamline study conducted. They include but are not limited to; Different tracers can show different concentration level at breakthrough wells.Wells further away from injectors can show earlier breakthrough than well close to injectors, as it is dependent on reservoir connectivity.In analysing model breakthrough times, the measurable tracer in the field needs to be used, as opposed to the actually first breakthrough seen in the simulator. As below such concentration cannot be measured.Operational changes in wells need to be captured properly in the simulator, in order ensure improve prediction of tracer breakthrough times.