Improved Understanding of Reservoir Connectivity in an Evolving Waterflood with Surveillance Data

Abstract

Abstract Establishing connectivity among various injectors and producers is a key to improving the understanding of a reservoir under waterflood. This understanding improves the estimates for ultimate recovery and also helps to better define the future development plan. In deepwater turbidite reservoirs, numerical flow-simulation models are used to make performance predictions, with reservoir connectivity as one of the key uncertainties. In the initial phase of field development, interwell tracers were used to assess the connectivity. As more wells were drilled, updates were required for the simulation models. Instead of waiting for the next phase of an ongoing tracer program, both rate-transient analysis (RTA) and capacitance-resistance model (CRM) were used to understand connectivity. The input for both RTA and CRM are the rates and pressures, which are being gathered with real-time surveillance. This paper presents a case study to compare findings from the use of interwell tracer data with the results of CRM based on dynamic data. Another study element demonstrates the use of RTA in identifying and estimating the volume of thief zone. Attempts are made to use CRM and RTA to predict connectivity based on performance prior to experiencing water breakthrough. These case studies demonstrate the application of RTA and CRM in ongoing waterfloods. The CRM concurred with the initial tracer results and helped to understand the change in pressure distribution with time as the field was being developed. We learned that the use of CRM can be a viable alternative to an interwell tracer program to reduce uncertainty related to injector-producer connectivity. CRM also helped in understanding the efficiency of the injectors, which is important in a facility with limited water injection capacity. The ease of use of CRM and RTA makes them useful as screening tools in the process of developing a detailed flow-simulation model.