Dynamic Modelling for Reservoir Connectivity Analysis in Mature Fields

Abstract

Reservoir connectivity remains a critical and growing area of research and application in the petroleum industry, as most discoveries go through development to maturity. This becomes highly imperative for reservoir management decisions in highly fractured compartments or stacked reservoirs with faults across them. In most field cases and especially for a highly faulted region like the Niger Delta, there are some uncertainties around connectivity primarily due to seismic data and resolutions as regards the technology available at acquisition. The primary aim of this work is to use dynamic modelling to ascertain connectivity in mature reservoirs. This work applied the standard workflow for Reservoir Connectivity Analysis (RCA) in evaluating four (4) stacked reservoirs in the RAINBOW field, onshore Niger-Delta using dynamic modelling of the MBAL multi-tank option. Various scenarios were analyzed with the integrated data – geology, production and reservoir pressure history, fluid and rock properties to select the most likely scenario. For this analysis, a new diagnostic plot was introduced for evaluating transmissibility, which improved the clarity in decision making. Using the prevalent economic parameters, a quick evaluation was done to understand the impacts of the reservoir management decisions on the viability of this approach. From the results, two of the four reservoirs are observed to be dynamically connected. The analysis shows that a new perforation extension opportunity is a quick return decision that can yield considerable returns, while new infill opportunities as the optimal decision. Also, the effects of transmissibility on the reservoirs affect the Net Present Values of the decisions. Therefore, this improved workflow approach can be recommended as a quick win when sufficient time and resources are not available for opportunity maturation. Further work is also required to integrate this understanding to build a simulation model for robust benchmarking.