A number of geological and engineering parameters influence and control the performance and ultimate recovery from an oil reservoir. These are commonly interlinked and the relative importance of each can be difficult to unravel. These variables include geological parameters such as depositional environment, which has long been considered to be a key factor influencing the production characteristics of fields. However, quantifying the importance of any single factor, such as depositional environment, is complicated by the impact of the other variables (geological and engineering) and their numerous interdependencies. The main aim of this study is to unravel the impact of the depositional environment and primary facies architecture on reservoir performance using an empirical study of oilfields from the Norwegian Continental Shelf. A database of 91 fields, with a total of 7.8 Bbbl (billion barrels) of oil in place, has been built. Within this a total of 93 clastic reservoirs were classified into three gross depositional environments: continental, paralic/shallow marine and deep marine. The reservoirs were further classified into eight depositional environments in order to provide further granularity and to capture their depositional complexities. A further 28 parameters which capture other aspects that also impact production behaviour, such as reservoir depth, fluid type and structural complexity, were recorded for each reservoir. Principal component analysis (PCA) was utilized to explore the importance of sedimentological-dependent variables in the dataset, and to determine the parameters that have the strongest influence on the overall variability of the dataset. PCA revealed that parameters associated with field size and depth of burial had the most influence on recovery factor. Gross depositional environment and other stratigraphic-dependent parameters were the most significant geological factors. Fluid properties, such as API gravity and average gas/oil ratio, were unexpectedly among the less important parameters.
Bibliographical noteFunding Information:
This work was funded by the SAFARI III University of Aberdeen (RG13517-10) and the Petroleum Technology Development Fund (PTDF).
Acknowledgements We are very grateful to reviewers for their thoughtful comments which have improved the manuscript.