Reservoir characterization and production history matching of Lower Cretaceous, Muddy Formation in Ranch Creek area, Bell Creek oil field, Southeastern Montana, USA

Abstract

A detailed reservoir description of the Lower Cretaceous, Muddy Formation is an essential step for developing an enhanced oil recovery project in the Bell Creek field. The applied workflow highlighted the role of key uncertainties that were not well addressed during early field development. Several challenges including: stratigraphic subdivision of the reservoir units, rock type complexity due to diagenesis, petrophysical data inconsistencies and fluid contacts have historically led to hurdles in reservoir development. The study outlines a workflow used to refine some of these uncertainties and generate better inputs for the geologic model. The Ranch Creek area at Bell Creek field in southeastern Montana is characterized by a clastic sequence of mixed rock types from argillaceous mudstone, bioturbated tightly cemented to clean sandstone lithofacies. Delta front to fluvial-dominated depositional settings as well as diagenesis played an important role in controlling reservoir quality and geometry. The facies complexity is reflected in rock properties controlling fluid migration through the reservoir intervals. The lithofacies types are clustered into three groups, each with characteristic porosity and permeability cutoffs based on core plugs measurements. The rock types are linked to depositional facies obtained from the petrographic description and calibrated to the clusters derived from the capillary pressure measurements. Porosity distribution within the reservoir is biased to the diagenetic complexity and rock type distribution. Three different transforms were used to calculate the corresponding permeability to the modeled porosity parameter. Modeling water saturation for the three different capillary pressure groups using the J-function approach improves the spatial distribution of water saturation and leads to a better match of historical oil and water production rates.Reservoir simulation is crucial to improve our understanding of rock and fluid properties. Integrating dynamic data such as pressure, production and injection rates helps to reduce the uncertainty of the model parameters. The history match process highlights the need to tune the output of the static model. Iterations on the static model, focused on changing the ranges of key parameters such as facies volume fractions and water saturation that contain the most uncertainties. Applying these changes results in an improved geological model that provides a better data to the simulation model. The tuned static model was able to reproduce the observed results. Matching the pressure data supports a valid oil-in-place which is consistence with historical volumetric calculation.