RESERVOIR MODELING FOR A CO2 EOR INJECTION TEST, LOUDEN OIL FIELD, ILLINOIS BASIN

To test the efficacy of CO2 as a flooding medium for enhanced oil recovery (EOR) in the Illinois Basin (Illinois, Indiana, and Kentucky, USA), the Midwest Geological Sequestration Consortium has scheduled field testing of three different techniques at sites of existing petroleum production in the Basin. The first test, a “huff-n-puff” well-stimulation technique, was carried out at the C. Owens #1 well in Louden Field, Illinois, in mid-2007. A huff-n-puff technique entails injection of CO2 into a producing well (the “huff” phase), which is followed by a rest period (a “soak”) before the well is returned to production (the “puff”).

In order to design a successful injection test, it was first necessary to create a realistic reservoir model and injection simulation. However, in mature oil fields, such as the ones that populate the Illinois basin, it is difficult to obtain the necessary data. Louden Field is one such field, having been discovered in 1937 and developed in the 1930s and 1940s. Though wells form a tight, 10-acre grid, not all were logged and few have more modern logs than an SP-Resistivity suite. For 138 wells in the model area, 62 logs and 17 cores were available.

To generate a useful reservoir model it was necessary to convert SP – the log curve most nearly independent of hydrocarbon content – to a sand/shale curve through normalization. The normalized data were then used to geostatiscally characterize the reservoir.

The normalized SP curves were cross-plotted against the seventeen core analyses to obtain regression curves relating SP to permeability and permeability to porosity. Estimated values at cored wells were overwritten by analytical values during modeling. The permeability and porosity models were submitted to reservoir simulation along with the reservoir model after upscaling.

The results of these reservoir models were then passed on to injection simulation in order to predict the test parameters (amount of CO2, distribution of CO2 plume, oil recovery). The results of the injection test will then be compared to the modeling results to analyze the effectiveness of the modeling approach.