Analysis of Production Data to Improve Characterization of In-situ Megascopic Reservoir Permeability

Abstract

Abstract Critical to the success of history matching a 3D geoscience model is a link between coreplug scale permeability and megascopic scale reservoir permeability. This paper demonstrates how permeabilities derived from production data, complement core data and relatively short duration pressure transient analysis. Heterogeneities related to the fluvial deltaic geology typically result in spatially heterogeneous permeability patterns that degrade or enhance the average permeability within the area of influence. In addition, the area of influence may not be radial (or elliptical) but rather a network of connected channels. The analysis presented for a tight multi-layered gas reservoir in the South Australian Cooper Basin shows aspects that reflect all these issues, hence contributing to further understanding of the dynamic behaviour of the 3D geo model. Modelling the productivity and recovery in such heterogeneous reservoirs requires a good understanding and representation of the permeability distribution. Analytical work done prior to simulation of a multi-layered tight gas field in the South Australia Cooper Basin shows that the permeability is a function of the scale of measurement. Advanced decline methods such as the Inverted Decline Curve1 (IDC) and the Reciprocal Productivity Index2 (RPI), which use production data, have been used to estimate the megascopic reservoir permeability. This is particularly useful when little pressure transient analysis is available. Furthermore, conventional pressure transient analysis can be misleading in tight gas reservoirs with heterogeneities. Specifically, the work done demonstrates how to: Interpret core and well test kh relative to production kh Generate effective kh values for a megascopic scale in fluvial environments Take into account hydraulic fractures and/or naturally occurring high permeability streaks Properly upscale the core plug permeability to the megascopic simulation scale so that the reservoir simulation model can be pre-conditioned thereby reducing the time required to achieve a history match.