--> Abstract: Modeling Variable-Saturation Multi-Relative Permeability Layered Systems in Transition-Zone Environments; #90063 (2007)

Datapages, Inc.Print this page

Modeling Variable-Saturation Multi-Relative Permeability Layered Systems in Transition-Zone Environments

 

Byrnes, Alan1, Saibal Bhattacharya2 (1) Kansas Geological Survey, Lawrence, KS (2) Kansas Geological Survey, KU, Lawrence, KS

 

Fundamental to reservoir modeling is the assignment of petrophysical properties to geomodel cells. Imbibition oil-water relative permeability (kr) measurements performed on Pennsylvanian-age oomoldic limestones and Mississippian-age moldic-porosity mudstone to grainstone lime-dolomites show residual oil saturation after waterflood, Sorw, increases with increasing initial oil saturation, Soi due to increasing oil trapping in fine pores, consistent with the Land-defined trapping characteristic. The trapping characteristic is lithofacies- and porosity-specific. As Soi decreases with depth in the transition zone, proper modeling of kr requires a family of kr curves that reflect changes in kr with changing Soi.

 

Simulations in a vertically finely-layered model utilizing a family of kr curves exhibit higher oil and water recoveries than predicted from models utilizing kr curves with a constant Soi and Sorw. In the transition zone, proximity to the oil-water contact leads to lower Sorw(Soi) and higher Sw causing both higher oil and water recoveries.

 

Transition-zone systems further illustrate a larger issue with upscaling. Simulation studies demonstrate that systems comprising layers of different kr cannot be rigorously upscaled using static kr properties because kr is a function of how the saturation was achieved. Results indicate that relative permeability is not a state function, as it is widely applied in simulation, but is dependent on the saturation distribution which upscaled systems may not fully represent. Simulation study results demonstrate that fluid recovery from transition-zone dominated reservoirs is critically influenced by Soi, Sorw, and kr models and kr upscaling methodology. Proper modeling provides significant improvement of IOR and EOR implementation.

 

AAPG Search and Discover Article #90063©2007 AAPG Annual Convention, Long Beach, California