Log Derived Novel Approach to Estimate Vertical Permeability in Multilayer Reservoirs Through Resistivity Anisotropy in the Natih Formation, Oman

Abstract

Abstract For an oil reservoir subjected to waterflooding, the horizontal and the vertical permeabilities have a considerable influence on the areal and vertical sweep efficiencies. Particularly, in the case of the Natih formation – a multilayered and heterogeneous reservoir located in Oman – a new workflow is proposed to enhance vertical permeability estimation through resistivity anisotropy modeling, while also correcting Sw (water saturation) for resistivity anisotropy effects. Permeability anisotropy (Kh / Kv) was measured in core samples taken from a vertical pilot hole and correlated with the resistivity anisotropy (Rv / Rh) inferred from inversions of LWD resistivity data in offset lateral wells. Also, by modeling phase-shift and attenuation resistivity measurements in the laterals, true horizontal resistivities are estimated and then used to obtain a more accurate Sw (water saturation). The Rv / Rh ratio of the laterals is used to infer Rv / Rh in the verticals and to generate the final Kh / Kv transform in the laterals. The corrected Sw as well as the estimated Kh / Kv ratio in the laterals are used as inputs to the field reservoir modeling. Traditionally, when history matching waterflooded Natih reservoirs, the vertical permeability was assumed to be a fraction of the horizontal reservoir permeability. A Kv / Kh ratio, typically in the range of 0.1 to 0.3, was applied throughout the model. In the absence of core measurements, the Kv / Kh ratio was iteratively adjusted through trial and error, until a reasonable match of the historical production and injection data was achieved. The process was time-consuming, especially when history matching a large field. Through this new workflow, vertical permeability was estimated as an independent and variable reservoir property on a foot-by-foot basis along each lateral of the field. The new independent permeability was tested as an input in a dynamic model of a Natih reservoir in Oman, yielding results comparable to the traditional approach while significantly reducing the time to achieve history match.