Predicting Pore Types in Khuff Reservoirs: A Step Towards Improved Permeability Predictions

Ian Billing
Saudi Aramco
[email protected]

The Upper Permian Khuff-B reservoirs of Saudi Arabia are prolific gas producers, with much of the production coming from stacked grain-rich shoals. While such stacked shoals often have excellent primary inter-particle porosity, this is frequently superseded by a phase of pore cementation and grain leaching. Thus, ooid- and peloid-rich reservoirs are frequently characterized by oo- or pel-mouldic pore systems. Such pore systems may contain significant hydrocarbons but connectivity may be limited due to the isolated nature of these pores. The detection of the relative abundance of poorly connected oo-mouldic porosity is essential for applying accurate porosity-permeability transforms. This study investigated the use of conventional wireline log data to identify relative percentages of well-connected and poorly-connected pore systems in Khuff-B reservoirs. Detailed core descriptions were carried out on two calibration wells, producing digital tracks of five major classes of pore types. Careful depth matching and correlation with reprocessed wireline logs created a highly constrained dataset. A combination of sonic, density and neutron logs produced predictive algorithms which identified calibrated zones of poorly connected porosity. Blind-testing of these same algorithms on a wireline log suite from a third Khuff-B well produced a very good match. Ongoing work is now investigating the transform from pore-type distribution to improved permeability predictions. These new algorithms act as a proxy for creating boundaries to the permeability distributions; the resultant transforms match the datasets significantly better than a simple linear transform method, and can be routinely used as part of the process of Khuff-B reservoir assessment.

AAPG Search and Discovery Article #90077©2008 GEO 2008 Middle East Conference and Exhibition, Manama, Bahrain