oa Log-based permeability prediction in a complex carbonate hydrocarbon field, Kangan and Dalan Formations, Iran

The excellent dataset available from this field has been utilized in the search for a log-based permeability model. The resulting model required input from the ‘basic’ suite of logs available in all production wells and the sonic log was therefore not included. The main depositional lithofacies of the Kangan and Dalan Formations in the area of interest are ooid-peloid grainstones and lagoonal wackestones-mudstones. Diagenesis (dissolution, cementation, dolomitization) has modified initial pore textures and ideal reservoir rocks (grainstones) have partly been transformed into low permeability-high porosity oomoldic rocks. Dolomitisation of more muddy lithofacies enhanced the reservoir quality and high permeability- moderate porosity conduits for fluid flow were formed. The key to successful prediction of permeability thus becomes the identification of pore textures. This was done through the application and adaptation of concepts developed by J. Lucia. The resulting model was developed using logs and core data-descriptions from three extensively cored wells. The model was subsequently applied to a number of production wells, where production logs confirmed the predicted permeability profiles. Nuclear magnetic resonance and sonic measurements can also provide independent estimates of permeability along a wellbore. In the cored wells, these data exist, and interpreted permeabilities are available. Results are presented and discussed in a rock-fabric context.