Estimation of Permeability in a Naturally Fractured Carbonate Reservoir, Lessons Learned

One of the most important petrophysical properties of an oil/gas reservoir, which directly controls hydrocarbon production, is permeability. In cases when pore-perm data of core is properly classified, a good semi-log function can be fitted to each classes. We observed that non of facies based methods can classify the data accurately. In this paper, lessons-learned from prediction of permeability in a naturally fractured reservoir in SW Iran is presented. The studied reservoir formation is Sarvak. It is a carbonate formation which mainly deposited in marine shelf environment in Albian.

In pore-perm cross-plot, two populations can be distinguished. The first, which is in 45 fitting line and the second one is in upper-left part of the cross-plot. It seems that the second group shows an abnormal trend. By removing core plug samples with fractures, the remaining samples are those representing matrix permeability. We used Lucia Rock-Fabric Number concept to estimate permeability in un-cored wells. Performance of the prediction was appraised in a blind test well. In this well, matrix permeability was predicted with high R2=0.78, which is acceptable in a heterogeneous carbonate reservoir.