Integration of NMR, Conventional Logs and Core Data to Improve Formation Evaluation of a Gas Reservoir in Kangan and Dalan Formation, Iran

Conventional log based reservoir characterization of a gas reservoir in the Kangan and Dalan formations have recently been improved by application of the nuclear magnetic resonance log (NMR). Important reservoir properties such as permeability, pore size distribution and capillary pressure curves can be estimated from NMR. These parameters are simulated directly in the laboratory on core samples recovered from the reservoir. Due to high cost associated with coring and some technical problems, few wells in any given field are cored. The only problem of NMR measurements in gas reservoirs is that in gas-bearing zones, total NMR porosity read much less than actual porosity due to low hydrogen index of the gas. This problem was solved by integration of NMR porosity with conventional well logs such as density and sonic and compared with core porosity. Improved porosity calculation lead to better core independent permeability estimation on the wells logged with NMR. NMR T2 distribution was calibrated with laboratory derived pore size distribution in 7 samples and a constant scaling factor was derived for each rock type to predict a pseudo pore size distribution from NMR. Logarithmic mean of pore size distribution in 4 wells with NMR was integrated with conventional logs in an artificial neural network to predict a pseudo pore size distribution logarithmic mean (PPSDLM) in the wells without NMR. PPSDLM and conventional well logs were involved to an electrofacies modeling to predict electrofacies in the reservoir for characterization of heterogeneity of the reservoir in 3D geological model. NMR permeability was also imported to the model as an associated log to predict facies base permeability. To test the permeability prediction, estimated permeability was compared with core derived permeability on 2 cored wells to see how well, estimated permeability fitted the actual core permeability.