The objective of this paper is to calibrate porosity and permeability derived from NMR logs to core measurements and thus allow improved porosity and permeability prediction from new and historic NMR logs. Permeability, and to a lesser extent porosity, are the most important parameters in determining the potential production of a reservoir. Measuring these parameters accurately is challenging in carbonate formations due to heterogeneity and the wide variety of pore classes, such as interparticle porosity, moldic porosity, vuggy porosity, and microporosity. Traditional density-neutron and sonic tools used to measure the porosity are strongly dependent on lithology which might yield incorrect porosity measurements. NMR logs, while largely independent of matrix properties, can potentially underestimate microporosity. No wireline tool is capable of continuous permeability logging.
In this study, we provide an assessment of NMR measurements at laboratory and reservoir conditions to allow improved estimation of porosity and permeability. The formation under investigation is largely limestone, with negligible dolomite or clay – but has a large range of pore structures. Several samples from different rock types were selected for this study. These were investigated with computed tomography scanning, thin section analysis, and high pressure mercury injection capillary pressure (MICP) tests. NMR T1 and T2 measurements on fully formation water saturated samples were then conducted at laboratory and reservoir conditions. Finally, to define the bound and free water, and determine appropriate T2cutoff(s) for each rock type, samples were desaturated to irreducible water saturation (Swi) using a centrifuge technique, followed by NMR T1 and T2 measurements at Swi at reservoir conditions. Additionally, the laboratory NMR measurements are compared to wireline log NMR measurements at the same depths.
Pore size distributions from NMR measurement agree well with those derived from MICP tests. The NMR measurements on desaturated samples showed that Swi and the corresponding T2cutoff vary from sample to sample. Possible controls on this will be discussed. The results from this study showed that T1 measurements confirm T2 responses: both are strongly dependent on temperature, but less dependent on pressure.
A strong 1 to 1 correlation between helium porosity, φHE and NMR porosity, φNMR was observed demonstrating that in this formation under recording of microporosity is not an issue. Based on NMR data, permeabilities were calculated using both the Coates and SDR models. It was found that the Coates model gives good agreement with permeabilities measured on core plugs by the standard laboratory techniques.