A New NMR Interpretation Technique Using Error Minimization With Variable T2 Cutoff

Abstract

A New NMR Interpretation Technique Using Error Minimization With Variable T2 Cutoff M. Oraby, SPE, Halliburton Energy Services, N. Chafai, Sonatrach Oil Company, Algeria, R.B. Hussing and D.R. Massengill, Parker & Parsley Petroleum USA Inc., J.S. Clark and D. Pletcher, Halliburton Oil Producing Company Abstract A complete formation evaluation requires determination of irreducible fluids, movable fluids, and formation permeability. Analysis of NMR measurements to determine these parameters is strongly dependent on formation lithology and mineralogy, especially in complex and heterogeneous reservoirs. Therefore, NMR measurements must be combined with measurements from conventional logs for comprehensive formation evaluation. In this paper, an interpretation method that combines these data is discussed and is applied to shaly sand and carbonate formations. An important advantage of the method presented in this paper is its capability to generate a variable T2 cutoff for NMR interpretation that depends on the volumetrics of the calculated formation lithology. In NMR analysis, the T2 cutoff is required to divide effective porosity into irreducible and movable porosity. In traditional T2 interpretation, the T2 cutoff is usually a constant value throughout a formation; however, the methodology proposed in this paper uses a variable T2 cutoff calculated on a foot-by-foot basis from log- derived lithology. Comparisons between the constant and variable T2 cutoffs and their effect on formation evaluation are included, and recommendations are discussed. An error minimization technique is used to solve a system of equations derived from the log data. In this system of equations, the NMR effective porosity is used as the formation effective porosity. To evaluate the goodness of the fit between the original logs and the reconstructed logs, the chi-square of the difference between logs is calculated. In this methodology, weighting factors can be applied to the log data, and geological and petrophysical constraints can be imposed. Introduction Nuclear magnetic resonance (NMR) technology has proved to be essential in modem formation evaluation. Many NMR applications have been introduced over the last few years in different areas of formation evaluation and have added great value to oil and gas production. In low-resistivity-pay applications, Oraby and Eubanks applied NMR technology to improve determination of pay zones in East Texas. Similarly, Sutiyono applied NMR measurements to improve identification of low-resistivity pays in the Attaka field of Indonesia. Kenyon et al. used NMR technology to estimate pore-size distribution in microporous, cherty sandstones. Their results are compared to quantitative analysis of digitized thin-section images. Kenyon et al. incorporated NMR lab measurements in their study of depositional textures in the carbonate Thamama group in Abu Dhabi. Loren and Robinson discussed a theory that relates the NMR relaxation time of water and hydrocarbon confined in rock pores to the pore-size distribution of the rock. In the area of residual-oil measurements, Horkowitz et al. used the mud-doping technique using manganese ions (Mn++) to suppress water signals and maintain the oil signal in NMR measurements. The oil signal is then used to estimate residual-oil saturation. Akkurt et al. developed new techniques to apply when logging natural gas reservoirs with NMR tools. The differential spectrum method (DSM) and the shifted spectrum method (SSM) are introduced to identify gas zones and fluid contacts in multiphase reservoirs and to correct for gas effects on porosity measurements. In most of the cited works, researchers either assumed a simple formation lithology or intentionally neglected the rock effects on NMR interpretation and data analysis. Rocks may not have direct effects on NMR measurement techniques, but they do have considerable impact on NMR interpretation methodology. In complex and heterogeneous reservoirs, the simple methods traditionally employed in interpreting NMR measurements can lead to incorrect assessments of reservoir performance. Few papers studied the effect of formation lithology on the NMR measurements. P. 175^