Learnings From Impact and Implications of Signal-To-Noise in NMR T1-T2 Logging of Unconventional Reservoirs

Abstract

We study the impact of signal-to-noise ratio (SNR) on nuclear magnetic resonance (NMR) T1-T2 maps across data sets acquired in multiple wells of an unconventional field under various logging and processing conditions. The mean and standard deviation of NMR porosity error between continuous moving-pass and stationary measurements are used to obtain insights into the impact of SNR on accuracy and precision. In a proof-of-concept experiment, we introduce a novel semi-analytical smeared-peak (SASP) technique that compensates for the over-regularized smearing due to poor SNR, of T1-T2 relaxation responses of different fluids. The SASP approximation to de-smear volumes of different fluid types is validated with field measurements from multiple wells. The uplift of the SASP technique in improving fluid volume interpretations is apparent in the in-situ calibration of low-SNR moving-pass NMR measurements with high-quality stationary measurements. The learnings show that logging protocols that are designed to increase SNR by combining specific acquisition parameters with processing strategies, within acceptable compromises, are mandatory for reliable NMR characterization of unconventional reservoirs.