NMR relaxometry interpretation of source rock liquid saturation — A holistic approach

Abstract

Fluid types and saturation are essential factors in assessing oil recovery resources and potential in organic-rich mudstones. One of the methods for analyzing tight rocks' saturation is low-field nuclear magnetic relaxometry (NMR). Interpretation of NMR results, particularly the data of T1-T2 maps, delivers different fluid saturation and estimates hydrogen populations. Although previous research proposed a qualitative interpretation of T1-T2 distributions, there is still no obvious scheme that can be directly applied in routine petrophysical studies. However, current NMR-based workflows and manual interpretation routines are subjective and laborious. We attempt to fill this gap by identifying the best interpretation scheme for three distinct types of organic-rich source rock of the Bazhenov formation and modifying a well-known T1-T2 map interpretation scheme. Experimental results include NMR 2-MHz T1-T2 datasets (original and differential maps for samples in as-received, dried, and saturated states) with the corresponding Rock-Eval pyrolysis parameters and mineral composition from X-ray diffraction. The analytical part compares the existing schemes and modifies the interpretation scheme based on the rock petrophysical background. The paper reviews the features of the manual interpretation and previously established schemes and proposes a modified scheme best fitted to the petrophysical and geochemical data. In addition, we study the impact of mineral composition and geochemical properties on T1-T2 signal interpretation and analyze cross-correlations. The study confirms the presence of a significant amount of clay-bound water from a detailed analysis of the corresponding regions in T1-T2 regular and differential maps. In the paradigm, we propose a semi-automated processing of NMR T1-T2 data using a custom interpretation scheme composed of regions from multiple interpretation schemes and consider their modification for high-quality results.