Distinguishing fractures from matrix pores based on the practical application of rock physics inversion and NMR data: A case study from an unconventional coal reservoir in China

Our main scheme in this study was distinguishing between fracture porosity and matrix porosity in coalbed methane reservoirs through a novel approach which is the joint usage of NMR transverse relaxation (T2) measurements and rock physics modeling based on Levenberg-Marquardt (LM) algorithm. For this purpose, NMR T2 relaxation curves of 34 water-saturated coal samples, prepared and processed in the laboratory, were measured and the pore size distribution inside them was investigated. Subsequently, matrix and fracture porosity were separated based on the threshold T2 relaxation time (90–110 ms) which was achieved through our particularly designed fracturing experiments (this is different from the common T2 cutoff). The T2 measurements were performed in the laboratory using our NMR machine. After that, a rock physics scheme based on Levenberg-Marquardt algorithm was applied to independently estimate matrix porosity and fracture porosity from the samples' statistic mechanical properties including compressional wave velocity (Vp), shear wave velocity (Vs), bulk modulus (K), shear modulus (G), Young's modulus (E), and Poisson's ratio (ν) which were all carefully measured in the laboratory. Afterward, both types of the abovementioned porosities were comprehensively characterized and the obtained achievements were listed. Once finished with this step, the 3D structure of the entire reservoir was extracted using the recorded data inside 32 drilled wells, and then the established models were upscaled and the unique and independent contour maps of fracture porosity and matrix pore porosity were drawn over the entire reservoir area.