Abstract
One of the most uncertain aspects of producing gas shale is the prediction of production with time. Typically, the production qualitatively exhibits a two-component decline. It drops at a fast rate just after the well is stimulated (fractured), then settles with slow declining long-term production. Quantifying this decline is a crucial aspect in determining the economic viability of the well and making revenue predictions.
In this work, we reproduced in the laboratory the depletion of a gas shale reservoir using three different shale core plugs. Starting from a sample saturated with methane gas, we monitored the depletion for several days by continuously acquiring the nuclear magnetic resonance transverse relaxation time (NMR-T2) signal arising from the methane in different pore types. The results showed that we were able to develop an experimental workflow to quantify methane saturation in a shale sample down to fractions of porosity units and monitor production of gas over several days, which allowed the determination of production rates of different T2 environments in shale. We were also able to predict reasonable permeabilities with a highly simplified model in the range of 1.4 nD to 145 nD. The results also showed different behaviors of production rates despite the very similar native T2 distributions measured from the samples.
This workflow using NMR time dependent monitoring allowed us to extract an extremely rich dataset, demonstrating that permeability in unconventional reservoirs is an extremely complex phenomenon not accurately describable with a single number.
