A Novel Method to Identify Pore Fluids Using Nuclear Magnetic Resonance (NMR) Logging in Tight Reservoirs During Oil-Based Mud (OBM) Drilling

Abstract Identifying pore fluids in tight reservoirs during oil-based mud (OBM) drilling present significant challenges. Traditional methods, such as crossplotting porosity and resistivity or overlapping multiple curves, often fall short due to extremely low hydrocarbon saturation. The differences in log responses between hydrocarbon-bearing formations and fully water-saturated layers are minimal. However, nuclear magnetic resonance (NMR) logging can play a crucial role in formations drilled with water-based mud (WBM), and various methodologies have been proposed. Unfortunately, in formations where OBM is used for drilling, existing methods are ineffective because they do not account for the impact of OBM filtrate on NMR T2 spectra. This study aims to enhance the applicability of NMR logging for pore fluids identification in tight formations drilled with OBM. We conduct NMR experiments on 10 representative core samples retrieved from the Qingshuihe Formation in the southern margin of Junggar Basin. These core samples are analyzed under two saturation conditions: fully water saturated and gas-driven water states. The third saturation state, gas-displacing- OBM filter is not absolutely finished at present. W extract the T2 distributions from in-suit field measured NMR logging to represent this state. We seek to clarify the NMR response mechanism under the OBM displacing condition. For fully water saturated rocks, the NMR T2 distribution demonstrates a short distribution, with relaxation times ranging from 0.03 to 10 ms. When large pore spaces occupied by OBM filtrate, the resulting T2 spectrum show a bimodal distribution, spanning from 1.0 to 200 ms, with the second peak correlating to the bulk relaxation time of the OBM filtrate. Following the injection of natural gas into the pore spaces, the NMR T2 distribution still evolves into a bimodal pattern. The left peak represents the surface relaxation of irreducible water, while the second peak corresponds to the bulk relaxation of the OBM filtrate and natural gas, but the bulk relaxation time is much longer. We conduct two T2 cutoff values-4 ms and 25 ms-to classify the NMR T2 spectra into three segments. Three parameters, S1, S2, and S3, are derived from the NMR T2 spectra, representing the ratios of small, medium, and large pore volumes to the total pore volume of the rock. A new parameter, termed the pore fluid index (Fluid_index), is introduced to indicate the presence of different pore fluids. Criteria are developed to identify gas-bearing formations, water-saturated zones, and dry layers. Field applications in the southern margin of Junggar Basin demonstrate that the proposed method is effective for our target tight reservoirs in the Qingshuihe Formation. The accuracy of pore fluid identification using NMR logging have significantly improved. Once this technique is widely adopted, it will facilitate better evaluation of tight reservoirs in OBM drilling, thereby greatly expanding the applicability of NMR logging.