Reverse Time-Lapse Technique for Moveable Gas Identification

Abstract

The Barik, Miqrat and Amin formations are deep, tight reservoirs of the Haima Supergroup that provide the majority of gas production in the Sultanate of Oman. The Miqrat formation is a feldspathic sand/shale sequence with complex pore structure and occasional bitumen presence. In the area of interest, it occurs at a depth of approximately 5000 m. Average porosity varies from 5 to 9%, average permeability for Lower Miqrat does not exceed 0.1 mD. In general, Archie equation derived saturation in low porosity rocks is subject to medium to high uncertainty. Therefore the most common challenge in the petrophysical evaluation of tight reservoirs is the determination of gas saturation and fluid type identification. In an effort to improve the reliability of saturation calculation and fluid typing, several different methods were tested including cased-hole Pulsed Neutron Logs (PNL). The classical sigma interpretation was found to be too sensitive to input parameters and did not provide significant improvement to saturation determination in the complex Haima lithologies. An important breakthrough was made when the dynamics of the mud filtrate invasion process in these reservoirs was understood. During open-hole logging usually very little or no gas effect is observed on logs with negligible or no density-neutron separation. The reason is considered to be deep mud filtrate invasion pushing moveable gas beyond the depth of investigation of radioactive logs. One or two months later, the filtrate in the invasion zone dissipates with gas returning to the near wellbore formation. The best match between log calculated moveable gas saturation and production test data was obtained using a reverse time-lapse technique, with PNL cased-hole logs compared to baseline open-hole neutron measurements. The changes in neutron porosity with time can be attributed to moveable gas saturation. Careful neutron log quality control and normalization across non-reservoir and known water-bearing sections is required. Knowing the hydrogen index of gas, we can calculate the moveable gas saturation from the difference in neutron log response. In contrast to the sigma approach, an accurate rock matrix model is not required. This paper describes the Reverse Time-Lapse technique: a novel application of the classic time-lapse technique between open-hole neutron and cased-hole PNL. The case studies demonstrate that this technique is applicable for completion decision making and field-scale development planning.