Mud Gas Samples: The First and the Last Chance to Access Reservoir Fluid Properties, A Gulf of Mexico Case

Abstract

Mud gas logging is used to test reservoir rocks for presence of hydrocarbons and other properties while drilling. Mud gas sampling enables conducting laboratory tests for more advanced fluid type assessment. Additionally, if well operations prevent fluid sampling downhole, mud gas samples may be the only fluid samples available from the targeted formation. While mud gas samples are inappropriate for many laboratory tests, which require high quality downhole fluid samples (e.g. phase behavior, flow assurance, advanced geochemistry), mud gas provides a signature that can be correlated to specific fluid properties in formations where both fluid and mud gas data are available. These correlations can then be used to infer fluid properties for reservoir zones where downhole fluid samples are not available. In the Gulf of Mexico Tornado field (Green Canyon 280), Pliocene sand reservoirs B5, B6, and B7 have significant lateral and vertical fluid property variations (35°–25° API gravity, 1.3‐ to 2.5‐wt% sulfur, 0.1‐ to 4.7‐wt% asphaltene, 1096–4209 scf/bbl GOR (gas‐oil ratio)). Mud gas samples were collected and analyzed for wells T2 and T2ST which penetrated both B5 and B6 sands. We were able to integrate data from mud gas and downhole fluid samples across the field, and reduce the risk associated with fluid property understanding in T2‐B5 where downhole fluid sampling failed. Because T2 is the only well located in a separate fault block, the findings have implications for the petroleum system evaluation. In T2ST‐B5, downhole samples display a gas condensate fluid (31.9° API gravity, 3694‐scf/bbl GOR, 0.1‐wt% asphaltene, 1.4‐wt% sulfur). The underlying B6 sand is characterized by a black oil fluid across the field (average 27.8° API gravity, 1290‐scf/bbl GOR, 1.5‐wt% asphaltene, 2.0‐wt% sulfur). One other B5 fluid sample was collected in Tornado field (T1ST well) and indicates a black oil, but with a distinctly higher GOR and saturation pressure than B6 fluid. No mud gas samples are available for T1ST‐B5. In T2 and T2ST, B5 and B6 mud gas samples show composition and 13C values compatible with a predominant thermogenic gas, with a possible biogenic methane contribution (average wetness 8.6 and C1‐13C ‐57.7‰), not altered by in‐reservoir alteration processes. Well T2 has no significant changes between B5 and B6 mud gas composition and 13C values. In T2ST‐B5, mud gas wetness decreases and shows lighter methane 13C values by 2.2‰. Because mud gas samples were collected using a simple degasser, small variations are expected between mud gas 13C data and formation fluid 13C values. However, the same isotopic shift was recorded between T2‐B6 and T2ST‐B5 downhole fluid samples, suggesting that the observed isotopic shift in the mud gas samples is representative of formation fluid variations. Moreover, the T2ST‐B6 mud gas indicates similar composition and 13C values to T2‐B5 and T2‐B6. Mud gas and downhole fluid data therefore suggest that T2‐B5 fluid is a black oil, possibly close to the B6 fluid type in the Tornado field. A more complete record of mud gas analysis with corresponding fluid data would help refining T2‐B5 fluid type. Using advanced quantitative degasser with controlled temperature and volume would also reduce the uncertainty on the fluid characterization. In conclusion, the fluids in T2‐B5 sand, which appear to be similar to B6 fluids, are proposed to be black oil with a characteristic 2.2‰ offset in C1 13C values compared to the gas condensate found in T2ST‐B5. The related implications for the reservoir charge history are discussed.

AAPG Datapages/Search and Discovery Article #90349 © 2019 AAPG Hedberg Conference, The Evolution of Petroleum Systems Analysis: Changing of the Guard from Late Mature Experts to Peak Generating Staff, Houston, Texas, March 4-6, 2019