Palaeozoic oil-source correlation in the Tarim Basin, NW China: A review

Oil-source correlation in the cratonic region of the Tarim Basin, NW China has long been controversial. Current knowledge of the potential source rock distribution, end-member selection, oil-source correlation and impacts of secondary alteration processes have been reviewed. Two source rock systems from the Cambrian-Lower Ordovician (epsilon-O-1) and Middle-Upper Ordovician (O2-3) potentially contributed to oil accumulations in the cratonic region. Geochemical correlations suggest that oils are dominantly derived from the O2-3 source rocks, while geological evidence and sulfur isotopic compositions supports the epsilon-O-1 source rocks as a main contributor. Such inconsistency is rooted in the selection of endmembers and their characterization. Changes in some biomarker parameters, such as sterane distribution, gammacerane index and ratio of tricyclic terpanes to pentacyclic terpanes, are mainly controlled by thermal maturation and biodegradation rather than source input and depositional environment variation. No clear boundary between the two source systems can be established using biomarkers except for a few diagnostic components. Some epsilon-O-1 source signatures, such as even/odd predominance of n-alkanes, unusual tricyclic terpane distribution, unusually enriched C-13 carbon isotopic compositions of oils and kerogens, and the occurrence of combustion related polycyclic aromatic hydrocarbons, largely result from an abnormal heating influence either from igneous intrusion or hydrothermal fluid activity. Local thermochemical sulfate reduction can also remove most source-related signatures. These so-called markers in currently selected end-members result from extensive secondary alteration processes and do not reflect source input differences. While more research is needed for full reconstruction of oil-source correlation in the cratonic region of the Tarim Basin, the main source rocks most likely reside in the epsilon-O-1 succession rather than O2-3. (C) 2016 Elsevier Ltd. All rights reserved.