METHODS TO ESTIMATE EROSION IN SEDIMENTARY BASINS. (30th March 2021)

• Record Type: Journal Article
• Authors: Iyer, Karthik; Hartz, Ebbe H.; Schmid, Daniel W.
• Is Part Of: Journal of petroleum geology. Volume 44:Number 2(2021)
• Page Start: 121
• Page End: 144
• Publication Date: 2021-03-30
• Subjects: erosion -- vitrinite reflectance -- sonic velocity -- numerical model -- Barents Sea -- sedimentary basin -- petroleum systems; Petroleum -- Geology -- Periodicals; Aardolie; Electronic journals; 553.28
• DOI: 10.1111/jpg.12782
• Languages: English
• ISSNs: 0141-6421

Abstract

Abstract : Net erosion, the difference between the present‐day and maximum burial depths of a reference unit, may have a major impact on hydrocarbon prospectivity in a sedimentary basin. Erosion may affect all the components of a petroleum system, from source rock to reservoir to seal. In most cases, vitrinite reflectance (VR), temperature and sonic velocity data, which are often readily available, can be used to determine net erosion in a region based on the thermal and mechanical evolution of sedimentary layers with burial. This paper revisits these methods and discusses the determination of net erosion from these datasets. Furthermore, it is shown that a closer look at the data is warranted if the estimates derived from complementary VR/temperature and velocity datasets significantly diverge. Such differences can be reconciled by critically examining the datasets and the regional geology, resulting in erosion estimates from both datasets which are within 100 m of each other. Lastly, a fully automated, process‐driven method combined with multi‐objective optimization algorithms and that takes all three datasets into account is showcased while determining net erosion for three wells located in the Norwegian Barents Sea. One of the benefits of this method is that it explores a wide range of likely scenarios that would best match the different datasets. Furthermore, the method can also automatically flag datasets that are inconsistent with each other by returning an overall low fit score. These datasets can then be critically examined to determine their reliability and to arrive at a more consistent erosion estimate, reducing the error margin to about 100 m.