About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

Journal of Sedimentary Research (SEPM)

Abstract


Journal of Sedimentary Research
Vol. 89 (2019), No. 3. (March), Pages 187-198
DOI: 10.2110/jsr.2019.10

Enhanced Reservoir Characterization Using Hyperspectral Core Logging

Ralf R. Haese, Grant Jiang, Scott Ooi, Jay R. Black

Abstract

High-resolution records of rock properties derived from wireline and core logging together with seismic data are commonly used for the development of geological models for hydrogeology, sedimentology, and facies studies as well as in the context of resource exploration. The exploration of sedimentary reservoirs benefits from continuous mineral logs in order to define and discriminate lithotypes. Such a lithotype classification accounting for the mineral composition can then be used, for example, in reactive-transport models predicting changes in water composition and fluid–rock reactions driven by the injection and long-term partitioning of CO2 into gas, fluid and mineral phases. Automated recording of mineral spectral reflectance in the very-near and short-wave infrared (VNIR/SWIR, 280–2500 nm) and the Thermal InfraRed (TIR, 6000–14500nm) wavelength ranges offers rapid, semi-quantitative analysis of hydrous and anhydrous minerals in cores and cuttings. This study uses TIR reflectance from two sedimentary rock cores from the Darling Basin (NSW, Australia) to a) derive logs of lithotypes with the end members being sandstone, mudstone, and calcareous sandstone and/or mudstone and b) integrate hyperspectral and wireline logging data. Good agreement is found between lithotypes determined by XRD and hyperspectral logging and between lithotypes and porosity. However, there is a poor agreement between absolute mineral abundances measured by the established quantitative techniques of XRD and QEMSCAN when compared to the semi-quantitative analysis of the hyperspectral logging. The abundance of mica correlates with gamma radiation, but the slope of the correlation differs significantly between the two cores possibly due to different sediment provenance or diagenetic processes. Kaolinite was found by hyperspectral logging in many intervals whereas it was below XRD detection limit, which can be explained by a high ratio of mineral area-to-mass ratio and possible mineral overgrowth leading to disproportionally high reflectance. It is concluded that hyperspectral core logging is a rapid and effective way to determine variable lithologies at high resolution; however, it is not replacing quantitative mineral analysis.


Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24