Hyperspectral images of fossil coral cores reveal a heterogeneous nature previously unseen and difficult to quantify. This heterogeneity can wreak havoc on geochronology and paleoclimate analyses.
Researchers have shown for the first time how hyperspectral analysis (analysing waves beyond the visible light spectrum) can be used to effectively and non-destructively map out minerals within fossil coral samples.
Spectral imagery has been delivered via satellites for decades, and hyperspectral analysis is a common tool in the mining and minerals industry, where it provides a high-resolution record of mineralogical changes and their relative abundance. Geologists can use this information to enhance their understanding of economic mineralisation, such as gold. However, in a novel study undertaken by Dr Richard Murphy from the Centre of Field Robotics at the University of Sydney in collaboration with researchers at the Geocoastal Research Group at the University of Sydney, the University of Queensland, Queensland University among other institutions, researchers applied the technology to coral specimens.
In a traditional study of mineralogy, samples would have to be cut from the core and pulverised for X-Ray diffraction (XRD). As a result, individual corals, especially those smaller in size, would be unavailable for further analyses, such as dating or taxonomy. Instead, this hyperspectral approach used a short wave infrared sensor (SWIR 1000-2500) to obtain a spectral image with a 0.33 mm resolution, leaving the original sample completely intact.
Coral cores from One Tree Reef in the Southern Great Barrier Reef and Papua New Guinea were acquired for the study and imaged using a Specim hyperspectral camera. Using Automated Feature Extraction (AFE), the spectral signatures were then classified as calcite, aragonite, or dolomite giving scientists an unbroken record of coral growth and mineralogy throughout the samples. The AFE technique allows very small changes in mineralogy to be detected, which are important to the interpretation of environmental change.
The high-resolution images showed a greater degree of heterogeneity than would be possible using the traditional point sampling method and proved to be a useful tool to geochemically estimate not only how altered the original coral skeleton is, but also to determine whether terrigenous (i.e. land-derived) clays are present. Using this method, scientists can reliably identify parts of the core that would be most pristine for purposes such as radiometric dating and paleoclimate reconstructions, allowing them to know at what time in the geologic past these corals were growing.
This project is part of a larger ongoing collaboration that seeks to understand how reefs evolved during previous climate change.
You can read this research, published in Geochemistry, Geophysics, Geosystems here.
