QUANTITATIVE PREDICTIONS OF REE ABUNDANCES IN CARBONATITES USING REFLECTANCE SPECTROSCOPY

Rare earth elements (REE) became a strategic raw material in the 21
^st
century and carbonatite-related deposits frequently carry potentially economic levels of these critical metals. To determine major and trace elements (including REE content) conventional analysis of bulk rock geochemical samples are conducted by ICP-OES and ICP-MS. However, these analytic methods are costly and time-consuming. In contrast, modern Remote Sensing (RS) - e.g. proximal remote sensing, mineral or imaging spectroscopy - has become a novel tool for detecting and quantifying geological materials as they offer an efficient, nondestructive and cost-effective way not only to identify different minerals and/or chemical elements but also model their abundances. In this study, we investigated whether multivariate regression modelling (PLSR) allows quantitative prediction of bulk REE concentrations in natural carbonatite samples collected in India (Samalpatti, Sevattur, Amba Dongar), USA (Iron Hill, Lemitar Hills) and East Africa (Tororo, Sukulu, Oldoinyo Dili, Panda Hill).