Detection of Bastnäsite-Rich Veins in Rare Earth Element Ores Through Hyperspectral Imaging

Abstract

Spectral indices are a well-established remote sensing tool showcased in many studies pertaining to fields such as agriculture, forest ecology, geology, soil sciences, vegetation, water resources, and urban development. In the case of geology, several indices already exist for the purpose of highlighting regions with interesting lithologies, such as iron oxides, aluminous clays, carbonates, and mafic to ultramafic minerals, based on signature absorption features in their reflectance spectra. However, in the literature, most indices often depend on the few, wide spectral bands used by spaceborne multispectral imagers such as Landsat 8 OLI, Terra ASTER, and Sentinel-2 MSI. Hyperspectral imagers are able to glean more detailed compositional information of a rock surface by operating with a myriad of narrow spectral bands, though the usage of these bands in spectral indices for mining and exploration purposes remains largely unexplored. We propose a new bastnäsite index capable of detecting the presence of rare earth elements and mapping their relative abundances across rock surfaces. Using four reference bands and four absorption bands in the visible to near infrared wavelength range, we apply this novel technique in a laboratory setting to ore samples extracted from the carbonatite body localized in the Sulfide Queen mine in Mountain Pass, California. Visual inspection is used to compare an index map to corresponding reflectance spectra for areas of known high and low rare earth element abundances, demonstrating its potential as a fast and straightforward means to assess rare earth element ore grade remotely.