Integrating hyperspectral analysis and mineral chemistry for geoenvironmental prediction

Hyperspectral drill core scanning technology (e.g., CoreScan®), which uses visual nearinfrared (VNIR), shortwave infrared (SWIR), and longwave infrared (LWIR) data, is being increasingly used for geological domaining of ore deposits. Advantageously, this technology can identify carbonate-group minerals that can effectively neutralise many mine wastes. The chemistry of neutralising minerals in drill core can also be routinely analysed by multiple techniques (e.g., portable X-ray fluorescence (pXRF), laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS),) but hasn’t been integrated with hyperspectral data for deposit-scale geoenvironmental characterisation. In this study we integrate hyperspectral mineral data with a newly developed LAICP- MS line-scan method to characterise major and trace element chemistry of neutralising minerals in drill core. We demonstrate how this data can enable effective geoenvironmental domaining of ore deposits with examples from a porphyry Au-Cu deposit in Australia. We validate Corescan® data using X-ray diffractometry (XRD) and a series of acid-base accounting tests to define geological domains with high and low acid neutralising capacity (ANC). The distribution and abundance of trace elements are defined in these domains using by LA-ICP-MS. A new geoenvironmental domaining index (GDI) is developed using Corescan® data, which can assist with deposit-wide characterisation. The rapid and cost-effectiveness of hyperspectral core scanning and LA-ICP-MS techniques makes them critical emerging technologies for routine geoenvironmental risk domaining using drill core. Here we emphasize that integrating these techniques potentially enables best practice ARD management at the beginning of the life-of-mine cycle by allowing early forecasting of the geoenvironmental properties of future wastes.