Mineralogical and geochemical characterization of mine tailings in the King river delta, Western Tasmania: Implications for long-term stability of trace elements

Nearly a century of mining activities at the Mount Lyell Cu-Zu-Au mine has resulted in the accumulation of ∼14 Mt of fine-grained tailings at the mouth of the King River, Western Tasmania. The tailings consist primarily of sulfides ( i.e., pyrite, chalcopyrite, molybdenite, galena), silicates, and amorphous phases such as Fe-oxyhydroxides and sulphates. The long-term weathering of these anthropogenic-deltaic sediments has influenced the geochemical distribution and mineralogical hosts of trace metal(loid)s that are of both environmental concern and economic interest (i.e., Cu, Co, Au, Ag, Mo, As). In this study, the mineralogy, element deportment, and metal(loid) enrichment of variably altered delta sediments were investigated to understand how natural weathering processes influenced the environmental mobility and amenability to reprocessing these waste products. Micro-scale compositional and textural domains were assessed using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), where metal(loid) relationships revealed a complex multi-stage diagenesis within the sediment. Generation of grain-scale geochemical domains through machine learning cluster analysis support the interpretation of in-situ mineralogical growth zones that influence the distribution of trace metal(loid)s. Eight clusters were identified by their unique mineralogical and chemical composition, revealing that metal(loids), initially bound to primary sulfides, have been remobilized and sequestered by authigenic material. This study demonstrates that the complex hydrodynamic and geochemical processes within the delta continue to influence the mobility and long-term stability of legacy contaminants within this deposit. Results demonstrate that machine learning cluster analysis provides a robust tool to understand the distribution of trace metal(loid)s within unconsolidated mining-impacted sediment that have been altered by complex hydrodynamic and geochemical processes. Knowledge generated in this study is relevant for assessing the feasibility of various remediation strategies, including reprocessing of tailings as an alternative management option. • Machine learning cluster analysis aided on the mineralogy and geochemical characterization of tailings. • Metal(loid)s have been mobilized from sulfides and sequestered by authigenic materials. • Mineral sinks for metal(loid)s were Fe-oxyhydroxides, Fe-sulphates and Si-FeO x . • Decades of weathering have altered the geochemical stability of mine tailings.