GEOSCAN Search Results: Fastlink

GEOSCAN Menu


TitleGeochemical characterization of the Central Mineral Belt U ± Cu ± Mo ± V mineralization, Labrador, Canada: application of unsupervised machine-learning for evaluation of IOCG and affiliated mineral potential
 
AuthorAcosta-Gongora, P; Potter, E GORCID logo; Lawley, C J MORCID logo; Corriveau, LORCID logo; Sparkes, G
SourceJournal of Geochemical Exploration vol. 237, 106995, 2022 p. 1-21, https://doi.org/10.1016/j.gexplo.2022.106995 Open Access logo Open Access
Year2022
Alt SeriesNatural Resources Canada, Contribution Series 20210478
PublisherElsevier
Documentserial
Lang.English
Mediapaper; on-line; digital
File formatpdf; html
ProvinceNewfoundland and Labrador
AreaLabrador; Canada
Lat/Long WENS -61.3333 -59.3333 55.5000 54.0000
Subjectsgeochemistry; Science and Technology; uranium; Central Mineral Belt
Illustrationslocation maps; photographs; diagrams; cross-plots
ProgramTargeted Geoscience Initiative (TGI-6) Digital Geoscience and Method Development Project
Released2022 04 04
AbstractThe Central Mineral Belt (CMB) in Labrador, Canada, hosts multiple U (±base±precious metal) showings, prospects and deposits in metamorphosed and variably hydrothermally altered Neoarchean to Mesoproterozoic, igneous and sedimentary rocks. Previous work has recognized U mineralization locally associated with Fe-Ca and alkali metasomatism typical of metasomatic iron oxide and alkali-calcic alteration systems (IOAA) that host iron oxide-copper-gold (IOCG) and affiliated primary critical metal deposits. However, to date little is known on the type, extent and temporal or genetic relationships between the diverse Fe, Ca and alkali metasomatism and the regionally distributed U mineralization. Combined unsupervised machine-learning and classification of alteration from a large geochemical dataset distinguish the main alteration phases in the CMB, identify compositional changes related to U mineralization, and infer lithological/mineralogical information from samples with censored, limited and/or inaccurate metadata. Weak to intense Na and Na+Ca-Fe (Mg) metasomatism in the southwest (Two-Time and Moran Lake areas) and eastern (Michelin area) portions of the CMB pre-dates U mineralization and Fe-oxide breccia development, similar to albitite-hosted U and IOCG deposits globally. For the first time, we report the occurrence of intense Na-alteration in the Two Time deposit. Classic REE and spider diagrams highlight preservation and disruption of protolith primary signatures of normally immobile elements. Principal component and cluster analysis indicate significant variations in Fe-Mg±Na contents in the rocks from combinations of Na, Ca, Fe, and Mg-rich alteration while protolith REE signatures can be locally preserved even after pervasive albitization-hematization. Cluster analysis identifies mineralized felsic and mafic rocks in the Michelin deposit and Moran Lake area, facilitating inference of relevant lithological/mineralogical information from samples lacking or with limited meta-data. The methods outlined provide rapid and relatively inexpensive means to optimize identification of mineral systems within large geochemical datasets, verify drill core or field observations, highlight potentially overlooked alteration and refine economic mineral potential assessment. Based on our results and previous work, we suggest the mineral potential of the southwestern and eastern CMB needs to be re-assessed with modern exploration models for IOAA ore systems and their iron oxide-poor variants.
Summary(Plain Language Summary, not published)
The Targeted Geoscience Initiative (TGI) is a collaborative federal geoscience program that provides industry with the next generation of geoscience knowledge and innovative techniques to better detect buried mineral deposits, thereby reducing some of the risks of exploration. This contribution applies unsupervised machine-learning and geochemical classifications to a large, public geochemical dataset to distinguish the main alteration types in the Central Mineral Belt of Labrador, identify compositional changes related to uranium mineralization, and infer lithological/mineralogical information from samples with censored, limited and/or inaccurate metadata.
GEOSCAN ID329342

 
Date modified: