Defining and mapping hydrothermal footprints at the BIF-hosted Meliadine gold district, Nunavut, Canada

• Multivariate hydrothermal footprints are defined at BIF-hosted gold deposits.
• We apply a conditional probability-based model to map fuzzy footprints.
• pXRF spectrometry is sufficient to define and map multivariate fuzzy footprints.

The geochemical and mineralogic signature, or hydrothermal footprint, at Banded Iron Formation (BIF)-hosted gold deposits has great potential as an exploration tool at district- to deposit-scales. The Meliadine Gold District (MGD) is one of Canada's largest emerging BIF-hosted gold districts (2.8 Moz contained Au in reserves, plus an indicated and inferred resource of 5.8 Moz Au). Rocks are variably altered (silicified ± sulphidized ± sericitized ± carbonatized ± chloritized) adjacent to BIF-hosted replacement-style gold mineralization and auriferous greenstone-hosted quartz (± ankerite) veins cutting mafic volcanic rocks, interflow volcaniclastic rocks and turbiditic successions. Hydrothermal altered and veined rocks provide a visual guide to ore and are spatially associated with anomalous pathfinder element concentrations (Au–As–Te–Bi–Sb). These geochemical anomalies are typically defined using a preferred threshold concentration for each element, or ratio, of interest. However, the conventional approach inadequately accounts for the multivariate nature of ore signatures and the inherently imprecise boundary between barren and mineralized rock. Herein favourable pathfinder element enrichment and hydrothermal alteration intensity are integrated and mapped using a conditional probability-based model in an effort to further highlight the complementary nature of multivariate datasets and to define fuzzy footprints. Key hydrothermal alteration mineral and element assemblages associated with gold are enriched from 10s to 100s of metres adjacent to ore zones and provide a vector to gold ore. We demonstrate that the accuracy and precision of portable X-ray fluorescence (pXRF) spectrometry on drill core surfaces is sufficient to map these multivariate and fuzzy hydrothermal footprints from the rock record, but at a fraction of the cost and time compared to conventional whole-rock analyses.