eCite Digital Repository

Identifying common protoliths for altered rocks using clustering and classification of geochemical data at the Minto Copper-Gold Mine, Yukon, Canada

Citation

Hood, S and Cracknell, M, Identifying common protoliths for altered rocks using clustering and classification of geochemical data at the Minto Copper-Gold Mine, Yukon, Canada, Extended Abstracts of Gold' 17, 21-23 February, Rotorua, New Zealand, pp. 52-56. ISBN 978-1-876118-01-3 (2017) [Refereed Conference Paper]

Copyright Statement

Copyright 2017 Australian Institute of Geoscientists

Official URL: http://www.sjsresource.com.au/BookingRetrieve.aspx...

Abstract

Rocks in metalliferous ore deposits are typically classified as either ore or waste rock. An alternative distinction, more helpful for discriminating alteration haloes around ore, is protolith rock and altered rock. Metasomatic alteration occurs during hydrothermal ore deposition when mineralising fluids are in disequilibrium with surrounding host rocks. Geologists in brownfields areas commonly seek to interpret ore body geometry and surrounding alteration halos using the spatial distribution of whole rock chemical analyses. However, modelling the relative highs and lows in geochemical data does not give a complete representation of alteration geometry; weight percent values provide no direct information about the nature and magnitude of mass transfer during hydrothermal alteration. The compositional data of rocks are inherently multivariate and the concentrations of all constituents are related to each other by the process of closure. For example, in a simplified three element system, reducing one element will lead to an apparent enrichment of the other two.

Here we present a workflow to process a whole rock geochemical dataset and produce labels for groups of similar least-altered protolith rocks, and then apply these labels to groups of altered rocks. This method involves quick means of exploratory data analysis and lends itself to iterative refinement, relying on a geologist to use their knowledge of a particular site.

Item Details

Item Type:Refereed Conference Paper
Keywords:supervised classification, unsupervised clustering, geochemistry
Research Division:Information and Computing Sciences
Research Group:Artificial Intelligence and Image Processing
Research Field:Pattern Recognition and Data Mining
Objective Division:Mineral Resources (excl. Energy Resources)
Objective Group:Mineral Exploration
Objective Field:Precious (Noble) Metal Ore Exploration
Author:Hood, S (Mr Shawn Hood)
Author:Cracknell, M (Dr Matthew Cracknell)
ID Code:120200
Year Published:2017
Deposited By:CODES ARC
Deposited On:2017-08-15
Last Modified:2017-09-28
Downloads:0

Repository Staff Only: item control page