A Continent-Scale GIS-Based Assessment of the Distribution and Potential for Sandstone-Hosted Uranium Deposits

This chapter describes the steps required to design and construct a GIS-based fuzzy logic mineral prospectivity model for sandstone-hosted uranium deposits at a continental scale. The example of Australia (land mass ~7.69M km²) is used, due to a relatively high density of pre-competitive data available for this continent, as well as its hosting a considerable number of economically significant deposits with >5,000 tonnes of contained U₃O₈ resources (e.g., Beverley, Four Mile, Manyingee, Oobagooma, Mulga Rock). The output generated from the fuzzy logic model documented in this study successfully 'rediscovers' proven sandstone-hosted uranium provinces and confirms the veracity of the current exploration targeting model for this type of uranium deposit. In addition, the result highlights several favorable geological regions within 'geologically permissive tracts' that should be prospective for as yet undiscovered sandstone-hosted uranium deposits. We demonstrate the importance of suitable input layers (i.e., scale-dependence of mappable proxies, relevance to the formation of the deposit type sought), re-classification of pertinent data into appropriate groupings, and the development of a meaningful inference network that combines individual input layers to produce a realistic result. The methodology and approach employed herein is readily transferrable to other continents or geological provinces of interest, provided a-priori assessments of suitable input data and their limitations form an integral part of the development of the fuzzy logic mineral prospectivity model. If done correctly, GIS-based fuzzy logic mineral prospectivity modeling provides an extremely powerful visualization and decision-making tool for the explorationist. (author)