Global to local scale geographic information system processing strategies for targeting orogenic gold deposits

[Truncated] A major challenge in modern mineral exploration is to identify critical relationships between geological characteristics and economically significant mineral deposits, and to apply these in conceptual targeting. With increasingly global exploration, it is important that GIS tools for prospectivity analysis apply to all scales of exploration from global, craton, regional to local. This allows spatial controls of mineralisation to be better defined, at these scales and hence improve our exploration criteria. To achieve this, a number of parameters have to be integrated to produce the best targeting methodology for each scale. This study examines the effectiveness of a variety of GIS methods and exploration criteria for orogenic gold at different scales.

The geological features used to estimate prospectivity were identified using a new approach to modelling GIS fuzzy membership functions and economic deposit target sizes to create and test input GIS layers. This approach was found to be effective for global- to regional-scale targeting strategies, where sufficient numbers of deposits are available for statistical analysis of spatial relationships between geological map features and sites of known deposits. This study resulted in the development of; 1) a robust measure of the strength of GIS input layers based on both deposit frequency and total contained gold (TCG), 2) a methodology for incorporating economic criteria e.g., 31 t TCG (1 Moz Au) in the conversion of raw data to fuzzy membership values for GIS input layers, 3) an approach to identifying the relationship between the size of gold deposits at a regional-scale, based on a variety of GIS deposit layers, i.e., a) all deposits (dep-all), b) deposits < 6 t TCG (dep-min), c) deposits > 6 t TCG, with deposits > 31 t TCG reduced to 31 t TCG (dep-cut), d) deposits > 18 t TCG (dep- max), 4) an improved method of calculating the capture efficiency ratios that quantifythe predictive power of candidate fuzzy membership GIS input layers, and 5) abinominal statistical test to determine confidence levels for the significance of spatial relationships based on both deposit frequency and TCG.