- Prof. Eric Grunsky, Department of Earth and Environmental Sciences,University of Waterloo, Waterloo, Ontario, Canada
- Date: Jun, 10, 2015 12:00 am
- Place: Sala d'Actes del Institut de Ciències de la Terra Jaume Almera(ICTJA)
- Location: C/ Solé i Sabarís s/n, Barcelona
- Contact: Neus Otero (UB)
Multi-element geochemistry surveys of rocks, soils, stream/lake sediments, regolith and weathered materials are usually carried out by government surveys and mineral exploration companies at continental, regional and local scales. The chemistry of these materials is defined by minerals and their subsequent modification by comminution, dissolution and weathering. A database of geochemistry, with up to 50 elements and sufficiently low detection limits, represents a multi-dimensional geochemical space that can be studied using multivariate statistical methods from which patterns are described (process discovery) reflecting geochemical/geological processes. These patterns form the basis from which predictive maps, using probabilities are created (process validation). When assembling large regional datasets from various sources, it is imperative to understand the nature of the sample media, methods of sample preparation, laboratory digestion procedures and analytical instrumentation methods. Problems that are typically associated with the interpretation of multi-element geochemical data include the problem of closure, missing values, censoring and adequate spatial sample design.Effective analysis and interpretation of the information that exists in geochemical datasets can be obtained from the use of multivariate statistical procedures and spatial mapping of the results. Geochemical and geological processes can often be recognized through the use of data discovery procedures such as the application of principal component analysis. The use of classification and predictive procedures can be used to confirm lithologic variability, mineral ore zone alteration and zones of mineralization. Studies of multi-element geochemical surveys of lake/till sediments from Canada, and soils from Australia and the USA show that predictive maps of bedrock, buried tectonic assemblages, climate, soil humidity and ecosystems can be identified. Kimberlite phase discrimination is demonstrated using drill hole lithogeochemistry from the Star kimberlite field in Saskatchewan, Canada.