Application of Multivariate Statistical Analysis on Soil Multi-Elements Geochemical Data of the Wassa Gold Project Western Region, Ghana
No Thumbnail Available
Date
2014-10
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Ghana
Abstract
The main aim of this dissertation was to identify the geochemical element associations, pathfinder elements for gold and delineate gold targets in the Wassa project area. Multivariate statistical techniques using Principal Components (PCA) and Hierarchical Cluster (HCA) analyses techniques were used to study the chemical characteristics of three hundred and eighty eight (388) soil samples which were each analysed for 30 elements. Descriptive statistics, correlation matrices, probability plots, and factor score models were used in the identification of element associations and target generation. Soil samples were taken from the B –horizon of the soil profile of the study area and analysed for Au by conventional fire assay-atomic absorption spectrometry (FA-AAS) and ICP-MS (inductively coupled mass spectrometry) for the trace elements/base metals (As, Cu, Ni, Pb, Zn, Ag, Al, Ba, Bi, Ce, C0,Cr, Ga, Hf, K, Mg, Mn, P,S, Sb, Ti, Zr, Na, Be, Ge, Li, Hg, B and Fe). Data reduction was achieved by means of Principal Component analysis, factor analysis and cluster analysis. Principal Component analysis identified four geochemical associations within the dataset. The first factor (PCA 1) which accounted for 37.65% of the total variance was characterized by very high loadings of Ag, Al, Bi, Cr, Ga, Hf, Zr, Hg, Pb and Fe, which are linked to the underlying felsic intrusive rocks such as granodiorite, tonalite and quartz-feldspar porphyry. The second factor (PCA 2) which accounted for 22.83% was characterized by high loadings of Cu, Ni, Zn, Ba, Ce, Co, Mg, and Mn, which are linked to the underlying mafic rocks such as basalts and diorites. The third factor (PCA 3) which accounted for 10.35% is characterised by high loadings of As, Ni, K and Sb, which are linked to metasedimentary rocks such as schist and phyllites. The fourth factor (PCA 4) which is the gold factor also accounted for 5.90% of the total variance and was characterized by Au with positive contribution from Ag, Zn, Mn, Fe, As, Sb and Hg.
Cluster Analysis also reduced the whole data to three clusters. The second cluster or the gold group consisted of Au, As, K and Sb. Pathfinders for gold in the area under investigation
were identified as Au, Fe, Mn, Ag, Zn, As, Sb, and Hg. Six gold targets were identified, which were in a northwest-Southeast trend (NW-SE) of the Wassa project area
Description
Thesis (MSc) - University of Ghana, 2014