Statistical analysis of stream sediment geochemical data from Oyi drainage systems, Western Nigeria

Abstract

Sixty-one stream sediment samples were collected at sampling intervals that varied between 1.5 km and 2.0 km along the channels of River Oyi and its tributaries. The samples were air-dried, disaggregated and sieved to obtain the minus 80 mesh (177 microns) fraction. Half a gram of each of the sieved samples was digested with aqua regia, and then analysed for twenty-four elements using Inductively Coupled Plasma Mass Spectroscopy (ICP-MS). The results of concentrations of twenty-four elements treated with both univariate and multivariate statistical analytical techniques revealed that all the elements analyzed except Co, Cr, Fe and V had log-normal distributions. The cumulative probability plots of the elements showed that Mn and Cu consisted of one population. Ni, Pb, Ag, La, Th, Co, Sc, Ga, Cr, Tl, Au, Bi, U, Sr, Fe, and Zn had two populations while the remaining elements were made up of three populations. The Simple Person correlation analysis revealed that there were strong positive correlations between the pairs of the following elements: Fe-Mn-Ni-Co-Sc-Tl-Ba-VMo; La-Pb-Th; Zn-Cu-Ag-Au-Sb; Bi-Cr and Cr-La, with 'r' values ranging from -0.74 to 0.98. These correlations were significant at 95% and above confidence level. R-mode varimax factor analysis produced a Four-Factor model, accounting for 90.08% of the data variability with the following metal associations: Ga-Fe-Ni-Co-Sc- Mn-Tl-Ba-V-Mo-Zn-Cu-Ag-Au; Pb-La-Sr-Th-U-Cd; Cu-Ag-Au-Sb and Bi-Cr. The results of the statistical analyses suggested the occurrence of potential mineralization containing Cu-Zn-Ag-Au in the gneisses of the study area.

Keywords: Stream Sediment, Multivariate Analysis, Cumulative Probability, Precision.