地理情報学と地球統計学: 概要

Statistical Evaluation of Soil Geochemical Data from Dagbala-Atte District in Igarra Schist Belt, Southwestern Nigeria

Adepoju MO, Adekoya JA, Odeyemi IB

The results of 27 trace elements, namely, Ag, As, Au, Ba, Bi, Cd, Co, Cr, Cu, Fe, Ga, Hg, La, Mn, Mo, Ni, Pb, Sb, Sc, Se, Sr, Te, Th, Tl, U, V and Zn in 49 soil samples collected on a uniform 3 km × 2 km grid pattern over Dagbala-Atte District were subjected to some univariate and multivariate statistical evaluation. The univariate methods used include frequency distribution and dispersion boxm plots while the multivariate methods employed are correlation and factor analyses. Frequency distribution and dispersion box plots revealed that most elements showed mixtures of populations and are log-normally distributed in varying degrees at 95% confidence level. Eventually, the univariate statistics was applied to compute the threshold values for all the elements in the District, whichm allowed the determination of anomalous concentration for each element. Pearson correlation coefficients indicate fair to very good correlation between pairs of elements. Four factors, similar to the subjective groupings obtained from the correlation matrix, were derived from R-mode factor analysis. This four-factor model has been interpreted as reflecting underlying lithology, potential mineralization and physico-chemical conditions in the environment. Results of a study of the correlation matrix and factor analysis of the soil geochemical data from Dagbala-Atte District indicate that most of the anomalous values of the elements could be attributed to mineralization such as U-Th in rare-metal pegmatite in the northeastern part of the district underlain by granitic-gneiss, Cu-Ni mineralization associated with minor Au at the boundary between the granitic gneiss and the metasediments, and base metal (Pb- Zn) and Ag mineralization, possibly in form of minor veins, in the southwestern part of the district underlain by metaconglomerate.