Multivariate Statistical Interpretation Of Soil Quality Data In The Context Of Public Health

摘要

In this study, the distribution patterns of physical (sand%, loam%, clay%, pH and organic matter‰) and chemical (calcium carbonate (CaCO_3), sodium (Na~+), magnesium (Mg~(+2)), phosphorus (P), boron (B), iron (Fe), manganese (Mn) and zinc (Zn)) parameters of soil commonly assayed in analytical chemistry laboratories, are reported. Three representative areas (lowland (LL), semi-mountainous (SM), and coastal (C)) and 170 sample sites were selected, 510 soil samples were collected and analyzed for a 3-years period (2002-2004), in the prefecture of Larissa, Thessaly, central Greece. In each studied site, the kind of cultivation is also reported. Soils in LL area of the prefecture of Larissa were found to be rich in organic matter, Mg, B, but poor in Fe, Mn and Zn, when compared to the other two sites. The studied SM area of Larissa's prefecture constituted from clayey soils showed big concentrations of Mg, Zn, high value of clay and low value of sand, when compared to the other two studied areas. The studied C area is characterized by the sandy soils near the coastal areas of the prefecture of Larissa. In each studied area, the parameters are principally separated into two main clusters, each of them divided additionally into sub-clusters with slight differences. Six latent factors explain 78.1%, 74.3%, and 74.6% of the total variance of the LL-site, SM-site, and C-site, respectively, which is an indication for the factor analysis model adequacy.