Multivariate geostatistical methods to identify and map spatial variations of soil heavy metals

Abstract.

This study characterized and mapped the spatial variability patterns of seven soil heavy metals (Cd, Cr, Cu, Hg, Ni, Pb and Zn) for further environmental monitoring, remediation and planning by multivariate geostatistical methods, to delineate sources of soil heavy metals in a rice paddy field in the north of Changhua County, Taiwan. The variograms and cross-variograms of soil heavy metals were modeled in three spatial structures which included the nugget effect and two spherical models with ranges of 288 and 1,200 m for short-range and long-range spatial scales, respectively. Multivariate spatial variations of soil heavy metals can be detected using factor analysis based on coregionalized coefficient matrices for different scales. Moreover, the multivariate geostatistical analysis results indicated that factor loadings of the regionalized factors of soil heavy metals changed depending on the spatial scale. The loadings of each factor became stable for distances larger than 1,200 m. The maps of spatial components of soil heavy metals reveal that the short-range variation sources of Cd, Cr, Cu, Ni and Zn were highly correlated to the industrial plants in this study area. The long-range variations of Cd, Cr, Cu, Ni and Zn strongly corresponded to the locations of irrigation systems and industrial plants.