Use of multivariate statistical analysis to determine the relationship between the magnetic properties of urban topsoil and its metal, S, and Br content

Multivariate statistical analysis has been used for detailed examination of the relationship between the magnetic properties of Xuzhou urban topsoil, for example concentration-dependent properties (mass magnetic susceptibility (χ), susceptibility of anhysteretic remanent magnetization (χ _ARM), saturation isothermal remanent magnetization (SIRM), soft remanent magnetization (SOFT), and frequency-dependent magnetic susceptibility (χ _FD)) and feature-dependent properties (S _{−100 mT} ratio, SIRM/χ ratio and F _{300 mT} ratio), and the concentrations of metals (Ti, Fe, Cr, Al, Ga, Pb, Sc, Ba, Li, Cd, Be, Co, Cu, Mn, Ni, V, Zn, Mo, Pt, Pd, Au, As, Sb, Se, Hg, Bi, Ag, and Sn), S, and Br in the soil. The results show that SIRM/χ ratios correlate best with the heavy metals (Hg, Cr, Sb, As, and Bi) which are mainly derived from coal-combustion emissions whereas χ _FD correlates best with the metals (Al, Ti, V, Be, Co, Ga, Mn, and Li) which principally originate from soil parents. Concentration-dependent magnetic properties (χ _ARM, χ, SIRM, and SOFT) correlate well with elements (Se, Pb, Cu, Zn, Fe, Ag, Sc, Ba, Mo, Br, S, Cd, Ni, etc.) which are mainly derived from road-traffic emissions. For the same chemical element, χ _ARM, SIRM, and SOFT values are frequently better correlated than χ values, and χ _ARM values are the best indicators of the concentrations of these elements associated with traffic emissions in this study area. In addition, S _{−100 mT} ratios significantly correlate positively with Se, Sc, Pb, Cu, Zn, Mo, and S whereas F _{300 mT} ratios only correlate positively with Pt and negatively with Fe. These results confirm the suitability of different magnetic properties for characterizing the concentrations of heavy metals, S, and Br in Xuzhou urban topsoil.