Assessing the fractionation and bioavailability of heavy metals in soil–rice system and the associated health risk

This study developed a method to build relationships between chemical fractionations of heavy metals in soils and their accumulations in rice and estimate the respective contribution of each geochemical speciation in the soils from the Yangtze River Delta, China. In contaminated areas, residue and humic acid-bound fractions in soils were the main phases for most heavy metals. The mobility of heavy metals was in this following order: Cd > Pb ≈ Zn > Ni > As ≈ Cr > Hg. Transfer factors calculated by the ratios of specific fractionations of heavy metals in the soil–rice system were used to assess the capability of different metal speciation transfer from soil to rice. The carbonate and Fe/Mn oxyhydroxides bound phase had significant positive correlations with total metal concentrations in rice. Hg uptake by rice might be related to the exchangeable and carbonate-bound fractions of soil Hg. Results of PCA analysis of transfer factors estimated that the labile fractions (i.e. water soluble, exchangeable and carbonate bound) contributed more than 40% of the heavy metal accumulations in rice. Effect of organic matter and residue fraction on metals transfer was estimated to be ~ 25 to ~ 30% while contribution of humic acid and Fe/Mn oxyhydroxides-bound fractions was estimated to be ~ 20 to ~ 30%. Modified risk assessment code (mRAC) and ecological contamination index (ECI) confirmed that the soil samples were polluted by heavy metals. Soil Cd contributed more than 80% of mRAC. Contrarily, the main contributors to ECI were identified as As, Hg, Pb and Zn. The average values of total target hazard quotient (TTHQ) and Risk total were above 1 and 10 –4 respectively, implying people living in the study area were exposed to both non-carcinogenic and carcinogenic risk. As and Pb were the main contributor to high TTHQ value while As, Cd and Cr in rice contributed mostly to Risk total value. Spatial changes of ecological risk indexes and human health risk indexes showed that the samples with high TTHQ values distributed in the area with high values of mRAC. Likewise, the area with high ECI values and with high carcinogenic risk overlapped.