Zinc phytoavailability after remediation in soils contaminated by sphalerite-containing pyritic sludge

Zinc can be toxic to plants growing on soils in areas of the Guadiamar River valley (southwestern Spain) affected by the spillage of pyritic sludge in April 1998. The shoots and the soil around the roots of two wild plants (viz. Amaranthus blitoides S. Wats., November 2000; and Xanthium strumarium L., June 2001) growing in the sludge-affected areas were sampled with the purpose of relating Zn phytoavailability to soil properties. The soils were calcareous and non-calcareous Entisols and Inceptisols which, after remediation, contained ploughed-in residual sludge and unevenly distributed industrial lime. Chemical extracts from the soils suggested that much of the sphalerite (ZnS) originally present in the sludge had weathered and Zn was partly bound to carbonates and Fe oxides, the total Zn concentration ranging from 37 to 2407 mg kg ^–1. To identify the soil properties that influenced Zn phytoavailability under controlled conditions, the soil samples (n=63) were homogenized and oilseed rape (Brassica napus var. Karola) was pot-grown on them in a growth chamber. The concentrations of Zn in oilseed rape shoots and roots were below phytotoxic levels, with mean ± standard deviation values of 142 ± 128 and 244 ± 328 mg kg ^–1 dry matter, respectively. Citrate/bicarbonate-extractable Zn in soil (Zn _cb) was found to be the best predictor for the Zn concentration in both shoots and roots. Also, the Zn _cb/Olsen P ratio exhibited a high predictive power for Zn in shoots as the likely result of the Zn-P interaction in soil. The shoot Zn concentration in the wild plants, generally lay below phytotoxic levels (the mean ± standard deviation values were 261 ± 255 and 200 ± 228 mg kg ^–1 dry matter for Amaranthus blitoides and Xanthium strumarium, respectively) and was not correlated with soil properties – by exception, there was slight correlation between the Zn concentration in Amaranthus blitoides and Zn _cb/Olsen P. Such a lack of correlation can be ascribed to the local small-scale soil heterogeneity caused by remediation practices. The Zn concentration in wild plants growing on CaCO ₃-poor soils was weakly correlated with Zn _cb/Olsen P; no similar correlation was found in CaCO ₃-rich soils, however. The wild plants growing on CaCO ₃-poor and CaCO ₃-rich soils differed little in Zn concentration; this suggests that further addition of lime to reduce Zn phytoavailability may be unjustified.