Comparison of EDTA and sequential extraction tests for phytoavailability prediction of manganese and zinc in agricultural alkaline soils

Twenty-eight soil samples with a wide range of chemical and physical properties were collected from typical uncontaminated soils of Central Spain, characterised in general by their medium or high calcium carbonate contents. The extractability and solid-phase fractionation of manganese (Mn) and zinc (Zn) in the soils was determined by the single extraction procedure proposed by the BCR for calcareous soils (EDTA extractability) and two different sequential extraction procedures: a modification of the five-step Tessier method (procedure A) and the three-step BCR method (procedure B). The EDTA extractable Mn levels were relatively high (on average 30%), but they were less in the case of Zn (on average 3.7%). Manganese was mainly released in the residual fraction when using procedure A (47%) and in the reducible fraction when using procedure B (47%). Zinc in these soils was dominantly associated with mineral lattices (procedure A: 79%; procedure B: 91%). The fractionation methods for the prediction of phytoavailability of Mn and Zn for barley in a greenhouse experiment were not significantly better than EDTA plus soil properties. Inclusion of soil properties in the best-fitting multivariate regression models improved the relationship for metal concentration in plants. Values for R2 of the models ranged from 0.37 to 0.95.