Advantages and limitations of chemical extraction tests to predict mercury soil-plant transfer in soil risk evaluations.

In this study, we compared the size of the mobile Hg pool in soil to those obtained by extractions using 2 M HNO3, 5 M HNO3, and 2 M HCl. This was done to evaluate their suitability to be used as proxies in view of Hg uptake by ryegrass. Total levels of Hg in soil ranged from 0.66 to 70 mg kg(-1) (median 17 mg kg(-1)), and concentrations of Hg extracted increased in the order: mobile Hg < 2 M HNO3 < 5 M HNO3 < 2 M HCl. The percentage of Hg extracted relative to total Hg in soil varied from 0.13 to 0.79 % (for the mobile pool) to 4.8-82 % (for 2 M HCl). Levels of Hg in ryegrass ranged from 0.060 to 36 mg kg(-1) (median 0.65 mg kg(-1), in roots) and from 0.040 to 5.4 mg kg(-1) (median 0.34 mg kg(-1), in shoots). Although results from the 2 M HNO3 extraction appeared to the most comparable to the actual total Hg levels measured in plants, the 2 M HCl extraction better expressed the variation in plant pools. In general, soil tests explained between 66 and 86 % of the variability of Hg contents in ryegrass shoots. Results indicated that all methods tested here can be used to estimate the plant total Hg pool at contaminated areas and can be used in first tier soil risk evaluations. This study also indicates that a relevant part of Hg in plants is from deposition of soil particles and that splashing of soil can be more significant for plant contamination than actual uptake processes. Graphical Abstract Illustration of potential mercury soil-plant transfer routes.