The effectiveness of strategies to reduce cadmium (Cd) availability for crop uptake can be assessed using various measures of Cd availability, such as Cd concentration in pore water and Cd extracted with salt solutions. This study evaluated the performance of multi-surface modelling (MSM) to predict dissolved Cd in two Irish tillage soils treated with lime, zinc (Zn) and spent mushroom compost (SMC). Predictions were assessed against Cd measured in three solution media, i.e., 1 mM Ca(NO3)2 and 0.1 M CaCl2 extractions, as well as in soil pore water. Results indicate that reactive soil organic matter (SOM) may be underestimated using a single 0.1 M NaOH extraction in the investigated soils, leading to substantial overestimation of dissolved Cd by MSM, particularly at higher pH. Repeating the 0.1 M NaOH extraction three times substantially improved model predictions. Additionally, using reactive Cd determined by isotopic dilution instead of 0.43 M HNO3 improved model predictions for one of the soils that was rich in manganese (Mn) oxides, revealing a possible role of Mn oxides in determining the reactive Cd fraction. After optimizing reactive SOM and reactive Cd, residuals between predicted and measured Cd showed an increasing trend along with increasing solution pH and decreasing dissolved Cd. This is likely related to Cd binding to high affinity sites due to uncertainties in the binding parameters for these sites and/or to slow desorption kinetics for Cd bound to these sites. Despite significant variations in solution properties, including higher dissolved Ca and reactive dissolved organic matter (DOM), 1 mM Ca(NO3)2 extracts exhibited similar extractable Cd levels, model performance, and Cd speciation compared to soil pore water especially at higher pH. Thus, 1 mM Ca(NO3)2 can be a reliable proxy for soil pore water in assessing Cd availability for crop uptake in soils with circumneutral to alkaline pH.
Read full abstract