Plant phosphorus availability index in rehabilitated bauxite-processing residue sand

Abstract

Background and aims Soil phosphorus (P) indices that have been originally developed and applied to agricultural soils for predicting P uptake by plants were examined in a pot experiment to determine the most suitable index for P availability in bauxite-processing residue sand (BRS). Methods Pot trials with ryegrass were established using BRS that had been amended with various organic (greenwaste compost, biochar and biosolids) and inorganic (zeolite) materials and different levels of di-ammonium phosphate fertiliser. Soil P availability indices tested included anion-exchange membrane (AEM-P), 0.01 M calcium chloride (CaCl2-P), Colwell-P, and Mehlich 3-P. Results AEM-P was found to most closely reflect the available P status in BRS across all treatments, and had the strongest associations with plant P uptake compared to Colwell-P, Mehlich 3-P and CaCl2-P. AEM-P was more closely correlated with P uptake by ryegrass than other P indices, while Colwell-P was closely related to leaf dry matter. Interestingly, a strong inverse relationship between plant indices and pH in BRS growth media was observed, and an adequate level of plant P uptake was found only in 15 year-old rehabilitated BRS with pH <?8.0. Conclusions AEM-P was found to be the most suitable index for evaluating P availability in highly alkaline BRS and pH was an important parameter affecting uptake of P by ryegrass. Importantly, time is required (> 5 years) before improved uptake of P by plants can be observed in rehabilitated residue sand embankments.