Improving phosphorus availability in Andisols and Oxisols

Abstract

Low phosphorus (P) availability limits plant growth in many soils, particularly in Andisols and Oxisols, due to their large content of minerals that strongly sorb P (e.g. Al/Fe oxyhydroxides, allophane). Because of the strong P retention, P fertilizer requirements are high in these soils. Strategies to increase the efficiency of P fertilizers – and reduce P rates needed to obtain maximal yield – remain key to reducing the pressure on limited rock phosphate reserves. To develop management practices or fertilizer formulations that enhance P availability and fertilizer efficiency in strongly P-sorbing soils, a better understanding of the chemical reactions of P in these soils is needed. This work aimed (i) to examine the chemical behaviour of soil P and added P to plant uptake in strongly P-sorbing soils and (ii) to compare the effect of different P fertilizer types (granular/fluid/nano-sized) as a strategy to increase the efficiency of P fertilizers. A laboratory incubation experiment was conducted to evaluate the diffusion and lability of P from granular and fluid fertilizers applied to Andisols and Oxisols using the isotopic dilution technique and a novel visualization method. In all soils, fluid fertilizers enhanced P diffusion, but not P lability, i.e. the amount of added P that remained in isotopically exchangeable form. In the Oxisols, a greater percentage of added P remained isotopically exchangeable when added as granular monoammonium phosphate (MAP) (41% labile) than when added as fluid MAP (25% labile). In the Andisols, no significant difference was observed in the percentage of labile P between both fertilizer types (circa 25% labile). Given these results, it was hypothesized that there would be no agronomic benefit from the application of fluid P fertilizer in these soils. A subsequent pot trial was conducted to assess the uptake of P by wheat (Triticum aestivum) from granular and fluid fertilizers using the indirect isotopic dilution method in two Andisols, two Oxisols, and a calcareous soil (where fluid P has been proven more effective). This pot trial indeed showed no significant difference in dry matter yield, P uptake and the percentage of P derived from the fertilizer in the plant (%Pdff) between granular and fluid MAP in the Andisols or Oxisols, while there was a significant increase with fluid fertilizer in the calcareous soil. Hydroxyapatite nanoparticles (Ca₁₀(PO₄)₆(OH)₂, n-HAP) were also tested as a potential P fertilizer, based on the hypothesis that nano-sized particles can potentially move in the soil and reach the plant roots through the transpiration flow. Because of the strong adsorption and subsequent fixation of soluble P in this type of soils, nanoparticulate P could potentially have a benefit over soluble fertilizers. Column studies showed some leaching (5%) of n-HAP in the Andisol but very little in the Oxisol. In contrast, bulk-sized HAP did not move in either of the soils. A pot trial using the isotopic dilution procedure evaluated P availability for wheat from n-HAP, bulk-sized HAP, and triple superphosphate. For Andisols and Oxisols, P uptake and %Pdff differed significantly from P treatments as follows: TSP > n-HAP > bulk-HAP. Thus, while sparingly-soluble fertilizer in nanoparticulate form (n-HAP) performed better than its bulk counterpart, it was less efficient than soluble fertilizer (TSP). It was hypothesized that the difference between n-HAP and bulk-HAP was due to the difference in rate of dissolution, but that the n-HAP has no direct effect on the uptake and only contributes via dissolution. The pot trial showed that n-HAP did not have an agronomic benefit over soluble granular fertilizers, but the possible contribution of nanocolloidal P to P uptake was still further investigated in hydroponic experiments. Phosphorus bioavailability is related to its concentration and speciation in the soil solution. Free orthophosphate is the form of P taken up by plants; but colloidal P constitutes an important fraction of total solution P in oxide- or allophane-rich soils and its bioavailability has not been previously considered. The uptake of P by wheat seedlings was measured from radiolabeled non-filtered (colloid-containing) and 3-kDa filtered (colloid-free) soil-water extracts from Andisols and Oxisols. In the Andisol extracts, P uptake was up to seven-fold higher in the non-filtered solutions than in the corresponding 3-kDa filtered solutions. It is hypothesized that labile humic/fulvic-Fe/Al-P complexes increased the diffusive transport flux of free P to the roots. In the Oxisol extract, no difference in P uptake between both solutions was observed. Also, the diffusional flux of P measured with the diffusive gradient in-thin films (DGT) method was larger in the nonfiltered than in the 3-kDa filtered solutions. These results are the first observation that natural colloidal P is not inert and can contribute to plant P uptake. This work has shown that increasing soil available P and fertilizer efficiency in soils where strong adsorption reactions control P availability is very challenging. However, the observed contribution of colloidal P to plant P uptake for Andisols is a finding that may lead to the development of new management practices to enhance the release of P-containing colloids into solution as a complimentary strategy to P fertilization in these strongly Psorbing soils. Although in this study hydroxyapatite nanoparticles offered no advantage over conventional soluble P fertilizers for plant growth, this does not imply that nano-sized P fertilizers can be ruled ineffective. The addition of labile nanocolloidal P that is mobile in soil and contributes to P uptake is still a worthwhile fertilizer strategy to investigate.