Impact of phosphate fertiliser derived fluorine on soil microbiology and white clover (Trifolium repens L) : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy (PhD) in Environmental Sciences at Massey University, Palmerston North, New Zealand

Abstract

Fluorine (F) is a significant contaminant in most phosphate fertilisers and fertiliser-derived F is accumulating in New Zealand agricultural soils as a consequent of phosphate fertiliser application. There is potential for soil fluoride (F⁻) to detrimentally affect soil biological functions such as nitrogen fixation by Rhizobium leguminosarum, and to alter soil properties. Fluorine accumulation in soil may require changes to future land use and management practices. The aim of this thesis is to investigate whether phosphate fertiliser-derived soil F has a detrimental effect on soil microorganisms. A novel analytical method for soil F analysis was developed to measure the total soil F concentration based on extraction with dilute NaOH. The relative error between a novel 4 mol L⁻¹ NaOH extraction and the conventional fusion method was < 2 for organic-matter and volcanic parent material derived soils but was > 2 for recent and pallic soils. Precision of the 4 mol L⁻¹ NaOH extraction method, measured through repeat analysis of three further soils (n = 270), was calculated as < 9% Relative Standard Deviation (RSD). To define a standard method to quantify the bioavailable F concentration in soil, samples were extracted with water, 1 mol L⁻¹ HCl, 0.01 mol L⁻¹ CaCl2, 0.01 mol L⁻¹ KCl, and 1 mol L⁻¹ NH4Cl. Compared to water, 0.01 mol L⁻¹ CaCl2 had high relative recovery (of bioavailable F) in soils which have elevated Fe and Al content. Therefore, 0.01 mol L⁻¹ CaCl2 is recommended to measure the bioavailable F concentration of New Zealand pastoral soils. There is no data available which describes the toxic effect of bioavailable F on R. leguminosarum in New Zealand soils. A laboratory incubation experiment and MicroResp 96-well format respiration-inhibition assay were conducted to investigate the effect of F on R. leguminosarum and white clover. Rhizobium leguminosarum growth was not significantly suppressed by F⁻ concentrations less than 100 mg L⁻¹. The normal rod-shaped bacterium cell of R. leguminosarum was morphologically altered when exposed to F⁻ concentrations above 100 mg L⁻¹. The IC10 values determined for F⁻ toxicity to R. leguminosarum were higher than 100 mg F⁻ L⁻¹. Pottle-based experiments showed that white clover growth was not significantly supressed at a F⁻ concentration < 70 mg L⁻¹, while healthy nodules were formed up to a F⁻ concentration of 100 mg L⁻¹. Light and TEM micrographs of nodules revealed that the Rhizobium-white clover interaction was not influenced by F⁻ concentrations up to 100 mg L⁻¹. To assess the potential effects of lime and compost amendment on the bioavailability of F, laboratory F⁻ adsorption/desorption experiments were conducted. Results revealed that at pH < 6, F⁻ adsorption significantly (p < 0.05) increased as a function of compost application. At soil pH > 6, F⁻ adsorption was not significantly (p > 0.05) influenced by compost. Lime application increased the soil pH and maximum F⁻ adsorption was recorded at soil pH between 5.5 – 6.8. These results showed that soil pH significantly influences (p < 0.05) F⁻ desorption and this should be considered in the management of pastoral soil with elevated F. A pot trial was conducted to quantify the effect of added F (equivalent to 0 - 50 years of F accumulation via the continuous application of phosphate fertiliser) on soil properties, soil microbial activity, white clover growth, and R. leguminosarum in an Allophanic soil. F addition (0 – 385 mg kg⁻¹) significantly (p < 0.05) increased soil pH and Dissolved Organic Carbon (DOC) from 5.18 to 5.53 and from 270.5 to 331.3 mg kg⁻¹, respectively. The CaCl2-extractable F concentration increased from 4.95 to 12.67 units as a function of added F. Microbial biomass carbon and soil enzyme activities, and white clover growth and interaction with R. leguminosarum, were not influenced by added F⁻ up to the highest concentration used in this study. White clover variety Merlyn and Tribute shoot F concentration was increased from 4.9 to 19.9 mg kg⁻¹ DM and 5.12 to 16.68 mg kg⁻¹ DM, respectively, however these concentrations are not expected to represent a risk to grazing livestock. This study highlights that the 4 mol L⁻¹ NaOH extraction method is a simple and accurate technique to measure the total F concentration for soils which have high Fe, Al and organic matter content. Water extractable and 0.01 mol L⁻¹ CaCl2-extractable F concentration are recommended to measure the bioavailable concentration of F in New Zealand soils. Current New Zealand bioavailable F concentrations are orders of magnitude lower than the F⁻ concentration assessed to be toxic to R. leguminosarum and white clover, and this suggests there is no imminent risk of soil F to R. leguminosarum. Compost is not recommended as an amendment for soils which have a pH above 6.0 to minimize the bioavailable soil F⁻ concentration. Lime application is suitable in such soils to minimize the bioavailable soil F⁻ concentration through altering soil pH. The major fraction of added F is immobilised by Allophanic soil and this effectively reduces the available F concentration to plants and soil microorganisms. Future work is recommended to investigate the uptake mechanism of bioavailable F into white clover shoots and roots. However, there is no evidence to suggest that F concentrations in New Zealand soil are a risk to New Zealand’s pasture-based farming systems.