Phosphorus bioavailability from land-applied biosolids in south-western Australia

Abstract

The annual production of biosolids in the Perth region during the period of this study was approximately 13,800 t dry solids (DS), being supplied by three major wastewater treatment plants. Of this, 70% was typically used as a low-grade fertiliser in agriculture, representing an annual land use area of around 1,600 ha when spread between 5 and 7 t DS/ha. Loading rates of biosolids are typically based on the nitrogen (N) requirements of the crop to be grown, referred to as the N Limiting Biosolids Application Rate (NLBAR). A consequence of using the NLBAR to calculate loading rates is that phosphorus (P) is typically in excess of plant requirement. The resultant high loading rates of P are considered in the guidelines developed for the agricultural use of biosolids in Western Australia, but lack research data specific to local conditions and soil types. Regulatory changes throughout Australia and globally to protect the environment from wastewater pollution have created a need for more accountable and balanced nutrient data. Experiments presented in this thesis were undertaken to ascertain: the percentage relative effectiveness (RE) of biosolids as a source of plant available P compared with inorganic P fertiliser; loading rates to best supply P for optimum crop growth; P loading rates of risk to the environment; and the forms of P in local biosolids. Therefore, both the agronomic and environmental viewpoints were considered. Anaerobically digested and dewatered biosolids produced from Beenyup Wastewater Treatment Plant, Perth with a mean total P content of 2.97% dry weight basis (db) were used in a series of glasshouse, field and laboratory experiments. The biosolids were sequentially fractionated to identify the forms of P present and likewise in soil samples after applying biosolids or monocalcium phosphate (MCP).The biosolid P was predominantly inorganic (92%), and hence the organic fraction (8%) available for mineralisation at all times would be extremely low. The most common forms of biosolid P were water-soluble P and exchangeable inorganic P (66%), followed by bicarbonate extractable P (19%) and the remaining P as inorganic forms associated with Fe, Al and Ca (14%). Following the application of biosolids to a lateritic soil, the Fe and Al soil fractions sorbed large amounts of P, not unlike the distribution of P following the addition of MCP. Further investigation would be required to trace the cycling of biosolid P in the various soil pools. The growth response of wheat (Triticum aestivum L.) to increasing rates of biosolids and comparable rates of inorganic P as MCP, to a maximum of 150 mg P/kg soil was examined in the glasshouse. The percentage relative effectiveness (RE) of biosolids was calculated using fitted curve coefficients from the Mitscherlich equation: y = a (1-b exp–cx) for dry matter (DM) production and P uptake. The initial effectiveness of biosolid P was comparable to that of MCP with the percentage RE of biosolids averaging 106% for DM production of wheat shoots and 118% for shoot P uptake at 33 days after sowing (DAS) over three consecutive crops. The percentage residual value (RV) declined at similar rates for DM production in MCP and biosolids, decreasing to about 33% relative to freshly applied MCP in the second crop and to approximately 16% in the third crop. The effectiveness of biosolid P was reduced significantly compared with inorganic P when applied to a field site 80 km east of Perth (520 mm annual rainfall). An infertile lateritic podsolic soil, consistent with the glasshouse experiment and representative of a soil type typically used for the agricultural application of biosolids in Western Australia was used.Increasing rates of biosolids and comparable rates of triple superphosphate (TSP), to a maximum of 145 kg P/ha were applied to determine a P response curve. The percentage RE was calculated for seasonal DM production, final grain yield and P uptake in wheat followed by lupin (Lupinus angustifolius L.) rotation for the 2001 and 2002 growing seasons, respectively. In the first year of wheat, the RE for P uptake in biosolids compared with top-dressed TSP ranged from 33% to 55% over the season and by grain harvest was 67%. In the second year, and following incorporation with the disc plough at seeding, the RE for P uptake by lupins in biosolids averaged 79% over the growing season compared with top-dressed TSP, and by grain harvest the RE was 60%. The residual value (RV) of lupins at harvest in biosolids compared with freshly applied TSP was 47%. The non-uniform placement of biosolids (i.e. spatial heterogeneity) was primarily responsible for the decreased ability of plant roots to absorb P. The P was more effective where biosolids were finely dispersed throughout the soil, less so when roughly cultivated and least effective when placed on the soil surface without incorporation. The RE for grain harvest of wheat in the field decreased from 67% to 39% where biosolids were not incorporated (i.e. surface-applied). The RE could also be modified by factors such as soil moisture and N availability in the field, although it was possible to keep these variables constant in the glasshouse. Consequently, absolute values determined for the RE need to be treated judiciously. Calculations showed that typical loading rates of biosolids required to satisfy agronomic P requirements of wheat in Western Australia in the first season could vary from 0 to 8.1 t DS/ha, depending on soil factors such as the P Retention Index (PRI) and bicarbonate available P value.Loading rates of biosolids were inadequate for optimum P uptake by wheat at 5 t DS/ha (i.e. 145 kg P/ha) based on the NLBAR on high P sorbing soils with a low fertiliser history (i.e. PRI >15, Colwell bicarbonate extractable P <15 mg P/kg). On soils of PRI <2 mL/g however, biosolids applied at identical loading rates would result in high concentrations of available P. Further work on sites not P deficient would be necessary to validate these findings on farmed soils with a regular history of P fertiliser. The sieving of soil samples used in the field experiment to remove stones and coarse organic matter prior to chemical analysis inadvertently discarded biosolids particles >2 mm, and thus their was little relationship between soil bicarbonate extractable P and P uptake by plants in the field. The risk of P leaching in biosolids-amended soil was examined over a number of different soil types at comparable rates of P at 140 mg P/kg (as either biosolids or MCP) in a laboratory experiment. Given that biosolids are restricted on sites prone to water erosion, the study focussed on the movement of water-soluble P by leaching rather than by runoff of water-soluble P and particulate P. In general the percentage soluble reactive P recovered was lower in soils treated with biosolids than with MCP, as measured in leachate collected using a reverse soil leachate unit. This was particularly evident in acid washed sand with SRP measuring 14% for biosolids and 71% for MCP, respectively, although the differences were not as large in typical agricultural soils. Specific soil properties, such as the PRI, pH, organic carbon and reactive Fe content were negatively correlated to soluble reactive P in leachate and thus reduced the risk of P leaching in biosolids-amended soil.Conversely, the total P and bicarbonate extractable P status of the soils investigated were unreliable indicators as to the amount of P leached. On the basis of the experiments conducted, soils in Western Australia were categorised according to their ability to minimise P enrichment and provide P necessary for crop growth at loading rates determined by the NLBAR. Biosolids applied at the NLBAR to soils of PRI >2mL/g with reactive Fe >200 mg/kg were unlikely to necessitate P loading restrictions. Although specific to anaerobically digested biosolids cake applied to Western Australian soils, the results will be of relevance to any industry involved in the land application of biosolids, to prevent P contamination in water bodies and to make better use of P in crop production.