Removal of dissolved reactive phosphorus from municipal and dairy factory wastewater using allophanic soil : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Soil Science, Massey University, Palmerston North, New Zealand

Abstract

Many of New Zealand’s sewage treatment plants (STPs) and rural factories discharge treated or partially treated sewage, which is rich in dissolved reactive phosphorus (DRP), into rivers and streams. A large number of these STPs are not able to comply with the current DRP river standards because conventional treatment methods are cost-prohibitive. There is an abundance of Allophanic soils with high phosphorus (P) sorption capacities located in the central North Island of New Zealand that have potential for use as low-cost filter material for removing DRP from wastewaters. For Allophanic soil filters to be a viable treatment option, the soil, in addition to having a high P sorption capacity, should be both accessible and plentiful. The main aims of this study were to assess and improve the effectiveness of Allophanic soil filters at removing DRP from wastewaters and to evaluate the agronomic value of P-enriched soils as a P source for plant growth. It also sought to contribute to a better understanding of the feasibility and important design characteristics of fullscale soil-based treatment systems. Five quarry sites in the Waikato Region were soil sampled to identify soils with high P retention values. Only the Te Mata Quarry (TQ) soil in the, northwestern Waikato Region, had a high P retention value at or close to 100% as assessed using the standard (5 g) anion storage capacity (ASC) test. The modified (1 g) ASC test revealed P retention values of 47 – 91% for samples taken from different soil depths at TQ. All of the soil depths down to 600 cm, except for the 125 – 175 cm depth, had modified (1 g) ASC test values >58%. This indicated that the TQ soil had P sorption capacities that would potentially make it a suitable material for filtering DRP from wastewater and, therefore, it warranted further evaluation using real wastewater. Wastewater pH has a marked influence on the P sorption capacity of soil filters, with the sorption capacity expected to increase as wastewater pH is decreased, from being alkaline to acidic. The laboratory soil column experiment quantified the effect of the level of acid dosing and the type of acid used on the capacity of soils to remove P from wastewater. Columns of soil, taken from a quarry at Ohakune (OQ), and treated with wastewater adjusted to pH 5.5 removed the greatest amount of DRP. A total of 8.9 mg P/g oven-dried soil was removed at an average removal efficiency of 75%. In comparison, the soil columns treated with wastewater without pH adjustment, removed only 4.5 mg P/g oven-dried soil at the same removal efficiency of 75%. This highlights the merits of lowering wastewater pH to increase DRP removal capacity. The performance pilot-scale soil filters at the Dannevirke STP and Fonterra Te Rapa WTP were evaluated, under field conditions, for a total operational period of 440 and 376 days, respectively. Each filter contained the OQ soil and had a surface area of 1 m². The OQ soil had an overall P removal efficiency of 67% and 71% at the STP and WTP sites, respectively. The OQ soil filters at Dannevirke STP removed a total of 6.4 mg P/g oven-dried soil, while the OQ soil filters at the Fonterra Te Rapa WTP removed a total of 1.87 mg P/g ovendried soil. This discrepancy in performance was due to the difference in wastewater type and pH adjustment, initial P concentrations, and soil pretreatment (i.e. the soil used at Dannevirke was sieved). A cost/benefit analysis suggested that if the STP was 225 km from the soil source then the cost of acid dosing is about ten times greater than the cost of supplying additional soil to achieve the same amount of P removal. Therefore, it is unlikely that acid dosing will be cost competitive for most wastewater treatment sites in the central North Island of New Zealand. The wastewater treated soil (WTS) obtained from the Dannevirke STP pilotscale filter experiment was evaluated for its agronomic effectiveness in a glasshouse pot experiment. The ability of WTS to supply P for ryegrass growth (Lolium multiflorum) was compared with a soluble phosphorus source (monocalcium phosphate, MCP). The WTS was highly effective at increasing available P in the soil, as measured by the Olsen P soil test, ryegrass yield and ryegrass P uptake. The soluble fertiliser P value of WTS was estimated to be equivalent to 61% of MCP applied at the same rate. Therefore, the results show that WTS is an effective P source for plant growth and its application to soil has the potential to recycle both the soil and the P it contains.