Land application has become a widely applied method for treating wastewater. However, it is not always clear which soil-plant systems should be used, or why. The objectives of our study were to determine if four contrasting soils, from which the pasture is regularly cut and removed, varied in their ability to assimilate nutrients from secondary-treated domestic effluent under high hydraulic loadings, in comparison with unirrigated, fertilized pasture. Grassed intact soil cores (500 mm in diameter by 700 mm in depth) were irrigated (50 mm wk(-1)) with secondary-treated domestic effluent for two years. Soils included a well-drained Allophanic Soil (Typic Hapludand), a poorly drained Gley Soil (Typic Endoaquept), a well-drained Pumice Soil formed from rhyolitic tephra (Typic Udivitrand), and a well-drained Recent Soil formed in a sand dune (Typic Udipsamment). Effluent-irrigated soils received between 746 and 815 kg N ha(-1) and 283 and 331 kg P ha(-1) over two years of irrigation, and unirrigated treatments received 200 kg N ha(-1) and 100 kg P ha(-1) of dissolved inorganic fertilizer over the same period. Applying effluent significantly increased plant uptake of N and P from all soil types. For the effluent-irrigated soils plant N uptake ranged from 186 to 437 kg N ha(-1) yr(-1), while plant P uptake ranged from 40 to 88 kg P ha(-1) yr(-1) for the effluent-irrigated soils. Applying effluent significantly increased N leaching losses from Gley and Recent Soils, and after two years ranged from 17 to 184 kg N ha(-1) depending on soil type. Effluent irrigation only increased P leaching from the Gley Soil. All P leaching losses were less than 49 kg P ha(-1) after two years. The N and P leached from effluent treatments were mainly in organic form (69-87% organic N and 35-65% unreactive P). Greater N and P leaching losses from the irrigated Gley Soil were attributed to preferential flow that reduced contact between the effluent and the soil matrix. Increased N leaching from the Recent Soil was the result of increased leaching of native soil organic N due to the higher hydraulic loading from the effluent irrigation.

There has been a recent, rapid increase in both land application of dairy shed effluent in Southland, New Zealand, and the microbial load in ground and surface waters. We investigated the fate of faecal coliforms, a host-specific Salmonella bacteriophage, and a non-reactive chemical tracer (Br–), when applied to large intact lysimeter soil cores (500 mm diam. by 500 mm high), to determine the pattern of microbial transport through typical Southland soils. The soils were a poorly drained Fragic Perch-gley Pallic Soil, and a well-drained Typic Firm Brown Soil. A depth of 25 mm of dairy shed effluent containing faecal coliforms and spiked with bacteriophage and Br– was applied to the soil at a rate of 5 mm/h followed by ~1 pore volume of simulated rainfall applied at 5 mm/h. Resulting leachates, collected continuously over ~1 pore volume, were analysed for the microbial and bromide tracers. The microbial tracers moved rapidly through both soils, peaking early in the leachate at ~0.15 pore volume and then tailing off in a pattern indicative of bypass flow. Bromide moved more uniformly through the soils but peaked at ~0.5–0.8 pore volume. The microbial flow pattern observed indicates that the structure in these soils makes them vulnerable to leaching of microbes into local surface and ground water. The large difference between the rate of microbial and chemical tracer transport indicates chemical tracers should only be used with caution to model microbial transport parameters.

Microbial breakthrough curves of 12 soils, generated by the application of dairy shed effluent followed by continuous artificial rainfall for one pore volume at 5 mm h(-1) onto large undisturbed soil cores, have been ranked as high, medium, or low potential for microbial bypass flow. The ranking is based on the position of the peak in the breakthrough curve. Knowledge of soil properties that affect microbial transport through soil gained from the microbial breakthrough curves was linked to soil classes, or to their accessory properties, of the New Zealand Soil Classification. Spatial depiction of the ratings has been achieved via the national 1:50,000 scale soil map. Soils with a drainage impediment or those with well developed soil structure have a high potential for microbial bypass flow, whereas soils from tephra and Recent Soils with less developed, porous, soil structure have a low potential for microbial bypass flow. The risk rankings should be considered as maxima because management may change some rankings.

The ability of New Zealand soils to renovate dairy-shed effluent following application to land is being evaluated. We investigated the pattern of transport of faecal coliforms, a host-specific Salmonella bacteriophage and a non-reactive chemical tracer (Br-), when applied to large, intact lysimeter soil cores (460 mm dia. × 520-700 mm high) of three contrasting soils. The soils were imperfectly drained Ultic and Granular Soils and a well-drained Recent Soil. A depth of 25 mm of dairy-shed effluent containing faecal coliforms and spiked with bacteriophage and Br-was applied to the soil at a rate of 5 mm h -1 followed by up to 1 pore volume of simulated rainfall applied at 5 mm h -1 . This application rate is generally much slower than the soil's saturated hydraulic conductivity except in the Ultic Soil where saturated hydraulic conductivity is slower. Resulting leachates, collected continuously, were analysed for the microbial and bromide tracers. The phage tracer moved rapidly through all soils, peaking early in the leachates and then tailing off in a pattern indicative of bypass flow. Faecal coliforms also moved rapidly through the Ultic and Granular Soils but numbers were much lower or not detectable in leachate from the Recent Soil. In contrast, bromide moved uniformly through Granular and Recent Soils but peaked early at about 0.5-0.8 pore volume.The microbial data suggest the soil structure in the Ultic and Granular Soils makes them vulnerable to leaching of microbes into shallow water bodies.

Land treatment is the preferred option for the disposal of wastewater in New Zealand. We applied secondary-treated municipal wastewater to 4 contrasting soils (a Gley, Pumice, Recent, and Allophanic Soil) at the rate of 50 mm per week, for 4 years. Amounts of N and P in applied wastewater, leachates, and removed in herbage were measured every 1–4 weeks, and a range of soil chemical, biochemical and physical characteristics measured by destructive sampling after 2 and 4 years. After 4 years, leaching losses amounted to 290–307 kg N on the Gley and Recent Soils, representing approximately 22% of the N applied. Leaching losses from the Allophanic and Pumice Soils were 44 and 69 kg N/ha, respectively, representing <5% of that applied. More than half of the N leached was in organic forms. Leaching losses of P were <5 kg P/ha on the Pumice and Allophanic Soils (< 1% of that applied), 41 kg P/ha from the Recent Soil and 65 kg P/ha from the Gley Soil (8% and 13% of that applied, respectively). After 4 years, the total C and microbial C content in the A horizon of the irrigated Recent Soil were, respectively, 47% and 44% less than non-irrigated cores. All irrigated soils showed a rise in pH of up to 1 unit, and all had a marked increase in the exchangeable Na+ which reached 4–22% ESP. After 4 years, the saturated and near saturated hydraulic conductivity of the Gley Soil had declined from 567 and 40 mm/h to 56 and 3 mm/h, respectively. Allophanic and Pumice Soils are to be preferred over the Recent and Gley Soils for effective treatment of wastewater and to minimise the loss of nutrients to the wider environment.

No abstract

The ability of soil to function as a barrier between microbial pathogens in wastes and groundwater following application of animal wastes is dependent on soil structure. We irrigated soil lysimeters with dairy shed effluent at intervals of 3–4 months and monitored microbial indicators (somatic coliphage, faecal enterococci, Escherichia coli) in soil core leachates for 1 year. The lysimeters were maintained in a lysimeter facility under natural soil temperature and moisture regimes. Microbial indicators were rapidly transported to depth in well-structured Netherton clay loam soil. Peak concentrations of E. coli and somatic coliphage were detected immediately following dairy shed effluent application to Netherton clay loam soil, and E. coli continued to leach from the soil following rainfall. In contrast, microbial indicators were rarely detected in leachates from fine-structured Manawatu sandy loam soil. Potential for leaching was dependent on soil moisture conditions in Manawatu soil but not Netherton soil, where leaching occurred regardless. Dye studies confirmed that E. coli can be transported to depth by flow through continuous macropores in Netherton soils. However, in the main E. coli was retained in topsoil of Netherton and Manawatu soil.

A well-drained soil in N-fertilized dairy pasture was amended with particulate organic carbon (POC), either sawdust or coarse woody mulch, and sampled every 4 wk for a year to test the hypothesis that the addition of POC would increase denitrifi cation activity by increasing the number of microsites where denitrifi cation occurred. Overall mean denitrifying enzyme activity (DEA), on a gravimetric basis, was 100% greater for the woody mulch treatment and 50% greater for the sawdust treatment compared with controls, indicating the denitrifying potential of the soil was enhanced. Despite diff erences in DEA, no diff erence in denitrifi cation rate, as measured by the acetylene block technique, was detected among treatments, with an average annual N loss of ~22 kg N ha −1 yr −1 . Soil water content overall was driving denitrifi cation in this well-drained soil as regression of the natural log of volumetric soil water content (VWC) against denitrifi cation rate was highly signifi cant (r 2 = 0.74, P < 0.001). Addition of the amendments, however, had signifi cant eff ects on the availability of both C and N. An additional 20 to 40 kg N ha −1 was stored in POC-amended treatments as a result of increases in the microbial biomass. Basal respiration, as a measure of available C, was 400% greater than controls in the sawdust treatment and 250% greater than controls in the mulch. Net N mineralization, however, was signifi cantly lower in the sawdust treatment, resulting in signifi cantly lower nitrate N levels than in the control. We attribute the lack of measured response in denitrifi cation rate to the high temporal variability in denitrifi cation and suggest that diff usion of nitrate may ultimately have limited denitrifi cation in the amended treatments. Our data indicate that manipulation of denitrifi cation by addition of POC may be possible, particularly when nitrate levels are high, but quantifying diff erences in the rate of denitrifi cation is diffi cult because of the temporal nature of the process (particularly the complex interaction of N availability and soil water content).