Pasture as a treatment system for high rate application of effluent

The basic aim of this project was to study the effectiveness of pasture as an aerobic treatment system for high rate piggery effluent application. Two management systems were compared. The first was year round applicaLion of effluent and the second employed separation of the effluent into particulate matter and supernatant then application of the sludge to pasture and storage of the supernatant during those months where the pasture cannot accept extra hydraulic loading. The stored supernatant was applied to pasture during the summer months when the soil water levels were low. Separating out the high BOD containing solids components of the effluent before storage considerably reduced the oxidation pond area required. An above ground settling tank was easy to operate and after 24 hours settling the tank produced a supernatant with a total solids content of 1172 mg/1 and a sludge of 9993mg/1 total solids. An oxidation pond was chosen to store the supernatant to keep odour problems at a minimum. The pond worked well with the low turbidity effluent allowing good light penetration resulting in dense algal growth and thus producing an aerobic layer that reduced odour. Large losses of water and nutrients occurred from the pond due to evaporation, seepage, losses in gases and fixation by the clay forming the bottom and walls of the pond. These losses may have been compounded by difficulties in obtaining representative samples from an unmixed pond. Soil on the effluent treated plots showed a larger increase in phosphorus content than potassium which would reflect the higher phosphorus levels in the effluent applied. Pasture growth was not adversely affected by the high application rates. Pasture growth was greater on the effluent treated plots than the control plots during summer due largely to the irrigation effect. Acceptance of effluent soiled pasture by sheep was poor when new sheep were introduced, however, acceptance increased with time after introduction. Runoff was measured by draining the .45 hectare plots with levy banks and surface drains to a dumping tank situated in a ditch at the base of each plot. The number of dumps were automatically recorded and surface water samples taken daily during periods of runoff. Runoff was higher in 1981 which had a higher rainfall than 1982, 604mm compared to 453mm. Total runoff on the stored plots was 48% and 46% of the runoff on the fresh plots in 1981 and 1982 respectively. Even the stored treatment showed a large increase in runoff over the control plots due to extra hydraulic loading onto saturated soil during the winter months. The ability of pasture to filter nutrients was very high. Approximately 1.3% of the BOD, 4.2% of the nitrogen, .8% of the phosphorus and 1.3% of the potassium applied to the stored plots was lost in runoff. However the quality of the runoff was not good enough for direct discharge into streams. Drainage pipes were laid in the plots at a depth of 30cm. Pollutants in the drainage water were generally in low concentrations. Runoff predicted by a water balance model WBAL3 (Rosenthal et al, 1976) was compared to runoff actually measured for each plot. The model accurately predicted the weeks in which runoff occurred. It also satisfactorily predicted the volumes of runoff although it showed a tendency to overestimate. The model can also be used to determine when and how much effluent irrigation could be applied throughout the year without causing increased runoff and therefore has a place as a management aid for high rate effluent application.