Efficiency of reed beds in treating dairy wastewater

At dairy farms in the UK, dirty water is produced from dairy wash water, runoff from soiled yard areas and in some cases silage effluent. It typically contains organic matter, solid matter, nitrogen and phosphorus compounds and pathogens. The purpose of this paper is to test the hypothesis that reed beds can be used to treat high-strength agricultural effluent of 5-d biochemical oxygen demand (BOD5) up to 6000 mg l−1, following either dilution with treated water or pre-treatment such as intensive aeration. The experimental process was designed to treat up to 1/50th of the average volume of 25 m3 d−1 of wastewater produced on the farm. When 500 l d−1 wastewater had been pre-treated in an intensive aeration unit BOD5 removals of the order 98.6% occurred in the aeration plant, 75.1% in the reed bed and 99.6% in the whole system incorporating the reed bed. However, the supply of a smaller volume of 50 l d−1 of raw dirty water diluted with a recycle of 1100 l d−1 of treated water to the reed bed only, led to removal rates of BOD5 as low as 58%. A mathematical model based on the tidal flow of water and first-order BOD5 removal was applied to predict the outlet concentration of BOD5 from the reed beds. During the summer months the correlation between predicted and actual BOD5 was high, as indicated by a correlation coefficient (R2) of 0.882, but in the winter months the correlation was poorer, with a correlation coefficient of 0.763.