Nitrification in bulk water and biofilms of algae wastewater stabilization ponds.

Nitrogen removal in wastewater stabilization ponds is poorly understood and effluent monitoring data show a wide range of differences in ammonium. For effluent discharge into the environment, low levels of nitrogen are recommended. Nitrification is limiting in facultative wastewater stabilization ponds. The reason why nitrification is considered to be limiting is attributed to low growth rate and wash out of the nitrifiers. Therefore to maintain a population, attached growth is required. The aim of this research is to study the relative contribution of bulk water and biofilms with respect to nitrification. The hypothesis is that nitrification can be enhanced in stabilization ponds by increasing the surface area for nitrifier attachment. In order to achieve this, transparent pond reactors representing water columns in algae WSP have been used. To discriminate between bulk and biofilm activity, 5-day batch activity tests were carried out with bulk water and biofilm sampled. The observed value for Rnitrbulk was 2.7 x 10(-1) mg-N L(-1) d(-1) and for Rbiofilm was 1,495 mg-N m(-2) d(-1). During the 5 days of experiment with the biofilm, ammonia reduction was rapid on the first day. Therefore, a short-term biofilm activity test was performed to confirm this rapid decrease. Results revealed a nitrification rate, Rbiofilm, of 2,125 mg-N m(-2) d(-1) for the first 5 hours of the test, which is higher than the 1,495 mg-N m(-2) d(-1), observed on the first day of the 7-day biofilm activity test. Rbiofilm and Rnitrbulk values obtained in the batch activity tests were used as parameters in a mass balance model equation. The model was calibrated by adjusting the fraction of the pond volume and biofilm area that is active (i.e. aerobic). When assuming a depth of 0.08 m active upper layer, the model could describe well the measured effluent values for the pond reactors. The calibrated model was validated by predicting effluent Kjeldahl nitrogen of algae ponds in Palestine and Colombia. The model equation predicted well the effluent concentrations of ponds in Palestine.