Identification of polyphosphate accumulating bacteria from pilot- and full scale nutrient removal activated sludges

General removal of phosphorus (P) from wastewater was introduced in Scandanavia in the late 1960's. At that time it was believed that P alone was limiting to algal growth and that the sole removal of P would solve the problem of eutrophication. However, we now know that both P and nitrogen (N) contribute to this deleterious effect and as such, much research has been conducted concerned with both the biological and chemical removal of these nutrients from sewage effluents. Enhanced biological phosphorus removal (EBPR), which is basically the biological accumulation of soluble P (as polyphosphate or poly-P) from the bulk liquid in excess of normal metabolic requirements, still tends to be sensitive to many external parameters and, as such, is subject to fluctuations. This makes it extremely difficult for wastewater treatment installations to achieve and maintain full compliance with strict discharge regulations. A more comprehensive understanding of the microbial community within the mixed liquor of a wastewater treatment system is therefore required which will ultimately assist in improving system design and performance. Chemical and civil engineers, when designing biological wastewater treatment systems, consider only the processes (biological or chemical) taking place within the reactor/s with little or no regard for the individual microbial species or the entire microbial community involved. Process design appears to be tackled empirically from a 'black box' approach; biological reactions or processes occurring within a system such as wastewater treatment are all lumped together and attributed to a single surrogate organism ie., the response of the surrogate to certain stimuli accounts for the total system response. This is similar to an analogy which Professor George Ekama (Dept of Civil Engineering, UCT), a leading scientist in wastewater treatment and process design, refers to where engineers, if, for example, are confronted with modelling the dynamics of carbon dioxide utilisation ofa forest, would recognise the accumulative system response and not give cognisance to each individual tree's contribution. It is true that if one had to consider every microbial species present in a highly organised community such as activated sludge, process models, designed to make quantitative and qualitative predictions as to the expected effluent quality from a particular design, would become increasingly complex and superfluous. It is evident from the countless accomplishments that engineers have succeeded, to a certain degree, in modelling wastewater treatment systems. One only has to consider the tremendous success of biological P (bio-P) removal and nitrification/denitrification processes at full-scale. However, there are limitations to this empirical approach and EBPR processes occasionally deteriorate in phosphate removal efficiency. In order to further optimise biological processes, whether they be organics oxidation, bio-P removal, nitrification or denitrification, biological community analyses will have to play a more significant role in design. The better microbial community structure and function is understood, the better the control and management of the system. With the advent of improved microbial identification and enumeration (to a certain extent) techniques (in situ), it was considered significant to investigate the mechanism ofbio-P removal and to elucidate which bacteria are actively responsible for this process. To this end, experimental work was conducted in two phases: \xAE laboratory, where samples of mixed liquor were obtained from a full-scale wastewater treatment facility exhibiting biological nutrient removal (BNR) characteristics and @ pilot plant, where an enhanced culture ofpolyphosphate accumulating organisms (PAO's) was developed and probed using molecular identification and enumeration techniques (as well as a cultivation-dependent approach). During phase \xAE of experimentat