Abstract
The numbers of salmonellas in raw sewage entering a treatment plant varied hourly and diurnally; their peak concentration preceded the peak influent waste water flow into the plant by about 2 hours. Salmonellas were detected in all raw influent samples collected from 2 sewage works and the mean population level at the daily peak period was about 3000 organisms/l. On average, 70-80% of the salmonellas were removed in 6-7h during primary sedimentation when upwards of 76-82% of solids were removed, Biological treatment and secondary sedimentation removed a further 70-100% of the pathogen. Compared with the activated sludge process, the trickling filters were less efficient in removing salmonellas and they were adversely affected by increased loading following rainfall. The number of salmonellas in the final effluents varied from 0-50/1 in the activated sludge plant (overall salmonella removal efficiency = 99.83%) and 1-2500 /1 in the trickling filter plant (overall salmonella removal efficiency = 98.70%); these results suggest that greater emphasis should be placed on the actual quality of the effluent rather than on percentage removal efficiencies. The performance of both processes were discussed in relation to the current E.E.C. guidelines concerning the level of salmonellas in surface waters intended for both drinking and swimming. A total of 43 different serotypes were isolated, and of these, only 21 were common to both towns. The common serotypes included, in order of decreasing frequency, S. typhimurium, S. bredeney, S. virchow, S. reading, S. st. paul and S. agona. The isolation of several species from the sewage coincided with the occurrence of clinical cases of salmonellas infection in the human population in the town concerned. Removal of salmonellas from waste water was studied in a laboratory model of the activated sludge process. It appeared to occur in two phases. During the first phase, the salmonellas were rapidly adsorbed to the activated sludge floes and was completely removed from the liquid in under 15 hours. There-after, a slower elimination was observed which was due to predation by ciliate protozoa. This was confirmed by inhibition of predation in the presence of manoxol which inactivated the ciliate protozoan populations. The rate of salmonellas destruction increased with increasing temperature up to 25°C. At 30°C, the ciliate protozoa were destroyed by heat and the salmonella removal rate fell dramatically, A batch anaerobic digestion process was efficient at eliminating salmonellas in raw sludge. A 7-log10 reduction was achieved in under 9 days but the pathogens survived for considerably longer periods when the proportion of raw sludge in the digester was reduced. The bactericidal agent was most effective at low pH (< 5.0) and appeared to comprise the volatile fatty acids. Simple storage of raw sludge in open vessels also led to reductions in their salmonella content but was markedly less efficient compared to the anaerobic digestion process. The reasons why intermittently fed anaerobic digesters do not attain maximum pathogen kills were discussed in the light of the experimental results obtained, and proposals were made to modify the mode of operation of digesters to improve the efficiency of the process with regard to the elimination of the salmonella group and other microbial pathogens.