Elimination of micro-organisms by water treatment processes : a review

The Dutch Drinking Water Decree prescribes that drinking water companies should demonstrate that their drinking water is microbiologically safe by performing a quantitative microbial risk assessment (QMRA). The required level of safety is a risk of infection of 10-4 per person per year. Such a QMRA requires knowledge about (i) the concentration of pathogens in source water and (ii) the efficacy of water treatment to eliminate these pathogens. Consequently, there is a need for a scientific database with the elimination capacity of all relevant processes used in drinking water treatment for viruses, bacteria and pathogenic protozoa. In the joint research program of the Dutch Drinking water Companies (BTO-programme) a project was defined to create such a database. An increasing amount of knowledge about the microbiological efficacy of drinking water treatment is published in the literature. This literature is collected and evaluated. Many publications describe experiments on the elimination of micro-organisms that are conducted in the laboratory under well-defined conditions with lab-strains of micro-organisms. Experience of Kiwa and of many others show that the efficacy of (well-controlled) full-scale treatment processes is usually lower than expected from a direct extrapolation of laboratory experiments. This can be due to several factors, such as the variability of the conditions in practice (feed water quality, temperature etc.), hydraulic differences between well-mixed laboratory vessel and a large flow-through reactor and the difference between micro-organisms in the laboratory and in the environment (survival state, attachment to particles etc.). This does not disqualify laboratory experiments. These provide a first impression of the efficacy of a process and in some cases data collection of full-scale conditions is not possible. Furthermore, for disinfection processes these studies are necessary to assess dose-response curves for selected pathogenic micro-organisms under standardised conditions. In addition laboratory experiments are needed to study the effect of conditions such as temperature, pH, turbidity etc. on the efficacy of a process. Combining these results with full-scale observations will optimise the process of risk assessment. The overall aim of this review was to produce a default value for the Microorganism Elimination or Inactivation Credit (MEC or MIC) of full-scale treatment processes and a description of the effect of water quality parameters and process control parameters on the elimination or inactivation capacity. The literature data are valued according to their representation of full-scale conditions. - For physical processes the calculated default value of MEC is weighted on the basis of the resemblance with full-scale conditions. - In the dose requirement table described for certain MIC values of a disinfection process, effects of microbial and process conditions on the dose-response curves assessed for spiked and pre-cultured organisms are included. Third edition in 2007 The report describes the state-of-the-art and will be updated periodically to incorporate the progress in research. In the last and second edition the literature on conventional treatment (coagulation and floc-removal plus rapid granular filtration) was evaluated and the chapters on coagulation and flocremoval, slow sand filtration and UV disinfection were updated. In the third edition a chapter on rapid granular filtration is added and the chapter on slow sand filtration was updated