Water Journal : Water Journal December 2011
refereed paper microbiology water DECEMBER 2011 83 shows that data are unsuitable, the designer of the program is forced to consider why such data are unsuitable and how data that are to be collected are best reported (Step 8) to optimise future data sharing and use. While not directly related to the attributes of the monitoring program itself, Steps 3 and 4, which focus on the testing laboratory and communication between the program designer and laboratory personnel, are important in assuring the quality of generated data. Decision making about sampling issues, such as where samples should be collected, sampling technique and sample preservation, are also important but have not been considered as part of the schema outlined. Step 1: Defining the Purpose of Monitoring Microbiological monitoring programs are undertaken for many reasons, ranging from compliance with water quality regulations, where the end product is tested; operational monitoring, where process control is assessed and verified at different points in the water treatment train; and investigative monitoring as part of the risk management process, where treated water and/or source water quality may be assessed. For water quality compliance monitoring, many of the decisions about the monitoring program are pre-determined. For this type of monitoring the analyte is prescribed, as are the frequency of testing and the number of samples per annum per population supplied. Regulations may also require that a particular method is employed for testing and prescribe sampling locations. However, this is not the case for operational monitoring or investigative monitoring programs. For these types of monitoring programs, there is scope in choosing the design. Defining the purpose of monitoring is central to monitoring program design; hence, it is important that it is carefully considered. In some instances a number of monitoring options may be suitable for purpose and the selection of one approach over another may be determined by budgetary constraints or, perhaps, on the basis of standardisation with prior collected data. Careful consideration of the purpose of monitoring may show that the use of a surrogate micro-organism or analyte, rather than monitoring for the micro- organism of interest, is both suitable for purpose and affordable. Alternatively, based on the intended purpose of monitoring, investigation may show that the use of surrogates is inappropriate and there are no data supporting their use, requiring review of the budget amount to allow for monitoring of the target micro- organism, assuming that a testing method is available. Where the purpose of monitoring is to quantify the risk associated with a particular scenario or to compare two scenarios, the aim of the monitoring program should be defined in terms of the planned statistical analysis. Where a judgment is to be made about whether the scenario meets pre-set guideline limits, the acceptance criteria should be specified. These additional specifications will ensure that appropriate consideration is given to method detection limits and to the appropriate use of enumerative versus qualitative methods. Specification of the type of statistical analysis to be performed prior to data collection also assists to ensure that sufficient samples are analysed or, if this is not possible due to budget constraints, that there is a re-consideration of the budget amount and/or the scope of the monitoring program at the planning stage. Careful consideration of planned statistical analysis or pass/fail judgements may show that pilot monitoring is required. A pilot monitoring program can be used to determine the likely prevalence and levels of micro-organisms in the water type in question, thereby assisting to ensure that in the 'full' program, sufficient numbers of positive samples are collected (and budgeted for) and/or the sensitivity of the method employed is adequate. Step 2: Are Other Data Available and Useful? Monitoring programs are expensive, hence it is preferable that high quality data are generated which, ideally, can be shared. The usefulness of data for purposes other than for which they were initially collected depends upon the integrity of the method(s) used for generating data, the documentation accompanying the data about the method used (e.g. reference to a standard method or, if a standard method was not employed, exact details of the method employed, including its operating characteristics); the water type tested (e.g. physico- chemical properties, catchment type, pollution inputs, rainfall levels) and, data being made accessible. Data may be available from the scientific literature, from organisational records (e.g. where prior monitoring programs have been undertaken) and, increasingly, because of electronic data storage capabilities, from data repositories that have been specifically set up for this purpose. A search for relevant data in advance of conducting a water quality monitoring program has merit for many reasons. Where suitable data are already available, data may be used as a substitute for, or a supplement to, data that are planned to be collected. This allows for a cost- effective use of resources. Where data are available but scrutiny shows that data are unsuitable, the designer of the program is forced to consider why such data are unsuitable and how data that are to be collected are best reported to optimise future data sharing and use. In addition, Step 1: Summary • Are program objectives too broad or ambitious? Can project objectives be refined? • Type of data analysis (QMRA, comparative, pass/fail with guideline etc)? • Budget adequacy? • Consider target micro-organism(s)? • Surrogate? • Viability/infectivity/genotyping? • Information to be recorded for future data sharing? Step 2: Summary • Are data available that can be used to substitute/ supplement data to be collected? • If data are able to be used (e.g. as 'default' values), what are the assumptions made (if any) about their suitability and is there supporting information to justify any assumptions made? • Does data provide information about the prevalence and/ or levels of micro-organisms in the water type of interest and, thereby, assist in scoping the monitoring program? • Where data are unsuitable, what information is missing? Use this as a prompt to ensure that all necessary information is recorded with results in the execution of this monitoring program.
Water Journal April 2012
Water Journal November 2011