John Fawell looks at some of the safety issues facing the world of water supply - and some of the possible solutions
The maintenance and assurance of drinking water quality and safety requires that we exercise continual vigilance to maintain the defences against known threats. However, it is equally important to remain vigilant for emerging threats that become apparent because of natural and anthropogenic changes in the environment and new knowledge revealing previously unconsidered or unknown threats. Some of the requirements are driven by public perception and the need to demonstrate safety in order to maintain consumer trust. Sometimes it seems that the latter is the primary driver, causing much frustration amongst professionals because, in communicating with consumers, it often seems as though the two sides are speaking different languages.
The publication of the third edition of the WHO guidelines for drinking water quality and the articulation of the water safety plan (WSP) approach, along with the water industry's Bonn initiatives, provided renewed impetus to the consideration of quality and safety. The approach derives from the application of hazard assessment and critical control points (HACCP), widely adopted by the food industry. This has resulted in a look with fresh eyes at the requirements in the drinking water directive - regarding all contaminants, not just those for which regulatory values are given.
WHO's WSP concept asserts that, in order to achieve the primary objective of 'providing good safe drinking water that has the trust of consumers', there is a need to manage water supply from catchment to tap. The process involves an assessment of the hazards and risks throughout the system with mitigation of the risks, taking into account the potential to affect or create other hazards and risks. A key feature is the measurement of operational parameters which demonstrate that the barriers are working efficiently at all times and provide early warning of potential problems.
Hazard identification will almost certainly give rise to the identification of microbiological and chemical hazards and risks for which there are no standards but which will require careful consideration.
The introduction of WSPs will result in the identification of problems that require solutions, but it will also provide significant opportunities for water professionals to take greater control over their systems. It will also act as a focus for knowledge capture at a time when experience is being lost from the industry, and it will help to provide reassurance to consumers.
While WSPs provide a valuable approach which builds on much of current water supply practice, there are issues that need to be considered. In the WHO WSPs the targets for public health are based on acceptable risk, to try and provide a means of comparing risks and thus facilitate better prioritisation of action and investment. When one is starting from a blank page this is a valuable tool and may well prove to be a key part of future decision-making for drinking water - but it may be difficult to handle in a risk-averse society such as we have in western Europe.
The benefit of the approach was shown by groups in the Netherlands and the USA in comparing the risks from Cryptosporidium with the risks from ozonation, with a clear demonstration that even the most conservative assessment of the risks from ozonation (bromate formation) was insignificant compared to the risks from Cryptosporidium. The introduction of WSPs will bring its own problems, such as the need for better means of continuously monitoring treatment and distribution and, if greater transparency is actually to be realised, better communication with consumers on a range of topics, including risk.
The emergence of Cryptosporidium as a significant threat to drinking water, and the regulation that ensued from it, made the industry reconsider many practices with regard to water treatment. It brought recognition that the multibarrier concept is still an important part of assuring drinking water safety. It also prompted research into a number of microbial contaminants of which knowledge was limited, particularly with regard to the efficiency of various treatment processes.
Cholera and typhoid, the major killers in 19th-century cities, are not an issue in developed countries at present. This is due to a concerted effort on three fronts: sewage treatment, drinking water treatment and health surveillance. Outbreaks as a consequence of chlorination failures became rare in public supplies but chlorination became an increasingly important barrier. Until the advent of Cryptosporidium, which was not susceptible to control by chlorine, large supplies were considered to have conquered waterborne disease. The position in small supplies is potentially different, however, and there remain many small supplies throughout Europe. In 2002, an incident in a small municipal supply in Canada (Hrudy et al 2003), in which several consumers died from waterborne E. coli O157, showed that other pathogens can still be a critical threat.
It is essential that all the barriers are optimised and that it is fully recognised by suppliers (and their staff) that drinking water safety is not just about Cryptosporidium and E. coli. The absence of indicators in treated water does not necessarily show that the water is always safe; the samples provide a tiny snapshot of the total amount of water supplied, and the potential for microbial contamination is highly variable. The same applies to the absence of Cryptosporidium oocysts, since their absence may not necessarily mean that other pathogens are absent. In particular questions remain as to the efficiency of virus removal from many small supplies and some larger supplies. This will also be an issue for assessing the safety of wastewater re-use in the future. Appropriate indicators that will specifically reflect viruses are needed, to confirm that a particular treatment train is capable of their removal, and to identify appropriate operational parameters.
Barriers to infection
Other pathogens have also been identified as possible waterborne threats. Understanding their routes to drinking water, and how barriers can be put in place and maintained for optimal prevention of contamination, remains an important requirement for research and practice. Among these organisms are some that have been primarily associated with food and some that remain of uncertain transmission but which are now considered as possible drinking water threats. A number of these pathogens are particularly important because there can be long-term consequences for those infected, and they cannot be considered as just another tummy bug. These include Campylobacter and Helicobacter. Other pathogens of potential interest are microsporidia and Mycobacterium avium Complex (MAC). One of the significant issues is determining the best means of control. For example Legionella is, not surprisingly, found at low numbers in distribution systems. It is not appropriate to set standards for Legionella in drinking water but if drinking water is a significant route of inoculation of distribution systems in buildings it may require that we find ways of limiting its presence in water supplies as well as limiting growth in plumbing systems.
Another challenge that remains is the more efficient maintenance of storage reservoirs and distribution systems. There remains the potential for ingress by pathogens as a consequence of breaches in the system and pressure drops and pressure surges. Research is currently being directed at the risks to consumers in different circumstances. However, if a significant risk is shown from this source, it will be important to determine how that risk can be detected and dealt with, without causing problems with consumer acceptability. Distribution systems remain a difficult area for the industry at a time when pressure to solve problems is directed towards capital spend rather than operational maintenance and preventive actions.
Chemicals on the rise
Chemical contaminants have risen in profile in developed countries but this seems to have been largely a concern due to adverse consumer perception. WHO considers that there are relatively few chemicals on the list of high-priority contaminants: arsenic, fluoride, nitrate (under some circumstances) and lead from plumbing have been shown to cause health effects through drinking water. One newly recognised contaminant for which there is uncertainty regarding potential health effects is uranium, a kidney toxin.
There is a need for clinical epidemiological studies in larger exposed populations to determine whether apparent risks are real. Arsenic and fluoride can pose problems for some supplies in western Europe, but we have the means of removing these natural contaminants. However, they remain very significant problems in many parts of the world. Nitrate remains a contentious issue but the problem is well understood and the key is how to achieve better means of controlling inputs when the benefits of control may not be seen for many decades. New methods for more efficient and cost-effective removal and replacement of lead pipes in domestic premises remains a requirement that presents a challenge, although phosphate dosing has proved to be a successful palliative.
Research continues to identify possible threats and challenges for both wastewater and drinking water. The two sides of the managed water cycle are becoming closer with the increasing recognition of the need for considering wastewater re-use as a means of conserving drinking water resources. In addition, there is increasing pressure on wastewater quality for environmental reasons. The traditional use of surface water as a source of drinking water and the indirect re-use of wastewater by discharge to surface waters continues to pose challenges. While heavy metals were once a concern, the new issues are hormones and endocrine disruptors, pharmaceutical residues and personal care products.
Much of what we use in a domestic setting may finally be present in wastewater, and many of these compounds are of importance for modern life. Control at source will, therefore, be difficult. Collection and treatment of sewage has played an important role in defeating waterborne disease; however, this also means collecting and treating wastewater that contains excreted natural hormones and the excretory products of pharmaceuticals. While removal of the more lipophilic compounds, such as the hormones, is relatively easy, a significant number of these substances are polar and may be less susceptible to traditional drinking water treatment.
One of the problems is understanding the range of substances that can be present and determining the likely concentrations that might be in abstracted water. Already there are calls for a precautionary approach, even though the data so far do not indicate any discernible risk to consumers. In addition, the range of substances is constantly changing as new compounds are introduced and as there are changes in prescribing fashions and in the pattern of purchase of over-the-counter preparations.
The challenge also relates to consumer concerns and perceptions, often exacerbated by media exaggeration and hyperbole. It is not practical to demonstrate control by measuring an ever-increasing number of substances in final water, so we need to demonstrate what particular treatment processes, and combinations of processes, can achieve in terms of removal. It is also important to develop measures for operational control, to demonstrate that treatment is working at its optimum at all times and that, if problems begin to arise, they will be identified at an early stage and corrected. This reflects the change in approach, from the retrospective monitoring that has grown up over the past two decades to the more proactive approach favoured by WSPs.
Other challenges are emerging that require a more holistic approach. Research into the health effects of the unwanted by-products of the reaction between disinfectants and, usually, naturally occurring organic and inorganic matter continues to raise questions. For many years the concern was potential carcinogenicity, which looks increasingly unlikely to be a significant issue. Latterly, though, research has focused on the potential for adverse birth effects, including low birth weight, stillbirth and malformations. The outcome of this research is still uncertain, but finding means of lowering by-products without exchanging one hazard for another is an important requirement. In Europe we have a standard for total trihalomethanes but WHO has guidelines for a wider range of by-products. It is still not entirely clear how good, or bad, THMs are as an indicator of the by-products, and we may need to look more closely at others such as the haloacetic acids.
The dangers of over-reliance
Although many of the epidemiological studies have used THMs as an indicator, they are not the only by-products that are implicated. Certainly there is a need for relatively urgent focused research on this topic. This is another example of the danger of too much reliance on statutory standards without sufficient consideration of the wider issues. However, such thinking also requires that regulators are able to work within a sufficiently flexible framework so that the investment needed is properly considered and allowed for.
One new potential threat requiring research is n-nitrosodimethylamine (NDMA), which may be formed during chlorination. This is a probable human carcinogen of some potency. Early consideration of the risks indicates that if standards or guidelines are required, these are likely to be in the region of a few nanograms per litre. The formation may be particularly pertinent for chloramination but there are only limited data on the likely occurrence in drinking water and on the conditions under which nitrosamines may form. Such understanding is vital for a proper and proportionate response.
A perennial problem facing water suppliers using surface waters is that of blue-green algae. These organisms are capable of forming large and rapid blooms in still or slow-flowing waters when the conditions are right. Many species are also capable of producing toxins as well as substances that can interfere with coagulation or that give rise to unpleasant musty tastes in drinking water. One problem is the range of toxins; the list grows as more research is undertaken. The only toxin for which WHO has produced a guideline value is microcystin-LR, but this is not the only toxin of concern. While research continues on how toxins can be removed in drinking water treatment, there is a need to develop novel approaches to preventing blooms in drinking water sources. The range of toxins includes a number of diverse chemical structures, and there is little doubt that further toxins will be identified. This is a classic example of a problem that constantly changes, with a need for a response in terms of prevention and, where prevention is not possible, remediation through treatment.
In spite of the enormous progress made over the past 25 years, drinking water supply and wastewater treatment face a range of challenges. Climate change will bring a range of new hazards and problems, such as changes in the toxins from cyanobacteria, but we will find that many are not easy to predict. Pressure on water resources will inevitably lead to changes in the way we manage the water cycle - already wastewater re-use is being considered very seriously as one of the solutions, bringing the problems of drinking water and wastewater closer together.
Microbiological threats from pathogens that are emerging as potential waterborne problems are still an issue, particularly for small supplies. We are still uncertain as to whether some pathogens are a threat from drinking water, and the move to wastewater re-use will uncover new microbiological problems. The introduction of WSPs will result in the identification of problems that have not been considered previously, and one area that will be affected will be distribution and the operation of distribution systems. New chemical contaminants continue to be identified, even though some may have been present for a considerable time. These can arise from natural sources, from wastewater, or as an unwanted and unexpected by-product of the water supply process. The risk assessment process in developing WSPs will also reveal chemical threats of which we may have been previously unaware, and will create a requirement for new and better means of continuous operational monitoring to control water systems. There will, therefore, continue to be a need to develop new and flexible approaches to dealing with these contaminants and problems - approaches which will permit continued delivery of water that is safe, acceptable and affordable.
This article is based on a paper presented at Cranfield University's Developments in Water Treatment and Supply conference, which took place in York, July 2005.
© Faversham House Group Ltd 2005. edie news articles may be copied or forwarded for individual use only. No other reproduction or distribution is permitted without prior written consent.