The World Health Organisation and its partners have set guideline values and health-based targets, helping define what concentration of hazardous chemicals and pathogens constitutes a ‘safe’ concentration for human consumption. So we have an agreed definition of ‘safe’ water, but what do we actually mean by ‘safe’ and how safe is our water?

Unfortunately, most water quality objectives are not routinely measured, so we don’t really know if our water meets those goals. The problem for most chemical, radiological and physical water quality parameters is that they are often costly to measure; as a result, most potential contaminants are not measured at all in most water supplies. Even those that are measured can only be measured occasionally.

The problem is even more complex for pathogens.

We cannot routinely detect pathogens at concentrations that are usually considered low enough to be safe. Simultaneously, pathogens are acute in their action, making gaps between testing much more problematic than they are for usually much slower-acting chemicals. We rely on E. coli, and sometimes other microbial indicators, but the relationship between those indicators and pathogens is imperfect, particularly in treated water.

In both cases, sampling programmes rarely have their monitoring frequencies determined by a first principles statistical analysis, to determine what would be required to provide us with adequate assurances that hazardous substances were below guideline values. Most monitoring programmes involve somewhat arbitrary weekly, monthly, quarterly, annual, or even five-yearly testing frequencies. Furthermore, recontamination during distribution of water can occur and the vast majority of taps and containers are never sampled.

The Water Safety Plan (WSP) approach goes most of the way to addressing these issues.

Water Safety Planning promotes a detailed system understanding and qualitative risk assessment to determine what contamination might be expected; coupled with the implementation of reliable barriers to prevent the transmission and consumption of such contaminants. But we still have some gaps when it comes to the detail of WSPs.

We are often faced with having to decide just how effective our barriers need to be. For example, we need to think about how much chlorine is enough; how low our filtered water turbidity should be; and how often we should inspect water storages or test backflow prevention devices. The problem is that more rigour and a more quantitative WSP entails more cost, and that creates a need to justify those costs.

Water safety varies between systems

Smaller, less sophisticated and less centralised water supply systems typically have much less invested in their WSPs than better resourced systems, making the estimates of what is required to make the water ‘safe’ even less quantitive and evidence-based. The safety of water is likely to vary between systems simply because some are more conservative than others, but we often don’t have the evidence to know by how much.

If money were unlimited, the solution to this problem would be to conduct extensive contaminant monitoring to support a more quantitative assessment of risks and controls in developing WSPs. A much more affordable approach, that can get us close to that goal and help us decide how safe our water really is, involves estimating contaminant concentrations in particular types of water using data from comparable sites and locations. But there aren’t many comparable sites from which data is readily available at present.

In most countries, there hasn’t been a systematic source water quality characterisation programme to help answer these questions. Under these circumstances we need to consider how conservative we should be. We need to decide whether, and how much, to invest in source water characterisation, and how far to go with respect to the effectiveness and reliability of our barriers. To some extent, we are guessing how safe our water is because we don’t really have the evidence. Even the best WSP can only make the water adequately safe if we truly understand what the risks are.

Pooling data would help us focus on areas of greatest need

Hopefully water professionals can find ways to work together to share and pool our, often very closely guarded, water quality data. We could make good reliable water quality monitoring datasets much more readily available. We could build on this body of existing data by filling gaps through systematically designed and targeted monitoring programs to help us to better characterise water from logical categories of types of water, focusing on where the greatest data needs are.

Doing so would help us provide a much more quantitative evidence base to support our WSPs so that we are better able to answer the question: How safe is our water?

Addressing the challenges of water safety, and sharing innovative solutions related to water and sanitation safety planning, are prime objectives of the Global Water Safety Conference, taking place in Palawan, Philippines, from April 25-28, 2016.

Don’t miss the newly launched Water Safety Portal – your online network to support implementation of Water Safety Plans (WSPs), where you can find resources, share experiences, and keep up-to-date with news and events: https://www.wsportal.org/