Water quality matters

Dragan Savics’ speech at the 71st Vakantiecursus in Delft

Last Friday I delivered an invited talk on “Quality matters” during the Vakantiecursus day at TU Delft. The traditional annual meeting of who’s who in the Dutch water sector provides an opportunity for people to meet colleagues, discuss various issues, exchange best wishes and help start the new year on a positive note. I heard that this year’s attendance exceeded more than 350 water professionals and that the meeting was the 71st time people gathered in early January to kick-start another successful year for the Dutch water industry.

As the title of my talk suggested, I was asked to talk about water quality. We in the West often take our drinking water for granted. Most of us only think of water when there isn’t enough of it (i.e., droughts), when there is too much of it (i.e., floods), or when we are worried about its quality and its potential impact on us and our loved ones (i.e., health). A prime example of the latter and how things can go awfully wrong is the city of Flint, Michigan. The situation started unfolding in 2014 when some bad decisions and money-saving measures caused one of the worst public health crises in the United States. Flint is the city where the great car maker General Motors was born more than 100 years ago, which was because of that called “Vehicle City”. However, since its heyday in the sixties and seventies, the car industry has been deserting Flint, and the city’s economy declined so much that in 2014 a decision was made to change the supply of water to Flint from Lake Huron (and the Detroit Water and Sewage Department) to the cheaper Flint River. This decision led to an unanticipated situation where insufficient treatment of Flint River water, which is of different quality to the Lake Huron water, caused lead service pipes to release their lead into the tap water. The highest lead level recorded in Flint was 13,000 parts per billion (ppb) in 2015, far higher than the federal guideline of 15 ppb (10 ppb in the EU). Needless to say, there were health problems associated with the deteriorating water quality. There were and also are still a number of ongoing lawsuits against Flint and state officials. Fortunately, this is a rare example where the governing system failed to uphold people’s right to access to clean and wholesome water. However, the most important outcome is that water lead levels have improved considerably in the last few years.

Although the overall situation with water quality in the Netherlands is good, there were a few issues in the past with chemicals and compounds released by industrial processes, e.g. Pyrazole in the River Meuse (2015) and GenX in the surface water around Dordrecht (2016). The situation with Legionella is also something that has been closely monitored, especially since the serious outbreak in Bovenkarspel in 1999. RIVM reports a rising incidence of diagnosed Legionellosis in the Netherlands over the recent years, with 2017 being the year with the highest number of incidences ever reported. Antimicrobial resistance is a rising concern as the role of water as a possible reservoir and melting pot for gene transfer, and the emergence of new resistance genes is not fully understood. Among the scientific “known unknowns” and “unknown unknowns”, there is also a growing problem related to these persistent and bioactive contaminants – micropollutants, including pharmaceuticals,  pesticides, cosmetics, industrial chemicals, microplastics, to name but a few.

Now two questions could be posed: 1) What can water utilities do to avoid such situations and 2) What is the current state-of-the-art in assessing how good is the tap water we are drinking?

To answer the first question, we can learn from the Flint experience. It seems obvious that, at the time of the switch to Flint River water, a proper investigation had not been conducted on how the Flint water quality could impact on the existing water infrastructure. If sound research had been done, there could have been steps taken in the treatment process to avoid the release of lead into the water. The key failing is probably that of governance. By strengthening regulation and governance processes, a situation like Flint could be avoided. It is interesting to note that in a research conducted by Chapman University, among the ten top fears in 2017, about 50% of American people reported being afraid or very afraid of the pollution of drinking water while the top fear was a corruption of government officials (74%).

The answer to the second question is a bit more complex. There is a big shift in the monitoring of water quality and the push to know both the “knowns” and “unknowns”. For example, monitoring of microbial water quality has moved a long way from plate count methods and the use of a microscope to DNA profiling. An example of such approaches is the rapid RT-PCR method for detecting E.coli presence in drinking water. The new method provides answers within 4-6 hours, which is much faster than the standard plate count that takes several days. It is not surprising that the project to develop this method won the latest BTO (the Dutch and Flanders joint research programme for the water utilities) Innovation prize during the annual BTO meeting in November 2018 at KWR. Monitoring of chemical water quality has also evolved to cover not only target analysis, but also non-target screening. High-resolution mass spectrometry (HR-MS) is becoming a tool of choice for non-target screening and with the use of ToxCast (US EPA’s Toxicity Forecaster) to prioritise for suspect screening data combining occurrence and hazard data. The approach allows automation and high throughput.

If we are to prevent situations such as the one that occurred in Flint, we need to:

  • Consider the entire system (e.g., governance, economics, science and engineering), rather than focus on one of the components and potentially miss a crucial risk.
  • Use more data for analysis by taking advantage of the proliferation of sensors, sensing technologies and broader screening.
  • Invest in better data analysis.
  • Use both scientific approach and common sense.