Dead Daphnia flag mystery water contaminants
Researchers in The Netherlands show how water fleas can detect new toxic compounds in river water.
On March 19, 2004, alarms went off at the Evides water company’s monitoring station on the Meuse River in the south of The Netherlands. Half of the Daphnia water fleas in the station’s aquarium had died. The search for what caused the deaths—a true detective story—is described in research posted today to ES&T’s Research ASAP website (DOI: 10.1021/es052035a). The research shows that combining continuous biomonitoring with chemical analysis can help utilities to detect toxic mixtures of contaminants that conventional technologies would miss.
The water fleas in Evides’s aquarium are akin to canaries in a coal mine—their deaths are a quick indicator that something could be seriously wrong with the water quality. Evides immediately stopped pumping water from their reservoir 9 kilometers downstream from the monitoring station on the Meuse. Evides provides drinking water to 1.3 million people and uses the reservoirs for storage.
Although the biomonitoring systems managed by Evides and other utilities generally trigger alarms three or four times a year, causing temporary pump shutdowns and mostly uneventful inspections, this was the third alarm that dying Daphnia had set off in just over a month. Evides’s technicians were befuddled as to the cause and handed over the case to experts at Kiwa Water Research, an institute in Nieuwegein (The Netherlands).
When standard analytical methods failed to detect anything unusual in the water samples, Kiwa researchers turned to the most sophisticated tool in their arsenal. Using quadrupole time-of-flight mass spectrometry, they determined that the culprit in the Daphnia deaths was 3-cyclohexyl-1,1-dimethylurea, a compound they had never seen before in the Meuse.
“This is one of the few success stories,” says Ariadne Hogenboom, an analytical chemist at Kiwa and the paper’s coauthor. Researchers at Kiwa handle water contamination cases for many countries in the EU, but they are not always successful in identifying previously unknown contaminants.
However, the mystery wasn’t solved yet. The researchers found a maximum of 5 micrograms per liter of the urea in the water samples in which the Daphnia died. But when exposed to the same levels of the compound in follow-up tests, the tiny crustaceans did not seem to be affected by it. So the researchers looked at the samples more closely. “We found that the compound was never alone in water; it was found either in combination with HMMM [hexa(methoxymethyl)melamine] or isoproturon,” says Corina de Hoogh, the paper’s lead author. This suggested that the Daphnia responded to the toxic effects of these compounds together, she says, even though “these compounds may be in very [low concentrations], so they are not even detectable.”
The researchers did not get a chance to test their suspicion that the compounds’ mixture was toxic, but no other explanation for the Daphnia’s deaths seems to exist, de Hoogh says. And without the combination of continuous biomonitoring and chemical analysis, they would never have known about the deadly mix present in the water. This is why it is crucial for water companies to use the techniques together, she says. “What we can detect [with chemical analysis] is a fraction of what we find in surface water,” she says, but biotests clearly indicate when something more needs to be done. “Both techniques are not really new; it’s the combination that’s unique.”
Joel Allen, an environmental scientist at the U.S. EPA, agrees. “The uniqueness is that they were able to take an alarm from an on-line toxicity monitor and translate that into a specific compound that was causing the toxicity,” he says. This process occurs rarely, he says. The use of on-line biomonitors is limited—the method is barely used in the U.S., although Allen is trying to change that—and not every alarm is worth pursuing. The Kiwa research demonstrates how the combination of on-line monitoring and analytical tools can be effectively used, he says. “It starts to help build the case that these systems, when they are deployed, are doing what we expect them to do.” Biomonitoring is currently used in Germany, France, and Italy, de Hoogh says. EPA is researching the technology, Allen adds.
For all its benefits, the combination system hasn’t helped investigators find the source of the contaminant. Because of the urea’s patented uses and the HMMM levels that accompanied it, the Kiwa researchers speculate that it came from a hydraulic fluid drainage.
At this point, not much else can be done, Allen says, because a one-time contamination event is not enough to trigger remedial action. According to Walter Giger of the Swiss Federal Institute of Aquatic Science and Technology, water monitoring data must show many occurrences of a pollutant before it can move into a regulatory framework, unless it is known to have a high toxicity.
Right now, Kiwa researchers are investigating another series of alarms at Evides’s monitoring station, and de Hoogh thinks that they could be caused by a similar culprit. “If we see it again this year between March and May, and again next year, then it’s more urgent to look for the source,” she says.
