Watching Water Quality

E-mail Melanie Martella

Water is a precious resource (although we in New England would like to request a stop to the excessive rain we've experienced so far this summer) and maintaining the quality of our water supply is extremely important. Monitoring and security the water supply is important not just for public health reasons (epidemics of cholera and typhus swept the U.S. in the 1880s, killing tens of thousands of people), in the wake of the September 11 attacks, the need to protect against malicious tampering came to the fore. Now, a software suite developed by the EPA is being used to help spot contamination.

Meet the CANARY
"Regan Murray: Advancing Crucial Clean Water Standards" in the Washington Post describes how Regan Murray, an EPA statistician, and a team of government scientists, developed the CANARY software suite to provide real-time, automated analysis of water system sensor data. More specifically, it can spot recurring patterns in the data and identify abnormalities. In addition, they've developed another program, TEVA-SPOT, that identifies the optimal sensor placement, within a water system, to register contamination. Although the impetus was the potential for terrorist assaults on the water supply, contamination from other sources does happen. For instance, excessive rainfall can flush more crud into the water supply. The software analyzes the normal water quality parameters and uses historical data to determine the water quality's normal range of variation over time.

This sounds, to me, like the condition-based monitoring used for machine health. By monitoring vibration on the machine you get an early warning of trouble ahead if things start trending out of normal, allowing you to take preventative or ameliorative measures. Similarly, these water quality monitoring approaches don't just monitor the water quality—they can give you information on the health of the water supply system itself, which is also valuable.

I suspect, further, that having these combined sensor/software systems in place provides a platform for future adaptation and extension. Pharmaceuticals and other trace contaminants are finding their way into our water supplies and, as far as I know, nobody has come up with a good method for monitoring or removing them.

Suggested Articles

Iowa State University researchers are working with NSF grant

Brain Corp. reported a sharp increase in autonomous robot usage in 2Q

Nvidia DGX accelerators helped train system from 150,000 chest X-rays with inference results in less than a second