At Pamela Park in Edina, Chris Wennen , a graduate student in water resources science at the University of Minnesota, stands knee-deep in a pond, examining what resembles an enormous film canister. The device is a wireless sensor that monitors water quality. It is one of five implemented in the Minnehaha Creek area. Wennen said the sensors measure the levels of pollutants such as nitrate and chloride, as well as the amount of oxygen, pH, temperature, turbidity and depth of the water. âÄúThese are the first wireless sensors to record runoff data in an urban area,âÄù said study leader and civil engineering professor William Arnold. Arnold and colleague professor Miki Hondzo have been working on the project for the past three years with help from graduate students, including Wennen. The sensors take measurements every two hours and every 30 minutes during a rainstorm, when runoff is highest, Arnold said. The data is recorded in a solar-powered box on the shore and sent via cell phone to the UniversityâÄôs computer lab in St. Anthony Falls. There, water experts analyze the data and predict where and when pollutants âÄî including pharmaceuticals and pesticides âÄî will hit. Before sensor technology, âÄúgrab samplesâÄù were taken physically twice a month. By providing researchers and recreational water users with up-to-the-minute statistics on the water supply, Arnold and his team hope to protect human and animal populations. Arnold is currently working to collaborate with a similar project in Iowa to compare runoff and water treatment practices in urban and agricultural areas. âÄúThese are vital statistics for human health, and vital statistics of the environmental body,âÄù said Deborah Swackhamer , program director of the Water Resources Center at the University. âÄúThe sensors let you assess how you are managing the resource. They allow us to access water at a finer scale.âÄù The close examination has yielded surprising results, Arnold said, particularly in stormwater ponds, which collect runoff from streets and yards and filter it through the soil. The stormwater ponds being monitored include two near Minnehaha Creek and at a second site near Shingle Creek . Sensors have detected caffeine at the sites, an indicator of sewage leakage. âÄúThe caffeine is a marker that other, more dangerous, pollutants may be leaking in,âÄù said Wennen. âÄúHormones like estrogen could feminize the male geese, birds and fish so that they are unable to reproduce.âÄù Although the detection of caffeine is worrying, it proves that the ponds are fulfilling their function by taking out pollutants. âÄúWhat we found was surprising. The stormwater ponds remove pollution we didnâÄôt expect them to,âÄù Arnold said. The sensors will change location in six weeks, said Arnold. The next location to be examined is where Shingle Creek meets the Mississippi River. By looking at various locations, researchers can monitor leaks in sewage lines and recognize what preventative methods are working. The research group plans to expand from five sensor sites to 100 in the next five to 10 years. âÄúThe challenge is always funding,âÄù said Wennen. The project has received a total of $592,000 in funding from federal and local agencies and the University, Water Resources Center spokeswoman Nina Shepherd said. The researchers are hopeful for the expansion of the project. âÄúThis is a pilot program,âÄù Swackhamer said. âÄúIn the future, this is how monitoring will be done.âÄù