Researchers deploy water quality forecasting system around the globe

Our water resources are undergoing kind of unprecedented demand, and they're also experiencing increasing variability in response to a suite of pressures. Our Center for Ecosystem Forecasting team has developed kind of a modeling system forecasting framework which we can easily apply to a suite of different lakes. One of our kind of closest partners is with the Western Virginia Water Authority in Roanoke, Virginia. We give them every morning forecasts of the water quality in three of their drinking water reservoirs. By working with Virginia Tech and getting information telling us what is going to happen, not what has happened, it helps us to prepare for what type of treatment methods we need. How is the best quality water going to be delivered to our customer? Society has come to really use weather forecasting to guide decision making. And that recipe, we apply to water quality and other aspects of the environment. So that leads to the Center for Ecosystem Forecasting needing to have a strong data collection component where we put sensors in the environment that collect real-time data so that now we have what we would call a digital twin, a version of, say, a water quality reservoir that's running into the future and providing insights on where the water quality is headed over the next few days to multiple weeks in the future. We are now making water forecasts for 15 lakes and reservoirs around the globe, both in the United States, in Ireland and in Australia. And each of those partnerships, we're developing forecasts that are specific for each of the decision-making needs. The main thing that we do in this research station is monitor the migration of diatribus fish. We have fish traps that count every single fish, so we know exactly how many fish are moving, we can count them. But the idea would be to use these forecasts that we can develop in this system and validate them and then move them to catchments for areas that don't have the infrastructure that we have. I think that's where it can be really useful. Adelaide as a city has about 1.4 million and all of Adelaide's drinking water comes from the River Murray. If we can get a handle on the likely response of the lakes in terms of water level and salinity to incoming flows from the River Murray, and then we can make daily decisions about what we might be able to release out of the barrages because we can choose to open and shut gates. It's very powerful to have a bit of insight as opposed to being reactive all the time. We can be a bit more adaptive. We're training the next generation of forecast producers and also the forecast users. Over 20,000 students have learned about the concept of environmental forecasting through modules that have been developed through the center. We're bringing together researchers from data science and statistics and decision science, geosciences, ecosystem modeling, a range of different researchers coming together that I think is a very special contribution of being here at Virginia Tech, but also it's kind of the work that's needed and the expertise that's needed to be able to make forecasts that are truly useful and meaningful for communities. It's really exciting that we get to work at the intersection of application and innovation. And being at an academic institution gives us that freedom to take on really hard problems that may work or may not work, but someone needs to really try them.