by Adam Hinterthuer
Early next year, as soon as Madison’s lakes thaw from their winter freeze, CFL research professor Paul Hanson and graduate student Sophia Skoglund will start scanning the weather forecast for a calm, sunny day.
And, when conditions are just right, they, along with a handful of other UW-Madison research labs and one Wisconsin Department of Natural Resources (WDNR) airplane, will head out on – and above – Lake Mendota to launch an ambitious new project.
Dubbed “LakeView,” the project will harness the power of hyperspectral imaging to help researchers get a clearer view of water quality dynamics in Lake Mendota, Lake Wingra and a couple of small stormwater ponds in the Madison area.
Hyperspectral imaging is essentially a way to use a high-tech camera mounted to a satellite or airplane to produce extremely detailed maps of a landscape. It has been used for years by scientists like geologists looking for oil deposits or agronomists wanting to precisely apply pesticides or fertilizers to fields.
But, says Hanson, the technology hasn’t made the same waves in bodies of water – especially inland lakes. And that’s something he and his team hope to change.
“Inland lakes have this fundamental problem in that they look like black, featureless bodies with current imaging, so they’re typically not analyzed in the same way as terrestrial systems. All of the algorithms scientists have created to interpret these images are for terrestrial or marine systems, not lakes.” That, Hanson says, is where LakeView comes in.
Their plan is to have the WDNR airplane – outfitted with hyperspectral imaging equipment – fly between 15 to 20 missions next year during conditions of clear skies and calm water. At the same time the plane is up in the air looking down on the lake, research crews will be out in boats collecting real-time data on everything from water clarity to dissolved oxygen levels to the different species of algae present in the lake at that specific moment in time.
“When we look at the surface of a lake, we might note the clarity or color and then wonder why it looks that way,” says Sophia Skoglund, a first-year graduate student working on the LakeView project with Hanson. Different things in the water reflect and absorb light at distinct wavelengths, Skoglund explains, and those differences create the colors we associate with things like the green of an algae bloom or brown, muddy tint of sediment erosion.
As the WDNR airplane is flying overhead, Skoglund will be out on the lake collecting data so that scientists can see how the images the LakeView camera takes correspond with real-time conditions in the water. By comparing the data they collect from the lake to the hyperspectral images, the team will be able to develop computer algorithms that can “read” the images and capture what is going on in the water. Eventually, says Skoglund, whole teams of researchers like hers won’t need to be out on a lake to get real-time data. One satellite or airplane fly by will be enough to get a sense of current conditions.
The LakeView project is designed, in part, to “unlock the power of upcoming hyperspectral satellite missions,” planned by NASA over the next one to five years, Skoglund says. Those flights will greatly increase the number of lakes captured in these giant global imaging datasets. If researchers can learn beforehand how to read what those images are telling them about the condition of lakes across the world, it will be a powerful tool.
“We have information on probably only one percent of U.S. lakes, so this is a huge opportunity to do a much broader scale assessment,” Hanson says. “The problem is we don’t know how to interpret the data yet.” But, thanks to LakeView, he says, “that’s where we’re headed.”