Scientists have collected boatloads of data about Lake Minnetonka for many decades, but one question has always been elusive: How does the sprawling lake behave?

A University of Minnesota scientist is providing new answers that he and others say will pinpoint lake problems and identify the most cost-effective ways to solve them.

Shane Missaghi, a graduate student in water resources science, spent the past three years calibrating a computer model to understand one of Minnesota's most complicated lakes. Unlike a more traditional bowl-shaped lake, Minnetonka's 26 bays and 125 miles of shoreline make it seem like many separate lakes, with water quality that varies from excellent to very poor.

Missaghi's work tries to make sense of it all, looking at how each bay interacts with other bays and with the lake as a whole. It can track parcels of water in three-dimensional blocks as they move through the lake, and how they change in clarity, temperature and other features and as they flow from the more polluted western areas to the cleaner and deeper eastern half.

"We can look at it from east to west, or top to bottom, and we can see the water quality in any part of the lake," Missaghi said. He can also take a section of lake that has good fish habitat, and check over time to see whether it shrinks or expands as wind, rain, water flow and sedimentation affect it.

The customized model is already producing results.

It has shown that inflow of water from Six Mile Creek, Painter Creek and other streams into Minnetonka plays a larger role than previously thought in stirring up the water in Halsted Bay and other locations, and adding nutrients such as phosphorus that feed unwanted algae.

"We had an idea that creeks influenced the lake, but now we can actually pinpoint where to make improvements along the creeks," said Kelly Dooley, water quality technician for the Minnehaha Creek Watershed District.

Researchers know that planting vegetative buffer strips or rain gardens or restoring natural shorelines can reduce harmful runoff, but Dooley said the key is knowing where to do that to achieve the maximum benefit for the lowest cost. Having a more sophisticated notion of Minnetonka's dynamics is a major advance, she said. "You don't want to put taxpayers' money into an area that may not actually be beneficial," she said.

Missaghi has used the district's water quality data on Minnetonka to fine-tune the accuracy of his model. He also mapped the bottom of Minnetonka more precisely than has been done before.

The model could be used for other lakes in the state because certain climate factors are common, he said, but it would need to be adjusted for each lake's dimensions, watersheds, wind patterns and other characteristics.

Model is just the start

Karen Jensen, senior planner at the Metropolitan Council Environmental Services, said that having a computer model for Minnetonka, while important, is just the start. The real benefit comes from using that model to test hypothetical changes, she said.

An accurate computer model allows planners to try out different solutions in select places, Jensen said, and to predict what would happen.

"Let's say you want to investigate where to put in rain gardens, and how big to make them," she said. The cost of buying land and planting the gardens is known, the computer can estimate the benefits in reduced runoff, and decision-makers can determine whether it's worth the money, she said. "Otherwise, there's not a lot of guidance on where to spend money so it'll do the most good for the lake," Jensen said.

Computer models have become steadily more sophisticated during the past 20 years as computers have become more powerful, Jensen said. But she warned that they're only useful if they've used actual monitoring data from years of water sampling to confirm their accuracy.

Missaghi said models are most helpful with the physics and chemistry of how a lake behaves, and less so in understanding some biological processes such as algae growth that can be more complicated.

Knowing the dynamics of Minnetonka may also bolster fish stocking success, Missaghi said. Based on the timing and weather conditions, it's possible to estimate where the flow will take fingerlings released in a certain area, he said, and where they'll end up in two or three weeks. That would be the optimal place to restore natural shoreline and aquatic plants, he said, to provide habitat as young fish grow.

Besides the more immediate applications, Missaghi said he'll be using the model to predict "what-if" scenarios of future weather, such as how Minnetonka will behave in years of drought or excessive rain, or if average air temperatures rise by 2050 as predicted by some climate-change researchers.

"This has become a cornerstone for lake management," he said. "We've got to have it."

Tom Meersman • 612-673-7388