POLSON — As the days shorten and the temperatures drop, Flathead Lake is beginning its fall “turnover,” an annual process in which cold, dense water begins mixing nutrients into the more biologically active layer of warm water on top as lake temperatures even out.

From a waterfront office in the Flathead Lake Biological Station, postdoctoral researcher Shawn Devlin is working to develop a highly sophisticated model of the lake that will be able to predict the impacts of turnover and other processes within it — revolutionizing scientists’ understanding of the largest natural body of fresh water west of the Mississippi.

“I was tasked with a whole-lake picture, in terms of how we can use our long-term dataset, 35 years’ worth of data that has been collected over the years on pretty much all the components of the food web below fish,” Devlin explains. “It’s a really big task to try to put them all together, to try to link them in a real way.”

Yet he’s well on his way, armed with an enviable trove of scientific data, meticulously documented since 1970 by outgoing station director Jack Stanford and his teams of University of Montana students and professors.

Flathead Lake is one of the most-studied bodies of water in the world, renowned for the quality of its water, the diversity of wildlife and natural processes affecting it and the relatively pristine condition of much of its sprawling basin. But despite a rich history of groundbreaking research, scientists remain with many questions unanswered.

“For example, we know that in the system, nitrogen deposition and nitrogen loading has been going up over the past 25 years, but we haven’t seen a result of that that we can really link to this process,” Devlin says. “We know it’s happening, so one of the things we want to look at is why this isn’t having an impact, or if it is, where in the food web it’s happening.”

The model on which Devlin has spent most of his time thus far is focused on the physical processes at work in the lake: variations in the Flathead River’s flow rate, the depth and geological features of the lake bed, air temperature, wind speed and direction, humidity levels and the amount of sunlight hitting the waters’ surface.

It also takes into account that Flathead Lake is so large that its currents are affected by the earth’s rotation — known as the Coriolis effect, which also determines which direction toilets drain in the northern and southern hemispheres. In the lake, it causes a pair of massive gyres to form periodically, churning cooler water up toward the surface.

A second model will be used to make predictions about the organisms living in the lake, examining biochemical processes that drive the populations of single-celled plankton and their ripple effects all the way up through the food web.

As Devlin puts it, the physical model “sort of sets the stage, and [the food web model] is the script, it tells us who is playing and how they interact.”

The biological station operates four data-collecting buoys on the lake, that continually record meteorological data. When he plugs those inputs into the physical model, Devlin can generate a three-dimensional map that plots the temperatures throughout Flathead Lake. He says that on-the-ground testing has shown that 97 percent of the time, the modeled results have proven accurate within one degree.

“When papers are published on [the Estuary, Lake and Coastal Ocean Model], they’re really happy if they have 70 percent agreement,” he says. “And what’s most important to our work is temperature... that’s really important for the biology out there, because temperature predicts different rates of production and different rates of uptake and regulates the microorganisms out there in a big way.”

The importance of temperature in the lake is borne out throughout the year.

As the ice recedes and spring ushers in increasing sunlight and rising mercury, the process of stratification begins. The warmer, less dense water remains on top of the colder, denser water, and that division becomes increasingly rigid as the lake’s surface temperature peaks in the low 80s by August.

“As that density increases, the mixing between the two layers becomes basically nonexistent, and that has big impacts as far as nutrients,” says Devlin. “There’s a set amount of nutrients [in the top layer], and as the algae grow they deplete the nutrient pool... Essentially, they get blocked off from most of the nutrients in the lake, and that’s one of the big reasons Flathead Lake stays so pristine throughout the year.”

While seasonal variations in air temperature and humidity are fairly easy to predict year to year, he noted that wind is the most variable input. A strong wind event displayed graphically on the model shows warm water pushed to one side of the lake, ultimately forcing the less dense water down and churning up colder, nutrient-laden water from below.

Even so, he said his model is already sufficient to use those variations to make basic predictions of how changes in global temperatures will affect the physical conditions of the lake. Devlin has already used his lake model in combination with existing climate change models (20 of them) to calculate possible scenarios 100 years into the future, generally divided into “optimistic” and “dire” predictions.

“What we see in the optimistic perspective, the lake is going to stay pretty cold down deep, but what we do see is a substantial amount of the surface is going to increase in temperature,” he says. “Over those next hundred years we’ll see substantially warmer water in the surface layer and this summer is a relatively good window into it.”

The effects could extend beyond a more agreeable swimming season. Based on his predictions, warm temperatures would extend deeper into the water column and persist for a larger chunk of the year, reducing the available habitat to cold water dwellers like lake trout by an estimated 15 percent.

“What’s important about that is what 15 percent they are losing,” Devlin points out. “The 15 percent is in the illuminated surface waters, where all the phytoplankton and zooplankton and most fish live. So they’re really losing a pretty substantial portion of their habitat, the most important portion of their habitat in some regards.”

As he continues to hone his sophisticated tool, Devlin said his ultimate hope is that it can be used to more accurately develop policies and regulations, such as maximum nutrient loading levels set by environmental agencies.

“I’d like to make this available as an academic tool, and also to help inform policy. I’ve had some communications with the [Montana Division of Environmental Quality],” he said. “That’s the perfect example of where science can help inform policy, and then everybody benefits.”

As the food web model becomes more developed, Devlin says he will also be able to predict, to an extent, what would be the biological impacts of lake trout removal, or of zebra mussel introduction or of a chemical spill from a truck crash or train derailment.

“Of course, that’s something that we never hope to see, but that’s the beauty of the model — it allows us to come up with these experiments that we could never do, like dumping a bunch of phosphorus into Flathead Lake.”

Aside from the considerable time investment in developing the lake model, it takes a lot of expensive resources as well. An anonymous $1 million donation, which was provided as matching funds to incentivize a community-driven donation drive, helped pay for his powerful desktop computer and other equipment.

“The temperature calculations alone are about seven hours with a really fast computer, and when you add [the ecological model software] it’s going to bump up to about 10 times that,” he said. “We really depend on private funding here, and the fact that this community really cares about Flathead Lake.”

Reporter Samuel Wilson can be reached at 758-4407 or by email at swilson@dailyinterlake.com.