It’s been a tough century for fish on Wisconsin’s Bayfield peninsula.

Their problems started when the region was logged in the late 1800s. Without the shade of a forest canopy to slow the pace of spring thaws, trout streams surged with melt water, overwhelming newly hatched trout and depositing layers of sand over the gravel beds where fish spawn.

But even when forests returned, the fish continued to struggle. The spring floods persisted, turning trout streambeds into “a sand desert,” explains Dennis Pratt, a fisheries biologist with the state Department of Natural Resources.

So why didn’t the return of trees help save the fish? To answer that, Pratt turned to CALS forest ecologist David Mladenoff MS’79 PhD’85, who began to suspect the reason for the rapid melt was in the trees.

“I spent a lot of time thinking about what could be different about the pre-settlement forest, and the big thing was the big evergreen conifers that had dominated the landscape,” Mladenoff says. “We lost those large conifer trees like white pine, red pine, hemlock and cedar.” In their place, aspen and other trees grew, making for a younger, more open forest. Mladenoff theorized that the conifers’ dense branches and evergreen needles kept snow from reaching the ground and created more shade, extending the time needed to melt off the snowpack.

To test the theory, graduate student Jordan Muss picked 33 sites on the peninsula where relatively pure stands of nine tree species live. For the next three winters, he visited each site to measure the snow on the ground and calculate how much water it held. He says he found significantly more snowpack underneath deciduous species than under the conifers, supporting the notion that the amount of snow on the ground is directly linked to the density of the canopy above. Mladenoff plans to incorporate the findings into a model that predicts how forests change over time under certain management practices, which could help forest managers identify species or cutting strategies that lead to less snow on the ground and a longer melting period.

“When any resource manager has a tool like that model, it gives them better options to set land management goals that also protect the fishery,” says Pratt. “This tool is giving them a better opportunity to both set that goal and get there. It’s a real positive thing for both the forest and the fish.”