�պ�����

It’s eight degrees Fahrenheit. Off trail, at 2,042 feet of elevation, on the side of Camel's Hump, Professor Steve Keller takes off his gloves, pulls a Dell tablet out of his backpack, unrolls a wire, and plugs it into a scrawny maple tree. Well, actually, into a tiny sensor hanging under a white plastic funnel hanging off the side of the maple. The winter sunshine feels beautiful and the sky glows with a preternatural blue. The trees stand still, a mix of beech and sugar maple, plus a few yellow birches, all silent, their elbows clothed in new snow.

Keller has come here to take the temperature of the forest—and to show me the red spruce trees that he and generations of �պ����� scientists have been studying on this mountain since 1964. At this elevation there aren’t many spruce. “But there’s one—there,” Keller says, pointing to a brave and solitary evergreen tucked into the understory.

“One of the interesting trends that we're seeing is a bit of a rebound in red spruce at these lower elevations,” Keller says. As this forest has recovered from the ravages of acid rain and a long history of land clearing, “we're seeing some spruce come back in—back down the slope—to where they were missing before.”

Keller waits as the data from the sensor downloads onto his tablet—months of temperature and relative humidity readings. He presses the screen and points to a spiking line running across it. “See, it got down to just below zero on Christmas Eve,” he says, tracing his finger down a steep drop in the graph.

Mostly though, it’s up—for both temperatures and trees. Over recent decades, the average temperatures here are rising. These warmer conditions are, generally, pushing trees upslope and northward. On Camel's Hump, the ecotone—the complex boundary between the mid-slope hardwood forest and the high-elevation spruce/fir forest—seems to be caught in a tug-of-war between the recovery of spruce from decades of damage and the rising heat that pushes trees to migrate toward the summit.

Trees do migrate. If you stood at the summit of Camel's Hump 13,000 years ago, you would have witnessed a bulldozed landscape of rubble and bedrock, left behind by retreating glaciers. Slowly, a treeless tundra grew. Then, at the same time the first Paleoindian hunters were arriving in the Champlain Valley—12,000 years ago—so were trees, wind-blown pioneer species like paper birch and black spruce. Over centuries, a forest formed, dominated by spruce. White pine and hemlock started to show up about 10,000 years ago. Around 8,000 years ago, beech, chestnut, and Vermont’s beloved sugar maples moved in and began to get a foothold. In a drying period 4,000 years ago, the conditions were favorable for oak, which expanded its range. For millennia, trees have marched up river valleys and climbed mountains—their seeds dropped on the ground, carried by rodents, washed by streams, tossed by storms—generation upon generation, chasing a suitable climate. 

Now many tree species are losing the race. Historical research estimates that trees in New England, on average, can disperse about one-tenth of a mile per year. If they’re booking it, maybe as fast as three-tenths of a mile. But today the climate is warming much faster than that—shifting at four to six miles per year. Under a business-as-usual scenario of greenhouse gas emissions, the climate of Vermont is projected to warm by five, six, seven or more degrees Fahrenheit by the end of the century—becoming like that of, perhaps, West Virginia.

“The trees can’t keep pace,” says Keller, a professor in �պ�����’s Plant Biology department and an expert on tree genetics. “Climate change is already causing stress. I'm talking about reduced growth, reduced carbon sequestration, more susceptibility to extreme events like drought and heat waves, less ability to fend off pests,” he says. “Our local forests will become more and more maladapted if we don’t do anything. So how do we help?”

He and other �պ����� researchers are at the vanguard of a growing number of ecologists, foresters, and land managers who think part of the answer may be “forest assisted migration”—moving the seeds or seedlings of trees from where they live now to where they might have a better shot at thriving in a warmer future. Helping the trees to walk.

Keller tucks the tablet back into his pack and starts his own walk higher up the mountain, kicking up clouds of snow, to look for some more red spruce. The native range for this species stretches from North Carolina to the coast of Newfoundland. Regional modeling suggests that red spruce is especially vulnerable to decline driven by climate change. Isolated “sky islands” at the summit of peaks in the South are at risk of blinking out entirely, and spruce will face stiff competition from hardwood trees in a warmer, wetter Vermont. Plus, they’re especially slow to migrate since they can live 300 years or more.

Keller’s years of research here on Camel’s Hump and all along the eastern United States aims to improve the odds for red spruce. “If you look at red spruce and you assume that all members of the species are alike, you’re missing a lot. The trees differ across their range, across populations,” he says. “There may be unique genetic diversity within the range, pre-adapted to future climates, to future environments. So it becomes a matching problem: if we want to have a healthy spruce forest in New England, where would we look to find the genetic adaptations that will be well matched to the climate of New England in 2100?”

To find out, Keller and his colleagues and students grew more than 5,000 spruce seedlings with funding from the National Science Foundation. First, they collected seed and genetic information from mother trees at 65 locations throughout the spruce’s range—up the spine of the Appalachians, across New England, and into Canada. Then they germinated the seeds and grew seedlings in a �պ����� greenhouse for a year. In the spring of 2019, they transplanted the trees to three locations with different climates: Asheville, N.C., Frostburg, Md., and Burlington, V.T.—1,700 seedlings at each place in raised bed gardens.

Then they watched, while the trees went through two growing seasons and one winter, collecting data on survival, growth, bud break and bud set, height, nutrient concentration in their needles, and other measures of whether the seedlings were thriving. They wanted to know how the seedlings were affected by the difference between the climate of their garden site compared to the climate of their mother tree. They discovered that juvenile spruce like to grow in the same climate as their mothers, and their growth trailed off as they got into different climates. 

But, of course, the current climate of the mother tree will—very soon—not be its future climate. Using this insight—combined with U.S. Forest Service tree inventories, DNA sequencing, and climate modeling—Keller and his collaborators have developed tools to estimate how far—and from what source—red spruce seed should be moved to give the next generation its best chance of doing well in a warmer future. On a new app that Keller and colleagues created, users can select a period for which they are planting, say the years 2071-2101, dial in moderate or severe greenhouse gas emissions—and then enter a location, say Mount Katahdin in north-central Maine. The app churns away and suggests that red spruce seed with good adaptations for planting might be found on the eastern slopes of Mount Mansfield in Vermont. For Camel's Hump, in 2100, it suggests a spruce forest near the Finger Lakes in New York.

As temperatures rise and Vermont experiences more rain, intense storms, and severe droughts—the conditions will improve for some trees and be worse for others. Forest ecologists expect more southern-adapted trees, like shagbark hickory, black cherry, and red oak, will increasingly find conditions in the state that suit their needs. Other species, including balsam fir, yellow birch, black ash, and sugar maple, “will be negatively impacted,” the 2021 Vermont Climate Assessment reports. But even if the conditions are cozy for a particular species, it must be there to benefit. “The climate might be perfectly suitable for red oak in a given area,” says Professor Tony D’Amato, director of the Forestry program in the Rubenstein School of Environment and Natural Resources, “but it's just not able to get there quick enough to capitalize on that new environment.”

Unlike California or other spots in the West, forests in the eastern United States have the delightful quality of just growing back after they are harvested, or burn, or get knocked down in a hurricane. There is not an extensive tradition of planting trees in the North Woods. Mostly you get whatever trees volunteer to grow—and for areas that are being replanted the adage has been “local is best,” meaning sourcing seed only from nearby. It may become useful, even necessary, for landowners, timber companies, and conservation land trusts to start thinking about how they will introduce genes, trees, and even whole suites of species from farther afield that can keep forests healthy or even forests at all. In some spots, like the invasive-choked, deer-chomped, pest-threatened woods of Chittenden County, there is reason to be concerned that an “alternative stable state”—in the anodyne jargon of scientists—will emerge in the coming decades: the weeds will win and there will be few trees. 

“From my perspective as a scientist, the ‘local is best’ paradigm is—if it isn't already antiquated for a particular species—it will be within our lifetimes. And certainly within the lifetimes of the forests we're talking about,” Keller says. “As the climate continues to change, this approach will become less and less viable.”

Peter Clark leads the way with a machete, down a slippery hill, and into a very young, quarter-acre patch of woods at one of �պ�����’s research forests in Jericho, Vt. He’s already hacked a path through the raspberry underbrush and over a muddy ditch. “Haven’t been down here in a while and thought I’d lay out the red carpet,” he says. Soon, Clark—a �պ����� post doctoral scientist—and Tony D’Amato are pointing out unusual trees that they planted here as part of an experiment five years ago while Clark was earning his Ph.D. in D’Amato’s lab at the Rubenstein School. 

“Here’s an American chestnut,” Clark says, pointing to a sapling thinner than my arm. “It’s a disease-resistant one from the American Chestnut Foundation.” Chestnuts were once a keystone species, with billions of them growing, 10 feet wide, from Maine to Mississippi—until they were almost entirely wiped out by a blight in the first half of the 20th century. Last year, Clark and his colleagues published the results of a study, in the journal Forest Ecology and Management, that showed promise for these hybrid chestnuts to be used in restoring northeastern forests—and for assisted migration outside its historic range, which reached its northern limits in the Champlain and Connecticut River valleys of Vermont.   

“This is red oak,” Clark says. “We’re on the northern range limit of red oak here. You see it around here, but not much.” Then they both crane their heads back to look up at several spindly trees that tower over the others. “These big screamers right here are black birch. This is a true assisted migration species. It's not found here,” Clark says. Black birch is common in southern New England and makes it into southern Vermont, but not this far north at this elevation. Then Clark and D’Amato turn to look at a much smaller, bedraggled-looking sapling. “And here’s bitternut hickory—another assisted migration species,” he says. “Model projections say that it’ll fare pretty well, but we've had a lot of trouble getting it to succeed in our experiments.”

Forest assisted migration, in some form, has been happening for thousands of years. There is evidence showing that Native Americans, ancestors of today’s Abenaki people, moved and cultivated oaks, chestnuts, butternuts, and other large-seeded trees, valuable for food, moving them north and elsewhere long before Europeans arrived. 

Much more recently, in the early 2000s, a group of conservation activists began transporting seeds and seedlings of a critically endangered conifer tree species—Florida Torreya—from its only remaining locations on a river in Florida to land in North Carolina and farther north, even attempting to grow them in New Hampshire. This is a form of assisted migration sometimes called “species rescue” that these conservationists see as a pressing ethical obligation to prevent extinction. In contrast, some ecologists see this as a fool’s errand, taking a species far outside its current range, where it may have little chance of long-term survival or bring havoc to local ecosystems. Monterey pine is endangered in its native habitats in California—but after being introduced and widely cultivated in Australia, it’s become invasive, damaging wildlife habitats.

Assisted migration (sometimes called assisted colonization) has sparked strong controversy and some outlandish ideas. Moving polar bears from the melting Arctic to Antarctica only seems like a good idea if you’re a polar bear, not so much if you’re a penguin. And, conversely, “moving tree species outside of their native range may be a lot harder than people might expect,” says Professor Jane Molofsky, Steve Keller’s colleague in the Plant Biology department and an expert on the evolution of invasiveness. “Trees have a range for a reason. And it often has to do with the soil. There's a lot of interconnections between species, especially in the mycorrhizal associations that occur under the plants. Unless you're migrating with your fungi, your mutualists, it may not be successful. As you go from a deciduous forest in Vermont up to a Canadian forest, which has a layer of needles, soil conditions may be very different.”

The devil is in the forested details. On one end of a spectrum is what scientists call “assisted population migration”—moving trees, usually from warmer locations to cooler ones, within their existing range. Like transporting red oak from southern Vermont, where it’s plentiful, to northern Vermont, where it’s rare. “That’s low risk,” Carrie Pike, a tree geneticist with the U.S. Forest Service, told me. 

Somewhat more controversial: moving trees outside their current range—but just a little bit. That’s called “assisted range expansion,” where the goal is to plant trees that could have migrated on their own—in less than the time of an ice age—but may not arrive as soon as they’re needed or that might be blocked from dispersing by, say, farmland or cities. “It’s doing what occurred historically over hundreds of years of change,” says D’Amato, “which, unfortunately, is now happening over decades—so the trees just can't track it.” Clark and D’Amato have been testing this approach here at the Jericho Forest and, in a much bigger experiment, at three locations in northern New England. “That’s medium risk,” Pike said. And most controversial is “species assisted migration.” Picking up red oak seedlings, say, and plunking them down hundreds of miles north in central Quebec where they would be strangers in a strange boreal land. “High risk,” said Pike.

But what, really, is risky? “Doing nothing is risky,” says D’Amato. Unlike efforts to rescue a beleaguered single species—by moving it to a new home—forest assisted migration aims to keep alive the hyper-complex ecosystems that millions of species call home—roaming moose and flying squirrels, rare mosses and unknown insects. “Trees are great,” says D’Amato, “but get enough trees together to form a forest and, wow, something amazing happens, something spiritual.”

Globally, many forests are facing catastrophic disturbances and declines. And in the North Woods, “the risk profile may have fundamentally changed too,” Clark says. “In the absence of exploring some radical ideas to adapt our forests we may be putting them at greater risk. Assisted migration is focused on keeping forests as functioning healthy forests.”

A crucial question, then, is: can the functions of a forest be maintained if the trees are replaced by new species? For example, Vermont forests dominated by balsam fir are highly vulnerable to climate change. “If we’re going to lose balsam fir in the Nulhegan Basin in the Northeast Kingdom—it’s a deep-crowned conifer—can we plant species there that would fill a similar ecological niche?” wonders D'Amato. He and colleagues are working on an experiment there, after decades of industrial management, to plant white pine, red spruce, eastern hemlock, and northern white cedar, all of which are found in that landscape but were mostly eliminated by intensive logging. They’re including both local seeds and more southern genotypes—including six sources of spruce from West Virginia.

Models suggest that the Northeast could gain 10 to 20 species of climate-adapted trees—if we just waited, say, 300 years for them to wander in. Vermont forests will likely become more biologically productive for the next 50 or 100 years—even as individual species struggle—before higher summer temperatures, drought, and loss of soil nutrients bring dramatic declines in health. “The goal is to try to get ahead of that,” D’Amato says, “so that when those declines happen, pockets of climate-adapted species are on the landscape—perhaps as refugia, as seed sources, from which a new forest can grow.” 

In a thousand years, the North Woods will likely recover its footing, find a new assemblage of trees, build novel natural communities, and get on just fine. But we can’t wait that long. Forests are natural infrastructure that humanity depends on. “We want—and need—mature trees and all the benefits they provide,” D’Amato says, “carbon storage, clean water, wildlife habitat, cool shade, wood products, recreation—I could make a long list.”  Letting a rare tree species disappear into the night of extinction may be a tragic failure; letting forests decline is an existential threat. 

D'Amato’s overall goal is to develop tools to help forests and forest landowners thrive, or at least survive, in the face of climate change. In some places, that might mean improving a woodland’s resistance to change, like fighting to keep sugar maple growing well on a Vermont hillside as long as possible, thinning, selecting, planting, and weeding on behalf of the glory (and market) that is maple syrup. Great while it lasts; tough sledding when maple can no longer hack the heat.

In other forests, it might mean building resilience. “Vermont's forests are pretty simplified,” D’Amato says, so increasing the diversity of conservation land by planting and protecting populations of rare native trees—or increasing the structural complexity of a woodlot by letting old trees remain—can give forests pathways to adapt to rapid change. 

And—as part of a project called Adaptive Silviculture for Climate Change—D’Amato, Clark, and a large team of colleagues from �պ�����, Dartmouth College, and the U.S. Forest Service Northern Research Station have been exploring the promise and peril of helping forests transition into new arrangements of species, adapted for a warmer future. Assisted migration has been a central tool in these experiments. 

“In just seven years, the mindset of a lot of scientists and foresters has changed,” D’Amato says. “Foresters have been moving trees around for a long time, white pine, oaks, other species. But a lot of that work has been in aid of commercial forestry. There hasn’t been much research on moving trees in aid of ecosystem function.” 

Within the 27,000-acre Dartmouth Second College Grant in northern New Hampshire—a managed forest dominated by sugar maple, yellow birch, and beech—the team cleared a group of one-acre and quarter-acre plots, harvesting most of the trees to mimic the natural disturbances you might find in these woods: ice storms, wind throws, small fires. At two �պ����� research forests in Wolcott and Washington, Vt., additional quarter-acre plots were prepared. Then, in the spring of 2018, scientists and land managers planted 4,675 seedlings at the New Hampshire site and more than 400 at each of the �պ����� forests. The team chose nine species for testing. Six were locally present, but minor, species—red oak, white pine, hemlock, black cherry, bigtooth aspen, and red spruce—that the team thought could do well in a warmer future and could replace some of the ecological roles of dominant trees as they get slammed by climate change. Three were the species Clark and D’Amato showed me at the Jericho Forest: chestnut, bitternut hickory, and black birch—more southerly trees not found on these sites. 

They tracked the survival and growth of all the seedlings. After three years, just over half of them were still alive. The trees from outside their range didn’t do as well as the local trees. A severe drought, extreme winter cold, and deer and moose browse put stress on all the young trees. And it highlighted a curious reality researchers call “ecological memory,” where the past state of a forest resists efforts to introduce new species, giving locally adapted trees a competitive advantage, in the short run, over the transplants. That’s why D’Amato, Clark, and the team are doing this research: to go beyond computer modeling and greenhouse experiments to understand real-world outcomes in forests at the scale of what commercial landowners encounter. “There may be extra work needed to give assisted migration species a leg up in the early years,” says Clark. 

And what else is needed? “Humility,” says D’Amato. “There is so much we don’t know. We have to accept that we can’t build a new forest, that we can’t manage our way out of many of our troubles. Mostly, we just have to let the forest change. And we need a diversity of forestry approaches to see if they help it to adapt.”

This is what Bill Keeton calls “risk spreading.” “The fundamental problem is the difficulty in predicting the future,” says Keeton, a professor of forestry and forest ecology in the Rubenstein School and a fellow in �պ�����’s Gund Institute for Environment, “so we should not put all our eggs in one basket.” Like D’Amato, he sees both benefits and risks in assisted migration, but he sees limitations to its wide-spread use. “Novel species assemblages are likely when the climate changes dramatically. We might have completely different natural communities in the future. So where should tree species be moved to be ready for the future? That question makes assisted migration really tricky.” That’s why Keeton recommends pairing it with other approaches, including an expanded network of protected lands that “encompass more geophysical diversity, like topography and soil—and are better connected,” he says. “This will give species room to move on their own and sort themselves—and will be more adaptive to change.”  

On a warm and rainy afternoon, Clark and Miriam Wolpert ’20, a technician in D’Amato’s lab, are sorting through 6,000 red oak acorns at the �պ����� Aiken Forestry Sciences Lab on Spear Street. Wolpert holds up one Ziploc bag after another while Clark describes where these seeds came from: Newport, Vt., the Massachusetts border, Ohio, Pennsylvania, Delaware, Virginia. Wolpert picks one acorn out of a bag and places it in a tube of soil. Then another and another. These seeds, from every cold hardiness zone down to North Carolina, will be sprouted and planted out at several �պ����� research sites in Vermont—looking for oaks that might be well-matched to a warmer future. As Clark and Wolpert carefully cover each acorn with dirt, it seems so hopeful to imagine these seeds, from perhaps Tennessee, growing into sturdy red oaks that bring shade and new life to a corner of a Vermont forest. That will be beautiful—but only if we move fast enough to save ourselves from the heat first.