Last updated: January 20, 2025
Article
One Way to See the Future of Alaska’s Unparalleled Forests: Look at Their Past
A surprising result in a natural climate change experiment inspires hope of a resilient future for Alaska’s old-growth, temperate rainforests.
By Ben Gaglioti
Image credit: NPS
If scientific predictions hold true, the climate of Olympic National Park in Washington State could shift 900 miles north in the next century.
Glacier Bay National Park in Alaska would seem as temperate as coastal Washington is now. Glaciers would retreat, fires may become common, and new wildlife would arrive. How would long-lived, stationary organisms like trees cope with these shifts? Scientists try to answer that question in a number of ways. Most of them have logistical drawbacks, like the high maintenance costs for lengthy experiments.
My colleagues and I took a new approach by studying forests growing next to glaciers. We did this because those forests have endured extreme temperature changes in the past and recorded them in their tree rings. As Markus Stoffel, a global change scientist at the University of Geneva in Switzerland, said, “These types of natural experiments provide insight into how forest ecosystems respond to the kind of rapid and radical temperature changes we can expect in the future.” Stoffel was not involved in our study, though he did publish a commentary on it.
Our results show that when faced with large temperature swings, the forests stayed unexpectedly stable. This suggests that vegetation replacement, forest dieback, or changes in tree composition are less likely to occur in response to radical climate change than most land managers might predict. Through ongoing research, we’re trying to understand the causes and limits of these forest-stabilizing effects.
The Future Is Hard to Predict
The most common approach to anticipating how ecosystems will respond to sudden climate shifts is to observe how a species performs under a range of natural conditions. For example, how does the growth and reproduction of a pine tree species vary in dry versus wet habitats, or relatively warm versus cool summers? This information can be combined with climate forecast maps to predict where the population of a species will live and thrive in the future. But this empirical method offers no way of knowing whether a species' climate sensitivity will remain the same as changes come faster and extend beyond historical extremes.
The limitations of these methods mean there are few real-world examples to use as ecological guides.
To get around this, some researchers have taken a more hands-on approach to looking at the ecological impacts of climate change. These climate manipulation experiments observe how plants respond to simulated conditions. Researchers expose the plants to increasingly warm or dry conditions by using open-topped greenhouses or rain diversion systems.
Image credit: Allison and Treseder, Warming and drying suppress microbial activity and carbon cycling in boreal forest soils, Global Change Biology, Volume 14, Issue 12, pages 2898-2909.
Climate manipulation studies have been key in understanding how grassland and tundra vegetation might respond to large and sudden changes in climate. But these studies often require lots of money and labor to maintain them long enough to see useful outcomes. Controlling the air temperatures in forest communities is particularly hard because of their high canopies. The limitations of these methods mean there are few real-world examples to use as ecological guides for anticipating forest responses to radical and rapid temperature change.
A Cool Idea Emerges
About 27 percent of Glacier Bay National Park is covered by more than 1,000 glaciers. Many of these are alongside old-growth, temperate rainforests. This type of rainforest also clings to the damp, coastal mountains of Canada, Chile, and New Zealand. It’s considered critical for global diversity and carbon storage.
In glacier-rich regions like Glacier Bay, cold, air-conditioned winds cool the forested areas bordering the large masses of ice. In June 2018, my colleagues and I learned about those cooling effects firsthand on a two-week field trip. We were studying the ice-age history of the remote, Outer Coast region of Glacier Bay. On a backpacking portion of the trip, we couldn’t help but notice an incessant, cool breeze. We felt it while camping on the strip of gravel that separated the La Perouse Glacier from the nearby old-growth forest.
Image credit: courtesy of Ben Gaglioti
As we searched for fossil plants in the ice-age sediments nearby, the massive glacial “AC unit” forced us to bundle up in nearly all the clothes we had. Carrying everything on our backs required that we pack light, so they weren’t many. The first night was clear. Without the blanketing effects of the usual cloud cover in Southeast Alaska, it grew cold. I cursed my choice to pack only a summer sleeping bag. I had figured our trip was in June after all.
In addition to affecting everything in their vicinity (including us), glaciers can also be sensitive responders to climate change. Essentially, they’re giant conveyors of ice. High-elevation snowfall that persists for years eventually gets compacted into a blob-like mass of ice. This mass slowly flows downhill into warmer climates, where it melts and “calves”—splits off in chunks—into the sea. The glacier’s ice budget is determined by the amount of ice entering it from snowfall versus the amount of ice exiting through meltwater and calving.
Changes in climate influence the bottom line of a glacier's ice budget, altering the size of the glacier’s footprint. Colder conditions caused most glaciers on Earth to become larger during the Little Ice Age. That was the cold period between the years 1300 and 1900 that caused crop failures and social unrest in Europe. It was marked by slower growth rates in ancient trees around the world. Since about 1900, an anthropogenic warming trend has reversed the effects of the Little Ice Age and accelerated glacial retreat. Today, nearly all glaciers on Earth are shadows of their former selves. Their ice margins have retreated well back from the visible footprints they occupied 10, 50, or 100 years ago.
Image credit: NPS
Thinking about the Little Ice Age while we camped at the La Perouse Glacier gave me an idea. After a long night of restless, pondering sleep—common in the midnight sun backcountry—I posed the idea to my colleague Dan Mann, a geologist at the University of Alaska. I suggested that maybe the advance and retreat of the La Perouse Glacier had altered the air temperature in the nearby forest. This would have caused the trees to record outsized climate trends in their tree rings. As the sun caught a rufous hummingbird, we discussed the idea over a breakfast of oatmeal and coffee.
Evolving Questions—and a Baby Porcupine
As we waited for the breakfast water to boil, the sun lit up the textured green crown of the nearby forest. We talked about how the trees we were staring at—a hundred miles from the nearest town—could be useful to society’s current climate quandary. The La Perouse environment offered a ready-made natural climate change experiment. One that wouldn’t require decades of careful monitoring or maintaining greenhouses. As I lugged my heavy pack through wet brush later that day, it did cross my mind that those experiments are usually a stone’s throw from relatively lavish research stations that offer catered meals and saunas.
The next several days followed with the normal tempo of ups and downs that can typically be experienced in the backcountry on the Outer Coast of Glacier Bay. One night, a powerful windstorm arrived at 3 a.m. to shred one of our tents. We had to shoo an emaciated black bear with dental problems away from camp.
Would the La Perouse trees have felt the microclimate effects that accompanied the glacier’s fluctuations since the Little Ice Age?
On a foggy trek through thick, old-growth forest, we realized we had walked in a mile-long circle. This was after recognizing a distinct moose-chewed log that we had already passed (“better turn the GPS back on”). But our trip had many high points as well. There were terrific swimming holes with stunning backdrops, and close encounters with fascinating creatures. We met a baby porcupine the size of a hamburger that ferociously popped up the three spines on its back as we passed by.
We also found vivid traces of what the Outer Coast may have been like for the first Americans as their population expanded from Siberia into the New World. Based on our fossil finds, there were blueberries, clams, and maybe even a few trees to greet them. A constant reel of scenery, weather, and mysteries were intermixed with our discussions about the La Perouse forest.
The next week, as the trip wound down, Mann and I arrived at a series of questions we wanted to pursue. Would the La Perouse trees have felt the microclimate effects that accompanied the glacier’s fluctuations since the Little Ice Age? If so, when exactly did these temperature changes occur and in what season? In what ways would we measure how the nearby forest responded to these amplified temperature trends?
Image credit: courtesy of Ben Gaglioti
Explorers and Ghosts
Like bushwhacking around the Outer Coast, putting scientific ideas into practice can be slow. To answer the questions that had emerged on the 2018 backpacking trip, my colleagues and I spent the next six years collecting environmental data around La Perouse Glacier. First, we needed to map the extent of its cold microclimate. So, a month after our June trip, Phillip Wilson—a college student with a penchant for the Alaskan wilderness—and I returned to La Perouse on Independence Day weekend. We installed a network of sensors that recorded air temperature every hour along a transect one kilometer long (a little less than two-thirds of a mile), perpendicular to the ice margin.
Image credit: NPS
Second, we needed to reconstruct the glacier's position over the past several centuries. That way, we could estimate how and when the glacier‘s movements affected the adjacent forest. Fortunately, the La Perouse Glacier was a good place to do that. This was partly because, despite its remote location, several 19th and 20th-century explorers had recorded the glacier’s position. One such explorer was John Muir, who championed the idea of national parks in their infancy. Muir capsized his dory near La Perouse Glacier while on the Harriman Expedition in summer 1899. He took the opportunity to hike around the ice margin and describe the La Perouse forest.
Another useful source of the glacier’s history came from a “ghost forest.” That’s one that the advancing ice of the La Perouse Glacier had run over during the Little Ice Age. We found several patches of the ghost forest on the Outer Coast. Through cross-dating, the ghost forest provided a precise timeline of the most recent glacial advance during the Little Ice Age. Cross-dating matches the ring-width patterns of living trees, whose ages are known, with the ring-width patterns of dead trees of unknown age. We used the tree rings of nearby living trees to date when gravel pushed by advancing ice smothered and killed ghost-forest trees. The gravel preserved the trees as the ice covered the forest. Then the gravel eroded away as the glacier retreated, giving us access to the ghost trees.
Image credit: courtesy of Ben Gaglioti
Ancient Temperature Extremes
Over the six years of our study, the temperature sensor data showed that the glacier had a cooling effect within about 600 yards of the ice edge. This effect was most pronounced in summer. The area within 100 yards from the glacier’s margin experienced summers that were around 3-4°C (7-9°F) colder than sites 1,000 yards away.
By combining the ages of the ghost forest trees with eyewitness accounts and historical maps, we reconstructed a timeline of glacial advance and retreat since 1830. The narrative went like this: After being in a retracted position well away from the living forest before 1830, the glacier began advancing. The ice margin then came to within 600 meters of the living forest around 1850. It continued advancing to within about 50 meters of the forest a few years before John Muir saw it in 1899. After a series of minor retreats and advances, the ice margin began retreating in earnest around 1950. Like glaciers around the globe, the retreat of La Perouse ice has accelerated in the last 20 years and continues today.
Over the last 50 years of glacial retreat and anthropogenic warming, the La Perouse forest probably experienced one of the fastest rates of warming on the planet.
To determine the history of La Perouse forest temperatures since 1830, we combined the information from the temperature sensors with the glacial timeline reconstruction. This enabled us to estimate how the air temperature in the forest had changed as the nearby glacier and its cold microclimate approached the forest and then retreated away from it. The resulting data described the cooling effects that Mann and I felt while backpacking in 2018. The results were impressive.
Our work showed that the rates and magnitudes of summer temperature changes in the nearby forest were three to four times greater than those experienced by other forests in Alaska. Over the last 50 years of glacial retreat and anthropogenic warming, the La Perouse forest probably experienced one of the fastest rates of warming on the planet. The rate of temperature change in the La Perouse forest was similar to what scientists predict will occur in other forests during the coming century. This convinced us that studying the La Perouse forest could show how such forests might respond to abrupt climate change in the future.
Image credit: courtesy of Ben Gaglioti.
The Portfolio Effect
After we published the results of this initial work, I presented our research at a science conference in San Francisco. My friend and colleague Sean Brennan was also there, and we chatted about science over pizza and beer. We talked a lot about the “portfolio effect” in ecosystems. This is the idea that communities of organisms can respond to environmental change in diverse ways across a landscape. And these different responses can cancel each other out to instill stability in the broader ecosystem. It’s similar to the idea of diversifying a financial portfolio to minimize the risk of loss. To sustainably manage natural resources, scientists and land managers have worked to preserve those components of a system that optimize the portfolio effect.
Localized bust years were compensated by boom times in the next stream over, so the larger salmon population of the whole watershed didn’t change much from year to year.
Through research on salmon in Southwest Alaska, Brennan and his colleagues had recently shown that fish populations from some small creeks had boom times. But, for whatever reason, other populations in adjacent creeks went bust in the same year. Localized bust years were compensated by boom times in the next stream over, so the larger salmon population of the whole watershed didn’t change much from year to year. It was a classic example of the portfolio effect. Brennan’s work showed the importance of keeping every neighboring salmon population intact to allow for the portfolio effect and preserve the region’s famous salmon fishery. If just one creek was polluted, that effect could be lost, and the region's fishing economy could be compromised.
Image credit: U.S. Fish and Wildlife Service
Brennan’s perspective was inspiring. Could trees and forests be like fish and rivers? Could the overall growth of a forest as a whole remain stable over time because the growth rates of trees belonging to different species diversified the forest’s growth rate “portfolio”? Testing this in the La Perouse forest could add a new dimension to the growing body of research on the portfolio effect. And would that portfolio effect be strengthened or weakened as climate change accelerates?
When the glacier was close enough to accentuate the temperature trends in the forest, the growth rates of different tree species diverged.
My colleagues and I attempted to answer these questions by measuring tree rings in the five conifer species that grow in the La Perouse forest. We studied those trees that had experienced large glacier-affected temperature swings. What we found was surprising and fascinating. When the forest experienced relatively small climate changes, before the glacier was close enough to amplify them, there wasn’t much of a portfolio effect in the La Perouse forest. The trees grew in relative synchrony year after year. The overall growth of the forest varied widely depending on the region’s air temperature each year.
But when the glacier was close enough to accentuate the temperature trends in the forest, the growth rates of different tree species diverged. Their boom-and-bust responses canceled each other out, which stabilized forest-wide growth rates. This was a surprising result because most ecologists, including myself, probably would have predicted that larger swings in temperature would trigger larger swings in forest-wide growth rates.
Image credit: U.S. Forest Service / Marion Glaser
More Resilient Than We May Think?
These results are somewhat encouraging for a society that may soon be overwhelmed by the compounding impacts of climate change upon ecosystems. The forest's portfolio effect instilled stability, but only when climate changes were large and fast. It would be as if your stock portfolio automatically diversified and protected itself from loss only when big swings in the markets occurred. Could this portfolio effect help stabilize forests during the accelerated episodes of climate change predicted over the coming decade? Does the portfolio effect work for biological communities other than forests?
“Understanding ecosystem responses to major environmental stressors is acutely relevant today,” said Martin Hutten, a terrestrial ecologist at Glacier Bay National Park. “This research observed resilience to rapid, long-term temperature change in a pristine forest. Natural experiments such as these offer useful insights in how forests may respond to rapid future change.” Hutten also said that studying these climate gradients “can help us predict how diversity and ecosystem services in parks may respond to climate change in the future.”
We can test whether a forest with fewer tree species responds more sensitively to accelerated climate change because it lacks the diversity of growth responses that lead to the portfolio effect.
My colleagues and I hope to explore these ideas in similar, natural, climate change experiments in other glacier-adjacent forests. Like those in Glacier Bay, Wrangell St. Elias, and Olympic National Parks, as well as in protected areas in Switzerland and New Zealand. In the La Perouse forest, tree species diversity was a key component of the portfolio effect. By replicating this study in other national parks, we can test whether a forest with fewer tree species responds more sensitively to accelerated climate change because it lacks the diversity of growth responses that lead to the portfolio effect.
We could also test whether monoculture forests composed of individual trees having diverse growth responses also experience the portfolio effect. Our ultimate goal is to understand which national park forests may be resilient and adaptable despite the oncoming roller coaster of climate extremes.
Ben Gaglioti is a paleoecologist and professor at the University of Alaska Fairbanks. He studies how Alaska’s ecosystems responded to past climate change. Image courtesy of Ben Gaglioti.