# News this Week

Science  21 Aug 2015:
Vol. 349, Issue 6250, pp. 772
1. # This week's section

### How the gas giants got so big

To get their giant girths, Jupiter (shown) and Saturn had to form quickly from the sun's disk of gas and dust—within a few million years. Most theorists believe that they grew from rocky cores, about 10 times the mass of Earth, which then gravitationally sucked up the remaining gas in the disk. Models have successfully shown pebbles of dust clumping together into planetesimals that gobble up the rest of the pebbles and form rocky cores, but they also generated hundreds of cores, not just a few. Now, a new solar system model published on 19 August in Nature solves this problem, by slowing down the pebble-eating process so that all but a few cores are kicked out of the plane of the solar system. The model predicts the formation of one to four giant planets within just hundreds of thousands of years. As a byproduct, it also generates a few ice giants—Neptune and Uranus—and a Kuiper belt, the region of small icy worlds in which Pluto resides. http://scim.ag/gasgiants

### ‘Godzilla El Niño’ is coming

Conditions are pointing to a strong El Niño event for the winter of 2015 to 2016, and possibly one of the strongest on record, according to climatologists with the National Oceanic and Atmospheric Administration (NOAA). The forecasted event is strong enough to earn comparisons to both martial arts star Bruce Lee (at NOAA's ENSO blog) and Godzilla (by NOAA climatologist Bill Patzert, to CBS News). El Niño events occur every 2 to 7 years, when trade winds over the equatorial Pacific Ocean weaken and sea-surface temperatures increase, which can, in turn, dramatically alter weather patterns around the globe. Last week, NOAA scientists noted that sea-surface temperatures have been near or exceeding 2°C above normal since July. The most recent powerful El Niño event was in 1997 to 1998, which pushed global temperatures to new highs and triggered extreme events around the world, including widespread drought and flooding.

16%—Amount by which existing estimates overstate China's 2013 greenhouse emissions due to fossil fuel burning, according to a study in Nature.

## Around the world

### Columbus

Crowdsourcing animal research

The Beagle Freedom Project (BFP), an animal advocacy group, is crowdsourcing hundreds of public records requests to target cat and dog research across the United States. Supporters visit the group's “Identity Campaign” website, “adopt” a dog or cat at one of 17 public research universities, and learn how to file a Freedom of Information Act request with the school to get information on the animal. More than 1000 people have participated, resulting in about 100 public records, says Identity Campaign coordinator Jeremy Beckham. Some of the records, BFP claims, suggest that an Ohio State University lab violated National Institutes of Health rules concerning the use of dogs in biomedical research. The university denies those charges—and provided Science with evidence to the contrary. But BFP's effort could cause headaches for animal researchers, says Michael Halpern of the Union of Concerned Scientists in Washington, D.C. http://scim.ag/ICanirights

### London

Severe weather threatens crops

Weather-related crop disasters will become more likely with climate change, as drought, flooding, and heat waves strike fields more often, warns a report released 14 August by the Global Food Security (GFS) program, a network of public research funding agencies in the United Kingdom. Dozens of scientists, policy wonks, and industry experts examined the global food system and its vulnerabilities to severe weather. They used existing models of how crops respond to temperature, precipitation, and other factors to determine that by 2040, severe crop failures—previously estimated to occur once a century—are likely to happen every 3 decades. If a “plausible” worst-case scenario of drought simultaneously hitting four key staples—wheat, rice, corn, and soybeans—were to occur next year, it would likely triple the price of grain, the researchers suggest. http://scim.ag/foodweather

### Washington, D.C.

Push for chronic fatigue research

Patient advocates and scientists joined forces this week to boost research funding for the mysterious and debilitating disease chronic fatigue syndrome (CFS), also known as myalgic encephalomyelitis (ME). In an open letter to U.S. senators, the group, #MEAction, noted that ME/CFS affects an estimated 836,000 to 2.5 million people in the United States, and stated that research into treatments should be “proportional to and commensurate with other diseases with similar patient populations.” Through changes to an authorizing bill for the National Institutes of Health (NIH), the group aims to increase research funds available for ME/CFS from $5.4 million to levels similar to diseases that cause comparable disability, such as multiple sclerosis. It also wants to transfer responsibility for the disease from an isolated office within the U.S. Department of Health and Human Services to the National Institute for Neurological Disorders, a part of NIH. http://scim.ag/_MEAction ## Findings ### Support for ‘exercise hormone’ New evidence lends support to the existence of irisin, a hormone secreted from muscles and believed to mediate the effects of exercise on metabolism, in humans. First described in 2012 by a team led by researchers at Harvard Medical School, irisin appears to circulate in blood and prompt energy-storing white fat cells to behave like energy-burning brown fat cells. But other groups have suggested that humans produce only negligible quantities of the protein due to a genetic mutation, and that common tests for irisin are inaccurate (Science, 20 March, p. 1299). Last week in Cell Metabolism, the Harvard group used mass spectrometry to sort human blood protein fragments by mass. They found evidence of the hormone at levels similar to insulin and suggest the approach could validate other tests. http://scim.ag/irisin ### Gut bacteria linked to eye disease Some microbes that naturally dwell in our intestines might be bad for our eyes, triggering autoimmune uveitis, one of the leading causes of blindness. In this disease, T cells that target ocular proteins invade the eye and spur damaging inflammation. Although these cells occur even in the blood of healthy people, they won't enter the eye unless they first encounter their target proteins—so the mystery is how the invading T cells happened upon proteins that are still locked up in the eye. Now, a new mouse study in Immunity suggests an answer to the conundrum: Still-unidentified bacteria in the intestines produce proteins that closely resemble the eye proteins. Some T cells, after encountering the microbial proteins in the gut, may attack similar molecules elsewhere, such as the eyes. Identifying these molecular mimics of the eye proteins and the bacteria that spawn them could help researchers develop new treatments for autoimmune uveitis. http://scim.ag/eyeproteins 2. Forest Health # Battling a giant killer 1. Gabriel Popkin* The iconic eastern hemlock is under siege from a tiny invasive insect. On a frigid morning this past March, arborist Will Blozan snuck behind a small church here and headed down into a gorge thick with rhododendron. He crashed through the shrubs until he spotted the gorge's treasure: the world's largest known living eastern hemlock tree, known as the Cheoah. In 2006, Blozan had climbed the nearly 50-meter-tall giant and calculated that it contained 44.29 cubic meters of wood—then a record. Blozan would later discover two even larger hemlocks in the nearby Great Smoky Mountains National Park. Both of those champions, however, are now dead. So are millions of other hemlocks across eastern North America. They've been reduced to leafless gray skeletons by the hemlock woolly adelgid (Adelges tsugae), a tiny sap-sucking insect about the size of a pinhead. Originally from Japan, the adelgid has spread from Georgia to Maine in recent decades, entering new hemlock stands every year. Left unchecked, it kills nearly every tree it attacks. Paradoxically, large, seemingly vigorous trees like the Cheoah often go fastest. For years, forest managers have been in a fierce fight against the adelgid, and the battle has recently expanded to new fronts. The Cheoah and hundreds of thousands of other hemlocks are still alive because they have been treated with insecticides. But that's an expensive and labor-intensive tactic, so scientists are trying out more sustainable strategies. They're rearing and releasing predatory insects that eat the adelgid, and even looking for rare hemlock genes that might help them breed resistant trees. Eastern hemlocks, warns forester Jesse Webster of the Great Smoky Mountains park, “are in intensive care.” Like the family of a gravely ill patient, ecologists are also preparing for the possibility that these efforts will fail, and the eastern forest will lose one of its defining species. TSUGA CANADENSIS is one of eastern North America's largest native conifers. It has been called the “redwood of the east” and the “queen of the conifers.” A healthy tree resembles an evergreen waterfall; overlapping layers of short, downy needles cascade from the crown almost to the ground. Biologists believe the species diverged from its Asian cousins some 23 million to 40 million years ago. When Europeans arrived nearly 500 years ago, hemlocks dominated valleys in the 3000-kilometer Appalachian corridor and grew as far north as present-day Canada and as far west as Minnesota. Loggers once targeted the tree, but typically not for its wood, which is brittle. Instead, they prized its coffee-colored bark, which is rich in the tannins traditionally used to tan leather. Hemlocks nurture an ecosystem that has evolved nowhere else. Their defining characteristic is deep shade; unlike many pines, hemlocks keep needles on their lower branches, creating a thick canopy that blocks up to 99% of sunlight. Few plants grow in the gloom, but a hemlock seedling can bide its time for decades or more, waiting for a sunlit opening. Hundreds of species of insects, mites, and spiders appear to live primarily or exclusively in hemlock forests, and some aquatic invertebrates eat the hemlock needles that fall into mountain streams. Many migratory birds seek out the trees. The oldest known specimen was 555 years old when dendrochronologist Edward Cook measured it in 1991; just four other eastern tree species are known to live longer. “There's no tree, certainly in the east, that has anything like that kind of complete control” over its environment, says David Foster, an ecologist and director of the Harvard Forest in Petersham, Massachusetts. THE HEMLOCK'S MIGHTY GRIP is now being loosened by the diminutive A. tsugae. Hemlock woolly adelgids “are bizarre little things,” says entomologist Lynne Rieske-Kinney of the University of Kentucky in Lexington. An adult is about a millimeter long, with a threadlike proboscis that can be three times as long as its body. They can easily catch a ride to new trees on the wind or on birds and mammals. Once an adelgid pierces the base of a hemlock needle and starts sucking out starch, starving the tree, it stays put and envelops itself in a distinctive white fluff—hence the “woolly”—to protect itself and its eggs. The adelgid reproduces asexually in North America, rapidly spawning genetically identical offspring. All you need to get an outbreak, Rieske-Kinney explains, is “a susceptible host and one insect.” Nobody knows for sure how the Asian species got to North America, but all evidence points to ornamental hemlocks that were imported from Japan. The invader was first documented near Richmond in 1951. It went essentially unnoticed until 1986, when Mark McClure, then a scientist with the Connecticut Agricultural Experiment Station in New Haven, warned that it was killing trees in his state. Two years later, researchers found adelgids in Virginia's Shenandoah National Park, and by 1994 most of the park's hallmark hemlocks were dead or dying. “That was a big wake-up call,” says entomologist Rusty Rhea of the U.S. Forest Service's Southern Research Station in Asheville, North Carolina. Alarmed, federal biologists moved to stop the invasion. But there was one problem, recalls recently retired Forest Service entomologist Brad Onken: They knew nothing about the adelgid. “We had to really start from scratch.” AS THE TREES KEPT DYING, researchers scrambled to locate and inventory important stands. Blozan and his colleagues, for instance, used aerial photos and ground surveys to locate 75 hemlocks that were taller than the previously known record of 48.8 meters. All stood within a 100-kilometer-wide area that includes parts of North Carolina, Tennessee, Georgia, and South Carolina; most were within the Great Smoky Mountains park. Blozan and colleagues measured the trees and found that, contrary to theory, hemlocks reached their largest sizes at the southern edge of their range, not the geographical center. Unfortunately, the large southern trees also proved among the most vulnerable to the adelgid, because the region's relatively mild winters didn't keep the insect in check. To fight infestations, some researchers tried soaps and oil-based insecticides, with only modest success. One problem is that the adelgid attacks hemlock needles from below, where aerial spraying doesn't reach. In 2003, however, the chemical company Bayer gained the necessary approvals to sell neonicotinoid insecticides—primarily used in agriculture—for use in forests. Unlike soaps and oils, neonicotinoids are systemic; trees pull the chemicals into their tissues and can become toxic to insects for 5 years or more. The compounds gave forest managers a powerful new weapon, but they came too late for the 75 “superlative” hemlocks, Blozan says. The insects, aided by a major drought, weakened them beyond saving. Other trees have been luckier. On a rainy spring morning earlier this year, Great Smoky's Webster bounded up a streamside trail in the Low Gap conservation area near Cosby, Tennessee, to show off the results of an aggressive chemical campaign against the adelgid. Massive hemlocks towered overhead. Most had lower branches killed by the insects, but the upper branches were lush, thanks to workers who have injected insecticides into—or drenched the soil around—more than 200,000 trees since 2002. The effort has saved some of eastern North America's last remaining old-growth temperate rainforest, Webster says. But the chemicals have downsides. They can harm invertebrates, so their use is limited, especially near streams. (Neonicotinoids have been implicated in harming bee populations, but bees do not pollinate hemlocks and so are unlikely to be affected.) They also aren't cheap; costs are declining, but treating a single 30-centimeter-diameter tree requires$1.20 worth of chemicals plus labor, and projects often involve thousands of trees. And, in the end, insecticides do not permanently eliminate adelgids; they only reduce the numbers. “Chemical treatments are just a Band-Aid,” Rhea says.

THE LONG-TERM CURE, Webster believes, is down the road at a facility called an insectary. Here, personnel from the park, the Forest Service, and the University of Tennessee, Knoxville, are cultivating four species of adelgid-eating beetles—part of a larger effort to develop biocontrol methods.

Hammering some wet hemlock branches with a stick, Webster manages to collect one of the attack insects—a torpid 3-millimeter-long black Laricobius nigrinus beetle—on a white canvas “beat sheet.” Thousands of its cousins, whose ancestors lived in the Pacific Northwest, have been released at 120 sites in the park, and at hundreds of other places in the 20 states now infested with the adelgid. The goal is to establish permanent populations of beetles that do nothing but hunt adelgids.

It took decades to get this far. The adelgids exploded in eastern North America partly because few native insects eat them. So scientists looked abroad for predators, in the adelgid's home range. The first promising candidate was a tiny ladybird beetle that McClure and a Japanese scientist found in 1992, munching adelgids on a local hemlock species. After testing to ensure that the beetle wouldn't harm North American ecosystems, forest managers began releasing it by the millions in the United States. “There was a lot of optimism,” recalls Scott Salom, an entomologist at the Virginia Polytechnic Institute and State University in Blacksburg, who has led much of the biocontrol research. But hope faded as the introduced insects disappeared without making a dent in adelgid numbers.

Other candidates show greater promise. A few biocontrol insects, including the beetle Webster found at the insectary, have established stable, but small, wild populations. But rearing the insects has been challenging. Salom's technicians spend days trying to coax finicky beetles to reproduce in rooms cooled to 18°C or below. Recent cold winters have complicated matters by killing the adelgids that, ironically, researchers now need to feed their beetles. That's one reason scientists are exploring other biocontrol options, including a silver fly native to the Pacific Northwest and a fungus that may kill adelgids. A discouraging precedent tempers hopes. In the 1900s, scientists introduced more than 30 different insects to the eastern United States and Canada to try to control the balsam woolly adelgid (Adelges piceae), a related insect that devastates native fir trees. Each introduction failed, and the balsam adelgid is still a serious pest. “There are dozens of reasons why biocontrol programs don't work quite as planned,” Rieske-Kinney says. “Biological control is not for the faint of heart.”

A FEW SCIENTISTS think the hemlock's future rests not with finicky beetles, but with special genes. As early as the mid-2000s, scientists noticed that a few hemlocks appeared to resist adelgid infestations longer than their neighbors. Now, the hunt is on for genes that might be responsible. Ecologists Richard Casagrande and Evan Preisser of the University of Rhode Island, Kingston, have collected cuttings from what some call “putatively resistant” hemlocks found at isolated sites in New Jersey, Connecticut, and elsewhere, and they are now growing seedlings in Rhode Island and North Carolina. Some do seem to be less susceptible to the adelgid, Preisser says. That may be because the trees contain unusually high levels of terpenes, chemicals that can provide resistance to insects, says ecologist Joseph Elkinton of the University of Massachusetts, Amherst.

Others are skeptical about the effort. “In my estimation, there's not much hope that we're going to find genetic resistance to this insect,” the Forest Service's Rhea says. He and others note that the hemlock has relatively little genetic variation through its range, making resistant outliers unlikely.

The Forest Service is trying another approach: figuring out how to protect hemlocks from other stresses, so they are more likely to resist the beetle. Biologists from the service and North Carolina State University (NC State) in Raleigh are studying how well hemlocks grow under different conditions, including sunlit or shaded, and with and without deer fences, weeding, and fertilizers. They are also evaluating combinations of chemical and beetle treatments. As Bud Mayfield, a U.S. Forest Service entomologist at the Southern Research Station in Asheville, explains, “maybe one thing won't let you grow hemlocks, but four things will.”

The service is also preparing for the worst-case scenario: the extinction of eastern hemlock in the wild. It is funding Camcore, a tree breeding and conservation group housed at NC State, to collect seeds and genetic material from eastern hemlocks and the related, but rarer, Carolina hemlock. The group has planted trees in places likely to remain adelgid-free, including Chile and Brazil, in case they are needed to help seed reintroductions. “If everything else fails,” says NC State biologist Robert Jetton, “we're the hemlock insurance policy.”

ONE GROUP OF ECOLOGISTS reacted to the adelgid not by trying to save the hemlock, but by trying to learn from its death. In 1995, when ecologist David Orwig started his first research position at the Harvard Forest, the adelgid was working its way north toward the 1500-hectare research preserve, which holds extensive hemlock stands. Rather than invest in insecticides or beetles, Orwig and his colleagues decided to take the rare opportunity to study what happens when an ecologically important tree disappears. “At this site, I think you could learn more, unfortunately, by watching the demise of hemlocks than by trying to save them,” Orwig says.

The adelgid's arrival also represented an experimental ecologist's dream, says Foster, Orwig's colleague. It's “the gentlest way that you can alter the abundance of something,” he says. “You're not directly disturbing anything else; you're just selectively affecting the one species.”

Because Harvard Forest is part of the National Science Foundation's Long Term Ecological Research Network, the researchers were already collecting preinfestation baseline data. The floor of Prospect Hill, the forest's main hemlock study site, sprouts a garden of instrumentation, including moisture sensors and baskets for capturing leaf litter. Metal bands called dendrometers record the growth of almost a thousand trees. A 34-meter tower rises above the canopy to measure gases emanating from the forest. As a result, Prospect Hill now hosts one of the world's most comprehensive studies of forest death.

During a recent visit to the site, Orwig noted that some trees have survived longer than expected. But from the tower, he pointed out the yellowish needles that are the hallmark of a dying hemlock. “This should be a sea of bright green right now,” he lamented. “This is just kind of sick.”

The Harvard researchers haven't just watched their hemlocks decline in real time; they've also accelerated the process. In 2003, they girdled or felled some healthy trees, then watched what happened. As the trees decayed, wood and nutrients entered streams and the trunks began releasing carbon. In newly sunlit clearings, black birch, white pine, and other saplings shot up.

In a 2014 book that summarizes their findings, Orwig and his colleagues predict that dying hemlock forests will store less carbon for a decade or two, but that fast-growing young trees will eventually reverse the trend and soak up carbon even faster than the absent hemlocks. Stream flows might decline, because deciduous trees use more water. But local forest biodiversity is likely to increase, because deciduous forests typically support more species. At a broader scale, however, the researchers believe the New England landscape will become more homogeneous as hardwoods replace hemlocks. In short, they say, the forest will live on, changed but still functional.

THANKS TO REPEATED chemical treatments, the Cheoah hemlock is still standing. But Blozan says that in big trees like this one, with complicated branching structures, water and chemicals—including insecticides—can be slow to reach some branches. Indeed, one of the Cheoah's four forks has lost most of its needles. But on the whole, the tree looks healthy. “It's the last of its kind,” Blozan says wistfully.

He takes some photographs and records health data for the Eastern Native Tree Society, a group that he helped found. At worst, the documentation will help memorialize a remarkable life. But Blozan and others hope it will ultimately enable them to look back on the time when they began to turn the tide against the adelgid. “They'll come back,” he predicts. “It's just going to take a long time. Tree time.”

Correction (2 September 2015): Figure 4 has been updated.

• * in Highlands, North Carolina; photography by Katherine Taylor. Gabriel Popkin is a freelance writer in Mount Rainier, Maryland.

3. Forest Health

# The new North

1. Tim Appenzeller

Stoked by climate change, fire and insects are remaking the planet's vast boreal forest.

For 7 weeks last year, Yellowknife was besieged by smoke. In the vast evergreen forests encircling this small city in Canada's Northwest Territories, years of drought had set the scene for a historic fire year. Across the territories, 3.4 million hectares burned—an area equal to the state of Maryland, and seven times the annual average. The smoke darkened the sky, stung eyes, and filled Yellowknife residents with “a sense of panic,” says Frank Lepine, who manages wildfire response for the Northwest Territories government.

When the snow fell and the fires died, Lepine's army of firefighters—about 1000 strong at one point—could stand down. But for scientists the work is just beginning. The fires, they say, were an extreme example of the forces transforming the boreal forest, a stronghold of spruce, pine, and other conifers that rings the top of the planet, spanning northern Canada, Alaska, Russia, and Scandinavia. With its millions of square kilometers of pristine timber, thick carpet of moss and needles, and organic-rich frozen soil, the boreal forest stores more carbon than any other land ecosystem. And more than any other forest, it is bearing the brunt of climate change, warming roughly twice as fast as the rest of the planet. The effects on its cold-adapted trees are already evident. “We are on the cusp of a transformation,” says ecologist Michelle Mack of Northern Arizona University in Flagstaff.

The early signs can be seen from space, where orbiting sensors that monitor photosynthesis show that much of the boreal belt is “browning”: not literally turning brown but losing its vigor. They can be seen on the ground, in tree-ring studies that show trees are struggling to grow during the increasingly hot summers. They can be seen in insect outbreaks that are killing trees hundreds of kilometers farther north than they did 20 years ago. Most dramatically, the transformation can be seen in forest fires so fierce and voracious that they kick the forest into a new state, with a different mix of species and untold impacts on wildlife and climate.

“Fire is a catalyst for change,” says Mike Flannigan, a fire specialist at the University of Alberta in Edmonton, Canada. Across the boreal, fires are burning as never before, favored by heat and drought that dry out the forest floor and spawn thunderstorms that bring lightning but little rain. In Canada, the total average area burned each year has more than doubled since the 1970s, Flannigan says—and that's in spite of more effective firefighting. In Alaska and Siberia, too, fire is on the rise. But it is a change in the nature of the fires, as much as their extent, that is transforming the forest.

The trees of the boreal, after all, are used to fire. The dominant species in Alaska and much of Canada, black spruce, maintains an aerial storehouse of seeds, locked in cones that form a distinctive tuft at the treetop. When a fire singes the cones and melts their resin, they spring open, releasing years of seeds all at once—an adaptation known as fire-mediated serotiny. Normally, this seed rain ensures that black spruce comes back strong after a fire, outcompeting other species. But the most severe fires can break this stranglehold and open the way to a new kind of forest.

When ecologist Jill Johnstone was a graduate student at the University of Alaska, Fairbanks, she set out to study how the boreal forest regenerates after fires of differing severities. A convenient laboratory was at hand: burned areas near Fairbanks left by recent fires. The severity of the fires had been low; they had spared much of the organic layer of moss and duff that carpets the floor of black spruce stands. Johnstone decided to simulate a more severe fire by taking a propane torch and burning off the organic layer to various depths, in some cases all the way to the bare, silty soil beneath. Then she sowed the seeds of spruce, aspen, and other trees, mimicking what happens after a natural fire.

Where the organic layer remained, she found that black spruce held the advantage. The charred duff “is a really bad seedbed” for most tree species, she says. “It's black, it dries out, and it heats up in the sun to 40°C. To regenerate trees you need to have a lot of seeds”—exactly what black spruce provides after a fire. But the spruce's advantage disappeared in plots where she burned off the organic layer. On the exposed soil, cooler and moister than the duff, aspen germinated in greater numbers.

Boreal fires increasingly resemble Johnstone's propane torch, searing the forest floor to bare soil. These days, “people talk about megafires,” she says. “The fire weather is shifting.” Fires are also recurring more frequently, in some cases sweeping across areas that burned as little as a decade before, consuming any organic material left on the forest floor. Today, Johnstone says, late-summer fires “burn and burn until the snow falls. They'll burn pretty much everything in their path—they'll blow right through old fire scars.”

As her small-scale experiments suggested, black spruce is losing out. The clinching evidence came after Alaska experienced its biggest fire year in modern history in 2004. Some of the more than 700 fires just singed the forest floor, others blowtorched it away, giving Johnstone, Mack, and their colleagues a chance to compare how the forest recovered over large areas. Four years after the fires, they surveyed 90 burned sites and found that whereas those with an intact organic layer were thick with baby black spruce, sites where the fires had left bare soil were typically covered with seedlings of trembling aspen and paper birch. At those sites, Johnstone, Mack, and their colleagues wrote in 2010, the “legacy lock” of black spruce was broken.

To a satellite looking down on the forest, aspen and birch appear brighter than black spruce, says Scott Goetz, a remote-sensing expert at the Woods Hole Research Center in Falmouth, Massachusetts. By matching up a record of Alaska's most severe fires over the past 50 years with satellite measurements of forest brightness, he and colleagues including Mack confirmed the pattern seen on the ground: The most severe fires led to the most extensive regrowth of deciduous trees. Other researchers predict that deciduous stands, which accounted for less than half of interior Alaska's forests in 2001, will expand to cover two-thirds of the forested area by 2020.

Once established, researchers say, the broad-leaved trees are unlikely to be dislodged, as they grow faster and are less prone to burning than the conifers they replaced. For black spruce, the shift spells the end of a long reign, Johnstone says. “Black spruce have been pretty stable on the landscape for about 5000 years—we can tell from pollen records.”

Not every year brings a forest-transforming megafire. But warming temperatures are applying their own, steadier pressure to the forest. Again, the effects are apparent from space, in data from sensors that monitor specific wavelengths of light absorbed by chlorophyll. The resulting false-color maps show much of the boreal belt in green, indicating vigorous photosynthesis. But at its heart, especially in North America, are regions where photosynthesis—the vital function of a forest—has slowed over the past 30 years. They are depicted as patches of brown.

When Goetz and his colleagues reported the browning in 2005, he says, “it certainly got people's attention.” Modelers had expected the warming of the north, together with a fertilizing effect from rising carbon dioxide, to trigger a surge of forest growth. Instead, wide swaths—roughly a quarter of Alaska's forest, Goetz estimates—are languishing.

The browning isn't always obvious at ground level, but land-based measurements corroborate the remote sensing. Collaborators of Goetz's took core samples from black and white spruce, another common tree in the boreal, across much of Alaska, then analyzed tree rings to track the trees' growth history. In nearly every sample from the interior of the state, they reported in 2011, the rings had narrowed over the past 30 years, and the density of the annual increments of wood had risen. Those are signs that the increasingly hot summers are causing the trees to run short of water. To stem water loss, they are narrowing tiny pores, or stomata, on their needles—choking off their intake of carbon dioxide and slowing photosynthesis.

If the trend continues, the trees will begin to die, says Glenn Juday, a forestry expert at the University of Alaska, Fairbanks, who collaborated with Goetz on the 2011 study. In a new tree-ring study, Juday and his colleagues studied growth trends in white spruce at sites across Alaska. They found that trees in the interior of the state, where summers are hotter and drier than near the coast, are struggling to keep pace with climate warming. (Annual average temperatures at Fairbanks are up 1.5°C over the past 50 years.) Conditions are now nearing the trees' physiological limits, they fear. “Coming generations won't see the same large, old conifers,” Juday predicts.

Besides stressing trees directly, warming is favoring their enemies: the insect pests that are exploding across the boreal forest. Warming is enabling them to expand their ranges by accelerating their life cycles, helping them survive the winter, and weakening host trees. One of the most dramatic cases involves the mountain pine beetle. Until the late 1990s, the scourge was confined to lodgepole pine in British Columbia, west of the Canadian Rockies. Scientists had hoped that two barriers would restrain its spread: the Rockies and a vegetation boundary just to the east of the mountains, where the beetle's favored lodgepole pine gives way to a related species, jack pine.

Those hopes were dashed, says insect specialist Barry Cooke of the Canadian Forest Service in Edmonton. In the mid-2000s, the beetle staged “a mass exodus,” spilling eastward into Alberta and attacking jack pine. So far its range has extended 300 kilometers to the east and more than 1000 kilometers to the north, Cooke estimates. “The mountain pine beetle is poised to go all the way to Newfoundland” on Canada's eastern seaboard, he says.

The beetle joins a host of other boreal pests on the march: the spruce beetle, western spruce budworm, Douglas-fir tussock moth, hemlock looper, and willow leaf blotch miner, to name a few. And like severe fires, Cooke says the worst insect outbreaks may drive long-lasting ecological change, as the denuded forests regrow in a different form, better adapted to the new normal. “We will see irreversible changes. Whereas insects used to play a temporary role, they will become agents of permanent change.”

Researchers have varying visions of what the future boreal forest will look like. “I think there will be shrublands, even grass,” Flannigan says. “My gut instinct is that the forest may be gone in some places.”

“I think we're going to see some interesting changes regionally,” says Carissa Brown, a former student of Johnstone's who is now at Memorial University of Newfoundland in St. John's. But she's cautiously optimistic: “The boreal forest is a very resilient system.”

Juday suggests it may respond by migrating. Although his recent study showed that trees in the center of Alaska are suffering, it revealed a surge of growth near the Bering Sea coast, where summers are cooler and wetter and the forest gives way to tundra. The productive heart of the forest is shifting, Juday says. “An early stage of biome shift is underway.”

Whatever the new shape of the forest, the change will ripple through wildlife communities. Caribou, for instance, like evergreen black spruce stands, which are carpeted with lichen, an important forage. But in stands of deciduous aspen and birch, the lichen never gain a foothold among the fallen leaves. “That takes from the caribou their source of winter forage,” Brown says. Moose, on the other hand, are fond of aspen shoots and may thrive as the broadleaf forest expands.

The transformation will also affect climate, for better and worse. Fires, for example, turn wood and needles into climate-warming carbon dioxide—a positive feedback. But Mack, who has studied carbon flows in the changing forest, says that forest regrowth ultimately takes up the lost carbon—especially when aspen, which grows fast and quickly locks away carbon in wood, takes over. The spread of aspen could even restrain climate change through a pair of negative feedbacks: The aspen canopy reflects more sunlight than spruce, and broadleaf forests are less flammable.

Bad news, however, may be unfolding below the forest floor. Underlying much of the boreal forest is permafrost, deep-frozen soil filled with thousands of years of organic matter—a massive reserve of carbon. The duff layer beneath a spruce forest, up to a meter thick, insulates the permafrost from summer warmth. But when a fire burns off the duff, the permafrost can start to thaw and release carbon dioxide or methane, another potent warming gas. After a severe fire in the Yukon, Brown recalls, “we even had a hillside fall away” as the frozen soil softened. “You could actually smell all that carbon being released.”

Nature is providing plenty of opportunities for further study. The scars of last year's fires in the Northwest Territories are now a field site for Johnstone, Mack, and other researchers. And this year's fire season promises a generous new crop of natural experiments. In the Northwest Territories, the burning season it got an early start thanks to a phenomenon that used to be rare: a dozen “holdover fires” that started last year and then smoldered beneath snowbanks through the long winter. With the spring thaw, the flames rekindled, and the transformation of the boreal forest began anew.

Correction (24 August 2015): The graphs in this article have been corrected with the accurate numbers for the average area burned per year (in thousands of hectares) in Canada and Alaska.

4. Forest Health

# Second Act

1. Elizabeth Pennisi

Forest ecologist Robin Chazdon is helping show that regenerating tropical forests aren't wastelands.

When ecologist Robin Chazdon began studying tropical forests in the 1990s, she took the road less traveled. Whereas many researchers were scrambling to study undisturbed forests before they disappeared, she focused on what grew back once the trees were burned or logged. Rather than working in the forest's shaded understory, an ecosystem celebrated in Hollywood films, she labored in scraggly deforested plots in the broiling sun, covered head to toe to keep prickly bushes and biting chiggers at bay.

For decades, Chazdon worked in relative obscurity, barely scraping together funding for long-term studies of these so-called secondary forests. Her findings challenged some prevailing views: that tropical forests wouldn't regenerate, and that second growth was a biological wasteland. And over time, Chazdon and like-minded colleagues began building a case that, although protecting intact tropical forest was essential, second growth couldn't be ignored in efforts to protect the environment and human livelihoods. Secondary forests are “very dynamic places where nature is reasserting itself,” Chazdon says. “It's an elegant thing to behold.”

The rest of the world is beginning to see her point. Chazdon has “done a huge amount to elevate the visibility of secondary forests,” says tropical community ecologist Stephen Hubbell of Princeton University. Now that humanity has cleared or damaged at least three-quarters of the world's primary forests, governments and conservation organizations are increasingly turning their attention to the “junk” that regrows. Thanks in part to Chazdon's work, many now see secondary forests as key to restoring biodiversity and performing important ecosystems services, such as providing clean water and sequestering carbon. And last year, nations attending a United Nations climate conference set a goal of reforesting 350 million hectares of degraded land—an area larger than India—by 2030.

Reaching that goal, however, will require resolving some thorny issues. Some advocates insist “reforestation” should mean recreating, as closely as possible, the original forest. But others think planting rows of oil palms or timber trees should qualify. There are also disagreements over which areas should be targeted for reforestation, and whether people can and should accelerate the process with costly tree-planting programs. Some worry that efforts to promote second growth could undermine efforts to preserve intact forest.

For Chazdon, 58, the rising interest in secondary forests has prompted a second act of her own. After 27 years as an academic at the University of Connecticut (UConn), Storrs, she's taken a leave of absence, moved to Colorado, and shifted much of her attention from collecting and analyzing data to influencing policy—most notably in Brazil, which has made tropical reforestation a centerpiece of its efforts to combat climate change. “She has the drive, the personality to be a major player” in policy, says Peter Raven, president emeritus of the Missouri Botanical Garden in St. Louis.

But Chazdon is aware of the risks. She worries that policymakers might think she's been in the ivy tower too long, whereas scientists might look askance at her entanglement in policy. She'd like to make policy work her full-time job, but has no offers yet. “I'm facing a very uncertain next few years,” she admits. “And that's weighing heavily.”

STROLLING THROUGH A WOODED GLEN on the family ranch 2 hours southwest of Denver, Chazdon examines the new growth on a chest-high lodgepole pine. She impulsively gives the young tree a hug, telling it that despite the risk of drought and disease, it may one day be a giant. She identifies with a forest, she says: “When I just stand in there, I can feel the photosynthesis flowing.”

That affinity developed early. Despite growing up in Chicago in the late 1960s, family camping trips and summer camps turned her into an environmentalist. She fell in love with the tropics after a field course in Costa Rica in 1976, her sophomore year of college. “I felt like it was a second home,” she recalls. “That was very empowering.”

As a graduate student at Cornell University, she returned to Costa Rica and did her dissertation research at La Selva, a field station run by the Organization for Tropical Studies. Trying to understand how understory palm trees could grow in the deep shade of a mature rainforest canopy, she spent days using a homemade sensor to measure the light that filtered through the leaves. She discovered that flecks of sunlight were the secret to palm success: Eighty percent of the plants' productivity was fueled by these temporary patches of light.

Her studies of photosynthesis continued for years, but each time she returned to Costa Rica, more forest had disappeared, cleared by loggers, farmers, and developers. So once she moved to UConn in 1989, she decided to shift gears. Chazdon and Julie Denslow, an ecologist who studied forest dynamics and is now with the U.S. Forest Service's Pacific Southwest Research Station in Hilo, Hawaii, began to track how shrubs and trees were returning to plots on abandoned pastures purchased by La Selva and on nearby farms. The work would ultimately lead to a landmark 25-year project.

At the time, many ecologists doubted a tropical rainforest would ever grow back—they thought the soils were too fragile and would erode away before new roots could take hold, or too nutrient-poor to sustain regrowth. In La Selva, however, Chazdon and her colleagues found that tropical forests can make a comeback. They documented that, gradually, biodiversity returns, with a mix of plants reestablishing the understory and canopy layers that support key ecosystem services. Even species with commercial potential can regain a foothold.

A site's “ecological memory” helps shape what returns. Residual seeds that survive in the soil, waiting for a chance to sprout, are part of this biological memory bank, as are trees that remain standing nearby. Visits by bats, birds, and other seed dispersers also play a role in determining which plants re-emerge, as does the site's history of use.

At the time, few paid much mind to these findings: Tropical plant succession wasn't a sexy topic. So attracting funding was a challenge. “We don't do secondary forests,” one funder told Chazdon as she scrambled to find money after a grant from an early backer, the Andrew W. Mellon Foundation, expired. They were “too messy,” said another. “That was the low point,” Chazdon recalls. Still, she persevered, often with her husband, the recently retired ecologist Rob Colwell, and their two children in tow.

In 2007, just when she thought she had finally exhausted her funding opportunities, she enlisted Brazilian and Mexican researchers in a successful bid for a National Science Foundation (NSF) grant. In part, it aimed to use what had been learned over decades in La Selva to examine the validity of a common, less time-consuming approach to studying forest regrowth, known as “chronosequence” studies. Unlike long-term projects that track changes at a single site for decades, chronosequence studies—which have become a backbone of regeneration science—take a simultaneous snapshot of a set of plots in the same area, each at a different stage of regrowth. The goal is to get a quick read on how local forests might regenerate—without waiting years for the answer. “The assumption is that what's happening in the younger forest [plots] is what happened in the older forest [plots],” says forest ecologist Jess Zimmerman of the University of Puerto Rico, Río Piedras, another pioneer of studying tropical secondary forests. The long-term studies indicated that young forest plots did not necessarily reflect what older forest plots were like in their past. So researchers need to be careful about the conclusions they draw from chronosequence studies, the researchers concluded in June in a paper published online in the Proceedings of the National Academy of Sciences.

Chazdon says the work underscores perhaps the most important message to emerge from La Selva and related studies: that each regenerating site “tends to have its own path,” even when they share similar soils and climate. That's because chance plays a big role in what regenerates in the forests both short- and long-term. The research shows “you can reforest,” says Stefan Schnitzer, an ecologist at Marquette University in Milwaukee, Wisconsin, “but you still don't know what you will get.”

As policymakers come to grips with that ecological uncertainty, they are finding Chazdon's recent book, Second Growth: The Promise of Tropical Forest Regeneration in an Age of Deforestation (University of Chicago Press), all the more valuable. Five years in the writing and published last year, the tome is a kind of guide to restoration, synthesizing decades of research and explaining how tropical forests can come back on their own—and what to do if they don't. “It's an opus; it covers all you would want to know and could imagine you want to know about secondary forests,” says Thomas Rudel, a rural sociologist at Rutgers University, New Brunswick, in New Jersey. “There's nothing quite like [it].”

THE BOOK, SECOND GROWTH, ARRIVED at a timely moment, just as large-scale forest restoration was gaining momentum. In 2010, nations that had signed the United Nations' Convention on Biological Diversity set a goal of restoring 15% of the world's ecosystems by 2020. The following year, ministers from many countries issued the Bonn Challenge, which called for widespread reforestation. Then at last year's U.N. meeting, they upped the ante in a statement known as the New York Declaration on Forests, setting the 350-million-hectare goal. “I was thrilled—the international dialogue is not just about deforestation anymore,” Chazdon says. “It changes the vision.”

But the new vision is still blurry—and Chazdon thinks she can help achieve clarity. The Food and Agriculture Organization of the United Nations' official definition of reforestation is “very imprecise,” she notes. By its criteria, replacing a natural forest with plantations of introduced trees to make fuel and wood, or soak up carbon, could qualify as reforestation.

Not surprisingly, many conservationists oppose that idea, arguing that monocultures provide fewer of the ecological benefits of less homogenized forests. “Such projects have a long history of failure and they do nothing to restore ‘health’ to forests,” Hubbell argues. Some advocates say projects should count in official tallies only if they aim to restore a forest to some original state—typically a long and difficult task.

Chazdon isn't a big fan of monoculture plantations, but believes that to reforest “is not just to create a forest like before.” There are now too many people in need on the planet to allow for the return of unmanaged forests in very many places, she says. And she notes that the discovery of ancient pottery shards and earthworks in tropical forests once considered pristine shows that people have long played a role in shaping landscapes. “I used to be a little bit more idealistic,” she says. “But it's not realistic to have it all natural forest.”

Reforesters face another pressing question: to plant or not to plant. There's a long history of planting trees to speed the process along, Chazdon notes. But her work has shown that, if left alone, some forests come back on their own—with less effort and cost. And although active managers often replace mixed forests with a single species, or introduce exotic species, she notes that more passive strategies can restore something closer to the original species mix.

Chazdon concedes that natural regeneration can be a long process. It “isn't just a Band-Aid for a photo shoot after 2 years,” she says. And she notes that it may make sense in some places to plant some native species, particularly commercially valuable ones, to kick-start regeneration—in part to give local people a greater financial incentive to protect nascent forests. “We must meet the needs of the people or we are not going to be able to protect the landscape,” she argues.

Chazdon's approach has attracted particular attention in Brazil, where Environment Ministry officials have pledged to reforest some 12 million hectares of land as part of the nation's climate commitments. With official support, she'll be spending at least 3 months a year in Brazil, helping researchers and policymakers figure out how best to harness passive reforestation approaches. “Before we spend a lot of money on active restoration, let's first take advantage of the free help of nature,” says ecologist Pedro Brancalion of the University of São Paulo, who is involved in the effort.

DECIDING EXACTLY WHERE to grow new forests is another source of friction. Since 2009, the World Resources Institute (WRI), an influential think tank based in Washington, D.C., together with researchers at the International Union for Conservation of Nature and the University of Maryland, have been developing global and regional maps that highlight more than 2 billion hectares of land that could be reforested. When Chazdon borrowed the maps and displayed them at a workshop last year, however, some researchers were stunned—and infuriated. That's because the maps identify some endangered grassland ecosystems, including portions of Africa's savannas, as having “high reforestation potential.” (That's because they, too, have climates suitable for trees.)

The maps “provided a clear illustration of the fact that grassy biomes are undervalued and misunderstood as ‘degraded’ ecosystems,” says one critic of the effort, ecologist Joseph Veldman of Iowa State University in Ames. Earlier this year, he, University of Cape Town vegetation ecologist William Bond, and others criticized the map in a letter to Science (30 January, p. 484), and they have another critique in press at BioScience.

Chazdon would like to see the maps reworked and has been corresponding with both sides toward that goal. In the meantime, she's busy trying to catalyze discussion and consensus elsewhere. She and two colleagues have founded People and Reforestation in the Tropics, an NSF-funded networking effort aimed at getting policymakers, landowners, and scientists talking about what reforestation means, how to implement it on a large scale, and how to monitor the impact on local people.

It's a potentially contentious conversation, but colleagues believe she's got the people skills to hold her own—and produce results. “She has the rare ability to bring disparate communities together around a common cause,” says Toby Gardner of the Stockholm Environment Institute. “She's standing with two feet in science, but she communicates it in a way that policy people like me can use,” says Lars Laestadius, a senior associate at WRI.

For Chazdon, the dive into policy is a chance to put decades of science to practical use—and to try to make sure that reforestation is done right. And it is rooted in her belief that once-ignored secondary forests can play a phoenixlike role in restoring global forest health. “Things are dying and things are coming back to life at the same time,” she says. “It fills me with a lot of hope.”