Читать книгу Early Warming - Nancy Lord - Страница 7

PART ONE

MY SALMON HOME: KENAI PENINSULA

On a mid-May Friday, the Ninilchik River on Alaska’s Kenai Peninsula was running high, fast, and dark. Sue Mauger, stream ecologist for the nonprofit Cook Inletkeeper, and I stood on the muddy bank, among piles of moose turds, willow bushes close-cropped by those same moose and just beginning to tint into green, and a crushed pop can. A pair of harlequin ducks beat past us, low over the water and heading upriver, the male in its colorful clown plumage, the duller female sporting white cheek patches like silver dollars.

Somewhere in that muddy water whipping past us, the first king salmon of the year were likely forcing their way upstream. In another week, Memorial Day weekend would launch the sport fishery on the Ninilchik and neighboring rivers, and barbecue grills throughout the region would be put back to work. The economy of the Kenai Peninsula, in fact, largely runs on salmon—not just the sportfishing that occurs along the rivers and in the inlet, but also commercial fishing by seiners, drifters, and setnetters and the subsistence and personal use fishing that nets Alaskans food for their freezers. The early king salmon would be followed by the big push of red salmon in the rivers that lead to lakes, then late-run kings, pinks, and silver salmon later in the summer.

We had driven north from Homer for forty miles in an effort to collect a “TidbiT,” a temperature data logger, left in the Ninilchik River since the previous October, and replace it with a new one to begin to record the spring and summer water temperatures. Mauger pointed toward the middle of the river, where the logger, an item not much larger than a quarter, was housed in a piece of PVC pipe and anchored to the river bottom with rebar. The whole apparatus was well out of sight below the surface, in a low spot behind a rock, where fishermen should not have snagged it and, Mauger hoped, where ice and logjams should not have scraped against it. This was her first visit to the river since fall, and she was eager to get the logger back to the Homer lab to download its winter’s worth of temperature data.

In chest waders and long plastic gloves, Mauger stepped off the bank and braced herself against the current. A slight and athletic woman who has run marathons and had just been telling me about a triathlon she was entering in Anchorage the next day, Mauger worked her way over the mud-and-gravel bottom, careful not to lose her footing in the swift current. The tannin-brown water reached her knees, then her waist, and she was nowhere near the logger’s location. She turned and shuffled back to the bank. “This is what they call ‘bankful,’” she said. Indeed, the river filled its banks from one side to the other, about forty-five feet away. On both sides the willows were hung with grasses and other debris, deposits from even higher water levels a few weeks earlier.

Spring that year—2008—had been late and cool, and a surprise snowstorm in April had left snowpack still melting in the hills. There were even late snow patches in the shadowy places along this lower stretch of the river, just a mile from where it emptied into Cook Inlet. The morning was overcast and cool—thirty-eight degrees Fahrenheit when we’d left Homer—and rain during the night had swelled the river.

The late spring and cooler temperatures had local people saying things like, “There goes the global warming theory.” I was also used to seeing a popular bumper sticker that read ALASKANS FOR GLOBAL WARMING and knew there were plenty of Alaskans who were suspicious of the underlying climate change science but also thought that warmer temperatures could be a good thing—more shorts-and-sandals weather, a longer gardening season, lower winter fuel bills. Certainly the difference between weather and climate had more than a few people confused, and erratic weather events—even sometimes involving colder temperatures and more snow—were hard for many to connect to a warming planet.

Besides the climate change effects of increased greenhouse gases in the atmosphere, we needed to contend with natural cycles, including what scientists call the “El Niño Southern Oscillation.” The shifting in the equatorial Pacific from warmer masses of water to cooler ones, and back again, affects the wind patterns that carry warmer or cooler air to Alaska. We were just coming off of a strong La Niña, which, according to the models, should be followed by an El Niño and warmer than average winters.

If air and water temperatures were not alarming all Alaskans, there was one thing almost everyone cared about passionately—our wild salmon. Salmon evolved as cold-water animals, and Alaska today supports tremendous runs of five species. In some recent years the commercial fisheries have caught upward of two hundred million salmon valued at hundreds of millions of dollars at the dock. The sport fishing industry, which claims twelve thousand jobs, boasts hundreds of millions of dollars of additional value to the state’s economy.

And salmon are anadromous, meaning they travel up rivers and streams, and sometimes into lakes, to spawn, and are thus sensitive to both marine and freshwater conditions.

Cook Inletkeeper was all about protecting the Cook Inlet watershed and the life forms that depend on it and since 1998 had been monitoring twelve sites on four Kenai Peninsula salmon streams for pollutants, along with water temperature, pH, dissolved oxygen, turbidity, and nutrients.

The biggest surprise was finding that in recent summers stream temperatures on the lower Kenai often exceeded state water-quality standards set to protect salmon spawning and the survival of eggs and fry. And water temperatures were trending upward, tracking air temperature increases.

Here at the Ninilchik River, the number of summer days that water temperatures had exceeded state standards for the upper limit of egg and fry incubation (55.5 degrees Fahrenheit) increased from fifty-six days in 2002 to seventy days in 2005. The number of days that the standards for the upper limit for fish migration and spawning (59 degrees) were also high—to more than fifty days. Other Kenai Peninsula streams showed similar temperatures and trends. In 2005 the Anchor River close to Homer topped the 55.5-degree limit on eighty-eight days and the “do not exceed” temperature of 68 degrees on six days.

Mauger is well aware that a few years of data mean, by themselves, very little. (In fact, in the couple of years after 2005, the stream temperatures were a little less alarming, before shooting back up in 2009.) She also knew that increased stream temperatures could be due to a number of factors. Like a detective, she’d set out to discover associations and meaning.

Stream temperatures can increase as a result of land development—the cutting of shade trees, paved surfaces where water warms before running off, water withdrawals. It was true that much of the spruce forest on the Kenai Peninsula had died from a beetle infestation, but an analysis of the subsequent logging that occurred on different watersheds, with changes in stream shade, could not account for the warming. Neither could the very small amount of development that dotted woods and fields with cabins and connected them by roads and trails.

What did correlate with the warming streams was warming air. At the Homer Airport, air temperature records go back to 1932—a long time for Alaska, where most baseline data is sorely lacking. Those records show that most of the warming has occurred since 1977, following worldwide trends, with summer air temperatures warmer by two degrees Fahrenheit and December and January temperatures by four degrees. Beginning in 2005, Cook Inletkeeper started hanging temperature data collectors in trees near the stream data loggers and found that the water temperatures generally tracked the air temperatures. That is, when we have a warm summer, the water temperatures are warmer, and when we have a cooler summer, the water temperatures are cooler. (The relationship also changes with water volume, which relates to the amount of rain and snowmelt.) With that relationship established, Mauger and others can “backcast” to estimate earlier stream temperatures as well as begin to predict future stream temperatures if air temperatures continue to rise.

We were not, Mauger emphasized to me, at temperatures where we were seeing any visible changes in survivorship. That is, fish were not dropping dead in the streams. The effects could be more subtle—stress that might affect growth or susceptibility to disease. It was time, she said, before we had catastrophic results, to “get people comfortable with using the data. The trajectory that we’re on gets us there. What we don’t know is the time scale.”

Mauger had, in the last couple of years, begun making public presentations about her work to various audiences.¹ She’d presented at science conferences and to conservation groups, to chambers of commerce and fishing organizations. In 2007 she’d been an invited guest to speak to the Alaska Climate Impact Assessment Commission appointed by the state legislature. The commission, a “balanced” group of government employees, industry representatives, and citizens, included members who did not believe that human activities contributed to global warming, and Mauger described the group as “a little bit of a hostile audience.” (One person in particular wanted to argue the science with her, and others accused her of having “no proof” that salmon were being affected.) As we turned back from the river, Mauger half grimaced, half chuckled. “I didn’t feel like I had a very aggressive message, so I was kind of surprised by the feedback I got.”

Her recommendations to the commission, about what should be done in Alaska, had been straightforward:

1. Collect stream temperature and flow data in key watersheds across the state.

2. Incorporate temperature data (both stream and marine) and climate information into salmon management models and plans.

3. Encourage actions to increase watershed resiliency to climate warming.

One of the points she tried to make in all her presentations was, she acknowledged, the need to control greenhouse gas emissions. But even if we stopped burning fossil fuels immediately, past actions had already committed us to a certain amount of environmental change, and she wanted people to think about what that meant. “We need to understand what that change is going to look like on small, regional scales,” she explained. By starting to collect stream temperature data, “we’re helping communities understand what role fishing will play in their future. This project is about getting people the information they need to make good economic decisions.” It was up to those people, then, to decide what they wanted to do, both to ameliorate the negative effects of climate change and learn to live with them.

Mauger and I returned to Inletkeeper’s Ford hybrid and drove the short road down into the village of Ninilchik. There, a huddle of homes crowded the lowest bend of the river and a small boat harbor accessible only at the highest tides.

Ninilchik has a unique history among Alaskan towns, having been settled as a retirement colony for Russians and their “creole” families (the results of Russian and Alaska Native unions) who chose to stay in the country when control passed from Russia to the United States in 1867. Many of the residents today are direct descendants of those settlers, and the onion-domed Russian Orthodox church on the hill is surrounded by a graveyard filled with more of those Russian names. From where we stopped by the river to check a stream gauge, I spotted the notched-log house once owned by a friend, who, when he stripped the inside for a bit of modernizing, found old Russian newspapers used as insulation. The Russian colonists, the Dena’ina Athabascan people who preceded them, and all comers since have relied on the river and its salmon for their lives and livelihoods. It was unthinkable that the people of Ninilchik should have to turn away from that history and dependence.

What does it mean when stream temperatures exceed state standards for salmon spawning and rearing? Nobody quite knows. Alaska’s temperature standards are actually a modification of research-based recommendations from the Pacific Northwest, not based on any Alaska-specific data—and not adjusted regionally to match Alaska’s huge geographical spread. A former governor, Frank Murkowski, had eliminated the Department of Fish and Game’s Habitat Division, so there was no one home to figure out what effects, if any, warmer temperatures might actually be having on Alaska’s wild salmon.

We do know, from testing and experience elsewhere, that higher temperatures can have profound effects on salmon and other cold-water fish. Higher temperatures can reduce growth rates (when fish have to put more energy into respiration and metabolism), reduce the survival of eggs and fry, affect the timing of out-migration (reducing marine survival), increase disease, and make fish more vulnerable to pollution (since some chemicals and metals increase in toxicity with higher temperatures). Warmer waters can also influence food supplies, vulnerability to predators, and competition with other species including exotics that move into waters made more hospitable to them. And we do know that when fish (and living organisms generally) are stressed by any one process, they’re less able to deal with other stressors.

Along the West Coast of the United States, salmon runs have been in serious decline, with the collapse of California’s Sacramento River Chinook runs resulting in a complete closure of sport and commercial salmon fishing off the coast of California and most of Oregon. The reason for the collapse was not clear. Some biologists pointed to unusual weather patterns that disrupted the upwelling of nutrient-rich water in coastal waters that normally supported the marine food web. Other people pointed to water quality issues related to damaged habitat, agricultural pollution, and altered flows and temperatures related to development and water diversions. Greenhouse gas-induced climate change might be just one factor among a series of environmental insults.

In the Sacramento River, engineering efforts had already been made to try to keep temperatures from becoming overly warm; in 1996 a temperature control system was added to the Shasta Dam to release deep, cold water from the lake bottom. As early as 1976 and 1977, thousands of Sacramento River salmon died when water temperatures rose to sixty-two degrees Fahrenheit. (Recall that Alaska’s “do not exceed” temperature is sixty-eight degrees, and that in 2005 our Anchor River exceeded that on six days.)

Michael Healey, a professor emeritus at the University of British Columbia, is a nationally recognized expert in both the ecology of Pacific salmon and the design of resource management systems. He has noted that the effects of climate change on salmon are of major concern to resource managers and that the degree of warming expected in both freshwater and marine habitats over the next century “will have uncertain but potentially devastating effects on salmon and their ecosystems.”² By focusing on the sockeye salmon of British Columbia’s Fraser River (the most valuable—commercially and ecologically—salmon river in Canada), he has developed a model for examining the cumulative effects of climate change on the many stages in the salmon life cycle and across generations. What he lays out is not pretty.

First, he notes that in recent years large numbers of adult salmon that enter the Fraser River have failed to make it to the spawning grounds and many that did died without spawning, and that a late run has been entering the river weeks earlier than in the past. Extremely poor survival in the Fraser in 2004 was linked to exceptionally high temperatures. Sockeye returning to the Fraser have been both smaller than in the past and with lower energy reserves, suggesting that “energetic exhaustion” may be one cause of the observed mortality, perhaps along with temperature stress and disease.

For his analysis, Healey recognized eight stages in the life of the sockeye and detailed the effects of increasing temperature (drawn from previously published scientific studies) on each stage, then looked at how the effects on each stage affected performance at later stages. He has shown that global warming has negative effects on productivity at every stage and that the effects at one stage carry through to the next and then generationally. He further considered how the effects of high temperatures at each stage might be mitigated or adapted to—biologically and by management and policy decisions. For example, management options might include releasing cool water from reservoirs upstream of spawners, fertilizing lakes to make up for nutrient deficits related to the mismatch of plankton bloom times, and preventing overfishing. Policies might include reserving adequate stream flows in salmon rivers, managing predators, and establishing “salmon first” for the use of estuarine habitats.

Healey wrote to me, “I had been stuck a bit trying to find anything positive to say, but I think I have a found a thread (melting Arctic is opening new habitat for salmon).” Given time, he says, salmon will naturally colonize the Arctic, but he fears that their rate of moving into new habitats might be too slow to keep ahead of global warming. It was his opinion that managers and policy makers should be thinking now about assisting salmon to colonize the Arctic (by transplanting them or developing freshwater nurseries), not only to keep salmon in the world but also to retain the option of eventually restoring populations to the south, when the climate there is favorable again.

Alaska has its own example of temperature-stressed salmon not making it back to their spawning beds. A five-year study of Yukon River salmon infested with Ichthyophonus linked the microscopic parasite (commonly called “ick”) to warmer stream temperatures.³ Before the mid-1980s, Ichthyophonus, which causes “white spot disease” in fish, had never been reported from the Yukon River. Today, it infects more than 40 percent of the river’s adult Chinook salmon.

This disease, which can be fatal to fish, doesn’t harm people but makes the salmon meat mealy, with an odd smell and unpleasant texture, fit only for feeding to dogs. Richard Kocan, a fish disease expert from the University of Washington, has linked the emergence of the disease to increased river temperatures. Average Yukon River temperatures have been rising for three decades. Since 1975 June water temperatures at the village of Emmonak, on the river’s delta, have increased from less than fifty-two degrees to fifty-nine degrees Fahrenheit with July temperatures even higher.

Kocan found that the number of infected fish, and severity of the disease, was highest in the study years with the highest temperatures and highest during the times of summer when the water was warmest. Laboratory studies have also shown that Ichthyophonus thrives as a host’s temperature increases (as would be the case of cold-blooded salmon in warming water). Kocan’s peer-reviewed study concluded, in the conservative manner of science, that water temperatures above fifty-nine degrees appeared to correlate with increases in Ichthyophonus infection, and that rising average water temperatures in the Yukon River in the last three decades may be an important cause of increased disease and mortality among Chinook salmon. He believed—but did not have good data to support—that as much as 20 percent of the Yukon’s Chinook were dying from the disease en route to their spawning grounds. Ichthyophonus has also been detected, in limited surveys, in salmon in other rivers, and in other fish species. Kocan considers it to be a classic emerging disease—defined as a disease that has either newly appeared in a population or has been known for some time but is rapidly increasing in incidence or geographic range. In this case, the triggering factor for its emergence appears to be warming waters.

To put it simply, Ichthyophonus may have always been with us, but a warming climate may be redistributing it, allowing it to flourish in parts of the north where cold temperatures once acted as a barrier to its spread and where species, themselves stressed by warmer temperatures, lack individual or evolutionary resistance.

And more of the same may be on the way.

In a spitting rain, Mauger and I headed back south on the highway, past an eagle’s nest in a bare cottonwood right next to the road, to our last stop, at Stariski Creek. Stariski is a smaller watercourse than Ninilchik but still an important salmon river, with spawning runs of king, silver, and pink salmon, along with steelhead and Dolly Varden trout. At Stariski, culverts running under the highway had been recently replaced with a bridge, and Inletkeeper was hired by the Alaska Department of Transportation to check turbidity around the project. The project included a fancy new boardwalk that led from the road to the river, so that even someone in a wheelchair could easily roll to streamside for fishing. The whole bridge project, including riverbank stabilization with rock and willow plantings, was necessitated by not one but two “hundred-year (that is, expected to occur only once in a century) floods” in 2002. Those major floods had dramatically reshaped channels, scoured beds, and undercut banks.

The landscape remodeling brought home to me those other likely effects of global warming—not just an increase in air and water temperatures, with their direct implications, but lower stream flows in summer (reducing habitat areas and increasing stream temperatures even more), stream-scouring floods in the fall (wiping out eggs and egg-laying habitat), changes in the timing of freeze and thaw cycles, and sedimentation.

Sedimentation could come from sources previously unthought of. Alarms had sounded recently about Skilak Lake, a critical rearing habitat for red salmon that ascend the nearby Kenai River.⁴ There, the glacier that feeds the lake is melting more rapidly, depositing more ground-up rock “flour” into the water. The resulting turbidity means sunlight can’t penetrate the lake water as far, which means less photosynthesis, less plankton production, less food for the young salmon.

Downstream of the Stariski bridge, Mauger took a water sample to check turbidity and filled out a data sheet to describe the current river conditions, high and fast. I watched a pair of mergansers, looking like passive wooden decoys, take a wild ride downstream. Then we crossed the highway and walked upstream, past riverbanks restored with fabric and willow plantings and a section where more recent erosion had undercut a bank. We walked through dead, flattened grasses and twisted alders and into the shade of cottonwoods, where snow patches still lingered. Mauger pointed away from the river. “It’s interesting,” she said. “This area’s never been logged, but it’s all open.” Indeed, the “forest” was more grass than trees; most of the spruce trees were broken off, leaving splintery stumps at various heights, and the deadfall of their tops lay under and over the twining grasses.

This was a story that Mauger and I knew all too well. In the warmer temperatures of the last couple of decades, spruce bark beetles in the region had flourished. They not only survived the winters that had previously kept them in check with cold temperatures, but were able to complete their life cycle in a single year instead of two. They loved the hot summers that enticed them from their galleries beneath the bark and propelled them to new trees. The infestation, which eventually killed thirty million trees (decimating four million acres of spruce forest—that is, a land area larger than all of Connecticut)—was the largest insect infestation ever documented in North America. (It was recently overtaken by an even larger attack of pine beetles in British Columbia, also linked to climate warming.)

Spruce bark beetles (Dendroctonus rufipennis), like the Ichthyophonus salmon parasite, are a natural part of the ecosystem in our region—thought perhaps to be the instrument of forest succession, rather than fire—but their success in attacking and killing nearly every adult spruce tree across an entire landscape is unprecedented in either historic or prehistoric (judged by tree-ring evidence) times.⁵ I well remember the summer “flights” in the 1990s, when a series of overly warm days would release swarms of beetles, like a biblical plague, that would, literally, drive people to the shelters of their closed homes. And I remember the march of death across the landscape—the forest turning red as needles died, then gray, then splintered as the dried-out trunks shattered in winter winds.

Parts of the peninsula were logged, the dead trees turned to chips and shipped to Asia, while others were left “natural” as habitat for insect-eating birds, for building new soil, for eventual regrowth. In both cases, our woods were being replaced with grasslands; it’s thought that the tall native grasses may, for a long time, keep any trees from gaining a foothold.

The changed landscape can, of course, have profound effects on water resources and salmon. Trees provide shade, which can help cool rivers. Trees also provide woody debris, beneficial to salmon streams for breaking up the flow and providing resting and hiding spots for fish.

Mauger waded into the raging stream to take another water sample. She pointed out, against the far bank, a pole and instrument she uses, when she can reach it, to measure stream height. A dead tree had fallen against it, and next visit, when she hoped to be able to wade across, she’d bring a chainsaw to clear the area.

All around us, grasses and other debris were hung in the willows, showing how high the water had been earlier. Mauger surveyed the banks, alert to another climate change threat: invasive species. Two that had devastated habitats elsewhere on the peninsula were northern pike—a toothy, predatory fish that can take down ducks and muskrats while also gobbling up young salmon—and exotic grasses. Neither had moved in naturally, as some species did, mile by mile, as the climate changed. They had been introduced—pike by people who valued them as a sport fish, and exotic grasses by being mixed with grass seed, seed packets, or imported hay. Still, in a hospitable climate, they can flourish. Such “weedy” species can outcompete native ones.

Reed canary grass (Phalaris arundinacea)—“a huge deal,” Mauger said—grows into mats that turn flowing streams into marshes. It’s been used for revegetation of roadsides precisely because it spreads rapidly and builds sod that helps with erosion. All along the West Coast, from California to British Columbia, it’s destroyed wetlands and salmon habitat. Plant specialists once thought that in Alaska’s cold the grass wouldn’t produce viable seed, but recent surveys had found infestations in at least 259 locations on the Kenai Peninsula, including along salmon streams.

“It’s all additive,” Mauger said, capping her water sample. Warmer temperatures, more flooding, greater drying, less shade and debris, human alterations to the river corridor and uplands, invasive species—“the system is getting hammered.”

We made our way back along the river, past the fallen forest and the eroded muddy banks, to the highway. We drove back toward Homer through more dreary rain, and I asked Mauger whether, with all that she knew about consequences, she considered herself an optimist or pessimist about the future.

She hesitated a few seconds before drawing a breath. “I do have optimism that some of the repercussions of climate change can be minimized. We’re going to do that with having better information. That’s what this project is about—getting the information that we need so that we don’t have a collapse in fisheries with no warning.”

She paused again and then added that she was also hopeful that, after so much delay and denial from our political leaders, our national politics would change enough, soon enough. “To make the transitions that we need to make, I think it will be a painful ten years. But we have to start now.”

That summer, the stream temperature work Mauger had pioneered on the four Kenai Peninsula rivers was being extended, with state funding, to the rest of the Cook Inlet watershed—forty-eight sites in all. Mauger had worked all winter on the protocols so that data would be collected in a consistent, reliable manner. “We want to try to identify what types of streams are likely to warm fastest, and what types of systems are likely to remain coolest, so we can make some decisions about where to study more about the habitats and the fish. This is really a first cut at looking at different stream sizes and types, the role of wetlands, the role of lakes.” The plan was, after that, for the same monitoring system to be employed by partners and volunteers in the major salmon streams of Bristol Bay (the location of the most valuable salmon runs) and elsewhere in the state.

In this way climate data would be broken out of the broad models and brought down to a local, real-time, and real-place level, to empower communities with the tools and data they need to protect salmon habitat and watershed health. Biologists and land use planners, it’s hoped, will use the data to identify streams most vulnerable to change, then apply it to decisions about further research, habitat protection, water use, and restoration activities. In that ideal world, fishery managers will incorporate temperature information into their modeling of run strengths and escapements (the numbers of fish allowed upstream to spawn). They will also use everything they know about stock structures and life histories to maintain genetic diversity within Alaska’s salmon, knowing that such diversity is critical to the ability of salmon populations to respond to climate change; elsewhere in the world, where individual stocks have disappeared due to overfishing or habitat loss, there’s been little left for filling voids. (This need for genetic diversity, of course, applies to all species.)

We drove along and then across the swollen Anchor River, and Mauger offered up an example of practical application, one way that stream temperature data had already been used in decision making. She’d worked with a university student to identify and map “temperature refugia” along the Anchor. Areas of cool water—shaded by banks with overhanging vegetation—would be most essential to maintaining the river for salmon, and the local land trust added that information to their conservation priorities for working with willing land owners on maintaining vegetative cover in those key areas.

I was listening and making notes, but I was thinking about that eagle in the nest we’d passed, in the cottonwood tree right beside the highway. We used to think that eagles were shy birds that needed plenty of undisturbed space around them; now we know they don’t—or that some have adapted to our busy presence.

Eagles—like bears, belugas, other large and small birds, trout, microbes, even the trees enriched by the spawned-out salmon carcasses that are carried into the forest by the birds and the bears—depend on the salmon. So, of course, do we. I can hardly imagine my home place without them. How would we live?

Salmon are adaptive; we know this. The five Alaska species have managed to survive in this part of the world for six million years, through periods of warmth and cold. Over the course of time, individual stocks have been challenged by change—whether it came from glaciers and ice sheets that overran streams or tied up water in ice, from volcanoes that buried streams in ash or mud, from rock slides, floods, earthquakes that raised or lowered the land and its streambeds. Some stocks perished, while others survived, adapting to conditions and colonizing new habitat.

The challenge, this time, looks to come from climate change that modifies both freshwater and marine conditions on a large scale, and rapidly. Despite all of Alaska’s bragging about our sustainable salmon management, we may find ourselves up the proverbial creek. This time, the degree and speed of change may be more than salmon, as species, can adapt to.

Freshwater—not just in streams but also in lakes, wetlands, and the water table, and in its various forms of precipitation and storage, including in glacial ice—is absolutely key to the Kenai Peninsula landscape and everything that landscape supports. Despite the high water I witnessed in May, I knew that the western peninsula was drying, and I wanted to understand both what that might mean and how the people who decide what it might mean do their work. On a coolish July morning, I joined my neighbor Ed Berg for a day trip into the heart of the Kenai National Wildlife Refuge, where he works as the refuge ecologist. We were joined by Dick Reger, a retired geologist who volunteers with Berg just for the fun of it, for a trip to the middle of the five Finger Lakes. Or Middle Finger Lake.

“We definitely got the idea why it’s called that,” Reger laughed as we readied gear at his cabin. A few days earlier, the two men, both in their late sixties, had canoed the length of that longest of the Finger Lakes against the wind.

Now, Berg, wearing an orange field vest with multiple bulging pockets and with his ever-ready hand lens on a cord around his neck, waited patiently while Reger showed me how to look at aerial maps with a stereoscope and pick out their three-dimensional features. Reger put on his own orange vest—with the six pens and pencils lined up in their pocket compartment—and tossed his lunch into a backpack. And then we were off, bumping down Swanson River Road into the refuge, nearly two million acres of federal land characterized by mostly scrubby forest, lots and lots of small lakes and two really big ones full of red salmon, plenty of wetlands, and on the southeast side, the Harding Icefield and its glaciers. This northern part of the refuge, near the town of Sterling, had been entirely burned in a 1969 forest fire and was now largely covered by skinny birch trees.

Berg and Reger explained as we drove that this lowland had once been lake bottom; during the last ice age, ice sheets had formed dams and impounded freshwater. The lakes that pock the refuge today, including the Finger Lakes, are known as “kettle lakes,” created from the melt of giant ice blocks left by retreating glaciers and filled by precipitation and groundwater; streams do not flow in or out of them, thus making them very useful for studying effects of climate change. Over time, they’ve been good recorders of what Berg calls “available water”—that is, precipitation minus the water that’s lost to evapotranspiration (the combination of water transferred to the atmosphere by evaporation and from the leaves of plants). The record is long; although glaciers remain in nearby mountains, the land here has been free of ice for about eighteen thousand years.

Our goal was to again canoe the length of the lake, for further investigation of a key geological feature called an ice-shoved rampart. The mystery of that berm of earth, and others within the refuge, might, the men thought, be unlocked into an understanding of past climate—and thus be useful for imagining a future.

As soon as we were out of the truck and lathering on mosquito repellent, Berg and Reger were examining and debating, with tangible excitement, oddities in the bark of some birch trees. This supreme inquisitiveness is what I love about Berg, whom I’ve known for years—since the time he was a carpenter—and from whom I’ve twice taken a geology class at our local college. Before the carpenter period of his life, Berg had been both a geophysicist and doctor of philosophy, and when he tired of hammering, he became a botanist. Berg is also particularly skilled at making science understandable to the public. His study of the spruce bark beetle and its warm-weather success at devastating Kenai Peninsula forests has been oft-reported in the popular press, in which he tends to be very quotable.⁶ “Beetles take no prisoners,” he once told reporters during a tour of the refuge. “It’s a Mafia-style execution.”

The woods, as we slipped through, were full of carefully observed insects, seedlings, bird calls, and a spruce tree clawed by a bear.

At the edge of the lake, we maneuvered into the water the canoe they’d previously left there. Reger, lingering in the shallows, examined acorn-sized freshwater snails; on the earlier trip he’d collected some for his home aquarium, and now he told me of the snail’s Asian origins and its passage—likely by birds—across Bering Strait. Berg, examining a sedge, quizzed Reger about its species.

Reger said to me, “Ed and I appreciate the same sorts of things. We have different backgrounds. He shows me this, and I show him something else. Everyone else thinks I’m a crackpot.”

The two men paddled, and I rode like an Egyptian princess in the middle of the canoe. A slight breeze rippled the water, but there was no real wind. The lake, long and narrow but indeed “kettlelike,” lay within steep sides and a surround of higher, forested land.

We stopped on a small island, and the two men went into high gear, digging holes and mixing soil samples with spit in their hands, referring to a soil color chart. This was not the day’s project but a side stop to explore how long the island had been an island as opposed to lake bottom, which would say something about water levels and climate. The men engaged in a vigorous discussion about soil “platyness” and how sand is winnowed from silt and the age of the vegetation on top. The island, they could tell, was wave-flattened, and the sandy soil on top had to have been deposited by waves that had washed over the island, perhaps not that long ago. There was essentially no soil on the island—just that sand over the silt that had been lake bottom. The trees—some birch, a few short spruces, alders—were sparse, and the ground beneath them was mostly mossy, with a few wintergreen plants and dwarf dogwood and, in a moist spot near the shore, reddish sundews with their tentacles and seductively dewy tips.

After the 1969 fire that destroyed so much forest in the refuge, Berg said, he would have expected the water level in the area to rise, not fall; trees would be drinking less, so more water would stay in the lakes. But, in general, the refuge’s lakes and wetlands were drying in the warmer temperatures and greater evapotranspiration associated with human-induced climate change.

In the expected cycles of warming and cooling, this part of Alaska should have been cooling. “Glaciers should be advancing,” Berg said, as he paused to wipe his brow with a bandana. “Climate change may be overriding the natural order of things.”

Then it was back into the canoe and only a short paddle to the south end of the lake, where we clambered out and approached a small berm with a hole dug into it and a pile of sandy soil to one side. This was it—our ice-shoved rampart, among the carpet of dwarf dogwood flowers and the dead leaves of the last year.

The men went back to work with a shovel, enlarging the hole they’d begun two days before on the landward side of the berm and from which they’d taken away a bit of woody evidence. Berg dug, and Reger squeezed and tasted soil samples, and then Reger dug and Berg talked to me about what they were doing. He spoke in his slow and always precise manner, in sentences that unfolded complete and orderly thoughts.

“What we’re doing here at Middle Finger Lake is excavating a berm that’s about two feet high, parallel to the shore, and it’s back thirty or forty feet from the shore. The base of the berm is about eight and a half feet above the present surface of the lake. These berms are formed by the ice bulldozing the lake sediments. This happens in the spring when the lake ice is breaking up and getting blown around by the wind. We find them on the south and southwest sides. The remarkable thing about these is that they’re so far above the present lake level—on other lakes we’ve found them twenty or twenty-five feet above the modern shorelines. Sometime in the past, the lake levels were very high—much higher than in the historic period. So we’re excavating this berm—or ice-shoved rampart—in hope of finding some pieces of wood we can date with radiocarbon. That will give us an idea about when these berms were formed.”

That—knowing when the lake levels were so high, the ice thick, and the northeast winds strong—could be matched to other climate data from that time period to tie regional effects to global ones and suggest linkages among conditions. An ability to “backcast,” as Sue Mauger was doing with stream temperatures on a shorter time scale, could help with forecasting a climate future.

On their previous outing, Berg and Reger had collected two small pieces of wood that, from their placement between bulldozed lake material and forest floor, appeared to have been at the bottom of two different rampart-building “shoving events.” Today, the goal was to get to a layer of soil underlying the whole berm. If they could find something woody there, they might get a maximum date—an upper limit on how old the berm might be.

Slate-colored juncos rustled through the underbrush, sparrows chattered, Reger paused to pinch some soil and what might have been part of a disintegrating root. He and Berg debated the theory they were working with—that the ramparts were created during the Younger Dryas period, some twelve thousand years ago. Younger Dryas was considered to be “a cold snap” after a warming period following the last big ice age, but there was some evidence from lake sediments and peat cores that it had been a wet period here on the Kenai.

That was the theory, but the first three radiocarbon dates they’d gotten from ramparts had dated from just fifteen hundred to five thousand years ago.

The two men argued the theory back and forth—whether the samples were good or might have been from deep roots, what the soil profiles told them, what it might mean if the ramparts were younger than they thought. Reger was leaning now to a mid-Holocene age. “That tells us a story, too.”

“A stranger story,” Berg said.

“It’s not that we’re arguing,” Reger said to me. “It’s just on the table.”

Reger was standing in a pretty big hole now. “Look what I see down here—a thin possible silt layer.” He and Berg studied the dirt, talked excitedly in technical terms, showed me—I could kind of see it—where the rampart ended and layers of overridden forest floor began. In another minute Reger was scraping, like an archaeologist, around a chunk of wood, and then a second one. Berg drew in his field notebook—the hole, the dimensions, notes on soil colors. He drew pictures of the little chunks of wood and then carefully wrapped them in aluminum foil.

Months later, when the radiocarbon dating was completed on those chunks of wood and twelve others from the ramparts at six different refuge lakes, the “stranger story” would be told. All the samples dated within the last fifty-two hundred years, in that recent interglacial period we know as the Holocene, when climate was thought to be reasonably stable. Why would there have been such high water at that time? What might have been going on with the climate then, here in this part of Alaska and globally? Were there other data sets that could support such a finding? Lake sediment studies in the refuge have suggested that the land in question had generally cooled and become wetter over the last nine thousand years, but there was, until now, no record of such extreme wetness.

Berg’s new theory posits that the ice-shoved ramparts associate with a large regional climate trend, perhaps involving “a series of stormy, high-precipitation anomalies that have occurred over the last 5,200 years, reflecting major changes in the North Pacific weather system.”⁷

We ate our sandwiches, and I wandered through the woods for a while, on animal trails that followed two smaller (and presumably more recent) ramparts that lay between the one we’d dug into and the shoreline. Had I come across any of these berms in the woods on my own, I would have guessed them to be glacial moraines or eskers, features I was more familiar with. In among the birch and the sweet-smelling balsam poplar stood a few blackened tree stumps, from the fire forty years earlier. Open areas around them were filled with bursts of purple fireweed.

I took a turn filling in the hole and then swatted more mosquitoes and listened to the far-off wavering call of a loon while Reger filled his plastic sandwich bag with lake plants to take home to his snail aquarium.

We boated back along the east side of the lake, watching for the indentation where we would portage to the next lake and then find a trail to the road.

I thought about the process of science—its posing of questions, all the tedious data collecting, the accumulation over time of observation, test results, reviews of results. The scientific process was slow and incremental, and conservative; it didn’t respond well to crises.

I tried to think as a geologist might, back through time and the processes that work on landscapes. Imagine a woody plant on a forest floor, twelve thousand years ago. Or five thousand years ago. What kind of a world did either date define? Twelve thousand years ago humans were just coming across, or along the coastline of, the Bering Land Bridge, land exposed because so much water was tied up in ice caps and glaciers. Five thousand years ago our ancestors were primarily hunters and gatherers, although, at least in Asia, farming was developing on a largish scale; some clever beings invented both the wheel and systems of writing. In both those time periods the amount of carbon dioxide in the atmosphere (measured from Antarctic ice cores) was around 250-280 parts per million. Today atmospheric CO₂ exceeds 392 parts per million and is continuing to rise; levels of it and other greenhouse gases are higher than they’ve been at any time in at least eight hundred thousand years, which is as far back as ice core records go. Never in those eight hundred thousand years years did CO₂ levels increase at a rate anywhere near what we’re experiencing today. Eight hundred thousand years ago Homo sapiens was still five hundred thousand years years away from evolving; our species has never had to cope with what are, indeed, unprecedented conditions—that is, unprecedented in our human history.

From his position in the stern, Berg talked about the Aleutian Low—that low-pressure center south of the Aleutian Islands in winter, characterized by high winds—that plays a major role in atmospheric circulation. If he and others could learn when the Kenai was particularly wet, climate modelers might be able to link that to other conditions at the time—for example the intensity of that Aleutian Low. Or to the advancement or retreat of glaciers in our part of Alaska, or to periods of intense storm activity in the Arctic. “That’s the practical way this information will be used,” Berg said. “This will help the climate modelers calibrate the models. In order to predict the future, we need to know the past. We need to run the model backward.”

More locally, knowing something about climates of the past and conditions associated with them should inform decisions about land and water use, species conservation, and fire protection. Would it be a good thing to encourage beavers to build dams and store more water in the future? Would it be smart to reserve water for salmon streams, as opposed to using it for industrial or agricultural use? Should forest fires be allowed to burn, or should they be controlled?

We cruised past a flock of golden-eye ducklings, all paddling furiously with no parent duck in sight. When we got over that excitement, Berg told me about other research related to “available water.” Various studies in the refuge, including analyzing photos that go back to the 1950s, have shown that wetlands are shrinking at accelerating rates. Ponds have disappeared, shrubs are filling in, and black spruces are expanding into areas that had once been too wet for them.⁸ (I see this myself. Flying over, as I do when traveling to Anchorage, I look down on the “bathtub rings” around drying ponds, as well as the new, dark growth of little spruces pushing into open areas.)

Local meteorological records have shown a 60 percent decline in available water in the Kenai lowlands between 1968 and 2009; onethird of that, Berg has calculated, is due to higher summer temperatures and increased evapotranspiration and two-thirds due to lower annual precipitation. “That’s a big change,” Berg said. “That’s 60 percent less water to recharge groundwater, fill up lakes and rivers, and be used by plants and animals. That’s pretty dramatic.”

With less water, areas once dominated by herbaceous plants—that is, leafy plants lacking woody stems—have been converting to shrubland at increasing rates—more than 12 percent per decade—and previously unforested areas are becoming forested at similar rates. Peat cores taken from the drying wetlands found no history of woody plant cover; that is, the sedge and sphagnum moss fens have dominated for eighteen thousand years. Only since about 1850 have those lands been drying—a drying that has greatly accelerated since 1970.

Let me repeat that: The current invasion of Kenai wetlands with shrubs and trees is unique in the last eighteen thousand years, and it is accelerating.

On the local level, the drying meant we were likely to see more—and more damaging—fires. Berg said, “What were fire breaks in the past are becoming fuel bridges.” A warmer climate, generally, will result in more fire activity. If the wetland areas that have acted as natural firebreaks between grasslands and forest dry out, they will no longer help control fires but will connect and speed them through the refuge and the rest of the Kenai lowlands. So far, the refuge has been experiencing more early-season fires—as early as April—in the grasslands that have replaced beetle-killed forests, and more fires have been ignited by lightning strikes. Lightning used to be a rare phenomenon on the peninsula, but has been increasing with the warming that builds thunderhead clouds. In 2005, six hundred lighting strikes started twenty-two Kenai Peninsula fires.

The conversion of our wetlands to a landscape of shrubs and trees is also significant for the global climate because of the major role wetlands play in the cycling of both CO₂ and methane. In the cycle that takes place everywhere on earth, trees and plants take up great quantities of carbon dioxide, release the oxygen, and store the carbon in their cells. (Wood is one-half carbon.) That carbon stays stored not just in the living vegetation but in woody debris and soil, until it’s released by decomposition or burning. Methane (CH₄) is another form of carbon, created in the absence of oxygen. (When it’s released to the atmosphere, where it’s a powerful greenhouse gas, it eventually converts to CO₂ and water.)

Northern peatlands hold a tremendous amount of CO₂ and CH₄, equivalent to somewhere between a third and a half of that in the atmosphere. In the anaerobic (lacking oxygen) situation of wet soils, the gases stay in place. As the soils dry, the microbes get to work, decomposing the organic matter and giving off CO₂. But at the same time, the plants and trees that fill in the wetlands perform more photosynthesis, which takes in CO₂. The net result of this carbon “flux” (the transfer or rate of exchange between carbon “pools,” as between, in this case, organic matter and the atmosphere) is not yet well understood.

A couple of months later, I returned to the refuge with a grad student, Sue Ives, helping her place a clear plastic box over vegetation plots and measure carbon flux with a gas analyzer machine. It was another example of small science steps, the necessary accumulation of place-specific data. It took most of the day to set the box over twenty plots marked out along a gradient—five very wet and mossy ones near the edge of a lake, five slightly drier ones with sedges and bog rosemary among the mosses, five dominated by brushy dwarf birches, and finally the driest, where small black spruce trees had begun to sprout up among cranberry and cloudberry plants—and to take readings with different cloud cover-imitating screens wrapped around the box.

So far it was looking as though, at least in summer, the drier plots were both respiring and photosynthesizing more than the wetter plots—with photosynthesis significantly outpacing respiration. The drier plots were sequestering more carbon than the wet ones.

On the surface, that seemed like a good thing—a landscape change that allowed for holding on to more carbon, keeping it out of the atmosphere. But summer measurements were, of course, not the whole story. In winter there would be little photosynthesis, but the microbes in the soil under the insulating snow would still be working, still be respiring. And there were other considerations: More plant growth would create a darker surface in winter, decreasing the albedo (surface reflectivity) and increasing the absorption of solar heat. The implications of more heat are earlier snowmelt, more drying, more plant and microbe activity, and more respiration and release of CO₂.

And this was just one type of wetland, in one place and set of conditionos.

In the next chapter, we’ll look more closely at the role of carbon in vegetation, soils, and permafrost—what makes a particular landscape a carbon “sink” or a carbon “source.”

Late that day, dark rain clouds rolled in over the mountains, threatening the whole day’s work. Suddenly, we heard bugling. From the north, long skeins of sandhill cranes were coming our way, high in the sky and stretching for miles. The enormous flock—hundreds of birds—passed to our east, their calls echoing across the landscape, and I remembered the first time I saw cranes, my first fall in Alaska. I had stood then in the yard of a homestead house in the hills outside Homer and watched them spiral up into a great cloud, and I had thought they were magnificent. They were no less magnificent thirty-odd years later, flying high in their multiple twisting V formations. The two of us standing on a boardwalk in a bog stopped working to watch them pass, and then we turned to more bugling and another tremendous flock, spread out like musical notes across shifting staffs. And then, a third time—but now the birds came directly over our heads, and lower, so that we could see their long necks extended and their legs trailing, broad wings slowly flapping, the feathers on the wing tips separate and pointing like fingers. There were thousands, and the noise was deafening.

The cranes were flying from their nesting grounds on the Yukon-Kuskokwim Delta and the tundra areas of northern Alaska and Siberia. They scatter in those places—wetland places, peat bogs and muskegs, wet tundra near water—into their nesting pairs and raise one, sometimes two, “colts” while they fatten on insects, seeds, frogs, even small rodents. They were going to wintering grounds, in the southwestern United States, Texas, Mexico, where they—birds with the longest fossil record of any bird still flying, going back perhaps ten million years—have adapted well, in recent time, to feeding on the waste grain in agricultural fields.

I wondered where they’d settle for the night, what field or wetland they’d find, and how different that field or wetland might be in another year, ten years, within the lifetime of a long-lived crane.