NEWS FEATURE CLIMATE CHANGE 2007 NATURE|Vol 446|12 April 2007

A world melting from the top down Despite years of speculation, little can be said for sure about the future of the Arctic’s permafrost. But that’s no grounds for complacency, reports Gabrielle Walker.

ome of Phil Camill’s trees are drunk. has brought with it a trebling of the thaw rate. For decades environmentalists have worried Once, the black spruce trees on the In some places the permafrost’s perimeter is about the possibility of this great putrefaction. plots of woodland that he monitors in retreating by 30 centimetres a year1. If this trend It has become perhaps the most cited exam- Snorthern Manitoba stood as straight continues, Camill estimates that no permafrost ple of the biogeochemical feedback that could and honest as pilgrims. Now an ever-increas- will be left in any of his five sites by the end of drastically worsen the effects of anthropo- A. COOPER/ALAMY ing number of them loll about leaning like the century. genic climate change. The idea is that humans lager louts. The decline is not in the moral Thawed-out permafrost has already under- increase levels of carbon dioxide in the atmos- standards of Canadian vegetation, but in the mined buildings, highways and other infra- phere, warming the permafrost, which in turn shifting ground beneath their roots. Once it structure from Alaska to Siberia. The damage is releases yet more carbon, warming the world was all hard, solid permafrost. Now much of it one of the most visible effects of warming tem- — and the permafrost — further still in an has thawed into a soggy sponge that no longer peratures on human activities. But the effects ever-escalating positive-feedback loop. provides a steady footing for the trees. Some on natural systems are to some extent more However, although such feedback has been contrive to grow at screwball angles; others have worrying. Buildings can be rebuilt, asphalt discussed for almost as long as the threat of drowned and been replaced by floating mats of relaid and agricultural practices changed global warming has been taken seriously by mosses or sedges. “It is really easy to tell when through adaptation, given the right policies scientists, the lack of firm data on the subject the permafrost has gone,” says Camill. “The and priorities (see page 716). But changes in is striking. “There is a lot that we don’t know at vegetation changes right before your eyes.” the vegetation and, crucially, in the soils of the this point,” says Walter Oechel from San Diego Camill, an ecologist from Carleton College frozen northern landscapes might not be so State University in California. “People haven’t in Northfield, Minnesota, has used those easy to cope with. The soils of the Arctic are quite pulled the whole picture together yet changes to trace the rate at which the perma- crammed with organic matter — a frozen res- — but what we do know is that the potential frost is disappearing. In their desperate attempts ervoir of beautifully preserved roots, leaves and amounts are huge and very, very scary.” to buttress themselves upright, his leaning other raw material that may contain as much spruces put on extra wood on the downslope carbon as the whole atmosphere. They are The big picture side of their trunks. Counting the asymmet- quite unlike soils from more temperate regions, There is no doubt that the Arctic is heating up. rical tree rings that result and measuring the which are mostly made up of the parts that the Vladimir Romanovsky from the University of distance of each tree from the current bound- bacteria cannot digest. “We are unplugging Alaska Fairbanks has collated borehole- and air- ary of the permafrost gives a measure of the pace the refrigerator in the far north,” says Camill. temperature data from throughout the Arctic2. of change. The results are shocking. An average “Everything that is preserved there is going to He found that only one region in the Arctic warming across his sites of 1.3 °C since 1970 start to rot.” had not warmed over the past 30 years — and

718


  
 
NATURE|Vol 446|12 April 2007 CLIMATE CHANGE 2007 NEWS FEATURE

in the 1990s even that region joined the trend. flourish and decompose the defrosted organic its future course is even harder. In 2005, two Some places are warming at more than twice matter. The probability that the active layer researchers from Boulder, Colorado — David the global average rate. Romanovsky recently will deepen — putting a larger stash of carbon Lawrence from the National Center for Atmos- received a US$1 million grant to take this moni- up for grabs — or that the permafrost will thaw pheric Research and Andrew Slater from the toring work further with a network of stations completely depends on the type of vegetation Cooperative Institute for Research in Environ- in North America and Russia. and soil. Thus, thawing can accelerate rapidly mental Sciences — published the results of the The effect that this warming will have on the if a fire passes through a dried- first attempt to project the fate permafrost and its stored carbon will vary from out forest in the uplands, or if of the permafrost through the region to region. Not surprisingly, the most the soil contains enough ice “We are unplugging twenty-first century using the dramatic signs of thaw have come from the that thawing causes it to col- the refrigerator climate predictions of a gen- fringing, southerly regions of discontinuous lapse, creating a crater-scarred in the far north. eral circulation model (GCM). permafrost — such as Camill’s research sites ‘thermokarst’ landscape. Their results made dramatic — where the frozen layer is only a few metres Researchers expect to see the Everything that is headlines. Using figures from thick and average temperatures are already first signs of thaw in a deepening preserved there is one of the ‘high emissions’ sce- within a whisker of the melting point. In the of the active layer. But compli- going to start to rot.” narios developed by the Inter- colder Arctic the permafrost can be hundreds cations from terrain, vegetation governmental Panel on Climate of metres thick and it is harder to know what and other local conditions mean — Phil Camill Change (IPCC) in the model to expect. In principle, the thawing might be that data need to be collected resulted in 90% of the northern quite slow, with the warmth at the surface being continuously over several dec- permafrost disappearing by transmitted gradually into the colder depths. ades to pick up such a trend. Unfortunately, 2100. Of 10.5 million square kilometres of per- In practice, things are probably more complex, the relevant measurements so far have been mafrost around today, only about 1 million made and in places more precipitous. patchy and sporadic. In 1998, various organiza- it through the century. Even more worryingly, tions involved in permafrost research banded running the model with a ‘low emissions’ sce- Patchy progress together to receive funding as the Circumpolar nario still wiped out 60% of the permafrost, sug- Permafrost is defined as ground in which the Active Layer Monitoring (CALM) programme, gesting that severe losses are inevitable no matter temperature is less than 0 °C for at least two hosted at the University of Delaware in Newark, which policies are followed. And the model did successive years. But the ground in question to tackle the problem. The network now includes not take into account any further warming from does not have to be at the surface. In most places 125 active sites and has participants from 15 carbon given off in the thaw3. the top part of the soil thaws during the countries. Still, it is unlikely to bear fruit in the summer, providing plants form of spotting unequivocal Thick and thin and microbes with an trends for some time. The model has since attracted some criticism, ‘active layer’ above If measuring the most notably because its permafrost is a mere the permafrost damage done so 3.4 metres thick throughout the Arctic, which in which far is hard, is far from the hundreds of metres present in they can predicting some regions. However, Lawrence says that he has since re-run the model for a mid-way emis- sion scenario and dealt with some of the other criticisms at the same time, and the results remained more or less the same. He says that the model captures many important aspects of the Arctic system — including the hydrology and physical properties of the soil — and that the point is not so much the actual percentage of loss but the overall principle that a fright- eningly large amount of permafrost could be vulnerable to quite small changes in climate. “In this field you have to accept that we won’t have a perfect knowledge of what’s happening up there,” he says. “But we should be able to capture the fundamental properties. And so far the model shows that a major change is going to happen to the Arctic.” Perhaps the most intriguing of the com- plications that Lawrence’s original work did not address is the suspicion in the minds of some researchers that the permafrost itself is putting up a defence against the thaw — a set of negative feedbacks. For instance, Oechel points out that warmer temperatures lead to a thicker layer of moss on the surface of his

research sites in the Alaskan tundra. Because 1.0. V. (CAPS), SYST. PERMAFROST ACTIVE-LAYER 1998. CIRCUMPOLAR ASSOC., PERMAFROST INT. UNEP/GRID-ARENDAL; REKACEWICZ, P.

719


  
  NEWS FEATURE CLIMATE CHANGE 2007 NATURE|Vol 446|12 April 2007

moss is a superb insulator, especially added in they are still sources, not when dried out by surface warming, sinks, overall5. Another factor is the the thickened vegetation helps to northern march of the treeline. As shield the frozen soil beneath from the region warms, the growth of new the warmth above. In one of the forests over what was once tundra north–south transects he studies on could also help to reduce the net Alaska’s North Slope, the warmest, release of carbon. southernmost section has both the However, even if the surface eco- thickest layer of moss on the surface system starts to take up more car- M. HOSHINO/MINDEN PICTURES/FLPA M. HOSHINO/MINDEN and the shallowest active layer. bon, the old, dead material frozen in Frederick Nelson at the Univer- the soil still needs to be considered. sity of Delaware points to another What’s more, the overall amount of self-preserving feature of perma- carbon emitted is not necessarily frost. The base of the active layer, the whole story. Torben Christensen he says, can become especially icy at Lund University in Sweden also because water draining down there has data on the carbon balance that will pool above the impermeable span several decades, in his case layers below. The richer this layer from a low-lying mire in the patchy is in ice, the more difficult it is to permafrost of northern Sweden6. As thaw, preventing the active layer the permafrost has steadily thawed, from deepening further4. Even snow the ground has grown soggier; the cover does its part. Like moss, snow previous hummock vegetation has is an effective insulator. But because been replaced by sedges, which are it falls in the winter, it works in the Mosses act as insulators for underlying permafrost. better at tolerating wet roots. The opposite direction, shielding the water seals the underlying soil from soil from cooling further in air temperatures bout of warming in the late 1970s, he saw the the air, which means that decomposition has that can be as low as –40 °C. Camill noticed region emit a mighty pulse of carbon dioxide to proceed without the benefit of oxygen. that the snow at some of his sites has thinned — a pulse that is now showing signs of tailing This slows things down, and has reduced the in the past few decades, which may cause the off. He suggests that one possible reason for the region’s carbon emissions by 13% since 1970. soil to grow colder in winter than it used to decline is that the initial orgy of decomposition However, the carbon that is released by this and thus store up protection against the heat spurred by the warming released nitrogen-con- oxygen-free decomposition comes out in the of the following summer. Increased forest taining nutrients into the soil. The plants have form of methane, rather than carbon dioxide, cover can have a similar effect, causing snow now responded to the additional nutrients by and methane packs a far greater warming to be thinner beneath the trees than it would growing more and taking up carbon dioxide in punch than its oxidized sibling. So the overall have been on open ground. “At face value you the process. Oechel thinks that in the summer greenhouse effect of the mire has actually gone would expect the permafrost to start thawing some of his sites now take up more carbon than up by a disturbing 47%. really rapidly as temperature rises, but these they emit, although when winter emissions are feedbacks can keep it around longer than you Back to biology would expect,” says Camill. These are the sorts of issues that future attempts to model the process will need to take Pushing the boundaries into account. Lawrence says that the next phase But these effects can’t last forever — and of this work will be to treat the biology more although they might forestall thawing, the sud- carefully, incorporating carbon and nitrogen den change in conditions when one or more of cycles into the models, allowing the vegeta- them fails might lead to quicker thawing there- tion to respond to the changes in climate, and after. A few really hot summers could break modelling sources of methane. According to through the ice barrier. And Oechel is already Nelson at least half a dozen groups around the V. ROMANOVSKY, UNIV. ALASKA FAIRBANKS UNIV. ROMANOVSKY, V. worried about his mosses. He has noticed that world are planning to develop their GCMs to they are highly sensitive to direct sunlight, and address at least some of these aspects of the now that the Arctic has fewer cloudy days the permafrost thaw, and the plan is to have a mosses could well begin to suffer. Loss of any or much more extensive set of predictions in the all of these protections would allow any thaw to next IPCC report, in five years time. accelerate. “It is difficult to push permafrost over Knowing more about the thaw, though, will the threshold of thawing,” says Romanovsky. allow useful predictions of carbon emissions “But after tipping it will go by itself”. only if researchers can also quantify the amount Uncertainties about thawing obviously com- of organic matter in the soils. A series of work- plicate the question of how much greenhouse shops under the auspices of the Global Carbon gas the permafrost will emit — but they are Project and the International Polar Year is aim- not the only complications. Oechel has been ing to answer those questions. Current estimates tracking the carbon balance of his Alaskan Thawing ice under the surface can cause the of the amount of carbon that might be in play tundra for several decades. After a serious landscape to collapse. range from 350 gigatonnes to more than 900 Gt;

720


    NATURE|Vol 446|12 April 2007 CLIMATE CHANGE 2007 NEWS FEATURE

Fire and ice: decomposition under thaw lakes is releasing large amounts of methane, which bubbles to the surface and adds to carbon emissions.

by way of comparison, the atmosphere contains of Alaska Fairbanks and colleagues have been greatly in the past few decades, leading to a rise 750 Gt or so. The estimates at the high end of the tracking the methane that bubbles out of thaw in methane emissions of nearly 60%. And that scale are based on the discovery of a new, vast lakes in northern Siberia. These lakes don’t nec- methane seems to be coming from the depths S. ZIMOV pool of buried carbon — a type of wind-blown essarily arise from global warming. Any local of the permafrost: its lack of carbon-14, an iso- soil called yedoma, which was laid down over disturbance can trigger a temporary thaw in tope continually made in the atmosphere that large tracts of northern Siberia in the ice ages. the permafrost. As the ice melts, the ground takes thousands of years to decay, suggests that sinks and fills with water. The lakes then tend the organic matter beneath the lakes has been Rich sources to migrate across the landscape, eroding away stored away for a very long time. Stuart Chapin, also from the University of their margins, and can last as long as a thou- This and the Arctic’s other warning signs Alaska Fairbanks, says that sand years; their sideways make it increasingly urgent that researchers the yedoma soil is extraor- “It is difficult to push motion allows them to eat resolve their remaining questions about the dinarily rich in carbon. Last through permafrost much fate of the permafrost. And those answers year, he and his colleagues permafrost over the more quickly than would a won’t come a moment too soon. “We have estimated the range and pos- threshold of thawing, steady heating pulse head- been asleep at the switch,” says Oechel. “If you sible thickness of this layer but after tipping it will ing down from the surface. look at the things that were said in the 1970s and calculated that it alone And decomposition in their about the Arctic’s response to increasing CO2, could contain 450 Gt of car- go by itself.” oxygen-free depths and the the place we were off is not that we overstated bon, compared to the esti- — Vladimir Romanovsky thawed sediment beneath or were overly pessimistic, but that we were mated 350–450 Gt in the rest will produce methane, not not aggressive enough about the predictions. of the Arctic7. This number, carbon dioxide. To me, the precariousness of the situation is says Chapin, is probably “only good to within How much methane nobody realized until now clear. We are in a world of hurt.” ■ a factor of two”; but even half of such a huge recently — mainly because it bubbles out from Gabrielle Walker is the author of amount would be significant, whereas twice random parts of the lake and disappears unno- An Ocean of Air. as much hardly bears thinking about. Camill ticed into the air. Walter and her colleagues points out that humans release around 9 Gt of managed to catch this methane in the act, by 1. Camill, P. Clim. Change 68, 135–152 (2005). 2. Romanovsky, V., Burgess, M., Smith, S., Yoshikawa, K. & carbon per year from fossil fuels and deforesta- noting where the bubbles emerged when the Brown, J. EOS Trans. AGU 83, 589–594 (2002). tion. “If just 1% of [the possible 900 Gt in the lake froze over in the winter, and then leaving 3. Lawrence, D. M. & Slater, A. G. Geophys. Res. Lett. 32, yedoma] is decomposed in a warmer world it instruments to catch the emissions throughout L24401 (2005). 8 4. Shur, Y., Hinkel K. M. & Nelson, F. E. Permafrost Periglac. would be as if we doubled our current rate of the following year . They calculate that bub- Process 16, 5–17 (2005). emissions. That’s what is alarming.” bling lakes from northern Siberia are already 5. Oechel, W. C. et al. Nature 406, 978–981 (2000). Ominously, the first signs that parts of the responsible for nearly four million tonnes of 6. Johansson, T. et al. Glob. Change Biol. 12, 2352–2369 (2006). continuous permafrost might now be thawing methane a year, and that the amount is on the 7. Zimov, S. A. et al. Geophys. Res. Lett. 33, L20502 (2006). have come from lakes that overlie this carbon- rise. Warmer temperatures mean that the lake 8. Walter, K. M., Zimov, S. A., Chanton, J. P., Verbyla, D. & rich yedoma. Katey Walter from the University area in Walter’s study region has increased Chapin, F. S. Nature 443, 71–75 (2006).

721