Keine Kipppunkte beim arktischen Meereis: Was schmilzt, kann auch wieder gefrieren

Das arktische Meereis ist in den letzten Jahrzehnten geschmolzen. Das passt gut mit dem Wärmeplateau zusammen, auf dem wir uns gerade befinden. Alles andere wäre eine Überraschung gewesen. Ende 2017 sah die Entwicklung wie folgt aus (Arktis, blaue Kurve):


Abbildung 1: Entwicklung der Meereisbedeckung in der Arktis (blau) und Antarktis (rot). Quelle: Climate4You.


Seit 1980 hat sich die arktische Meereisbedeckung von 12,5 auf jetzt 10,5 Millionen Quadratkilometer verringert, was etwa 16% entspricht. Zuletzt hat sich das Eis wieder vergrößert (Abb. 1). Einige alarmistisch veranlagte Zeitgenossen hatten mit einem viel schnelleren Rückgang gerechnet. Laut einigen Vorhersagen, sollte das arktische Meer schon heute eisfrei sein. So hatte Al Gore dem Eis 2008 nur noch 5 Jahre gegeben, bis es vollends verschwindet. Eine Fehlprognose wie wir heute wissen. Genauso schlimm waren die Vorhersagen von Peter Wadhams., der nach 2016 einen eisfreien Nordpol postulierte. Ganz falsch. Ein Mann der bewusst die Pferde scheu machte.

Till Wagner und Ian Eisenman untersuchten 2015 in den Geophysical Research Letters, wie es zu diesen krassen Fehlprognosen kommen konnte. Die Autoren fanden, dass man sich zu sehr in Kipppunkte verliebt hatte, die in der Realität meist gar nicht existieren, aber sehr gerne von Modellierern (z.B. beim PIK) angenommen werden. Abstract:

False alarms: How early warning signals falsely predict abrupt sea ice loss
Uncovering universal early warning signals for critical transitions has become a coveted goal in diverse scientific disciplines, ranging from climate science to financial mathematics. There has been a flurry of recent research proposing such signals, with increasing autocorrelation and increasing variance being among the most widely discussed candidates. A number of studies have suggested that increasing autocorrelation alone may suffice to signal an impending transition, although some others have questioned this. Here we consider variance and autocorrelation in the context of sea ice loss in an idealized model of the global climate system. The model features no bifurcation, nor increased rate of retreat, as the ice disappears. Nonetheless, the autocorrelation of summer sea ice area is found to increase in a global warming scenario. The variance, by contrast, decreases. A simple physical mechanism is proposed to explain the occurrence of increasing autocorrelation but not variance when there is no approaching bifurcation. Additionally, a similar mechanism is shown to allow an increase in both indicators with no physically attainable bifurcation. This implies that relying on autocorrelation and variance as early warning signals can raise false alarms in the climate system, warning of “tipping points” that are not actually there.

Der Hang zum Fatalistischen wird den Modellierern jetzt zum Verhängnis. Sie hatten viel zu lange angenommen, dass der Rückgang des arktischen Meereises irreversibel wäre. Alles was geschmolzen ist, käme niemals wieder. Eine falsche Annahme. Natürlich kann Eis das schmilzt, auch wieder gefrieren. Eigentlich trivial. In einer Pressemitteilung erklärte die Scripps Oceanic Institution diesen eigentlich intuitiven Sachverhalt:

Research Highlight: Arctic Sea Ice Loss Likely To Be Reversible

Scenarios of a sea ice tipping point leading to a permanently ice-free Arctic Ocean were based on oversimplified arguments. New research by Till Wagner and Ian Eisenman, scientists at Scripps Institution of Oceanography, UC San Diego, resolves a long-running debate over irreversible Arctic sea ice loss. 

Ever since the striking record minimum Arctic sea ice extent in 2007, the ominous scenario of a sea ice tipping point has been a fixture in the public debate surrounding man-made climate change and a contingency for which Arctic-bordering countries have prepared. For decades, scientists have been concerned about such a point of no return, beyond which sea ice loss is irreversible. This concern was supported by mathematical models of the key physical processes (known as process models) that were believed to drive sea ice changes. The process models forecasted that increased global warming would push the Arctic into an unstoppable cascade of melting that ceases only when the ocean becomes ice-free. 

Implications of a permanently ice-free Arctic for the environment and for national and economic security are significant, driving deep interest in predictive capabilities in the region. Wagner and Eisenman’s research was co-funded by the Office of Naval Research (ONR) and by the National Science Foundation. It supports the goals of the Navy’s U.S. Arctic Roadmap, which calls for an assessment of changes in the Arctic Ocean to clarify the national security challenges for future naval operations as this strategic region becomes increasingly accessible. “The Navy has broad interest in the evolution of the Arctic,” said the ONR’s Frank Herr. “Sea ice dynamics are a critical component of the changing environmental picture. Our physical models lack important details on the processes controlling ice formation and melting, thus ONR is conducting a series of experimental efforts on sea ice, open water processes, acoustics, and circulation.”

During the past several years, scientists using global climate models (GCMs) that are more complex than process models found sea ice loss in response to rising greenhouse gases in their computer simulations is actually reversible when greenhouse levels are reduced. “It wasn’t clear whether the simpler process models were missing an essential element, or whether GCMs were getting something wrong,” said Wagner, the lead author of the study. “And as a result, it wasn’t clear whether or not a tipping point was a real threat.” Wagner and Eisenman resolve this discrepancy in the study in an upcoming Journal of Climate article,  “How Climate Model Complexity Influences Sea Ice Stability.”

They created a model that bridged the gap between the process models and the GCMs, and they used it to determine what caused sea ice tipping points to occur in some models but not in others. “We found that two key physical processes, which were often overlooked in previous process models, were actually essential for accurately describing whether sea ice loss is reversible,” said Eisenman, a professor of climate dynamics at Scripps Oceanography. “One relates to how heat moves from the tropics to the poles and the other is associated with the seasonal cycle. None of the relevant previous process modeling studies had included both of these factors, which led them to spuriously identify a tipping point that did not correspond to the real world.” “Our results show that the basis for a sea ice tipping point doesn’t hold up when these additional processes are considered,” said Wagner. “In other words, no tipping point is likely to devour what’s left of the Arctic summer sea ice. So if global warming does soon melt all the Arctic sea ice, at least we can expect to get it back if we somehow manage to cool the planet back down again.”

Auch am Hamburger Max Planck Institut für Meteorologie nimmt man nun Abstand von den Kippunkten. Dirk Notz beschrieb im PAGES Magazin 2017 die wichtige Rolle der natürlichen Klimavariabilität:

Arctic sea ice seasonal-to-decadal variability and long-term change



This short overview presents some recent work on the variability and long-term evolution of Arctic sea ice area. For space constraints, the focus was only on September sea ice coverage as this is the month with the strongest observed trends. The discussion can be summarized as follows:

– On seasonal time scales, atmospheric internal variability and its imprint on sea ice renders skillful predictions of September sea ice coverage more than two months in advance inherently difficult.

– On annual time scales, negative feedbacks stabilize the sea ice cover. There is no “tipping point” beyond which the loss of the remaining summer sea ice becomes unstoppable.

– On decadal time scales, internal climate variability can cause a substantial acceleration or temporary recovery of the sea ice cover that renders the evaluation of individual model simulations based on their short-term trends impossible.

– On longer time scales, internal variability causes a substantial spread in possible 30-year long trends supporting for the production of large model ensembles. Nevertheless, the impact of anthropognic forcing on the long-term sea ice evolution is clear, with an average loss of 3 m2 of September sea ice cover per metric ton of anthropogenic CO2 emission.

Auch Runge und Kollege beschäftigten sich 2016 in Nature Climate Change mit fehlgegangenen Klimaprognosen:

Detecting failure of climate predictions
The practical consequences of climate change challenge society to formulate responses that are more suited to achieving long-term objectives, even if those responses have to be made in the face of uncertainty1,2. Such a decision-analytic focus uses the products of climate science as probabilistic predictions about the effects of management policies3. Here we present methods to detect when climate predictions are failing to capture the system dynamics. For a single model, we measure goodness of fit based on the empirical distribution function, and define failure when the distribution of observed values significantly diverges from the modelled distribution. For a set of models, the same statistic can be used to provide relative weights for the individual models, and we define failure when there is no linear weighting of the ensemble models that produces a satisfactory match to the observations. Early detection of failure of a set of predictions is important for improving model predictions and the decisions based on them. We show that these methods would have detected a range shift in northern pintail 20 years before it was actually discovered, and are increasingly giving more weight to those climate models that forecast a September ice-free Arctic by 2055.

Modelle, die ein schnelles Verschwinden des Arktiseises postulieren, sind offenbar wenig vertrauenswürdig. Erst ab 2055 könnte es soweit sein, falls die Erderwärmung ungebremst fortschreiten sollte. Wie geht es nun wirklich weiter? Wissenschaftler nehmen heute an, dass sich das Meereis im arktischen Atlantik in den nächsten zehn Jahren wieder vergrößern wird, wie Eos 2016 unter Hinweis auf Yeager et al. 2015 berichtete. Grund: Die Ozeanzyklen.

Atlantic Sea Ice Could Grow in the Next Decade
Changing ocean circulation in the North Atlantic could lead to winter sea ice coverage remaining steady and even growing in select regions.

Weiterlesen in Eos.

Bei den Prognostikern ist zwischenzeitlich eine goße Diskussion ausgebrochen. Weshalb lagen sie dermaßen falsch? Welche Komponenten ihrer Modelle haben versagt? Eos berichtete über den Diskussionsstand hier. Scafetta & Mazzarella forderten 2015, dass die natürliche Variabilität in den Modellen nachgebessert wird:

Our results imply that the climate is regulated by natural mechanisms and natural oscillations that are not included yet in the climate models

Anstatt der vormals befürchteten stetigen Eisschmelze, gehen Forscher nun von einer starken Beeinflussung durch die natürliche Klimavariabilität aus. Hierzu eine Pressemitteilung der University of Colorado at Boulder von 2015 (via Science Daily):

Erratic as normal: Arctic sea ice loss expected to be bumpy in the short term
Arctic sea ice extent plunged precipitously from 2001 to 2007, then barely budged between 2007 and 2013. Even in a warming world, researchers should expect such unusual periods of no change — and rapid change — at the world’s northern reaches, according to a new paper.

„Human-caused global warming is melting Arctic sea ice over the long term, but the Arctic is a variable place, said Jennifer Kay, a fellow of the Cooperative Institute for Research in Environmental Sciences at the University of Colorado Boulder and co-author of the new analysis out today in Nature Climate Change. Natural ups and downs of temperature, wind and other factors mean that even as sea ice slowly melts, random weather can mask or enhance the long-term trend. For example, even in a warming world, there’s still a one-in-three chance that any seven-year period would see no sea ice loss, such as in 2007-2013, the new analysis shows. And the chaotic nature of weather can also occasionally produce sea ice loss as rapid as that seen in 2001-2007, even though the long-term trend is slower.

Neither time period should be used to forecast the long-term future of the region, Kay and her colleagues concluded. Some commentators tracking sea ice trends have used the recent „pause“ in sea ice loss to claim that human-caused climate warming is not occurring; others previously used the rapid decline from 2001-2007 to speculate about ice-free Arctic summers by 2015. Neither claim is warranted, the authors report. „To understand how climate change is affecting the Arctic, you cannot cherry pick short stretches of time,“ Kay said. „Seven years is too short.“ The research team, led by Neil Swart of Environment Canada, analyzed both long-term records of Arctic sea ice observations and an extensive dataset of results from global climate models. From the model runs, they could calculate the chances that certain types of scenarios could play out in a slowly warming Arctic: For example, just how likely is it that sea ice would not decline during a seven-year stretch? The team focused on September measurements of sea ice, which is when the extent reaches a yearly minimum. By early October, Arctic sea ice generally begins growing again, a seasonal response to colder temperatures and shorter days.

The researchers determined that a seven-year period is too short to accurately capture long-term sea ice trends in the region. Even given long-term melting, there’s a 34-percent chance of randomly getting an unusual period of no change or even growth in sea ice, and a 5-percent chance of a period of very rapid loss, similar to the decline in 2001-2007. The team also increased the time period of analysis, to see if longer spans of time would be long enough. In about 5 percent of model simulations, there were even 20-year time periods with no loss of sea ice, despite strong human-caused warming. „It is quite conceivable that the current period of near zero sea-ice trend could extend for a decade or more, solely due to weather-induced natural variability hiding the long-term human caused decline,“ said Ed Hawkins, a co-author and researcher at the National Centre for Atmospheric Science, University of Reading. „Human caused climate warming has driven a decline in Arctic September sea-ice extent over the past few decades,“ the new paper reports, and „this decline will continue into the future.“ But understanding how and why natural variability affects sea ice trends should help scientists better predict how sea ice will evolve in upcoming years and decades, with implications for natural ecosystems, shipping routes, energy development and more.

Paper: Neil C. Swart, John C. Fyfe, Ed Hawkins, Jennifer E. Kay, Alexandra Jahn. Influence of internal variability on Arctic sea-ice trends. Nature Climate Change, 2015; 5 (2): 86 DOI: 10.1038/nclimate2483

Die neue Prognosestrategie: Beobachtete Daten, empirische Zusammenhänge und Ozeanzyklen verstärkt in die Berechnungen einbeziehen. Die Zeit der physikalischen Freistilmodellierungen ist nun endgültig vorbei. Beispiel Barents Sea, Onarheim et al. 2015:

Skillful prediction of Barents Sea ice cover
A main concern of present climate change is the Arctic sea ice cover. In wintertime, its observed variability is largely carried by the Barents Sea. Here we propose and evaluate a simple quantitative and prognostic framework based on first principles and rooted in observations to predict the annual mean Barents Sea ice cover, which variance is carried by the winter ice (96%). By using observed ocean heat transport and sea ice area, the proposed framework appears skillful and explains 50% of the observed sea ice variance up to 2 years in advance. The qualitative prediction of increase versus decrease in ice cover is correct 88% of the time. Model imperfections can largely be diagnosed from simultaneous meridional winds. The framework and skill are supported by a 60 year simulation from a regional ice-ocean model. We particularly predict that the winter sea ice cover for 2016 will be slightly less than 2015.

Ist schmelzendes Meereis eigentlich so schlimm? Forscher fanden nun, dass die Schmelze das Leben in den freigegebenen Meeresgebieten anfacht. Presemitteilung der University of Southern Denmark von 2017:

Melting sea ice may lead to more life in the sea

Every year an increasing amount of sea ice is melting in the Arctic. This can start a chain reaction, which leads to increased production of algae and hence more food for creatures in the sea.

When spring arrives in the Arctic, both snow and sea ice melt, forming melt ponds on the surface of the sea ice. Every year, as global warming increases, there are more and larger melt ponds.

Melt ponds provide more light and heat for the ice and the underlying water, but now it turns out that they may also have a more direct and potentially important influence on life in the Arctic waters.

Mats of algae and bacteria can evolve in the melt ponds, which can provide food for marine creatures. This is the conclusion of researchers in the periodical, Polar Biology.

Own little ecosystems

– The melt ponds can form their own little ecosystem. When all the sea ice melts during the summer, algae and other organisms from melt ponds are released into the surrounding seawater. Some of this food is immediately ingested by creatures living high up in the water column. Other food sinks to the bottom and gets eaten by seabed dwellers, explains Heidi Louise Sørensen, who is the principal author of the scientific article, continuing:

– Given that larger and larger areas of melt ponds are being formed in the Arctic, we can expect the release of more and more food for creatures in the polar sea.

Heidi Louise Sørensen studied the phenomenon in a number of melt ponds in North-Eastern Greenland as part of her PhD thesis at University of Southern Denmark (SDU). Bo Thamdrup and Ronnie Glud of SDU, and Erik Jeppesen and Søren Rysgaard of Aarhus University also contributed to the work.

Food for seals and sea cucumbers

In the upper part of the water column it is mainly krill and copepods that benefit from the nutrient-rich algae and bacteria from melt ponds. These creatures are eaten by various larger animals, ranging from amphipods to fish, seals and whales. Deeper down, it is seabed dwellers such as sea cucumbers and brittle stars that benefit from the algae that sink down.

For some time now, researchers have been aware that simple biological organisms can evolve in melt ponds – they may even support very diverse communities. But so far it has been unclear why sometimes there are many organisms in the ponds, and on other occasions virtually none.

According to the new study, ‘nutrients’ is the keyword. When nutrients such as phosphorus and nitrogen find their way into a melt pond, entire communities of algae and micro-organisms can flourish.

From the Siberian tundra

Nutrients can find their way into a melt pond in a variety of ways, For example, they can be washed in with waves of sea water; they can be transported by dust storms from the mainland (for example, from the Siberian tundra); or they can be washed with earth from the coast out on the ice, when it rains.

Finally, migratory birds or other larger animals resting on the ice can leave behind sources of nutrient.

– Climate change is accompanies by more storms and more precipitation, and we must expect that more nutrients will be released from the surroundings into the melt ponds. These conditions, plus the fact that the distribution of areas of melt ponds is increasing, can contribute to increased productivity in plant and animal life in the Arctic seas, says Professor Ronnie N. Glud of the Department of Biology at SDU.

Warmer and more windy

There are further factors that may potentially contribute to increased productivity in the Arctic seas:

• When the sea ice disappears, light can penetrate down into the water.
• When the sea ice disappears, wind and storms can stir the water up, bringing nutrients up to the surface from deep water. When it gets warmer on the mainland, this creates more melt water, which can flow out into the sea, carrying nutrients in its wake.

Ebenfalls 2017 berichtete Deutschlandfunk Nova über Algenblüten unter dem Eis. Der Spin der Meldung ist überraschend positiv:

Diese Art von Algen im Meer nennt man auch Phytoplankton. Und sie sind sehr wichtig – denn sie regulieren das Klima. Die Algen nehmen beim Wachsen Kohlendioxid auf und geben Sauerstoff ab.


Foto: Greenland Travel. Verwendung mit freundlicher Genehmigung