Polar bear, Arctic sea ice all-but doomed: Misleading Nature cover story misleads the media, public

Last week Nature published a study, “Greenhouse gas mitigation can reduce sea-ice loss and increase polar bear persistence” (subs. req’d). The journal had a pretty sensational cover, with a polar bear and the compelling headline, “Staying Alive: Cut greenhouse-gas emissions now we can still save the polar bear.”

If you missed Nature, you probably saw the headlines:

I really wish any of that were realistic, not so much because the polar bear is a critical linchpin species, but because the loss of Arctic ice in the summer may well trigger even more rapid warming (see “Tundra 4: Permafrost loss linked to Arctic sea ice loss” and below). But in fact a much more reasonable AFP headline would be “Arctic ice cap on verge of runaway melting: study.” The NSF release should read, “Polar bear extinction now likely.”

I understand that journalists typically don’t read studies closely, but Nature ought to know better. Perhaps, as we will see, it is just a matter of climate scientists of being utterly divorced from the reality of our energy and political systems. Still, in reading the study and its supplementary information, I am puzzled why Nature published the article as written and especially why it chose to sensationalize it on the cover.


Let’s set aside, for now, the fact that the study focuses on sea ice area, not volume. This is key figure in the paper:

Figure 3: September sea-ice extent (50% concentration) recovers from a RILE [rapid ice-loss event] in a 2020 greenhouse gas commitment realization:

In the 2020 commitment realization, which was integrated from the same initial state as the A1B reference realization, greenhouse gas concentrations followed the A1B scenario until 2020, and were fixed thereafter. RILEs occurred in both realizations during the decade of the 2020s. In contrast to the reference run (red line), the substantial sea-ice recovery in the 2020 commitment scenario (purple line) supports the concept that RILEs represent natural sea-ice variability superimposed on a secular warming-induced sea-ice decline, rather than tipping points. All lines represent 10-year running averages compiled from the annual data.

Hurray! We’ve saved the Arctic and the polar bear!

All we need to do is get on that “2020 commitment scenario.” That shouldn’t be any problem, should it? How hard can it be to freeze GHG concentrations at 2020 levels? It can’t be that hard, can it, because Nature’s own news story on the article doesn’t tell its vast number of readers who are not expert in energy or GHG scenarios any more than this:

This paper provides reason to hope that the previous predictions of declines in polar bear populations can be avoided if concerted efforts are made to reduce greenhouse-gas emissions

The National Science Foundation release is a tad clearer:

The new Nature paper indicates that if greenhouse gas emissions were reduced substantially in the near future, rapid ice losses would be followed by substantial retention of the remaining ice through this century — and partial recovery of the ice that disappeared during the rapid ice loss.

In fact, if you wanted to stabilize concentrations the 2020 levels, you would in fact have to cut GHG emissions about 60% to 70% almost immediately and have further cuts over time (as I discuss here and RealClimate discusses here).


Nothing close to that is gonna happen — in case that wasn’t obvious to pretty much everybody on the planet other than the authors of this study, the peer reviewers, and the editors at Nature.

It is well known that the difference between stabilizing concentrations and stabilizing emissions is probably the the biggest source of climate confusion among even people who are pretty knowledgeable (as I discuss here based on the work of John Sterman, Director of the MIT System Dynamics Group at the Sloan School of Management, here).

It is simply inexcusable that Nature and the authors don’t explain this fact in very clear language.

The A1B scenario, for the record, is close to the one that we’re on through 2050, so the figure above makes clear that, on our current emissions path, sea ice is likely to plummet in the near future.

The study has a number of other limitations. The word “volume” or “thickness” never appears in it — even though sea ice has three dimensions and many cryoscientists believe olume is a more important indicator (see “Arctic Death Spiral 2010”). The study never mentioned the black carbon, although that is considered to be a major contributor to sea ice loss. The study concludes:

Our general circulation model outcomes did not reveal thresholds leading to irreversible loss of ice; instead, a linear relationship between global mean surface air temperature and sea-ice habitat substantiated the hypothesis that sea-ice thermodynamics can overcome albedo feedbacks proposed to cause sea-ice tipping points.

… the hypothesis that the climate system contains tipping elements means that habitats supporting cold-dependent species could disappear abruptly and irreversibly when a particular global mean surface air temperature (GMAT) is exceeded. It has been proposed that existing greenhouse gas emissions already have committed the earth to temperatures that will rise above the tipping point for loss of perennial Arctic sea ice.

Huh. That figure above looks awfully like a tipping point to me, once you get realistic about near-term emissions reductions. The tipping point notion is NOT that the ice vanishes abruptly when a certain temperature is exceeded, unless you are trying to define tipping point so narrowly that it has no utility. The tipping point is that the ice can’t plausibly be stopped from vanishing once a certain concentration level is hit because of the tremendous lags in the climate system.


I mean, if you want to get technical here, there could just about never be a tipping point because humanity could, theoretically, go to zero emissions — and then negative emissions (i.e. sequestration) over a very short period of time. But again, it is misleading to the public to leave the impression that, say, there is any plausible chance we would stabilize concentrations at 2020 levels.

I also think it strange that an article focusing on rapid ice loss events never cites “Accelerated Arctic land warming and permafrost degradation during rapid sea ice loss.” The lead author is David Lawrence of the National Center for Atmospheric Research (NCAR), who I interviewed for my book and recently interviewed again via e-mail about his recent work. The study’s ominous conclusion:

We find that simulated western Arctic land warming trends during rapid sea ice loss are 3.5 times greater than secular 21st century climate-change trends. The accelerated warming signal penetrates up to 1500 km inland”¦.

In other words, if we get rapid sea ice loss in the near future (as the Nature study projects), the recent trend in sea ice loss may triple Arctic warming, causing large emissions in carbon dioxide and methane from the tundra this century. That would seem to support the tipping point theory.

The study has an NCAR coauthor. It seems to me if the Nature authors don’t believe the earlier NCAR study is accurate, they need to explain why, not ignore it.

The study asserts:

The perception that nothing can be done to avoid catastrophic losses and ultimate disappearance of polar bears was exemplified in 2007 when the general media proclaimed polar bears were irreversibly doomed.

Well, it cites precisely one article for the final claim: “Kizzia, T. Alaska polar bears called doomed. Anchorage Daily News A1 (September 8 2007).” And the weird thing about citing and criticizing that article is that it quotes heavily from a study by the lead author of the current study, Steven C. Amstrup, of the USGS:

Shrinking sea ice will leave only a remnant surviving population of the world’s polar bears in the islands of the Canadian Arctic by mid-century, according to a breathtaking new study by the U.S. Geological Survey. Two-thirds of the world’s polar bears, including those along the coasts of Alaska and Russia, will have disappeared.

The loss of summer sea-ice habitat will be so profound for bear populations that regional efforts to protect them, such as restricting subsistence hunting or Arctic oil and gas development, will not be able to prevent their disappearance, the government scientists said.

Moreover, the bears’ doom is irreversible, the study said. Even a dramatic effort to reduce greenhouse gas emissions would not be enough to halt the near-term warming trend and save the coastal bears. The species might manage to survive in its remnant outposts if long-term warming trends are reversed, scientists said.

“Things could be turned around so that they don’t disappear completely,” said Steve Amstrup, the biological study team leader for the USGS. On the other hand, Amstrup said, climate-warming models chosen for the study tended to be conservative, so the bears might disappear faster than predicted.

“As the sea ice goes, so goes the polar bear,” Amstrup said….Amstrup rebutted the idea that polar bears could survive by adapting to land-based hunting. He said studies have shown the bears to be very inefficient hunters of land animals, which in any case do not provide the kind of rich nutrition polar bears seek.

So the new study’s findings, read realistically, aren’t much different than that of the 2007 study, which I discussed here: “Will polar bears go extinct by 2030? “” Part I.” Indeed, the new study makes perfectly clear that if we stay anywhere near our current emissions path, most of the polar bear population will be wiped out by midcentury.

The only ‘good news’ in this study “Greenhouse gas mitigation that keeps GMAT rise below 1.25”‰°C combined with traditional wildlife management could, it seems, maintain polar bear numbers at sustainable although lower-than-present levels throughout the century.” Note that GMAT is measured from 1980–1999 levels. So you’re talking about keeping total warming from preindustrial levels at below about 1.8°C, which requires keeping concentrations at or, more likely, below 450 ppm. That means we need to peak around 2015 to 2020 at the latest, then drop at least 60% by 2050 to at most 15 billion tons (4 billion tons of carbon), and then go to near zero net carbon emissions by 2100 (see “How the world can stabilize at 350 to 450 ppm”).

I say:

  1. Let’s do it
  2. Don’t hold your breath, Ursus maritimus.

As is pretty clear from this discussion, the Nature analysis is over-idealized. Because they focus their model on sea ice area and not volume, I’m more inclined to buy Admiral Titley’s data-driven projection based on three dimensions than the study’s model based on two (see “Arctic Death Spiral 2010”).

Rear Admiral David Titley, the U.S. Navy’s chief oceanographer and director of its climate change task force, “predicts an ice-free Arctic in late summer by 2020.” By mid-century, he thinks we could be seeing 2 to 3 months of ice free conditions.

“The survival of polar bears as a species is difficult to envisage under conditions of zero summer sea-ice cover,” concludes the 2004 Arctic Climate Impact Assessment, by leading scientists from the eight Arctic nations, including the United States. Another 2004 study, by Canadian scientists, agreed:

[G]iven the rapid pace of ecological change in the Arctic, the long generation time, and the highly specialised nature of polar bears, it is unlikely that polar bears will survive as a species if the sea ice disappears completely.

I think the Nature cover should have looked more like this: