Reply to comment

Modeling Ice Age's End Lessens Climate Change Worries

Two articles in the July 17 edition of Science describe efforts to model Earth's rapidly changing climate at the end of the last glacial period, between 21 and 11 thousand years years ago (ka). After a year and a half of number crunching on Oak Ridge National Laboratory's Jaguar supercomputer, the first results indicate that climate experienced cooling 17 ka, during the Heinrich Event 1 (H1), followed by an abrupt warming at the onset of the Bølling-Allerød Warming 14.5 ka. These abrupt climate changes were accompanied by large changes in the “ocean conveyor belt”: the Atlantic meridional overturning circulation (AMOC). The results suggest that this transition can be viewed simply as the North Atlantic climate response to rapidly changing glacial meltwater flow. The findings call for a paradigm shift in our understanding of abrupt climate change and weakens the threat of “irreversible tipping points” so popular with climate change extremists.

The time from the Last Glacial Maximum (LGM, ~21 ka) until the Holocine warming had firmly taken control of Earth's climate (~11 ka) is the most recent period of rapid climate change in the long history of the Pleistocene Ice Age. A general observation, made by many researchers, is that glacial periods tend to end abruptly with a rapid transition from cold to warm. Scientists hope to understand more about what triggers sudden climate swings by studying the end of this most recent glacial period. Glacial termination did not take place as a single event—there were several wild swings in climate as the frozen Earth changed to the more temperate climate we now enjoy. Two major events during the transition were the Bølling-Allerød Warming (BA) and the subsequent Little Dryas cooling. To many, these events suggest that Earth's climate is a bi-stable system that can switch between stable warm and cold modes. This latest climate simulation suggests that the bi-stable “tipping point” hypothesis is not true.

In a perspective article in the same issue of Science, Axel Timmermann and Laurie Menviel of the University of Hawaii's International Pacific Research Center ask the question “What Drives Climate Flip-Flops?” Quoting from their perspective on Liu et al's report:

Around 14,600 years ago, the atmospheric circulation over the North Atlantic region flipped within just a few years to another state; also, Greenland temperatures skyrocketed by >10°C over several decades, terminating a cold phase known as Heinrich Event 1. The global impacts of this Bølling-Allerød transition have been well documented with climate proxy records such as sediment cores and ice cores, but the physical conditions that triggered the transition remain controversial. The temperature evolution from the Heinrich Event 1 to the Bølling-Allerød and the subsequent Younger Dryas cold phase (see the figure) is strikingly similar to the Dansgaard-Oeschger cycles that dominated Northern Hemispheric climate between 60,000 and 30,000 years ago. Hence, unraveling the processes that triggered the Bølling-Allerød transition may also help to elucidate the mysterious, tantalizingly regular Dansgaard-Oeschger cycles.

A. Timmermann et al., Science July 17, 2009.

Previously, long transient simulations have not been carried out using coupled atmosphere-ocean general circulation models (CGCM), which include the most advanced climate physics and are currently being used for future climate projections. In the paper “Transient Simulation of Last Deglaciation with a New Mechanism for Bølling-Allerød Warming,” Z. Liu of the Nanjing University of Information Science and Technology,et al. report on their analysis of data from extraordinarily detailed modeling runs using a state-of-art CGCM: the National Center for Atmospheric Research Community Climate System Model version 3 (NCAR CCSM3). Their results suggest a causal linkage between rapid climate change and heat transport by the AMOC.

Heinrich events were first described by marine geologist Hartmut Heinrich. During these events, huge armadas of icebergs broke off from glaciers and drifted across the North Atlantic. Scientists know this because they have found “ice rafted debris” in ocean floor sediments. Glacial icebergs contain rocks and dirt scraped up off the land the glaciers move over. As they melt, this rock debris is dropped onto the sea floor. Scientists studying marine sediments have found six distinct debris layers in cores of mud retrieved from the sea floor. These layers indicate six distinct events, which are labeled H1-H6. The last such event, at the onset of the Holocene warming, is called H1 or Heinrich Event 1.

Heinrich events. Figure by Leland McInnes.

In the simulation, a Northern Hemispheric freshwater forcing scenario was created in which the discharge of meltwater from the retreating glacial ice sheets during H1 suddenly stops. Thus, Liu et al. are able to simulate an abrupt recovery of the AMOC that triggers the transition from H1 conditions to the Bølling-Allerød. The results are in good agreement with paleoclimate reconstructions based on climate proxy records. The rapid AMOC recovery described by Liu et al. also involves an overshooting effect (see the figure above) that was noted in previous climate model simulations. To investigate the possibility that meltwater flux (MWF) was responsible for the change the researchers devised the following set of modeling scenarios:

The MWF was then reduced in two scenarios: a linear decrease to zero at 14.2 ka (DGL-B) and a constant flux (of 15 m/ky) until a sudden shut-off at 14.67 ka (DGL-A). Because the meltwater termination scenarios DGL-B and DGL-A represent the slowest and fastest possible MWFs, the two corresponding experiments represent two end members for simulations under more realistic MWF.

Though the recovery time was different in the two scenarios, the AMOC in both experiments peaked at ~19 sverdrup at the onset of the BA, or ~6 sverdrup greater than the glacial-state transport (~13 sverdrup). The sverdrup, named in honor of the oceanographer Harald Sverdrup, is a unit of measure of volume transport where 1 sverdrup = 106 m3/s. It is the equivalent of approximately 264 million US gallons per second. To comprehend the magnitude of water flow being described here consider that the total input of fresh water into the ocean from all the world's rivers is equal to about 1 sverdrup. Overall, the resulting AMOC is characterized by a deeper and stronger circulation, which is comparable with that during the Holocene.

Figure 2 from Liu et al., Science July 17, 2009.

Results from experiment DGL-B (in °C) are shown in Fig. 2 from the paper above. AMOC at (A) GLA, (B) H1, and (C) BA. Temperature at (D) GLA and temperature changes from the glacial state for (E) H1-GLA and (F) BA-GLA. Along with these changes in AMOC flow, a seesaw effect in surface temperature was predicted up until H1, followed by global warming peaking at the BA. The BA warming was dominated by a maximum warming at northern high latitudes while the warming from H1 to BA was global, with the maximum warming relative to H1 exceeding 20°C in the North Atlantic and Arctic. The final conclusions of the modelers are:

In contrast to previous mechanisms that invoke AMOC multiple equilibrium and Southern Hemisphere climate forcing, we propose that the BA transition is caused by the superposition of climatic responses to the transient CO2 forcing, the AMOC recovery from Heinrich Event 1, and an AMOC overshoot.

In other words, the change wasn't the AMOC suddenly jumping to a new climate equilibrium, it was a number of different factors acting in concert to perturb the system, causing several wild swings as Earth's climate transitioned to a warmer state. The listing of CO2 first among the causes of these ancient swings in temperature is misleading. Certainly the 40 parts per million by volume increase in atmospheric carbon dioxide that accompanied Heinrich Event 1 was a contributor to the Bølling-Allerød warming and further accelerated the deglaciation. The important point is that it was not the trigger for the event, nor did the increase in carbon dioxide prevent a return to colder conditions during the Younger Dryas. In fact, where the increase in CO2 came from is uncertain. “Its origin remains a mystery,” state Timmermann and Menviel.

According to previous work (see “Atmospheric CO2 Concentrations over the Last Glacial Termination , in Science 5 January, 2001) by Eric Monnin et al., “sudden CO2 increase could have been caused by changes in thermohaline circulation.” Other possible sources are from the increase in methane during the BA warming, which would rapidly react in the atmosphere to produce more CO2. Again according to Monnin et al., the increase in methane “is thought to have been caused by an intensified hydrological cycle during the B/A warm phase, which led to an expansion of wetlands in the tropics and northern latitudes.” Certainly this fits with recent findings that changes in sea-level drive changes in CO2, not the other way around (see “Ice Ages & CO2, Part II – Rising Sea-levels in Tahiti”). Once again, CO2 only plays a supporting role in climate change.

An hypothesis has been advanced by a number of scientists describing how a megaflood from the Laurentian inland ice could have been responsible for the Bølling-Allerød warming. Cold fresh water from a glacial lake under the Laurentian ice sheet entered the Gulf of Mexico, forcing warm surface water into the Gulf Stream, which restarted the AMOC in the North Atlantic. This in turn caused the BA warm phase. Many of the events required for this scenario to play out have support in geological data. Such an occurrence is not incompatible with the modeling study's results, unfortunately (or perhaps fortunately) there is no way to test this hypothesis directly since North America is no longer buried under glacial ice. Deglaciation events during the last glacial are shown in the figure below. The bold line marks the ice margin at the end of each time interval. Arrows mark meltwater routes.

(a) 16.1 – 15.35 ka, (b) 14.7 – 13.85 ka, (c) 11.4 – 10.85 ka.

Regardless of its source, it seems likely that large changes in MWF are needed to trigger abrupt changes like the Heinrich Event, the Bølling-Allerød warming and the Younger Dryas. More importantly from a modern point of view, the conditions necessary for a sudden shift in climate no longer seem to be present and the risk of a catastrophic “tipping point” event seems remote. Previous work with simplified models had suggested that gradual changes in MWF could trigger a state change in the AMOC: the infamous tipping point hypothesis. This new result using more a more complex model—and a heck of a lot of computer cycles—seems to indicate otherwise. According to Liu et al.: “Our results suggest that the current generation of CGCMs, like CCSM3, may not be able to induce an abrupt onset of BA warming under a gradual forcing. Is the current generation of CGCMs deficient in generating the abruptness of climate changes? Is the AMOC hysteresis a fundamental feature of the real-world AMOC as suggested in intermediate models, or not essential as suggested in current CGCMs?” The bottom line, as stated by the modelers themselves: “Current observations are insufficient to address these questions unambiguously.”

Are scientists any closer to being able to accurately model Earth's climate? Not really, at least not at the precision and time scales required for IPCC like predictions and certainly not any time soon. According to Timmermann and Menviel: “Even completing the CCSM3 simulation by running it into the present will require another 2 to 3 million CPU hours on the Jaguar supercomputer.” In other words, to bring the computation up to the present would require more than 340 years of continuous supercomputer time at ORNL. Currently even the most complex models are not up to the job of making accurate short term predictions, predictions on a scale of decades or even a few hundred years. Liu et al. are using climate models properly, to provide insight as to what mechanisms are at work and how they might interact. The results from their model are in no way a minute by minute recreation of how the last glacial period actually ended.

As Timmermann and Menviel say at the end of their perspective, “Ultimately, breakthroughs in our understanding of Earth's climate evolution will come from close interactions between paleoproxy experts, paleoclimate modelers, and climate dynamicists. It is time to train a new generation of scientists familiar with all these fields.” Perhaps that is also the solution to the global warming debacle, the arrival of a new generation of better trained, more widely knowledgeable climate scientists. A new generation of climate scientist who understand that climate models provide insight not proof.

Be safe, enjoy the interglacial and stay skeptical.


  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.