24 September 2013

Is climate change already dangerous? (4): Tipping points and climate modelling

by David Spratt

Fourth in a series

A tipping point may be understood as a step change, or passing of a critical threshold, in a major earth-climate system component, where a small perturbation (a small push or change) unleashes a bigger change in the component.  Potsdam Institute Director, Prof. Hans Joachim Schellnhuber, says that tipping points “identify the most vulnerable components (tipping elements) of the Earth System, the critical warming thresholds where the respective Earth System elements flip into a qualitatively new state”.  These elements include ecosystems, major ocean and atmospheric circulation patterns, the polar ice sheets, and the land- and ocean-based carbon stores.

This process is often tied to positive feedbacks, where a change in a component leads to other changes that eventually “feed back” onto the original change to amplify it.  The classic case in global warming (or, in reverse, cooling) is the ice-albedo feedback, where decreases (increases) in the ice cover area change surface reflectivity (albedo), trapping more (less) heat and producing further ice loss (gain).
In some cases, passing one threshold will trigger further threshold events, for example where substantial releases from permafrost carbon stores increase warming, releasing more permafrost carbon but also pushing other systems, for example parts of the Antarctic ice sheet, past a threshold point.

Once a tipping point is crossed, it is irreversible (under natural conditions) within certain time frames, so the consequence is to significantly affect the earth’s climate and ecosystems, for example by raising temperatures or greenhouse gas levels, or changing the efficiency of the land and ocean carbon sinks.  Given enough time and the right conditions, most processes (but not extinctions, for example) can be reversed.

In a period of rapid warming, most major tipping points once crossed (ice sheet loss, large-scale land carbon store releases such as permafrost) are irreversible on human time frames running to a few generations, principally due to the longevity of atmospheric CO2 (several thousand years). Large-scale human interventions in slow-moving earth system tipping points might allow a tipping point to be reversed (for example, a large-scale atmospheric CO2 drawdown program, or solar radiation management).

There is discussion, for example, that Arctic sea-ice loss is “easily reversible” in a cooling world, but that is easier said than done.  That would require greenhouse gas levels to be reduced significantly, below the level equivalent to the temperature at which the sea-ice system tipped in 2007, to produce a sufficiently cooler world.  This would be around 300–325 ppm CO2, compared to the present level of 400 ppm, so it is not so “easy” in the real world.

The scientific literature on tipping points is relatively recent, with a significant contribution by Lenton, Held et al. in 2008 on “Tipping elements in the Earth’s climate system” in an issue of the journal Proceedings of the National Academy of Sciences devoted to the subject. However, our knowledge is limited because “a system-level understanding of critical Arctic processes and feedbacks is still lacking” (Maslowski, Kinney et al.) and “no serious efforts have been made so far to identify and qualify the interactions between various tipping points” (Schellnhuber).
 
Climate models are not yet good at dealing with tipping points. This is partly in the nature of tipping points, where a particular and complex confluence of factors suddenly change a climate system characteristic and drives it to a different state. To model this, all the contributing factors and their forces have to well identified, as well as their particular interactions, plus the interactions between tipping points. Duarte, Lenton et al. conclude that “complex, nonlinear systems typically shift between alternative states in an abrupt, rather than a smooth manner, which is a challenge that climate models have not yet been able to adequately meet”.

The classic case was the Arctic sea ice “big melt” in 2007. Many models, including those on which the 2007 IPCC report had relied to conclude that Arctic sea-ice was pretty much likely to remain till the end of the century, did not fully capture the dynamics of sea-ice loss. Thus when in 2007 the summer sea-ice extent dropped radically compared to previous years, some model-oriented researchers exclaimed that the Arctic was melting “a hundred years ahead of schedule”.  

Even today, papers are still being published with modelling that suggests a sea-ice free Arctic will not occur till mid-century. Given the observations, it’s difficult not to conclude that given a choice between their models and real-world observations, some modellers will always choose the former. 

In an overview of the current state of Arctic climate research, Maslowski, Kinney et al. conclude that: “Model limitations are hindering our ability to predict the future state of Arctic sea ice”, and that the majority of general climate models (GCMs) including those used in IPCC (2007) “have not been able to adequately reproduce observed multi-decadal sea-ice variability and trends in the pan-Arctic region”, and their ensemble mean trend in September Arctic sea-ice extent “is approximately 30 years behind the observed trend”.

For example, what would be the impact of a sea-ice-free Arctic summer and the consequent amplified regional warming on the stability of the Greenland Ice Sheet (GIS)? Research does not yet provide a robust framework for considering such questions, yet most scientists if asked for their expert elicitation would probably say that it is hard to imagine the GIS doing anything other than melting at an accelerating rate and passing a critical tipping point in such circumstances.

The sea-ice model that has performed best (acronym NAME), is one of a new range of more specialised regional climate models developed by Dr Wieslaw Maslowski and colleagues. Maslowski is highly regarded, in part because his position at the American Naval Postgraduate School has given him unique access to half a century of Arctic sea-ice thickness scans from polar US military submarines. Maslowski told BBC News:
In the past… we were just extrapolating into the future assuming that trends might persist as we’ve seen in recent times. Now we’re trying to be more systematic, and we’ve developed a regional Arctic climate model that’s very similar to the global climate models participating in IPCC assessments. We can run a fully coupled model for the past and present and see what our model will predict for the future in terms of the sea ice and the Arctic climate. 
He emphasizes “the need for detailed analyses of changes in sea ice thickness and volume to determine the actual rate of melt of Arctic sea ice”, and concludes that:
The modeled evolution of Arctic sea ice volume appears to be much stronger correlated with changes in ice thickness than with ice extent as it shows a similar negative trend beginning around the mid-1990s. When considering this part of the sea ice–volume time series, one can estimate a negative trend of −1,120 km3 year−1 with a standard deviation of +/-2,353 km3 year−1 from combined model and observational estimates for October–November 1996–2007. Given the estimated trend and the volume estimate for October–November of 2007 at less than 9000 km3, one can project that at this rate it would take only 9 more years or until 2016 +/-3 years to reach a nearly ice-free Arctic Ocean in summer. Regardless of high uncertainty associated with such an estimate, it does provide a lower bound of the time range for projections of seasonal sea ice cover.
The point cannot be emphasised enough that the best-performing Arctic sea-ice model projects 2016 +/-3 years to reach a nearly ice-free Arctic Ocean.

Arctic sea ice volumes estimates from observations and from the NAME model
(Maslowski, Kinney et al., 2012, Figure 9)
 The non-linear problem still plagues many Arctic GCMs, and indeed parts of the IPCC process which largely excludes tipping points and carbon cycle feedbacks from consideration, exemplified by the 2007 IPCC’s reticence on sea level rises. Several fundamental projections found in IPCC reports have consistently underestimated real-world observations in at least eight key areas.  In its February 2007 report on the physical basis of climate science, the IPCC said that Arctic sea-ice was responding sensitively to global warming: ‘While changes in winter sea-ice cover are moderate, late summer sea-ice is projected to disappear almost completely towards the end of the twenty first century.’ And apparently the forthcoming 2013 IPPC AR5 has omitted consideration of permafrost feedbacks – another glaring example of that body’s scientific reticence (Romm, 2012).
  • Concluding post: Summing up