The IPCC released a 23-page report summarizing the work of the fourth phase. Newspaper headlines suggested the document was more apocalyptic than the third summary of 2001. The reality is more complex.
Since 2001 huge numbers of scientific papers have been published on climate change. Major research efforts have been devoted to understanding the process by which increasing greenhouse gas (GHG) volumes will affect the world climate. But a careful comparison of the 2001 and 2007 synthesis reports does not immediately show substantial increases in the degree to which climate scientists feel confident about their predictions and their understanding of climate systems. For example, the width of the range of predictions of temperature rise has increased. Scientists acknowledge greater uncertainty about the impact of the sliding ice on sea level rise than they did in 2001.
The increased uncertainty does not extend to all aspects of the report. The 2007 authors say that they are more confident about climate changes at the regional level. In other words, they feel that they understand better how temperature and rainfall changes are going to affect various different parts of the world. Nevertheless, the degree of confidence in the main predictions appears to be no greater than six years ago.
What is going on? There are several possible hypotheses. First, as the IPCC process has become better known, it may be individual scientists have argued for greater caution in forecasts. Perhaps the process has become more political, with scientists from countries lagging in the response to climate change more likely to restrain the language of their colleagues. The IPCC report writing process is collegial, and individual scientists can alter the tone of what is written.
The last few days have not seen any evidence of merit to this first hypothesis. A small number of scientists have voiced concern that the report is not hard-hitting enough. But I believe that the same phenomenon occurred in 2001. I see no evidence of any systematic change.
A second hypothesis is that the participating scientists are now even more clearly aware of the complexities of climate than they were in 2001. Refinements to climate models may be showing greater unpredictability. This hypothesis is consistent with the language of the 2007 document, particularly on temperature rise and sea level increases.
What are the main changes between 2001 and 2007?
Temperature change up to now
In 2001, the IPCC reported its estimate that global temperature had increased by 0.6 degrees in the period 1901 to 2000. The updated figure for the hundred years to 2005 is 0.74 degrees, a substantially larger number. Both reports say that the increases have been highest in the high northern latitudes and greater over land than ocean.
Temperature changes in the future
The third assessment report (TAR) suggested that the range of temperature increase to 2100 is between 1.4 and 5.8 degrees. The second report from a few years earlier had said 1.0 to 3.5. The rise between the two older reports was due to an IPCC expectation that sulfur dioxide emissions would be lower in the future than it had previously expected. (All other things being equal SO2 emissions tend to decrease temperatures.)
By the time of the 2007 report, the IPCC had widened the range of the forecast for the end of the century to 1.1 to 6.4 degrees, suggesting a fall in scientists’ confidence in their ability to predict the future. The 2007 document says that the ‘uncertainties and upper ranges for temperature’ arise because the IPCC models suggest ‘stronger climate-carbon cycle feedbacks’.
This comment from the IPCC is the most important remark in its 2007 work. The organization is saying that the changes induced by global warming, in terms of temperature, rainfall, and wind, are tending to interact with sources and sinks of atmospheric CO2. For example, lower rainfall in the Amazon basin might be reducing the CO2 take-up of the forest. Or increasing ocean water temperatures may be reducing the ability of the seas to absorb CO2.
The 2001 report makes no mention of any link between carbon cycle changes and global temperatures. The 2007 document reflects the growing unease that land and ocean CO2 sinks are already beginning to lose their effectiveness. To repeat, this is the most important implication of the 2007 document.
The 2007 report should have also perhaps have noted another reason for concern. The IPCC temperature estimates are based on six scenarios for the future growth of emissions. The scenarios with the highest emissions growth are, of course, those with the highest expected temperature rise. Since 2001, the rate of emissions growth has been much faster than the policy-makers and scientists expected under any of the six scenarios. If this continues, the most likely expected atmospheric concentration of CO2 will be at the top end of the range, implying that the temperature rise will be nearer 6.4 degrees than 1.1 degrees.
To summarize, the IPCC’s range of possible temperature increase has widened and shifted slightly upwards. Within this range, the rapid growth of world emissions will mean that the likely eventual outcome is towards the higher end of the range of possible temperature rises.
For the next two decades, the mid-point estimate of temperature rise is approximately the same in both reports. Temperatures are forecast to rise by about 0.2 degrees per decade.
Sea level rises
In 2001, global mean sea level was projected to rise between 9cm and 88cm between the years 1990 and 2100. The melting of the Greenland ice cap was forecast to contribute ‘ a few cm’ to sea level during the 21st century.
In 2007, the scientists appear to have reduced the range of possible sea level rises, and its midpoint. The IPCC quotes a spread of 18cm to 59cm. It therefore looks as though scientists are more confident that they can forecast the rate of thermal expansion (the dominant cause of sea level rise thus far) and speed of Greenland and other ice melt.
This would not be an entirely correct impression. The IPCC report explicitly excludes the impact of sea levels on changes in the rate of flow of ice into the sea. (A discussion of this phenomenon as it applies to Greenland ice is contained in Carbon Commentary issue 1.)
The IPPC says the following in 2007:
There is better understanding than in the TAR that the risk of additional contributions to sea level rise from both the Greenland and possibly Antarctic ice sheets may be larger than projected by ice sheet models and could occur on century time scales.
The IPCC goes on to say that it does not therefore provide a ‘best estimate’ or ‘upper bound’ to sea level rise because of the lack of knowledge of what determines the rate of Greenland ice flow. As with future temperature rises, the UN body has indicated that as it improves modeling in one area, other uncertainties develop.
By contrast, the IPCC now appears more confident about rainfall level changes. In 2001, it said that ‘Globally averaged annual precipitation is projected to increase during the 21st century’ and indicated also that this increase will be accompanied by larger year-to-year variations.
Australia, Central America, and southern Africa were expected to show consistent decreases in winter rainfall.
By 2007, the IPCC is saying that it has more confidence in projections than in the TAR. In its precise language, it now says that it is very likely that precipitation will increase in high latitudes, whereas in 2001 this was just ‘likely’. There is also a very likely increase in heavy precipitation events. This is one of the few instances where the IPCC has felt able to express substantially more confidence in its forecast than in 2001.
In 2001, the synthesis report did not mention this effect of global warming. It is now perceived as a serious problem. Higher acidity will destroy corals, with the consequent effect on local ecosystems. A large percentage of the world’s population has a diet that includes fish species that live and feed in the protected waters of coral reefs. Higher acidity will also inhibit the formation of calcium carbonate shells on marine creatures. These shells sequester carbon by falling to the bottom of the ocean when the organism dies. So higher acidity means that oceans will become less effective carbon sinks.
The latest report expresses ‘medium confidence that approximately 20-30% of species assessed so far are likely to be at risk of extinction’ with more than 1.5 to 2.5 degrees of warming. With more than 3.5 degrees of temperature change, the IPCC’s model projections indicate a possible figure of 40-70%. The 2001 report mentions species extinction, but does not provide percentage estimates for varying temperature rises.
In the wake of the 2007 report, many commentators have fastened on the use of the words ‘abrupt’ and ‘irreversible’ to refer to possible catastrophic impacts of global warming
It may be interesting to note how the 2007 report varies from its predecessor in its use of language. Here is what was said in 2001:
Greenhouse gas forcing in the 21st century could set in motion large-scale, high impact, non-linear and potentially abrupt changes in physical and biological systems over the coming decades and millennia, with a wide range of associated likelihoods.
Some of the projected […] changes […] could be irreversible, but there is an incomplete understanding of some of the underlying processes.
(The use of italics is mine.) The 2007 document says:
Anthropogenic warming could lead to some impacts that are abrupt or irreversible, depending upon the rate and magnitude of the climate change.
Climate change is likely to lead to some irreversible impacts.
The 2007 report is slightly less tentative, particularly with regard to irreversibility, but the attention paid to the two words by the press and television may simply reflect the greater simplicity of the language in 2007 rather than any underlying change in the scientific assessment.
Six years of further study of climate change appears to have increased our knowledge of the ‘known knowns’ but decreased our comprehension of the ‘known unknowns’ and made us far more aware of the possibility of significant ‘unknown unknowns’.