bubbles_BAS-10006664_In blog image

Ice cores contain air bubbles from which we can measure the CO2 concentration of the atmosphere over 800,000 years. Courtesy British Antarctic Survey.

Eric Wolff is a Royal Society Research Professor at the Department of Earth Sciences, University of Cambridge. Eric is one of the foremost investigators of Quaternary and recent climate records using ice cores. He was scientific leader of the European project that produced 800,000 year records of climate and greenhouse gases from an Antarctic ice core and his other prominent contributions include his re-interpretation of sea salt in ice cores as a proxy for sea ice, and pioneering work on how impurities affect the  physical properties of ice.

Eric was the Royal Society Chair of a new publication from the Royal Society and US National Academy of Sciences on ‘Climate Change: Evidence and Causes’. Read the full document here.

We all want to know what’s going to happen in the future. But sometimes the discussion about the future of climate can seem a little theoretical. The problem is that, unless you have a time machine, you can’t observe the future. However, we can observe the past: this is the task of palaeoclimate scientists such as myself. I use ice cores (cylinders of ice drilled from the Antarctic and Greenland ice sheets). Those cores, and the other materials used by my colleagues, give us some solid facts on which to base our understanding of the situation we are in now.

The first thing to say is that humans are carrying out a unique experiment on the planet. Ice cores show that the amount of carbon dioxide (CO2) in the atmosphere is now over 30% higher than at any time from 800,000 years ago until the 20th century, and is still rising quickly. CO2 is a greenhouse gas, and greenhouse gases exert a strong control on climate. Indeed it’s almost impossible to explain major episodes in Earth’s history, such as the entry into and exit from snowball Earth, without invoking changes in the strength of the greenhouse effect. Similarly, the small changes in solar energy, due to Earth’s changing orbit, that occurred about 20,000 years ago could not have pulled us out of the last ice age without the help of rising greenhouse gas concentrations that amplified a modest regional signal into a large global one.

The past can also give us a feeling for the meaning of climate numbers. It is predicted that, if emissions of greenhouse gases are not curbed, the average temperature on Earth could rise by as much as 5°C from its preindustrial value by 2100. Maybe that doesn’t sound like a lot. But it turns out that the global average change estimated for the warming from the last ice age to the present is also about 5°C. That led to a wholesale change in the land surface and vegetation, and in the circulation of the ocean and atmosphere. If there had been human societies they would certainly have been disrupted.

While 5°C is at the top of the range of possible warmings at the end of the century, major consequences, such as significant sea level rise, are expected with smaller warmings. The past gives us a warning there. About 120,000 years ago, Earth was in an interglacial (the warm period between ice ages) similar to that of the last 10,000 years. It was actually slightly warmer than today for at least part of the time, particularly near the poles, probably because of a combination of a different orbit and a change in ocean circulation. Anyway, whatever the cause, it seems that sea level was at least 5 metres higher than today, implying substantial loss of either the Greenland or West Antarctic ice sheets, or both. This tells us that, if temperature in the polar regions is raised for thousands of years, we really should expect an irreversible rise in sea level, that people in the year 3000 may not thank us for (maps of how sea level may effect different regions can be found here).

Finally the past has something to say about precisely that question of the length of warming we can expect. Unfortunately while it is easy for us to put CO2 into the atmosphere rather fast by burning fossil fuels, it looks as if it takes a very long time for natural processes to take it all out again. The past provides us with one experiment where we can test this statement. About 55 million years ago, in what is known as the Palaeocene-Eocene Thermal Maximum (PETM), we find evidence that a huge amount of carbon was released into the atmosphere in at most a few thousand years. The exact cause is not known, but the amount released was comparable to that being released by humans in the 20th and 21st centuries. After the rapid release, the carbon did decay away but it took around 100,000 years to do so.

These are just a few examples of where the past gives us confidence that our understanding is good. Of course what we would really like is an analogue for the future. The PETM is the most obvious one, but unfortunately it’s too far back in time for us to have any detail about the effect that the large carbon release had on climate. So if we do want to know what will happen in the future, we have no choice but to use models of how Earth’s climate system works. However, it’s easy to think that they are theoretical and that somehow Earth will deal with whatever we throw at it. So it’s worth remembering that there are solid observations that show us that the climate system really does react, sometimes quite dramatically, to changing greenhouse gas concentrations.

  • David

    In your report you write “Evidence does not indicate pronounced long-term changes in the Sun’s output over the past century”. Can you provide any references for this? Thanks :)

    • TotalMassRetain

      See, for example, Lockwood, Proceedings or Royal Society Feb 2010. 10.1098/rspa.2009.0519Proc. R. Soc. A 8 February 2010 vol. 466no. 2114 303-329
      The IPCC AR5 report has more, plus references.

      • David

        Great! Thanks . I had a look at Lockwood and he confirms the increase in solar activity from the maunder min onwards, and of course the recent decline. Further he says (rspa.2007.1880) “All the solar parameters show significant change over the 20th century”. I think that’s a fair statement. Also, the decoupling of TSI and temperature is clear from 1980 onwards. However, I don’t think that the statement “Evidence does not indicate pronounced long-term changes in the Sun’s output over the past century” is correct, nor can I find any references that support this.

    • Eric Wolff

      David, Firstly it is worth looking at Figure 2 of our report, which show that there has been no upward trend in solar energy over the period for which we have the best data, and during which a significant temperature trend exists. For the earlier period (when the reconstructions of solar activity are less direct), the most recent compilation is in the IPCC AR5 report (downloadable from http://www.ipcc.ch/), see chapter 8, Figure 8.11. This will refer you on to the individual reconstructions. They tend to show only small variations apart from the 11 year cycle, and the net change from about 1900 to 2000 is close to zero. I guess we’d have to discuss how large a chnage has to be before it is “pronounced”. In my view changes that are small in relation to the 11 year cycle (which does not show up as a significant influence on global temperature) are not pronounced, and the net change over the last century is much smaller than that..

  • Colin Porter

    Professor Wolff, as scientific leader of the European project that produced the 800,000 year antarctic ice core record, you must have known from the outset that carbon dioxide concentration changes lagged temperature changes by an average of eight hundred years. So why did you not speak out when that disgracefully distorted publication, “An inconvenient truth” tried to imply that the reverse relationship predominated.

    • TotalMassRetain

      Are you suggesting that the reason your car emits CO2 is because of some past warming event rather than the combustion processes in your car’s engine?

      “Inconvenient Truth” was a polemic, not a peer reviewed scientific product. The point in there though, and repeated by Prof Wolff above is that once in the air this CO2 amplifies warming (whether that’s CO2 released by oceans or from the burning of fossil fuels in vehicles and power stations). That’s basic physics. Disagreeing with that is evidence you don’t think the climate obeys the laws of physics.

    • TotalMassRetain

      Whilst we’re on the subject of misleading or factually erroneous polemical documentaries, I presume you are equally, if not more, outraged by the many more and much greater lies and errors in Durkin’s “Great Global Warming Swindle”? If not why not?

    • Eric Wolff

      Colin, the period of major CO2 and temperature change for which the best age control exists is the transition from the last glacial maximum to the Holocene, 21000 to 10000 years ago. Both the CO2 and temperature increase took about 6000 years, and ran in parallel. This is exactly what we expect if this is a feedback where temperature change, through various mechanisms, causes CO2 to increase, and CO2 increase causes temperature to rise further (ie both relationships are true). I believe you are referring to the paper by Monnin et al (2001) who estimated that the temperature led by 800 +/- 600 years. As you can see, 800 years is a small proportion of the 6000 year period in which the increase occurred. Given the natural timescales in the transport of heat round the ocean, and of CO2 from the deep ocean, such a lag would not be surprising, and would not invalidate the contribution of CO2 to the warming over the majority of the transition. However, the latest work has refined this estimate. It has done this by two independent methods. Parrenin et al (2013) used a new method to determine the difference in age of the ice and the air at a given depth at Dome C, while Pedro et al (2012) used ice cores at sites with a high snowfall rate to reduce the uncertainty in the age
      difference. They both concluded that the lag is actually, within uncertainty, zero: ie the temperature and CO2 move together. I suggest reading the very
      readable Perspectives article by Brook (Brook EJ (2013), Leads and lags at the end of the last ice age, Science 339, 1042, doi:10.1126/science.1234239).

  • TotalMassRetain

    One common climate “myth” that keeps being raised is that of the interglacial “lag” exemplified by this question raised on the Google discussion forum:

    “Is it true that global warming has been happening over and over again, for tens of thousands of years, and higher CO2 concentrations in the atmosphere are an EFFECT, and not a CAUSE?”

    Prof Wolff confirms that the first part of the question is correct (100% of scientists agree that the earth’s climate has been constantly changing for billions of years due to natural factors). The second part references the “lag” which again Prof Wolff addresses this with CO2 amplifying the temperature increase.

    People do need to carefully think about the implications of this question, because if they are asserting that increased CO2 in the atmosphere is ONLY caused by increased temperatures, they are basically saying that the cause of CO2 being emitted by their car’s exhaust is due to some past warming process not due to the combustion processes in the car’s engine. That would be a ridiculous assertion.

    Having then established that there are many ways for CO2 to get into the atmosphere, but the dominant process driving the observed 40% rise since 1800 has been man’s fossil fuel emissions. Prof Wolff provides several lines of evidence to justify that conclusion. Once in the atmosphere that CO2 MUST cause warming according to the laws of physics. Claiming that it does not and has not is proclaiming one does not accept that the climate obeys the laws of physics.

  • TotalMassRetain

    I put some posts several hours ago that have yet to appear. What’s happening? Are the moderators asleep on the job? Why does it take so long?

  • Trevor Jones

    My real difficulty with climate models is that they appear to be intrinsically non-falsifiable, and I always understood that falsifiability was central to the scientific method. What evidence would lead you to reject the current models as being unsatisfactory, given that they all overestimate the cessation in warming in the last 17+ years – as we see from data in IPCC AR5?

    • Eric Wolff

      My understanding is that individual model runs do show slowdowns of the sort seen in the last decade or so. However, they do not predict (or claim to
      predict) when they will occur, and of course when averages of many runs are
      taken the slowdowns and accelerations disappear. However, you are right in the sense that “all models (of climate and everything else) are wrong, but some are less wrong than others.” The models cited in IPCC are simply the best representation (translated into computer code) of our current knowledge. I cannot think of any better way to assess the future than that, and we cannot by definition assess their performance in the future until the future occurs. However, they are tested against past climate change, and this can sometimes reveal missing processes that need to be added to the models to make them less wrong.

    • TotalMassRetain

      Climate models merely instantiate the known laws of physics. So, if faced with the situation that the models do not appear to satisfactorily fit the observations there could be several reasons:

      1) The laws of physics are wrong
      2) the climate models do not correctly instantiate the laws of physics
      3) there are physical processes going on that the models do not include, either correctly or at all

      1) would only be considered if there was independent, corroborative evidence. I am not aware of any
      2) is doubtless being investigated all the time and improvements made but no obvious issues

      so we are left with 3 as actually happens with most sciences if there is a discrepancy between theory and observation, especially in a very complex and in many respects poorly understood science.

      What isn’t valid is to assert “the models are wrong, therefore the alternative [manmade climate change is false] must be true” as actually the models do falsify that proposition. That would be like saying about Copernicus’ theory “it fails to predict the movement of the planets accurately, so Ptolemy must be right”. When all that was needed was for Kepler’s discovery that planetary orbits are ellipses to be included and it matches perfectly (to the precision of the day). However, to those still wedded to Aristotle’s “motion in perfect circles” around a single point of motion this was not acceptable until Galileo showed that this was invalid (via experiment and observation). The climate is, of course, much more complex that the motion of the planets.