Past, present and future temperatures


Check out these two graphs of past, present and future climate change:

They both provide a compelling visual picture of potential future warming in the context of past temperatures.

I wanted to use one of these graphs for a climate presentation recently, and there are pro’s and cons to each. Most importantly (to me):

– The bottom figure includes the “constant composition” scenario (C3 in yellow), which I find a useless distraction. It has no realistic value, as it implies an abrupt and arbitrary CO2 emission reduction, but continued aerosol emissions. (Aerosol particles have a short atmospheric residence time, so emissions have to be sustained in order to keep the concentration constant. OTOH, CO2 has a very long lifetime and with roughly 70% lower emissions its concentration would remain constant). It shows the amount of unrealized warming, but it most definitely is not an (even remotely) plausible scenario, and thus should not be presented as such.

– As for the proxy temperatures going back to the Middle Ages (or even further), I prefer the (mini) spaghetti graph of the lower figure over the top figure, which features only the Mann et al (2008) reconstruction.

Steve Easterbrook reminded me of my dilemma with his post comparing the top figure with one on which the bottom figure is based. He notes a few other differences:

– Different scenario’s are used in each (e.g. the top fig includes the “fossil intensive” A1FI (sort of business as usual) scenario, whereas the bottom fig doesn’t. A2 is worst case shown in the bottom graph, whereas it’s the middle of the road in the top graph.)

– Different temperature baseline (pre-industrial versus 1995-2004 period)

The top figure is from the Copenhagen Diagnosis; the bottom figure from Chapman and Davis (2010).

I chose to show the top graph, mainly because of the bothersome inclusion of the C3 “scenario” in the other one.

Tags: , , , , , ,

38 Responses to “Past, present and future temperatures”

  1. Dana Says:

    Considering that so many “skeptics” want us to remain on the business-as-usual A1F1 scenario, and that’s the one which our CO2 emissions are closest to thus far, it’s a useful one to show.

    Too bad the Copenhagen Diagnosis didn’t show more reconstructions.

  2. TimG Says:

    The A2 scenario is nonsense and should be ignored because it presume slow economic growth AND rapid expansion of population (peaks a 15 billion people vs 9 billion for the A1 scenarios).

    Basically A2 is impossible because if economic growth is slow then there will be no way to feed those people which reduces population growth the hard and nasty way.

    The cynic in me suspects they tried a low economic growth scenario and discovered the results weren’t scary enough – so they inflated the population projections instead.

    B1 and A1 scenarios are interesting because under the A1 growth projection poverty as we know it today will be gone by 2050-2060. By 2100 the average GDP per capita will be twice what Americans enjoy today. That pretty much eviscerates the argument that the ‘poor’ won’t be able to pay for adaptation.

    Ironically, the economic growth in the A1 scenario is 50% more than the B1 scenario which implies that the price we would pay for reducing emissions will be a huge increase in the amount of poverty worldwide. The lower B1 income would also make it more difficult to pay for adaptation measures which will exacerbate the harms. It is quite likely that risking more harm but being richer will result in less human suffering than reducing harm but being poorer since nothing we can do will reduce the hypothetical harm to zero.

    Of course, these are all economic projections and everyone here knows that economic projections are almost guaranteed to be wrong. What many don’t realize the scary IPCC results depend entirely on those economic projections and if those projections are off then the scary outcomes may never materialize even if we do nothing about emissions.

  3. Bart Says:


    Your final statement is at least partly untrue:

    Future temperature projections depend on two things:
    – Future climate forcing (dependent on our collective behavior as embodied in the emission scenario’s)
    – Climate response to that forcing (i.e. climate sensitivity and climate response time)

    The former is largely unpredictable, at least for long timeframes, and ultimately depends on choices we make (will we walk the red, yellow or green path?).
    The latter is to an extent predictable, though surrounded with considerable uncertainty (approximated by the lighter coloured bandwith around the darker coloured central estimate).

    Hence these scenario’s are presented in the form of “if (we chose this path) then (this will happen)” scenario’s.

    If (we do nothing about emissions), then (we’ll probably end up somewhere not too far from the red path in the upper graph). I prefer to go somewhere else instead.

  4. Marco Says:

    The cynic in me tells me TimG did not bother to read the background of the various scenarios, and thus comes with the claim that “The cynic in me suspects they tried a low economic growth scenario and discovered the results weren’t scary enough – so they inflated the population projections instead.”

    Here you go, TimG:
    and more specifically:

  5. Tom Fuller Says:

    Well, it seems as if we’ve spent the last year and three months arguing about the credibility of these projections without coming any closer to either an agreement or a solution.

    The controversy over the paleoclimatic record offers real evidence that the left hand side of those figures is nowhere near as flat and unchanging as the charts suggest. The controversy over Argo, UHI, siting of Chinese, Russian and Australian temperature measurement stations questions the rapid rise towards the end.

    And the fact that we do not know the sensitivity of the atmosphere to a doubling of concentrations of CO2 means that the projected rise going forward is a guess.

  6. TimG Says:


    I have read those pages a few times.

    There is absolutely nothing in this explanation that explains how slow economic growth and reduced global trade could support a population of 15 billion. I am saying the scenario is impossible for that reason. If globalization slows down or reverses then global population will decline the hard way.

    There is also no scenario that covers the ‘slow growth but stabilizing population scenario’. Why not? I can think of no reason to exclude that option as plausible outcome.

  7. TimG Says:


    If you were given a choice:

    A) Make 60K/year but face a 50% chance that your house would be destroyed and would need to be rebuilt.

    B) Make 40K/year face a 40% chance that your house would be destroyed and need to be rebuilt.

    Which would you pick? I think most would go with option A.

    You could play with percentages but the chance of disaster under Scenario A is not 100% nor does Scenario B ensure that disaster will not occur. But we do know that, by definition, people will be much wealthier if we go with Scenario A.

    It seems to me that Scenario A is the better choice if we want to maximize human welfare.

  8. toto Says:

    There is absolutely nothing in this explanation that explains how slow economic growth and reduced global trade could support a population of 15 billion.

    India supports a population four times as large as the US, and growing, with one-quarter of the GNP.

    A grim scenario for sure, but there is no physical impossibility there.

    Tom Fuller: are you working on your Gish Gallop skillz, or have you just decided to become a living advertising board for ? ;)

  9. TimG Says:


    India supports its population only because of the global trade flows that gives it access to technology and markets. If those trade flows dropped India would be in deep trouble. Specifically, India could never feed its people without large scale imports of oil and fertilizer and the export of goods used to pay for those imports.

    So yes. A 15 billion scenario in a world with reduced global trade flows and slow economic growth is an impossibility when you look at what needs to be done to support the populations that exist today.

  10. Marco Says:

    TimG: the question is WHERE the “slow growth” is located. A2 scenarios are *globally* slow-growth scenarios. That’s the first point. Second, the A2 scenarios also have food stability as the primary area of focus for technological advances, exactly because of continuing population growth.

    Moreover, B1 would be your “slow economic growth” and “low population growth” scenario. They have the same GDP by 2100.

  11. Bart Says:


    You’re welcome to increase the error bars of proxie based temperatures by a factor of two or three. The trend in thermometer measurements has no appreciable issues as has been shown ad nauseam (trend is insensitive to including e.g. good/bad sited stations according to Watts, light/dark stations, high/low population stations, low/high windspeed), and the uncertainty in the climate sensitivity is reflected in the lighter coloured margin of variability of future warming.

    You don’t have to know everything perfectly to still make a useful prediction.

  12. TimG Says:


    1) I know what the words in the SRES say. I am saying that they do not add up to a remotely plausible scenario. If a society generates the wealth required to achieve ‘food stability’ it will experience slow population growth. If population grows without economic growth it will not have ‘food stability’.

    2) If B1 is the slow economic growth/slow population scenario then you are agreeing that CO2 mitigation is not a requirement for that scenario. i.e. if the IPCC economic projections are wrong the CO2 catastrophe may never materialize.

  13. Eli Rabett Says:

    Since Eli can buy insurance for his burrow (and a mighty comfy place it is) for $2K/yr or so, he hosies the $60K.

  14. Marco Says:

    Correction: not B1, but B2 has slow economic growth and low population growth. B1 actually has a higher economic growth but a population of a mere 7 billion by 2100.

    And I don’t see why the B1 would indicate mitigation is not a requirement. Both B scenarios explicitely contain environmental protection (which includes CO2 mitigation).

  15. William Says:

    The controversy over the paleoclimatic record offers real evidence that the left hand side of those figures is nowhere near as flat and unchanging as the charts suggest. The controversy over Argo, UHI, siting of Chinese, Russian and Australian temperature measurement stations questions the rapid rise towards the end.

    The controversy over temperatures is all bluster and no substance. The satellite record, now that it has been debugged to a large extent, shows much the same temperature trend as the surface stations.

  16. TimG Says:


    When I asked why there is no ‘low growth – low pop’ scenario I was assuming BAU on the CO2 emissions front. If B1 requires CO2 emission reductions then my original question remains: why was this scenario omitted from the SRES? It is certainly more plausible than the low growth high pop A2 scenario.

  17. Scrooge Says:

    I never thought Eli lived in Florida but the next hurricane could bankrupt the state because of insurance like that. Sorry for off topic.

  18. Marco Says:

    TimG, it seems you really do not understand the scenarios. They are not scenarios which “require CO2 emission reductions”, but scenarios in which choices RESULT in CO2 emission reductions. And this scenario is not omitted from the SRES, it is explicitely part of the SRES (how else would I have known what the B1 and B2 scenarios are?).

  19. Hans Says:

    Good to see the hockeystick is still unbroken on this blog.
    I wonder why the Little Ice Age has dissapeared.

    Some people predict that we will colonize the moon before 2050.
    Others predict that the North pole will be icefree within 10 years from now.

    Many more worry about the price of food and whether they will earn enough to feed their children.
    Good luck with your presentation.

  20. TimG Says:


    You are playing word games. The B scenarios are sold as the ones where the world buys into the CO2 mitigation meme. The A scenarios are sold as the ones where nothing special is done about CO2. The B scenarios may not specifically mention CO2 controls but that is the intent. Just like the A scenarios (specifically A1F1) are not called BAU but that is the intent.

    That said, what I am looking for is a BAU with lower growth projections. If that is what you think B1 is then fine. I withdraw my comment about there being no low-growth low-pop. But this also confirms my previous point that the AGW catastrophe meme depends entirely on extremely dubious 100 year economic projections. IOW – we could do nothing and no catastrophe will appear. This is what happened with the ‘population bomb’ scare in the 70s – BAU technological and economic progress meant the hypthetical problem never materialized.

  21. Sailrick Says:


    Your comment about choosing between making $50k a year in one scenario or $60k in another is an absurd oversimplification of the problems the world faces from climate change. How about no water to drink, no food to buy with your $60k, millions of displaced people, rising sea levels threatening shoreline infrastructure, cities and agriculture. What is most absurd about it is that you are only concerned about how much you will earn in your lifetime, or someone elses hypothecal lifetime, if you are an older person already. Do you think global warming will just come to an end in one lifetime? How many times will adaptation have to occur over the next 200-400 years, if no mitigation is done now? I find your observations to be shallowly selfish.

  22. TimG Says:


    Climate change presents no challenges that not been faced before by humans. They can and will be overcome, however, those challenges will be a lot easier to overcome if people are wealthier.

    More importantly, numerous economic studies have attempted to quantify the cost of dealing with these issues. The number varies depending on assumptions but we are talking about no more than 5% of world GDP 100 years from now. 5% is a price I think is worth paying if that means people will be 50% wealthier in 2100.

    Also. do you really argue that we have been better off today if people in 1611 decided to forgo economic development in order to limit carbon emissions? Why is the argument any less absurd when we are talking about sacrifices today for people in 2411?

    Lastly, you really need to separate the abstract concept of mitigation with the practical reality. In abstract it makes sense to talk about mitigation. In practice, the options are not there or they cost too much. This means that money spent on mitigation will be money wasted. If money is available it would be much better if it was invested in adaptation rather than mitigation.

  23. Marco Says:

    TimG, if you’re so eager to distort the background of the scenarios, go ahead.

    I do have a quibble about your “adaptation” investment: what are we supposed to adapt to? The answer is “extreme weather”, but we don’t know what and WHERE that extreme weather is going to be. Northeast Australia would likely have to adapt to long periods of drought, and when there is rain, it’s extreme rainfall. How d’ya adapt to that, mate?

  24. Sou Says:

    Bart, have you thought about creating a thread where TimG can post to their hearts content? A boor hole like RC would do it.

  25. Mel Says:

    TimG, I run a small business, and from our perspective, the cost of mitigation is a serious concern. Over a 2-year period, we introduced energy efficiency measures and were able to reduce our natural gas consumption by 67%. The shift to air-source heat pumps resulted in an increase in hydro costs of about 30%. Our overall saving in energy costs came out to about 25% and this is a big number in our bottom line. Return on investment will be less than 9 years. Three local companies made money by selling and installing equipment that saves us money. The increases in energy efficiency reduced our CO2 emissions by 10 -12 tonnes per year. This is not an “abstract concept”. Nor did not cost too much – it is saving us money. The “practical reality” is that money we spent on mitigation was not “wasted”, it was used to increase our energy efficiency which is smart. To argue that we should not have done this and instead spent the money on adaptation strategies would be – from a business perspective – stupid.

    From our experience, the business case for increasing energy efficiency and reducing CO2 emissions is clear. If mitigation is beneficial on the scale of an individual business, why would it not be beneficial for the larger economy?

  26. Tom Fuller Says:

    Bart, your chart is visually compelling. Does it correspond to reality as far as the past, present or future?

    Is the period covered long enough to adequately show the range of variation experienced?

    Does it adequately capture the range of uncertainty in both measurements and projections? There seem to be no margins for error for the instrument record.

    Do you as a professional scientist working in the field actually believe that projections of a 7C rise by 2100 are realistic or plausible enough to warrant inclusion? I personally believe that’s Day After Tomorrow hysteria. Educate me.

  27. TimG Says:


    1) You traded natural gas consumption for electricity consumption. If your electricity comes from coal then you may not have reduced CO2 production by even one gram. You may have actually increased your emissions.

    2) The business case for better efficiency exists independently of CO2 emissions. If such savings were so easily obtainable then the question is why didn’t you invest the money earlier?

    If your answer is ‘subsidies’ then you will need to include a portion of your taxes in your calculation of the real cost of the upgrades before you can make any claim about whether the investment was economic.

    3) What percentage of your total emissions did you eliminate? 10%? 20%%? How much would it cost to reduce your emissions by 50% or 80%? I think you will find that the incremental cost of additional emission reductions increases exponentially once you have dealt with the easy stuff.

    That said, if there is money to be saved there is nothing stopping business owners from making the investments today. We should not need an anti-CO2 policy to convince people to do that. IOW, the fact that you found a way to save money with energy efficiency is not a justification for anti-CO2 policies. It is, at most, a justification for a program to educate business owners about the potential for energy savings.

  28. Mel Says:

    TimG, Apologies in advance for a cryptic response. I’m not trying to be snarky – I’m doing this between clients.

    1) Our business is in BC, Canada – there are no coal-fired power installations in the Province. 14% of the up-front costs were covered by incentives and a good chunk of this came from the furnace company. On the other hand, the BC Provincial government is subsidizing the oil and gas sector in a hell of a big way with our tax dollars.

    2) We have owned the business since 2004 so this is one of our early big investments. In retrospect, we should have done it earlier. In my mind, energy efficiency and CO2 are linked. I presented our results last night to the Chamber of Commerce along with several other businesses and all demonstrated that reductions in CO2 emissions saved money. We all could have said that improvements in efficiency saved money. Same thing. As I understand it, most people who are arguing for CO2 reductions are suggesting a big part of the solution should be improvements in efficiency.

    3) Our natural gas consumption represented the vast majority of our GHG emissions – transportation for our business is minimal.

    Lastly, we did it because it made economic sense on the scale of our business AND because we have good reason to believe that advertising our reductions in CO2 emissions will attract more clients. Most Canadians accept the findings of the scientific community (national academies of science, AGU, AMS, GSA, AAAS, AMS, APS, ACS, etc.) that indicate we need to reduce global CO2 emissions.

  29. TimG Says:


    Thanks for the clarifications.

    First, the BC government collects gasoline taxes, resource royalties and now a carbon tax. The subsidies exist in the form tax credits for resource exploration but these subsidies are paid back with royalties. It is not reasonable to claim that fossil fuels are subsidized in BC. The subsidies that you got for your capital investments have to come from somewhere and must be included when you calculate the ROI for your investment. In short, all the subsidies do is increase your tax bill.

    Second, you appear to be extremely happy that you could something good for the environment while saving money. The trouble is that is only true because the government did not force you to meet specific targets. You were free to identify potential savings and act only on those that would also save you money. I am sure your tune would change if the government came back and insisted that you must reduce your emissions to by 50% or even 80% (the current BC target). You would likely find that meeting the higher targets would be rediculously expensive or technically impossible even with your limited CO2 emissions. I am sure you would also change your tune if you happened to run a business where even a modest 15% reduction in CO2 emissions is too expensive to contemplate.

    The harm caused by unrealistic targets and timetables is why CO2 regulations are bad for the economy. Short term success stories like yours mean nothing as long as these targets are in the legislation. That is also why I say it wrong to justify CO2 reductions because of side effects like saving money on energy. Saving money on energy is something that can be justified on its own and there is no reason to risk the harm caused by heavy handed CO2 regulations.

    Lastly, excess hydro power in BC is sold to the US where it does displace coal fired emissions. If residents of BC consume more power that means there is less for Americans. This, in turn, results in an increase in CO2 emissions even though you think you are getting power from a CO2 free source. These kinds of secondary effects show up all of the time and it one of the reasons why I think emissions accounting games are waste of time and money.

    Finally, Canadian support for CO2 reductions is a mile wide and an inch deep. They only support it if it costs them nothing. How much business would you attract if you had to add a 10-20% surcharge to your prices to cover the cost of reducing CO2?

  30. Bart Says:


    Is the period covered long enough to adequately show the range of variation experienced?

    As to the proxy covered period, I don’t know. As I wrote in thje post, I would have preferred a spaghetti plot for the pre industrial period, as in that case it wouldn’t depend on one (out of many) reconstructions. Failing that, I told the audience that the error bars may be larger than depicted here, e.g. the MWP may be a bit warmer according to some. But that’s not the focus of this graph. The reason I was initially hesistant to show this graph is that I’m wholly uninterested in a hocky stick discussion overshadowing the big picture which this graph is about.

    For the instrumental period, make the black line twice (recent) or three times (pre-1950) as thick and I think that covers it (kind of).

    For the projections, I think the variation within one scenario (light coloured bandwidth) provides a realistic picture of the likely outcome based on today’s knowledge of the climate system (climate sensitivity). By looking at a collection of scenario’s (the different colours) I think the landscape of potential future paths is quite well covered. I haven’t looked much into the individual scenario’s, so I don’t really have an opinion about which one is more likely. Ultimately they represent a choice. Whether the upper red one (A1FI) is implausible I don’t know. I sure hope so. But don’t underestimate the human potential to do both very smart and very stupid things.

  31. Bart Says:

    TimG, Mel,

    Interesting discussion but veering far off topic. Let me know if you want to continue discussing efficiency/CO2 reduction measures, in which case I’ll open an open thread for that purpose. Let’s keep this thread for discussions of past, present and future temperatures. Thanks!

  32. matt Says:

    Both graphs are quite frightening, but neither of them show the full range of outcomes for varying climate sensitivities. The ippc source shows a range for each scenario: The top graph in particular overlaps the spread indications so only the worst cases are shown.

    The estimates for climate sensitivities are very much determined by the variation in sulphate forcing history assumed for the models (or for empirical determinations, which have the same problem.) See for example, where very different sulphate forcings are included for each model. The truth is, we don’t really know how good our estimates of climate sensitivity are at the moment.

    GLORY is due for launch on the 23rd, and with a bit of luck the GLORY and ARGO data will narrow those outcome ranges down significantly within a few years. Then we’ll be in a better decision to judge what we need to do. In the meantime, everything that would help that we SHOULD BE DOING ANYWAY, climate change or not, should be pushed hard. “No regret” steps like those outlined by Mel earlier need to become commonplace.

  33. Bart Says:


    The graph you link to is the same as the right hand part of the lower graph in my post: The scenario’s have the same spread in expected future temperature change. The other graph in my post has the same spread in the scenario’s. So I don’t see how you can support your claim that they don’t show the full range of outcomes for varying sensitivities, whereas presumably you claim that the graph you linked to does? They both show several scenario’s, and they are by definition “if (we do this), then (this may happen)” style scenario’s, and since they show the model spread of each of these, they’re not “worst case” only.

  34. matt Says:

    Bart, maybe I’m completely misunderstanding the IPCC graph I linked to, but aren’t the grey bars to the right of the 2100 AD vertical showing the upper and lower limits of the scenario projections, with different climate sensitivities? The same graph and bars is shown here, with a bit more explanation:
    So for example, your top graph for the worst case A1F1 scenario appears to show a trajectory ending at between 5 and 6.5 degrees positive anomaly by 2100, whereas the bars actually show a range of 2.2 to 6.2 with a best estimate of 4. Not that that is good news of course, but your top graph at least appears to predict the high ends of the range rather than the “best estimates”.

  35. Bart Says:


    Good point, there are indeed different estimates of uncertainty. From SPM Fig 5 (i.e. the last link you provided):

    Solid lines are multi-model global averages of surface warming (…). Shading denotes the ±1 standard deviation range of individual model annual averages. (…) The grey bars at right indicate the best estimate (solid line within each bar) and the likely range assessed for the six SRES marker scenarios. The assessment of the best estimate and likely ranges in the grey bars includes the AOGCMs in the left part of the figure, as well as results from a hierarchy of independent models and observational constraints. {Figures 10.4 and 10.29}

    Fig 10.4 is similar but extended to 2300 for 2 scenario’s (shown e.g. by Steve Easterbrook in his linked post). Fig 10.29 provides other estimates and grey error bars, which are the

    AOGCM mean plus 60%, minus 40%

    which pretty much encapsulate the range of values found by different means.

    Note also that the reference point for the top graph in my post is pre-industrial, i.e. almost 1 deg lower than the base period for the IPCC graph. It is thus not the case that the envelope of most likely values is shifted upwards in this figure.

  36. matt Says:

    Ah, I missed the shifted base period! I do still contend that those “likely ranges” are distressingly broad and the consequences and action required are rather diffferent for the top and bottom extremes. I’m optimistic that the GLORY mission will do a lot to clarify matters, and IPCC5 will demonstrate far more progress on IPCC4 than IPCC4 did on the TAR. Hopefully those ranges will narrow onto values near the bottom extremes rather than the opposite.

    That said, even without the proxy reconstructions or the model projections, just eyeballing the uptick on the direct observations is quite sobering.

  37. Biodiversity, extinction and climate change « My view on climate change Says:

    […] much to add. It’s a very important aspect though, especially when taking into account that projected future (rates of) warming are much stronger than what we’ve seen so far. Not to mention that once a species is lost, it’s lost practically forever. I would think […]

  38. A Tale of Two Graphs | Serendipity Says:

    […] Feb 3 2011: Bart does a similar comparison with a third graph, which does a better job of the pre-industrial reconstruction, but still suffers from all the other […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: