The significance of past climate change
Posted on 21 April 2010 by John Cook
A common skeptic argument is that climate has changed naturally in the past therefore humans aren't causing global warming now. Interestingly, the peer-reviewed research into past climate change comes to the opposite conclusion. When I try to explain why to people, I usually get blank, confused stares. I gave a presentation to a roomful of engineers this week and after explaining the significance of past climate change, complete with slides of climate sensitivity PDFs and examples of positive feedbacks, the result was a long, silent pause. I asked if anyone understood what I'd just talked about. A few asked some follow-up questions which made it clear they didn't. So I'm reworking my whole explanation of past climate change in an attempt to make it as clear and simple as possible. Comments, particularly on anything confusing or unclear, are welcome!
In the past, climate has changed, sometimes very dramatically. This has gone on long before SUVs and coal fired power plants. If climate can change on its own, couldn't current global warming be natural as well? To answer this, first you have to ask why climate has changed in the past. It doesn't happen by magic. Climate changes when it’s forced to change. When our planet suffers an energy imbalance and gains or loses heat, global temperature changes.
This can happen in a number of ways. When the sun gets brighter, the planet receives more energy and warms. When volcanoes erupt, all the particles suspended in the atmosphere reflect sunlight and the planet cools. These effects are referred to as external forcings because by changing the planet's energy balance, they force climate to change.
Looking at the past gives us insight into how our climate responds to external forcings. Using ice cores, we can work out past temperature change, the level of solar activity plus the amount of greenhouse gases and volcanic dust in the atmosphere. From this, we can determine how temperature has changed due to past energy imbalances. What we have found, looking at many different periods in Earth's history, is that when the Earth gains heat, positive feedbacks amplify the warming. This is why we've experienced such dramatic changes in temperature in the past. Our climate is highly sensitive to changes in heat.
What does that mean for today? Rising CO2 levels are an external forcing. They're causing an energy imbalance and the planet is building up heat. From Earth's history, we know that positive feedbacks will amplify the CO2 warming. So past climate change doesn't tell us that humans can't influence climate. On the contrary, the past tells us that climate is highly sensitive to the CO2 warming we're now causing.

Arguments



























- The role of greenhouse gases in the past climate (or where they come to the atmosphere) might need an explanation.
- You should explain what positive feedback means and the factors which are positive feedbacks.
- The part about CO2 being an external forcing might call for additional explanation - I don't think it's readily clear for someone not familiar with the issue.
Which bit of 'this' are you referring to? And why do you use a word like 'believe'?
In the past . . .
What happened as Surface Temps rose, due to CO2 on occasion, with regards to the rates of convection and evaporation?
Did they increase, decrease, or stay the same?
If convection and evaporation rates increased as surface temps increased, did that cool the surface, heat the surface, or did the surface temps remain the same?
Thanks
The basic point was to say that there are a number of feedbacks - both positive and negative. Finding the net feedback by adding up all the individual feedbacks is a complex job. But you can cut through all that in one fell swoop by looking at past change. By just comparing temperature change to changes in the energy balance, you determine the net feedback without having to know all the individual feedbacks.
So I skipped all those details in this blog post in the effort of a simpler argument. But going through the explanation too fast leaves people wanting more details. Where to draw the line?
The past geological record indicates that changes on a global scale are invariably very slow. We're talking tens of thousands, to hundreds of thousands and even millions of years in the vast majority of cases. For something like acidification of global oceans by volcanoes, and indeed the majority of mass extinction events (except in rare cases of eg bolide impacts), that is the sort of time frame the geological record indicates is invariably required.
To take a few examples, the output of greenhouse and other gases by Siberian Traps volcanism at the end of the Permian occurred on a major scale over several hundred thousand to several million years, and that is how long it took to cause major climate changes such as (possibly) acidifying the oceans, the collapse of coral reef ecosystems, and mass extinction. It was not a ‘rapid process’ when compared to the scale of human lifetimes. The break-up of the Gondwana supercontinent is likely implicated in eg the Mid-End Triassic mass extinction-continents do not break up 'rapidly'. This occurred over millions of years - with possible stress-related tipping points etc, as rift-related volcanism increased over very long time periods. Many other examples from the geological record indicate much the same thing, (eg oceans don't ‘acidify’ within short human timescales when similar amounts of c02 have been added to the atmosphere in the sort of time scales involved as is currently the case).
This is also the major reason people such as the gradualist Charles Darwin were so skeptical of the presence of 'mass extinction' events in the geological record in the first place, and also I suspect why the person on the street is with current climate change and the extreme predictions around it as well. It's like worrying about continental drift changing the climate within this century.
This is one of the major skeptical arguments, that major global changes such as those predicted by the IPCC to occur within the next century by human emissions of greenhouse gases will NOT occur, not so much because the concepts and theory is wrong, but because the scale of time involved is far too short, and that the geological record provides very good support for this contention. Academics and other pro AGW advocates get the concepts largely right, but get the time scales involved largely wrong.
On the other hand the current rapid Climate Change -- when humans are now one of our planet's larger animal species -- is a highly risky for both us and most of the other living species; though as an extinction event it will probably be no worse than many in the past. Arguably it's made worse by the unprecedented fact that humans are responsible for creating the problem this time.
It would be interesting to conjecture (is it possible to know for sure?) the extinctions and changes to the nature of life on the planet as a result of past climate changes, and then extrapolate how the current climate change being instigated by humans could alter the nature of life on the planet today.
Of course one of the problems in doing this is that global heating is only one of many serious issues now being caused by human influences on life's existence on our planet. Pollution; deforestation; industrial agriculture; fresh water diversion; resource depletion; over-population -- to name just some of the most obvious -- are all key components in the environmental timebomb we're creating.
While a number of false arguments try to use paleoclimate to deny AGW, I must admit I have never been convinced that the paleoclimate record is terribly useful to the debate. Fundamentally, there is nothing it gives us that is not more clearly communicated by contemporary data.
For example, the "Hockey Stick" shows that the contemporary rate of warming diverges enough from the last few thousand years to raise a smoking gun type of question, but the graph itself is not an explanation -only a way of flagging an anomaly.
Even if the hockey stick was proven wrong, it wouldn't matter - because arguments about medieval warming, past CO2 levels, higher ancient temperatures etc. all miss the point:
In the contemporary world, we can identify the rate of global warming, the mechanisms of warming and the predicted (and confirmed) effects.
Until there is a better explanation for the colossal body of evidence of the last 100 years, the opponents of the AGW thesis will try and drag our attention into the distant past, where the ground under any given argument is weaker.
I get that argument sometimes and my response is that it wouldn't matter if there was either a more rapid or larger warming at some point in the distant past - there weren't billions of people living near the sea then. "The Planet" is not endangered - it is a small percentage of the planet's flora and fauna, plus a big percentage of its human beings. Of course, it's like asking for anesthetic at the dentist and being told you shouldn't have it because the dentist's grandfather had to suffer more than you!
1: 'Climate has changed in the past so how do we know humans are causing it now?'
2: 'How do we know that the climate change we are causing is going to be significant?'
I think you can really handle the human causation question using CO2 alone (past correlations of CO2 and temperature, satellite and ground measurements confirming energy imbalance in CO2 absorption spectra, various proofs of increased CO2 levels being human caused, et cetera)... possibly following that up with explanation of why it ISN'T the Sun, cosmic rays, volcanoes, martian death rays, or whatever. All of which can be handled by showing lack of correlation - some match past changes, but none match the current.
Once you have established CO2 as a major 'control knob' in determining temperatures then you can get into feedbacks amplifying that effect and what the historical record tells us about those.
Basically, handle one issue at a time. Don't start talking about positive feedbacks before they've bought into CO2 increasing temperatures at all.
Rather than list all the possible primary drivers (forcings), start by describing a couple of periods when the climate changed in the past from different forcings. For example, Milankovitch and volcanoes as the initial forcing.
For each of the two periods in turn, describe in simple terms what the earth was like before the change (eg ice free Antarctica) and what it was like afterwards (eg lots of ice everywhere). Describe the forcing that led to the change (and define the terms - engineers most probably use slightly different jargon from climate scientists).
Then move to today and show the evidence that the climate is changing (temperature, ice, oceans, sea levels etc). And explain that the forcings that caused the prior changes you discussed are absent, and the only primary driver or initial forcing operating today is CO2 / greenhouse gases.
Then show how greenhouse gases also acted as a positive feedback, amplifying the changes you discussed in the two first examples (warming or cooling).
You can deal with the 'but couldn't it be due to ...' in the question time / discussion - keep your detailed slides on hand for when specific questions are asked.
If there are particular points you want to elaborate on (eg role of water vapour/clouds which some often ask about), you can probably do this during the discussion by saying, 'someone often asks me xyz' and then whip out your extra slides :).
(As a broad (over?) generalisation, engineers are often clever, but their thought processes tend to be linear rather than lateral. Simple works best. One step after another, with a bit of repetition and reiteration of key points along the way. No offense meant to engineers, it's just an observation from working with groups of engineers in a variety of contexts.)
The following publication is rather good, and will no doubt be more useful than my off the top of the head idea: CRED guide
Try that line of argument in a court of law against a arsonist, by saying that forest fires have always happened naturally; it won’t fly.
Indeed, GHG had an important role to play in many past climate changes (even though their concentration changed without human involvement). Looking at the past actually strenghtens the evidence for a climatic effect of GHG.
Climate change history is like precedent in law - you have a good chance that a previous finding or two might establish where things stand 'when deciding subsequent cases with similar issues or facts'.
Yes; the climate has always changed, man-made climate change is just a new precedent. Looking at 'precedents' or past forcings helps us with what might come next.
All indications are that climate hasn't changed significantly for the best part of 10,000 years, so for it to suddenly change in the last 150 years just at the same time as mans influence on the biosphere has become significant, strongly implies this is the most likely cause. This is without any knowledge of greenhouse gases or other causes, it is a pure statistical explanation. The other more direct experimental evidence such as radiation exchange simply increases this probability further.
As you can clearly see, temperature is much more variable when it is cold. In other words: climate sensitivity diminishes with increasing temperatures.
(1) The Dansgaard-Oeschger events in your graphics are measurements of rapid temperature changes in Greenland cores. They are barely recognizable (and asynchronous) in Antarctic cores. Thus they are not representative of global warming/cooling events.
(2) In fact these events are recognised likely to arise from major rapid shut down and restarting of the major ocean circulations (thermohaline circulation) that strongly participiates in bringing heat (or "thermal energy" if we're stil being pedantic about that!) from the equator to the high Northern latitudes.
In other words they are indicative of major redistribution of Earth heat as opposed to global scale warming or cooling events.
(3) Thus they have nothing to do with "climate sensitivity" which is the equilibrium response of the Earth's global temperature to changes in radative forcing.
(4) The fact that D/O events (and similar large scale jumps/drops in high Northern latitude temperatures) aren't apparent in the Holocene part of the record you posted, is that these events only occur under conditions that major ice sheets occupy the high N. latitudes (that's part of the likely mechanism of the D/O events). Since we're in an interglacial period when these major ice sheets have melted away, we can't have D/O events.
It's got nothing to do with temperature "being much more variable when it's cold" or "climate sensitivity diminishing with increasing temperature".
One should really address these issue in terms of what we know!
Also, is it really possible to derive solar activity from ice cores? I’m sure there are other proxies (carbon-14 comes to mind), but ice cores?
I would add in your response to Péter that this interglacial is remarkably stable when compared to the previous ones. So his assertion "temperature diminishes with temperature" has little supporting evidence.
Besides, some events like PETM support a self-reinforced warming hypothesis.
I agree with you. Paleoclimatology is just one of the many lines of evidence to climate sensitivity. And if you pick one specific period -like the Medieval Warm Period- it's just one tiny fraction of this.
So even if there had been a strong MWM... not much would change in the science.
And I don't know if this goes beyond your intent, but it might be worth mentioning the rate of change now compared to past warming events.
What you already have, though, is pretty clear I think. I'm not a scientist or an engineer, but it makes sense to me. (I am a fairly well-read layman, though)
Do you have a link to this schematic? I am an engineer and would be pleased review it as well. I find your answer lacking since it has been dumbed down. It is, with all due respect, meaningless in 'simplified' form.
What I am especially interested in is the detail of the positive feedbacks and whether you have satisfied the negative feedbacks sufficiently.
Regards
Unfortunately, that means you'll need a few different presentations instead of just one, and you should ask the group that's sponsoring your presentation what the audience is likely to be before (and maybe ask the audience too and tailor your presentation accordingly).
what was the reason for the MWP? there seems to be a new paper on this topic that points to this happening in the Indo-Pacific so it appears this event wasn't localized to Greenland. were we in a solar optimum at that point? i've heard the MWP debunked as a localized event but there are other papers such as this that show it happened other places on the Earth at the same time.
I've also seen graphs of vostok ice cores showing we are due for another glacial period and i'm surprised i've never seen comments related to the fact that maybe we humans are counteracting that normal temperature decline wiht our GHG production. the up shot being, even if you subsribe to the AGW theory, it might not be a bad thing for the next 1000 years or so since plus 1 to 2 degrees C is easier to adapt to than negative 8 to 10 degrees C.
forgive me if this has been already discussed ad nauseaum.
Another aspect, I have found difficult for engineers and many other scientists to grasp is: How a system where all feedbacks sum up to a net "positive" does not constitute a run-away system. This is not and easy thing to explain and typically requires too much time in a short talk. But may constitute a stumbling block for engineers in particular.
You may already be familiar with this presentation in June 2008 to the Tällberg Forum by David Wasdell: Planet Earth - We Have a Problem
In case not, you might want to review it - he does an excellent job on feedbacks, and you may find some ideas therein.
I find this an excellent presentation, not only for it's content, but also for the passion with which it is delivered.
Thanks for all your excellent work!
you should always refer to the original scientific papers. The link you provide on the MWP has been "adapted" to hide the rise and erroneously draw to the conclusion that "Medieval Warm Period was about 0.4°C warmer than the Current Warm Period.". Nothing similar can be found in the original fig. 2b. On the contrary, it explicitly shows that the average 1997-2007 SST is higher than any other period in the last 2000+ years.
I have, and not just from the skeptical optimists. ;-) The next ice age isn't due for tens of thousands of years. The earth has cooled by about half a degree C since the end of the climb out of the last glaciation 10k years ago. We've countered that and more within a century. Our immediate concern is the next hundred years or so.
Here is the issue I suggested earlier regarding rate of change. We can adapt to long-term slow changes much more comfortably than relatively fast changes in the near-term (relatively speaking). Perhaps in 10 000 years, if we haven't warred ourselves to extinction or succumbed to a devastating plague, we may be knowledgeable enough to fashion some kind of thermostat for the planet that doesn't interfere adversely with long-term, possibly necessary climate changes. We're not yet wise enough to deploy any form of geo-engineering.
i've heard the MWP debunked as a localized event
The language is too strong. We think the Earth was generally warm, but datasets all over the world show 'medieval' warmth at different times, as much as 500 years apart. And most of the data we have is from the Northern Hemisphere. The MWP may or may not have been a global event, but it would seem there is some evidence for that. Whether or not the warmth for some sequent decades in the past was comparable to the last few decades is the qualified assessment most discussed (probably not).
Here's a map of data sets often deployed by skeptics. Check the warm dates for each of the time series. Ironically, skeptics don't realize that they're buttressing the 'not global' argument when they reference this - they don't investigate much further than the message.
http://pages.science-skeptical.de/MWP/MedievalWarmPeriod.html
The map, by the way, documents a small number of paleo data sets (47). There are now hundreds. No doubt these have been selected to buttress the message. Ironic then...
See this thread or this paper
#32 Riccardo, that figure on CO2 science appears to match the figure 2b in the paper. the graph in 2b does show temperatures during the MWP that are above modern temperatures. here is the graph from the Nature website. what rise are you talking about and why would you call the graph on the CO2 Science website a lie? i didn't read the article on the Nature website but instead just went to the graph that you referenced.
thanks all (barry at #33 too) for the quick feedback - just another reason why this is the place to come when I have questions.
in the original fig.2b the line labeled "1997-2007 mean annual SST" is higher than any other line for the whole period shown. So, when co2science says that following the paper it was 0.4 degree warmer, it is blatantly false.
You may be right that this is not a lie, they possibly can't read a graph.
thingadonta- rates. The "Heinrich" events as we moved out of last ice age look to have dramatic temperature swings in the scale of decades. However, we have no reason to believe that such swings are feasible in an interglacial from natural causes.
You could point out that over a very short period of time mankind has removed and burned an enormous quantity of fossil fuels that had been comfortably buried for, roughly, the past 500 million years of earth history. I think if you can graphically show an estimate of how much fossil fuel by weight has been burned since the industrial revolution and compare that with the rise in atmospheric CO2 concentration over the same period, it might convince some. Past climate change is used to help us understand what changes we will likely see for the future. When climate has changed in the past scientists have struggled to understand the cause. This time the cause is obvious.
One more question. I did not know that ice cores could reveal solar activity. What in the frozen ice gives us that information?
C14 and Be10 production are proxies for solar activity. I believe Be10 is used in ice core.
I would suggest not using words like "highly". That word has little scientific value.
Thanks for the help on that!
He has repeated classic denialist rubbish in the past while putting the opposite spin on it - like that climate models don't include the role of water vapour (utter nonsense). He also frequently says misleading things about his CV to suggest that he is a climate scientist when in fact he is a psychotherapist.
After he accused the IPCC of political corruption in the New Scientist (or rather Fred Pearce did on his behalf, quoting him) a letter "From the co-ordinating lead authors of Working Group 1 of the IPCC Fourth Assessment Report" signed by 20 major climate organisations was posted in protest, pointing out major falsehoods in Wasdell's claims.
It can be viewed here: http://www.newscientist.com/article/mg19325960.900-climate-with-care.html
It seems fairly straightforward to me but I supposed I've been following the science for a year or so. As you say it depends on audience and what you can assume about them.
One thing that jumped out is that you switch between energy/heat/temperature alot. You could think about just talking about one of them. I think the clearest is talking about energy building up in the system. I don't know whether by sticking to one term it makes the ideas flow better. With one line to at the start or end to say that energy build up = rising temperatures.
Two things:
From eyeballing the adapted "CO2 science" graph (with my ruler on the monitor I get 0.27 deg warmer in the MWP than today based on matching the red peak to the blue peak so I don't know how they got that.
But when I quickly read over the journal paper as far as I can tell it was incorrect method anyway to interpret figure 2b or at least to compare it to modern times as they author's state a few things:
...reconstruction suggests that at least during the Medieval Warm Period, and possibly the preceding 1,000 years, Indonesian SSTs were
similar to modern SSTs....Contrary to the Indonesia SST reconstruction, however, the Northern Hemisphere temperature reconstruction does not estimate temperatures as warm as modern at any time during the past two millennia.
The author's then continues to say: We note that the high-amplitude variations resulting fromthese hypothesized changes in G. ruber seasonality also preclude accurate estimates of the rates of SST change in the past and a meaningful comparison to the rate of SST increase during the past
decade.
Reading this, makes me instantly skeptical of the claim at "CO2 Science" that the Medieval Warm Period was warmer than the Current Warm Period. I infer that they intentional (not just that they can't read graphs) with their lack of transparency over how they interpret the science.
This would APPEAR to overrule the presentation's observations that 1) climate changes took place in the past due to forcings and 2) CO2 is a forcing.
What is the right approach to this? My understanding (I would certainly welcome Cook's corrections) is that the degree to which the climate is sensitive to forcings has itself also been highly variable. Specifically, that with the Late Triassic configuration of continents and corresponding ocean currents, even a rather large CO2 forcing had a surprisingly small effect; but with modern ocean currents, it is well-expected to have a MUCH larger effect.
So if that understanding is correct, then it certainly should be included in the presentation for engineers.
On the contrary, climate sensitivity has been surprisingly consistent in the past, even going back millions of years. So if someone asked about CO2 being much higher in the past (noone did bring that up at the talk), I would say that the sun has been getting steadily brighter over Earth's history. Millions of years ago when CO2 was much higher, solar output was lower. The combined effect of sun and CO2 show good correlation with climate. For more details, see CO2 was higher in the past.
In fact, it's because the sun was cooler in the past that CO2 was so much higher. CO2 acts as a natural thermostat for our climate, regulated by rock weather which is the process of removing CO2 from the atmosphere by chemical reactions. When it's cooler, rock weathering activity slows so there's less removal of CO2 from the air. This means CO2 builds up, warms up the planet. Then as it gets warmer, rock weathering activity increases which removes CO2 out of the atmosphere.
This process is a natural way of keeping our climate within a certain temperature range. It means that if the sun was cooler, temperatures get cooler so rock weathering slows down, increasing CO2 levels. This is a fascinating process but of course, I didn't go into that much detail at the talk - I just mentioned that climate sensitivity has been consistent in the past.
But when I turned to the diagram (and Trenberth's original paper, whose title I forget), what I saw was something very different: I could not find evidence of any such violation (of course), but I DID find evidence of some very disappointing mistakes in presentation that made it difficult to verify that the First Law is satisfied by his numbers.
So what is the implication of all this for this article? I would say two things 1) do NOT assume a deep understanding of Thermodynamics, not even from engineers 2) make sure that quantities, variables, scales etc. are CLEARLY marked in the diagrams, and that simple questions like which way how much heat/energy flux flows are easily visible on the diagram. Diagrams like the one in post#27 in this thread are damn near useless.
As you've mentioned, though, perhaps the issue is that you got bogged down in the details. A different order of presentation may be required, where you present a simplified overview, then explore a few areas in more depth, perhaps having some of those detailed explanations in reserve to answer questions.
Also, from personal experience - if you were presenting to a bunch of students, then the 'blank looks' & shortage of questions aren't that uncommon while people are digesting the information just presented.
Explaining AGW in terms of radiative heat transfer, though, is something any fourth-year engineering students should understand (heat transfer was 3rd year subject when I studied at UQ, but that was nearly 20 years ago...)
I think a major problem with people's misinterpreting the significance of palaeoclimate is the naturalistic fallacy: what's natural is good for us, or the Nietzschean corollary (recently popularized by the Koch brothers in the Smithsonian exhibit)that what doesn't kill us makes us stronger. What the general public doesn't usually understand is the extremely rapid rate of current changes compared to past changes, as well as the fact that the most extreme palaeoclimates would not permit the survival of our current civilization.
There's also the logical fallacy that bverheggen noted (about past and present causes) that seems so elementary (to me) I'm amazed that people keep making it. Nevertheless, I've lost count of the number of times that I've used the forest fires/arson metaphor.