Arguments
What does past climate change tell us about global warming?
What the science says...
| Select a level... |
Basic
|
Intermediate
| |||
|
Greenhouse gasses, principally CO2, have controlled most ancient climate changes. This time around humans are the cause, mainly by our CO2 emissions. |
|||||
Climate Myth...
Climate's changed before
Climate is always changing. We have had ice ages and warmer periods when alligators were found in Spitzbergen. Ice ages have occurred in a hundred thousand year cycle for the last 700 thousand years, and there have been previous periods that appear to have been warmer than the present despite CO2 levels being lower than they are now. More recently, we have had the medieval warm period and the little ice age. (Richard Lindzen)
At a glance
Just imagine for a moment. You fancy having a picnic tomorrow, or you're a farmer needing a dry day to harvest a ripe crop. So naturally, you tune in for a weather-forecast. But what you get is:
“Here is the weather forecast. There will be weather today and tomorrow. Good morning.”
That's a fat lot of use, isn't it? The same applies to, “the climate's changed before”. It's a useless statement. Why? Because it omits details. It doesn't tell you what happened.
Climate has indeed changed in the past with various impacts depending on the speed and type of that change. Such results have included everything from slow changes to ecosystems over millions of years - through to sudden mass-extinctions. Rapid climate change, of the type we're causing through our enormous carbon dioxide emissions, falls into the very dangerous camp. That's because the faster the change, the harder it is for nature to cope. We are part of nature so if it goes down, it takes us with it.
So anyone who dismissively tells you, “the climate has always changed”, either does not know what they are talking about or they are deliberately trying to mislead you.
Please use this form to provide feedback about this new "At a glance" section. Read a more technical version below or dig deeper via the tabs above!
Further Details
Past changes in climate, for which hard evidence is preserved throughout the geological record, have had a number of drivers usually acting in combination. Plate tectonics and volcanism, perturbations in Earth's slow carbon cycle and cyclic changes in Earth's orbit have all played their part. The orbital changes, described by the Milankovitch Cycles, are sufficient to initiate the flips from glacials (when ice-sheets spread over much of Northern Europe and the North American continent) to interglacials (conditions like the past few thousand years) and back – but only with assistance from other climate feedbacks.
The key driver that forces the climate from Hothouse to Icehouse and back is instead the slow carbon cycle. The slow carbon cycle can be regarded as Earth's thermostat. It involves the movement of carbon between vast geological reservoirs and Earth's atmosphere. Reservoirs include the fossil fuels (coal/oil/gas) and limestone (made up of calcium carbonate). They can store the carbon safely over tens of millions of years or more. But such storage systems can be disturbed.
Carbon can be released from such geological reservoirs by a variety of processes. If rocks are uplifted to form mountain ranges, erosion occurs and the rocks are broken down. Metamorphism – changes inflicted on rocks due to high temperatures and pressures – causes some minerals to chemically break down. New minerals are formed but the carbon may be released. Plate tectonic movements are also associated with volcanism that releases carbon from deep inside Earth's mantle. Today it is estimated by the U.S. Geological Survey that the world's volcanoes release between 180 and 440 million tonnes of carbon dioxide per year - as opposed to the ~35 billion tonnes we release.
Epic carbon releases in the geological past
An extreme carbon-releasing mechanism can occur when magma invades a sedimentary basin containing extensive deposits of fossil fuels. Fortunately, this is an infrequent phenomenon. But it has nevertheless happened at times, including an episode 250 million years ago at the end of the Permian Period. In what is now known as Siberia, a vast volcanic plumbing-system became established, within a large sedimentary basin. Strata spanning hundreds of millions of years filled that basin, including many large coal, oil, gas and salt deposits. The copious rising magma encountered these deposits and quite literally cooked them (fig. 1).
Fig. 1: schematic cross section though just a part of the Siberian Traps Large Igneous Province, showing what science has determined was going on back then, at the end of the Permian Period.
Now laden with a heavy payload of gases, boiled out of the fossil fuel deposits, some of the magma carried on up to the surface to be erupted on a massive scale. The eruptions – volcanism on a scale Mankind has never witnessed - produced lavas that cover an area hundreds of kilometres across. Known as the Siberian Traps, because of the distinctive stepped landforms produced by the multiple flows, it has been calculated that the eruptions produced at least three million cubic kilometres of volcanic products. Just for a moment think of Mount St Helens and its cataclysmic May 1980 eruption, captured on film. How many cubic kilometres with that one? Less than ten.
Recently, geologists working in this part of Siberia have found and documented numerous masses of part-combusted coal entrapped in the lavas (Elkins-Tanton et al. 2020; fig. 2). In the same district are abundant mineral deposits formed in large pipes of shattered rock as the boiling waters and gases were driven upwards by the heat from the magma.
Fig. 2: an end-Permian smoking gun? One of countless masses of part-combusted coal enclosed by basalt of the Siberian Traps. Photo: Scott Simper, courtesy of Lindy Elkins-Tanton.
It has been calculated that as a consequence of the Siberian Traps eruptions, between ten trillion and one hundred trillion tons of carbon dioxide were released to the atmosphere over just a few tens of thousands of years. The estimated CO2 emission-rate ranges between 500 and 5000 billion tonnes per century. Pollution from the Siberian Traps eruptions caused rapid global warming and the greatest mass-extinction in the fossil record (Burgess et al, 2017). There are multiple lines of hard geological evidence to support that statement.
We simply break into those ancient carbon reservoirs via opencast or underground mines and oil/gas wells. Through such infrastructure, the ancient carbon is extracted and burned. At what rate? Our current carbon dioxide emissions are not dissimilar to the estimated range for the Siberian Traps eruptions, at more than 3,000 billion tons per century. The warning could not be more clear. Those telling you the climate's changed before are omitting the critical bit – the details. And when you look at the details, it's not always a pretty sight.
Last updated on 14 February 2023 by John Mason. View Archives
@446, that's why I quoted IPCC and GRIDA that abrupt glaciations are possible, volcanic eruptions and solar activity were given just as examples of possible causes that we can not model reliably
@449, I understand that climate models can not predict abrupt changes, that's why I am worried about abrupt climate changes that we can not predict, that's my whole point from the very start
@448, well, changes from glacials to interglacial are normally quite rapid (much more rapid than 20th century warming), so rapid warming is not something completely new for the climate system
Before the ice ages started (when CO2 dropped below today's levels) the climate was a lot more stable. I am not talking about extreme CO2 levels, just a 100, at most 200ppm more.
@442, well, ice ages started when CO2 levels were about today's, so it is entirely possible
skeptic123 @428, last and most offensive first! The GISP2 ice core data represents a regional record only, not a global record. Are you seriously trying to suggest that global temperatures vary as rapidly as do regional temperatures (and regional temperatures with one of the most rapid rates of change of temperature found on the planet)? Further, are you seriously trying to suggest that the magnitude of temperature change from a regional record is also to be found in the global record? If so, you have largely disqualified yourself from the conversation on the basis of complete ignorance of basic relevant facts. If not, you have certainly disqualified yourself from the conversation on the basis of deliberately presenting evidence in a form you know to be misleading.
Taking the former, more generous interpretation, consider this graph of eight full holocene regional temperature proxies:
Individual proxies show rapid variation in temperature of considerable amplitude. Of those, GISP2 (light blue) shows the greatest variation, having the highest peak holocene temperature anomaly, and the lowest most recent temperature anomaly. Because peaks in various records rarely coincide, and some records are always out of phase with others (ie, have troughs where the others have peaks), the arithmetic mean of all 8 proxies shows both much less absolute temperature variation, and much lower rates of temperature change than do individual proxies. Consequently, presenting a single proxy (let alone the most variable proxy) as representative of either absolute magnitude of global temperature change or of rates of temperature change over the holocene is fundamentally misleading (whether from ignorance of the effects of regression to the mean, or intent to decieve).
(As an aside, the overall decrease in the mean temperature over the holocene is largely an artifact of a NH bias in the individual proxies (ie, there are more NH than SH proxies presented), a problem also with Marcott et al. An unbiased sample is likely to show much less, or possibly no decline over that period.)
The same basic problems afflict the Vostock proxy record (blue in the above graph). The absolute temperature magnitude shown in the Vostock record is approximately twice the absolute variation in the global record. Further, periods of rapid decline rarely coincide with other regional proxies so that periods of rapid decline in the Vostock record will coincide with much slower decline (or sometimes even increases) in a global record. Further, your quote from the caption of the Vostock graph that you show is misleading out of context, and not supported by the evidence in any event.
In particular, while rapid temperature changes can occur over only a few decades, the trend over successive decades will often greatly slow or reverse direction. The consequence is that multi-century temperaturetrends are typically very slow. This can be seen in a scatter plot of time intervals vs temperature change in the Vostock record:
(Larger version)
While there are some very rapid short duration changes, they are seen to quickly reverse themselves. The result is that changes over a century or more are at rates of -1C per century or less. Typically much less. As the transition from inter-glacial to glacial in the Vostock record requires a temperature change of approximately -6C, that means transitions from interglacial to glacial cannot occure in less than 500 years or more. Indeed, based on a pixel count of the graph of the vostock record, the most rapid interglacial to glacial transition (taken as the interval between 0C and the bottom of the first trough below -4C, or to -6 C, which ever is shorter) takes 6250 years (approx 240 thousand years ago). The next most rapid, and most recent took thirteen thousand years.
Finally, the TAR quote references Alley 93, which analysed early icecore data from Greenland. The rapid transition it found was the Younger Dryas, which was primarilly a North Atlantic phenomenon, and which involved much slower transitions in temperature when averaged across a number of diverse locations. (In 1993, only Greenland proxies were available back so far in time.) It is, therefore, obsolete, having been disproved by more recent data.
(I've run out of time, and will return to the CO2 issue later.)
skeptic123 @452 - In fact climate models predict more abrupt temperature changes as a result of volcanic activity than actually occur. Your claim, therefore, is the exact reverse of the truth.
@453 As I have just shown, changes from interglacial to glacial are relatively gradual, with the most rapid representing a mean rate of temperature change at Vostock of less than 0.01 C per decade, ie, a tenth of the rate of the current trend (which the AGW deniers insist on calling a pause). Transitions from glacial to interglacial are typically much more rapid, and projected temperature increases over the coming century being much more rapid again, with the equivalent temperature increase to that between interglacial and glacial within 200 years on the outside (with BAU).
@454 Actually, the transition from glacial to interglacial typically occurs with CO2 concentrations 20-40 ppmv below preindustrial levels. No such transition has ever occurred with CO2 concentrations at modern values.
@455, my mistake about GISP2, it was more an example of smoothing
as for the Vostock data, it was given as example of CO2 levels, since CO2 is fairly evenly distributed, it should still be valid even if it is regional
about the rapid changes, a 2-3 degrees drop for a century even if it reverses itself would still have some quite bad effects
@456, 452 - I claim that we can not predict volcanic eruptions, not the effects of them
@456, 453 - ok, so where do the statements from IPCC and GRIDA about rapid glaciations in a matter of few decades come from?
@456, 454 - so, why did the ice ages epoch start in the first place then, CO2 concentrations before it started were above current levels. It is of course possible that the CO2 got sucked somewhere first, but do we know of such an event
@445, So you are promoting geo-engineering of the planet?
skeptic123 @457, if you only presented GISP2 as an example of an unsmoothed record, you deliberately presented an irrelevant example in that Marcot et al is not a smooth of the GISP2 record.
Presenting the Vostock graph as an example of rapid declines in CO2 level is even worse for your case given that declines in CO2 level are less rapid than are declines in temperature in the Vostock record. (Ie, my points are even stronger as applied to CO2.)
A 2-3 C drop on global temperature over a century or two would indeed be no fun at all. However, the likelihood of such an event with 280 ppmv of atmospheric CO2 is very low. We know this because only three events in the holocene come close to such a situation. The largest of these, the 8.2 kiloyear event was the result of a sudden spike in sea level resulting from the release of damned melt water durring the melting of the laurentide ice sheet. That cause is not applicable for stable preindustrial conditions or indeed at anytime except following the end of a glacial. The 5.9 kiloyear event and the 4.2 kiloyear event where much smaller and preceded the rise of CO2 concentrations to 280 ppmv. The most recent similar event was the Little Ice Age, with global temperatures declining by just 0.2 to 0.4 C.
With respect to the LIA, if that is the sort of unexpected climate variability we have to worry about, you need to make the difficulty argument that a low risk of a 0.3 C decline in global temperature is more threatening than a very high probability of a 3-5 C rise in global temperature. You also need the face the fact that these events seem to have a period of about 1500 +/- 500 years, and with the most recent event starting less than 500 years ago, the next event cannot be expected for at least another 500 years.
In short, it is absolute folly to not intervene to prevent an almost certain, large and very rapid event starting now because of the low risk of a small, relatively slow event that may occur 500-2000 years from now. Yet that folly is the basis of your argument.
@ 447, the counter to this is that California isn't a global indicator no matter what those guys handing out cds on the boulevarde say !
@459, not geoengineering, just an increase of CO2 by another 1-200ppm to be on the safe side. The effects of geoengineering as a broad term are not well known while I hope you would agree the effects of CO2 increase are pretty well known
@460, 457, for GISP2 I meant the green line, but I agree I was wrong, it's not smoothing, it's worse, it's polynomial fit
the Vostock data was an example that quick CO2 decline is possible, I wasn't trying to relate it to the temperature
I don't have the actual data about past temperatures, so I can't quote numbers, that's why in 440 I said that I have to rely on people who do have it and I will quote them again
"Over the last 400,000 years the Earth's climate has been unstable, with very significant temperature changes, going from a warm climate to an ice age in as rapidly as a few decades."
www.grida.no/publications/vg/climate/page/3057.aspx
@461, I agree :)
@462, do you mean 1-200 ppm from now or from then? Why shouldn't we go up 300 ppm from now to be really safe? (Are you sure you are prepared to be questioned about this ad hoc policy?)
skeptic1223@458.
Within your mercurial argumentation & hypotheising you write:-
A bit of light reading for you. While the conclusions from such studies as presented within the linked document remain incomplete, knowledge of what is under examination may stop painfully simplistic assertons about CO2 levels and the causes of glaciation.
@466, thanks, this is a nice article, and a great support for my thesis. So, according to the article the ice ages started at levels of CO2 close to today's due mainly to atmospheric and oceanic conditions. Let me quote
"These preconditions—moisture plus an Arctic nucleus for cooling—would have made the climate system highly susceptible to ice sheet growth. Even modest changes in the global environment would have been sufficient to tip the scales and lead to the onset of major Northern Hemisphere glaciation.
Just such a change occurred between 3.1 and 2.5 million years ago, as Earth’s axis fluctuated so that the planet’s tilt toward the sun was less than today’s angle of 23.45 degrees."
So, what's stopping the same thing happening today, I mean not due to an axis tilt but for example due to a major volcanic eruption or decline in solar activity, the other conditions seem to be the same.
@465, 1-200ppm from now, that number is just an example, when the ice ages started CO2 concentration was similar to today's, before that the climate was relatively stable, so some level of CO2 above today's should make matters safe
skeptic1223@467.
You write "So, according to the article the ice ages started at levels of CO2 close to today's..."
Where exactly does the article make this assertion?
@469, it doesn't directly, however it states that the reason for the glaciation were atmospheric and oceanic conditions plus a tilt in the Earth's axis, it doesn't say anything about a sudden drop in CO2 and we know that 3 million years ago CO2 concentrations were the same as today's, so.
"Scientists believe that the most recent period with a 400 ppm level of carbon dioxide in the atmosphere was the Pliocene, between five million and three million years ago, according to the Scripps Institution of Oceanography, which keeps track of the Keeling Curve."
www.livescience.com/29437-carbon-dioxide-record-broken.html
skeptic1223 @468.
Advocating increasing C02 to prevent a future ice age sounds like someone saying "Let's burn our house down today to prevent the possibility of flooding due to rising sea levels next year."
skeptic1223 - As numerous posters have pointed out, there is really no evidence supporting a risk of sudden glaciation right now. The ice age cycles have in the past been driven by Milankovitch forcings, orbital changes, episodes like the Younger Dryas cooling event were as far as we can tell due to conditions (giant ice dams releasing) that simply don't exist today, and were regional, not global.
Our emissions have essentially removed any chance of a glaciation for the next Milankovitch cycle - we're just not at risk of an ice age. On the other hand, the warming we've already committed to points toward mass extinctions (as climate change exceeds the speed of species movement to keep up with their environments), considerable impacts on agriculture, disease, weather, and many many other impacts.
You've presented no evidence whatsoever supporting your hypothesis, and therefore your suggestion that we need another 100-200ppm of CO2 is just absurd. There's no risk of sudden cooling, and multiplying the impacts of warming seems just foolish.
skeptic1223@470.
"It doesn't directly," you write. Indeed it doesn't make that assertion at all.
Now you are asserting that "we know that 3 million years ago CO2 concentrations were the same as today's." You provide a quote from a livescience blog but is your assertion supported by this livescience blog? The answer is "no".
If you could be bothered to examine the Scripps posting referenced by your livescience blog, you will see that it talks of "The Pliocene is the geologic era between five million and three million years ago. ... It is trickier to estimate carbon dioxide levels before then(800kybp), but in 2009, one research team reported finding evidence of carbon dioxide levels ranging between 365 and 415 ppm roughly 4.5 million years ago." So this citation is actually a little early for our purpose and the range of CO2 level has mostly been left behind by today's anthropogenic emissions.
Perhaps you can find some support for your assertion elsewhere. But then do bear in mind the actual age we are interested in is the point that Arctic glaciation kicked off which I believe was 2.7Mybp.
skeptic123 @463, looking at the Vostock data another way, I examined in detail each of the four transitions from inter-glacial to glacial in the Vostock data. From each such transition, I determined the minimum interval for the transition, calculated as the time from the most recent relevant datapoint above -2 C to the time of the first datapoint following which was below -4 C. The relevant intervals were, from oldest to newest:
3426 years (average rate of decline = -0.006 C per decade)
4399 years (average rate of decline = -0.005 C per decade)
2654 years (average rate of decline = - 0.008 C per decade)
2028 years (average rate of decline = -0.01 C per decade)
(Note: the rates of decline are for a regional temperature value. Global values would be about half that.)
No matter how many times you repeat the quote, the data does not support it. Further, science proceeds by evidence, not by out of context quotation (which is rather the mark of pseudoscience).
Now, it is possible that the scientists who made that claim define an "ice age" as any period with a temperature anomaly less than x, where x is -0.5, or -0.1, or some other arbitrary value. It is also possible that they consider temperatures of x + 0.5 C as "a warm climate". In that case, what they say is true, trivial, and so vague without the specification that that is how they interpret their words, and a specification of x as to be useless. If you want to use the quote, it is therefore incumbent on you to find out the exact interpretation the authors give to the words.
Absent that effort, however, it remains that the data directly contradicts the claim supposedly based upon it. So, if you are not prepared to make the effort to provide the context of the quote on whose authority you rest, it is incumbent on you to follow the data. Failure to do so simply demonstrates that you accept data only if you think it supports your position. Worse, it shows that you present data that in fact falsifies your position as supporting it, and refuse to acknowledge the detailed examination that shows that the data refutes your position. Again, those are the hallmarks of pseudoscience.
Long experience has shown that debating with pseudoscientists is completely unprofitable in that their positions are not based on reason and evidence, and therefore cannot be altered by either reason or evidence. I also think that the complete divorce of your opinions from actual data is sufficiently evident to any interested readers that I do not need to spell it out again. Consequently I will ignore your responses in future until such time that you start correcting your position based in the actual data.
[JH] Your above comment is actually directed at skeptic1223.
skeptic1223's over-enthusiasm for linking present-day CO2 levels to some threashold CO2 level which allowed the era of ice-ages to kick off appears to have run its course. Yet he did present one aspect of climate that seldom gets discussed.
skeptic1223 began his input way up @420 with the observation that orbital forcing had become a bigger feature of climate since the inception of the Arctic glaciation by creating higher ampitude oscillations in the global temperature record. The graphic below (derived from Lisiecki & Raymo (2005) Fig 4) was presented @423 by way of illustration.
Such apparent increases in ampitude do support the idea of an increase in climate sensitivity in some manner as in the long run orbital forcing cycles are constant in size. Such forcing is large at specific latitudes while small globally. To suggest this increased ampitude (and thus increased sensitivity) is somehow a function of CO2 levels is of course wrong. But it does beg the question - why doesn't this increased ampitude in oscillation make ECS vary with temperature/Arctic glaciation?
My view of this is that to calculate the sensitivity of climate to orbital forcing from what appears a hysteresis loop could be used to show big sensitivity changes but these would not be very helpful for our purpose of finding ECS under AGW. Thus the slow feedbacks of albedo, CO2, methane, etc. are considered as forcings rather than feedbacks to allow a meaningful ECS to be calculated.
The reference cited by AR5 in this matter is PALAEOSENS Project Members (2012) which states:-
This accounting of slow feedbacks as forcings is a reason for ECS only accounting for fast feedbacks.