Arguments
CO2 lags temperature - what does it mean?
What the science says...
| Select a level... |
Basic
|
Intermediate
|
Advanced
| ||||
|
CO2 didn't initiate warming from past ice ages but it did amplify the warming. In fact, about 90% of the global warming followed the CO2 increase. |
|||||||
Climate Myth...
CO2 lags temperature
"An article in Science magazine illustrated that a rise in carbon dioxide did not precede a rise in temperatures, but actually lagged behind temperature rises by 200 to 1000 years. A rise in carbon dioxide levels could not have caused a rise in temperature if it followed the temperature." (Joe Barton, US House of Representatives (Texas) 1985-2019) - Full Statement
At a glance
Antarctic ice-core data today provide a continuous record on temperature and atmospheric composition that goes back for some 800,000 years. The data track the last few glacial periods and their abrupt endings, with rapid transitions into mild interglacials. But in some of the ice-cores, temperature rises first and is followed, a few hundred years later, by rising carbon dioxide (CO2) levels.
Certain purveyors of climate-myths seized on this observation, claiming it to be “proof” that carbon dioxide doesn't cause climate change. Wrong, wrong, wrong. But how? The answer lies in a beer-can.
In fact, you can do this one yourself. You need two cans of any fizzy beer. On a nice summer's day, take one out of the fridge and place it outside in direct sunshine for a few hours. Leave the other where it is. Then open the two at the same time. The warm one will froth like mad, half-emptying the can and making a mess. What is left in the can will be horrible and flat. Conversely, the one straight from the fridge will just give a “pfft” noise and will be pleasant to drink, being cool and fizzy.
What's that got to do with this myth? Well, you have just demonstrated an important point about the solubility of CO2 in water. CO2 gives fizzy drinks their fizz and it is far more soluble in colder water. As the water warms, it cannot hold onto as much CO2 and it starts to degas. Hence that flat lager.
Exactly the same principle applies to the oceans. When global warming is initiated, both land and the oceans start to warm up. On land, permafrost starts to thaw out, over vast areas. Carbon dioxide (and methane) are released, having been trapped in that permafrost deep-freeze for thousands of years. At sea, that “warm beer effect” kicks in. Thanks to both processes, atmospheric CO2 levels rise in earnest, amplifying and maintaining the warmth. That rise in CO2 thereby caused more of the gas to be released, warming things up yet more in a vicious cycle, known as a positive feedback. Other feedbacks kick in too: for example as the ice-sheets shrink, their ability to reflect Solar energy back out to space likewise decreases, so that heat is instead absorbed by Earth’s surface.
The trigger for the initial warming at the end of an ice-age is a favourable combination of cyclic patterns in Earth's orbit around the Sun, leading to a significant increase in the solar energy received by Earth's Northern Hemisphere. That's no secret. Glacial-interglacial transitions are caused by several factors working in combination – triggers and feedbacks. We've understood that for a long time.
And when you think about it, saying CO2 lagged temperature during glacial-interglacial transitions so cannot possibly be causing modern warming is a bit like saying, “chickens do not lay eggs, because they have been observed to hatch from them".
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
That CO2 can lag behind but amplify temperature during a glacial-interglacial transition was in fact predicted as long ago as 1990. In the paper The Ice-Core Record: Climate Sensitivity and Future Greenhouse Warming by Claude Lorius and colleagues published in the journal Nature in 1990, a key passage reads:
"The discovery of significant changes in climate forcing linked with the composition of the atmosphere has led to the idea that changes in the CO2 and CH4 content have played a significant part in the glacial-interglacial climate changes by amplifying, together with the growth and decay of the Northern Hemisphere ice sheets, the relatively weak orbital forcing and by constituting a link between the Northern and Southern Hemisphere climates."
This was published over a decade before ice core records were accurate enough to confirm a CO2 lag. We now know that CO2 did not initiate the warming from past ice ages but it did amplify the warming. In fact, about 90% of the global warming followed the CO2 increase.
Antarctic ice cores reveal an interesting story, now going back for around 800,000 years. During this period, changes in CO2 levels tend to follow changes in temperatures by about 600 to 1000 years, as illustrated in Figure 1 below. This has led some to disingenuously claim that CO2 simply cannot be responsible for the current global warming. Unsurprisingly, such a claim does not tell the whole story.
Figure 1: Vostok ice core records for carbon dioxide concentration and temperature change.
The initial change in temperature as an ice-age comes to an end is triggered by cyclic changes in Earth’s orbit around the sun, affecting the amount of seasonal sunlight reaching Earth’s surface in the Northern Hemisphere. The cycles are lengthy: all of them take tens of thousands of years to complete.As both land and oceans start to warm up, they both release large amounts of CO2 into the atmosphere, from melting permafrost and from warming ocean water, since CO2 solubility in water is greater in cold conditions. That release enhances the greenhouse effect, amplifying the warming trend and leading to yet more CO2 being degassed. In other words, increasing CO2 levels become both the cause and effect of further warming. Once started, it’s a vicious, self-reinforcing cycle - an excellent example of what science refers to as a positive climate feedback.
Indeed, such positive feedbacks are necessary to complete the shifts from glacial to interglacial conditions, since the effect of orbital changes alone are too weak to fully drive such variations. Additional positive feedbacks which play an important role in this process include other greenhouse gases like methane - you may have seen videos of that gas bubbling up through icy lakes in permafrost country and being ignited. Changes in ice sheet cover and vegetation patterns determine the amount of Solar energy getting absorbed by Earth’s surface or being reflected back out to space: decrease an ice-sheet’s area and warming will thereby increase.
The detailed mechanisms for the above general pattern have of course been investigated. In a 2012 study, published in the journal Nature (Shakun et al. 2012), Jeremy Shakun and colleagues looked at global temperature changes at the commencement of the last glacial-interglacial transition. This work added a lot of vital detail to our understanding of the CO2-temperature change relationship. They found that:
1) The Earth's orbital cycles triggered warming in the Arctic approximately 19,000 years ago, causing large amounts of ice to melt, flooding the oceans with fresh water.
2) This influx of fresh water then disrupted ocean current circulation, in turn causing a seesawing of heat between the hemispheres.
3) The Southern Hemisphere and its oceans warmed first, starting about 18,000 years ago. As the Southern Ocean warms, the solubility of CO2 in water falls. This causes the oceans to give up more CO2, releasing it into the atmosphere.
4) Finally, CO2 levels may lag temperature in some ice-core records from Antarctica, but in some other parts of the world the reverse was the case: temperature and CO2 either rose in pace or temperature lagged CO2. Figure 2 demonstrates this graphically and shows how things are never as simplistic as purveyors of misinformation would wish.
Figure 2: Average global temperature (blue), Antarctic temperature (red), and atmospheric CO2 concentration (yellow dots). Source.
Last updated on 14 February 2023 by John Mason. View Archives
The article states that CO2 is a feedback that assists Milankovitch cycle warming / cooling. Many commentaters here have also stated that Milankovitch influences are weak, and need a feedback to have any effect on climate. Can someone explain why?
The following graph is of Milankovitch forceing changes in the all-important high latitudes, and they are up to 100 wm2, or 25% of the average insolation strength. In what way is this large change in insolation 'insignificant', requiring a feedback to assist it?
https://www.cabrillo.edu/~rnolthenius/climate/Denialists/D-Ruddiman/milak&temp.jpg
[Rob P] - See this post: Milankovitch Cycles, by Chris Colose.
Let me see if I can make that graph display properly:
[img]https://www.cabrillo.edu/~rnolthenius/climate/Denialists/D-Ruddiman/milak&temp.jpg[/img]
And other question, if I may.
If the world warms, and increasing CO2 acts as a positive feedback, why does the temperature eventually stop increasing, and reverse? Surely more CO2 means ever increasing temperatures.
Especially as Milanokovitch cycles are not very influential, and need a feedback, as many commentaters have said here.
Conversely, when the world cools and reducing CO2 is less able to warm the world, why does the temperature not continue cooling and end in a snow-ball Earth?
Surely a strongly positive feedback CO2 would be very unstable, leading to runaway temperatures both ways. But this is not what we see, because temperatures are tightly controlled withing 10ºc.
Thanks
[Rob P] - The flaw in your reasoning is the assumption that CO2 would keep increasing. The leading hypothesis is that carbon stored in the Southern Ocean is released by warming. Once that ocean carbon store is exhausted, no more extra CO2 is reaching the atmosphere. So no more push in the direction of warming.
There is a recent paper that seems to be making a stronger case for this hypothesis: Southern Ocean buoyancy forcing of ocean ventilation and glacial atmospheric CO2. Here's some slides from a lecture given by the papers authors.
The graphic which tatelyle was trying to post @477 is below.
tatelyle, you're talking about a runaway warming or cooling, which can happen only if the feedbacks are greater than 1. It is obviously not the case. See the "positive feedback means runaway warming" post, use the search engine.
tatelyle @476, the Earth is effectively a sphere. No amount of rotations, changes of tilt or axial wobbles will change the total amount of energy recieved by a sphere by radiation from a distant source. All that will change is the location on the surface that recieves the insolation. It follows that while changes in obliquity and precession may change the insolation in July at 65 degrees North substantially, it cannot change the total amount of energy recieved by the Earth anymore than the day night cycle can. Thinking that it can is just a more subtle version of the error that thinking that just because it happens to by night time at my current location, therefore the global average energy recieved from the Sun at the moment is zero.
In contrast, changes in eccentricity can change the annual average insolation. That is because with a more eccentric orbit, the Earth spends more time at a greater distance from the Sun. That more than compensates for the briefer time spent closer to the Sun. This effect, which is the only Milankovitch cycle that actually changes global, annually averaged insolation, is very small. If my calculations are correct, the maximum change is of the order of 0.2% or about 0.5 W/m^2, and is typically much smaller than that.
That being the case, milankovitch cycles can only significantly effect global temperatures if there are, not just feedbacks, but differences in th feedbacks that depend on location and season. If the winter feedback to the change in insolation in the NH was as strong as the summer feedback, milankovitch cycles could not result in glaciations. Likewise, if the NH feedback was as weak as the SH feedback, the milankovitch cycles could not result in glaciations.
And of course, if there were no feedbacks, there could be no variation in the strength of feedbacks based on latitude and season.
As it happens, the feedback that shows the most seasonal and latitudinal difference is the albedo feedback. This is, primarilly because in the NH, lower latitude snow falls on land and can accumulate, wheras in the SH it falls on water and melts. As a result, the NH albedo feedback is stronger. Further, because of seasonal variation in insolation, the albedo feedback is also seasonal, being stronger in the summer than in the winter. That means that an albedo feedback by itself could, in principle, induce glaciations from milankovitch cycles. In practice, however, the size of the albedo feedback is fairly well known from knowledge of the size of glaciation. It turns out to be about half of what is needed to actually explain the temperature variations. The total temperature variations can be fully explained only by assuming an additional, greenhouse feedback. As that feedback is predicted by radiative theory in any event, it would be obtuse to insist that it does not exist, thereby requireing a new radiative theory (which has not been supplied by the "deniers") and that we add significant complications to our theory of milankovitch cycles to make room for this non-existent new radiative theory.
tatelyle @478, from the wording of your comment, I assume you are making the error discussed here. For what it is worth, a feedback is positive if its incremental gain (g) is positive. Because feedbacks are iterative, ie, they respond to temperature increases resulting from feedbacks just as much as they respond to those from forcings, the final response of a feedback (f) is given by the formula, f=F/(1-g), where F is the Forcing, and f and g have already been defined. A simple look at that formula tells you that feedbacks do not "runaway" unless g is greater than or equal to 1, which is not the case.
You may, however, by raising the issue that atmospheric CO2 concentration can be a feedback on temperature, and temperature is a feedback on atmospheric CO2. Independent of the factors discussed by Rob P, CO2 concentration increases as a linear function of temperature, all else being equal. In contrast, global temperature increases with the log of CO2. That means that, all else being equal, the CO2 increase loop will quickly self damp. You would require the CO2 concentration to increase exponentially with temperature to get a runawy effect (until oceanic CO2 was effectively exhausted).
Tom Curtis @481.
The figures I have read elsewhere do conform with your calculated 0.2% for the global annual insolation variation (resulting from obliquity which is varying from roughly zero to a little above 0.06). Mind, the graphic tatelyle was linking to above (as @479) originates from the web-page here that for some reason gives the figure for global annual insolation variation a lot higher at 0.7% although the graphic they show for this (their fig 3) does look to be the right shape. Still, they aren't the only folk to get their numbers wrong on this matter. This Washington Edu slide show manages to miss off the decimal point in their insolation figure, twice, giving a 0.18% change in insolation from obliquity as a rather toasty 5Wm-2.
MA Rodger @483, thank you for drawing my attention to this graph:
That shows an approximate 2 W/m^2 total range of fluctuation, but is calculated for TOA insolation. Converting for current albedo brings that down to 1.4 W/m^2, or approx three times my calculated range. The key point is that is still a tiny forcing. If total anthropogenic forcing since the pre-industrial could be limited to 1.4 W/m^2, global warming would not be a problem. Conversely, if we asssume that the temperature differences between glacial and interglacial were due to a feedback on the global annual average value (calculated as 1.4 W/m^2), then the climate sensitivity would be approximately 13 C per doubling of CO2.
Setting that aside, however, I notice that Chis Colose calculates a similar value to mine:
John Baez finds only a 0.167% range (or 0.4 W/m^2 with current albedo):
Baez also shows the derivation of the result.
Even the Washington Edu slide you link to gets the percentage change about right (0.18%), but messes up in calculating the change in forcing that results.
Tom Curtis @484.
I would say your 0.2% is sound. Another person calculating this value is Tamino who arrives at 0.18% and 0.61Wm-2 for obliquity ranging "between a minimum of near zero, and a maximum of slightly less than 0.06" and prior to albedo considerations. My take on the www.climatedata.info is that they calculated the discal forcing (as graphed in their fig 3 that you show) but then they forgot to divide by 4 when they applied that as a global forcing to calculate the percentage, which should therefore be 0.175%.
It is usual, when discussing Milankovitch cycles, to dismiss any effect of obliquity on global, annual mean insolation based on the fact that the Earth is a sphere. I did it myself @481. As the Earth is an oblate spheroid, however, it is not strictly accurate. In particular, on the equinoxes, the Earth presents its minimum aspect to the Sun, showing an eclipsed area of 1.2737 x 1014 m2 to the Sun. As the Earth moves to the solstice (either winter or summer) it presents its maximum aspect, showing an eclipsed area of 1.2744 x 1014 m2. That represents a difference of 0.05% in recieved sunlight, or approximately 0.12 W/m2. That seasonal variation is much less than the 6.8% (16.2 W/m2) seasonal variation due to the eccentricity of the Earth's orbit.
As obliquity changes, the maximum eclipsed area (ie the insolation at the solstice) also changes, although the minimum eclipsed area (ie, insolation at the equinoxes) does not. Just as the variation due to the Milankovitch cycle for eccentricity (~0.175%) is much smaller than the seasonal variation, so also is the variation due to the Milankovitch cycle in obliquity much smaller than the seasonal variation. Specifically, it represents only a 0.01% variation in the solstice insolation. As the equinoctial insolation does not change, the variation in global, annually averaged insolation due to changes in obliquity will be much less than 0.01%.
For perspective, these factors are also less than the difference of treating the Earth as a perfect oblate spheroid, or allowing for the additional interception of sunlight due to continents, mountains and, of course, the atmosphere.
Thank you for your explanations. However:
>>Milankovitch cycles require feedbacks.
I understand all your arguments, that overall insolation cannot change much. However, the critical season and region for Ice Age modulation is the NH summer at northerly latitudes, because of the large NH landmass, as you say. It is this insolation that will decide whether the winter ice sheets melt, and so it is this region that will modulate Ice Ages.
Thus obliquity, precession and eccentricity must ALL effect ice sheet extent. Obliquity will obviously increase summer warming at high latitudes. Precession will also increase summer warming at high latitudes, when synchronised with obliquity. And it is the NH summer melt that is important, rather than the global energy budget. And therefore it is high latitude NH insolation and albedo that are important.
And as that Milkanovitch graph shows, the critical summer NH insolation can change by up to 25% during the Milankovitch cycle. Not a few wm2 here or there, but a whopping 90 wm2. So why would the 4 wm2 provided by CO2 be significant, in comparison to the 90 wm2 of the Milankovitch cycle?
>>Once that ocean carbon store is exhausted, no
>>more extra CO2 is reaching the atmosphere.
Thank you, but I do not follow that argument. CO2 solubility can proceed all the way up to 60oc, so why would the 'CO2 store' be exhausted at say 20oc? If the average southern ocean temperature in an Ice Age is 10oc, then the reducing solubility of CO2 in the ocean should continue at almost the same rate all the way to 20oc or more. So why would the outgassing of CO2 stop during an Interglacial, just because the southern ocean has reached its current 15oc temperature?
http://www.rocketscientistsjournal.com/2006/10/_res/CO2-06.jpg
Thank you,
[Rob P] - "Thank you, but I do not follow that argument."
Read the links I provided to you earlier, particularly the recent Watson et al (2015) paper. Here's the abstract:
Citing denier blogs lends the opposite of credibility.
[PS] debunking such nonsense however is however largely what this site is about.
tatelyle @487:
Causing "winter ice sheets" to melt is a feedback. Further, it is a seaonal and regional feedback. If the seasonal and regional feedbacks in all areas and seasons were equally strong, then that feedback would be balanced by other, opposite feedbacks because the changes in global annually averaged insolation balance out (for all intents and purposes). It would follow that there would be no net change in GMST. That is the point of the arguments above. Ergo, when you say "I understand all your arguments" but "feedback", which is essentially what you have said, you clearly do not understand the arguments at all.
The Earth has 46 million kilometers squared of area North of 55 North, out of a total of 510 million kilometers squared. That represents 9% of the Earth's total surface area. Further, Summer constitutes just 25% of the year. Ergo, your 90 W/m^2 at 65 North in Summer is equivalent to 2 W/m^2, globally averaged. So why would the 3.7 W/m^2 globally averaged forcing from a doubling of CO2 by more significant than the 2 W/m^2 globally averaged milankovitch forcing when we ignore everything but the milankovitch forcing in Summer in the high NH? Or the less than 0.4 W/m^2 globally averaged forcing if we include the full milankovitch forcing?
Your original question was why do we need to infer strong feedbacks given that the seasonal/regional milankovitch forcing was strong. Even only considering that forcing (ie, ignoring the opposite milankovitch forcings in other regions or seasons), the globally averaged mean annual temperature would only change by 0.65 C without feedbacks. In practise it changed by about 5 C. Further, the SH temperature changes have the same sign, and approximately match the size of the NH temperature changes. That would be impossible unless at least one of the feedbacks was necessarilly global in nature, and quite strong (to counteract the opposite signed local milankovitch forcings in the SH).
[PS] rate matters too. The change of forcing at 65N only per century is about 2 orders of a magnitude smaller than rate of change of in CO2 forcing globally.
Thank you again for your replies, but they are still not making complete sense. You said:
>> feedback would be balanced by opposite feedbacks because the
>>changes in global annually averaged insolation balance out.
But that is not true is it? - because as you said the NH and SH are very different, and so there is no 'balance' or equivalence between them.
And I can prove this to you because in the graph in post 479, the Interglacial temperature rise closely follows northern hemisphere Milankovitch forcing, and not the southern hemisphere forcing (which would happen 12,800 years later). So although the Ice Age becomes a global phenomina, it is triggered and forced by northern hemisphere Milankovitch forcing and only by northern hemisphere forcing. That is an indisputable fact. So Interglacial forcing and warming is asymmetric and regional, not global.
So there is no 'balance' as you call it. Ice Ages are caused by NH conditions, and so your spreading out the NH forcing across the globe, and thereby reducing it to just 2 wm2 is completely erroneous. Firstly, the NH summer melt is 40% of the year, as you know. And secondly, the SH is completely irrelevant, as we has just been demonstrated. So the Milankovitch forcing is actually 6.5 wm2, before any feedbacks have been added.
So why would CO2 be important in this process?
Your assertion that CO2's influence is significant is based upon a doubling of CO2, or 3.7 wm2. But that assertion is completely erroneous and misleading. The initial CO2 increase at the beginning of the Interglacial warming is about 20 ppm, and this equates to less than 1 wm2. And why would less than 1 wm2 have any influence on world temperature? So what will be the primary driver - 6.5 wm2 of Milankovitch warming, or less than 1 wm2 of CO2? And since a warmer NH summer during an Ice Age is going to produce melt, exposing dirty old ice, the primary feedback is more likely to be reducing albedo than increasing CO2.
>>Citing denier blogs lends the opposite of credibility.
What a silly comment. Why are comments like this tolerated on a serious blog? I did not cite anything, I borrowed a standard graph that is present on hundreds of web pages. So do you have an issue with that graph? Do you consider it incorrect or misleading in any way? And if not, then why the silly comment?
Actually, I would like to take back that 6.5 wm2 that I calculated above.
Spreading the summer Milankovitch forcing all over the winter period makes no sense whatsoever. The ice-sheets are in perpetual lock-down for the winter, and nothing happens, so the ambient temperature is irrellevant. It is only the summer melt that is meaningful. And when I was doing research on the Himalayan glaciers, melting only occurred during sunny days, while during cloudy days and nights the glacier remained frozen.
And if we erase the winter dilution that was suggested, the Milankovitch forcing on Ice Ages is more like 16 wm2. And it may well be more than that, for I see no reason for spreading this forcing around the whole northern hemisphere. It is sunny days that cause most melting on a glacier, rather than raised ambient temperatures. We were in shorts and singlets during sunny days, even at 18,000 ft. So I think that this 16 wm2 should be doubled to something more like 32 wm2, to take into account the local or regional aspect of this melting process. That is the true increase in forcing of Milankovitch cycles on northern ice-sheets over the whole Interglacial warming cycle - 30 to 40 wm2.
tatelyle.
You ask difficult questions but that is because you appear to seek answers that are simpler than the situation merits. Do also bear in mind that the full mix of forcings that result in ice age cycles is not yet fully determined which makes an informed answer a tad harder still.
Examine that graphic @479. At first glance, the last ice age shows a reasonable match between Antarctic temperature and July insolation for 65ºN. But look again. There is a good wobble-for-wobble match but there is also a ramping down of temperature over the ice age. This ramping shows there is something else at work other than NH insolation. And look at the graph for the ice age previous to the last one. The wobbles are a far poorer match but the ramping down remains a strong feature. Further, if the ice age cycles were solely about NH insolation, why didn't the insolation maximum 170ky bp, a maximum larger than that which saw the last ice age replaced by the holocene, why did it not have the power to create an inter-glacial warming?
Things get more interesting when we look at the Greenland temperature record. The graphic below is Figure 1 from EPICA (2006) which compares Antarctic temperatures with NGRIP from Greenland.
As well as insolation, we have ice albedo, CO2, CH4 and dust and added to that climatic effects. A quick back-of-the-fag-packet calculation suggests the annual energy flux arriving at 65ºN from insolation, something like 100Wm-2 ±7Wm-2, is about the same size as the net energy from the South warming latutides at 65ºN, mainly through the atmosphere.
Answers on how ice ages happen are not simple. So are you then still happy to assert "So although the Ice Age becomes a global phenomina, it is triggered and forced by northern hemisphere Milankovitch forcing and only by northern hemisphere forcing. That is an indisputable fact."? (My bold)
Rodger,
Thank you for your considered reply and the additional datasheet.
When I said the trigger for Interglacials was NH Milankovitch (and an indespitable fact), I think this is a correct statement. The forceing trigger is Milankovitch, but this does not exclude feedback assistance following the initial trigger.
I was trying to counter the previous statement by Tom that 'Causing ice sheets to melt is a feedback'. Sorry, that is wrong. There needs to be a trigger, before feedbacks can assist in that process. Tom was trying to undermine the value of NH Milankovitch forcing, and I think I have successfuly countered that argument.
As to the subsequent feedback enhancement, you name a few possibilities, but the question then becomes 'which is the primary feedback'? And I think the obvious answer is albedo. A reduction of albedo from ice (say 50%) to soil (say 10%) is highly significant, giving a local summer daytime absorption increase of about 180 wm2. That is significant, especially if we now know the primary forcing-driver is a regional NH event (melting ice sheets), which subsequently allows worldwide warming. Remembering that all net warming is tropical, which spreads out over the globe, with the higher latitudes being constant net losers of energy.
And if albedo is the primary feedback for Interglacial warming, then the reason for the lack of response at 170k bp is equally obvious. In this scenario, higher latitude NH Milankovitch warming requires an albedo response to be successful, but the major dust storms did not occur until 150k bp. So the brilliant white 170k bp ice sheets could not melt at any insolation value; while the 140k bp ice sheets were dirty, primed and ready to go.
Tate
[PS] Can you please reference your "obvious answer"? Hansen & Sato 2012 by contrast calculate albedo and GHG forcings to be approximately equal and match observed temperature well (Fig3). Furthermore it is not obvious how albedo can create a NH-SH teleconnection.
>>Can you please reference your "obvious answer"?
Seriously? Are we not allowed to think for ourselves?
Can you see a flaw in my logic or argument?
>> Hansen & Sato 2012 by contrast calculate albedo
>>and GHG forcings to be approximately equal.
Yes, but Hansen has made a very basic error in this paper, that should have been picked up in peer review. Why was it not?
Hansen claims an equivalence between albedo feedback and sea level, and this is the basis for his albedo feedback calculation. So Hansen has smeared the albedo feedbacks out across the whole globe, much as Tom tried to do in post 489. But as I demonstrated in posts 490 and 493, the Interglacial initiation or trigger is SOLELY a function of increased northern hemisphere (NH) Milankovitch insolation, and so the feedbacks that assist this initiation are likely to be NH feedbacks as well. The southern hemisphere (SH) has nothing to do with the Interglacial warming process, as the Ice Age record in post 479 clearly demonstrates. Increased insolation in the SH never triggers an Interglacial, only the NH can do that.
So Hansen's smearing of albedo feedbacks across the entire globe is incorrect, and we can therefore double his 'insolation equivalent' calculation from the outset. So Hansen's 4 wm2 figure for albedo feedbacks suddenly becomes 8 wm2. But as I pointed out previously, the albedo trigger effect is mainly a local ice sheet condition in the higher latitudes, rather than a hemispherical response, mainly caused by surface dust and dirt, so there is no point adding the entire NH into this initial calculation. Ambient temperature is not the key to ice-sheet melting, direct insolation onto the ice sheet is, as was patently obvious on the Himalayan glaciers. But if that is so, then we can restrict the albedo feedback to the ice sheets themselves, and therefore multiply the albedo forcing-feedback by 5 (the area above 55º N). And this results in a true albedo feedback-forcing of 40 wm2 (about half the Milankovitch forcing value, over the complete Interglacial warming). Which is exactly what I said in post 491, but we have approached this new calculation from a different direction - Hansen's erroneous paper.
>>Furthermore it is not obvious how albedo
>>can create a NH-SH teleconnection
Because of the hemispherical heat transport process, as in the following (simplistic) diagram, with tropical heat being exported to the poles. NH ice sheet melting will cause a large net increase in net energy absorption, not only on the NH ice sheet itself but also at the retreating margins. Each annual km of ice retreat at the terminus (that is about the rate of ice retreat), will have its albedo changed from about 60% to about 30% with sandy soils, and then to about 15% as plants reestablish themselves. I make that a 4 x 10(12) watts increase in absorption at 50ºN, each year. That is a large increase in the whole-earth energy radiation budget, that will be transported to the SH pole as easily as to the NH pole.
And once the world has begun to warm, there may then be some minor assistance to the process from CO2 and CH4. Correct me if I am wrong but my back of a fag-packet calculation for a 5,000 year ice retreat gives 4 x 10(12) w increase for albedo changes at 50ºN, annually, while CO2 influences averaged over 5,000 years are only 4 x 10(11) w. So CO2 on its own, without CH4 or H2O, is order of magnitude less in influence.
http://www.goes-r.gov/users/comet/tropical/textbook_2nd_edition/media/graphics/tropics_surplus_heat2.jpg
[PS] Seriously, yes. Handwaving away a paper that is doing the hard maths is little better than sloganeering. If you can fault Hansen and Sato, then publish it. Other commentators are giving you guidance on your logic. Since I am moderating, I will try to stay out of it.
tatelyle @493.
Can we then be clear then - Ice ages are triggered. Following this they are forced by feedbacks.
Concerning the trigger, the evidence suggests the trigger is NH insolation, in the late pleistocene this being synchronised to the 100ky orbital eccentricity cycle while in the early pleistocene this was synchronised to the 41ky orbital obliquity cycle. The trigger mechnanism is thus probably more a question of what primes the trigger than what the trigger happens to be.
Nothing about ice ages is straightforward. Even the events that began the ice age cycles at the end of the pliocene, events bound up with the closing of the Panama isthmus, events which caused both warming and cooling, even those events are not simple and obvious.
So to say "the trigger for Interglacials was NH Milankovitch ..., I think this is a correct statement." is not an end to the 'trigger' story. (Note you yourself @493 are speculating that dust was not present 170kybp preventing an interglacial occurring but was 150kybp for the creation of the Eemian.)
Then concerning feedbacks, once the ice age is in progress, the idea that feedbacks will drag the climate through the cycle is perhaps simple to grasp. But, while feedbacks can be listed out, the mix of feedbacks cannot be determined accurately enough to be certain of that mix. Thus your question @493 'which is the primary feedback'? is a bit presumptive, even as a genuine question. Why should there be a single primary feedback? All the evidence points to there being no single primary feedback. So the answer you yourself give to a misplaced rhetorical question "And I think the obvious answer is albedo."is not speculative. It is wrong.
Thus the moderator response pointing the need for reference (that you protest about @494) is no great surprise to me.
And the remainder of #494 reminds me so much of why I wrote the first paragraph @492. I would humbly suggest you revisit it.
tatelyle @493:
The brazen assertion that "Causing ice sheets to melt is a feedback" is wrong evidently needs some clarification. Specifically, you commented on the origian articles claim that:
The inclusion of the two sentences is necessary to establish context, specifically that the articles claim about feedbacks related not to icesheet melt directly, but to the warming of the climate, ie, the increase in Global Mean Surface Temperature.
With regard to the claim in the article, you asked the question:
Given the clear context of the claim you were discussing, you were asking for an explanation as to why feedbacks must be inferred to explain the "... warming of the climate", ie changes in GMST. At least, if you were not, you were instead indulging in the rhetorically odious, and vacuous tactic of bait and switch. But, just to reinforce it, you yourself clearly reference effects on climate. Once again, ice sheet melt is not a change in climate, still less the more precise change in GMST required by context.
Formally, A is a feedback on B if and only if:
1) A is not B;
2) Changes in B cause changes in A; and
3) Changes in A cause changes in B.
As such, it seems straightforward that melting of icesheets are a feedback on global GMST. Specically, melting of ice sheets are not GMST (clause 1); changes in GMST can cause icesheets to melt (clause 2); and melting of ice sheets changes GMST by changing albedo (clause 3). Ergo, the only way you can deny that the melting of ice sheets is a feedback in the context of this discussion is to deny that we are talking about GMST (ie, genuinely be indulging in the odious tactic of bait and switch), or deny that either the melting of ice sheets can effect GMST, or that changes in GMST can effect the melting of icesheets. Neither of the later two seems at all probable. The first (denying that melting icesheets effects GMST flies in the face of all the ice core evidence), while denying the second amounts to denying that the changes of state of water are controlled by temperature.
I require urgent clarrification of what claim exactly you are basing your denial that melting of ice sheets is a feedback. Are you employing bait and switch tactics (ie, discussing feedbacks on the melting of ice sheets rather than as required by context and your prior statements, feedbacks on changes in GMST) - or are you making extraordinary claims without evidence (ie, denying either of the two relevant cause/effect relationships).
I also request that other commentors not engage with Tate until he has clarified what is the exact basis of his absurd rejection that ice sheet melt is a feedback. Prima facie, and absent clear clarification such, engagement is futile in that he is merely using rhetoric to mask confused thinking rather than honestly engaging in the discussion.
If a gun is fired it doesn't mean the target will be hit.
Many things happened to make the gun fire and many things happened after the gun was fired.
This is complexity: now measure it!
>>The trigger is synchronised to the 100ky orbital eccentricity.
No it is not. Count the years. The trigger is synchronised in the later Ice Ages with increasing NH Milankovitch insolation, which is directly linked to the 25,700-year precessionay cycle. So the cycle length is either 102,000 years or 128,000 years, as you can see.
Big difference. Wrong cycle equalls wrong explanation.
>>'which is the primary feedback' is presumptive.
Not sure why. There are always greater and lesser feedbacks. But now we have identified the correct cycle that regulates Interglacials, we are in a better position to identify the primary feedback. And since Interglacials are SOLELY triggered by increased NH insolation events we can be pretty sure that CO2 is not the primary feedback. If Co2 was thenorimary feedback, Interglacials could also be triggered by SH warming. But they are not.
The exclusivity of NH insolation triggers strongly suggests that albedo is the key factor and the key feedback regulating Interglacials. And as we know the actual high latitude insolation and the actual ice sheet insolation increase is an order of magnitude greater that CO2 can manage.
Actual insolation increase 90 wm2 on the ice sheets (over whole Interglacial).
Actual albedo insolation increase 170 wm2 in northern latitudes (over the whole interglacial)
(Assumptions. NH insolation 460 wm2. Cloud albedo -80 wm2. Increase in albedo between ice and vegetation 50% or 170 wm2.)
So which is going to be the primary feedback? CO2's 4 wm2, or albedo's regional 170 wm2?
Tom Curtis @496.
I think you are asking for more than tatelyle is capable of providing. Consider his track record down this thread & make your own judgement. Below I collate the interchange that led to your comment @496. It is interesting how often tatelyle actually argues against what is his own misunderstanding. Even tatelyle's attack on Hansen & Sato (2012) leads him to his back of a fag-packet calculation @494 which looks to me as though he perhaps also mistakenly multiplied by the number of cigarettes in the packet, because with the model tatelyle tries to use the albedo effect seems to confirm Hansen & Sato. But, of course, that would never do because with tatelyle it is ABC - anything but carbon.
☻ TC@481 "(M)ilankovitch cycles can only significantly effect global temperatures if there are, not just feedbacks, but differences in th feedbacks that depend on location and season."
☻ tl@487 to TC "(You say ) Milankovitch cycles require feedbacks?
I understand all your arguments, that overall insolation cannot change much. However, the critical season and region for Ice Age modulation is the NH summer at northerly latitudes, because of the large NH landmass, as you say. It is this insolation that will decide whether the winter ice sheets melt, and so it is this region that will modulate Ice Ages."
☻ TC@489 to tl "Causing "winter ice sheets" to melt is a feedback. Further, it is a seaonal and regional feedback. If the seasonal and regional feedbacks in all areas and seasons were equally strong, then that feedback would be balanced by other, opposite feedbacks."
☻ tl@490 to TC "(you say) feedback would be balanced by opposite feedbacks because the changes in global annually averaged insolation balance out? But that is not true is it? - because as you said the NH and SH are very different, and so there is no 'balance' or equivalence between them.
So although the Ice Age becomes a global phenomina, it is triggered and forced by northern hemisphere Milankovitch forcing and only by northern hemisphere forcing. That is an indisputable fact."
☻ MAR@492 to tl "Answers on how ice ages happen are not simple. So are you then still happy to assert "So although the Ice Age becomes a global phenomina, it is triggered and forced by northern hemisphere Milankovitch forcing and only by northern hemisphere forcing. That is an indisputable fact."? (My bold)
☻ tl@493 to MAR "I was trying to counter the previous statement by Tom that 'Causing ice sheets to melt is a feedback'. Sorry, that is wrong. There needs to be a trigger, before feedbacks can assist in that process. Tom was trying to undermine the value of NH Milankovitch forcing, and I think I have successfuly countered that argument."
☻ TC@496 to tl "The brazen assertion that "Causing ice sheets to melt is a feedback" is wrong evidently needs some clarification."
tatelyle @498.
Thank you for your response to #495. It is always refreshing to meet new ideas.
I am assuming that you have not yourself "counted the years." For my own part, I have in the past been content to take the word of others for the recent ice age cycles being 100,000 years long, this being a well-known and unchallenged finding. Thus your contrary assertion would certainly tumble much scientific work, for example, Abe-Ouchi et al (2013) which is titled "Insolation-driven 100,000-year glacial cycles and hysteresis of ice-sheet volume." That is to say, your assertion would certainly tumble such work if your assertion were based on a shred of truth. So I did some measuring. I see no evidence for a 102ky/128ky cycle. My measurements are cursory but I feel they are adequate. They show cycles in orbital eccentricity to average 97ky(s.d. 7ky) and the ice age cycles to average 100ky(s.d. 10ky). I conclude that your bold assertion "equalls" wrong presumption.
I would be interested to learn why you assert "There are always greater and lesser feedbacks." Certainly, as a general rule systems are driven by simple mechanisms rather than complex ones but it is well known that 'general rules' do not apply "always".
Sadly, the remainder of your comment tumbles into incoherence. I am, for instance at a loss with what you mean by the phrases "key factor" and "key feeback" or what you mean by 'regulating interglacials.' You rely on a finding of your own calculation (presented less than confidently "correct me if I am wrong" @494 ) but you may have already noted @499 that I refute that "back of a fag-packet calculation"@494. Indeed (and here adding substance to the moderation Response@494), within the ice age dynamics it would be a surprise if any realistic calculation of Δ(NH ice albedo forcing) were greatly different to the Δ(global ice albedo forcing), which is the value presented by Hansen & Sato (2012): a surprise as all here do appear in agreement that these two quantities are effectively equivalent. The other 'back of a fag-packet calculation' you present (@498) is more of a mystery despite the list of assumptions provided. It is not at all clear by what means you convert peak 65ºN Milankovitch insolation figures for July into a figure for "over the whole interglacial" or indeed how you define that term "over the whole interglacial". I think these exemplars of the error & the lack of precision that you have managed to entwine together @498 adequately illustrate the level of incoherence @498.
To paraphrase my comment @492 - You create bizarre answers but that is because you appear set on answers that are simpler than the situation merits. Do also bear in mind that the full mix of forcings that result in ice age cycles is not yet fully determined which makes any informed response to you a tad harder still.