Wednesday
Oct232013
by 
Bishop Hill
Bishop Hill Fast freeze
Oct 23, 2013 Climate: Surface Isn't the Arctic refreeze quick this year? The extent figures are rapidly approaching the 2000s average, something that has only happened in the first half of recent years. Yet we're still in autumn.
Interesting times.
Reader Comments (118)
It is derived from the thermodynamic equations and tested under laboratory conditions. It correctly predicts the amount of downwelling IR in the field.
It is included in the models, of course, but it is an input parameter, not an output.
If you can find a copy of this, it might help. It's an old paper and I havent seen it online. Do you have library access?
Kiehl, J.T., and Ramanathan, V. 1983. CO2 radiative parameterization used in
climate models: Comparison with narrow band models and with laboratory
data. Journal of Geophysical Research, Vol. 88, pp. 5191–5202.
EM
That's not what I meant. This paper just describes the absorption and emission - the radition field. it doesn't show how that translates to real heating of a surface which is another matter. The forcing equation used by Hansen and others explicitly states a temperature change but uses the assumption that a net change in IR emitted causes the temperature increase.
Nevermind that it is an idealised situation, it needs to be tested. You cannot assume that just because the field exists then it must cause heating.
Micky H Corbett.
The basic lab test for back radiation was done back when I was a student. God knows where it was written up.
Stripped of the details:-
Take a long tube with IR-transparant ends. Fill with gas of various CO2 contents. Shine IR through one end. Use an IR spectrometer to measure spectrum and intensity of transmitted radiation and back radiation.
EM says "Milankovich orbital variations produce a variation of about 4W/m^2. at 65N latitude. This amplifies to 5C between glacial and interglacial, amplifid by a change of CO2 between 200ppm and 280ppm. "
So how do you get back to an ice age again? Based on your numbers, once the CO2 has risen, the CO2 forcing would be the same magnitude as Milankovitch, therefore you could not return to an ice age.
Also, please explain how CO2 amplifies the temperature change, and yet physically lags temperature by 800 years at all times in the ice core data?
Thinkingscientist
When you pass the peak of an interglacial insolation begins to drop.
Temperature drops a little.
The temperature drop allows more CO2 to dissolve in the oceans, causing the CO2 concentration to drop.
The reduced CO2 concentration reduces forcing and the temperature drops a little more.
More CO2 is taken up by the oceanic carbon sink and temperature drops a little further.
The feedback continues until the temperature has fallen enough for extensive ice sheets to reform and increased albedo then causes a further temperature drop into glacial period conditions.
Regarding the supposed lag between temperature and CO2 , I see no problem. These are processes involving equilibria gradually shifting over tens of thousands of years. An 800 year lag is small by comparison. Remember that you are changing the temperature and CO2 content of 1.3 billion cubic kilometres of seawater.
There are also lines of evidence that the 800 year lag is an artefact of the way snow converts to firn and then permanent ice. Gas bubbles migrate upwards as the firn hardens and softens from year to year.This mean that the gas trapped at a particular year is from an earlier date than the ice around it. Hence the apparant lag.
This links to an old and invalid sceptic argument against AGW. The problem is that glacial cycles and modern conditions are not comparable.
The Holocene is the product of a long term insolation driven temperature warming. This was amplified by a feedback cycle in which increasing temperature drove increasing CO2 concentration; which amplified the temperature change. The peak was some 8000 years ago. The last two millennia have shown a gradual cooling trend as Northern Hemisphere insolation started to reduce.
The modern warming is a case of a long term cooling trend being reversed by an industrially induced increase in CO2, without a preceding temperature increase. Now CO2 increase is driving increased temperature. The same feedback amplification is occuring as in the early Holocene, but the trigger is different.
Entropic Man:
You say, "Milankovich orbital variations produce a variation of about 4W/m^2."
Sorry, not even close. Even at today's low eccentricity for the earth's orbit of 0.0167, the difference between the solar constant at perihelion and aphelion is 90 W/m2. At 65 latitude at the solstice, the trig gives an average of about a third of this, so 30 W/m2. Even with a third of this just reflected from clouds, that's 20 W/m2.
In the Eemian, the eccentricity was 0.043, and the difference in the solar constant between perihelion and aphelion was 230 W/m2, so averaging 77 W/m2 by trig and 50 W/m2 including cloud albedo.
But what's an order of magnitude between friends?
Curt
Link, please.
Curt
Both the Holocene and the Eemian are interglacial periods. The Eemian ended 114,000 years ago and preceded the last glacial period, the Winsconsin.
If you've done these figures yourself, please calculate an equivalent figure for a glacial period. Try the peak of the Winsconsin glaciation, 50,000 years ago.
Em
I'm not questioning that. Co2 experiments show that you get isotopic emission. That's fine. The issue is where have experiments been done where similar IR radiation heats a surface. Not a surface inside the same cylinder but able to transfer energy by convection and surface conduction like a container of water.
The only ones I've seen are for higher power densities which isn't really the same situation.
I'm not disagreeing with the emission. What I'm saying is what happens next and how has this been tested?
Do you understand what i mean?
EM: Here's a nice graph of the variation in the eccentricity of the earth's orbit, with references:
http://www.museum.state.il.us/exhibits/ice_ages/eccentricity_graph.html
Lots of sources for the present eccentricity, e.g.:
http://nssdc.gsfc.nasa.gov/planetary/factsheet/earthfact.html
At the present eccentricity, at the perihelion, solar irradiance (to a plane at the earth's distance from the sun) is:
(1 + 0.0167)^2 * 1365 = 1410 W/m2
At the aphelion, the solar irradiance is:
(1 - 0.0167)^2 * 1365 = 1319 W/m2
That's a difference of 91 W/m2. So compared to the present, when the northern summer solstice is near the aphelion, half a precession cycle ago, about 11,000 years ago, the solar irradiance at the northern summer solstice was over 91 W/m2 greater than now. This actually understates the difference, because the eccentricity was somewhat greater then. But the key point is that there was far more northern hemisphere spring/summer insolation when we were coming out of the last glacial period than there is now.
Now for the geometry: At 65N, the daily surface insolation varies from a noon peak of about 70% of this value (cos[65-23]=0.74) to a midnight low of 0% as the sun barely dips below the horizon. Let's call it 33% on average for 30 W/m2. If one third of this is reflected back into space, that's 20 W/m2 average greater surface insolation then than now. This dwarfs the 1.7 W/m2 kicker fro CO2 now (both values before feedbacks).
Comparable calculations for the Eemian:
(1 + 0.043)^2 * 1365 = 1484 W/m2 at perihelion
(1 - 0.043)^2 * 1365 = 1254 W/m2 at aphelion
So there is a 230 W/m2 difference in summer-solstice solar constant over half a precession cycle. By the same geometric calculations, that's about 50 W/m2 average difference in surface insolation at 65N at the summer solstice.
You asked about 50,000 years ago. Eccentricity then was slightly less than now - about 0.014. This yields a 76 W/m2 difference between aphelion and perihelion, so about a 16 W/m2 difference at 65N surface insolation over half a precession cycle. And this was not enough to bring the earth out of a glacial period.
Micky H Corbett
I'm not sure where isotopic emission came from? The interaction of radaition with surfaces has been under invesastigation for the better part of a century, to the point where most of it has long settled into te toolkit and its origins have been forgotten.
There's a standard sceptic trope, often requoted, that its all models. There is solid measurable science behind all this, taught to every undergraduate physics student.
Curt
May I remind you of the chronology.
The Eemian interglacial began around 130,000BP and lasted to 114,000BP.
The Wisconsin glacial period followed the Eemian and continued until 12,500BP.
The Holocene has continued since.
The periods in which we are interested are thw two transition periods as the Eemian cooled into the Wisconsin and the Wisconsin warmed into the Holocene.
Looking at the CO2 record, the transitions produce a rapid response.
http://en.wikipedia.org/wiki/File:Co2_glacial_cycles_800k.png
It would be interesting to be able to compare these with parallel insolation and temperature data. I note particularly the rapid increase in CO2 at the start of the three most recent interglacials, as if a step change of some kind took place. The most interesting question is whether there were similar steps in temperature and insolation, or whether a tipping point occured only in the behaviour of CO2.
@ Doug McNeal
thanks for your post. It is important to keep everything in perspective. Too early to call. Needs another 15 years or so. Just a pity that the alarmists called it too early"
Is 44,015 years a long enough run for Dolphinlegs ?
Unprecedented recent summer warmth in Arctic Canada
Gifford H. Miller, Scott J. Lehman, Kurt A. Refsnider, John R. Southon, Yafang Zhong
DOI: 10.1002/2013GL057188
Geophysical Research Letters 2013
Abstract
Arctic air temperatures have increased in recent decades, along with documented reductions in sea ice, glacier size, and snowcover. However, the extent to which recent Arctic warming has been anomalous with respect to long-term natural climate variability remains uncertain. Here we use 145 radiocarbon dates on rooted tundra plants revealed by receding cold-based ice caps in the Eastern Canadian Arctic to show that 5000 years of regional summertime cooling has been reversed, with average summer temperatures of the last ~100 years now higher than during any century in more than 44,000 years, including peak warmth of the early Holocene when high latitude summer insolation was 9% greater than present. Reconstructed changes in snow line elevation suggest that summers cooled ~2.7 °C over the past 5000 years, approximately twice the response predicted by CMIP5 climate models. Our results indicate that anthropogenic increases in greenhouse gases have led to unprecedented regional warmth.
EM - I am well aware of the chronology, and I don't know why you think I wouldn't be. I am simply pointing out how much greater the insolation changes are (at 65N at the solstice, the usual metric for Milankovitch effects) over a precession cycle in these periods than the 4 W/m2 you asserted.
In the case of the Eemian, you look to be to low by over an order of magnitude, so you would get far bigger swings across a precession cycle then that you assert (and the Eemian looks to have lasted only half a precession cycle).
Any follow-on effects from changes in CO2 are of a much smaller magnitude.
Em
Isotropic emission means that when you absorb from one direction you emit in all directions with equal probability. It's what every atomic or molecular emission does. You are correct in saying that interactions with surfaces have been studied; however they continue to be studied i.e. laser interactions, because we are finding out new things.
But you seem to be arguing that it is all settled...and what I am saying is okay show me the characterisation curve. Not the theory as you have done, not the radiation field. Where is the curve of the exact effect, or how the temperature varies per material for varying power densities. And as I've been looking for this for some time and as the Met Office is continuing to try and look for it in the observations, it doesn't exist. Otherwise it would be front and centre. Instead we get the logarithmic curve.
What we need is a something like what you read in Kaye and Laby for the evaporation temperature of xenon under certain conditions. Or like a sputter yield curve for carbon using Argon ions.
A precisely measured curve for temperature change due to IR "forcing" of a surface in an atmosphere, going from power densities of 100s W/m2 down to 10s and less (these may not be measurable though). You control the environment and wait for temperatures to stabilise. Repeat. What you will get are is the forcing/temp relationship and the losses. You can then make a reasonable guess at a relationship (it may be the log curve) with errors. Then we can see if all of the predicted 4W/m2 would actually get into the surface and cause heating. it might or it might be that some gets in, maybe 60%.. Or maybe there is a threshold energy of 10W/m2 as it is not a surface in a vacuum and so there is no effect at all. This type of thing is quite common in interface effects. The data will guide this process.
You would get all this from experiment. Then we can see what is going on and make a reasonable prediction of what will happen. But until then this is real doubt. Which is why I am sceptical about it.
I think anyone reading this can understand what I mean.
Curt, Micky H Corbett.
This is getting too technical for me! Goodnight.
EM: Bottom line, you can arm wave all you like, but if CO2 lags temperature at all measured time scales, and given the forcing figures given above, there is NO ROLE for CO2 in ice ages andn inter-glacials. NONE, Nada, nothing. Like RealClimate, the mechnism you propose is physically impossible and clearly shown to be so by a back of the envelope calculation.
You state "The temperature drop allows more CO2 to dissolve in the oceans, causing the CO2 concentration to drop. The reduced CO2 concentration reduces forcing and the temperature drops a little more. More CO2 is taken up by the oceanic carbon sink and temperature drops a little further."
In other words, temperature drives CO2, not the other way around. And yet you still claim that CO2 causes temperature to change. Your model is physically impossible. You cannot claim to have a physical process driving something by cause and effect when your "cause" clearly lags the effect by 800 years. Its also clearly too weak a process, and if we based the numbers on yours, once the warming increases ("accelerated" by your CO2 effect) and gets to a certain level, in your model, we could not possibly get back to an ice age, because the CO2 would result in an increased forcing that exceeded your figures for Milankovitch cyclical warming and cooling. Your (meaning others generally, such as RC, don't take it personally) model is physically impossible. And the figures given by others show that the orbital forcing is way larger than any putative CO2 effect. The model that CO2 "amplifies" is quite frankly bollocks.
Why do people believe this stuff? Its not like the principles are hard physics.
The final point to make, EM, is that for ice ages to work, you don't need CO2 at all. In fact it makes the model worse, not better. And you have gone to great lengths to justify the inclusion of CO2 effects, but have effectively argued that the drop in temperature causes the <CO2 to drop. If CO2 forces temperature, your description of the cause and effect is backwards.
And CO2 lags (follows) temperature in ALL observations at ALL time scales over which it has been measured. Not just ice core, also in modern satellite data - see Bart's woodfortrees graph for further evidence.
Your statement "There are also lines of evidence that the 800 year lag is an artefact of the way snow converts to firn and then permanent ice. Gas bubbles migrate upwards as the firn hardens and softens from year to year.This mean that the gas trapped at a particular year is from an earlier date than the ice around it. Hence the apparant lag." I am using the age corrected gas data from Vostok ice core, which corrects for this effect. The age correction does not explain the 800 year lag that still exists when you run a cross-correlation between the CO2 (using the published gas age) and the temperature proxy.
Its funny how CO2 adicted scientists always find a way to bring in CO2 effects that are only apparent after they have made some tenuous correction to the data. Even modern temperatures: the 1990's are not as hot as the 1930's until after the data are adjusted. And I keep asking why the corrections are a systematic straight line making older temperature measurments colder and modern ones hotter. If the corrections are systematic, what is the explanation for the corrections themselves? But I digress...