In this essay I return to the SCM paper that I discussed last week, considering this time what it says about atmospheric carbon dioxide, which for the orthodox is the ‘control knob’ of the planet’s atmosphere. As is well known, the actual amount of the gas in the atmosphere is very small indeed, about 400 parts per million, but is supposed to do its work through the greenhouse effect, by which radiatively active gases, of which CO2 is the most prevalent, radiate energy in all directions, including the surface, thereby warming it. In the orthodox view, again, as CO2 increases so does warming increase, multiplied by ‘climate sensitivity‘, the supposed positive feedback from water vapour and clouds (more clouds, therefore more warming still).
The graph of carbon dioxide concentrations in the atmosphere, that measured at the Mauna Loa observatory on the island of Hawai’i, is well known.
The red line shows the seasonal variations in CO2, while the black shows the average. There is little doubt that the level is rising, at least at Mauna Loa. In summers the plants get as much CO2 as they can, while in winters they are dormant and use little; these conditions explain the jagged line. Most of the jagged effect comes from what happens in the northern hemisphere, where most of the land mass, and the plants, are situated.
So far so good. The SCM paper sets out clearly and simply how and where carbon dioxide is measured, and it takes a dim view of the reliance of climate scientists on the data so gathered. Carbon dioxide, it argues, is part of the carbon cycle, with carbon atoms from CO2 being transferred to various ‘reservoirs‘. There is a constant flow of exchange between reservoirs. During these transfers, reservoirs that release CO2 into the atmosphere are called ‘sources‘, and those that consume it are called ‘sinks‘. The sea, for example, is both a source and a sink. The carbon cycle makes it difficult to determine the life span of CO2 in the atmosphere, which is why scientists focus more on its concentration.
The composition of the atmosphere at any place and at any time depends on pressure, temperature and the extent to which other gases are diluted in water vapour. To obtain a measurement that is not dependent on these parameters, you have to measure the number of CO2 molecules in one million molecules of dry air. This measurement is expressed in ppm (parts per million)…
Although Mauna Loa is accepted as the standard for atmospheric measurements of CO2 (mostly because the observatory is at 3394 metres and is thought to be relatively free of human or other biological influences, and measurements are taken only from readings at night-time, when the flow of air is from high to low) there are other stations and, says the paper, they provide different outcomes. There is much greater variation at Barrow in Alaska than at Mauna Loa, while the variation at the South Pole is very small. Nonetheless, as I see it, the fact that the trend lines at these four sites are the same suggests not only that the general trend is accurately measured, but that CO2 is also a reasonably well-mixed gas.
The SCM paper is less sanguine. It points out that measurements are taken not only at baseline observatories like Mauna Loa, but at tall towers, and at the surface. It would like much more data, and has tables showing that there is variation at all of these sites. It argues that the distribution of sensors is poor, with most being in the USA or Europe (true), while the data are not only insufficient, but they have been interpolated (and the raw data are not available). I’ll come to my reactions at the end.
Then the paper considers the work of Ernst-Georg Beck, in which samples of air are subject to chemical analysis in laboratories. These measurements are taken mostly in rural areas or on the outskirts of towns and cities. This work has been going on for almost two centuries, and its results show much greater variability than, for example, those from Mauna Loa. But the paper doesn’t like them either, for reasons of location, though it notes that the data collected by this method do not make it possible to conclude that CO2 concentrations are rising.
Nor is the paper happy with ice-core measurements, for reasons that will be familiar to anyone who has studied this issue. Briefly, the gas can dissolve, while the deeper the crystals the more compacted they are, which greatly widens the error bars around timing. Nonetheless, the rising and falling trends associated with ice ages are clear enough. The paper is equally unhappy with analyses using the stoma of fossilised plants, because the errors are around 60 ppm.
All in all, the paper is deeply critical of what is done. For my part the criticisms seem well-founded, but the conclusions drawn from them are over-stated. What do we want to know? That is the first question, to which the answer probably centres on whether or not the concentration of the gas is rising, and if so, by how much. I think that the conclusion we can draw from the data is that the proportion of carbon dioxide in the atmosphere has been rising steadily, at least since the Mauna Loa and companion stations have been established, which is 1960.
The second question is probably whether or not the increase is or ought to be a worry, and the paper makes an off-hand remark to the effect that if increased CO2 leads to warming, then France could do with a good deal of it. My own interest here is in the flow of the gas between sources and sinks, but the paper doesn’t get into that game of estimation at all, though I would have thought it something that ought to have been considered. I’ll leave that to another post on another day, with just the comment that on the evidence I have seen, the human contribution to that flow is likely to be small.
Finally, the paper does not refer to, and could not have referred to, the NASA orbiting CO2 observatory, which has very recently produced this image.
Most of the southern hemisphere is not shown, which is a pity, and the colouring is of course a form of (in this case tendentious) artistic licence. The range shown is about 10ppm, or about 2.5 per cent, which is further evidence to support the view that CO2 is indeed well-mixed. I don’t think that too much should be made of the fact that Australia shows a low CO2 score, as does much of Russia.
Once again, I appreciated the assembly of methodology that the paper was able to produce with respect to measuring carbon dioxide. But, in contrast to my reaction to the CSM paper’s approach to measuring temperature, I felt that its ‘critical analysis’ was a bit OTT, finding fault but making too much of the consequences of the faults. In a later post I’ll look at the CSM paper’s analysis of the measurement of sea-levels, a subject dear to my heart.
Footnote: In truth, we simply don’t know a great deal about the sources and sinks of CO2, and the flow among them. Everything, reasonably enough, is based on estimates. Tom Quirk has a fascinating piece about one aspect of it here.
Three or more years ago I came upon an internet site showing the findings of 37 different studies as to how long CO2 lasted in the atmosphere. The period was from 1957 to (I think) 2003, and the results varied from a few months to 20-plus years, the one exception being the IPCC finding of 100+years. Some of the research teams had made several findings, some were one-offs. The aggregate of all these was about 7 years.
I’ve never been able to find the site again, and yet I know it was no chimaera. But a good, clear post, Don. My only addition to it would be what my Icelandic Chemistry-wiz mate, Einar, tells me about the heat-retention capability of the CO2 molecule, whuch is that it is a poor retainer of heat, which I assume means that it is an inefficient building block if you are constructing a greenhouse around our globe.
John Tyndell, the 19th century physicist, who discovered the infrared absorption properties of gases noted CO2 as by far the weakest absorber of all the infrared absorbing gasses he discovered. The strongest he found was water vapour – some 40X to 50X CO2!
The 1000+ years CO2 remains in the atmosphere advocated by Tim Flannery around 2007 is nonsense, the IPCC says 200 years, but I have heard a report saying it is 9 years( or lower).
Like equilibrium climate sensitivity, no one really know; so much for the “settled science” mantra.
The key tenets of climate science are settled, in particular that global warming is currently occurring, and that cause of that global warming is anthropogenic GHGs.
There are still plenty of open questions in climate science.
How do you square a short net persistence of CO2 in the atmosphere with the trend line in the Mauna Loa concentrations? In particular what do you think that implies about the rate at which emissions are going into the atmosphere? I’ll answer for you: if the persistence is 9 years, and there is roughly x additional concentration units added per year, then the Mauna Loa graph should be flat, albeit at 9x units higher than the natural baseline.
The implications: If the slope is increasing roughly linearly beyond the 9 years then the amount of CO2 going into the climate system is increasing exponentially.
Perhaps you could explain how a trend plotted on a linear scale can demonstrate an exponential increase.
Actually I wasn’t quite correct – to maintain linearity, for the first few cycles the amount added per unit time outgrows exponential growth but then exponential growth overtakes. This is all very approximate using discrete maths not calculus. If decay rates are calculated and so on the results will probably look quite different.
Beginning of first cycle: y added. By the beginning of next cycle y has been removed; to have linear growth 3y needs to have been added. By beginning of next cycle 3y has been removed, so 6y needs to be added to keep increase linear, etc etc. The cumulative totals added are y, 4y, 10y, 20y, 35y, 56y, 84y, 120y, 165y
Why do you claim that the amount of CO2 removed from the atmosphere increases exponentially?
Neither your reasoning nor your mathematics make any sense.
If you assume that any “pulse” of CO2 added to the atmosphere is completely removed after a cycle of say 9 years, as commenter JMO claims, and you have a linear increase in CO2, and if you simplify everything to discrete calculations where you assume that all the CO2 added or removed happens as a pulse at the start of the cycle, then the calculation I did seems sound. If it’s not could you please tell me what is wrong?
I will explain again: if at the zeroth cycle the amount of CO2 is normalised to zero (because the baseline doesn’t matter), at the start of the first cycle the amount is y, then it will be 2y at the start of the second cycle, 3y at the start of the third etc.
But we are making the assumption that if amount ny is added at the start of some cycle, where n is a natural number, then ny will be removed by the start of the next cycle.
Hence if we have y at the start of cycle 1, 2y at the start if the second cycle and y was also removed, there must be 2y + y = 3y added at the start of the second cycle. By the start of the next cycle 3y must be removed, result must be 3y so 6y needs to be added etc etc. The amounts added are: y, y +3y =4y, y+3y+6y = 10y, … .
ie we have added at each cycle y, 4y, 10y, 20y, 35y, 56y, 84y, 120y, 165y,… .
At first this grows faster than exponential because in exponential growth, doubling time is always constant. Doubling occurs from 10y to 20 y, so for exponential growth you’d have 2.5y, 5y, 10y, 20y, 40y, 80y, … (i.e. it doesn’t even go back to y), so prior to the doubling from 10y to 20y the growth in our simple system is faster than exponential , after 20y it is slower.
This is just a simple explanation that ignores differential equations; it’s not really instructive to do DE calculations here because they can be quite messy. It’s meant to be heuristic more than anything.
Since you are adding CO2 at an exponential rate, you cannot have it being removed at a linear rate.
Well, if you are adding CO2 at an exponential rate, what does that imply about the rate of removal if all CO2 added is removed after 9 years?
Care to pinpoint the problem with the calculation?
“we are making the assumption that if amount ny is added at the start of some cycle, where n is a natural number, then ny will be removed by the start of the next cycle”
Why?
So my calculation is in response to the original claim made by JMO who says that any CO2 added to the atmosphere (ny) is removed by the beginning of the next cycle (i.e. in 9 years or less as JMO claims).
So what IS your opinion of the David Evans analysis and model???? Or have you not looked at it!!
I haven’t checked it out, actually it would be handy if you could provide a link/s so I can take a look when I have a bit of time.
Btw the simple calculation I did in this thread is really just a way of pointing out that if we assume roughly linear growth in CO2 concentration AND that ALL CO2 added to the atmosphere is removed in some fixed time period (such as 9 years) regardless of the amount of CO2 added, these assumptions imply something unpalatable: that the growth in CO2 concentration added to the atmosphere to maintain linear growth is far beyond what is happening in the real world.
The incorrect assumption is that all CO2 added to the atnosphere is removed in a fixed time period. A more realistic assumption would be that CO2 is removed from the atmosphere at a certain rate that depends on a number of variables.
file:///C:/Users/Don/Downloads/synopsis-of-basic-climate-models.pdf
I hope to read your comments
That’s your local file address, do you have a URL?
Your best bet is to go to JoNova.com.au and scroll down to it. A lot if interesting stuff going on there you might enjoy
Don, the project home page is
http://sciencespeak.com/climate-basic.html
Read the Summary there, download the Summary document or Synopsis document, links to blog posts, etc.
Incidentaly this is just a synopsis of 19 posts with all the detailed arguments and mathematics to support them. If you are able to cope with the complex maths you should read all 19 posts. It was a bit above me I’m afraid so the synopsis was appreciated. Cheers
Hi Bobo
Just to clarify. The time how long C02 stays in the atmosphere can be equated to a half-life; ie if the 9 years is correct (which I do not advocate or claim) then after 9 years about 1/2 of the original CO2 molecules emitted in the atmosphere 9 years ago would be absorbed out of the atmosphere. Would this affect your calculations?
Sure it would. I can have another crack but not for a few days, I do have a bit on my plate at present, I’ll do the calculation on the weekend.
Hi JMO,
I’ve done some calculations assuming the rate of removal of CO2 occurs with some fixed half-life.
Suppose that at time s, the rate of CO2 emission is E(s). The infinitesimal amount of CO2 added at time s is E(s)ds. At some later time t, the amount of this infinitesimal “pulse” of CO2 remaining in the atmosphere is
where t_{1/2} is the half-life. The total amount of CO2 present at time t is
If E(s) is constant, say 36 billion tonnes/year (the 2013 figure) and there is a half-life of 9 years, we get
where alpha is E(s); the amount of CO2 in the atmosphere then looks like:
If on the other hand the amount of CO2 with time is always observed to be linear, then the integrand must be constant, but the denominator of the integrand is a function of t, so it must also be constant (which implies that E(s) must be constant too). The only way this can occur is if the half life t_1/2 is effectively so large that it is essentially infinite.
Ok, images haven’t been placed properly but they are in order:
first equation is infinitesimal amount of CO2 added;
second equation is total amount of CO2 at time t;
third equation is the solution to the second equation if the rate of emission, E(s) is some constant alpha;
the graph describes the total amount of CO2 with time in years (ignore negative years) and the y axis is the amount of CO2 in tonnes if we assume constant rate of emission of 36 billion tonnes per year.
To clarify again:
1st equation is the amount remaining of the infinitesimal pulse of CO2 E(s)ds that was added at time s at some later time t
Graph assumes the half-life of CO2 is 9 years.
The amount of CO2 removed at each cycle in my calculation is: At cycle 1: 0. cycle 2: y. Cycle 3: 3y. Cycle 4: 6y. Cycle 5: 10y. ….
So in my calculation the removal isn’t occurring at a linear rate.
“The key tenets of climate science are settled, in particular that global warming is currently occurring, and that cause of that global warming is anthropogenic GHGs.”
Now that is quite an extraordinary statement and will require extraordinary evidence for anyone of sceptical disposition to accept. Correct me if I am wrong but for the last 18 years the patchy surface observations show limited warming while the satellite observations of the lower atmosphere show no warming at all. This is with ever increasing emissions of CO2 as you say.
To address your global warming question, it appears you aren’t clear what global warming actually is. It’s ok, most people don’t seem to know.
It simply means that there is more radiative energy flowing into the climate system from space (i.e. the sun) than is escaping.
Where does this accumulating energy go? Into the hydrosphere, atmosphere, land, cryosphere and biosphere.
In order to determine whether global warming is occurring, that is, whether there is a net flow of energy into the system, you need to add up the changes of energy in each of these components. That quantity is well approximated by just adding up the changes in thermal energy in the components, and you get the following graph:
http://www.easterbrook.ca/steve/wp-content/IPCC-AR5-WG1-Box-3.1-Fig-1.png
Global average surface air temperature is an estimation of the temperature of a layer of air about 50cm thick about 1.2m above the surface. It provides an estimation of how much thermal energy is in that thin layer of air wrapping around the earth. But it only tells a small part of the story: it ignores the thermal energy in the remainder of the atmosphere, and the oceans, land, ice.
As you remarked in another comment on this page, you seem to understand that thermal energy accumulating in the climate system does not guarantee accumulating energy over particular short time-frames in this global surface layer, because the thermal energy is not well-mixed over short time frames.
I explained why global warming is occurring in the previous comment and pasted a diagram of the amount of thermal energy accumulating in the climate system. This is roughly equal to the amount of global warming occurring.
So what is the cause of this?
The smoking gun suggesting that GHGs are to blame is upper stratosphere and mesosphere cooling.
In particular, this cooling rules out increasing solar irradiance as the cause, because increasing irradiance would cause warming of these layers.
SO how do GHGs cause cooling of these layers?
The increasing opacity of the troposphere to CO2 spectrum thermal IR photons (caused by the increasing scattering of outgoing photons by the additional CO2 in the atmosphere) leads to a reduction of these thermal IR photons reaching the upper stratosphere from underneath. But the concentration of CO2 in the upper stratosphere and mesosphere are also higher as a result of increasing CO2 emissions, and this means that these layers have increased their emissivity to CO2 spectral photons. Due to the higher emissivity, existing CO2 spectral photons are lost to space more quickly. But there is less replenishment of CO2 spectral IR from the troposphere/lower stratosphere, so the amount of thermal IR photons in the upper stratosphere/mesosphere decreases. The consequence of this is cooling.
The fact that CO2 is a “poor retainer of heat” is completely irrelevant to the fact that it is an active greenhouse gas.
CO2 molecules are strong absorbers of thermal infra red at certain frequencies and strong emitters too. An energised CO2 molecule – one that has just absorbed a photon – only stays in that state for a very small fraction of a second before reemitting the photon in some random direction.
The higher the concentration of CO2 molecules in the atmosphere the more often a CO2 spectral photon is scattered, and the longer it takes to escape the atmosphere into space.
This increasing delay causes the output of radiation from the atmosphere into space to become less than the radiative input into the climate system from the sun.
Eventually a new thermal equilibrium will be reached, where the radiative output matches the radiative input, but in order for that to happen, it is a physical certainty that there will have to be a higher amount of thermal energy in the climate system, which results in higher temperatures.
Yes, more thermal energy in the climate system, but what happens next is the unknown. More energy will mean enhanced convection which inturn will produce more wind and evaporative cooling, enhanced conduction of heat away from the surface and probably more cloud. Have they been able to model such phenomena accurately?
The earth system has been around a long time, I would be highly surprised if negative feedbacks didn’t dominate such a system.
Increased thermal energy (i.e. heat in) divided by heat capacity = increased temperature.
The term climate change applies to certain changes in convection.
There’s still a lot that is not well understood about the convection consequences of global warming, but over time, kinetic energy of convection degrades to thermal energy.
As for your negative feedback remark, if you refer to the following graph that shows the accumulation of thermal energy in the climate system, you will see that negative feedbacks are not yet dominating, otherwise thermal energy would not be accumulating:
http://www.easterbrook.ca/steve/wp-content/IPCC-AR5-WG1-Box-3.1-Fig-1.png
Agree highly likely negative feed backs will “dominate” a closed system. But highly unlikely they will have any appreciable effect in time span relevant to humans.
And AGW is a known, feebacks are unknown!
“This increasing delay causes the output of radiation from the atmosphere into space to become less than the radiative input into the climate system from the sun”.
According to NOAA web site, there has been a net increase (not a reduction) in amount of outward long-wave radiation leaving the top of the atmosphere across equatorial areas over the past 10 years.
That’s interesting, but you really need to be looking at the radiative output of the entire outer atmosphere into space, not just a localised region.
Bobo you should go visit jonova and read David Evans analysis of this. Hard work but illuminating if you can cope.
Are you referring to the notch delay theory of his? I had a brief look, will have a more in-depth look at some stage, already though I spotted a key issue – he hypothesises that global warming is currently being caused by solar irradiance, but this is wrong because upper layers of the atmosphere are actually cooling, not warming as would be expected if increasing insolation was the cause.
bobo, my work clearly says (multiple times) that the warming is *not* due to solar irradiance, and points out that it *cannot* be due to solar irradiance. For example, see the section under the headline “It’s Not Variations in Direct Heating by the Sun” at http://joannenova.com.au/2015/11/new-science-20-its-not-co2-so-what-is-the-main-cause-of-global-warming/.
The notch-delay theory is based on evidence that the TSI signals, but does not cause, the warming. The warming force is delayed by one sunspot cycle (half a solar cycle of ~22 years), following the TSI.
See http://sciencespeak.com/climate-nd-solar.html.
At least you are aware of bozo now, and it seems he is well worth ignoring!! Looking forward to your next missive. Cheers
Hi David, many thanks for your comment.
At Don Amoore’s urging I’ve actually spent a few hours reading through your notch delay theory on Jo’s website in the last day, and I posted a comment with a few thoughts on your theory in the comments section at
http://donaitkin.com/the-puzzle-of-energy-policy/
If you have time I would be greatly appreciative if you could read it and address any serious misunderstandings.
Presumably in your comment you are referring to what you’ve called “Force X” in Jo’s pages? Why, if Force X has warming effect, is unmodulated cooling observed in the upper atmosphere? Exactly what sort of matter do you claim interacts with Force X? If Force X causes cloud albedo modulation as you surmise might be the case, why isn’t there any 1/(11 year) signal in the surface temp spectrum? If there are albedo peaks, shouldn’t there be correlated cooling in surface temps? Is there any modulation in global rainfall figures?
What you seem to be saying is that changes in filtered TSI are too small a forcing to be observed in surface temp data, so essentially you’re using TSI as a proxy for solar magnetic field. Some confusing things are going on here:
1) You claim that TSI 11 year variation would be strong enough to cause peaks in surface temp if Force X cooling wasn’t coinciding. Therefore Force X cooling must exactly cancel out 11 year warming cycles in TSI. Since Force X interacts with matter differently to insolation, it is remarkable that the sum of the energy absorption resulting from these two different forcings kills off any 11 year variations in surface temp. In other words, there seems to be a remarkable coincidence: if instead Force X had slightly different interaction with matter or had a slightly different amplitude, then one of the Force X or TSI 11 year peaks would dominate, and presumably there would be some 1/(11 year) signal in the surface temp spectrum.
2) in your model you are filtering TSI but not filtering magnetic field. But changes in filtered TSI are not strong enough to entrain the surface temp even though the 11 year time translation of filtered TSI is correlated with surface temp. But the peaks in solar magnetic field seem correlated with the next (or prior) TSI 11 year cycle peak. So essentially you are saying that the surface temp is correlated to filtered (1/(11 year), >1/(5 year)), translated (by 11/2 years or 11/2 years plus some 11 year integer multiple) solar magnetic field amplitude. If so, is the 11 year TSI cycle “cancelling out” the 1/(11 year) spectral peak for solar magnetic field amplitude?
3)If the ~1 W/m^2 variation in the TSI 11 year cycle is strong enough (in the absence of Force X) to cause variations in surface temp, why isn’t the filtered TSI strong enough when the change in filtered TSI forcing over the last few decades should be a significant proportion of 1 W/m^2?
bobo,
We are rolling out a revamped version of the notch-delay
solar theory on Joanne’s blog at the moment (resuming after Paris ends). It is de-mathified, relying on observable scientific phenomena and reasoning almost
entirely. Rather than answer your questions by rewriting large slabs of the upcoming posts (and the two that are already up) in comments on Don Aitken’s blog, why don’t you wait and see what arrives on Joanne’s blog — looking through your comment I’m confident that will all be answered and a lot more in the upcoming series (“new science” posts 22 to 28). The
project home page is
http://sciencespeak.com/climate-nd-solar.html
Btw, the solar relationship is between the TSI and surface
temperature — not the upper atmosphere.
If you are interested in how global warming works and what is going on in the upper atmosphere, I recommend you look at the other (separate) thing going on: see
http://sciencespeak.com/climate-basic.html
Basically we discovered the poor assumption responsible for the whole AGW imbroglio. The models, from 1896, have overestimated the sensitivity to CO2 by a factor of 5
to 10 due to a poor assumption built into the architecture of all climate models. Nice pictures of upper atmosphere on pages 17 and 18 of the synopsis:
http://jo.nova.s3.amazonaws.com/guest/david-evans/synopsis-of-basic-climate-models.pdf
David, thanks for the links. I’ll keep an eye on Jo’s updates.
You say that your notch delay model only effects surface temps but if there are periods of increased albedo from low altitude clouds then there should be a specific signature: surface temps should cool during albedo peaks and upper atmosphere should warm because there is a lot more solar irradiance passing through (due to the larger amount of low altitude reflection).
If there are no modulations then Force X is just an arbitrary (unmodulated) forcing that will be correlated to CO2 or indeed any forcing that causes surface temps to rise as they have. The filtered TSI has an upward trend as does surface temp but as the saying goes, correlation does not imply causation.
I am very interested in your estimate of CO2 sensitivity so when I’ve read your notes I will post my thoughts. I had a quick flick through and discovered what appears to be an error: in figure 4 of the notes in your last link, you have the water vapour emissions layer higher in altitude than the CO2 emissions layer. You justify this with the “observed emission spectrum” link on page 7 to a diagram of nimbus data on Jo’s page which was from WUWT:
https://wattsupwiththat.files.wordpress.com/2011/03/gw-petty-6-6.jpg
As you can see in this data the 15 um line corresponding to CO2 has a blackbody temperature of 215K, while the H2O line at around 18um corresponds to a temperature of 260K. The CO2 emissions layer must therefore be higher because it is cooler.
Bobo: Again, rather than preempt and repeat the upcoming blog posts, I just say to wait for them.
You say you “discovered what appears to be an error: in figure 4 … you have the water vapor emissions layer higher in altitude than the CO2 emissions layer.” No. As it says in the caption to figure 4: ” (Although the CO2 emissions layer is in the stratosphere around the center of its blockage at 15 µm, averaging by wavelength across the whole CO2 blockage gives an average height around 7 km, out in the wings of the blockage, which also happens to be where the main changes due to increasing CO2 are occurring. Hence this depiction.)”
As CO2 increases, the only parts that really matters is the CO2 in the upper troposphere. Yes the emission layer is mainly in the stratosphere, but its emissions barely change — they are already small, below 220K, see Nimbus emission spectrum. The changes in the emission spectrum are in the wings — see the last diagram here:
http://www.barrettbellamyclimate.com/page21.htm
Just average the emission temperature of the entire CO2 blockage around 15 um, by frequency, and you get a temperature corresponding to about 7 km. For more details on emission layers, see
http://joannenova.com.au/2015/10/new-science-14-emission-layers-which-pipe-is-the-biggest/
David, I don’t think it makes sense to obtain an average blackbody temp for CO2 with “wing” data because the higher frequency wing has increasingly dominant H2O spectra and the lower frequency wing has mixed in surface spectra. You need to look at the lower bounds in regions that are strongly absorbing.
Even if you choose not to use the 15-21um window for H2O and use say 20-25um you still obtain 250K for water (I am ignoring the narrow spike at 24um because it is not being consistently reproduced over an essentially non-trivial support, but even if you choose this very narrow spike the cooling temp is still higher that the stable minima near 15um for CO2).
Correction: I should say 17-21 um window for water
Bobo: Please see
http://joannenova.com.au/2015/10/new-science-14-emission-layers-which-pipe-is-the-biggest/
Note 1 notes that the CO2 “temperature and the corresponding height is not used in the calculations here; it is included only for completeness.”
The 7 km is a depiction only, but correctly indicates where the crucial changes to the emission spectrum are occurring,
David, it’s an important feature of the climate system even if you aren’t using it for calculations. For example, the higher the virtual surface of CO2 is the lower the levels of OLR are (colder emission => less energy flow), which effects the radiation budget. The more CO2 in the atmosphere, the higher the virtual surface of CO2. A still higher “surface” for H2O reduces the influence of forcing due to changing CO2.
Suppose that what you are saying was fine, that the virtual emissions surface of CO2 was lower in altitude than for H2O. Then there would be a relative peak around 15um compared to the water spectra at 17-25um (with respect to black body spectral curves), because the CO2 molecules in the “outer surface” are at a higher temp than water molecules at the corresponding surface. I had a quick look at the link you posted but could not find a solid justification of your claim.
Bobo: Yes, you are quite right that the higher CO2 emission layer goes in the troposphere (out in the wings, away from 15um), the less it emits to space (other way around in the stratosphere of course, but it makes little difference there because the temperature gradient is much less steep there). We quantified this in the outgoing longwave radiation (OLR) model; see
http://joannenova.com.au/2015/10/new-science-15-modeling-outgoing-radiation-olr/
The OLR model estimates how much the OLR (heat to
space) changes with changes to the heights of the emission layers, the
lapse rate, the surface temperature, the cloud fraction, and the CO2
concentration.
Which link, what claim?
Btw, the water spectrum continues out to 70 um or so, way past 25 um.
“Which link, what claim?”
The link I referred to was
http://joannenova.com.au/2015/10/new-science-14-emission-layers-which-pipe-is-the-biggest/
Your claim that I am referring to is that H2O virtual emission surface is higher than that of CO2.
I’m interested in reading about your OLR model at some point but I think the relative altitude of the H2O and CO2 virtual emissions surfaces can be deduced solely from the Nimbus data, or at least at that point where the data in the diagram was collected.
Of course, the emissions surface of H2O is expected to vary with latitude, because the latitude will roughly determine the altitude where most water condenses out.
Bobo, I never claimed that (or at least, not without the qualifiers, which you omit). I depicted it in the diagram that way, and have explained why.
David, OK, sure. First a correction re your model: when I claimed modulating albedo should have a specific signature (albedo peak => cooling surface T, warming upper atmosphere T), I think I was wrong: you have specifically stated that there is no “ripple” in surface T. However I still think albedo modulation should produce upper atmosphere T modulation. But as requested I’m happy to wait to see what Jo puts up.
I had a look at the page at the Barrett Bellamy website:
http://www.barrettbellamyclimate.com/page21.htm
Here are the pertinent diagrams you referred to (yellow is 220K, aqua 240K, purple 260K):
380ppm CO2:
http://www.barrettbellamyclimate.com/userimages/MOD380.jpg
760ppm CO2:
http://www.barrettbellamyclimate.com/userimages/MOD760.jpg
1000ppm CO2:
http://www.barrettbellamyclimate.com/userimages/MOD1K.jpg
I was mystified by the lack of movement of the bottom of the well as CO2 increases and so I did a calculation and found that the CO2 emission surface increases by about 4km if CO2 rises from 380ppm to
1000ppm (if you want to see the details I can TeX up a few of the equations I used). Things become clear if we look at
http://www.goes-r.gov/users/comet/tropical/textbook_2nd_edition/media/graphics/vertpro_temp_prof.jpg
The lack of temp change of the bottom of the well means that the MODTRAN simulation is being run in mid latitudes and the CO2 emissions surface remains in the tropopause the entire time. This means that in fact the CO2 emission surface temperature does not change at all (at mid latitudes) as CO2 is increased from 380ppm to 1000ppm.
So why are the wings shifting in Barrett’s simulations? On one wing increasing CO2 is increasingly blocking surface spectra, and importantly, on the other wing, increasing CO2 is increasingly blocking H2O spectrum. If the H2O emissions layer was on top of the CO2 layer, the CO2 layer would not be shifting at all, because it remains fixed in the tropopause, so we would observe no change whatsoever on the water spectrum wing. (This is not referring to the stratospheric CO2 spectral contribution which is the narrow spike in the middle of the well.)
You also mention
“The 7 km [CO2 emissions layer height] is a depiction only, but correctly indicates where the crucial changes to the emission spectrum are occurring,”
All the emissions surface is is the altitude of the atmosphere where, if you get any closer to earth, no spectral radiation is able to penetrate the rest of the upper atmosphere to be detected by the Nimbus satellite, and it is in fact 11-15km in altitude not 7km, and the H2O emissions surface is 6-8km (I seem to be using a different definition of emissions surface to you). I calculated the amount of CO2 above the emissions surface to be about 16 mol/m^2. The key to CO2 forcing with increasing concentration is not a reduction in the wing emissions temperature as Barrett and yourself suggest, but instead it is related to the extra diffusion “distance” required for CO2 spectral energy that is imposed by the increasing altitude of the emissions layer, i.e. if there is less than 16mol/m^2 of CO2 above a certain point then some CO2 spectrum is escaping directly into space; the less CO2 present above a given altitude (assuming CO2 above is already <16mol/m^2) the greater the fraction of CO2 losing energy by radiating directly to space. But increasing CO2 atmospheric concentration "deletes" lower altitude direct emitters, so the slow grind of collisional energy exchange persists to higher and higher altitudes.
Here are some figures for cooling of upper atmosphere from various channels of the AMSU satellite.
C10 (weight function peaking at 20 km altitude):
http://data.remss.com/msu/graphics/C10/plots/RSS_TS_channel_C10_Global_Land_And_Sea_v03_3.png
C12 (weight function peaking at 30 km altitude):
http://data.remss.com/msu/graphics/C12/plots/RSS_TS_channel_C12_Global_Land_And_Sea_v03_3.png
C14 (weight function peaking at 40 km altitude):
http://data.remss.com/msu/graphics/C14/plots/RSS_TS_channel_C14_Global_Land_And_Sea_v03_3.png
So what you see is that the rate of cooling becomes more pronounced with altitude; warming due to increasing solar irradiance could not possibly have this effect.
bobo
CO2 has 3 strong absorption lines at 2.8, 4.2 and 15 microns IR wavelength (IR ranges between 0.7 to 100 micron wavelength);there is also a weak absorption line at 2 microns.
The bulk of Earth’s radiation into space is between 8 to 13 microns with an absolute cut-off at 5 microns. So the higher energy absorption lines 2, 2.8 and 4.2 microns are irrelevant for global warming (in fact they have a cooling effect since without CO2 those wavelengths would hit and heat up the Earh’s surface anyway which in turn heats up the lower atmosphere).
It is the 15 micron line (at the much lower energy end of Earth’s emission radiation spectra and outside the bulk of Earth’s radiation into space whicht (slightly) warms the lower atmosphere. If you Google ” Wien’s displacement law calculator” you will find the temperature of a black body radiator emitting a peak IR wavelength of 15 microns has a temperature of….wait for it…-80 degrees (yes, minus) centigrade. A lot of warming in that one bobo!
It gets better, this -80C black body has broad IR spectrum whereas CO2’s 15 micron emission line is only 3 microns wide (4 absolute max), so it is like comparing the heat from an incandescent light bulb (broad spectrum) to an equivalent light output (lumens) of LED bulb (narrow emission spectrum (say at 1/6 the wattage for the same visible light). So bobo, we have even less warming.
It gets better still, water vapour (40-50X stronger IR absorber than CO2) also absorbs the 15 micron line (albeit less strong than CO2) so a lot of absorbing of the 15 micron is taken up by water vapour (which has on average a far higher concentration than a measly 400ppm). As there is only a finite energy at the 15 micron line and much of that is taken up by water vapour, there is even less warming attributable to CO2.
Bobo, I suggest you have a Bex and a good lie down, CAGW ain’t going to happen for a while yet …if ever.
It is well-known that H2O is a stronger greenhouse gas than CO2. But the vast majority of the atmosphere, if you take a vertical profile, is below zero celcius, so most water vapour condenses out before it gets very high:
http://www.periodni.com/gallery/atmospheric_composition_and_profiles.png
In the volume fraction graph you can see that above about 15km in altitude there is about 100 times as much CO2 in the atmosphere as water vapour. The graphs don’t include a dependence on latitude which, if shown, would show that CO2 would be the dominant GHG in the polar regions at even lower altitudes.
So if your figures are correct, that water is 40-50 times as strong a GHG as CO2, then CO2 is still the dominant GHG in most of the atmosphere.
The blackbody temperature of the earth is actually below zero, so if there were no non-condensible GHGs in the atmosphere such as CO2, the earth’s oceans would be frozen.
You wrote:
“If you Google ” Wien’s displacement law calculator” you will find the temperature of a black body radiator emitting a peak IR wavelength of 15 microns has a temperature of….wait for it…-80 degrees (yes, minus) centigrade.”
You misunderstand that statement. What that means is that if you look at the IR spectrum of a blackbody at -80C, the peak will occur at 15 microns. A blackbody at room temperature will have peak output at a shorter wavelength but it will still radiate more 15 micron radiation than the -80C body.
You wrote
“It gets better still, water vapour (40-50X stronger IR absorber than CO2) also absorbs the 15 micron line”
Water is a weak absorber there, CO2 is a very strong absorber in that band, CO2 closes off that section of the transparency window of H2O to thermal IR.
“CAGW ain’t going to happen for a while yet”
I haven’t made any arguments about a runaway greenhouse effect, that is highly unlikely, I’ve suggested in my comment that a new equilibrium will be established which will give rise to a warmer biosphere. Runaway GHA implies a lack of equilibrium.
Thanks Bobo for those graphs, yes they do show H20 vapour drops of at altitude and CO2 stays constant. As I understand when we discuss global warming we are referring to the lower atmosphere which I also understand is lower than 10km altitude where water vapour is more prevalent and therefore its far stronger “greenhouse” properties must be accounted for. This is clearly demonstrated where clear dry nights cool quicker and are cooler than humid nights.
And yes, if there is no “greenhouse” effect and no extra heating due to gravitational atmospheric compression then Earth’ s temperature would be around -18C. On the other hand if we add much more atmosphere to equate with Venus 92 atmospheric pressure, Earth temperature would be about 400 C ! (Venus with 97% CO2 and nearly 30% closer to the Sun is 470 C. Conversely if we moved Earth to Venus’s orbit it would reach mid to high 60s C, Venus temperature at one atmosphere (some 50km altitude) drops to low mid 70s – so not much difference considering our CO2 is 0.04% and Venus is 97%.
Also, I know a -80C black body radiator has a broad IR spectrum peaking at 15 microns. I did emphasise this by comparing an incandescent bulb (black body-(BB)- radiator) with an LED bulb (emission line emitter).
And I am aware a BB at room temperature will emit a higher intensity through the IR range (including 15 micron line) and both peak and cut off wavelengths are higher up the IR scale (which would be about 10 and 6 microns (assuming a room temp of 20C) respectively. What I am getting at is there is only so much energy at the 15 micron line and even though that energy increases as the temperature rises, its proportion of the total energy intensity emitted by the BB reduces (think of the area under the Wien’s displacement curve) . Therefore its effect on global warming is reduced proportionally because higher energy IR kicks in (Energy = Plank’s constant x frequency) which warms Earth’s surface which in turn warms the lower atmosphere by convection.
I disagree with H2O being a weak 15 micron absorber, it is at least 60% ( if not 70% ) of CO2. I did acknowledge this by saying ” (albeit less strong than CO2) “.
An interesting fact is that the ability for a CO2 molecule to absorb 15 micron radiation at low altitudes is lower than at higher altitudes, because equipartion principles imply that the 15 micron energy level has a higher probability of already being populated due to the higher frequency of collisions with other molecules in air.
Another way to think about the important effect of CO2 at higher altitudes is that “surface of emission” of the outer atmosphere (i.e. the altitude at which atmosphere becomes opaque to incoming 15 micron radiation) actually increases in altitude if more CO2 is added to the atmosphere. As a consequence the “surface of emission” is cooler, so the amount of 15 micron radiation emitted into space decreases, which of course shifts the radiant energy budget of the outer atmosphere.
“because equipartion principles imply that the 15 micron energy level has a higher probability of already being populated due to the higher frequency of collisions with other molecules in air.”
and also because there is a greater chance of 15 microns worth of energy being imparted in a given collision due to the higher temperatures
I have heard this explanation of absorption and re-emission many times. It does not explain how the atmosphere warms. We are measuring the temperature of air. Temperature of a gas is an expression of the kinetic energy of the molecules in the gas. Now if the mechanism involved was only absorption and re-emission then it would be only the CO2 and H2O molecules involved in the process. Oxygen and Nitrogen would not get very excited about the emissions from CO2. So how do they get “warmer”? They do so by interacting mechanically with CO2, thus absorbing some of the kinetic energy from the molecules. When a CO2 or H2O molecule absorbs a photon they get excited and bump into a neighbour O2 or N2 molecule (much more likely to find one of those than the very few kindred spirits) and thus transfer some of their excitement to these molecules. The CO2 and H2O is left with less energy than is required for emitting a photon, but still with higher kinetic energy than they had before the photon hit them. This is what really happens, not that pure absorption and re-emission stuff. And that is how energy is transferred to the bulk of the atmosphere.
Yes, the absorption/re-emission of CO2 explains the (slight)warming of the lower atmosphere. The 15 micron photon excites the CO2 molecule which either re-emits the photon or collides with O2 or N2 molecule, transferring kinetic energy. However we are only dealing with a finite amount of energy (which is also absorbed by water vapour – to keep Bobo happy- albeit a lesser % absorption) .
O2 and N2 in the atmosphere do not absorb IR (well O2 has a very weak absorbtion line at 7 microns) are warmed essentially by initial conduction and then convection from the warmed land mass and oceans from solar radiation (hot air rises, cool air sinks).
The upshot is CO2’s 15 micron absorption interferes with Earth’s radiation into space, but a heck a lot of other IR wavelengths at higher intensity and energy and many more IR wavelengths at lower energy pass out to space unimpeded. But remember also that CO2 other absorption lines (2, 2.8 and 4.2 microns), which are outside Earth’s radiation into space, act as a cooling effect by partially shading us from those wavelengths from the Sun ( a reverse “greenhouse” effect).
Think of a large house with all it many windows open on a frosty night – CO2 would have the effect of closing one window!
Anders,
You’re correct that there is more going on, at low altitudes the collision rate is so high that there is a much higher likelihood of the 15 micron bending mode being damped by transferring its energy in a collision to kinetic energy of another molecule, hence raising the temperature of the air. However there is also a much higher probability of the 15 micron energy level of a CO2 molecule being excited by the transfer of energy in a thermal collision with another molecule, and even without absorbing 15 micron photons, the CO2 will be weakly emitting 15 micron radiation that originated as thermal kinetic energy that was transferred from another molecule in a collision.
At higher altitudes the probability of direct photon emission goes up dramatically due to the lower frequency of collisions, and at all altitudes the presence of additional CO2 increases the emissivity of the local body of air of 15 micron radiation.
Do you understand why air samples taken in rural areas are going to have a great variation in CO2? For the simple reason that the samples could be taken close to sinks, e.g. plants absorbing CO2, or sources, e.g. vegetation matter rotting. The advantage of Mauna Loa as a sampling station is that it is a long way from highly variable sources and sinks.
Yes, I do understand about sampling in rural areas, and Mauna Loa’s advantages are set out in the essay.
So to summarize. You accept that the “proportion of carbon dioxide in the atmosphere has been rising steadily since 1960” I agree, a 25% increase to be precise.
[…] foray into the SCM document on ‘climate change’ that I have investigated twice before, here and here. But before I get into it, readers in New South Wales will have noticed that the Government of […]
I came across a comment on another website about the relationship of CO2 to temperature which I thought passing on:
‘CO2 levels have increased 9.5% (35 ppm) in the 17 years from 1997 through 2014, while they only increased 7.4% (25 ppm) from
1980 through 1997. Hiatus or not, what does this say about the sensitivity of surface temperatures to atmospheric CO2 levels? Clearly there are important factors involved in global surface temperature variation other than atmospheric CO2 levels.’
That seems sensible to me.
Don, you have made it pretty clear that you reject all surface temperature data, so I don’t really know what you mean by “seems sensible to me”. Could you please state with maximal clarity what you mean?
You say that you only consider satellite data. The weight function of interest is TLT:
https://upload.wikimedia.org/wikipedia/commons/2/2e/Weighting_Function.png
As you can see TLT is very different to surface air temperature. Surface air temperature is measured 1.25 to 2m above the surface.
Bobo,
I don’t reject all surface temperature data. There are many posts in which I consider the problems of measurement. I simply prefer the satellite data as an indication of global warming. Surface air temperature, as you say is measured about 1.5 meters above the surface, but only on ground. The 70 per cent of the globe that is sea has quite different and diverse forms of measurement for temperature, another reason I prefer satellite data. It depends on what it is you are interested in.
Don, ocean buoys measure air temperature as well as sea temperature.
Your opening comment is not clearly articulated so I don’t really know what you are trying to say. What do you mean by sensitivity? ECS?
I think your comment is related to notions such as thermal inertia of the oceans/climate system, heat capacity, changes in CO2 forcing with changes in CO2 concentration. Perhaps a bit of reading about those things might clarify your thoughts a little.
ECS. Ocean buoys of the Argo kind don’t go back much more than ten years, and the others go back to 1960, but are in no sense a good sample of the oceans.
Don, trying to make estimations of ECS from short time intervals is extremely difficult. All you can really do is ascertain what the lower bounds are likely to be.
Measurements of surface air temperature over the ocean has a much older history than the Argo program, older also than 1960.
If the Argo data isn’t good enough, I’m not sure why you would use surface air temp data at all.
Why don’t you use total heat content of the climate system as a metric to determine whether global warming is occurring?
Bobo,
You are pointing to the problems that underlie the whole AGW scare. Without ECS or TCS there is no scare. It is the IPCC that has fathered the ECS notion. You should talk to that august body and point out the error of its ways.
The measurement of temperature over the seas is confounded by the sampling problem. Vast areas aren’t measured at all except by extrapolation.
I’ve written about this on several occasions. For my part, the surface data are so bad that it’s almost senseless to talk about global average temperature. But unless one does one can’t take part in the discussion at all, other than talk about what the satellites and balloons tell us. And their time length is short.
Seems to me you’ve just fallen for another example of cherry-picking. Did you ask yourself why those dates? You only have to look at that 20-year satellite plot from the other day (reposted here) to realise that if you want to start dicing and slicing the timeseries, you can pretty much “prove” anything you want depending simply on where you slice. And of course CO2 isn’t the only thing involved in global surface temperatures; that’s hardly insightful. You can clearly see the 2007/2008 and the 2015 El Niños (Los Niños?) in the data below.
The years are almost the totality of UAH measurements. As you say, you prove anything (almost) if you pick beginning and end years. Why, exactly, did you start with 1995?
I was reusing yesterday’s graph, which was in response to your earlier claim “there has been very little warming over the last twenty years.” (2015 – 20 = 1995)
At any rate, it sounds like dlb is working on a newer one for us with the UAH V6 beta data.
OK, understood.I have to say that I vary between ‘a
couple of decades’, ’18 years or so’ or similar less-than-precise measures. I’ll do better next time.