From the department of “I told you so and I have an experiment that precedes this to prove it” comes a paper that proves Bill Nye’s faked ‘greenhouse effect’ experiment is also based on the wrong ‘basic physics’. Remember when I ripped Bill and Al a new one, exposing not only their video fakery, but the fact that experiment fails and could never work? Well, somebody wrote a paper on it and took these two clowns to task.
The Hockey Schtick writes:
Oh dear, the incompetent & faked attempt by Bill Nye to demonstrate the greenhouse effect for Al Gore’s Climate “Reality” Project has also been shown by a peer-reviewed paper to be based upon the wrong “basic physics” as well. According to the authors, Nye’s experiment and other similar classroom demonstrations allegedly of the greenhouse effect:
“All involve comparing the temperature rise in a container filled with air with that of the same or a similar container filled with carbon dioxide when exposed to radiation from the Sun or a heat lamp. Typically, a larger temperature rise is observed with carbon dioxide and the difference is attributed, explicitly or implicitly, to the physical phenomena responsible for the climate change. We argue here that great care is required in interpreting these demonstrations and, in particular, that for the case of the demonstration described by Lueddecke et al., the results arise primarily from processes related to convective heat transport that plays no role in climate change.”
Bill Nye the propaganda guy experiment for the Climate Un-Reality Project
According to the paper, Nye’s experiment
“demonstrates an entirely different phenomenon: The greater density of carbon dioxide compared to air reduces heat transfer by suppressing convective mixing with the ambient air. Other related experiments are subject to similar concerns. Argon, which has a density close to that of carbon dioxide but no infrared absorption, provides a valuable experimental control for separating radiative from convective effects.“
Not only did the authors find that addition of the non-greenhouse gas Argon had similar heating effects to CO2, the Argon control actually heated up slightly more than in the greenhouse gas CO2 experiment, definitively proving that such experiments assume the wrong “basic physics” of radiation were responsible for the heating observed, instead of the limitation of convection due to CO2 having a greater density compared to air.
Nye’s experiment not only limits convection by addition of denser CO2, it completely eliminates convection by enclosing the CO2 in a bottle with the top on.
According to the authors,
“It has been known for more than a century that the warming of air in a real greenhouse results primarily from entirely different physics—mainly that the glass prevents mixing between the warm air inside and the cooler air outside, and therefore suppresses convective heat transfer between the interior and the exterior; the infrared absorption of the glass plays a much smaller role. We show here, via experimental data and a simple theoretical model, that the effects observed in the demonstration described in Ref. 1 arise from a similar restriction of convection rather than from radiative effects. In this case, it is the density difference between carbon dioxide and air, rather than the presence of a solid barrier, that suppresses mixing of the gases. Although the details differ, similar considerations apply to other demonstrations that have been reported.”[including Nye’s ‘greenhouse effect’ enclosed in a glass bottle]
Thus, Nye’s experiment, in addition to the video fakery and incompetence, is not even wrong on the “basic physics” of the greenhouse effect.
As the authors point out in the conclusion,
“Although not an accurate demonstration of the physics of climate change, the experiment we have considered and related ones are valuable examples of the dangers of unintentional bias in science, the value of at least a rough quantitative prediction of the expected effect, the importance of considering alternative explanations, and the need for carefully designed experimental controls.”
The paper in the American Journal of Physics:
Climate change in a shoebox: Right result, wrong physics
Paul Wagoner , Chunhua Liu and R. G. Tobin
Classroom experiments that purport to demonstrate the role of carbon dioxide’s far-infrared absorption in global climate change are more subtle than is commonly appreciated. We show, using both experimental results and theoretical analysis, that one such experiment demonstrates an entirely different phenomenon: The greater density of carbon dioxide compared to air reduces heat transfer by suppressing convective mixing with the ambient air. Other related experiments are subject to similar concerns. Argon, which has a density close to that of carbon dioxide but no infrared absorption, provides a valuable experimental control for separating radiative from convective effects. A simple analytical model for estimating the magnitude of the radiative greenhouse effect is presented, and the effect is shown to be very small for most tabletop experiments.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
“Going from higher pressure to lower pressure chills the CO2 even though the lower pressure is still higher than ambient.”
Yes, “chills the CO2” meaning its temperature decreased. To what level? Cooler than ambient? No. Start with gas at ambient temperature of 25C and 1 bar. Now compress to 2 bars at equal volume. Temperature rises. Reverse the adiabatic process. Lower pressure to 1.5 bar. Temperature decreases. Is it higher or lower than 25C? Higher.
The result will be different if the system interacts with its surrounding such as poor heat insulation resulting to big heat loss. The process is no longer adiabatic and reversible.
SMS
CO2 will warm slower than air because its molar heat capacity is 28 J/K whereas air (N2 and O2) is 21 J/K.
Aussiebear
If you reduce CO2 to 400 ppm, your glass jar must be at least 4 meters tall to have a chance of detecting “greenhouse effect.” Four meters is approx. the full IR absorption length by CO2 at 400 ppm.
“CO2 will warm slower than air because its molar heat capacity is 28 J/K whereas air (N2 and O2) is 21 J/K.”
This was what I was thinking (and apparently impartial experiments have shown). If I’m not mistaken this would mean that CO2’s primary warming effect would be limited to reducing cooling. In ‘lay’ terms the CO2 acts like a capacitor for heat, with the molecules remaining excited for longer before releasing residual heat (err… cooling down).
That should mean that so long as night time temperatures can reach their ‘optimum low’ (the same temperature they would hit with, say, 300PPM) that AGW by means of CO2 is falsified. Designing an experiment around this seems to be a much more simple endeavor with many factors not requiring an accounting. Multiple identical boxes each with different concentrations of air/CO2/Ar, etc… all within close proximity to each other and simply watch what a few days of normal weather does the temperatures in the boxes relative to each other.
I honestly don’t know what the results would be. We wouldn’t need anything special but boxes carefully designed to try and not interfere with the natural processes being studied that can remain air tight with multiple shielded thermometers within each box. I’ve been thinking about this experience off-and-on again for quite a few months and it may soon be the case that I will have the means (time, skill, & location) to pull it off myself. Any ideas on material choices and layout?
deebodk says:
August 11, 2014 at 4:04 pm
“On a loosely-related note, Bill Nye [greases] the cover of the latest Popular Science magazine with the caption ‘Bill Nye Will Save Science in America – or Go Down Swinging’….”
Fixed it fer ya.
I think that the experiment is limited by using equipment that is readily available in a secondary school.
The observation that it is not easy to show the atmospheric greenhouse effect (if any) is difficult to undertake should be sympathetically understood.
The main point is that some attempts seem valid but use flawed physics and can be ruled out.
Tom T says
August 11, 2014 at 12:44 pm
I believe that Dr. Spencer has also chimed in on these worthless experiments pointing out that CO2’s radiative properties are both a function of concentration and distance. At such short distances it is simply impossible to measure the greenhouse gas effect of CO2 with a thermometer. Even in major controlled laboratory experiments using long distances it takes extremely accurate instrumentation to measure the actual GHG effect of CO2.
I refer you to the work of Hoyt C Hottel on radiant heat transmission from nonluminous gases such as CO2 and water vapour.He developed smoothed plots of CO2 and water vapour emissivities at a total pressure of 1 atm, based on direct measurements of total emission. For CO2 emissivity depends on gas temperature Tg, the product term PcL (CO2 partial pressure and mean beam length) and the total pressure.Similarly for water vapour the gas emissivity depends on gas emperature, concentration and distance.
This work is set out in Heat Transmission by WH McAdams 3rd ed chapter 4 and is widely used is design calculations for industrial furnaces.
Stokes,
Sure! I have numerous patents on products that have generated millions of dollars in profits for my employers. My designs have broken new ground in several different industries: personal protective equipment (gas masks, rubber gloves), sports equipment (racquets), and the automotive industry. BSE in Mechanical Engineering, University of Michigan, 1981, not the highest rated school in this discipline, still only second behind MIT where my dad went.
You seem reluctant to attempt to answer direct questions. Once again, could you assess my criticisms of Nye’s procedure? Nye is a mechanical engineer as well, but has gone over to the dark side, choosing to conform to the Team’s contentions rather than remaining rigorous to his training.
Concerning radiative physics, my class in Transport of Heat and Mass, as well as standard ME Thermo and Circuit Analysis, gave me a background, and I have read extensively to understand what CO2 can and cannot do in the atmosphere. It seems to me that Trenberth’s 342 W/m2 from CO2 at the tropopause could not possibly heat Earth’s surface, as it would be absorbed and thermalized many thousands of feet above. Care to reflect?
So, by hook and crook I was able to get a copy of the following paywalled paper:
‘Climate change in a shoebox’: a critical review
M Bert`o, C Della Volpe, and L M Gratton
Eur. J. Phys. 35 (2014) 025016
The paper itself is a bit critical of the Tobin paper, but I just can’t figure out why. They raise the same concern as Nick Stokes that the thermometer in the original paper is placed at the bottom of the vessel so it had no choice but to measure the convective effects, but I really have no idea why, because they placed many thermometers at places all over their apparatus at various heights and locations with respect to external lab features (on the side of the box toward the wall, away from the wall, etc.) and as far as I can tell from the data they’ve plotted the measured change in temps are smaller at the top of the vessel than they are at the bottom (which makes sense for many reasons) but the *relative* changes in temperature are essentially the same. They use three instead of two gasses, argon (mass 40), carbon dioxide (mass 44), and nitrous oxide (mass 44), and what they find is that the argon/air delta in their setup conditions is about 2.64C, which they attribute to a purely convective contribution, and the deltas for carbon dioxide and nitrous oxide are both approximately 0.25C higher, which they attribute to a purely radiative absorption effect.
As I said, they are critical (thus “critical review”) of the earlier experiment, but I really have no idea why. They say that a careful experiment can separate the convective and radiative effects and the original authors were too dismissive of this as a lab demonstration possibility. But the original authors in fact clearly state that the difference in the two is over an order of magnitude different (0.25C is an order of magnitude lower than 2.64C the last time I checked) so isn’t a credible experiment to easily establish the greenhouse effect in a simple Gore/Nye-like demonstration. Taken at face value the results obtain in the new paper are perfectly consistent with and support rather than refute the original paper.
But there’s a little more to it than that…
In my opinion, or a paper that tried to show up the original and come running to the rescue of shoebox greenhouse effect demonstrations it did a poor job and left more questions than answers. I cannot find a non-paywalled version, but if anyone can so all the others can look it would be VERY helpful. The key figure I’ll describe is a temporal plot where they turn the light source on over an air-filled box and plot the temperature as a function of time to show it come to equilibrium. Then after equilibrium is established they dump argon from a room-thermalized container into the test container, presumably filling the test container with pure argon which then re-thermalizes under the lamp. As in the original experiment an easily observed spike in temperature occurs and is measured. Then, after peaking, the temperature starts to go back down because the test box is open at the top and the argon slowly diffuses out of the box to eventually result in a box filled once again with air and back at (or near) the original equilibrium point. After equilibrium is observed they then dump a room-thermalized container of carbon dioxide into the box and repeat the same temporal measurement and finally once the box is presumably just room-equilibrated air again they repeat with nitrous oxide.
They talk about multiple experimental repeats and statistical analysis of the results, but they miss what to me is the big elephant in the middle of their plot. Lacking actual numbers and only looking at the plot itself it is very evident that the time from argon addition to peak argon temp is 2/3 that of the same measurement for carbon dioxide and nitrous oxide and the time from peak temp back down to equilibrium for argon is again about 2/3 that of carbon dioxide and nitrous oxide. There is absolutely nothing magical about the radiative effects present in the polyatomic gasses that can explain that. The argon goes up much faster and comes down much faster.
There are two very obvious effects that would contribute to this observation. One is that argon has a mass of 40 and the other gasses have a mass of 44. So the diffusion rate of argon based on mass alone is about 5% faster than the mass-based diffusion rate of the polyatomic gasses. What that means is that argon is disappearing from the vessel measureably faster and the total integrated heat that argon’s convective behavior in the container contributes to heat trapping is more than a bit lower than in the cases of the polyatomic molecules. In fact, the 5% higher diffusion rate of argon could easily result in more than a 5% integrated difference in trapped heat (remember, the measured difference is only 10%) and the small dipole of nitrous oxide and the relative strong Van der Walls forces present in carbon dioxide and nitrous oxide might also help make up that total difference by decreasing he diffusion rate of these gasses even more.
I’m not suggesting there is *no* radiative effect due to the IR absorption lines of carbon dioxide and nitrous oxide, but I think that this paper clearly makes the case it is directly measure to be more than an order of magnitude below the convective effect – INDEPENDENT of where the thermometer is placed in the container – and a careful examination of results and experimental conditions suggests the radiative effect is possibly even quite a bit smaller than reported.
You know what I’d have loved to see in this paper? The same measurement but with krypton.
I hate auto-correct! van der Waals…
I’m still getting my head around the claim that a CO2 concentration of 400 ppm (or 0.0004) will cause ever increasing temperatures. And yet here we are at 400ppm and temperatures have not increased for 17 years and counting. That’s a total headscratcher for the honest scientist.
Merrick
Argon gas warmed faster than CO2 and N2O because its molar heat capacity is less than half of the latter two gases. Ar = 12.5 J/K vs. CO2 and N2O = 28 J/K. Convective heat transfer will take place if the container is open, exposed to air. In closed container, it’s radiative and conductive heat transfers. Unless the container is as big as a room, it will have convective currents inside.
Michael Moon
“It seems to me that Trenberth’s 342 W/m2 from CO2 at the tropopause could not possibly heat Earth’s surface, as it would be absorbed and thermalized many thousands of feet above.”
The ‘back radiation’ is 324 W/m^2 not 342. You’re right, it is absorbed actually in just several meters. But it is re-emitted and re-absorbed continuously until it reaches the surface. It also reaches space since it travels in both directions, up and down.
BTW I’m impressed with your credentials. I’m not a climate scientist but climate scientists seem to think I know enough about science. Today I got an invitation to review a climate change paper by a group of IPCC scientists. It’s unusual for a ‘denier’ to be doing this. I’m thinking whether to accept or decline.
‘Back radiation’ is a misnomer. Radiative heat transfer is the net result of an interchange between two ‘bodies’. The net result for interchange between gas containing CO2 and water vapour and its bounding surface is Q/A = sigma*(eg*Tg^4-ag1T1^4). The net transfer depends on the fourth power temperature difference between gas and surface (Tg^4-T1^4); which will be negative ie no ‘back radiation’ as T1 > Tg for land surface T1 and troposphere Tg.
There is still back-scattered radiation, no matter the thermal balance. If I emit 15um wavelength IR into a gas mixture which will reflect 25% of that 15um back as 20um, to which I am especially sensitive, then that back-scattering effect may be critical in raising my energy levels. Indeed, if I merely reflect 15um IR but “absorb” 20um, it could very well be that the gas mixture is the only reason why I am warmer than ambient…
Radiative heat transmission is computed as a net transfer between two emitter/absorbers; magnitude depends on Stefan- Boltzman fourth power difference equation with heat flowing from the hotter to the colder body. This includes band emissions and total spectrum emissions.
“Radiative heat transmission is computed as[…]”
Irrelevant when speaking of actual processes rather than the macro-scale view point. Colder objects STILL emit radiation. That radiation STILL interacts with warmer objects. Therefore, radiation from cold to hot is still occurring – no matter of the given fact that the warmer body will be emitting greater radiation in most (all?) cases.
I think a highly radioactive colder body could warm a warmer low radioactive body, but I have no evidence to support this hypothesis. It may require a very minute quantity or special circumstance, but I believe it to be possible.
We know that cold bodies radiate, all bodies above absolute zero radiate and the emissive power W depends on its temperature only: W = sigmaT^4 (Stefan Boltzman Law).
The radiation is two way from the cold to the hot and from the hot to the cold; but the net transfer is from the hot to the cold Q(h-c)=Aesigma(Th^4-Tc^4)
looncraz wrote “There is still back-scattered radiation, no matter the thermal balance.”
I wonder if I will ever find “back-scatter” in a physics book? Probably not. the warm object has no idea neither does it care where its heat came from, thus there is no “front” and no “back”. There is only radiation (scatter).
“then that back-scattering effect may be critical in raising my energy levels. Indeed, if I merely reflect 15um IR but “absorb” 20um, it could very well be that the gas mixture is the only reason why I am warmer than ambient.”
Well yes, if ambient is NOT absorbing 20 micron but your sample gas IS absorbing 20 micron, then depending on how much 20 micron is actually present you will obtain a temperature differential, all other things being equal. Of course, all other things usually aren’t equal but there you go.
“I wonder if I will ever find “back-scatter” in a physics book? Probably not.”
Yes, of course you would, though usually as one word: http://en.wikipedia.org/wiki/Backscatter
I learned about it in a very low level physics course… and also in common use amongst painters 😉
looncraz says “Yes, of course you would,” (find backscatter in a physics book, cites Wiki…)
Well, okay, true enough, but not as it is seems to be used climatism.
“is the reflection of waves, particles, or signals back to the direction from which they came”
Backscatter is not re-radiation. CO2 absorbs infrared, presumably heats up slightly, and emits a photon of its own, possibly or probably at a different wavelength, after holding the energy for a small amount of time. It might also convect its heat (and appears to be the principle mode).
My comment expressly remarked on a CO2 molecule NOT knowing or caring where its energy came from — there can be no “back” when the molecule has no concept of direction. When it is ready, it will emit a photon.
Backscatter is a reflection phenomenon, not an absorption phenomenon.
Clouds probably backscatter AND absorb (followed by convection and radiation).
Well anybody who thinks that the kind of apparatus used in these Bill NYE type of experiments are quite immune to pseudo BB radiation effects from to 1,000 to 3,000 K sources used, has rocks in their head.
Take the flasks for example. Just what kind of glass are they made out of ??
Any ordinary (battleship plate) glass is almost certain to have plenty of water in it. Yes I do mean in the glass itself.
Consequently, almost any ordinary glass, is strongly absorbing in the 1micron region. Get out your Edmunds Scientific catalog and look for the transmissions for some common optical glasses; like BK-7 for example. You have to know where to look, if you want a glass that doesn’t have the water dip in the IR transmission.
Why do you suppose your house or car windows get hot,.
So IMHO, the Nyescam is simply the vessel getting hot, and the thermal conductivity of the air mixture, being different. And if you don’t get convection in your jar, please photograph it, and make accurate dimensional drawings of your whole system; and send them to me. I have clients, who will pay almost anything to get any sort of illumination system, that is a perfectly uniform radiation beam.
You don’t even need to get your hands dirty, with specific heats or molecular weights. First of all, you need to get two identical systems, to heat the same, before changing one of them.
“the Nyescam is simply the vessel getting hot, and the thermal conductivity of the air mixture, being different.”
Thermal conductivity of air is higher than CO2 and Ar. Air = 0.024 W/m-K; Ar = 0.016; CO2 = 0.015
Air will transfer heat faster. Warms faster.
“And if you don’t get convection in your jar, please photograph it”
The light-heat source must be at the bottom of the closed jar to get convection. If gas at top is warmer than gas at bottom, no convection. Hot gas will not rise because it is already on top. In small closed containers, the temperature difference of gas inside is small so convection is negligible.
Dr. Strangelove,
You presumably haven’t seen the paper and the data, so it’s difficult to draw conclusions from what you haven’t seen and what I can only strive, imperfectly, to clearly describe in text. You are correct in pointing out that argon has a much lower heat capacity than the other gasses, but it doesn’t influence the observations I described. What happens in the experiment is this: the apparatus is at equilibrium with air in it, at a temperature that, importantly, is above the ambient temperature in the room. The gas which is being introduced is at thermal equilibrium with the room, not the apparatus, before it is added. What that means is that, necessarily, the temperature of the apparatus briefly declines when the test gas is added. But the thermal mass of the gas is so small compared to the apparatus itself (not at all surprising) the data as presented show no decrease in measured temperature. That is, the temporal resolution of the data as presented (it was taken at 2 Hz, so they have it, they just display decimated data) shows absolutely no dip. The heat capacity of the gas would have it’s greatest effect in how far this temperature drop occurs and how fast the drop is reversed.
The observed rise in temperature, because the thermal mass of the apparatus is so many orders of magnitude greater than the thermal mass of any of the gasses contained in the apparatus, is impacted by the heat capacity of the gas almost negligibly. And, even if the heat capacity of the argon describes why the argon temperature rise peaks faster (the heat capacity plays a minor, not major, role here) it absolutely does NOT explain why that temperature spike then returns back to the ambient temperature of the air-filled apparatus so much faster for argon that for the other gasses. The only thing that explains that observation is that the argon has to be leaving the apparatus much faster (i.e., observational timescale) than the other gasses. Therefore the time integration of the convective effect, which everyone agrees is the major player in the temperature rise, is much less for the argon than it is for the other gasses. And the difference between the amount of “extra” heat captured by the apparatus for argon and the other gasses doesn’t just automatically get placed in the radiative greenhouse column on the ledger sheet.
As an aside, all of the arguments that are being made about how this experiment has to be run with a 288K, or whatever temperature, blackbody are, to borrow a phrase, “not even wrong.” That is in itself a valid experiment, but it isn’t an experiment that tests the so-called greenhouse effect (so-called because of the many earlier discussions in which we all agree that greenhouses generally work not on the radiative properties of the glass walls, but on the fact that the glass prevents gas convection). That test which employs an *internal* heat source only would be a test of the *insulative* effects of various gasses – due to mechanical and thermodynamic properties of the gasses as well as any potentially active spectral modes in the relevant regions of the IR. The effect that the Nye-like experiments are trying to demonstrate is how incoming energy, in the form of primarily visible light (like from the Sun) are converted to heat by the “ground” and how gasses between the “ground” and the visible light source then, to varying degrees, *radiatively* interact with the “ground” to affect the equilibrium temperature. The problem related with the light source is really more along the lines that there is one heck of a lot more IR, relatively, in the source spectrum of a light bulb than there is at the bottom of the Earth’s atmosphere coming from the Sun and that the container also interacts with the light source in those parts of the spectrum in ways that don’t accurately reflect the Sun-Earth-Atmosphere system. Those are valid points to make in a critical analysis of the experimental setup, but removing the visible light source from the experiment doesn’t “fix” those problems, it makes the experiment a different experiment altogether that doesn’t tell us anything useful about the so-called greenhouse effect and the overall impact of greenhouse gasses in the Earth’s atmosphere.
“The effect that the Nye-like experiments are trying to demonstrate is how incoming energy, in the form of primarily visible light (like from the Sun) are converted to heat by the “ground” and how gasses between the “ground” and the visible light source then, to varying degrees, *radiatively* interact with the “ground” to affect the equilibrium temperature”
Then why are they using heat lamps? Seems a more proper experiment would be to use a low-IR light source shining into a container with a thick layer of dirt on the bottom and then compare temperatures over time. As it stands, the heat lamps are just heating the glass (which then gets hot) thus convecting heat into the gas contained therein (as well as some spectrally irrelevant radiance effects), thus providing no useful information what-so-ever other than specific heat comparisons.
It is a completely useless experiment for trying to show the GHG effect of CO2. To do that one would need a much larger volume, a proper (diffused, spectrally accurate) photon source, and the container’s influence minimized. Five containers of an identical atmosphere. Same humidity, nitrogen, oxygen, etc… Remove all CO2 from all containers. All containers are placed in yet another master container to isolate external effects. This outer container contains either no atmosphere or an inert atmosphere and has a homogenous temperature.
Test Containers:
1. 0PPM CO2
2. 200PPM CO2
3. 400PPM CO2
4. 800PPM CO2
5. 1600PPM CO2
Note: Each container will have a non-heated control duplicate (so 10 containers total), 5 are for testing.
Preparing the containers:
On the bottom of each container will be placed uniform gravel, six inches thick, perfectly level. Next, silica sand will be evenly distributed over the gravel (the sand would be 1 inch thick on a perfect surface), followed by twice as much purified soil (by loose volume). After this the relative humidity in each container will be raised to 100% and the containers atmospheres will be intermixed and deprived of most or all of its CO2 via whatever means available. The air returning to the containers must be sterilized and kept at a relatively stable 288k, as well as reducing the humidity to normal-earth levels (global average). This process should continue for long after the majority of atmospheric CO2 has been removed – until soil moisture in all containers is equalized and the atmospheric analog contains the appropriate water vapor levels. 1.0 bar must be maintained throughout this process.
It is now time to differentiate the containers by adding CO2 into the containers in pairs. One pair is already (0PPM – or as close as it gets) and it sealed. The remaining containers’ atmospheres are intermixed during the entire time when they are trying to reach their CO2 concentrations to distribute any errors evenly (such as a trace impurity). The second pair (200PPM) is sealed and the remainder keep gaining CO2 (while maintaining 1.0bar, temperature, and humidity) and sealing off once they hit their target concentrations. The containers are left to rest in a dark, homogenous environment, maintained at 288k with a controlled atmosphere and tested for proper isolation after a few days (all containers should show the same temperatures in pairs in the very least). Finally, the testing can begin.
Now, heating should occur for 12 hours on, 12 hours off. Heating should be PURELY by the same spectra of radiation as seen in the lower troposphere. The control containers are shielded from the direct influence of this radiation, but remain in the same isolating atmosphere. During the experiment it should not be known which containers contain what concentrations of CO2 nor which container is relaying information to the control center (double blind) and the experiment should be able to run unattended for a few MONTHS.
Sensors and cameras will monitor all factors of interest (absorption spectra, temperature, humidity, pressure, convection, condensation, and many atmospheric qualities). On the note of temperatures, we need to know more than just one or two. We need to know the actual temperatures of every layer in the containers, including the temperatures of the gravel, sand, and soil. From this, we can calculate how much energy has been stored in the containers.
My guess is, with 12 hour cycles, there will be no significant net difference after about 400PPM. The walls of the containers will convect the heat to the 288k buffer fully over night (the walls don’t need to be specially insulated, that is the purpose of the homogenized common external buffering atmosphere held at 288k). All that needs to happen for CO2-induced warming to be proven is that the temperatures in the containers cannot reach the same thermal equilibrium as their lower-CO2 counterparts overnight because CO2 keeps the energy trapped – or that the higher CO2-containing containers heat up more during each and every day.
My expectation is that the extra CO2 slows down the heating process, delaying the point at which the greatest temp is reached, and increasing nighttime temperatures until equilibrium is reestablished. At some point CO2 should be able to prevent nighttime temperatures from reaching equilibrium – and that will equate to thermal runaway, but I think the atmosphere would no longer be breathable by that point.
Dr. Strangelove,
I just realized I have a better way of explaining why heat capacity is not the issue in what the data show in the Eur. J. Phys. paper. In this case, the heat capacity of the gas would primarily influence the *rate* at which the temperature rises in the vessel. This also would eventually impact the equilibrium temperature of the vessel, but the experimental data show that the *rates* of temperature increase are, within the ability to intuit from the figure, essentially the same for all of the gasses. And this is consistent with the fact that it is not the gas itself that is being raised but the apparatus as a whole, and the rest of the apparatus has a MUCH higher thermal mass that any of the gasses so the thermal mass of the gasses is a minor effect at best. What is *different* between the gasses is the amount of *time* that the temperature increases and the amount of time that the temperature decreases, coupled with the fact that the apparatus never comes even remotely close to equilibrium again until all the gas is gone (i.e., the apparatus is back to air-filled). That’s a detail that may not have been clear and again points out the difficulty of fully and completely describing with words what is best seen with the eye.
So, to be clear, the argon as well as the carbon dioxide and nitrous oxide all leave the vessel LONG before anything close to a new “greenhouse” equilibrium is established. The only thing approximating equilibrium that the apparatus ever experiences is when the gas is air in the container. So, because the gas residence time is *short* compared to the intrinsic time needed to measure the new equilibrium state the total amount of time the gas stays in the vessel is at least as critical as every other property of the gas. And the residence time of the argon in the vessel, as the figure clearly shows, is at least 10% less than the residence time of the other gasses, probable more, so a temperature increase of argon that is at least 90% that of carbon dioxide and nitrous oxide suggests that the less than but almost 10% contribution to the radiative effects of the IR gasses is almost certainly an invalid conclusion.
I’m going to try to send Anthony a copy of the paper to see if he thinks posting just this figure is within Fair Use so everyone can see what I think the problem is with this paper and why, indirectly, it in fact strongly bolsters the argument that measuring the radiative effect of gasses in this way is very difficult.
This again?
The equilibrium temperature difference (and the rate at which it is approached) occurs when the heat escaping balances the incoming power. Assuming that the incoming power is constant (which I doubt), the rate of escape depends on the thermal resistance between the heated body and the surroundings. Thermal resistance has (in these circumstances) three components: diffusive conduction, radiation, and convection. These work in parallel across any one interface, and in series between the heated object and the gas, across the gas, and between the gas and the container/surroundings.
Conduction in gases is low, the thermal resistance is high. A blanket works by stopping air moving so that only conduction is left as a transport mechanism.
Radiation depends on the difference in fourth powers of the temperature of the object and its surroundings. The surroundings are at nearly the same temperature as the object – the experiment is usually done in a warm room rather than open to the cold sky – and the small difference between the temperature of the room and the temperature of the gas warmed by radiation is small. In addition, the thickness of CO2 in a small bottle is tiny, compared to that of the full depth of the atmosphere, and even at 100% the CO2 is too thin to provide much of a barrier.
The third effect, convection, tends to dominate. Not only because the thermal resistance is much, much lower, but also because it is non-linear. Increases in radiative or diffusive resistance tend to be cancelled by a drop in convective resistance, that takes up the slack.
The thermodynamic properties of gases differ. Air and CO2 have different densities, specific heat capacities, viscosities, and thermal conductivities. The heat has to conduct across the static boundary layer into the bulk of the gas where it can convect, it has to convect to the top of the jar, and then it has to cross another static boundary layer into the jar (or the gas has to leak out of the jar through the cracks). The different physical properties of the gases affect all these processes.
If you want to avoid all the arguments about how much radiation CO2 absorbs at what wavelengths, there’s a very easy answer. Instead of using CO2, just use liquid water.
It’s a lot cheaper, easier to manage, easier to see what’s going on, is transparent to visible light, and incredibly absorptive to thermal infra-red. It’s about 20,000 times more powerful a greenhouse agent than air. Simply stick your thermometer in a jar of water and shine your light on it, and the radiated heat from it will be blocked from escaping entirely. Where you need metres of CO2 to have an effect, about 20 microns of liquid water will do.
Why muck about with marginal, difficult-to-detect effects? Why not go for the most easily available material with the most extreme case of the property you’re looking for? You’re after something transparent to visible light and that absorbs and re-emits thermal infra-red, and water’s got that in spades. 20,000 times the effect ought to make the experimental procedure really easy.
Merrick
I have not seen the experimental set-up. I just rely on your description of it. If the Argon gas is cooler than the air inside jar, the experiment is flawed from the start. There is no control because gas and air are at different temperatures. Obviously we cannot accurately replicate the atmospheric greenhouse effect in a jar. The best these lab experiments can do is detect LWIR absorption by CO2. It has been done long ago with more sophisticated equipment. Nye-like crude experiments are just for the entertainment of hobbyists.
Dr. Strangelove,
I wouldn’t say that the argon gas being cooler than the air inside the container invalidates the experiment. It just defines a different starting point for the experiment. All of the gasses are introduced at that similarly cooler temperature with respect to the container upon introduction. True, as you point out, the argon should heat faster because it has a lower heat capacity, but the heat capacity of all the gasses is substantially less than the heat capacity of the container, so this impact, while absolutely present, is a minor effector of the results that needs to be noted but it doesn’t invalidate the results.
In going over the paper again, the small difference in heating between carbon dioxide and nitrous oxide does seem to suggest they have accurately measured the radiative component, which clearly is at least an order of magnitude lower than the convective component, so this experiment really isn’t, as everyone has agreed, a way to demonstrate the radiative “greenhouse” effect in a simple classroom experiment. And, as Anthony said, Bill Nye is still not even wrong.
Michael 2 says:
August 15, 2014 at 4:28 pm
Backscatter is not re-radiation. CO2 absorbs infrared, presumably heats up slightly, and emits a photon of its own, possibly or probably at a different
The term photon can be misleading when used in general discussions as it implies something that is constant and unchanging when it is really a packet of energy whose magnitude depends on the frequency of the emitted radiation. Short wave or gamma rdiation gives rise to high energy photons as opposed to infra red or long wave radiation producing lower energy photons. Black body radiation comprises photons from across the spectrum but are they discreet and what distinguishes one quantum of energy from its neighbour. For instance a photon of frequency f1 with energy h*f1 has a neighbour with energy h*(f1+df1); how large is df1 is it f1*0.1 or f1*0.000001. In classical physics it was assumed that energy was taken up continuously which gave rise to the ultra violet catastrophe in physics during the late 19th C. Plancks Law solved this telling us that energy can only be taken up in discreet packets or quanta with each quantum having energy of h*f. Since h is a constant how large is the frequency ‘quantum’ or change between one photon and its neighbour? I for one don’t know but constantly remind myself that the term photon should be qualified by referring to it as a frequency photon or wavelength photon in discussions otherwise its significance can soon be lost.
I’ve read several just right stuff here. Certainly price bookmarking
for revisiting. I wonder how a lot attempt you place to make this sort
of fantastic informative website.