Guest Post by Willis Eschenbach
I’ve been wanting to [take] another look at the relationship between net top-of-atmosphere (TOA) radiation changes on the one hand and changes in temperature on the other. As a necessary prelude, I finally have gotten round to an oft-postponed task, that of looking at the thermal lag in different areas of the planet. This is the lag between a change in TOA radiation and a corresponding change in surface temperature. Figure 1 shows my results.
Figure 1. Lag in months between the changes in TOA radiation and the corresponding changes in temperature.
As always when working with CERES data, I find that every graph brings surprises. For example, the wet land areas respond more quickly to changes in TOA radiation than do the dry land areas. Not sure I understand that. Also, you can see the quick response in the Inter-Tropical Convergence Zone (ITCZ) just above the equator in the Pacific. This suggests that the water-thunderstorm cycle transfers the energy more readily to-from the surface, with less thermal lag.
Now, I admit that there are more accurate ways of calculating the lag. However, my method gives the lag to the nearest month, so the error won’t ever be more than half a month. Here is the relationship between raw temperature and TOA radiation at a randomly chosen point in the Atlantic:
Figure 1a. Scatterplots of net TOA radiation versus surface temperature at 35°N, 10°W. Letters in the graphs are the initial letters the months, JFMAMJJASOND. All trends given are in degrees C per 3.7 W/m2 of TOA radiation change (the change expected from a doubling of CO2). Click image for larger version.
As you can see, the actual best fit for the lag is slightly more than two months, but not as large as three months. However, we won’t get much error using the 2 month lag.
In any case, having that lag data, I was able to see what the trend was between temperature and TOA radiation given the lag in each cell. Figure 2 shows that result. I have given the trend in temperature per 3.7 watts per square metre (W/m2) increase in TOA radiation. This is the change in the amount of downwelling TOA radiation that the IPCC says will result from a doubling of CO2. In other words, the graph shows the immediately change in temperature that (might) theoretically result from a doubling of CO2 IF ALL ELSE STAYED UNCHANGED. Of course, things wouldn’t stay unchanged, because change is the rule, not the exception …
… in any case, Figure 2 below shows the relationship between TOA radiation and temperature, with the trend calculated using the gridcell-by-gridcell lag:
Figure 2. Change in temperature per 3.7 W/m2 change in net TOA radiation. All gridcell trends have been calculated using the thermal lag indicated in Figure 1.
This is the raw trend of the relationship between the net TOA radiation and the appropriately lagged temperature response.
However, we still have a problem. This is that because of the thermal lag, the surface temperature doesn’t get as high as it would if there were no lag and the thermal response were instantaneous. I discussed in Time Lags in the Climate System how we can estimate the reduction in amplitude due to the thermal lag. In general, as you might expect, the longer the thermal lag, the smaller the change in temperature.
Once I’ve adjusted the amplitudes upwards to allow for the reductions in amplitude caused by the thermal lag in each gridcell, I get the results shown in Figure 3. This is an estimate of the Transient Climate Response (TCR), which is the short-term response to an increase in net TOA forcing.
Figure 3. Best estimate, transient climate response (TCR) in degrees C per doubling of CO2 (3.7 W/m2 change in TOA radiation.) It differs from Figure 2 in that the amplitude has been increased as a function of the thermal lag.
As is expected, this adjustment increases the overall expected thermal response to the change in net TOA radiation. Note that this adjustment reduces the land/ocean difference, but without removing it entirely. I take this as support for the method being used.
So what can we learn from this? Well, first off, the transient climate response (TCR) value of 0.44°C per 3.7 W/m2 shown in Figure 3 is pretty small compared to the IPCC values. For the 18 climate models that reported results in the IPCC Fourth Assessment Report (AR4), the mean TCR is 1.8 ± 0.1 °C (std. err. of mean) per doubling of CO2. My results in Figure 3 are less than a quarter of their results.
Finally, how does the transient climate response (TCR) relate to equilibrium climate sensitivity (ECS)? The ECS is the long-term response to a change in TOA radiation after all succeeding adjustments have occurred. ECS is the sensitivity that people usually discuss.
Well, the relationship shown by the AR4 models linked above is that the ECS is about 1.82 ± 0.1 (std. err. of mean) times as large as the transient climate response (TCR). With a TCR from Figure 3 of 0.44 degrees C per doubling of CO2, that would put the equilibrium climate sensitivity at 0.8 ± 0.03 degrees C per doubling of CO2.
By comparison, the average of the 18 AR4 climate models linked to above is an equilibrium climate sensitivity of 3.2 ± 0.2 degrees C per doubling of CO2. This is about four times larger than my results.
Please note that I make no overarching claims regarding the accuracy of these measurements and estimates. They represent my best effort in my continuing quest to understand the relationship between the TOA radiation and the temperature.
Regards to all,
w.
Usual Note: If you disagree with someone, please have the courtesy to quote the exact words you think are incorrect so that we can all understand your objection.
Math Note: Using the method in my post linked above, the multiplication factors to increase the amplitude of the temperature cycle based on the individual gridcell thermal lags are 1.0, 1.7, 2.9, and 4.8 for lags of 0, 1, 2, and 3 months respectively. In other words, amplitudes are not increased for 0 months lag; for a one month lag, all amplitudes would be multiplied by 1.7; and so forth.
I have calculated the length of the lag by using the cross-correlation function (ccf) and selecting the lag with the greatest correlation. I understand that there are likely better ways to establish the more precise value, but using the ccf was quick and did what was required.
Data Note: I’m using the CERES EBAF dataset. For the surface temperatures I’ve converted the (calculated) upwelling surface longwave radiation dataset to the corresponding blackbody temperatures. The data starts in March 2000 and ends in February 2014.
Thanks, Willis.
Enjoy your holiday weekend.
The issue I have with this is the original assumption that TOA changes are directly correlated to temperature changes up to 3 months later. There is nothing in this article to explain why this is assumed or counter explanations for the evidence used to make this assumption.
Study fig. 1a a little.
Might the difference be in the proposed increase in water vapor creating more reradiated warmth than CO2 increase alone?
Yes, in my meagher understanding. I see how an increase in cosmic radiation due to the decrease in heliospheric density might excite more nucleic cloud formation at the TOA and cause an overall cooling. I think I’ve witnessed that from may location this past winter and spring.
I have often wondered if the duration of contrails from jets is only temperature and humidity dependant.
I also wonder if noctilucent clouds might be an indicator of cosmic radiation levels at the TOA. Can any body set me straight here?
That’s not what I meant. Willis calculates reduced sensitivity compared to IPCC. I am wondering if the major difference has to do with IPCC’s fudge factor related to increased water vapor due to increase evaporation from longwave radiation evaporating off the ocean surface.
He states, “Well, the relationship shown by the AR4 models linked above is that the ECS is about 1.82 ± 0.1 (std. err. of mean) times as large as the transient climate response (TCR).”
The model includes a water vapor response, thus amplifying the affects of increased CO2. I think it is possible Willis has shown that the amplification does not exist.
Water vapor does not have a positive feedback. Even IPCC gives clouds a negative feedback of -20 W/m^2, ten times opposite the 2 W/m^2 RF of CO2. Water liquid and vapor run the climate. GHGs are a bee fart in a hurricane.
The models include this water vapor amplification and there are studies that attempt to demonstrate it. Whenever I run across such studies I first look for reanalysis, a method that takes raw data and “improves” it. Next I look for modeled results presented to look like observations. Third I look for any attempt to rule out natural drivers of observed phenomenon. The last one is most concerning and reveals evidence of bias. Bias is something that should never see the light of day in the mind of a real scientist. Unfortunately, bias is more often than not the driver of research.
http://www.pnas.org/content/111/32/11636.full.pdf
Willis I have trouble following some of this, but thought I’d just ask, is your finding of ECS somewhere in the ballpark of Lindzen, Lewis Curry, Spencer, Michaels etc. And why do you think that is, in comparison to the IPCC models?
BTW the first sentence needs one less look to make sense.
[edited. .mod]
I greatly appreciate your analysis. One point: The statement shortly following Figure 3 “So what can we learn from this? Well, first off, the transient climate response (TCR) value of 0.44°C per W/m2 shown in Figure 3 is pretty small compared to the IPCC values.” left the 3.7 out of per W/m2. It is shown correctly elsewhere and a simple overlook, but needs correction.
[So, you recommend it be changed to “Well, first off, the transient climate response (TCR) value of 0.44°C per W/m2 shown in Figure 3 is pretty small compared to the IPCC values of 3.7 deg C W/m2. ” ? .mod]
No, he means it should be 0.44°C per 3.7 W/m2
Agree it should be 0.44°C per 3.7 W/m2 ?
Thanks, Leonard, fixed.
w.
I can’t make out the months that well in your charts but it looks like one group of months lag at three months (summer through fall?) and the other lags at two months (winter to spring?).
Thanks Willis, I really enjoy your work.
“…the trend in temperature per 3.7 watts per square metre (W/m2) increase in TOA radiation.” (3.412 Btu/h per W)
Out of what? ToA total 340 W/m^2 +/- ? 2? ? 10? 3.7 = 1.1%!!!
How can anybody claim that level of accuracy? And from space!! Much of my triple decades career involved measuring thermodynamic parameters on power plant equipment and systems. Up close & personal. ASME PTCs have entire sections on the treatment of inaccuracies and uncertainties
IMHO there is no way anybody can measure 3.7 W/m^2 at ToA or mm sea level changes or 0.5 C temperature anomalies with any significant certainty.
The limitation and misapplication of significant figures is a whole ‘nother story..
See what I wrote in a comment, just 2 and 1/2 weeks ago, on another blog, about the “climate sensitivity”:
on Judith Curry, climate sensitivity and climate policies
It is entirely obvious to me that there is no global-warming greenhouse effect due to increasing CO2.
The question always remains regarding the accuracy of CERES TOA measurements. The following Hansen paper discusses the problems:
http://pubs.giss.nasa.gov/docs/2011/2011_Hansen_etal_1.pdf
From the paper:
“The notion that a single satellite at this point could measure Earth’s energy imbalance to 0.1Wm−2 is prima facie preposterous. Earth emits and scatters radiation in all directions, i.e., into 4 steradians. How can measurement of radiation in a single direction provide a proxy for radiation in all directions? Climate change alters the angular distribution of scattered and emitted radiation. It is implausible that changes in the angular distribution of radiation could be modeled to the needed accuracy, and the objective is to measure the imbalance, not guess at it. There is also the difficulty of maintaining sensor calibrations to accuracy 0.1Wm−2, i.e., 0.04 percent. That accuracy is beyond the state-of-the art, even for short periods, and that accuracy would need to be maintained for decades.”
“The precision achieved by the most advanced generation of radiation budget satellites is indicated by the planetary energy imbalance measured by the ongoing CERES (Clouds and the Earth’s Radiant Energy System) instrument (Loeb et al., 2009), which finds a measured 5-yr-mean imbalance of 6.5Wm−2 (Loeb et al., 2009). Because this result is implausible, instrumentation calibration factors were introduced to
reduce the imbalance to the imbalance suggested by climate models, 0.85Wm−2”
The instrument “calibration” factor is pure baloney. All it does is artificially gives you the result the model suggests. What the hell kind of science is that? My take is that scientists do not even know what they are measuring in the TOA with all the emr going in all directions.
In my opinion this is not quite fair to Loeb et al.who concluded from their study that present state-of-the-art satellite technology lacks the precision to close the energy budget at TOA.
In 2012 an update was published in Nature Geoscience on the subject of uncertainties in the observations of solar and terrestrial energy flows (radiative flux). The authors stated:
“‘The net energy balance is the sum of individual fluxes. The current uncertainty in this net surface energy balance is large, and amounts to approximately 17 Wm–2. This uncertainty is an order of magnitude larger than the changes to the net surface fluxes associated with increasing greenhouse gases in the atmosphere (Fig. 2b). The uncertainty is also approximately an order of magnitude larger than the current estimates of the net surface energy imbalance of 0.6 ±0.4 Wm–2 inferred from the rise in OHC. The uncertainty in the TOA net energy fluxes, although smaller, is also much larger than the imbalance inferred from OHC.”
Graeme L. Stephens et al, An update on Earth’s energy balance in light of the latest global observations. Nature Geoscience Vol. 5 October 2012
URL:http://www.aos.wisc.edu/~tristan/publications/2012_EBupdate_stephens_ngeo1580.pdf
Further, the comment by Hansen was earlier. By 2011 Hansen had adjusted downwards his estimate of global energy imbalance to 0.6 W-m2 which was adopted by Stephens in 2012. in their 2012 paper, Stephens et al. cite Hansen’s figure +0.6 W-m2 for energy imbalance. This estimate of the energy flux into the oceans is from James Hansen et al. (2011) Earth’s energy imbalance and implications, Atmos. Chem. Phys., 11, 13421-13449, 2011). URL:
http://www.atmos-chem-phys.net/11/13421/2011/acp-11-13421-2011.pdf
In 2012 Loeb et al had reduced this to 0.5 W-m2. Norman G. Loeb, John M. Lyman, Gregory C. Johnson, Richard P. Allan, David R. Doelling,Takmeng Wong, Brian J. Soden and Graeme L. Stephens. Observed changes in top-of-the-atmosphere radiation and upper-ocean heating consistent within uncertainty. (Nature Geoscience Vol 5 February 2012)
URL: http://www.met.reading.ac.uk/~sgs02rpa/PAPERS/Loeb12NG.pdf
In effect, these papers taken as a whole are saying that, using satellite measurements as their basis of measuring radiation at TOA, the net global energy imbalance measured at the surface was not significantly greater than zero during the decade considered.
What does it all mean? Because of the lack of precision in measurements, the net incoming energy may be greater than, equal to, or less than the energy leaving the Earth. NASA cannot say for sure if the Earth is warming, remaining the same, or cooling. The observations are not precise enough to tell us for sure that some other climate phenomenon does not completely counteract the warming effects of greenhouse gases, including the warming effect of CO2.
Obviously, given the stance of the POTUS, these scientists cannot come right out and say, “Mr President, we don’t know whether the Earth is warming, cooling, or maintaining its thermal balance.”
But that is what their papers have said.
Great information, thank you.
Side note: The earth is exothermic at a rate of 44 terawatts, which I think is about .08 W/m^2. This is cooling the earth at a rate of about 100°C every billion years. This is 0.0000001 °C / year.
.5 W/m^2 / .08 = 5.81. So, the earth may be warming at a rate of 0.0000005 °C / year +/- ??.
Satellite measurements have been a great disappointment and hindrance to the warmists. It has never supported their models or assumptions. Most recently satellite data shot down Hansen’s hypothesis that the halt in global warming was due to increased atmospheric turbidity. The data has to be “improved” in some way to be acceptable to them, usually by running the data through a computer model. Here Hansen is shown adjusting straightforward satellite data to match his models rather than adjusting his models to match data because to do the latter would be to admit that catastrophic global warming is unlikely.
Willis, a question, do you think TOA is affected by the amount of water vaper?
“For example, the wet land areas respond more quickly to changes in TOA radiation than do the dry land areas. Not sure I understand that. ”
I think our questions are related, with dry land you’re heating rocks, wet land you’re boiling water. and I think it’s related to the temperature response to changes in air mass, it’ll change 10 – 20F as the air changes from originating in the Tropics to Canada and it does this in basically hours.
And I believe it based on the water vapor the air mass carries.
Even if you are correct, the planet is ~71% ocean.
“the wet land areas respond more quickly to changes in TOA radiation than do the dry land areas”
Perhaps because water vapor is the most active greenhouse gas, so the heat is more easily retained. Desert areas are known for getting very cold at night because the daytime heat radiates away to the clear night sky.
You are completely dismissing direct conduction and focussing entirely on radiation.
Willis,
Very interesting, but not yet convincing. Why should we assume that the slope of the line is the climate sensitivity? Why the ellipses in Fig 1a? Without understanding that, how can we know how to properly analyze the data?
One way to test this might be to apply the method to the output of a GCM and see if it gives the same sensitivity as known from the model.
>> Why should we assume that the slope of the line is the climate sensitivity?
I’ve also found it confusing because climate sensitivity descriptions/discussions usually link it directly to CO2. However, it’s actually defined the way Willis is calculating it. Climate Sensitiviy = dT/dR.
>> Why the ellipses in Fig 1a?
The ellipses are caused by the fact that although the radiation has dropped (August), the temperature are reaching their maximum. When he increased the time lag to 2 months, it matches up. This matches exactly my experience growing up on the shores of Lake Erie, as the Lake increases the time lag.
VikingExplorer,
OK, seasonal lag makes sense as the cause of the ellipses.
Yes, Climate Sensitiviy = dT/dR. But T is global average temperature, not local temperature, and R is radiative forcing, not TOA radiative imbalance. So my question remains: Why should we assume that the slope of the line is the climate sensitivity? I am pretty sure it is not, for reasons given below.
>> Why should we assume that the slope of the line is the climate sensitivity?
No need to assume anything. It’s a graph of T vs R. Climate Sensitiviy = dT/dR. If you’ve ever taken Calculus 1, you would know that the first derivative is the slope of the line.
Thanks Willis!
Good stuff once again
Question,
How do we determine the saturation point with respect to IR, more specificaly LWR and CO2?
I think RGBatduke spoke of such once.
It seems important to the energy budget, hence my curiosity.
I’m not clear what you mean by the “saturation point”. Absorption by well-mixed GHGs doesn’t get “saturated”, although it’s a common misconception.
w.
Typo in the second paragraph after Fig.2: This is that the because of the thermal lag, the surface temperature doesn’t get as high as it would if there were no lag and the thermal response were instantaneous.
I think the emboldened text needs to be deleted.
[Fixed, thanks. ~mod.]
Wet things conduct heat better than dry things. Take something hot out of the oven with a dry mitt and then try it with a damp mitt. Wetlands usually consist of shallow water. Wet mud will be warmer than dry mud.
Very interesting and educational post. It does make sense that more water in the atmosphere would facilitate Heat transfer, because water is a good conductor of heat.
Also, the plots basically indicate that the heat of summer usually comes a bit later than peak radiation. In other words, as we all intuitively know, June 21 is not the hottest day of summer. It being as long as 2+ months later is probably due to picking a spot in the ocean. Land heats up quicker, as every sailor knows.
The climate sensitivity discussion was very interesting. Of course, the controversial part is whether CO2 would actually cause the 3.7, because you say, all other things are not equal.
Willis,
I have a problem with your use of net TOA radiation. In a simple radiation balance model, we have
dH/dt = F – (lambda*delta_T)
where F is the forcing and (lambda*delta_T) is the change in outgoing IR due to a change in T. So if you want to get lambda, you need to plot outgoing IR vs. T.
For the planet as a whole, net TOA radiation is equal to dH/dt so one could, in principle, plot net TOA radiation vs. T to get lambda, but that only works for the planet as a whole. At any given location there will also be horizontal transport that is not included in the above equation, so the equation is no longer valid.
Horizontal heat transport will generally be from hotter areas to cooler ones. That will reduce the temperature of the former and increase the temperature of the latter, thus compressing the temperature range and making the slopes of your plots lower than they otherwise would be. So I think your estimate of TCR is only a lower bound.
Mike M,
Your question is highlighting an issue that many of us have discussed before. You are modeling the earth as a flat surface heated by a cool sun. Many of us have been the arguing that the real situation is a rotating/spinning earth heated by a hot sun.
Willis is taking empirical data and very elegantly showing what the climate sensitivity must be. During summer, the data shows that incoming radiation exceeds outgoing, which naturally causes the surface to heat up. The opposite in winter.
Radiation balance doesn’t matter because planets have a tremendous amount of energy. Mercury is warming up while Jupiter is cooling off, but ever so slowly. It would take eons to achieve radiation balance.
VikingExplorer,
“You are modeling the earth as a flat surface heated by a cool sun.” That is ridiculous. I am doing no such thing and can not imagine why any sane person would do that.
“Many of us have been the arguing that the real situation is a rotating/spinning earth heated by a hot sun.” Of course that is the real situation and that is what is in all climate models.
“During summer, the data shows that incoming radiation exceeds outgoing, which naturally causes the surface to heat up.” Yes, but that has at best a tangential relation to climate sensitivity.
“Radiation balance doesn’t matter”. Nonsense. In any case, Willis’s entire calculation is based on radiation balance.
>> Nonsense. In any case, Willis’s entire calculation is based on radiation balance.
What law of nature is inducing radiative balance?
Why haven’t Mercury or Jupiter achieved it, after all this time?
How did Willis use “radiation balance” to compute the values for Figure 1?
I didn’t say anything about climate models, since I have no first hand knowledge of them. I was reacting to an implication of what you wrote.
Take two numbers 1 and 3. The average of 1 and 3 is 2. 2 ^ 4 = 16. However, 1^4 = 1 & 3^4 = 81. The average of those two numbers is 41. IOW, any sane person realizes that (avg T1-n) ^ 4 is not equal to avg( T1^4 + T2^4…)
And yet you wrote with apparent insanity:
I was giving you the benefit of the doubt to say you just modeled the earth wrong. The alternative is that you are really ignorant of math.
No, Willis did it exactly correct. Only Net TOA radiation can effect temperature. You are claiming that the water level of a lake is dependent on incoming water only. The reality is that only Input – Output can affect the water level.
>> Yes, but that has at best a tangential relation to climate sensitivity.
Nonsense. Climate sensitivity is defined exactly the way Willis is calculating it.
The horizontal transport question is valid. Willis admits it’s all pretty rough but just to get a ball-park figure. One may argue that it will lower sensitivity in tropics and boost it in temperate regions. In an ugly, rough as guts, approach it *could* all average out. That would need some justification…
The reason that there are loops is the phase lag, as others have said. If if was a simple lag there would be thin ellipse for 2 and 3 months. The fact that it’s twisting is because it’s not that simple.
What happens is that there is an in-phase relationship between forcing ( rad ) and dT/dt, ie temp and rad are orthogonal. The induced temp increase ( δ T ) at any point in time is the integral of rad(t) . Once the system starts to adjust neg. feedbacks kick-in. This is the increase in outgoing LWIR.
Assuming ( as climatologists always do ) that this is linear system this would lead to an exponential response to an changes in radiative forcing. Now you can’t straighten out an exponential with a simple lag. Spencer and Braswell (2011) discusses the problem of separating the two effects, without knowing how to do it.
The problem is that there are complex reactions that are orthogonal mixed in together. Just aligning the peaks, which is what Willis’ method if effectively doing, is not going to distinguish them, and thus the answer is not what he interprets it as being.
Douglas and Knox also tried this Lissajou figure approach, paradoxically in a paper where they also did a very thorough mathematical exponential calculation. They apparently missed the contradiction. More detail on all of this in the following link:
http://judithcurry.com/2015/02/06/on-determination-of-tropical-feedbacks/
The above article shows how this exponential reaction can be fitted to tropical TOA radiation budget around Mt. Pinatubo eruption.
It finds the tropics to be very insensitive to radiative change.
I don’t think you understood my comment. I’ll try once more.
You are probably referring to the fact that Rout = k Tout ^ 4. This implies that if the earth was a flat surface, there would be a parabola shape in a graph of radiation vs TOA temperature.
However, the earth isn’t flat. The orbit, along with the tilt, result in a large amount incoming radiation in the summer, and a large amount radiation going out in winter.
So, although TOA Radiation-out is changing like a parabola, it’s being subtracted from a much larger sinusoidal like input value. The NET difference is the ellipses in figure 1a.
You write “The induced temp increase”, but it’s important to never reverse cause and effect with SB. Temperature causes radiation, not the other way around. The characteristics of the thermodynamic system determine it’s temperature, and whatever temperature the TOA happens out to be, it radiates at that temperature according to SB. However SB is not a substitute / shortcut to analyzing the thermodynamics of the system using the principles of classical thermodynamics.
moderator, can you please delete the last post since I missed a end blockquote.
[done -mod]
There is no law of nature inducing radiative balance. Radiative balance is just a consequence of either gaining or losing energy. The temperature of earth reflects the total Energy of earth, not the rate of change in energy. AGW folks assume no time delay to achieve radiative balance, but in fact, Mercury and Jupiter have been working on it since they were formed.
Mike,
Thanks for the discussion of the phase diagram. I will have to think about it. I agree that one can not properly interpret the diagram without having some idea of how to untangle the effects.
But as the the horizontal transport, I must disagree when you say “One may argue that it will lower sensitivity in tropics and boost it in temperate regions. In an ugly, rough as guts, approach it *could* all average out.” That might well be the case for averages. But for slopes, horizontal transport causing hot areas to get cooler and cold areas hotter are changes in the same direction: towards smaller slopes. So the errors will not average out.
> Also, you can see the quick response in the Inter-Tropical Convergence Zone (ITCZ) just above the equator in the Pacific.
I don’t get it. Your heretic theory as I have understand it is that tropical ocean climate is regulated by emergent effects of thunderstorms and squall lines and that heat transfer from the tropics to the poles is much more efficient than what the models show.
Why would we accept a thin line of quick response, surrounded by significantly less efficient response in the rest of the tropics?
amplitudes are not increased for 0 moths lag
shurely shome mishtake?
Lagging moths? Thanks, fixed.
w.
Willis
Your above analysis of CERES satellite data says
Your finding shows remarkably good agreement with the finding of Sherwood Idso from his eight “natural experiments” that he published in 1998.
Idso’s paper is here and says
A similar result has also been obtained by
Lindzen & Choi from ERBE satellite data
and by
Gregory and from balloon radiosonde data
http://www.friendsofscience.org/assets/documents/OLR&NGF_June2011.pdf
Richard
Mods
I twice attempted to make a post but it vanished both times. If you find it when you check the ‘bin’ I would be grateful if you post the first version and delete the repost.
And if you don’t find it in the ‘bin’ please be so kind as yo say so I can post it again.
Thanking you in anticipation
Richard
One of my earlier posts also vanished. Don’t want to double post so I’m waiting.
Thank you for some very interesting graphs Willis
However, the question is how fast you see the effect of the imbalance in the TOA radiation.
In IPCC’s definition of transient climate response, they use a gradual increase in greenhouse gases over a period of 70 years as a measure. That means that 70 years is their short-term effect. The equilibrium occur approximately 100 years later.
You operate with only a few months, and then the sensitivity will of course be smaller.
/Jan
Willis writes “As always when working with CERES data, I find that every graph brings surprises. For example, the wet land areas respond more quickly to changes in TOA radiation than do the dry land areas. ”
My guess is the rapid and efficient transfer of energy to the TOA by water vapour’s latent heat of vaporisation works better over land because there is enough water there to do it but not so much as to cause the cloud cover that feeds back against this process.
Willis,
I have made a similar calculation using the seasonal variations for the two hemispheres. The climate sensitivity in my calculation ends up even lower than yours. Please have a look in this paper to see if you can spot the difference: http://www.klimatupplysningen.se/wp-content/uploads/2014/11/seasonal_variations_0_6.pdf
Changes in TOA radiation vary little for the globe as a whole so the above article is primarily of interest for seasonal changes.in individual regions.
For climate change across decades and centuries what matters is changes in the proportion of TOA insolation able to reach the surface as global cloudiness and global albedo vary.
Downward radiation is a bit of a red herring. What matters is either TOA insolation or the proportion that reaches the surface.
Downward radiation fails to warm the surface or reduce the rate of cooling of the surface and so has zero net thermal effect at the surface.
The reason is that the rate of photon emission declines relative to temperature as one descends trhrough the mass of an atmosphere.
The greater the density the more collisional activity occurs instead of photon emission so downward directed IR is statistically less likely to be re-emitted as a photon every time it is reabsorbed by another molecule during its descent into greater density.
The net effect is that all downward IR is absorbed into collisional activity before it reaches the surface and its energy then causes an enhancement of convective overturning whereby that downward IR becomes potential energy (not heat) within the convective columns of ascending and descending air.
That is one reason why Willis’s Thermostat Hypothesis, although incomplete, being limited to the tropics, hints at a fundamental truth.
Convective activity does indeed negate radiative imbalances but since the hydrostatic balance of the atmosphere is mass dependent any effect from GHGs could never be measured.
Hydrostatic balance involves the upward pressure gradient force matching the downward gravitational force in order to keep the mass of an atmosphere suspended off the surface.
The energy required for such balance is collisional activity derived from kinetic energy at the surface.
That kinetic energy is prevented from radiating to space by the declining amount of photon emission as one descends trhrough the mass of the atmosphere and the evidence of that decline is the dry adiabatic lapse rate slope.
Everyone thinks that a surface at 288K beneath an atmosphere radiates at 288k but it does not because of the photon emission constraint caused by mass density.
Earth’s surface actually emits photons at a rate commensurate with 255K.
GHGs do try to interfere with that 255K rate of emission but their blocking effect is negated via convective adjustments that I have described elsewhere.
Essentially GHGs block certain wavelengths from escaping to space but the energy so blocked is converted to enhanced convective overturning in which it becomes potential energy rather than sensible heat.
It reappears as sensible heat at the surface beneath the nearest descending column, is converted back to the full range of wavelengths and leaves to space from the surface past the GHGs that initially blocked it.
The rate of convective overturning increases until the effect of GHGs is negated but the effect is too small for us ever to measure it as compared to solar and oceanic effects.
Good to hear from you Stephen.
EM Smith has a post you would probably be interested in.
https://chiefio.wordpress.com/2015/05/20/polar-night-jet-and-uv-driven-albedo/
“the wet land areas respond more quickly to changes in TOA radiation than do the dry land areas”
Hmm… while clearly visible in the tropics, this statement does not hold for the temperate zones on your map. From the N of Oregon up to southern Alaska for a width of 100 to 300 miles you have a rain forest that is almost never dry… yet it does not demonstrate any distinction in your map – nor does any of the temperate zones demonstrate any kind of variation.
That is because the temperature is low. Absorption and emission are temperature related. Lower temperature means lower absorption/emission. Kirchoffs Law variant. That is a standard calculation with NASA and most scientific instrument makers today.
I really have to get off my lazy @ss and post about this. There seem to be a lot of people who are confused about how emission and absorption works.
“N of Oregon up to southern Alaska….”
I think Willis was referring to tropical rain forests. A forest in a regions where it rains a lot is not the same thing and is a very different situation.
Right, I think Alex is right. Rout = k Tout ^ 4. Therefore maximum radiative loss occurs at high noon, in the summer time. Maximum convective/conductive loss occurs when there is maximum atmospheric instability and maximum water content. Therefore, it seems to make sense that the maximum heat transfer / minimum response time would be where the data indicates.
“Therefore maximum radiative loss occurs at high noon, in the summer time. ”
Peak surface temp is about 5:00-6:00 pm, not noon.
Carry on 🙂
Right, good point, I stand corrected.
And actually that would give a delay from max incoming solar of about 6 hours, noon could be max rate of change though, I’d have to look.
Max cooling rate is say 10pm or so, and continues until sunrise. And for surface temps it warms in summer due to the difference between the length of day vs night. In the northern hemi the max rate of temp change is in March or April, max cooling rate is about October.
Willis, What is tthe effect of changes in irradiation between the Earth’s perihelion and aphelion of 22 watt/m2 on lag estimates? Your article:
http://wattsupwiththat.com/2014/10/25/changes-in-total-solar-irradiance/
Willis is satisfied with an average on everything. I have nothing against him. It’s just the way it is.
Alex May 24, 2015 at 2:19 am says:
In fact, I dislike averages, and have stated so repeatedly. I use them only as and when I have to. And I have nothing against me either. It’s just the way it is.
w.
Tony, no reason to bring that up again. It was all cleared up here
Tony May 24, 2015 at 2:14 am Edit
I don’t think that the perihelion/aphehelion changes affect the lag. The lag is a function of the material making up the surface, and is not a function of instantaneous insolation.
w.
Willis: Perhaps some questions will prompt a deeper understanding of your data:
1) Why does a radiative imbalance exist at 35N,10W or any other location at most times of the year?
The location doesn’t warm up or cool off as quickly as seasonal changes in incoming radiation. (If it was always in equilibrium, there would never be an imbalance.)
2) What controls how fast or slow a location warms up or cools off in response to changes in incoming energy?
The location’s heat capacity. (It can’t be ECS, because the E refers to equilibrium and 35N,10W is has a radiative imbalance almost all of the year. It can’t be TCR, because this refers to the temperature response to a 3.7 W/m2 change in radiation over 70 years, not tens of W/m2 change in one month.)
3) What is the heat capacity at 35N,10W?
About 10,500 kJ/m2/K for the atmosphere. We know that seasonal changes in the ocean temperature can be detected down to about 100 m, because surface winds and waves physically mix the top of the ocean. The greatest change is on the surface. On the average, the effective depth of the mixed later is about 30-50 m. The total heat capacity is 125,000-209,000 kJ/m2/K.
4) How fast can a +100 W/m2 radiative imbalance warm (or a -100 W/m2 cool) this much air and water?
100 W/m2 is 263,000 kJ/m2/month. A 100 W/m2 imbalance will change temperature by 1.2-1.9 degK/month.
5) Approximating, every year 35N,W experiences about 3 months of winter with a -100 W/m2 imbalance, three months of summer with a +100 W/m2 imbalance, and six months of spring and fall with much less imbalance. How much should we expect temperature to rise and fall every year?
3.6-5.7 degK, with the larger value for the 30 m mixed layer.
6) Don’t you think this is a far better explanation than climate sensitivity for why 35N,10W warms and cools about 7 degK every year? Do seasonal temperature changes and radiative imbalances have anything to do with climate sensitivity? (You might want to think in terms of the climate feedback parameter, the reciprocal of climate sensitivity. When surface temperature rises 1 degK, how much does OLR (+reflected SWR) increase. The radiative imbalance at any particular location at different times of the year is not the same thing as the change in OLR + reflected SWR.)
You answer this question.
Frank, your numbers look fairly plausible to me. However, when you say:
You missed the point. Climate sensitivity is not a cause or an explanation of this warming. The cause of this response is exactly what you described. Climate sensitivity is a defined emergent property of the climate.
Climate Sensitivity is defined as temperature response to changes in radiative forcing, where radiative forcing is the NET TOA radiation imbalance.
The seasonal temperature swing is a perfect opportunity to measure it.
VE, You should like this then, it’s the seasonal slope calculated by year and then plotted.
Lots of data based on NCDC GSoD here http://sourceforge.net/projects/gsod-rpts/
Some stale words on it here http://www.science20.com/virtual_worlds
VinkingExplorer and Willis: Climate sensitivity is the temperature response to a radiative forcing, not to a radiative imbalance. Equilibrium climate sensitivity is the temperature response after the planet has responded to a forcing so that a radiative imbalance has been reduced to zero. You are confusing a radiative forcing with a radiative imbalance.
The current forcing from anthropogenic GHGs and aerosols is about 2 W/m2, but the current global average radiative imbalance is about 0.4 W/m2. If we wish to calculate an approximate TCR from this data, we use the forcing (a little more than half of a doubling of CO2), not the current imbalance, The result is about 1.3 degK, not 5 degK!
Climate sensitivity is the equilibrium (ECS) and near-equilibrium (TCR) response that develops over several decades to a forcing. Seasonal changes are associated with massive radiative imbalances and therefore have nothing to do with climate sensitivity.
“Seasonal changes are associated with massive radiative imbalances and therefore have nothing to do with climate sensitivity.”
But you can use the changing ratio of day to night to show surface response due to a change in forcing. It’s the only way we can shutter the Sun, and have a way to actually measure the balance point.
Frank, I’m aware that some people claim that Radiative Imbalance = Radiative Forcing + Radiative Feedbacks.
However, as someone who strives to approach science agenda-free, without any preconceived notions, I consider this breakdown into hypothetical components to be tied inexorably to the AGW hypothesis. A valid analysis needs to start with first principles.
Analogy: In discussing a car traveling down the road, I note that speed = dP/dt. A critic says: no, we define speed to be composed of scalar velocity + the extra speed provided by a horde of invisible flying insects pushing with all their might.
Wikipedia defines radiative forcing as the difference of insolation (sunlight) absorbed by the Earth and energy radiated back to space
Radiative imbalance = Radiation coming in minus radiation going out. IOW, it’s the same thing.
“Feedback” is a term from control systems, which I’m trained in (I once implemented the control systems for a locomotive). However, AGW folks seem to be using the term in order to add FUD and overwhelm people with artificial complexity. In a closed loop control system, the output signal is measured, processed and brought back as “feedback” to be combined with the input. In this context Error = Input – Feedback (terminology varies). The error signal is then used to adjust the output signal.
The problem is that the concept of “Feedback” is exclusively associated with closed loop control systems. Unless some divine inspiration tells us conclusively that the Earth has been designed, we have to assume that earth is an open loop system. Note that an open loop system has NO feedback. [I’m speaking conceptually, which does not imply that there aren’t phenomena that increase or decrease the output]
The bottom line is that the input into our climate system is Radiative Imbalance. Since radiation is the only method of heat transfer affecting the earth, only radiative imbalance can result in heating or cooling. If we arbitrarily define the output as the surface temperature of air, and then define CS as dT/dR, then we can measure that property just the way that Willis did.
VikingExplorer says, May 25, 2015 at 1:18 pm:
“Wikipedia defines radiative forcing as the difference of insolation (sunlight) absorbed by the Earth and energy radiated back to space
Radiative imbalance = Radiation coming in minus radiation going out. IOW, it’s the same thing.”
No, it isn’t. Frank is correct. The increased ‘radiative forcing’ is causing the ‘radiative imbalance’. The Earth system will have to counter the increase in forcing to restore radiative balance. This it can only do by warming.
Ultimately, a strengthened ‘radiative forcing’ manifests itself at the surface as an intensification of atmospheric DWLWIR. This occurs because, as the atmosphere’s IR opacity increases from us putting more and more CO2 into it, the ‘average depth of atmospheric downward radiation’ is becoming shallower, that is, the DWLWIR is coming from on average warmer layers of air. This development is parallel to the raising of the ERL (Earth’s ‘effective radiating level’ to space), which is basically the opposite depth, the ‘average depth of atmospheric upward radiation’. The more opaque the atmosphere gets to outgoing terrestrial IR, the deeper into the troposphere this level will be situated, and hence the outgoing radiation will be coming from on average cooler and cooler air layers.
At least, this is how the theory goes …
Kristian, you wasted a comment, because I covered that in my first two paragraphs. Inexorably tied.
>> The Earth system will have to counter the increase in forcing to restore radiative balance.
Also covered this with by asking in other comments “What law of nature is inducing radiative balance?”
Earth is not in radiative balance, there is no closed loop control system, and there is no intelligent controller.
Viking Explorer wrote: “Wikipedia defines radiative forcing as the difference of insolation (sunlight) absorbed by the Earth and energy radiated back to space”
If you look carefully at the Wikipedia page for radiative forcing, a reviewer has added “citation needed” to the definition you quote – a warning that the author of this page may have made up this definition rather than obtaining it from a reliable source. Check the IPCC’s definition a little further down the page:
“Radiative forcing is a measure of the influence a factor has in altering the balance of incoming and outgoing energy in the Earth-atmosphere system and is an index of the importance of the factor as a potential climate change mechanism. In this report radiative forcing values are for changes relative to preindustrial conditions defined at 1750 and are expressed in Watts per square meter (W/m2).”
In other words, radiative forcing is the CHANGE in incoming and outgoing energy caused by a FACTOR that has changed since preindustrial conditions. A familiar “changing factor” or “forcing” is the doubling of CO2, but there are many other GHGs and aerosols that have changed. “Forcings” cause climate to change by directly effecting the rate at which radiative energy enters or leaves the planet. Forcing will gradually change the planet’s temperature, but forcing refers to the change in flux of energy entering and leaving the planet – not to temperature change and all of the consequences (feedbacks) that follow.
How does one convert a change in CO2 – usually measured in ppm – to a “forcing” measured in W/m2? Using the absorption spectrum of CO2 measured in the lab, observed temperature, pressure and composition of the atmosphere (which changes with altitude), and the observed types and distribution of clouds, one CALCULATES how the LWR flux from the surface to space will be changed by a doubling of 2XCO2 (or by some other forcing) – ASSUMING nothing else changes. The calculated change is caused by both absorption and emission of LWR. (One also calculates the change in the flux coming from space to the surface, which is zero at the edge of space and starts growing due to emission from CO2). The best calculations say doubling CO2 will reduce the net outward flux by 3.7 W/m2. So a radiative forcing is a CHANGE in the amount of GHG or aerosol since 1750 converted into units of W/m2 by means of a calculation. (One can treat a change in solar output as a forcing, but it must be divided by 4 – ratio of the surface area of a circle to a sphere – and multiplied by 1-albedo, before comparing to other types of forcing.)
A radiative imbalance between incoming and outgoing radiation is OBSERVED by satellites from space (or inferred from a change ocean heat content OBSERVED by ARGO).
A radiative forcing is CALCULATED from observed changes in GHGs and aerosols – ASSUMING NOTHING ELSE CHANGES. In the real world, things change: temperature changes, that temperature change will change absolute humidity, clouds, lapse rate, and surface albedo. These feedbacks also effect incoming and outgoing radiation.
Both forcing and feedback effect the rate energy enters and leave the planet: Forcing is caused by a change in GHG or aerosol measured in W/m2. Feedback is the change in rate at which energy enters and leaves the planet caused by a change in surface temperature. Feedback is measured in W/m2/degK. With different units, you can’t add them together.
The earth isn’t “designed” to have feedbacks. The fact that warmer air can hold more water vapor and water vapor reduces the rate at which OLR escapes to space causes water vapor feedback. The fact that a warmer planet will have less ice and snow on its surface, which will reflect less SWR, causes ice-albedo feedback. Putting a microphone to close to a speaker carrying the microphone’s output produces feedback.
A simple calculation shows that the earth that behaved like a blackbody would have to warm about 1 degC to increase OLR by 3.7 W/m2, the net change in radiation expected for 2XCO2. If the feedbacks from that 1 degC of warming further reduced OLR by 1.85 W/m2/degK, that would cause a further 0.5 degK increase in surface temperature. That 0.5 degK would reduce OLR by 0.925 W/m2, thereby reducing temperature 0.25 degK, etc. If you sum the infinite series the total warming is 2 degK.
VikingExplorer says, May 25, 2015 at 5:10 pm:
“Kristian, you wasted a comment, because I covered that in my first two paragraphs. Inexorably tied.”
You apparently didn’t read my last line: “At least, this is how the theory goes …”
Within the framework of the rGHE hypothesis, Frank is correct. There is a difference between ‘radiative forcing’ and ‘radiative balance’. And within the framework of the rGHE hypothesis, there is a radiative balance that has to be maintained.
… between ‘radiative forcing’ and ‘radiative IMbalance’.
Frank says, May 25, 2015 at 8:30 pm:
“How does one convert a change in CO2 – usually measured in ppm – to a “forcing” measured in W/m2? Using the absorption spectrum of CO2 measured in the lab, observed temperature, pressure and composition of the atmosphere (which changes with altitude), and the observed types and distribution of clouds, one CALCULATES how the LWR flux from the surface to space will be changed by a doubling of 2XCO2 (or by some other forcing) – ASSUMING nothing else changes.”
This is the problem. You cannot assume this. There is no way you can expect that all other forcings within the Earth system remain unchanged while the rise in CO2 is allowed to quietly increase the total forcing on the surface, moreover, that all other – non-radiative – processes remain statically in place as well. Just have a look at the CERES data for the global surface between 2000 and 2014/15. The absolute atmospheric content of CO2 increased by 8.4%, or about 26% of the total rise since the mid 19th century. Total global tropospheric WV + cloud cover fraction/cloud total water content went up significantly as well. But the total average atmospheric DWLWIR to the global surface still went down, not up. All because of cloud distribution. Clouds simply appear to control more or less completely the changes in total ‘radiative forcing’ in the Earth system. Furthermore, the radiative cooling ability of the Earth’s global surface – its ‘net LW’ – strengthened considerably (by ~1.5 W/m2 in 15 years!) over that same period of time. In other words, there is no hint anywhere of any tightening up of atmospheric radiative insulating effect on the surface. No “enhanced greenhouse effect” to be observed. On the contrary! The OLR at the ToA simply and slavishly tracks tropospheric temps over time, clearly an effect, not a cause.
Kristian wrote: “This is the problem. You cannot assume that nothing else changes [when CO2 doubles]. There is no way you can expect that all other forcings within the Earth system remain unchanged while the rise in CO2 is allowed to quietly increase the total forcing on the surface, moreover, that all other – non-radiative – processes remain statically in place as well.
You fail to understand that the amount of forcing (change in energy entering and exiting) for a change in CO2 using laboratory-validated radiative transfer equations can’t be calculated without making this assumption! This calculation REQUIRES you input how temperature, density, composition, and clouds change with altitude. You can do a simplified version of these calculations (MODTRAN) at this website in a few minutes:
http://climatemodels.uchicago.edu/modtran/
This is why climate scientists go through such a convoluted process: GHG change, forcing change assuming no feedbacks, temperature change assuming no feedbacks, feedback (in W/m2) from no-feedbacks temperature change, temperature change from first round of feedback, feedback (W/m2) from two rounds of temperature change, second round of feedback and third temperature change … and sum the infinite series.
Radiative transfer calculates the forcing from a change in GHGs and aerosols without feedback. Only climate models attempt to calculate what happens in the presence of feedbacks – and climate models require parameters to describe what happens on a sub-grid cell scale: cloud formation, precipitation, convective towers, evaporation, turbulent flow …. Radiative transfer equation have been validated in the lab and are trusted. AOGCM’s can’t be validated in the lab and the chaotic behavior of weather and climate make them impossible to properly validate in a few decades. So climate scientists still rely on the forcing/feedback concept. The effect of some feedbacks can be studied from space.
Frank,
The models don’t seem to realise that increasing density beneath the height of downward photon emission from a CO2 molecule progressively reduces the probability of a further photon emission as successive re- absorptions and re-emissions occur during the descent.
Bit by bit, from one CO2 molecule to the next that increasing density progressively absorbs DWIR fom GHGs into increasing collisional activity with depth.
The energy of DWIR from GHGs becomes incorporated into the vertical temperature profile as potential energy within convective overturning.
It never reaches the surface as radiation.
Instead of warming the surface or reducing the rate of cooling at the surface it simply distorts the lapse rate slope to the warm side which results in enhanced speed for the whole convective cycle. We can see that especially well with the moist adiabatic lapse rate slope but the same principle applies for any GHG or aerosol.
What is measured by pyregeometers and interpreted as downwelling radiation from the sky is simply the temperature along the lapse rate slope that the instrument is set to focus on. That is dependent on optical depth.
Under a clear sky they measure a cold temperature high up. Under a cloudy sky they measure the warmer lower temperature at cloud height. They do not really measure DWIR at all.
The models are incorrect to assume DWIR as having any effect on the surface in the first place. All DWIR does is enhance convective uplift and overturning from all points within the vertical column and not at the ground.
The error arises from the all pervasive mistake of assuming that a surface at 288K beneath an atmosphere of 1 bar pressure releases photons at the same rate as such a temperature at the top of the atmosphere.
In fact the surface still radiates to space at 255K and the other 33K of the surface kinetic energy is engaged in constantly maintaining the hydrostatic balance of the atmosphere. That 33k is locked in and cannot radiate out.
Until that is factored into the models they will never reproduce the extraordinary atmospheric stability observed in nature.
Steve Wilde wrote: “The models don’t seem to realise that increasing density beneath the height of downward photon emission from a CO2 molecule progressively reduces the probability of a further photon emission as successive re- absorptions and re-emissions occur during the descent.
Bit by bit, from one CO2 molecule to the next that increasing density progressively absorbs DWIR fom GHGs into increasing collisional activity with depth.
The energy of DWIR from GHGs becomes incorporated into the vertical temperature profile as potential energy within convective overturning.
It never reaches the surface as radiation.”
Pure fantasy. CO2 molecules don’t know anything about their surroundings or energy flux though the atmosphere. In the troposphere and most of the stratosphere, they are in local thermodynamic equilibrium, meaning that the fraction of CO2 molecules in an excited vibrational state depends on the local temperature, not the local radiation. They emit photons at a rate proportional to their density, absorption cross-section and B(lamba,local_T). They absorb a fraction of the incoming photons which is proportional to their density (n) and absorption cross-section (o). In mathematical terms, the Schwarzschild equation:
dI/ds = n*o*B(lamba,T) – n*o*I_0
Radiative transfer calculations apply this eqn to the outward and downward flux of LWR through the atmosphere. Those calculations don’t produce the situation you describe, but do produce the situation scientists measure in the field. (You claim they don’t understand how their instruments work, so their measurements are wrong.)
Steve Wilde wrote: “In fact the surface still radiates to space at 255K and the other 33K of the surface kinetic energy is engaged in constantly maintaining the hydrostatic balance of the atmosphere. That 33k is locked in and cannot radiate out.”
More fantasy. Scientists have studied the influence of pressure on the absorption cross-section of the lines in the spectrum of CO2. Laboratory measurements show how the cross-section (o) used in the Schwarzschild equation varies with pressure and temperature. Anything that is “locked in” at 1 atm pressure is already included in the measured cross-section.
Frank,
The correct way to see the situation is to recognise that on average every radiating molecule from surface to space is radiating at 255K as per the S-B equation.
Every such molecule then carries an additional load of internal energy ‘worth’ 33K. That additional energy is derived entirely from conduction.
For molecules at the surface that 33K is in the form of sensible kinetic energy and at the top of the atmosphere that 33K is in the form of non-sensible potential energy.
The DALR marks the point of transition between kinetic and potential energy up the entire vertical column.
Any molecules that find themselves too warm relative to their height will rise and any that are too cool will fall. The hydrostatic process makes that inevitable.
If radiative molecules become warmer than 255K then they rise and in the process cool both by additional radiation to space and conversion of KE to PE until they are radiating at 255K once more.Whilst rising they distort the DALR to the warm side.
If radiative molecules become cooler trhan 255K then they fall and in the process warm both by additional absorption of IR from the surface and conversion of PE back to KE until they are radiating at 255K once more. Whilst falling they distort the DALR to the cool side.
It is the concept of radiative molecules ever being able to upset thermal equilibrium that is a fantasy.
Steve WIlde wrote: “Any molecules that find themselves too warm relative to their height will rise and any that are too cool will fall. The hydrostatic process makes that inevitable.
If radiative molecules become warmer than 255K then they rise and in the process cool both by additional radiation to space and conversion of KE to PE until they are radiating at 255K once more.Whilst rising they distort the DALR to the warm side.”
What you ignore is that kinetic energy can be transported by collisions without any molecules converting KE to PE or back. Solids conduct heat via collisions without any molecules changing places and swapping KE for PE. The hypothesized thermogravitation effect assumes that vertical energy flux in a gas occurs by molecules changing altitude. However, heat in a gas can be transported upward merely by collisions. Random collisions distribute kinetic energy equally in all directions. You can see this happen in a two-dimensional molecular dynamics simulation of gases at:
http://physics.weber.edu/schroeder/md/InteractiveMD.html
There is an option for turning gravity on our off. Turning gravity on converts some potential energy into kinetic energy, creating both a density, pressure and temperature gradient. But the temperature gradient quickly dissipates through collisions, leaving the pressure and density gradient at equilibrium.
Frank,
Rising and falling molecules will also transfer some energy via collisions so I agree you can add that in. The basic principle still applies though. I was really considering the adiabatic aspect but energy exchange via collisions is diabatic.
Frank,
Assuming CO2 forcing without feedbacks is one thing. Assuming that if there’s a change in other forcings – or in non-radiative processes, like evaporation/convection – then this can ONLY be a feedback to the CO2 forcing. As if the rest of the climate system does nothing but stand fixed in breathless anticipation of CO2’s bidding, incapable of acting independently. The data from the real Earth system shows that this is not the case at all. That’s what I was trying to tell you … MODTRAN, meet Reality.
“Assuming CO2 forcing without feedbacks is one thing. Assuming that if there’s a change in other forcings – or in non-radiative processes, like evaporation/convection – then this can ONLY be a feedback to the CO2 forcing, as if the rest of the climate system does nothing but stand fixed in breathless anticipation of CO2’s bidding, incapable of acting independently. The data from the real Earth system shows that this is not the case at all. That’s what I was trying to tell you … MODTRAN, meet Reality.”
There’s a “…is another” missing there somewhere.
Kristian wrote: “Assuming CO2 forcing without feedbacks is one thing. Assuming that if there’s a change in other forcings – or in non-radiative processes, like evaporation/convection – then this can ONLY be a feedback to the CO2 forcing. As if the rest of the climate system does nothing but stand fixed in breathless anticipation of CO2’s bidding, incapable of acting independently. The data from the real Earth system shows that this is not the case at all. That’s what I was trying to tell you … MODTRAN, meet Reality.
MODTRAN is reality – it is the product of thousands of laboratory measurement of the absorption of GHGs and quantum mechanics. However, alone it tells us nothing about climate sensitivity, which is the ratio of eventual equilibrium warming to forcing. MODTRAN doesn’t tell us whether climate sensitivity for 2XCO2 (a forcing of 3.7 W/m2) is 5 or 0.5 degC, because we don’t know how much the earth must warm before OLR and post-albedo SWR are equal.
You may want to consider the difference between “forced variability” (or forced change) and “unforced variability”. Chaotic changes in temperature, clouds, winds, ocean currents, lapse rate, evaporation and precipitation are a normal features of our weather. The average water molecule that evaporates remains in the atmosphere for about 9 days. Most weather fronts and storms dissipate in the same period of time. Rossby waves can be stationary for a month. Changing ocean currents produce SST anomalies like El Nino that last about a year (and possibly the AMO or PDO that last decades). When it is windier or warmer than normal, there will be more evaporation than usual and more clouds and rainfall whenever that air is lifted. These are all forms of “unforced variability”. They change how heat (and clouds) are distributed within the atmosphere, surface and ocean. This distribution does effect how much radiation enters leaves the planet (via albedo and radiative cooling to space).
Averaging over at least 30 years defines climate – something that we believe doesn’t change unless influenced or forced by some phenomena not normally present in our climate: changes in solar output and orbit, volcanic eruptions (examples of “naturally-forced variability”) and “anthropogenically-forced variability”. All of these FORCINGS directly influence how much radiation enters and leaves the planet and are only indirectly connected with the distribution of heat and clouds by unforced variability.
Some phenomena that exhibit unforced variability also vary systematically with surface temperature: evaporation, dew point, convection and clouds (cumulus in summer, stratus in winter) and snow cover. These are feedbacks.
So we have forced variability, amplified by feedbacks, immersed in unforced variability (which, when the variability involves temperature, already includes feedbacks). Distinguishing between forced and unforced variability is difficult.
Frank says, May 27, 2015 at 9:40 pm:
“MODTRAN is reality”
Hahaha!
Of course it isn’t. It’s “Reality” in a purely radiatively driven, static layer model world, where you can freely isolate individual components of a potential radiative budget and hold everything else (all other ‘forcings’) unchanged and not care at all that in the real Earth system, the heat delivered from the surface to the atmosphere is transported up through the tropospheric column by way of convection (bulk movement of air), not radiation.
Sorry, Frank, but this isn’t how the world works. You cannot simulate the real Earth system in the laboratory or in a radiative transfer model.
And the data from this real world of ours shows I’m right and you’re wrong.
Kristian
Correct but don’t forget that the heat so taken up comes back down in the next descent which is why the surface becomes 33K warmer than S-B.
The mistake that the mainstream has made is to try to integrate the thermal effect of conduction and convection into the S-B equation when the S-B equation applies only to the underlying radiative exchange.
S-B is perfectly correct. 255K is the portion of surface temperature supplied by the radiation coming in from space exactly as per S-B.
The additional 33K has to be treated separately as a completely independent energy loop within the atmosphere.
That energy loop must raise the surface temperature above S-B because conduction and convection are slower than radiation.
Furthermore the kinetic energy required to sustain hydrostatic balance is unable to radiate out because as fast as it is returned to the surface in descent it is taken up again in ascent.
So, S-B is perfectly correct, the surface is at the correct temperature in S-B terms but one must add another 33K to account for the energy tied up in convective overturning.
The mainstream has incorrectly decided that a surface at 288K is in breach of S-B but it is not. Conduction and convection simply supplement the S-B calculation as a separate and independent phenomenon.
“… the heat so taken up comes back down in the next descent …”
No, Stephen. The heat doesn’t come back down. It is radiated out to space as OLR.
It is well established that descending air warms at the dry adiabatic lapse rate as it descends.
That reduces convection from the sun warmed surface below the descending column so that the surface can warm up above S-B.
“It is well established that descending air warms at the dry adiabatic lapse rate as it descends.”
That’s not what you said, Stephen. You said “… but don’t forget that the heat so taken up comes back down in the next descent …” The ‘heat so taken up’ was a direct response to my “… in the real Earth system, the heat delivered from the surface to the atmosphere is transported up through the tropospheric column by way of convection (bulk movement of air), not radiation.”
The whole point of the climate engine after having reached its dynamic equilibrium (steady state) is to get the incoming solar heat back out to space as efficiently as possible (and/or needed). It does so by first transferring the surplus energy from the surface to the massive atmosphere above by way of conduction, evaporation and radiation, secondly by the surplus energy naturally (upon absorption) making the surface air rise up through the tropospheric column towards the tropopause, before it’s finally being radiated to space from air and cloud levels aloft. The descending air doesn’t bring the surplus energy back to the surface, Stephen. Whoever gave you that silly idea …? It descends specifically so that it can pick up more surplus energy (the next round, so to say) from the surface, the heating end, and dump it at the cooling end (at or towards the tropopause). That’s how convective circulation works. Tell me you know this …
The energy used to lift the gases against gravity and which is converted to PE by expansion is no longer available as heat to be radiated to space. That part of the energy involved in uplift is what returns to the surfsace as heat in the descent.
Surely you must know that?
Check this out:
https://books.google.co.uk/books?id=PkScAQAAQBAJ&pg=PA341&lpg=PA341&dq=Convection+KE%2BPE&source=bl&ots=98UwR5ltNJ&sig=zwv3jPoguI0D1SPypVgR-F1c75U&hl=en&sa=X&ei=lFJoVbSCEIHjUvmKgLAJ&ved=0CFIQ6AEwBw#v=onepage&q=Convection%20KE%20PE&f=false
and click on the word ‘convection’ highlighted in yellow for page 341.
You could browse some of the other material there too.
Kristian wrote: “In other words, there is no hint anywhere of any tightening up of atmospheric radiative insulating effect on the surface. No “enhanced greenhouse effect” to be observed. On the contrary! The OLR at the ToA simply and slavishly tracks tropospheric temps over time, clearly an effect, not a cause.”
There is a GHE. The surface is emitting and average of 390 W/m2 of OLR, but only 240 W/m2 reaches space. Radiative transfer calculations show this is due to absorption and emission of LWR as it passes through the atmosphere. The same calculations show that the surface should receive an average of about 333 W/m2 of DLR, which is observed with pyrometers (assuming you believe in these instruments).
Mean global surface temperature (not the temperature anomaly) rises 3.5 degC during summer in the NH because its heat capacity is lower than the SH. CERES and ERBE has observed the mean GLOBAL change in OLR and reflected SWR) associated with this seasonal change in mean GLOBAL temperature. You can see the results here:
http://www.pnas.org/content/110/19/7568.long
See Figure 1B. When the earth warms 3.5 degK, it doesn’t emit as much additional OLR through clear skies as expected for a blackbody. This happens because the increase in water vapor and decrease in lapse rate that develop when the earth warms together reduce OLR. (Climate models says water vapor reduces OLR about twice this much, while the lapse rate reduction negates half of the reduction.) These feedback exist.
Note that the temperature change is so large and repeats every year, so the error bars associated with these measurements are far smaller than when we try to detect a change of in radiation of about 1 W/m2 associated with a temperature change of a few tenths of a degC. You cite a 1.5 W/m2 change over 15 years without any mention of confidence intervals. In this paper, we are dealing with a 10 W/m2 change in a year measured with the same equipment every year for 10 years.) Climate models do a good job of reproducing OLR from clear skies, a lousy job with SWR from clear skies (seasonal snow cover), and a lousy both with from cloud skies.
“Radiative transfer calculations ”
For global surface stations that collected more than 360 daily samples per year (69 million samples in total ), when you subtract the average for tonight’s falling temp from today’s rising temp for all measurements, it is negative, 50 of the last 74 annual averages is negative, 30 of the last 34 years annual average is negative.
There is no evidence in the surface record of a loss of nightly cooling.
Frank says, May 26, 2015 at 5:15 pm:
“There is a GHE.”
There sure is an ‘atmospheric insulating effect’ on the surface. However, this effect comes as the result simply of the MASS of the atmosphere.
Kristian wrote: “There sure is an ‘atmospheric insulating effect’ on the surface [not a GHE]. However, this effect comes as the result simply of the MASS of the atmosphere.
The thermogravitational effect is the creation of people who forget that energy is transferred mostly by molecular collisions – not by changing PE to KE when the mean free path of a gas molecule near the surface is only 1 MICRON. (Heat is conducted through solids exclusively by such collisions.)
You can experiment with a 2D-molecular dynamics simulation of the molecular behavior of gases at the link below. It allows you to pause, “turn on” gravity, and watch what happens. Falling molecules pick up KE from PE, creating a density, pressure and temperature gradient. The density and pressure gradients persist; but the temperature gradient is dissipated by collisions.
The atmosphere is 3D not 2D so the outcome is different.
In 3D the convective columns can redistribute energy around a sphere rather than just in a box and so any energy taken up in ascent as PE is returned to the surface as KE elsewhere and radiated to space from the surface so that the temperature decline with height is maintained.
If higher molecules were able to contain the same KE as the surface PLUS the PE applicable to their height then hydrostatic balance would never be achieved and the atmosphere would be lost to space.
Throughout the vertical column the upward pressure gradient force would permanently exceed the force of gravity so the atmosphere would drift off into space.
To keep an atmosphere there must be a point of hydrostatic balance where the upward pressre gradient force equals the downward force of gravity but that could never happen if collisional activity caused an ‘excess’ of KE in higher molecules.
Therefore the radiative only scenario is entirely false.
Kristian: On Earth, there is a “turbopause” at about 100 km and it wouldn’t be unreasonable to claim that a thermogravitational effect exists above this altitude. This is where molecules begin to separate according to molecular weight, with the lightest gases being enriched at the highest altitudes. If my calculations are correct, the mean free path between collisions is about 1 meter around the turbopause, but it is difficult to say because local thermodynamic equilibrium doesn’t exist and temperature has lost its thermodynamic meaning
https://en.wikipedia.org/wiki/Turbopause
But has it lost it’s quantum electrodynamics?
Frank says, May 28, 2015 at 11:12 am:
“The thermogravitational effect …”
I am not talking about “the thermogravitational effect”, Frank. That is pure nonsense. I’m talking about the massive atmosphere as an insulating layer on top of the solar-heated surface.
Kristian wrote: “The OLR at the ToA simply and slavishly tracks tropospheric temps over time, clearly an effect, not a cause.”
Of course. Most of the photons escaping to space are emitted from troposphere. When the troposphere is warmer, it emits more OLR and DLR.
The current anthropogenic forcing is about 2 W/m2, but warming of the troposphere has reduced the current average global imbalance to less than 0.5 W/m2 (according to ARGO).
Radiative forcing and radiative imbalance are NOT the same thing. Climate sensitivity is the change in temperature in response to a radiative FORCING, not a radiative imbalance. Willis is doing a great disservice to readers of WUWT by confusing these concepts. Above I clearly explained why 35N,10W responds to the annual change in radiative IMBALANCE with about a 6 degK in surface temperature without any mention of climate sensitivity. No response from Willis. Willis’s seasonal temperature analysis has nothing at all to do with climate sensitivity, since no FORCING has been described. A radiative imbalance is not a forcing! These are seasonal changes – The forcing is SOLAR! The angle of the sun rays changes with the seasons and surface irradiance changes with the cosine of that angle. The solar forcing also changes with cloud cover. If Willis complied the right data, he could produce the solar forcing at 35N,10W. Unfortunately, ocean and air currents rapidly move heat to and from 35N,10W. The temperature change at 35N,10W is not a result of the seasonal change in solar forcing or the radiative imbalance at the TOA above 35N,10W.
Frank says, May 26, 2015 at 5:48 pm:
“Of course. Most of the photons escaping to space are emitted from troposphere. When the troposphere is warmer, it emits more OLR and DLR.”
Yes, Frank. That’s what happens in the REAL world. But in the rGHE model universe, over time, the troposphere is not allowed to emit more OLR as it grows warmer than when the ‘forcing’ first started increasing IF the cause of the warming is an “enhanced rGHE”. That’s the whole point. That’s the working mechanism for warming.
Also, OLR and DLR are very different entities. The former is Earth’s radiant HEAT to space. The latter is NOT the atmosphere’s radiant heat to the surface. The DLR is merely the smaller, downward component potential of the surface’s radiant heat (‘net LW’) to the atmosphere (and space).
This is why, while the OLR actually just represents the Earth system’s continuously restored heat balance with its surroundings, the DLR (the “back radiation”) is meant to represent the actual atmospheric ‘radiative forcing’ itself, the actual postulated rGHE/AGW warming mechanism. If we don’t see it grow, the atmospheric ‘forcing’ isn’t growing. This is why I referred you to the CERES data, Frank.
“Radiative forcing and radiative imbalance are NOT the same thing.”
I already agreed with you on that one, if you didn’t notice.
“The DLR is merely the smaller, downward component potential of the surface’s radiant heat (‘net LW’) to the atmosphere (and space).”
This is my thoughts as well, which makes me wonder where the polar DLR is coming from with such a very cold emitting surface.
Kristian wrote: “But in the rGHE model universe, over time, the troposphere is not allowed to emit more OLR as it grows warmer than when the ‘forcing’ first started increasing IF the cause of the warming is an “enhanced rGHE”. That’s the whole point. That’s the working mechanism for warming.”
Let’s approach the problem from a different perspective. Instead of asking how much warming a forcing will cause, let’s ask how much a surface warming will change the TOA radiation balance. MODTRAN and other radiation transfer programs say that about 3.7 W/m2 less OLR will reach space after an instantaneous doubling of CO2. Let’s ignore all of the details of what happens next and jump ahead to a time when changes FORCED by an instantaneous doubling of CO2 have ceased (and only unforced variability remains). What can we say about this new equilibrium? We know outgoing OLR and incoming post-albedo SWR will be in balance – that the planet will have warmed enough to eliminate the 3.7 W/m2 imbalance that was created when CO2 doubled. How much warmer on the average will the planet have to be for this to happen? About 1 degK, if the earth behaves like a simple blackbody at 255 degK. 1.15 degK, if we assume the planet warms equally everywhere and nothing else changes (no feedbacks).
However, things will change. Absolute humidity will increase upon warming. That will cause the lapse rate to decrease. Surface albedo due to seasonal snow cover will decrease. No one knows for sure what clouds will do. All of these factors influence how much surface warming will be needed to increase OLR plus reflected SWR by the needed 3.7 W/m2.
We are discussing the climate feedback parameter: W/m2 change at the TOA per degK of surface warming (W/m2/K). The reciprocal (K/(W/m2)) is climate sensitivity, though we usually divide by 3.7 W/m2/doubling of CO2 to get an answer in terms of degK/doubling.
Kristin wrote: “Also, OLR and DLR are very different entities. The former is Earth’s radiant HEAT to space. The latter is NOT the atmosphere’s radiant heat to the surface. The DLR is merely the smaller, downward component potential of the surface’s radiant heat (‘net LW’) to the atmosphere (and space).
This is why, while the OLR actually just represents the Earth system’s continuously restored heat balance with its surroundings, the DLR (the “back radiation”) is meant to represent the actual atmospheric ‘radiative forcing’ itself, the actual postulated rGHE/AGW warming mechanism. If we don’t see it grow, the atmospheric ‘forcing’ isn’t growing. This is why I referred you to the CERES data, Frank.”
Most of our discussion has been highly rational. This is not. Photons emitted by GHGs in the atmosphere don’t “know” whether they are DLR or OLR. They are absorbed elsewhere in the atmosphere following the usual rules or else reach they surface or space. (The usual rule is the differential form of Beer’s Law which can’t be integrated because the density of GHG changes with altitude: dI/ds = noI_0.) Photons are emitted upward or downward such that dI/ds = noB(lamba,T). The amount of energy that each photon transfers is hv and little DLR is reflected at the surface. Pretending OLR and DLR behave differently is absurd. MODTRAN works for both.
TOA OLR is what reaches space. TOA DLR is zero, so it doesn’t contribute to the planet’s radiative (im)balance. Surface OLR and surface DLR are different. They contribute to the surface energy (im)balance, not the planetary one.
CERES doesn’t measure forcing – the change since 1750 in the amount of a material (GHG or aerosol) which perturbs energy flux into and out of the planet – expressed in terms of W/m2 by means of radiative transfer calculation. CERES measures radiative imbalance. Radiative forcing grows as GHGs increase. The radiative imbalance does not. Global warming reduces the imbalance that GHGs create. So you can’t prove that radiative forcing has stopped increasing by looking at the radiative imbalance. You can see a plot showing both forcing and imbalance after Pinatubo at the link below. Since the planet cooled substantially while aerosols were at their peak, the imbalance began to decrease within a month of the peak and the imbalance was eliminated a year later at a time when 50% of the aerosol forcing was still present.
http://www.drroyspencer.com/wp-content/uploads/Pinatubo-revisited-SAGE-tau-and-ERBE.gif
Frank says, May 28, 2015 at 2:17 am:
“Let’s approach the problem from a different perspective.”
No, let’s not. What you (and SoD) suggest is only fogging up the real issue here – the atmosphere’s increased IR opacity reducing the OLR, forcing the surface to warm so that the OLR can grow back to where it was, once again balancing the budget. THIS is the rGHE warming mechanism. See also my reply to SoD further downthread.
“Most of our discussion has been highly rational. This is not. Photons emitted by GHGs in the atmosphere don’t “know” whether they are DLR or OLR.”
Frank, then you obviously don’t understand what I’m talking about.
The OLR is an actual HEAT flux, Earth’s heat to space. DLR is not an actual heat flux. It is but one constituent potential part of a heat flux, the radiant one from the surface up/out to the atmosphere (and space).
If you don’t get this fundamental difference, there is no point continuing this discussion.
Here’s the difference between the annual average of today’s rising temp and tonight’s falling temp for stations that took more than 360 daily sample per year, and the number of records
YEAR DIFFERENCE in degrees F SAMPLE COUNT
1940 0.027957973 40450
1941 -0.010104032 37104
1942 0.007376309 50974
1943 -0.006600669 106368
1944 0.000894331 171413
1945 0.002708585 109356
1946 -0.016936611 75818
1947 0.015605421 104547
1948 -0.009773913 196738
1949 0.011653284 274738
1950 0.001123508 294791
1951 0.020635422 301060
1952 -0.014110651 366071
1953 -0.008804188 380160
1954 0.010772869 396199
1955 -0.007817724 361934
1956 0.0087034 355229
1957 -0.028805471 396449
1958 0.003690311 497221
1959 0.001327244 451085
1960 -0.026174748 508024
1961 -0.001203715 511500
1962 -0.007567744 514658
1963 0.011024797 507837
1964 0.005702056 485246
1965 0.009888569 335812
1966 -0.017190748 393037
1967 -0.002150335 397752
1968 -0.012093387 362322
1969 0.004224134 416322
1970 -0.010334386 486444
1971 0.008302247 176121
1972 0.007295899 172782
1973 -0.009157925 564178
1974 -0.003228731 805208
1975 -0.020917758 792671
1976 -0.038319245 1111465
1977 0.026638833 860841
1978 -0.016594529 1093975
1979 0.016303773 1028032
1980 -0.017638483 1129689
1981 -0.005171018 1099474
1982 -0.011377151 1055440
1983 -0.012833048 1166200
1984 -0.003854703 1220950
1985 -0.004487731 1185677
1986 -0.002162743 1254703
1987 -0.003036479 1235016
1988 -0.006528295 1365931
1989 0.002728288 1265629
1990 -0.008999233 1247673
1991 -0.008292409 1171457
1992 -0.01383395 1304978
1993 -0.005153482 1277117
1994 0.007434471 1298317
1995 -0.008586358 1293354
1996 -0.003513682 1318816
1997 0.008396956 1321324
1998 -0.021962934 1169739
1999 -0.020323599 1147533
2000 -0.029961211 1582673
2001 -0.002299226 1455055
2002 -0.011770051 1534148
2003 0.005673355 1562356
2004 -0.000420242 1769217
2005 -0.004098412 1928381
2006 -0.008332224 2058850
2007 -0.011856501 2070282
2008 -0.007146907 2324740
2009 -0.004842814 2401806
2010 -0.001665325 2506477
2011 -0.002674635 2529280
2012 -0.019303489 2632177
2013 -0.00327971 2488421
9999 is the all years average.
9999 -0.004151764 69864812
Either ECS is less than a year, or there’s no loss of nightly cooling.
Kristian wrote: “The OLR is an actual HEAT flux, Earth’s heat to space. DLR is not an actual heat flux. It is but one constituent potential part of a heat flux, the radiant one from the surface up/out to the atmosphere (and space).”
OLR and DLR are the same thing – radiative energy. However, when you use the term “heat” in a thermodynamic sense, “heat” is energy transferred. In the case of radiation, it is the NET energy transferred from hot to cold by radiative fluxes running in both directions: roughly 390 – 333 = 57 W/m2 upwards from the surface to the atmosphere and space. Unfortunately, “heat” is not always interpreted in the strict thermodynamic sense and means different things to different people.)
Unlike OLR, DLR is absorbed by something (the surface) that gains and loses energy from more than one source and by more than one mechanism, so one can’t say that DLR by itself caused warming. A change in temperature is caused by the net result of all of these processes.
When you use the terms OLR and DLR, however, you talking about one-way fluxes – and by the technical definition – they are not “heat”. They are radiative energy, which behaves by the same rules whether it is traveling up or down. Saying OLR is a heat flux is an oxymoron and implying that DLR is different only makes things worse.
It appears as if we agree on the climate feedback parameter approach to climate sensitivity.
Frank says, May 28, 2015 at 5:51 pm:
“OLR and DLR are the same thing – radiative energy.”
No, Frank, they are not the same thing. They are both made up of radiative energy, yes, but the former is a thermodynamic HEAT flux (a ‘net LW flux’), the latter is not. Therefore they are not directly comparable. Whether you believe it or not, the radiant flux from Earth as read by the satellites is actually the ‘net LW flux’ from the Earth to space. The UWLWIR from the ToA is countered by DWLWIR from space, but since the DWLWIR in this case is so microscopically small, it hardly affects the magnitude of the ‘net flux’ at all. In other words, the UWLWIR part of the radiative exchange between Earth and space is effectively equal to the ‘net LW flux’, that is, the outgoing HEAT, the OLR.
This is why OLR is equivalent only to the ‘net LW flux’ (the radiant heat) from the surface, that is, the DWLWIR minus the UWLWIR: [345 – 398 =] -53 W/m2. This is the radiant energy actually transferred in the thermal exchange between the surface and its surroundings. Of this total flux, 20 W/m2 on average go straight to space through the atmospheric window, 33 W/m2 are transferred as radiant heat from the surface to the massive atmosphere on top.
So, under an increased ‘radiative forcing’, we should not be able to observe a decrease in OLR at the ToA. It should remain stable (flat) as the layers of troposphere underneath, plus significantly the surface at the bottom, warm. We should be able to observe an increase in DLR to the surface.
The funny thing is, we do observe an overall stable OLR average since 2000, but this is in spite of a non-warming troposphere. In fact, the non-warming troposphere (the stable temp average) seems to be the direct cause and the stable OLR average the direct radiative effect, the opposite of what the ‘warming by increased RF’ hypothesis dictates – flat OLR during rising tropospheric temps:
However, we do not observe an increase in DLR to the global surface since 2000, as we should with a steady increase in total RF. We observe rather a reduction!! And parallel to that, we observe also an increase in upwelling radiation from the surface (but obviously an increase not caused by, not as a response to, an increase in the downwelling), so that the ‘net LW’, the radiant heat from the surface up, has increased robustly in 15 years (by 1.5 W/m2, from -52.5 on average to -54 W/m2 on average):
As would be expected if the surface warmed, because it got warmer having nothing to do with Co2 forcing!
And, surface stations show exactly this, it cools more at night than it warmed during the day.
[html “pre” and “/pre” coding inserted for table. .mod]
Frank wrote: ““OLR and DLR are the same thing – radiative energy.”
Kristian replied: “No, Frank, they are not the same thing. They are both made up of radiative energy, yes, but the former is a thermodynamic HEAT flux (a ‘net LW flux’), the latter is not. Therefore they are not directly comparable. Whether you believe it or not, the radiant flux from Earth as read by the satellites is actually the ‘net LW flux’ from the Earth to space. The UWLWIR from the ToA is countered by DWLWIR from space, but since the DWLWIR in this case is so microscopically small, it hardly affects the magnitude of the ‘net flux’ at all. In other words, the UWLWIR part of the radiative exchange between Earth and space is effectively equal to the ‘net LW flux’, that is, the outgoing HEAT, the OLR.”
To avoid confusion, let’s discuss TOA_OLR, surface_OLR, surface_DLR, and TOA_DLR. These are all fluxes of radiative energy. Since you don’t define what you mean by “thermodynamic heat”, I’ll repeat my explanation.
Surface_OLR – Surface DLR is a measure of the amount of thermodynamic heat transferred from the surface to the atmosphere and space. NEITHER one ALONE is thermodynamic heat, since heat is the NET flux (not the larger flux). The smaller flux does not “cancel” part of the larger flux when speaking of radiative energy or photons. Heat is also transferred from the surface to the atmosphere by convection. Individual molecules transfer kinetic and latent energy both up and down, but we can’t measure two fluxes (as we can will radiation). So all we talk about a one-way flux of energy which is the NET result of what many molecules do (and therefore heat).
TOA_OLR – TOA_DLR is the amount of heat transferred from the earth and atmosphere to space. Since TOA_DLR is negligible, TOA_OLR is both a one-way flux of radiative energy and heat – net transfer of energy. (Given the fact that TOA_DLR – aka as the cosmic microwave background – is composted of microwaves instead of infrared, I don’t know if it can be detected by the detectors on CERES.)
You’ve cited observations. Over short periods of time (several years to several decades), unforced variability can obscure variability forced by man or nature: 1) Take Willis’s “Spot the Volcano” posts. Unforced variability (including autocorrelation) in monthly GMST is large enough to mask the temperature change of the large, but rapidly decaying, forcing produced by volcanos. Pinatubo supposedly caused about 0.6 degC of cooling at its peak influence, but is barely apparent in graphs of monthly data. (If Willis had smoothed the monthly data, the unforced variability would have been suppressed and the volcanic signal would have been more apparent.) 2) The deep ocean contains a massive reservoir of cold water and chaotic fluctuation in the rate of exchange of heat between the surface and the deep ocean has the ability to produce fluctuations surface temperature. During the 97/8 El Nino and subsequent La Nina (which interfere with upwelling of deep water off of South America among other things), GMST rose and fell 0.5 degC. We know a lot about ENSO, because we have experience many of them. Other oscillations (such as the AMO and PDO) may or may not have a large effect over several decades. Some think the AMO produces a 65-year oscillation in GMST of about 0.25 degC, but we haven’t seen enough cycles to know.
So the first step in evaluating the meaning of the hiatus is to compare it to expectations of what natural variability might produce. The IPCC did not do an appropriate job of this when evaluating the meaning of the warming in the 1980’s and 1990’s. The second step is to ask how much change you expect to see during the hiatus, if the conventional view were correct.
Climate models predicted warming of 0.2 degC/decade or more, so the hiatus represents at least 0.3-0.4 degC of missing warming. Given the modest size of the 1920-40 warming, 1950-70 cooling, 1978-1998 warming, I’m reluctant to assign ALL of the difference to such decadal unforced variability. Factor in the slightly weaker sun and slight increase in volcanic aerosols and perhaps the difference has been explained. Most likely the models are running hot.
You cited DLR not increasing. How much was DLR supposed to increase as CO2 rose from 370 ppm to 400 ppm (8% increase) during the hiatus. One figure I’ve seen is that DLR should increase ONLY 0.8 W/m2 for a doubling of CO2. (Overlapping bands with water vapor negate most of the influence of rising CO2 in the lower troposphere, but not in the drier upper troposphere.) The hiatus period involves 1/10 of a doubling of CO2, or less than 0.1 W/m2! You can try your own experiments with MODTRAN.)
What factors influence DLR the most? Cloud cover (60% of the sky) AND cloud altitude. Average DLR is about 333 W/m2. Do you seriously want to invalidate AGW on the basis of changes of less than 0.5% in this signal which varies with clouds!
Finally the CERES Surface_EBAF data (DLR) you cite is produced by radiative transfer calculations, not direct observation. There is no composite global record for changes in DRL measured at the surface. CERES looks at cloud tops and estimates temperature of the lower cloud surface and how much DLR they emit. In clear skies, OLR is used to estimate atmospheric temperature and humidity and then how much DLR is emitted in the opposite direction. They may also make use of UAH or RSS data.
The change you expect to see in DLR during the hiatus is ridiculously small, natural variability is high, and the “observed” change is uncertain. Drawing any conclusion is a bad joke!
As we have discussed earlier, TOA_OLR depends on atmospheric and surface temperature and a hiatus in the latter will produce a hiatus in the former.
What does the hiatus mean? A 15-year hiatus in warming in GCMs simulating the growing forcing occurs about 2% of the time. The models either over-estimate climate sensitivity or underestimate unforced variability or both. Forcing and observed warming (energy balance models) suggest that climate sensitivity is about 2/3 as large as the IPCC’s models predict.
Willis et al,
my contribution to climate science, a new Law of nature, has now been more than 95% verified through 40+ years of observational data on climate alarmism from the Ige Age evangelists in the 70’s to today’s Thermageddon teams, and it goes like this:
“The CO2 sensitivity is at any given time inversely proportional to the fact sensitivity”
PS: This Law of nature is commonly known as “Telehiv’s Law of Climate Cognition”.
https://climategrog.wordpress.com/2014/03/08/on-inappropriate-use-of-ols/
The first two panels of fig 1 have visibly wrong slopes fitted. This is due to regression dilution as explained in the link.
The latter two will also suffer from this to a lesser extent.
Fitting a linear trend to noisey data will almost always underestimate the slope. This is why it is customary in climatology to plot rad vs temp and not temp vs rad. It produces a lower slope and thus higher climate sensitivity.
This was discussed in the appendix of Forster and Gregory 2006 but they avoided putting it in the main text or in the results.
Mike,
Very interesting link. The one study of this sort that has impressed me is that of E.-S. Chung et al. (An assessment of climate feedback processes using satellite observations of clear-sky OLR, GEOPHYSICAL RESEARCH LETTERS, VOL. 37, L02702, doi:10.1029/2009GL041889, 2010). Might you have any words of wisdom on that one?
Wills, you say:
“I’ve been wanting to [take] another look at the relationship between net top-of-atmosphere (TOA) radiation changes on the one hand and changes in temperature on the other. As a necessary prelude, I finally have gotten round to an oft-postponed task, that of looking at the thermal lag in different areas of the planet. This is the lag between a change in TOA radiation and a corresponding change in surface temperature. Figure 1 shows my results.”
Considering how the solar heat travels through the Earth system, this seems a strange approach to me, using the “net ToA radiation”. I can’t see what useful results that would provide, concerning real lags between radiation and temperatures.
The “net ToA radiation”, after all, is made up of two components: 1) the incoming SW (from the Sun), and 2) the outgoing LW (from the Earth).
To get a proper overview of how the Earth system and the radiative processes work, I really feel you need to keep the two apart, seeing how the one (the former) is a direct ’cause’ of change and the other (the latter) a direct ‘effect’ of (response to) change.
THE SOLAR HEAT CYCLE (diurnal/annual):
The solar SW enters the Earth system > the Earth system absorbs the SW as heat > temperatures go up > the warmer Earth system responds by emitting more LW > the LW goes out to space through the ToA as heat > temperatures go down.
HEAT IN = HEAT OUT. Temps UP > Temps DOWN.
Here’s “net SW” in through the ToA vs. tropospheric temps:
The correlation is far from perfect, but this is quite expected, because there is hardly any direct warming response between incoming SW and the largely transparent troposphere. The troposphere is rather mostly warmed from below, by surface processes, effectively oceanic ones. Still, there’s a fairly distinct lead-lag pattern to be discerned, where the tropospheric temps lag the incoming SW by 1-4 months, centred around 2-3.
The “OLR” out through the ToA vs. tropospheric temps:
The correlation here is, as you can see, much tighter. Again, as would be expected, because OLR is to a large extent a direct effect of tropospheric temps. The surface is warmed by the Sun, responds and cools back by dumping its surplus energy into the troposphere as heat, mostly via evaporation/condensation, in turn warming the troposphere (tropospheric temps ‘always’ lag surface temps). The troposphere finally radiates Earth’s acquired heat from the Sun back out to space, cooling back down. We see this as the OLR through the ToA.
If you look carefully at the graph above, you will see that the red OLR curve lags the blue tropospheric temp curve by 0-1 months.
What is most interesting here is that there is no hint anywhere of an “enhanced greenhouse effect” in operation. The OLR curve simply follows the tropospheric temp curve loyally from 2000 to 2015, the only noteworthy deviations from the general fit occurring in 2008 and 2010, caused by well-understood, ENSO-related processes (strongly reduced and/or lowered cloud cover during La Niña 07/08, strongly increased and/or elevated cloud cover during El Niño 09/10).
Sorry. Willis.
Of course 3.7 watts per square metre (W/m2) increase in TOA radiation actually has nothing to do with a Doubling of CO2 does it?
The Solar Flux hit a peak about 2 weeks ago at 165 sfu. The Solar EUV directly warms the Ionosphere, Troposphere, and the Ozone layer. This creates heat that produces down-welling air masses, which, when they reach the ground are called “high pressure areas”. It takes about 3 weeks {when the Sun is in the Northern Hemisphere} for the effects to reach Antarctica; and the heat to be slowly dissipated. One can watch Antarctica warm by monitoring the various Antarctica Stations, such as, Vostok.
Also, some of the Solar EUV makes it through the atmosphere and can be measured on the landmasses or oceans by an increase in UV intensity. The UV directly reaching the landmasses or oceans warms the upper surfaces especially of the Oceans. This makes evaporation increase. The stored energy in the very high humidity atmospheric air masses is well known.
Directly under the Sun {+20degrees latitude, now}, the air masses that are forced downward {by atmospheric expansion, due to heat} are replaced by upward flowing air masses, as indicated by “low pressure areas”. The low pressure areas take the surrounding landmass heat and moisture {especially over the Oceans}, and move it upward.
The Solar Flux is now at 98 sfu and falling. Central America and Southern USA will see a dissipation of the extensive rain and thunderstorms!
This is a direct coupling of more powerful storms to the Solar EUV {measured by the Solar Flux}.
If you disagree, what is your Theory??
Please note that in Iraq substantial changes in wetland occurred during the Saddam regime,nresulting in strong temperature increases, not related to forcing.
Thanks Willis,
It is, again, surprising that no one has done this before with the available data.
The 0.44C/W/m2 is a little higher than what I was expecting based on previous attempts to to calculate this number but the data is the data.
Bill Illis,
You wrote “It is, again, surprising that no one has done this before with the available data”. Why do you say that? In fact, a fair number of papers have attempted this with much more care than Willis (though still mostly insufficient care). Mike (a different one) has kindly provided the following link with a bunch of references: https://climategrog.wordpress.com/2014/03/08/on-inappropriate-use-of-ols/
Very interesting. Most amazing to my eyes is the rapid response of Antarctica across the board. Sure, it’s land, but nearly completely covered with ice. Why should it respond as quickly as tropical land areas? Why should it have the highest response on the planet to the 3.7w increase?
Maybe this is a CERES artifact, or maybe it speaks to the puzzling seesaw problem…
Again, I feel that this is due to the rapid and efficient transport of energy via water vapour through the atmosphere. The Antarctic has plenty of water available through sublimation…enough to transport the energy and be responsive to surface temperature change but not enough to saturate with cloud and feedback against the process.
Ok, but why not the Arctic as well?
I think it kinda is. Although its not quite so clearcut…maybe the extra water around makes it less so.
Willis my personal gut visualization or how it really works asks this question or wonders.
In the lag plot, what is happening. My own believe is that radiation cooling rate is greatly influences by the dust content of air. For heat absorbed by polarizied molecules most of the re-radiations occurs from solid particles. Just like if you heat water in a pan, bubbles first form at surface defects on the surface. Or if you see the fine line of bubbles in beer in a glass.
An IR thermometer is a photon counter. So many photons per degree calibrated. What mass of what produces photons for solids vs gasses vs liquids.
Also if I point an IR gun up. I am reading the emission of photons from an expanding cone of air. How many tons of mass? In dry air I will see a temperature dozens of degrees below the local temp of air. As humidity goes up the temp read will approach the local air temp.
What is the mass emissivity of gases of gasliquids transiting liquid liquidsolids transiting and solid.
I would proffer there are orders of magnitude differences in emissivity.
How thick would I need a solid film with is quarter inch or ??? radius would be needed to provide the sensor with enough photons to read the current local temperature. This gives an idea of the size of the differences in mass emissivity between the different phases of matter.
So would a correlation of the delay with global dust conditions reveal any insight ?
“My own believe is that radiation cooling rate is greatly influences by the dust content of air. ”
Nightly cooling is reduced by 2-3x as temps get near the dew point .
“An IR thermometer is a photon counter. ”
Not exactly it measures the accumulated energy of the photons times the flux at those wavelengths, mine measures from 8u to 14u, so it also has to have a calibration to account for the missing part of a BB spectrum.
” In dry air I will see a temperature dozens of degrees below the local temp of air. As humidity goes up the temp read will approach the local air temp.”
On humid days I measure a Tsky that’s 50F or more colder than the ground, which is near air temp, on cool dry days I’ve measured as much as 105F colder Tsky.
To be fair you do need to at least add the 3.7W/M^2, but at Tsky of -40F that’s less than about 3F.
Yes it counts photons of a very narrow wave number. Each photon’s energy is a function of it wave number.
The sensor generates a voltage that is the sum of or accumulation of or count of the number of photons.
One can say an IR thermometer is a photon counter. That is in effect exactly what it does. An IR thermometer is a counter of a subset of photons with particular wave numbers. Empirically the count of photons correlates to or is calibrated to a voltage of the IR sensor and that is then presented as a temperature.
One would expect Nightly cooling rate to decrease as temps approach the dew point as heat of vaporization comes into play.
The point of my observation was that in Firgure 1 above the lag over equatorial Africa and west into the Atlantic ocean shows a 0 lag and I seem to recall that from some other earlier post show dust that was a very high dust area.
Global warming is about a global slowing of the cooling rate of the Earth. I believe rate of retransmission or the emissivity of non idea atmospheric gases is controlled by dusts. If one increases the retransmission rate on increases the cooling rate as 50% goes up and some percent of that is instantly gone.
tom Watson
Your last para is right on the money. If the CO2 concentration is increased everywhere, the number of transmitters in the air is increased and the emission into space is more efficient.
Once this is admitted, the remaining argument is that, well, at the lower level the temperature will have to be higher, but that presupposes that there are no transport mechanisms that move heat from the surface to high altitude without being radiated. Convection of water vapour does exactly this, thunderstorms being the most visible example.
Energy retained by CO2 and water vapour, the two main GHG’s, does not follow the modeled path of serially bouncing photons around getting higher and higher like the photons in the centre of the Sun that take 100,000 years to reach the surface. The convective transport mechanisms largely bypass the lower atmosphere radiative pathway.
AGW was a good guess but the net effect is undetectably small, so far. The major failure was to overlook clouds (induced cooling) and the vertical transport of latent heat.
Willis –
I have a possible explanation, and a note to compare.
I think the climate Establishment has a poor vocabulary, and may need some help. In that spirit, some relevant clarifications (mostly for the general reader, so bear with me):
Network analysis includes “delay”s and “lag”s; the former all-at-once, but not immediately, the latter the converse. Both (confusingly) are expressed as time. Our climate system has both, on many timescales. I think your method estimates average or most-probable time, so is properly a lag (as you say).
Except that you infer a time. You’re analysing a cyclic (seasonal) response, so are estimating a *phase-shift*. The significance is that this is always ambiguous (plus 3 months = minus 9, etc). This might muddle discussion of causation.
You’re considering top and bottom air in a CERES cell. These are two points in a circulation or loop. In a closed loop, causation becomes ambiguous. Your post might be read as “OLR change causes SST change” (I’m sure you mean nothing of the sort), but no-one should attempt to read more than “is associated via the loop” into it.
Thunderstorms will be essentially local loops (~hours, km), giving your low/zero phase-shift. A reasonable explanation for your “wet land” finding is: Wet tropical surface (les upthread) = lush vegetation = high transpiration + cloud nuclei = T/S = close-coupling (many Earth IR images show the ITCZ as not prominent, but the three tropical jungle zones as masses of cloud).
Elsewhere the loops may be large, so top and bottom air may be only distantly related (in space, also with delays and lags incl those from the oceans). The Hadley circulation alone gives a ~1 month delay down to the surface.
I have a very simple basic-physics model of the Hadley circulation, with (almost) no free parameters. It offers natural explanations for ELR, Hadley subsidence rate, and limiting SST; also a climate sensitivity: Close to 0.5K/doubling. It’s a steady-state model, so this is an ECS estimate. I discounted the model as its ECS was below any likely TCR, but your finding changes that. The IPCC ECS estimate has AFAIK no good real-world constraint on it. Pending that, it may be that TCR ~ ECS, and the CERES data may approximate a steady-state atmosphere such as I modelled. Your estimate would then be of ECS. If your data is conveniently arranged by zone, I would very much like to see your tropics-only “TCR”.
Willis — Using your approach you show that the ECS is only one quarter of what IPCC claims for doubling of carbon dioxide. This is a step in the right direction but you are not yet there. The actual sensitivity per doubling is zero. This follows directly from the existence of a hiatus that is now 18 years old. No matter how long you have to wait for CO2 to double, temperature does not change and subtracting the temperature at doubling time from initial temperature gives zero, period. This observation invalidates the Arrhenius greenhouse theory used by IPCC and sends it into the waste basket of history, there to join phlogiston waiting for company.The correct greenhouse theory to use is the Miskolczi greenhouse theory, MGT. Its prediction is straightforward: adding carbon dioxide to air does not warm the atmosphere. When this came out in 2007 the global warming gang had it blacklisted. You could not mention it and the grad students were kept ignorant of it. But all scams will eventually come out. The agent responsible for exposing this one is the hiatus/pause. I am afraid that the climate “scientists” still trying to deny it do not deserve to be called scientists. That is because they lack an elementary understanding of how the scientific method works. According to MGT, carbon dioxide and water vapor form a joint optimum absorption window in the infrared whose optical thickness is 1.87 as determined by Miskolczi from first principles. If you now add carbon dioxide to the atmosphere it will start to absorb in the IR just as Arrhenius says. But as soon as it starts, water vapor begins to diminish, rain out, and the original optical thickness is restored. The added carbon dioxide will of course keep absorbing but reduction of water vapor keeps total absorption constant and no warming is possible. If you now want to claim that the hiatus is temporary as has been repeatedly done by using movable goalposts, I have news for you. There was another hiatus in the eighties and nineties that lasted for 18 years, the same length as the current one has lasted. Now you have two hiatuses to deal with. You have not heard about this one because the climate monopoly of GISS-HadCRUT-NCDC has covered it up with fake warming as I discovered it while writing my book [1]. That is a scientific fraud that needs to be investigated, by people not connected with Climategate in any way. To see the hiatus and the fakery that covers it up, take a look at figures 15 and 32 in my book. I have periodically called attention to it but have been ignored by the establishment. As to the reality of the hiatus, Miskolczi [2] studied behavior of global temperature at length using NOAA radiosonde measurements. He discovered that atmospheric absorption of IR was constant for a period of 61 years while carbon dioxide increased by 21.6 percent. There was no warming, which makes this NOAA data set a complete parallel to the hiatus we have right now.
[1] Arno Arrak, “What Warming? Satellite view of global temperature change” (CreateSpace, 2010)
[2] Ferenc M. Miskolczi, “The stable stationary value of the Earth’s global average atmospheric Planck-weighted greenhouse-gas optical thickness” Energy & Environment 21(4):243-262 (2010)
The Miskolczi paper is available in pdf from , eg, here
http://www.friendsofscience.org/assets/documents/E&E_21_4_2010_08-miskolczi.pdf
21 pages of somewhat demanding argument (well for me anyway), but don’t let that put you off.
He references an earlier paper which will explain some of the definitions :
http://owww.met.hu/idojaras/IDOJARAS_vol111_No1_01.pdf
On second thoughts, do let it put you off. Miskolczi has been rejected, rebutted and refuted by many reliable sources.
Judith Curry has reviewed it here
http://judithcurry.com/2015/01/08/miskolczi-discussion-thread/
Thanks, MikeB. Couldn’t agree more.
w.
Thanks again for all the work.
Two issues, though:
(1) It’s not clear how the adjustment of amplitude for lag in the other post applies to this one. That one’s were lags between depths, while these are not.
(2) I don’t quite understand how the results above are necessarily inconsistent with IPCC values. Mathematically it’s entirely consistent for, e.g., a two-box model to exhibit an amplitude at high frequencies (such as one cycle per year) much lower than its TCR or ECS (if TCR is defined as the temperature change reached in response to increasing concentration at 1% per year until it has doubled). I don’t have time to do the math now, but it looks as though a model characterized by appropriately weighted time constants of around 1/40 and 50 years would come close.
Joe Born: Interesting.
Willis, I got a question. If the lag time for TOA response is only two months, do you see any implications for the TCR/ECS argument? How much warming do you think is in the pipeline and for how long?
Willis:
I have an idea. It is a bit Unorthodox but that is my middle initial (the ‘northodox’ is silent).
Suppose you didn’t accept the IPCC value of 3.7 Watts, which is as we know feeds inappropriate modelling that fails at the starting gate. We need some better measure of a possible future no matter how the IPCC misrepresents it. The 3.7 figure is a ‘net impact’, really, and it would be fairer to use the net effect value at the start of an analysis such as you have done.
We have a situation in which the model temperature forecasts are all above reality, and we know that at least half the models are totally wrong because they are on the high side of the median and will never be correct. 3.7 is input the radiative value, but we know it cannot be that high on a net effect basis, were other model issues to be corrected. Instead of correcting the models (which we can’t), change the value 3.7 to a lower net value that is more likely to give a model output no higher than the present model mean. It is a poor man’s way of running the models with half the resulting effect.
It is manifestly clear that the 3.7 Watts is incorrect for (at least) the reason that the thunderstorm vent hypothesis is valid, and perhaps so are other major considerations ignored. So, how about taking the IPCC models and dropping everything that comes from all models that predict values over the median? This cutoff point is arbitrary, but no more arbitrary than models which are that far wrong. This would leave the lower half. I have not looked, but suppose there were half the number of models remaining.
As time passes and the pause continues, additional models could be ruled out and the 3.7 value reduced for your input variable .
Then using the new value of 1.85 Watts, say, which might have a chance of being inside the zone of the model forecast ranges and reality, re-run the scenarios. From what I understand of your math, no, wait, I don’t understand your math so I will keep quiet. It is just that the final result is likely to be reduced quite a bit but will it still be 1/4 of 1.85? If not, you might get convergence on some value.
The reason to do this is to try to start with a forcing that is halfway realistic on a net basis, not to use a figure jiggered and poked to be as high as possible to raise the model mean. Starting with 3.7 is agreeing to debate on turf of their choosing, turf we already know is probably in La-La-Land.
Thanks.
“Suppose you didn’t accept the IPCC value of 3.7 Watts, which is as we know feeds inappropriate modelling that fails at the starting gate. We need some better measure of a possible future no matter how the IPCC misrepresents it. The 3.7 figure is a ‘net impact’, really, and it would be fairer to use the net effect value at the start of an analysis such as you have done.”
well, the problem is 3.71 watts is not a net impact. It is the no feedback response to doubling.
Try as you might to get rid of this number you will not. It’s working science. Any engineer who works with LBL would laugh at you if you suggested it wasnt correct. The entirety of radiative theory and practice would have to change and since it is used successfully day in and day out by engineers nobody would listen to you.
What you should really look at is what Willis actually calculated. Once you do you will see why it is wrong on its face. And then you will understand why these musings never make it further than a blog post.
Steven Mosher May 24, 2015 at 3:18 pm
Well, I don’t see why it is “wrong on its face”. So how about you get off your high horse, lose the paternalism, and explain to us all just where I went wrong?
As to whether my “musings never make it further than a blog post”, three of them have been peer-reviewed and published in the scientific journals, and another one is now in peer-review … can you say the same about your blog postings?
So you can gently insert your “never make it further” meme in some orifice or another, it’s total BS. Why do you try this mud-slinging? I doubt that you are truly out of scientific points … but when you start throwing mud, it sure looks like you’re out of ammunition.
Gotta say, Mosh, once again your habit of posting drive-by comments with extravagant unsupported claims and untrue ad-hominem accusations is not doing your reputation any good. Not sure why I care, it’s your reputation after all, but I do care ..
w.
Mosher has a good point. What you have done here, if I understand correctly, is to calculate the lag time between the seasonal variation in insolation and the seasonal variation in temperature.
We have all experienced that there is a lag, because we see that the warmest summer days in the Northern hemisphere normally occurs in July or August, although the maximum insolation is on midsummer June 22nd. Similarly will the coldest winter days normally occur in January or February, not December 22nd.
We also know that this lag is longer over oceans than over land. I think Willis has made a good job job in quantifying and graphing this interesting lag globally.
However, for several reasons, I do not think that the correlation between seasonal insolation and temperature can be used as even a rough estimate of the climate sensitivity.
One is that the insolation is oscillating. That means that the temperature lag is shortened. When we for example experience higher summer temperatures in the end of August, than we had in the end of June we do so in spite of the fact that the insolation has decreased considerably.
A situation more equal to the lag time from global warming would be to imagine if the axial tilt of the globe had been stopped in midsummer. The expected temperature in the Northern Hemisphere would then increase for every month after midsummer. July would not be so much different from ordinary July months, but August would definitely be warmer and September still warmer, and so on.
In this thought experiment, which is more similar to increased greenhouse effect, we would then see a much higher increase in the summer temperature than what we actually do, and therefore have a higher effect on temperature of more insolation.
That is the main reason why we cannon conclude anything about climate sensitivity from these measurements. Well where are other major reasons, such as the short timescale, but I think the reason above is enough.
/Jan
Mosher writes “well, the problem is 3.71 watts is not a net impact. It is the no feedback response to doubling.”
And in your mind, that makes it a minimum, doesn’t it, Steve.
Steven
My comment was addressed to Willis who I have confidence will understand the suggestion and how to implement it. You have missed my point entirely as shown by your condescension.
I was not challenging the calculated value of the 3.7 Watts, please re-read…no, don’t bother. I will put it into simpler wording for you.
I was pointing out that when a suite of models have 3.7 Watts as the input value, they produce predicted outcomes that are so far from reality we are very safe in eliminating the worst half of them in attempts to find the effective value of all inputs. I have suggested a way of correcting the models making parallel computations by eliminating 50% of the models themselves, while simultaneously producing a trial input value Willis needs to run his calculations. [In case you want to get all pedantic on me, I suggest eliminating those above the median output value, not exactly 50% of the models.]
This is a conceptually sophisticated yet straightforward way to try to help resolve the mismatch between the average of models as a group, and reality.
First, the 3,7 Watts is not working alone, it is an input that can be combined with something left out of the model: Clouds, specifically the thunderstorm cooling hypothesis. If the models had included the thunderstorm cooling the effective value of (3.7 x thunderstorm cooling) would be the same has including thunderstorms and perhaps 1.85 Watts. Engineers have no problem with such generalisations when thinking out loud on a napkin.
You have not criticised the models for leaving thunderstorm cooling out, I noticed. I have however included it not by accessing and analysis and rejigging all 73 models, but by the simple mechanism of reducing the 3.7 by half to compensate for the (still unknown precise effect of) thunderstorms and using that as a correction factor for the obviously wrong models that forget to consider the cooling through the Hot Spot by convective mass transfer to high altitude. An engineer will also have no problem with this approach because they think conceptually and quantitatively.
Engineers know full well that +3.7 times an erroneously omitted negative value can be approximated by +3.7/n as a starting point, for example n = 2, correcting by at least half the unskilled suite of models used by the IPCC.
I have further suggested that it may have to be 3.7/4 in order to rein in the models to the point that their lower bounds of confidence encompass the actual temperatures recorded. Certainly for the upper half of models running hot, measurements are outside the 95% confidence range.
If you can’t keep up, please stop commenting and leave it to the engineers, dishwashers and pilots who take the time to learn how to read, analyse and, especially, think. It wastes my time re-explaining things on a simpler level.
Willis: I a saying that it was unfair to your initial idea to accept the IPCC’s 3.7 Watts, when we all know the models leave ‘large things out’. If your outputs are non-linear, there may be convergence which feeds back to the input Wattage that would give the ‘right’ model output matching reality.
Model + 3.7 + omitted variables = reality = models +3.7/n
Thanks.
an Kjetil,
Huh? The lag is a measure of the time difference between when the radiation turns down (e.g. June 21) and when the temperature turns downwards. This does NOT affect the slope calculation because our seasons are long enough. It would be distorted if the lag was longer than 6 months, so that Dec 21 came before the temperature started to cool off.
The slope of a graph of T vs R is NOT distorted in the way you claim. The slope is dT/dR and that is Climate Sensitivity.
For Jan – just adding to your statement, “We also know that this lag is longer over oceans than over land. I think Willis has made a good job job in quantifying and graphing this interesting lag globally.” That greater lag being probably due to the greater heat sink capacity of the ocean.
Thank you for comment Viking, but the lag is shortened.
Let me explain by an example from the lag in summer heat.
Let’s fist look at a hypothetical place “A” where there is no lag at all.
In “A” June 22nd would on average be the hottest day of year because the insolation is largest that date. The June 23rd would be a little bit colder and June 24th still a little bit colder, on average.
Let us then look at a place “B” with exactly two months lag. In “B”, August 22nd would on average be the hottest day because it is exactly two months after the day with largest insolation.
August 23rd would on average be a little bit colder and August 24th still a bit colder.
But the temperature on , let us say August 24th is not only dependent on the insolation two months before, it is a result of the insolation the same day and every day before in a very long time. And the last days are those which count most.
The main reason August 24th would be a little bit colder than August 22nd is that the insolation on August 23rd and August 24th is lower than August 22nd and August 21st.
If we, as I described in my thought experiment above, had a situation where the tilting of the globe stopped at midsummer, so we had constant midsummer insolation, the August 24th would on average become hotter than august 22nd, i.e. the lag would be extended.
That’s all.
It is also worth noting that the IPCC does not count lag in months; it is expected to take at least 100 years before the temperature stabilizes after the radiation increase has stopped.
/Jan
Jan Kjetil,
You seem to be coming at this from a place (like CS?) without much exposure to RL circuits, sinusoidal inputs and phase shifts. They are central to Electrical Engineering.
Short version: Take a look at this image As a voltage is applied, the current doesn’t start flowing right away. It lags the voltage signal. The current signal is the integral of the voltage signal. When the voltage signal is zero, the slope of the current signal is zero ( V = dI/dt ).
In an analogous way, Energy (Joules) in earth is integrating (adding up) the TOA net Power (Watts). P = dE/dt.
If a location on Earth was like a pure inductor, the phase angle would be π/2 or 90 degrees, which would be 3 months. The real situation is of course complex, but is similar to an RL circuit. In some places, it’s dominated by R (fast response), while other places, it’s dominated by L (slow response).
The time lag is a function of the attributes of the system, or as Willis says “The lag is a function of the material making up the surface”.
Here you can see an animation that shows that the time constant T = L / R (attributes of the system).
In short, your statement “but the lag is shortened” [by the fact that input is oscillating] is false.
Viking,
You are talking about circuits with sinusoidal inputs.
That is a good model for the oscillating seasonal insolation, but the problem is that the climate response is not responding to an oscillating force at all, and therefore it will be wrong to use conclusions from the oscillating model to a totally different reality. The climate forcing is more like a step function which rise and ultimately stabilize on a higher level.
The lag in Equilibrium Climate Sensitivity after Transient Climate Sensitivity is shown in TAR fig. 9.2: http://www.ipcc.ch/ipccreports/tar/wg1/345.htm
As you see, if the CO2 stabilize after 70 year, it will still take another 100 years to stabilize the temperature. In Willis oscillating model the time horizon is only three months.
/Jan
Let me suggest another reason why there a couple months delay in peak temp after peak isolation, the Sun is still north of the equator, and so there still a large amount of hot humid air available to move north, and it’s this air that maintains the warming.
As evidence of this, the rate of change in day to day min temp, switches from positive to negative in late June or so, at the same time the Sun starts it’s trip back towards the south.
The maximum rate of warming is March, April, and the max rate of cooling is around October.
This is also consistent with nightly cooling on annual average being larger that day time warming, the excess heat is from this same tropical evaporation, somebody should let Kevin know we found his missing heat.
Also aiding this is the time constant of the ground cooling.
You really should take a course in control system theory. Or you could watch the animation I linked to. Then, you might learn that what you just wrote is simple ignorant hogwash. The characteristics of the system don’t change based on the type of input. The animation clearly shows the response to a step input, which is what your argument has now shifted to. (what happened to the “lag is shortened based on input?).
Even with a step input, the time constant is a property of the system, not the input. The response to a sinusoidal input reveals the characteristics, and so we can now predict what the response would be if it was presented with a step input.
However, the bottom line is that we’re talking about analyzing data here. There is no NET TOA data that shows a step input. Even if CO2 warming was taking place, it would not be a step input, but rather a change in the shape of the sinusoidal input.
Viking,
I have thought it through and I must realize that I have made an error here. The dynamics of a system does not depend on the input.
However as I said, I had several reasons to think that this model which describes the dynamics in points on the surface is unusable to determine the climate sensitivity of the Earth as a whole, and here is my next:
The heat in a certain point on the surface can basically disappear in two ways, either as radiation out in space or as mixing with other parts of the air and oceans.
On the other hand, the heat from the Earth as a whole can only disappear as radiation out in space.
Therefore the dynamics for a point is very different from the dynamics of the globe as a whole.
/Jan
As with so much of Willis work, it is very interesting how he takes existing data and views it in different ways.
But, I feel its more important worth is that it will act as an enabler, by triggering research by others into this data and rounding out these points and ideas into a greater whole.
Well done Willis.
What’s up w/ my posts?
Do you really want my frank opinion?
As to why they disappear? or?
“This is not the definition of ‘greenhouse effect’ that is commonly used in climate science, whereby it relates (only?!!) to the atmospheric emission and absorption of infrared radiation.”
This definition is incomplete. SWIR/LWIR are bee farts in a hurricane compared to water vapor Without water vapor you don’t have a greenhouse, you have an oven.
IPCC AR5 TS.6 Key Uncertainties is where climate science “experts” admit what they don’t know about some really important stuff. IPCC is uncertain about the connection between climate change and extreme weather especially drought. IPCC is uncertain about how the ice caps and sheets behave. Instead of gone missing they are bigger than ever. IPCC is uncertain about heating in the ocean below 2,000 meters which is 50% of it, but they “wag” that’s where the missing heat of the AGW hiatus went, maybe. IPCC is uncertain about the magnitude of the CO2 feedback loop, which is not surprising since after 18 plus years of rising CO2 and no rising temperatures it’s pretty clear whatever the magnitude, CO2 makes no difference.
http://www.writerbeat.com/articles/3713-CO2-Feedback-Loop (1,575 reads)
Barring some serious flaw in science or method, Miatello’s paper should serve as the death certificate for AGW/CCC.
So what about Miatello?
http://principia-scientific.org/publications/PSI_Miatello_Refutation_GHE.pdf
sure, why not.
Mike May 24, 2015 at 3:16 am
I would certainly argue that any energy exported from a gridcell reduces the thermal response, and vice versa. As to whether it would average out, the amount exported has to agree with the amount imported, doesn’t it?
Mike, first, thanks for a most interesting analysis of the Pinatubo eruption.
However, regarding the question of the Lissajous type of analysis that I’m doing, my analysis depends on a curious fact. This is that the exponentially decayed version of a sine wave is merely a lagged and rescaled version of the input signal. Which means, of course, that the exponentially decayed version of a sine wave is itself a sine wave.
As a result, doing a straight lag of the type I’ve shown in the Lissajous figures above wouldn’t be improved by using an exponential-lag based model, because that gives the same results as the simpler model I used.
Where your method comes into its own is in analyzing a single pulse, such as your nice work with Pinatubo.
Regarding the question of the asymetrical “figure eight” shape of the figures, I ascribe that to differences in the heating and cooling curves of the ocean. I suspect this is due to the fact that the ocean is heated in 3-d but only cools from the surface.
I say this in part because the corresponding graphs over the land have a much more linear form, and the land is both heated and cooled via the surface.
However, that is speculative in nature.
Again, thanks for your Pinatubo analysis, I need to review it in more detail.
w.
Willis
Your 2 month ocean lag findings corresponds to David Stockwell’s model of Pi/2 (90 deg) lag for a cyclic temperature oscillation. i.e. 3 months for the annual 12 month cycle (with less over land.) See “Key Evidence for the Accumulative Model of High Solar Influence on Global Temperature”.
[Attempt to post] (I’ve been getting a notice that my comment can’t be posted.)
Willis Eschenbach:
I had thought TCR as used by the IPCC was the temperature increase that results from increasing CO2 concentration by 1% per year until it doubles. If so, the connection between the “TCR” Mr. Eschenbach has calculated and the IPCC’s quantity is not clear.
yields:
per kelvin, 
per kelvin, 
per kelvin, and 
per kelvin. If 
is the forcing level for doubled CO2 concentration, we can see by inspection that ECS = 
.
, where 
, we get a sinusoidal temperature response of 
K: a small amplitude and a two-month lag. If instead we apply a ramp stimulus 
for 70 years, we find that the temperature at the end of that period, i.e., what the IPCC calls TCR, is 1.9 K.
To appreciate that, consider the “two-box” model defined by the following system of equations:
Eliminating
Suppose we adopt the following parameter values:
If we then solve the above equation for a sinusoidal stimulus
Now, I picked those parameters hastily, and it turns out they imply a heat capacity only about a tenth of what I would have expected. But I’ll bet that one could pick other, more-realistic parameter values and get results that are not dissimilar.
So it’s not a self-evident fact that the IPCC values are inconsistent with the quantities Mr. Eschenbach has calculated.
I would have lost that bet. If the problem is constrained by the lag and magnitude values Mr. Eschenbach observed and the ECS and TCR values from the IPCC, the problem is fully constrained: the two-box model I assumed has only four parameters.. To get greater heat capacities would require that lag, magnitude, ECS, and/or TCR change–or that the model be changed by adding another path by which heat leaks from the system.
CAGW sits on that infamous, figurative, three legged stool.
Leg one: Anthropogenic CO2 sources contributing to the atmospheric CO2 concentration between 1750 and 2011 are based on cherry picked assumptions, estimates, and wags to generate the desired result when in fact nobody really knows with certainty. (APS workshop) Could be natural sources, but IPCC’s mandate prohibits consideration.
Leg two: Compared to the climate’s natural ebb & flow of heat/energy over time (i.e. power & watts) the 3.0 W/m^2 of GHGs is inconsequential. Especially when the ocean/water vapor cycle is considered.
Leg three: The dire predictions of warming, ice melting, sea levels rising are based on GCMs (RCP 2.6, 4.5, 6.0, 8.5 W/m^2) which the pause/hiatus has rendered worthless. Even IPCC admits uncertainty about the GCMs value.
These three points are all that really matter and CAGW theory joins cold fusion, phlogiston, and the geocentric flat earth.
Mr. Eschenbach,
A nice data point for sanity checking.
It is a pity that there’s no data from 1880 to compare with; I only say this because I believe the warming to date is 0.7 degrees C – and we had had a similar CO2 forcing increase since then (?)
Shouldn’t the caption text that starts with “All trends given are in degrees C per 3.7” be associated with figure 2 instead of 1?
“shurely shome mishtake?”
water vapour should increase in the warmer summer months while co2 should decrease because of photosynthesis but water vapour should decrease in cold winter months as co2 increases because of photosynthesis stopping. can co2 really have a lagged effect on increasing temperature when its maximum effect is during the colder winter months when temperatures are falling.
When the air is dry, temps fall like a rock at sunset, everywhere the air is dry it’s the same.
Modern warming isn’t from Co2, it’s from the ocean cycles, and remember we didn’t discover the PDO until after CAGW was hypothesized as the cause.
Willis,
Perhaps too late..
I think you have misunderstood what climate science is specifically trying to calculate (as collated in various IPCC reports).
The question is, if the temperature is perturbed, what is the change in top of atmosphere radiation? For example, if surface temperature increased 1K, what is the TOA radiation (flux) change?
If there was no positive or negative feedback, the 1K increase would lead to an increase in outgoing longwave radiation (OLR) of 3.6 W/m^2 (example in Clouds & Water Vapor – Part Eight – Clear Sky Comparison of Models with ERBE and CERES). Various results (including the example paper shown in the linked article) show that OLR doesn’t increase by 3.6 W/m^2, instead something like 2 W/m^2, indicating some positive feedback.
What you are looking at in your article is the change in surface temperature from a change in net TOA radiation (OLR – absorbed solar radiation). This is quite a different perspective (causality in the other direction) so obviously you are not going to see the same result.
Given the respective heat capacities involved I suspect it would be amazing to see a large change in the surface from a TOA net change. That said, I have just in the last 2 weeks been downloading and trying to understand the raw data from CERES & AIRS (along with the “model based” NCAR reanalysis data) for the last 14 years and don’t feel like I yet have a useful interpretation of your results.
scienceofdoom:
What you are looking at in your article is the change in surface temperature from a change in net TOA radiation (OLR – absorbed solar radiation). This is quite a different perspective (causality in the other direction) so obviously you are not going to see the same result.
Even if they were speaking of the same causality direction, though, Mr. Eschenbach’s calculations would not be very relevant to what I understand TCR and ECS to be. The earth is like a complicated low-pass filter, and Mr. Eschenbach measures its response to a high-frequency stimulus. TCR is its response to a 70-year ramp, while ECS is the DC response. My comment here illustrates that by reference to a “two-box” model. You may prefer different parameters.
That second paragraph was supposed to be a block quote. I must have used the wrong tags.
Joe Born, May 26, 2015 at 2:57 am:
“Even if they were speaking of the same causality direction, though, Mr. Eschenbach’s calculations would not be very relevant to what I understand TCR and ECS to be.”
Well, most of all, Eschenbach’s use of ‘net ToA radiation’ is mixing cause and effect into one parameter, so is only confusing the issue and is not providing any useful results. You can’t determine climate sensitivity to an increase in CO2 and/or WV looking at these data. The ‘net SW input’ (solar) leads lower tropospheric temps by 2-3 months on average, while ‘LW output’ (terrestrial OLR) lags lower tropospheric temps by 0-1 months on average. The OLR at ToA simply tracks tropospheric temps as a true radiative effect should and does, no gradual divergence either upward or downward over time to be observed.
SOD,
What if there is no change in surface temperature but instead a change in lapse rate slopes ?
Are you aware of the concept of hydrostatic balance?
Do you realise that a surface beneath an atmosphere radiates less photons as a result of an increase in collisional activity and therefore that a surface at 288K beneath 1 bar pressure of atmospheric mass actually only radiates as if the temperature were at 255K without such an atmosphere?
Stephen Wilde,
I think we have attempted discussion of climate science before. I suggested equations were important for helping us distinguish between entertainingly bad ideas and science. Equations are found in physics textbooks and are used as reference points between correct science and fantasy.
I recommend interested readers to follow that exchange.
If however you are not the author of the comments in that article let me know and I may review your comment here.
scienceofdoom, May 26, 2015 at 1:03 am:
“The question is, if the temperature is perturbed, what is the change in top of atmosphere radiation? For example, if surface temperature increased 1K, what is the TOA radiation (flux) change?
If there was no positive or negative feedback, the 1K increase would lead to an increase in outgoing longwave radiation (OLR) of 3.6 W/m^2 …”
Peculiar. There is no overall increase in OLR allowed if an “enhanced GHE” caused that 1K temp rise of yours. The 3.6 (or 3.7?) W/m2 increase in outgoing LW in this case would be back up to the original level. A sudden doubling in CO2, for instance, would reduce the OLR at the ToA by 3.7 W/m2, forcing the surface to warm by 1K for the Earth system to be able to once again put out those final 3.7 W/m2 and balance the budget – back to square one, only 1K warmer. In the real world, the CO2 increase is incremental, so we never get to see that sudden plunge in OLR, because the Earth system continuously adjusts by gradually increasing its temperature parallel to the increased forcing, keeping the OLR trend flat over time while the surface and the troposphere warms.
SOD,
That was me but I have moved on since and now see where you have been going wrong.
Perhaps you could reply to my above comment here so that I can assess the validity of your response.
I don’t think one needs equations at this point. The dry adiabatic lapse rate slope clearly shows how the rate at which photons are emitted declines with increasing density through an atmosphere to the surface.
For a non GHG atmosphere at 1 bar pressure the surface must be at 288k in order for it to radiate to space at 255k past the conducting and convecting barrier of atmospheric mass.
Stephen Wilde:
“..and therefore that a surface at 288K beneath 1 bar pressure of atmospheric mass actually only radiates as if the temperature were at 255K without such an atmosphere?”
This statement is not correct. Thermal radiation from a surface is governed by the Stefan-Boltzmann equation, r = %epsilon;σT^4 – the factors are the material properties (emissivity) and the temperature of the surface.
Therefore, the atmosphere pressure above is irrelevant for the emission of radiation from a surface. There is no possibility of confusion about this. No possibility of confusion, that is, for people who can look up an equation and understand an equation.
..my last comment should have: r= εσT^4
SOD,
The S-B equation does not apply to a surface sandwiched between two grey bodies which are exchanging energy by conduction and convection.
It must be applied from a point external to the object being observed so that there is no conduction or convection passing between the two locations.
On that basis it must be applied from a point in space which is near enough to a vacuum for present purposes.
Earth radiates 255K to space exactly as predicted by S-B.
S-B tells us nothing about the pattern of heat distribution within an atmosphere surrounding a planet nor about the temperature of the surface beneath that atmosphere.
Climate ‘science’ has negligently applied the S-B equation to a surface beneath a conducting and convecting atmosphere. It was never designed for that purpose.
Conduction and convection are slower than radiation so there will always be temperatures higher than S-B within a planet / atmosphere system and the temperatures will reflect the mass density distribution.
Stephen Wilde:
“The S-B equation does not apply to a surface sandwiched between two grey bodies which are exchanging energy by conduction and convection.”
Hilarious.
Here are two textbook extracts for emission of thermal radiation.
Whenever I produce multiple textbook extracts on the emission of thermal radiation, no one ever produces a textbook which says anything different.
After you produce your textbook with a different equation I will comment further. Otherwise there is no point.
Emission of thermal radiation is in the form of photons.
You seem not to have understood your text books.
Collisional activity between molecules transfers kinetic energy by conduction directly between them through contact which reduces the total energy of the warmer molecule. That reduces the probability of photon emission from the warmer molecule.
The same packet of kinetic energy cannot be used to emit a photon if it is being transferred directly between molecules by contact.
On average, overall, the frequency of photon emission declines as one moves down through atmospheric mass because collisional activity is increasing in its stead.
Consider this:
i) At the top of the atmosphere there is little or no collisional activity so molecules at 255K radiate to space at 255K
ii) Inside a solid body ,beneath its surface, there is little or no emission of photons , it is all collisional activity in the form of conduction but the surface will radiate at 255K in the absence of an atmosphere.
It follows that if one then adds an atmosphere one moves from a vacuum to a solid over a greater distance so that conduction takes over progressively from photon emission as density increases.
The DALR draws that line through the atmosphere from the vacuum of space to the solid surface below.
Conduction, in progressively taking over from photon emission, causes the surface temperature to rise. Less photons are emitted from a surface beneath an atmosphere RELATIVE TO THE TEMPERATURE. The surface still radiates 255K to space but it must be at 288K to hold the weight of the atmosphere off the ground in hydrostatic balance AT THE SAME TIME.
Convection within the mass of the atmosphere smears the temperature difference between the surface and space across a distance equivalent to the height of the atmosphere.
Thus the mass induced greenhouse effect which was the standard explanation from established science 50 years ago.
Unfortunately, the astrophysicists who took over climate science know nothing about the effects of non radiative energy transmission within atmospheres.
So, SOD,
How exactly do you propose that a molecule can emit photons at the same rate both with and without collisional energy transfers going on at the same time and a convective circulation to maintain as well?
Where in the S-B equation is any consideration given to ongoing conduction and convection?
How can you and your chums possibly show that it is correct to apply S-B to a surface beneath the conducting mass of a convecting atmosphere?
How much of your site has been a waste of space and of the time of readers and yourself?
The correct way to see the situation is to recognise that on average every radiating molecule from surface to space is radiating at 255K as per the S-B equation.
Every such molecule then carries an additional load of internal energy ‘worth’ 33K. That additional energy is derived entirely from conduction.
For molecules at the surface that 33K is in the form of sensible kinetic energy and at the top of the atmosphere that 33K is in the form of non-sensible potential energy.
The DALR marks the point of transition between kinetic and potential energy up the entire vertical column.
Any molecules that find themselves too warm relative to their height will rise and any that are too cool will fall. The hydrostatic process makes that inevitable.
If radiative molecules become warmer than 255K then they rise and in the process cool both by additional radiation to space and conversion of KE to PE until they are radiating at 255K once more.Whilst rising they distort the DALR to the warm side.
If radiative molecules become cooler trhan 255K then they fall and in the process warm both by additional absorption of IR from the surface and conversion of PE back to KE until they are radiating at 255K once more. Whilst falling they distort the DALR to the cool side.
SteveW wrote: “The correct way to see the situation is to recognise that on average every radiating molecule from surface to space is radiating at 255K as per the S-B equation.”
Scientists measure the absorption spectrum at different pressures and temperatures because the line-shape of the absorption bands changes modestly with pressure (collision broadening) and temperature (Doppler broadening). It is laughable to suggest that we don’t know how pressure effects absorption and emission.
SteveW wrote: “Where in the S-B equation is any consideration given to ongoing conduction and convection?”
The molecules radiating according to the S-B equation don’t “know” that they are part of a larger system that may or may not also be involved in convection and conduction. Molecules conduct when they collide and gain or lose kinetic energy. Some collisions create an excited vibrational state. SInce collisional excitation and relaxation is much faster than emission or absorption of a photon in the troposphere and stratosphere, molecules radiate at a rate that depends on temperature and excited state energy (Planck’s function B(lamba,T) ). Convection represents motion of a large group of molecules and vertical convection requires in unstable lapse rate. All three processes operate INDEPENDENTLY according laws that have survived numerous experimental test.
Furthermore, molecules don’t have any way of sensing the local pressure so that they can determine how frequently to radiate a photon. The Boltzmann distribution of molecular speeds for a given temperature controls the likelihood that a collision will produce an excited vibrational state, so temperature determines the fraction of molecules that are in an excited state. Pressure is a bulk property that emerges when a large number of molecules are confined in a particular volume.
Molecules don’t conspire to produce a predetermined outcome (such as the whole atmosphere radiating as if it were 255 degK); circumstances and laws produce a particular outcome. People conspire to produce a pre-determined outcome.
SteveW wrote: “How much of your site has been a waste of space and of the time of readers and yourself.”
In this post at WUWT, Willis doesn’t recognize that he has mistaken radiative imbalance for radiative forcing when calculating an perverted form of TCR, a measure of climate sensitivity. When his mistakes are pointed out by someone who understands them, there is no response. You won’t find such mistakes and irresponsibility at SOD. You will find references to articles and information likely to be reliable.