Guest Post by Willis Eschenbach
The CERES dataset contains three main parts—downwelling solar radiation, upwelling solar radiation, and upwelling longwave radiation. With the exception of leap-year variations, the solar dataset does not change from year to year over a few decades at least. It is fixed by unchanging physical laws.
The upwelling longwave radiation and the reflected solar radiation, on the other hand, are under no such restrictions. This gives us the opportunity to see distinguish between my hypothesis that the system responds in such a way as to counteract changes in forcing, and the consensus view that the system responds to changes in forcing by changing the surface temperature.
In the consensus view, the system works as follows. At equilibrium, what is emitted by the earth has to equal the incoming radiation, 340 watts per metre squared (W/m2). Of this, about 100 W/m2 are reflected solar shortwave radiation (which I’ll call “SW” for “shortwave”), and 240 W/m2 of which are upwelling longwave (thermal infrared) radiation (which I’ll call “LW”).
In the consensus view, the system works as follows. When the GHGs increase, the TOA upwelling longwave (LW) radiation decreases because more LW is absorbed. In response, the entire system warms until the longwave gets back to its previous value, 240 W/m2. That plus the 100 W/m2 of reflected solar shortwave radiation (SR) equals the incoming 340 W/m2, and so the equilibrium is restored.
In my view, on the other hand, the system works as follows. When the GHGs increase, the TOA upwelling longwave radiation decreases because more is absorbed. In response, the albedo increases proportionately, increases the SR. This counteracts the decrease in upwelling LW, and leaves the surface temperature unchanged. This is a great simplification, but sufficient for this discussion. Figure 1 shows the difference between the two views, my view and the consensus view.
Figure 1. What happens as a result of increased absorption of longwave (LW) by greenhouse gases (GHGs), in the consensus view and in my view. “SW” is reflected solar (shortwave) radiation, LW is upwelling longwave radiation, and “surface” is upwelling longwave radiation from the surface.
So what should we expect to find if we look at a map of the correlation (gridcell by gridcell) between SW and LW? Will the correlation be generally negative, as my view suggests, a situation where when the SW goes up the LW goes down?
Or will it be positive, both going either up or down at the same time? Or will the two be somewhat disconnected from each other, with low correlation in either direction, as is suggested by the consensus view? I ask because I was surprised by what I found.
The figure below shows the answer to the question regarding the correlation of the SW and the LW …
Figure 2. Correlation of the month-by-month gridcell values of reflected solar shortwave radiation, and thermal longwave radiation. The dark blue line outlines areas with strong negative correlation (more negative than – 0.5). These are areas where an increase in one kind of upwelling radiation is counteracted by a proportionate decrease in the other kind of upwelling radiation.
How about that? There are only a few tiny areas where the correlation is positive. Everywhere else the correlation is negative, and over much of the tropics and the northern hemisphere the correlation is more negative than – 0.5.
Note that in much of the critical tropical regions, increases in LW are strongly counteracted by decreases in SW, and vice versa.
Let me repeat an earlier comment and graphic in this regard. The amounts of reflected solar (100 W/m2) and upwelling longwave (240 W/m2) are quite different. Despite that, however, the variations in SW and LW are quite similar, both globally and in each hemisphere individually.
Figure 3. Variations in the global monthly area-weighted averages of LW and SW after the removal of the seasonal signal.
This close correspondence in the size of the response supports the idea that the two are reacting to each other.
Anyhow, that’s today’s news from CERES … the longwave and the reflected shortwave is strongly negatively correlated, and averages -0.65 globally. This strongly supports my theory that the earth has a strong active thermoregulation system which functions in part by adjusting the albedo (through the regulation of daily tropical cloud onset time) to maintain the earth within a narrow (± 0.3°C over the 20th century) temperature range.
w.
As with my last post, the code for this post is available as a separate file, which calls on both the associated files (data and functions). The code for this post itself only contains a grand total of seven lines …
Data (in R format, 220 megabytes)
Robert Clemenzi says:
January 8, 2014 at 6:22 am
I don’t believe that for one minute. Citation?
My bible in these matters, “The Climate Near The Ground” by Geiger, says that going the other way, downwelling radiation, the situation looks like this:
Layer thickness Percent share of downwelling radiation
1st 87 metres above the ground — 72%
Next 89 metres above the ground — 6.4%
Next 91 metres above the ground — 4%
So the majority (72%) of the downwelling radiation comes from the first 300 feet of atmosphere above us, and 82% comes from the first thousand feet. Given that, the idea that the upwelling radiation is absorbed in a single foot of atmosphere seems highly unlikely.
w.
Phil. says:
January 8, 2014 at 7:33 am
Nonsense. If that were the case, the upwelling radiation measured by CERES would be on the order of 390 W/m2, the average radiation of the surface.
But the CERES data says the outgoing IR totals 240 W/m2, not 390, and that’s just what we’d expect. And I can produce for you a very nice map showing exactly how much of the surface radiation is absorbed in different parts of the world. It matches very nicely with the amount of our favorite GHG, water vapor.
Sorry, but you seem to have misread something, because your claim is completely false.
w.
“I hope that when people ask me questions, they quote whatever they are talking about”
I agree. Of course I don’t care much for the way many of the websites work. People should be able to highlight some words and then go something like CTL r and have the quote appear in the reply box. Maybe there’s a way to do that here, but I don’t know what it might be.
Willis top post: ”In response, the albedo increases proportionately, increases the SR. This counteracts the decrease in upwelling LW, and leaves the surface temperature unchanged.”
This would imply in effect the Tmean is forever fixed at ~288K (the thermostat set point) and anomaly about the set point would be zero mean with fixed hysteresis. Thermometer evidence shows Tmean is not a fixed set point. Anomaly reports show monthly differences unlike thermostat fixed hysteresis.
Comments?
Stephen Rasey says:
January 8, 2014 at 8:42 am
As you are the second person to comment on this, I’ve changed the diagrams for greater clarity.
w.
Dave Dardinger says:
January 8, 2014 at 9:18 am
Highlight the words you are interested in, press CONTROL-C, go down to the comment box, press CONTROL-V …
w.
Trick says:
January 8, 2014 at 9:20 am
I would comment, but I’m totally unclear what you mean. Bear in mind what I said above:
w.
“Chop and change as you wish …”
Thanks, I added two more colors but I still only get six in the legend range, yet I can’t see where the number of intervals is assigned.
However, I see the legend numbers are not too accurate, lots of crude rounding going on.
maxcolor=.25,mincolor=-1,roundto=2,legendlabel=””
There is a world of significance difference between -0.6 and -1, can you say exactly what interval is getting coloured as “-1” in your graph?
I think this would really be a lot better if it had a continuous colour scale (or at least 10-20 nuances). As it stands it could be over-selling the result.
No knocking it, there is clearly some good information presented but I’m sure you don’t want to give a false impression that there are large swathes with CC near -1 if that’s not the case.
Isn’t there a question here about the direction of causality? Willis is interpreting the anti-correlation between upwelling SW and LW as support for his theory about cloud formation acting as a thermostat (a theory that I find compelling). But there is also a simpler explanation for this anti-correlation. Where clouds block incoming solar the planet below warms less, leading to less outgoing LW. It seems likely to me that this direction of causality (where cloudiness is the initiating cause) dominates the data, making it hard to say anything about what causality might be going on in the other direction (where cloudiness is the effect).
Would be similar to the correlation between temperature and CO2, where the paleo-data is dominated by the direction of causality where rising temperatures cause CO2 to bubble out of the oceans, making it difficult or impossible to discern causality in the other direction. CO2 certainly COULD be having a warming effect, and we know on theoretical and experimental grounds that it should have a small forcing effect, but the paleo-data gives us almost no information about its net effect.
I think the thermostat hypothesis is correct but I’m not sure that this particular anti-correlation provides much or perhaps even any evidence for it.
“This would imply in effect the Tmean is forever fixed at ~288K”
Do you understand what correlation coefficient means. If not, you will need to understand the post and the graph.
Willis, as per your request, this is the comment from MikeB that I was referring to:
About half of TSI is longwave in the first place
You probably say this because someone told you that half of the incoming solar radiation is in the infrared. But this is the near infrared, it is not longwave infrared. The proportion of solar radiation with wavelength greater than 5 microns is negligible in comparison to the radiation emitted from the Earth’s surface itself. It’s safe to say that if we detect radiation shorter than 4 microns then it is from the Sun (or a rocket engine or a furnace) and that infrared radiation above 5 microns is from the Earth or its atmosphere.
All warm bodies emit electromagnetic radiation. The distribution of that radiation accords with Planck’s Law and depends only on the body’s temperature and its emissivity. To find where the peak emission will be simply divide body’s absolute temperature into 3000. For example, a body at a typical Earth temperature of 300K will have a peak emission of 3000/300 = 10microns. On the other hand the Sun, with a surface temperature of 6000K, will emit its peak radiation at 3000/6000 = 0.5 microns. This is Wien’s Law (or more exactly an approximation to it. Use 2897 instead of 3000 for a precise answer).
How then does a material convert shortwave to longwave?
You can see from the above that a material will emit according to its own temperature. Since the Sun at 6000K does not manage to heat the Earth to 6000K but only to, say, 300K, then the Earth radiation will be LW and the Sun’s radiation is SW.
Are you saying that cloud formation changes the frequency of the Earth’s emitted energy, or are you simply saying more clouds reflect solar SW radiation, but are transparent to emitted LW radiation?
Willis Eschenbach says:
January 8, 2014 at 9:16 am
Phil. says:
January 8, 2014 at 7:33 am
“I think there is a misunderstanding in Willis’s CERES analysis. The upwelling LW channel produced by CERES is Surface IR which is confined to the window wavelength range of 8-12 μm, this range is unaffected by GHGs.”
Nonsense. If that were the case, the upwelling radiation measured by CERES would be on the order of 390 W/m2, the average radiation of the surface.
No, because the window wavelength range of 8-12 μm is only part of the emissions from the surface. CERES explicitly states: “Each CERES instrument measures filtered radiances in the shortwave (SW; wavelengths between 0.3 and 5 μm), total (TOT; wavelengths between 0.3 and 200 μm), and window (WN; wavelengths between 8 and 12 μm) regions.” They also perform numerous calculations on the raw data so there is scaling going on.
But the CERES data says the outgoing IR totals 240 W/m2, not 390, and that’s just what we’d expect. And I can produce for you a very nice map showing exactly how much of the surface radiation is absorbed in different parts of the world. It matches very nicely with the amount of our favorite GHG, water vapor.
They term it “Surface Upwelling Longwave Radiation (rlus) Wm-2” so it is not subject to absorption by GHGs.
Sorry, but you seem to have misread something, because your claim is completely false.
Not according to the CERES site. Perhaps you’re using a different product, which data are you using?
Using this -3.33 W/m2/K value for the cloud feedback drops CO2 climate sensitivity to 0.75C per doubling from the theory’s 3.0C per doubling.
degrees kelvin. Natural variability alone per century is at least this much, and we don’t know how to predict the background natural variation “at all”.
Which is right at the lower bound, from the sound of it, of the new values snuck into AR5.
I’d suggest that this be stated, however, as a range.
This is the sort of thing that one does have to wonder about. Again, looking at the fluctuation-dissipation theorem one should actually be able to find the time-signature of causality in this, although it is going to be much more difficult because heating in one place (say, the tropical ocean) can easily cause cooling somewhere else because of lateral transport of the water vapor before the clouds form. This is obviously the case for nearly all of the clouds forming over the land masses, for the monsoon, etc. There is substantial bulk transport of both latent heat, LWIR emission (rom the clouds “created” from warmed ocean water elsewhere) and albedo.
The best place to look for the signal probably is the monsoon. Those are persistent long-time scale phenomena, and one would expect to see a consistent variation of SW/LW radiation from precisely this lagged heat transport from their primary oceanic vapor sources. Of course extracting any kind of signal from their substantial chaotic variation would be very difficult, and there could easily be other factors that are equally important obscuring the signal).
My recollection is that this is very much like what Roy Spencer did in a short time study of much the same thing (but only in the context of specific regional weather). There the causal time signature was very clear, and he also found the negative lagged correlation suggesting natural negative feedback. It would be interesting to see if this holds globally — instead of doing static correlation do lagged correlation and see if fluctuations are correlated in a lagged manner.
rgb
“I think the thermostat hypothesis is correct but I’m not sure that this particular anti-correlation provides much or perhaps even any evidence for it.”
Since >SST will cause evap, will cause cloud, it’s rather chicken and egg.
The primary driver must be changes in isolation even in the tropics. Perhaps something can be gained for looking at the annual cycle rather than dumping it. If more cloud happens when there’s more insolation, it implies a feedback.
Otherwise we could look at non seasonal changes in insolation:
http://climategrog.wordpress.com/?attachment_id=310
Greg says:
January 8, 2014 at 9:38 am
“Not too accurate”? “Crude rounding”? Take a breath, old son, and back off the insults, your complaint generator is set way too high. I’m happy to field comments and questions, that’s what science is about … but calling my work “not too accurate” and “crude”? Not polite, particularly when you don’t understand what you are looking at.
I use “roundto” to set the number of decimals displayed in the legend. If you don’t do that, you get numbers on your graph like “0.1544398264” in the display, and a graph containing something like that looks … well … crude. However, the variable “roundto” doesn’t affect the calculations, just the display.
In addition, if you think increasing the number of decimals displayed would make the slightest difference to the legend in Figure 2, think again. See below for why.
Consider Figure 2 above. The legend is accurate to 2 decimals. In fact it’s accurate to a hundred decimals, because when you divide the given range (from -1 to 1) into 5 equal intervals, you get 0.4000000… for the width of each interval. Since R doesn’t print trailing decimal zeros, this comes out at “0.6”, “0.2”, etc. So changing the roundto to 5 wouldn’t change the displayed output.
Regarding your other question, the colors show points on the scale, not intervals. The legend (with colors for say 1, 0.6, 0.2, etc) shows the color that is associated with that specific number.
All the best,
w.
If there would be only albedo increase as a feedback to the forcing induced by more longwave absorption then the model presented here could be valid. But there are other phenomena that need to be taken into account in the feedback mechanisms such as Planck’s response, change of lapse rate, water vapour, cloud coverage.
Anyway, it’s good to see that as much goes out as comes in, otherwise we would be freezing or frying.
@Willis:
I’m getting the picture from your comments that you can’t use CERES data for an energy balance, but your comment:
“In the CERES data, both the incoming flux and the outgoing flux are averaged 24/7 over their particular gridcell. They are not general measurements of the total global flux. As a result, there is no such error as the one you imagine.”
also doesn’t make much sense in this context. Isn’t the reason that there is an energy imbalance because the observations don’t cover all wavelengths? If not, then there has been a major error in communication of your results, because I doubt most people just from reading this article could could pass a quiz on where the energy is going, or even if CERES data indicates and energy imbalance. Further clarification may be required vis-a-vis why an energy imbalance in CERES data isn’t an imbalance in total solar energy flux for the earth’s surface.
Rob Ricket says:
January 8, 2014 at 9:57 am
Thanks, Rob. I don’t know that I explicitly said any of those, but in order:
• Cloud formation does indeed change the frequency of the emitted energy.
• More clouds do reflect more SW
• However, clouds are the opposite of transparent to emitted LW. In fact, they are generally considered as blackbodies for IR purposes, as they absorb ≈ 100% of incident longwave radiation.
w.
Willis Eschenbach says:
January 8, 2014 at 10:35 am
• However, clouds are the opposite of transparent to emitted LW. In fact, they are generally considered as blackbodies for IR purposes, as they absorb ≈ 100% of incident longwave radiation.
Each droplet in the cloud will absorb ~100% of the incident light and also scatter an equal amount of the incident light.
However, clouds are the opposite of transparent to emitted LW. In fact, they are generally considered as blackbodies for IR purposes, as they absorb ≈ 100% of incident longwave radiation.
And they strongly emit LWIR as well, and do so high in the atmosphere where there is much less of a GHG “blanket” between them and the TOA/space. As such they are active transport heat sinks: LWIR incident on the ocean surface is absorbed in the top millimeter, transformed almost entirely into latent heat by causing surface evaporation. The evaporated water vapor with its latent heat is vertically transported and cools via conduction with the lapse rate of the surrounding air and via radiation. When it condenses into clouds it gives off the latent heat, now much higher up in the atmosphere. In the case of upper troposphere clouds, the latent heat is relatively quickly lost. Down lower it may still have a substantial amount of atmosphere and lapse rate to work through.
This is why simple one slab or two slab atmospheric models don’t do very well. From some papers recently reported on WUWT, GCMs substantially underestimate the cooling potential of e.g. thunderstorms via the mostly neglected mechanism of direct, rapid transport of head-laden warm moist air aloft to where it is rapidly lost via radiation on the top of most of the GHG layer. Plus, of course, the modulation of albedo, plus the additional latent heat removal at the warmed Earth’s surface when rain falls and re-evaporates from the ground beneath. It’s basically a heat engine and runs by transporting heat from down low to up high.
rgb
Greg says:
January 8, 2014 at 6:01 am
And what is the sky temperature on a clear, dry day? (Well away from direction of sun)
That is away from the sun. The sun is low on the horizon and I measured directly above.
Don’t take offence Willis, I’m trying to make a suggestion to improve what you are showing , not to give offence.
“Regarding your other question, the colors show points on the scale, not intervals. The legend (with colors for say 1, 0.6, 0.2, etc) shows the color that is associated with that specific number.”
I realise what the scale shows , my question is what the graph shows. Clearly it is not just the points that get exactly -1.00000000, that R wants to print as “-1” that get coloured blue.
It seems that the legend label “0.6” is a truncated 2/3 , so what I was wondering was what _interval_ of values get the coloured on the graph.
My guess is -0.67<x<-1 but it could be perhaps half way between -0.67 and -1
I doubt any of the cells actually correlate at -1.0 , so it would be more informative and perhaps less misleading (depending upon exactly what the values turn out to be) to have more graduations.
If I could fix it easily I would have done it and posted the result for you to check out but R is a bitch to work with and I don't have a day free to waste on it's enigmatic and incomplete documentation, to fix something you can probably do in two minutes on code you wrote.
Neither do I have time couch everything in flowery language and conditionals , so please take comments in the spirit they are intended and not get shirty. As you know, I am supportive of what you are suggesting, I'm not trying criticise it other than to improve it and make it more convincing to others who will want to break it.
best regards.
Willis 9:33am: “..I’m totally unclear what you mean. Bear in mind what I said above: This is a great simplification, but sufficient for this discussion.
Global surface temperature 1.5F increased since 1880:
http://climate.nasa.gov/
Thermometer measured surface Tmean isn’t observed unchanged; interested in comments on why there isn’t compensation by albedo observed per your view top post Fig. 1: ”In response, the albedo increases proportionately, increases the SR. This counteracts the decrease in upwelling LW, and leaves the surface temperature unchanged.”
Kevin, “That is away from the sun. The sun is low on the horizon and I measured directly above.”
Well since you said you’d measured cloud, it would not seem to be the “clear dry day” I mentioned. OK, I’ll spell it out. We know emissions are proportional to T^4 where T is absolute temperature.
Space is about 3 kelvin. The difference between -30 C (~243)^4 and and 273^4 is not so great compared to 3^4. Now the atmosphere will still be emitting more than 3K but you can start to get my point. On _dry_ day (not just a gap between clouds on a day with high humidity) the sky can be much colder that what you related. A lot of what you measured as “-30” was thermal emission from water vapour a potent greenhouse gas.
Also there is substantial thermal inertia in the ground , so even when the sky clears don’t expect its temp to plummet in 10 minutes and start to draw conclusions about downward radiation.
Trick,
I sort of doubt Willis is proposing that it’s impossible for global surface temperatures to change at all, merely that there are mechanisms which regulate temps and keep them within certain boundaries.