Update (March 24th) at the end of this post – a kind of response from Feldman.
There was much ado recently about a new paper published in ‘Nature’ (“Observational determination of surface radiative forcing by CO2 from 2000 to 2010″ by Feldman et al.) claiming to have observed a strengthening in CO2-specific “surface radiative forcing” at two sites in North America going from 2000 to the end of 2010 (a period of 11 years) of about 0.2 W/m2 per decade, and through this observation further claiming how they have shown empirically (allegedly for the first time outside the laboratory) how the rise in atmospheric CO2 concentration directly and positively affects the surface energy balance, by adding more and more energy to it as “back radiation” (“downwelling longwave (infrared) radiation” (DWLWIR)), thus – by implication – leading to surface warming.
In other words, Feldman et al. claim to have obtained direct empirical evidence – from the field – of a strengthening of the “greenhouse effect”, a result, it would seem, lending considerable support to the hypothesis that our industrial emissions of CO2 and other similar gaseous substances to the atmosphere has enhanced, and is indeed enhancing still, the Earth’s atmospheric rGHE, thus causing a warming global surface – the AGW proposition.
From the abstract:
“(…) we present observationally based evidence of clear-sky CO2 surface radiative forcing that is directly attributable to the increase, between 2000 and 2010, of 22 parts per million atmospheric CO2.”
“These results confirm theoretical predictions of the atmospheric greenhouse effect due to anthropogenic emissions, and provide empirical evidence of how rising CO2 levels (…) are affecting the surface energy balance.”
So the question is: Do these results really “confirm theoretical predictions of the atmospheric greenhouse effect due to anthropogenic emissions”?
Of course they don’t. As usual, the warmists refuse to look at the whole picture, insisting rather on staying inside the tightly confined space of their own little bubble model world.
Let us first see what they mean by “surface radiative forcing”. From the paper:
“Surface forcing represents a complementary, underutilized resource with which to quantify the effects of rising CO2 concentrations on downwelling longwave radiation. This quantity is distinct from stratosphere-adjusted radiative forcing at the tropopause, but both are fundamental measures of energy imbalance caused by well-mixed greenhouse gases. The former is less than, but proportional to, the latter owing to tropospheric adjustments of sensible and latent heat, and is a useful metric for localized aspects of climate response.”
There are definite problems with some of the things stated in this paragraph, especially how they assume a direct proportional connection between the energy balance at the ToA and at the surface.
But we’ll put this aside for now. Let’s rather see where the above definition actually ends up taking us. “Surface radiative forcing”, according to Feldman et al., is all about quantifying “the effects of rising CO2 concentrations on downwelling longwave radiation” (DWLWIR).
The two sites used in their study are 1) at the North Slope of Alaska (NSA), close to Barrow, and 2) on the Southern Great Plains in Oklahoma (SGP), in Lamont, about 30 km NE of Enid.
Here are their results:
Figure 1. This is a cropped version of Figure 4 from the paper, stating: “Time-series of surface forcing. a, Time series of observed spectrally integrated (520-1,800 cm-1) CO2 surface radiative forcing at SGP (in red) with overlaid CT2011 estimate of CO2 concentration from the surface to an altitude of 2 km (grey), and a least-squares trend of the forcing and its uncertainty (blue). (…) c, As for a but for the NSA site.”
Looks like a strong case. Especially the NSA fit is impressive (the SGP forcing appears to be sagging a bit after 2007, doesn’t it?).
This blog post, however, is here to provide some much needed context to these findings. Context that Feldman et al. really should have explored themselves, but of couse wouldn’t. Because the context basically renders their study irrelevant (as simply an artefact of an extreme “cherry-picking” exercise) and their grand conclusion at best somewhat rash, at worst a prime example of pseudoscience.
These are the (visually suggestive) plots that the ‘Climate Establishment’ prefers for people to see, in order for them to silently draw their own inevitable conclusions about what really causes ‘global warming’ and what doesn’t:
Figure 2. Global temperatures between 2000 and spring 2014 according to GISS (central red curve) versus atmospheric CO2 concentration (upper blue curve) and sunspots as a proxy for solar activity (lower light cyan curve).
Got to be CO2, right? Most likely somewhat augmented by an accompanying positive feedback from H2O (not depicted here). If anything, natural factors like the Sun (as one can well observe) and maybe volcanic aerosols plus internal decadal variability like ENSO (not represented), has pulled the opposite way, making the rise in global temperatures evidently less than the increased “radiative forcing” from the clearly enhanced “rGHE” as a consequence of our evil industrial emissions alone would have it.
This is their carefully chosen narrative. This is the story they wish to convey to the world. This is the visual impression they want to leave in people’s minds. Repeat, repeat and repeat yet again. So that we don’t forget. Never forget. Never question. This is what they want us to believe, firmly, with all our hearts.
And Feldman et al. seem only too eager to bolster this narrative. Their study might just be seen as a work aspiring to become some kind of capstone to this meticulously assembled AGW structure, one final ‘piece of the puzzle’ to complete the storyline.
Too bad, then, that it’s ultimately void of real, objective science …
The basic premise behind the idea of the “enhanced greenhouse effect” as a cause of ‘global warming’ is that the more IR-active (IR absorbing and emitting) substances are released into the atmosphere, the more intense the DWLWIR (the “atmospheric back radiation”) to the surface will be. This effect is meant to be directly proportional, only inversely so, to the corresponding parallel effect of a less intense radiative flux from Earth to space. They both represent increased “radiative forcing”, only the former at the surface, the latter at the ToA. The increased “back radiation” will, by adding energy to the surface (!!), reduce its ‘radiative heat loss’, meaning, its ability to cool by radiation. Which will lead to piling up of energy and thus … warming.
This is what should happen as we keep spewing our evil industrial emissions out into the atmosphere … according to theory.
But what about reality?
Reality very much appears to tell a different story:
- From March 2000 to May 2014, average atmospheric DWLWIR (LW down, “back radiation”) to the global surface of the Earth decreased by about 0.63 W/m2 (-0.44 W/m2 per decade) (Figure 3).
- During that same time, average insolation (SW down) to the global surface of the Earth increased by 0.51 W/m2 (+0.36 W/m2 per decade) (Figure 4).
This is actually the exact opposite development from what those curves presented in Figure 2 would seem to suggest, that atmospheric “radiative forcing” is increasing while solar “radiative forcing” is decreasing.
It goes without saying, then, that the ‘Climate Establishment’ never presents the plots in Figures 3 and 4 to the general public. Only the ones in Figure 2. First rule of biased (pseudo)science: Presenting ‘observed reality’ over ‘hypothetical reality’ is never a good idea when your goal is to preserve and promote a hypothetical (imaginary) storyline …
But atmospheric CO2 is going up globally, right? So should contribute to an overall rise in the global atmospheric DWLWIR and hence its global “surface radiative forcing”. Right?
True. If the world were in fact that simple …
To see how important (or unimportant) – in the grand scheme of things – this ‘contribution’ from CO2 to the total atmospheric DWLWIR really is, let’s start by having a look at the evolution in total DWLWIR to the surface from 2000 to 2010 at or around the two sites used in Feldman’s study.
For this we will use data from the CERES EBAF-Surface Ed2.8 product (same as in Figures 3 and 4 above). Be advised that this differs substantially from the ToA product in that its radiation data to a much larger extent is simply derived mathematically from observed proxy variables obtained from various other sources and used as input to a standard radiative transfer model. These variables or parameters significantly include atmospheric temperature, humidity and cloud profiles. (This is how NASA decribe the estimation/computation process of CERES surface data.)
The interesting (and quite convenient) thing here, though, is that the CERES team specifically uses ARM facilities and their surface-based instruments, such as the ones at the North Slope and in Oklahoma, to validate its output products, notably through efforts such as CAVE (‘CERES/ARM Validation Experiment’).
There is therefore no immediate reason to doubt the consistency between the findings of DWLWIR from CO2 in Feldman et al. 2015 and the CERES estimates of total DWLWIR at the same sites.
Here’s Feldman’s claimed increase in “surface radiative forcing” from a rise in atmospheric CO2 plotted against the total DWLWIR at the two sites in question:
Figure 5. Southern Great Plains, Oklahoma.
Figure 6. North Slope, Alaska.
Note how what appears as white, near horizontal lines in the two graphics above actually represent the CO2 ‘contribution’ to the total “radiative forcing”, rising by about 0.2 W/m2 from end to end in each case. Comparing this value to the scale along the y-axis to the right puts it in its proper context. The mean seasonal range in total DWLWIR is ~30 W/m2, two orders of magnitude larger than the total rise in CO2 “forcing” from 2000 to 2010. What’s more, any apparent overall correlation between the trends in total DWLWIR and the CO2 part of it seems to be coincidental at best, and certainly not indicative of any +CO2 ⇨ +DWLWIRtot ’cause and effect’ relationship.
In summary, the whole hypothesis of the “enhanced greenhouse effect” seems very much to be resting on the premise of ‘All Else Being Equal’: “If air temperatures and the water cycle (including convective power) didn’t in any way change in response to other physical processes, and if the solar input remained constant, then (and only then) a simple increase in atmospheric CO2 content would potentially be able to force a net energy imbalance on the surface, making energy pile up and thus induce warming.”
In other words, ‘Climate ScienceTM’ assumes that all changes in all other climatic processes on Earth come only as mere ‘feedbacks’ to an initial change in atmospheric CO2 concentration, meaning, they are simply there to either modify or amplify the original effect of the induced “radiative forcing” from rising CO2 – if the CO2 “forcing” increases, then the rest of the climate system can only respond to this increase to make the ultimate increase lesser (neg. feedback) or greater (pos. feedback) in extent. Never override it. Never do ‘other stuff’ independent of the “radiative forcing”. The final result of the originally imposed energy imbalance can thus – in this model world – never end up being zero and definitely not negative. The corollary of this of course being that an increase in atmospheric CO2 must induce surface warming of some magnitude, no matter what happens elsewhere. Because what happens elsewhere is always mere ‘feedbacks’ to what happens here.
Looking at Figures 5 and 6 above makes you quickly realise what an absurd approach to how the real world works this really is.
The three primary variables of 1) air temperature, 2) air water vapour content, and 3) air column cloud distribution, couldn’t care less what CO2 is up to. They all operate on a completely different scale, vary on a completely different level and they thus all fundamentally respond to a completely unrelated set of physical mechanisms. Figure 3 tells us how total atmospheric DWLWIR to the global surface has gone significantly down since 2000, despite the fact that the total atmospheric CO2 content simultaneously went up globally by more than 7.5%.
There is simply no connection …
CO2 is but the fly riding on the back of an elephant, in this case a three-headed one. The fly can always try, but it can never tell the elephant where to go.
Why did global DWLWIR from the atmosphere to the surface decline over the last 14 years (Figure 3), while the global UWLWIR back out from the surface at the same time saw a significant increase (Figure 7)?
What kind of twisted logic would even try to still explain the former as somehow the cause of the latter in such a situation? The ‘logic’ of the ‘Climate Establishment’ surely would … If it only cared to even have this issue addressed, that is (it’s a total non-issue to the warmist high-priests, firmly swept under the rug, kept safely under lock and key from the public eye).
Both the rise in the blackbody emission (UWLWIR) of the global surface and the parallel decline in global “sky radiance” (DWLWIR) contribute to the quite massive increase in the net surface LW flux since 2000*, the ‘net LW flux’ effectively being the ‘radiative heat loss’ of the global surface. IR-absorbing substances accumulating in the atmosphere (for they clearly are) in other words do a surprisingly poor job at reducing the surface’s ability to cool by radiation, seeing how this ability has rather been robustly strengthened over the last 14 years.
*The increase [growing more negative, it’s heat loss, after all] is -0.25 W/m2 (-0.17 W/m2 per decade) for clear-sky conditions (clouds excluded), -1.52 W/m2 (-1.07 W/m2 per decade) for all-sky conditions (clouds included).
So how will people ever get to see all this and understand its implications? As long as no one’s telling them about these basic, observable facts of the natural world at present?
Most of what one calls atmospheric DWLWIR to the surface is a ‘consequence’ simply of the temperature of the air in the lower part of the troposphere, which in turn is simply a ‘consequence’ of the surface temperature beneath:
To this base block of “sky radiation” comes – almost exclusively over land – variation from differences in water vapour not directly related to surface temperatures; over the ocean, tropospheric water vapour content is extremely closely correlated to the sea surface temperature (with the latter clearly the cause, the former the effect), over land less so, the Sahara-Sahel a good case in point.
Finally, on top of this, comes the cloud radiation, what is normally referred to as LW CRE (‘longwave cloud radiative forcing’). Globally, this contributes no more than ~29 W/m2 to the total atmospheric DWLWIR ‘flux’ (according to CERES), a mere 8.4% of the full 345 W/m2.
But what will become clear is that it’s still the clouds that make all the difference, being by nature much more responsive and prone to variability (through both extent, type and vertical distribution), than both global temps and the accompanying WV content.
When it comes to change, clouds are the main player without a doubt.
Figure 9. Global clear-sky DWLWIR, from air temps and WV; anomalies.
Figure 10. Cloud LW forcing, the global cloud contribution to the total atmospheric DWLWIR; anomalies.
While the air temp+WV “forcing” has stayed remarkably stable (Figure 9), cloud “forcing” makes a major downward swoop over the period (Figure 10) and in effect proves to be responsible for the entire drop in total DWLWIR globally since 2000, comfortably trumping every other effect and, as it appears, working quite independently from all other contributing factors (including CO2).
How does this intriguing change in LW CRE spread spatially across the globe?
Figure 11. Cloud LW forcing, a zonal breakdown of the global signal in Figure 10; top: NH (90-20N), middle: tropics (20N-20S), bottom: SH (20-90S).
I will leave these plots for the reader to ponder …
Higher up I stated the following: “[‘All Else Being Equal’,] a simple increase in atmospheric CO2 content would potentially be able to force a net energy imbalance on the surface, making energy pile up and thus induce warming.”
How come such an imposed energy imbalance from an increase in atmospheric CO2 is only potential? Why isn’t it a given?
Well, empirical tests have shown quite conclusively that even pure (100%) CO2 as opposed to regular dry air (containing just 0.035-0.04% CO2) is not capable of slowing – not even in the slightest – the total heat transfer through the medium by simple absorption of IR within certain spectral bands. Much has to do with the simultaneous IR emission by the gas. Furthermore, heat transfer from a heated surface through a fluid to the ultimate heat sink at the top of it is governed almost exclusively by convection. Whatever stimulates the movement of bulk air between the heating end down low and the cooling end up high, like on Earth, will help boost the cooling ability of the heated surface at the bottom. IR-absorbing gases in the air directly contribute to the warming of the air above the surface by radiative heat transfer. When this air warms, it automatically and instantly expands and floats up, adding to convective uplift and thus the efficient removal of the energy transferred from the surface to the air, making room for new energy to arrive.
Water is a special case in this regard, since it possesses a comparatively high ‘heat capacity’, so will maintain a fairly large thermal inertia (it needs to absorb more IR to warm as much as CO2 for instance, which has a ‘heat capacity’ even slightly lower than N2 and O2). But H2O is still a fierce facilitator of convective uplift and thus of surface cooling (regardless of its radiative properties).
Below are two plots by Willis Eschenbach from observations in the tropical Pacific showing how the relative humidity of the air above the ocean surface is almost perfectly inversely correlated to the temperature of that same air. Knowing how the temperature of the air is a direct, though lagged, response to the temperature of the ocean surface below it, and knowing how the temperature of the ocean surface strongly affects its evaporation rate plus how the temperature of the air strongly affects its capacity to hold evaporated water, this would, on the face of it, seem to be a strange finding indeed. After all, higher temps should give a higher specific humidity (total air WV content) and thus a fairly stable relative humidity (air WV content as related to the air temperature – more WV, but also higher temp, stable rel. humidity).
What these observations show in an elegant manner, however, is that something negates the piling up of WV (and thereby surface energy) in the air (the increase in specific humidity) as temps go up (it likewise negates the depletion of WV in the air (the reduction in specific humidity) as temps go down). Something simply brings the water evaporated up and away from the surface. Convective uplift. The higher the temperature, the more powerful the evaporation rate, the more powerful the resulting convective uplift:
Figure 12. (From Willis Eschenbach, using TAO data.)
To sum up:
The world is not a simple, linear place. It is a complex and highly interconnected place. Focusing solely on radiative heat transfers within the Earth system and then try to draw conclusions about what overall (net) effects these will have on temps is quite simply proof you’ve lost your touch with reality.
If the Earth’s surface warmed from 2000 to 2014, it sure wasn’t the result of an “enhanced greenhouse effect”. Because the “greenhouse effect” (rGHE) as defined by the warmists themselves has significantly weakened over this particular period. The atmosphere as a whole now imposes a considerably smaller “radiative forcing” on the surface than it did 15 years ago.
And isn’t that jolly good to know …?
Update (March 24th): Commenter Rafael Molina Navas contacted Daniel R. Feldman of the ‘Lawrence Berkeley National Laboratory’ and the lead author of the study under discussion here, about this blogpost, and was kind enough to share with us his reply. It is not a terribly informative answer, the way I see it, but I’ll post it here anyway for you to read:
“Thanks for your interest and I appreciate you sending these to me.
Kristian’s blog is very detailed and he has thought about these issues for a while. That being said, our study was focused on the effect from CO2 alone, which we found pushes the system towards a warmer state, by radiative forcing, and that blog post appears to conflate what we found with a large number of other effects. It is important to note that while there are certainly other feedbacks in the climate system, the forcing from CO2 is largely independent and separable from these. The feedbacks on CO2 forcing tend to enhance the effect of rising CO2, as cited in the paper, so I think Kristian has a sign issue there.
It would take a significant amount of time to formulate a detailed response to every claim made in the post (which unfortunately I don’t have time for), but suffice it to say, there are numerous issues there.
For example, I should note that the ARM observations are superior to the CERES products for surface forcing, because we’re measuring at the surface and CERES is a satellite instrument. I applaud the CERES team for putting together a surface flux product, but the surface stations have direct measurements. Also, we are looking essentially continuously at two sites rather than how CERES observes these sites occasionally and at the same time of day (CERES is sun-synchronous). Also, we have independent, in situ measurements from weather balloons, and are able to identify clear-sky scenes with very sensitive radar and lidar measurements, which are not available for the CERES products.
Hope that helps a little. Right now, I have numerous projects with pressing deadlines, but I would be happy to respond to peer-reviewed criticism.”