This post contains three addenda to the next post; additional/further explorations that I feel have more of a tangential than a fundamental bearing on the main argument laid out there, still, I would say, providing some definite extra depth, scope and context to it. The figure numbering here will simply carry over from the main post (ending with number 31.), and all figures referred to in the text or captions below (but not in direct quotes) numbered somewhere between 1. and 31. will be from that post, unless otherwise noted.
The three addenda are:
I – A net flux composite
II – What do the models say?
III – ASR and cloud albedo
A net flux composite
A little bit about tracking Earth’s ToA net flux (‘heat balance’) anomaly from 1985 until today. We have the same two associated but separate radiation flux datasets on hand (ERBS Ed3_Rev1 and CERES EBAF Ed4) as we did when we established our two OLR composite records in the main post. However, this time around, calibrating the mean levels of the two distinct sets of flux anomaly data across the 1999-2000 gap that separates them, and to find any kind of real confidence in the result, is way harder. Not qualitatively, though; 1) we’re quite certain that they – the mean anomaly levels – should be generally positive on both sides of the splice, and 2) we are pretty confident that the mean anomaly during the first four years of the record, 1985-88, was fairly close to zero, the net ToA flux hovering around neutral. The problem is distinctly a quantitative one; in determining the magnitude of the mean anomaly adjustment required to calibrate the latter part of the record (the CERES data) to the former (the ERBS data).
What distinguishes a ‘Net flux composite‘ from an ‘OLR composite‘ is in the way we “anchor” it.
- The OLR, Earth’s heat loss to space, will over time, as we firmly established in the main post, simply be a direct function of tropospheric temps, more or less a mere radiative expression (effect) of them.
- The Net flux, however, ASR minus OLR, Earth’s heat balance, will best translate into an Earth system heating (or cooling) rate, i.e. the total net storage of internal energy over some specified time frame, like a month or a year. We can use OHC data to calculate such a rate, and in turn determine (“anchor”) the net flux anomaly from this.
The main problem with the OHC→net flux approach is what depth layer to go by. Do we use the 0-700m layer as our guide? Or the 0-2000m layer? Or some other one? The question you need to ask yourself is this: How deep into the ocean can we really expect a ToA net flux signal to penetrate?
Bearing all this in mind, I hereby present a possible, but by no means necessarily accurate in its cross-splice calibration, near-global ToA net flux composite record (ERBS+CERES) from 1985 to 2017, sort of based, though kind of loosely, on NOAA/NCEI’s global OHC data (0-700m) in Fig.1:
Figure 32. It seems likely that the net flux underwent a multiyear decline between the high level around 1995-98 and a closer-to-neutral stretch around 2000-2003/2007 (however you prefer to read the graph above), with the bottom ostensibly reached some time around 2001-2002, before it slowly, but steadily started growing back (from 2002-2003 already, but especially from 2008, and even more consistently so from 2011 onward).
We know from the main post, looking at the ERBS and the CERES data separately, that an increase in ASR is the clear reason behind both confirmed periods of rising positive net ToA imbalance over the last 32-33 years; first the one from 1988-89 to the mid-to-late 90s, and then again the one from 2002 to the present high level (2012-2017), and that a parallel increase in OLR was very much what kept it from running away in both cases.
My estimate represented in Fig.32 is, however, quite a conservative one when compared to the (in “Mainstream Climate Science (MCS)” circles) generally agreed upon assessment of the post-2000 mean anomaly level of the positive ToA imbalance. A relevant case in point, here’s what the CERES team themselves have determined for their newest version (Ed4):
“A one-time adjustment to shortwave (SW) and longwave (LW) TOA fluxes is made to ensure that global mean net TOA flux for July 2005–June 2015 is consistent with the in situ value of 0.71 W/m2.”
That 2005-15 mean level is about 0.27 W/m2 higher than my estimate in Fig.32.
Another example, Allan, 2017, appears to be placing the mean value of this particular interval at more or less the same level as the CERES team. His Fig.1d:
Its caption reads:
“(…) top of atmosphere net radiation for a combination of satellite and surface observationally-based estimates, atmosphere-only climate models using prescribed observed sea surface temperature and sea ice (AMIP) and a the ERAI reanalysis over the period 1979-2016 (3 month smoothing is applied).”
It appears the difference simply lies in how much water column one considers reasonable to include in determining an existing ToA net flux. In the end, it all must come down to personal preference, because this question isn’t one that can be fully resolved simply through objective deliberation. I would say that including water masses below 700 – even 300-400 – meters of depth is meaningless in this regard; there is no immediate, first-order connection between a change in the radiative balance at the top of the atmosphere and what happens with the temperature a kilometer or two below the surface of the ocean. If anything, if warming of the water masses in the upper parts of the ocean (0-700 (300-400) m), where the connection and energy exchange with the atmosphere and the Sun is much more direct, were to gradually propagate (diffuse) downwards, which in itself is a perfectly plausible scenario, this would surely be a slow and creeping process involving considerable lags, naturally making any observed warming signal at depth, say, 700-2000m below the surface, utterly out of sync with the presently operative net flux at the ToA.
Others would probably argue against this …
Either way, “MCS” seems very much in favour of using the deeper layer (0-2000m) when setting the post-millennium (CERES) ToA net flux mean level anomaly. An average of +0.71 W/m2 for the period 07/2005 – 06/2015 seems to square quite well with the general rise in NOAA/NCEI’s official OHC curve for the 0-2000m depth layer since 2005. My estimate, as it turns out, conforms much better with the progression of the 0-700m curve. Go figure!
The interesting thing here, though, is that, even if they’re right and I’m wrong, the explanation behind the positive net flux value remains the same: A sustained high level of solar heat input (ASR) to the Earth system from 1988-89 onwards. Stating a larger positive net flux value after 2000 simply necessitates a larger solar heat surplus, a higher mean ASR level, after 2000 than what I’m suggesting, since we after all can be pretty certain that the OLR didn’t just increase with the tropospheric temps from 1985 to 1999 (ERBS) and from 2000 to 2017 (CERES), but also from the 90s to the 00s, from the ERBS era to the CERES era.
What do we see?
We see the gap (ASR over OLR) opening up in 1989, we see it widening during the 1993-97 period, before it drops and starts closing up again, a combined result of a weakening of the incoming solar heat flux (–ASR) and a simultaneous strengthening of Earth’s own outgoing heat flux (+OLR), the latter effect, as we now know and understand, a direct radiative (Planck) response to tropospheric warming, the tropospheric warming in turn a system response to the, after all, quite strongly positive net ToA imbalance that persisted all the way through the 90s. We finally see how the gap reopens, slowly, but steadily, from 2002-2003 on, and how since 2012 it has once more reached a rather substantial magnitude, yet again all thanks solely to the increase in ASR, and only countered somewhat by the concurrent, but much smaller, temperature-driven rise in OLR.
What do the models say?
It almost goes without saying that ‘Mainstream Climate Science (MCS)’ has its very own stated version of how ‘global warming’ came about. The interesting thing to notice in this regard is how this particular version was established, and how one seeks to justify, not just its validity, but in fact its professed preeminence over all other ‘versions’ out there. Their approach is an unusual one. In their efforts to explain what one might view as a concrete physical phenomenon, global warming, they have decided to rely exclusively on the output of model simulations and to fully disregard all real-world observations directly pertaining to it. Their claims about the world on this particular issue thus come purely as the result of conjecture. Based on certain theoretical considerations and nothing else.
They want us to turn a blind eye to the readily available, empirically obtained data by simply brushing it off as somehow “not up to standard”. They normally go to great lengths to try and make this point. To simply reject the ‘usefulness’ of the data. From Trenberth and Fasullo, 2009 (link below):
“(…) for observations of the Earth’s radiation, changes in instrumentation [and calibration] do not allow changes to be detected between the Earth Radiation Budget Experiment (ERBE) and more recent Clouds and the Earth’s Radiant Energy System (CERES) observations [Fasullo and Trenberth, 2008a]. A continuous record from Earth Radiation Budget Satellite (ERBS) is compromised by a discontinuity (late 1993) [Trenberth et al., 2007].”
And Donohoe et al., 2014 (link below):
“Reduced OLR (…) seems the likely cause of the observed global energy accumulation, although the limited length of satellite TOA radiation measurements precludes determination of the relative contributions of ASR and OLR by direct observation.”
These are quite standard, run-of-the-mill examples of attempting, with easy cop-outs, to avoid the whole problem of ‘actual-observational-data-is-here-to-test-whether-the-models-are-in-fact-onto-something-or-not’. “Oh, sorry. Can’t use it. Too uncertain. There. Done. Back to our models, our only reliable source of understanding.” The irony …
Dr. Leon Festinger had this to say about this almost pathological unwillingness to acknowledge data and observations that go against one’s own ideas:
“A man with a conviction is a hard man to change. Tell him you disagree and he turns away. Show him facts or figures and he questions your sources. Appeal to logic and he fails to see your point.
- We have all experienced the futility of trying to change a strong conviction, especially if the convinced person has some investment in his belief. We are familiar with the variety of ingenious defenses with which people protect their convictions, managing to keep them unscathed through the most devastating attacks.
But man’s resourcefulness goes beyond simply protecting a belief. Suppose an individual believes something with his whole heart; suppose further that he has a commitment to this belief and he has taken irrevocable actions because of it; finally, suppose that he is presented with evidence, unequivocal and undeniable evidence, that his belief is wrong: what will happen? The individual will frequently emerge, not only unshaken, but even more convinced of the truth of his beliefs than ever before. Indeed, he may even show a new fervor for convincing and converting other people to his view.” 
So what, then, are the models saying?
The models, of course, operating within a realm fully detached from the real – observationally constrained – one, a realm where theory is rather always circularly allowed to verify itself, are essentially free to make up their own story, their own truth, as they go along. Completely unaffected by the course of events actually observed to play out in the real Earth system. And this story – the model story – is then also, quite naturally, the official story of ‘MCS’ on the ‘global warming’ issue.
And so, as you can gather, the models are the ultimate source and alibi, the only source and alibi, of the ‘SW(+ASR)-warming-is-just-a-positive-feedback-to-LW(–OLR)-warming’ argument.
This argument was notably promulgated in two rather salient papers, already referred to above, one specifically referencing the other, and both now leaned heavily on by today’s MCS when it comes to this particular issue. These papers are:
To outsiders, the whole argument feels a bit cheap, more like a ‘factoid’ or a ‘talking point’, like a seemingly plausible ad hoc explanation concocted on the spot to salvage an original hypothesis from what at first glance would look like pretty detrimental real-world observations; akin to the whole “every and any weather (and climate) pattern is consistent with the idea of anthropogenic global warming” meme.
The claim is made, however, that the ‘LW warming → SW feedback’ causal relationship thing is simply an emergent feature of the climate models, and so it would appear that the argument in question indeed and quite naturally, albeit surprisingly to many, perhaps, arises directly from the underlying theoretical framework of the “AGW” idea itself. It was there all along, we just weren’t quite aware of it …
The two papers do, however, disagree somewhat on what the SW feedback is supposed to be about …
Trenberth & Fasullo are certain it’s ultimately about cloud feedback:
“As greenhouse gases and associated radiative forcing increase [from 1950], the models amplify the effect through increases in water vapor. Hence both clear sky and all-sky integrated OLR decrease initially (Figure 1 [my Fig.36]) and there is a net heating of the planet. Snow and ice also begin to decrease and reduce albedo at high latitudes (Figure 2), thereby providing further amplification through ASR increases – the ice-albedo feedback. (…) From 1950 to 2000, the net result globally is a decrease in OLR and a net heating. Only after about 2050 does the increase in temperature become large enough to overwhelm the increasing greenhouse effect (Figure 1) (…).
(…) In most models, the late 21st century planetary imbalance is not dominated by the ice-albedo effect, but rather stems from changes in clouds (Figure 3) and aerosols. From 1950 to 2000, increases in sulfate aerosols decrease the ASR by increasing reflected solar radiation (RSR), and this is slightly offset by a modest decrease in clouds. (…) integrated all-sky ASR anomalies become positive by 2040 owing mainly to decreasing cloud amount and this continues throughout the 21st century (Figure 3).”
While Donohoe et al. for some reason seem to feel compelled to dismiss this notion altogether:
“Although the differences in λSW [the (generally positive) shortwave radiative feedback] across the CMIP5 models are primarily caused by differences in SW cloud feedbacks, the ensemble average value λSW = 0.6 W m−2 can be attributed to two robust and well-understood consequences of a warmer world: (i) the enhanced SW absorptivity of a moistened atmosphere and (ii) the enhanced SW reflection associated with less extensive snow and sea ice cover. (…) the positive λSW of the CMIP5 ensemble average and the resulting energy accumulation by enhanced ASR under GHG forcing can be expected based only on the robust physics of the water vapor feedback and the surface albedo feedback in the absence of any changes in clouds.”
What the two do agree on, though, is 1) that warming in the models always starts out as a result of reduced OLR (through an “anthropogenically enhanced GHE”), and 2) that this LW warming does not cause a positive ASR (SW) feedback to amplify the original warming as a consequence of reduced cloud albedo any time during the first 50-60 years after the initial perturbation – there is simply not enough warming up to that point. It only comes considerably later (Trenberth & Fasullo), or not at all (Donohoe et al.) …
Either way, Donohoe’s Fig.1 is instructive:
Figure 35. Fig.1 from Donohoe et al., 2014. Caption reads:
“(A) Idealized response of global mean radiation at the top of atmosphere to an instantaneous greenhouse gas forcing (green dots) assuming no shortwave feedback and a radiative adjustment e-folding time of 20 years. The green line shows the OLR response (anomaly from pre-industrial), and the shaded green area shows the LW energy accumulation. (B) As in (A) but in response to an instantaneous SW forcing (red dots), with the red line showing the ASR response. In this case, the net energy accumulation is the difference between the SW energy accumulation (the shaded red area) and the LW increase (the hatched green area where the hatching indicates that the LW response leads to a cooling of the climate system). (C) The ensemble average radiative response in the CMIP5 4×CO2 simulations. The shaded area represents the energy accumulation by SW (red) and LW (green) anomalies and the hatched area indicates energy loss by enhanced OLR. The dashed red and green lines show the predicted ensemble average ASR and OLR response from the linear feedback model (…). (D) As in (C) but for the CMIP5 ensemble average radiative response in the 1% CO2 increase per year simulations (with linear increase in forcing, as shown by dotted lines).”
And so is Trenberth & Fasullo’s Fig.1, even though it presents the argument (essentially equal to Donohoe’s 1D scenario) in a bit more elaborate manner:
Figure 36. Fig.1 from Trenberth and Fasullo, 2009. Caption reads:
“Integrated perturbations in (a) total net radiation RT, (b) –OLR, and (c) ASR relative to 1900 to 1950 in 1024 Joules (YottaJoules, YJ) (RT = ASR–OLR) for the clear sky (dashed) and all-sky (solid) for the A1B scenario. The range of plus and minus one standard deviation is shaded grey for all-sky. Values are given for the 1990s and 2090s for all-sky, clear sky and their difference, the CRF.”
Anyway, let’s at this point have a look at how the models actually do simulate the world;
at “The Model Story of Global Warming.”
(Data can be found here.) First, the Net flux:
Figure 37. Notice the y-axis. It spans a range of 5 W/m2. The positive net imbalance starts opening around 1970, after having hovered comfortably around the zero line during most of the previous 100+ years, volcanoes the only (short-term) disruptions to what appears to have been a relatively fine-tuned balance. Note, however, that even though there’s a slight upward movement in the net curve to be discerned already from 1970 onwards, it only makes its first substantial leap up across the 1976-77 transition. Coincidence?
Figure 38. Yup, we can very much appreciate from this chart why MCS is simply forced to a) reject the validity of the calibration across the 1993 gap in the ERBS dataset, even though they both do (Trenberth et al. above) and don’t (e.g. Allan, 2017; Loeb et al., 2012), and b) deny the correctness of the OHC data over the 1977-88 period, even though they both implicitly do (here) and explicitly don’t (at all other times), as the former shows a distinct ±1 W/m2 rise in the near-global net flux between the 1985-88 period and the 1993-97 period (Fig.6), and the latter clearly showing no positive overall change whatsoever in global OHC, indicating a ToA net flux anomaly level of essentially zero, from 1977 to 1988 (Fig.2).
The models, according to Fig.38, claim a steady rise in the positive net flux value all the way from ~1970 and up to ~2000, when for some reason, it apparently takes a 15-20 year break, at an elevated level approximately 0.6 W/m2 above neutral.
Knowing now how the models picture a gradual opening up, from 1970 to present, of a positive net heat imbalance for the Earth system, what is most insteresting for us to figure out is how they view the relative contributions to this ‘net heating’ process of the two (concrete/measurable) opposing heat fluxes that counterbalance to produce the (after all, abstract/merely calculated) net flux between them. That’s the ASR, the incoming heat flux (Sun→Earth), and the OLR, the outgoing heat flux (Earth→space), at the ToA.
Figure 39. Notice again the y-axis. It now covers a range of 8 W/m2, not just 5. The green curves in Fig.37 and 38 are simply expressing the arithmetic difference between these two curves: ASR (gold) minus OLR (red).
How do we interpret this graph? What does it tell us? Let me highlight the sections most worth noting:
Here’s how MCS apparently reads this graph (recalling that Donohoe et al. quote from the main post):
“(…) given a present GHG forcing of about 2.8 W/m2, the increase in global surface temperature of about 0.85 K above preindustrial temperatures, and the observational estimate of λLW, Eq. 2 [−OLR = FLW + λLW TS] suggests an anomalous OLR of ≈ −0.8 W/m2, implying that OLR is still contributing to global energy accumulation.”
“This apparent discrepancy can be attributed to the effects of tropospheric aerosols, which are acting to reduce global warming (and thus, OLR) through a negative SW radiative forcing on the order of 1 W/m2 (although with large uncertainty). Eq. 2 [ASR = FSW + λSW TS] and our observational estimate of λSW then suggest an anomalous ASR of ≈ −0.2 W/m2 in the current climate. Altogether, these estimates imply that the current global energy accumulation is still dominated by decreased OLR.”
And, lo and behold! What do the models say? The 2000-2012 mean ASR level appears to be some 0.2 W/m2 below the (granted, slightly uneven) 1860-1960 mean level, while the 2000-2012 mean OLR level seems to be about 0.8 W/m2 below that same base level. So there you have it: The ASR contribution to the current ToA imbalance is (still) somewhat negative, while the OLR contribution is strongly positive (+0.8 W/m2).
That’s how the world works. That’s how we’ve gotten our ‘global warming’. Right?
Well. That … depends. What do we actually see in Fig.40?
A few peculiar items:
We see a sudden drop of about 0.7 W/m2 in the mean level of both OLR and ASR between 1960 and 1965. The funny thing about this is that, since they are directly inversely related, and since they drop by the exact same amount, at the exact same time, this abrupt and significant change in both contributing heat fluxes doesn’t show up as a change in the net heat at all (Fig.38). Even so, one is inclined to interpret this synchronous drop as somehow the result of a specific assumption on the part of the models (and/or of their architects, the modellers) that at this particular point in time there was 1) a massive pulse of (anthropogenic) CO2 injected into the atmosphere, strongly and suddenly affecting Earth’s heat loss to space (–OLR), and, simultaneously, 2) a massive pulse of (anthropogenic) sulfate aerosols dispersed through the atmosphere, strongly and suddenly affecting Earth’s heat gain from the Sun (–ASR).
We see, starting around 1970, that the ASR gradually rises back up, but that it still as recently as during the 2000-2012 period had yet to fully reach its former, pre-1960 (pre-perturbation?) level; that has only happened as of the last few years. What caused this gradual re-intensification? A ‘brightening’ of the Earth resulting from the progressive dissipation of the atmospheric aerosols that likely caused it to dim down in the first place? Or is what we see somehow simply the canonical consequence of positive SW feedbacks to global warming kicking in, like the reduction in ice albedo that both Trenberth and Donohoe discuss above? I guess we’ll have to assume a mixture of the two …
We see that, while the ASR flux naturally regrows in strength after what one might assume is the initial perturbation, the OLR flux doesn’t. It remains more or less at the same sunken level all the way from 1963 till today, including the three deep temperature-induced depressions of the Agung (1963), El Chichón (1982), and Pinatubo (1991) volcanic eruptions. It stays flat even as we know from observations that the world grew considerably warmer over the period. What we see, then, is simply the hypothetical “Greenhouse Warming Mechanism” in operation. Inside the model framework. It is supposed to work (create warming, that is) by holding back Earth’s heat loss to space as the atmosphere grows ever more opaque to outgoing surface radiation, which means that, under such circumstances, keeping the heat loss at the same level over time requires an ever higher temperature.
The main problem when it comes to the ASR/OLR plots in Figs.39 and 40 is how to resolve the cause-and-effect relationship.
Which of the two fluxes is actually causing the positive ToA imbalance to open? Which of the two fluxes is actually causing the Earth system to warm?
Let’s start with the initial 17 years post 1960, the light-gray shaded section of the diagram in Fig.40. What do we know about this period? We know that the Earth … cooled. (In fact, this is the latter part of the period of general global cooling that started all the way back in the 1940s (Fig.3).) Surface temps:
So we can be pretty sure that the sudden and substantial reduction in OLR that occurred around 1960 didn’t produce any ‘global warming’ whatsoever at least up until the end of 1976.
But how come the Earth didn’t just not warm, but in fact cooled? If a reduction in OLR – which did, after all, happen, according to the models – is meant to produce warming? Well, it can only produce warming insofar as the ASR, the opposing – incoming – heat flux, doesn’t simultaneously decrease also, by at least the same amount. And as we can see in Figs.39 and 40, this is exactly what happened, once again according to the models.
However, this leaves us with the same question: What caused Earth to cool during this period? If the models are right in claiming that the ToA net heat was never consistently negative …
The NATURAL, and most obvious, explanation of the cooling, and of the 1960-76/77 section of the graph in Fig.40, would be the following; but it would require the golden ASR curve to drop further down than the red OLR curve, to create a negative ToA imbalance, and a certain (slight) delay between the two, the ASR leading, the OLR trailing:
1) The incoming heat from the Sun is, for some reason or other, reduced (–ASR);
2) global temps drop as a natural response to (effect of) the reduction in solar heating …… …. (–T);
3) Earth’s heat loss to space is weakened as a natural radiative response to (effect of) the …. dropping global temps (–OLR).
–ASR → –T → –OLR;
root cause → primary effect → secondary effect / negative feedback
Which would, if true, be the exact opposite course of events from what we’ve observed over the last three decades, since 1988/89, as handily described in Donohoe et al.’s 1B scenario:
“(…) consider a hypothetical step change in solar insolation (Fig.1B): ASR is increased, and energy accumulates until the climate warms sufficiently that OLR balances the ASR perturbation. In this case, the net energy accumulation (shaded red area in Fig.1) is a consequence of increased ASR and opposed by the increased OLR (hatched green area in Fig.1).”
Further, we see the ASR gradually increasing from 1970 onwards, while the OLR stays put. How do we explain this in terms of cause and effect?
Reduced OLR evidently didn’t cause any warming prior to 1977. If it was low, it was because the temps were low. So what produced the observed jump in global temperature going from 1976 to 1977? Well, there was a certain buildup of ‘ocean heat’ from 1970 to 1976, but this very much appears to be ENSO-related (five out of seven years in deep La Niña conditions). And there was no tropospheric warming going on to affect the ToA fluxes; quite the opposite. We see a big upward shift in the positive net flux in Fig.38 across the 1976-77 transition, but that’s clearly the result of a sudden rise in ASR, not in a sudden drop in OLR (Fig.40); the OLR at this point can simply be assumed to remain in a slump due to the low surface and tropospheric temps.
So how, when and where is there ANY evidence of “Greenhouse” warming prior to 1977? And how, when and where is there ANY evidence of the same after 1977?
Because after we know what’s happened … It’s in the freely available observational data, after all. Collected from the real Earth system. Global OHC data (NOAA/NCEI) from 1977 to 1988, ToA radiation flux data (ERBS+CERES) from 1985 till today.
BOTH the ASR and the OLR went up. Directly contradicting the models. Which claim the ASR went up, but not the OLR.
So this is what MCS needs to be doing, and indeed does, to defend its models from the data. They need to say we DON’T know. Only the models really know. They need to assert, and do indeed assert, that:
- The observed increase in ASR post-1976/77 is somehow just a positive SW feedback to “Greenhouse” (LW) warming (–OLR) (to this end, they also have to implicitly deny the validity of the OHC data from 1977 to 1988), and
- claims of observed increase in OLR post-1976/77 is somehow wrong (or the data showing this is way too uncertain to be ‘useful’).
But are Trenberth/Fasullo and Donohoe et al. right to distrust the validity of the radiation flux datasets, and thus simply dismiss them altogether? Or are they just pulling their claims of concern out of their … behinds?
“(…) for observations of the Earth’s radiation, changes in instrumentation [and calibration] do not allow changes to be detected between the Earth Radiation Budget Experiment (ERBE) and more recent Clouds and the Earth’s Radiant Energy System (CERES) observations [Fasullo and Trenberth, 2008a].”
Loeb et al. (2012) beg to differ.
“A continuous record from Earth Radiation Budget Satellite (ERBS) is compromised by a discontinuity (late 1993) [Trenberth et al., 2007].”
And Wong et al. (2006) disagree:
“Comparison of decadal changes in ERB with existing satellite-based decadal radiation datasets shows very good agreement among ERBS Nonscanner WFOV Edition3_Rev1, HIRS Pathfinder OLR, and ISCCP FD datasets. (…)”
Even Trenberth’s own (self-referencing) link (leading to the IPCC) appears to disagree:
“These conclusions depend upon the calibration stability of the ERBS non-scanner record, which is affected by diurnal sampling issues, satellite altitude drifts and changes in calibration following a three-month period when the sensor was powered off (Trenberth, 2002). (…) However, careful inspection of the sensor calibration revealed no known issues that can explain the decadal shift in the fluxes despite corrections to the ERBS time series relating to diurnal aliasing and satellite altitude changes (Wielicki et al., 2002b; Wong et al., 2006).”
The basis for Trenberth’s concerns was simply sorted out by 2006, with the development of the current Ed3_Rev1 version.
Donohoe et al.:
“(…) the limited length of satellite TOA radiation measurements precludes determination of the relative contributions of ASR and OLR by direct observation.”
I mean, how cheap an excuse is this? Limited length!? We have high-quality satellite ToA radiation flux measurements going back 33 years! That’s more years than what’s needed to establish a ‘climate normal’. We can easily determine from these measurements the “relative contributions of ASR and OLR by direct observation,” don’t you worry about that.
ASR and cloud albedo
As to what caused the marked observed increase in ASR from the 80s to the 90s, the most (probably the only) likely explanation is a drop in cloud albedo:
This matter is discussed at length in the following document: “Assessment of Global Cloud Data Sets from Satellites” (GEWEX (WCRP), Nov 2012).
They specifically point out the conspicuous correlation between the changes in cloud cover reported by ISCCP and the changes in reflected shortwave during the same period reported by ERBS:
“3.5.3 Long-Term Variations and Radiative Flux Constraints [pp.66-67]
(…) constraints can be placed on the variations of cloud properties (geographic, seasonal, synoptic, diurnal) using the observed variations of the TOA fluxes. Going further, comparing in detail the observed top-of-atmosphere and surface radiative fluxes and their variations to calculations of radiative [fluxes] that account not only for the cloud properties but also for the atmosphere and surface properties as well can provide an even stricter test of the accuracy of the cloud properties and their effects on Earth’s radiation budget: This exercise has been carried out for the ISCCP product, where the agreement between the ISCCP-based calculations (ISCCP FD product) and the long ERBS-based record of top-of-atmosphere radiative fluxes show excellent quantitative agreement (Zhang et al., 2004). This study also shows good agreement with another calculation that uses the ISCCP products in a different way and with the ERBE and CERES TOA flux products.”
The authors conclude on the validity of the ISCCP global mean cloud amount (CA) data, as depicted in the Appendix Figure 2.1 (p.127):
“2.6 Conclusions [p.131]
A number of factors that might cause spurious changes in ISCCP global monthly mean total CA [cloud amount/fraction] have been investigated: the most that can be said nowis that the slow variation of global CA shown in Figure 2.1 may be somewhat exaggerated in magnitude, especially the peak values in the late 1980s, but that this variation cannot be dismissed as completely spurious. None of the hypothetical causes of spurious total CA changes is large enough to explain this variation. This conclusion is also supported indirectly by the fact that the anomalies in top-of-atmosphere radiative fluxes, calculated based on the ISCCP cloud properties, are in excellent quantitative agreement with those determined from the long-term ERBS instrument record (Zhang et al., 2004; Norris 2005). Moreover, given the stability of the ISCCP radiance calibrations, the variations of cloud types also appear to be reliable for changes larger than about 0.01-0.02.”
“3.5 Long Term Variations
3.5.2 Anomalies [p.64]
Detailed investigations (Annex 2) on possible sources leading to spurious changes in the ISCCP CA time record show that, although they can change the magnitude of the slow CA variations by about one third, they cannot account for all of the variation.”
During the most recent 17-18-year period, this is how the global cloud cover anomaly evolved, according to MODIS: