We all know how powerful clouds are. Just stand outside on a patchy day — feel the goosebumps. These megaton floating conglomerates of water act as vast shields — they cover 60% of the surface of Earth, and even a small change makes a big difference. While changes in the total amount of direct sunlight coming off the sun are very small, the changes to the amount of reflecting surfaces floating above Earth are, proportionally, at least twice as large, and possibly much much more influential. The IPCC includes changes in sunlight (TSI), so it does not make sense to ignore the larger and more powerful changes in the Earth’s albedo (fraction of sunlight that is reflected) due to “external” factors (due to factors other than feedbacks to surface warming). Both contribute to the amount of sunlight heating the surface, or “absorbed solar radiation” (ASR) (before feedbacks).
There are lots of reasons clouds might change that are not included in standard climate models. Just for starters — cosmic rays may seed cloud formation. Aerosols released by plants, plankton and marine life do — some aerosols are included, but new varieties are turning up in studies. We know the solar magnetic field influences cosmic rays. Who knows what other effects solar factors have on clouds — through changes in spectral properties (like UV versus visible light), or through the solar wind. The IPCC admits they are weak in this area, saying “Clouds and aerosols continue to contribute the largest uncertainty…”.[IPCC, AR5, p592 Ch 7]. They also admit that different models handle them in different ways and they have “low confidence” in many aspects of cloud feedback. But if influences on clouds are a forcing, and they have been omitted, it turns their models inside out.
David compares the data on variation of albedo to the observed variation in total solar radiation, and finds that the former has at least twice the impact on surface warming. Obviously, any alternative climate model has to include “EDA” – external driven albedo — and since it is externally-driven, it is, by definition, a forcing.
We haven’t seen this comparison done elsewhere, though it may have been.
10. Externally-Driven Albedo (EDA)
Having discussed the main errors in the conventional basic climate model — heavy reliance on unverifiable partial derivatives, omission of feedbacks that respond to climate drivers directly rather than to surface warming, and applying the solar response to non-solar climate influences — this blog series now moves on to its second part, constructing an alternative, improved basic model for estimating the sensitivity to increasing CO2. (The third and final part of the series will be aimed at finding what did cause recent global warming, where the search will result in the notch-delay hypothesis.)
Albedo changes that are not in response to surface warming — those presumably driven by external forces — will come to play a potentially important role in climate modeling, so we need to discuss them before we embark on the alternative basic model. The externally-driven albedo changes omitted from the conventional basic climate model and the big computerized climate models (GCMs).
Albedo is the reflection of incoming radiation from the Sun back out to space, mainly by clouds and ice. Reflected radiation does not heat the Earth. The amount of albedo is very significant — about 30%. That is, only about 70% of the sunlight (or to be more exact, of the total solar irradiance or TSI) incident on the Earth contributes to heating the Earth. We call that 70% the absorbed solar radiation (ASR).
The significance of albedo is that it only takes a small change in albedo to make a big difference to the amount of ASR, and thus to the surface temperature. It’s like a tap on a fire hose. Unfortunately albedo is difficult to measure, and data only covers a short period.
The only changes in albedo accounted for in the conventional climate models are changes occurring as a result of feedbacks in response to surface warming. For example, surface warming reduces the area covered by highly-reflective ice and snow, so the albedo decreases and ASR increases. Or, surface temperature affects plant life which affects ground cover and thus albedo. Other effects of surface warming are thought to influence the amount and types of clouds, which also change albedo. We accept the best estimates of the effect of surface warming on albedo in AR5, which are expressed as the albedo feedbacks to surface warming in the final diagram of the conventional model, in Fig. 2 of post 9 (where the albedo feedbacks are separated from the non-albedo feedbacks).
“Externally-driven albedo” (EDA) is albedo that is not attributable to feedbacks in response to surface warming. Possibly due to non-terrestrial influences, EDA includes any changes in albedo due to modulation by the Sun. For instance, cosmic rays are widely suspected of encouraging cloud formation, or any number of solar effects might change the proportions of ozone and the relative heights of the tropopause at the poles and equator, which affects jetstreams and the amount of clouds. EDA might also involve terrestrial agents, such as cloud changes owing to aerosols released by plankton, or volcanic eruptions could change the lower stratospheric temperature which in turn changes albedo.
The total change in albedo is the change in EDA plus the change in albedo feedbacks in response to surface warming.
The proportional-variation argument here suggests that changes in EDA may be a larger influence on surface warming than the direct effect of changing TSI — because the proportional variation in albedo, even after taking out the variation due to surface albedo feedbacks, is much larger than the proportional variation in TSI. Hence we will include EDA in our alternative model.
The increase in the no-feedbacks ASR, ΔANF, is the increase in ASR that is independent of surface warming; see Fig. 2 of post 9. It can be partitioned into the part due to increasing EDA, and the part due to increasing TSI (via the direct heating effect of TSI):
The ASR at any time is
because the TSI is estimated for 1 AU (i.e. at the average distance of the Earth from the Sun) while the cross section of a sphere is only a quarter of its surface area, so each square meter on Earth only receives a quarter of the TSI at 1 AU on average. In the starting steady state:
Let the EDA increase by ΔαE, and the TSI increase by ΔS, during the time to the final steady state. Then the increase in ASR due to EDA and the direct heating effect of TSI during that time is
to first order. Comparing this to Eq. (1), we get the obvious result that
So how do the changes in ASR due to EDA and TSI compare, in a relative sense?
TSI, known as “the solar constant” until 1979 when satellite measurements began, has varied less than 1.8 W m−2, or 0.13%, over the 400 years since sunspot records began, according to the reconstruction of Lean (2000)  with the background corrections of Wang, Lean, and Sheeley (2005) . Reconstructions favored by the IPCC (e.g. Kirvova’s, smilar to Svalgaard’s) say the variation is even smaller.
Albedo data is sketchy, but sufficient for a lower bound on proportional variation.
Palle, Goode, and Montanes-Rodriguez (2008)  found some agreement between earthshine, the ISCCP FD product, and satellite (CERES) observations: from 1984 to 1998 the first two sources (CERES started in 2000) showed a fall in smoothed reflected solar radiation (upwelling SW) of ~1% or ~1.0 W m−2, then a rise by almost as much to mid-2000, then a roughly constant level to 2005 when the data stops. Some confirmation comes from Pinker, Zhang, and Dutton (2005) , who found that ASR increased from 1984 to 1998 by ~0.16 W m−2 per year, suggesting a fall in reflected solar radiation of ~2.2 W m−2 (the 11-year smoothed TSI rose ~0.3 W m−2 in that period, accounting for only (0.3/4) * 0.7 or ~0.05 W m−2), before levelling off then decreasing after 2000. We conclude there was a fall in reflected radiation of at least ~1.0 W m−2, which, since total reflected radiation is ~100 W m−2, corresponds to a change in albedo of ~1% (from say 30.0% to 29.7%). This is being conservative: even the CERES data from 2000 to 2004 (Stephens et. al., 2015 ) shows variation of ~2 W m−2 in reflected radiation, or albedo variation of ~2%.
Albedo feedback from 1984 to 1998 due to surface warming is responsible for fαΔTS, or ~(0.4±0.5) × 0.35 or 0.14±0.18 W m−2, leaving a fall of at least 0.86±0.18 W m−2 in reflected solar radiation.
Thus the proportional variation in EDA from 1984 to 1998 was at least ~0.86±0.18% (perhaps from 30.0% to 29.74±0.05%). These figures can only understate the relative variation of EDA, due to the relative shortness of the period from 1984 to 1998, compared to the last 400 years for TSI.
Plugging these proportional variations into Eq. (6),
ASR directly drives surface temperature, so the effect of changes in EDA on surface warming is at least twice as great as the direct effect of changes in TSI, and possibly much more. (And if the reconstructions that say the TSI variations are smaller are correct, then the relative influence of EDA is even larger.)
Given that the direct effect of changes in TSI is worth taking into account in climate models, so is EDA.
[1^] Lean, J. (2000). Evolution of the Sun’s Spectral Irradiance Since the Maunder Minimum. Geophysical Research Letters, 2425-2428.
[2^] Wang, Y. M., Lean, J. L., & Sheeley, N. R. (2005). Modeling the Sun’s Magnetic Field and Irradiance Since 1713. Astrophysical Journal, 522-538.
[3^] Palle, E., Goode, P. R., & Montanes-Rodriguez, P. (2008). Inter-annual variations in Earth’s reflectance 1999 – 2007. Journal of Geophysical Research.
[4^] Pinker, R. T., Zhang, B., & Dutton, E. G. (2005). Do Satellites Detect Trends in Surface Solar Radiation? Science, Vol. 38, 6 May 2005, 850 – 854.
[5^] Stephens, G. L., O’Brien, D., Webster, P. J., Pilewski, P., Kato, S., & Li, J.-l. (2015). The albedo of Earth. Reviews of Geophysics, 53, doi:10.1002/2014RG000449.