Earthshine project reports on Earth's reflectance over the last 16 years

In the early days of WUWT, I reported on this project here: Earth’s Albedo Tells an Interesting Story Now data has been release in a new paper. From the INSTITUTO DE ASTROFÍSICA DE CANARIAS (IAC)

New data on the variability of the Earth’s reflectance over the last 16 years

The measurements of this key climate variable, obtained from telescopes on the ground, agree with the measurements made from satellites and show a period of stability

Composition of the earthshine, the light reflected from the Earth to the night-time face of the Moon, and the figure showing the monthly mean apparent albedo anomalies from December 1998 through December 2014. Anomalies were calculated over the mean of the full period, positive anomalies are shown in red and negative in blue. Averaged standard deviation (error) of the monthly data is also indicated in the lower right corner for simplicity. Only months with at least 5 nights of observations are shown. From November 2005 to August 2006 several months of earthshine data are missing due to the replacement of the dome of the solar telescope, while the new automated telescope under a separate dome was not yet operational. CREDIT Daniel López/IAC.

Click for full size image. Composition of the earthshine, the light reflected from the Earth to the night-time face of the Moon, and the figure showing the monthly mean apparent albedo anomalies from December 1998 through December 2014. Anomalies were calculated over the mean of the full period, positive anomalies are shown in yellow and negative in blue. Only months with at least 5 nights of observations are shown. From November 2005 to August 2006 several months of earthshine data are missing due to the replacement of the dome of the solar telescope, while the new automated telescope under a separate dome was not yet operational. CREDIT Daniel López/IAC.

The Earth’s albedo, or reflectance, is a fundamental atmospheric parameter having deep implications for temperature and climate change. For that reason, experiments have been performed to monitor it over the past two decades to reveal how it evolves. One of these, in which the Instituto de Astrofísica de Canarias is a participant, has brought up to date the observations made since 2007 and adds to, and gives a new systematic record of the Earth’s albedo covering the period between 1998 and 2014 from California. This study, whose first author is IAC researcher Enric Pallé, has been published in Geophysical Research Letters, and shows that although the albedo fluctuates both monthly and annually, but there is no trend over those years.

The effect of the albedo is important for life on Earth because it affects the energy budget of the planet: it plays a key role in the process by which the planet achieves an equilibrium between the solar radiation which enters the atmosphere and the radiation emitted as heat into space. The albedo is defined as the fraction of sunlight that the Earth reflects back directly into space, and it varies according to cloud composition, ice, snow, and the nature of the surface onto which the sun’s rays fall. The term comes from the Latin “albus” (white), and a rise or fall in the albedo will have consequences for the planet’s global warming or cooling.

To check how the albedo has evolved, since the 1990’s it has been measured from space with instruments such as ERBE and CERES, whose estimations are made using absolute photometric measurements, which could be affected by any degradation in the instrument while in space.

However, a complementary way to measure the reflectance, which does not suffer the same calibration errors since it is a relative measurement, is from the ground, using telescopes that observe the so-called earthshine (the light reflected from the Earth to the night-time face of the Moon). This method has been used during the period 1998-2014 from the Big Bear Solar Observatory (BBSO) in California, and also, since 2007 from the Earthshine telescope at the Teide Observatory (Tenerife), to measure this climate parameter. These observations were aimed at increasing the temporal coverage of the measurements, and reducing the measurement errors.

The result of this study is that, applying strict quality criteria to the measurements of the earthshine, and after re-calibrating the measurements taken from the CERES instrument in space, the variations found in the value of the albedo not only agree in magnitude but also show identical, small annual variations over the 14 years that the two types of observations overlap. Philip Goode, lead earthshine researcher at BBSO explains that “Although the measurements that we have made of the albedo over the past 16 years show monthly and annual variations, there is virtually no change in the long term. This also coincides with a stabilization in the mean temperature of the planet”, he says.

Even so, although a consensus has been reached about the results of the two types of measurements, the origin of the anomalies is not completely clear. “We need to continue the experiments to measure this phenomenon accurately and see where we get to in a few more years. For example, the construction of a global network of robotic telescopes around the world or the launching of a micro-satellite dedicated to the study of earthshine will give us data to improve our knowledge of changes in the albedo, and see how they affect the climate”, concludes Pallé.


BBSO is operated by New Jersey Institute of Technology.

Article: “Earth’s albedo variations 1998-2014 as measured from ground-based earthshine observations” by E. Pallé (IAC-ULL), P. R. Goode (Big Bear Solar Observatory, Institute of Technology, EEUU), P. Pilar Montañés-Rodríguez (IAC-ULL), A. Shumko (Big Bear Solar Observatory, Institute of Technology, EEUU), B. González-Merino (IAC-ULL), C. Martínez Lombilla (IAC-ULL), F. Jiménez-Ibarra (IAC-ULL), S. Shumko (Big Bear Solar Observatory, Institute of Technology, EEUU), E. Sanroma (IAC-ULL), A. Hulist (IAC-ULL), P. Miles-Páez (IAC-ULL), F. Murgas (IAC-ULL), G. Nowak (IAC-ULL), S. E. Koonin (Center for Urban Science & Progress, New York University, EEUU). Geophys. Res. Lett., 43, doi: 10.1002/2016GL068025.


The Earth’s albedo is a fundamental climate parameter for understanding the radiation budget of the atmosphere. It has been traditionally measured not only from space platforms but also from the ground for 16 years from Big Bear Solar Observatory by observing the Moon. The photometric ratio of the dark (earthshine) to the bright (moonshine) sides of the Moon is used to determine nightly anomalies in the terrestrial albedo, with the aim of quantifying sustained monthly, annual, and/or decadal changes. We find two modest decadal scale cycles in the albedo, but with no significant net change over the 16 years of accumulated data. Within the evolution of the two cycles, we find periods of sustained annual increases, followed by comparable sustained decreases in albedo. The evolution of the earthshine albedo is in remarkable agreement with that from the Clouds and the Earth’s Radiant Energy System instruments, although each method measures different slices of the Earth’s Bond albedo.

Full paper (paywalled)


newest oldest most voted
Notify of

Long term stability would seem to track the “pause” in global temperatures over the same time period.

george e. smith

So on some given night, the sun is shining on some other place on earth; ” over there ” with whatever clouds, snow, grass, whatever, happen to be over there at that time, and the moon is ” up there ” above us moonshiners and off at some angle from the sun direction, so the moon of course is not uniformly illuminated, but any region on this side of the moon will be reflecting some of the earthshine that falls on it, and that amount reflected will depend on where on the moon surface it is, and off course the moon perturbates from time to time, as to just where it is pointing.
So that reflected reflected sunshine now comes back to the earth; hey! we’re over here, not over there, so those clouds/CO2/H2O/aerosols above us (we’re the guy with the telescope) so our atmospheric absorption, is not the same as theirs “over there” where the sun is.
So why do I think this is a pretty rinky dink way to measure how much solar EM radiant energy, is actually making it past all of these barriers and actually landing on the ground (solid and liquid ground) to warm the earth, on the sunny side.
Well yes I admire their tenacity and all that, but it’s a poor substitute for actually measuring what really lands on the earth.
That of course would take a radiometer network UNDERNEATH all those clouds ” over there ” where the sun is shining.
BUT ! it is nice to find out that even with such a wishy washy method, there doesn’t seem to be any reason for the earth to have warmed up over the last 16 years, which is a relief to find out, because we know it actually hasn’t warmed up during that time.
I wonder if this third or fourth party indirect proxy for a stand in for a real measurement, can also tell us how much the global average atmospheric CO2 abundance in ppmm has been during those 16 years.
If chemical process engineers designed their process control feedback loops using ” Climate Science ” proxy methodologies, we might expect a sizeable (probably logarithmic) increase in the instances of destructive chemical plant explosions.
But it is nice that they are now telling us, what they have accomplished so far.
Is the real paper paywalled, or can we get to read it ?

Jeff Norman

Climate “science” does not encourage the dissemination of measured data. Please revert to the modelled data which only shows a positive feedback.


“Within climate change science almost all the research (97%) refers to modeling in some way.”
From a recent article published by the CATO Institute:

Is 16 years a long enough time to get enough data to draw any inference, let alone conclusion?


It’s long enough to say that, for the past 16 years there is no trend. That’s important because I am assuming that the GCMs show a steadily decreasing albedo parameter.

I think the only thing that it shows is that there is not a catastrophic collapse of the reflectivity of the earth. The often repeated discussions about how arctic sea ice decrease is catastrophically changing the albedo seems to be pure bunk. But I could have told you that. The only month when the arctic gets significant sunlight, there isn’t ever much ice. The change doesn’t have any significance when discussing global albedo.


It is when you mind is already made up and closed.

george e. smith

What data and what conclusion are you trying to get ??
16 years is exactly long enough to get 16 years of data of anything you could possibly measure, or want to measure.
From that data you can determine exactly what happened to that metric during the 16 years in which you gathered that data.
Any time interval is generally sufficient to gather enough data on absolutely anything that happened during that time interval, and you cannot obtain enough data to determine what happened for some longer time interval than that for which you made data observations.


If the 16 years show no trend in the albedo, this means that albedo variations have not been what has been countering the warming by CO2. Something else has, or else there would be no pause.


Or else it wasn’t CO2 that was doing the warming. Just as it didn’t do any warming between 1940 and 1975 or, according to the IPCC before then either.
Makes you wonder what magic powers it acquired in 1975.


No “magic” powers.
Just overcame the negative forcing of sulfate aerosols.
And the “global dimming” of dirty industry post WW2 that lasted into the 80’s….comment imagecomment image


I love how you use past data to try and prove that the models are accurate, especially when they carefully tuned hundreds of variables to get the projections to match past data (and still didn’t do a very good job.)
On the other hand, when the same models make projections into the future, they have failed.

That model breaks down at the “Pause”. If it’s half right, then it’s half wrong, and cherry-picking a chart that stops at 1990 proves nothing.


“That model breaks down at the “Pause”. If it’s half right, then it’s half wrong, and cherry-picking a chart that stops at 1990 proves nothing.”
It shows what I posted it up to show.
That it ends in 1990 matters not a jot.
THE GHE of anthro CO2 did not outweigh the -ve forcing of sulphate aerosols and “global dimming” until into the eighties.
And thereby rebutting this assertion….
“Just as it didn’t do any warming between 1940 and 1975”

Mark - Helsinki

That’s not observed attribution Toneb, ROFL. It’s observed warming adjusted to hell and models that have been brought in line with observations.
The sulfate question is also actually unknown, and lastly WIki is not a source.

Mark - Helsinki

*models brought into line with adjusted observations

Mark - Helsinki

Your logic is completely flawed and you are seeing what you want to see

Mark - Helsinki

It only makes sense if one is ludicrous to claim, and you are, that all warming is man made and all cooling is natural. Typical Guardian reader logic.
It seems there are some scientists who say man made CO2 is influencing warming since 1850 and others who say (IPCC) 1960.
Also the claim “And the “global dimming” of dirty industry post WW2 that lasted into the 80’s….” is baseless
Where is the data to show this? It’s cant be a temperature graph or a model btw

Gary Hladik

Nylo, that was also my first thought, assuming of course that the measurements are “robust”. Another possibility is that the “pause” is really the measurement error. Or maybe it’s something else entirely.
Gee, this science stuff is HARD. 🙂

george e. smith

I think you are confusing warming with changes in CO2.
For the subject 16 year period, the paper says, there has been no change in albedo.
For the subject 16 year period, the data says there has been no change in the Temperature.
For the subject 16 year period, the paper says the CO2 has increased circa 24 PPM.
Ergo The Temperature MAY be tracking the albedo; but it most certainly is NOT tracking the CO2, nor for that matter is it tracking the global human population.


Since I’m lazy and don’t want to go read the paper etc., does anyone know what wavelengths they are measuring? It seems to me that reflectance at different wavelengths would be important.

True. The whole global warming story is that the IR is absorbed by the increased CO2, and therefore can’t leave the earth system like it did before we F-ed everything up.

just all the white light from the moon that gets through the earth’s atmosphere.


Thanks, Leif!


What’s the wavelenght of white light again, lsvalgaard?
(I’m guessing he meant viable light ; )

white light is that mixture of all wavelengths that we can see:


(Yeah, I meant visible but I had just come in from glaring sunlight and couldn’t actually see my screen very well ; )

MarkW May 19, 2016 at 2:42 pm
Since they are measuring from the bottom of the atmosphere

Well 7800 feet up anyway.


What about all the blue light scattered in the atmosphere each way? Sun looks yellow, moon looks white…

george e. smith

The Radiation reflected from the moon would include the solar spectrum radiation that is reflected from earth (albedo) plus the LWIR radiant emission from the earth that is emitted to space, to stop it overheating.
So the solar spectrum portion of it, would have 98% of the power between 0.5 and 8.0 times the peak wavelength , so that is about 250 nm to 4.0 microns, for the solar insolation peaking at about 500 nm.
The earth LWIR emission on the other hand has a spectral peak at about 10 microns, so the 98% spectrum is from 5.0 microns to 80.0 microns.
So those two do not overlap, so a filter can easily stop the 5-80 micron LWIR leaving the solar spectrum component.
Now at night time, there wouldn’t be direct sunlight scattering in the night time atmosphere, but the radiation returning from the moon would presumably scatter very much the same as it did when originating from the sun.
Both the moon and the sun subtend an angle of about 30 arc minutes, so they would be quite similar spectra, just quite different intensity.
So the blue component should scatter in pretty much the same way that the sun does, and since it is illuminating the whole night time sky, the apparently black sky is really sky blue just like in daytime.
So at the surface, the moonshine spectrum should have lost some more of the short wave part of the spectrum, due to the Raleigh and Mie Scattering that we see in daylight.
But generally those should be fairly constant factors.
But that doesn’t alter the apparent fact reported in this paper, that for the 16 years of their observations neither the albedo, nor the Temperature show any significant change, whereas the CO2 has clearly increased by some 24 ppmm over that time.
That suggests that albedo controls the Temperature, and CO2 does not. At least that’s how it was for that 16 years during the pause.

Lance Wallace

The paper did not seem to give a precise definition of the wavelengths being monitored. However, the last paragraph indicated that they will be measuring “color albedo” in the future. they do reference a 2005 paper that looked at albedo as a function of lambda. That paper (Montanez 2005) used a single night (actually only about 5 hours) of monitoring spectroscopic variations from Palomar. The range of frequences was from 480-980 nm. The albedo dropped sharply from about 0.3 at 500 nm to 0.25 at 600 nm and then stayed pretty much the same out to 980 nm.


Since they are measuring from the bottom of the atmosphere, it’s not likely that UV or IR are included.

george e. smith

Roger that. Just what moonshine are they drinking ??

Kevin Kilty

I am a bit surprised by the alleged correspondence of absolute measurements from satellite and the indirect measurement of earthshine, because the weighting as a function of latitude has to be very different. What this suggests to me is that the albedo must be dominated by what occurs in the tropics.

“Alleged correspondence” is a good choice of words.
Are the measurements from the individual recording stations and methods shown? Or are they blended together in some sort of minimum error weighting model?
CERES satellites if I am not mistaken are in Sun synchronous orbits, so they take their measurements at a constant solar time over the ground. The TERRA satellite, for instance passes the equator southbound at 10:30 am solar local time on every orbit.
10:30 am is a lousy time to measure the earth’s albeto from cloud cover.
There are two other CERES satelites, one is only 20 minutes behind the other, the other is an afternoon pass, like at 1:30 pm local time.
Willis has pointed out how afternoon clouds in the tropics can change the albeto in late afternoon quickly, even from the low sun angle late afternoon reflecting off of towering cumulous clouds.
So I wonder just how good the CERES data really is at albedo measurement and how closely the agree with Earthshine.

Michael J. Dunn

The Moon is far enough away, and the reflected light is sufficiently non-specular that it smears across the Moon. What you are concerned about has no effect.

george e. smith

The moon’s reflectance is clearly not the same everywhere on the moon.
Unless the moonshine measurements are made ONLY within +/- 24-48 hours of Full moon, I would NOT trust the values.
This method is too many proxies removed from measuring what lands on earth under the albedo mirror.

I think the concern Kevin and myself is on the “Alleged Correspondence”, emphasis on “Alleged” between albedo estimates separately from CERES and Earthshine off the moon.
It might indeed be that Earthshine is sufficiently averaged via smear that it is the better of the two classes of measurement. But Earthshine is a non-dense form of measurement in that it is dependent upon the phase of the Earth and the relative axial tilt as a function of time of year. Only a small fraction of the month is the moon sufficiently “NEW” and sees a Full Earth to get a good average. Does it see the Earth morning or afternoon? (It matters). The moon’s maximum declination with respect to the earth’s equator can change by 10 degrees over the course of a score of years from +/- 28.5 deg to +/-18.5 deg. (see “Lunar Standstill Couple that with the hemispheric spring season of blinding white ice, north pole and especially south pole, and there are a lot of degrees of freedom in the lunar Earthshine measurement.
Conversely, CERES seems to be designed to minimize the variability of albedo as a function of the time of local day and seems ill-suited to the task of measuring the earth’s overall albedo an any given time of day. CERES is well able to measure the albedo changes of any particular point at 10:30 am. But can it determine the albedo change at 3 pm?
or 5pm,
or 8pm,
or 8am.?


Unless the moonshine measurements are made ONLY within +/- 24-48 hours of Full moon, I would NOT trust the values.

It was stated that there was monthly variability in the data.
With the exception of small variations due to libation, the same part of the moon’s surface is always towards Earth. It should be possible to measure variation in reflectance across the surface and compensate for whatever section of that surface is being observed.
I have not read the paper to see whether this is done. Neither, it seems, have you.

Michael J. Dunn

Dear George,
First, a hat tip to Greg, for pointing out that we always see the same side of the Moon (libration being only a small effect), so integrating over the face of the Moon should be a suitable approach.
Second, a guffaw at your suggestion that we should attempt these measurements only at full Moon–when the Moon only has the dark side of the Earth to reflect! (Really embarrassing when you out yourself like this.)

What are the +/- W/m^2 variation boundaries? How do they compare to IPCC AR5’s CO2 RF of 2 W/m^2? AR5 credited cloud CRE with -20 W/m^2. A +/- 10% matches/exceeds CO2’s RF.


The paper is also available at this non-paywalled site:
Figure 4 gives you the values: the top panel suggests that over the zero trend period measured, that albedo variation was +3.5W/m2 to -3W/m2.

Gary Pearse

By eyeball, it seems obvious that albedo has increased in the latter 5years or so. It would have become an actual scientific paper if they’d also layed the other albedo data over a longer period beside it. I’m curious about a few decades ago when there was warming. Such an important negative feedback should be getting all the attention it can get.

Thomas Homer

If you were to throw a brick into a brook of running water, the brick would trap/reflect some amount of water for some amount of time. But the throughput of water would remain the same.
Isn’t that analogous to the claim that CO2 traps/reflects some amount of Earth’s radiation for some amount of time, but the throughput would remain the same? So these measurements of Earth’s Albedo would also remain consistent.
Now we need those that make the claim about CO2 having a “greenhouse gas” property to show how this physical property is measured. By definition, physical properties can be measured. If it can’t be measured, it’s not a physical property.


It’s not analogous. Albedo works on completely different wavelengths than does CO2.
It’s true that the total energy over some arbitrary time period will remain pretty close to the same.
That is, at any given instant total energy in may not exactly equal total energy out, but over time, they will average out to about equal, with the net difference approximating the change in temperature over the same period.
More clouds would reflect more light. To tie that into your analogy that might be the equivalent to someone digging an alternate channel and diverting some of the water away from your brick.

Thomas Homer

Thank you Mark – why aren’t we measuring the Earth’s reflectivity/emission for the “completely different” wavelengths that CO2 does “work on”? We’ll call that Albedo Prime. Does my analogy hold for Albedo Prime?


Because we care about all energy, not just the energy in the CO2 bands.
One of the stations that’s been measuring albedo the longest by measuring light reflected from the moon, is at the bottom of the atmosphere, which means that UV and IR are blocked by the atmosphere.

Thomas Homer

“Because we care about all energy, not just the energy in the CO2 bands.”
The whole entire premise of CAGW is that CO2 is trapping energy. Perhaps we should focus on measuring that?
What we’ll find is that the throughput of these CO2 bands remains unchanged, just like the water in the brook. Now the claim is that CO2 is trapping some amount of energy for some amount of time. Since the energy in the CO2 bands is moving at the speed of light,
let’s see how long light is trapped in a room full of mirrors, and that is our order of magnitude.


You need to factor in changes in TSI.


That is, you if you wish to calculate how much energy is actually reaching the surface.


If I understand the measurement, they are measuring the ratio of the sunlit and earthlit sides of the moon. So TSI falls right out of the ratio. The technique is an automatic correction for source (sun) intensity.

not really, as the change in TSI is much smaller than the error bar for the albedo.

Peter Sable

Preceding, and here comes, oodles of wild speculation on various correlations on not NEARLY enough data to speculate upon. Let’s do as the author suggests and measure it better. Then wait for 60 years of data to cover the ocean multi-decadal cycles, and THEN draw some conclusions.

The Earthshine data is one of the sets of observations that led me to consider the possibility of changing Earth’s albedo via changes in the meridionality of the jet stream tracks on the basis that more meridionality results in longer lines of air mass mixing and thus more clouds.
Such meridionality does seem to be solar related:




For interests sake the earth to moon albedo stats should be compared to satellite records.
Measuring the brightness of the Dark Side of the Moon and calculating out the Earths albedo maybe cheap but is it accurate.
I’m more interested in Earths albedo in day/night.


Just for accuracy, the dark side of the moon is not dark. It is the side of the moon that faces away from the earth, so does not receive earthshine, by definition.
You meant measuring the brightness of earthshine visible in moon shadow, on the side that faces the earth.
And the albedo measurement is only reflective of Earth’s day side, obviously.


Since there is only one station that is measuring this data (Big Bear Observatory (now that’s tough field work)) that means that the same side of the earth is always being tested. To get better data there should be two or three observatories as evenly spaced as possible.

george e. smith

Not so.
It is earth’s day side reflectance plus earth’s night side transmission, and each of the two is not the same.
Remember they are recording photons that have been to the earth twice, and visited the moon on the way . That’s not a good way to measure the photons that visit the earth once AND STAY HERE !!


There is no dark side of the moon, really.
Matter of fact, it is all dark.
The moon is blacker than freshly laid asphalt.


I’m afraid Earthshine is a terrible way of measuring Earth’s albedo.
Albedo was measured by ISCCP satellite between 1994 and 2000, and showed a marked decrease in Earth’s albedo from about a +3% anomaly to 0% anomaly. Apparently nobody trusts that data anymore and has been ditched. It is in black in this figure from Pallé et al., 2004comment image
In blue in the same figure is the Earthshine albedo calculation from the same article.
There was a big uproar because Earthshine data showed a big increase of albedo at a time of cooling and thus provided an alternative explanation to GW.
So the data was reanalyze by Pallé et al., 2008
To show that the increase in albedo was actually quite a lot less.
This figure from Tallbloke shows the end result. Anything later than 2000 is only Earthshine, and the 2008 data (light blue) shows a marked reduction in the trend.comment image
However a new satellite, CERES, started measuring albedo in 2000, as reported by Loeb et al., 2012.
They found albedo is pretty constant. We can see albedo changes both global and by latitude in this figure:
from Seidel et al., 2014
Now Pallé et al., 2016 say that the Earthshine data is in good agreement with satellite data. It is not. Earthshine data has to be adjusted a lot, and shows a lot more variability than satellite data. Even their own figure 4 shows that the agreement between both is not very good. They however claim that it is good and get away with it.
What the albedo data says is that a 1% anomaly is actually huge in terms of energy. However the Earth is strongly buffered so that the short term changes in albedo that are being measured do not translate to changes in temperature. Over the long term, small changes in albedo that we cannot accurately measure and we have not been measuring long enough to detect could potentially have a bigger effect than changes in atmospheric GHG concentration.
Earthshine data appears pretty much useless and could lead to mistakes if trusted.

george e. smith

Well you’ve got that right. Earthshine photons have come here twice, and been to the moon as well.
And as for their recent studies showing no albedo variations, I think that sort of jibes with there being no warming either for the same time. Ergo no change in earth’s energy capture from the sun.

My problem with CERES data is that it is acquired by three satellites in two solar synchronous orbits.

CERES data comes from a device on solar synchronous satellites (TERRA, AURA, AQUA) which are 720 km high, 99 min. orbits. TERA only sees the earth at solar 10am to 11am (depending on Latitude) making its equatorial pass at 10:30 am. AURA and AQUA are in the same orbital plane train, 8 minutes apart making sunlight equatorial passes at 1:30 pm. (Rasey – 1/5/2014)

The CERES satellites do an excellent job of looking at changes in the earth and atmosphere globally, but only at two times per day: 10:30 and 1:30 pm. But can they measure changes in cloud cover over the years at 3pm, 4pm, 5pm? No. those times are beyond the horizon from the satellite in the tropics.
The CERES dataset contains data at other times, but these data come from the GOES geosync satellites.
Perhaps GOES has the instrumentation to properly measure the albedo changes over the years. But then, why would it be called the “CERES” dataset?

Actually, Earthshine has been quite useful. Earthshine found an albedo increase that the satellites had missed. Both Earthshine and the satellite projects ended up being improved – and the albedo increase stayed.'s_reflectance_1999-2007
It is always a good idea to measure things from different angles and in different ways, so Earthshine does have value.
More useful information here: (see “Can Earth’s Albedo and Surface
Temperatures Increase Together?”).


I agree that it is a good idea to measure things in different ways even if one of the ways is shitty. In this case in my opinion the Erthshine is the shitty one.
They explain the huge peak in albedo in 2003 that they detect and satellites do not as sparse data (your second reference), and the increase in albedo and temperatures as due to changes in the ratio of high-clouds / low-clouds (also in your second reference). Another example of “science” that can explain one thing and the opposite and accommodate any result and still be right.
In your first reference, if you care to look for it, the first figure is shameful. Top Earthsine and bottom CERES do not look at all any similar, One has positive slope and the other negative, yet the author concludes that they “do not differ much.” It is not a good journal, but still one wonders how these things get passed peer review.

Thanks for the added detail.

Do you understand that the Earth shine data is limited to the spectral window from 0.36 to 1.0 microns while the CERES data that you are displaying is for a much wider spectral range? You are comparing oranges with apples – hence your analysis is essentially worthless. The Earth-shine project is not measuring the Earth’s total albedo – and it has never claimed that it has. It only measures the ratio of reflected light from the Earth to incoming light in the visual and near infra-red.

Of course. Read the linked papers.


It is not my analysis that we are talking about, astroclimateconnection. It is the people responsible for the Earthshine data that are doing the apple to oranges comparison when they transform the Earthshine data into changes in % of total albedo or albedo anomaly, and changes in W/m2 to compare their data with satellite data. See for example fig 2 in: image?w=700
You should take your complain to them.


That’s a very strong AMO signal on the ISCCP data.


The relation between polar caps and albedo has to be discussed. An AGW proponent please step forward.
According to AGW, the two melting poles, now 1 Mio sqkm less than 30 years ago in their extend, should lower the albedo because of no more sunlight reflectance from the missing ice surface. The global overheating is connected to the dark water surfaces, which, now free of ice, absorb more heat instead of reflecting it into space.
Mosher, step forward. Cheers JS


Global albedo is mainly atmospheric albedo. Surface albedo has a very small contribution to global albedo, and most of it comes from Antarctica. The changes in global albedo due to Arctic melting are imperceptible. The effect of changes in albedo on ice melting both on glaciers and sea ice are so far inferred, not demonstrated.

george e. smith

Albedo is global. Reflectance may be local. Albedo is one number for the entire earth. 35% or thereabouts.
Not much sunshine at the poles to even worry about so how much gets reflected from there is irrelevant.


Albedo is not a number. It is a surface property, and can be measured and calculated for any part of it, like for example on a latitudinal axis:
Antarctica is a significant contributor to global albedo unlike most of the rest of the Earth’s surface.

george e. smith

There seems to be a very good reason for all that ice at the polar regions.
I think it might have something to do with there not being very much sunshine getting to the poles.
And if there is not much sunshine getting to the poles, then there wouldn’t be very much reflected sunshine from the poles contributing to earth’s overall albedo of 0.35 or so.
So nyet on polar ice affecting earth albedo much.
And just think how bloody cold the poles would be, if it wasn’t for all of the huge amount of heat convected to the poles by ocean currents like the gulf stream and atmospheric currents. Ergo, there’s even less sunshine at the poles than you would think, so even less albedo component from polar ice.

bit chilly

it would be really good if you could go over to the arctic sea ice forum and explain that george e.smith. they seem to have terrible trouble understanding that.


the entire average albedo of Earth tends to be measured at 0.30 or so (with variations substantially from that at different times. what matters is the sunlit side hemisphere. ocean tends to run around less than 0.04 albedo and land typically from 0.12 to as much as 0.4 (sand and snow) but it’s only 1/3 of the surface. About 62% of the Earth has cloud cover at any particular time. This has higher albedo reflection than even snow and sand. When you calculate it out, assuming albedo = 0.3, then the atmosphere albedo (including atmospheric scattering) provides about 0.22 of the albedo and the surface visible through the clouds or between the clouds provides about 0.08. What is absorbed in the Earth system is TSI * (1.0 – albedo). While you are right about Palle’ having a heck of a mess to deal with figuring out albedo from Earthshine, there has been very little albedo data acquired over the years yet this absorbed power is actually relevant to Earth’s power balance while TSI is not directly relevant because albedo is totally variable in the short term and perhaps in the long term. The simple use of TSI assumes that albedo is a constant. Of course the great ‘minds’ of agw talk about albedo variation only in terms of human corruption due to land use changes – total BS since it’s 70% ocean to begin with. The other problem with this though is that it is not just cloud cover but also cloud reflectivity that affects most of Earth’s albedo. Lindzen’s iris effect (as I recall) was totally associated with reflectivity of the clouds rather than the cloudy fraction of Earth. It also does provide that opportunity of chaotic undeterminable behavior since it is both cloud cover fraction and cloud reflectivity that affects the albedo and that means there is this huge variety of internal and external factors that can affect it and it is ulikely that these factors can truly be segregated and quantified


I bet my last dollar if this albedo chart showed a trend that aligned with the CAGW meme, it would be lauded everywhere, esp. MSM.and CSIRO. Doom and gloom is what those in power need / want to keep the status quo. O/T but… It would be nice to see governments only pass new legislation by removing an equal number.

Ian Wilson

I will stick my neck out and make the following prediction about the mean monthly apparent albedo anomaly:
Not withstanding the large error bars associated with the mean monthly apparent albedo anomalies, and the gap in the albedo anomaly data between June 2005 and December 2006, there appears to be rough correlation between the 6 month retarded SOI and the monthly apparent albedo anomaly.
Please see:
If this correlation has any validity then I would predict that the Earth’s mean monthly albedo anomaly will remain in negative territory from late 2014 till the end of 2015.
The next few of years of data from the Earth-shine project should be very interesting if it does.


Does not look very balanced to me, appears weighted to the upside in recent years.


comment image

Full paper here: .
Palle, E., et al. “Earth’s albedo variations 1998–2014 as measured from ground‐based earthshine observations.” Geophysical Research Letters (2016).
As for long-term changes in albedo, I draw you attention to Fig. 4 at the end of the paper and to this summary on page X-3.
“On shorter time scales, large inter-annual albedo changes are observable, for example over the 2008-2010 interval there isa strongly increasing trend in albedo. However, this trend begins to reverse after 2010. Thus, taken in isolation, over the past several years, 2007-2014, the Earth’s albedo, as registered by earthshine, has shown a net trendless interval.”
In my opinion seven years is not “long-term”. Another Pallé paper is much longer term.
Changes in Earth’s Reflectance over the Past Two Decades
“The decrease in Earth’s reflectance from 1984 to 2000 suggested by ISCCP data in 5RPk”) corresponds to a change in ?p* of some –0.02, which translates into a decrease of the Bond albedo by 0.02 (?p*/p* =?A/A) (- ) and an additional SW absorption, R, of 6.8 W/m (R = ?A × C/4, where C = 1368 W/m is the solar constant). This is climatologically very significant. For example, the latest IPCC report (‘ &) argues for a 2.4 W/m increase in CO longwave forcing since 1850. Our observational ES data extend from 1999 to 2003 and indicate a clear reversal of the ISCCP-derived reflectance trend starting in 1999 up through 2003. The increasing trend in reflectance corresponds to approximately 5 W/m , bringing the mean reflectance anomaly back to its 1980s values. Only the ES data are currently available to signal this reversal; it will be interesting to see how the proxy behaves when ISCCP data are available beyond mid- 2001.”
(poorly copied, pls refer to the paper)

That 1368 w/m^2 is wrong and the reduction of 8 w/m^2 matters. The IPCC got all the numbers to match.I wonder how they did that? It’s 1360 w/m^2. If 1368 were the right number a decrease of 1 w/m^2 would wipe out the entire global warming regime. The only way I can see that they would address the 1360 w/m^2 is to ignore it. Oh, and that 1368 is used in the temperature constructions on Mars and Venus, probably others as well. By the IPCC’s own reckoning, the forcing of co2 is 0.66 w/m^2. What do you think it does to this ” latest IPCC report (‘ &) argues for a 2.4 W/m increase in CO longwave forcing since 1850.” . Bottom line is that the IPCC is off in their calculations, anywhere, again depending on which paper you reference for TSI, form 2 to 3 C… lower not warmer.

Pamela Gray

What matters is how much energy reaches oceanic locations that act as super cell batteries. We don’t need to know about land surfaces. Ring the Earth with anchored ocean sensors in the equatorial band that measures incoming solar energy at the ocean surface, as well as below the surface.


Interesting treatment. So from the arguments, I get that the earth “shines”. Produces it’s own light,, called albedo. Black magic? The mirror produces it’s own light? I gotta see that. There is no effect calculated from that ball of light that illuminates us? Interesting. It doesn’t vary in its output? But the earths shine is noteworthy as the sun’s output is negligible. Where is Denmark again?


What might account for noisy and quiet episodes?


How could this “science” be “settled” considering this important factor is just now being defined/quantified with measurements? This “science” is such a rush to judgement joke. What if they find fall in albedo can explain all the warming, and has nothing to do with CO2? The very fact all the funding has gone to collecting data on CO2 and not the sun, absorption, H2O and other key variables pretty much proves this isn’t a real science. It is a propaganda data generator. They crank out data like communist agracultural bureaucrats did to claim they were meeting the absurd quotas established by Stalin, Pol Pot, Castro, Mao and Kim Jung. They were all lies to support the communist lie.


Fiscal meltdown.

Proud Skeptic

16 years?
Maybe after observing it for another 16 centuries we will have enough data to draw rational conclusions.