Zombie asks a question

Zombie, of San Francisco’s “Zombietime” fame, writes in with a question that he has graciously allowed to be given to our readers.

He writes:

I’m preparing to write an essay on the following hypothesis:

Solar power installations, especially in desert areas, replace light-colored high-albedo sand/rock ground surface with very low albedo black solar panels. The “side effect” (in fact, the whole purpose) of solar panels is therefore to capture radiant energy coming from the sun that would otherwise reflect back into space. Because this energy is then converted into electricity, which is then used to power devices and inevitably degrade into atmospheric heat (which does NOT as easily radiate back out into space), the overall result of large solar panel installations is to heat up the planet more than it would be heated without the solar installations.

But of course the solar-energy advocates will say that the solar installation is replacing a carbon-burning power plant, which produces greenhouse gases that the solar facility does not.

The question I seek to answer is: 

Has anybody ever actually sat down and calculated whether the CO2 greenhouse effect caused by a carbon-based power plant generating one megawatt of electricity is more or less than the warming effect caused by the lowering of the earth’s surface albedo from a black-panel solar installation with power output large enough to completely replace the carbon-burning power plant?

I suspect that no calculations of this type have ever actually been done, and that solar panels may in fact contribute more to global warming than anyone previously realized — and in fact may cause just as much warming as the power sources they replace.

I have searched but cannot find such a study; but the reason I’m writing to you is that I have some vague memory of this thesis once being discussed on WattsUpWithThat — although I no longer can track down where exactly.

So I ask: Do YOU (without any time investment) remember offhand where or when this hypothesis was discussed on WattsUpWithThat? And if not, do you think this is a worthy line of investigation?

I know this is a somewhat vague question, but your guidance is invaluable!

— zombie


newest oldest most voted
Notify of

Since the effect of CO2 is zero, the warming effect of solar panels are infinitely greater.

Jack McGrats

I have also entertained that question and it will be interesting if someone ever sat down and did the calculations. Another similar question would be: what kind of imbalance in the earths composition is created by removing heat via thermal means, ie using earth’s interior heat to warm a home, etc?

Terry Hoffman

This is an excellent question but part of a larger one:-) where is the thermodynamic equivalent of a free body diagram? Just as CO2 respiration is ignored, I suspect that the heat generated from biological processes is significant but ignored by the geologists, politicians and lawyers who mostly represent CAGW levels to use indolent minorities.

Dave D

Amatori: I think you mean approaching zero, beyond our current technologies ability to differentiate, a third decimal place, well, yeah, OK – zero.

Doug Danhoff

I believe that the “warming effect” of additional CO2 is actually a cooling factor in the atmoshere.
When you consider that CO2 does not produce any heat, (It is not a heat source) It re-emits energy as it recieves it, and a bit more than half is radiated away from the earth in each exchange, creating a negative balance. When this is considered in light of the logarithmic nature of additional CO2, I would agree with you that a gas fired power plant would be less warming than the solar farms that are supposed to replace them.
Good topic…. sorry I do not have a lead to the former debate on WUWT.

Eric Eikenberry

Heat generated from biological processes is insignificant, as the ice ages have repeatedly shown. Heat generated from CO2 is insignificant for the same reason mentioned above. Neither tail can wag the dog, nor will they ever.

I did not calculate it but instinctively it seems to me that the environmental problem of solar panels is that it will reduce the surrounding areas temperature significantly, thus causing alteration of the habitat around it.
The better the absorption, the less heat reaches the land.
In Israel there is a crazy plan on its way to erect a huge solar plant in the Negev desert.
People are beginning to realize that like the wind farms, this will cause a grave damage too.
Hopefully common sense will prevail.


This is a great point. I wonder what the answer will be. I suspect that when the IPCC estimates of climate sensitivity are used then the “CO2 effect” will be larger than the Solar Panel effect. I say this because we know that they have got their sensitivities too high already, which is why their models produce higher simulated temperatures than have been observed.
@Amatør1, let’s use the IPCC estimates – even if we disagree with them. We could also determine the sensitivity at which solar panels break even.

The difference is that a CO2 molecule yields useful energy only once, but can contribute to warming the atmosphere by absorbing and back-radiating infrared photons multiple times, for as long as it stays in the atmosphere. Estimating the exact number of those back-radiated photons is not trivial, but I would expect their cumulative energy to be orders of magnitude greater than the amount of energy generated by the combustion of the carbon that yielded the CO2.

Barry Cullen

Absolutely brilliant question! I’m looking for an old envelope or napkin to do the calculations.
Amator1 is close but there is still 5 – 7% energy absorption available (maybe a little more due to band broadening) by CO2 bands in LWIR going from ~300 ppm to 1,000,000 ppm (100%) CO2. The increased energy absorption just from doubling CO2 to 560 – 600 ppm is small.


Neat question. This is the reason I like to come to WUWT. Can’t wait for the answer.

Werner Brozek

I read that a 1% or 2% increase in cloud cover effectively nullifies a doubling of CO2. So if your solar panels effectively “decrease cloud cover” so to speak, then it would seem that the warming effect is larger than the cooling effect. Another area that could be looked at is how does the darkness or lightness of our highways affect global warming or cooling? Since the areas of our roads is much larger than the areas of solar panels, this may be another project to undertake next.

jai mitchell

Modifications to the surface albedo through the deployment of cool roofs and pavements
(reflective materials) and photovoltaic arrays (low reflection) have the potential to change
radiative forcing, surface temperatures, and regional weather patterns. In this work we
investigate the regional climate and radiative effects of modifying surface albedo to mimic
massive deployment of cool surfaces (roofs and pavements) and, separately, photovoltaic arrays
across the United States. We use a fully coupled regional climate model, the Weather Research
and Forecasting (WRF) model, to investigate feedbacks between surface albedo changes,
surface temperature, precipitation and average cloud cover. With the adoption of cool roofs and
pavements, domain-wide annual average outgoing radiation increased by 0.16 ± 0.03 W m−2
(mean ± 95% C.I.) and afternoon summertime temperature in urban locations was reduced by
0.11–0.53 “C, although some urban areas showed no statistically significant temperature
changes. In response to increased urban albedo, some rural locations showed summer afternoon
temperature increases of up to +0.27 “C and these regionswere correlated with less cloud cover
and lower precipitation. The emissions offset obtained by this increase in outgoing radiation is
calculated to be 3.3 ± 0.5 Gt CO2 (mean ± 95% C.I.). The hypothetical solar arrays were
designed to be able to produce one terawatt of peak energy and were located in the Mojave
Desert of California. To simulate the arrays, the desert surface albedo was darkened, causing
local afternoon temperature increases of up to +0.4 “C. Due to the solar arrays, local and
regional wind patterns within a 300 km radius were affected. Statistically significant but lower
magnitude changes to temperature and radiation could be seen across the domain due to the
introduction of the solar arrays. The addition of photovoltaic arrays caused no significant
change to summertime outgoing radiation when averaged over the full domain, as interannual
variation across the continent obscured more consistent local forcing.

–you’re welcome


Energy is neither created or destroyed, merely transported. Sure the desert area will be slightly cooler (because some energy has been removed locally) and some other place has been heated (because of the final thermal degradation) but the overall effect will be, by definition zero.


oebele bruinsma says:
December 10, 2013 at 11:17 am
A quick search delivered this: http://www.clca.columbia.edu/13_39th%20IEEE%20PVSC_%20VMF_YY_Heat%20Island%20Effect.pdf

Good find. From the abstract of the 2nd reference in that paper:
If photovoltaics (PV) are to contribute significantly to stabilizing the climate, they will need to be deployed on the scale of multiple terawatts. Installation of that much PV would cover substantial portions of the Earthʼs surface with dark-colored, sunlight-absorbing panels, reducing the Earth’s albedo. How much radiative forcing would result from this change in land use? How does this amount compare to the radiative forcing avoided by substituting PV for fossil fuels? This analysis uses a series of simple equations to compare the two effects and finds that substitution dominates; the avoided radiative forcing due to substitution of PV for fossil fuels is approximately 30 times larger than the forcing due to albedo modification. Sensitivity analysis, including discounting of future costs and benefits, identifies unfavorable yet plausible configurations in which the albedo effect substantially reduces the climatic benefits of PV. The value of PV as a climate mitigation option depends on how it is deployed, not just how much it is deployed—efficiency of PV systems and the carbon intensity of the substituted energy are particularly important.

Tom in Florida

What must also be considered is the efficiency decline of the solar panels as they become covered with desert sand.

I have not done or seen such a calculation, but I have seen articles suggesting we paint roads or rooftops white (which is probably dumb for other reasons) so we should identify greenies and ask them to crunch the numbers. I would pay to see figures on how many ounces of coal have been saved by all the US wind “farms”. Spiegel claimed that all wind, pv, biogas etc in Germany has not saved a single gram of CO2… I wonder… Can greenies convert grams of CO2 into ounces of coal?

Roger Dewhurst

These solar panels convert sunlight into heat, one way or another, with extraordinary efficiency. Thus I suggest that the OP is entirely correct in his supposition.

J Martin

Perhaps If the rate of road building has declined over recent years that may show a better correlation with temperatures than co2. /sarc
Perhaps the link sought is this;

Rob Dawg

Bright sunny places become dark electricity eventually to heat places. Why is there a question?

David, UK

I’m no professional scientist, but it seems obvious to me that solar panels – in the numbers that exist and are likely to exist in the foreseeable future – would have an infinitesimally tiny effect on global temperature – much like the much bigger UHI effect (said effect being local rather than global).
I suppose it is worthy of investigation if there is a possibility that this tiny effect is nevertheless greater than the effect that the panels are supposedly designed to mitigate.
If I’m talking BS, I apologise!


The local endothermic, cooling effect is compensated for when the produced energy is used.
“Has anybody ever actually sat down and calculated whether the CO2 greenhouse effect caused by a carbon-based power plant generating one megawatt of electricity is more or less than the warming effect caused by the lowering of the earth’s surface albedo from a black-panel solar installation with power output large enough to completely replace the carbon-burning power plant?”
Solar panels are expensive because a lot of energy is needed to create them. This energy comes in the form of electricity from coal power plants in China. Solar panels are made in China because coal-generated electricity is especially cheap over there. Meaning: Solar panels probably do not save any CO2 output but increase total CO2 output.
Economic inefficiency translates into energetic inefficiency. If solar panels were useful / energetically efficient, they would not need to be subsidized.
Production processes are constantly optimized; even without subsidies the cost of solar panels halfed once per decade due to ever decreasing energy and resource usage in production. It is not a natural law that solar panels increase the total CO2 output, it is just so that it is currently the case; and will likely not be the case anymore in one or two decades.
You will recognize the shift to that situation by the removal of the subsidies.

X Anonymous

ray pierre @ real climate did this in response to freakonomics, and found it to be a non issue. Levitt is at his uni….

If you take research such as this:
and calculate the expected energy for the solar cell, you may be able to get a fair understanding of the heat load transmitted to underlying surfaces, lost to the atmosphere, and converted to electricity or other conversions. In any case, there’s some data available.
It may also be worthwhile to investigate some building science research.

Brent Seufert

I’ve often contemplated a similar question, that is: Given the green’s penchant for complaining about the disruption of the environment by the Alberta oil sands mining (at least the part that is still above ground) how much land is being disrupted by these utility scale solar installations, and for how long, compared to the oil sands mining. Keep in mind, the oil sands mined land is recovered to natural state one cannot tell the difference from surrounding undisturbed lands.

If you’re in a very sunny place (Like Phoenix) I do think it makes sense to color a rooftop white, especially on a commercial building where the default choice would be a flat, black tar roof. The effect is small, but measurable. It’s all about albedo.
And let me assure you, you don’t want to be in a black car with a black interior in Phoenix in the summertime!

Chris B

There’s a similar problem with wind turbines altering normal surface wind patterns, with unintended consequences.

Bloke down the pub

My solar pv system is fitted on my roof which is covered with concrete tiles. As such I don’t expect there’d be much difference in the albedo.

puzzled @ ted palmer. Redistributing the same energy multiple times increases the heat content? Have we finally invented the perpetual motion device?


“solar installation with power output large enough to completely replace the carbon-burning power plant”
Fools errand.
It is not possible to have a solar array on Earth that ‘could replace a carbon-burning power plant’. Carbon-burning power plants operate 24 hours a day, a solar array on Earth can not.


Archaeos Pteryx says:
December 10, 2013 at 11:37 am
“I have not done or seen such a calculation, but I have seen articles suggesting we paint roads or rooftops white (which is probably dumb for other reasons) so we should identify greenies and ask them to crunch the numbers.”
One of Barack Hussein Obama’s henchmen, Steven Chu, proposed this, together with a mandated switch of the economy to glucose.
Somehow the Chicago machine never jumped on the glucose bandwagon though.
“Spiegel claimed that all wind, pv, biogas etc in Germany has not saved a single gram of CO2…”
Next, they probably demanded the government double its efforts. Would be typical German logic.
German Greens to this day demand that more efforts be made to slow down Climate Change to protect the drought-stricken German trees. (Don’t worry. Germany is the wettest country I know, just not in the mind of a German Green. Some like it moist.) But how do you slow down something that stopped 17 years ago? German Greens never explain.

Mark Bofill

Holy smokes, that’s true! Mainstream climate science tells us it’s not about atmospheric temperature, but about the energy budget of the system. If that energy isn’t being reflected back into space, you can bet your bottom dollar that it will end up as heat in the system at the end of the day. If solar ever got into use on a really massive scale…
…oh. Well, hypothetically speaking I guess it could eventually be an issue. 🙂

Reg Nelson

O/T, but I had a similar common sense question regarding “The Missing Heat”. The current theory is the that the planet is actually warming, but this increased warming is hidden in the ocean.
My question is this: If the planet is warming, and two thirds of the planet is covered by water, wouldn’t this lead to greater evaporation, which would 1) cause a decrease in SST (evaporative cooling) and 2) increase albedo — in the form of cloud cover?
Or would this process be constrained by humidity?

Hope you didn’t miss this little gem from last nights Colbert Report on the comedy channel;
“A Case For Climate Engineering” author David Keith discusses his proposal to fight climate change.


jai mitchell says:
December 10, 2013 at 11:34 am
“To simulate the arrays, the desert surface albedo was darkened, causing
local afternoon temperature increases of up to +0.4 “C.”
A sunny afternoon in a desert, and darkening the albedo leads to 0.4 deg C.
I don’t know what Barack Hussein Obama’s government employees computed there but it looks like they dropped two decimals. Maybe not that experienced with the Celsius scale?


Difficult question! Why bother? Sounds strange but ever considerer the question you ask for any construction material? Or for that mater any dark material. There is a totally different problem, CO2 dose nothing for warming so thats not the problem. Cleaner environment is more reasonable to look at. And yes solar panels are cleaner, not considering the production of it. But there is alway a but and yes its the same problems as whit wind mills they don’t work all the time. And the moment they not work cane be longer then the moments they work or the production is less. And less production means less power. You need your power so an conventional power plant needs to take over. There is your biggest problem a power plant needs to run beside the mill and the solar collector to make power. Shutting down a power plant is no option because it takes hours to even days to start up a plant. Even lowering the output of an power plant to say 25% of capacity cost more power to restart and work up to 100% and cost more fuel and so makes more CO2. In total you could say the effect is negative.


1000 MWh coalplant has side effect of cooling process steam to water . Operating efficiency is typically 30% so. 1000 MWh electric power is 30% total of app. 3400 MWh of heat.


This question has indeed been addressed, with “back of the envelope” calculations. If you can look past the usual nauseatingly arrogant and condescending tone of RealClimate, please see Ray Pierrehumbert’s open letter to Steve Levitt, as it is still an interesting read:

the proportion of the Earth covered by solar panels is vanishingly small, and will remain very small in the future. So the panels’ “extra heating” effect will pale in comparison to CO2-induced warming…assuming that warming is non-zero, of course. 😉
Reg, yes ocean warming leads to more evaporation, and more precipitation. This effect is already contained in the climate models. The effect on cloud formation, though, is a huge wild card. The reason why is that all ascending air in clouds has to be balanced by descending air, which is usually cloud-free. So an increase in the atmospheric hydrologic cycle can lead to either more or less cloudiness. I predict more (a negative feedback on warming), the models say less (positive feedback on warming)…but no one really knows for sure.

Regardless of the amount of overall heating or cooling effect of CO2 vs. industrial heating vs. heat island effect… A photoelectric solar panel generating an electrical current is generally going to provide that electrical current to a electrical machine [solid state circuits are machines of sorts]. These machines do work [the physics definition]. In the process of doing work, there are losses [heat] and changes in potential energy [work done]. So a very basic assumption would be that whatever possible heating effect is equal to the net electrical panel output [efficiency of the panel multiplied by the incident convertible photon energy] minus the net work done by the electrical current which should equal the losses [heating]…
eff * incident photon – electrical work done = delta t
This would be a very simplistic start, but it clearly shows that the overall heating effect of the PV unit plus the electrical system it’s connected to would not be directly equatable to the captured incident photon energy…
However, since determining the comparative heating or cooling effect of all industrial processes is a nearly impossible feat – I doubt any conclusion beyond the very simplistic statement above would be possible…
It’s also good to note that not all photons hitting the panel meet the requirements to be eligible for capture. This is in addition to the loss of the non-100% efficiency of the panel — which refers to the efficiency of the conversion of eligible photons only… not all photons are created equal…

Reg Nelson

Thank you, Dr. Spencer, for the explanation.


I had cause to do a similar calculation a while ago when I wondered whether carbon offset forestation would actually warm the earth by decreasing the albedo. The conclusion was that it actually cools the earth, under reasonable assumptions.
The electrical power generated by a 1 meter square panel is at most (1365/4)*0.2, as the panels are around 20% efficient. So how much coal is burnt to produce that much power? Wiki says that coal has about 32MJ/kg energy content. Every year the panels generate 0.2*(1365/4)*365*24*60*60 Joules of electric power, or about 2000 MJ. At reasonable efficiency, coal is producing 10MJ(e)/Kg. So, the Panel saves about 200Kg of coal (say Carbon) a year. There are roughly 900GT of Carbon in the atmosphere. so the proportionate rise in carbon is 200Kg/900GT=0.2T/9×10^11=2.2*10^-13. For the sake of argument the panel operates indefinitely, so that we will just see a linear rise in temperature forever, which I make around 6*10^-13 degrees C per year at a climate sensitivity of 2K/doubling.
Say, at worst, each square metre of panel takes the albedo from 70% reflection to 0% reflection. Crazily, this raises ‘radiative temperature’ by about 70C. Let’s just go with that for a minute, as it is obviously pessimistic in the sense that it minimizes the benefit of the panel. 70C over 1 square metre averages about 1.5*10^-13 K over the earth’s surface.
So there you have it. The panel will raise the remperature of the earth by 1.5*10^-13 K, and then cool it by 6*10^-13 per year thereafter. There is a net cooling after about 3 months.
Of course, that’s not the conclusion most of us were hoping for, so you might want to consider the energy used to make the panel. However, I’ve had enough fun for one night!

I read somewhere that the energy used for all the stages from mining the raw materials, processing, manufacturing, logistics/transportation, installation and the maintenance of a square mile solar panel array is 60-70% grater than the energy produced by the solar array over a 50 year period.

I have posed exactly the same question and have no ability
to determine the answer, but I did several calculations,
using published data about California solar farms.
The first thing to be aware of with solar power is that
rated capacity means nothing – those ratings usually
quoted are not realistic – they overstate outputs by 12%.
There are also inversion losses when going from DC to AC
There are also deterioration losses due to age , anywhere up
to 1% per year.
I suggested that someone with ability to obtain surface
temps from satellite measurements do so for some of the
California solar farms to see how they are affecting the
local climate.
I did , however, calculate the land required to produce gross
output equal to a modern nuclear plant of 1500 MW (90% capacity)
and estimated that such a solar farm required around 80,000 acres.

Those who are saying the effect should be zero due to conservation of energy are neglecting the point the questioner raises about solar panels changing albedo. Energy that would have been reflected back into space is now being turned into heat here on earth, whether on site or at some other location. The difference created by this change in albedo is what needs to be calculated.

Jim S

Won’t a large array of solar panels cast a shadow on the ground and thus induce cooling?

Conrad is right.

Jim, yes it casts a shadow, but as Conrad has alluded to, the reflective ground is now replaced by the black (absorptive) solar panel.