I am just curious if solar panels, because they absorb radiation, cool the atmosphere. If there were enough of them, would the temperature be different? I'm not a physicist or student.


marked as duplicate by Kyle Kanos, Jon Custer, John Rennie thermodynamics May 26 '17 at 5:04

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


So this is an interesting question I think, and the answer turns out to be 'no' and in fact that they might make the atmosphere slightly warmer (but such warming would be tiny compared to the effects of reducing carbon emissions of course).

One good way of thinking about this (see the comment by JMac) is that the aim of solar panels is to capture energy from the Sun which would otherwise at least partly escape to space by reflection. Essentially all of this captured energy ends up as heat, and thus the effect will be some warming.

So if you think simply about this: if there were no solar panels, then some radiation from the Sun would reach the ground, at which point two things happen: some of it gets reflected back, and some of it will heat the ground. In turn some of the reflected radiation will be absorbed by the atmosphere and will heat it, while some of it will escape to space where we can stop worrying about it. So there are two warming effects from the radiation:

  • the ground warms a bit, and will warm the atmosphere as a result;
  • some proportion of the reflected radiation is absorbed in the atmosphere and warms it.

(Modelling these phenomena is one of the easier sort of things climate models do: even here things can get exciting because the albedo -- reflectivity -- of the ground changes in complicated ways depending on what else is going on.)

In the presence of a solar panel, then the panel will absorb some proportion of the incoming radiation and turn it into electricity. So that radiation will then not reach the ground or be reflected, and there will be a slight cooling effect as a result.

But, wait, we don't just store that electricity in batteries: we use it to do work. And that work ends up as heat, almost entirely (some of that work might be running lights, and some of that light might escape to space thus not contributing to heating, but this will probably be a seriously small proportion.

So although the solar panels might directly cause a little cooling, the electricity they generate causes warming. And almost certainly that warming is greater than the immediate cooling (because more of the energy is turned into heat and less escapes to space as visible light than would have done by reflection in the absence of solar panels).

But again note that these effects are tiny: solar panels might cause some tiny amount of warming but it is utterly negligible compared to the carbon-induced warming that does not happen because of them.

One question worth asking is how tiny are the effects? To answer this we can do a simple-minded computation.

The radius of the Earth, $R\approx 6.4\times 10^6\,\mathrm{m}$, so the total surface area of the Earth, $A\approx 5.1\times 10^{14}\,\mathrm{m}^2$. And let's assume that the Earth is at $T=283\,\mathrm{K}$ (it is a little warmer in fact I think). And, finally, let's assume it's a perfect blackbody (which it isn't, but this is good enough). So the total power radiated from the Earth is

$$\sigma T^4\times A \approx 1.9\times 10^{17}\,\mathrm{W}$$

This is the energy it radiates to space by virtue of being warm (of course there is incoming radiation from the Sun which is keeping it warm, and some internal heating). Note that this is about $370\,\mathrm{Wm^{-2}}$, which is plausible: direct sunlight is about three times this.

How does this compare with human power usage? Well, according to Wikipedia (which has pointers to the source of this information), total human energy consumption in 2013 was about $5.7\times 10^{20}\,\mathrm{J}$, which corresponds to about $1.8\times 10^{13}\,\mathrm{W}$. So now we can calculate what the temperature difference is, which is:

$$\left(\frac{1.9\times 10^{17} + 1.8\times 10^{13}}{\sigma A}\right)^{1/4} - \left(\frac{1.9\times 10^{17}}{\sigma A}\right)^{1/4} \approx 0.007\,\mathrm{K}$$

In other words the total direct warming as a result of all human power use is somewhat under a hundredth of a degree. So the differences we are considering will be significantly smaller than that.

  • $\begingroup$ I'll add that this warming makes sense from a distilled simplistic standpoint. We are trying to capture solar energy. Compared to the normal operation of Earth, we want to try and minimize the solar energy that is escaping to get more efficiency. This seems like a pretty intuitive reason why it works. A big factor is how much less we can expect that to heat us compared to consumable fuel sources. Those are extracted from Earth directly, we are turning chemical bonds into heat; we aren't just absorbing more energy, we are releasing stored energy from the planet. Capturing the energy... $\endgroup$ – JMac May 25 '17 at 15:46
  • $\begingroup$ .. that is already hitting the planet is intuitively going to have a smaller effect. (but this all depends on the scales they are done at, and is in no way a thorough analysis) $\endgroup$ – JMac May 25 '17 at 15:47
  • $\begingroup$ @JMac: yes, of course, I feel silly now! I'll add a note. (But note that almost all of this is in the noise, including the direct warming from burning fossil fuels: everything is dwarfed by the indirect warming from greenhouse gases.) $\endgroup$ – tfb May 25 '17 at 16:00
  • $\begingroup$ Your last paragraph might be true, but is not supported by any facts you present. And it is not needed. If you burnt fossil fuels to achieve the same power output, you would have the heat of burning that fuel (times some efficiency factor that means you need additional heat) plus the heat of the energy dissipation. No "carbon emission" or greenhouse effect needs to be invoked. $\endgroup$ – Floris May 25 '17 at 16:08
  • $\begingroup$ @Floris: I was not trying to compare the direct effects of solar panels with the direct effects of fossil fuels, but with nothing. I was then just pointing out that both are tiny compared to the indirect effects (which, indeed I do not present facts to support, but they are reasonably well known!), given that I suspect this was behind the question. $\endgroup$ – tfb May 25 '17 at 16:22

Not the answer you're looking for? Browse other questions tagged or ask your own question.