I know that refrigerated air just moves heat around, and creates more by using fossil fuel energy in the process. If we wanted to actually reduce the amount of net heat energy in an area on the planet, and not just move it around, how?

How does the Earth do it? Is it all in radiating infrared back up through the atmosphere?

Can we humans help with this in a way that doesn't just need more carbon based energy? Can we have thermoelectric (Seebeck) generators powering infrared spotlights pointing out into space or something?

  • $\begingroup$ The Second Law of Thermodynamics states that any work you do on Earth will just make the Earth overall hotter, even if it makes portions of the Earth cooler. Of course, the Earth is losing heat constantly, but the problem is that it's gaining more heat (allegedly) than it's losing. A better approach might be to reduce the amount of heat (primarily solar radiation) the Earth receives. $\endgroup$
    – user967
    Commented Jun 25, 2018 at 0:19
  • $\begingroup$ If you reduced the amount of greenhouse gasses in the atmosphere enough, then temperatures would gradually decline. $\endgroup$
    – naught101
    Commented Jun 25, 2018 at 6:03
  • $\begingroup$ @naugt101: Small correction: Temperatures would instantly decline. A new radiative balance would be found at the speed of light. $\endgroup$ Commented Jun 25, 2018 at 11:42
  • $\begingroup$ We have the technology to create clouds which would increase the Earth's albedo (reflectivity) so less light would reach the surface. Of course, the heat generated in creating clouds might exceed the savings. $\endgroup$
    – user967
    Commented Jun 25, 2018 at 18:12
  • $\begingroup$ From my POV, the Greenhouse Effect shall be understood first. One thing is the noise we hear from the "supposed" main actors, and another is the main ones. CO2 or CH4 maintain some of the radiation, but if you look for cooling the planet, it is better not to receive that energy. As acting at the Sun is not an option, our best shot is clouds: high ones warm the Earth, and low ones, cool it. Human beings can create low ones. Energy is not an issue as our best "cloud generators" are power plants and huge cooling systems. Check link $\endgroup$ Commented Jan 31 at 9:18

2 Answers 2


On earth, heat can leave an object through conduction, convection or radiation.

In space, conduction and convection do not work because there is insufficient matter. So the only way for heat to leave the Earth as a whole is by radiation.

This works; the planet loses a lot of heat by radiation. However, at the moment we lose slightly less than we gain, hence global warming. You are correct that the only means of losing heat from the system is radiating it (as infra-red, because that's the wavelength that corresponds to the temperature of the earth's surface). However, while the surface of the planet radiates infra-red, greenhouse gases reflect or absorb a portion of it, reducing the amount that escapes the atmosphere.

So one way to help the planet to radiate heat is to reduce the concentration of greenhouse gases in the atmosphere.

More direct ways are tricky. My instinct is that things like setting up infrared spotlights pointing up would probably lead to producing more heat on earth than they get rid of, but I'm not totally sure of this - I don't see a fundamental reason that it couldn't work[1], if they could be made efficient enough; but I don't know whether that level of efficiency is remotely plausible.

[1] I don't see a conflict with the 2nd law; it's potentially a "moving heat around" mechanism that increases total entropy, but decreases that on earth.

  • $\begingroup$ Thank you, very concise. You're right it's still just moving heat around, but at least it's moving it into space. It's probably a better idea as you mentioned to use Seebeck generated power to capture carbon (or plant trees in deforested areas, which do it). $\endgroup$ Commented Jun 26, 2018 at 18:01
  • $\begingroup$ @zack you might want to focus less on Seebeck power; AIUI that works off a difference in temperature, and having too much heat doesn't mean we have that difference. Where we do have a difference at a large scale, other methods tend to be more suited to that scale (see e.g OTEC electricity generation) $\endgroup$ Commented Jun 26, 2018 at 21:36
  • $\begingroup$ OTEC sounds interesting, I'm looking into it now. Unfortunately as a practical DIY limitation I'm about as far inland as you can get (although on a large scale it sounds very workable), but you're right, lots of heat doesn't mean you can power something with heat exchanging if everything everywhere is hot, e.g. Phoenix, AZ. Even a concentrated solar turbine would need the ability to bleed off heat of some fluid, near impossible in an inferno such as the Sonora. $\endgroup$ Commented Jun 27, 2018 at 6:16

What is the best way to actually make the Earth lose heat?

TL;DR: Stop pumping so much carbon dioxide into the atmosphere.

The easiest approach in terms of human effort would be to let the Earth warm up a tiny bit. Thanks to the Stefan-Boltzmann law, a tiny increase in the Earth's effective temperature will easily rectify the ~0.6 watts/meter2 imbalance in the Earth's energy budget. If the Earth was a perfect black body radiating at an effective temperature of 252 kelvins, all that would be needed would be a 0.17° Celsius increase in the effective temperature to make that energy imbalance disappear. Easy, right?


Thanks to feedbacks, that small increase in the Earth's effective temperature requires a significantly larger increase in the Earth's surface temperature, and that's ignoring the fact that humanity is currently making the energy imbalance worse. Temperatures are currently rising and will continue to rise for a while due to the large amount of CO2 humanity has already pumped into the atmosphere.

Continuing to pump ever increasing amounts of CO2 into the atmosphere (the "business as usual" scenario) will result in a 4° to 9° C increase in the Earth's surface temperature by the end of the century, with the uncertainty of the increase depending largely on the climate sensitivity to a doubling in CO2 levels. Even the bottom of that range represents a disaster. The only way to avoid this disaster is to stop pumping so much CO2 into the atmosphere.

Can we have thermoelectric (Seebeck) generators powering infrared spotlights pointing out into space or something?

You wouldn't want to use infrared spotlights. You'd want the radiation to be in the visible range. The atmosphere is opaque in the infrared.

Playing "what if", suppose every last bit of energy currently consumed by humanity (18 terawatts) is used to power massive air conditioners, with lasers venting the generated heat into space. Suppose we somehow manage to get 17 times as much cooling as energy consumed out of the system, resulting in 300 terawatts of cooling. Dividing that by the Earth's surface area results in 0.6 watts per square meter -- i.e., the current energy imbalance.

This is of course utterly unrealistic. We aren't plowing or harvesting our fields, manufacturing products, transporting goods, driving to and from work, heating or cooling our houses. To make this unrealistic "what-if" scenario worse, the hydrocarbons used to produce that energy are exacerbating the energy imbalance.

The best way to make the Earth lose heat is to stop pumping so much carbon dioxide into the atmosphere.

  • $\begingroup$ Well, I agree, the best way to deal with it is to stop pumping CO2 and other greenhouse gases. Or to replant trees in deforested areas, which absorb CO2. But, my technological solution is an attempt to improve on such simplistic ones I've seen like "pump warm ocean water under the surface using thousands of floating wind turbines" (true example, they've got a website and CGI) which just banks heat immediately, potentially causing MUCH worse ecological damage (and hurricanes) when it finally pumps enough down there to change the undersea temps. This is a net loss of energy on Earth. I think. $\endgroup$ Commented Jun 26, 2018 at 18:07
  • 1
    $\begingroup$ I'm not sure that "not as crazy as this crazy idea" is much of a recommendation 😉 $\endgroup$ Commented Jun 26, 2018 at 21:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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