1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Clouds: Cool Off or Heat Up?

  1. Jul 9, 2011 #1

    Quick question: Let's say you have a large open oceanic area with no wind...
    The sun's radiation hits the water, and heats it, causing some to evaporate. The evaporation creates water vapor. Water mist/vapor, and the ocean itself, heat up the local air and cause an updraft. The vapor is carried upwards to lower pressure air, cooled, and forms clouds.

    The water vapor heated the local atmosphere up because it acts very strongly as a greenhouse agent and absorbs a lot of heat the air wouldn't have...
    But it also forms clouds, which reflect the sun's radiation.

    So does it cool off the area by reflecting heat, or does it heat the area up by absorbing it? Is the net effect to cool off the local air or to heat it up?

    Thanks in advance.
  2. jcsd
  3. Jul 9, 2011 #2
    The net effect can be either. It depends upon both the short-wave and long-wave albedos of the cloud. These are functions of cloud height, cloud elevation, cloud thickness, mean droplet size, time of day, time of year, what types of condensation nuclei are present, ice content--if any, and a host of other factors.

    The precise role of clouds in the Earth's heat budget is still poorly understood and the subject of vigorous debate. The general assumption is that they have a net cooling effect worldwide and yearlong.

    We are skirting a forbidden subject (climate change) here.
  4. Jul 9, 2011 #3
    I feel a bit uncomfortable being first to answer and admit that I really don't know.
    But also my answer might help someone to add more info.in their reply.
    Seems to me that the local air would warm. If you don't mind, let me work through it as follows.
    For one thing, I think the initial warming of a parcel of air near the hot earth surface causes expansion and does work on the environment - the atmosphere is pushed up a bit. Then as a parcel of air rises it cools at about 5.5 F / 1000 feet. I think this cooling tends to bring the water molecules closer, as heat is given off to the environment. In competition with this is decreasing pressure that comes with higher altitude. Lower pressure on the parcel tends to make the water molecules move further apart. I think this would tend to do work on the atmosphere - tends to take in heat or tends to cool the local air. But the effect of the pressure decrease only causes the dew point to decrease 1 F /1000 feet. Since clouds form in this example, I guess I would think the contraction of the parcel due to heat loss is greater than the expansion due to pressure. I think the expansion due to heat loss is overcome by contraction due to condensation to form clouds. Since contraction is more my guess is that there is a net warming effect on the local air.
    Now once clouds form, don't they reflect more of the sun's radiation into the atmosphere?
    Finally, can anyone tell me why my knee is starting to ache. It just started raining outside. It seems unbelievable how quick this response is, but I guess I am a believer.
    Last edited: Jul 9, 2011
  5. Jul 9, 2011 #4
    Hmm. I guess I'll be a bit more specific. I was a bit worried about scaring off forumites with the words "sci-fi", but I'm putting a lot of work into making this believable, and even a pleasure, to even the most highly educated reader :smile:.

    I'm just wondering what would happen on an Earth-like moon (significant N2/02 atmosphere, H20 vastly present), orbiting at about a zone where it receives a very similar amount of thermal radiation from it's star as Earth does, but with a much thicker atmosphere. The moon has large oceans and an atmosphere with roughly 2x the pressure as ours (2x as dense?); so this changes the scenario a little bit and de-relavates it from climate change.

    Is it more clear how this would effect a system where the atmospheric pressure is 2x as great and the water is far more abundant?
  6. Jul 9, 2011 #5
    In general, the surface of your moon would be cooler than that of the Earth. Denser clouds mean greater albedo. Greater albedo means lesser heat budget.
  7. Jul 9, 2011 #6


    User Avatar
    Gold Member

    It may have only just started raining now, but I'll bet the mercury has been falling for hours as the front moved in. Your knee is sensitive to atmo pressure, not moisture. :wink:
  8. Jul 11, 2011 #7


    User Avatar
    Gold Member

    I asked Jeff Kiehl (of Kiehl and Trenberth) this question a year ago and he replied that there have been large field campaigns that included ships, aircraft and satellites to measure the transport of water vapor, and that this is still a very great topic of research in the atmospheric sciences. He was willing to say however that the total effect of upper troposphere moisture (gas phase and solid) is to warm the planet, not cool it.

    On the other hand http://terra.nasa.gov/FactSheets/Clouds/ [Broken] says "Until recently, scientists did not know whether clouds had a net cooling or heating effect on global climate. Clouds reflect solar radiation, which tends to cool the climate, but they also help contain the energy that the Earth would otherwise emit to space, which tends to warm the climate. Measurements made in the 1980s by NASA's Earth Radiation Budget Experiment (ERBE) satellite demonstrated that clouds have a small net cooling effect on the current global climate."

    These aren't in conflict when one takes into account that a large fraction of Earth's surface has clear sky containing water vapor. If the skies were not cloudy or gray, the planet's albedo would be a lot less but water vapor would still trap radiation from the surface, which could make for even more warming that we have now. If on the other hand clouds covered the planet, so much sunlight would be reflected that the heat that did make it to the surface would hardly seem enough to keep the planet above freezing.

    Or so it might seem. However if the emissivity of the clouds at FIR wavelengths plus the reflectivity of the clouds at solar wavelengths equalled unity (true by Kirchhoff's law of radiation for any given wavelength and hence true to the extent that reflectivity and emissivity don't depend on wavelength, which of course they do) then the radiation not reflected by the clouds would also tend to be retained by the clouds, i.e. albedo and emissivity would cancel and just below the clouds the temperature should be 278.7 K (the temperature of a black body at Earth's distance from the Sun). GHGs above the clouds would increase this a fair bit.

    So with no cloud cover, water vapor in the atmosphere should heat the planet, while with 100% cloud cover and assuming as big a decrease in emissivity at FIR as the increase in reflectivity at solar wavelengths (reasonable since both reflection and emission from water droplets are mainly scattering phenomena) it should still heat the planet.

    It is a nice question what impact if any cloud cover has on global warming. With the above reasoning in mind I see no compelling apriori reason why it should have any significant impact, other than the evidence of the above-mentioned ERBE measurements.
    Last edited by a moderator: May 5, 2017
  9. Jul 12, 2011 #8
    Fairly good... Except I have one problem with that modeling:

    The mainly implies a little bit is missing from that sim, and while it's reasonable, it doesn't seem to match ERBE's observation.

    Although if I were actually doing some scientific application I would try to find out more definitively, my purpose in asking is to make a sci-fi setting believable to people who might otherwise notice physical impossibilities, so in terms of believability, it looks like I could go either way.

    Unless the situation drastically changes for a much thicker atmosphere. Lower gravity would mean the atmosphere is more expanded, so there would be far greater atmosphere above the clouds to absorb heat.

    Still, this is a very interesting topic for any application.

    Is it fair to assume the cloud cover would be much greater over the oceans if temperatures were higher?
    It would make sense that clouds would cool off the planet, and hotter temperatures would cause more to form, creating a negative feedback system.

    One thing I've never understood, is that if the planet operates on positive feedback systems, such as melting of the polar ice caps, what keeps the climate steady at all?

    It seems that to keep from going to the extreme all the time it would need more negative feedback systems, such as clouds forming as temperatures rise and reflect radiation, lessening the heat budget.

    But, Ack, though that's what I think I'm no expert.
  10. Jul 12, 2011 #9


    User Avatar
    Gold Member

    Although reducing gravity would expand the atmosphere, it would not change the number of molecules in it, which is the only factor relevant to Beer's Law governing absorption by the atmosphere of radiation at each wavelength. A less direct way of arriving at the same result is to use the formula g/c_p for dry adiabatic lapse rate: halving g halves the dry lapse rate while at the same time doubling the height of the atmosphere at any given pressure. The saturated adiabatic lapse rate is also linear in g, with 1/c_p replaced by a more complicated expression in half a dozen parameters, see the Wikipedia article on lapse rate. End result: no change in surface temperature as g is varied, though one can expect the altitude of clouds to vary inversely with g.

    That would be nice. The past hundred years have seen a one-degree rise in global temperature. However much cloud cover might have increased over that period, it would appear not to have been enough to prevent this rise. We know a lot more about global temperature change in the past century than about change in global cloud cover, see e.g.
    http://meteora.ucsd.edu/~jnorris/reprints/02_Norris_and_Slingo.pdf [Broken]

    Who says it's steady? The two temperature plots at
    show wide variations in planetary temperature over the past half billion years (including a likely snowball-Earth period around 450 Ma) and one-tenth of that (including the disastrous PETM at 55.8 Ma). In a complex system like Earth, with time constants ranging from decades to billions of years, "steady" is a relative concept.

    Feedback can drive a system over the linear portion of its response curve, but no response curve can be linear forever and eventually something saturates. The system may stop there if it's in equilibrium, or swing back if not. A complex system like a planet can be expected to exhibit a mixture of chaotic, oscillatory, and exponentially decaying behaviors. In principle one should be able to explain the past billion years of climate swings; in practice we are nowhere near being able to do so based on our present understanding of the energy flows in the many layers of the planet, from its core out to the top of the atmosphere.
    Last edited by a moderator: May 5, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook