[Seany (OP)]

Aren't we getting closer to the sun? OR is it because of the heat which is absorbed from the Earth's ground which is keeping us warm?

[another_someone]

The Sun is approximately 93 million miles away, so being 5 miles up in the air is not sufficiently closer to make much difference.

In part, it is as you say, the heat of the Earth that warms the air at ground level; but as that warm air rises, so it expands (as the air pressure at higher altitude is less than at ground level), causing adiabatic cooling.

[Seany (OP)]

Hmm OK.. So even though warm air rises.. it expands.. and thus.. I'm getting confused

[another_someone]

Hmm OK.. So even though warm air rises.. it expands.. and thus.. I'm getting confused

Think of the classic bicycle pump - as you compress the pump, the air heats up - but conversely, as you expand air in the pump, so the air will cool.  This is also how most refrigerators work - they compress the working fluid on the outside of the refrigerator, which then gets hot, but loses the heat to the surrounding air, and then take the working fluid inside the refrigerator, let it expand, thus cooling down, and sucking heat out of the refrigerator box.

At ground level, air pressure is about 1Kg cm^-2.  At 10,000 feet, the pressure is about 70% of the pressure at ground level, and so the so all things being equal, the temperature (measured in absolute terms - i.e. in Kelvin) would have dropped 30% of the temperature at ground level.  In fact, all things are not equal, so the temperature at 10,000 has only dropped about 7% (if ground temperature is about 15C, then the temperature at 10,000 feet would be about -5C).

At 30,000 feet, air pressure is less than ¹/₃ that, so all things being equal, the temperature would have dropped to ¹/₃ the temperature at ground level.

If this adiabatic cooling did not take place, you would not have snow on the mountain tops when the valleys below may be baking hot.

[Seany (OP)]

Ah.. OK... Thanks George []

[chris]

We've discussed this, and answered it pretty thoroughly, in a previous thread. Here's the link to that discussion:

http://www.thenakedscientists.com/forum/index.php?topic=6239.msg63362#msg63362

Chris

[lyner]

Actually, the 'rising so it expands so it gets cooler ' argument, whilst correct in detail, doesn't really say WHY. I love ENERGY as a way of explaining things.
How about this?
A mass of warm air has a bit more internal energy (heat) than the surrounding air. It is displaced by the  cooler air and work is done in raising it to a height as it  expands.  (pressure times volume change) The energy for this work comes from the internal energy  in the mass of air so its temperature will drop. The air will continue to rise until  its temperature is the same as the surrounding air- no more upthrust. At this height, the Gravitational Potential Energy it has gained will have come from the thermal energy that has been lost.
BUT it is not necessary for any convection to take place for there to be a temperature gradient. You would have a similar situation even if were to stop convection currents - say you had a VERY low density glass layer around the Earth.
The atmosphere is almost transparent to a lot of the (particularly visible) radiation that arrives from the Sun - so the surface  of the Earth gets most of it, absorbs it and its temperature is, consequentially, high (say 300K).
Objects (like soil)  at 300K radiate a significant amount of Infra Red, which  then can be absorbed by the atmospheric gases. The gases will also re-radiate IR.
For a body of gas at any height, there will be IR  both from above and below it. This body of gas will establish a temperature which makes it in thermal equilibrium  with its surroundings i.e. it is losing the same amount of energy that it is receiving. The higher altitude it is, the less energy it will be getting - so the lower its temperature will be.
I have a feeling that the average temperature gradient would be the same with or without convection. Any ideas?

[another_someone]

I have a feeling that the average temperature gradient would be the same with or without convection. Any ideas?

No, I don't think it can be, for two reasons.

Firstly, a system that has convection currents must have higher entropy than a system that is purely dependent upon radiation and conduction (i.e. it will have more large scale order in that you will have the various convection currents creating structure within the atmosphere).  I would guess that this would not only be extracting energy from the thermal gradient, but also from the Earth's rotation.

Secondly, the convection currents also provide transport for water and dust which would not happen in the event of a pure radiation and conduction system.

I accept that these are real world issue, and it does not necessarily address what a theoretically ideal system on a non-revolving planet, with 360° sunshine, that has a uniform hard smooth surface, might do.  On the other hand, such a supersymetric environment would have difficulty creating convection currents in the first place.

[lyner]

That sounds fair to me.
The easiest model for a planet is a solid  black surface with no atmosphere.
After that, it just gets more and more complicated!
btw, the discussion which involves adiabatic changes in the rising air  seems to ignore adiabatic changes in the air which is falling, to displace it.

[another_someone]

btw, the discussion which involves adiabatic changes in the rising air  seems to ignore adiabatic changes in the air which is falling, to displace it.

That would certainly affect the speed of heat transport, but would it effect the temperature gradient?  After all, if air at one level of the atmosphere is 30% less dense than air in another part, the temperature gradients (only taking into account adiabatic heating) should only concern itself with the relative pressures, not whether the air went up or down when changing pressure).