Gases are governed by the rule that:
PV = nRT
So, this would suggest that when pressure (P) is reduced, so is temperature (T). However this is only true of a fixed volume (V). But when air rises, it expands, so both its pressure goes down and its volume goes up. (n & R are constants)
So, another way that is used to attempt to explain why the temperature of the gas goes down, is to suggest that it is due to “work done”: because whilst it is expanding it is doing work, this is why it loses energy and therefore heat. That again sounds superficially attractive, but in actual fact when the parcel of air rises it is also gaining potential energy. And moreover, all the surrounding gas is also “doing work” on the other bits of gas around it. So, not only is the parcel of air we are considering doing work against other parcels, but other parcels are doing work against the parcel we are considering. So where is all this “work” going – it can only go into other parcels of air. There can be no net change in energy!
Lapse rate is just loss of potential energy
However, once we start thinking about potential energy we have the solution. Imagine if you will a balloon of gas slightly lighter than air (assume the balloon itself is negligible weight). Because the balloon is lighter, it will rise upward. But the gas within the balloon as a finite mass, so in order to make this rise, there has to be a gain in potential energy. But also as the balloon rises, the pressure drops, the balloon expands and the air within the balloon does work (force x distance = work). This work comes from the thermal pressure of the gas. So, as the air molecules expand they lose heat. And so this is where the potential energy is coming from. If we assume the balloon is big enough to contain 1kg of air, then the transfer of energy is as follows:
Potential energy = energy lost from air = temperature change x specific heat capacity
mgh = T Cp m
(where m is mass in kg, g is gravitational constant 9.8, h is height m, T is temperature C, Cp is specific heat capacity in Joules Centigrade-1 kg-1)
Simplifying we get:-
T/h = g/Cp
The specific heat capacity of air at -50 to 40C is 1.005 kJ Centigrade-1 kg-1
So lapse rate T/h = 9.8/1005
This is the lapse rate of dry air. (Moist air has a different lapse rate because it condenses out water droplets which effectively makes Cp much larger)
Breaking the “second law” that heat flows from hot to cold
The second law of thermodynamics is wrongly interpreted as meaning that heat MUST flow from hot to cold (that’s not what the actual “law” says). Therefore some argue that because air gets cooler as we get higher, that this must mean there is a heat flow.
This is false.
To explain why imagine the situation shown to the right. Here the molecules of the atmosphere are free to move up and down, but they cannot move from side to side. Imagine first a column without gravity (there is now really no “up” or “down”, but I will use these terms). An atom impacts a lower one at say height h1 and it moves “up” to point h2 where it hits another molecule. On average the energy it gets from the “lower” one will all go to the “upper” one. If we then assume some variation in the size of molecules to disperse energy through the column, eventually the energy will tend to equalise so that they are all effectively vibrating with the same temperature. This means there should be a uniform temperature “up” the column.
However, now imagine a column with gravity. Now if the molecule of weight m hits the lower one at h1 and then hits the higher one at h2, then because of the difference in height between h1 and h2 it will lose potential energy equivalent to:
m g (h2 – h1)
Therefore slightly less of the vibrational energy is given to the molecule above than was received from the molecule below. Therefore as we rise h meters up the column, the energy drops as:
Vibrational Energy drop = m g h
This vibrational energy is another name for heat, so, the temperature drop is equal to T Cp m. In other words, the heat conducted from the bottom of the gas upwards drops at the lapse rate. That is to say, heat energy is gradually lost both to conduction and potential energy as it is transferred up the column. So, in the absence of any other heat flows, the column will eventually stabilise with a lapse rate of g/Cp thus we have a thermal gradient which is stable with no heat flow.
However if we introduce other heat flows, the lapse rate changes the behaviour of the air.
Rising hot air
If air lower in the column is heated by e.g. incoming solar energy, its temperature will rise. As it is surrounded by air at the same pressure, and PV = nRT, that temperature rise will cause an expansion of the air. It will then become less dense than the surrounding air and so it tends to rise. However, so long as the air profile drops per the lapse rate, this packet of air will always be lighter than the surrounding air (another way to view it, is that it has more energy than the lapse rate) and so it will continue upward, either until it encounters a layer that has less energy than the lapse rate, or it loses the excess energy by conduction to the surrounding air, IR heat loss or condensing water droplets.
In theory, if there is no way to lose energy but conduction, it will continue upward effectively heating the whole column marginally until it replaces the very topmost molecules of air.
Infra-red breaking the lapse rate
However, whilst there is a lapse rate for conductive heat, so that the temperature drops as me move upward as heat is lost to potential energy, the same is not true of infra-red which is carried by electro-magnetic waves which have no weight and so energy is not needed to rise up against gravity.
So infra-red energy will tend to equalise the temperature causing the higher levels to warm more than the lapse rate and the lower levels to cool to be lower than the lapse rate. When this happens, the lower air becomes denser than the lapse rate and less likely to rise and conversely the upper air becomes less dense than the lapse rate and more likely to float higher up. So, now, even if some air is slightly heated, it may initially rise and expand, but because the IR has broken the lapse rate profile, it quickly reaches a layer of air which is less dense at that temperature.
And the more IR is a significant conveyor of heat, the more likely it is that IR will break the lapse curve.
So, as long as there is heating of the lower atmosphere large enough to cause rising hot air to carry enough heat to offset the infra-red transfer, the atmosphere will retain the lapse rate. However, this only happens when the atmosphere is dense enough to block the movement of IR. Eventually as we rise up through the atmosphere, the rate of transfer of IR become greater than the potential to transfer heat through convection and the lapse rate breaks down and the IR heat transfer breaks the lapse rate, equalises the thermal gradient and then effectively blocks upward movement.
This is effectively happens at the top of the troposphere (or at least once the air loses its moisture).
But also there is IR heat loss to space. This is stronger the higher we go, and so this should tend to reduce the temperature of the upper atmosphere. But then other effects come into play and it is far too complicated to work out what is happening from theory.
Greenhouse gases and lapse
From this we can conclude:
- The lapse rate is the rate of loss of energy to potential energy as a packet of air rises.
- The lapse rate will occur naturally in a transparent atmosphere which has no interaction with IR or visible light.
- If lower layers of the atmosphere are heated, they tend to rise until they encounter a layer at a lower temperature than the lapse rate. So, this mechanism tends to cause convection from the ground up to where the lapse rate breaks down.
- The lapse rate breaks down because of IR heat loss (or gain).
Far from the lapse rate being a result of IR interactive gases (aka greenhouse gases), as has been suggested, greenhouse gases tend to break down the lapse rate. In the bulk of the atmosphere, they tend to move the thermal gradient away from the lapse rate and toward an isothermal region which tends to repress convection, but near the top the IR heat loss to space tends to make the upper layers with more of a “window” to the cold of space, colder, this tends to make convection easier. So, what creates the lapse rate is the loss of potential energy. And far from greenhouse gases creating the lapse rate, the effect of “greenhouse” gases is complicated causing both and increase in gradient as IR is lost to space AND a decrease in gradient as IR tends to equalise the temperature within the atmosphere.
Your reasoning is refreshingly sensible and correct sir. The lapse rate is indeed a result of gravitational containment and does not require radiation for its existence.
I would like to add two things though. Firstly, photons carry energy. Energy has effective mass and all of matter and energy feel gravity. Harvard Tower Experiment; photons travelling 27m vertically in Earth’s gravity were red-shifted as they gained potential energy and blue-shifted travelling downwards. Hence all photons carrying energy in a gravity field must pay some price for gaining altitude in a gravity field. Vertical photon exchanges must to some degree carry information about the presence of the gravity field.
Secondly, data from Earth and Venus lapses show no change (reduction) in the lapse rate where opacity is highest, other than following the changing of Cp with composition and temperature. So, although logically correct, the idea that radiative exchange ‘should’ act in a manner to reduce the lapse rate, the data shows no evidence of their presence other than they exists and therefore must conform to and maintain the existing gradient.
The dry lapse rate is supposedly adiabatic, meaning there is no loss or gain of energy. A mole of gas at 1 bar pressure and at room temperature has the same amount of energy as a mole of gas at 35,000 ft altitude and -50C. The only difference is the volume of the gas. Take that mole of gas at 35,000 feet back down to sea level, and it will be at room temperature again. No energy has been gained or lost in the process. Temperature is a measure of how concentrated the energy is, not a measure of the total amount of energy.
If it’s true that the lapse rate is totally adiabetic, then you have to conclude that the atmosphere does not lose any energy to outer space until you get well above the troposphere. All of the temperature differences of increasing altitude and lower pressure are due to the established gas laws, not because the gasses are losing energy to outer space. The atmosphere doesn’t radiate significant energy to outer space until you reach almost the top of the atmosphere.
Well put. I use the analogy of looking into a wood from the edge … the only trees you see are those closest to you. Likewise, looking into the atmosphere, the IR molecules we can “see” – the ones that are emitting IR to space – are the ones closest to us. And when the wood is thick (IR doesn’t travel far), then doubling the density of trees just means we can see slightly less far into the wood. So, beyond a certain density – for all reasonable purposes, the edge of the wood is the practical limit of our observation. Likewise, in most IR bands, the practical emission point is “at the top” of the atmosphere as you suggest.