Greenhouse theories compared

There are now several different models trying to explain or calculate the Greenhouse effect. None of them are perfect. The best models are too complex and not even understood by some top scientists who have failed to spot the importance of pressure. Others like the “standard” model are so dumbed down as to be laughable. And clearly my own model (hereafter called Haseler’s lapse rate model) is no better because although physically much much more realistic than most models, it is clearly too difficult for many to understand.

So without going into detail, I thought I would try to describe the differences and the pros and cons of each model.

Layer-by-layer Numerical Model

The best model of the atmosphere is a detailed layer by layer model of the atmosphere whereby for each layer, for every frequency, for all molecules types, the net incomings and outgoing IR energy and heat flow from convection, etc.


This will undoubtedly produce the most realistic model of the atmosphere


Extremely difficult to understand to such an extent that some features of the atmosphere like pressure effects are not obvious even to some experts.

Greenhouse model

The greenhouse model says that greenhouse molecules like CO2 act like a greenhouse and trap heat.


They do trap heat


Greenhouses trap heat largely because they prevent the heated air escaping. This is entirely irrelevant when comparing a planet with an atmosphere and one without, because heat can’t escape from either of them by convection.

IR Heat Trapping Model

The standard way to describe the “Greenhouse” effect is that “Greenhouse” gases are transparent to incoming sunlight but opaque to outgoing IR. Thus they let heat get to the surface, but block it being emitted.


This model is extremely simple to understand and explains how a transparent gas could cause warming by using the analogy of a greenhouse.


There are so many it’s difficult to go through them all. But a few are:

  1. Greenhouses and Greenhouse gases work by entirely different principles.
  2. Greenhouse gases not only absorb radiation, but emit radiation IN EQUAL MEASURE. So, on it’s own the theory is totally unscientific.
  3. This model doesn’t even acknowledge the key role of convection
  4. It ignores the importance of the gas temperature.
  5. Because the model is inherently unstable it implies even a minuscule increase or decrease in “greenhouse” gases would result in runaway warming or cooling.

Ned Nikolov’ Pressure “model”

Ned’s “model” isn’t so much a model as the accidental discovery of the importance of pressure in creating the greenhouse effect, with a lot of cherry picking to enable a ridiculous conclusion that is physically impossible: that heat is not lost from the atmosphere by IR.


It demonstrates the importance of pressure in the greenhouse effects. In effect it is the same as “Haseler’s lapse rate” model (below), except that Ned’s model inherently assumes the temperature of the very topmost molecule is fixed. It works because on most planets, there is very little difference between the “top” of the atmosphere as a physical entity, and the “top” in terms of the height from which radiation leaves the planet.


Whilst it demonstrates one aspect of the Greenhouse effect, otherwise it has no scientific or mathematical validity and this important work is unfortunately, doomed to be either ignored or worse laughed at. In particular it has no validity at all for a largely IR transparent atmosphere.

Haseler’s “lapse rate” model

This model focusses on the radiation leaving the earth and it uses the simple approximation that for stability, the average temperature of the radiation from the “topmost” molecules/surfaces must be that of a similarly situated blackbody. It then uses the principle of the “lapse rate”, the drop in temperature air rises in the atmosphere, to explain the temperature difference between the surface and the “top” of the atmosphere.


On a planet like the earth with a relatively dense atmosphere, the emissions can be approximated by a “height”. It also highlights the importance of convection & pressure in creating the greenhouse temperature. Furthermore, it treats greenhouse gases as both emitters and absorbers of IR. In summary it is far more physically realistic than other models. Also, because the change of height of the “top” of atmosphere can be estimated, it provides pseudo-mathematical predictions of the effect of changing greenhouse gas concentration & pressure – by equating this to a change in radiant height (and therefore temperature) of the outgoing IR.

The model can also be used on liquid “atmospheres” (except for it to be valid, there needs to be a source of internal heating to drive the lapse rate).


  1. What appears to be a very simple model seems to be too complex for many people.
  2. It glosses over the mechanisms driving the lapse rate and the importance of water in that cycle. The “lapse rate” is used as if it is a constant whereas in reality it is very complex.
  3. It ignores heat trapping, without which there would be no outgoing flow and so no lapse rate. (See below)
  4. Others (see below)

Haseler’s more complex “lapse rate” model

One of the big problems with the simple lapse rate model is that a significant amount of radiation exits the planet from the surface of the earth. This doesn’t fit with the simple model which assumes the “topmost” molecules can be approximated as a single emitting layer in the atmosphere that moves up or down depending on the concentration of greenhouse IR emitters like CO2, water vapour or cloud droplets.


More realistic


Even more difficult to understand and less useful at qualitative predictions

Heat trapping

In the “heat trapping” model – which is the model usually described as the “Greenhouse effect”, heat trapping is correctly described as causing the greenhouse effect, but this model is very wrong when it suggests that the scale of the greenhouse effect is determined by heat trapping.

One way to understand the important of heat trapping is to use the analogy of a pile of sand:

PileOfSandIt is well known that if sand is poured onto one spot, that it will create a pile with a characteristic angle. This angle will remain the same if either:

  1. More sand is piled on the top and allowed to fall down the sides
  2. Sand is removed from the base and sand is caused to fall down from the top.

The lapse rate is equivalent to the angle of the sand. Like the sand slope, the atmosphere will have a stable lapse rate slope if:

  1. More heat is added to the bottom of the atmosphere
  2. Heat is lost from the top of the atmosphere
  3. AND … that the rate of heat flow up through the atmosphere by convection is greater than the rate of redistribution of heat that tends to cause the atmosphere to thermally equalise at one temperature.

It is this last one where the heat trapping mechanism is so important. Without some kind of power source to drive the convection currents (effectively pouring sand on the top of a sand hill), the atmospheric temperature will stabilise. But likewise without heat loss from the top of the atmosphere the lapse rate will also disappear (and this is where the normal “Greenhouse model” and Ned’s models completely fail and where Haseler’s model is so good.

Where’s the Pressure Effect in the standard Model

I got asked this question by an atmospheric physicist and the answer is simple: the pressure is in their model, but they describe it in terms of a height of the atmosphere. In other words, their atmosphere has a tropopause at around 10km or 0.1bar. Pressure and height are interchangeable and so whilst this parameter is expressed as a height it is also a measure of pressure. And if they were modelling an atmosphere where the tropopause (~0.1bar) was at 100km, then they would have a 100km high tropopause.

What’s wrong with measuring heat flux at the surface

There’s nothing wrong with measuring the heat flux near the surface, but it’s pretty pointless and adds an unnecessary layer of complexity to any model.

The reason my model works is because most of the IR radiation is “locked up” in the atmosphere and cannot escape until it gets to around the 0.1bar level. So, yes you can estimate the heat flows throughout the atmosphere, but in reality this whole flow can be expressed in the simple temperature “slope” of the lapse rate. And  it is only at the “top” of the atmosphere (where the lapse rate breaks down = tropopause) that you get anything interesting happening. And because it’s because you get a discontinuity in the lapse rate or convection heat flow, that you then know that the entire heat flow must be radiation**. And because we know the radiation flow must be that of a black body, we then know that the heat flow in the region dominated by convection heat flow must be the same. However, since all we are interested in is the temperature drop from the “top” to the ground, and this is given by the lapse rate, we don’t need to know the detailed breakdown of the heat flow in the region dominated by the lapse rate.

**Haseler’s model presumes no convection above the tropopause. This is a good approximation, but it contains a small error as there is a small amount of convection through the stratosphere & above – but its small enough to ignore except in a very detailed model.

This entry was posted in Climate. Bookmark the permalink.

13 Responses to Greenhouse theories compared

  1. Pablo says:

    An atmosphere without water vapour or other radiating gases will still have the lapse rate of 9.8C/km due to the acceleration of gravity at 9.8m/sec/sec.
    The presence of water vapour enables surplus energy at the surface to be radiated to its compatriot molecules aloft to cool the surface air and warm the higher levels thus shrinking the thermal gradient.
    This means that stable air is usually 3.3C/km warmer than it should be for its height (an increase of potential temperature of 3.3C/km) and has a lapse rate of 6.5C. So at a 10km high tropopause the air would be 16.5C warmer and the surface air 16.5C cooler by intermolecular radiation. ( NASA’s 33ºC warming?)
    Mixing by turbulence and wind in the boundary layer during day time always corrects that discrepancy back to the gravitational lapse rate so usually up to the cloud base there is no potential temperature increase with height (ie. the 9.8ºC/km lapse rate).

  2. oldbrew says:

    ‘Pressure and height are interchangeable and so whilst this parameter is expressed as a height it is also a measure of pressure.’

    Yes — max height (equator) = max temp and min height (poles) = min temp.

    • Scottish-Sceptic says:

      But the “blackbody” temperature at the poles and equator is different. And I bet there are changes in lapse rate as well. But in theory the ground temperature should be determined by lapse rate, “radiation height (which is affected by cloud cover & greenhouse gases) and blackbody temperature.

  3. HOw does one get a password for this site? The first post is password protected.


    • Scottish-Sceptic says:

      The password is “Crap” because it’s a crap article which never got past a very very early draft. It’s there only because it contained the source of a graph I was discussing on twitter and a video of the conference where it was presented.

  4. oldbrew says:

    Re: ‘max height (equator) = max temp and min height (poles) = min temp.’
    Mercury, Venus, Mars – same result. Venus poles are even colder than Earth…

    ESA Finds a Frigid Surprise Hiding at Venus’ Poles
    Venus may be boiling hot, but its poles are very, very cold.

    Logically the declining tropopause height between the equator and the poles implies a declining pressure at the surface as you go towards a pole, due to less atmosphere above.

    Obviously the angle of the Sun gets shallower too, which is why the equator-pole tropopause height gradient exists in the first place.

    • Scottish-Sceptic says:

      Wow! My first thought is to question the accuracy of the measurement. However as it’s more interesting to assume the finding is correct I will give it some thought.

      Just a quick thought comes to mind … is it possible there is some kind of convection current, whereby warmed atmosphere at the equator is then cooled at the upper surface by IR emission and then it moves to the poles where it is subducted. This would mean the poles don’t have the normal lapse rate gradient and hence the colder area.

      • oldbrew says:

        Don’t know, but there’s also ‘a huge, double-eye atmospheric vortex over the south pole’ to ponder.

        Plus the Venus atmosphere rotates in ~4 days, compared to 243 days for the planet itself.

        ESA: ‘High velocity winds are known to spin westwards around the planet, and to take only four days to complete a rotation. This ‘super-rotation’, combined with the natural recycling of hot air in the atmosphere, would induce the formation of a vortex structure over each pole. But why two vortexes?’

        • Scottish-Sceptic says:

          You know that point in Jurassic park where the girl looks at the screen and says: “This is a Unix system I know how to use this”. This is just a simple question of rotational flows – I know how to do this – at least I would if I hadn’t been drinking 80% proof Austrian rum.

        • Scottish-Sceptic says:

          As my wife tries to discuss who bought the Happy NewYear pigs I’m thinking about this and how to apply to poles of Venus (which was not a Euphemism)

        • Scottish-Sceptic says:

          Ok, here’s a rough line of thought. The rotation is affected by the Coriolis force, but there is also a hydrostatic force which will tend to cause movement toward the poles. Thus two sources of rotation – which may interact to give two vortices – perhaps one ascending and the other descending. That’s as good as I can do NewYear’s eve after a few drinks.

Leave a Reply to Scottish-Sceptic Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>