Guest post by James Hansen.
In the 1970s, I made up my mind that rising CO2 levels were the cause of rising temperatures. Nobody really knew whether this was cause-and-effect (and if so, which causes which); correlation (a third factor causes both) or pure coincidence, but I decided to stake my career on it. I didn’t have much evidence, so I twisted the facts to fit my logic. I did a Diagonal Comparison, reversed cause and effect and then covered my tracks a bit by using technical jargon.
There is of course a Greenhouse Effect. Broadly speaking, all else being equal, the thicker the atmosphere, the warmer the hard surface will be. The atmosphere is like a blanket. It warms up and cools down slowly, it keeps the hard surface a bit cooler by day time and a lot warmer by night. Surely people can work this out intuitively? What the atmosphere’s constituent gases are is pretty irrelevant, all that really matters is their specific heat capacity (which is not much different for nitrogen, oxygen or CO2), ozone is a special case (that actually does have an effect). But ‘Greenhouse effect’ is neat shorthand, so I hijacked the term and made up my own explanation for it.
Luckily, CO2 levels and temperatures have continued rising since my 1981 essay, so everybody heralds me as a visionary genius and assumes that my logic was sound. I’m really surprised that nobody ever noticed the sleights of hand, not even those we deride as Climate Skeptics or even Science Deniers who are out to debunk it all.
Nonetheless, like any great showman or huckster at the end of his career, I now want to tell the world how I did it (not least because regular commenter Dinero posted a link to my original essay here, and our blog host saw straight through it and will blow the gaff if I don’t).
Start with some basic physics and maths
As any physicist knows, you can calculate the expected temperature of the visible surface of a planet by working out the average amount of solar radiation each m2 gets from the Sun (over the planet’s day); deducting the amount that is reflected back to space; dividing that by the Stefan-Boltzmann Constant (5.67/10^8); taking the fourth root of the result; and that’s your temperature in degrees K.
Here are the workings:
The Stefan-Boltzmann Constant is derived from observation and trial and error; you can tweak the figure for albedo a bit (nobody knows the exact values) and so of course it all stacks up and is a good match for measured temperatures.
Don’t forget that we take (white) cloud cover into account when calculating how much radiation is reflected straight back to space, so the end-result of our calculations is the expected temperature of the cloud cover (to the extent that the planet has cloud cover) at that altitude (i.e. the altitude of the visible surface)…
– The altitude of the visible surface of Venus is 65 km, the mid-point of the cloud cover, which is between 50km and 80 km altitude.
– The average altitude of the visible surface of Earth is 5 km altitude. Earth’s visible surface is two-thirds cloud cover, mainly in the upper half of the atmosphere – effective altitude 7.5 km – and one-third hard surface or open ocean – altitude zero – so the overall weighted average altitude of the visible surface is at an altitude of approx. 5 km.
– The average altitude of the visible surface of Mars is 1 km. Mars has very little cloud cover, but there are dust storms. On average, the visible surface is barely above the hard surface, call it 1 km for sake of argument.
Of course, simply referring to this as “the expected temperature of the visible surface, be that cloud cover or hard surface, at the average altitude of the visible surface” would have made the next logical leap too transparent, so I used the term “effective temperature” instead, a term whose meaning is not immediately obvious.
Trick 1 – the Diagonal Comparison
Having shown that I can plug numbers into equations and get the right answer, I then did a Diagonal Comparison. I compared the temperature of the visible surface (at a high altitude, now referred to as “effective temperature”) with the actual measured hard surface temperature (at zero altitude) and called the difference The Greenhouse Effect, which I explained with the hand wave “the excess… is the greenhouse effect of gases and clouds”.
It’s like saying that tomatoes are cheaper than potatoes, because 1 lb of tomatoes costs £1, but 5 lbs of potatoes costs £1.50!
Some more basic physics and maths
Having established the expected temperature at a certain altitude, it’s not difficult to estimate the temperature of the hard surface and to get your estimate to match up to measured temperatures.
The atmosphere gets cooler the higher up you go; which is another way of saying that it gets warmer as you descend to ground level. This is the lapse rate (which we call ‘adiabatic lapse rate’ to sound clever). The lapse rate is actually easy to work out. You just have to understand that the kinetic energy (heat) in air lower down is converted to potential energy as it rises (and vice versa); heat is ‘used’ to lift the air, this giving it more potential energy (and vice versa as it falls). You calculate the lapse rate as gravity ÷ specific heat capacity of the air.
How to work out the lapse rate:
[What tickles me is that I openly admitted that water vapour and clouds reduce the lapse rate by one-third. Now that my followers have been forced to admit that the effect of CO2 must be far lower than they want it to be, they are falling back on the Positive Feedback effect of water vapour (and ignoring the inevitable clouds, which more than cancel it out).]
How to estimate the expected hard surface temperature – you start with the estimated effective temperature from above; multiply the lapse rate by the altitude of the visible surface and add them together. Stands to reason – the thicker the atmosphere (and lower the hard surface is relative to the cloud cover), the higher the temperature. The end results are very close to measured temperatures (unsurprisingly, because you can tweak all the variables a bit):
Trick 2 – reversing cause and effect
Of course, using the proper chain of logic and cause and effect (as above) would have given the game away, so I reversed the logic and swapped cause and effect – I had to show that a higher hard surface temperature led to less radiation going into space (seriously, I’m surprised anybody fell for this).
To quote from my seminal work again, “The mean surface temperature [on Earth] is 288 K. The excess [‘effective temperature’ less hard surface temperature] is the greenhouse effect of gases and clouds, which cause the mean radiating level to be above the surface… the atmospheric composition of Mars, Earth and Venus lead to mean radiating levels of about 1 km, 6 km and 70 K and lapse rates of 5K/km, 5.5 K/km and 7 K/km*”
(* the current agreed lapse rates differ slightly, but I deserve credit for being pretty close)
If you read it through properly, what I actually did was to start with the temperature of the hard surface and subtract the lapse rate to arrive at the temperature at the altitude of the visible surface, which in real life is the first thing you work out, not the last. The figures 1 km, 5 km and 70 km are our starting points for the whole calculation, not the end result!
Also, it’s nonsense to work out the lapse rate by dividing the altitude of the visible surface by the difference in temperature between hard surface and visible surface (as I did). You can work out the lapse rate independently (see above) and then you multiply that by the altitude etc.
It’s like a green grocer saying, “Well, 5 lbs of potatoes cost £1.50, so the price for 1 lb must be 30p”, when in fact, he decides the price per lb first; you tell him how many lbs you want to buy; and then he works out the total price.
Covering my tracks
I referred to the altitude of the visible surface (i.e. the cloud cover, which anybody can understand) as “mean radiating level” and/or “the flux-weighted mean altitude of the emission to space”.
What also amazes me, looking back, is that instead of explaining how to estimate the temperature of the cloud cover based on solar radiation coming in (the common sense approach), I started the section headed Greenhosue Effect as follows “The effective radiating temperature of [a planet] is determined by the need for infrared emission from the planet to balance absorbed solar radiation”.
Well of course the temperature of the layer that absorbs radiation must be the same as the temperature of the layer than emits radiation – it’s the same layer! But it distracted people from the fact that I wasted two whole columns explaining how to work out how high the cloud layer is (to the extent that a planet has a cloud layer), which is easy – it’s the top half of the troposphere, on Earth as it is on Venus, the dust clouds on Mars are more at surface level.
You don’t believe me? Just read third column, page 1 and first column, page 2 of this.
Please don’t kick yourselves too hard, a whole generation of scientists and ‘activists’ fell for it as well!
With best wishes,