Some readers may recall that back in June 2010 I tried to build a simple computer simulation model of the Earth and its atmosphere. I started out looking at a square metre on the equator of an airless planet, being warmed throughout the day by sunlight, and radiating heat back into space and conducting it into and out of the ground.
The programme chugged round and round, working out the heat gain from the sun as the sun angle changed every few minutes, and the surface temperature Ts at which this solar heat gain was balanced by radiative heat loss to space, and conductive heat loss into the ground.
After that, I added an atmosphere on top, and a few reflective clouds, and it all started to get complicated as I tried to figure out how the atmosphere might absorb and re-radiate heat from the Sun and the Earth. I ended up studying photon football, and going round and round in circles.
Yesterday, I read about a couple of physicists who’d adopted a similar approach, only using the kind of differential calculus that I can’t do (and which is why I always build simulation models of everything, because the maths is much easier). Like me, they’d started out working out what the mean temperature of an airless planet Earth would be, and came to the conclusion that the usual estimate of 255 degrees Kelvin (-18 degrees C) was way too high. Their figure was about 155 K – about 100 degrees lower than the generally accepted figure. It would be a revolution in physics if they were right. But in the comments under the WUWT presentation of it, a lot of people were very sceptical, and saying that they’d got the maths wrong and the physics wrong and maybe pretty much everything else wrong too.
It wondered if I could maybe adapt my little model to look not just at a single square metre of ground on the equator of the Earth, but at a whole set of square metres of clay running from the equator to the pole, and that way make my own estimate of the mean planet temperature.
Nobody else, as far as I could see, had built a simulation model. They were all doing it with heavy duty maths.
So I worked out a simple approximation for the reduced solar heat gain with increasing latitude, and after a few glitches got the modified code working. It took a few months for the temperatures to settle into a stable cycle.
On the right are my results. The green area shows the surface temperatures throughout the day, ranging from the equator to the pole. And the mean temperature of the airless planet over the whole day came out at 247 degrees K.
Which is only about 8 degrees lower than the generally accepted figure of 255 K, and not too bad for a bit of a rough approximation. And it’s certainly nowhere near 100 degrees lower.
So I reckon they’ve probably got the maths or the physics wrong too.
But I’m glad they published their results in a public forum. This is how I think science should be done, with people publishing things and other people criticizing it, all freely and openly, and not locked away in university laboratories only talking to like-minded people, and refusing to reveal their data and their methods.
I’ll post a little comment on the WUWT forum, linking to this page of mine, and that will be my rather belated contribution to the debate.
And maybe then I’ll try to polish up the code a bit, and stick an atmosphere back on top of it again, and see if that warms the surface a bit (it needs to be 30-40 degrees warmer to match actual earth temperatures), or whether it just ends up with photons bouncing all over the place again.