<Unfortunately this isn't what you want. What you want are pressure versus temperature curve fits for water and carbon dioxide phase transition (the mathematical basis for the diagrams).

how silly of me, of course, the diagrams themselves are based on the peicewise equations. . . not the other way around.

This information can be found in thermodynamic reference books at a university research library (engineering library). These curve fits are calculated through statistical mechanics using a Gibbs (or Helmhotz) free energy solver. I have one of these routines in my atmospheric entry simulation code. This routine is based upon the Gordon and McBride code. The Gordon and McBride FORTRAN source code can be downloaded from the NASA Glenn Center.

yikes, fortran *shudder*.

Get on NASA Glenn's search engine and lookup Gordon and McBride then follow the links. The Gordon and McBride code is among the best but I do NOT recommend your getting this code (it would be like killing an ant with a hand grenade).

very interesting though I wonder how it would run on my 750 Athlon. But from what you wrote it sounds like I would have to learn to adapt it to generate the peicewise equations which is probably way over my head right now. It makes sense to jut use the already generated one form the text book rather than remodeling them.


You're screwed. Planetary emissivity is strongly based upon cloud cover. Cloud cover is determined through weather simulation. As you know, weather simulation is a hard problem.

well I can say I sorta saw this coming. I was planning on just calculating info for things up to the point of where an atmosphere doesn't escape, then just drop it with a big note that says 'not a dirty iceball like the others'

The question you didn't ask:



How do I calculate atmospheric heat storage through water vapor, carbon dioxide and methane green house effects?



when I origionally started on this I got a lot of the info from
http://cix.co.uk/~cicarage/planets/science.html he just gets users to enter in a random greenhouse effect, and he's a meteorologist ;-)

I like his disclaimer:
For better results, try modifying our comprehensive Unified Climate/Forecasting Model. It is only 300,000 lines of Fortran, with 200 pages of input panels and 100+ developers. It does run on a high-end workstation, but we tend to use a 880 processor, 128 Gb Cray T3E, so I can't it running on this Web site in the near future :-)

Very hard to do. Many Ph.D.s have been devoted to solving this problem. Simply calculating the surface temperature of a planet through radiative equilibrium using the Stefan-Boltzman law will only work on planets without atmospheres, i.e. Moon, Mercury, asteroids, etc. Do the naive calculation for the Earth and you'll find the surface temperature is way to low.

-21 degrees celsius! :-D (from my notes)

Why? Because of the water vapor green house effect due to the ocean. Water vapor is a neat temperature regulator. If the atmosphere is clear then heat is absorbed through water vapor. However if too much heat is absorbed then clouds form from the ocean thus blocking out the sunlight (self regulating). Also the ocean and ice caps act as nice thermal reservoirs.

interestingly the 'starform' package based on the accrete source code does return a cloud cover percentage. I have no idea what they ar basing this on, can't be part of the origional accrete source, but I will look at it closer tomorrow and see. it can't be very good. probably good for a hardscifi computer game but not much else.

Isn't nature marvellous?

I think Nature has secured the careers of scientists for many centuries to come ;-)

-Justin Kelly