So, I'm attempting to calculate the habitable band of a solar system based off of a star's luminosity and radius (currently discounting multistar systems).
This article quotes a lot of factors, but notes at the bottom that a band can be calculated by finding the expected temperature at a given distance from the star based on luminosity: if the
(
Read more... )
However, I believe that the habitable band IS calculable, just as the Roche limit is, and the gravitic neighborhood perturbation that indicates a satelite's destiny as a planet or a dwarf is.
Hell, I may have to randomly assign vulcanism and surface qualities, but once I do average planetary albedo pre-water vapor is calculable, which means I can get an average planetary temperature, and from there I can find out what atmosphere it'll hold, which will tell me if temperature will spiral up, down, or form a nice maintaining feedback due to water vapor and CO2 released by said vulcanism.
So, if calculating the habitable band is too hard, it means I haven't broken down the problem enough. This post was an appeal to peers to see if I was missing something obvious before I took steps to correct for errors I'm seeing, mainly.
I'll save rolling dice for determining which planets have what elements (within a standard distribution of galaxy average), not for when modeling something is "hard". =P
Reply
Leave a comment