I have had another thought. I think I read somewhere that you set the sun for a particular time of day and time of year and it gets the right angle. If you know the time of year have you thought about using that colour control to make the trees much more yellow for certain latitudes based on time of year so that N hemispheres go yellow in autumn ?
Yes, that's the plan once the basic stuff is in place. There should be a coarse model computing the mean temperature for given latitude, altitude and current earth's inclination, later augmented by a differential map that brings the effects of things like ocean currents, mountain range air current shields and such.
From the mean temperature we'll also derive probabilities of vegetation types occurring - survival of plants depends mainly on minimum temperatures during year that the plant can withstand; I guess the minimum can be roughly computed from the mean value somehow, or it will be computed&stored separately. Mean can be used for snow coverage and for many other things, so it will be computed in parallel with terrain data in sufficient resolution, since it's expected to be used a lot. Together with precipitation maps. Basically, the idea is to provide the data so one can query the values for mean and min/max values at given place, and use it for vegetation and weather phenomena and whatever, in further development.
