Climate in Universe Sandbox ²

[Dan Dixon]

Naomi, our climate scientist, just wrote up a detailed explanation of how we're handling the climate simulation in Universe Sandbox ².

Read about it here:
http://universesandbox.com/blog/2014/10/climate/

Let us know if you have any questions...

[TaroTaoYin]

nice touch there. Looks really good.

[gabriel.dac]

That's the best blog post you have written so far. 

From what I read, the climate simulation really is accurate enough. I imagine it must be really hard to implement such feature because there are lots of details and parameters that can be changed by the player (CO², atmosphere mass, etc), so it must be nearly impossible to make something totally accurate, but it seems to be good enough.

I've been worrying about... the clouds. When the Earth gets hotter (you replace the Sun and add Sirius, for example) I assume the surface would no longer be visible because of the them. I see the oceans drying, but where does all that water go to? Also, there is a problem. When the oceans evaporate, the atmosphere will get denser, and that should change the atmosphere mass properties in the right-click menu. Also, a denser atmosphere would be better at trapping the heat (like in Venus), I guess.... and that would make Earth even hotter. Well, as I said, the climate sim doesn't have to be perfectly accurate anyway

_{edit: I posted another feature suggestion in the other thread, so maybe you wanna check it out... http://universesandbox.com/forum/index.php/topic,13729.30.html}

[nosirrbro]

I just hope for climates on all planets and making the point on the planets that are the poles be dynamic. Other than that its perfect!

[WitheHole18]

This Sound cool..
The new cloud simulation and climate revision..
Good job.

[Plutonium]

Сan there be an atmospheric opacity function? That would be very interesting in modelling the impact of different types of opaque gases in atmosphere.

[Bill3000]

The paper you linked doesn't indicate how you calculate greenhouse inputs in your model; according to your article you use the idealized greenhouse model, but how do you link the greenhouse gas amount to the emissivity?

Also when will the climate simulation be applicable to planets that are not earth (e.g. random rocky planets for other systems)

[Naomi]

Thanks for checking it out. There is a long list of things to add to complicate what we're doing with climate for Earth.

Climate on Other Planets
Generalizing this to any old planet is actually a pretty hard problem. We specify a lot of parameters which are Earth-specific: they are calibrated to give us realistic output for our planet, with which we are quite familiar. Gradually we'll move towards informing those parameters more with other physical information that you could specify, but it'd be pretty imprecise at the moment.

To illustrate how much one planet could differ from another, I'm working on extending the model to work on Mars. This is more complicated than just changing all of the model parameters to be Mars-appropriate. On Mars the primary constituent of the atmosphere also condenses out at the poles in winter. The atmosphere is so thin that you need to account for the feedback between the mass the atmosphere loses to condensation and the condensation temperature itself. And the heat transfer involved in the phase change is important too. You can see how an arbitrary planet with an arbitrary atmosphere could easily have some other important process that we're not considering.

Greenhouse Gas and Emissivity
The way that the outgoing infrared energy is connected to the CO₂ amount is twofold. We adjust one of the constants using an empirical relationship - it comes from running more complicated models of radiation transfer and fitting a function to the results. Turns out that the outgoing IR energy falls off like a logarithmic function as CO₂ concentration increases. This too is Earth-specific, derived from models of CO₂ in the context of the rest of our atmosphere composition in realistic concentrations. Second the temperature also feeds back and effects the value, as you'd expect because bodies emit as a function of their temperatures. I'll add a better explanation of this to the FAQ soon.