The idea of thermal inertia from the hydrosphere is really neat and I believe will help temper (pardon the pun) the more extreme environmental behavior possible from eccentric orbits. 

Actual implementation could be as simple as creating a maximum temperature change rate based on the size of the planet, hydrosphere percent, and any other factors Steve wants to include such as distance from the star and brightness of the star.

I doubt there are many life forms which can survive such large changes in temperature, regardless of atmospheric composition, so you could start by excluding any forest/jungle type of terrain.

I don't see it being inherently problematic if  it changes from a hot desert to a cold desert, but the sudden appearance and disappearance of mountains or jungles would be.
So maybe the solution could just to be having a hot, medium, and cold environment types and it switches between them, depending on whether it passes through the 100 and 0 degree marks?

As for temperature inertia, the planet core has a fixed (for game timescales at least) temperature which would provide some sort of buffer to temperatures falling too much and changing too quickly. However that might be a factor you don't want to start adding just for the sake of it.

As a final note, any infrastructure for the colonists would need to be adapted to both extreme heat and extreme cold, is it worth having a minimum 3.0 colony cost for this? Adapting a house or car to cold weather and hot weather normally involve rather different designs, so if you want to incorporate both features into the same design it will be less efficient.

Quote from: alex_brunius on September 16, 2021, 07:48:41 AM

Would it make sense to add some inertia to the hydrosphere type or have larger hydrospheres provide some inertia to planet temperatures?

It's probably not very realistic to have planetary sized bodies of water swap from frozen solid to steam and then back again in a matter of days.

Interesting idea. In the past, the change from liquid to vapour was instant because it was always the result of terraforming and therefore likely to be permanent. Maybe the change should be gradual in the same way as condensing (0.1 atm per year). That would solve the problem of drastic changes in short orbits while leaving the option of different conditions in long orbits.

Liquid to ice and back is less of an issue because it only affects albedo and doesn't introduce a sudden 2.0 colony cost (and I already include albedo changes in the min/max temperature and min/max colony costs). I could perhaps track the percentage of water that is frozen, rather than having separate states, but in reality the water to ice process happens quite quickly so probably not worth it.

Temperature changes could be gradual as well. Planets heat up because radiation from their star hits them, and heat is simultaneously escaping the planet by radiating away. If more heat is radiating away than is radiating in, then the planet is cooling, otherwise it is heating up. One interesting quirk is that the amount of radiation emitted is proportional to the fourth power of the temperature, so hot objects radiate away heat faster than cold objects, all else being equal. (On the other hand, emission doesn’t really become an efficient process until you hot enough to glow in visible light).

With some interia in the temperature, the planet in your example would probably have a narrower range of temperatures. Especially since it spends a lot less time near the periapse, and more near the apoapse.

Perhaps For extra fun, you could actually implement the single–layer model, where the ground absorbs part of the incoming radiation and reflects the rest (due to albedo), the ground emits radiation, part of which is absorbed by the atmosphere. The atmosphere also emits radiation, some of which goes into space and some of which is reabsorbed by the ground. Normally you solve this for an equilibrium condition, but with an elliptical orbit the numbers wouldn’t always balance. You would need to track the atmospheric temperature separately from the ground/hydrosphere temperature, which could be a lot of fun. Plus you would get to decide what temperature new gasses from the aether are brought in at when terraforming!

Another thing you could do is to take into account the latent heat of the hydrosphere as it freezes and melts. As you know, water that is freezing stays at 0℃ the whole time. The heat that is in the water is emitted to the surroundings, but instead of dropping the temperature of the water, this causes a phase change instead. Likewise, when melting the ice absorbs heat but the temperature does not go up. If that planet has any significant amount of water, it would buffer the temperature quite a lot: as the planet heats out towards the apoapsis, the temperature of the ground, the hydrosphere, and the atmosphere would begin to drop. The hydrosphere would begin to freeze, dumping heat into the atmosphere and helping it to stay near 0℃. Thus the colony cost would not go up right away.

What about just adding a new environment/hydrosphere for sufficiently eccentric planets? IE on that planet instead of swapping back and forth between them it would just say Environment: Extreme Variations and Hydrosphere: Seasonal?

My concern with giving inertia to the hydrosphere is that it might complicate terraforming. Hydrosphere changes albedo, which normally isn't an issue since it shifts instantly, so if you're adding greenhouse gases to a planet at some point you'll get a message the hydrosphere has melted and that bumps the temperature up further. But if there's a delay, that could mean terraforming too quickly would result in you getting the planet up to a pleasant temperature, then the hydrosphere melts and the temperature jumps past the high end of tolerance.

If I add inertia it would only be between liquid and vapour. See my answer a few posts above on this topic.

I been considering a different hydrosphere state, but I think there are too many variations. For example, a planet that is mainly liquid that loses it during a small portion of the orbit is very different from a planet that only has liquid for a small portion and both are different from the planet above with all three states within a few days.

At the moment I am leaning toward adding the same limit on evaporation as exists on condensation (0.1 atm per year), plus making the dominant terrain fixed if the orbital period is six months or less (unless terraformed). That dominant terrain would be based on all possible environmental conditions during the orbit.

I don't know the exact math, but shouldn't atmospheric pressure have an insulatory effect regulating the extreme temperature radiation.  I know on a daily scale this is important in comparing min/max day/night temperatures between earth, Mars and mercury, and I would assume the effect would be present though likely not as strong for longer time scales for orbital eccentricity.

Could maybe even add another -isium type gas that reduces temperature variation.

Quote from: Steve Walmsley on September 16, 2021, 08:40:30 AM

Liquid to ice and back is less of an issue because it only affects albedo and doesn't introduce a sudden 2.0 colony cost (and I already include albedo changes in the min/max temperature and min/max colony costs). I could perhaps track the percentage of water that is frozen, rather than having separate states, but in reality the water to ice process happens quite quickly so probably not worth it.

Does this mean adding ice will lower albedo in 1.14/2.0?

Water to turning to ice and vice versa already changes albedo in v1.13.

As for vegetation... There is no option to have any sort of soil on anoxic planet. No life - no sod - no soil. There will be plenty of rocks, sand and mineral dust, yet to have a soil or even clay (except on oceanic abyssal flat) - you need at least thousands of years of ubiquitous terrestrial vegetation.

So I think it will be cool to have this feature: a boolean field Soil Layer, possible for the same planets that have oxygen, mandatory for homeworld planets, can be erazed with backing (I'd say average surface temp 1000C), cannot be restored during playtime, mandatory for <1.50 CC.

Two things: If you can't create a soil layer, how could you ever fully terraform Mars and other planets?

And what about alien biologies? Even if they'd be incompatible with human/Earth biology, large alien plants could still provide the cover that the terrain modifiers represent.