Climate help, or advice

[Ares Land]

I've decided to re-do my conworld maps; and review maybe a few things about climate.

The thing is, I'm not entirely sure about the climate zones. Here's a rough schematic for two continents (I hasten to add, not the actual map, which I'll rework from the schematics once I have a better idea about climate zones among other things).

climatedraft_2.png (212.84 KiB) Viewed 3660 times

The things I'm not sure about:

It looks like (1) should be humid continental; though with the mountain ranges to the west, I'm not sure how dry it gets; though it gets some moisture from the cold ocean to the north-west.
Same question about (2); maybe it's too dry for humid continental?
(3) ought to be Mediterranean, I think? Not sure how far west the climate zone would extend.
(4) should be mostly steppe, but here I wonder how far the steppe should extend. Right up to the western mountain range?

[Arzena]

My quick thoughts:

1) and 2): without knowing how tall those mountains are it's hard to say. If they are relatively short then I see a situation like the Appalachian Mtns. in the States where there's ample green on both sides of the ranges. If they're taller, however, then I'd expect rain shadows to dry out the leeward sides of those mountain ranges.

3) seems fine the way it is; I'd expect it to be Mediterranean climate too.

4) Given that this area is in the northern hemisphere, winds should blow northeastward from the ocean towards the mountains. Therefore the eastern portion of the steppe following the winds should be wetter and more forested than steppe proper. Deciduous temperate forest zone?

[Ares Land]

Thanks! For (1) and (2) the mountain ranges are as high as the Alps, or higher, so I wonder about the rainshadow too.

[Arzena]

If they're as tall as the Alps then I'd expect tall-grass prairie, something like the more humid regions of the American Great Plains on the eastern slopes of those mountains with forest returning the closer one gets to the coast / windward side of those mountains across the strait.

[WeepingElf]

For (3) I'd expect a climate like in southern China (I have seen the term "Sinic" for it), which differs from Mediterranean primarily in having monsoons; otherwise, no objections.

[zompist]

I haven't done this for awhile and could easily be wrong, but applying the rules from the PCK...

I think you have Subtropical going too far south, to 15 degrees. I'd expect desert in the top half of the Hadley cell (thus ~ 15 to 30).

Maybe monsoon forest on the east side of both continents.

I think area 2 would be subtropical (i.e. like SE China or the SE US). North of it could be a strip of Marine.

I agree about area 1 being Humid Continental.

The southern coast of the main continent would be pretty wet, with at least a coastal strip of subtropical. I don't think (3) would be Mediterranean, because it's exposed to the ocean to the south. Y'all have Africa below you in the actual Mediterranean.

I think you'd get wetter conditions in the east of area 4, but desert in the west near the mountains. A lot depends on exactly how high the mountains are... the Appalachians don't form a rainshadow, but the Alps probably would. You could get pockets of Humid Continental if there are passes or lower stretches of the mountains.

[Ares Land]

Thanks everyone, that's all great.

The southern coast of the main continent would be pretty wet, with at least a coastal strip of subtropical. I don't think (3) would be Mediterranean, because it's exposed to the ocean to the south. Y'all have Africa below you in the actual Mediterranean.

I'd say the eastern part of (3) should be Mediterranean? I might tweak the map around to make it that way.

I came across a tutorial on running actual exoplanet climate simulations: https://worldbuildingpasta.blogspot.com ... i.html?m=1. I'm trying it right now (because it's an incredibly cool thing to try!).
I'll report on the end results (after a first run the model broadly agrees with you guys, but I'll keep refining and adjusting.)

[bradrn]

I came across a tutorial on running actual exoplanet climate simulations: https://worldbuildingpasta.blogspot.com ... i.html?m=1. I'm trying it right now (because it's an incredibly cool thing to try!).
I'll report on the end results (after a first run the model broadly agrees with you guys, but I'll keep refining and adjusting.)

I tried exoplasim (on Lērisama’s recommendation, actually) and almost immediately ran into this issue: https://github.com/alphaparrot/ExoPlaSim/issues/45. Please let me know how you go! It is incredibly frustrating because I have a cool conworld idea where the normal climate rules wouldn’t work…

[Ares Land]

I came across a tutorial on running actual exoplanet climate simulations: https://worldbuildingpasta.blogspot.com ... i.html?m=1. I'm trying it right now (because it's an incredibly cool thing to try!).
I'll report on the end results (after a first run the model broadly agrees with you guys, but I'll keep refining and adjusting.)

I tried exoplasim (on Lērisama’s recommendation, actually) and almost immediately ran into this issue: https://github.com/alphaparrot/ExoPlaSim/issues/45. Please let me know how you go! It is incredibly frustrating because I have a cool conworld idea where the normal climate rules wouldn’t work…

I've really followed the steps given in the worldbuildingpasta tutorial. You'll want to follow the instructions for Ubuntu (I've tried on my usual Fedora machine but got into various issues I didn't feel like debugging.) They even provide steps for the topography, and a script that takes the results and converts them into a neat climate zone map.
So far I've gotten results that seem to make sense. (In fact it's not too different from what people suggested here, so besides the cool factor
I'm trying it with higher resolutions right now, to see how precise it can get before it breaks.

What I think helped most though I've kept all parameters to Earth values. I'll try changing things a bit later (though it's not a very alien planet... so I won't alter things much.)

[bradrn]

I've really followed the steps given in the worldbuildingpasta tutorial.

Hmm, I started with the steps in the official tutorial — I figured that I should validate it for a simple case before moving onto something more complicated. The flipside is that, since my results didn’t match for that simple case, I’m reluctant to use it for anything more complicated.

You'll want to follow the instructions for Ubuntu (I've tried on my usual Fedora machine but got into various issues I didn't feel like debugging.)

I’m on Arch (well, an Arch derivative); surely it doesn’t matter though?

[Ares Land]

I've really followed the steps given in the worldbuildingpasta tutorial.

Hmm, I started with the steps in the official tutorial — I figured that I should validate it for a simple case before moving onto something more complicated. The flipside is that, since my results didn’t match for that simple case, I’m reluctant to use it for anything more complicated.

You'll want to follow the instructions for Ubuntu (I've tried on my usual Fedora machine but got into various issues I didn't feel like debugging.)

I’m on Arch (well, an Arch derivative); surely it doesn’t matter though?

Oh, yes, I suppose so. I ran into some packages issues with python that I didn't feel like getting into, but they were most likely benign. TBH I was actually working on that during a couple of very boring meetings

[bradrn]

Well, I decided to try your link anyway. We’ll see if my results agree with his or not, I guess…

[Ares Land]

One result I'm reasonably happy with, trying for 128x64 resolution.

climatedraft_3.png (90.25 KiB) Viewed 3339 times

It's not bad really, though overall the planet looks a bit too cold. (It's a bit surprising to have ice caps that far south, though some of these are actually high mountains).
You'll see things like patches of oceanic climates in the tropics, but it's a high-altitude plateau so I can accept it.

The model places desert a bit farther norther than I expected (but that's alright); where I had mediterranean it places subtropical, mediterranean, and even, oddly enough oceanic).

TBH I'd half-assed the topography map. I'm going to make a more precise one and see what the model does with it.

The model ran more or less without issues; though it has a quite a tendency to crash. There's a 'timestep' parameter that tends to improve things.

[bradrn]

The model ran more or less without issues; though it has a quite a tendency to crash. There's a 'timestep' parameter that tends to improve things.

I haven’t yet managed to follow the worldbuildingpasta tutorial without it crashing halfway through. I’ll try adjusting the timestep and see what happens.

[Ares Land]

The model ran more or less without issues; though it has a quite a tendency to crash. There's a 'timestep' parameter that tends to improve things.

I haven’t yet managed to follow the worldbuildingpasta tutorial without it crashing halfway through. I’ll try adjusting the timestep and see what happens.

The smallest resolution looks more stable; but what did the trick was setting a timestep of 30 minutes instead of 45.

[bradrn]

The model ran more or less without issues; though it has a quite a tendency to crash. There's a 'timestep' parameter that tends to improve things.

I haven’t yet managed to follow the worldbuildingpasta tutorial without it crashing halfway through. I’ll try adjusting the timestep and see what happens.

The smallest resolution looks more stable; but what did the trick was setting a timestep of 30 minutes instead of 45.

OK, thanks; I’ll try it.

[Lērisama]

It's a nightmare getting it not to crash before it stablises. We eventually gave up and Frankensteined together a few of the individual year maps¹.

¹ Which is my recommendation. Run all scripts year by year and save the intermediate data, or enough of it to make a map with that pasta tool. It might just be paranoia, but it seemed to help somehow.

[zompist]

One result I'm reasonably happy with, trying for 128x64 resolution.

climatedraft_3.png

It's not bad really, though overall the planet looks a bit too cold. (It's a bit surprising to have ice caps that far south, though some of these are actually high mountains).
You'll see things like patches of oceanic climates in the tropics, but it's a high-altitude plateau so I can accept it.

The model places desert a bit farther norther than I expected (but that's alright); where I had mediterranean it places subtropical, mediterranean, and even, oddly enough oceanic).

Huh, I don't trust that map much. Does it know about mountains? It doesn't seem to account for either eastern mountain range.

I'm not sure it understands monsoons? Monsoon areas adjoin the oceans. It has areas that (I think) should be monsoon but aren't, and other that are unlikely (far inland).

Overall, feels like to too much Humid Continental and not enough steppe.

And yeah, feels too cold. It's easy, doing this stuff, to concentrate on present-day Earth; remember we've had ice ages and also superhot epochs. And the Hadley cell boundary isn't set in stone.

[bradrn]

For what it’s worth, the guy on the linked blog has tested it with Earth and it seems to reproduce our observed climates reasonably well.

Does it know about mountains? It doesn't seem to account for either eastern mountain range.

As Ares Land has already mentioned, it depends on the quality of the topographic map it’s given.

[Ares Land]

Mountain ranges do affect the model; I've seen it by playing around with the topography map. For now my working hypothesis is that the topography map should be improved. I wanted to test the model first, so I went for quick and dirty; but editing a greyscale map by hand takes quite a bit of work, really.
Another constraint is that you have to work with the low resolution of the model (I understand it's hard to get it to work with anything higher than 128x64 for the whole planet); wordbuildingpasta provides a workaround (adjusting temperatures based on a higher-res topography map) but so far I'm not entirely convinced by the results.

And yeah, feels too cold. It's easy, doing this stuff, to concentrate on present-day Earth; remember we've had ice ages and also superhot epochs. And the Hadley cell boundary isn't set in stone.

Good point; that map reminds me a bit of how I picture an ice age. For some reason the lower resolution was warmer.

The low-resolution model was better in some ways but failed to predict the desert on the eastern continent (and I'm really convinced there should be desert there).