The Köppen–Geiger climate classification made simpler (I hope so)

[Charerg]

If it increase the aridity there, it will also increase it somewhere else.
-12 will be too low for some places.

It wouldn't be such a bad thing if the steppes extended further north in Canada, since I think that is the case in real world Köppen maps. I think the error comes because we only use data from extreme months: the continental climates tend to have too low T_ann values because of this.

[Azélor]

It wouldn't be such a bad thing if the steppes extended further north in Canada, since I think that is the case in real world Köppen maps. I think the error comes because we only use data from extreme months: the continental climates tend to have too low T_ann values because of this.

Maybe a little more arid yea.

I don't think using the extreme month has a big impact. The temperature follow a bell curve. Temperatures changes more quickly in spring and fall but more slowly in summer and winter. Overall, I don't think the yearly average would be very different.

[Charerg]

Maybe a little more arid yea.

I don't think using the extreme month has a big impact. The temperature follow a bell curve. Temperatures changes more quickly in spring and fall but more slowly in summer and winter. Overall, I don't think the yearly average would be very different.

I'd say there definitely is some difference, because if there wasn't we should be able to sort out the Db/Dc boundary. It would be interesting to compile an "annual mean temperature" map using the categories we use and compare it with an actual map of annual mean temps.

[Charerg]

Ok, I've now updated the script.

Here's the output:
genClimates_test.png

Note:
I made the slight modification of including the [200+ / 100-50] precipitation combo with Af instead of Am, since that was the case in the prior version as well (and overall the combo is closer to Af prec pattern, I think).

Overall it does match real-world Köppen maps better than the previous version: Australia is more arid, and the cold steppes extend further north in North America and Central Asia. The updated script is in the attachments, as usual.


Installation:
Overwrite the older version of the script if you're using it. Otherwise, follow the instructions in this post.

[Azélor]

Yes I think this is an improvement from the previous version.

Next?

[Azélor]

Trying to figure out a simple way to map precipitations (July).
test.png

So ideally, each colour represent if an area is wet or dry and the reason why. Sometimes a darker colour is used to represent a wet yet drier area.
I'm not sure where this is going.

[Charerg]

Yes I think this is an improvement from the previous version.

Next?

I guess we could work on adding the 10-13 °C temp category. That should eliminate the awkward Cc-Db transitions that are very common right now and largely sort out the Cc distribution. I feel after that we could consider the main tutorial largely finished, unless you want to develop the 8-step precipitation version further?

Although I guess the development focus could be shifted towards refining the instructions, since I think we're in a pretty good spot with the climate generation even with just the 6 precipitation levels, now that the aridity thresholds have been sorted out (that really did fix a lot of the most problematic regions like Australia).

[Azélor]

I guess we could work on adding the 10-13 °C temp category. That should eliminate the awkward Cc-Db transitions that are very common right now and largely sort out the Cc distribution. I feel after that we could consider the main tutorial largely finished, unless you want to develop the 8-step precipitation version further?

Maybe.
No, we don't really need 8 categories.

Although I guess the development focus could be shifted towards refining the instructions, since I think we're in a pretty good spot with the climate generation even with just the 6 precipitation levels, now that the aridity thresholds have been sorted out (that really did fix a lot of the most problematic regions like Australia).

Improving the instructions for the precipitations is a good idea.

[Charerg]

Maybe.

I think it's definitely worth pursuing that improvement, since the Cc distribution is the biggest flaw with the climate generation. Here's the 10x10 temp table where I've highlighted the "bad transition" from Cc to Db:

10x10 Temp Table.png

And the 11x11 table, where the transition is now from Cc to Dc (as it should be):

11x11 Temp Table.png

[Azélor]

How do you put it on the map?
Select about 30% of the area?