Biome placement for fantasy worlds

[mbartelsm]

This is a continuation of Azelor's wonderful guide

BiomesGuide.pdf

The purpose of this tutorial is to help people who want a realistic world to go from climates to biomes, which may sometimes prove more useful.

Changelog:

1.0.1:
— Fixed a color map that was missing one color

[mbartelsm]

The result of this tutorial can be seen below:
BiomeMap.png


Climate map for thos curious
ClimateMap.png


Built from the following world map
Topography.png

[Azélor]

I did some quick montages were I included your colors scales with the data I used in my tutorial. It was made hastily but I think it could be interesting for comparison.
I rotate your tables by 180 degrees to go from high to low (left to right/ top to bottom)

The precipitation table was imported from Excel via a print screen (I know how awful it is ). The categories at the top and those of the left have only min/max with the max capped at 200 (because more rain over 200mm had little impact with Earthlike temperatures on the climates).
For the other categories, the bottom right is the minimum, the top right is the maximum and the other two are the average.

* I don't remember if the lowest category is supposed to 0-10 or 0-12,5, because I did this table a couple of weeks ago.

So normally, unless I'm missing something, the precipitation categories, once combine (summer and winter) should cover all the possible combinations of the Holdridge system. Except for the rainiest one, over 4000 were there would only be two categories instead of three. Meaning that the rain forest would be the same as wet forest. Like if there is a big difference.

mbmartin copie.png

mbmartin copie 2.png Just to be sure, you used 5 different colors here or 6?

This is the yearly average temperatures. I used this to determine the aridity index in another Excel sheet.

temp copie.png

temp 2.png

Now about you results, aside form having some weird climates (having deserts near the polar circle), I see that there are many inaccuracies between the climate and biome maps. Some deserts disappear to become forests and vice-versa?

Also, I think the coniferous forest is misleading. These foresst can also be found with milder climates like in California and British-Colombia, not just in cold places. Personally, I would rather use boreal or taiga to define that specific biome.

Btw, as a graphic designer, do you sometimes have to design things that are made to be used by color-blind people? I was just thinking that these people would find our tutorials pretty annoying since they are highly dependent on using the right colors.

[mbartelsm]

I did some quick montages were I included your colors scales with the data I used in my tutorial. It was made hastily but I think it could be interesting for comparison.
I rotate your tables by 180 degrees to go from high to low (left to right/ top to bottom)

The precipitation table was imported from Excel via a print screen (I know how awful it is ). The categories at the top and those of the left have only min/max with the max capped at 200 (because more rain over 200mm had little impact with Earthlike temperatures on the climates).
For the other categories, the bottom right is the minimum, the top right is the maximum and the other two are the average.

* I don't remember if the lowest category is supposed to 0-10 or 0-12,5, because I did this table a couple of weeks ago.

So normally, unless I'm missing something, the precipitation categories, once combine (summer and winter) should cover all the possible combinations of the Holdridge system. Except for the rainiest one, over 4000 were there would only be two categories instead of three. Meaning that the rain forest would be the same as wet forest. Like if there is a big difference.

Pretty much, there is something that's been troubling me though, and it's the fact that even though the combinations don't cover the whole system, the do extend beyond it, especially as you progress to colder climates. It bothers me because, with my limited climatic knowledge, I don't understand the implications of having such combinations nor the reason they are not included in the first place.

mbmartin copie 2.png Just to be sure, you used 5 different colors here or 6?

I messed up there, it's supposed to be 6 different colors.

This is the yearly average temperatures. I used this to determine the aridity index in another Excel sheet.

temp copie.png

temp 2.png

I'll update my notes with this, thanks for sharing. There is one problem though (in fact, the biggest problem about Holdridge's system), and it is that I didn't use average temperatures, but average biotemperatures. Biotemperatures being defined as the range of temperatures that fall within the 0° to 30°C range, and average temperature being the sum of the monthly averages that fall in that range, divided by 12. The problem with using biotemperatures is that we have only two temperature maps available, one for January and the other for July, not the whole year.

I tried to solve the problem by assuming a sine wave pattern for a year's temperature change, as can be seen of wikipedia's climate charts
Untitled-1 copy.png
With the help of the sine function I assigned each month a multiplier value with january and july at the peaks and did some simple math to find the average temperature for each month. After that I calculated the biotemperatures for each possible combination of temperatures, the result below.
Untitled-1.png
And the partly expanded table (the top values are the multipliers, the bottom ones are the month's guessed avg temperature).
Untitled-1.png
For obvious reasons, my temperatures differ slightly from yours, but that's something I'll fix for the next version.

There is a trend I noticed in climate charts that I want to try to reflect, and that is that the curve for winter seems to be steeper than that of summer, that is, the coldest period is shorter in duration than the warmest period. I don't quite understand why but it seems to be present in the great majority of charts I've seen.

Now about you results, aside form having some weird climates (having deserts near the polar circle), I see that there are many inaccuracies between the climate and biome maps. Some deserts disappear to become forests and vice-versa?

This is really a problem with the way I decided to simplify Holdridge's System. Because I merged many categories it looks as if some biome transitions are too sudden. The deserts near the polar circles are actually cold deserts, and the forest desert transition is actually a forest-woodland-shrubland-desert transition. That is something I'm planning on fixing by adding at the very least a woodland category.

Also, I think the coniferous forest is misleading. These foresst can also be found with milder climates like in California and British-Colombia, not just in cold places. Personally, I would rather use boreal or taiga to define that specific biome.

You are absolutely right, I think boreal forest describes the biome much better. Do you suggest something for mixed forests as well? it's the category I'm most uncomfortable with.

Btw, as a graphic designer, do you sometimes have to design things that are made to be used by color-blind people? I was just thinking that these people would find our tutorials pretty annoying since they are highly dependent on using the right colors.

I apologize for that, this was made kind of in a hurry as I only really have free time during the weekend and wanted to finish the first version before monday. It goes without saying that it still needs a lot of improving.

EDIT: I think I can upload a new version today, with some of the more immediate problems fixed. A more complete update will have to wait probably until saturday because of my work.

[Azélor]

So the problem with the missing combinations is: since you used 5 greens instead of 6, you ended up with missing 7 of the temp/rain combinations. The whole 5th row is missing.

I'll update my notes with this, thanks for sharing. There is one problem though (in fact, the biggest problem about Holdridge's system), and it is that I didn't use average temperatures, but average biotemperatures. Biotemperatures being defined as the range of temperatures that fall within the 0° to 30°C range, and average temperature being the sum of the monthly averages that fall in that range, divided by 12. The problem with using biotemperatures is that we have only two temperature maps available, one for January and the other for July, not the whole year.

I tried to solve the problem by assuming a sine wave pattern for a year's temperature change, as can be seen of wikipedia's climate charts

So my table of average temperatures is not compatible with this variable. Since it's just (Min average + Max average)/2

Do you suggest something for mixed forests as well? it's the category I'm most uncomfortable with.

I have mixed feelings about it too. Presumably, I live in a mixed forest biome but most of the trees are deciduous trees https://en.wikipedia.org/wiki/Easter...owland_forests

I apologize for that, this was made kind of in a hurry as I only really have free time during the weekend and wanted to finish the first version before monday. It goes without saying that it still needs a lot of improving.

It was just by curiosity, not a request.

[mbartelsm]

It was just by curiosity, not a request.

I'm sorry, it seems that I misread the question yeah, colorblind people would probably be pretty annoyed by our tutorials, but I don't know of a better way of doing it since we are technically using colours to perform calculations.

So the problem with the missing combinations is: since you used 5 greens instead of 6, you ended up with missing 7 of the temp/rain combinations. The whole 5th row is missing.

That isn't it, I just checked and my photoshop work file has the correct colors.

It seems the problem stems from combining the temperature layers. An odd result in the temperature categories chosen and the biotemperature calculations is that sometimes the biotemperature layer will skip categories, resulting on it skipping biomes.
Untitled-1.png Untitled-1.png

Looking at the raw combined temperature layer it looks like there isn't a way to easily fix the problem, so the solution is gonna have to be manually painting missing temperatures. The problem appears on mountain in the tropic where the temperature maps are very similar and thus generate strong overlapping (such as an area skipping 2 or 3 categories) and in mid latitudes between the second and third categories (seen above).

[mbartelsm]

Ok, there are a bunch of problems, some were the fault of the process to make the biome map, others were in the formulas and data I was using and some are a part of holdridge system itself.

Problems with the process:
- The colors I had chosen were conflicting heavily while one was painting them, which caused the map to miss a lot of biomes
- The categories I had defined were misleading and sometimes way too broad

Problems with the data:
- I had to tweak Azelor's data a bit in order to be able to generate more accurate results, this change was mostly done to the precipitation categories
- I revised some of the formulas to more accurately represent the categories and their true values.

Problems with Holdridge System:
- The use of bio-temperatures is troublesome and misleading as one approaches higher temperatures, since the formula treats temperatures above 30° as 0° then some of the hotter places end up categorized as cold, for example a desert that varies between 28° and 35° will have a biotemperature of less than 5°, meaning it would be categorized as a boreal desert, when, in fact, is a tropical or subtropical desert.

This means I'll have to tweak the model to make it more accurate for our purposes, it will probably end up looking like the bastard child of Whittaker's biome classification and Holdridge's Life Zones.

---

UPDATE:
Ok, so I ran a few tests, the model is now much closer to the climates map, I rethinked the biotemperature equations to only divide the temperatures by the number of months below 30°C, that way hot deserts are still represented as such while remaining accurate for the lower temperatures.

I also added a new temperature category to account for the subtropical/warm temperate half step in the holdridge model. That way there is no need to resort to selecting climates with the wand in order to get a half accurate result.

As a result of the tweak in the numbers and equations, there is now "access" to a new, wetter category for tropical forests, which would be similar to the central amazonic jungle where water level can almost reach up to the canopies of the trees. The number tweak also made deserts true to their climates.

The only thing I still have left to do is rethink the categories so that the map makes sense, right now I'm working with a super simplified category system made for test purposes.

[Azélor]

I think that replacing the high temperature with 30 instead of 0 should make more sense. That's what I did and I find it give a similar results to the climates.

[arch-fiend]

sorry for resurrecting an old thread but i've been playing around with all of these climate and biome guides i've been seeing on here (well actually only pixie, azelor, and mbartelsm) and i just feel like sharing what I've been coming up with. so a few things i should note is that first i have very simplistic geometry for this map, the mountain ranges are very wide and completely consistent in their elevation (not very realistic but i felt like being lazy for that.) the second thing i should note is that by the time i made it to the section on plotting precipitation patterns on azelor's guide i simply could not figure out how it should work, perhaps if i had more experience doing this kind of thing it would make sense but i felt like azelor was kind of vague on how far humidity should be pushed inland starting with the green/white map. after the green/white map i could start to make sense of things but by that point i was following instinct as much as the guide. i must commend you on suggesting using the paint-bucket tool in the way you suggested for this guide mbartelsm, when i was doing azelor's climate map i was simply color selecting and filling which took hours compared to minutes yours would take, but i blame that completely on my inexperience, i could probably knock azelor's map out in a few hours now to instead of the 3 days it took. you will also notice that i used the traditional koppen colors for the climate map.

prometheus maps.png

[Azélor]

zelor was kind of vague on how far humidity should be pushed inland

Yes, that is not the strongest point of the tutorial.

n suggesting using the paint-bucket tool

I usually just paint the border unless it's necessary to paint the whole thing by hand (when it's small, for example)
and fill the rest with the bucket.
I haven't mentioned it because I guess it's like you said, there are some time-saving tricks that you learn overtime.