Climate divisions and food yields, a framework and overview

[Sulphurologist]

1. Climate Maps comparison

Now that we are receiving all the juicy details of topographic modifiers in the game, and lots of people are discussing it, I'd like to show how Tinto's classes relate to the full Köppen Classification.
Köppen uses 30 categories, which Tinto (understandibly) wittled down to 8.

Considering climate in game represent both climate phenomena (winters, monsoons, droughts,...) AND represents food availability/development potential, I went down the rabbit hole, and delved deeper into crop yields for each climate zone.

You can view my full statistical analysis for each crop in my google doc here, with graphs (which I'll also highlight futher down the post).
All GIS files can be found on my GitHub page, including non-food files for Tobacco, Coffee, Cocoa, Flax, Sugarcane, Olives and Tea.

[I'll update files/typing errors int he following days]

Spoiler: All Topography modifiers as we know them

Tropical

Subtropical

Mediterranean

Oceanic

Continental

Arctic

Cold Arid

Arid

Desert

Sparse

Grasslands

Farmlands

Woods

Forest

Jungle

Wetlands

Flantlands

Plateau

Hills

Mountains

Population capacity

+50%

+100%

+150%

+50%

+50%

-50%

10k

25k

50k

100k

+50k

+25k

+50k

-80%

Development growth

-10%

-25%

-10%

+10%

-20%

-25%

-50%

-30%

-25%

-30%

-70%

Life expectancy

-5

-5

-5

Attracts pops?

Yes

Yes

Yes

Yes

Yes

No

No

Yes

Winters

None

Mild

None

Mild

Normal

Severe

Mild

No

Precipitation?

No

No

Wheat Production

-10%

-10%

Food Production

-0.33

+10%

+33%

+10%

-10%

-10%

-20%

Unique weather?

Sandstorms

winter blocked

Road build time

+10%

-10%

+10%

+25%

+50%

+200%

+75%

+25%

+50%

+200%

RGO build time

+50%

+33%

+50%

+25%

+25%

+25%

+100%

RGO max size

+10%

Army movement cost

+10%

+10%

+25%

+50%

+100%

+50%

+25%

+50%

+100%

Attacker Dice Roll

-1

-1

-1

-1

-1

-1

-2

Max battle frontage

-2

-3

-4

-3

-1

-3

-4

Blocks adjacent vision

Sea

Sea+Land

Sea+Land

Sea

Sea+Land

Sea+Land

Tinto's Climate World Map	Tinto's 8 categories colored in on Köppen	the Köppen climates in full spectrum	The games climates according to my personal taste in regards to gameplay.

Spoiler: Climate Legend Comparison side-by-side

2. Individual crops, and their inherent potential for each Köppen zone
We can view the potential performance of crops in 3 ways:

1. We focus on climate alone
   • We will not factor in 'high input' of fertilisers, as we are not considering soil suitability either
   • pure theoretical performance of a crop, accounting for all climate factors (evaporation, rain,...)
2. We focus on climate + soil suitability WITHOUT fertilisation (= Attainable Yield, Low Input)
   • More in line with what would be realistically attainable in the late medieval period
3. We focus on climate + soil suitability WITH fertilisation (=Attainable Yield, High Input)
   • This is what we in our modern society should be able to extract with maximised efforts

The FAO provides us with with a very extensive map viewer on all crop-related factors, such as climate and soil.
I can highly recommend reading on their methodology in the documentation.
Especially from p 147 onwards is very interesting (with detailed tables of yields for each crop within AEZ-category.

• The first map in Theme 1 is the "agro-ecological zones" (AEZ) map of the FAO. This is an alternate climate categorisation including soil characteristics, specifically for agriculture.
  • This is basically 'the climate map for crops'
  • Some categories are remarkably similar despite their differences in the Köppen zone (e.g. Oceanic and Continental, Arid and Cold Arid)
  • Includes 'irrigated' zones such as the Nile, Mesopotamia,...
  • Includes 'unsuitable terrain' as a category

Keep the following things in mind:

1. Crop performance is dependent on five factors:
   1. Soil suitability
   2. Climate
   3. Artifial irrigation
   4. Use of fertilisers
   5. The crop itself

In an ideal situation, specific crop fertility could be tailored in the game to each crop type, based on the first 4 factors:

• the inherent soil suitability (base)
  • no matter what else, if plants can't take root, you can forget about all other factors
  • Currently not implemented in game whatsoever
• the climate (modifier)
  • no matter how good the soil, if it is too dry or too wet for that crop, the crop will struggle
  • Has great potential to be expanded upon
• artificial irrigation (neutraliser)
  • Places that struggle with too much or too little water access, can be irrigated to alleviate (or even remove) the climate problems
  • River / irrigation presence could be checked to 'erase' climate/terrain negatives
• Fertilisation (amplifier)
  • Some climates or soils can yield surprising amounts of produce with the right investments
  • Development level (amount of 'input') and specific technologies ('fallow fields') could impact produce output in specific terrains and climates.

Some examples:

• great soil, terrible climate, no potential for irrigation: Tarim basin, Gobi desert
• great soil, terrible climate, good potential for irrigation: Nile river valley, Indus river plain, Mesopotamia
• unsuitable soil, great climate, good potential for irrigation: Upland Korea

DISCLAIMER: Irrigated places are NOT considered in these maps! (So the valleys of the Nile, Indus,... look unrealistically barren)

• EDIT - IMPORTANT NOTE: this assumes that all places on earth use the same inputs and have access to the same technologies/logistics. This is obviously NOT the case in real life!
  • Some regions will IRL have results similar to the map of 'high inputs' (e.g. SE England), while some regions despite their best efforts will be limited towards the results of the 'low inputs' map (e.g. central Spain, tropics).
  • Historical technologies of each region/climate will ensure that MOST regions cannot achieve the results of the "high inputs" map, even in modern times.

Food Crop	Theoretical Climate Potential	Actual Attainable Yield - No fertiliser ("Low Input")	Actual Attainable Yield - High fertiliser ("High Input")	Histogram of Crop Yield for each climate, for each version
Wheat
Sturdy Grains (barley, buckwheat, oats, millet, sorghum, rye)
Rice (dry- and wetland rice)
Legumes (phasoleus, chickpea, cow pea, pigeon pea, gram, groundnut)
Soybeans
Maize
Tubers (White Potato, Sweet Potato, Yam, Cassava)

EDIT: Yield Achievement ratio.
Despite the high attainability of crops for most climates, reality dictates that due to limitations in technology, logistical issues regarding transport, labor and materials, access to fertilizers, impractical seasons that make it difficult to have consistent growing seasons,... mean that a lot of nations will not achieve the "high inputs" map.

This difference is called the 'yield achievement ratio', and here's the map for how well regions can achieve the theoretical models.

3. Analysis of Crop Yields per Köppen Zone

In GIS, we can calculate 'zonal statistics', where we calculate the mean yield of each crop for a unit of surface area.
We can then compare all the yields to see which Köppen zone is the most diversely fertile ("general food"), or if specific regions lack the requirements for a specific crop (like rice in Oceanic).

The results and graphs are made available in the google sheets link.
All GIS files, statistics files and images can be found on my GitHub page.

	Climate only	Climate + Soil suitability - Low Input	Climate + Soil suitability - High Input Needs redo because of the potatoes map
Sorted according to Köppen Classification
Sorted according to Tinto Categories

4. Conclusions

• The top 5 most productive climates (Attainable Yield - Low Input) are:
  • Dfa - Hot-summer humid continental
  • Cfa - Humid Subtropical
  • Dwa - Hot summer continental
  • Dfb - Warm-summer humid continental
  • Cwa - Dry winter humid subtropical
• Potatoes should perhaps be renamed to "Tubers", as its cousins are very intensively grown in (Sub)Tropical regions. Edit: maybe potatoes is better after all... Potatoes and yams are tubers, whereas taro and cocoyams are derived from corms, underground stems, and swollen hypocotyls. Cassava and sweet potatoes are storage roots and canna and arrowroots are edible rhizomes.
• Soybeans do not appear to behave very differently from all other legumes, so imo could be merged with legumes.
• The treatment of merging "BSk: Cold Semi-Arid" into "Cold Arid", and "BSh: Hot Semi-Arid"into "Arid" is too punishing for the semi-arids.
  • EDIT: The grasslands/woods/farmlands topography could perhaps represent the semi-arids, but should allow for migration then.
    • Maybe invert the logic: make desert/sparse the condition that doesn't attract migrants. Although I agree Arctic should keep the modifier.
• The current Continental category is perhaps too broad, especially looking at crop yields and winter severity. A colder and more temparate variant wouldn't hurt.
• The Tropical Savannah climate is noticably more fertile than its other Tropical counterparts, a category perhaps worth splitting.
  • EDIT: the vegetation modifiers could be utilised to make the Savannah region more suitable.

 

[korandder]

So on climate factors alone, the PC category Mediterranean is not too different from Subtropical but Mediterranean is inferior once you fold in soil quality. This is not surprising given that the only climate difference between Mediterranean and Subtropical is whether you have more rain in the winter or the summer.

Another thing I would expect to have an effect on yields is the length of the growing season. If you are some where mild enough were you would have multiple plantings in the same year you should usually produce more than if the growing season is shorter and you can only make one planting.

 

[SPb enjoyer]

Our topography-climate king has spoken!

 

[Tissaia]

Really interesting analysis. I'd like to hear more from the devs about any existing restrictions or modifiers related to food production. Judging by this exchange from a few days ago a lot is still in the air in that department:

Why only wheat has a malus on arid and not rice for example?
Or is rice not possible at all on arid?

might be a good idea

One question to you (since you were involved in the works related to climate during Tinto Maps):
In some areas it looks like not only number of Koppen categories was reduced, but also some borders between them were moved. I'm specifically thinking here about the line between oceanic and continental in Central Europe. It seems to me as if for that area current Koppen map was used rather than 1901-1930. Do you know anything about it?

PS. I would be against changing potatoes to "tubers". Everybody wants to bring potatoes to Europe, nobody cares about "tubers". I even had to check what that word means.

 

[Dev2150]

That's a lot of effort making this!
I would love to see this implemented!

 

[racnes]

I'm all for more climate-terrarin types than currently implemented in game. It adds more realism

 

[POD012]

PS. I would be against changing potatoes to "tubers". Everybody wants to bring potatoes to Europe, nobody cares about "tubers". I even had to check what that word means.

And several people probably had to check what "legumes" and "sturdy grains" meant, too. A lot of categories of agricultural goods probably have to be coalesced for gameplay reasons, or at least would offer little benefit split. Having the groups here be based on their general (botanical, nutritional) characteristics and affinity for different soils and climates would make the most immediate sense to me.

 

[korandder]

Another factor to consider in pre-modern agricultural yields is the availability of animal power. The main effect of having animals available is to pull a plough, with those with access to more or stronger animals (oxen as opposed to horses) being able to pull a deeper plough. Ploughing is key to break up and turn over the earth, destroying the root systems of weeds, disrupting the life cycles of pests and making it easier for seeds to germinate. Animals could also be used to help thresh grain, either by trampling it or in the later stages of the period covered by PC, powering a threshing machine. Note that these effects are stronger for growing grain than they are for rice which is more dependent on investments in irrigation and the like.

Having animals around also provides a source of manure, the main type of fertilizer used in pre-modern societies as well as being pulling carts to help transport agricultural goods.

There was a range of agricultural productivity in pre-modern societies. Poor subsistence farmers would produce significantly less than a rich landowner aiming to sell to the market. Bret Devereaux has a series of blog posts on pre-modern agriculture.

 

[Sulphurologist]

So on climate factors alone, the PC category Mediterranean is not too different from Subtropical but Mediterranean is inferior once you fold in soil quality. This is not surprising given that the only climate difference between Mediterranean and Subtropical is whether you have more rain in the winter or the summer.

Another thing I would expect to have an effect on yields is the length of the growing season. If you are some where mild enough were you would have multiple plantings in the same year you should usually produce more than if the growing season is shorter and you can only make one planting.

Indeed, length of growing season is a big one! The length of the 'longest dry period' (*insert immature joke here*) in growing season is apparently very influential: too dry for too long is obviously bad, but no dry periods are also detrimental.

 

[Sulphurologist]

One question to you (since you were involved in the works related to climate during Tinto Maps):
In some areas it looks like not only number of Koppen categories was reduced, but also some borders between them were moved. I'm specifically thinking here about the line between oceanic and continental in Central Europe. It seems to me as if for that area current Koppen map was used rather than 1901-1930. Do you know anything about it?

Yes they are still going through the climate map to adjust them to the 1901 map, I think it'll be a while till they adjust everything.

PS. I would be against changing potatoes to "tubers". Everybody wants to bring potatoes to Europe, nobody cares about "tubers". I even had to check what that word means.

If potatoes were to be separate then sweet should be a separate good, too. The influence it had on China and East Asia in general was immense!
Sweet potato also was massively produced in East Asia, and Cassava everywhere in the tropics (nicknamed 'the bread of the tropics'). Yam was important in Western Africa.

To that end they represent the same thing though: starchy 'roots' from the New World, that increased food availability significantly.
Similarly I wonder why soybeans are separated from legumes, and why dates were split from fruits. They will probably have a flavor function, but still!

 

[Sulphurologist]

Another factor to consider in pre-modern agricultural yields is the availability of animal power. The main effect of having animals available is to pull a plough, with those with access to more or stronger animals (oxen as opposed to horses) being able to pull a deeper plough. Ploughing is key to break up and turn over the earth, destroying the root systems of weeds, disrupting the life cycles of pests and making it easier for seeds to germinate. Animals could also be used to help thresh grain, either by trampling it or in the later stages of the period covered by PC, powering a threshing machine. Note that these effects are stronger for growing grain than they are for rice which is more dependent on investments in irrigation and the like.

Having animals around also provides a source of manure, the main type of fertilizer used in pre-modern societies as well as being pulling carts to help transport agricultural goods.

There was a range of agricultural productivity in pre-modern societies. Poor subsistence farmers would produce significantly less than a rich landowner aiming to sell to the market. Bret Devereaux has a series of blog posts on pre-modern agriculture.

That is indeed a massive aspect, and don't forget the use of human urine from urban regions!
The biggest development though is the use of fallow crops (like clover, some forms of sorghum,...) that reintroduce nutrients into the soil. (The 'input' factor).
Variable crops were discovered to be useful throughout the centuries, but it was a technology that gradually expanded in the 14th and 15th centuries through both Europe and Asia iirc.

 

[zeruosi]

I still can't understand why the climate in Hebei and Guanzhong is drier than other places. This is obviously not very reasonable. Because from the map, this area does not appear to be drier than other regions. Instead, it is filled with rivers and swamps.
The Qiqihar region, marked as having the same climate, can be seen on the map to be relatively arid at a glance.

The Guanzhong Plain is not drier than other surrounding areas, and those areas that appear slightly yellow when enlarged are all farmland. And the surrounding areas of Qiqihar are semi-arid salt lakes and sand dunes.
The difference can be clearly seen from the map.

 

[zeruosi]

If potatoes were to be separate then cassava should be a separate good, too. The influence it had on China and East Asia in general was immense!

In my opinion, the influence of cassava in China is minimal. Cassava prefers environments with high temperatures and humidity. While it may have had a significant impact in Southeast Asia, in China, it is sweet potatoes rather than cassava that have had the main impact.
Sweet potatoes require less irrigation and are more tolerant of cold temperatures. Sweet potato vines and leaves can also be consumed, and do not have certain toxins like cassava.

Similarly I wonder why soybeans are separated from legumes, and why dates were split from fruits. They will probably have a flavor function, but still!

Perhaps because soybeans are the main raw material for tofu, other widely grown legumes may not have as high a protein and oil content as soybeans.(At least I haven't seen tofu made from chickpeas)

 

[Sulphurologist]

In my opinion, the influence of cassava in China is minimal. Cassava prefers environments with high temperatures and humidity. While it may have had a significant impact in Southeast Asia, in China, it is sweet potatoes rather than cassava that have had the main impact.
Sweet potatoes require less irrigation and are more tolerant of cold temperatures. Sweet potato vines and leaves can also be consumed, and do not have certain toxins like cassava.

Perhaps because soybeans are the main raw material for tofu, other widely grown legumes may not have as high a protein and oil content as soybeans.(At least I haven't seen tofu made from chickpeas)

I'm sorry, I meant Sweet Potatoes! Cassava was only introduced to China in the 1820s, my bad!
Cassava was very intensively grown in SE Asia, in the more tropical part.

Id like to include other tubers, like how the Hakka peoples were able to settle themselves deeper into Fujian and Huang He regions, where Rice and Wheat would not grow as well, but maize, Sweet potato and potato grew very well, but especially sweet potatoes. (Charles C. Mann, 1493 - ca p234)

Edit, the naming is more complex than I thought... Potatoes and yams are tubers, whereas taro and cocoyams are derived from corms, underground stems, and swollen hypocotyls. Cassava and sweet potatoes are storage roots and canna and arrowroots are edible rhizomes.

 

[Sulphurologist]

I still can't understand why the climate in Hebei and Guanzhong is drier than other places. This is obviously not very reasonable. Because from the map, this area does not appear to be drier than other regions. Instead, it is filled with rivers and swamps.
The Qiqihar region, marked as having the same climate, can be seen on the map to be relatively arid at a glance.

The Guanzhong Plain is not drier than other surrounding areas, and those areas that appear slightly yellow when enlarged are all farmland. And the surrounding areas of Qiqihar are semi-arid salt lakes and sand dunes.
The difference can be clearly seen from the map.

Well, not dissimilar how the Nile valley, indus valley and Mesopotamia are very fertile, despite the Arid climate.
There's access to surface water and ground water, which allows for irrigation.

 

[GenericUsername1444]

This is amazing!

It's nice to have a professional with us for such things.

 

[korandder]

I wonder if Root Vegetables would be a better name than tubers. This would also include various Old World crops like turnips, parsnips and onions that are less nutritious than potatoes but were significant food crops in varying places at varying points.

 

[Clever_dog98]

Soil suitability should definitely play a role as well, or at least especially fertile soils should get a boost. The black soil in Ukraine and Southern Russia, for example, can provide bountiful harvests of wheat, but was historically underutilised because nomads lived on that land. It's only when settled nations started pushing the nomads back and no longer suffered damaging raids that you see rapid population growth. It would be cool to see this represented in-game somehow.

 

[zeruosi]

I wonder if Root Vegetables would be a better name than tubers. This would also include various Old World crops like turnips, parsnips and onions that are less nutritious than potatoes but were significant food crops in varying places at varying points.

These should be vegetables rather than staple crops, and garlic has something to say!

 

[Sulphurologist]

I wonder if Root Vegetables would be a better name than tubers. This would also include various Old World crops like turnips, parsnips and onions that are less nutritious than potatoes but were significant food crops in varying places at varying points.

I think the old world root crops should remain separate from the New World ones.
Otherwise the impact of potatoes on the old world is nullified.

I was also thinking "hey, why is there no vegetables good", but the low caloric value of it doesn't mean much on a grand scale, and we sort of have fruit to compensate for that, in a way. Dairy is also included in livestock after all.

I'm fine with the term potatoes to avoid confusion, but the image and flavor text needs to include some sweet potatoes too to illustrate the diversity of the crop imo.