Land Biomes in Thrive

[Nein99]

There hasn’t been any discussion yet that I can find on how land biomes / patches will be defined in Thrive.

Recently I’ve embarked on a quest to define what future soundtrack themes we will likely need (forum post on this soon), and part of this requires knowing what biomes will be in the game. So, I’ve taken it upon myself to find a biome classification system we can use, and further consolidate that into a few biome types we can implement that represent the variety of land biomes on Earth.

My initial instinct was to use the Koppen climate classification system, but even though I love it, it’s pretty unintuitive and clunky to translate into layman video-game biomes.

Köppen climate classification

I think a better option for Thrive would be the Holdridge life zones system.

Holdridge life zones

In my opinion, the Holdridge system is much more intuitive and easy to translate to a patch map.

I’ve somewhat arbitrarily further consolidated the 38 Holdridge biome types into 12 more general biome types that we can use in Thrive:

Thrive Biomes

Which specific Holdridge biomes translate to which Thrive biomes may need tweaking, but I think that about 12 is the right number of distinct land biomes to portray. For context, we currently have 10 water biomes in the game.

This consolidation from 38 to 12 land biomes doesn’t mean we lose any resolution. Since we should have temperature and rainfall numbers for each map tile, we can always differentiate between, for example, a Subpolar dry tundra and a Subpolar wet tundra if we wish. I believe that we should show this extra nuance in the patch details screen or somewhere similar, and that the unique conditions in each patch should influence autoevo and gameplay. However, in the broader patch map view, those two biome types will simply be “Tundra” and have the Tundra icon.

I believe this classification system allows a good balance of real science representation and gameplay that’s simple to understand. I’d love to hear others’ thoughts on this, especially from our Theory and Game Design teams.

[Thim]

My first thought is that a lot of these zones seem to make assumptions on the plant life present. That’s probably true on Earth today, but what about in the past before flora we have today? What does it look like in Thrive if the rainfall says this patch is a rainforest, but no species that can create a canopy exist?

[Nein99]

Great question.

I think in the absence of “Trees”, the forest biomes would instead be scrubland, tundra, steppe or desert. It’s a forest if some definition of tree-like organisms exist, otherwise we would set a fallback biome for each biome type.

In the absence of any plants we could call “shrubs”, any shrubland biomes would instead be tundra, steppe or desert.

In the absence of any plants or topsoil, every biome would be some form of desert. Or, instead, we could perhaps add a “Barrens” biome, indicating that this land patch has no topsoil or plant-like organisms at all.

[Nein99]

Okay, so I’ve added three ‘Special biomes’ that aren’t covered by Holdridge for various reasons, that I think are important.

Wetlands are an extremely important ecosystem, but are based partly on geography, which is why the Holdridge system misses them.

Barrens are meant to represent land that has not undergone primary succession and does not have vegetation or soil. This will either be fresh volcanic islands, or any land before the evolution of land plants.

Caves are a unique environment also determined by geography, which is why Holdridge misses them.

This brings the land biome total to 15, or 14 excluding Barrens.

Prompted by @Thim I’ve also loosely defined what each biome becomes in the absence of trees, shrubs and any land plants.

As noted, grass actually evolved pretty recently in Earth’s history, emerging between 70 and 114 million years ago. For comparison, trees evolved around 400 million years ago, and shrubs likely earlier (though not too much earlier). Unless we want to ignore this fact in Thrive, this means that the Steppe biome or any grasslands would likely not exist until long after trees evolve in a given playthrough.

As @Juuzo_Lenz214 mentioned in the discord, we’ll also need some additional aquatic biomes for the 3D stages. I’ve gona ahead and added a few special aquatic biomes for this purpose.

The important aquatic environments I could think of that aren’t already present in the microbe stage are rivers, lakes, coral reefs, and kelp forests (called Aquatic Forest here to avoid referencing an Earth species).

This brings the total aquatic biomes to 14, the same number as land biomes (excluding barrens). Symmetry! Combined, that’s 28 biomes.

*Edit: I can’t count! There’s 11 biomes in the microbe stage, meaning 15 total aquatic biomes, and 29 total biomes excluding barrens

[Thim]

Ok, this still leaves us with the question “what is a shrub?” but I think we can cross that bridge when we get there. Two more questions off the top of my head:

1. If thick ice is directly on the ground with no layer of liquid water in-between, is that still an “aquatic” ice shelf, or is that something else?
2. What about mountains and similar features?

[C009]

@Nein99

I recommend instead of doing Before/After, use timeline fashion where barrens are leftmost and instead of said before/after, something like periods, but that’s just for readability. Anyways good job at classifying biomes depending on climate.

I would also add island to the list you posted above.

We also need to distinguish between biome and biome feature, e.g. river deltas - would it be biome or something contained within biome like coastal, estuary?

[Nein99]

If thick ice is directly on the ground with no layer of liquid water in-between, is that still an “aquatic” ice shelf, or is that something else?

I think that would be a Polar Desert. An ice shelf is specifically over water.

What about mountains and similar features?

Mountain biomes like alpine forests are accounted for by Holdridge. If you look at ‘alditudinal belts’ on the right, it shows which Holdridge zones exist at which altitudes.

this still leaves us with the question “what is a shrub?” but I think we can cross that bridge when we get there.

Defining “shrubs” and “trees” is totally arbitrary. I think the simplest definition could be “a shrub is a sedentary photosynthetic organism with X height” and “a tree is a sedentary photosynthetic organism with Y height”.

[Nein99]

I recommend instead of doing Before/After, use timeline fashion where barrens are leftmost and instead of said before/after, something like periods, but that’s just for readability. Anyways good job at classifying biomes depending on climate.

Good point. I’ve changed it to be chronological:

I would also add island to the list you posted above.

Islands are just tiny landmasses that have one or more biomes, such as tropical rainforest. I don’t believe having an Island biome makes sense, as the unique species and conditions on an island would already be represented by the fact that species populations and resource availability are simulated per patch.

We also need to distinguish between biome and biome feature, e.g. river deltas - would it be biome or something contained within biome like coastal, estuary?

I believe river deltas would be Wetland, probably a subtype of that biome (like how the Holdridge zone ‘Subpolar dry tundra’ would be a subtype of the Tundra biome). There may be some other environments I’ve overlooked.

[Deus]

One thing about the aquatic biomes is that I think certain patches we have currently, like hydrothermal vents, caves, and tide pools will become unique features of larger patches rather than entire biomes themselves. The implications of having an immense patch of hydrothermal vents and tidepools instead of the sea floor or coast they belong I think would be inconsistent with the more broadly defined patches.

I think we could benefit from making our own Holdridge zone diagram which account for the important attributes we will track and how we designate different levels of that attribute. For example, if I assume moisture levels and temperature are the most important characteristics in defining a land patch for Thrive, how many levels of moisture will we recognize - Super-Arid, Arid, Moderate, Sub-Tropical, and Tropical? Then after doing that with temperature, seeing how, say, a super-arid hot environment would be designated - perhaps a desert, as Holridge notes.

Like in other phenomena with Thrive that are very broad and diverse, we should probably start with a top-down approach in defining potential environments, going from as broad of descriptors to the narrowest level of detail we deem significant to represent. I think looking at things from that perspective could help us understand our spread of potential patches better and better determine what level of detail we want to go with in this topic. It could also help us understand how we’re, for example, designating which environments will allow which forms of biota to grow.

[Nein99]

You raise great points.

The implications of having an immense patch of hydrothermal vents and tidepools instead of the sea floor or coast they belong I think would be inconsistent with the more broadly defined patches.

I actually disagree for most of the microbe stage biomes. I think that because the Vents are a unique and very different environment than the surrounding Seafloor, they should count as a biome even in the 3D stages. My definition of a biome (vaguely) is a distinct environment with distinct resources for life. Vents have all sorts of compounds that the broader seafloor doesn’t.

Where I do agree with you is the Tidepool and Estuary. In the 3D stages, tidepools would probably just be part of the Coastal patch. I think the Coastal biome should persist into the 3D stages, as in my mind it represents the boundary between land and sea anywhere that isn’t a “wetland”, including the shoreline and the shallow water immediately next to it. This could perhaps extend to the edge of a continental shelf, I’m not sure. The relative shallowness of the water is what distinguishes the Coastal patch from Epipelagic.

I believe river deltas would be Wetland, probably a subtype of that biome

Estuary patches should be Wetlands in the 3D stages, at least once land vegetation arises. This is something I overlooked.

Underwater caves should usually have a different spread of compounds than the rest of the ocean, so I believe those should be a unique patch as well. (Though there’s no reason that small caves can’t generate underwater in other patches. Perhaps some of these could allow travel between a water patch and an underwater cave patch representing a whole cave system.)

I agree that making a modified/extended Holdridge zone diagram for Thrive might be beneficial. However, I’m not certain how the diagram itself it could be nicely extended to account for geography-dependant biome types like caves and wetlands. Caves aren’t even a surface biome. I’m a huge fan of elegance and consistency, but real-life biomes sort of evade simple classifications based on 2 or 3 statistics, since no planet is a smooth and featureless sphere with easy-to-predict conditions. For this reason, I think we’ll always have to have a few ‘Special’ biome types. I would love to eat my words if a game designer or theorist could design such a system however.

if I assume moisture levels and temperature are the most important characteristics in defining a land patch for Thrive, how many levels of moisture will we recognize - Super-Arid, Arid, Moderate, Sub-Tropical, and Tropical? Then after doing that with temperature, seeing how, say, a super-arid hot environment would be designated - perhaps a desert, as Holridge notes.

I’ve intentionally not decided what exact condition thresholds will exist between Thrive biomes, as I am leaving that for game designers closer to when these biomes are actually implemented. I’m creating a general framework that others can refine and that I can use to predict what soundtrack themes we’ll need. These are questions that should definitely be answered, however.

I’ve added notes to the Tidepool and Estuary patches on how they should change in the 3D stages, and I’ve also changed ‘Coral Reef’ to ‘Reef’. As @Buckly pointed out in the discord, reefs aren’t just limited to corals – oyster reefs exist!

[harmonyjpetersen]

I’d also like to point out that these are also distinctly earth (and mostly modern-earth) temperature ranges.

Earth has had much hotter and much colder temperatures, and there were many biomes that once existed that no longer do, as evidenced by the lack of tar pits so common during the dinosaur era; not to mention non-Earth biomes, such as the very real possibility in the search for life of tidally locked planets with habitable twilight zones. Molten seas might not have been considered a normal Earth Biome (well, until recently, Kurzgesagt had a nice video on the updates there… ) so for biome selection options, we might want to make sure to widen the range a bit to temperature extremes beyond our norm.

[Nein99]

I agree with @harmonyjpetersen that we’ll need to extend this system to account for environments on planets with exotic or early-Earth conditions. For now, though, I’m focusing on how to represent Earth’s modern environments, and a robust classification system that can handle all that should hopefully be not too difficult to extend to account for alien environments.

A community forum member (Poodelicus) has given some great feedback on what we’ve discussed so far: Biomes - #3 by Poodelicus - Future Game - Thrive Community Forum

I incorporated much of their feedback, and in my response talked about some other things such as tar pits as well. Here is my response, reposted to the dev forum for posterity:

This is some excellent feedback, thank you for your input!

First, my initial reasoning for having a ‘Special Land Biomes’ category is that some environments, like wetlands and caves, cannot be defined with humidity and temperature alone, but are also defined by their geography. You’re right that Wetlands are a transition zone between aquatic and land biomes, meanwhile caves aren’t on the surface at all, instead being totally unique spaces underground. With that said, I will admit that I don’t care too much for having a ‘Special’ category of biomes that don’t fit in the system, and I’ll integrate some of your proposed solutions at the end of this post.

The ‘Special Aquatic Biomes’ category, now that I look at it again, doesn’t hold up to my original logic. This category is for aquatic biomes I added that aren’t in the microbe stage, but that division is totally arbitrary. I realize that we lack a way to quantify boundaries between aquatic biomes, with the exception of the -pelagic biomes which are defined by depth. If we define an ‘Aquatic Biome’ as a habitat dominated by liquid water, there is at a glance no reason for having two separate categories.

It seems that in the real world, aquatic biomes are usually divided broadly into ‘marine’ and ‘freshwater’ categories. I think this is a good place to begin. For a specific definition of ‘freshwater’, I am using this definition from USGS:

Freshwater - Water that contains less than 1,000 milligrams per liter of dissolved solids.

I am considering brackish (semi-salty) water to be saltwater. Ideally, we will track the salt content in each aquatic patch, just like moisture in land patches. 1,000 mg/L is roughly the same as 1,000 ppm, and ‘dissolved solids’ can be simplified to mean salt.

Based on all of this, we can divide the aquatic biomes as follows:

The Marine category includes saltwater biomes. The Tidepool patch (which to my understanding is different from Tidal Wetlands) is a subtype of the Coastal biome. When the player reaches the 3D stages, ‘Tidepool Coastal’ would be a Coastal patch with an abundance of tidepools, which at the scale of the 3D stages would be terrain generation features rather than entire patches. Things like tar pits (which harmonyjpetersen mentioned on the Dev forum) would also be terrain generation features, and perhaps could also constitute a biome subtype(s) if they are abundant enough.

I think Wetlands, though diverse, are still important enough and easy enough to define as a group (transition environment between water and land) that they should be a major biome type. As such, I’ve added Saltwater Wetland and Freshwater Wetland to their respective categories. Each will have subtypes based on the kinds of species present (woody vs. herbacious plants), the pH level, salinity, and so on, just like how terrestrial biomes will have subtypes based on moisture, temperature, native species, etc.

Based on Holdridge, a ‘Scrubland’ patch might be ‘Tropical Desert Scrubland’ or ‘Tropical Thorn Woodland’; based on our eventual system, a ‘Wetland’ patch might be a ‘Brackish Estuary Wetland’ or a ‘Seasonal Freshwater Floodplain’. As with land biome subtypes, I am leaving the specific numerical boundaries between wetland biome subtypes to be determined by more knowledgeable theorists and the people who will implement these biomes into the game, but this is a good start.

You may notice that the Underwater Cave biome is missing from the Aquatic categories. This is because I’ve removed the Special categories entirely, and made both the Cave and Underwater Cave biomes part of the new Subterranean category. Barrens, and all of the former land biomes, now belong to the Terrestrial category.

This means there are now four biome categories:

• Terrestrial
• Subterranean
• Marine
• Freshwater

I think this is altogether much more grounded approach to classification.

In review, from your suggestions:

I believe it would be better in Thrive to define them either A: in a separate category (Aquatic, Terrestrial, Wetlands, and Subterranean), or, better yet, as a set of modifier types.

I’ve incorporated your suggestion to make Subterranean its own category, as I think that makes sense. I’ve also renamed the Land category to Terrestrial. As for the aquatic biomes, I felt that Marine vs. Freshwater was a better major distinction than ‘Wetlands’ and ‘Aquatic’, because in real life it seems that’s the primary distinction people go with, and as you mention in your other suggestion, I believe that the various kinds of Wetlands work better as biome subtypes or modifiers.

Finally, since I’ve been throwing around words like ‘patch’ and ‘biome’ a lot, for my own sanity I’ve added a Definitions section at the top of the spreadsheet for all the important terms. The definitions themselves may need some work.

Here is the spreadsheet where you can view everything put together: Thrive Biomes - Google Sheets