Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[wierd]

I sent NW_Kohaku a link to a short book on organic composting just a little bit ago. (Free on gutenberg!)

My mom also does extensive organic horticulture, and has for years.  I haven't had the opportunity to read this thread properly yet, but will offer advice after doing so too.

Currently too busy making industrial blueprints for my employer.



OK, reading the first bits with a quick glance, basically you need to impose limits without breaking the system spectacularly. This means you need crop management.

An example of modern failures of crop management, is the over-use of nitrogen fertilizers. They allow explosive growth of green plants, by permiting ready protein synthesis, but this also affects soil microbes, which use the nitrogen to destroy the soil's carbon based organic sponge, or humus. This results in a lower and lower capacity for the soil to retain and provide nutrients over time, resulting in less and less nutritious produce over time. Ideally, farmers would replenish this lost humus through liberal application of compost, but supplying compost at those scales to sustain massive intensive agriculture is simply not economically feasible.  This has resulted in the steady decline of fertility in modern agricultural regions over time. In short, the "green revolution" using synthetic fertilizers is simply not sustainable.

For a more historically accurate picture, we can look at crops as being one of 3 basic types: soil buildng, soil neutral, and soil depleting.

 Legumes and certain grasses are soil building. These are things like clover, alfalfa, and tall prarie grasses. These plant forms either fixate soil nutrients directly, such as legumes, or produce copious amounts of organic sponge raw materials, like gras straw and as such, build up the sod layer, making it more useful for agriculture.

Soil neutral plants only make use of an established sod layer, but don't disrupt the soil's equilibrium in a serious fashion. Most weed plants and leafy herb type plants are of this variety. They have requirements for different types of soils, (for instance, sorrels and mallows tend to thrive on poor soils, while cooking herb plants prefer richer soils.) But don't really change the biome significantly, and form a homeostasis with their environment when competing with other similar species in the ecosystem.

Soil depleting plants essentially "rape" the soil, draining it of vital reserves of nutrition, and force the farmer to have to correct the imbalance by growing a soil improving crop soon after.  These are things like cotton, especially, but also includes things like major vegetables, and some cereal grain crops.

As previously stated, the major modern deleterious factor is the destruction of humus in the soil by intensive cultivation coupled with nitrogen fertilizers. All plants require nitrogen to produce protiens within their cells, which is essential for them to live. Nitrogen from the air won't do, because it is too difficult to overcome the high energy triple bond of the N2 dimer, at least fr the plant. This is where soil building plants like clover and alfalfa come into play. As legumes, they serve as a symbiont for a very vital soil microbial fauna that is able to take nitrogen dimers from the atmosphere, and combine it with oxygen and alkaline earth ions to produce nitrate compounds, which they share with their symbiotic plant partner. (The plant provides them with a stable growing environment, reliable moisture supplies, and simple sugars for food.) The maximal benefit of these crops comes from composting the while plant bodies of these plants over the year, by tilling the plants under at the end of the season, and then allowing the field to lie fallow. This returns the nitrogen and soil humus reserves to the soil, with a surplus from the fixation processes of the legumes.

Depending on the aggressiveness of the depletions of the actual crops grown per crop, build, fallow cycle, differing amounts of building and fallowing will be required for sustained agriculture.

Coupling that with seasonal restrictions, and entire year fallowing times, ad you will quickly find that you need much more cropland that DF currently requires to provide stable food production in the long run, without destroying the soil. (Completely depleted soil is almost impossible to get started again, without very intensive and laborious efforts. In fact, fully delepted soils won't even grow weeds for cover, and will become blowing deserts!)

What needs to be done here, is to segregate surface crop cultivation, from subsurface crop cultivation, because the type of organism cultivated in those growing conditions have VERY VERY different requirements. Up until now, I have been referring only to surface horticulture.

Mycological species are NEVER producers. *all* species are soil depleters! Since they do not get any of their energy from sunlight, 100% of their biological energy comes from the decomposition of already existing organic materials.

In the case of DF, the native biome for these organisms are vast underground cavern systems with interspersed volcanic vents and cavern lakes. As such, the primary producers of this environment are likely to be autochemotrophs, which as far as I know, are all monocellular. These organisms usually have a sulfur respiraton cycle, and make use of mineral and geochemical energy sources to produce primary nutrients, and build up organic humus of sorts through their accumulation, and decay on the bottoms of these cavern system lakes and rivers. Periodic flooding must somehow trasport these organically rich silt deposits onto the cavern layers adjacent, where secondary fungiforms make use of it to produce food products for cavern organisms, who transport the nutrients further afeild via their droppings. Each time, energy will be lost from the natual processes of the organisms involved, so the primary concentration of nutriton would be found in the cavern lakes and rivers as silt.

Sustainable agriculture of subterrene crops would therefore require either periodical inundations of the croplands with fresh layers of cavern silt, or direct introduction of organic materials from the surface, such as compost, to sustain the nutritional and biological requirements of these crops. This means that unless very elaborate methods to artificially induce seasonal flooding with cavern water sources is enacted, underground crop production would be tied to the hip of surplus nutrients produced above ground, and transported below, to be sustainable.

We will have to conjure up numbers for the rates of consumption, ad production of these nutrients to establish a growing chart.

[NW_Kohaku]

Well, for starters, as I've tried to explain to some of the others in this thread, even I do try to keep things as relatively simple as possible while accomplishing the goals of the thread.  Hence, I've fudged together a few concepts into single variables that simplify things a little more than realistic.  (Even doing so, I'm still at 12 variables...)

For example, I have a single "biomass" variable that represents both only-partially-decomposed material (especially for growing mushrooms) at high values, heavily carbonaceous humus at mid-range values, and then a dearth of carbon at low values.  Likewise, colloid soil sizes and drainage are all reduced into a single variable gauging a general sense of sandy vs. heavily packed clay soils. 

That said, maybe there's a case to be made for something like splitting compost into its own variable, so that the notion of using something like a plump helmet to build up humus for later crops could be more heavily explored, but I'm still not terribly sure how well that would work, mechanically.


Almost all of the research I did for this thread was on organic farming since, almost definitionally, that's all dwarves are going to have access to.  (Barring some mining for apatite or a few other ways to find non-renewables naturally and magma farming.)  Hence, Haber-Bosch nitrogen overabundance isn't going to be a major factor. 

The only fertilizers that are going to be sustainably available will be the kind you compost yourself, and so it's going to be heavily carbonaceous.


We're going to assume that subterranean crops are simply not based upon any known (or even necessarily rational) ecosystem, and as such, I've taken to saying that the underground crops don't run on photosynthesis, but "xenosynthesis" instead, and have their own magic-based ecosystem, where magic levels are tracked, and you can actually have "magic pollution" or possibly "magic erosion".

This is mostly done because there's just no way you can justify having decomposition of materials from the surface or chemosynthetic organisms that far from magma vents creating such a robust ecosystem.  (Especially since that "ecosystem" already contains things like amethyst men, cave dragons, and magma crabs that do things like live inside magma and spit rocks.  Plus, the underground mushrooms already include the Nether Cap, which magically keeps temperatures always at exactly 0 Celsius.  There's no point in pretending underground crops will be anything but magic.)

Aside from that, even the chemosynthetic creatures we're going to assume can somehow symbiotically live inside of a "plant" that can be cultivated the way that the tube worms symbiotically use those monocellular chemosynthetic organisms.

Ultimately, when it comes to underground ecosystems, much like Discworld, we're just going to have to desperately backfill rationalizations into an originally arbitrary and obviously magical system to have something that makes contiguous sense.


Beyond that, much of what you and I have written are in agreement.  I have yet to really read anything in the ebook you sent, however.  (And my eyes are starting to fail me for the night, so that will have to be a weekend endeavor.)

With that said, I'm less concerned with absolute accuracy and more concerned with accomplishing something that is automatable, not routine, diverse, interesting, not completely front-loaded on information for a new player, and which leads to the potential for any given fortress to be radically different because of the choices players made regarding how they would farm with as much realism as I can cram in beyond that.  If we get even half of this stuff into the game, it's already a far better teaching tool for how real-life agriculture works than any system in any game shy of an out-and-out farm simulation game. 

[wierd]

At the end of chapter 2, and the beginning of chapter 3, it begins dealing with carbon to nitrogen ratios of various plant materials, and covers the issue of supplying sufficient nitrogen to a composting system in depth. It lists several types of common leaf and plant materials for reference, and their ratios.  That should be somewhat helpful in building a model, at least for surface crops.


As for subterene crops, there is a potential-but-crazy mechanism that could account for at least some of the missing energy, and allow for a wider area of nutrient distribution from the volcanic vents.

Sometime last year, a group of researchers discovered something unusual about LEDs. They can operate at apparent overunity, if held at just the right voltage, by tapping thermal oscillations to push electrons over the bandgap. This means that more photonic energy is produced than comes from the electronic energy supplied, and the LED gets slightly cooler as a consequence.  Here's a press release from phys.org

There seems to exist an exotic subterene myconoid species that would seem to fit the bill here: Nether Caps.

Through some mysterious, and unexplained mechanism, they are always just barely above freezing. Open magma vents in the cavern systems would cause a very intense heat distribution, especially with volcanically active cavern lakes and rivers, which would introduce copious quantities of water vapor into the cavern atmosphere, which has a very high specific heat.  Yet, somehow, the caverns arent a hellish, nightmarish, hot and toxic environment, like they should be.  If we assume that these are our 'xenosynthetic' secondary producers, we have a potential source of both cavern climate control, atmospheric processing, and primary nutrient fixation in the cavern ecosystem. (Essentially, the "fungus" uses a special organic semiconductor molecule (many species actually produce organic semiconductors, including humans. Melanin is one such beast.) to make use of a low level ion gradient between the humus layer of the cavern silt, and it's own internal humors, by absorbing dangerous acidic compounds from the atmosphere, and using the voltage difference that makes to supply an initial electrochemical charge to that complex, which then makes use of the copious heat generated in the cavern by the volcanic activity to drive additional positive entropy reactions via this "over unity" mechanism.

Hoorah modern science.

The fruiting bodies of fungiforms are by no means the true manifestation of such organisms-- which are really teeming networks of mycelial fibrils that entwine and permeate their substrates.  Fruiting bodies are just mass aggregations of these mycelial fibrils that bind together in a communal mass between two genetically different mycelial cultures, which is who the fungus reproduces sexually to produce genetically diverse spores.  If we assume that the activity manifest at high levels in the nethercap wood, is conducted at widespread, but low levels in the soil beneath, out of sight, we have a direct food source that parasitical plant species could tap.

[wierd]

We really dont need to track certain variables, like mineral nutrition, really.

All we really need, are total energy in (sunlight, thermal, geochemical, et al) energy cost of reaction per trophic level, (EG, the growth of the plump helmet has to be factored as well, when figuring its energy content), total energy out, and energy capacity.

EG, we can 'fudge' soil depletion as a simple factor of how much energy the soil can supply, VS what can be supplied by sunlight, and what gets stored in a crop's biomass upon harvest.

EG, ratweed does not deplete the soil, because it can get 100% of its energy needs from the sun, but needs a baseline energy capacity to grow on the farm plot. It does not retain a very large amount of energy, which is why it produces a crappy alcohol. This also makes it a poor composting material, since it doesnt supply much energy in the compost.

while RopeReed may be soil depleting, requiring more energy than it can get from sunlight, but having a very high energy density per produce unit-- which is why it makes AWESOME alcohol.

We could say that bladeweed is a soil improving crop. It absorbs all the energy it needs from the sun, is implied to be nitrogen fixating, and produces a large amount of medium energy density biomass. Not suitable for alcohol, because most of the energy is protien. Instead makes alternative herbivore food, in addition to yucky green dye.

We could say similar things about hide root, and imply that it is like fibrous rhizome. Not suitable for consumption, but high in biomass energy and thus, good for composting.

Longland grass should probably be soil depleting, being a foodcrop that is high energy, and also an alcohol source of reasonable quality. Requires rich soil

Wild strawberry should be a low yeild crop, to make up for its relative high density-- while being soil neutral. This is in line with actual wild strawberry plants, which produce very few fruits.

prickle berry and fisher berry should be soil depleting.

Sunberry should be RADICALLY soil depleting, and prolific. I mean, practically murders soil. balanced by being so horrible to the soil, that you cant grow much else besides ratweeds afterwards.

Sliver barb should be soil building, with a radically low soil nutrient requirement. (useful for building up nearly dead soil) It should also be low yield. Its rarity means using it to revitalize soil likely only happen on blasted, horrible, dead and reanimating biomes, to which it is naturally indigenous. Valuable black dye is useful to trade for medium energy density crops, for immediate composting to help improve the soil more quickly.

whip vine can be the exception that proves the rule. I always imagine Kudzu as the basis for it. A high protein fodder, it is highly abundant, but is slightly soil building, but only just barely so. Makes really good compost, makes good dried herbivore food, makes a protein rich flour, is high sugar, and makes good wine. Balanced by being difficult to get. Requires already rich soil to grow.

This reduces the required variable count down to just 2 variables. Energy cost, and energy capacity.

[zwei]

In gist, you want system that allows easy farm setup for early fort and easy to set up fire-and-forget, but grossly inefficient farm.

But if player want to reap benefits, he should expend major effort and deal with issues that frequently arise.

I think that it is similar to forges: you can either use standart forge, but it is inefficient and short trem solution. In the end you want magma forges which come with their own set of dangers and challenges and possibly inefficiencies too (trade making fuel for long trips to magma sea forges or build expensive and labor intensive pump stack).

Frankly, given amount of variables discussed here opposed to simple fuel in case of forges, first and most important part of this suggestion should be new noble, "Lead Farmer", which would periodically check fields, available crops and make suggestions to alter what is grown to increase yield, demands on fertilizer production and import of seeds of plants he would deem necessary for production.

"Farm plot #11 currently growing Longland grass should be planted with Bladeweed next season to restore soil."

His suggestions should not be perfect (but improved by Analytical Ability, Intuition and Creativity attributes and Organizer, Farmer and Herbalist skills) so that farming minded player could make his crop cycles much better at cost of learming stuff behind them.

[10ebbor10]

This reduces the required variable count down to just 2 variables. Energy cost, and energy capacity.

Sadly, that reduces the gameplay options quite a bit, turning the farming in yet another simple optimalization thingy. (Since both variables are so interlated that they can basically be seen as one)

As for the new Noble Lead farmer. I don't think it will be so usefull, after all it's a suggestion that discourages people to learn more about the underlying system, as well as being rather complicated to implement. I'd just integrate it with the interface (ie, allowing us to sort things by different nutrition value, or even a more complicated searching thing that allows to search for multiple things at the same time*).

* Ie, I want a crop that grows quickly, in winter, doesn't use much of X, Y,Z, replensishes A and B and suchlike.

[NW_Kohaku]

In gist, you want system that allows easy farm setup for early fort and easy to set up fire-and-forget, but grossly inefficient farm.

But if player want to reap benefits, he should expend major effort and deal with issues that frequently arise.

I think that it is similar to forges: you can either use standart forge, but it is inefficient and short trem solution. In the end you want magma forges which come with their own set of dangers and challenges and possibly inefficiencies too (trade making fuel for long trips to magma sea forges or build expensive and labor intensive pump stack).

Frankly, given amount of variables discussed here opposed to simple fuel in case of forges, first and most important part of this suggestion should be new noble, "Lead Farmer", which would periodically check fields, available crops and make suggestions to alter what is grown to increase yield, demands on fertilizer production and import of seeds of plants he would deem necessary for production.

"Farm plot #11 currently growing Longland grass should be planted with Bladeweed next season to restore soil."

His suggestions should not be perfect (but improved by Analytical Ability, Intuition and Creativity attributes and Organizer, Farmer and Herbalist skills) so that farming minded player could make his crop cycles much better at cost of learming stuff behind them.

I remember discussing suggestions like that, but it's been 2 years since this thread started...  Going back over the thread, though, Khym Chanur had a suggestion that I adopted where players can look at a screen where you can know what crops your home civ grew in what cycle, as well as the farming techniques of other known civs.  (Which only helps if you are in semi-familiar conditions, of course.)  I know I talked about things somewhat similar over in Class Warfare, as well, since reintroducing guild leader nobles that do things was part of the whole point, there.

Part of the problem, though, is that making a Lead Farmer that's worth listening to would be extremely complicated to program, as it would require making assessments of farming as a math problem when I've specifically set up farming not to be a simple math problem.

Having a mechanic where you can see, at the "I'm choosing what to plant" screen what crops the soil will currently support, what crops will take what amount of fertilizer before being able to be supported, and what crops are just not fit for the current soil or environment, no matter how you try to dink with it, is both easier to accomplish and also perhaps a bit more informative to the player for understanding how the whole thing works. 

(That is, rather than just telling you what to do, you know what you can do easily, and what is more difficult to do, and why it's more difficult.)

You might want to look at the (admittedly rather complex) Interface section where it covers the scheduler, and also see the mock-up I made for it:

Spoiler: Scheduler Window (click to show/hide)

[wierd]

The idea about tieing into energy available as the kicker, implies that available sunlight will be different based on region.

For instance, a temperate region is temperate, because it has seasonal variation in intensity of sunlight, which alters the climate, which is why there is a winter.

If you couple this with a moisture requirement, and better weather simulation, you will directly end up with fertile temperate bands, and equitorial deserts, like is currently seen on earth, but with a much simplified system.

It is important to remember that the dwarves themselves wont know about the carbon to nitrogen ratios of their crops-- only that certain crops leech the soil, where others build it back.

The idea is to carefully tweak the sunlight requirements of plants to be "optimal" and "easy" only on temperate biomes. This way, embarking on that freezing glacier really does mean "you know what, you arent gonna grow food here."

Same with embarking on an evil desert-- low soil energy, low moisture == VERY HARD to build up the soil, and as such, VERY HARD to survive on.

The idea is that something like whipvine would be slightly soil building in the summer of a temperate biome, but aggressively soil depleting in the winter, while still retaining it's basic energy needs. The plant's needs dont change, the environment changes with the seasons, and with it the energy supplied by the sun. Also, some crops shouldnt be plantable in the winter anyway. Berries for instance.  I'd restrict winter months to longland grass and hide root only, for instance.

This could be further tied in with weather, as a local variable on sunlight and moisture availability. Rain increases soil moisture, but blocks out the sun for awhile, making the overall energy needed to grow the plants have to come from the soil's reserves.  (this isnt exactly accurate, but would make the farming system much more variable, and less "Fire and forget")

[NW_Kohaku]

As for subterene crops, there is a potential-but-crazy mechanism that could account for at least some of the missing energy, and allow for a wider area of nutrient distribution from the volcanic vents.

Sometime last year, a group of researchers discovered something unusual about LEDs. They can operate at apparent overunity, if held at just the right voltage, by tapping thermal oscillations to push electrons over the bandgap. This means that more photonic energy is produced than comes from the electronic energy supplied, and the LED gets slightly cooler as a consequence.  Here's a press release from phys.org

There seems to exist an exotic subterene myconoid species that would seem to fit the bill here: Nether Caps.

Problem with that is that Nether Caps are not just infinite heat sinks, they're also infinite heat fountains - they are always 0 Celsius, even if outside temperatures are well below freezing.  That is, in sufficiently cold terrain, it would make sense to make hostile environment clothing/tents with an outer shell of nether wood, since it would actually insulate you from the extreme cold.  (And in extreme heat, it's of course just as good.)

The seeming canonical explanation for how Nether Caps work is that they are literally portals to a "Nether Dimension", which serves as a perfect heat exchanger to a well of infinite mass and specific heat capacity.

I.E. it is an "autotroph" in that it can introduce energy to an ecosystem, but it relies upon magically bridging in energy (or, usually, OUT energy) from another dimension. 

Plus, keep in mind, the wood stays 0 Celsius even after the mushroom is chopped down and dead for quite some time.  The heat-exchanging capacity of the wood is not a biological process. 

In fact, the tiny bridges to the nether dimension may not even be formed by the nether cap itself.  It may just be a fungus that grows around mollecular-scale bridges to the nether dimension that already exist, finding some way to xenosynthesize energy from the heat exchanging properties, or something of chemosynthetic value that they can extract through the portals.

[wierd]

That's just quibbling over the mechanism--- I just tried to supply a possibly plausible (yet absolutely absurd at the same time!) one for how the cavern ecosystem could be sustained over large areas, and climate regulated.

Substituting your explanation, we have other issues-- it means that the nethercap concentrates this "netherworld portal" meta-material in its fruiting bodies, which would then deplete the environment of this exchange medium.  We would have to make the nether caps much more prolific, and outright delicious to cavern flora to provide a mechanism to redistribute it back into the cavern ecosystem, in order to balance that out.

(EG, imagine-- Dwarves harvest the loverly nethercaps to make dwarven icechests, because you just add water! This means that they are taking the meta-material out of the environment with thier economical activities, and slowly killing the cavern ecosystem by doing so.)

As for how to biologically capitalize on an infinite heat sink/fountain metamaterial--- the two thermoelectric effects already have this covered. (Seekbeck and Peltier)

This would be a less radical explanation for the infinite heat fountain characteristic of the nethercap besides being a netherworld metamaterial absorber--- The absurd degrees of energy conversions of the two above thermoelectric effects can be tied directly with the carnot efficiency, and its TOTALLY BONKERS WRONG values in the DF universe. (EG, the same crazy physics that enables purpetual motion water machines enables purpetual motion thermal engines too, and must-- because they are both in reality just heat engines when you get down to it.)

[NW_Kohaku]

At the end of chapter 2, and the beginning of chapter 3, it begins dealing with carbon to nitrogen ratios of various plant materials, and covers the issue of supplying sufficient nitrogen to a composting system in depth. It lists several types of common leaf and plant materials for reference, and their ratios.  That should be somewhat helpful in building a model, at least for surface crops.

Well, thing is, this thread already has a model.  There are plenty of concepts here to work with.  I'm more looking for flaws in the model, or ways to accomplish the same goals with less complexity.

We really dont need to track certain variables, like mineral nutrition, really.

All we really need, are total energy in (sunlight, thermal, geochemical, et al) energy cost of reaction per trophic level, (EG, the growth of the plump helmet has to be factored as well, when figuring its energy content), total energy out, and energy capacity.

...

This reduces the required variable count down to just 2 variables. Energy cost, and energy capacity.

As I've said to many of the other people who want to reduce this down to just one or two variables, there's more to this than just trying to make something simple that gets a message of "soil fertility" across, I'm also interested in the gameplay impact of these choices.

In other words, I don't want reality, or a "close enough fudge" of reality, I want something that makes for a good game.

To repeat the basic explanation for why this suggestion is the suggestion it is, I am trying to accoplish the following goals:

Alright...

I started just writing out a new one, and wound up creating something still kind of long, but close to as manageable as I could get without completely editing out large chunks of the conversation.  (Actually, I cut out the parts that were lead-ins to other possible overhauls, like magic biome types, and dietary models.)

The tl;dr Version:

Spoiler: The problems this tries to solve (click to show/hide)

The purpose of this thread is to solve several problems with the current game:

• Problem 1: Farming now is, generally, easy and boring.
  
  • Farming needs to be more interesting, with more crops that are different.
  • Current crops (especially above-ground) are nearly indistinguishable in how they are grown.
  • Farming should include more physical types of farms and farming techniques.
  • Different climates and biomes make no difference in how you farm or what you farm.
• Problem 2: Farming is currently a front-loaded challenge
  
  • We have a "learning cliff" currently - you have to know everything to get anywhere in the game.
  • Once you know everything, the game is too easy, all problems have been solved.
  • The goal should be to make something simple to start with, so any random new player survives their first year, but progressively more difficult as time goes on, challenging those who have mastered the basics
  • It should be possible for players to get by without knowing terribly many tricks, but with rewards for the player who truly masters the system, in much the same way as dwarfputing.
• Problem 3: Deep without micromanagement or pointless complexity
  
  • All automatable things should be automated.
  • Anything that is a simple "what's the quickest way to make the most food" math problem is a calculation, not a choice, and that is automatable.
  • Once all automatable functions are out of the way, high-level management choices should be made as a matter of a strategy, based upon comparing difficult-to-compare-mathematically things, such as short-term payoffs vs. long time setting up but better long-term payoffs.
  • The game needs to do the basic math calculations itself, but leave the grand strategy to players.
  • To keep players engaged, occasional challenges should arise to bring player attention back to the farms, but not in a micromanagement way.
• Problem 4: Creating a Medieval Fantasy Ecosystem Simulation
  
  • While abstracted, the game should leave players with the sense that this is an actual ecosystem, where the changes they make impact their world.
  • This can mean clear-cutting a forest leading to soil erosion that means no more "free trees", for example.
  • This can also empower players to purposefully alter the ecosystem to better suit their needs if they have a mastery of the concept.
  • If the game models real-world systems, even if abstracted, it creates learning opportunities through play.

The reason this thread is long is because it is arguing for why most simple solutions do not actually address all of these problems, and why a marginally more in-depth solution is needed.  Specifically:

Spoiler: The solutions proposed (click to show/hide)

• Solution 1: Multiple soil quality variables
  
  • Multiple variables for soil quality (instead of just a simple "fertility") mean that different methods of fertilization can be used without being a simple calculation. ("What gives me the most bang for my buck?") You cannot rely upon a single fertilizer type, you have to use several.
  • If you have differing amounts of different kinds of fertilizers, crops that consume different nutrients become more or less attractive from a strategic perspective - depending on certain types of crops means depending on keeping the flow of certain fertilizers moving.  This is a strategic, non-micromanagement choice, whereas a single fertility value simply requires automating the re-fertilization.
  • Some soil quality variables can be easy to change (like how recently plants have been watered) but others much more difficult (like if the soil is highly acidic), and some may be specific to the embark site (no growing tropical plants in an arctic tundra).  This means different embarks are going to require different crop types and farming methods that support them.
  • Several of these soil variables, especially the nearly impossible to change ones, may add variables to the game, but not terribly much complexity, as players will have few reasons to actually worry about soil acidity or salinity once you have already picked a plant suited for your soil.
  • Pollution variables can exist as a means of making players choose between short-term gains and long-term damage to their fields
• Solution 2: Pests
  
  • Pest creatures, from fungal infections to caterpillars eating leaves to locust swarms, can exist as a "disaster" form of macro-problem for players to deal with, keeping farming challenging and slightly unpredictable.
  • They can be controlled through both limiting the number of crops of a certain type (reducing the exposure to a single crippling pest infestation) and using natural pesticides - the predators of those pest creatures, or fragrant herbs that repel bugs.  (Birds, bats, and spiders eat insects, many herbs use pheromones to repel insects or attract predators of pests like wasps that lay their eggs inside caterpillar bodies.)
  • Since pests are not seen on-map, the relative population numbers of pests can be tabulated invisibly and infrequently, saving on processor power and not appearing complex to players.
  • Solutions to pests can be easy to understand, if not necessarily easy to implement - bat or bird nests near crops with birds/bats in them will keep pests down.  You then need to acquire and care for those creatures, however.
  • Since pests, especially diseases, are attracted to large, single-crop farms, it is a problem more for players who keep using simple farming methods for years, so it doesn't overwhelm players on their first few years, but challenges players in established forts to diversify.
• Solution 3: Different methods of obtaining Fertilizers
  
  • Crop rotations with crops like beans that are nitrogen fixators is a common and simple solution to one type of fertility deficiency, but not all.
  • Decomposing organic waste (including corpses, grinding down bones, and manure) into fertilizer is the most sustainable and reliable method.
  • Some fertilizers can be mined or traded for, but they are of finite or unreliable supply.  Minable Sulfur can be used to change soil acidity, and make some soil suitable for different crops long-term, however.
  • Exotic fertilizers, like magma, can be used, but require more elaborate engineering, and may cause soil pollution that limit what crops are viable
• Solution 4: Water Management
  
  • Water is consumed by farms.  This means you need to constantly bring more water into your fortress.
  • Water generation from off-map or aquifers would be slowed down, to make there be limits on how much water you can use.  This would mean desert embarks are tougher.
  • More elaborate systems are more efficient but take more time to set up and reward player engineering skills with better farm yields.  Simpler versions like bucket-watering are faster to set up, but consume more labor.  Newbie methods are not "wrong", and don't lead to dying, but there is strong incentive to learn.
• Solution 5: Polycultures - plants affect other plants
  
  • Plants can be planted with/near other plants, such as the fragrant herbs that repel some types of pests, or ground covers that reduce water consumption.
  • Plants like beans that add nitrogen to the soil can be planted with some crops instead of just in a cycle - supplying more nitrogen even as the crop grows.
  • It is an advanced farming method for advanced farmers, but not necessary for basic farming - an "advanced level" for people who master the system.
• Solution 6: More farmable materials
  
  • If you can farm trees for wood, plants for cloth, flammable oils, medicinal herbs, and many other resources besides just food (basically, anything that isn't mined can be farmed), then there is a reason to constantly upgrade farms - you want more than just basic survival on food, you want more stuff.  This lets the complexity of a farm for food become easy, but having to sustain more farms for more crops for more uses means that you are encouraged to farm more complex methods to gain greater yields for non-essential products.
• Solution 7: Nature plays by the same rules (but abstracted more)
  
  • Non-farm soil tiles have a generalized region that shares the same fertility and pollution values as farms.  This means that when grass/shrubs/trees grow, it takes nutrients from the soil.  When they burn or are decomposed, it returns nutrients to the soil, but if you harvest from the land without returning, then the soil loses fertility just like a farm.
  • Trees are therefore not an infinite resource anymore - you may have to farm them, yourself to get greater yields.
  • Wild animal and livestock populations play by pest rules, to a degree - you have control over what is attracted to your land, and livestock can get parasites.
  • You can eventually turn a land into a place where nothing grows or functionally change its biome if you work hard enough at it.  Garden of Eden or Hellscape, choice is yours.

I'm still trying to cut off some of the most common arguments, even in tl;dr mode, so I could edit it down more from there if I wanted to, but then have to keep going back to explaining why this increases challenge over time...


Also, as one last thing, Toady's already planning on throwing something like this into DF, as seen on the dev pages:

Farming Improvements

• Soil moisture tracking and ability to moisten soil (buckets or other irrigation)
• Soil nutrient requirements for plants and nutrient tracking to the extent the farming interface can provide decent feedback for you, fertilizers can reflect this
• Harvestable flowers and fruit growing on plants, ability to plant trees
• Weeds
• More pests

When asked about it, Toady also said that he was interested in making the game as realistic as it could be while still being manageable, as his grandparents were farmers, and he had an interest in the topic. 

So, by and large, I'm not actually suggesting many concepts beyond what Toady was already considering doing, but that I'm suggesting ways of implementing the concepts in a way that makes the game non-micromanagey and as interesting as possible for players.

A simple couple variables doesn't accomplish the goal of not making this game nothing more than a routine optimization strategy, as 10ebbor10 said. 

Each variable introduces the chance for each fertilizer to actually be differentiated. (I.E. what you were talking about with the difference between chicken waste and horse waste, and the difference between compost and hot manure - that takes variables to differentiate what those mean.) This means that you can start creating plants that need different soil nutrient concentrations, and are better-or-worse suited to different fertilizers, which are not infinite, and which make for more fluid choices as each fertilizer resource you have starts having different values to you based upon their relative abundance.  (I.E. You were talking about how carbonaceous fertilizers could not be viably produced in enough abundance to make up for the losses in soil fertility due to modern agricultural practices.) 

Adding variables that separate out water, biomass(carbon), NPK, as well as environmental variables like energy source, soil acidity, drainage/CEC, and finally pollutants like hot manure, heavy metals, and soil salinity all make the "math problem" of trying to make "what can I plant to get the greatest return in food per the least unit time and area" something that cannot be easily "solved" with a single perfect answer for the sheer number of changing variables.

[wierd]

Actually, that doesnt make it into strange voodoo either, just adds unnecessary mathematical overhead.

Sustainable horticulture is already about know what your crops want-- farmers have built cheatsheets for soil management and crop cycles for years. They are called farmers almanacs. (though most are little more than pitiful astrology in a pulp paper wrapper.) Crop management *IS* just an optimization strategy, that once you get down, is very easy to keep up.

Adding weather effects to screw up the reliability of growing conditions would introduce far more FUN than an overly complicated growing algorithm.

Observe:

Let's say we are dealing with wild strawberry.  We will say it needs 4000 units of energy per tile to produce a crop, each crop is seasonal, so it's a 3 month period, making each month deliver approximately 1250 units of sunlight per tile, under "ideal" crop conditions. (Since this is a soil neutral crop.) In rolls a series of summer rain clouds that blankets out the sky's sunlight partially for the majority of the summer.  Instead of the 4000 units of sun, we only got 3000 units, or less.  That energy has to be made up for, so the soil neutral crop suddenly becomes a soil depleter, because the sun wasn't out. This means your soil will deplete if you don't watch the weather, and make adjustments!

This, even with the highly simplified system suggested.

The weather is unpredictable, and random based on the biome embarked upon-- Dust storms would be especially terrible for crop growth-- and would make evil biome embarks all the more difficult--

[Wastedlabor]

Has it been suggested before to let plants mutate so the fort selects which subspecies to farm?

Subspecie properties wouldn't be obvious, so it would make sense to devote farming to several of them lest it turns out a variant is a lot less fertile or has a syndrome. It has the reward that great subspecies could be produced and, at some point in DF development, traded in to allow the civilization to become stronger. Of course you could make a fatal mistake and doom your civilization by exporting some seeds of "oblivion plump helmets".

[NW_Kohaku]

That's just quibbling over the mechanism--- I just tried to supply a possibly plausible (yet absolutely absurd at the same time!) one for how the cavern ecosystem could be sustained over large areas, and climate regulated.

It's not, it's an important conceptual difference, even if the mechanism is actually similar when you go beyond that.

It means recognizing what it means to talk about xenosythetic lifeforms, especially when we start expanding the magic plants available for farming... or, in other words, exactly what you start getting to in the next paragraph:

Substituting your explanation, we have other issues-- it means that the nethercap concentrates this "netherworld portal" meta-material in its fruiting bodies, which would then deplete the environment of this exchange medium.  We would have to make the nether caps much more prolific, and outright delicious to cavern flora to provide a mechanism to redistribute it back into the cavern ecosystem, in order to balance that out.

Yes, that's exactly my point.

Keep in mind what it means to have "Nether Particles" that fungi grow around as an energy source - there's more than one dimension that we can link to, and it means that it's possible to conceptualize "Good" biomes as being based upon "Good Particles" that flood a region with Good Magic. 

Sunberries, for example, only grow in good regions, same with feather trees and bubble grass.  (Whose names, incidentally, are rather literal.  Feather trees have actual feathers and bubble grass actually is made of bubbles, making Good biomes look something like a Dr Seuss book.  Sunberries have what appear to be miniature magical suns inside them, and their alcohol is "liquid sunshine".)

Part of my point is that you can't just treat this as just another thing you can explain purely by known physics, there's something obviously magic going on, here, and your explanation has to take routinized magic into account.  (Hence, the references to Discworld.)

Sunberries can't just be strawberries that eat up more soil nutrients, they have to tap into some external magical energy source, and when taken out of that energy source, will not grow. 

If nether caps grow on nether particles, then there might be a finite number of nether particles in the caverns, and there may be a serious difference between what happens when there's lots of nether caps and when there aren't any nether caps in the area.  What happens if destroying nether wood releases nether particles into the air, and it causes "nether magic levels" in the area to spike?  What if nether caps are like a sink that keeps nether magic from going wild, and destroying all the nether caps alters the magical ecosystem? Does that start releasing shades into the caverns where unchecked nether magic levels are starting to spike?  Does it suddenly make the whole cavern colder? Does it make magic-hungry megapreditors like forgotten beasts come, attracted by the sweet scent of released magic?

If we are to treat caverns as a magic ecosystem, then the "plants" down there are the foodstuff of the amethyst men, cave dragons, and magma crabs.  All things with obviously magical biologies.  They eat not just nutrition (or in an amethyst man's case, they probably don't even need that) but the magic, or at least the magic-energy-synthesizing organisms the same way that all the creatures in deep sea volcanic vents rely upon symbiosis with chemosynthetic monocellular organisms.

The xenosynthesis concept means taking things beyond their first step of assuming arbitrary game mechanics are just arbitrary and you don't have to think them through any further - it means that you take seriously the concept that there are multiple energy sources in the game universe, and what their ramifications are.

[NW_Kohaku]

Has it been suggested before to let plants mutate so the fort selects which subspecies to farm?

Subspecie properties wouldn't be obvious, so it would make sense to devote farming to several of them lest it turns out a variant is a lot less fertile or has a syndrome. It has the reward that great subspecies could be produced and, at some point in DF development, traded in to allow the civilization to become stronger. Of course you could make a fatal mistake and doom your civilization by exporting some seeds of "oblivion plump helmets".

Yes, mostly in conjunction with procedurally generated plants in the first place.  (That is, rather than just using wild berries, and treating them as though they were domesticated, you can really only efficiently farm fully domesticated plants, meaning trade with the elves and humans for surface crops, not just plant wild berry seeds if you want good crops.)

Although it mechanically requires some means of informing the player of what, exactly, these things they can't look up on the wiki actually are and do in a sane way.  (Which requires a sort of "Civpedia" like interface built into the game.)