Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1] 2

Author Topic: Ecology and more complex behavior for a more realistic animal population growth?  (Read 1047 times)

TiagoTiago

  • Bay Watcher
    • View Profile

Have the animals vary their rate of reproduction based on how often they eat, have them actually eat things, herbivores will go after plants (some even only a restricted set of plant species), carnivores will chase fight and eventually eat other animals (depending on  the species, even attacking as a pack) etc; and perhaps also have them potentially eating stuff that was meant for the dwarfs if it's avaiable, there is no dwarfs currently with the food in direct sight line and if the animals haven't been properly trained to not eat dwarf food, but the propensity to go after dwarf food instead of intended animal food and wild food would vary depending on species, training and hunger level.


Also, have the avaiability of food and the presence and number of other animals of the same species and of other species affect where the animals decide to go, so if there is too many in an area the ones near the edge of the group will move further away looking for somewhere less crowded and with more avaiability of food.

If the ratio of herbivores to plants or the ratio of predator to prey is too high, there will be the risk they'll eat up too much food and eventually kill the ecosystem or at least severely imbalance it, resulting in shortage of food for them, so for species that tend to be too devastating and in areas with too little wild food the player will have to artificially control the  animal population, and if desired also try to help the ecosystem grow to a sustainable level, perhaps even reduce the amount of hunting, tree and plant collecting, or depending on the situation, hunting certain species more.

After a bit of tweaking i imagine this should allow for population of animals to become relatively stable and sustainable.
« Last Edit: March 24, 2010, 09:35:25 pm by TiagoTiago »
Logged
---
Notification emails don't always reach me, please let me know if you think it has happened (like if i don't reply a thread i'm expected to reply for more than a week)

Andeerz

  • Bay Watcher
  • ...likes cows for their haunting moos.
    • View Profile

I have no idea if this suggestion has been suggested before...

At any rate, I love this suggestion!  Hmmmm... there have to be simple ecology/population equations/simulators in existence that could give us a good idea of how to go about doing this.  I wonder how complex the math would have to be...  I'll see if I can consult with some of my ecology friends.
Logged

Mel_Vixen

  • Bay Watcher
  • Hobby: accidently thread derailment
    • View Profile

Could you do something about the readability of the op?
Logged
[sarcasm] You know what? I love grammar Nazis! They give me that warm and fuzzy feeling. I am so ashamed of my bad english and that my first language is German. [/sarcasm]

Proud to be a Furry.

TiagoTiago

  • Bay Watcher
    • View Profile

Could you do something about the readability of the op?

better now?
edit: oops, lol, weird, i don't remember wiriting inside the quote....
« Last Edit: March 24, 2010, 09:48:01 pm by TiagoTiago »
Logged
---
Notification emails don't always reach me, please let me know if you think it has happened (like if i don't reply a thread i'm expected to reply for more than a week)

Andeerz

  • Bay Watcher
  • ...likes cows for their haunting moos.
    • View Profile

http://en.wikipedia.org/wiki/Ecosystem_model
http://en.wikipedia.org/wiki/Population_ecology
http://en.wikipedia.org/wiki/Evolutionary_game_theory
http://en.wikipedia.org/wiki/Malthusian_growth_model
http://en.wikipedia.org/wiki/Matrix_population_models
http://www.sosmath.com/diffeq/first/application/population/population.html

Quite pertinent to the discussion at hand!!!  Seriously, check 'em out!  In some respects, DF could ideally be an ecological model of epic (and dwarfy) proportions.

Hmmm... ok, suggestion time.  How are we going to model population growth and decline during world-gen and behind the scenes in the world during fort mode?  I am no ecologist, but I propose we follow the BIDE model suggested in the "Matrix population models" article in wikipedia and elsewhere.  Easier said than done, I know, but perhaps we could get it to work!

So, the BIDE model is basically as follows.  The population of a particular species at time "t", or N(t), one season after time "t-1", in a particular area can be represented by the following equation:

N(t) = N(t-1) + B − D + I − E

Where N(t-1) is the population at time "t-1", B is the number of births since t-1, D is the number of deaths since t-1, I is the number of immigrants since t-1, and E is the number of emigrants since t-1.

The tricky part is determining what factors contribute to each of the BIDE variables and how to represent them numerically.  Some examples of factors for B and D:

For B (births): babies per litter, litters per time increment per individual, gestation time   

For D (deaths): predation, disease?, old age, infant mortality

Hmmm... I'm gonna have to think about this for a while.
« Last Edit: March 25, 2010, 03:14:44 am by Andeerz »
Logged

Andeerz

  • Bay Watcher
  • ...likes cows for their haunting moos.
    • View Profile

Oooo!  Idea!

So, IRL, ecologists studying predator/prey relationships and population dynamics extrapolate models from real-life population data.  This would be a wonderful way to use Arena mode; as an ecology laboratory!!!  Just stick animals from a biome in a large area, track their interactions, and make a mathematical model of these interactions.  :D :D :D  I would be willing to do the grunt work and devise the models, Toady!!!

One more thing... for Toady and anyone else that might be interested just in case, here's a primer on Ecology and Population Dynamics that is relatively short and well written:

http://www.cas.muohio.edu/~stevenmh/Stevens-2009-part.pdf

If I have time, I will read it and see what kind of model I might be able to come up with or suggestions and stuff and stuff.

edit: dammit... it's missing the important chapters, I think, like the parts about predator-prey relationships.  Oh well.
« Last Edit: March 26, 2010, 02:57:59 am by Andeerz »
Logged

Footkerchief

  • Bay Watcher
  • The Juffo-Wup is strong in this place.
    • View Profile

We've had several ecology threads before:
http://www.bay12games.com/forum/index.php?topic=24910.0
http://www.bay12games.com/forum/index.php?topic=4624.0
http://www.bay12games.com/forum/index.php?topic=42441.0

It would be helpful if this discussion takes into account the ideas in those threads.
Logged

Andeerz

  • Bay Watcher
  • ...likes cows for their haunting moos.
    • View Profile

Thanks, Footkerchief!  The last thread you listed is particularly relevant.  The other two, sadly, don't go into much depth nor do they provide much in the way of a model.  I'm not saying I or anyone else in this thread is doing any better, though.

Hmmm... the third thread's mention of Sim Earth deserves some exploration...
Logged

Silverionmox

  • Bay Watcher
    • View Profile

SimLife is relevant as well.
Logged
Dwarf Fortress cured my savescumming.

BigD145

  • Bay Watcher
    • View Profile

What we don't need is evolution. With world generation stopping at 200-300 years, there's just not enough time to see anything all that noticeable.
Logged

Silverionmox

  • Bay Watcher
    • View Profile

We'll still need to keep an eye on it, though. The megabeasts disappearing too soon is a typical case of natural selection :)
Logged
Dwarf Fortress cured my savescumming.

TiagoTiago

  • Bay Watcher
    • View Profile

actually, evolution can happen in a matter of decades (like what happened with the domesticated foxes on Russia, i think it was Russia, not sure), and for bacteria and virii("viruses"?) you just skip one dose of antibiotics and already a bunch of more resistant strands evolve

But i agree that actually modeling evolution realisticly would probably be too much (would probably involve reworking the creature definition system to deal with raw genes instead of abstracted features, and it would be hell to tune up to have good odds of a enjoyable result), perhaps at most have the appearing of certain species only happen after an amount of time another (supposedly more primitive) species has been around, basicly a pre-determined evolutionary tree but with the final result still being dependent on which species survived together up until "now", and since we got faries and other magical stuff, allow for certain species to beat the odds and come to exist without ancestrals, and some extinct ones to be magicly jurassic-parked back to life.
Logged
---
Notification emails don't always reach me, please let me know if you think it has happened (like if i don't reply a thread i'm expected to reply for more than a week)

cameron

  • Bay Watcher
    • View Profile

one of the things to keep in mind about evolution is that though now 200-300 years is the normal length of a world, all the end year conditions can be set rather high, and one or two thousand year old worlds could eventually become the norm
Logged

NW_Kohaku

  • Bay Watcher
  • [ETHIC:SCIENCE_FOR_FUN: REQUIRED]
    • View Profile

A more basic problem with evolution isn't the timeframe, but rather, the way that creatures are handled.

In real-life evolution, the development of any major organ has to be weighted against that organ's cost in minerals and energy, which in turn means more food - if that organ doesn't help that creature get more food or breed more frequently, it is not worth its cost, no matter what its other benefits, and it will not develop.  (Vestigal organs being an oddity/artifact.)

In DF, there are only two major factors: being tough enough to survive combat, and, much more importantly, breeding speed.

Why would creatures not evolve impossible physiologies that enable them to survive, like [no_pain] and [no_fear]?  Why would non-breeding creatures like Bronze Collosi ever stand a hope of surviving the onslaught of fast-breeding, pain-immune, giant super-groundhogs?  Especially since the short maturity rate of most animals compared to larger creatures ensures that evolution in a one-year-to-adult creature happens twelve times faster than a twelve-years-to-adult creature? Also, what keeps herbivores weaker than carnivores? 
Logged
Personally, I like [DF] because after climbing the damned learning cliff, I'm too elitist to consider not liking it.
"And no Frankenstein-esque body part stitching?"
"Not yet"

Improved Farming
Class Warfare

TiagoTiago

  • Bay Watcher
    • View Profile

cats would take over the world
Logged
---
Notification emails don't always reach me, please let me know if you think it has happened (like if i don't reply a thread i'm expected to reply for more than a week)
Pages: [1] 2