Assignment 1
In this assignment, we worked with the model from Tutorial 3 for Netlogo. Netlogo uses “turtles” and “patches”. In this model, they represent “free thinking agents” and “food sources”, or, as I’ll call them, “birds” and “grass”. The agents behave in a “bottom up” fashion, without a pacemaker agent giving order to the system, and form patterns on their own.
There is a variable starting number of birds, and a starting full screen of grass. Birds are born with a certain variable amount of energy (“energy-at-birth”), expend energy when moving (-1), and gain a variable amount of energy when they eat grass (“energy-from-grass”). Grass will regrow, but it is random and based on a probability percentage (I created a variable for mine, and called it “grass-regrow-probability”).
The variable parameters can be modified and the model run several times to observe the patterns that develop. The same settings do not guarantee the same outcome each time, in fact, it is incredibly unlikely that the same thing will happen twice, even with the same settings.
How does the “energy-from-grass” parameter influence the emergent patterns observed in the model simulation?
The “energy-from-grass” parameter affects how long birds are able to live (because their energy decreases by 1 with each move) as well as if they are able to reproduce (because they need 50 energy to reproduce). With a high “energy-from-grass” parameter, birds can live longer and reproduce more, but soon, there are so many birds that there is not enough grass to feed them all.
I worked with the model to show this. In this run of the model, the energy gained from grass was manually varied as the model ran. With the “energy-from-grass” setting being higher, grass gives the birds high energy, resulting in more bird births and longer lives. An interesting outcome of this is that birds then eat off the grass, then birds die down as a result of not enough grass, then grass is able to regrow as bird populations lessen. With the “energy-from-grass” setting being lower, it causes quick bird death, hardly any reproduction, and massive grass regrowth.
Ran with the following settings:
Initial number of birds: 1
Energy gained from grass: (variable)
Birth energy: 2
Grass growth probability: 1%
How does the “birth-energy” parameter influence the emergent patterns observed in the model simulation?
The “birth-energy” parameter affects how much energy birds have to start with when they are born. If the birds start with high energy, they can reproduce immediately, causing a surge in the bird population, which quickly causes grass to die out, which causes birds to die out. If the birds start with low energy, the other settings can make up for it as they can gain energy from the abundant grass at the start of the model. Even with a low “energy-from-grass” setting, birds can live.
A sudden surge in population can occur if there is enough grass or a higher “energy-from-grass” setting to offset the competition from birds, paired with a high birth rate.
I ran the model to study this. In this run of the model, the starting energy at birth was varied as the model ran. At the beginning of the run, the starting energy was high (100), and a surge of more birds was born. However, soon the competition for food caused many birds to die out, and new birds were not born as quickly to replace them. Lowering the birth energy over the course of the model did not help the problem, of course, as birds then were not born with enough energy to reproduce immediately, and had to eat enough grass in order to reproduce. The competition from other birds prevented this from happening, and the population reached an equilibrium and then stayed about the same.
Ran with the following settings:
Initial number of birds: 1
Energy gained from grass: 12
Birth energy: (variable)
Grass growth probability: 1%
Is the emergent pattern sensitive to the initial population size?
I believe that the pattern is not sensitive to the initial population size, because the abundant grass at the start of the model allows even one bird to eat enough grass (even with the “energy-from-grass” setting set very low) to allow them to reproduce and reach the same population levels as a higher starting population would achieve.
I ran a few models to study this, and some interesting things happened.
Initial Population 1 – First Run
In this run of the model, all of the birds died early on.
Ran with the following settings:
Initial number of birds: 1
Energy gained from grass: 12
Birth energy: 2
Grass growth probability: 1%
Initial Population 1 – Second Run
In this run of the model, “waves” occur of birds eating grass and the grass dying off, then birds dying down as a result of not enough grass, then grass able to regrow as bird populations lessen.
Ran with the following settings:
Initial number of birds: 1
Energy gained from grass: 12
Birth energy: 2
Grass growth probability: 1%
Initial Population 60
In this run of the model, all of the birds died early on.
Ran with the following settings:
Initial number of birds: 60
Energy gained from grass: 12
Birth energy: 2
Grass growth probability: 1%
Initial Population 80
In this run of the model, “waves” occur of birds eating grass and the grass dying off, then birds dying down as a result of not enough grass, then grass able to regrow as bird populations lessen.
Ran with the following settings:
Initial number of birds: 80
Energy gained from grass: 12
Birth energy: 2
Grass growth probability: 1%