Courses/CS 461/Winter 2006/Chris Lemcke/Week 1

From CSWiki

[edit] Tutorial Comments

The program really has a lot of examples. I look forward to seeing what each one is about.

I opened the wolf-sheep preditation model, and it starts out with 100 sheep and 50 wolves scattered randomly, and the whole area covered in grass.

I start the model, and the wolf and sheep populations start swirling around a lot, first with a lot of sheep, then a lot of wolves, then almost none of either, then the wolves die out and the sheep take over a swarm the place.

I like that the go button is also the pause button: makes is easy to start and stop quickly.

When I ran the model again a couple of times, I got three different types of results: 1) My initial result, with an arbitrarily large amount of sheep. 2) After the initial burst in both populations, they both die almost immediately. 3) Very rarely, instead of dying out completely, a very small number of wolves survive the initial drought of sheep. The sheep, however, sufficiently outnumber the surviving wolves so reproduce at a massive scale. Then the wolves follow suit and wipe out the sheep at a scale far greater than the previous burst of population. I suppose that if the same conditions applied, it might be possible to sustain this pattern, with greater and greater bursts of population, but I would guess that would be highly unlikely.

After about 100 tics, the sheep population has declined and the wolf population has risen. When I turned the grass on, however, the sheep and the grass created a sustained relationship (the wolves died out quickly). This happened multiple times. Sometimes, however, either the sheep or the wolves would die out almost immedietly, and so the ecosystem would stop.

Whenever I setup the model with grass on at the start, the ecosystem balanced itself out quite soon and stayed relatively stable.

[edit] Final Comments

So far, NetLogo seems to be a very useful tool. As a simulator, it is quite easy to use. I have not experienced the code, yet.