![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
cdaveA post on the scientific preprint paper blog on how the credit crisis spread, reminded me of my masters dissertation.
It was on attempting to model solar flare statistics using a self-organised cirtical forest fire model. In laymans terms we had a program that updated a grid with spaces, trees, and fires following some simple rules. This paper sums up the sort of thing we were supposed to be thinking about.
The most interesting results we got out of model in the end came when we started playing with underlying rules. If you let a few of the trees wander about into empty spaces beside them, you still get self organising behaviour.
This sort of model had been using in the past to look at measles epidemics (Spread of fire ~ spread of disease).
One thing we didn't get a chance to try, was adding an occasional long range correlation. In other words giving a tree a friend on the other side of the grid. Turning into the kind of small world network, where there's only 6 degrees of separation between any tree and Kevin Bacon.
I couldn't find this in the literature anywhere. So thought that if I'd stayed on at Warwick I might have been able to write it up, and get published.
I just had a quick look now and it seems that while I was thinking this, someone else had already done this, written it up, and it was awaiting publication. I'm quite glad I didn't end up wasting a lot of time on this now.
In an entirely unrelated matter, if anyone buys me this t-shirt, they will be thoroughly irradiated.
It was on attempting to model solar flare statistics using a self-organised cirtical forest fire model. In laymans terms we had a program that updated a grid with spaces, trees, and fires following some simple rules. This paper sums up the sort of thing we were supposed to be thinking about.
The most interesting results we got out of model in the end came when we started playing with underlying rules. If you let a few of the trees wander about into empty spaces beside them, you still get self organising behaviour.
This sort of model had been using in the past to look at measles epidemics (Spread of fire ~ spread of disease).
One thing we didn't get a chance to try, was adding an occasional long range correlation. In other words giving a tree a friend on the other side of the grid. Turning into the kind of small world network, where there's only 6 degrees of separation between any tree and Kevin Bacon.
I couldn't find this in the literature anywhere. So thought that if I'd stayed on at Warwick I might have been able to write it up, and get published.
I just had a quick look now and it seems that while I was thinking this, someone else had already done this, written it up, and it was awaiting publication. I'm quite glad I didn't end up wasting a lot of time on this now.
In an entirely unrelated matter, if anyone buys me this t-shirt, they will be thoroughly irradiated.
