Tutorials: STELLA

PRESENTER: Throughout this class, we're going to be using a series of computer models to explore some things about climate and our consumption of energy and the economic consequences of all these things. And so I thought it would be helpful to begin with an explanation of what a STELLA model is. And so you can see how they're made and how they work and so forth.

So STELLA models begin with some kind of a conceptual diagram of the system that you're going to try to represent. So over here is a drawing. A simple drawing of a water tub that's got water in it. It has a faucet pouring water in. And water leaves the tub via this drain here.

So there are some mathematical relationships here between the area of the base of the tub and the amount in there, and the depth of water in the tub. The depth of water in the tub then controls the velocity of water going out the drain. That times the drain area tells you the amount of water that leaves in a given period of time.

So anyway, there's the conceptual model. Here's what it looks like in STELLA. This square thing here, we call a reservoir of stock. That's where the water in the tub is recorded. And if I click on that, you can see we've entered 0.5 cubic meters of water. That's the starting amount of water in there.

This entity here is called a flow. It's just like a pipe that allows water to flow into the tub. And that has a specified rate. It's 0.25 cubic meters per minute. And that's a constant value.

The drain, on the other hand, is not a constant value. It's defined as the velocity at the drain times the area of the drain. The area of the drain is defined over here by another equation that involves the drain radius. The velocity of the drain here is this thing. It's something called Torricelli's law. Gravity here is a constant. The area of the base of the tub is something that is a constant that we'll start off with as 0.5. And the depth of water here is just the water in the tub divided by the area of the base of the tub.

So that's the way the model is set up. So you build the thing here. And you enter in the equations and values for all these things. And then, you're set to go.

And so, we can run this model now. And what that means is we start the clock ticking. We are going to pour in water at a given rate, a specified rate, through the faucet. We'll let it drain out according to that equation. And we'll see what happens.

So here, we're going to run this for 30 minutes. And we just click the Run button here. And you can see very quickly it calculates this. The blue curve here is the water in the tub. Remember, we started off with 0.5 cubic meters. And it rises up to something close to about 0.9 cubic meters.

And see how it rises up quickly at first, but then it levels off and approaches a steady state. And that's because the water leaving the drain is rising and rising and rising until it becomes essentially equal with the amount of water flowing in the faucet. So at that point in time, the water coming in is equal to the water going out, so there's really no change in the amount of water, the volume of water in the tub here itself.

This interface to the model here includes some little knobs and things that you can adjust. So I can change the drain radius. We can see what happens if we change the radius of the drain a little bit. Increase it there. There it levels off at a much lower value. 0.5 cubic meters in the tub if I decrease that somewhat to there.

You can see what happens there. It doesn't even, in this time period, level off entirely. It's already up to more than three cubic meters. So we're going to have a lot of water in that tub at that point. And you can reset those things by hitting the [INAUDIBLE] buttons there.

So this is how you operate these models. This is how you interact with them. Just remember that if you're looking at an interface like this, just know that behind the scenes, underneath the hood, there is something like this. This diagram that lays out the logic and the structure of this system that the computer uses to mathematically solve the evolution of this system over time.