We're going to talk about models and data. Fire-EdUp works with vegetation models that are really used by the RFS. In the examples directory, we can open up the valid model list. And this is a list of all the valid vegetation models we can work with. We've already seen button grass in these examples, but we can have spinifex grasslands, grasslands, wet eucalypt forests, and so on. There's quite a few to choose from. So if we wanted to simulate a different vegetation model for a different environment, we could just copy one of these from the list. I'll choose wet eucalypt forests and paste it to where we had button grass. Now if I run this basic code, I can dial in some settings. I'm going to set 50 tons per hectare, 10% moisture fuel content, a nice 24 degree day at 27% relative humidity, and a pretty gentle breeze of 13 kilometers per hour.

The whole point here is to observe the FBI that we can print. If you don't have the FBI printing, you may just have to uncomment that line that we saw in the first example. And under these conditions, I've got an FBI of 72. Cool. What happens if we go back down to the button grass model? If I run the script now, we have a different FBI and that kind of makes sense. In a wet eucalypt forest, the fire is going to behave under the same conditions a bit more severely than in a button grassland. So I would say have a play with finding different environmental conditions and fuel conditions and changing the vegetation model to observe the impact that just being in a different location with different vegetation can have under the same environmental and fuel conditions. Let's talk a bit more about the data that comes off each of these sliders. You can access that under the data variable.

If I comment the FBI line with alt three or adding that hash and uncomment print data, we're back to familiar territory that we saw in the first example where we have these plots coming through of all the data and we can see as we change the data. This is all a bit messy, though. There's a lot of text on the screen. What if we only cared, for example, about the temperature? I'll just stop the script. We can access just the temperature component of that data variable with the following code. We can open a square bracket and type in temperature as a string and then close the square bracket. And I'll pad this out so it's a bit easier to read. Now when I run the script, we're getting just one variable. And this can make it a lot easier for you to do whatever logic calculations you would like. As I lower that temperature slider, we can see that only one value is on the screen. And I've got a plot of just the value of that slider.

I can still work with the other sliders and my model, but the only thing being printed to the shell is the temperature. And so we can include some simple logic. If data temperature is greater than 30-degrees, we can print it's too hot. Now when I run the script again, when it's only 26-degrees, we have nothing. But as I slide that slider over 30-degrees, we now have a message coming through. So you can see we can extract just the value we want out of that data variable. It also stands to reason that we could overwrite the data that we put into our models. We could disregard the slider value and set some kind of constant. For example, we would do that by assigning. We would assign data temperature to some value. In this case, we could assign it to 30-degrees. This would ignore the slider and it would insert that constant variable into our Fire-EdUp calculations.

I'm not sure why you would want to do this, but the code will let you do it if you want to. Okay. So we've seen how to select different models and we've seen how we can access specific data from our Fire-EdUp platform so we can make decisions based off that data or even manipulate it. Let's turn back to models for a moment. You can see that our valid model list has a few entries in it, but maybe we want to create a model for an area that isn't really captured here. We can see the existing models use these things called weights that are represented as a number. So one to 1.5 and these represent how severely a fire behaves in that model environment. What if we wanted to create a model for a very, very tropical, very wet place, like a constantly damp rainforest that would intuitively have a pretty low fire danger rating because fire would not be able to spread on very damp or like soft rotting wood.

So just as an exercise, let's try to create our own model with an appropriate model weight. That's this number here. In my main code, I'm going to delete where the model is selected and I'll instead create a model. Create model. And this is where we get to name our model and assign it some weight. So I'm going to call this a damp rainforest and it's going to have a pretty low weighting. I'm going to select 0.3. So we can save the script and run the script. And now if I set my platform to have a very high fuel load, we can still see there is some fire danger here. It's still in the extreme state, but it's much more resistant to fluctuations in FBI. The FBI is only 60, even though it's a very hot day and we have a lot of fuel. In our other models, if we had a very high fuel load on a hot day like this, we would have easily seen this as a catastrophic fire danger rating.

So even with extremely unfavorable conditions, this damp rainforest is a lot more resistant to fire danger. And of course it's your model, so you get to choose what's going on here. Maybe you want to create a model for a fireworks factory and use a value of 10, in which case this would be an extremely sensitive model. We can see under these conditions that I have, the FBI is enormous. It's 900, which is silly, but this is how models work. You get to choose the tuning parameters to create the model that you want. A quick note about slope. You can think of it as just another condition that we can input. Here we have slope degrees and we've got a very, very gentle three degree slope. It makes intuitive sense that for steeper hills, a fire will behave a lot more aggressively because the upward burning fire can access fuel uphill much easier.

So I encourage you to run a little experiment for a given set point, maybe the conditions that you have around you today that you can program into the sliders. Have a go at playing with the slope degrees and seeing what kind of difference it makes to fire danger. What do you observe for higher degree angles versus lower degree angles? I've set my Fire-EdUp platform for the weather that I'm experiencing today with pretty sensible values. And if I run the code with a slope degrees of three, then I have an FBI of about 45, 44, 45. If I change that now to a rather steep 45-degrees, then we can see the FBI has shot up by a count of about 10. And so there you have it. We've been able to play with the more advanced features of the fired up platform. We were able to access individual data from the sliders for our own logic, change the slope way, change the vegetation model, and even prescribe our own vegetation model where one didn't exist before. Well done for following along and I'll catch you next time.