 We're doing experimental burns today with the aim to establish the fire behaviour prediction model for croplands. So in the past we were utilising the grassland fire behaviour model to model how fires will go in the crops. This time we're trying to establish if there is a difference between fires behaving in the croplands and in the grasslands. So here we're trying to burn under conditions when we have a control problem. So it's no point establishing the model for the lower end of the equation where we put those fires reasonably easy out. So we're trying to have conditions that are really hot and dry with reasonably strong winds. So we can find out how fast those fires spread. The reason why we want that to occur is first of all that we can train our firefighters to become even better in extinguishing them. Also so we can have a targeted messaging, be on a timely messaging for our communities. This is part of the national fire danger rating project as well. So overall jurisdictions across Australia including Commonwealth are building the new fire danger rating system. So we're burning today, it's about 38 degrees 15% humidity and the winds are around 35 km per hour from the north. So we really need to capture that higher end of the dataset because this is a type of condition when we will have a problem with controlling those fires. Within CFA's bushfire research and development team we've been working for quite some time on fire propagation in grassland fuels. Initially this work stemmed from the War Commission finding that we need to improve our understanding of grassland curing and how partially cured grasslands fire spread through these grass types. So what we did was a series of experimental burns with a similar experimental design as we're doing today. And we've replicated this study up and down the eastern seaboard so we've got a pretty good understanding of fire propagation in improved pastures. And really what we wanted to know which we thought was critically important was our understanding of how fires compare from improved pasture into crop lands and also different crop variations. So natural unharvested wheat through to harvested wheat through to harvested and bale wheat. Obviously all have different fuel arrangements and therefore it affects the fire propagation through each of the different types. Some of the surveys are very interesting. We have a motto, you got to burn to learn. Each time we light fires we learn something new about how fire propagates in different fuels. That challenges the prior knowledge we had or assumptions we had about the fire in these kind of fuels. In this vegetation there was a fair bit of ignorance from our side about what were the drivers and how wind and fuel moisture were influencing the fire or how fast the fire would spread. We haven't analyzed the data but the first idea is that unharvested crops will spread much faster than harvest crops. It has to do with the structure of the vegetation of the crop as a fuel that feeds the fire. That is going to change from an unharvest to an harvest condition. We've all experienced fires at all varying degrees of speed and all that but there's no documentation, no data recorded ever over crop. We go to crop fires every year. We burn thousands of hectares of stubble but we've never timed it. We've got no real concept of how quick the fire is actually burnt. So this has given us real time data so we can say it's going to go from point A to point B in a certain amount of time. So my role has really been centred around assisting in terms of suppression resources to ensure that the research can be conducted safely and that there's no risk of any fire escape from the burns here that we've been doing. Anecdotally there's been a fair bit of talk of the spotting that occurs within wheat, short distance spotting and I guess it's largely from the research burns we've seen here today. We've seen very little in the way of spotting within that unharvested wheat under today's conditions. So a lot of the work that's been done today will lead to better predictive models. So we'll end up with information on rates of spread, flame heights that will assist us in terms of providing better predictions that will ultimately inform communities of what their risk is for a particular day in those sorts of environments. For the crews I think it's provided a really unique opportunity today to sit back and watch fire. So rather than being actively involved in having to suppress things they've had the opportunity to sit back and observe and actually see fires develop and spread. And a lot of the comments from the crews here have been around just how quickly, from an ignition to being a fully established fire just how quickly that can occur. It's been a good opportunity for the crews to observe that and also an opportunity for them to talk to some of Australia's top fire scientists and just share experiences with them. So it's been a really great day today. We had more than 100 volunteers from different parts of the state coming and supporting this research and without them we wouldn't be able to do the work.