Sure. Thank you for that question.
It's a multistep process. It starts with actually monitoring the environment. The data that we start with is what's happening with water levels and flows. We run the stations. It's a bit of a collaborative program, so there are stations that are of interest to the provinces, like Ontario and Quebec, and ones where the water is moving interprovincially or internationally where we have the most interest. We collect all the data, though, and it's supported through a cost recovery program. We maintain the data quality control. We maintain the data flow to make sure that the information—the data itself—gets to the provincial governments very quickly.
Quebec is a little different from the rest of the provinces because they collect their own data and we acquire it and make it available. That's a small difference, but it doesn't materially change the speed at which the data is shared.
It starts with data, and then, in those two jurisdictions, the provinces have their own flood forecasting centres. We feed the forecasting centres in two ways. The weather part of the enterprise is continually doing the forecasting of what the conditions are, such as how much rainfall there is. We look, on a season by season basis, to see whether we're going to have a wetter season ahead of us. There's a fair amount of uncertainty with a seasonal scale prediction versus a daily weather forecast, but nevertheless that's part of what we will give them. We will update that on a monthly and then weekly basis as it starts to get to the spring freshet season.
The other thing we do is track the amount of precipitation that has happened in the winter. The snow pack, the rate of melt and the intensity and duration of rainfall are all the major conditions that determine whether we're going to have a flooding kind of event like the ones we had in the region for those two years you've described.
That's the principal engine. You spoke about modelling generally. I described the weather modelling enterprise. It's a very complex atmosphere, ocean and ice model. The last component that we're working on now through our science is to bring in the hydrological modelling component. Our vision is that, within a few years, we hope to provide the same kind of predictive outputs to the provinces and territories as we do with weather forecasting. The science isn't quite there.