Cloud-Based Urgent Computing for Forest Fire Spread Prediction

Fire is a natural element of many ecosystems. Nevertheless, every year forest fires cause damages to biodiversity, atmosphere, and economy activities. Forest fire simulation have improved significantly, but input data describing fire scenarios are subject to high levels of uncertainty. Unknown parameters need to be adjusted, and, in this work, an input data calibration phase is carried over following a genetic algorithm strategy. The calibrated input is then pipelined into the actual prediction phase. This two-stage prediction scheme is leveraged by the cloud computing paradigm, which enables high level of parallelism on demand, elasticity, scalability, and a low-cost. We use a sound goodness-of-fit function together with an adaptive evaluation technique capable of reducing the overall calibration time by 60% when compared to current data-driven approaches. We have also set up an experimental study to validate the platform against a challenging wildfire with promising results