HESS-IM: A Uncertainty Reduction Method that Integrates Remote Sensing Data Applied to Forest Fire Behavior Prediction

MÉNDEZ-GARABETTI, MIGUEL; BIANCHINI, GERMÁN; CAYMES-SCUTARI, PAOLA

Communications in Computer and Information Science

Springer Science and Business Media Deutschland GmbH

Año: 2021 vol. 1444 p. 17 - 30

1865-0929

Natural disasters alter the stability and living conditions in the environment. Natural disasters are caused by different types of hazards, which can be of natural or man-made origin, the latter in turn can be intentional or unintentional. Forest fires cause great losses and harms every year, some of which are often irreparable. Among the different strategies and technologies available to mitigate the effects of fire, the forest fire behavior prediction may be a promising strategy. This approach allows for identifying areas at greatest risk of being burned, thereby permitting to take decisions in advance to reduce losses and damages. In this work, we present a Hybrid Evolutionary-Statistical System with Island Model (HESS-IM), a new approach of the uncertainty reduction method that integrates remote sensing data applied to forest fire behavior prediction. HESS-IM uses hybrid metaheuristics under a collaborative approach as an optimization technique, satellite images for application to real cases, statistical analysis, and heterogeneous high-performance computing to generate predictions in the shortest time possible.