Progress Report 2: Resilience and Adaptation to Climatic Extreme Wildfires (RACE Wildfires)

This is the second progress report of the international project funded by the National Research Council of Canada called Resilience and Adaptation to Climatic Extreme Wildfires (RACE Wildfires). In this second phase, the research performed included two main tasks: 1) developments concerning the modelling of smoke and 2) development of analysis methods concerning validation datasets for wildfire evacuation. Visibility in smoke is a key aspect in terms of safe evacuation in wildfire scenarios. As valid results of evacuation modelling tools would rely on an accurate representation of the impact of smoke on people, physical accuracy is required. Therefore, the rendering of smoke needs to be physically based while still being computationally inexpensive so that it can be run in a multi-physics tool in real-time. This report presents an approach for rendering smoke with a single in-scattering term which allows for smoke and light interaction over multiple wavelengths. In addition, analysis methods concerning validation datasets for wildfire evacuation models are presented and discussed. This includes both traditional regression methods as well as approaches based on machine learning.