Applying GIS to predict high-risk wildfire zones within Wood Buffalo National Park
Wildfire is a common occurrence within Canada's boreal forest, and can pose a threat to human health, industry, infrastructure, and endangered species. Recently, wildfire modeling has emerged as a valuable method of predicting the behaviour, spread, and extent of wildfire over a given area. In this study, Geographic Information Systems (GIS) were used to develop a simplified model of wildfire risk within Wood Buffalo National Park, which is located on the border of Alberta and Northwest Territories. A multi-criterial evaluation (MCE) approach was used to integrate several variables that were found to relate to wildfire risk. Using this approach, three main components related to wildfire were identified; spread potential, ignition potential, and fuel potential. Integrating these components and their related variables, the wildfire model output was created identifying zones of risk; 20% of park land was identified as high risk for wildfire, 46% as medium risk, 26% as low risk, and 8% as no risk. The model output also indicated that the southern part of Wood Buffalo National Park is generally more susceptible to wildfire due to lower amounts of precipitation and higher temperatures. To evaluate the model, the final wildfire risk output was then compared to areas burned during the 2016 wildfire season. The model was effective in predicting these wildfire occurrences, and could be applied to future fire management situations.