PCMDI - IPCC

Forest fires are a global hazard. Antarctica is the only continent entirely free from the combinations of fuels, weather conditions and ignition sources, which provide a potential wildfire hazard. Some regions such as California, Australia and the Mediterranean exhibit particular susceptibility to forest fires, and research throughout the 20th century established a strong correlation between meteorological variables and fire occurrence and extent. Since then, a number of fire-weather indices have been established which allow the seasonal and daily forecasting of fire hazards in these regions. The past decade has seen significant interest in the impact of anthropogenic climate forcing on the occurrence of forest fires, focused on the most fire-prone regions, predicting overall increases in fire-weather in Australia, Spain, Portugal and South-west U.S.A. An increase in fire-susceptibility has also been predicted for parts of the boreal forests of Canada and Russia. However, no global overview has been presented, and little interest is given to regions with a currently low fire danger, which may or may not experience an increase in forest fire occurrence in the future.
This study asks ‘can we expect a change in the global distribution of forest fires due to anthropogenic climate change?’ This question is of great significance due to the environmental, social and economic damage caused by forest fires. This damage may possibly be even greater in regions previously free of major fire hazards as they lack the management experience of current fire-prone regions. The increased occurrence of fire in forested regions not adapted to periodic fire seasons may also have significant ecological impact. A secondary purpose of this study is to evaluate the applicability of a regional fire weather index to the global climate model. This study uses the ECHAM climate model data for the 4th IPCC Assessment along with the McArthur Forest Fire Danger Index (FFDI) to compare the global distribution of forest fire weather in 2080 under a 2xCO2 climate with that of today. The model results are validated against observed forest fire data for the past 20 years. This validation enables the study to be evaluated as a model for predicting the impact of anthropogenic climate change on the global distribution of the forest fire hazard. Validation of the model suggests that it is useful as a predictive tool on a global scale, and also on a regional scale in Africa, China, Canada and Europe. For other regions alterations to the model are required to produce a more useful output. This study predicts a significant overall increase in global forest fire danger, with significant increases in Australia, Africa, South America, China, Canada and Europe. Other regions are predicted insignificant changes in fire danger.