Fuel and fire behavior analysis for early-season prescribed fire planning in Sudanian and Sahelian savannas uri icon

abstract

  • Early dry-season prescribed fires can reduce fuel loads and thus prevent or mitigate the severity of late, high-intensity fires that spread widely in savanna ecosystems and damage woody plants. However, due to the lack of scientific knowledge regarding fuel characteristics and fire behavior in West African savannas, this practice can have effects that are diametrically opposed to those desired and may threaten the environment. There are three crucial parameters that must be considered when planning early-season prescribed fires: the ignition probability, the rate of spread of a fire and the amount of fuel consumed. In this study, 231 early-season prescribed fires were conducted in three savanna ecosystems in Senegal in order to characterize these three fundamental parameters.
  • Early dry-season prescribed fires can reduce fuel loads and thus prevent or mitigate the severity of late, high-intensity fires that spread widely in savanna ecosystems and damage woody plants. However, due to the lack of scientific knowledge regarding fuel characteristics and fire behavior in West African savannas, this practice can have effects that are diametrically opposed to those desired and may threaten the environment. There are three crucial parameters that must be considered when planning early-season prescribed fires: the ignition probability, the rate of spread of a fire and the amount of fuel consumed. In this study, 231 early-season prescribed fires were conducted in three savanna ecosystems in Senegal in order to characterize these three fundamental parameters. Logistic regression analyses revealed that fuel moisture content and relative humidity are good predictors of ignition probability. Multiple linear regressions were used to investigate the relationships between fire rate of spread, fuel consumption or fire intensity and fuel and weather conditions. Readily usable nomographs for forest managers were created based on those relationships that proved to be significant. Kruskal-Wallis tests performed to compare the observed rates of fire propagation with those predicted using BehavePlus showed no statistically significant difference between them
  • Logistic regression analyses revealed that fuel moisture content and relative humidity are good predictors of ignition probability. Multiple linear regressions were used to investigate the relationships between fire rate of spread, fuel consumption or fire intensity and fuel and weather conditions. Readily usable nomographs for forest managers were created based on those relationships that proved to be significant. Kruskal Wallis tests performed to compare the observed rates of fire propagation with those predicted using BehavePlus showed no statistically significant difference between them. (C) 2012 Elsevier Ltd. All rights reserved.

publication date

  • 2013
  • 2013
  • 2013