A series of 23 experimental fires was carried out in a Pinus elliottii forest fuels plantation of the Experimental Station of Rio Negro-PR, Rio Negro, Universidade Federal do Paraná, in July 2008. The objective of this research was to develop fire behavior prediction models for fire management purposes. Observations and measurements of the forest fuels and meteorological conditions were carried out in the local prepared for the fires and fire behavior characteristics (fire spread, fire intensity and flame height)were measured during the fires. By correlating both groups of variables (environmental versus fire behavior) it was obtained significant association between these variables. Regression models developed using the collected data are able to predict fire behavior in terms of forest fuel consumption, air temperature, relative humidity and wind speed. It was also observed that the flame height is strongly associated with fire intensity and it can be used to estimate this intensity in the conditions which the data were collected.
Controlled burning; Fire behavior; Forest fires; Forest fuels
