Wednesday, August 4, 2010

PS 53-52: CANCELLED - Time evolution of landscape patterns for different prescribed fire ignition methods

Robert L. Kremens1, Matthew B. Dickinson2, and Anthony S. Bova2. (1) Rochester Institution of Technology, (2) US Forest Service

Background/Question/Methods Prescribed fires have become an essential tool both for management of fire-dependent ecosystems and to reduce fuels to lower landscape flammability. Prescribed fires may be ignited using different methods including by helicopter (heli-torch or ‘ping-pong balls’), manually (firefighter and drip-torch), and with motorized vehicles (ATV). Because the pattern and areal rate of ignition, and thus the heat energy release patterns and magnitude, vary over a wide range for these different ignition methods, it would be expected that the ecological effects would also vary widely.  We have developed a method to monitor the heat release and patterns at a landscape scale (up to 2500 ha). We fly an airborne infrared sensor repeatedly over the fire and thus generate a time-sequence ‘movie’ of the heat release and patterns generated by the fire. We use in-fire instruments to absolutely calibrate the data collected by the airborne sensor.  We have analyzed the landscape patterns produced by two different ignition methods (ridge ignition by drip-torch and well-distributed ping-pong ball ignition by helicopter) on prescribed fires on dissected terrain in the Daniel Boone National Forest in Kentucky and Tar Hollow State Forest in Ohio, respectively.

Results/Conclusions

We used the landscape ecology tool Fragstats 3.3 (McGarigal et al.) and ENVI 4.5 (ITT Space Systems) to describe the differences in patterns. We analyzed the time series to understand the nature of the evolution of the fire, as well as the time integral to understand the overall effect.  There were significant differences in the aerial rate of growth and heat release patterns generated by the fires using the two ignition methods.  Ignition by helicopter was found to produce a greater mean and coefficient of variation in heat release that showed low correspondence to topographic patterns  whereas the ridge-ignition fire, being mostly backing spread, closely tracked the underlying topography. We will discuss the details of this analysis and provide insights on the fire effects differences created by these two ignition methods.