Graduation date: 2007
Fuel management has been used as an effective local strategy to reduce the
undesirable consequences of wildfires. Many efforts toward scheduling of fuel
management activities across a broader landscape have been proposed, with the hope of
achieving larger landscape-scale management effects. However, scheduling of fuel
management treatments across the broader landscape is limited by understandings of
how individual management activities aggregate to larger scales and how they affect the
behavior of wildfires. Since full coverage of a landscape with fuels management
treatments is unlikely, it is necessary to examine the effects of a spatial pattern of
individual management activities at the landscape scale.
In this research, four spatial patterns of fuel management activities – dispersed,
clumped, random, and regular – were tested to investigate their potential for reducing
the risk of severe wildfire. A new methodology was developed for optimizing fuel
management patterns across a landscape based on a heuristic technique and GIS
databases. To quantify the cumulative effects of fuel management patterns for
disrupting the progress of wildfires, overall flame length, fireline intensity, and fire size
were measured for simulated fires, using a fire growth simulation model, FARSITE.
The management scenarios generated from the scheduling model presented a
variety of dispersion and treatment sizes, but also evenly distributed the harvest volume
through the multi-decade time horizon. The optimized spatial patterns were qualified
through visual examination as well as a statistical assessment.
Through this research, I have learned that the efficiency of fuels management
activities for reducing severity of wildfire is primarily influenced by treatment size, type,
and intensity. Most importantly, treatment types and intensity are the critical factor to
disrupt human-caused wildfires. The regular pattern seemed to be the most acceptable
for either random ignitions or hypothetical human-caused ignitions. It provided the
highest frequency in which simulated fires could contact the treated units, and higher
treatment intensity measured by amount of harvested volume from a unit area. To
enhance the results of this research, we suggest that one should utilize more feasible
management prescriptions for post-fire fuel conditions, and expand ignition sources to
other type of human-caused ignitions or natural-caused ignitions.