A simulation-optimization model for selecting the location of fuel-breaks to minimize expected losses from forest fires

Abstract Fuel-breaks function to restrict the spread of wildfire through a forest and thereby reduce the damage to values lost when fires escape. The challenge of selecting optimal locations of fuel-breaks arises from the uncertainty over how future fires might originate and spread spatially in the future. In this research, we evaluate a new approach to addressing this uncertainty through the formulation and application of a decision support model using simulation-optimization. The problem is formulated as a combinatorial simulation-optimization problem with the objective of minimizing fire-risk subject to a limit on the total area to be used as fuel-breaks. A stochastic simulation model of fire-spread was designed to estimate the fire-risk of each candidate solution produced by a metaheuristic search algorithm. The model was tested on a forest of 220,000 ha in north-western Ontario. The objective function (total fire-risk) of the solution found using the simulation-optimization model was significantly lower than the objective function of a solution found using a greedy heuristic—a spatially blind approach, by which the highest risk stands were iteratively removed until the allowable limit on fuel-breaks was reached. This significant difference indicates that the spatial relationships between selected fuel-breaks is an important factor in reducing a forest's total fire-risk. Hence, the spatial layouts of the best solutions were visually analyzed using GIS-software to illustrate patterns that result in reduced fire-risk at the landscape-scale. We conclude that this modeling approach, while computationally intensive, can be used to support improved spatial decisions on the location of fuel-breaks.

[1]  Kurt Mackes,et al.  Costs for Reducing Fuels in Colorado Forest Restoration Projects , 2003 .

[2]  Carl N. Skinner,et al.  Basic principles of forest fuel reduction treatments , 2005 .

[3]  Holger Vogt.  A new method to determine the tool count of a semiconductor factory using FabSim , 2004, Proceedings of the 2004 Winter Simulation Conference, 2004..

[4]  Abhijit Gosavi,et al.  Simulation-Based Optimization: Parametric Optimization Techniques and Reinforcement Learning , 2003 .

[5]  Averill M. Law,et al.  Simulation Modeling and Analysis , 1982 .

[6]  David L. Martell,et al.  A 500 hPa synoptic wildland fire climatology for large Canadian forest fires, 1959–1996 , 2002 .

[7]  Hong S. He,et al.  An object-oriented forest landscape model and its representation of tree species , 1999 .

[8]  M. E. Alexander,et al.  Chisholm/Dogrib Fires Research Program QuickNote #6: Incorporating Breaching and Spotting Considerations into PROMETHEUS--The Canadian Wildland Fire Growth Model , 2004 .

[9]  W. Hargrove,et al.  Simulating fire patterns in heterogeneous landscapes , 2000 .

[10]  L. Mearns,et al.  Climate Change and Forest Fire Potential in Russian and Canadian Boreal Forests , 1998 .

[11]  Paul Rogers.  Optimum-seeking simulation in the design and control of manufacturing systems: experience with OptQuest for Arena , 2002, Proceedings of the Winter Simulation Conference.

[12]  Susumu Morito,et al.  Exploration of a minimum tardiness dispatching priority for a flexible manufacturing system—a combined simulation/optimization approach , 1993, WSC '93.

[13]  Michael C. Fu,et al.  Simulation optimization via simultaneous perturbation stochastic approximation , 1994, Proceedings of Winter Simulation Conference.

[14]  M. Finney.  FARSITE : Fire Area Simulator : model development and evaluation , 1998 .

[15]  Sarah McCaffrey,et al.  Science basis for changing forest structure to modify wildfire behavior and severity , 2004 .

[16]  David L. Martell,et al.  Statistical quality control analysis of forest fire activity in Canada , 2002 .

[17]  G. Richards.  An elliptical growth model of forest fire fronts and its numerical solution , 1990 .

[18]  Scott L. Stephens,et al.  Evaluation of the effects of silvicultural and fuels treatments on potential fire behaviour in Sierra Nevada mixed-conifer forests , 1998 .

[19]  Hong S. He,et al.  Simulating forest fuel and fire risk dynamics across landscapes—LANDIS fuel module design , 2004 .

[20]  Jan W. Van Wagtendonk,et al.  Use of a Deterministic Fire Growth Model to Test Fuel Treatments , 1996 .

[21]  Brian Peter,et al.  Estimating harvest schedules and profitability under the risk of fire disturbance , 2005 .

[22]  M. Finney.  Design of Regular Landscape Fuel Treatment Patterns for Modifying Fire Growth and Behavior , 2001, Forest Science.

[23]  Jack P. C. Kleijnen,et al.  Optimization of simulated systems: OptQuest and alternatives , 2007, Simul. Model. Pract. Theory.

[24]  Fred W. Glover,et al.  A Template for Scatter Search and Path Relinking , 1997, Artificial Evolution.

[25]  Glen W. Armstrong,et al.  Estimating the cost of land base changes due to wildfire using shadow prices , 2003 .

[26]  A. Shabbar,et al.  The Association Between Circulation Anomaliesin the Mid-Troposphere and Area Burnedby Wildland Fire in Canada , 1999 .

[27]  Andres Weintraub,et al.  Assessing the impact of stand-level harvests on the flammability of forest landscapes , 2007 .

[28]  Paulo M. Fernandes,et al.  A review of prescribed burning effectiveness in fire hazard reduction , 2003 .

[29]  Susan G. Conard,et al.  Changing fuel management strategies - The challenge of meeting new information and analysis needs , 2001 .

[30]  Philip N. Omi,et al.  Effect of thinning and prescribed burning on crown fire severity in ponderosa pine forests , 2002 .

[31]  C. Holling,et al.  Command and Control and the Pathology of Natural Resource Management , 1996 .

[32]  Frederick S. Hillier,et al.  Introduction of Operations Research , 1967 .

[33]  Sierra Nevada Ecosystem,et al.  Sierra Nevada Ecosystem Project final report to Congress , 1997 .

[34]  K. Hirsch,et al.  Fire-smart forest management: A pragmatic approach to sustainable forest management in fire-dominated ecosystems , 2001 .

[35]  M. Flannigan,et al.  Future Area Burned in Canada , 2005 .

[36]  S. Pyne.  Introduction to Wildland Fire: Fire Management in the United States , 1984 .

[37]  Farhad Azadivar,et al.  Simulation optimization in strategic location of semi-finished products in a pull-type production system , 1996, Winter Simulation Conference.

[38]  Henry Carey,et al.  Modifying WildFire Behavior - The Effectiveness of Fuel Treatments The Status of Our Knowledge , 2003 .

[39]  M. E. Alexander,et al.  Canadian Forest Fire Danger Rating System: An Overview , 1989 .

[40]  Marc-André Parisien,et al.  Using Landscape-Based Decision Rules to Prioritize Locations of Fuel Treatments in the Boreal Mixedwood of Western Canada , 2006 .

[41]  Craig Loehle,et al.  Applying landscape principles to fire hazard reduction , 2004 .

[42]  Mark A. Finney,et al.  The challenge of quantitative risk analysis for wildland fire , 2005 .

[43]  Sierra Nevada Ecosystem,et al.  Assessments and scientific basis for management options , 1996 .

[44]  Yu Wei,et al.  An optimization model for locating fuel treatments across a landscape to reduce expected fire losses , 2008 .

[45]  Jennifer Johnson,et al.  A FireSmart approach to integrated fire and forest management in the boreal forest region of Canada , 2004 .

[46]  B. M. Wotton,et al.  Climate Change and People-Caused Forest Fire Occurrence in Ontario , 2003 .