Forest Science 47(2) 2001 219Design of Regular Landscape FuelTreatment Patterns for ModifyingFire Growth and BehaviorMark A. FinneyABSTRACT. Patterns of disconnected fuel treatment patches that overlap in the heading fire spreaddirection are theoretically effective in changing forward fire spread rate. The analysis presented heresought to find the unit shape and pattern for a given level of treatment that has the maximum effecton forward spread rate. This occurs when the treatment units cause the fire to spread through themat the same rate as it spreads around them. Simulations suggested that these treatment patternsreduce the spread rate or fireline intensity over much of the area burned, even outside the treatmentunits where the fire was forced to flank. The ideal patterns are theoretically scale independent, allowingfor flexible application across heterogeneous landscapes. The topology of these patterns hasimplications for designing landscape-level fuel treatment patterns and for understanding spatialdynamics of fuel patterns across landscapes. FOR. SCI. 47(2):219–228.Key Words: Fuels, fuel treatments, fire behavior, landscape patterns, fire modeling.Mark A. Finney is Research Forester, USDA Forest Service, Rocky Mountain Research Station, PO Box 8089, Missoula MT 59807—Phone: (406)329-4832; E-mail: [email protected]. The paper was written while the author was a research scientist with Systems for Environmental Management,PO Box 8868, Missoula MT 59807.Acknowledgments: This research was funded by the USDA Forest Service Rocky Mountain Research Station, Fire Sciences Laboratory, Fire EffectsResearch Work Unit, under Research Joint Venture Agreement INT-96095-RJVA. The author thanks Kevin Ryan for his support of this work, and JimAgee for his comments on a draft. Three anonymous referees also improved the paper with their helpful reviews.Manuscript received November 29, 1999. Accepted June 1, 2000.This article was written by a U.S. Government employee and is therefore in the public domain.THE GOAL OF FUEL MANAGEMENT IS to preemptively modifywildfire behavior through changes to the fuel com-plex. Fuel management has received increasing inter-est for mitigating fuel hazards (U.S. Department of theInterior and Department of Agriculture 1996, U.S. GeneralAccounting Office 1999), some of which were created bynearly a century of fire suppression on millions of acres in thewestern United States (Arno and Brown 1991). Fuel treat-ments are intended to help limit wildland fire sizes andseverity by directly mitigating fire behavior and indirectly byfacilitating suppression. Prescribed burning and mechanicalthinning can lower fire spread rates and intensities within thetreated area (van Wagtendonk 1996, Helms 1979), at leastuntil fuels and vegetation reaccumulate. Fireline construc-tion can be faster and more effective (fewer escapes) whenheavy concentrations of brush and logs are removed, andspotting from torching trees is limited.Treating all fuels across an entire landscape is practicallyimpossible, however. Limited funding, inadequate road ac-cess, variable land ownership, and regulations often restrictprescribed burning, smoke production, or timber harvesting.Fuel management on a landscape scale tends to be limited inthe amount of a given treatment, location of treatments, andthe kinds of treatments permitted. Priorities for treatment areoften based on local hazards, ecological objectives, conve-nience, cost, land ownership, or accessibility. These priori-ties are not necessarily topological or spatial as is fire growthand behavior and they do not prioritize the layout of treatmentunits with an explicit consideration of fire growth amongadjacent units. With all the limitations on treatment locationand continuity of treatments across a landscape, it is logicalto address how the spatial arrangement of treatment unitsaffects fire growth.Two basic strategies for landscape-level fuel managementare to contain fires and to modify fire behavior. Linear fuelbreaks (Agee et al. 2000, Weatherspoon and Skinner 1996)have been proposed to help contain fires. Fuel breaks areintended to reinforce defensible locations and facilitate sup-220 Forest Science 47(2) 2001pression action by indirect tactics including backfiring (Green1977, Omi 1996). It is assumed that undesirable fire effectsare limited by reducing fire sizes. The fuel breaks themselvesare only burned along the zone of suppression, not by the fire.By contrast, a spatial arrangement of treatments that prima-rily modifies fire behavior would involve area-based ordispersed patterns (Martin et al. 1989). Fire effects andbehaviors are modified wherever the fire encounters thetreatment units. Suppression is facilitated by allowing anytactic (direct, indirect, or parallel attacks) to adapt to changesin collective fire behavior.For fire modification, it is clear that the greatest reductionin fire size and severity occurs when fuel treatment units limitfire spread in the heading direction. The heading portion of afire (moving with the wind or slope) has the fastest spread rateand highest intensity compared to flanking and backingportions (Catchpole et al. 1982). The heading fire also holdsthe most potential for initiating crown fire and spotting,which makes suppression much more difficult.To disrupt the spread of the heading fire, there are threebasic geometric treatment patterns offering varying degreesof overlap: complete, none, or partial (Figure 1). The case ofcomplete overlap by multiple treatments (Figure 1a) has theeffect of producing a harmonic mean spread rate amongmultiple fuel types as the fire burns sequentially through thestrips (Fujioka 1985, Martin 1988):hfr fr=++111 22...(1)where fi is the fraction of the total distance across the ith fueltype having a characteristic spread rate ri. Here the effectiveheading spread rate h is proportional to the time spent in eachfuel type; the fire must spread through the treatment stripsperpendicular to their orientation (e.g., no flanking). Thisarrangement would rarely be practical for treating large andvariable landscapes. It would require extensive area to betreated and continuous land ownership and access. By con-trast, a treatment pattern with no overlap of the treatmentunits (Figure 1b) may not change the forward spread rateacross the landscape; fire can burn unfettered through thecorridors between treatment blocks. A random or arbitraryarrangement of treatments would closely resemble