There are both economic and ecological justifications for fuel treatments, but based solely on economic objectives to reduce suppression costs on treated acres, there is evidence to support fuel treatments for cost-effective hazard reduction. Choosing a cost-effective fire and fuels management strategy requires an examination of available treatment methods and the suppression cost-savings of fuel load reductions. Prescribed burns and mechanical treatments alter the fuel load in a stand so that when a wildfire ignites in a treated stand, it spreads slower, burns with less intensity and severity, and is less costly to suppress. The cost-effective level of fuels treatment occurs where the marginal cost of treatment plus the expected post-treatment suppression cost is equal to the expected marginal cost of suppression without treatment. Cost data collected for the Deschutes and Ochoco National Forests indicate that cost-effective fuel treatment levels are consistent with per acre treatment costs between $20 and $115, depending on the initial stand conditions, fuel model and the probability of a fire. These per-acre treatment costs are inclusive of both prescribed fire and mechanical treatments. The lowest cost fuel treatments were wildland fire use for resource benefit, followed by management ignited prescribed fire. The most costly fuel treatments were pre-commercial thinning followed by grapple-piling; however even these mechanical treatments were cost-effective in areas with a high probability of wildland fire. On the Deschutes and Ochoco National Forests, in the period 1995 through 2000, 91% of total fire management spending went to suppression and 9% went to fuels treatment.

The fire season of 2000 heightened awareness of wildland fire and fuels management across the United States. The fires that blazed across the western United States burned over 8.4 million acres; in many areas dense forest stands were left blackened and charred, entire valleys filled with dark smoke, and ash floated down from the sky into towns located miles away from the flames. The fires of 2000 burned in spite of an estimated $1.6 billion spent on fire suppression.

After nearly a century of wildland fire suppression, forests throughout the interior West contain extremely high levels of fuel loads that, in many areas, exceed the historic range of variability. The result is a significant increase in the number of severe fires and an increase in the number of acres characterized by conditions that enable fires to ignite easily, burn with greater intensity and severity, and spread more rapidly and extensively. Aggressive suppression is becoming more costly while the economic, social and ecological benefits of systematic fire suppression are increasingly uncertain. It is now the responsibility of the US Forest Service, together with other federal agencies and the public, to determine the optimal balance between suppression and fuel treatments for fire hazard reduction.

The 1995 Federal Wildland Fire Policy marked a fundamental shift in federal wildland fire management from static to strategic. Federal agencies moved from a policy focused almost exclusively on prevention and suppression at any cost, towards a policy based on a broader range of management options, including the use of prescribed fire and fuels management, and the balancing of cost and benefits. Two of the primary objectives of the revised federal policy include the development of a strategic fire management plan for every burnable acre of federal land, and the re-introduction of fire into fire-adapted ecosystems. Fuels treatment is essential in order to successfully achieve these policy goals.

The objective of fuel treatments for hazard reduction is to reduce fuel loads (i.e., the quantity of fuel) and/or change the fuel profile (i.e., the spatial arrangement of fuels), thereby minimizing the risk of severe, high-intensity wildland fires. This objective is based on the premise that altering the fuel load will protect and sustain natural resources, particularly vegetation, wildlife habitat, and watershed integrity, increase the safety of wildland firefighters and people living in the wildland-urban interface areas, and reduce the suppression costs associated with high-intensity wildland fires. The optimal level of fuel treatment requires an analysis of the tradeoff between the costs and benefits of treatment.

The focus of the present analysis is on the cost-effectiveness of hazardous fuel treatments in the Deschutes-Ochoco National Forest in terms of reduced wildland fire suppression costs. The objectives of the present study are to:

* Examine trends in suppression costs and fuel treatments following the 1995 Federal Wildland Fire Policy;
* Gather qualitative and quantitative data on available fuel treatments;
* Collect cost data for fuel treatments;
* Determine the cost-effective level fuel treatment based on suppression cost-savings;
* Identify the treatment mix that can be executed at this level; and
* Compare the cost-effective level of treatment to the present level of treatment.

I will use treatment cost data collected from the Deschutes and Ochoco National Forests and suppression cost estimates from the National Fire Management Analysis System (NFMAS) to determine if there is economic evidence to support the use of fuels treatments for cost-effective hazard reduction.

DESCHUTES-OCHOCO NATIONAL FOREST

The geographic focus of the present study is the Deschutes-Ochoco National Forest. The Deschutes-Ochoco National Forest is located on the eastside of the Cascade Range and, excluding the Crooked River National Grassland, encompasses 2,450,000 acres of Central Oregon. Alone, the Deschutes National Forest (DNF) totals nearly 1,600,000 acres, of which 182,740 acres are designated wilderness. The DNF stretches for 100 miles along the eastern slope of the Cascade Mountains. Elevations range from 3000 feet to 10,358 feet at the summit of South Sister. The Ochoco National Forest (ONF) encompasses 850,000 acres, of which 36,200 acres are designated wilderness. The ONF extends eastward from the Crooked River National Grassland to the Ochoco Mountains. Together, the two Forests contain the headwaters of the Deschutes River, the Metolius River, and the North Fork of the Crooked River. The four primary plant association groups within the Forests include lodgepole pine, ponderosa pine, mixed conifer wet and mixed conifer dry. Up until the early 1990s, timber sales drove the commodity management focus on both National Forests. More recently, however, timber sales have been replaced by recreation as the primary management focus.

FIRE ECOLOGY

Wildland fire is a naturally occurring disturbance essential to ecosystem functioning. Fire stimulates nutrient recycling and creates the conditions required for the reproduction of fire dependent species. The frequency, intensity, and severity of wildland fire is described by and ecosystem’s fire regime, which varies spatially and temporally. Fire frequency, or fire return interval, refers to the number of years between two successive fire events in a given area. Historic fire intervals are estimated using fire scars on trees, plants that germinate after a fire, or from charcoal residue. (Agee1993) Fire intensity and severity are two related terms that refer to how hot a fire burns and how much vegetation is killed, respectively. Low severity fires are generally characterized by <10% canopy mortality, medium severity fires by 10%-70% canopy mortality and high severity fires by >70% canopy mortality. Much of central Oregon, and approximately half of the DNF and ONF, is characterized by a low-intensity, high frequency fire regime. Historical records indicate that in these areas, fuels rarely reached high levels because frequent fires regularly consumed surface fuels and pruned trees. (Agee 1993)

The build up of fine fuels and fuel ladders can significantly increase the intensity and severity of wildland fires. Fine fuels increase the rate of spread and understory fuels provide a ladder for fire to move up into the forest canopy, transforming a surface fire into a crown fire. When fuel loads are within the historic range of variability, it is assumed that fire frequency and intensity will also be within the historic range of variability. Conversely, if fuel loads exceed historic averages fire intensity and severity may increase. Historical records indicate that fuel loading on the DNF is far greater than it was a century ago. The following chart illustrates this change:

Table 1.1: Fuel Loads and Stand Density on the Deschutes National Forest

1893 1993
Trees per acre: 10-30 0-100
Average Diameter (Inches): 17 7
Average Age: 225-275 < 100
Average Tree Spacing (Feet): 35 x 35 8 x 8
Average Percent Shrub Cover: 0-5 30-70
Average Small Trees Per Acre: < 20 100 +
Average Down Logs Per Acre: 2-5 10-30

Source: DNF Integrated Natural Fuels Management Strategy 1998.

The changes observed in Table 1.1 are the result fire exclusion combined with the effects of timber extraction, road construction, and grazing. Natural and human caused ignitions in forests with altered fuel and forest structure, similar to those described above, can result in high intensity fires, excessive fuel consumption, large tree mortality and other adverse ecological effects.
After almost a century of fire suppression, land managers on the Deschutes-Ochoco National Forest presently face the challenge of managing wildland fire and fire risk in forests with significant fuel loads. Fires will burn in fire-adapted ecosystems regardless of human action or federal policy. It is not a question of whether or not a fire will burn, but how many acres and where? Given the certainty of fire presence along with the observation that following an ignition, fire behavior is a function of weather, topography and fuels, the importance of fuels management is unquestionable: it’s the only variable in the fire equation that land managers have the power to influence.

ECONOMIC THEORY AND US FOREST SERVICE POLICY

Historically, economic considerations have had only minimal influence on fire management and federal fire policy. For nearly a century, the US Forest Service enforced a strict policy of aggressive suppression on all wildland fires regardless of cost. The USFS has engaged in deficit spending during forest fire emergencies followed by supplemental appropriations to cover all excess expenditures since 1980. This fiscal strategy reinforces the notion that economic considerations are secondary to fire suppression goals and objectives. Unrestricted spending for emergency suppression eliminates incentives to develop cost-effective, efficient fire management strategies, and encourages the misallocation of resources from planning and fuels treatment to emergency suppression.

There were, however, attempts at the beginning of the 20th century to improve the efficiency and cost-effectiveness of USFS fire management. In 1914 the USFS Chief, Henry Graves, proposed that the amount spent on fire suppression should be proportional to the value of the resources being protected and the cost to protect them. (Pyne 1982) The proposal put forth by Graves is consistent with the definition of efficiency, as applied to fire management, which states that net benefits to society are maximized where the marginal cost of fire suppression is equal to the marginal benefits of suppression. In 1916, Roy Headly proposed the "least cost plus" theory, which stated that the sum of damages and suppression costs should be kept at a minimum. (Pyne 1982) Unlike Graves, Headly’s proposal does not use the efficiency criterion, but it does introduce the notion of cost minimization. The weakness inherent in both of these economic theories -- one that continues to challenge environmental economists -- is the difficulty associated with calculating the value of a continuous stream of both market and nonmarket benefits from environmental amenities. The value of recreation, wildlife habitat, watershed services, and aesthetic services, all unpriced nonmarket environmental amenities, are difficult to quantify. Because the value society derives from these amenities is influenced by wildland fire, determining the optimal level of fire suppression and risk reduction is problematic. Stephen Pyne (1996) describes the division of power between economic and political forces over the development of federal fire policy:

"For most of its evolution, fire management has been dominated by historical circumstances, political mandates, and non-economic criteria. Economic analysis could rationalize this process, but it could neither direct it nor substitute for it. Economic theory could occasionally advise, but it could not command." (Pyne 1996, p.326)

Throughout history, economists viewed the National Forests as economic institutions. Within these institutions, they advocated a fire management strategy that recognized the tradeoff between the cost of fire suppression and the value of the resources protected from fire. For example, it would be senseless to spend $1 million on suppression to prevent the destruction of a stand whose timber was valued at only $100,000. Politically minded officials, however, viewed the National Forests as political institutions upon which public opinion was formed. For this group, it was the responsibility of the Federal Government to protect the nation’s natural resources regardless of cost. Wildland fire protection, from the destructive flames of what Gifford Pinchot termed the "Dragon of Devastation," was justified without consideration of forest science or economic rationality.

More recently, the economics of fire policy and management has begun to emerge as an applied field of study. An example of this surge of interest is reflected by the 1999 Symposium on Fire Economics, Planning and Policy: Bottom Lines held in San Diego, California. The ecological repercussions of past management have resulted in diminishing marginal returns to suppression: it takes more and more dollars to prevent less and less damage. Diminishing returns to suppression have increased the importance of marginal analysis as applied to fire management. The majority of economic research has, however, focused on suppression and only limited attention has been given to hazardous fuel treatment. This is primarily because up until 1995 fuel treatments were, for the most part, limited to slash burns following commercial timber extraction. Subsequent to the Federal Wildland Fire Policy of 1995, and its emphasis on hazardous fuels reduction and strategic fire management planning, the use of and interest in natural fuel treatments has increased significantly.

FUELS MANAGEMENT

Past management practices including fire suppression, timber extraction, road construction, and grazing have increased the risk of high-severity wildfire in the Deschutes-Ochoco National Forest. However, with the use of hazardous fuel treatments the risk of a high-severity fire and the cost of suppression following an ignition can be reduced. Hazardous fuels are characterized as either activity fuels or natural fuels. Activity fuels include fuel loads that are the direct result of human activity, such as slash generated from timber extraction. In contrast, natural fuels are naturally occurring and include anything from fine fuels, like pine needles, grass, and leaves, to woody twigs, branches and even trees over 3 inches in diameter. The most commonly used methods of hazardous fuel treatments are prescribed burns and mechanical treatments.

Prescribed fire for fuel reduction in the National Forest can be divided into three categories: prescribed fire in natural fuels, prescribed fire in activity fuels, and prescribed wildland fire use for resource benefit (WFURB). A prescribed burn is generally defined as a management ignited fire in natural and/or activity fuels under clearly defined environmental parameters with expected outcomes. Prior to the ignition of a prescribed burn, the fuel is often gathered into piles, which can be constructed by hand, with a bulldozer, or with an excavator. WFURB is defined as a naturally ignited fire that is allowed to burn under clearly defined environmental parameters according to a unit’s management plan. WFURB management strategies range from limited or modified suppression to monitoring with no suppression. The reintroduction of fire into fire-adapted ecosystems called for in the 1995 Federal Wildland Fire Policy favors prescribed fire as a fuel treatment. However, it is important to note that the use of fire as a management tool also requires consideration of risk of escape, wildland urban interface areas, smoke, and emissions. Because of these health and safety considerations, environmental conditions consistent with specified parameters often result in narrow "burn windows" or opportunities to burn. In areas where prescribed fire for hazard reduction is not feasible, often because of excessive fuel loading in sensitive areas, mechanical treatments may be used alternatively or in preparation for prescribed fire.

Mechanical treatments include piling, chipping and thinning. Mechanical piling is generally executed with a bulldozer or an excavator. Grapple piling is the term given to the construction of piles with an excavator that moves along skid trails left from a timber extraction. Once constructed, these piles can be burned under a wider range of environmental parameters and, therefore, within a wider burn window. The most common objective of thinning (precommercial, noncommercial, and commercial) for fuel hazard reduction is to decrease stand density. While thinning does not decrease the amount of fine fuels, which determine a wildfire’s rate of spread, it does reduce the amount of large fuels often responsible for a wildfire’s intensity and severity. In areas where fire reintroduction is a management objective, mechanical treatment is often a necessary precursor to fire entry and once executed, the opportunities for fire entry increase. (Deschutes National Forest Integrated Natural Fuels Management Strategy 1998)

METHODS

Initially, I planned to use historical data on suppression costs on units with and without fuel treatments to determine the cost-savings associated with the relevant treatment. However, given limited data availability and data inconsistencies this was not possible. At this point there is only anecdotal evidence of suppression cost savings following fuel treatments. In the absence of empirical evidence, treatment cost estimates from fire management officers and suppression cost estimates from the National Fire Management Analysis System (NFMAS) provided an alternative data set for analysis.

The National Fire Management Analysis System (NFMAS) estimates suppression costs according to fuel model classifications. Based on the degree to which a treatment changes the fuel model, NFMAS estimates can be used to calculate the suppression cost-savings of fuel treatments. For example, given a 1-acre unit that is classified as a fire behavior fuel model 8, the NFMAS suppression cost per acre estimate is $5339. If the probability of ignition is 6%, then the expected suppression cost on the 1-acre unit is $320.34. This figure represents the total cost of fire management on the unit. Fuel treatment X, which costs $50 per acre, will change the fuel model from 8 to 9. The NFMAS suppression cost per acre given a fuel model 9 is $3427 per acre and given a similar 6% probability of ignition, the total cost of fire management with treatment is $255.62. In this example the suppression cost-savings generated from treatment is positive, $64.72, and treating the unit minimizes the expected cost of fire management.

Table 1.2: Suppression Cost-Savings Example

Suppression Cost * Probability of Ignition Treatment Cost Expected Cost of Fire Management
Without Treatment $5339 x 6% $0 $320.34
With Treatment $3427 x 6% $50 $255.62
Expected Cost Savings $64.72

The cost-effective level of fuels treatment, depicted in Figure 1 as Q*, minimizes the cost of fire management. In this figure the expected cost of fire management without treatment, MC’, is constant. The expected cost of fire management with treatment, MC’’, is upward sloping and includes the cost of treatment plus the expected cost of suppression after treatment. MC’’ is not constant and depends on the number of acres treated and the treatment method used. The positive slope of this curve indicates that as the number of acres treated increases and the cost of treatment increases, the marginal cost of fire management increases at an increasing rate. Under an optimal management strategy, low-cost fuel treatments that achieve the greatest fuels reductions and generate the greatest suppression cost-savings would be executed first. High-cost fuels treatments that achieve less significant fuel load reductions and have smaller suppression cost-savings would follow these initial treatments. In other words, the fuel reductions projects with the greatest net benefits would be given priority. However, interviews with fire/fuels managers revealed that the low-cost fuels treatments were given priority and the benefits (i.e., reduced suppression costs, reduced fire hazard) were not considered. In Figure 1, the cost-effective solution occurs at Q*, where MC’=MC’’. Up to this point, expected suppression cost-savings exceed treatment costs and beyond this point treatment costs exceed expected suppression cost-savings.

Figure 1: Cost-Effective Fire Management

[not shown]

DATA

Data on aggregate fuel treatment costs and suppression costs for the Deschutes and the Ochoco National Forests were gathered from the Budget and Financial Management Department at the USFS Pacific Northwest Regional Office in Portland, Oregon. Aggregate cost data were collected from Statement of Obligations records for FY1995 through FY2000. Also from the Pacific Northwest Regional Office, Fire and Aviation Management provided annual figures from FY1995 to FY2000 on natural and activity fuels treatment area for the DNF and the ONF.

Fire management officers on the five Ranger Districts comprising the DNF and ONF provided fuel treatment cost estimates. These cost estimates were collected from survey responses and interviews with six fire management officers. Fire management officers are responsible for fire readiness and effective use of personnel and equipment in the protection of Forest resources and improvements. The Fuels Management Survey was divided into four parts: fuel treatments, resource targets, fuel management planning, and suppression costs. However, due to time constraints on the part of fire management officers, responses were completed and collected only for the fuel treatment portion of the survey. Resource targets, fuel management planning and fire management on the Forest were discussed in interviews with the fire management officers.

Suppression cost estimates were from the 1999 National Fire Management Analysis System (NFMAS) runs for the Central Oregon area. The NFMAS data were included in a USFS Technical Report, Determining the Current Condition of Fire Regimes on the Prineville District, published in spring 2001 by Brian Scholz of the ONF. The fuel model for a given area may be one of the thirteen basic types described by Albini (1976). Table 1.3 provides a description of the relevant fuel models upon which USFS NFMAS estimates were made.

Table 1.3: Fuel Models for Fire Behavior

Fuel Model Description
8 Closed Timber Litter – tightly compacted short-needled conifer litter, not much branch/log fuel
9 Hardwood Litter – loosely compacted long-needle pine or hardwood litter
11 Light Logging Slash (<90 tons/ha) -- most needles fallen, compact; partial cuts and clear cuts

Source: Albini 1976.

RESULTS AND DISCUSSION

Suppression costs and fuel treatment costs between FY1995 and FY2000 on the DNF and the ONF are depicted below in Figure 2. In this period, average annual suppression cost on the Deschutes was $17,892,223.17, ranging from $7,869,035.12 up to $26,288,796.03. On the Ochoco, average annual suppression cost was $3,164,358.93, ranging from $1,480,362.04 up to $7,214,167.19. The annual variation in suppression costs over this relatively short time period is largely the result of climatic variables such as annual precipitation and temperature. For example, the fire season of 1996 was particularly hot and dry, these conditions…suppression costs spiked in that year.

Between FY1995 and FY2000 fuel treatment costs on the DNF and the ONF have not increased significantly. Given the reduction in hazardous fuels called for in the 1995 Federal Wildland Fire Policy, it would be reasonable to expect an increase in fuel treatment costs as the number of acres treated increased. This, however, has not been the case. For the entire period, spending on suppression represented approximately 91% of the annual fire budget while hazardous fuels treatment captured only 9% of the budget, as shown in Figure 3.

Figure 4 depicts the number of acres treated in the same period and indicates that the total area treated between 1995 and 2000 has not increased significantly on either one of the Forests. Breaking down total treatment area based on fuel type, activity fuels and natural fuels, provides significant insight into these trends. Activity fuels treatment area has decreased on the Deschutes and the Ochoco as a result of reduced timber extraction. Natural fuels treatment area, however, has increased.

It is of primary importance to determine whether or not the current level of fuel treatment is cost-effective. If total treatment costs exceed total suppression cost-savings, then too much area is being treated. But, if suppression cost-savings outweigh treatment costs, then not enough area is receiving treatment. In the second case, increasing the area treated would result in cost-savings and would improve cost-effectiveness.

The cost-effective level of fuel treatment depends on two variables: pre-treatment fuel model and the probability of a fire. Table 1.4 outlines the cost-effective level of treatment as indicated by fuel treatment cost-per-acre. In characteristic fuel model 11 units, treatments with costs-per-acre between $20 and $80 are cost-effective, with respect to the probability of a fire. This means that for a treatment that can be executed at a cost of up to $80, the expected suppression cost-savings exceed the treatment cost. In characteristic fuel model 8 units, treatments with costs-per-acre between $40 and $115 are cost-effective, also with respect to the theoretical probability of a fire. Increasing the probability of a fire increases the likelihood of realizing the suppression cost-savings of fuel treatment. When the probability of a fire is great, more expensive fuel treatments become cost effective. Decreasing the probability of a fire decreases the likelihood of realizing returns on the investment in hazardous fuels reduction and more expensive fuel treatments might no longer be cost effective. Given a small probability of a fire, only the least expensive fuel treatments are cost-effective.

Table 1.4: Cost-Effective Treatment Costs

Treatment Cost Per Acre In Characteristic Fuel Model 11 Units / Treatment Cost Per Acre In Characteristic Fuel Model 8 Units
Prob(fire) = .06 $80 $115
Prob(fire) = .04 $55 $75
Prob(fire) = .02 $20-$30 $40

Cost estimates collected from fire management officers on the DNF and the ONF, described in Table 1.5, indicate that both prescribed fire and mechanical treatments can be achieved within the within the range of cost-effective fuel treatments. On average, prescribed fire is observed to have a lower average cost-per-acre than mechanical treatments. The lowest cost fuel treatment is wildland fire use for resource benefit followed by management-ignited prescribed fire. The most costly treatments are mechanical and include small diameter pre-commercial thinning followed by grapple piling. However, both the low-end grapple piling and thinning cost estimates are included among the cost-effective treatments in the above scenarios in areas with a high probability of fire.

Table 1.5: Fuel Treatment Costs

Treatment Cost Per Acre:
Management Ignited Prescribed Fire: $40-$150
Prescribed Natural Fire: $20-$59
Slash Burn: $70-$175
Dozer Piling: $55
Small Diameter Thinning: $70-$360
Grapple Piling: $80-$145
Hand Piling: $100-$250

It is important to note that not all cost-effective solutions are efficient. The socially optimal level of fuels treatment maximizes net benefits and is, by definition, the efficient solution. The focus on minimizing the cost fire management encourages the use of low-cost, high-fuel-load-reduction treatments, but without considering the full benefits of available treatments and multiple management objectives, such as ecosystem health and fire reintroduction, certain treatment methods and vegetation association groups may be overemphasized and even overused. The present analysis does not include the benefits of fuels treatment beyond reduced suppression costs. Including additional benefits would increase the optimal level of fuels treatment. Since the present analysis includes only a fraction of fuel treatment benefits, it represents a low-end range estimate of cost-effective treatment levels.

CONCLUDING REMARKS

Cost data collected from the Deschutes and Ochoco National Forests and suppression cost estimates from the National Fire Management Analysis System (NFMAS) indicate that there is economic evidence to support the use of fuels treatments for cost-effective wildland fire hazard reduction. Based on the suppression cost-savings of fuel treatments, the cost-effective level of treatment is inclusive of treatments ranging from $20 to $115 per acre, depending on initial fuel model and probability of a fire. The present level of fuels treatment on the DNF and the ONF falls within this range. In spite of this evidence and the management goals put forth in the 1995 Federal Wildland Fire Policy, treatment costs and total treatment area on neither the Deschutes National Forest nor the Ochoco National Forest have increased in the period FY1995 to FY2000. However, in spite of this evidence and the management goals put forth in the 1995 Federal Wildland Fire Policy, treatment costs and total treatment area on the Deschutes National Forest the Ochoco National Forest have not increased in the period FY1995 to FY2000. Including estimates of the full range of social and ecological benefits of hazardous fuel treatments suggests that increasing the treatment level to include higher cost treatment methods would result in increases in net benefits. Future research is needed to quantify the full benefits of hazardous fuel treatments, particularly nonmarket benefits, beyond reduced suppression costs.