California Fires

California Fires

Fire Ecosystem and Occurrence

With its Mediterranean climate and ecosystems comprised of vegetation that evolved with episodic fire, wildland fires are an episodic certainty in virtually all of the state’s 58 counties. Most fires, and virtually all of the large ones, occur during the droughty summer and fall months when the moisture content of down/dead wood and herbaceous material as well as live vegetation fuels is low. In a typical year, most of the hectares that burn are not forested per se—rather, they are covered in grasses (mostly exotic invasive, annual grasses that cure as early as May and that track ambient humidity the rest of the year) or in chaparral, a plant community composed of a mixture of evergreen shrubs such as species in the genera Ceanothus (e.g., deer brush) and Adenostoma (e.g., chamise) that burn readily, and under some weather conditions, fuel fires that move at great speed (35 km per hr is not uncommon) and burn with great intensity. By far the greatest losses of life and property in California in modern times have been associated with fires fueled by chaparral, a vegetation type that covers nearly 6 million acres of the state (Fried et al 2004), most of it in southern California and along the central California coast. For example, the 2003 and 2007 southern California fires, the 1993 Topanga fire, and the 1990 Paint fire in Santa Barbara burned primarily chaparral. Sometimes, a volatile mixture of underbrush and urban forest composed of exotic trees such as Eucalyptus sets the stage for catastrophic results under adverse weather, as occurred in the 1991 Oakland Hills Tunnel fire (Table 1)

This is not to say that all California wildfires happen outside of forests; on the contrary, some of the largest have burned at least partly through parts of California’s 33 million acres of forested lands, including the Zaca, Marble Cone, Day, McNally and Stanislaus fires (Table 2).

Over the period 2001-2005 on non-federal land, about 9 percent of the 172 thousand acres burned annually were timberland (commercial forest land capable of producing 20 ft3/ac/yr). The percentage on federal lands is undoubtedly greater, however statistical information is not readily available.

Fire Types

It can be helpful to think of different types of California fires: (1) resource fires, where the impacts that generate the most concern are ecological (e.g., loss of habitat, loss of soil or soil productivity, sedimentation of streams) or economic (e.g., loss of timber or recreation value), and (2) wildland urban interface (WUI) fires (where the greatest concerns center on saving human lives, homes, and infrastructure). The fire management problem takes on a different flavor for these two fire types, and operations and tactics are certainly different.

For example, on resource fires (particularly on federal lands), while efforts are made to reduce resource losses, a variety of fire management approaches may be practiced, from pre-fire fuel treatments designed to reduce fire intensity (and thus, the severity of fire impacts) to Appropriate Management Response, a philosophy of focusing protection effort on portions of the fire perimeter where high resource values are threatened and allowing fire to burn, under close observation, in places where the benefits of fuel consumption by the fire exceed the potential damages produced by the fire. The typical response to WUI fires is to pull out all the stops to contain them as quickly as possible; sometimes, weather and fuels dictate a virtually exclusive focus on protecting life and property (to the detriment of containment success), such that fires may grow much larger than they would in the absence of WUI.

Either type of fire can occur in nearly any of California’s diverse vegetation types, from those with extremely long return intervals (e.g., redwood forest) to forests dominated by surface fire regimes (e.g., ponderosa pine forests), to mixed severity fire regimes (e.g., some mixed conifer forests), to crown fire regimes (most prominently, chaparral, but also true fir and lodgepole pine). A great many fires start in dry grass, partly because ignitions occur easily in this fuel and because it is so ubiquitous along roads, where a great many fires start, and partly due to exotic invasive annual grasses having long ago replaced nearly all the less ignitable perennial bunch grasses that once covered the state.

By far the majority of fire ignitions are human-caused. On state-protected wildlands (the non-federal portion of the landscape), the breakdown of causes for the 5700 fires that occur each year, on average, is 7 percent arson, 88 percent accidental and 5 percent lightning, While the share of lightning ignitions is undoubtedly greater on the generally higher elevation federal wildlands, human causes still dominate statewide.

Fire Fighting

California has long been renowned as fielding the most professional fire suppression organization in the world. With an annual budget nearing 1 billion dollars, the California Department of Forestry, recently rechristened CALFIRE, manages a well-distributed, heavily capitalized and professionally staffed force of fire engines (over 1100), bulldozers (58), air tankers (23), Airtactical planes (13) and helicopters (11) along with 15 to 20 person hand crews supervised by the state Department of Corrections and the California Youth Conservation Corps (39). These initial attack forces respond to wildfires reported on non-federal lands and successfully suppress about 99% of them within tightly defined size and time limits (typically 50 to 100 acres and 2 hours).

In southern California, these forces are in most cases managed by counties under contract with and funded by the state. Federal fire protection forces such as the U.S.D.A. Forest Service respond to fires on federal lands with similar suppression success, and mutual aid agreements ensure that all firefighting resources in the state pitch-in whenever and wherever they are needed, with reimbursement of costs worked out after the fact.

Much of the emphasis in fire management in California is on initial attack, also referred to as preparedness; when successful, firefighting costs are far lower than when fires grow beyond initial attack into extended attack or campaign fires. Because of the size of the investment in firefighting infrastructure and the sometimes staggering operational costs, over the past decade, CALFIRE has implemented models to evaluate alternative initial attack deployments in an attempt to improve overall efficiency and efficacy of firefighting, and to promote equity in fire protection accomplishment around the state (Fried et al. 2006).

Fuels Management

In the past few years, increasing attention has been paid to fuels management, typically removal of live and/or dead vegetation via mechanical treatment and/or prescribed fire, as a strategy to slow fire spread and reduce fire intensity and consequent impacts. Analytic systems have recently been developed to assess the economic implications of implementing such fuels management over broad forested landscapes, and it appears that depending on assumptions about treatment choices and economic opportunities to convert mechanically removed material into saleable products, enough revenue could be generated to more than offset the costs of fuel treatment in California (Daugherty and Fried 2007). This is largely due to the fact that treatments which achieve significant crown fire hazard improvements require the removal of at least some trees of a commercial size that garners substantial revenue as timber, thereby defraying the costly removal of the small trees that comprise “ladder fuels” (Barbour et al. in press).

What is less clear is the extent of the benefits that could be ascribed to such management in terms of fire outcomes. Analysis has demonstrated that under extreme fire weather conditions, fire would likely occur as surface fire on 72 percent of the forest. Active crown fire would be expected on only 7 percent of the forest, and passive crown fire on 20 percent (Fried and Christensen in press).

Fuel treatments designed to reduce crown fire risk are difficult to justify in forests where models suggest that even extreme weather conditions are unlikely to lead to crown fire. Moreover, fire in the chaparral systems where most of the fires have occurred that led to lost lives and ruined homes is nearly always “crown” fire, and we know of no case where chaparral fuel treatment has ever generated positive net revenue. As shown in figure 1, the areas of high crown fire potential in forests are concentrated in the northern part of the state while the areas of chaparral are found mainly in southern California (Figure 1 - see attached file at bottom of page).

An alternative to attempting to treat fuels in all wildlands is to focus on reducing hazard at the WUI, where most of the economic value at risk is concentrated, and in California, this approach has been pursued with vigor. The state, in particular, has encouraged and financially supported the implementation of fuel treatments in WUI, often providing labor in the form of conservation camp crews to accomplish mechanical fuel treatment. In comparison to fuel treatments out in the non-WUI forest though, such treatments can be quite costly, focused as they often are on removing shrubs and the small trees that constitute “ladder fuels”, an activity that is highly unlikely to generate positive economic return. Moreover, logistical constraints and the high visibility of fuel treatment activity make almost any kind of fuel treatment in the WUI far more costly than in wildland forests.

There is also increasingly strict enforcement of defensible space requirements for residential property in many counties (e.g., maintaining a 300 foot wide buffer around homes free of flammable vegetation installation). The fact that defensible space laws have been in effect for many years and that mechanical fuel treatment has been practiced around WUI areas for decades makes these fuel management approaches familiar, and likely explains why California residents are far more accepting of these fuel treatment approaches than residents of fire-prone areas elsewhere in the U.S. (Vogt et al. 2005). However, enforcement of defensible space requirements varies by county, and where not strictly enforced, homeowners who do comply can remain at risk due to the inaction of neighboring landowners. On the windiest days, even a defensible space ten times the size of current requirements cannot ensure the survival of a home.

Future Prospects

Portending greater difficulties in managing future wildfires in California, the area of WUI and the number of homes within it grew substantially over the 1990s, to 7 million acres and 5 million homes (Hammer et al. 2007). Two kinds of WUI are recognized: interface, where high density housing, such as a subdivision, abuts wildland vegetation, and intermix, where relatively low density housing is intermingled with wildland vegetation.

As of 2000, the 5.2 million acres of intermix WUI accounted for 74 percent of WUI area but for only 32 percent of WUI homes. However, the area of intermix WUI grew nearly 12 percent between 1990 and 2000, accounting for most of the overall growth in WUI acres (nearly 97 percent). Growth in intermix WUI areas also accounted for the greatest percentage increase in number of homes: 25.1 percent, a figure far in excess of the 9.2 percent growth in housing across all California. Larger human populations usually result in more fire ignitions, and they certainly pose a challenge for managing a wildfire. Some county planning departments are proactively encouraging clumped development with large, vegetation free-buffers, and carefully thought-out emergency evacuation plans.

Indications are that climate change will not make fire management in California any easier. Predictions of fire behavior under climate change scenarios designed to reflect mid-century conditions absent change in the greenhouse gas trajectory show increases in fire spread rate and intensity which would have an impact on fire outcomes much greater than one might expect from the magnitude of these increases. Even modest shifts upward in the spread rate distribution translated to as much as a 100 percent increase in the number of fires which escape initial attack and have the potential to become large, damaging fires. Most of the impacts seen are in grass and chaparral vegetation—precisely where most of California’s WUI growth is occurring.

Conclusions

Clearly, California faces a number of challenges relating to wildfires. In addition to the fuels management, enforcement of defensible space, WUI growth and climate change aspects already mentioned, there is strong interest in restoring natural, usually defined as pre-settlement, fire regimes, especially on public lands. There are active conversations about who should bear the costs of fire protection, especially for new development, and about the feasibility of prescribed fire as a fuels management tool from the standpoint of responsibility for prescribed fires that escape and the constraints of air quality management.

Being prepared to respond to massively multiple fire situations spawned by lightning storms or arson binges remains a challenge, and post-fire effects and restoration remain active areas of concern and research (e.g., whether to stabilize slopes with rapidly establishing exotic grasses, propensity for debris flows on steep slopes post-fire, and productivity impacts on future forest stands resulting from loss of soil and/or soil nutrients).

Literature Cited

Barbour, R.J., J.S. Fried, P.J. Daugherty and R. Fight. [in press]. Predicting the potential mix of wood products available from timbershed scale fire hazard reduction treatments. Forest Policy and Economics.

Daugherty, P.J. and J.S. Fried. 2007. Jointly optimizing selection of fuel treatments and siting of biomass facilities for landscape-scale fire hazard reduction. INFOR: Information Systems and Operational Research 45(1):353-372.

Fried, J.S., C.L. Bolsinger and D. Beardsley. 2004. Chaparral in Southern and Central Coastal California in the Mid-1990s: area, ownership, condition, and change. Gen. Tech. Rep. PNW-RB-240. Portland, OR: Pacific Northwest Research Station, Forest Service, U.S. Department of Agriculture.86 p

Fried, J.S. and G.A. Christensen. In press. Crown Fire Hazard. In, Christensen, G.A.; Fried, J.; Campbell, S. (tech. eds.) California’s forest resources, 2001-2005, Gen. Tech. Rep. PNW-GTR-xxx. Portland, OR: U.S. Forest Service, Pacific Northwest Research Station.

Fried, J.S., J.K. Gilless and J. Spero. 2006. Analyzing initial attack on wildland fires using stochastic simulation. International J. of Wildland Fire 15(1):137-146.

Hammer, R.B., V.C. Radeloff, J.S. Fried and S.I. Stewart. 2007. Wildland-urban interface growth during the 1990s in California, Oregon, and Washington. International J. of Wildland Fire 16:255-265.

Vogt, C.A., G. Winter and J.S. Fried. 2005. Predicting homeowners’ approval of fuel management at the wildland-urban interface using the Theory of Reasoned Action. Society and Natural Resources. 18(4):337-354.

Figure 1. Most forests where crown fire is likely are in the northern half of the state; most chaparral can be found in southern California and on the central coast. Fire type is predicted for forested areas only by kriging FVS-FFE predictions for nearby FIA plots. (Click on attachment at the bottom of the page)