Southern California has experienced a series of extraordinarily large and destructive fires over the past decade. The 2003 firestorm, which was widely considered a hundred-year event, was rapidly followed by the 2007 firestorm. These fires incurred billions of dollars in costs associated with fire fighting, damage to property from fire and erosion, and impacts on human health and ecosystem services. The 2003 Cedar fire burned more than 270,000 acres in San Diego County, was responsible for the death of 14 civilians and one firefighter, and destroyed over 2800 buildings. The cost of firefighting alone exceeded $27 million. Insured losses in the Cedar Fire were at least $1.06 billion. The fire caused significant health problem for many who lived beyond the fire’s immediate reach; hospital and emergency room admissions for respiratory diseases increased by 34% to 67% during and after the fire. The October 2007 Witch fire was of comparable magnitude, burning over 197,900 acres in San Diego, damaging 1650 structures and causing 2 deaths. The arson-caused Station fire continues to burn as we finish this proposal. The Station fire has consumed more than 160,000 acres, resulted in the death of two firefighters, and is responsible for delaying the deadline for this proposal. The largest and most destructive fires in Southern California occur during intense Santa Ana wind events (Westerling et al. 2004). An improved ability to forecast how these fires are likely to change in the future is critically needed. This ability will require a multi-disciplinary approach that accounts for climate change effects on regional winds and ecosystems, other global change impacts on ecosystems at the wildland/urban interface, changing patterns of land use, and responses of land and fire managers to this evolving system.

Southern California’s landscape

Southern California provides an excellent natural laboratory for investigating the impacts of fire on ecosystem function and air quality, and for understanding the interactions between fire, climate, and vegetation dynamics. California’s Mediterranean climate, with a mild, wet winter and spring growing season followed by a predictable, warm, summer drought, and then extraordinarily strong, dry, and hot Santa Ana winds in autumn, creates the potential for large and intense wildland fires. Southern California’s landscape is climatologically and ecologically diverse; the orographic precipitation gradient associated with elevation supports a wide diversity of vegetation types (grasslands and shrublands in the coastal plains and foothills, larger shrubs, evergreen hardwood trees, and tall conifers in the mountains, and smaller and more sparse trees and shrubs on the desert side). Wildland fires are an inevitable component of many of the upland ecosystems (Fig. 1), and much of the vegetation is strongly adapted to periodic crown fire. Development in Southern California is spread out on the coastal plain and inland valleys at lower elevations, creating large areas at the wildland-urban interface.

Projected climate change in southern California

California will face intensifying climate change in coming decades (2009 Scenarios Project by the California Energy Commission). All model projections for California indicate increased temperature, with more pronounced warming in summer than winter (Field et al. 1999; IPCC 2001, Snyder et al 2002; Hakkarinen and Smith 2003; Hayhoe et al. 2004; Cayan et al. 2008; Diffenbaugh et al. 2008). Predictions of precipitation are less certain, though there is a growing consensus that variability willincrease and mean precipitation may decrease (Dai et al. 2001; Seager et al 2007; Cayan et al., 2008; Diffenbaugh et al. 2008; Kim et al. 2009). Southern California is at the southern edge of the winter storm track. Winter precipitation increases markedly from south to north in California, and a small shift in the mean location of the winter storm track would have a dramatic effect on precipitation. A change in storm track position or a change in the frequency and intensity of ENSO events could cause extreme climate change in Southern California. The increase in temperature will likely increase evaporation, thus amplifying any decrease in precipitation and causing more severe droughts. Dealing with extreme climatic events will be an important aspect of the state’s adaptation to climate change (CNRA 2009).

Expected effects of climate change on Santa Ana winds

Despite their importance for climate and fire weather, few studies have examined the variability of Santa Ana winds on time scales longer than a few months. Miller and Schlegel (2006) used global pressure gradients to forecast Santa Ana Winds, and found a shift in Santa Anas from earlier (Sept-Oct) to later (Nov- Dec) in the season, which they suggested will increase the extent of CA coastal areas burned. A more recent study found a decline in the incidence of Santa Ana conditions over the 44-year ERA40 reanalysis period, with 30% fewer events from 1991 to 2001 relative to 1959-1969 (Hughes and Hall 2009a). Additional analyses indicted a future decrease under the SRES-A1B emission scenario in the frequency and intensity of SA events in the mid-21st century compared with that in the late 20th century (Hughes and Hall 2009a).

Expected effects of climate change on ecosystems

cological models forecast that climate change will have a major impact on California’s ecosystems. Lenihan et al. (2003, 2008) used a dynamic vegetation model (MC1) to simulate the response of vegetation distribution, carbon, and fire to historical and forecast climate (IPCC projections of 1% increasing greenhouse gas concentrations per year). The increasing temperature under the IPCC scenarios caused a shift in dominance from needle-leaved to broad-leaved vegetation in some parts of the state. The area of some ecosystems, coastal sage in particular, decreased markedly. The response to changing precipitation was complex, involving both the effect of soil moisture on vegetation productivity and changes in woody plant–grass competition mediated by fire; the model indicated fire will play a critical role in the response of vegetation to altered precipitation. Fire may slow the encroachment of woody vegetation in grasslands under wetter conditions or hasten the transition from woody communities to grassland under drier conditions.

Expected land use patterns over the next decades

California has experienced rapid population growth, with a 2% annual growth rate (Potter et al 2007), resulting in rapid and extensive urbanization and loss of natural vegetation and cropland (Beck and Kolankiewicz 2001). 13% of the land urbanized in 42 of the state’s 58 counties from 1984 to 1990 occurred on irrigated prime farmland; 48% occurred on wildlands or fallow marginal farmland (Charbonneau and Kondolf, 1993). Future development will inevitably push into the coastal mountain ranges and into Riverside and San Bernardino counties (Sanstad et al. 2008; CAT 2009), further increasing the wildland-urban interface and thus wildfire risks to ecological and social systems. More than 9 million acres (9.1%) of California is currently considered a flammable and developed environment (FRAP 2008).

Historic fire regime and future prediction

Several studies have estimated the impact of future climate on wildfire in Western United States. The number of days of high fire danger has been predicted to increase in the 21st century across western US, mainly due to changes in relative humidity (Brown et al., 2004). Recent work (Westerling and Bryant 2008) shows that large fire risk (the probability of fires greater than 200 ha) is highly sensitive to precipitation in future climate scenarios, with decreases of up to -29% predicted for wetter scenarios and increases by up to 28% for drier scenarios. Additional work has indicted that wildfires in Western US will increase 50% by 2050, based on a temperature increase of 1.6oC (Spracklen et al. 2009).Neither of these studies considered the effects of changes in wind conditions, which may accompany climate change, even though the strength and direction of winds greatly influences fire behavior (Fried et al. 2004). The impacts of pre-fire species composition and fire weather on mortality, combustion completeness, and carbon losses also are not well understood. The longer- term impacts of fire severity and frequency for post-fire vegetation recovery have not been systematically characterized, and the modeling of vegetation recovery and succession after fire is still in its infancy. Most studies on the impacts of fires on air quality focused on individual fire events using point measurements, and much less is known on how future fires will affect air quality (Spracklen et al. 2009). All these call for an interdisciplinary study integrating field measurements, remote sensing, air quality and dynamic vegetation modeling.