Many factors increase susceptibility of forests to wildfire. Among them are increases in human population, changes in land use, fire suppression, and frequent droughts. These factors have been exacerbating forest susceptibility to wildfires over the last century in southern California. Here we report on the significant role that air pollution has on increasing forest susceptibility to wildfires, as unfolded in the San Bernardino Mountains from 1999 to 2003. Air pollution, specifically ozone (O₃), and wet and dry deposition of nitrogenous compounds from fossil fuel combustion, has significantly increased since industrialization of the region after WWII. Ozone and elevated nitrogen deposition cause specific changes in forest tree carbon, nitrogen, and water balance that enhance individual tree susceptibility to drought and bark beetle attack, and these changes contribute to whole ecosystem susceptibility to wildfire. For example, elevated O₃ and N deposition increase leaf turnover rates and leaf and branch litter, and decrease decomposability of litter. Uncharacteristically, deep litter layers develop in mixed conifer forests affected by air pollutants. Elevated O₃ and N deposition decrease the proportion of whole tree biomass in foliage and roots, the latter effect increasing tree susceptibility to drought and beetle attack. Because both foliar and root mass is compromised, carbohydrates are stored in the bole over winter. Elevated O₃ increases drought stress by significantly reducing plant control of water loss. The resulting increase in canopy transpiration, combined with [O₃ + N deposition]-induced decreases in root mass significantly increase tree susceptibility to drought stress, and when additionally combined with increased bole carbohydrates, perhaps all contribute to success of bark beetle attack. Phenomenological and experimental evidence is presented to support the role of these factors contributing to the susceptibility of forests to wildfire in southern California.

Diagram of factors

Many factors combine to increase forest susceptibility to wildfire in southern California, and most of these were set in motion decades ago. These factors include a rapid increase in human population and resource use; a shift from timber production to recreational forest use; fire suppression with subsequent forest densification; periodic, extreme drought; and bark beetle outbreaks. The contribution of air pollution to forest susceptibility to wildfire has not been studied extensively. In this paper, we will link air pollution to increasing forest densification, litter build up, drought stress, tree susceptibility to successful bark beetle attack, tree mortality, and increased forest susceptibility to wildfire (Figure on the right). A case study will be presented for the San Bernardino Mountain Range in the Transverse Range north and east of Los Angeles, California. We will focus on pollutant effects on ponderosa pine, which dominates the mixed conifer forest in the western part of the range.

In the late 19^th century, gold and other valuable minerals were discovered in the San Bernardino Mountains, and the population rapidly increased (Minnich 1988). The forest was logged for buildings, mine shaft support, and for fuel. In 1899, a severe drought occurred, water was limiting, and a premium was placed on reservoir development (Lake Gregory, Arrowhead, Big Bear). As the reservoirs were established, they became magnets for recreation use in the 1920s. With the shift from resource utilization to recreation, incursions of fire from the chaparral into the forest were suppressed, and forest density increased through the 1940s. In the 1950s, the Forest Service made an attempt to thin the forests, but, for aesthetic reasons, the mountain communities strongly opposed both branch trimming and stand thinning. As a consequence, the forest continued to increase in density, and trees grew increasingly closer to structures. In the 1980s, the community councils drew up “Forest Plans” that included branch trimming and thinning of trees within 30 m of valued structures (Asher and Forrest 1982). However, these recommendations were not followed or enforced. The region was, and is, highly susceptible to wildfire.

Long term record of precipitation

Although there is an increase in evapotranspiration from west to east, weather that results in precipitation in the San Bernardino Mountains is generally a regional phenomenon. We have used the longest record of precipitation for the range, collected over the last 120 years at Big Bear Dam (San Bernardino Water Management District), to identify the level of drought stress experienced by ponderosa pine from year to year. Moderate drought stress is defined physiologically as reduced cell turgor that generally results in reduced stomatal conductance (reduced water loss from the leaf), and lower cellular water potential, which allows the tissue to hold onto the water that is in the leaf more tenaciously (Levitt 1980). In 1994, a year of 80 percent of the average precipitation (preceded by an above-average precipitation year), ponderosa pine experienced moderate drought stress from mid-July through the end of the growing season (Grulke 1999). Severe drought stress is also accompanied by reduced cell turgor, reduced stomatal conductance, and reduced cell-water potential. The water potential is lowered sufficiently that cell solutes are concentrated enough to disrupt enzymatic function, and cell turgor is reduced enough and for a long enough duration that cell elongation growth is limited. Needles produced in years of severe drought stress are shorter. In 1996, a year of 60 percent of the average precipitation (preceded by an above-average precipitation year), ponderosa pine experienced severe drought stress from the end of June through the end of the growing season (Grulke 1999). Over the period of the long-term precipitation record, roughly 15 percent of the years had low enough total annual precipitation to result in moderate drought stress; 30 percent of the years had low enough total annual precipitation to result in severe drought stress. Using this rough index of the level of physiological stress, ponderosa pine experienced drought stress 45 percent of the years since 1883 when precipitation records were initiated (see figure on right).

Where O₃ exposure and nitrogen deposition reduce root biomass, trees are predisposed to drought stress. In general, low to moderate O₃ exposures (<60 ppb hourly O₃, averaged over 24 hours for the 6-month growing season) reduce water loss from trees. O₃ reduces photosynthetic rates, less CO₂ is required, and stomatal apertures are reduced to conserve water. However, under concentrations that are moderately high or higher, O₃ exposure modifies stomatal behavior in ways that increase drought stress.

For example, sugar maple was exposed to O₃ concentrations of 70 ppb during daylight hours (Tjoelker and others 1995). Early in the growing season and experiment, neither the net photosynthetic rate nor stomatal conductance was affected by the treatment. By midseason, there was a significant decrease in water-use efficiency—at the same level of carbon gain, seedlings growing in chronic O₃ exposure had twice the level of water use as had control seedlings grown in charcoal-filtered air. By late season, both net photosynthesis and stomatal conductance were suppressed in plants grown in chronic O₃ exposure. In a field study of sensitive and tolerant genotypes of Jeffrey pine exposed to the same ambient O₃ levels (~68 ppb O₃ averaged over 24 hours, for the 6-month growing season in Sequoia National Park), sensitive genotypes had lower water loss under moist, favorable conditions and higher water loss under dry, unfavorable conditions (Patterson and Rundel, 1989). Under favorable conditions, Jeffrey pine had less water loss, but because the stomatal apertures were smaller, there was also less photosynthetic carbon gain. Under unfavorable conditions (most of the day in the Sierra Nevada), sensitive Jeffrey pine had higher water loss, which would result in greater desiccation.

Although physiologists often report plant response under steady state (stable) conditions, the light environment in the forest is often dynamic. Understanding stomatal responses under rapidly changing environmental conditions with concurrent O₃ exposure can perhaps better explain why trees exposed to moderately high and higher concentrations of O₃ lose more water. In typical forest environments, foliage on a primary branch on the southern aspect of an open-grown tree receives flecky light two-thirds of the time (Grulke 2000). For example, the cutleaf coneflower (Rudbeckia laciniata var. digitata) is one of the most sensitive native plants to ambient O₃ concentrations in Great Smoky Mountains National Park. It persists in forest gaps and on forest-meadow margins, both with flecky light environments. Tolerant plants of cutleaf coneflower had normal responses to experimentally manipulated change in light from low to high and back down to low levels. However, O₃-sensitive plants had either no stomatal response or a muted stomatal response to changes in light level. The level of water loss from leaves with no or muted response to changing light level was high—they did not conserve water when light was low, and this failure to conserve water would contribute to desiccation. When humidity was lowered slowly, O₃-sensitive plants closed their stomata at much lower relative humidities than did O₃-tolerant plants, and this also contributed to greater desiccation (Grulke and others 2007). In a similar experiment with California black oak saplings (Quercus kelloggii) exposed to anthropogenic high O₃ in a natural stand, stomatal closure in response to abruptly reduced light level was slower in plants without additional N amendment, and N amendment partially mitigated the desiccating effects of high O₃ exposure (Grulke and others 2005).

Moderate to high O₃ exposure can also cause stomata to remain partially open at night. In experimental O₃ exposures, this was first observed in Norway spruce (Picea abies), (Weiser and Havranek 1993) and in birch (Betula pendula), (Matyssek and others 1995). Nighttime water loss rates were 25 percent as great as full daytime rates for Norway spruce, and 50 percent as great for birch. This was corroborated in ponderosa pine across the San Bernardino Mountain pollution gradient, with both higher O₃ and NO₂ and HNO₃ exposure. In the San Bernardino Mountains, nighttime water losses were 10 percent as great as full daytime rates (Grulke and others 2005). Because these studies were largely phenomenological, a new gas exchange system was designed and built to directly test known O₃ concentrations on single leaves. Chronic, moderate O₃ exposure (70 ppb O₃ for 8 hours per day for 1 month) significantly increased nighttime foliar water loss in California black oak and blue oak (Quercus douglasii) (Grulke and Paoletti 2005). Nighttime water losses were attributable directly to O₃ exposure and were 30 percent and 20 percent, respectively, as great as daytime rates in these species. Moderately high (or higher) O₃ exposure increases foliar water loss and increases tree susceptibility to drought stress.

Fig. 1 - Long term record of precipitation

Air pollution exposure (O₃ and N deposition) increases tree susceptibility to drought stress, and drought stress increases tree susceptibility to successful beetle attack. Dunning (1928) was one of the first to report a relationship between drought conditions and increased levels of tree mortality caused by western bark beetles. In the west, bark beetles reach epidemic proportions after 2 to 3 years of drought. The correlation between beetle attacks and climate can be diffuse because bark beetles may delay or prolong the exact time of tree mortality. However, mortality tends to increase in multiyear droughts, particularly in highly dense stands or those with pre-existing damage or stress. Figure 1 indicates that sequences of 2 to 3 years of drought have occurred 9 times (Table: List of drought years) in the last 120 years. We can document five bark beetle epidemics associated with those sequences.

Beetles are opportunists that attack trees in a weakened state. With only a few exceptions, either the host tree is killed by the colonizing bark beetles or the host resistance of the tree kills the attacking adults. To kill a tree, large numbers of bark beetles must successfully colonize it in a relatively short period of time (Paine and others 1984, Paine and others 1997). However, fewer beetles may be sufficient to kill a compromised tree (Paine and others 1984). The bark beetles most commonly responsible for tree mortality in the western San Bernardino Mountains are western pine beetle (Dendroctonus brevicomis) and mountain pine beetle (D. ponderosae). Western pine beetle can produce up to 4 generations in a year in southern California, where the mild thermoclimate permits populations of this species to expand rapidly when an abundance of susceptible host material is available for colonization.

Eggs are laid in the inner bark and the larvae excavate galleries. Pupation occurs in either the inner or outer bark, depending on the species of beetle. Adults emerge from the larval host tree and search for susceptible hosts. Healthy pines and firs respond by exuding pitch, which either pitches out the adults or blocks their progress. Resin production impedes bark beetle attack both physically and chemically. Oleoresin pressure, caused by turgidity of cells lining the resin ducts, forces preformed resin to the site of injury or invasion and results in a flushing action. The cell turgidity is derived from the transpirational stream, so if the tree is under moisture stress, the cells become increasingly flaccid, the resin pressure is reduced, and the effectiveness of the preformed resistance is compromised (Vite 1961). In weak trees with reduced resin pressure, the adults are able to initiate colonization and produce specific pheromones that attract other colonizing adults. Pheromone production ceases when the host tree ceases resin flow, (i.e., when the tree dies) (Raffa and Berryman 1983).

Severe drought and other stresses also reduce the photosynthetic capacity of trees and the levels of carbohydrates used for growth, defense, and tissue repair. This can have significant impact on the ability of the tree to induce an effective response to invasion (Paine and Stephen 1987a, b). Drought-stressed trees are also known to have elevated levels of free, translocatable proteins (Lei and others 2006), which are produced to generally increase cell osmoticum. We conjecture that increased bole carbohydrate content as a result of O₃ exposure + N deposition + drought and elevation of protein levels in response to drought enhance beetle fecundity. Pollutant-exposed trees may thus be primed for successful bark beetle attack.

The forest had been recently thinned early in the late 19^th century by commercial logging, so we would not expect to observe an epidemic beetle infestation in immature, low-density stands despite the drought stress experienced in the late 1920s (Minnich and others 1995). Human population in the Los Angeles Basin significantly increased after World War II, but air pollution levels were not quantified (or reconstructable) until 1963. From 1963 through 1980, peak 1-hour O₃ concentrations averaged 250 to 425 ppb (Lee and others 2003). From 1980 on, peak 1- hour O₃ concentrations were still high (>250 ppb), but cumulative O₃ exposures over the growing season began to decline. Through strong regulatory controls, O₃ concentrations declined further to tens of occurrences to only isolated occurrences of hourly concentrations exceeding 95 ppb from the mid-1990s to present. Throughout this time, N deposition continued to accumulate, and drought stress was exacerbated by chronic, if not acute, O₃ exposure.

Fig. 2 - Remote imagery of Cedar Pines Park

The most extreme drought recorded in 250 years was experienced in the hydrologic year 2002 (10/1/01 through 09/30/02) after 3 years of chronic drought. Results of 3 years of chronic drought (1999-2001) and extreme drought (2002) are shown in a sequence of imagery taken at 5 km above the forest canopy at the most polluted site in the western San Bernardino Mountains (Figure 2). After the chronic drought, bark beetles attacked, and there were clusters of tree mortality. However, the average stand tree mortality rate was near 0 percent. After the chronic and severe drought, tree mortality (primarily ponderosa pine) due to both bark beetle and drought increased to approximately 5 percent at the stand level. In the spring following the wet winter, bark beetle populations reached epidemic proportions, and 42 percent of the stand had died (ponderosa pine, white fir, and sugar pine). The stand was at high risk for an intense fire with high litter layers, high numbers of standing dead trees, and exacerbated drought stress. In autumn of that year, the Old Fire swept through the stand. Interestingly, not all of the red trees (standing dead trees with needles retained) were consumed in a crown fire because the highly dense understory was not in contact with the lower branches of the 100+ year- old-trees—the effects of O₃ exposure, N deposition, and drought had promoted lower branch abscission (Miller and others 1996b), so that the lowest branches were attached at 60’ or greater. Trees are still dying from bark beetle at this site (as of 7/06), but the rate of change is now statistically undetectable.

The role of air pollutants in increasing tree susceptibility to drought, successful bark beetle attack, tree mortality, and the susceptibility of forests to wildfire have not been studied extensively. Air pollutants, specifically strong oxidants and nitrogen deposition, contribute to increased litter accumulation and increased tree susceptibility to drought stress. It is well known that drought-stressed trees are more susceptible to successful bark beetle attack. The combination of chronic drought in 1999-2001 and acute drought in 2002 resulted in a bark beetle epidemic in the western San Bernardino Mountains. We contend that the severity of tree mortality in the western San Bernardino Mountains was significantly exacerbated by the higher air pollutant deposition in this region.