Yosemite National Park's resource managers are attempting to integrate ecosystem-based management techniques (such as prescribed burns) with the impact of 4 million visitors per year. We believe that the presence of root disease in conifers and hardwoods will play an important role in determining the direction of vegetation management in the park, particularly Yosemite Valley. Our preliminary surveys indicate that root disease is the most common disturbance associated with canopy gaps in Yosemite Valley (as compared to bark beetles or wind-throw). Since 1970, well over 200 canopy gaps associated with root disease have been recorded in Yosemite Valley, and approximately one-half of these have been mapped. In some developed areas of the valley such as campgrounds and lodging facilities, one-third to one-half of the forested area is associated with root disease infestations in conifers (authors, unpublished).

Root disease in forest trees has influenced management decisions within Yosemite Valley in at least three ways. First, wind-thrown trees with rotted roots have caused human fatalities as well as extensive property damage and liability settlements. Hazardous trees were first reported as a problem in the 1950s. Since 1973, tree failures have resulted in seven fatalities, 19 serious injuries and approximately $1 million in damages (National Park Service 1997). This includes damage to buildings, tent cabins, automobiles and power lines. The option of hazardous tree removal in developed areas of the park must be balanced among the often competing requirements for public safety, aesthetics and ecosystem function. Second, canopy gaps associated with root disease can have a potentially important effect on biodiversity. The composition of the dominant tree species may be altered, and regeneration of plant species in stand openings may be influenced. Third, mortality of trees associated with root disease will also increase fuel loads and influence fire behavior in the park. This is critical in light of efforts to restore historical influences on ecosystem processes in Yosemite Valley.

The problems currently associated with root disease in Yosemite Valley appear to be the result of a series of natural-resource management actions taken many years earlier. The current state of the forests of Yosemite Valley is another example of unintended consequences that may result from vegetation-management decisions.

Valley's natural history

Yosemite Valley is located within Yosemite National Park, in the central Sierra Nevada of California. The Merced River flows in a westward direction through the valley, creating a relatively flat, alluvial plain about 7 miles (11 kilometers) long and averaging 0.5 miles (0.8 kilometers) wide with an elevation of approximately 4,000 feet (1,220 meters). Sheer granite walls up to 3,300 feet (1,000 meters) high enclose the valley floor on three sides. The Mediterranean-type climate is characterized by hot, dry summers and cool, wet winters. The high cliffs of the valley cause some microclimatic differences on the valley floor. The south side tends to have shadier, cooler conditions during summer months and lower temperatures in winter, and the north side tends to have warmer, drier conditions throughout the year. Winter and spring floods are common, each time leaving another layer of fine silt and sand over as much as 50% of the valley floor (Heady and Zinke 1978).

Vegetation in the valley before the arrival of Euroamericans was predominantly scattered black oaks (Quercus kelloggii) and large ponderosa pines (Pinus ponderosa) intermixed with meadows of sedges, grasses and wildflowers (Gibbens and Heady 1964; Heady and Zinke 1978). This plant community existed in the presence of both natural wildfires and fires set intentionally by Native Americans. Evidence suggests that humans have inhabited Yosemite Valley at various times for at least 9,000 years. About 650 years ago, a more sedentary lifestyle associated with the acorn as a major food source seems to have been adopted by the Native Americans (National Park Service 1997). Fire and sometimes hand-eradication were used to clear small, invading trees from the meadows. This encouraged a flush of new growth in grasses and shrubs that provided highly nutritious food for wildlife, enhanced berry production, allowed for efficient acorn collection, and permitted easier foot travel.

Euroamericans came to Yosemite Valley in the 1850s. They planted crops and orchards, and created pastures by fencing and draining meadows. Fire was assumed to be destructive to agriculture and forestry, so meadow burning was phased out on the valley floor and wildfires were routinely extinguished. Beginning in 1864, the valley came under the supervision of a succession of state and federal entities. Park managers in the early part of this century had two often conflicting goals: to preserve natural conditions as much as possible and to encourage public use. Fire was thought to be essentially destructive and wildfires were suppressed. The glacial moraine was opened with explosives in 1879, reducing flooding and lowering the water table in portions of the valley. Agriculture continued throughout this period, but decreased with the advent of large numbers of tourists. Most of the meadows were grazed by cattle until 1924, when the last dairy herd was removed.

As a result of fire suppression policies and meadow draining, conditions improved dramatically for conifer germination and for seedling and sapling survival. Dense stands of conifers came to occupy much of the valley floor by replacing significant portions of meadows and oak woodlands. The predominant forest type in Yosemite Valley is now a typical west-side, mid-elevation, mixed-conifer Sierra forest containing dense aggregations of pon-derosa pine and incense-cedar (Calocedrus decurrens), with smaller numbers of black oak and other species. On the hotter, drier north side of the valley, scrub oak (Quercus chrysolepis) is present in the understory below mostly ponderosa pine. On the shady south side, white fir (Abies concolor) and Douglas-fir (Pseudotsuga menziesii) constitute a significant portion of the stand. Many of the pines have now reached sizes in excess of 40 inches (100 centimeters) in diameter and 150 feet (50 meters) in height. Many large oaks died or suffered extensive decay as a result of being overtopped and heavily shaded by taller conifers. Vistas from, the valley floor of sheer granite cliffs, domes and waterfalls have often disappeared behind a wall of intervening forest (Gibbens and Heady 1964).

Root-disease pathogens

There are several fungal root diseases native to Yosemite Valley. From a resource-management perspective, Heterobasidion annosum is currently the most important disease-causing pathogen. The fungus causes root decay and mortality of coniferous trees throughout the Northern Hemisphere (commonly known as annosus root disease). The fungus initially becomes established in a forest stand via airborne spore infection of living trees. The presence of fresh, still-living stump tops associated with tree felling increases the likelihood that H. annosum will be established. The fungus colonizes the stump and then spreads into surrounding trees through root-to-root contacts; H. annosum cannot grow through soil. Once infected, diseased pines tend to die quickly because they are predisposed to attack by the western pine beetle (Dendroctonus brevicomis) and other scolytid beetles. Because bark beetles associated with incense-cedars are generally less aggressive, these trees — when infected — tend to die slowly and suffer extensive root decay, which causes them to fall. The mortality of trees leads to the formation of openings in the forest canopy which continue to enlarge as the fungus moves into adjacent trees, following the death of a tree, H. annosum is capable of living as a saprobe. Receiving nourishment from the decaying tree, it slowly decomposes the underground root wood for a number of years. Because of this saprobic phase, the fungus has the capacity to infect regenerating conifers within a gap for 30 to 50 years. It is unlikely that H. annosum will infect trees already dead from other causes, as roots of these trees are rapidly colonized by competing fungi.

Armillaria mellea (commonly known as oak root fungus or shoestring root rot) also appears to be present throughout Yosemite Valley, particularly in association with oaks. A. mellea does not usually cause major losses in native stands, although it is an important disease of orchard and landscape trees in California. Lack of overstory mortality in natural forests does not necessarily mean absence of the fungus. Our studies in other forest ecosystems in California indicate that 50% to 100% of trees in forest stands may support A. mellea on their root systems, usually confined to small lesions on the outer parts (Baumgartner and Rizzo, unpublished). As native trees are cut, the fungus coionizes the dying root systems, decays the wood and builds inoculum. Under the right conditions, with a susceptible host and favorable environment, A. mellea will then attack adjoining trees. Unlike H. annosum, A. mellea can kill both hardwoods and conifers. Small forest-canopy gaps caused by tree mortality associated with A. mellea are now common in the valley.

Impact of past practices

While fungal root diseases are clearly part of the natural ecosystem in Yosemite Valley, human activities have influenced the establishment and spread of these diseases over the past 100 years. As the number of people visiting Yosemite and staying over-night night in the valley increased, projects were undertaken from the 1860s to the present to enhance the tourist experience. These have involved felling trees to open scenic vistas and accommodate building projects: campgrounds, tourist lodging, residences for park personnel and parking lots.

Droughts daring the 1920s and 1930s led to extensive outbreaks of western pine beetle and subsequent ponderosa pine mortality in Yosemite Valley. While timber production was not a goal within the park, the excessive tree mortality concerned park resource managers. Control of beetle outbreaks was attempted by felling beetle-infested trees during the winter months. The beetle-infested bark was then peeled off the top half of the log, piled alongside, and burned (Miller and Keen 1960). The fell-peel-burn method was used extensively during the 1920s and 1930s by managers, with the goal of saving as many of the remaining large pines in Yosemite Valley as possible.

These management actions produced thousands of stumps during this period. As described previously, freshly cut stumps are an important starting point for infection by H. annosum. Unfortunately, the importance of stump tops in the fungal infection process was not demonstrated until the 1950s. The potential impact of tree felling was not known at the time it was used extensively. By 1970,161 gaps associated with H. annosum had been identified in Yosemite Valley. These gaps were found primarily in campgrounds and other developed areas (e.g., Yosemite Lodge, Camp Curry, Ahwahnee Hotel, employee housing), but a number of root-disease gaps were located in the less-developed west end of the valley. In all cases, the gaps were initially associated with a stump remaining after tree felling.

J.R. Parmeter and colleagues at UC Berkeley and U.S. Department of Agriculture's Forest Service (Felix et al. 1974; Marosy and Parmeter 1989; Parmeter et al. 1978, 1979) documented and mapped 68 discrete gaps associated with H. annosum and created a database that has been used to follow progression of the disease over the past 30 years. Today these gaps range in size from just a few square yards to over 1 acre (4,000 square meters), with a mean gap size of 0.2 acre (712 square meters). By visiting these 68 gaps at 1 to 5 year intervals, we determined that the gaps enlarge radially from an initial stump at an average rate of 1.5 feet/year (0.47 meters/year). However, enlargement rates are variable, even within a gap, and may reach as high as 5 feet (1.5 meters) per year (table 1).

Recent research has revealed that the fungus does not spread indefinitely from the initial point of infection. Between 1965 and 1996, tree mortality ceased at 36 of 68 gaps. Such stabilization of gap enlargement is common throughout California east-side pine forests (Slaughter and Parmeter 1995). While still under study, reasons for this stabilization include a combination of factors acting along gap perimeters such as competing root-decay fungi, the presence of tree species not susceptible to infection by H. annosum, or trees that are already dead when the fungus first encounters their roots.

The effect of H. annosum as it moves through a stand is exemplified by the Sentinel Beach gap (table 1). As it enlarged to its current size of 0.63 acre (2,535 square meters), 106 trees were killed. This included a number of pines that were well over 40 inches (1 meter) in diameter. The pines tended to die much more quickly (within a few years of root infection), while cedars — with nonaggressive bark beetle associates – declined more slowly. Cedars generally die 10 to 30 years after adjacent pines die, giving more time for H. annosum to occupy a large portion of the still-living root system. The consequence of this is a rapid conversion of the stand to an overstory of incense-cedars, which also die over time, leaving openings in the forest canopy that are not occupied by trees. After 30 years, most of the gaps we surveyed had not returned to a closed forest canopy. In campgrounds, most gaps are now bare ground with no vegetation. Gaps in less developed areas are mostly occupied by herbaceous plants, shrubs and, less commonly, conifer regeneration. In no instance has regeneration in the gaps reached higher than 10% of the surrounding canopy height.

Once it became clear that root disease was associated with most conifer mortality observed in Yosemite Valley, park managers responded with a vigorous program of hazardous tree removal. Researchers developed a hazard rating system for incense-cedars (work on pines was never completed) to allow on-site workers to identify trees most likely to fail (Srago et al. 1978). As time passed, some canopy gaps ceased enlarging, while many others continued to sustain tree mortality and wind-throw at the gap edges. Discrete gaps often coalesced into large forest openings as tree mortality and removals continued. As part of a program to remove potential H. annosum-infected trees from the developed areas and prevent further spread of the disease, 247 trees were felled in the Yosemite Lodge complex between 1985 and 1987 (West 1989). There are now two large areas, each approximately 10 acres in size, within the lodge complex area that are virtually treeless. While the spread of the disease has apparently been slowed in this area, the presence of symptomatic cedars that were not removed during the original management action suggest that the fungus is still present and active on the site.

Many small root-disease gaps have also become established since the 1970s. These occur often but not always in locations adjacent to older, stabilized gaps. Some of these new gaps are not associated with any stumps, suggesting that there may be other ways for the fungus to spread throughout the valley forest besides airborne spore germination on freshly cut stump surfaces.

Complexities facing managers

The current goal of vegetation management in Yosemite National Park is to preserve, restore and perpetuate the natural processes that act upon the native plant life as part of natural ecosystem functioning (National Park Service 1997). Under this management protocol, vegetation in Yosemite Valley would be expected to gradually return to the open meadows and predominantly open oak woodlands present before the arrival of European immigrants. The extensive stands of high-density large pines and cedars that currently occupy the valley would be significantly reduced in area. Tree mortality is now considered as just one of numerous, ongoing ecological processes.

Gaps caused by root disease in the valley forest canopy are continually opening, allowing colonization by a variety of plant species (shrubs, grasses, herbs) not normally present in the shady understory of dense stands of large trees. These gaps also provide edge effects and increased food for wildlife. Gaps can re-establish vistas from the valley floor that had become obscured by trees. It could, therefore, be argued that naturally occurring root diseases do not necessarily have a negative effect on park vegetation. In fact, root disease may aid the park in its goal of reducing the density of conifers in the valley. Given this perspective, a fungus such as H. annosum may cause disease but also not be considered a “pest.” It would simply be part of an ecological process that promotes biodiversity.

From a different perspective, the campground manager's goals would include providing a safe environment for the public. When confronted with a large pine having just fallen and crushed a tent, he or she (and the tent owner) would probably not consider root disease as a positive part of the environment. The hazardous tree removal program is designed to prevent such an occurrence, and is an example of the trade-offs required when management goals conflict. To avoid wind-throw in developed areas of the valley and along major roads, trees must be removed when root disease causes substantial decay. Such decisions may initially be unpopular with the public because trees provide shade for campers, protection from wind and habitat for animals. Undeveloped portions of the park (most of the area outside Yosemite Valley) do not require such a program.

These examples demonstrate that like their 19th-century counterparts, resource managers in national parks today face conflicting goals: to preserve natural conditions as much as possible while encouraging public use. An additional example of the conflict between ecosystem function and public use is demonstrated by efforts to restore fire to Yosemite Valley ( see sidebar page 22 ). Planned ignitions (prescribed burns) have been used since the 1970s to reintroduce fire into the ecosystem. We do not have adequate data on the influence of fire on disease development. For example, residual trees may be more susceptible to infection by H. annosum following a prescribed burn, as fire scar wounds may act as entry points for the fungus. In addition, increased fuel loads within enlarging gaps may make it difficult to control fire intensity during prescribed burns. The reintroduction of fire into park ecosystems is another area in which trade-offs may be required when conflicting goals and competing environmental values are integrated into management plans and activities.

Knowledge of root disease is currently being integrated into park service planning. Following severe flooding of the Merced River in January 1997, many camping and lodging facilities in the valley were damaged or destroyed. As planning progressed for the relocation and construction of new campgrounds, motel units, cabins and dormitories, It was determined that root diseases should be considered. The current management policy for H. annosum is to avoid initiating additional infection sites. Mitigation activities include the treatment of all conifer stumps with Sporax (a borate chemical which protects stump surfaces from H. annosum spore germination) and the sifting out of potentially H. annosum-infected root chunks from soils excavated during construction. Revegetation planning emphasizes oak regeneration, which will minimize problems associated with annosus root disease. However, such revegetation plans may now strongly influence the development of Armillaria root disease. As trees are cut to clear ground for new buildings, buildup of A. mellea inoculum may strongly impact residual trees or newly planted trees. During the restoration period, oaks should be left undisturbed or, if necessary, totally removed, including the entire stump and root system. Post-construction, long-term monitoring will be essential if we are to assess the efficacy of these and other guidelines.

Given the progress of existing gap enlargement, the initiation of new gaps, the hazardous nature of root-rotted trees, and changing management goals, one might ask if the dense stands of large conifers now existing in the valley are sustainable over the long term. If the disease is allowed to run its course, what will be the long-term consequences? Currently we do not have sufficient information on regeneration of vegetation in disease-caused gaps. If conifers colonize these gaps, we do not know the probabilities that they will be killed by residual H. annosum before reaching maturity. We have an incomplete understanding of the genetic variability of the H. annosum and A. mellea within gaps associated with root disease. Research in the valley is planned to determine the origins, spatial distribution, genetic variability and potential for enlargement of existing root-disease gaps and initiation of new gaps. In the future we need to evaluate the impact of canopy gaps caused by root-disease fungi on the valley ecosystem and determine the implications of root disease on current and proposed management practices.