Climate Change 2001:
Working Group II: Impacts, Adaptation and Vulnerability
Other reports in this collection Pressures from diseases and insect herbivory

Insect outbreaks can be extremely important disturbance factors (Hall and Moody, 1994); during outbreaks, trees often are killed over vast areas (Hardy et al., 1986; Candau et al., 1998). Under climate change, damage patterns caused by insects may change considerably, particularly those of insects whose temporal and spatial distributions strongly depend on climatic factors. The ecological interactions are complex, however (see Box 5-10).

Box 5-10. Complex Interactions: North America's Southern Boreal Forests, Pests, Birds, and Climate Change

The eastern spruce budworm (Choristoneura fumiferana) is estimated to defoliate approximately 2.3 Mha in the United States (Haack and Byler, 1993) and affects 51 million m3 of timber in Canada annually (Fleming and Volney, 1995). Although the budworm usually is present at low densities, budworm densities can reach 22 million larvae ha-1 during periodic outbreaks (Crawford and Jennings, 1989). Outbreaks can extend over 72 Mha and last for 5-15 years, killing most trees in mature stands of balsam fir (Crawford and Jennings, 1989; Fleming and Volney, 1995).

Weather is thought to play a role in determining the budworm's range. Outbreaks frequently follow droughts (Mattson and Haack, 1987) or start after hot, dry summers (Fleming and Volney, 1995). Drought stresses host trees and changes plant microhabitats (Mattson and Haack, 1987). Moreover, the number of spruce budworm eggs laid at 25°C is 50% greater than the number laid at 15°C (Jardine, 1994). In some areas, drought and higher temperatures also shift the timing of reproduction in budworms so that they may no longer be affected by some of their natural parasitoid predators (Mattson and Haack, 1987). Weather, at least in central Canada, also may play a role in stopping some outbreaks if late spring frosts kill the tree's new growth on which the larvae feed.

Control of some populations of eastern spruce budworm may be strongly aided by bird predators, especially some of the wood warblers (Crawford and Jennings, 1989; but see Royama, 1984). Birds can consume as much as 84% of budworm larvae and pupae when budworm populations are low (approximately 100,000 ha-1), but once larvae populations exceed 1,000,000 ha-1, bird predation is unable to substantially effect budworm populations. This predatory action of birds works in concert with those of other predators, mostly insects.

The spruce budworm's northern range may shift northward with increasing temperatures—which, if accompanied by increased drought frequency, could lead to outbreaks of increasing frequency and severity that lead to dramatic ecological changes (Fleming and Volney, 1995). Increasing temperatures also might reduce the frequency of late spring frosts in southern boreal forests, perhaps increasing outbreak duration in some of those areas.

A changing climate also might decouple some budworm populations from those of their parasitoid and avian predators (Mattson and Haack, 1987; Price, 2000). Distributions of many of the warblers that feed on spruce budworms could shift poleward, perhaps becoming extirpated from latitudes below 50°N (Price, 2000). Replacing biological control mechanisms with chemical control mechanisms (e.g., pesticides) ultimately may yield a different set of problems; there are economic and social issues relating to large-scale pesticide application (see Section

In temperate and tropical regions, insect and disease outbreaks are reported mostly for plantation forests; relatively less is known about native forests (FAO, 1997a). In the boreal zone, insect-induced mortality was a significant part of the changing disturbance regime for Canada in the period 1920-1995 (Kurz et al., 1995). Insect mortality accounted for the loss of approximately 76 Mha in that period, with a near tripling of the average annual rate after 1970 (Kurz and Apps, 1999). Similar trends have been observed for Russian forests, where recent annual insect damage and disease mortality affecting as much as 4 Mha was reported by Shvidenko et al. (1995). In Siberian and Canadian forests, insect damage is estimated to be of the same magnitude as fire loss (Krankina et al., 1994; Fleming and Volney, 1995; Kurz et al., 1995; Shvidenko et al., 1995; Shvidenko, 2000). Changes in drought conditions appear to play an important role in insect outbreaks (Volney, 1988; Sheingauz, 1989; Isaev, 1997).

There is likely to be an increase in declines and dieback syndromes (Manion, 1991) caused by changes in disease patterns, involving a variety of diseases. For example, in temperate and boreal regions, there may be increased incidence of canker diseases in poplars and other tree species. Some canker diseases increase in severity with decreased bark moisture content brought on by drought (Bloomberg, 1962). As another example, Armillaria root disease is found throughout the world and causes significant damage on all forested continents (Kile et al., 1991), through mortality and growth loss. This disease—one of the largest threats to regeneration in the productive forest of the Pacific Northwest of North America—has surfaced as a result of past management practices (Filip, 1977) but may be exacerbated by changing climate. Under present conditions, Armillaria root disease causes losses of 2-3 million m3 yr-1 in the forests of Canada's Pacific Northwest (Morrison and Mallett, 1996). More recently, Mallet and Volney (1999) report a 43% decrease in annual volume increment and a 23% loss in annual height increment in lodgepole pine caused by Armillaria root disease. The incidence of Armillaria root disease can be expected to change under warmer or drier conditions. Significant damage has been observed in forests that have undergone drought stress (Wargo and Harrington, 1991). Morevoer, in regions such as the Pacific Northwest where the mean annual temperature presently is below the optimum (25°C) for Armillaria growth, a warmer climate is likely to result in increased root disease and rate of spread.

Other reports in this collection