Bark Beetle Infestation
The probability of beetle infestation was significantly influenced by spatial location, size of infestation in previous year, and size of 1- year-old fire within 1 km (P-value <<0.05). The estimated probabilities of bark beetle outbreaks, evaluated after controlling for all other significant predictors, demonstrated a spatial pattern with increasing probabilities of outbreak as one moves away from the coastal areas and into the eastern regions of Washington and Oregon (Figure 1). This pattern of declining forest health in the dry coniferous forests of eastern Oregon and Washington has been noted by others in recent assessments and syntheses (Gast and others 1991, Hayes and others 2001, Hessburg and others 1994, Jaindl and Quigley 1996).
One of the most important predictors when evaluating a bark beetle outbreak appears to be size of the beetle infestation in the previous year. The odds of a beetle outbreak seem to increase as the number of trees killed per km2 by bark beetles in the previous year increases (Figure 2). Although this outcome is probably expected, given the outbreak dynamics of insect populations in general, (e.g., Barbosa and Schultz 1987), it is important to know that it was included in the model and, consequently, is accounted for in a reasonable fashion. The other significant predictor was the size of fire in the previous year. The odds of an insect outbreak appeared to increase as the size of a fire increases from 0 (no fire) to ~750 ha after which the odds appear to decrease. The standard errors are large, especially for large fire sizes. This result is consistent with the expectation that as fire increases in size, the number of trees with significant crown damage and bole or root scorch increases, and, therefore, susceptibility to bark beetle attack may increase. However, very large fires may completely eliminate susceptible hosts in the area, resulting in less observable beetle damage when large fires are nearby.
Encyclopedia ID: p3560
