176
Lloyd Goldwasser, Scott Ferson, and Lev Ginzburg
effects in the population dynamics of the Northern Spotted Owl is to compare
model results by using different threshold levels.
Windstorms
In the recent past, windstorms have destroyed large areas of potential owl habitat
on the Olympic Peninsula. On a practical time scale, such destruction represents
an effectively permanent loss of owl habitat relative to the slow rate of regrowth of
the forest. To estimate the importance of these catastrophic windstorms, we simulated the population dynamics of the owl both with and without them.
According to Agee (1994; personal communication), approximatey three major
blowdowns occur per century on average in the Olympic Peninsula. Based on this
observation, we assumed that in any year there is a probability of 0.033 that a
storm will destroy some territories. Agee (1994) estimated that the 1921 storm,
which would cause moderate or greater damage to 13 of the current owl territories, was of approximately twice the severity of the average major storm, so an
average storm might remove seven or eight territories. However, because some
locations are more vulnerable to windstorms than are others, the effects of separate storms are likely to overlap rather than be distributed evenly over the landscape. For this reason, sites that have been struck by a windstorm are more likely
to be struck again. We modeled this overlap by having each subsequent storm
remove half as many territories than did the previous one. To maintain the mean
effect suggested by Agee (1994; personal communication), in the simulation, the
first storm removed 12 territories, the second 6, the third 3, the forth 1, and
thereafter 0. We treat this destruction of habitat as permanent, because within the
time scale of a century, this habitat does not return to the mature forest that
Northern Spotted Owls prefer (Gutierrez 1996). Once a storm has occurred in any
year over the course of a simulation, the affected territories remain unavailable as
owl habitat, and the number of territories is decremented accordingly.
Windstorms constitute only one class of catastrophic events that might befall
the Spotted Owl population. Historically, for instance, both storms and fires have
destroyed large areas of potential owl habitat in the Olympic Peninsula. We
neglected the impact of the latter on the persistence of the owl population for two
reasons. First, changes in silvicultural practices make it unreasonable to use
historical records to try to estimate the likelihood of occurrence of large fires.
Second, modern methods of fire control may now be relatively effective in extinguishing fires after they begin. However, potential biotic catastrophes are harder
to circumscribe by such reasoning. Disease outbreaks, for example, could directly
affect the owl population or could indirectly affect owls by reducing the availability of their prey.
Of course, it is difficult to predict what biotic interactions will be important in
the future, especially over the long term. For the purposes of this study, the
analysis of catastrophic events was confined to windstorms.
Lloyd Goldwasser, Scott Ferson, and Lev Ginzburg
effects in the population dynamics of the Northern Spotted Owl is to compare
model results by using different threshold levels.
Windstorms
In the recent past, windstorms have destroyed large areas of potential owl habitat
on the Olympic Peninsula. On a practical time scale, such destruction represents
an effectively permanent loss of owl habitat relative to the slow rate of regrowth of
the forest. To estimate the importance of these catastrophic windstorms, we simulated the population dynamics of the owl both with and without them.
According to Agee (1994; personal communication), approximatey three major
blowdowns occur per century on average in the Olympic Peninsula. Based on this
observation, we assumed that in any year there is a probability of 0.033 that a
storm will destroy some territories. Agee (1994) estimated that the 1921 storm,
which would cause moderate or greater damage to 13 of the current owl territories, was of approximately twice the severity of the average major storm, so an
average storm might remove seven or eight territories. However, because some
locations are more vulnerable to windstorms than are others, the effects of separate storms are likely to overlap rather than be distributed evenly over the landscape. For this reason, sites that have been struck by a windstorm are more likely
to be struck again. We modeled this overlap by having each subsequent storm
remove half as many territories than did the previous one. To maintain the mean
effect suggested by Agee (1994; personal communication), in the simulation, the
first storm removed 12 territories, the second 6, the third 3, the forth 1, and
thereafter 0. We treat this destruction of habitat as permanent, because within the
time scale of a century, this habitat does not return to the mature forest that
Northern Spotted Owls prefer (Gutierrez 1996). Once a storm has occurred in any
year over the course of a simulation, the affected territories remain unavailable as
owl habitat, and the number of territories is decremented accordingly.
Windstorms constitute only one class of catastrophic events that might befall
the Spotted Owl population. Historically, for instance, both storms and fires have
destroyed large areas of potential owl habitat in the Olympic Peninsula. We
neglected the impact of the latter on the persistence of the owl population for two
reasons. First, changes in silvicultural practices make it unreasonable to use
historical records to try to estimate the likelihood of occurrence of large fires.
Second, modern methods of fire control may now be relatively effective in extinguishing fires after they begin. However, potential biotic catastrophes are harder
to circumscribe by such reasoning. Disease outbreaks, for example, could directly
affect the owl population or could indirectly affect owls by reducing the availability of their prey.
Of course, it is difficult to predict what biotic interactions will be important in
the future, especially over the long term. For the purposes of this study, the
analysis of catastrophic events was confined to windstorms.