Km. Reynolds et Eh. Holsten, CLASSIFICATION OF SPRUCE BEETLE HAZARD IN LUTZ AND SITKA SPRUCE STANDS ON THE KENAI PENINSULA, ALASKA, Forest ecology and management, 84(1-3), 1996, pp. 251-262
Stand data from Lutz and Sitka spruce forest types occurring on the Ke
nai Peninsula were analyzed by tree-based classification and abductive
inference to develop decision models for classifying spruce beetle ha
zard. Model development and validation data sets contained 286 and 88
stand observations, respectively. The final decision-tree structure yi
elded 13 possible hazard outcomes based on total stand basal area, per
centage of total basal area composed of spruce, percentage of spruce b
asal area composed of trees with diameter > 25 cm, stand elevation, an
d stand aspect, Four paths in the decision tree led to low-hazard outc
omes (spruce basal area loss less than or equal to 10%); two paths eac
h led to low-medium, medium, and medium-high hazard outcomes (spruce b
asal area loss of less than or equal to 40%, 11-40%, and > 10%, respec
tively); and three led to high-hazard outcomes (spruce basal area loss
> 40%). Results of model verification were considered very acceptable
; in the worst case, predictions of high hazard were correct for 67% o
f the observations. Model validation results also were considered acce
ptable, but predictions of medium and high hazard showed a marked drop
from verification results. For comparison, the same analysis was perf
ormed using abductive inference to test a modeling method better suite
d to automatic processing of numerous stands for landscape-level analy
sis, There was a high degree of correspondence between predictions of
the two analytical methods.