For high q values, peak mortality tended to be 

 overestimated. By the third year, q usually becomes 

 smcdl due to the "high-grading" action of the beetle in 

 thinning a stand, resulting in greater overestimation of 

 tree mortality. However, the critical time during an in- 

 festation by the mountain pine beetle is at the point of 

 change from endemic to epidemic. The value q applied to 

 the larger diameter trees forecasts the pending infesta- 

 tion adequately in spite of the tendency toweird 

 overestimation. 



The model assumes optimum conditions for the life of 

 the epidemic. However, actual field conditions can cause 

 beetle populations to deviate from predictions. 

 Overestimation of tree mortality is not considered 

 serious in most cases, particularly in the larger diameter 

 classes. Epidemics usually begin in larger diameter trees 

 preferred by the mountain pine beetle, and the rate of 

 tree loss within these classes is critical. Thus, any factor 

 that affects brood survival (such as thick phloem [food 

 supply], which is correlated with larger diameters) will 

 affect the rate of tree loss and, in turn, successive 

 generations. 



Dispersion of the beetle is also affected by stand 

 characteristics such as species, age, stocking levels, 

 growth rates, and diameter class distribution; and by 

 site characteristics, including habitat type, soils, eleva- 

 tion, slope, and aspect. During the past decade the 

 system of environmental classification by habitat type 



developed by Daubenmire for the Northern Rocky Moun- 

 tain Forest Ecosystem has gained increasing acceptance 

 in other areas of the West. This concept stresses use of 

 the entire cUmax plant community as an environment£il 

 indicator that permits identification of environments 

 (habitats) with similar biotic potentials. All en- 

 vironments (habitats) with the potential to support ap- 

 proximately the same mix of stable (climax) plant 

 species are considered to be within the same habitat 

 type regardless of successional status of the vegetation. 



Recent data from the Forest Service Northern Region 

 show that the extent of lodgepole pine mortality caused 

 by the mountain pine beetle varies by habitat type 

 group, and by habitat type within groups. This type of 

 classification provides the framework essential for 

 organizing information to select alternative management 

 activities for habitat types. 



MODEL TESTS AND REFINEMENT 



Data from a mountain pine beetle infestation in the 

 Bechler River Drainage of Yellowstone National Peirk 

 (Klein and others 1978) were used to predict tree loss by 

 2-inch (5.1-cm) diameter classes (situation A, table 2; fig. 

 1). Trees were also grouped by 6- to 12-inch (15.2- to 

 30.5-cm), 14- to 16-inch (35.6- to 40.6-cm), greater than 

 16-inch (40.6-cm) diameters, and total stand (table 3; fig. 

 2 and 3). 



3 



