even chain distance. A 10 basal area factor variable 

 plot was installed at each location. All trees in the 

 plot were examined and classified alive or dead 

 based on the presence or absence of living foliage, 

 the percentage of basal circumference girdled by 

 fire, and insect infestation. The numbers of plots 

 examined were 321 in 1991, 198 in 1992, and 127 

 in 1993. Plot placement in 1992 and 1993 was along 

 the same roads surveyed in 1991, but they became 

 more widely spaced in succeeding years. Low eleva- 

 tions between 2,040 and 2,485 m delineated plot lo- 

 cations; therefore, trees consisted mostly of lodge- 

 pole pine and Douglas-fir. 



Tree observations included species, diameter at 

 breast height (d.b.h.), degree of fire injury, presence 

 of insect attack, and insect species. Tree injury was 

 measured as the percentage of basal circumference 

 in which the cambium was killed. Injuries attributed 

 to fire effects included trees with 100 percent stem 

 girdled by fire, while trees that were not 100 percent 

 girdled but showed signs of infestation by bark beetles 

 or wood borers had injuries attributed to insects. 

 Cambium injury was determined by removing small 

 sections of bark and visually inspecting tissues. Bor- 

 ing frass expelled from the bark was the usual sign 

 of insect infestation. Some bark was removed so that 

 the insects could be identified, and insect observations 

 were made on the lower 2 m of the trees. Insects were 

 classified according to whether they were primary or 

 secondary bark beetles, wood borers, or other insects. 



The year in which a tree died was estimated from 

 foliar characteristics. We estimated year of death only 

 for Douglas-fir and lodgepole pine because other tree 

 species were present in such small numbers that loss 

 by year would not be meaningful. 



For Douglas-fir, trees infested the year of the sur- 

 vey had green foliage and fresh boring frass on the 

 bole of the tree; the previous year's trees had mostly 

 red foliage with some needle loss occurring. Trees 

 infested 2 years previously had some red needles re- 

 maining, but most needles had dried and fallen. Trees 

 infested 3 years previously essentially had no needles 

 remaining, and some of the finer twigs had fallen. 



This method is not accurate in all cases because 

 Bedard (1950) stated that Douglas- fir varies greatly 

 in the elapsed time between Douglas-fir beetle (Den- 

 droctonus pseudotsugae Hopkins) infestation and 

 foliage discoloration. Some trees maintain red foliage 

 for a year or longer after beetles have emerged from 

 them. An occasional tree may exhibit some fading in 

 the fall following infestation, but it may appear mottled 

 and may be confined to a few scattered branches 

 (Furniss 1959). 



In lodgepole pine we used the characteristics pre- 

 sented by Cole and Amman (1969) to estimate how 



long since a tree had been killed by bark beetles: 

 (1) tree killed in current year— foliage green, fresh 

 boring frass, larvae or adults present; (2) tree killed 

 in previous year— foliage bright orange to straw color; 

 (3) tree killed in second year past— foliage dull orange 

 and most retained; (4) tree killed in third year past- 

 foliage dull orange to gray and most lost. This dates 

 trees back to 1988, the year of the fires and the year 

 that there was little infestation (Amman 1991). As 

 with Douglas-fir, some errors could occur. For exam- 

 ple, a tree with little foliage or on a dry site would 

 fade faster than a tree with a large crown or on a 

 moist site. 



An intensive study by Schmid (1976) shows that 

 the problem of fading in Engelmann spruce is more 

 complicated than in Douglas-fir and lodgepole pine. 

 Needles usually turn greenish-yellow and fall approxi- 

 mately a year after spruce beetles infest the tree. 

 However, some trees remain green until the fall of 

 the second year (Massey and Wygant 1954). A par- 

 ticularly complicating factor is that branches of 

 some infested trees develop new growth in the sum- 

 mer following the year of infestation (Schmid 1976). 

 The density of attacking beetles, and hence complete- 

 ness of blue stain fungal inoculation, are considered 

 factors affecting rate of needle drying and dropping 

 in spruce (Schmid 1976). 



Because there was no additional insect-caused mortal- 

 ity tallied in 1993, except for a small amount caused 

 by Douglas-fir beetle, 1993 data will not be included 

 in the figures. Percent basal girdling was observed 

 continuously and placed into classes, then data were 

 subjected to weighted regression analyses, because 

 of differing sample sizes, to show differences. 



Results and Discussion 



Fire injury and bark beetle infestation, not only of 

 fire-injured trees but also of uninjured trees, resulted 

 in considerable change to mortality estimates and to 

 the mosaics of fire-killed and green tree stands follow- 

 ing the 1988 fires. 



Tree Mortality by Cause 



Douglas-fir mortality consisted of 31.7 percent of the 

 1,012 trees examined, with 18.5 percent attributed to 

 delayed effect of fire injury, 12.6 percent attributed to 

 bark beetle and wood borer infestation, and 0.6 percent 

 due to unidentified causes (fig. 2). Mortality in trees 

 that were 100 percent stem girdled by fire was attrib- 

 uted to fire effects. Those that were not 100 percent 

 girdled, but became infested by bark beetles or wood 

 borers, had mortality that was attributed to insects. 



3 



