MADISON 

 Y— 0,2S2f0.277«>X ABLA/VAGL 

 V— 0.307*0, 30e-X ABLA/VASC 



Y— 0.22D*0.26-4-X ABLA/VASC-CARU 

 Y--0. 265*0. 2e6")( ABLA/VASC-VASC 

 Y— 0.248*0. 2E8-X ABLA/ALSl 



MADISON 

 Y— 0. 5a4-»0 , 638HX ABLA/VAGL 

 Y—0. 733*0- 737-X ABLA/VASC 



Y— 0.Sie*0 04O»X ABLA/VASC-CARU 

 Y— 0.e44*e,68H>X ABLA/VASC-VASC 



Y— 0.668*0. eee^x abla/alsi 



M<iDISON 

 Y— 0. ieS*0.209*>X abla/caru 

 Y— 0.277*0. 277-X ABLA/CaGE 



Y— 0, 1GE+0. IBB-X ABLA/CAGE-PSME 

 Y— 0.348*0. 3S4I.X ABLA/PIAL-VASC 

 Y— 0. IB3*0, 163».X PICO/CARU 



MADISON 

 Y—0 . 432+0 . 517*X ABLA/CARU 

 Y=-0. S33+0 . 533».X ABLA/CAGE 



Y"-0,5S8+0.55e<.X ABLA/CAGE-PSME 

 Y—B . 606+0 . 637-X ABLA/PIAL-VASC 

 Y— 0.E77+0.e43«X PICO/CARU 



Figure 6 — (con.) 



Stand data were then subjected to analysis of variance 

 and analysis of covariance for completely randomized 

 design and graphed to show lodgepole pine mortality by 

 habitat type over time (fig. 6). Analysis shows that the 

 percentage of lodgepole pine killed and volume loss vary 

 by habitat type. 



In some habitat types, tree mortality increased rapidly 

 and most susceptible trees and all volume are killed in a 

 relatively short time (fig. 6; ABLA/VASC-VASC, 

 ABLA/ALSI). In others, mortality may occur over a 

 10-year period and never exceed 30 percent of the stand 

 (fig. 6, ABLA/CARU, ABLA/LIBO-LIBO). All suscepti- 

 ble trees may be killed in other habitat types, but it 

 may require 8 to 10 years. These data provide guidance 

 as to which stands within those classed as high hazard 

 should receive priority management. For example, 

 management may be postponed until the next decade if 

 stand mortality does not exceed 20 to 30 percent over a 

 10-year period. Meanwhile, stands can be rated and 

 management implemented in the stands containing 

 habitat types where considerable tree mortality or 

 volume loss is predicted to occur over a short time. By 

 putting the higher risk stands under management, loss 

 would probably be prevented in some high-, many 

 moderate-, and many low-risk stands. 



INTEGRATION WITH FORPLAN 



The Forest Service currently uses FORPLAN, a linear 

 programing model (Johnson and others 1980), for land 

 management planning which is the basis for land use 

 allocations and scheduling of management activities. The 

 management activities and associated products, costs, 

 and environmental effects used in FORPLAN are 

 reflected in prescriptions for stands within analysis 

 areas. In the Forest Service Northern Region, analysis 

 areas £ire lamds that meet certain common classification 

 criteria; these lands are not usually contiguous. 

 Classification criteria include habitat type, timber size 

 class, slope class, and other characteristics. Prescriptions 

 describe specific mamagement practices used to manage 

 specific stands. 



One approach to modeling tree mortality caused by 

 the mountain pine beetle using FORPLAN has been to 

 predict susceptible areas within analysis areas and prob- 

 able mortality over two decades. Although it might be 

 possible to predict rate of loss caused by the beetles 

 throughout a forest, this information would be of little 

 value for adjusting yield tables if the locations of high-, 

 moderate-, and low-risk stands aire not identified within 

 analysis areas. The FORPLAN model would spread bark 



19 



