of the losses recorded through 1976 accumu- 

 lated from 1972 through 1974. 



Selected plot variables provide a general 

 description of the infested stands in 1971: 





Mean 



Range 



arpfl dpi* a orp (ft 2 ) 



95 



35-180 



Percent of basal area in oak 



56 



0-100 



Average dbh (inch) 



7.0 



4.7-10.8 



Stand age (yr) 



68 



25-105 



Site index (ft, upland oaks) 



59 



30-80 



Elevation (ft) 



1190 



620-1560 



Standing timber value per 







acre ($) 



132 



20-840 



Conversion standards developed by Men- 

 del and his coworkers (1976) were used to 

 estimate the value of timber lost on each plot. 

 These value standards incorporate species, 

 dbh, butt log grade, and merchantable height 

 for each tree. 



RESULTS 



Forest stand losses were summarized by 

 number of trees, basal area, timber volume, 

 and value from 1972 through 1976 (Figs, la- 

 4b). It is immediately obvious that losses, 

 however expressed, were not uniformly dis- 

 tributed among the plots. A small percentage 

 of the infested stands incurred major losses 

 while a large percentage of the stands in- 

 curred minor losses (that is, the distribution 

 is skewed). For example, volume losses over 

 the 5-year period averaged 2.5 cords per acre 

 (Fig. 3a). But 75 percent of the stands lost 

 less than 2.5 cords per acre, 20 percent of the 

 stands incurred no losses at all, and only 5 

 percent of the stands lost more than 10 cords 

 per acre. Volume losses averaged 12 percent 

 of total original stand volume (Fig. 3b). But 

 65 percent of the stands lost less than one- 

 tenth of their volume and only 5 percent of 

 the stands lost more than one-half of their 

 volume. 



The frequency distributions of stand losses 

 are highly skewed. In a skewed distribution, 

 the median (the value that divides the range 

 of values into two equal parts) helps describe 

 the "typical situation." Medians for the 5- 

 year stand losses are: 



30 trees per acre ( 10 percent of the original 

 stand of trees) 



6 ft 2 of basal area per acre ( 7 percent of the 



original basal area) 

 1 cord per acre (4 percent of the original 



volume) 



$4 per acre (4 percent of the original value) 



DISCUSSION 



Knowing how much and what kind of dam- 

 age to expect from gypsy moth outbreaks is 

 a must for planners of cost-effective control 

 programs. One 5-year case study of forest 

 stand losses in the Poconos will not provide 

 reliable and specific answers to all the ques- 

 tions about impacts of the insect. But it does 

 give us fresh perspective on what, in general, 

 to expect. 



Only a small percentage of infested stands 

 suffered major losses. This finding holds im- 

 portant implications for people whose job is 

 to make cost-effective decisions about con- 

 trol. Suppose, for example, that effective 

 gypsy moth control in infested stands re- 

 quired three successive annual treatments, 

 each costing $10 per acre — that is, a total out- 

 lay of $30 per acre. Our analysis indicates 

 that without treatment, 90 percent of the in- 

 fested stands would suffer timber value losses 

 of less than $30 per acre. So, from the stand- 

 point of timber value saved, the cost of treat- 

 ment would be justified on only 10 percent 

 of the infested stands (Fig. 5). 



Of course, decisions on gypsy moth control 

 are based on more than the value of timber 

 losses. Moreover, an operational decisionmak- 

 ing model must be able to help us determine 

 not only how many forest stands to protect, 

 but also which ones. And it must be able to 

 provide this information before an insect at- 

 tack. From the Pocono data, techniques have 

 been developed for predicting forest stand 

 losses attributed to the gypsy moth (Gansner 

 et al. 1978; Herrick et al. 1979). They can 

 be used to estimate losses from easy-to- 

 measure key characteristics of stand condi- 

 tion. We are working with the Northeastern 

 Area State and Private Forestry's Forest In- 

 sect and Disease Management staff to install 

 a system of field plots in Pennsylvania and 

 adjacent states in advance of gypsy moth out- 

 breaks. These plots will be used to monitor 



4 



