United States 

 Wg^) Department of 

 Agriculture 



Porest Service 



Northeastern Forest 

 Experiment Station 



Researcli 

 Note NE-311 



95 ^(^^^7 



Predicting the Rate of Change in 

 Timber Value for Forest Stands 

 Infested With Gypsy Moth 



David A. Gansner 

 Owen W. Herrick 



Abstract 



Presents a method for estimating the potential impact of gypsy moth 

 attacks on forest-stand value. Robust regression analysis is used to 

 develop an equation for predicting the rate of change in timber value from 

 easy-to-measure key characteristics of stand condition. 

 ODC 453—145.7 x 18.77 Lymantria dispar L. :624.4 



Recent studies in Pennsylvania 

 show that while most forest stands 

 come through a gypsy moth out- 

 break with tittle or no loss in timber 

 value, a few suffer heavy damage. 

 Knowing how much value^^loss to 

 expect for a stand and which stands 

 will suffer most would be'advan- 

 tageous to planners of cost- 

 effective control programs. In an ef- 

 fort to solve that problem, we have 

 developed a model for estimating 

 the potential hazard of impending 

 gypsy moth attacks— a simple equa- 

 tion that uses easy-to-measure key 

 characteristics of stand condition to 

 predict the rate of change in timber 

 value for stands that go through an 

 outbreak. 



Collecting the Data 



Changes in timber value over 

 an 8-year period were measured on 

 131 plots in Pike and Monroe 

 Counties, Pennsylvania. This area 

 was on the frontier of gypsy moth 

 infestations in the early 1970's. The 

 1/10-acre plots were established in 

 1971 at random in newly infested 

 woodlands. Severity and frequency 

 of gypsy moth attacks varied from 

 plot to plot. In general, the study 

 area had moderate to heavy defolia- 



tion from 1971 through 1973. Insect 

 populations collapsed in 1974 and 

 1975, built up again to some extent 

 in 1976 through 1978, and virtually 

 disappeared in 1979. 



In the fall of 1979 the plots 

 were revisited to record changes in 

 timber value occurring since 1971. 

 We began the study with more than 

 150 plots. But those that were cut 

 for timber and development or 

 sprayed to control gypsy moth 

 during the study period were 

 dropped from the analysis. So, 131 

 relatively undisturbed plots 

 remained for measuring the effect 

 of gypsy moth on stand value. 



Conversion standards devel- 

 oped by Mendel and his co-workers 

 (1976) were used to estimate timber 

 value. These value standards ac- 

 count for average regional 

 conversion costs and incorporate 

 species, dbh, butt-log grade, and 

 merchantable height for each tree. 

 Because our intent was to concen- 

 trate on impacts of the pest, we 

 held the value standards constant 

 over the study period. We thus 

 avoided effects of inflation and 

 changes in timber prices that oc- 

 curred during the period. 



\ " - 



..J 



'-^C5 CO 



^ V3 



-< 



Fitting the Model 



Robust regression, a weighted 

 regression technique, was used to 

 construct the model. Robust re- 

 gression minimizes the influence of 

 data points with large residual 

 errors, that is, large differences be- 

 ^tween predicted and actual values 

 .^' Of the dependent variable. The tech- 

 nique sacrifices the fit of suspect or 

 "outlier" data for a better fit of the 

 remaining more important infor- 

 mation. Robust regression usually 

 results in equations that are better 

 predictors (see Andrews, 1974). In 

 this application, the weight used 

 was 1.5 times the median of the ab- 

 solute values of the residual errors. 



The Dependent Variable 



The rate of change in timber 

 value expressed as a compound in- 

 terest percent was chosen for anal- 

 ysis as the dependent or predicted 

 variable. Timber value incorporates 

 a number of important attributes of 

 stand condition such as species 

 composition, tree size, and timber 

 quality. Compound rates of change 

 account for both the amount of 

 change and the time involved in the 

 change. So compound rates of 

 change in timber value provide use- 

 ful information for assessing the im- 

 pact of gypsy moth on forest stands 

 and for gauging the relative worth of 

 control projects. For the 131 plots 

 included in this study, changes in 

 timber value ranged from - $375 to 

 -I- $370 and averaged -i- $13 per acre 

 (+^2 percent) over the 8-year study. 

 When expressed as compound rates 

 of interest, changes in value ranged 

 from - 6.8 to -i- 9.1 and averaged 

 + 1.3 percent. 



1 



