I 



Equations for future growth estimation 



In this study, as expected, the variable d.b.h. contributed most to future basal- 

 area growth prediction. Simple correlation analysis showed the logarithm of d.b.h. to 

 be highly correlated with the logarithm of subsequent 10-year basal-area growth of 

 corresponding trees. The simple correlation coefficient was 0.74, corresponding to a 

 eouITi c i cjiit of determination in simple linear regression which explained percent of 

 the variation in basal-area growth, 



_ The addition of elevation, CCF, and the ratio of d.b.h. to average-stand-diameter 

 (d/D) , raised to 76.4 percent the amount of variation accounted for (equation 2, 

 table 1). As measures of stand competition effects and relative competitive status of 

 the subject trees, respectively, the variables CCF and d/D make worthwhile contribu- 

 tions. In the presence of elevation, they explain approximately 3.5 percent more 

 variation than did any other combination of three variables added to d.b.h. 



In addition to the measure of relative competitive status afforded by d/D, an 

 indicator of tree vigor (crown ratio) was also tested for its contribution. However, 

 because crown length measurements were not available at the beginning of the growth 

 period, the test was made using all independent variables measured at the end of the 

 growth period. The simple correlation coefficient between the logarithms of periodic 

 basal area growth and crown ratio was 0.36; however, when added to the equation con- 

 taining d.b.h., CCFj d/D, elevation, and site index, crown ratio had virtually no 

 effect--it raised explained variation by only 0.2 percent. 



It appears that crown ratio is not a useful prediction variable for basal-area 

 increment of individual tre£s in uniform, undisturbed stands of lodgepole pine when 

 such measures as CCF and d/D are available. 



4 



