-40- 



It should be noted that in a variable-radius cruise, the percent cruise may 

 be different for each species. See tables 2 and 3. 



In forest sampling, we deal with finite populations. The formula we 

 use to estimate the variance of the mean volume per acre for each stratum 

 includes a term known as the finite population correction (FPC) . The FPC 

 reduces the variance of the mean when a significant (usually > 5 percent) 

 portion of the population has been sampled. 



Let N = number of sample units in population 

 n = number of units actually sampled 



N - n 

 Then, FPC = " "^ 



N 



AC 

 In plot sampling N = and n = number of plots sampled. However, 



if Lt L 



in point sampling there is no easy was to calculate N. Therefore, mensura- 

 tionists have traditionally ignored the FPC when analysing point cruises. 

 This can be quite inefficient when tract sizes are small as is often the 

 case in New Hampshire. This problem can be easily overcome using the 

 previous result permitting the calculation of percent cruise when point 

 sampling. It can easily be shown that the FPC is simply a function of P 

 (percent cruise) . 



FPC = N - n ^ _N D_ = 1 _ ^ = 1 P 



N N N N 100 



INVENT uses this relationship to refine the variance estimates when point 



p 

 sampling as well as plot sampling. INVENT calculates FPC as 1 - y^ . 



The standard error due to the variation in volume between clusters, SE^, 



is calculated as follows: 



Y = mean volume /acre of stratum 



2 

 S = sample variance of volume 



n = number of sample units 



1 



then, SE^ = 



(100) (FPC) 



If the height of all volume trees was measured, then SE— is the standard 

 error of the volume estimate for the stratum, SE . If, however, the 



