Ht. Schreuder et Ms. Williams, DESIGN-BASED ESTIMATION OF FOREST VOLUME WITHIN A MODEL-BASED SAMPLE SELECTION FRAMEWORK, Canadian journal of forest research, 25(1), 1995, pp. 121-127
Equations for predicting tree volume are often developed using data co
llected either by a model-based method such as purposive sampling or b
y stratified random sampling so that an ''adequate'' number of trees f
rom each diameter class are sampled across the range of classes expect
ed in populations of interest. Such equations are then used together w
ith a design-based (probabilistic) sample such as variable radius plot
sampling from a specific population to generate estimates of total vo
lume. The probabilities of selection of the sample trees used in devel
oping the volume equation are ignored, may not be known, or may not be
appropriate for populations to which the equation are applied. Less b
iased and more efficient estimates of the population volume can be gen
erated by using known frequencies or estimated frequencies of the diam
eter classes in the population from the probabilistic sample used for
estimating total volume in the population. These frequencies are used
as weighting factors in the construction of population-specific volume
equations. We show a reduction in bias and increased efficiency in a
simulation study for several forest populations with strong linear rel
ationships between variables and reasonably well known error structure
. A model-based sampling procedure called pscx sampling or a large-sam
ple extension thereof is used to select sample trees for volume equati
ons. Such bias reduction did not happen for other populations with wea
k linear relationships and unknown error structure.