|
|
Forests are large and variable and the individual components (ie trees) tend to have a relatively small value. Consequently, samples or sub-sets rather than complete measurement of all individuals are often necessary.
Information derived from the sample can be used to make inferences or estimates about the total forest. Three general objectives in sampling include:
- To obtain an unbiased estimate of the population mean.
- To obtain as precise an estimate of the mean as is possible for the time and money spent.
- To assess the precision of estimate, i.e. standard error of the mean.
Effective forest sampling requires appropriate decisions about:
- The sample element. What is the individual in the sample - a single tree, a stand of trees, a plot, a stripline, a unit area of ground, or a point?
- The sample selection. Will elements be selected subjectively or objectively? If objectively, will each sample have the same probability of selection or will its probility of selestion change in a predetermined manner?
- The sample design. Will the sample elements be selected and measured in a single step, or will the sampling procedure be a series of steps and phases that depend on initial steps and assumptions? Can you make realistic assumptions about the structure of the forest (eg about the patterns of variation) and use this information to allocate the sample selection?
Together, these three aspects define a sampling frame.
|
