A Variable Area Plot Method of Assessment of Forest Condition and Trend

The properties of a variable area sampling technique, by which the observer varies the searchradius to obtain approximately a prescribed number of woody plants in each tier measured, were determined by (i) comparison of fixed area and variable area sampling of a computer-mapped shrub population; (ii) comparison of results from fixed area and variable area sampling of woody plants exceeding about 2m height in a rata-kamahi forest; (iii) two variable area surveys of woody plants exceeding 30cm height in a beech forest. Variable area sampling gave unbiased estimates of crown area and plant density in the computer-mapped population. These were as precise as estimates obtained by fixed area sampling when sampling intensities were equal. For the same number of plots, 20 m x 20 m fixed area sampling of density and basal area in rata-kamahi forest was more precise than variable area sampling in plots containing 30 stems per plot, but in terms of sampling intensity and time, variable area sampling was more efficient. Both sampling methods gave similar stem diameter frequency patterns. Two surveys of beech forest, conducted 11 years apart, showed that with about 70-80 single-tier plots, differences in basal area of about 15% can be detected using the variable area plot method. Variable area sampling is robust and suitable for ecological surveys of New Zealand's indigenous forests.