To assess the growth of the uneven-aged stands of Douglas-fir in the Interior Douglas-fir zone of British Columbia, and the factors influencing the growth, data from 92 permanent sample plots were obtained. Volumes ranged from 3 to 496 m3/ha, with a mean of 94 m3/ha. Annual growth rates varied from -1.8 to 9.5 m3/ha, with a mean of 3.3 m3/ha. Factors included in the analysis were the following: harvesting method (three classes), biogeoclimatic subzone (two classes), time since harvest, and the stand variables of trees per ha, basal area per ha, and quadratic mean diameter. Statistical analyses and visual examinations of scattergrams were applied to plot data compiled according to two utilization limits: trees with a diameter of 9.1 cm and larger, and trees with a diameter of 17.5 cm and larger.
For the 9.1-cm utilization limit data, results of analyses of covariance indicated significant differences among harvesting methods and biogeoclimatic subzones. The data were accordingly divided into four groups with statistically significant differences. A volume growth equation was next derived for each group using the stand variables as independent variables. For two groups (stands harvested using the diameter limit method, and unharvested stands in the dk biogeoclimatic subzone), the number of trees per ha was statistically significant while for a third group (unharvested stands in the xh biogeoclimatic subzone), the number of trees per ha as well as quadratic mean diameter were significant. Another group (stands harvested using the faller's selection method) had no significant independent variables. Considering this result, and since the data base for the faller's selection method included only nine plots with considerable variability in volume growth, this method was excluded from further analysis.
For the 17.5-cm utilization limit data, only three groups had statistically significant differences. One of these groups, stands harvested using the faller's selection method, was subsequently excluded from further analysis for the reasons stated above. For the remaining two groups (stands harvested using the diameter limit method, and unharvested stands), only the number of trees per ha was statistically significant.
While some trends are apparent, they are not well defined and the accuracy of the growth predictions is low; this result is attributed to deficiencies in the data and to natural variability within stands in the interior Douglas-fir zone.