Beyond planting scheme: tree diameter and micro-drilling resistance as predictors of wood density in a Norway spruce plantation

Initial planting scheme can influence long-term stand development and individual-tree growth, potentially affecting wood properties in Norway spruce (Picea abies (L.) H. Karst.) plantation. However, the extent to which wood-density variation reflects initial stand density rather than individual-tree development remains unclear, as does the additional predictive value of micro-drilling resistance beyond conventional tree measurements. This study evaluated wood density and micro-drilling resistance in a 45-year-old Norway spruce plantation established at four initial planting scheme (2,500, 3,300, 5,000, and 7,510 trees ha⁻¹), arranged in four blocks. A total of 313 trees were analysed. Linear mixed-effects models were used to assess the effects of planting scheme and diameter at breast height (DBH) on wood density and micro-drilling resistance while accounting for the experimental structure. Initial planting scheme had no significant effect on wood density (p = 0.332) or micro-drilling resistance (p = 0.383), and remained non-significant after adjustment for DBH (p = 0.752 and p = 0.231, respectively). In contrast, DBH was significantly associated with both properties (p < 0.001), with wood density decreasing as DBH increased. Predictive models based on DBH and micro-drilling resistance were evaluated using leave-one-plot-out cross-validation across the 16 experimental plots. The model combining micro-drilling resistance and DBH showed the best predictive performance (RMSE = 0.0187 g cm⁻³, MAE = 0.0147 g cm⁻³, predictive R² = 0.387), reducing RMSE by 8.8% relative to the DBH-only model. The combined model also yielded lower RMSE values in 14 of the 16 plots (paired Wilcoxon signed-rank test, p = 0.008). Adding planting-density treatment did not further improve prediction. Overall, the results indicate that individual-tree development is more strongly associated with wood-property variation than initial planting density at the stand age investigated. Combining micro-drilling resistance with DBH provides a useful minimally invasive approach for improving wood-density estimation in standing Norway spruce, although its predictive accuracy at the individual-tree level remains moderate.