In New Zealand, some sawmills are requesting logs above a stiffness grade, prompting forest managers to integrate stiffness-related data into their resource inventories. The IML PD400 resistograph provides a rapid and non-destructive means of sampling trees for density with studies demonstrating a high correlation between site-average estimates of basic density and site-average stiffness of board outturn at a sawmill.

OneFortyOne New Zealand has purchased a resistograph with the intention of using the estimates of basic density that it provides to segregate their stands for stiffness. This study investigated the sampling intensity necessary to achieve a probable limit of error (PLE) of 10-15 % for stand-level basic density estimates to help OneFortyOne operationalise the tool. High-intensity sampling was carried out across 15 stands that covered a range of environmental conditions. Simulations of the PLE equation were run in R with the sampling intensity systematically reduced to assess the influence on PLE.

Results suggested that sampling programs for stand-level estimates of basic density can be carried out at a very low sampling intensity. With only 10 total measurements across two sample plots, a PLE of less than 12.5% was achieved across the range of stands assessed. Increasing the sampling intensity to 30 total measurements across 15 plots returned a PLE of 2.5-5%. However, beyond this point, further increases to sampling intensity yielded diminishing returns.

Decisions relating to sampling intensity should be an operational call that takes into account the findings of this study alongside manager experience and knowledge of wood variability across a forest estate. Further research should be conducted to confirm the relationship between site-average estimates of basic density and site-average stiffness of board outturn. The IML PD400 and the processing software is a rapidly evolving space that will likely continue to be adopted as wood product customers demand higher quality logs.