Comparison of durability of treated wood using stake tests and survival analysis

Abstract

The stake test is widely used to evaluate the efficacy of wood preservatives. This test monitors the deterioration level observed in treated stakes partially inserted into the ground. The results are conventionally expressed as the relationship between deterioration levels and exposure periods. The preservative efficacy is compared among the stake groups treated with different retention levels based on the test results; however, there is no scientific basis for the comparison. We applied survival analysis to the conventional stake test to include a scientific basis to the test. Stakes impregnated with different types and retention levels of preservatives were subjected to deterioration at two test sites for approximately 30 years. The deterioration levels were monitored according to the conventional procedure and survival analysis was applied to the monitored data. Kaplan–Meier plots of the survival probabilities against the exposure periods indicated that there is a significant difference between the durability of the stakes treated with alkylammonium chloride (AAC-1) at K2 and K3 retention levels, whereas no significant difference was observed between those at K3 and K4 retention levels. Contrastingly, emulsified copper naphthenate (NCU-E) was found to be a reliable preservative, and the stakes impregnated with NCU-E showed a significant increase in durability in accordance with preservative retention. Alkaline copper quaternary (ACQ-1) also appeared to be a reliable preservative; however, the increase in stake durability after ACQ-1 treatment differed between the test sites. These results were verified using the modified Gehan–Breslow–Wilcoxon test with Holm’s p adjusting method.