Precision-Driven Classification of Coefficients of Variation in Clonal Eucalyptus Nursery Trials

The coefficient of variation (CV) is widely used to evaluate experimental precision; however, generic classifications may fail to capture the variability in clonal eucalyptus seedlings. This study aimed to propose a reclassification of CV values for morphological traits of Eucalyptus sp. seedlings at the dispatch stage, considering different plot sizes (PS). Data from 8200 seedlings were analyzed across six commercial clones and 41 evaluations of 200 seedlings each. Traits evaluated included plant height, stem diameter, shoot, root, and total dry mass, leaf number, leaf area, and slenderness index. Using bootstrap simulation, CV values were estimated for PS ranging from 1 to 100 seedlings. The relationship between CV and PS was fitted using a power model to estimate updated CV values and reclassify them. CV classification limits decreased progressively as plot size increased from 1 to 10 seedlings. Rather than defining a single optimal plot size, the proposed framework provides dynamic, trait-specific CV benchmarks tailored to the chosen PS. The proposed reclassification was more sensitive and appropriate than traditional criteria. In conclusion, adopting trait-specific CV intervals combined with PS provides more consistent evaluations of experimental quality, contributing to technical standardization in forest nurseries and eucalyptus breeding programs.