Comparing automatically generated and manually measured tree-ring transects of growth trends with Hawaiian sandalwood as an example species

Tree-ring measurements are a primary quantitative tool used in numerous scientific disciplines. Some species, however, exhibit morphological complexities leading to significant uncertainty in these measurements. Hawaiian Sandalwood (Santalum paniculatum) stems, for example, often develop asymmetric growth features that hinder tree-ring measurements. These features include faint-ring boundaries and wedging rings which disappear in portions of the cross-section. In this work we a use a novel two-dimensional transect methodology and our own open-source software, svg-dendro, to analyze particularly difficult cross-sections. Our method accomplishes this by first tracing all rings by hand and automatically generating a user-specified number of transects. On average, these traced measurements had more sensitivity to tree-ring variability without losing important equivalencies with the traditional binocular stereomicroscope technique (e.g., radii, skewness) as indicated by greater mean variance for ring number, mean tree-ring width, and standard deviation. All S. paniculatum samples had ring wedging, where certain sides of the stem had many locally absent tree rings but to different intensities. The new technique allows us to analyze the shift from complete rings with little to no wedging to rings with more wedging starting between the 19th and 40th ring, where deep stem lobes begin forming. The new method also reveals the difficulty in measuring these trees, as the wedging creates multiple lobes with different visible ring counts. This research suggests that this two-dimensional methodology would be best applied to non-circular trees with fewer incomplete rings, supporting the importance of species and population selection. Overall, we have developed an efficient and flexible means to measure otherwise unmeasurable growth features in tree samples through representing tree-ring boundaries as curves and developing software to sort and map transects.

Automated summary

Tree-ring measurements are crucial in many scientific disciplines, but some species like Hawaiian Sandalwood can exhibit complex growth features that make these measurements challenging. A new two-dimensional transect methodology and software called svg-dendro have been developed to analyze difficult cross-sections of Sandalwood stems. This new method has shown improved sensitivity to tree-ring variability and highlighted the challenges of measuring trees with wedging features.