Summary: Wood anatomy plays a key role in plants' ability to persist under drought and should therefore predict demography. Plants balance their resource allocation among wood cell types responsible for different functions. However, it remains unclear how these anatomical trade‐offs vary with water availability, and the extent to which they influence demographic rates. We investigated how wood anatomical trade‐offs were related to drought and demographic rates, for seedling communities in four tropical dry forests differing in their aridity indexes (AIs). We measured wood density, as well as vessel, fiber and parenchyma traits of 65 species, and we monitored growth and survival for a 1‐yr period. Two axes defined wood anatomical structure: a fiber‐parenchyma axis and a vessel‐wood density axis. Seedlings in drier sites had larger fiber but lower parenchyma fractions, while in less dry forests, seedlings had the opposite allocation pattern. The fiber–parenchyma trade‐off was unrelated to growth but was positively related to survival, and this later relationship was mediated by the AI. These findings expand our knowledge about the wood anatomical trade‐offs that mediate responses to drought conditions and influence demographic rates, in the seedling layer. This information is needed to anticipate future responses of forests to changing drought conditions.
Field deployable computer vision wood identification systems can play a key role in combating illegal logging in the real world. This work used 764 xylarium specimens from 84 taxa to develop an image data set to train a classifier to identify 14 commercial Colombian timbers. We imaged specimens from various xylaria outside Colombia, trained and evaluated an initial identification model, then collected additional images from a Colombian xylarium (BOFw), and incorporated those images to refine and produce a final model. The specimen classification accuracy of this final model was ~ 97%, demonstrating that including local specimens can augment the accuracy and reliability of the XyloTron system. Our study demonstrates the first deployable computer vision model for wood identification in Colombia, developed on a timescale of months rather than years by leveraging international cooperation. We conclude that field testing and advanced forensic and machine learning training are the next logical steps.
Summary: Trees experience changes in environmental conditions and biophysical constraints as they grow that may lead to shifts in functional traits. While ontogenetic shifts in leaf traits are relatively well understood, changes in wood anatomy traits from seedlings to adults are less clear, especially in tropical dry forests where drought strongly influences adult wood traits. We collected wood and leaf functional trait data for seedlings and adult trees in four dry forest sites along a rainfall gradient in Colombia. Using a Bayesian framework, we quantified intraspecific trait shifts between seedlings and adults, and we compared functional axes and trait correlations between these stages. Leaf traits shifted along a clear functional axis from more acquisitive seedlings to more conservative adults, but changes in vessel traits were more complicated. Vessel diameter increased across ontogeny as plant height increased. Wood anatomy traits were less tightly coupled for seedlings than for adults, such that the hydraulic safety–efficiency trade‐off observed in adults appeared to be weaker. Our results indicate that wood anatomy traits do not show coordinated ontogenetic shifts among traits because traits associated with hydraulic safety and efficiency are less integrated in seedlings. In dry forests, hydraulic trade‐offs become stronger as trees grow taller.