The need for a strict delimitation of early tree life stages in vegetation ecology

Posted on by

Prepared by Ricard Arasa-Gisbert, Víctor Arroyo-Rodríguez, Martín de Jesús Cervantes-López & Jorge A. Meave

Proposed standardized criteria (left Y axis) for defining seedlings (0 to < 50 cm height), saplings (50 to < 200 cm height and/or < 2 cm diameter at breast height, DBH) and juveniles (≥ 200 cm height and/or 2 to 9.9 cm DBH) of canopy trees. We also indicate the lower and upper observed measures (dots in boxplots within each panel) for height (a) and DBH (b) in studies evaluating the effects of defaunation (36 studies) and logging (34 studies) on these tree life stages. These 67 studies (3 studies evaluated both disturbances) resulted from a systematic literature survey and the application of multiple exclusion criteria. Note that some studies evaluated more than one tree life stage and that some studies did not include lower or upper limits of these stages for the same variable (i.e., DBH or height)

Imagine a study that tested a new drug on humans, taking 3-month-old and 15-year-old individuals and lumping them together into a single category called ‘infants’. Would you trust their results? Probably not. Well, something similar is currently happening in vegetation ecology.

Unlike adult trees, for which we have standardized measures in vegetation ecology (for example, a minimum stem diameter of 10 cm at breast height, that is, 1.3 m above the ground), we do not have the same for the early life stages of trees. In other words, we lack standardized quantitative criteria to decide when a seedling ceases to be a seedling and becomes a sapling, and to decide when a young plant is no longer a sapling and becomes a juvenile. For this reason, researchers often subjectively define what they consider to be a seedling or a sapling. This situation causes big problems when assessing the overall effect of a given disturbance factor (for instance, logging) on a variable of interest (for instance, species diversity). As a result of this ambiguity, one author may conclude that logging is good for seedlings (defined as young plants 0-30 cm tall), while another one may conclude that logging is bad for seedlings (defined as young trees 30-200 cm tall). Who is wrong? Nobody. These authors are simply clumping young plants that do different things in the forest (for example, that possess different sensitivities and strengths) into a single category, as if we were grouping newborns with teenagers in a single category, following the introductory example.

To provide a solution to this complication, we proposed a set of standardized measures or criteria based on biological and ecological rationales that allow us to qualitatively tell apart seedlings from saplings, and these from juvenile trees. We then tested the validity of these criteria by conducting a complex analysis (called meta-analysis) that examined the consequences of the regional loss of animals (defaunation) and of logging on some plant response variables (for example, species diversity). To this end, we grouped our data into (1) studies that matched our standardized criteria, (2) studies that did not match them, and (3) all studies together, regardless of the criteria they used. We observed a higher proportion of relevant effects when only considering the studies matching our standardized criteria for both disturbance types. The direction of the effects (positive or negative) changed in as many as 30–50% of the cases when comparing the results obtained for all data and for studies not matching our criteria with the results for studies matching our criteria.

These findings highlight the need for standardized measures when studying early tree life stages, as this increases the detectability of relevant effects that otherwise could go unnoticed when grouping plants that are quite different into a single category. This standardized protocol will be particularly useful in synthesis and reviews and sets a benchmark for future studies on plant community ecology, especially in species-rich ecosystems such as tropical forests.

This is a plain language summary of the paper of Arasa-Gisbert et al. (2024) published in the Journal of Vegetation Science (https://doi.org/10.1111/jvs.13313)