Search

Aim: Trait?based approaches are increasingly important in ecology and biogeography, but progress is often hampered by the availability of high?quality quantitative trait data collected in the field. Alternative sources of trait information include scientific floras and taxonomic monographs. Here we test the reliability and usefulness of trait data acquired from scientific floras against trait values measured in the field, and those in TRY, the most comprehensive global plant trait database. Location: Tenerife and La Palma, Canary Islands, Spain. Methods: We measured leaf area and specific leaf area (SLA) in the field for 451 native vascular plant species and compared them with equivalent trait data digitised from the most recent and comprehensive guide of the Canarian flora, and data sourced from TRY. We regressed the field?measured traits against their equivalents estimated from the literature and used the regression models from one island to predict the trait values on the other island. Results: For leaf area, linear models showed good agreement between values from the scientific flora and those measured in the field (r2 = 0.86). These models were spatially transferable across islands. In contrast, for SLA we found a weak relationship between field?measured values and the best estimates from the scientific flora (r2 = 0.11). Insufficient data were available in the TRY database for our study area to calculate trait correlations with other data sources. Conclusions: Scientific floras can act as useful data sources for quantitative plant trait data for some traits but not others, whilst the TRY database contains many traits, but is incomplete in species coverage for our study region, and oceanic islands in general.

Aim: The General Dynamic Model (GDM) links island biogeographical processes to island geological history. A key premise of the GDM implies that environmental factors shaping the ecology and evolution of biota on oceanic islands follow a hump-shaped trend over the island's life span and drive dynamics in carrying capacity, species diversity and endemism. An important component of the GDM is environmental heterogeneity (EH), but its effects on insular diversity remain poorly understood. Here, we first quantified EH, tested whether EH follows the expected hump-shaped trend along island ontogeny and evaluated how EH relates to plant diversity. Location: 135 oceanic islands of volcanic origin. Taxon: Vascular plants. Methods: We calculated 20 EH metrics focusing on topographic and climatic components of EH, and compared whole-island metrics (e.g. range) and moving-window metrics (e.g. roughness). Using linear mixed-effects models, we evaluated the trends of EH with island age and the EH–plant diversity relationship expected based on the GDM. Results: Our analysis revealed some EH components to be collinear, for example, elevation and temperature heterogeneity but also that EH metrics capture different aspects of EH, for example, climatic gradients versus climatic complexity. EH generally followed a hump-shaped trend with island age, peaking early during island ontogeny. Among the EH components, climatic heterogeneity had the strongest effect on plant species richness and elevational heterogeneity on endemism. Lastly, including EH metrics in GDMs (traditionally, only island age and area were included) improved their predictive power. Main conclusions: The EH metrics compared here captured various attributes of the environment that influence insular plant diversity. In line with the GDM, our results strongly support a hump-shaped relationship between EH and island age, suggesting that islands become highly heterogeneous early in their ontogeny. Finally, the contribution of EH to GDM-based models of species richness and endemism suggests that EH is a main driver of the diversity of oceanic island biotas. Refereed/Peer-reviewed