Search

Aim Determine if differences in species richness between currently climatically similar regions correlate to dissimilarities in functional diversity. Location Time period Europe and eastern North America. Neogene and Quaternary. Major taxa studied Methods Temperate trees of Europe and eastern North America. Using classic functional traits defining distinct ecological strategies, we described the trait spaces of European and eastern North American tree floras. The trait space of each region was described based on the occupied area, species accumulation pattern, and the trait-space size, dispersion and clustering. We then evaluated if the species richness differences between these two climatically similar regions correlate to dissimilarities in both occupied areas and species accumulation patterns; as well as differences in the trait-space size, dispersion and clustering. Results Main conclusions Differences in species richness between climatically similar regions do not necessarily result in functional dissimilarities. Regardless of eastern North American species having a larger trait-space, occupied areas and species accumulation patterns converged. Although in both regions species clumped towards the trait-space centroid, we observed statistically significant differences between Europe and eastern North America in the dispersion and clustering but not in size of the trait space. Moreover, traits from species in genera still present in eastern North America but lost from Europe during the Neogene and Quaternary were fully contained within the currently occupied European trait space. Positional convergence between European and eastern North American trait spaces suggests that this species richness anomaly does fully translate to the functional space. Our results suggest that species accumulation within a region occurs within a climatically restricted trait space, and not via trait-space expansion. Moreover, the consistent aggregation of species towards the trait-space centre aligns with the idea of a directional section towards a generalized morphology, which might provide the best way to interact with a broad array of environmental conditions.

As the earth system moves to a novel state, model systems (experimental, observational, paleoecological) are needed to assess and improve the predictive accuracy of ecological models under environments with no contemporary analog. In recent years, we have intensively studied the no-analog plant associations and climates in eastern North America during the last deglaciation to better constrain their spatiotemporal distribution, test hypotheses about climatic and megaherbivory controls, and assess the accuracy of species- and community-level models. The formation of no-analog plant associations was asynchronous, beginning first in the south-central United States; at sites in the north-central United States, it is linked to declining megafaunal abundances. Insolation and temperature were more seasonal than present, creating climates currently nonexistent in North America, and shifting species-climate relationships for some taxa. These shifts pose a common challenge to empirical paleoclimatic reconstructions, species distribution models (SDMs), and conservation-optimization models based on SDMs. Steps forward include combining recent and paleoecological data to more fully describe species' fundamental niches, employing community-level models to model shifts in species interactions under no-analog climates, and assimilating paleoecological data with mechanistic ecosystem models. Accurately modeling species interactions under novel environments remains a fundamental challenge for all forms of ecological models.