Abstract :
[en] Climate change has and will have a dramatic impact on species ranges. Terrestrial species have accordingly already migrated poleward at a median speed of 16.9km per decade since the beginning of the industrial era. However, many species are not equipped to efficiently track the geographic changes of the conditions matching their climatic niche, and are consequently prone to extinction. No less than about 20% of land plant species are hence threatened with extinction in the future, with major consequences on human food resources and health. In this context, Species distribution models (SDMs) offer an appealing framework to test the potential effects of climate change on species ranges. Like many biodiversity analyses, SDMs have traditionally been conducted at the species level. Cryptic speciation, which results in taxa that cannot rapidly be distinguished morphologically, but underwent divergent evolutionary histories, has been, however, increasingly reported, raising the question of whether SDMs should be fitted at the level of species (clade models), cryptic species or intraspecific lineages (subclade models). Projecting models through time further raises several questions and relies on several assumptions. In particular, projecting species potential ranges in the future based on their niche inferred from extant climate conditions onto future climatic layers involves that (I) species climatic niches are conserved through time (niche conservatism hypothesis) and that (ii) species are at equilibrium with their environment (i.e. their entire niche is filed), implying that they are not limited by their dispersal capacities, and are immediately able to colonize any newly suitable area.
Focusing on bryophytes, whose ecophysiological characteristics, such as poikilohydry and reliance on rainfall for water uptake, make them excellent candidates to study the impact of climate change, but which exhibit reduced morphologies, raising concerns about broadly defined morphological species concepts, we address here the following questions:
1 At which taxonomic level should SDMs be computed? We compare the extent to which model projections generated at the level of species differ from those obtained for intraspecific lineages. Modelling at the level of infraspecific lineages raises a second issue, which is associated with the very small sample sizes that typically characterize molecularly defined lineages, that is: how can ensemble of small models calibrated from very small datasets be evaluated? In the light of analyses of niche overlap, we finally determine whether models should be calibrated at the level of the species or intraspecific lineages.
2 Is there climatic niche conservatism in bryophytes, and how does the tendency for closely related taxa to share the same climatic niche vary at increasing taxonomic depth?
3 To what extent will such efficient dispersers as bryophytes successfully track the shift of their suitable areas during the next decades?
To address Q1, Ensembles of Small Models were evaluated by null models calibrated from randomly sampled presence points. We compared the extent of suitable area predicted by the projections of clade and subclade models. Niche overlaps were quantified using Schoener's D and Hellinger's I metrics, and the significance of these metrics in terms of niche conservatism or divergence was assessed by niche similarity tests. Combined predictions from subclade models contributed, on average, five times more than clade models to the total suitable area predicted by the combination of both subclade and clade models. Niche overlap was 0.71 on average, with evidence for niche conservatism in half of the species and no signal for niche divergence. Given the poor performance of models based on small datasets, we pragmatically suggest that, in the absence of evidence for niche divergence during diversification of closely related intraspecific lineages, SDMs should be based on all available occurrence data at the species level.
The hypothesis of climatic niche conservatism and its evolutionary ‘labillity’ was further tested at the level of an entire phylum of land plants, the Marchantiophyta, through analyses of the relationship between the spatial turnover of floras and macroclimatic variation. Phylogenetic turnover among floras was quantified through πst statistics. πst-through-time profiles were generated at 1 myr intervals along the phylogenetic time-scale and were correlated with current geographic distance and macroclimatic variation with Mantel tests based on Moran spectral randomization to control for spatial autocorrelation. The contribution of macroclimatic variation to phylogenetic turnover was about four-times higher than that of geographic distance. The correlation between phylogenetic turnover and geographic distance rapidly decayed at increasing phylogenetic depth, whereas the relationship with macroclimatic variation remained constant until 100 myrs. Our analyses reveal that changes in the phylogenetic composition among liverwort floras across the globe are primarily shaped by macroclimatic variation. They demonstrate the relevance of macroclimatic niche conservatism for the assembly of liverwort floras over very large spatial and evolutionary time scales, which may explain why such a pervasive biodiversity pattern as the increase of species richness towards the tropics also applies to organisms with high dispersal capacities.
Finally, we developed a newly designed spatially-explicit model of dispersal by wind in the context of changing climate and presented an example of application in the case of the European flora. A grid of pixel-specific environmental conditions and dispersal kernels, combining information on species dispersal traits, local wind conditions, as well as landscape features affecting dispersal by wind, was generated and used as input in simulations of species dispersal in the landscape under changing climate conditions. In European bryophytes, the median ratios between predicted range loss vs expansion by 2050 across species and climate change scenarios ranged from 1.6 to 3.3 when only shifts in climatic suitability were considered, but increased to 34.7–96.8 when species dispersal abilities were added to our models. This highlights the importance of accounting for dispersal restrictions when projecting future distribution ranges and suggests that even highly dispersive organisms like bryophytes are not equipped to fully track the rates of ongoing climate change in the course of the next decades.
