Home PhD dissertations

Plant diversity patterns across scales in a Mediterranean biodiversity hotspot; the interface between ecology and history in the Baetic-Rifan range

Rafael Molina Venegas. University of Seville. December 2014. Supervisors: Juan Arroyo Marín, Abelardo Aparicio y Sébastien Lavergne.
Summary:
The evolutionary and ecological processes that shape species diversity shift as spatial and temporal scales change and diversity patterns in assemblages may thus reflect these shifts. In this PhD Thesis, I explore taxonomic, phylogenetic and functional diversity patterns in plant assemblages on different geographical scales in the Baetic-Rifan range, a western Mediterranean biodiversity hotspot, as a means of inferring the relative contribution of each of the evolutionary and ecological processes responsible for the emergence and maintenance of plant biodiversity in this region. The Baetic-Rifan biodiversity hotspot is an ideal study system for exploring the interaction between such processes due to its extraordinarily diverse and well-known regional species pool, which includes both recent narrow endemics and the extant representatives of past floras, and its heterogeneous mosaic-like landscapes characterized by great geomorphological and lithological complexity.

In order to validate the results and conclusions reached in this PhD Thesis, I tested the influence of phylogenetic resolution and branch-length information on the quantification of phylogenetic structure, and explored the impact of tree shape (‘steminess’) on the loss of accuracy in phylogenetic structure quantification caused by phylogenetic resolution. Directional biases in phylogenetic structure quantification due to phylogenetic resolution and dating methods generally tended towards type II errors. Interestingly, I detected that the tree shape strongly influenced the loss of accuracy derived from the lack of phylogenetic resolution, which means that phylogenetic structure estimates based on poorly resolved phylogenies across biogeographical regions should be interpreted with caution.

I found that West Rifan plant assemblages resembled Andalusian ecoregions more than those of nearby northern Morocco ecoregions, which thus highlights the role played by the Strait of Gibraltar in the past as a migration route for plants from both sides of this strait. However, phylogenetic turnover in terminal lineages tends to occur right across the two landmasses of this hotspot, which thus suggests a common scenario of allopatric speciation among disparate angiosperm lineages that may have been facilitated by the intermittent joining of the two continents. Phylogenetic clustering reported on the eastern margins of the hotspot may be in large part due to syndrome-driven local diversification occurring in shrublands in the southeastern Iberian Peninsula. Diversity patterns of endemic taxa do not mirror the diversity patterns of the regional pool in which they are included. Specifically, although taxonomic turnover in regional endemics throughout the Baetic ranges was related to lithological conditions and geographical distance, climate and/or elevation explain better the turnover of non- endemic taxa. By contrast, the phylogenetic turnover of endemics was weakly related to lithological conditions, which suggests that edaphic specialist endemics have evolved from multiple lineages and highlights the role of the substrate in promoting differentiation in the Baetic ranges. I also found strong structuring of plant lineages along elevational gradients in the Baetic range, probably due to habitat filtering acting on life forms and character syndromes with strong phylogenetic signals. These results indicate that lineages belonging to the sclerophyllous syndrome, many of which are the extant representatives of the flora that once existed under the tropical-like Tertiary climatic conditions, seem to be ecologically restricted to the lowlands of milder climate, whereas non-sclerophyllous lineages are able to expand across higher elevation ranges where colder climatic conditions prevail. At community scale, elevation was the main factor in predicting phylogenetic and functional alpha diversity (maximum height and blooming time) in plant communities for both incidence-based and abundance data. However, their trends were opposing since phylogenetic diversity increases and functional alpha diversity decreases with greater elevation. I also detected a shift in deviances in local species abundances towards larger seed size and phylogenetic clustering at low elevation and divergent distributions and phylogenetic overdispersion at greater elevation. By contrast, turnover in incidence-based specific leaf area and leaf carbon:nitrogen ratio values were related to a pH and micronutrient gradient. These results highlight the multidimensional nature of the functional niche of species and suggest that different trait-mediated mechanisms are simultaneously at work in shaping the assemblage of local plant communities.