Does biodiversity self limit?

Pollination of endemic and non-endemic species in highly diverse plant communities

Three highly diverse flowering communities

Floras at the Mediterranean regions of the Iberian Peninsula and California and the sub-tropical Peninsula of Yucatan in Mexico are all characterized by high levels of species richness and endemism. High rates of endemism in the two Mediterranean regions reflect both historical factors of reduced extinction (glacial refugia) and ecological factors promoting endemism under stressful conditions, in addition to heterogeneous topography and nutrien-poor soils. Despite a more recent geological origin the Yucatan Peninsula has also a high endimism rate likely because historic influence from Caribbean, Central America and Southern Mexico regions.

On each region we focused on specific flowering communities rich in endemic species: the Baetic dolomitic outcrops within the Natural Park of Sierras de Cazorla, Segura y Las Villas, the serpentine seeps within the Donald and Sylvia McLaughlin Natural Reserve (Northern California) and the unprotected sub-tropical coastal scrublands in the northern coast of the Yucatan state. The Californian serpentine communities have been focal to many previous taxonomic and ecological studies so they can be considered as a reference within the study (See here).

In order to characterize the plant community, permanent square plots were set every 5 m along linear transects. Total number of plots per site ranged between 20 and 30. Within every plot, we recorded the identity of each plant species focusing on the flowering community of animal-pollinated plants and therefore both wind pollinated species (e.g.,grasses, pines,...), as well as species that never flowered during the study period (May-June 2010 in Cazorla, June-July 2010 in California and August -September 2011in Yucatan), were not included. That means that our estimates reflect one year flowering and could be consequently affected by the specific abiotic conditions conditioning seasonal flowering (e.g.,winter snow, precipitation, temperature). In addition, and due to the heterogeneous and specific soil type of the study sites in California and Cazorla, data sets were filtered before analysis in order to exclude species that were not strictly associated to the study habitat, but to other soil types intermixed, and that appeared within less than 20% of sampled plots in region.

Sample-based rarefaction methods were applied to the species incidence data to compare the species richness of study plant communities among regions, taking into account that habitat specificities precluded us to conduct the same sampling effort at all sites and regions. Results indicate that our sampling effort was sufficient to obtain stable species richness estimators at all the study regions. Californian serpentine seeps recorded intermediate species richness that the other two regions. The highest species diversity at Cazorla was due to reduced overlapping in the species identity among study sites. Indeed, some of the most typical dolomite species present at the study locations, as for instance Viola cazorlensis or Erysimun cazorlense, were never found within our study plots.