In addition to the degree of geographical isolation (sympatry, parapatry, allopatry and peripatry), ecology can be an important factor promoting diversification of lineages, both by niche divergence as well as niche conservatism. Tetraglochin is a genus of shrubs with six species distributed along the Southern and Central Andes, from Peru to southern Argentina and central Chile. Although monophyly of the genus as the identity of its species are well established, spatiotemporal framework for its diversification and the potential role of the ecology and geography in the speciation process remain unknown. In the present study we analyzed diversification times and historical biogeography of the genus, and conducted different climatic niche and geographical range comparisons among its species to determine possible patterns associated with speciation. Results support the Pleistocene diversification, early along the Southern Andes and the Patagonian Steppe and subsequently in the Central Andes. Climatic niche divergence did not prove to be a major factor promoting speciation, but rather the phylogenetic niche conservatism. Our analyses also favored the sympatric model of speciation, although patterns from geographical range evolution are difficult to interpret due to the lability on the ancestral distributions, and therefore micro-allopatric or parapatric divergence associated with the glacial-interglacial cycles and climatic oscillations throughout the Quaternary should not be discarded. Other potential factors associated with diversification of Tetraglochin are also discussed.
Tetraglochin ITS alignment
Nuclear ribosomal ITS aligned matrix in nexus format
Tetraglochin-ITS.nex
Tetraglochin trnLF alignment
Chloroplast trnL-F aligned matrix in nexus format
Tetraglochin-trnLF.nex
TETRAGLOCHIN_species_BEAST1.8.2
xml file for species tree analyses in BEAST v1.8.2 under the coalescent-based method implemented in *BEAST extension, using ribosomal ITS and chloroplast trnL-F regions, uncorrelated lognormal relaxed clock models, a yule process for the species tree prior, the piecewise linear and constant root for the population size model, and three calibrations points.
Tetraglochin_MCCT
Maximum clade credibility species tree (MCCT) estimated in BEAST v1.8.2 from nuclear ribosomal ITS and chloroplast trnL-F, using the multispecies coalescent method implemented in *BEAST, uncorrelated lognormal relaxed clock models, a yule process for the species tree prior, the piecewise linear and constant root for the population size model, and three calibrations points.
Tetraglochin_1000trees
1000 randomly subsampled posterior trees from the Bayesian species tree analyses in BEAST v1.8.2 using ribosomal ITS and the chloroplast trnL-F, uncorrelated lognormal relaxed clock models, a yule process for the species tree prior, the piecewise linear and constant root for the population size model, and three calibrations points.
Tetraglochin1000.tre
Tetraglochin_MCCT_pruned
Maximum clade credibility species tree (MCCT) estimated in BEAST v1.8.2 from nuclear ribosomal ITS and chloroplast trnL-F, pruned to include only the species of Tetraglochin
pMCCT.tetraglochin.tre
Tetraglochin_1000trees_pruned
1000 randomly subsampled posterior trees from the Bayesian species tree analyses in BEAST v1.8.2 using ribosomal ITS and the chloroplast trnL-F, pruned to include only the species of Tetraglochin.
p1000trees.tetraglochin.tre
FigS1
Figure S1. Distribution map of Tetraglochin and delimited area corresponding to the 50 km-buffered minimum convex polygon used in the niche analyses both along the environmental (E) and geographical (G) space.
FigS2
Figure S2.Results from the climatic niche analyses in the space environmental (E) space obtained using the PCA-env method. Left: Correlation circles showing the contribution of environmental variables on the first three axes of the PCA-env and the percentage of inertia explained. Right: PCAenv scatterplots of the first three principal components based on 21 climatic variables for the six Tetraglochin species represented by 388 occurrences.
FigS3
Figure S3. Biogeography of Tetraglochin species and their climatic niches along the environmental (E) space. Left: ancestral range estimation (ARE) on the Tetraglochin chronogram using the DEC model in BioGeoBEARS. States at nodes (squares) represent the area with highest ML probability before the instantaneous speciation event, whereas those on branches represent the state of the descendant lineage immediately after speciation. Squares with more than one letter refer to ancestral areas composed of more than one biogeographical area. Boxes to the left of taxon names indicate areas of tip species. Probabilities of ancestral areas are shown to the right of nodes. Right: Climatic niches for species of Tetraglochin produced by the three main axes of the PCA-env. For each species, the gray to black shading represents the grid cell density of the species occurrence (black being the highest density). The dashed line represents 50% of the available environment and the solid line represents 100%.
FigS4
Figure S4. Results from the niche comparisons between Tetraglochin species in the environmental (E) space. Niche overlap estimations among Tetraglochin species using the first three axes of the PCA-env. A. PCs 1-2. B. PCs 1-3. C. PCs 2-3. For A-C, left: pairwise niche overlap between each Tetraglochin species quantified using the Schoener’s D index, right: agglomerative hierarchical clustering UPGMA algorithm using niche overlap estimations. SA: Southern Andean distribution; CA: Central Andean distribution; SA-CA: Southern and Central Andean distribution.
FigS5
Figure S5. Bar plots of niche breadth and geographical range size values for Tetraglochin species. Niche breadth for each species was estimated as the sum of its variances on the first three principal components obtained in the PCA-env method. Geographical range size was estimated from binary (presence/absence) distributions maps for each Tetraglochin species derived from the SDM and using the maximum training sensitivity plus specificity as threshold.
FigS6
Figure S6. Results from the species distribution modelling (SDM) for Tetraglochin species. Maps show predicted suitable climatic conditions (logistic output) from the MaxEnt model for the six species of Tetraglochin, using BIO1 (Annual Mean Temperature), BIO2 (Mean Diurnal Range), BIO3 (isothermality), BIO4 (temperature seasonality), BIO9 (Mean Temperature of Driest Quarter), BIO12 (annual precipitation), BIO19 (Precipitation of Coldest Quarter), and PET (potential evapotranspiration) as climatic variables. Dots represent specimen occurrences used for the SDM analyses.
FigS7
Figure S7. Ancestral climatic tolerances plotted on the MCC species tree for variables analyzed in the geographical (G) space. Interior nodes represent the mean of estimated ancestral climatic tolerances using maximum likelihood and 100 random samples from the predicted niche occupancy (PNO) profiles. The 80 % density interval of climatic tolerances is indicated by a vertical dashed line, and the mean is indicated by the taxon label. BIO1: Annual Mean Temperature, BIO2: Mean Diurnal Range, BIO3: isothermality, BIO4: temperature seasonality, BIO9: Mean Temperature of Driest Quarter, BIO12: annual precipitation, BIO19: Precipitation of Coldest Quarter, and PET: potential evapotranspiration).
FigS8
Figure S8. Biplot of the first two principal components extracted in the phylogenetic principal component analyses (pPCA) using the weighted means of the PNOs for each species and environmental variables. BIO1: Annual Mean Temperature, BIO2: Mean Diurnal Range, BIO3: isothermality, BIO4: temperature seasonality, BIO9: Mean Temperature of Driest Quarter, BIO12: annual precipitation, BIO19: Precipitation of Coldest Quarter, and PET: potential evapotranspiration).
FigS9
Figure S9. Binary (presence/absence) distributions maps for Tetraglochin species derived from the SDM outputs using the maximum training sensitivity plus specificity as threshold.
FigS10
Figure S10. Geographical range overlap among species of Tetraglochin. Binary (presence/absence) distributions maps for each Tetraglochin species, derived from the SDM outputs and using the maximum training sensitivity plus specificity as threshold, were obtained to estimated the geographical overlap between species pairs . Green and blue cells indicate the presence of only one species, while red cells represent areas with sympatric distribution (presence of both species in the same cell).
Appendix1
Appendix 1. Taxon names, voucher material and GenBank accession numbers for trnL-F and ITS sequences respectively (trnL-F, ITS) used in the molecular study. Two accession numbers separated by dash corresponds to sequences of ITS1 and ITS2 respectively.
TableS1
Table S1. Species ocurrence data of Tetraglochin used in the niche analyses and geographic range comparisons.
TableS2
Table S2. Summary of the biogeographic event count for Tetraglochin using biogeographic stochastic mapping (BSM) in BioGeoBEARS. Simulations were conducted using the DEC model on the MCC tree and 100 randomly subsampled trees, and conducting 1000 BSMs on each tree (101000 BSMs). Mean numbers of different types of events estimated with standard deviations are shown.
TableS3
Table S3. Loadings on the first three components obtained from the principal component analysis (PCA-env) using bioclimatic variables extracted from the study area (50 km-buffered minimum convex polygon enclosing the Tetraglochin species distribution).
