Replicated Patterns of Population Genetic Structure Mirror
Community Structure Across the Hengduan Mountains
Biodiversity Hotspot
Deren A. R. Eaton Dept. of Ecology, Evolution, and Environmental Biology Columbia University
Co-authors: Jared Meek, Isaac Overcast, Patrick McKenzie, Jianjun Jin, Yue Yang, and Richard Ree
Community structure arises from the intersection of ecology and Biogeography: The spatial distribution of species
Population structure arises from the intersection of ecology and Phylogeography: The spatial distribution of alleles
Temporal hypothesis 1: Biogeographic patterns arise from
replicated phylogeographic patterns
Temporal hypothesis 2: Phylogeographic patterns are shaped
by existing biogeographic patterns
Plant Biodiversity, Genomics, and Reproductive Evolution
The Hengduan Mountains
The Hengduan Mountains
The Hengduan Mountains
Features/Topography of the Hengduan Mountains
Features/Topography of the Hengduan Mountains
Which barriers are most significant?
The genus in Pedicularis
Pedicularis L. (Orobanchaceae)
>600 species globally, >300 endemic to Hengduan Mountains
Extravagant floral diversity and convergence (shape and color)
(Ree 2005, Yang et al. In Prep)
Species rich assemblages of co-flowering spp;
(Eaton & Ree 2012; Tong & Huang 2016)
Extensive herbaria collections (KIB, HUH, FMNH)
High potential for reproductive conflict
Species interactions / Character displacement / Speciation
Reproductive Interference: Negative fitness consequences of reproduction being
disrupted by another organism. E.g., pollinator attraction and pollen placement.
Competitive exclusion evidenced by floral phenotypic overdispersion in assemblages (Eaton & Ree 2012).
Partitioning of shared pollinators
Mechanical isolation
The beak
of the galea directs pollen placement and pickup.
Isolation enables local adaptation
Differences among populations (within-species) may result from interspecific
interactions driving character displacement in local communities.
How are patterns of within-species gene flow shaped by geographic barriers?
How are patterns of across-species dispersal and community assembly shaped by geographic barriers?
Expansion from the north; structured by river basins
Populations are highly structured
Expansion from the south; structured by river basins
Populations are highly structured
Expansion from the south; structured by river basins
Effective migration surface: Prhi
Effective migration surface: Plon
Effective migration surface: Pcra
How do migration surfaces compare among species?
How do migration surfaces compare among species?
Effective migration surfaces across 12 species
Stacked migration statistics across 12 species
Stacked migration statistics across 12 species
Replicated phylogeography conclusions
Most spp migration surfaces are not concordant, but...
Tibetan Plateau: high migration, uniform among spp.
Southern HMR: low migration, variable among spp.
Three Rivers: high/med migration, uniform within and across.
Isolation by distance within mountain ranges
Pedicularis community migration/turnover
Data: 13,566 occurrence records of Pedicularis from GBIF + HUH + our collections.
HDBSCAN: cluster occurrences by distance & elevation = 211 assemblages
Assemblage motif membership (K=7) (ecostructure)
Pedicularis community structure (across 320 species) clusters by region
similarly to genetic population structure in the 12 widespread species.
Effective Assemblage Migration in fEEMs
Using site-by-species matrix in place of site-by-allele matrix. High
migration = low spp turnover across space; low migration = high spp turnover across space.
Population Genetic ~ Assemblage Migration
Patterns are highly concordant (orange=both high or both low).
Regions acting as barriers to species co-occurrence also
act as barriers to gene flow, and vice-versa.
Population Genetic ~ Assemblage Migration
Migration is lower across large elevation changes
For paths of length=5 on graph: sum migration x sum change in elevation.
Conclusions (and what's next)
Across 12 widespread species we found strong population structure.
But, spp often diverged over shared barriers in different orders.
And so most species exhibit different effective migration surfaces.
But when stacked, some consistent patterns emerge.
Across >200 assemblages we found strong community structure.
Community migration patterns mirror within-species genetic migration patterns.
Temporal hypothesis 1: Biogeographic patterns arise from
replicated phylogeographic patterns
Temporal hypothesis 2: Phylogeographic patterns are shaped
by existing biogeographic patterns
