skip to main content


Title: A global analysis of bats using automated comparative phylogeography uncovers a surprising impact of Pleistocene glaciation
Abstract Aim

Our work seeks to understand the global demographical response of bat species to the climate change that occurred at the Last Glacial Maximum (LGM).

Location

All continents except Antarctica.

Methods

MitochondrialDNAsequences were sampled from bat species throughout the planet where we could associate a georeferenced sample with a givenDNAsequence. Our investigation estimates the historical demographical response using over 12,000 samples from >300 nominal species of bats. CustomPythonand R scripts were written to aggregate sequence data from GenBank, locality information fromGBIF, and to associate these records to individual samples. We conducted approximate Bayesian computation to calculate the posterior probability of demographical bottleneck and expansion responses to the end of the Pleistocene, and then collected organismal trait data to identify traits that were associated with either demographical response. We also used R to estimate current and end‐Pleistocene species distribution models (SDM) for species where >10 georeferenced samples were available.

Results

Analysis of the genetic data indicate that some temperate insectivores responded to the end of the Pleistocene by undergoing a demographical expansion. However, the neotropical family Phyllostomidae experienced the most dramatic response, with many of its species undergoing demographical bottlenecks. Larger bats, and those with shorter forewings, were more likely to undergo a demographical bottleneck. In contrast with the results of the genetic data analysis, the automated SDMs all predicted range expansion since the LGM.

Main conclusions

Historical populations of Neotropical bats that rely on Angiosperms for resources (i.e., pollen, nectar, fruit) were negatively influenced by the climate change that occurred at the end of the Pleistocene. Our work highlights the utility of incorporating exploratory trait‐based analyses in phylogeography. It serves as an example of automated big data phylogeography, and suggests that repurposed data can lead to new insights about global biodiversity.

 
more » « less
NSF-PAR ID:
10061612
Author(s) / Creator(s):
 ;  ;  ;  
Publisher / Repository:
Wiley-Blackwell
Date Published:
Journal Name:
Journal of Biogeography
Volume:
45
Issue:
8
ISSN:
0305-0270
Page Range / eLocation ID:
p. 1795-1805
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    The persistence of populations declining from novel stressors depends, in part, on their ability to respond by trait change via evolution or plasticity. White‐nose syndrome (WNS) has caused rapid declines in several North America bat species by disrupting hibernation behaviour, leading to body fat depletion and starvation. However, some populations ofMyotis lucifugusnow persist withWNSby unknown mechanisms.

    We examined whether persistence ofM. lucifiguswithWNScould be explained by increased body fat in early winter, which would allow bats to tolerate the increased energetic costs associated withWNS. We also investigated whether bats were escaping infection or resistant to infection as an alternative mechanism explaining persistence.

    We measured body fat in early and late winter during initialWNSinvasion and 8 years later at six sites where bats are now persisting. We also measured infection prevalence and intensity in persisting populations.

    Infection prevalence was not significantly lower than observed in declining populations. However, at two sites, infection loads were lower than observed in declining populations. Body fat in early winter was significantly higher in four of the six persisting populations than duringWNSinvasion.

    Physiological models of energy use indicated that these higher fat stores could reduceWNSmortality by 58%–70%. These results suggest that differences in fat storage and infection dynamics have reduced the impacts ofWNSin many populations. Increases in body fat provide a potential mechanism for management intervention to help conserve bat populations.

     
    more » « less
  2. Abstract Aim

    We used genome‐scale sampling to assess the phylogeography of a group of topminnows in theFundulus notatusspecies complex. Two of the species have undergone extensive range expansions resulting in broadly overlapping distributions, and sympatry within drainages has provided opportunities for hybridization and introgression. We assessed the timing and pattern of range expansion in the context of late Pleistocene–Holocene drainage events and evaluated the evidence for introgressive hybridization between species.

    Location

    Central and southern United States including drainages of the Gulf of Mexico Coastal Plain and portions of the Mississippi River drainage in and around the Central Highlands.

    Taxon

    Topminnows, GenusFundulus, subgenusZygonectesFundulus notatus, Fundulus olivaceus, Fundulus euryzonus.

    Methods

    We sampled members of theF. notatusspecies complex throughout their respective ranges, including numerous drainage systems where species co‐occur. We collected genome‐wide single nucleotide polymorphisms (SNPs) using the genotype‐by‐sequencing (GBS) method and subjected data to population genetic analyses to infer the population histories of both species, including explicit tests for admixture and introgression. The methods employed includedSTRUCTURE, principal coordinates analysis, TreeMix and approximate Bayesian computation.

    Results

    Genetic data are presented for 749 individuals sampled from 14F. notatus, 20F. olivaceusand 2F. euryzonuspopulations. Members of the species complex differed in phylogeographic structure, withF. notatusexhibiting geographic clusters corresponding to Pleistocene coastal drainages andF. olivaceuscomparatively lacking in phylogeographic structure. Evidence for interspecific introgression varied by drainage.

    Main conclusions

    Populations ofF. notatusandF. olivaceusexhibited contrasting patterns of lineage diversity among coastal drainages, indicating interspecific differences in their Pleistocene southern refugia. Phylogeographic patterns in both species indicated that range expansions into the northern limits of contemporary distributions coincided with and continued subsequent to the Last Glacial Maximum. There was evidence of introgression between species in some, but not all drainages where the species co‐occur, in a pattern that is correlated with previous estimates of hybridization rates.

     
    more » « less
  3. Understanding megafaunal population dynamics is critical to testing and refining scenarios of how extinctions occurred during the terminal Pleistocene. Large‐scale, collections‐based, chronological and taphonomic analyses of midwestern Proboscidea suggest divergent population histories in mammoths and mastodons after the Last Glacial Maximum (LGM). Although extinction of both taxa occurred in the very late Bølling‐Allerød (B‐A) or early Younger Dryas (YD),Mammuthusis dominant during theLGMwith a decreasing presence in the region until extirpation.Mammut americanumhowever, is absent during theLGMbut becomes the dominant taxon during the subsequent B‐A. Trace and physical evidence of large carnivores in post‐LGMproboscidean assemblages is extremely rare, suggesting apex predators had minimal impact on mammoth and mastodon populations at this time. The ultimate mechanism(s) of extinction remain unclear; however, the wide geographical distribution of lateMammutand an increase in the frequency of multi‐animal death assemblages is consistent with a large, destabilized population experiencing periodic boom‐bust cycling at the end of the B‐A. We suggest this pattern is due to the collapse of trophic controls on proboscidean populations prior to theLGMand a subsequent system‐wide shift from top‐down to bottom‐up regulatory mechanisms in Proboscidea.

     
    more » « less
  4. Premise

    Abscission zones (AZ) are specialized cell layers that separate plant parts at the organ junction upon developmental or environmental signals. Fruit or seed abscission has been well studied in model species because of its crucial role for seed dispersal. Previous work showed thatAZlocalization differs among species of Poaceae and thatAZformation is histologically and genetically distinct in three distantly related grass species, refuting the idea of a broadly conserved module. However, whetherAZstructure is consistent within subfamilies is unknown.

    Methods

    Eleven species were selected from six subfamilies of Poaceae, and theirAZwas investigated using paraffin‐embedded, stained material. Observations were added from the literature for an additional six species. Data were recorded onAZlocation and whether cells in theAZwere distinguishable by size or lignification. Characteristics of theAZwere mapped on the phylogeny using maximum likelihood.

    Results

    Abscission zone anatomy and histology vary among species, and characteristics of theAZdo not correlate with phylogeny. Twelve of the seventeen studied species have anAZin which the cells are significantly smaller than surrounding cells. Of these, eight have differential lignification. Differential lignification is often associated with differential cell size, but not vice versa.

    Conclusions

    Neither smaller cells in theAZnor differential lignification between the AZand surrounding cells is required for abscission, although differential cell size and lignification are often correlated. Abscission zone anatomy does not correlate with phylogeny, suggesting its rapid change over evolutionary time.

     
    more » « less
  5. Background

    Children with the temperament of behavioral inhibition (BI) face increased risk for social anxiety. However, not all children withBIdevelop anxiety symptoms. Inhibitory control (IC) has been suggested as a moderator of the pathway betweenBIand social anxiety. This study uses longitudinal data to characterize development ofICand tests the hypothesis thatICmoderates associations between earlyBIand later social anxiety symptoms.

    Methods

    Children completed a Go/Nogo task at ages 5, 7, and 10 years as part of a longitudinal study ofBI(measured at 2–3 years) and social anxiety symptoms (measured at 12 years). To assessICdevelopment, response strategy (criterion) and inhibitory performance (d′) were characterized using signal detection theory. Latent growth models were used to characterize the development ofICand examine relations amongBI,ICparameters, and social anxiety symptoms.

    Results

    ICresponse strategy did not change between 5 and 10 years of age, whereasICperformance improved over time.BIscores in toddlerhood predicted neither initial levels (intercept) nor changes (slope) inICresponse strategy orICperformance. However, between ages 5 and 10, rate of change inICperformance, but not response strategy, moderated relations betweenBIand later parent‐reported social anxiety symptoms. Specifically, greater age‐related improvements inICperformance predicted higher levels of social anxiety in highBIchildren.

    Conclusions

    ICdevelopment in childhood occurs independent ofBIlevels. However, rapid increases inICperformance moderate risk for social anxiety symptoms in children withBI. Implications for theory and practice are discussed.

     
    more » « less