Search birdRS Box

Search birdRS blog posts

Browse the Blog Posts

Or scan through the blog archive below for items of interest as only the latest post is shown below, thanks.

Tuesday, 5 August 2014

Temperate origins of long-distance seasonal migration in New World songbirds. PNAS August 4, 2014: Paper Highlights

Temperate origins of long-distance seasonal migration in New World songbirds

Published online before print August 4, 2014, 
doi: 10.1073/pnas.1405000111 

PNAS August 4, 2014
Supporting Information

Benjamin M. Winger a,b,1, 
F. Keith Barker c,d, 
and Richard H. Ree a,b

a: Committee on Evolutionary Biology, University of Chicago, Chicago, IL 60637; 
b: Life Sciences Section, Integrative Research Center, Field Museum of Natural History, Chicago, IL 60605;
c: Department of Ecology, Evolution and Behavior;
d: Bell Museum of Natural History, University of Minnesota, St. Paul,MN 55108


Author's 'Significance' Quote
Determining where species arose and how they spread across the globe is paramount to understanding geographic patterns of biodiversity. For migratory organisms, this task has been difficult due to the complex geography of seasonal migration. One longstanding hypothesis is that migratory animals originated in the tropics and that migration evolved through shifts of breeding ranges to temperate regions. We tested this hypothesis in the largest radiation of migratory birds in the Americas, the emberizoid passerines. Contrary to expectations, we found that long-distance migration primarily evolved through evolutionary shifts of geographic range south for the winter out of North America, as opposed to north for the summer. Our results further suggest that seasonal migration promoted colonization of the tropics from North America.

Migratory species exhibit seasonal variation in their geographic ranges, often inhabiting geographically and ecologically distinct breeding and nonbreeding areas. The complicated geography of seasonal migration has long posed a challenge for inferring the geographic origins of migratory species as well as evolutionary sequences of change in migratory behavior. To address this challenge, we developed a phylogenetic model of the joint evolution of breeding and nonbreeding (winter) ranges and applied it to the inference of biogeographic history in the emberizoid passerine birds. We found that seasonal migration between breeding ranges in North America and winter ranges in the Neotropics evolved primarily via shifts of winter ranges toward the tropics from ancestral ranges in North America. This result contrasts with a dominant paradigm that hypothesized migration evolving out of the tropics via shifts of the breeding ranges. We also show that major lineages of tropical, sedentary emberizoids are derived from northern, migratory ancestors. In these lineages, the winter ranges served as a biogeographic conduit for temperate-to-tropical colonization: winter-range shifts toward the tropics during the evolution of long-distance migration often preceded southward shifts of breeding ranges, the loss of migration, and in situ tropical diversification. Meanwhile, the evolution of long-distance migration enabled the persistence of old lineages in North America. These results illuminate how the evolution of seasonal migration has contributed to greater niche conservatism among tropical members of this diverse avian radiation. 

The evolution of seasonal migratory behavior among animals
involves a suite of behavioral, physiological, morphological,
and neurological adaptations that enable migrants’ extraordinary feats of endurance and navigation. However, the evolution of migration also is an inherently geographic process during which a species’ breeding range and nonbreeding range (henceforth, winter range) become physically and ecologically separated. Understanding the evolution of migration therefore requires reconciling the fascinating adaptations of migratory individuals with the biogeographic factors that control the shifting boundaries of a species’ range. The field of historical biogeography has shed considerable light on the geographic histories of organisms but has largely ignored migratory species due to the difficulty of simultaneously reconstructing the evolution of the breeding and winter ranges, which in migratory species are often ecologically disparate and separated by long distances. Consequently, progress in our understanding of the evolution of migration has been impeded by a biogeographic conundrum: testing hypotheses on the evolution of migration requires knowledge of the geographic histories of migratory species, but the existence of migratory behavior in a lineage confounds our ability to infer these histories.
This difficulty in resolving the geographic provenance of mi-
gratory species not only has left incomplete our understanding of the geographic histories of many lineages that contain migrants but also has impaired our ability to discriminate among hypotheses on the selective forces that drive the evolution of migratory behavior. For over a century, the principal dichotomy among hypotheses on the evolution of bird migration has hinged on a question of geographic ancestry: does seasonal migration evolve through a geographic shift of the breeding grounds away from an ancestral year-round range, or via a shift of the wintering grounds? The most visible bird migrations occur between breeding regions at temperate latitudes to wintering areas at lower, more tropical latitudes. The dominant paradigm in the literature on the evolution of migration has imagined these long-distance migrations as evolving via shifts of the breeding range out of the tropics, driven by increased reproductive success and reduced competition in temperate regions. An opposing camp has hypothesized that migration evolves when species resident year-round in temperate latitudes shift their winter ranges to lower latitudes to increase survival during the harsh and resource-depleted temperate winters. Much debate has occurred over the selective forces that would make a tropical versus temperate ancestry of migratory birds more likely. However, due to the absence of historical biogeographic models capable of handling the complex geographic ranges of migratory species, previous studies have had difficulty determining which geographic shifts produced the distributions of migratory species observed today, as well as where migratory lineages originated. To address this challenge, we designed a phylogenetic model specifically for inferring the biogeographic history of migratory lineages. Our model is inspired by the dispersal–extinction–cladogenesis (DEC) model, in which geographic range evolves via discrete events of dispersal and local extinction along phylogenetic branches and via inheritance and subdivision at speciation events. Our model follows similar logic but is novel in jointly considering the evolution of both the breeding and winter range.
The discrete states of our model are presence–absence grids
whose cells signify breeding and wintering in three latitudinal
regions (Materials and Methods). These grids, which we refer to
as “dominos,” summarize the ranges of New World bird species
in each season (Fig. 1 and Table S1). Ancestor-to-descendant
transitions between dominos represent expansion or contraction
events of the breeding and/or winter range and describe broad
patterns of change in the geography of seasonal migration (Fig. 2
and Fig. S1). We focus on the evolution of long-distance bird
migration between summer breeding grounds in temperate North
America and winter grounds at subtropical or tropical latitudes.
This migratory system, known as Nearctic-Neotropical migration
(henceforth, Neotropical migration), involves the largest number of species of any avian migratory system in the New World. Applied to a phylogenetic tree, the domino model enables estimation of ancestral breeding and winter ranges (that is, ancestral dominos), as well as rates of ancestor–descendant transitions between dominos (Materials and Methods). Our estimation of transition rates results in an adjacency matrix that describes every possible geographic change throughout phylogenetic history (Fig. 3A). The complexity of this matrix illustrates the principal challenge of inferring the biogeographic histories of migratory species: many different sequences of geographic change could explain the evolution of a given migratory species’ range.
To overcome this challenge, we used graph theory and network
analysis to extract from the rate matrix the dominant pathways
of geographic evolutionary change that led to the evolution of
Neotropical migration (Fig. 3 B and C and Materials and Methods).
We applied the model in a study of the largest New World
radiation of migratory birds, the emberizoid passerines (super-
family Emberizoidea), for which we had a recent and comprehensive species-level molecular phylogeny. This lineage of ∼823 songbird species contains all New World warblers, sparrows, blackbirds, orioles, cardinals, buntings, tanagers, and allies. Most emberizoid diversity is comprised of nonmigratory, tropical species, mirroring the more general global trend of higher species diversity in the tropics than in temperate regions. However, all major lineages of Emberizoidea except the Thraupidae (tanagers) also contain Neotropical migrants (species that breed in North America and spend the northern winter in the Neotropics); in total, the group contains 120 species of Neotropical migrants, which together represent 25% of all Neotropical migratory bird species. The ancestral emberizoid is thought to have colonized the Americas via Beringia and thus have a northern origin in the New World. However, insight into the group’s biogeographic history in the New World has been complicated by the dilemma of Neotropical migration, particularly because multiple gains and losses of migration are evident throughout emberizoid history. Did major emberizoid lineages originate in the tropics, implying that Neotropical migration evolved in these lineages via shifts of the breeding ranges to North America? Or did migratory emberizoid lineages originate in North America and evolve Neotropical migration via shifts of the winter ranges toward the equator? What does the geographic history of Neotropical migration imply for the origins, geographic spread and diversification of this diverse, widespread radiation?

We used our model to examine the biogeographic events that accompanied gains and losses of Neotropical migration, as well as the influence of Neotropical migration on colonization of
the tropics.

Our study reconciles migratory behavior with geographic origins to untangle a complex geographic history. Despite early invasions of emberizoid lineages such as Thraupidae into tropical latitudes, our analyses demonstrate that long-distance migration between North America and the Neotropics evolved primarily via shifts of winter ranges from North America as opposed to breeding-range shifts from tropical latitudes (Figs. 3 and 4). These winter-range shifts occurred most notably in the ancestor to the largest clade of emberizoid migratory birds, Parulidae and Icteridae (starred blue branch in Fig. 4), and repeatedly in the Passerellidae (blue branches, Fig. 4). In contrast, shifts of breeding ranges out of tropical latitudes into North America during the evolution of Neotropical migration occurred more rarely and at a lower rate (Fig. 3). Apparent tropical origins of Neotropical migration led to small numbers of migratory species in Icterus orioles (Icteridae) (Fig. 4, node 1) and in the Cardinalidae (Fig. 4, nodes 2–4). Notably, neither Neotropical migration nor North American residency has evolved in the
largest emberizoid family, the Thraupidae, despite the widespread diversification of this lineage throughout the tropics and the evolution of shorter-distance intratropical migrations in
this group.
We do not interpret the emergence of Neotropical migration in Emberizoidea as de novo evolution of migratory behavior, as migration in some form likely traces much deeper in the avian tree of life. Rather, our study illuminates the geographic origins and history of a major migratory system during the hemisphere-wide radiation of this diverse New World lineage. Our results are striking considering that lineages containing Neotropical migrants have centers of species diversity in the tropics. Previous workers have often cited the taxonomic relationships between nonmigratory, tropical species and temperate migrants as evidence that migration evolved out of the tropics and have suggested myriad selective forces to explain the evolution of long-distance seasonal migration out of the tropics. For example, one hypothesis suggested that localized movements that evolve in the tropics, such as the search for ephemeral food resources by frugivores, may be “evolutionary precursors” to long-distance migration. That is, these hypotheses suggest that demographic and selective forces operating in tropical breeding regions drive the evolution of long-distance migration out of the tropics. However, our results suggest that traditional out-of-the-tropics hypotheses of bird migration do not explain the emergence of long-distance migration in the largest lineage of Neotropical migratory birds.
Rather, our results are consistent with hypotheses suggesting that shifts of the winter range are the primary drivers of the evolution of long-distance migration. Neotropical migration emerged early in Emberizoid history (Fig. 4), and age estimates for major emberizoid lineages date as far back as the early Miocene (15-20 MyBP), with the split from the sister family to Emberizoidea, the cosmopolitan Fringillidae, dated to the late Oligocene (20-25 MyBP). Thus, early shifts of winter ranges (e.g., starred node, Fig. 4) may have occurred in response to global cooling and increasing seasonality at high latitudes since the late Oligocene. As such, early shifts of winter ranges may not represent invasions of tropical habitats as much as tracking of tropical habitat to lower latitudes to escape harsh winter conditions while maintaining breeding at high latitudes. Plio-Pleistocene glaciations have clearly served to modify geographic ranges and migratory distances and routes and may have had an influence on more recent shifts of winter ranges out of North America in the Passerellidae (blue branches, Fig. 4). However, recent glacial cycles were not responsible for the initial development of Neotropical migration in Emberizoidea. The dominant biogeographic pattern throughout emberizoid history is one of colonization of the tropics from the north temperate region (Fig. S5), and our analyses revealed that this colonization occurred via three processes: fragmentation of widespread ranges, expansion of resident ranges of nonmigratory populations from temperate to tropical latitudes, and the expansion of a migratory species’ breeding range into a tropical area already occupied by their winter range (Fig. S6). Secondary
colonization of North America from the tropics occurred surprisingly rarely, given the high species diversity and deep history of Emberizoidea in the tropics. These results contrast with out-of-the-tropics models for the evolution of the latitudinal diversity gradient and instead are consistent with the hypothesis that tropical species have greater conservatism in ecological niche than temperate species. Our results further suggest that, in this diverse radiation of migratory birds, shifts of winter ranges out of North America during the evolution of migration served as a biogeographical conduit that bolstered the establishment of temperate lineages in the tropics and preceded in situ tropical diversification. Meanwhile, the evolution of long-distance migration out of North America also enabled the persistence and subsequent diversification of old lineages in temperate North America, by allowing species to escape inhospitable winter conditions each year. Thus, the complex geographic history of seasonal migration has at once promoted colonization of the tropics from the temperate zone and
enabled the persistence and diversification of incumbent lineages in temperate, seasonal environments.

No comments:

Post a Comment