Eawag_06930 The evolutionary diversification and biogeography of parrots (Aves: Psittaciformes): an integrative approach Inauguraldissertation der Philosophisch-naturwissenschaftlichen Fakultät der Universität Bern vorgelegt von Manuel Schweizer von Hasle bei Burgdorf BE Leiter der Arbeit: Prof. Marcel Güntert Naturhistorisches Museum der Burgergemeinde Bern Prof. Ole Seehausen Institut für Ökologie und Evolution Universität Bern The evolutionary diversification and biogeography of parrots (Aves: Psittaciformes): an integrative approach Inauguraldissertation der Philosophisch-naturwissenschaftlichen Fakultät der Universität Bern vorgelegt von Manuel Schweizer von Hasle bei Burgdorf BE Leiter der Arbeit: Prof. Marcel Güntert Naturhistorisches Museum der Burgergemeinde Bern Prof. Ole Seehausen Institut für Ökologie und Evolution Universität Bern Von der Philosophisch-naturwissenschaftlichen Fakultät angenommen. Der Dekan / Die Dekanin: Bern, 1.11.2011 Prof. Dr. Silvio Decurtins Table of contents TABLE OF CONTENTS Chapter 1 5 Introduction Chapter 2 25 Summary of chapters and synthesis Chapter 3 53 The evolutionary diversification of parrots supports a taxon pulse model with multiple trans-oceanic dispersal events and local radiations Chapter 4 67 Macroevolutionary patterns in the diversification of parrots: effects of climate change, geological events and key innovations Chapter 5 95 Nectarivory in parrots is a key innovation that triggered parallel adaptations and species pro- liferation through a nonadaptive radiation Chapter 6 133 Out of the Bassian province: historical bio- geography of the Australasian platycercine parrots Chapter 7 159 Phylogeny and biogeography of the parrot genus Prioniturus (Aves: Psittaciformes) Chapter 8 181 Disparity versus diversity and the role of ecological opportunity in the evolution of Neotropical parrots Acknowledgements 207 Erklärung 209 Curriculum vitae 211 3 Chapter 1 Chapter 1 INTRODUCTION In writing this PhD thesis I have attempted to gain an understanding of biogeographic and macroevolutionary patterns in the diversification of a particular species-rich avian group, namely parrots (Aves: Psittaciformes), by integrating mainly molecular genetic, but also morphological data in an ecological and historical context. A well resolved phylogenetic hypothesis is the prerequisite for predictions about evolutionary diversification processes and mechanism in a particular group. Therefore, the reconstruction of the phylogeny of the major groups of parrots provided the basis for this work. Biodiversity in the sense of species richness is neither randomly nor uniformly distributed on earth with some regions being much more diverse than others and different regions being inhabited by distinct species assemblages (Myers et al. 2000; Mittelbach et al. 2007; Gotelli et al. 2009; Lomolino et al. 2010). Moreover, species richness is highly unevenly distributed among taxonomic groups (Newton 2003; Ricklefs et al. 2007; Soltis 2007; Alfaro et al. 2009). Since the beginning of modern science, biologists have searched for explanations for these discernible spatial and taxonomic patterns of contemporary global diversity. The study of processes that generate and maintain biological diversity is still a major research issue at the interface not only between micro- and macroevolution, but also between different scientific disciplines such as evolutionary biology, ecology, paleontology, geology or climatology. One important goal is to infer the influence of ecological and evolutionary mechanisms on the relationships between time, dispersal and diversification (Wiens 2011). Understanding this is not only important in a historical context, but it is also prerequisite to predicting ecological changes and ecosystem function in the context of ongoing human-induced mass extinction (Ricklefs 1987). However, the timing and rate of diversification and their causes are still particularly poorly understood or controversially discussed for different organismic groups (e.g. Penny and Phillips 2004; Weir 2006; Bininda-Emonds et al. 2007; Brown et al. 2008; Nishihara et al. 2009). Species richness is basically the results of speciation, extinction and dispersal (Wiens 2011). These processes are dependent on time (e.g. clade age) or geographical area, but also influenced by historical events like continental drift, climate change or mountain orogenesis, by biotic factors like competition or ecological limits and intrinsic factors such as the potential to colonize new areas or to adapt to new ecological conditions (Hunter 1998; Price 2008; Thomas et al. 2008; Rabosky 2009; Lomolino et al. 2010; Yoder et al. 2010; Wiens 2011). Hence, diversity is the result of interactions between properties of the organisms themselves and their environment (Newton 2003). Diversification is promoted among other things by vicariance events, range expansions and ecological opportunities subsequent to habitat change, the colonization of new areas or the evolution of key innovations. In particular, evolutionary key innovations have often been considered as essential for promoting diversification (Heard and Hauser 1995; Hunter 1998). All these processes play an important role in different parts of this thesis and will be briefly introduced here. 7 Evolutionary diversification and biogeography of parrots Speciation and Extinction Several species concepts have been proposed to categorize the discernible discontinuities among groups of organisms observed in nature (e.g. Helbig et al. 2002). A discussion of these is clearly beyond the scope of this introduction. The so- called biological species concept puts the focus on reproductive isolation among groups of populations (Mayr 1942) and research on speciation has mainly focused on mechanisms leading to such reproductive isolation (Coyne and Orr 2004; Price 2008). It is now widely accepted that species arise in many cases by means of natural selection (Price 2008; Schluter 2009). The process of speciation can be viewed under different categorizations. One considers speciation in a geographical context, and a common classification scheme ranges from allopatric over parapatric to sympatric speciation (Price 2008). The formation of two species from geographically separated diverging populations is called allopatric speciation. It can proceed via vicariance when the range of an ancestral species is geographically separated by a barrier, e.g. through climatic or geographical events (Coyne and Orr 2004, see also below). Another mode of allopatric speciation is called peripatric speciation. In this case, the separation of an ancestral range of a species is reached through a founder event by the colonization of a new area (e.g. islands) usually by a small subset of the ancestral species population (Coyne and Orr 2004; Lomolino et al. 2010). Speciation can also take place in spatially overlapping populations (Coyne and Orr 2004). It is called sympatric speciation, when overlap between ancestral populations is extensive and interbreeding is possible throughout the process. In the case of parapatric speciation, however, areas of ancestral populations only partially overlap, making initial interbreeding possible (Coyne and Orr 2004; Price 2008; Lomolino et al. 2010). Another classification of the speciation process is concerned with the mechanisms and processes responsible for the evolution of reproductive isolation (Rundle and Nosil 2005; Price 2008). In this context, a distinction is made between ecologically-based and non-ecological or mutation-order processes that generate reproductive isolation (Rundle and Nosil 2005; Schluter 2009). In non-ecological processes, chance is thought to play an important role in the achievement of reproductive isolation through processes of genetic drift, founder events and population bottlenecks, hybridization or polyploidization (Coyne and Orr 2004; Rundle and Nosil 2005). Nevertheless, selection may be involved such as in speciation through some modes of sexual selection or the fixation of incompatible alleles (Schluter 2001; Rundle and Nosil 2005). However, this selection is not directly related to the environment, and fixed distinct mutations can be advantageous in both environments of the initial populations (Coyne and Orr 2004; Schluter 2009). In contrast, reproductive isolation can evolve directly or indirectly through divergent ecologically-based selection on traits between populations in contrasting environments (Schluter 2001; Rundle and Nosil 2005). This process is referred to as ecological speciation. It is a common mechanism by which new species arise, and strong connections between reproductive isolation and selection on phenotypic traits are often found (Schluter 2009). Ecological speciation can proceed in any geographical context (e.g. in sympatry or allopatry) and divergent ecologically-based selection can be triggered by environmental differences between populations, by differences in mate preferences between environments as well as by ecological interactions between populations that are basically sympatric (Rundle and Nosil 2005). Extinction can be considered to be the antagonistic process to speciation. The balance between speciation and extinction is expressed in diversification rates (e.g. Rabosky 8 Chapter 1 2009) and the net diversification rate is the difference between the rates of speciation and extinction. In the context of the current human-induced biodiversity crisis, extinction is usually perceived negatively. However, extinction is the final evolutionary