Mesoamerica Is a Cradle and the Brazilian Atlantic Forest Is a Museum of Neotropical
Total Page:16
File Type:pdf, Size:1020Kb
bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 1 Title page 2 3 Title: Mesoamerica is a cradle and the Brazilian Atlantic Forest is a museum of Neotropical 4 butterfly diversity (Lepidoptera: Nymphalidae: Brassolini) 5 6 Authors: Pável Matos-Maraví1,2,3*, Niklas Wahlberg4, André V. L. Freitas5, Phil J. DeVries6, 7 Alexandre Antonelli1,2,7, Carla M. Penz6 8 9 Full correspondence addresses: 10 1 Department of Biological and Environmental Sciences, University of Gothenburg, Gothenburg, 11 Sweden 12 2 Gothenburg Global Biodiversity Centre, Gothenburg, Sweden 13 3 Biology Centre of the Czech Academy of Sciences, Institute of Entomology, České 14 Budějovice, Czech Republic 15 4 Department of Biology, Lund University, Lund, Sweden 16 5 Departamento de Biologia Animal, Instituto de Biologia, Universidade Estadual de Campinas, 17 Campinas, SP, Brazil 18 6 Department of Biological Sciences, University of New Orleans, New Orleans, LA, USA 19 7 Royal Botanical Gardens Kew, TW9 3AE Richmond, United Kingdom 20 21 * Corresponding author (E-mail: [email protected], ORCiD id: https://orcid.org/0000- 22 0002-2885-4919) 23 1 bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 24 Abstract 25 Regional species diversity is ultimately explained by speciation, extinction, and dispersal. Here 26 we estimate dispersal and speciation rates in Neotropical rainforest biomes to propose an 27 explanation for the distribution and diversity of extant butterfly species. We focus on the tribe 28 Brassolini (owl butterflies and allies): a Neotropical group that comprises 17 genera and 108 29 species, most of them endemic to rainforest biomes. We infer a total-evidence species tree using 30 the multispecies coalescent framework. By applying biogeographical stochastic mapping, we 31 infer ancestral ranges and estimate rates of dispersal and cladogenesis at the scale of millions of 32 years. We suggest that speciation in Mesoamerica and the northwestern flank of the Andes have 33 only increased within the past 2 million years. In contrast, speciation in the Brazilian Atlantic 34 Forest has been constant throughout the past 10 million years. The disparate species 35 diversification dynamics may be partly explained by the geological and environmental history of 36 each bioregion. Importantly, the dispersal rates into the Atlantic Forest and Mesoamerica plus 37 NW Andes increased simultaneously in the middle-Miocene, suggesting that lineages from such 38 regions have had comparable times for speciation despite their decoupled diversification 39 dynamics. Diversification of extant Amazonian lineages, on the other hand, has episodically 40 increased since the late Miocene, including a rise in speciation rate during the Pleistocene. 41 Altogether, our results reveal a mosaic of biome-specific evolutionary histories within the 42 Neotropics, where species have diversified rapidly (cradles: e.g., Mesoamerica), have 43 accumulated gradually (museums: e.g., Atlantic Forest), or have alternately diversified and 44 accumulated (e.g., Amazonia). 45 2 bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 46 Keywords: Speciation; dispersal; multispecies coalescent; total-evidence analyses; ancestral 47 state inference; biogeographical stochastic mapping. 48 3 bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 49 Introduction 50 Terrestrial biomes are heterogeneous, and some areas accumulate greater diversity and endemic 51 species than others. An explanation for this biogeographical pattern includes three evolutionary 52 mechanisms: species origination, extinction, and dispersal (Goldberg et al. 2005, Jablonski et al. 53 2006). For example, a species-rich region can attain its present diversity by exceptionally high 54 rates of speciation, becoming a “cradle of diversity”, or by the persistence of old lineages, e.g., 55 via low species extinction rates, becoming a “museum of diversity” (Stebbins 1974). 56 Furthermore, a species-rich region might be a “source of diversity” when dispersal out of the 57 region is high, or a “sink of diversity” when regional species diversity accumulates from external 58 sources by dispersal into the region. Characterizing the interplay among speciation, extinction, 59 and dispersal is thus highly relevant for understanding the origin and evolution of total regional 60 diversity and endemism (Wiens and Donoghue 2004, Roy and Goldberg 2007). 61 62 The Neotropical region contains a great deal of the world‟s biodiversity and endemism. Most of 63 such diversity is concentrated in tropical rainforests in Central America, the Amazon basin, and 64 the Atlantic Forest in southeastern Brazil. However, there is no consensus on whether the 65 currently scattered rainforests were once connected, nor for how long (Jaramillo and Cárdenas 66 2013). It is also unclear whether rainforests represent centers of “ecological stability” that share a 67 common evolutionary history (Moritz et al. 2000). Mammals and birds in Neotropical 68 biodiversity hotspots, including the Brazilian Atlantic Forest, and rainforests in Mesoamerica 69 and the neighboring area Chocó (Myers et al. 2000) apparently originated at faster rates and 70 exported more lineages compared to surrounding areas, suggesting that such hotspots are both 71 cradles and sources of diversity (Igea and Tanentzap 2019). Some non-hotspots regions are, in 4 bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 72 contrast, museums of diversity given the old and gradual accumulation of species through time, 73 such as Amazonia for Troidini butterflies (Papilionidae) (Condamine et al. 2012). It thus seems 74 that the checkered spatial assemblage of rainforests acting as cradle and museum areas might 75 explain why the Neotropics is considered both museum and cradle of diversity (McKenna and 76 Farrell 2006, Moreau and Bell 2013, Antonelli et al. 2018a). However, the impacts of geological 77 and paleoenvironmental changes in the Neotropics together with the roles of speciation and 78 dispersal have seldom been jointly elucidated in individual rainforest hotspots. 79 80 Here we test whether there was significant biotic exchange among Neotropical rainforests during 81 the Neogene and Quaternary periods (i.e., the past 23 Ma), which might indicate physical 82 connectivity of biomes. If connection among biomes existed, we aim to quantify its timing and 83 magnitude in terms of dispersal and speciation of rainforest taxa. To do this, we estimate 84 dispersal rates and speciation of Neotropical butterflies that occur mainly in rainforest hotspots, 85 such as Mesoamerica, the northwestern slope (NW) of the Andes (including e.g., Chocó), the 86 Atlantic Forest, Amazonia, and the South American “dry diagonal” (Fig. 1). Butterflies in the 87 tribe Brassolini (Nymphalidae: Satyrinae) are particularly suited to address our question because 88 they are a monophyletic group (Freitas and Brown 2004, Wahlberg et al. 2009, Espeland et al. 89 2018, Chazot et al. 2019a), exclusively Neotropical, and have about one third of their species 90 diversity endemic to rainforest hotspots (see Penz 2007 for an overview). 91 92 Studies that focused on individual brassoline genera used adult morphology (Penz 2008, 2009a, 93 b, Garzón-Orduña and Penz 2009) and DNA data (Penz et al. 2011a, Shirai et al. 2017) to infer 94 species-level phylogenies, but several deep nodes had weak or even conflicting phylogenetic 5 bioRxiv preprint doi: https://doi.org/10.1101/762393; this version posted March 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. 95 support in tribal-level analyses (Penz et al. 2013). We infer a time-calibrated species phylogeny 96 of Brassolini by merging all previous datasets and generating new DNA and morphological data. 97 Specifically, we use this phylogeny to revisit and test the following predictions on diversification 98 and dispersal among three Neotropical rainforest biomes: 99 100 1. Amazonia is the main source of Brassolini diversity by means of dispersal, reflecting the 101 patterns from Neotropical vertebrates and plants during the past 60 million years (Antonelli et al. 102 2018b). Rapid species diversification of some butterfly groups occurred there since at least the 103 Oligocene (Matos-Maraví 2016), and old lineages might have gradually accumulated in 104 Amazonia since that time (Condamine et al. 2012). More recent diversification episodes 105 supporting Amazonia as alternating museum and cradle of diversity could have been triggered by 106 paleoenvironmental changes, such as the retreat of the Pebas wetland system (Antonelli et al. 107 2009, Antonelli and Sanmartín 2011, Chazot et al.