Introduction
Comprising South America, Central America, tropical Mexico and the Caribbean Islands, the Neotropics are arguably the most biodiverse region on Earth. This region not only includes the largest tropical rainforest, Amazonia, but also eight of the 34 known biodiversity hotspots (Mittermeier et al. 2011). The tropical Andes, in particular, is considered the most species-rich biome in the world for amphibians, birds, and plants, Mesoamerica and the Caribbean Islands are the richest regions for squamates, while Amazonia is seen as the primary biogeographic source of Neotropical biodiversity (Antonelli et al. 2018c). The age and underlying causes of the extraordinary Neotropical biodiversity represent one of the most debated topics in evolutionary ecology (Haffer 1969; Simpson 1980; Gentry 1982), but the mechanisms behind its origin and maintenance remain elusive (Leigh Jret al. 2004; Hoorn et al. 2010; Antonelli & Sanmartín 2011; Rull 2011a).
A long-held tenet was that the outstanding levels of Neotropical diversity were mainly generated during the Quaternary (i.e. the past 2.6 million years [Myrs]), as the product of recent environmental fluctuations (Haffer 1969; Rull 2011a). However, this hypothesis was challenged by the age of the most recent common ancestors (MRCA) of Neotropical taxa (Hoorn et al. 2010) and fossil studies showing a fossil record significantly more diverse in the Eocene than posteriorly (Jaramillo et al. 2006). Although recent studies show that modern climate (Moritz et al. 2000; Rull 2011a; Rowanet al. 2019) and intrinsic biological traits (Smith et al.2014) could have contributed to explain the maintenance of Neotropical diversity during the last few million years, megadiversity levels in the Neotropics probably existed before the Quaternary. Under this understanding, short temporal scales might be limited to study this pattern and historical eco-evolutionary factors need be investigated to understand the origin and evolution of the Neotropical diversity.
For a long time, the long-term stability and large extension of the tropical biome across the South American continent and tens of million years of continental isolation have been argued as the main factors promoting the gradual accumulation of lineages in the Neotropics (Wallace 1878; Stebbins 1974; Simpson 1980). Yet, new studies gathered during the last decade suggest instead that long-term environmental instability – geological and environmental perturbations together with intermittent land connections, mostly during the Neogene – have been responsible for the outstanding Neotropical diversity (Mittelbachet al. 2007; Hoorn et al. 2010; Rull 2011a; Antonelliet al. 2018b).
Among all potential environmental perturbations, the orogenesis of the Andes and associated geomorphological and ecological modifications has become paradigmatic for explaining Neotropical biodiversity (Gentry 1982; Hughes & Eastwood 2006; Hoorn et al. 2010; Luebert & Weigend 2014; Antonelli et al. 2018a; Esquerré et al.2019). Andean uplift began in the Central Andes ~65 Myrs ago (Mya) as a result of subduction of the Nazca plate along the Pacific margin, and in the Northern Andes ~23 Mya, with the collision of the Pacific plate. Uplift then intensified ~12-4.5 Mya (Garzione et al. 2008; Hoorn et al. 2010; Chen et al. 2019). The Andean orogeny deeply affected regional climate, hydrological conditions and landscape evolution at a continental level, with increased eastern rainfall and sediment flux into Amazonia (Hoorn et al. 1995; Armijo et al. 2015). This process resulted in the modern configuration of the Amazon drainage basin and fluvial system less than 10 Mya, and contributed to the formation of the “dry diagonal” Caatinga-Cerrado belt (Blisniuket al. 2005; Hoorn et al. 2010, 2017). This tectonic rearrangement also led to the closure of the Central American Seaway ~10 Mya (Jaramillo 2018).
The dynamic landscape caused by the Andean uplift has been postulated to have a major effect on Neotropical diversification by (i)increasing habitat and environmental heterogeneity (today all major biomes appear in the region, e.g. tropical forests, deserts, and high elevation grasslands), (ii) favoring isolation and thus allopatric speciation in montane populations separated by deep valleys, lowland populations on either side of the emerging mountains, and Amazonian populations separated by new riverine barriers (Flantuaet al. 2019), and by (iii) creating the longest latitudinally-elongated corridor for biotic montane dispersal (Antonelliet al. 2009; Luebert & Weigend 2014). As such, Andean uplift has been associated with the explosive radiation of plants, insects, and tetrapods (Weir 2006; Drummond et al. 2012; Lagomarsino et al. 2016; Pérez-Escobar et al. 2017; Pouchon et al. 2018; Chazot et al. 2019; Esquerré et al. 2019), and with increased rates of biotic interchange (Santos et al. 2009; Fjeldså et al. 2012; Antonelli et al. 2018c; Baconet al. 2018).
Analyses of the fossil record also suggest that Neotropical diversity has been strongly linked to temperature (Hoorn et al. 1995; Jaramillo et al. 2006). Global temperatures were warmer during the Cretaceous and early Paleogene, a period punctuated with hyperthermal events, such as the Paleocene-Eocene Thermal Maximum (PETM) ~56 Mya. The late Paleogene marks the onset of a long-term cooling trend that was accelerated at the beginning of the Quaternary leading to the glaciation-interglaciation climate of the past 2.6 Myrs (Zachos et al. 2008; Veizer & Prokoph 2015). In South America, global cooling promoted the expansion of open habitats and the subsequent establishment of fire regimes in the Cerrado savannas (Simonet al. 2009; Antoine et al. 2013). These past climatic events are also thought to have shaped Neotropical diversification (Pinto‐Ledezma et al. 2017). Plant diversity inferred from fossil morphotypes increased with warming periods during the Eocene, and decreased sharply with subsequent cooling (Hoorn et al. 1995; Jaramillo et al. 2006). Quaternary glacial cycles have been considered to promote fragmentation in rainforest ecosystems and high-altitude habitats (e.g. through altitudinal shifts in Andean vegetation zones), which in turn contributed to geographical isolation and diversification (Haffer 1969; Rull 2011a; Flantua et al.2019). Diversification of Neotropical clades have also been attributed to more ancient climatic events, such as the PETM and the cooling event subsequent to the Middle-Miocene climatic optimum (MMCO) (Hughes & Eastwood 2006; Lagomarsino et al. 2016).
These alternative but non-exclusive models of diversification in the Neotropics have been difficult to tease apart empirically for two reasons. Firstly, there has been a lack of large-scale comparative data across wide phylogenetic and ecological contexts. Secondly, it has been challenging to develop environmentally explicit diversification models linking changes in the physical environment and species diversification (Condamine et al. 2013, 2019). Some studies have focused on particular Neotropical clades to infer their triggers of diversification (Lagomarsino et al. 2016; Esquerré et al. 2019), but given the vast heterogeneity of the region, general insights can only be provided if patterns of diversification are shared among Neotropical lineages. Here, we use a comparative phylogenetic data set containing 150 well-sampled species-level molecular phylogenies and 12,524 species. Our data set represents ~60% of all estimated tetrapods and ~7% of the known plant Neotropical diversity, which we use to evaluate the timing and drivers of Neotropical diversification at a continental scale. Our results reveal an ancient pre-Quaternary origin of Neotropical diversity, as well as striking effects of climatic and landscape changes on Neotropical diversification, with varying degrees of importance and effect depending on organismal biology and identity.