**3.4. Integration of a historical biogeography perspective with modern phylogenetic approaches and ecological studies**

In addition to this, climate change might act as a trigger of species diversification after a dispersal or vicariance event. For instance, many Neotropical diversification have been attributed to quaternary climatic cycles [77], where cooling phases in the Amazon basin facilitated geographic spread of highland species toward lowlands with subsequent range fragmentation and isolation during warm phases [77–79]. In the context of studies linking biogeographical events and diversification, very few studies have been able to detect a positive correlation between past climatic changes and species (or phenotypic) diversification after a given dispersal or vicariance event. For instance, Gamble et al. [80] suggested that main events of cladogenesis in *Gonatodes* geckos coincided with a phase of climate cooling in the late Eocence and Oligocene. In addition, Antonelli et al [81] suggested that speciation in an orchid South American group was accelerated by climate cooling after the Middle Miocene Climatic Optimum, 15 million years ago. However, these two studies did not evaluate whether specific shifts in net diversification rates matched temporally with significant decreases in temperature (i.e., cooling) in each region. Although these two studies suggest that climate change might be a trigger of species diversification, they did not mention how cladogenesis can be affected by climate changes, particularly cooling phases. The link between climate change and diversification is still narrative and more research is necessary to establish first whether there is possibility to find a positive association between both phenomena and then to establish the mechanisms driving speciation by climate change oscillations. For the first part, some recently developed models are able to detect whether paleo-environmental variations (e.g., temperature fluctuations through Cenozoic; [82]) had a significant effect on speciation and extinction rates [43, 83, 84]. These new models allow to establish whether the association is strong and how effectively climate can influence speciation and extinction rates for many

Summarizing, both dispersal and vicariance scenarios can promote shifts in net species diversification rates in contrasting ways. In dispersal scenarios, it is expected as an early ecological niche differentiation through the lineage diversification of co-occurring species, as has been the case in Caribbean *Anolis* lizards [29] and Hawaiian spiders [17]. In vicariance scenarios, it is expected, that co-occurring species exhibit ecological niche conservatism in trait-niche axes, either in coarse-grain or fine-grain scales, or that these trait-niches likely evolved by a purely drift process. If ecological niche conservatism prevails after allopatric speciation, it is merely a by-product of the ancestral range fragmentation and therefore it is not possible that it promotes speciation (*contra* [25]). Otherwise, if there is some degree of niche differentiation between species this is likely due to either neutral evolution [6] or adaptation to a slowly changing environment [22]. The distinction between these two patterns (i.e., neutral versus adaptive niche evolution) will be crucial although it can be difficult to be detected using only phylogenetic comparative methods [44]. Furthermore, it is very likely that other ecological or non- ecological processes facilitating speciation occur in each one of these biogeographic scenarios [12]. This suggests that reproductive isolation between subpopulations as by-product of these processes (e.g., polyploidization, hybridization, and genetic drift) is not necessarily driven by historical dispersal or vicariance events. In addition, it is important to remember that inferring an early dispersal event (or vicariance) does not preclude that subsequent cladogenesis occurred by a successive series of vicariance events (or dispersals). For instance,

groups [85].

22 Pure and Applied Biogeography

An integration of evidence from multiple study fields is necessary to get a deep understanding about the underlying mechanisms of the link between historical biogeography and diversification. Therefore, it will be crucial to develop and integrate new methodological approaches to generate sound evidence of a link between historical biogeographic events and changes in macroevolutionary dynamics. I consider that the first step necessary is to adopt a phylogenetic perspective to study the historical biogeography of a taxa and infer dispersal or vicariance events through its evolutionary history. Many studies reviewed have adopted such perspective, particularly implementing more complex event-based methods [60, 61]. New statistical biogeographical methods incorporating increasingly complex models of range evolution are being developed [34, 58]. The next step is to develop increasingly statistical robust methods to detect shifts in diversification dynamics and evaluate whether these shifts coincide with past dispersal or vicariance events and are not confounded by the emergence of hidden innovation key traits [53, 75]. Although there is a current debate about the power of these methods to detect these shifts [42, 53, 55], the BAMM approach ([53]) seems promising to detect these shifts in diversification dynamics across a phylogenetic tree with high confidence. BAMM allows us not only to detect shifts in speciation rates but shifts in the diversification dynamics itself. For instance, BAMM could allows us to detect in which nodes the diversification dynamics follows a time-dependent process, where only the time for speciation influences speciation rates [69]; or whether it follows a density-dependent cladogenetic process, where the clade diversity is limited by a carrying capacity [51]. Even more, using BAMM we could establish whether a macroevolutionary regime is more prone to occur.

In addition to the statistical detection of these historical association, I suggest that it is necessary to test the specific predictions that might emerge from each biogeographical scenario. I outlined two specific predictions for phenotypic trait evolutionary trajectories after a dispersal or vicariance event (see above). In that case, it is necessary to collect information about ecologically relevant traits for species to test these predictions. This information can be collected from museum specimens (e.g., morphological traits) or field-based studies (e.g., habitat use, diet or foraging strategy). Using model-based phylogenetic comparative methods, it is possible to evaluate whether these traits evolved following a Brownian motion or Ornstein-Uhlenbeck model [20] or whether traits evolved under a pure drift process lacking its phylogenetic signal (e.g., a white-noise process [20]). Furthermore, recently developed methods allow to identify shifts in phenotypic trait diversification in nodes of a phylogenetic tree [14, 15, 53, 59, 67]. In these cases, it is possible to evaluate simultaneously whether specific shifts in species or trait diversification evolved as a response to a dispersal or vicariance event. I consider that testing these predictions allows us to establish potential ecological and evolutionary mechanisms driving biological diversity. Some recent studies have begun to adopt a combination of these new methodologies (e.g., Ref. [24]), but still more research is necessary.

Finally, studies undoubtedly need to include detailed research from the field of the genetics of the adaptation to new environments [8, 22]. These studies are useful to understand why some clades diversified extensively after a dispersal or vicariance events and others not. For instance, species widely distributed in archipelagos or species with disjunct distributions will be good candidates for these detailed studies. Information about mechanisms maintaining or disrupting genetic cohesion between populations or lineages is crucial to understand the role of biogeographical events in the generation of biodiversity.
