Tropical lands harbour the highest number of species, resulting in the ubiquitous latitudinal diversity gradient (LDG). However, exceptions to this pattern have been observed in some taxa, explained by the interaction between the evolutionary histories and environmental factors that constrain species' physiological and ecological requirements. Here, we applied a deconstruction approach to map the detailed species richness patterns of Actinopterygian freshwater fishes at the class and order levels and to disentangle their drivers using geographical ranges and a phylogeny, comprising 77% (12 557) of all described species. We jointly evaluated seven evolutionary and ecological hypotheses posited to explain the LDG: diversification rate, time for speciation, species–area relationship, environmental heterogeneity, energy, temperature seasonality and past temperature stability. We found distinct diversity gradients across orders, including expected, bimodal and inverse LDGs. Despite these differences, the positive effect of evolutionary time explained patterns for all orders, where species-rich regions are inhabited by older species compared to species-poor regions. Overall, the LDG of each order has been shaped by a unique combination of factors, highlighting the importance of performing a joint evaluation of evolutionary, historical and ecological factors at different taxonomic levels to reach a comprehensive understanding on the causes driving global species richness patterns.
30 October 2021 in Journal of Zoological Systematics and Evolutionary Research
The distribution and diversification processes of freshwater fishes have been influenced by tecto-volcanic and climatic events that have in turn promoted genesis, destruction, and numerous changes in the drainage networks, leaving complex phylogeographic patterns in the ichthyofauna of highly dynamic regions such as central Mexico. In this study, we evaluated the evolutionary history of the Aztec shiner Aztecula sallaei, which is distributed in currently isolated drainages in the highlands of central Mexico, performing phylogenetic, biogeographic, and phylogeographic analyses using three genes: the mitochondrial cytb and two nuclear genes, S7 and ACTB. The studied populations show three genetically divergent and structured lineages: two are exclusive to the Panuco biogeographic region and the third is widely distributed in the Upper Lerma, Middle Lerma, Cuitzeo, and Balsas (Zempoala Lakes) biogeographic regions, as well as one population in Panuco. This widely distributed lineage also presents a genetic structure that is congruent with these biogeographic regions supported by the cytb gene and partially supported by the S7 gene. The divergence time of these lineages was dated to the Pleistocene period (0.56–2.79 Mya) and was associated with temporally or spatially independent events of river capture, which promoted the exchange of fishes between the river headwaters followed by subsequent isolation, explaining the current distribution of this cyprinid in neighboring but currently isolated basins. Our results indicate that, in the highlands of central Mexico, tecto-volcanic activity has shaped the geographic ranges, diversification, and phylogeographic patterns of freshwater fishes.
Understanding the drivers of biodiversity in the world’s rivers, which are known to be hyperdiverse relative to their coverage area, has been an enduring goal in ecology. While regression-based empirical studies have identified a suite of environmental factors that are correlated with riverine fish biodiversity, these insights are often system-specific and inconsistent across regions. In contrast, a more limited body of studies have suggested that network connectivity of rivers affects fish biodiversity by limiting dispersal, and basin shape may modulate these relationships. The few theoretical papers that have explored basin morphology effects have tended to use extreme network shapes and inconsistent methods, thus limiting general insights. Here, we build on these results to demonstrate that river basin morphology, as measured by log aspect ratio, can alter both network connectivity and biodiversity in simulated, all-else-equal scenarios. First, we quantify variation in log aspect ratio across the world’s 100 largest rivers and use this empirical range of shape variation to guide synthetic experiments. In particular, we use Optimal Channel Networks (OCNs) constrained to basins with log aspect ratios within realistic range to study how shape alters connectivity profiles when node number and basin area are held constant. By coupling OCNs with dendritic neutral models, we demonstrate that variation in aspect ratio and concomitant changes in connectivity lead to substantial changes in simulated biodiversity. Finally, we use Earth Mover’s Distance to establish that basin-shape-induced changes in node-level connectivity distributions are predictive of transformations in node-level distributions of α and β diversity. Overall, elongated basins such as the Mekong River feature lower species richness ( α -diversity), higher turnover ( β -diversity), and less variable distributions of both quantities relative to a square reference basin. Furthermore, approximately one third of the world’s largest rivers are elongated enough to potentially feature statistically-detectable, shape-mediated variation in connectivity and biodiversity.
Speciation rates are a key driver of diversity patterns and are often used to explain the latitudinal diversity gradient (LDG). However, latitudinal variation in speciation rates at both assemblage and species levels remains poorly explored in freshwater fishes. This highlights a gap in understanding the mechanisms driving geographic biodiversity gradients in freshwater fishes. Here, we investigated the latitudinal speciation gradient in freshwater fishes, using a comprehensive database of freshwater fish distributions and phylogenetic relationships of Actinopterygian fishes at the global scale. We estimated speciation rates using three metrics (BAMM, DR, and ClaDS) and evaluated the latitudinal speciation gradient through spatial and phylogenetic regressions at assemblage and species levels. Finally, we analyzed those patterns based on the species assemblage’s phylogenetic diversity and structure. Our results show that areas and species with the highest speciation rates were in the tropics. However, the general assemblage pattern revealed a positive relationship between absolute latitude and speciation rates. This relationship is generally absent in tropical regions below 24.39° and became significant only at higher latitudes, particularly in the Northern Hemisphere. We did not find a significant relationship at the species level, mainly due to the strong influence of hyper-diverse groups like Cichliformes. Species-level findings showed the contribution of particular lineages to the speciation gradient as a whole, while the assemblage-level results emphasize the high speciation rates across the Northern Hemisphere, especially North America, potentially resulting from environmental filtering and dispersal events consistent with glaciation dynamics in the Pleistocene.