Phylogenomics establishes an Early Miocene reconstruction of reef vertebrate diversity

Oceans blanket more than two-thirds of Earth’s surface, yet marine biodiversity is disproportionately concentrated in near-shore habitats, especially coral reefs. Investigating the origins of the exceptional species diversity found on coral reefs is crucial for predicting how these ecosystems will respond to anthropogenic disturbances. Here, we use a genome-scale dataset to reconstruct the evolutionary history of the wrasses and parrotfishes (Labridae), which rank among the most species-rich and ecologically diverse lineages of reef fishes. We show that all major labrid clades experienced concurrent pulses of evolutionary innovation and rapid lineage diversification over a period of fewer than five million years during the Early Miocene. Analyses of historical biogeography, character evolution, and phenotypic diversification rates demonstrate that no single phenotypic innovation can explain this period of accelerated diversification in wrasses. Instead, modern labrid diversity is a mosaic of multiple concurrent adaptive and non-adaptive radiations that diversified ~20-15 million years ago. These results draw parallels to the evolutionary histories of many animal groups after the Cretaceous-Paleogene mass extinction, suggesting a marked change in marine biodiversity associated with the Miocene emergence of novel ecological opportunities that wrasses exploited. Our results corroborate recently reported fossil evidence for an Early Miocene extinction event in oceanic vertebrates that we tie to changes in coral reef faunal composition and ubiquity and support this period as a crucial time in the assembly of present-day oceanic ecosystems.

Methods
Please see the main text of the paper for details. Key permitting and tissue loaning information:

JT collected specimens under Colombian permit number 20182200001023 (Colombian National Natural Parks) and ANLA resolution 1070 28 (August 2015). We also thank Y.K. Tea for providing several cirrhilabrine tissue samples and for discussions related to this paper. We thank G.J Watkins-Colwell of the Yale Peabody Museum for assisting with DNA sample curation and transfer, J. Klunk from Arbor Bioscience for assistance with target capture sequencing, J. Hung for assistance with DNA extractions and quantification, S. Klanten, R. Robertson, and L. Liggins for providing several parrotfish and wrasse tissue samples, and collections staff at the Australian Museum, Museum Victoria, Te Papa Museum for assisting with tissue loans.