Comprehensive taxon sampling and vetted fossils help clarify the time tree of shorebirds (Aves, Charadriiformes)

Abstract

Shorebirds (Charadriiformes) are a globally distributed clade of modern birds and, due to their ecological and morphological disparity, a frequent subject of comparative studies. While molecular phylogenies have been instrumental to resolving the suprafamilial backbone of the charadriiform tree, several higher-level relationships, including the monophyly of plovers (Charadriidae) and the phylogenetic positions of several monotypic families, have remained unclear. The timescale of shorebird evolution also remains uncertain as a result of extensive disagreements among the published divergence dating studies, stemming largely from different choices of fossil calibrations. Here, we present the most comprehensive non-supertree phylogeny of shorebirds to date, based on a total-evidence dataset comprising 336 ingroup taxa (89% of all extant species), 24 loci (15 mitochondrial and 9 nuclear), and 69 morphological characters. Using this phylogeny, we clarify the charadriiform evolutionary timeline by conducting a node-dating analysis based on a subset of 8 loci tested to be clock-like and 16 carefully selected, updated, and vetted fossil calibrations. Our concatenated, species-tree, and total-evidence analyses consistently support plover monophyly and are generally congruent with the topologies of previous studies, suggesting that the higher-level relationships among shorebirds are largely settled. However, several localized conflicts highlight areas of persistent uncertainty within the gulls (Laridae), true auks (Alcinae), and sandpipers (Scolopacidae). At shallower levels, our phylogenies reveal instances of genus-level nonmonophyly that suggest changes to currently accepted taxonomies. Our node-dating analyses consistently support a mid-Paleocene origin for the Charadriiformes and an early diversification for most major subclades. However, age estimates for more recent divergences vary between different relaxed clock models, and we demonstrate that this variation can affect phylogeny-based macroevolutionary studies. Our findings demonstrate the impact of fossil calibration choice on the resulting divergence time estimates, and the sensitivity of diversification rate analyses to the modeling assumptions made in time tree inference.