Syntopic diet divergence inferred from metabarcoding in a pair of parulid warblers: <i>Setophaga virens</i> (Black-throated Green Warbler) and <i>Seiurus aurocapilla</i> (Ovenbird)

Abstract The role of diet in driving ecological differences across the radiation of parulid wood warblers has been a topic of substantial debate. The inferences made in Robert MacArthur’s original study of their niche differences relied heavily on observations from microhabitat partitioning. How these different foraging behaviors translated to distinct diets and, more specifically, individual prey items has been less clear. Here, we use fecal metabarcoding of 2 syntopic insectivorous warblers—Setophaga virens (black-throated Green Warbler) and Seiurus aurocapilla (Ovenbird)—to complement and expand previous work, and we address past limitations by densely sampling many birds at the same location over a short period of time. We found highly significant differences in diet composition using several multivariate measures of diversity. In an analysis of individual diet proportions of insect orders, S. aurocapilla consumed more beetles and flies (Coleoptera and Diptera), whereas S. virens consumed more “true bugs” (Hemiptera) and stoneflies (Plecoptera). At the arthropod species level, we found that both warblers readily consumed invasive spongy moths (Lymantria dispar), and we identified 9 other arthropod species that significantly differed between the warblers. Of those, for 3 spider taxa, we combined warbler diet information with observations from arthropod collections and showed that spiders, which were more likely to be encountered on the ground, were exclusively eaten by S. aurocapilla whereas those encountered in the canopy were more likely to be consumed by S. virens, fitting with the expected vertical foraging stratification of the warblers. We interpret these diet differences as likely due to these 2 warbler species “opportunistically” encountering different arthropod assemblages in distinct foraging strata as opposed to “preferentially” consuming different prey. Our research emphasizes the benefits of extending analyses to more distantly related taxa—beyond those considered by MacArthur—and suggests a need for similar fine-scale studies within genera to enhance our understanding of dietary dynamics.