On the Sangihe and Talaud islands of North Sulawesi, where forest cover has declined since 2010, communities look thinner and more uniform. The yellow-sided flowerpecker (Dicaeum aureolimbatum) and gray-sided flowerpecker (Dicaeum celebicum) on Sangihe both belong to the same genus, as do the elegant imperial pigeon (Ducula concinna) and gray imperial pigeon (Ducula pickeringii) on Talaud. Closely related species tend to forage together, a pattern the authors link to environmental filtering, the process by which local conditions favor particular lineages.

On Buton Island in Southeast Sulawesi, the team found an additional layer. The Sulawesi drongo (Dicrurus montanus), black sunbird (Leptocoma sericea) and yellow-sided flowerpecker all belong to the large order Passeriformes, whose presence reflects Buton's proximity to Sulawesi across shallow straits and the food resources the island offers. Karst and ultrabasic substrates (limestone and magnesium-rich soils) further shape the vegetation, favoring specific bird groups.

Why do evolutionarily distinct species matter?

The team applied the EDGE metric (Evolutionarily Distinct and Globally Endangered, a score that combines how isolated a species sits on the tree of life with its extinction risk) alongside the International Union for Conservation of Nature Red List. One standout: the Sulawesi maleo (Macrocephalon maleo), an endangered bird whose genus contains no other surviving species. Loss of the maleo would erase the last representative of an ancient branch.

Nearly half of all species in the maleo's wider family, the Megapodiidae (a group of ground-nesting birds found from Sulawesi to the Pacific), are classified as vulnerable to critically endangered. Forests that host such lineages therefore carry disproportionate conservation weight.

What does this mean beyond islands?

The authors note parallels elsewhere. On Mount Cameroon, closely related bird groups dominate both low and high elevations, a pattern linked to the history of African forests as they repeatedly expanded, contracted and broke apart. Selective logging in the Brazilian Amazon appears to prune away parts of birds' evolutionary history, and the conversion of Colombian forest to pasture has been estimated to erase around 2.3 billion years of avian evolutionary history.

The study has limits. It covers seven islands, and the authors describe a follow-up analysis on isolation between islands that has not yet been published. Results therefore remain preliminary, and patterns may not extend uniformly across all tropical regions.

The team frames its work plainly: birdsong tells only part of a forest's story. To judge what a forest protects, they argue, one must also read the family tree behind the sound.