Phylogenetic trees are increasingly used to infer the processes that shape biodiversity patterns, such as diversification, dispersal and trait evolution. However, collecting many samples and sequencing them for phylogenetic reconstruction is challenging and costly, and achieving high sampling fractions can be logistically impossible for species-rich groups or regions. When studying isolated environments such as islands, this issue applies to the species that make up the island community, as well as outgroup taxa when sampling from a large pool of mainland relatives is required. In this study, we use simulations of the island biogeography model DAISIE (Dynamic Assembly of Islands through Speciation, Immigration and Extinction) to investigate the best sampling strategy to minimize error in the estimation of the model parameters when there is a constraint on the number of species that can be sampled. We compare three different community-level sampling strategies for islands: (1) prioritizing species-rich island lineages; (2) prioritizing species-poor island lineages; and (3) random selection of island lineages. Furthermore, we explore the effect of the nature of the missing data within each (species-rich or species-poor) lineage by testing the impact of incomplete sampling of the oldest and youngest extant species within each lineage and assess the effect of excluding outgroup species or even entire island lineages. Parameter estimates of speciation, colonization and extinction rates of island lineages show slightly larger errors when the unsampled species belong to species-rich lineages. Within clades, we observed larger errors when the unsampled species were the oldest or outgroup species (i.e. mainland species sampled to determine the stem age of the clade). When sampling is limited by time and/or budget, our study suggests prioritizing sampling of the phylogenetically most distinct species from the most diversified island lineages along with their mainland relatives.

, , , , , ,
doi.org/10.1111/2041-210x.70058
Methods in Ecology and Evolution

Released under the CC-BY NC 4.0 (“Attribution-NonCommercial 4.0 International”) License

Staff publications

Dehayem, O., Brewer, R., Valente, L., Lens, F., & Etienne, Rampal S. (2025). Impact of sampling strategy on inference of community assembly processes in phylogenetic island biogeography. Methods in Ecology and Evolution, 2025. doi:10.1111/2041-210x.70058