Quantifying the unrecorded loss of avian phylogenetic diversity

Humans have drastically reduced avian diversity, with the majority of extinctions occurring on islands. Previous studies have quantified various aspects of this decline, including both taxonomic and phylogenetic diversity loss due to recorded extinctions. Other studies have estimated that unrecorded island bird extinctions – those that left no known fossil evidence – may represent hundreds of additional losses. However, these analyses have only focused on species diversity. In this paper, we bridge these two research efforts by estimating the phylogenetic diversity lost due to unrecorded island bird extinctions. Our findings suggest that the loss of phylogenetic diversity may be substantially smaller than expected, given the number of extinctions. Our results suggest that while unrecorded extinctions probably represented around 60% of all species extinctions, the majority of the phylogenetic diversity loss was likely caused by the recorded extinctions. The reason for this is that while extant island endemics are on average slightly more phylogenetically distinct than expected by chance, a disproportionate number of unrecorded extinctions are predicted to have been from islands in the eastern Pacific. Extant birds from this region generally have lower phylogenetic distinctiveness than other birds and the extinct species therefore likely did as well.