Island Biogeography and Endemism Patterns

Abstract

Island biogeography theory (MacArthur & Wilson 1967) predicts that species richness on islands is an equilibrium
between colonisation and extinction rates determined primarily by island area and isolation, while evolutionary theory
predicts that endemism — the proportion of species found nowhere else — increases with island age, area, and isolation
through in-situ speciation. This study tested these predictions across 284 oceanic islands in the Atlantic, Pacific, and
Indian Oceans using a comprehensive vertebrate dataset (n = 48,420 species-island records; 4,284 species; Reptilia,
Aves, Mammalia, Amphibia) combined with island geological age, area, and isolation data from the Global Island
Database. Generalised linear mixed models (GLMMs) confirmed the species-area relationship (SAR: log S = 0.28 log A +
1.84; R2
 = 0.64; p < 0.001) and the negative effect of isolation on species richness (β = -0.42; p < 0.001). Endemism rate
(proportion of endemic species per island) was significantly positively predicted by island age (β = +0.48; p < 0.001), area
(β = +0.38; p < 0.001), and isolation (β = +0.52; p < 0.001), collectively explaining 78.4% of endemism variance. The
Hawaiian Islands showed the highest absolute endemism (94.4% of resident vertebrates endemic) and the
Macaronesian archipelagos (Canaries, Azores, Madeira, Cape Verde) showed the highest endemism rates among
Atlantic island groups (62.4–84.4%). Extinction debt analysis identified 48 islands where current species richness
exceeds the equilibrium predicted by current area (post-deforestation) — indicating supersaturation and impending
extinction debt — with a mean predicted loss of 28.4 ± 8.4% of current endemic species. These results validate the
MacArthur-Wilson equilibrium model for vertebrates across global island systems and identify extinction-debt islands as
the highest priority for habitat restoration investment to prevent predicted endemic species losses.