Macroscale estimates of species abundance reveal evolutionary drivers of biodiversity

Evolutionary processes underpin the biodiversity on the planet. Theories advocate that the form of the species abundance distribution (SAD), presented by the number of individuals for each species within an ecological community, is intimately linked to speciation modes such as point mutation and random fission. This prediction has rarely been, however, verified empirically; the fact that species abundance data can be obtained only from local communities critically limits our ability to infer the role of macroevolution in shaping ecological patterns. Here, we developed a novel statistical model to estimate macroscale SADs, the hidden macroecological property, by integrating spatially replicated multispecies detection-nondetection observations and the data on species geographic distributions. We determined abundance of 1,248 woody plant species at a 10 km grid square resolution over East Asian islands across subtropical to temperate biomes, which produced a metacommunity (i.e. species pool) SAD in four insular ecoregions along with its absolute size. The metacommunity SADs indicated lognormal-like distributions, which were well explained by the unified neutral theory of biodiversity and biogeography (UNTB) with protracted speciation, a mode of speciation intermediate between point mutation and random fission. Furthermore, the analyses yielded an estimate of speciation rate in each region that highlighted the importance of geographic characteristics in macroevolutionary processes and predicted the average species lifetime that was congruent with previous estimates. The estimation of macroscale SADs plays a remarkable role in revealing evolutionary diversification of regional species pools.