Nature (2012 Jun 24) doi: 10.1038/nature11226

The larger the region, the more species live in it. This is the so called species-area relationship and it is one of the fundamental relationships in ecology. In their recent study published in Nature, researchers from the Charles University in Prague (Czech Republic) and Yale University (USA) have shown that, across whole continents, the species-area relationship is very different from what we had known from small areas. More surprisingly, the relationship is universal in a similar way to the universality of laws of physics. These findings will play an important role in estimations of total World's biodiversity as well as in estimations of rates of species extinctions with habitat loss.

In the time of ecological crisis it is crucial to know the natural laws and patterns in the global distribution of biodiversity. Perhaps the most important biodiversity pattern is the relationship between size of a region and number of species which live there - the so called species-area relationship. Traditionally, species-area relationship has been approximated by a simple power function (i.e. a straight line if both the area axis and species richness axis are expressed in a logarithmic scale). It has been used for extrapolations of diversity across spatial scales or to estimate numbers of species that will go extinct after a given area is destroyed. It used to be assumed that if we for example know the size of an area of a tropical rainforest which has been destroyed, we can use the species-area relationship to estimate the proportion of species which went extinct there.
In reality, the relationship is more complex than a simple power law and it also differs across taxa and regions, which would always complicate its usefulness.
It turns out that the situation is not as complicated. David Storch from Center of Theoretical Study and Department of Ecology, Faculty of Science, Charles University in Prague, together with Petr Keil and Walter Jetz from Yale University examined the distribution of all species of amphibians, birds, and mammals across all continental landmasses. They were surprised to find that the species-area relationship at these large scales follows simple and yet non-trivial rules. Instead of being linear in the logarithmic scale (that is power-law), it is upward-accelerating for all taxa and continents. Moreover, its curvature depends on mean species geographic range, so that taxa with smaller ranges - namely amphibians - reveal more prominent curvature and consequently higher slope of the relationship at large areas. When the authors expressed the area in units corresponding to mean species range of given taxon within given continent, all of the curves collapsed onto one universal relationship. The number of species for given area can thus be estimated using the knowledge of mean species richness for some given area, with only one additional information - mean range size of given taxon within the region.
Yet another surprise emerged when Storch et al. looked at the relationship between area and the number of species which are restricted exclusively to this area, i.e. which are endemic to it (the so called endemics-area relationship). These species are particularly relevant for extinction estimates as they will become globally extinct if the area is destroyed. The authors have shown that, at continental scales, it is the endemic-area relationship that follows a simple power law, so that the number of endemic (and thus potentially extinct) species is roughly proportional to the potentially destroyed area, indicating high risk of extinction from area loss.

Universal species-area and endemics-area relationships at continental scales (Nature)