The multiple scales of biodiversity

The claim that measures of global biodiversity dynamics are meaningless is based upon several methodological problems, including underrepresentation of tropical regions in ‘‘global’’ Phanerozoic data sets, inaccuracies in taxonomic data, non-equivalence of higher taxa among groups of organisms, and uneven sampling intensity across groups, environments, and time intervals. Some of these problems are inherent in the fossil record, whereas others lie in documentation and interpretation of the subject. But the subject of global biodiversity is perfectly legitimate, even if problems persist in evaluating its full history. Moreover, recognition of the methodological problems has resulted in notable improvements in the Phanerozoic diversity database (e.g., Adrain and Westrop 2000; Alroy et al. 2001). Here I offer three points: (1) The study of biodiversity is meaningful at many spatial and temporal scales, including the global scale. (2) Two major preservational biases limit the estimation of global biodiversity over the Phanerozoic, although estimates are still potentially meaningful. (3) Extrapolation of patterns and processes from a smaller scale to a larger one or from one clade to a more inclusive clade is inappropriate and justifies some criticisms of biodiversity studies at the largest scales. Below, I elaborate on these points from a perspective based in experience with vertebrates in terrestrial ecosystems.