Computer construction of species richness maps: Testing a new type of multifractal algorithm

We show how a new theoretical multifractal model provides means to generate virtual maps of highly variable spatial distributions of species richness. It should allow for various computer experiments in landscape ecology and the study of biodiversity. In this paper, the explicit distribution of species-representative individuals over a large range of scale leads to an original algorithm for the estimation of the Renyi dimensions of a multifractal measure. The method is successfully tested for simulated (S, A) data sets, where the variable S is simply the number of species found in a given domain of area A. This easy tool will help to characterize the spatial variability of multiscale density distributions in many fields, requiring only randomly sampled data at different locations and scales