RESEARCH PAPER © 2003 Blackwell Publishing Ltd. http://www.blackwellpublishing.com/journals/geb Global Ecology & Biogeography (2003) 12, 5 – 12 Blackwell Science, Ltd The coincidence of people and biodiversity in Europe MIGUEL B. ARAÚJO*† *Centre d’Ecologie Fonctionnelle et Evolutive, Centre National de la Recherche Scientifique, Montpellier, France, and †Macroecology and Conservation Unit, University of Évora, Portugal; E-mail: mba@uevora.pt ABSTRACT A positive correlation between human population density and species richness has been recorded across the tropics. Here I investigate whether this correlation holds true for Europe. Ana- lyses reveal a positive correlation between human population density and plant (rho = 0.505), mammal (rho = 0.471) and reptile and amphibian (rho = 0.556) species richness. The results are largely concordant with those obtained in similar studies for Africa. However, contrary to previous analyses, the corre- lation found between people and breeding bird species richness (rho = 0.186) was weak. Of three measures of endemism used, only combined European endemic species richness correlated with human density (rho = 0.437). Richness among combined restricted-range European endemics was not correlated (rho = 0.095) with human density, while richness among all combined restricted-range species was only weakly correlated with human density (rho = 0.167). The results partially support the idea of a correlation between people and biodiversity, although there are some important exceptions. Discussion of possible mechanisms underling the observed patterns is undertaken. Key words biodiversity, conservation priorities, endemism, Europe, human population density, species richness. INTRODUCTION The processes that generate diversity have been discussed in the literature by researchers with different perspectives (for reviews see, e.g. Huston, 1994; Rosenzweig, 1995; Maurer, 1999; Gaston & Blackburn, 2000). A number of patterns have been identified as well as competing hypotheses to explain them. Teasing apart mechanisms underlying an observed pattern, from indirect factors that correlate with it, is often difficult. For example, recent studies have shown that human popula- tion density was positively correlated with terrestrial verte- brate and plant species richness across sub-Saharan Africa (Balmford et al., 2001a,b). These observations coincide with those of Fjeldså & Rahbek (1998), who noted a tendency for human settlements to be located near the centres of bird endemism in the tropical Andes. Considered altogether these studies imply that areas preferred by humans might coincide with those preferred by many extant species. The mechanisms underlying such pattern of congruence are not fully under- stood and it is likely that the observed relationship between human density and biodiversity are due to covarying factors, such as climatic stability (e.g. Fjeldså & Rahbek, 1998), pro- ductivity (e.g. Waide et al., 1999), or landscape heterogeneity (e.g. Atauri & Lucio, 2001; van Rensburg et al., 2002). However, if human density was indeed related directly to biodiversity, then the relationship should reoccur with analyses of data from different parts of the world. In this study, I investigate the extent to which a positive correlation between human popu- lation density and plant, mammal, bird, reptile and amphibian species richness occurs across Europe. Local and regional deviations from a broad pattern of covariation between these variables are also explored. MATERIALS AND METHODS Species data included 868 960 occurrence records of Euro- pean higher plants and terrestrial vertebrates. These included 2294 plant species ( Jalas & Suominen, 1972–96), 186 mammal species (Mitchell-Jones et al., 1999), 440 breeding bird species (Hagemeijer & Blair, 1997) and 143 amphibian and reptile species (hereafter referred to as herptiles) (Gasc et al., 1997). Terrestrial vertebrates include all extant European species, whereas plants comprise ∼20% of the European flora. The grid is based on the Atlas Florae Europaeae (Lathi & Lampinen, 1999), with cell boundaries typically following the 50 km lines of the Universal Transverse Mercator (UTM) grid, except near the border of the six-degree UTM zones and at coasts (for details concerning data conversion see Williams et al., 2000). Human population density surfaces were created in two stages, the first involved generating high-resolution Correspondence: Miguel B. Araújo, Centre d’Ecologie Fonctionnelle et Evolutive, CNRS, 1919 Route de Mende, 34293 Montpellier Cedex 5, France. E-mail: mba@uevora.pt