Positive regional species–people correlations: a sampling artefact or a key issue for sustainable development?

Many studies are documenting positive large-scale species– people correlations (Luck, 2007; Schuldt & Assmann, 2010). The issue is scale dependent: the local association of species richness and people is in many cases a negative one (Pautasso, 2007; Pecher et al., 2010). This biogeographical pattern is thus important for conservation. If species-rich regions are also densely populated, preserving biodiversity becomes more difficult, ceteris paribus, than if species-rich regions were sparsely populated. At the same time, positive, regional species–people correlations are an opportunity for the biodiversity education of the majority of the human population and underline the importance of conservation in human-modified landscapes (e.g. Sheil & Meijaard, 2010; Ward, 2010). In our study (Barbosa et al., 2010), we tested whether a sampling artefact applies to the species–people correlation for carabids in Italy at three spatial grains of analysis. As the commentaries by Luck (2010) and McKinney (2010) point out, there has been little consideration of variation in sampling effort in previous regional studies of the coexistence of biodiversity and people. If sampling effort is positively associated with both species richness and human population, it could explain the correlation between the two (Cantarello et al., 2010). We agree with Luck (2010) that there is a need to further study the correlation of biodiversity and people for taxa other than vertebrates and plants, as this pattern is now well-documented for such well-studied taxa (recently, e.g. Fjeldså, 2007; Pidgeon et al., 2007; Fjeldså & Burgess, 2008; Lepczyk et al., 2008; Moreno-Rueda & Pizarro, 2009). At the level of European countries, a positive species–people correlation was found for ants, aphids, grasshoppers and stream macro-invertebrates (Pautasso & Fontaneto, 2008; Schlick-Steiner et al., 2008; Steck & Pautasso, 2008; Pautasso & Powell, 2009), but the data did not allow the inclusion of sampling effort in those studies. Apart from invertebrates, species-rich groups such as bacteria and fungi have been almost completely neglected from the point of view of the large-scale correlation with people (Pautasso & Zotti, 2009; Nemergut et al., 2010). We concur with Luck (2010) andMcKinney (2010) that it is important not just to document whether or not regional species–people correlations are present and artefactual, but also to investigate potential mechanisms and to study how the issue fits with the spatial distribution of protected areas. Unfortunately, protected areas have tended to be located far away from human settlements, thus ending up disproportionately located in relatively species-poor regions (e.g. Pautasso & Dinetti 2009). Regardless of the mechanisms underlying such three-way relationships among biodiversity, people and parks, positive regional species–people correlations imply that biodiversity conservation has to happen over the entire landscape, not just in wilderness areas. At the same time, the few existing protected areas in urbanized landscapes need to be defended as a matter of priority, not swapped with cheaper land away from civilization. In addition, research priorities related to the large-scale correlation of species and people include moving from species richness to other measures of biodiversity (e.g. abundance, genetic and phylogenetic diversity; Knapp et al., 2008; Pautasso & Chiarucci, 2008; Noël & Lapointe, 2010). Further, we need to know how the pattern is likely to develop over the next decades under various demographical, climate change and assisted migration (both for species and people) scenarios (Hoymann, 2010; Pautasso et al., 2010). Ultimately, if the large-scale correlation of biodiversity and people will be generally found not to be artefactual and to be present also for biodiversity measures other than species richness, we will have to make sure that news about this