LETTER Life-history traits predict species responses to habitat area and isolation: a cross-continental synthesis Erik O ¨ ckinger, 1 * Oliver Schweiger, 2 Thomas O. Crist, 3 Diane M. Debinski, 4 Jochen Krauss, 5 Mikko Kuussaari, 6 Jessica D. Petersen, 7 Juha Po ¨ yry, 6 Josef Settele, 2 Keith S. Summerville 8 and Riccardo Bommarco 1 Abstract There is a lack of quantitative syntheses of fragmentation effects across species and biogeographic regions, especially with respect to species life-history traits. We used data from 24 independent studies of butterflies and moths from a wide range of habitats and landscapes in Europe and North America to test whether traits associated with dispersal capacity, niche breadth and reproductive rate modify the effect of habitat fragmentation on species richness. Overall, species richness increased with habitat patch area and connectivity. Life-history traits improved the explanatory power of the statistical models considerably and modified the butterfly species–area relationship. Species with low mobility, a narrow feeding niche and low reproduction were most strongly affected by habitat loss. This demonstrates the importance of considering life-history traits in fragmentation studies and implies that both species richness and composition change in a predictable manner with habitat loss and fragmentation. Keywords Connectivity, fragmentation, Lepidoptera, meta-analysis, quantitative synthesis, species richness, species–area relationship. Ecology Letters (2010) 13: 969–979 INTRODUCTION Local species richness is predicted to decrease with decreasing area of habitat patches and increasing isolation from conspecific populations (MacArthur & Wilson 1967; Hanski 1999; Fahrig 2003). Not all species, however, respond equally to habitat loss and fragmentation, and it has been suggested that the variability in speciesÕ responses to habitat area and fragmentation can be explained by life- history traits (Henle et al. 2004; Ewers & Didham 2006; Prugh et al. 2008). When combinations of traits make species more susceptible to habitat loss and fragmentation, communities in small and isolated habitat patches will not only contain fewer species than large and well-connected ones, but will also differ in species composition. In highly fragmented landscapes, communities are generally pre- dicted to become increasingly similar in composition, and dominated by few generalist species with similar traits (McKinney & Lockwood 1999; Ekroos et al. 2010). How- ever, predicting such responses may not be quite so simple. For example, a recent study indicated non-univocal responses of different species groups, indicating potential interactions between niche breadth and dispersal capacity (Dormann et al. 2007). 1 Department of Ecology, Swedish University of Agricultural Sciences, P.O. Box 7044, 75007 Uppsala, Sweden 2 UFZ, Helmholtz Centre for Environmental Research, Depart- ment of Community Ecology, Theodor-Lieser-Strasse 4, 06120 Halle, Germany 3 Department of Zoology, Miami University, Oxford, OH, USA 4 Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA 50011, USA 5 Population Ecology Group, Department of Animal Ecology I, University of Bayreuth, Universita ¨ tsstrasse 30, D-95447 Bayreuth, Germany 6 Finnish Environment Institute, Research Programme for Bio- diversity, P.O. Box 140, FI-00251 Helsinki, Finland 7 Department of Entomology, Iowa State University, 424 Science II, Ames, IA 50011, USA 8 Department of Environmental Science and Policy, Drake Uni- versity, Olin Hall, Des Moines, IA 50311-4505, USA *Correspondence: E-mail: Erik.Ockinger@ekol.slu.se Ecology Letters, (2010) 13: 969–979 doi: 10.1111/j.1461-0248.2010.01487.x Ó 2010 Blackwell Publishing Ltd/CNRS