Recruitment limitation of forest communities in a degraded Mediterranean landscape Mendoza, Irene 1Ã ; Go´ mez-Aparicio,Lorena 2 ; Zamora, Regino 1,3 & Matı´as, Luis 1,4 1 Department of Ecology, Terrestrial Ecology Research Group, University of Granada, Av. Fuentenueva s/n, E-18071 Granada, Spain; 2 Instituto de Recursos Naturales y Agrobiologı´a (IRNASE, CSIC), PO Box 1052, 41080 Seville, Spain; E-mail: lorenag@irnase.csic.es; 3 E-mail rzamora@ugr.es; 4 E-mail lmatias@ugr.es; Ã Corresponding author; Fax: 134 9582 46166; E-mail imendoza@ugr.es Abstract Question: How does habitat degradation affect recruit- ment limitation and its components (seed limitation versus establishment limitation) of woody plant communities in a Mediterranean landscape? Location: 1600-1900 m a.s.l. in the Sierra Nevada National Park, southern Spain. The landscape is a mosaic com- posed of native forest and two degraded landscape units: reforestation stands and shrubland. Methods: We evaluated fruit production, seed rain, seed- ling emergence and seedling survival in two consecutive years with contrasting rainfall patterns. Seed and seedling data were used to calculate values of seed and establish- ment limitation. Results: In general, the woody community was both severely seed- and establishment-limited. Species were less seed-limited in the landscape units with higher adult density (i.e. shrub species in shrubland, Pinus spp. in reforestation stands). In contrast, degradation did not exacerbate establishment limitation, which was severe in all landscape units. This general pattern was modulated by the biogeographical distribution, dispersal type, and life form of the species. Boreo-alpine species were more limited in establishment than species with a typical Mediterranean distribution. Zoochorous species were less seed-limited in the landscape units preferred by dispersers (i.e. native forest). Tree species were more establishment- limited than shrub species, irrespective of the landscape unit. Seed limitation, and especially establishment limita- tion, varied among years, with establishment being almost nil in the very dry year. Conclusion: In the case of Mediterranean landscapes, when degradation from human impact involves a reduc- tion in the adult abundance of the woody plant community (trees and shrubs), seed limitation increases, although establishment limitation is generally high in all landscape units, especially for boreo-alpine species. Con- servation and restoration strategies should take into account our results showing that tree species were unable to recruit in an extremely dry year, because more aridity is expected under a climatic change scenario in Mediterra- nean ecosystems. Keywords: Acer opalus subsp. granatense; Land-use change; Pinus sylvestris; Regeneration dynamics; Seed dispersal; Sorbus aria; Taxus baccata. Nomenclature: Castroviejo et al. 1986; Valde´s et al. 1987; Blanca et al. 2002. Introduction Human impact is so widespread throughout the Earth’s ecosystems that most habitats undergo some form of degradation (Sanderson et al. 2002). Under this general context of habitat degradation, it is crucial to understand recruitment limitation ham- pering the natural capacity of plant populations to recover (Jordano et al. 2004). Studies addressing re- cruitment limitation have normally examined well- conserved habitats, i.e. tropical (Dalling et al. 2002; Muller-Landau et al. 2002; Svenning & Wright 2005), temperate (Schupp & Fuentes 1995; Clark et al. 1998), boreal (Eriksson & Ehrle´n 1992) or, more rarely, Mediterranean (Jordano & Herrera 1995; Hampe & Arroyo 2002; Rey et al. 2006) habi- tats. While these studies provide a general understanding of recruitment limitation when hu- man impact is relatively low, studies focusing on habitat degradation are rare, and generally centred on savannah-type ecosystems (e.g. Plieninger et al. 2003; Pulido & Dı´az 2005). The standard approach to analyse recruitment limitation includes two components (Eriksson & Ehrle´n 1992; Clark et al. 1998, 1999; Nathan & Journal of Vegetation Science 20: 367–376, 2009 & 2009 International Association for Vegetation Science