Colonization of abandoned lands by woody vegetation may be a great opportunity for ecosystem recovery given the current and future trend of land abandonment. This may help to reverse the generalized condition of ecosystem degradation of developed countries which is an urgent need. However, ecosystem recovery in abandoned lands can be seriously limited by biotic and abiotic factors. Indentify such factors and determine the relevant spatial scales at which they operate will help to understand natural patterns of colonization of abandoned lands and may be useful to guide restoration activities. We used Spanish juniper remnant woodlands and old fields recently colonized by the species to assess the variation in post-dispersal seed predation and environmental suitability to plant recruitment. These biotic and abiotic factors are two of the most limiting for vegetation recovery in abandoned lands, at the regional and local spatial scale. We found that recruitment was controlled by factors operating at different spatial scales in a hierarchical manner along different stages of the process. The regional scale was determinant for post-dispersal seed predation and seedling abundance which was, in turn, controlled by environmental suitability at the local spatial scale. Post-dispersal seed removal was higher in old fields than in mature woodlands, hence increasing seed limitation, a pervasive constraint for plant recruitment in abandoned lands. Environmental suitability for plant recruitment did not decrease as a result of previous farming uses at the regional scale, a common pattern in areas not subjected to intensive farming practices. Abandoned lands in Mediterranean areas seem to have a strong potential for ecosystem recovery being the biotic factors (e.g. seed availability) more limiting than the abiotic ones when non-intensive farming practices have been performed. This together with predictions of increase in land abandonment in low productive areas makes old fields especially relevant from an ecosystem recovery perspective. © 2014 Elsevier B.V.

Gema Escribano-Ávila was supported by a FPU-MEC doctoral grant from the Spanish Ministry of Education (http://www.educacion.gob.es/portadahtml). Funding was provided by the Spanish Ministry for Innovation and Science (http://www.idi.mineco.gob.es/) with research projects CGL2010-16388/BOS, Consolider Montes (CSD2008_00040), VULGLO (CGL2010-22180-C03-03), and CALCOFIS (CGL2009-13013), and by the Community of Madrid project REMEDINAL 2 (CM-S2009/AMB-1783).

Peer Reviewed