Post-fire restoration of Mediterranean forests: Testing assembly rules mediated by facilitation

In view of the importance of facilitative interactions between plants, nurse-based planting has been proposed as a restoration technique for Mediterranean vegetation. However, facilitation efficiency is known to depend on the environmental context and the particular pair of interacting species. Understanding these context- and pair-specific dependences is fundamental to understanding Mediterranean vegetation dynamics and to improving the use of nurse-based plantation for restoration. We assessed the effectiveness of nurse-based plantation and the significance for post-fire restoration of some assembly rules mediated by facilitation. In two nearby areas of different burning ages, we compared seedling establishment of 13 tall shrubs and trees planted in open ground and under nurses. Nurses and planted seedlings were selected from different life-forms. Tests of the assembly rules were provided by the partitioning of the statistical interaction effect between nurse and planted seedling life-forms in a two-factor design. Nurse-based plantation increased seedling survival 2-9 times compared to plantation in open ground, depending on the year. The higher efficiency of nurse-based plantation was consistent for the two burned areas and occurred in many species even in years with contrasting rainfall. We detected pair-specific differences in the efficiency of facilitation. This pair-specificity was partly explained by the dependence between life-forms of nurse and guest species, suggesting the existence of assembly rules. Our results confirm that nurse-based plantation would be an appropriate post-fire restoration technique in Mediterranean mountains under dry-subhumid climate, but suggest that attention to the life-form assemblage rules is needed for an efficient implementation of such technique.