Do gap dynamics really work for forest conservation and tree recruitment?

The understory vegetation plays a key role in sustaining the biodiversity of forest ecosystems and, therefore, should also receive attention in conservation and restoration plans oriented at the recruitment of forest trees. However, while extensive research has focused on the contribution of gap dynamics to tree seedlings recruitment and natural forest regeneration (1), the effect of treefall on understory composition has received less attention (2). This study aimed to assess the ability of the two most diverse forest recruitment approaches, i.e. ‘natural gap dynamics’ and ‘afforestation’, in maintaining or restoring the understory vegetation of mature meso-hygrophilous forests. The Regional Park of Migliarino, San Rossore, Massaciuccoli, which is located along the Tuscan coast (central Italy, 43°43′ N and 10°16′ E) close to farmland and urban settlements, was the study area. Phytosociological investigations were carried out in three vegetation types: MF, a mixed forest classified as association Fraxino-Quercetum roboris Gellini et al. 1986 (3), with over 200-year old Quercus robur L. trees; FG, gaps started to form in MF approximately 20 years ago by progressive trunk breakage and tree uprooting, and AF, a plantation of Q. robur set up around 1995 in an area bordering MF, which was used for centuries as pasture and then for farmland. Species richness and cover/abundance following Braun-Blanquet were determined on five 400-m2 plots for each vegetation type. Plant communities were compared by means of the Sørensen Similarity Index, biological and chorological spectra, Ellenberg’s ecological indicator values (EIV), and typical habitat of species. Overall richness was almost double in gaps than in forests, 77 species vs. 40 (MF) and 42 (AF). Species exclusive to gaps were 37 out of 77 and 48% of them were Therophytes mostly from ruderal habitats, 14% were aliens, and 13% were cosmopolite and subcosmopolite species. The ecogram showed a clear shift towards a grassland community (Fig. 1a). These findings challenge the theory that forest gaps promote the spread of understory species and that changes in composition are only temporary (1,2). They provide evidence that, in highly anthropized areas, gaps are primarily colonized by pioneer species from non-forest habitats and, thus, fail in preserving nemoral species and forest biodiversity. In contrast, after around 30 years, AF shared 59% of species with MF, among which typical nemorals like Carex sylvatica Huds. and Chamaeiris foetidissima (L.) Medik., and seedlings of Fraxinus angustifolia Vahl, and Ulmus minor Mill., besides those of Q. robur. Moreover, the ecograms indicate that in AF the ecological conditions had become comparable to those of MF (Fig. 1b). Our results suggest that afforestation can be an effective restoration strategy, if connectivity with natural systems is maintained for the supply of forest species propagules.