Interacting effects of forest edge, tree diversity and forest stratum on the diversity of plants and arthropods in Germany’s largest deciduous forest

Abstract

Major threats to global biodiversity include the continuous increase of forest fragmentation and the associated augmentation of forest edge zones. How much edge effects penetrate into the forest interior can be influenced by habitat structure with tree species composition weakening or strengthening edge effects. Here, we address for the first time forest edge and tree diversity effects and their potential interactions on the understory vegetation and arthropods, focusing on the Hainich National Park, Germany´s largest connected deciduous forest. A total of 12 transects extending from the forest edge up to 500 m into the forest interior were established – six of them in forest stands dominated by beech with a low tree species diversity and six in forest stands rich in tree species, containing up to nine deciduous tree species e. g. oak, ash, lime and maple. Understory vegetation and arthropods were studied along each transect. In the first manuscript (chapter 2) of this thesis we studied the understory vegetation along the edge-interior gradient. The herb layer plant species richness was influenced by an interaction of tree diversity and edge effects. In the high tree diversity forest stands herb species richness was not affected by edge proximity, whereas in beech dominated forest stands it strongly declined with increasing distance from the forest edge. This resulted in higher plant species richness in the forest interior of the high tree diversity level. The fraction of forest specialist species increased, while the fraction of forest generalists decreased from the forest edge towards the forest interior. The dominance of forest specialists was much stronger in the low tree diversity level. Plant community composition differed with tree diversity level and edge proximity and it was more variable in the high tree diversity forest stands and closer to the edge. Tree diversity mediated leaf litter thickness, which was identified as the most important predictor of plant species richness. The second manuscript (chapter 3) focuses on ground-dwelling arthropods (ground beetles, rove beetles and spiders) and the effect of body size and habitat specialisation on their response to tree diversity and forest edge proximity. While no consistent pattern was found for total species richness, the tree diversity and edge response acrIm doss all three taxa depended on habitat specialisation and body size. Neither tree diversity nor edge effects clearly affected the richness of forest species and body size was also not important. However, individual species suffered, whereas others were promoted by increased tree diversity. The species richness of habitat generalists strongly declined from the forest edge towards the forest center in the low tree diversity level. This effect was mitigated in the high tree diversity level (except for spiders) and the species richness of habitat generalists, and among these the small species in particular, benefited from increased tree diversity. Individual habitat generalist species were generally positively affected. Changes in environmental conditions and habitat heterogeneity induced by tree diversity and edge proximity are most likely the reason for the observed patterns. In the third manuscript (chapter 4) forest stratum as a third component was added to the study approach. We studied the forest in its full three-dimensionality by addressing edge and tree diversity effects on beetles across forest strata. Therefore, flight interception traps were installed both in the canopy and the understory along ten of the transects for a seven month period from April until November. Edge effects influenced beetle species richness and community composition on a large spatial scale extending up to 500 m into the forest interior. However, edge effects were weaker in the canopy than in the understory - likely a result of higher, edge-like microclimatic variability and harshness in the canopy. Tree diversity did not influence edge effects. The edge response of total beetle species richness was driven by habitat generalists, which strongly declined with increasing distance from the forest edge, whereas saproxylic and forest species only responded weakly. Habitat generalists and non-saproxylic species dominated the forest understory. The richness of saproxylic and forest species peaked in the canopy. Tree diversity enhanced beetle diversity across all strata and forest specialisation groups. Structural equation modelling revealed that pathways driving beetle richness differed across strata. Tree diversity, dead wood amount and (partly) canopy openness were the most important drivers in the canopy, whereas canopy openness, edge proximity and to a lesser extent tree diversity were important in the understory. In conclusion, in the canopy tree diversity effects were stronger and more direct than in the understory. Overall, we conclude that for a deeper understanding of forest fragmentation the relative importance of edge, stratum and tree diversity, but also species’ life-history traits (e. g. body size) and habitat specialisation should be considered. Increasing the abundance and diversity of deciduous tree species in Central European forests may help to preserve the regional species diversity of plants and arthropods. However, some forest species may rely on old-growth pure beech forests. These have received special attention in the UNESCO World Heritage sites “Primeval Beech Forests of the Carpathians and the Ancient Beech Forests of Germany”.