Changes in the nature of environmental limitation in two forest herbs during two decades of forest succession

Aims The environmental legacies of past land use generally have an important impact on the establishment and persistence of forest understorey plant species in post‐agricultural forests. Here we studied how the nature of such environmental limitation changes during the first decades of forest development. Location Deciduous forest ‘Muizenbos’, northern Belgium. Methods Two forest herbs with comparable life‐history traits, but contrasting rates of post‐agricultural forest colonization ( Primula elatior , Geum urbanum ) were introduced 18 yr ago. Introduction sites were ancient or post‐agricultural forest; seeds or adults were added to permanent plots where the understorey vegetation was initially cleared or left undisturbed. We measured chemical soil conditions and recorded the understorey plant community to characterize the sites. The effects of the treatments (land use, vegetation clearing) were tested on the present‐day occurrence and abundance of the introduced species across the plots. We also quantified whether the temporal changes in the total number of individuals per site was different between the two land‐use types and consistent within land‐use types. Results Land use had an important effect on soil and vegetation. In post‐agricultural forest, the P availabilities were many times higher, the plant community diversity was lower and the competitive species were significantly more abundant compared with ancient forest. The present‐day abundance of P. elatior was significantly higher in the ancient forest sites, both for planted and seeded plots, while for G. urbanum land use had no effect. The initial vegetation clearing had a persistent effect on the establishment success of both species. The 18‐yr‐long time series showed consistent changes in the total population sizes in ancient forest: P. elatior has been expanding and G. urbanum initially increased and then declined over the past decade. In post‐agricultural forest, temporal patterns were more divergent and partly related to differences in dominant tree species: differences in acidification rates and changes in insolation caused by canopy dieback have put the populations on new and unforeseen trajectories. Conclusions The longer the experiment is running, the more the individual sites show separate dynamics. In general, we showed that environmental limitation affects plant colonization at least during the first two decades of secondary forest succession, but also that as the post‐agricultural forests develop, the nature of environmental limitation changes.