Projected regional forest plant community dynamics evidence centuries-long effects of habitat turnover
La dynamique régionale prédite des communautés végétales forestières met en évidence des effets sur plusieurs centaines d'année après le turnover d'habitat
Lalechère, E. ; Jabot, F. ; Archaux, F. ; Deffuant, G.
Type de document
Article de revue scientifique à comité de lecture
Affiliation de l'auteur
IRSTEA CLERMONT FERRAND UR LISC FRA ; IRSTEA CLERMONT FERRAND UR LISC FRA ; IRSTEA NOGENT SUR VERNISSON UR EFNO FRA ; IRSTEA CLERMONT FERRAND UR LISC FRA
Résumé / Abstract
Questions Many Western European and North American countries have been experiencing strong reforestation in the last two centuries due to agricultural abandonment. Other land‐use changes, such as urbanization, can simultaneously trigger forest erosion. In this context of habitat turnover, forest understorey plant dynamics depend on the balance between immigration credit in post‐agricultural forest and extinction debt in ancient forest. The transient and final community properties following concomitant habitat creation and destruction are poorly known. In this contribution, we study the projected transient regional forest plant dynamics and identify the determinants of the relaxation duration at both landscape and patch scales. Location Seine‐et‐Marne region, France. Methods Species' incidences are projected in 9,208 patches of the Seine‐et‐Marne region characterized by strong development of post‐agricultural forests and a moderate erosion of ancient forests during the period 1840-2000. We use a metapopulation model with static biodiversity data and landscape history to project forest plant dynamics. We focus on 33 generalist forest species that are able to colonize both ancient and post‐agricultural forests. Results Our analyses reveal that (1) extinction in ancient forests slows the colonization process, which lasts between 250 and 990 years after habitat turnover, depending on species characteristics; (2) the species incidence in post‐agricultural forests converges towards that in ancient forests long before the overall species incidence stabilizes at the landscape scale (i.e. long before the relaxation time); (3) landscape‐ and patch‐scale relaxation times depend on species colonization ability and patch functional connectivity; (4) the colonization process is rather slow for ancient forest species; and (5) the incidence of ancient forest species becomes larger than the incidence of other species at the end of the colonization dynamics thanks to their strong persistence. Conclusions Ancient forest species may potentially benefit from a large immigration credit but need centuries to use it. On such time scales, many perturbations are likely to jeopardize the realization of this potential credit.
Journal of Vegetation Science, vol. 29, num. 3, p. 480 - 490