Fine-scale distribution and abundance of epiphytic lichens: environmental filtering or local dispersal dynamics?

Questions: How are the fine-scale spatial distribution and abundance of epiphytic lichens explained by factors related to environmental filtering and local dispersal? Are spatial distribution and abundance explained by the same underlying factors across sites for: (1) each species separately; or (2) groups of species with similar dispersal strategies? Location: Ten lowland deciduous forests at the southwest coast of Norway (60 N, 5 E). Methods: We investigated the spatial distribution and abundance, given occurrence, of 15 epiphytic Lobarion lichens in ten forest sites: six 1800-m 2 study sites and four 5000-m 2 study sites. We divided each site into a grid of 1-m 2 sampling units, marked all trees and recorded the abundance of individual lichen species. We assessed the relative impact of factors related to environmental filtering and local dispersal for each lichen species using multiple regressions and variation partitioning. Finally, to compare the results between species and between sites, we applied linearmixed effectmodels. Results: We found that the occurrence of lichen species on a tree is explained primarily by factors related to environmental filtering. The abundance of lichen species that occur on a tree is explained by a combination of environmental filtering and local dispersal, but the relative importance of these factors was found to vary greatly between sites. We found no differences in this respect between species with different dispersal strategies. Conclusions: Our results indicate that both environmental filtering and local dispersal dynamics are important processes explaining the distribution and abundance patterns of Lobarion lichens at fine spatial scales. However, spatial variations in environmental factors within sites interact with propagule distributions to produce a range of inter-site variation. Accordingly, the relative importance of these two structuring mechanisms varies among sites, particularly in the case of abundance patterns. Since single-site patterns are not readily generalized, we emphasize the importance ofmultiple study sites for evaluation of the role of different processes in shaping the spatial distribution patterns of species.