Biotic interaction effects on seedling recruitment along bioclimatic gradients: testing the stress-gradient hypothesis

QuestionsIs there a shift from positive to negative biotic interaction effects on seedling recruitment along two different stress gradients, temperature and precipitation (the stress-gradient hypothesis); do such interaction effects differ between species with different bioclimatic affinities? LocationBoreal, sub-alpine and alpine grassland in southern Norway. MethodsWe tested the stress-gradient hypothesis by comparing seedling recruitment in bare-ground gaps where vegetation has been removed vs in extant grassland vegetation in 12 boreal, sub-alpine and alpine grassland sites along a precipitation gradient. This was tested in (1) a seed-sowing experiment and (2) in naturally occurring recruitment of alpine, generalist and boreal species. ResultsEmergence of the sown alpine species was higher in the cold alpine than in the warmer sub-alpine sites, with no effects of precipitation or vegetation removal. The sown generalists also decreased in emergence towards warmer sites, whereas there was no effect of temperature on the sown boreal species. Vegetation removal, interacting with precipitation, increased the emergence of the generalist and boreal species sown at intermediate precipitation levels. In contrast, interactions between temperature and vegetation removal regulated the emergence of all groups of naturally occurring seedlings. Alpine and generalist species emerged at the highest rate in alpine sites, whereas boreal species had highest emergence in the lowlands. ConclusionFor all species groups, strong effects of vegetation removal show that competition from the extant vegetation dominates in controlling seedling emergence across all study sites and species. In generalist and boreal species, positive interactions between vegetation removal and temperature show that competitive interactions affect seedling emergence more strongly towards warmer climates, in line with the stress-gradient hypothesis. In contrast, alpine species show no such interactions. This suggests that species' adaptations to climate, in combination with environmental forcing, control seedling emergence along the bioclimatic gradients. Our results have implications for nature conservation, as we propose that disturbance from grazing animals can be useful to release competition and thereby increase seedling recruitment and biodiversity in boreal and alpine grasslands in a warmer future.