Positive and negative plant-plant interactions influence seedling establishment at both high and low elevations

Abstract

Abstract According to the stress gradient hypothesis (SGH), plant-plant interactions are expected to shift from predominantly negative (i.e. competition) to predominantly positive (i.e. facilitation) along gradients of environmental severity. The SGH has been particularly useful as a framework for understanding how plants interact with each other across elevation gradients, partly as a proxy for future responses to climate change. However, most experiments examine the net effects of interactions by growing plants in either the presence or absence of neighbours, thereby neglecting the interplay of both negative and positive effects acting simultaneously within communities. To partially unravel these effects, we tested how the seedling establishment of 10 montane grassland plants varied in the presence versus absence of plant communities at two sites along an elevation gradient. We created a third experimental treatment that retained the main hypothesised benefits of plant neighbours (microsite amelioration), while reducing a key negative effect (competition for soil resources). In contrast to predictions of the SGH, we found evidence for net positive effects of vegetation at the low site, and net negative effects at the high site. Interestingly, the surprisingly negative effects of plant neighbours at high elevation were driven by high establishment rates of low elevation grasses in bare soil plots. At both sites, establishment rates were highest in artificial vegetation (after excluding two low elevation grasses at the high site), indicating that positive effects of above-ground vegetation are partially offset by their negative effects. Our results demonstrate that both competition and facilitation act jointly to affect community structure across environmental gradients, while challenging the view that competition is weak at high elevation in temperate mountain regions. Consequently, plant-plant interactions are likely to influence the establishment of new, and persistence of resident, species in mountain plant communities as environments change.