Hierarchical patterns of symbiotic orchid germination linked to adult proximity and environmental gradients

1 The recruitment stage can be critical in determining plant population dynamics, as juveniles encounter a range of abiotic and biotic stressors in the environments where they land after dispersal, and often experience high mortality rates. Although both abiotic and biotic constraints on recruitment are often assumed to operate at multiple scales, these relationships are rarely quantified in ways that enable prediction of recruitment dynamics across environmental gradients. 2 In plants such as orchids, an important bottleneck on recruitment may arise from an obligate dependence on mycorrhizal fungi in order to reach photosynthetic stage. Very little is known, however, about the factors controlling the distribution of the fungi or how fungal distributions may constrain orchid recruitment and distributions. 3 In this study, seed introduction experiments were used to test for relationships between symbiotic germination success of an orchid, Goodyera pubescens, and both proximity to adult orchids and soil abiotic conditions. Experiments were linked to ongoing demographic monitoring across a large environmental gradient to place patterns of germination better in the context of longer-term demographic patterns. 4 A hierarchical Bayesian statistical framework was used to evaluate explicitly how recruitment varies across scales and how this variation is explained by key abiotic variables. This framework was further utilized to link hypothesis testing to model building, and thereby improve our ability to make predictions across a range of environmental conditions. 5 At a microsite scale (< 4 m(2)) strong evidence was found for decreasing germination success further from adult plants. At distances greater than 1 m from adults, increased germination success was associated with higher soil moisture, higher organic content and lower pH, although the strength of these relationships varied regionally. Patterns of symbiotic germination success were also highly correlated with observed recruitment patterns at the population level. 6 Explicitly linking plant recruitment to underlying abiotic gradients and key biotic interactions using a hierarchical, predictive modelling framework is essential for understanding basic plant population processes and building capacity to make predictions of how species and communities may respond to environmental changes.