Aim The aim of this study was to develop and evaluate a structural equation model explaining the species richness of bryophytes and vascular plants along multiple environmental gradients. Our primary interests in developing this model were: (1) to evaluate the effect of tree canopy along an altitudinal gradient on bryophyte and vascular plant richness; (2) to determine to what extent lithology was able to explain richness in the two groups of plants; (3) to assess whether anthropogenic disturbances decrease richness; and (4) to explore, comparing competing models, the causal links connecting spring area, discharge and spring complexity and to assess how these variables are related to richness.
Location Eighty-six springs were sampled in the south-eastern Alps of Italy, from the lowlands to high-mountain regions, and on different lithologies.
Methods A structural equation model (SEM) was used to test certain hypotheses about the direct and indirect effects of altitude, canopy, lithology, disturbance, spring complexity, discharge and spring area on bryophyte and vascular plant richness. Competing models were evaluated and bootstrap simulation was used to determine the stability of parameter estimates.
Results SEM analysis made it possible to disentangle the different effects of canopy and lithology on bryophyte and vascular plant richness: for the former it was demonstrated that the increased richness with altitude was related primarily to lithology, whereas the latter increased because of the reduction of canopy cover. In addition, the model predicted that spring discharge determined the size of the spring area and even the complexity of the spring; these latter two variables influenced to different degrees both bryophyte and vascular plant richness. Anthropogenic disturbances affected the richness of bryophytes more than that of the vascular plants.
Main conclusions Our study demonstrates that several similarities in the patterns of bryophyte and vascular plant richness are in fact induced by different environmental variables.