CONCLUSIONS

We here showed that functional traits of successful grassland species differed from those of less successful ones, but that the pattern depended on the success metric and the spatial scale it applies to. Low root-tissue density was the only trait that characterized successful species at every spatial scale, from being abundant in German grassland plots to being widely naturalized around the world. We showed that belowground traits are at least as important as the aboveground traits in explaining species success at the different spatial scales. The variation in importance and the sometimes-opposing directions of the effects of traits on species success at different spatial scales can explain why trait variation is maintained. Our study shows that, for Central European grassland species, variation in success is related not only to aboveground traits, but also to belowground traits.
ACKNOWLEDGEMENTS
We thank Vanessa Pasqualetto for weighing the seeds, Otmar Ficht, Maximilian Fuchs and Heinz Vahlenkamp for help setting up the experiments, Beate Rüter, Ekaterina Mamonova, Huy Manh Nguyen, Simon Gommel, Maximilian Rometsch and Anika Schick for help measuring the plant traits. We also thank the managers of the three Biodiversity Exploratories, Konstanz Wells, Swen Renner, Kirsten Reichel-Jung, Sonja Gockel, Kerstin Wiesner, Katrin Lorenzen, Andreas Hemp, Martin Gorke and Miriam Teuscher, and all former managers for their work in maintaining the plot and project infrastructure; Christiane Fischer for giving support through the central office, Andreas Ostrowski for managing the central data base, and Markus Fischer, Eduard Linsenmair, Dominik Hessenmöller, Daniel Prati, Ingo Schöning, François Buscot, Ernst-Detlef Schulze, Wolfgang W. Weisser and the late Elisabeth Kalko for their role in setting up the Biodiversity Exploratories project. The work has been (partly) funded by the DFG Priority Program 1374 ”Infrastructure-Biodiversity-Exploratories” (DFG-Refno.). Field work permits were issued by the responsible state environmental offices of Baden-Württemberg, Thüringen, and Brandenburg. We acknowledge funding from the German Research Foundation (DFG, grants KL 1866/12-1 and 264740629 to MvK and 323522591 to MR).
Figure 1 Estimates of trait effects on different success metrics of German grassland species from generalized linear models. On the y-axis are the nine traits used as predictors, with a linear term (white rows) and a quadratic (non-linear) term (grey rows) for each trait. The errors bars around the estimates are standard errors. Red points and orange points indicate significant (p < 0.05) and marginally significant (p < 0.10) negative model estimates respectively. Blue points and purple points indicate significant (p < 0.05) and marginally significant (p < 0.10) positive estimates respectively, and grey points indicate non-significant estimates (p > 0.10). The spatial scale of the success metric decreases from left to right. GloNAF: number of regions in which a species is naturalized (number of species, N=242); Euro+Med: number of regions in Europe and the Mediterranean basin in which a species is native (N=238); FloraWeb: number of grid cells in Germany in which a species is present (N=236); GPs Frequency: number of grassland grid plots in which a species is present (N=209); EPs Frequency: number of grassland experimental plots in which a species is present (N=240); GPs Abundance: mean species cover in grassland grid plots in which the species is present (N=209); EPs Abundance: mean species cover in grassland experimental plots in which a species is present (N=240). Delta R² was calculated according to