Positive interactions between nitrogen-fixing legumes and four different neighbouring species in a biodiversity experiment

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

The importance of facilitative processes due to the presence of nitrogen-fixing legumes in temperate grasslands is a contentious issue in biodiversity experiments. Despite a multitude of studies of fertilization effects of legumes on associated nonfixers in agricultural systems, we know little about the dynamics in more diverse systems. We hypothesised that the identity ofrget plant species (phytometers) and the diversity of neighbouring plant species would affect the magnitude of such positive species interactions. We therefore sampled aboveground tissues of phytometers planted into all plots of a grassland biodiversity-ecosystem functioning experiment and analysed their N concentrations, δ15N values and biomasses. The four phytometer species (Festuca pratensis, Plantago lanceolata, Knautia arvensis and Trifolium pratensis) each belonged to one of the four plant functional groups used in the experiment and allowed the effects of diversity on N dynamics in individual species to be assessed. We found significantly lower δ15N values and higher N concentrations and N contents (amount of N per plant) in phytometer species growing with legumes, indicating a facilitative role for legumes in these grassland ecosystems. Our data suggest that the main driving force behind these facilitative interactions in plots containing legumes was reduced competition for soil nitrate ("nitrate sparing"), with apparent N transfer playing a secondary role. Interestingly, species richness (and to a lesser extent functional group number) significantly decreased δ15N values, N concentrations and N content irrespective of any legume effect. Possible mechanisms behind this effect, such as increased N mineralisation and nitrate uptake in more diverse plots, now need further investigation. The magnitude of the positive interactions depended on the identity of the phytometer species. Evidence for increased N uptake in communities containing legumes was found in all three nonlegume phytometer species, with a subsequent strong increase in biomass in the grass F. pratensis across all diversity levels, and a lesser biomass gain in P. lanceolata and K. arvensis. In contrast, the legume phytometer species T. pratense was negatively affected when other legumes were present in their host communities across all diversity levels.

Original languageEnglish
JournalOecologia
Volume151
Issue number2
Pages (from-to)190-205
Number of pages16
ISSN0029-8549
DOIs
Publication statusPublished - 03.2007
Externally publishedYes

Bibliographical note

Acknowledgements We thank the people who helped design and set up The Jena Experiment, especially Ernst-Detlef Schu-lze, Christiane Roscher, Wolfgang Weisser and Jens Schumach-er, and those who helped with the management of the experiment in the first three years: the gardeners Steffen Eis-mann, Silvia Junghans, Heike Scheffler and Ulrike Wehmeier, and many student helpers, especially Manuela Bärwolff, Car-sten Möller, Katja Würfel and Christina Fischer, who assisted in the preparation of plant samples for isotope analysis. Thanks also to all the helpers during the weeding campaigns; and to Andy Hector for statistical discussions. At the Max Planck Institute for Biogeochemistry, special thanks go to Willi Brandt and Heike Geilmann in the isolab for the stable isotope analyses. The Jena Experiment is funded by the Deutsche Forschungsgemeinschaft (DFG, FOR 456), with additional support from the Friedrich Schiller University of Jena and the Max Planck Society.