Species richness and seeding density affect plant community composition and response to simulated drought

Background/Question/Methods

Extensive loss of temperate grasslands has greatly elevated the importance of grassland restoration for maintaining grassland biodiversity and ecosystem functioning. In this context, we investigated the effects of different restoration approaches (factorial combination of two seed mixture richness and sowing density levels; n=6, N=24; plot size 0.3 ha; sown April, 2006; South-Central Nebraska) on grassland plant community and ecosystem characteristics. As reduced and more variable summer precipitation is predicted to occur by the end of the century at our study site, we also erected rain-out shelters within all plots for the 2009 growing season to assess the impacts of, and recovery from, a one-year drought. Our primary questions were: 1) Are high initial seeding richness and seeding density associated with greater establishment of native, seeded species and reduced abundance of non-seeded (including exotic) species? 2) What is the effect of drought on recently restored grasslands, and do effects differ among seeding treatments? Response variables included canopy cover (abundance), stem density, grass flowering culm density, dormant-season belowground bud density, and aboveground biomass (ANPP), measured under rain-out shelters and in non-rain-out controls within each seeding treatment plot.

Results/Conclusions

Higher seeding richness and density both enhanced seeded species establishment and reduced non-seeded species establishment. Richness and abundance of seeded species were greater on plots sown with high richness than low richness, but non-seeded species richness, abundance, and stem and bud densities were lower. Similarly, species richness, abundance, and stem and bud density of seeded species were higher on plots sown at high density than low density, and species richness, abundance, and stem and bud density for non-seeded species were lower. Paradoxically, ANPP was lower on high richness than low richness plots, but this measure included both seeded and non-seeded species that had opposite abundance responses to richness treatments. The drought treatment reduced abundance and bud densities but not stem densities for seeded species irrespective of seeding treatment, but increases in stem density and grass flowering culm density from 2009 to 2010 were restricted to non-drought subplots. These legacies of drought for restoration establishment were associated with drought-mediated reduction of perennating belowground buds. While species rich mixtures and high planting densities appear to promote restoration establishment, this effect was not enhanced and at times attenuated by drought, so increases in drought frequency may limit the sensitivity of grassland recovery to different restoration techniques.