The role of seed banks in regeneration in established reconstructed tallgrass prairies

Background/Question/Methods

The recovery after a disturbance in tallgrass prairies can be driven by either seed or vegetative regeneration.  At remnant tallgrass prairies, recovery is dominated by vegetative regrowth.  At established reconstructed prairies, seeds may or may not play a role in recovery.  Disturbance is expected to increase the resources available to seedlings at both types of prairies, but it is not clear how much disturbance is required and what resource is most important.  We asked whether seeds, either added or those present in the seed bank, contribute to diversity following disturbance, and whether their contribution changes with disturbance frequency.  We tested that question at two 25-year-old, low diversity reconstructed prairies dominated by Andropogon gerardii.  After a fire or litter removal in spring 2011, we established treatments to compare the diversity among 1 m²plots with either all seedlings from the seed bank removed, seed bank seedlings allowed (control), or seeds added.  Each type of seed treatment was then clipped zero, one, or three times during the first growing season.  At four microsites within each plot, we measured available light, soil moisture, and nitrate against seedling abundance to determine the mechanism that links disturbance to seedling establishment. 

Results/Conclusions

Germination tests indicated that the seed bank contained only seven species at one site and eleven species at the second site and non-native species comprised 97% of all seeds.  After one growing season, seedlings of eight species from the seed bank were present and seven of the ten added species were present.  At both sites, more frequent clipping increased light availability, but did not alter available water and nitrate.  Effects of these resources on seedling abundance were site and weather dependent.  At the first site, which received above-average precipitation, seedlings were more abundant in microsites with lower soil moisture and nitrate levels.  The second site received below-average precipitation and seedlings were more abundant in microsites with less light.  Although seeds originating from the seed bank and from the added seed mix germinated, neither contributed to plot-scale diversity at the end of the first growing season.  This suggests that once established, reconstructions are also primarily maintained by vegetative regeneration.  Furthermore, given the high-proportion of non-native species in the seed bank, if managers attempt to increase local diversity through seed addition and management for seedling establishment, that management may lead to increased invasive species cover.