Effects of the seed bank and interseeding in reconstructed tallgrass prairies

Disturbances such as fire and mowing temporarily increase available resources for plants, opening a window of opportunity for establishment. During the recovery of vegetation after disturbance in remnant prairies, seedling establishment is often negligible compared to vegetative regrowth. It is unclear if this is the case in reconstructions. I tested the contribution of seedlings to diversity after disturbance in two, 25-year-old, low diversity reconstructed prairies by removing seedlings, allowing seedlings, and adding seed in 1 m 2 plots. Disturbance frequency effects were tested by clipping each treatment zero, one, or multiple times. After two growing seasons, in the wetter field site, seedlings removed plots had the lowest species richness and highest evenness, plots with seedlings from the seed bank were intermediate, and plots with added seed had the highest richness and lowest evenness. In the drier site, only adding seed impacted richness (higher) and evenness (lower). In both sites, the seeds present in the seed bank were over 80% from non-native species. Seedling establishment was quantified along with light, water, and nitrate availability within each seeded plot. In the wetter site, resource levels did not predict seedling numbers, but seedlings were more numerous in more frequently clipped, lighter plots. In the drier site, seedlings were more numerous in less frequently clipped plots which received less light and may have remained moister. Seedling establishment from any source never affected community diversity suggesting that while some seeds establish, reconstructions are primarily maintained by vegetative reproduction. The inconsistent effect of clipping highlights the need for management to adapt to varying precipitation, however, given the non-native dominated seed banks, any management intended to increase seedling establishment could result in increased non-native cover