In fall of 2010 in watershed N2B ( 39.088976°, -96.588599°), we established plant community plots to assess the potential ability of the riparian zone to shift to a grassland state based on cutting alone and cutting with replanting. The three treatments were 1) naturally open riparian grassland before the removal, 2) areas cleared of woody vegetation, and 3) areas cleared of woody vegetation and seeded with prairie plant species. The addition of the seeded treatment was designed to address if recovery of grassland vegetation is hindered by propagule limitation. The seeded and non-seeded removal plots were adjacent to each other and randomly assigned. In each community type, there were four plots, each of which was 10 m parallel along and 3 m perpendicular to the stream channel. Each plot had four plant composition transects along which we sampled four one m2 subplots along each transect. Vegetative cover of vascular plant species was determined using a modified Daubenmire scale (Gibson and Hulbert 1987).
We planted plots with seed collected from nearby species in fall 2010. Seeds were collected from 70 species (Appendix S1), with six species (listed by mass), Sorghastrum nutans, Andropogon gerardii, Antenneria neglecta, Elymus canadensis, Silphium laciniatum, and Silphium integrifolium, making up more than half of the total seed mass added. The seeding rate was relatively high (~38 kg/ha). This was about twice the recommended rate for grassland restoration, since we did not clean or test seeds for viability. While we did not clean seed (e.g., no pappus removed, seeds like those of purple prairie clover not removed from pods, wild coffee seeds not removed from fruits, etc.), Echinacea seeds were separated from the capitulum. Seed for each species was stored in paper bags at room temperature until time of seeding. Seeds were sown onto each plot and gently raked in on 28 Dec 2010 shortly after woody cutting. A paper bag was prepared for seed addition in each treatment plot with all species mixed together. Seed was hand broadcast, passing over the treatment plot at least three times in the process of emptying each bag to ensure even spread and that any smaller seeds settling to the bottom would be spread throughout.