Tallgrass prairie plant communities across twenty years of restoration

Ecosystem restoration projects need to measure progress toward project goals and deliver desired outcomes. This study examines longitudinal plant community data collected from permanent transects at the Nachusa Grasslands preserve in northern Illinois, USA. Managers established permanent transects for repeated plant community monitoring beginning in the mid-1990s. Native plant communities, including rare species, have persisted, or improved with management over two decades. Planted prairies have lower proportions of native species than native prairies but have generally maintained native-dominated communities and in some cases, increased presence of native species. Savannas have shown a distinct transition from shrub-dense communities to herbaceous understories dominated with native species. Restoration efforts at Nachusa Grasslands have been successful at sustaining unique native plant communities through management practices like prescribed fire, brush removal, and aggressive invasive species control. As a disturbance dependent ecosystem that has developed with human management over millennia, tallgrass prairie and savanna can thrive through restoration and active management.

Methods
Data was collected at The Nature Conservancy's Nachusa Grasslands (Franklin Grove, IL, USA). Between 1994 and 1996, managers set-up transects in 12 independent sites within the preserve to evaluate and monitor plant community changes in both native and restored prairie units across the preserve (see Bach & Kleiman 2021 for map and additional information). Sites were replicated within three habitat types: native prairie (n=4), restored planted prairie (n=4), and savanna (n=4). Transects in native prairie each included 15 quadrats (1m2) and transects in restored and savanna habitat each included 10 quadrats. All quadrats were spaced approximately 1 m apart on alternating sides of the transect tape. In 1994, we believe 0.5m2 quadrats were used instead of 1m2, although no documentation could verify one way or the other. Given that adjusting the number of species to a per meter basis does not reflect the true species richness at 1m2, we opted to keep the 0.5m2 data in the seven instances that it occurs. In all quadrats, plants were identified to species level. The transects were permanently marked and resampled over time. Due to differences in cover metrics used over the years, Bach & Kleiman (2021) consider species presence/absence only. Original field data was digitized and coalesced into this format. Plant species nomenclature, coefficient of conservatism, and native status were updated to match Wilhelm & Rericha 2017 Flora of the Chicago Region.

This data file includes the original cover values, analytical code reduces it to presence/absence for analyses used in the publication. Transects sampled fewer than three times between 1994 and 2016 were not included in the published analyses and are removed in the analytical code. Summarized data is also available at www.universalfqa.org at the transect level. Statistical code for analyses published in Bach & Kleiman (2021) is available at https://github.com/ebach/Nachusa_plants2021.