At nine burned Cedar Creek plots, halves where fertilizer stopped had 3.1 more plant taxa detected in at least two surveys, 2019–2022

We tested whether a known plant richness difference included repeat detections. We used Cedar Creek E002 field B data from nine plots with annual spring fire and the three highest past nutrient levels. Each plot had a half where fertilizer stopped after 2013 and a half where input continued. The primary count was the number of taxa detected in at least two of three late surveys: 2019, 2021, and 2022. Each survey sampled a different 0.3 square metre strip. Stopped halves had 6.89 taxa with repeat detections, compared with 3.78 on continued halves. The paired difference was 3.11 taxa, with a 95% plot bootstrap interval of 2.22 to 3.89. All nine differences were positive. Detection in all three surveys also gave a difference of 3.11 taxa. This class supplied 78.5% of the net annual richness difference. Species-level names only and a higher biomass threshold gave similar results. Nine plots without annual fire had a smaller primary difference of 0.44 taxa, with an interval of −0.44 to 1.22. This extends earlier SCI analyses of the same experiment. It is not an independent experimental replication. Repeat strip detections do not establish individual plant survival, new recruitment, or full community restoration. The package has unchanged source data, provenance, detection histories, code, figures, and independent arithmetic and reproduction checks.

grassland · nutrient cessation · plant richness · repeat detection · Cedar Creek · restoration · fire · reproducibility