Total biomass had a smaller diversity–stability slope than planted biomass at Cedar Creek in 2011–2021

We tested how biomass accounting changes the link between planted species number and stability at Cedar Creek, Minnesota. We used 154 grassland plots and paired samples sorted by species in all 11 years from 2011 to 2021. Stability was mean biomass divided by its standard deviation across years. The slope of log stability against log2 planted species number was 0.151 for planted biomass and 0.062 for total live biomass. The paired difference was −0.089 per richness doubling, with a nominal 95% plot bootstrap interval of −0.123 to −0.056. The total slope was 59% smaller. Species that were not planted raised the relative mean biomass most in low-diversity plots. An exact mean and variance identity explains the change in slope. Checks with common year blocks, linear trends removed, other periods, mixtures only, and repeated source rows retained a negative difference. An independent raw-data reconstruction matched all 3,388 primary biomass cells and all primary slopes. This result applies to sampled biomass at one managed site. It does not measure the effect of removing species that were not planted. Clipping noise, short time series, plot selection, other plot treatments, and weed records limit interpretation. The package contains unchanged source data, provenance, code, all estimates and resamples, two figures, and instructions to repeat the analysis.

biodiversity · grassland · temporal stability · biomass · Cedar Creek · planted richness · colonization · reproducible analysis