The biomass gain from planted richness had no clear link to April–June rain at Cedar Creek in 1997–2020

We tested whether planted species richness changed the link between spring rain and grassland biomass at Cedar Creek, Minnesota. We used 154 fixed plots and 24 years, from 1997 to 2020. Each year gave a slope of log live biomass against log2 planted richness. We fitted this slope against April–June rain, with linear and quadratic time terms. A station-month needed a valid rain value for every day. The rain coefficient was +0.0066 log biomass units per richness doubling per 100 mm. Its OLS 95% interval was -0.0063 to +0.0194. An interval from samples of plot histories and three-year blocks was -0.0132 to +0.0204. A 200 mm decrease in rain gave a fitted multiplier of 0.949 for the biomass ratio of 16-species plots to monocultures. The plot and year-block interval was 0.849 to 1.111. Daymet rain, including the missing station year 2021, gave a similar result. In sorted strips, the planted-species coefficient was negative and the non-planted term was positive. Both crossed-sample intervals included zero. The total-biomass data do not show a clear dry-year benefit from planted richness under this model. They do not establish no effect. This result concerns one managed site and annual rain totals. It is not a test of all severe droughts. The package contains source data, provenance, code, all results, figures, and independent checks.

biodiversity · grassland · rainfall · Cedar Creek · biomass · species richness · uncertainty · reproducibility