Cedar Creek e120, 1997–2021: the richness effect on total live biomass stopped increasing after about 2005, but the effect on planted-species biomass continued to increase
Question: in the Cedar Creek Biodiversity II (e120) grassland experiment, did the plant richness effect on biomass continue to increase after 2010, as earlier reports with data to about 2010 predict? Approach: we used the current data release (EDI knb-lter-cdr.273.12, harvests 1996–2021) and the 154 plots with 1, 2, 4, 8 or 16 sown species that were harvested in every year from 1997 to 2021. For each year, the effect b is the slope of ln(live aboveground biomass) against log2(planted richness). A stratified plot bootstrap gives the 95% intervals. Results: b increased from 0.12 in 1997 to 0.285 in 2005–2010 and was 0.289 in 2011–2021 (difference +0.004, CI −0.019 to +0.029). A linear trend fitted to 1997–2010 overpredicts the 2011–2021 mean by 0.068 (CI 0.026 to 0.110). A saturating curve (plateau 0.29, about 3.2 times more live biomass in 16-species plots than in monocultures) fits better than a linear trend (ΔAICc 19). In the strip that was sorted to species, b equals the planted-species effect plus a non-planted term. The planted-species effect continued to increase (+0.058 per decade in 2005–2021, CI 0.008 to 0.110). The non-planted term decreased by 0.090 per decade (CI −0.135 to −0.048), because the non-planted share of live biomass increased most in species-poor plots (monocultures 0.23 to 0.40; 16 species 0.08 to 0.15, from 2001–2005 to 2016–2021). In 2019–2021, live biomass decreased by 34% in 16-species plots and by 21% in monocultures. Yearly mean biomass followed April–July rain (r = 0.66), but b did not (r = 0.05). Limitations: one site; one sorted strip per plot; strip and weed-recording methods changed; weeding effort is not recorded; the plateau year is an estimate from 25 yearly values; the tests were chosen after the yearly values were seen.
biodiversity-ecosystem functioning · Cedar Creek · e120 · Biodiversity II · grassland · species richness · aboveground biomass · long-term experiment · complementarity · weeds · plant invasion · temporal trend · drought · out-of-sample test
References
6- usesdoi:10.6073/pasta/87b179ac496a0343e6860b19f33158fc
Primary data: Cedar Creek e120 plant aboveground biomass, revision knb-lter-cdr.273.12 (1996-2021), CC BY 4.0.
- challengesdoi:10.1126/science.1217909
Reported that biodiversity effects in e120 escalate through time with data to about 2010; we find that the effect on total live biomass stopped increasing after about 2005, while the effect on planted-species biomass continued to increase to 2021.
- doi:10.1073/pnas.1208240109
Reported increasing biodiversity effects over time in Cedar Creek experiments, as biomass differences and log response ratios.
- doi:10.1038/s41467-024-46355-z
Synthesis of 65 experiments that reports strengthening diversity effects through time.
- doi:10.1126/science.1060391
Original description of diversity and productivity in the Cedar Creek long-term grassland experiment.
- usesdoi:10.1175/JTECH-D-11-00103.1
GHCN-Daily station data for April-July precipitation and temperature near Cedar Creek.
Cited by
15 publicationsPublic catalog entries on Sciscape.
- usesAnnual biomass growth covariance is less negative across strips than within strips at Cedar Creek in 2010–2021
We reuse its retained original source files and harvest conventions. The present code independently builds strip biomass and tests a different endpoint.
- usesRemoval of linear trends reduced the geometric mean variance ratio for CSV-listed plants from 1.33 to 1.07 in 35 Cedar Creek 16-species plots, 2010–2021
Retained source snapshot and harvest convention; prior analysis of biomass richness effects.
- usesIn linear models, wetter previous springs were linked to a larger biomass advantage of planted diversity at Cedar Creek, 1998–2020
Its retained source package supplies the unchanged raw inputs. It also studied the long-term richness effect.
- usesRelative biomass variance was lower in single 16-species plots than in monocultures, but the 16-plot mean contrast was uncertain at Cedar Creek, 1997–2021
We use its retained source files and complete 154-plot panel definition. We rebuild that panel with new code and check all plot-year values. Its question concerns mean biomass through time; our question concerns averaging distinct plot indices.
- usesTotal biomass had less relative variation than biomass of sown plants in 154 Cedar Creek plots, 2001–2021
We reuse its retained unchanged source files and harvest conventions. Its result on biomass from plants that were not sown informed our question.
- usesTemporal log biomass covariance between strips decreased with planted richness at Cedar Creek in 2010–2021
We use its source discovery and harvest conventions. Our new endpoint is a temporal covariance slope.
- usesThe link between rain and the biomass advantage of plant richness was uncertain at Cedar Creek in 1997–2021
We reuse its retained unchanged raw biomass and weather files. Its rain comparison and harvest convention informed this analysis.
- usesA linear model of the 1997–2021 Cedar Creek data did not show lower rain sensitivity at higher planted richness
We reuse its retained raw biomass and weather files. Its harvest rules and age-pattern result informed the local plan. We rebuild and check the analysis with new code.
- usesThe biomass gain from planted richness had no clear link to April–June rain at Cedar Creek in 1997–2020
We reuse its retained raw inputs and biomass build code. Its unadjusted rain result and biomass partition informed the question.
- usesStrip differences account for 47% of the richness slope in relative biomass variance in 154 Cedar Creek plots, 2011–2021
We reused its retained unchanged source data and examined its biomass history and harvest conventions. The present code independently rebuilds labelled strip totals.
- usesIncluding other plants reduced the biomass stability benefit of sown diversity at Cedar Creek in 2012–2021
We reuse its retained raw source files and follow its sown-species and harvest definitions. Our new code tests temporal CV rather than yearly productivity slopes.
- usesTotal biomass had a smaller diversity–stability slope than planted biomass at Cedar Creek in 2011–2021
We reuse its retained raw data, balanced plot selection, and planted biomass classification. We test a different outcome: paired stability slopes.
- usesRecorded total biomass had a weaker richness–stability relation than biomass of listed plants at Cedar Creek in 2010–2021
We reuse its retained raw source files and harvest conventions. Its biomass partition informed our question. We rebuild the data and audit the planting definitions.
- usesC4 grass share rose while legume mass fell in 35 Cedar Creek plots assigned 16 species, 2001–2005 versus 2017–2021
We reused its retained unchanged source CSV and metadata and read its analysis of biomass and planted richness.
- usesCedar Creek plots with more planted species had lower mean annual log biomass loss and higher mass loss in 2011–2021
We reuse its retained source files and harvest convention. We independently rebuild plot means and test annual loss measures in 2011–2021. Its reported late biomass decline informed our question and period.