2021 ESA Annual Meeting (August 2 - 6)

Persistent decadal differences in plant communities restored under contrasting climate conditions

On Demand
Kathryn Eckhoff, University of Kansas;
Background/Question/Methods

The loss of native tallgrass prairie has increased the importance of successfully restoring this ecosystem to reverse ecosystem disservices caused by row crop agriculture. As interannual variability in precipitation increases in the US Midwest, understanding the effects of climate on the structure and function of reconstructed prairie is needed to achieve restoration goals. There is evidence that climate experienced in restoration establishment year causes variation in plant community composition. We examined the effect of interannual variability in climate on trajectories of community development in four prairies that were sequentially restored every other year using the same methods over the first five years of restoration (spanning a decadal establishment time period). Composition of the two oldest restorations, installed under contrasting climate conditions (average precipitation and drought), was compared for 9 years to determine whether initially different trajectories of community development persist over time. Plant species composition was measured by estimating the maximum percent cover of each species from spring and summer surveys each year in four subplots within each of four plots in each sequence. Cover was averaged over the subplots and analyzed with a repeated measures permutational multivariate analysis of variance (PERMANOVA), and visualized using non-metric multidimensional scaling (NMDS).

Results/Conclusions

The compositional trajectories of the four restorations moved in the same direction as age increased, but remained different from one another, with precipitation in the establishment year explaining their separation. Communities became more similar as restorations aged, but all communities in the fifth growing season remained distinct (PERMANOVA, P < 0.001). Compositional differences after five years of restoration persisted for nine years in the two oldest sequences (PERMANOVA, P<0.001), with communities showing relatively less change in their respective ordination space after the sixth growing season, suggesting more persistent community states rather than transient variation when restored under highly contrasting conditions. The oldest restored communities also appeared resilient to a severe drought that occurred during their seventh and ninth growing season. Persistent decadal differences between communities restored under different climatic regimes suggest initial differences in community composition can cause lasting differences in reconstructed prairie.