2018 ESA Annual Meeting (August 5 -- 10)

COS 100-8 - Can invasion be reversed by removing the main driver or has a regime shift occurred? A test case using a simulated wetland ecosystem

Thursday, August 9, 2018: 10:30 AM
238, New Orleans Ernest N. Morial Convention Center
Jason P. Martina, Texas A&M University, College Station, TX, William S. Currie, School For Environment and Sustainability, University of Michigan, Ann Arbor, MI, Kenneth J. Elgersma, Biology, University of Northern Iowa, Cedar Falls, IA and Deborah Goldberg, Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI
Background/Question/Methods

Plant invasions in the wetlands of the Great Lakes Region are known to be strongly influenced by nitrogen loading. The invaders Phragmites australis and Typha x glauca are often managed using fire, herbicide, and mowing with varying degrees of long-term success, likely due to the persistence of N loading. However, even if nitrogen loading decreased to pre-invaded levels the wetland ecosystem could potentially stay in an invaded state because a regime shift has occurred in nitrogen cycling. We studied the potential for such a regime shift using the MONDRIAN model, an individual based model that simulates growth and competition among individual ramets. We explored the success or failure of simulated invasion into a 3-species native community by both invaders across a range of N loading scenarios. These included constant (4 to 30 g N m-2 yr-1) or decreasing N loading where the system starts in a eutrophied state (30 g N m-2 yr-1) and ends in a low N loading state (4 g N m-2 yr-1) across a 50-year time period. We compared these scenarios to determine if invasion into a eutrophied wetland produced a regime shift, keeping N cycling higher even after a decrease in N loading.

Results/Conclusions

In the constant N loading scenarios, Typha and Phragmites were unsuccessful at low N loading, but formed monotypic stands at high N. An invasion threshold was observed between 8 and 12 g N m-2 yr-1 for both species. At a constant but low rate of N loading (4 g N m-2 yr-1), invaders were able to persist, but were a minor part of the modeled community at 115 g biomass m-2 yr-1 (~20% of community biomass). Interestingly, in the decreasing N loading scenarios, where the ending N loading rates were 4 g N m-2 yr-1 for ten years, the invader biomass was stable and intermediate between the 30 and 4 g N m-2 yr-1 N loading rates. Invader biomass was 410 g biomass m-2 yr-1 at the end of these simulations, which was 3.5 times greater than the constant N loading scenarios. These results suggest that a regime shift occurred and removing the main driver of invasion, N loading, did not return the system to a pre-invaded state, though invasion was reduced. The regime shift was likely caused by internal N cycling ramping up due to invasion.