Multiple global change drivers interact to control invasion dynamics in rare serpentine grasslands

Eliza Hernandez, Environmental Studies Program, University of Oregon, Eugene, OR, Andrew J. Muehleisen, Institute of Ecology and Evolution, University of Oregon, Eugene, OR and Lauren M. Hallett, Environmental Studies Program and Biology Department, University of Oregon, Eugene, OR

Background/Question/Methods
Global change drivers, such as nitrogen (N) deposition and climate change, are altering the availability of limiting resources in ecosystems worldwide. In California, nutrient-poor patches of serpentine grasslands have long served as refugia for native forbs from competition with invasive resource-acquisitive grasses. Increased resources from global change, however, can favor the invasion of acquisitive species. While N deposition acts as continuous resource addition, increased interannual rainfall variability under climate change also creates conditions that may favor either acquisitive or conservative species at different times: wetter years may promote invasive dominance and dryer years can suppress invaders. Here we test the interactive effect of N deposition and rainfall variability on the competitive success of a native serpentine forb, Plantago erecta, and an invasive annual grass, Bromus hordeaceus. In a greenhouse experiment we manipulated N (low, intermediate, high) and water (dry, wet) in a fully factorial design. Within each environmental treatment we manipulated the absolute and relative densities of Plantago and Bromus and measured per capita seed production. We used these data to fit Beverton-Holt competition models and simulate each species’ growth rate when rare (GRWR) when the other was at equilibrium under all nitrogen-water combinations.
Results/Conclusions
In invasion analysis, species coexist when species are mutually invasible, or each have a positive GRWR. We found that the native forb Plantago had a positive GRWR across all N and water combinations; in other words, it could always coexist. In contrast, the success of the invasive grass Bromus depended jointly on N and water. Bromus always had a negative GRWR under low N and a positive GRWR under high N, regardless of water. Under intermediate N, however, Bromus had a GRWR near 0 in dry conditions but had a positive GRWR in wet. Our results indicate that Bromus will always invade serpentine grasslands past a threshold level of N deposition, but at lower levels of deposition its invasion success depends on rainfall. These results parallel patterns we have observed in long-term data, in which Bromus invasion patterns have been cyclical but increasing over time. We conclude that wetter years provide a window of opportunity for Bromus proliferation under even moderate levels of N deposition, but that dry years can subsequently curtail the invasion. However, previous studies indicate that Bromus litter suppresses Plantago, suggesting that invasions in wet years may instigate feedbacks that ultimately maintain Bromus regardless of subsequent drought.