Grassland restoration under global changes – including nitrogen deposition and the spread of invasive species – requires tools that can be sustained and effective in the face of these drivers. This challenge is magnified in ecosystems characterized by high spatiotemporal variability, as management to address these novel drivers may affect a system differently across its range of variability. California serpentine grasslands epitomize this challenge: they host a high diversity of native species, are heterogeneous in space and time, and are experiencing high levels of atmospheric nitrogen deposition that can promote the conversion from systems dominated by native annual forbs to competitive non-native annual grasses. These changes generate high amounts of litter biomass from non-native annual grasses that prevents the recruitment of native plant species. Using a long-term monitoring dataset, we test the interactive effects of grazing and fire on native serpentine species following annual grass invasion and their litter accumulation. We assess management outcomes (burned-grazed, burned-ungrazed, and unburned-ungrazed) for their potential as restoration tools.
Results/Conclusions
A 2004 wildfire led to a reduction in annual grasses and a transient increase in native species (forb) richness. In 2008, a cattle-grazing treatment was reintroduced and crossed with the burn legacy, which sustained post-fire diversity and eventually led to native species recovery in unburned areas, with a caveat that a period of high precipitation promoted the growth of annual grasses and subsequent litter build-up. Our study indicates that serpentine seedbanks may persist despite a period of annual grass conversion, that short-term management such as fire can promote recovery from the seedbank, but that ongoing treatments like grazing are necessary to maintain native species diversity. Furthermore, Plantago erecta, which is a species of restoration concern as it is the primary host plant of the federally threatened Bay Checkerspot Butterfly, increased immediately following fire, whereas grazing reintroduction appeared to have a lagged effect in which P. erecta did not recover for six years. While short-term experiments that cover a small number of years may uncover successful management for particular weather conditions, we show the importance of long-term studies that assess intervention success across the range of natural variability. In the context of ongoing nitrogen deposition, we highlight that active, recurrent management is needed to control litter levels and maintain native plant diversity in this system.