Effects of repeated burning of cheatgrass (Bromus tectorum) dominated ecosystems on plant density, biomass and seed production: Implications for restoration

Background/Question/Methods

Restoration of sagebrush ecosystems dominated by cheatgrass depends on both controlling the invader and providing the conditions for native species establishment. Reducing available soil nitrogen (N) decreases cheatgrass growth and reproduction and native species opportunities for establishment. A mechanism for decreasing plant available N is repeated burning. Fire causes a loss of N by volatilization from burned plant material, but increases NH₄ in the short-term and NO₃ in the long-term. Increased available N results in greater cheatgrass growth, but is accompanied by higher cheatgrass N content and greater ecosystem N loss when burned in subsequent years. We hypothesized that repeated burning would decrease available N and lower establishment, growth and reproduction of cheatgrass. We conducted a repeated burning experiment in north-central Nevada on two Wyoming big sagebrush sites that are dominated by cheatgrass with some annual forbs. We used a completely randomized 2x4 factorial design with four replications of each treatment combination. The factors were two levels  of litter removal(removal or not) and four levels of burning and seeding (burning and seeding with cheatgrass, burning and seeding with wheat, burning only, and no burning or seeding). We analyzed the data with mixed model ANOVAs.

Results/Conclusions

Cheatgrass density increased from 2008 to 2011 except in plots seeded with wheat (df=3,114;  F=21.97;  p<0.0001).  The general increase was probably due to high soil moisture associated with relatively high precipitation from 2008 to 2011. Cheatgrass biomass also increased in all four years but only in control plots.  An increase in biomass occurred from 2008 through 2010 in plots that were burned, but biomass declined in 2011 (df=9,116 ; F=6.35; p<0.0001), indicating that individual plant size was decreasing in plots that were burned.  Cheatgrass seed production showed a similar pattern to biomass increasing in all four years in control plots, but decreasing  in plots that were burned in 2011 (df=9,116; F=2.91, p<0.0038).  These results support the predicted initial increase of cheatgrass following the first years of burning followed by a decline.  The mechanism was a decrease in growth and reproduction, and not lower establishment. Those plots seeded with wheat responded more quickly to the burning treatment, probably due to strong competition of wheat with cheatgrass.  Our results indicate that repeated burning may be a useful tool in the restoration of cheatgrass dominated sites, especially if accompanied by seeding with a strong competitor like wheat.