Effects of eleven year (2004-2014) garlic mustard (Alliaria petiolata) removal on native vegetation

Background/Question/Methods

The biennial plant, garlic mustard (GM), is considered to be a problematic invasive species in Eastern North American deciduous forests and may cause declines in native species abundance and richness.  Recent studies suggest that GM is an opportunistic species and increases in abundance under conditions of high deer densities, because deer browse native species, but not GM.

We evaluated intraspecific competition between 1^st- and 2^nd-year GM plants and the effect of removal of 2^nd-year plants on native vegetation in 240 study plots that were established in 2004. From 2005-2014, 2^nd-year plants were removed annually from one-third of the plots in an early treatment (March 4-8) and a late treatment (May 7-18). Removal was done after 2^nd-year and native plants were well established. Remaining plots were non-manipulated controls. Percent cover of all plant species was recorded from early- to late-May annually from 2004-2014.  To evaluate the effect of changes in deer density on GM, we used deer/vehicle collision data on adjacent roads as a surrogate for deer density.

Results/Conclusions

Early removal of 2^nd-yr plants resulted in a significant increase of 1^st-year plant cover compared to late removal and controls in 2006, indicating 2^nd-yr plants strongly compete with 1^st-year.  However, by 2007 and later, early- and late-removal treatments had significantly less cover of 1^st-year GM than the control.  For native species cover, there were significant time and treatment effects, but no two-way interaction.  Cover of native species for control, early, and late treatments were (mean±SE) 77.5±1.2%, 84.0±1.2%, and 82.6±1.2%, respectively.  In control plots, between 2004 and 2007, except 2005, there were significant negative correlations between GM and native species cover.  However, from 2007 to 2010 GM cover declined precipitously from 19.4±1.7% to 2.8±0.6%, and there were no significant correlations between GM and native species cover from 2008 to 2010.  In 2011, 2012, and 2013, there were modest increases in GM cover 3.7±0.5%, 5.8±1.3%, and 5.9±1.2%, respectively, and significant negative correlations between GM and native species cover.  These results suggest that modest increases in GM cover negatively affected native species.  We found a significant positive correlation between the regression coefficient of determination and proportion of 2^nd-year plant cover, suggesting 2^nd-year plants affect cover of native species more than 1^st-year plants.  Deer vehicle collisions were positively correlated with total cover of GM in control plots from 2006 to 2011, a period of rapid GM decline, supporting previous research that deer facilitate GM abundance.