Invasive plant removal method determines native plant community responses

1 Restoration of habitats invaded by non-native plants should include both the removal of invasive plants and re-establishment of native plant communities. To develop appropriate restoration strategies and quantify the effects of invasions, experiments that evaluate multiple removal methods and native community responses to those removal methods are needed. 2 We evaluated the response of native plant communities to removal of the invasive grass Microstegium vimineum (Japanese stiltgrass) in eastern forests in the USA. At eight field sites in southern Indiana, we applied three common removal treatments and compared native community responses among treatments and to untreated reference plots. 3 After 2 years of treatment, native community responses to Microstegium removal varied significantly among methods and plant functional groups in autumn 2006. Graminoid richness was greater when the invader was removed with hand-weeding, while graminoid biomass was lower in plots treated with post-emergent herbicide compared to reference plots. Forb richness was greater with hand-weeding and post-emergent herbicide compared to plots treated with post-emergent plus pre-emergent herbicides or untreated plots. Forb biomass was greater across all removal treatments. Overall native community diversity was 24% greater when the invasion was removed with hand-weeding and 21% greater with post-emergent herbicide compared to reference plots. No positive response in plant diversity occurred with post-emergent plus pre-emergent herbicide. 4 By spring 2007, graminoid percentage cover was greater with hand-weeding but not with herbicide treatments compared to untreated plots. However, forb cover was greater across all removal treatments compared to plots where the invader was not removed. The density of native tree seedlings was 123% greater in post-emergent herbicide treated plots than in untreated plots, indicating that the invasion was inhibiting tree recruitment. 5 Synthesis and applications. Our results demonstrate that multiple techniques can be used to control invasive plants but that the responses of native plant communities vary among removal methods. Further, greater native plant diversity and biomass following removal shows that invasions were suppressing native plant communities. Management of plant invasions should consider not only the effectiveness of removal methods but also how different methods influence native plant responses.