Selective method for invasive plant removal enhances restoration

Effective and sustainable restoration of habitats invaded by non‐native plant species requires both invader removal and recovery of native species. However, efficacy of removal methods and native species responses commonly depend on site conditions, spatial scale, and time, indicating invader removal and restoration approaches must consider environmental context. To better understand how selectivity of invader control methods and site conditions determine invader removal efficacy and resident community responses, we conducted an experiment with a widespread plant invader in the southeast United States (cogongrass, Imperata cylindrica ) and co‐occurring resident species across nine sites that varied widely in environmental context. We found that three applications of the commonly used broad‐spectrum herbicide glyphosate reduced invader cover by 97%, on average, across the sites over 2 years. Fluazifop‐P‐butyl, a selective grass‐specific herbicide was equally effective (95% reduction in cover) in removing the invader when only two additional treatments were applied (five treatments total over 2 years) but it also helped preserve or allow for recovery of resident plant cover. Furthermore, removal efficacy was similar across sites regardless of differences in environmental conditions. Higher resource availability, as estimated by initial invader biomass at the sites, resulted in greater resident species recovery, but only under the grass‐specific herbicide treatment. Our results suggest that refocusing invasive plant management efforts from removal alone to community‐wide native species preservation and recovery, in part by using more selective invader removal strategies, has higher initial management costs but could simultaneously ease invader impacts and enhance ecosystem restoration.