Ecological tradeoffs in non-native plant management

Controlling non-native plants in natural areas should, ideally, not only reduce target species' abundance, but also benefit broader management objectives such as conserving native species, improving wildlife habitat, and maintaining ecosystem function. In this context, the effectiveness and non-target impacts of control strategies, such as broadleaf herbicides, must be weighed against the impacts of non-native plants themselves. We undertook this relative assessment for Centaurea stoebe, one of the most widespread and heavily managed non-native plant species in the Intermountain West, USA. While effectiveness and plant community impacts of herbicide treatment for C. stoebe have been assessed, field-based experiments quantifying community-level impacts of C stoebe are rare. In a three-year experiment in sagebrush-grassland communities of southwest Montana, USA, we found that the broadleaf herbicide, picloram, was highly effective at reducing C stoebe, but also caused a significant loss of native forb cover and a significant increase in non-native grass cover, primarily Bromus tectorum. There was a significant increase in native forb cover in response to manual removal of C stoebe, which would seem to indicate C stoebe had been suppressing native forbs. However, there was an equivalent increase in native forb cover with no treatment. In some communities, C stoebe appears to have a negligible effect on native forb and grass cover and richness. Depending on management objectives, the loss of native forb cover and potential secondary invasion may outweigh the benefits of reduced target non-native plant abundance; thus, highlighting an ecological tradeoff of non-native plant management in natural areas. (C) 2012 Elsevier Ltd. All rights reserved.