Abstract

Invasive plants, especially non-native perennial grasses, are a critical threat to remnant prairies and oak savannas in the Pacific Northwest. Managers must control non-native plants without adversely impacting native species in fragmented prairie remnants. We describe results of a collaborative experiment replicated at 10 sites along a 500 km latitudinal gradient. Our objectives were to develop and test treatment combinations that reduce target non-native weeds with minimal nontarget impacts and increase native species diversity and abundance. By replicating experiments across the ecoregion, we tested strategies for widespread applicability. We compared four different combinations of seed addition and disturbance treatments comprising herbicide (sethoxydim and glyphosate), fire, and mowing. Each combination was created to target various factors likely to limit restoration in this system, including invasive species, litter accumulation, and limited dispersal of native species. After three years, the treatment combinations varied widely in their effectiveness. The most disturbance-intensive treatment combination (joint application of sethoxydim, burning, and postfire glyphosate) led to reduced abundance of non-native grasses and forbs without causing a decline in native species. Sethoxydim combined with fall mowing reduced non-native grasses, caused no change in non-native forbs, and increased total cover of native plants. In all cases, disturbance treatments reduced non-native cover to varying degrees but had no positive impact on native diversity except when seeds were added. Our results show that a combined treatment approach employing a variety of strategies codesigned by managers and ecologists is an efficient and effective way to improve degraded grasslands.

pdf