How effective are large-scale invasive species removal programs? A case study from the Berkshire/Taconic Plateau, western Massachusetts

Background/Question/Methods Large-scale invasive species removal programs are not common, and research on their efficacy is even rarer. Here we report data on native and invasive plant species abundance six years after The Nature Conservancy (TNC) began a glyphosate-based invasive plant removal program on the forest periphery of the Berkshire/Taconic Plateau region in western Massachusetts. In the late 1990's, increases in Berberis thunbergii (Japanese Barberry), Celastrus orbiculatus (Oriental Bittersweet), Alliaria petiolata (Garlic Mustard), Frangula alnus (Buckthorn), and Lonicera morrowii (Honeysuckle) abundance led TNC to pilot a removal program on 9,000 acres of federal, state, and privately protected land with the goal of reducing invasive cover to less than 5%. We asked whether this treatment, begun in 2002, had achieved its goals six years later. We determined invader percent cover using gridded quadrats on 10, 100m long randomly placed transects in an treated/untreated pair of otherwise comparable sites, and in two additional treated sites for which there was no appropriate untreated comparison. We also measured invader presence or absence over 200 1m² blocks adjacent to each transect for a total of 2000m²/site. Results/Conclusions Invasive species percent cover, as measured in 11 1m² quadrats per transect, was 4.2%±2.1 and 3.4%±1.7 percent, and not significantly different between adjacent and comparable treated and untreated sites (p=0.81). Similarly, 8% of the 2000 meter square cells adjacent to the untreated transects, and 17% adjacent to the treated transects, contained at least one invader. This difference was not significant (p=0.33). The two additional treated sites had mean invader percent cover of 24.9%±3.6, and a presence/absence cover of 62.9±6.8%. Of the five invaders, bittersweet and garlic mustard were most abundant in the treated sites, indicating that perhaps treatment was more effective for the other species. Garlic mustard's dominance in all treated sites indicates that an herbicide-based approach is unlikely to be successful in reducing its cover. Overall, invaders were significantly more abundant near the trail in the untreated site, and more abundant further from the trail in treated sites, but these patterns were not consistent by species. Our results indicate that only one out of three treated sites was below 5% invasive species percent cover six years after treatment. We suggest that either additional or different treatments will be required to meet this goal at a regional level.