Data Package Metadata   View Summary

Vegetation sampling: In 2017, we sampled all vegetation in each treatment plot (n = 12 per site) and one reference plot (n = 1 per site) at each forest site, prior to invasive plant removal. All plots were revisited for post-removal vegetation sampling in 2018 and 2019 to assess the responses of the plant community to the various treatments. Herb and tree species were identified, and individuals counted, whereas individual stems were counted for all shrub species. For vines, graminoids (grasses and sedges), and ferns, percent ground cover was estimated since it is often difficult or impossible to determine individuals. Spring ephemeral plants were counted every year from mid-March to mid-May, and all other plants were sampled in peak growing season starting in July.

Floristic Quality Assessment: A Floristic Quality Assessment (FQA) was performed for all treatment plots at each forest site following the equation for Native FQA described in Freyman et al. (2016). The FQA relies on coefficients of conservatism (<i>C</i>), originally developed by Swink and Wilhelm (1994), which are a value between 0-10 assigned for all plant species within a given region or state. Invasive plant species, by definition, have a <i>C</i> of 0. Low FQA values represent species that are disturbance and lower quality habitat tolerant (i.e., weedy or ruderal species). High FQA values indicate plant species that occur in pristine, undisturbed, or true remnant habitat (i.e., not disturbance tolerant). Floral databases for Pennsylvania and Delaware (McAvoy 2018), which includes native status of plants and coefficients of conservatism, were accessed and downloaded from the Universal FQA Calculator (https://universalfqa.org/; Freyman et al. 2016). Seven taxa were only identified to genus and were not included in FQA calculations.

Soil Seed Bank: Soil collection for seed bank evaluation was conducted in April 2018. At each site, 10 soil cores were taken at random points along each edge (0-20 m) of the removal zone’s perimeter, with each sample taken at least 1 m away from the previous one (n = 40 cores site-1). Soil cores were collected with an 8 cm diameter soil auger to a depth of 10 cm, which is a sufficient sampling depth for capturing viable seeds in forest soils (Kostel-Hughes et al. 1998, Leckie et al. 2000, Decocq et al. 2004, Wódkiewicz and Kwiatkowska-Falińska 2010). Soil was bulked and homogenized in the field using a 5 L bucket at each forest site. Soils were transported on ice to the University of Delaware’s Fischer Greenhouse and stored at 2 °C. In preparation for the seed germination study, soils were gently sieved (4.75 mm) to remove rocks and other debris, and no seeds larger than 4.75 mm were present in the soil. The resulting sieved soil was then divided into eight flats (54 cm × 26 cm) per forest site, with 1.5 cm of soil (2.3 L flat-1) placed over 3 cm of sterile potting mix (Metromix 852, Sun Gro Horticulture, Agawam, MA 01001). In addition, 6 control flats contained an equivalent volume of sterile potting mix (4.5 L flat-1). Each tray was watered to moisten both soil and potting mix, and soils were watered every other day, or when dry to the touch, with 700-800 mL of tap water for the duration of the study. Daytime and nighttime temperature (28 °C and 26 °C, respectively) and relative humidity (50% and 60%, respectively) were set to mimic peak growing season conditions in the region. Seedlings grew until they could be identified, or until their growth impeded subsequent seedling growth. Seedlings that required longer growth for positive identification were transplanted into individual containers until identified. Specimens of each species were pressed and archived for future use. In the winter, flats were moved from the growth chamber to a covered shade house at University of Delaware Botanic Gardens. This transfer assists with physiological dormancy for species that require a second cold stratification. All flats were covered for protection from weather, potential seed predation, and wind-blown seeds. The following spring (April 2019), flats were transferred back to a growth chamber under the same environmental conditions as the previous year. After a period of two weeks with no seed germination, flats were fertilized with 800 mL of 173 ppm potassium nitrate to mimic natural N addition to soils in field conditions (Benech-Arnold et al. 2000, Decocq et al. 2004, Finch-Savage and Leubner-Metzger 2006, Beauchamp et al. 2013). After two more weeks with no germination, soils were discarded. The study concluded by July 2019.