No abstract

The restoration of oak woodlands often requires removal and management of woody invasive plants. This can trigger germination of the soil seed bank, which can alter restoration trajectories. In degraded deciduous woodlands generally, it is unclear whether the soil seed bank will contribute native plant material in sufficient quantities to help achieve restoration goals and allow practitioners to rely on passive restoration without supplemental seeding. To support restoration decision‐making around passive or active restoration in three Rhamnus cathartica‐invaded forest preserves in the Chicago region, we asked: (1) Does the soil seed bank differ from standing aboveground vegetation at reference and unrestored sites? and (2) Can the species richness, Shannon diversity, floristic quality (measured by abundance‐weighted mean coefficients of conservatism), or density of germinable seeds of native species in the soil seed bank be predicted by a site's restoration status (reference or unrestored)? We found that species composition differed significantly between aboveground vegetation and the soil seed bank at reference and unrestored sites, with a significant interaction between restoration status and location. Despite high variation among the three forest preserves, restoration status also predicted native species richness, diversity, floristic quality, and seed density in the soil seed bank, with unrestored sites significantly lower than reference sites in all measures. Results suggest that reintroduction of native seeds will be necessary to fully restore desired native plant communities in oak woodlands following the removal of invasive woody plants.

Shifts in plant-community composition following habitat degradation and species invasions can alter ecosystem structure and performance of ecosystem services. In temperate North American woodlands, invasion by aggressive Eurasian shrubs has produced dense thickets with depauperate understory vegetation and increased rates of litter decomposition and nutrient cycling, attributes that could impair storage of carbon as soil organic matter (SOM). It is important to know if such impairment has occurred and, if so, the extent to which restoration can return this service. We used an oak-woodland restoration chronosequence in northeastern Illinois to contrast structural and functional attributes of unrestored areas dominated by Rhamnus cathartica (common buckthorn) with areas that had undergone buckthorn removal and ongoing, active management for less than 1 to 14 years. With increasing age, restored areas had higher understory plant diversity and cover (p < 0.0001 and 0.005, respectively) and higher litter mass (p = 0.018). These structural differences were associated with some evidence of reduced soil erosion (p = 0.027-0.135) but greater soil CO 2 efflux (p = 0.020-0.033). Total particulate organic matter (POM) in the soil increased with restoration age, which was driven by increases in the slow-turnover, mineral-associated SOM fraction. However, variance was high and relationships were only weakly significant (p = 0.082 and 0.083 for total POM and mineral-associated SOM, respectively). Our results suggest that, in addition to better documented biodiversity benefits, beneficial changes to ecosystem properties and processes may also occur with active, long-term restoration of degraded woodlands.