Date of Award

5-2013

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Environmental Science and Biology

Abstract

Sedge/grass meadow wetland restoration was conducted at three study sites located in about 4 ha of agricultural land recently acquired by New York State Department of Environmental Conservation (NYSDEC) adjacent to West Creek at its confluence with Braddock Bay in Hilton, Monroe County, New York. The restoration was conducted within a 75.35 to 75.60 m (IGLD1985) elevation range previously identified as capable of supporting sedge/grass meadow in Lake Ontario wetlands. This project consisted of an initial baseline survey during spring 2009, a seed-bank emergence study that began in September 2009 and terminated in early July 2010, restoration implementation during summer 2010, and follow-up after implementation during August 2010, 2011, and 2012. Data from other Lake Ontario drowned river- mouth wetlands and a study site at Kents Creek served as references.

Implementation at the three study sites began with disking in May 2010 to expose fresh soil and remove much of the old plant growth. Locally-sourced wetland seed mixes, plus seeds from Canada bluejoint grass (Calamagrostis canadensis) and tussock sedge (Carex stricta), were purchased, cold-stratified, and sown with shoulder-broadcast seed spreaders in June 2010 in the study site planted areas. Plugs of Canada bluejoint grass and tussock sedge were also hand-planted in the same areas. Sections of each disked site area were left unplanted and unseeded to serve as controls. At two of the sites, natural wetland remnants, near areas dominated primarily by river bulrush (Schoenoplectus fluviatilis) in 2009, were not disked, planted, or seeded.

Plant surveys were conducted in the study site planted, control, and natural wetland areas, as well as in the 2009 baseline survey, by sampling in randomly- placed 1m2 quadrats. Plant data (frequency and percent cover of species in 1m2 quadrats) were used to calculate Importance Values; species were classified according to the National List of Vascular Plant Species that Occur in Wetlands; statistical tests were performed to determine important species and total percent cover and species count differences among study site areas; and data from all three sites across all four years were analyzed by ordination using non-metric multidimensional scaling (NMDS) in sample x Importance Value matrices.

Fifteen of the 42 seeded/planted species, 38 remnant sedge/grass meadow associates, and 36 potentially problematic (agricultural weed) species were identified in study site community samples across years, and additional species continue to be found. Following restoration, seeded species diversity increased in each subsequent year, and potentially problematic species generally decreased each subsequent year. Drought conditions during 2012 likely affected survival of some wetland species with greater water demand. Control treatments on high-canopy, annual agricultural weeds by mowing at a height of about 30 cm also affected plant community changes. The seed-bank emergence study did not successfully predict ultimate community composition following implementation, likely because survival of plants from seed is often dictated by post-recruitment processes. Instead, seeded species, remnant vegetation, and nearby refuge populations seemed to contribute more to establishment in the planted areas than the original seed bank.

The NMDS ordination showed that the 2009 baseline plant communities had been displaced by 2010, likely as a result of implementation actions. The ordination also showed that overall communities in the planted areas at the three sites changed from year to year and largely converged with the unplanted controls by 2012, which suggests that remnant vegetation was highly influential and nearby refuge populations made contributions as seeded species spread throughout control and planted site areas.

Post-restoration sampling at the restoration sites identified 21 species that were found in the Lake Ontario drowned river-mouth wetland reference data base and six species sampled at the Kents Creek reference site. Reference data suggest that the restoration sites reflect sedge/grass meadow conditions but also contain many other species associated more commonly with disturbed sites.

The future plant community at these restoration sites will likely be dependent on survival and expansion of sedge/grass meadow species, as influenced by soil moisture and competition from remnant agricultural weed species. Prolonged drought could potentially extirpate many of the seeded/planted species, especially if those conditions occurred in successive years. Monitoring results showed that competition can be mitigated by repeated, well-timed mowing that cuts taller annual plants before seed set and opens the canopy for underlying sedges and sedge/grass meadow associates currently found beneath them.

Share

COinS