Term of Award
Master of Science
Document Type and Release Option
Thesis (restricted to Georgia Southern)
Department of Biology
Lissa M. Leege
Committee Member 1
Donald J. Drapalik
Committee Member 2
Lorne M. Wolfe
Committee Member 3
C. Ray Chandler
Committee Member 4
Stephen P. Vives
Feral horses range throughout Cumberland Island National Seashore, Georgia, and may have detrimental impacts on vegetation important in stabilizing the dune and salt marsh habitats. Areas accessible to horses were calculated with aerial photographs and GIS technology. During summer 2000,1 surveyed dunes and salt marshes of Cumberland to determine species composition and the extent of horse impact on vegetation. Species composition, species richness, percent cover, and grazed and ungrazed species across the 24 kilometers of coastline were sampled in 1 X 2m quadrats in two subhabitats: foredune/interdune and high/low marsh. Dune surveys measured and analyzed species richness, total percent cover of vegetation, and percent cover of species observed grazed. Salt marsh surveys measured and analyzed species composition, species richness, percent cover, and height of S. altermflora throughout the island. During the following summer (2001), six randomly selected 40 X 40m horse exclusion fences (one foredune, three interdune, and two salt marsh) were installed within the horse-accessible areas, and vegetation inside and outside was sampled over a 4-5 month period ending in October.
GIS analyses revealed that horses have access to 6,333 ha of the 9,633 ha comprising CINS. Of all horse accessible area of CINS, the dune and salt marsh account for 14 and 3%, respectively.
While grazing was observed on most grass species in the dune, the most frequently grazed species included Uniola paniculata, Spartma patens and Panicum amarum in the foredune. Species richness and percent cover in vegetated plots in the foredune averaged three species and 20% vegetation cover. Similarly, grazing in the interdune was observed on most grass species, and the top three grazing species included Conyza canadensis, Cyperas spp., and Uniola paniculata. The average number of interdune species and percent cover in vegetated plots varied throughout the island and was double that of the foredune (averaging 6 species, and 40% cover).
Horses grazed Spartina alterniflora in 29 % of the high, and 72 % of the low marsh subhabitats. Horses removed S. alterniflora throughout the island and percent cover of both total vegetation and S. alterniflora differed significantly across the island. While height of ungrazed S. alterniflora differed by island position, grazed height remained uniform across the island. Impacts based on percent tissue removal were observed throughout the island and were greatest in the north.
Foredune exclosure analyses revealed that vegetation cover is significantly reduced where horses selectively graze the important stabilizing species, U. paniculata. Though the interdune varied between exclosure site locations, grazing generally occurred on 11 of 17 grass species, and > 2 species of 11 sedge and rush. Comparative analyses since 1972 when over 300 head of cattle were present revealed species composition has shifted from a Paspalam vagmatum dominated foredune to one colonized by U. paniculata. Interdune species compostion was similar to that of 1972 dominated by Phyla nodiflora, and Hydrocotyle bonarensis, however, results indicate composition of other dominant species has changed. Salt marsh exclosure analyses revealed horses significantly reduced the growth and reproduction of S. alterniflora in areas where grazing occurs frequently.
Heavy removal of U. paniculata from the dune and S. alterniflora from the salt marsh could negatively impact Cumberland since dune and salt marsh ecosystems are important for reducing storm damage as well as providing habitat for both terrestrial and aquatic organisms.
Dolan, Peter Andrews, "Effects of Grazing by Feral Horses on Dune and Salt Marsh Vegetation at Cumberland Island, Georgia" (2002). Legacy ETDs. 855.