Date of Award

Fall 2000

Project Type


Program or Major

Natural Resources

Degree Name

Doctor of Philosophy

First Advisor

Frederick T Short


The small, fringing salt marshes that line the edges of estuaries in southern Maine and New Hampshire were the focus of this research. Although larger meadow marshes in New England have been studied extensively, little is known about the ecology of fringing marshes. Not only are fringing marshes much more numerous than meadow marshes in northern New England, they are often restored or created as mitigation for marsh impacts. Five ecological functions (primary production, soil organic matter accumulation, sediment trapping and binding, wave dampening and maintenance of plant diversity) were compared in meadow marsh and fringing marsh sites, and sometimes in areas where no marsh was present. Also explored were the relationships between these functions and several physical characteristics, including soil salinity, percent surface slope, elevation and size. Fringing marsh and meadow marsh sites differed significantly in terms of their physical characteristics, but functional indicator values were similar, with the exception of plant species richness and soil organic matter content.

A field experiment tested whether marsh surface slope or north-south orientation affects the growth of newly planted cordgrass (Spartina alterniflora) in fringing marshes. These experiments were not able to show that orientation or slope had an effect on plant growth.

Comparisons were also made between six constructed fringing marshes and a set of selected reference sites (matched to constructed marshes using principal components analysis) in the Great Bay Estuary. Four ecological functions (primary production, soil organic matter accumulation, sediment trapping and binding, and maintenance of plant diversity) were studied. Mean values for constructed site (n = 6) and reference site (n = 11) functions were significantly different. Because the age of the constructed sites ranged from 1--14 years, patterns of functional development could be examined. Using constructed marsh age as the independent variable and functional indicator values as dependent variables, non-linear regression analyses produced several ecologically meaningful trajectories (r2 > .9). These models illustrate that although indicators of some functions (primary production, sediment deposition, and plant species richness) should reach natural site values in less than ten years, soil organic matter content will take more than fifteen years to develop.