The vegetation of restored and natural prairie wetlands

Thousands of wetland restorations have been done in the glaciated midcontinent of the United States. Wetlands in this region revegetate by natural recolonization after hydrology is restored. The floristic composition of the vegetation and seed banks of 10 restored wetlands in northern Iowa were compared to those of 10 adjacent natural wetlands to test the hypothesis that communities rapidly develop through natural recolonization. Restoration programs in the prairie pothole region assume that the efficient-community hypothesis is true: all plant species that can become established and survive under the environmental conditions found at a site will eventually be found growing there and/or will be found in its seed bank. Three years after restoration, natural wetlands had a mean of 46 species compared to 27 species for restored wetlands. Some guilds of species have significantly fewer (e.g., sedge meadow) or more (e.g., submersed aquatics) species in restored than natural wetlands. The distribution and abundance of most species at different elevations were significantly different in natural and restored wetlands. The seed banks of restored wetlands contained fewer species and fewer seeds than those of natural wetlands. There were, however, some similarities between the vegetation of restored and natural wetlands. Emergent species richness in restored wetlands was generally similar to that in natural wetlands, although there were fewer shallow emergent species in restored wetlands. The seed banks of restored wetlands, however, were not similar to those of natural wetlands in composition, mean species richness, or mean total seed density. Submersed aquatic, wet prairie, and sedge meadow species were not present in the seed banks of restored wetlands. These patterns of recolonization seem related to dispersal ability, indicating the efficient-community hypothesis cannot be completely accepted as a basis for restorations in the prairie pothole region.