biological diversity. Under the Farm Bill, wetlands are restored on areas previously converted to agriculture through (1) easements on Farmers Home Administration (FmHA) lands and (2) enrollment of lands in the Conservation Reserve Program (CRP). Conservation easements established by FmHA are administered as a part of the National Wildlife Refuge System. The CRP program provides for costsharing of bottomland hardwood establishment on flood-prone croplands. Agreements with private landowners are for a minimum of 10 years.
Other support for bottomland forest restoration results from the Water Resources Development Act of 1986, the North American Waterfowl Management Plan, and the Environmental Protection Agency's policy of no net loss to wetlands.
Most bottomland forest restoration projects focus on techniques of planting and establishing forest species. Some of the most extensive research in this area has been conducted at the Southern Hardwoods Laboratory and in the Delta Experimental Forest (e.g., Johnson and Krinard, 1987; Krinard and Johnson, 1987; Krinard and Kennedy, 1987). A critical factor is to achieve adequate hydrological conditions for forest establishment and development. Other important factors may include substrate stability, availability of adequate soil rooting volume and fertility, and control of herbivores and competitive weeds (Clewell and Lea, 1989).
Restoration success is commonly judged, at least in the early phases, by the success of tree seedling establishment. For example, Allen (1990) reported densities ranging from 87 to 914 trees per acre in 10 stands of 4 to 8 years in age in the Yazoo National Wildlife Refuge. He found survival and growth of planted seedlings to be generally higher than those obtained from direct seeding.
The goal of duplicating an original forest stand in terms of species composition and age, structure, and function can only be approximated. Natural forests are themselves in constant flux. Also, land use activities may have modified soil or hydrologic conditions to the point that duplication of the original forest is impossible and an altered forest community is the only option.
Clewell and Lea (1989) recommend five criteria for judging success: