Survey procedures for this study were first used by Frayer et. al. 1983. The method was reviewed and approved prior to its use by statisticians from the Fish and Wildlife Service, Forest Service, Soil Conservation Service and the Army Corps of Engineers. It has been employed for a series of national (Frayer et al. 1983, Dahl and Johnson 1991, Frayer 1991) and regional wetland status and trend studies (Frayer et al. 1989, Hefner and Brown 1984).

The Southeast regional status and trends study consists of a stratified random sample of 2,204 plots. Each plot is four square miles, or 2,560 acres, and is permanently established for periodic reevaluation. The samples are stratified based on physical subdivisions (Hammond 1970) (Fig. 2), and state boundaries (Fig. 1) with the addition of a coastal stratum along the Gulf of Mexico and Atlantic coasts. The coastal stratum was added to incorporate estuarine and marine wetlands that extended beyond the continuous land mass. Sample plots were allocated to each stratum in proportion to the expected wetland density based on estimates by Shaw and Fredine (1956). Table 2 shows the number of plots within each state. Table 3 shows plot distribution within the physical subdivision strata.

Table 2.
Distribution of sample plots and mean dates of aerial photographic coverage, by state.

Table 3.
Distribution of sample plots within physical subdivisions (Hammond 1970).

Aerial photography was the basic information source. Two sets of photographs were analyzed for each study plot. The mean years of the aerial photography for the study were 1974 and 1983 (Table 2). This nine-year interval was used for calculating average annual wetland change estimates. The 1970's photography was primarily black and white at 1:80,000 or 1:60,000 scale, while the 1980's images were principally color infrared at 1:58,000 scale.

Aerial photographs were interpreted and cover types delineated according to procedures developed by the National Wetlands Inventory (U.S. Fish and Wildlife Service 1990a; 1990b). Wetlands, deepwater habitats and uplands identified on the photographs were assigned to one of 16 categories listed in Table 4 and described in Appendix A. All changes were determined to be either natural (e.g. scrub/shrub wetland succeeding to forested wetland) or human-induced (e.g. conversion of wetland to residential development or agricultural usage). Non-wetland areas were assigned to one of three general land-use categories: agricultural, urban and "other."

Table 4.
Wetland, deepwater and upland habitat categories used in this study.

Delineations on the interpreted aerial photographs were transferred to overlays on 1:24,000-scale U.S. Geological Survey topographic maps. Measurements of the various categories were made and acreages recorded for analyses. Changes in wetland area from the mid-1970's to mid-1980's for each sample plot were also determined from these maps, measured and recorded. Regional and state estimates were developed from the sample plot data using the statistical procedures presented by Frayer et al. (1983). As with previous status and trends studies by the Fish and Wildlife Service, this study is a quantitative measure of wetlands. No assessment of wetland quality other than changes in areal extent was made.

Norris Dam, Tennessee
LACUSTRINE
TENNESSEE VALLEY AUTHORITY