Characteristics of central North Dakota wetlands determined from sample aerial photographs and ground study

Lewis M. Cowardin, David S. Gilmer, Larry M. Mechlin

1981 · 16 citations · 5 references

Concepts

Abstract

Wetland characteristics were assessed from a systematic sample of 66 plots, 3.22 km2 each, drawn from a 10,041-km2 study area in central North Dakota. Each plot was visited once, and 8 sets of aerial photographs were obtained in 3 years. Density of wetland basins was 11.00 ha/ kM2, and area averaged 9.7 ha/km2. Seasonal and temporary wetlands were most abundant; semipermanent wetlands occupied the greatest area. Basin size was positively correlated with water permanence. Discriminant function analysis based on size and an index to wetness derived from photographs misclassified 33% of the wetland basins. Forty percent of the wetlands were tilled. Photography of sample plots is potentially useful for determining number of basins and wetland area, but precise classification of plant communities in this region would require ground study. Although a limited number of North Dakota wetlands have been studied in detail (Stewart and Kantrud 1972, Swanson et al, 1974, Winter and Carr 1980), there have been relatively few estimates of wetland characteristics covering large areas of the Prairie Pothole Region. Shaw and Fredine (1956) presented estimates derived from ground transects taken along roads in a dry year (U.S. Fish and Wildlife Service 1955). These data are not only out of date but also failed to detect many wetlands because of the dry conditions at the time of inventory. Stewart and Kantrud (1973) presented estimates of the proportion of various wetlands in North Dakota according to their classification (Stewart and Kantrud 1971), but their ground-based sample is limited in number of sample units and total area sampled. Ruwaldt et al. (1979) presented I Present address: U.S. Fish and Wildlife Service, Dixon Lab, Route 2, Box 5210, Dixon, CA 95620. 2 Present address: R.R. 3, Box 232, Macon, MO 63552. similar estimates for South Dakota. Gilmer et al. (1980) used a double sampling procedure based on complete enumeration of large wetlands from Landsat data supplemented with sample data obtained from aerial photography to obtain estimates of the number of ponds (wetland basins containing water). Although probably accurate, their estimate was not designed to account for dry basins at the time of the survey and does not separate ponds by class. At present the U.S. Fish and Wildlife Service (FWS) is conducting a complete inventory of wetlands in the Prairie Pothole Region. This inventory should furnish good estimates of the extent of wetland and the number of basins, but classification of the wetlands will not always be possible because it will be based on photographs taken on 1 date and without ground data. To evaluate annual changes in habitat conditions for a mallard (Anas platyrhynchos) production study, we sampled 2,324 wetland basins and obtained data on density of basins, area of wetland,

References

5