Wetland Birds in the Northern Great Plains

When the Wisconsin glacier retreated about 10,000 years ago, it left innumerable depressions scattered throughout the northern Great Plains. These depressional wetlands, called prairie potholes, contain water for various lengths of time in most years (Kantrud et al. 1989). Their size, permanence, hydrology, water chemistry, plant associations, and invertebrate communities vary widely among wetlands and, within a basin, through time (Cowardin et al. 1979).

These diverse wetlands support a breeding avifauna as rich and varied as the wetlands themselves. Johnsgard (1979) listed 72 breeding bird species associated with freshwater pond environments in the Great Plains. Other species, such as the northern harrier, marbled godwit, Le Conte's sparrow, and Nelson's sharp-tailed sparrow, are associated with grasslands but extensively use these prairie wetlands. Stewart (1975) identified 63 breeding bird species as wetland associates in North Dakota alone. Since 1975, several species could be added to Stewart's list (Faanes and Stewart 1982), including the reintroduced Canada goose (Lee et al. 1989) and several herons, egrets, and ibises that have expanded their breeding range into the state (Lokemoen 1979). Most wetland birds are short-distance migrants, wintering primarily north of the United States-Mexico border (Igl and Johnson 1995).

Wetland birds are not easily monitored by standard census techniques (Bibby et al. 1992). Dense vegetation reduces the visibility of some species. Many species lack territorial songs or rarely call; others that make diagnostic sounds do so primarily or only at night. Some species, such as rails, are notoriously elusive, even within a few meters of an observer (Burt 1994). Others are colonial, resulting in tremendous spatial variability in their numbers. Thus, no single technique works well for censusing all wetland species. Accurate censusing of wetland birds requires a variety of techniques, including nocturnal surveys, nest counts, intensive efforts involving walking or canoeing through marshes, and the use of recorded calls to elicit responses (Weller 1986). Recently, an informal group that monitors marsh birds has formed to address such issues.

Kantrud and Stewart (1984) surveyed breeding populations of wetland bird species other than waterfowl on 1,321 wetland basins in the prairie pothole region of North Dakota. Densities of each species were reported for six different classes of wetlands (as defined by Stewart and Kantrud 1971; Table 1). Four of the wetland classes (permanent, semipermanent, seasonal, and temporary) were distinguished by water permanence as indicated by the vegetative zone occupying the deepest part of the basin. Alkali wetlands were recognized by the occurrence of hypersaline surface water, and fens were identified by a characteristic zone of fen vegetation that develops where groundwater seeps saturate the soil. Most wetland species were found on semipermanent and seasonal wetlands, reflecting the variety of habitats within these wetland classes (Table 1). Although some species were found in all wetland classes, most species showed a preference for one or two classes.

Population estimates and trends of wetland bird species, exclusive of waterfowl, are limited. In 1967 Stewart and Kantrud (1972) conducted an extensive survey of breeding bird populations throughout North Dakota to obtain baseline estimates of statewide breeding bird abundance and frequency of occurrence. In 1992 and 1993 Igl and Johnson (1997) repeated the survey by using the same sample units and methods as the 1967 survey to examine changes in breeding bird populations. These data offer both overall population estimates and some indication of population changes from 1967 to now (Table 2). According to habitat affinities, wetland species composed the largest proportion (32%) of species and 22% of the observed breeding pairs during the three years covered in the two Igl and Johnson surveys. The species that declined in North Dakota were mostly grassland and wetland species, whereas increasing species were predominantly resident species and species associated with human structures and woody vegetation (Igl and Johnson 1995). Similarly, results from the U.S. Geological Survey Breeding Bird Survey for North Dakota showed that 23 of the 28 (82%) observed species with statistically significant decreasing trends in the state were associated with wetland or grassland habitats.

We obtained trends for abundance of 43 wetland birds from the Breeding Bird Survey (Robbins et al. 1986) for the central region (from the Rocky Mountains to the Mississippi River) during the entire survey period (1966-1994) and for two subperiods: early (1966-1979) and recent (1980-1994) (Table 3). Population increases outnumbered decreases for each of the three time intervals. The percentage of species with increasing trends was 67% during the early subperiod (1966-1979) and 55% during the recent subperiod (1980-1994). For the entire survey period (1966-1994), seven species increased significantly; these included mostly colonial-nesting species (American white pelican, double-crested cormorant, and three gull species). Five species decreased significantly; all of these frequently nest in emergent wetland vegetation (northern harrier, Franklin's gull, black tern, common yellowthroat, and red-winged blackbird).

These data are consistent with earlier reports showing that breeding bird populations in wetland ecosystems are as dynamic as they are rich. The divergent patterns observed among the species and studies reflect the species' disparate habitat requirements, geographic ranges, and unique responses to natural and anthropogenic changes in their environments. Determining the status and trends of wetland bird populations is a necessary first step toward the more daunting challenge of understanding the mechanisms that drive population changes.