The accumulation of greenhouse gasses in the atmosphere is expected to warm the earth's climate at an unprecedented rate (Ramanathan 1988, Schneider 1989). If the climate models are correct, within 100 years the earth will not only be warmer than it has been during the past million years, but the change will have occurred more rapidly than any on record. Many profound changes in the earth's environment are expected, including rising sea level, increasing aridity in continental interiors, and melting permafrost.

Ecosystems are expected to respond variously to a rapidly changing climate. Tree ranges in eastern North American are expected to shift northward, and seed dispersal may not be adequate to maintain current diversity (Cohn 1989, Johnson and Webb 1989). In coastal wetlands, rising sea level from melting icecaps and thermal expansion could flood salt-grass marshes and generally reduce the size and productivity of the intertidal zone (Peters and Darling 1985).

As yet, little attention has been given to the possible effects of climatic warming on inland prairie wetland ecosystems. These wetlands, located in the glaciated portion of the North American Great Plains (Figure 1), constitute the single most important breeding area for waterfowl on this continent (Hubbard 1988). This region annually produces 50-80% of the continent's total duck production (Batt et al. 1989). These marshes also support a variety of other wildlife, including many species of nongame birds, muskrat, and mink (Kantrud et al. 1989a).

Prairie wetlands are relatively shallow, water-holding depressions that vary in size, water permanence, and water chemistry. Permanence types include temporary ponds (typically holding water for a few weeks in the springs), seasonal ponds (holding water from spring until early summer), semipermanent ponds (holding water throughout the growing season during most years), and large permanent lakes (Stewart and Kantrud 1971). Refilling usually occurs in spring from precipitation and runoff from melting snow on frozen or saturated soils (Figure 2). Annual water levels fluctuate widely due to climate variability in the Great Plains (Borchert 1950, Kantrud et al. 1989b).

Climate affects the quality of habitat for breeding waterfowl by controlling regional water conditions--water depth, areal extent, and length of wet/dry cycles (Cowardin et al. 1988)--and vegetation patterns such as the cover ration (the ratio of emergent plant cover to open water). With increased levels of atmospheric carbon dioxide, climate models project warmer and, in some cases, drier conditions for the northern Great Plains (Karl et al. 1991, Manabe and Wetherald 1986, Mitchell 1983, Rind and Lebedeff 1984). In general, a warmer, drier climate could lower waterfowl production directly by increasing the frequency of dry basins and indirectly by producing less favorable cover rations (i.e., heavy emergent cover with few or no open-water areas).

The possibility of diminished waterfowl production in a greenhouse climate comes at a time when waterfowl numbers have sharply declined for other reasons (Johnson and Shaffer 1987). Breeding habitat continues to be lost or altered by agriculture, grazing, burning, mowing, sedimentation, and drainage (Kantrud et al. 1989b). For example, it has been estimated that 60% of the wetland area in North Dakota has been drained (Tiner 1984). Pesticides entering wetlands from adjacent agricultural fields have been destructive to aquatic invertebrate populations and have significantly lowered duckling survival (Grue et al. 1988).

In this article, we discuss current understanding and projections of global warming; review wetland vegetation dynamics to establish the strong relationship among climate, wetland hydrology, vegetation patterns, and waterflow habitat; discuss the potential effects of a greenhouse warming on these relationships; and illustrate the potential effects of climate change on wetland habitat by using a simulation model.

The extent to which intensive management of the waterfowl resource will be needed in the future strongly depends on whether a changing climate exacerbates the current problem of waterfowl decline. Should this occur, efforts outlined the recent North American Waterfowl Management Plan between the United States and Canada to reduce the current decline (Patterson and Nelson 1988) may need to be redoubled in coming years.