USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Waterfowl and Habitat Changes
After 40 Years on the Waubay Study Area


Abundances of North American duck species have been declining since 1955 (CWS/USFWS 1986).

Breeding pair densities in the Waubay study area during 1992-93 did not mirror population trends of breeding ducks in North America during the same time.

Yet, despite high pair densities in the Waubay study area, brood densities during 1992-93 were lower than during 1950-53, and clutch success was below levels needed to sustain stable duck population sizes (Klett et al. 1988).

Many landscape and biological factors affect clutch success and the recruitment rate of ducks. Factors reported to increase clutch success and recruitment include but are not limited to

The landscape in the Waubay study area has undergone numerous changes since 1950-53, and these changes may explain the observed changes in duck abundances in the study area.

Major Changes Since 1950-53

Brood Densities and Hen Success.  Brood densities of all duck species, except those of redheads and canvasbacks, were lower in 1992-93 than in the early 1950s, even though pair densities of most species in 1992-93 were similar or larger than those during 1950-53.

Redheads had been federally protected from hunting in the Central Flyway, at least in some areas, for 7 years and canvasbacks for 12 years since 1969 (Spencer Vaa, State Waterfowl Biologist, SDGFP, Brookings, pers comm). This may help explain their higher breeding populations.

During the 1950-53 period, many wetlands were in the open-water phase and over-water nesting habitat was limited. However, a sufficient amount of quality over-water nesting habitat has been present in or near the Waubay study area for the past 20 years. These factors may partially explain the apparent increase in the abundances of redheads and canvasbacks in the Waubay study area.

Hen success of all duck species was lower in 1992-93 than during 1950-53. Possible factors contributing to lower brood densities and hen success will be addressed in the rest of this section.

Annual Tillage.  Much of the prairie pothole region was settled in the 1880s. Drainage of wetlands, conversion of grasslands to croplands, intensive cultivation, and excessive grazing all followed (Duebbert and Frank 1984). These factors are detrimental to waterfowl productivity (Milonski 1958, Miller 1971, Higgins and Kantrud 1973, Higgins 1977; Kirsch et al. 1978, Klett et al. 1988). Several studies show that converting grasslands to croplands contributes greatly to waterfowl declines.

Annually tilled acreage decreased substantially since the early 1950s, mostly because of conversion of cropland to idle grasslands enrolled in the CRP. The conversion of annually tilled land to CRP grasslands since 1950-53 should have increased the clutch success and brood density of upland nesting duck species in the Waubay study area in 1992-93 over those in 1950-53. Instead, clutch success was poorer and brood densities were lower in 1992 and 1993, suggesting that other factors were depressing the reproductive potential of ducks.

Conservation Reserve Program Grasslands.  The importance of large, undisturbed fields of grassland to ducks that nest in upland habitats has been well documented. The number of ducklings hatched from nests located in large blocks of cool-season grasses mixed with legumes can be as much as six times greater than the number hatched from lands containing less suitable nesting cover (Duebbert and Kantrud 1974).

Upland nesting ducks recently have had high clutch success on blocks of CRP grasslands in eastern South Dakota (Luttschwager and Higgins 1992) and in south-central North Dakota and west-central Minnesota (Kantrud 1993). CRP grasslands provide dense residual cover that is important for nest sites of early nesting species (Sowls 1955, Martz 1967, Duebbert and Lokemoen 1976, Krapu et al. 1979).

CRP grasslands replaced a large percentage of the annually tilled land on the study area since 1950-53. However, 10.9% of the remaining cropland is now planted to row crops (corn and soybeans). Row crops are poor nesting habitats for ducks (Moyle 1964), and the negative effects of row cropping may offset the positive benefits of CRP grasslands.

The remaining cropland is planted to small grain. Small grain fields are better habitat for upland nesting ducks than other types of annually tilled land (Higgins 1977, Duebbert and Kantrud 1987).

The kind, quality, and amount of upland nesting cover in the Waubay study area has changed substantially since 1950-53, but the different habitats may be compensatory in relation to the potential recruitment of ducks.

Predator Community.  High clutch success, especially in wet years, is needed by all species of ducks to increase population sizes, but predators can severely depress clutch success.

Mammalian predation has been the leading cause of nest destruction in several studies (Duebbert and Lokemoen 1976, Stoudt 1982, Higgins et al. 1992, Kantrud 1993, Solberg and Higgins 1993a).

Predation primarily by raccoons was the leading cause of destruction of over-water nests in the study area in 1992-93. On upland sites, red fox (Vulpes vulpes) and striped skunks (Mephitis mephitis) were the primary cause of nest destruction.

Raccoons are relatively recent inhabitants of the prairie pothole region (Sargeant et al. 1993). Historically, they occupied wooded hills and river valleys in the southeastern portion of the region until the 1940s and 1950s when they expanded their range because humans had altered the landscape (summarized by Sargeant et al. 1993).

The expansion of raccoons into formerly unoccupied areas has had an impact on the clutch success of ducks that nest over water. Olson (1964) found that the average clutch success by canvasbacks was 77% in southern Manitoba in 1953 when raccoons were absent but only 21% during 1959-61 when raccoons were common.

Raccoon survival is enhanced during harsh winters because the animals can use many unnatural den sites (old buildings, rock piles, junk piles, shelterbelts; Fritzell 1978) for shelter and cereal grains for food (Cowan 1973, Greenwood 1981), and both were abundant in the Waubay study area in 1992-93. Raccoons also use dense patches of cattails as winter den sites.

The number of coyotes (Canis latrans) has recently increased because of bans on 1080 poisoning and hunting from airplanes and snowmobiles, plus less hunting and trapping in response to low fur values (USFWS 1978, Sargeant 1982). Coyotes are common in and around the study area, and coyote numbers in northeastern South Dakota have nearly tripled since 1983 (SDGF&P Animal Damage Control, Pierre, unpub data).

Demographics.  The number of occupied farms and landowners in the Waubay study area decreased since 1950-53. Buildings are still present on most of the abandoned farms and may provide possible den sites for predators.

At the same time, the average farm size in the Waubay study area has nearly doubled. These changes could have indirect effects on breeding waterfowl. Fewer residents in the study area may reduce hunting and trapping of predators. Furthermore, without an economic incentive for hunting or trapping predators, most people choose not to spend much time pursuing these activities (Sargeant 1982, Sovada 1993). These changes may be increasing predator populations in the study area and contributing to lower clutch success.

Wetland Vegetation.  Cattails dominated many of the semi-permanent wetlands in the Waubay study area in 1992-93, whereas hardstem bulrush dominated during the early 1950s (Evans and Black 1956).

This is a major habitat change. However, cattails seem to be adequate nesting cover for ducks that nest over water because the largest percentage of nests over water in 1992-93 were in wetlands with dense, monotypic stands of cattail (residual and new growth).

Because data on clutch success from the 1950s were unavailable, we can only speculate that over-water clutch success was higher during 1950-53.

Many of the cattail-dominated wetlands in the study area have several smaller wetlands nearby that drain into them. Thus, these wetlands remain in the degenerating and lake marsh stages of the wet-dry cycle described by van der Valk and Davis (1978b) for long periods of time. Many of these wetlands no longer return to the important dry and regenerating marsh phases (van der Valk and Davis 1978b) except during extreme drought conditions.

Static water levels are optimum for cattail growth and result in dense, monotypic stands. Once established, either during an extreme drought or partial summer drawdown, cattails spread vigorously by vegetative means. Clones from a single plant of Typha latifolia can spread over an area as large as 58 m² within 2 years of establishment (Grace and Wetzel 1981).

Wetlands with a 50-50 interspersion of open-water and emergent vegetation (hemi-marsh) provide better food and cover resources for waterfowl (Weller and Spatcher 1965).

Solberg and Higgins (1993b) used Rodeo herbicide in an attempt to create a hemi-marsh environment in cattail-dominated wetlands of northeastern South Dakota. Most class IV wetlands in the Waubay study area in 1992 were completely choked with cattails and contained few open-water areas. However, in aerial photographs from the early 1950s, many of these wetlands were hemi-marshes, suggesting that they were more beneficial to breeding ducks because they were enhanced by better wetland vegetation and associated invertebrate populations (Voigts 1976, Nelson and Kadlec 1984) during 1950-53.

Minor Changes Since 1950-53

Introduction of Row Crops.  Corn, soybeans, and sunflowers were not planted in the study area during the early 1950s (C. Evans, Lumni Island, Wash., and K. Black, Panama City, Fla., pers comm), but they were planted on 10.9% of the annually tilled land in 1992-93.

Several other studies also showed higher row crop acreages. Nomsen (1969) in Iowa, Vance (1976) in southeastern Illinois, and Taylor et al. (1978) in Nebraska reported substantial increases in row crops over time in their study areas. These changes have caused a shift from many diversified farms with small field sizes to fewer farms dominated by large fields of row crops (Taylor et al. 1978).

Row crops generally require larger amounts of fertilizers and pesticides (Grue et al. 1988). Several studies showed that pesticides reduced survival of ducklings (Sheehan et al. 1987, Grue et al. 1988, Martin et al. 1991, Dieter 1993). This may have occurred in the Waubay study area.

Introduction of Alfalfa.  Alfalfa was not planted in the study area during the early 1950s (C. Evans, Lumni Island, Wash., and K. Black, Panama City, Fla., pers comm) but was growing on 3.4% of the tillable land in 1992-93.

Alfalfa is a nitrogen fixer and is regularly rotated with other crops in the Waubay study area. Farris et al. (1977) reported that alfalfa had largely replaced other hay crops in Iowa during 1950-74.

Alfalfa is often chosen as a nest site by upland-nesting ducks, and it is an important livestock forage. Alfalfa is usually harvested as hay three to four times annually. Many hens are killed by haying equipment each year, and nearly all of the clutches are left exposed to predators after cutting has occurred.

Therefore, if a substantial portion of dabbling duck hens nested in alfalfa in 1992-93, the reproductive output from this habitat type was probably low.

Trees and Shrubs.  The percentage of land in the study area in trees and shrubs increased since 1950-53 because of new shelterbelts planted near agricultural fields to help prevent soil erosion and to protect farmsteads from the wind.

In contrast, aerial photographs revealed that most trees and shrubs in the study area during 1950-53 were native and grew near wetlands. Only a few shelterbelts existed.

The greater number of trees and shrubs and the maturation of older shelterbelts in the study area since 1950-53 may be indirectly lowering duckling numbers by providing numerous den sites for raccoons and striped skunks (Cowan 1973, Fritzell 1978).

Wetland Drainage.  Frayer et al. (1983) estimated that 223,799 ha per year of palustrine emergent wetlands were destroyed from the 1950s through the 1970s, with large losses in Nebraska, North Dakota, South Dakota, and Texas. Dahl and Johnson (1991) estimated that 1,335,510 ha of palustrine emergent wetlands were lost in the U.S. between the mid 1970s and mid 1980s.

Many wetlands in the study area were drained prior to 1992-93, but we found no evidence of new drainage. Wetland losses in the Waubay study area were greatest before 1954 and after 1968. This is in agreement with other estimates. Wetland losses in northeastern South Dakota during 1974-80 were lower (1.5%) than those in southeastern portions of the state (7.6%; USFWS 1980).

Many wetlands (45.2%) in the Waubay study area were protected by federal easements, federal ownership (maps at Waubay National Wildlife Refuge), or by private landowners whose land use practices did not require drainage. This may explain the relatively low loss of wetlands in the study area. Higgins and Woodward (1986) found lower drainage rates on wetlands protected by long-term federal easements.

Most drainage (85%) in the Waubay study area was for agriculture, although some earlier drainage (15%) was for road construction. Our findings are in agreement with Frayer et al. (1983) who found that 87% of the wetland losses between the 1950s and 1970s were associated with agriculture. Wetlands are often regarded as obstacles and are drained to gain cropland or to accommodate large modern farm machinery (Aus 1969).

Adequate wetland habitat for breeding pairs and broods still exists in the Waubay study area, but previous wetland drainage may be indirectly reducing the clutch success of ducks that nest over water. For example, flooding was the second leading cause of destruction in over-water nests in the study area in 1992-93.

Short, intense summer storms are a characteristic of this region, and a substantial amount of rain may fall in a short period. Smaller wetlands often drain into larger seasonal and semi-permanent wetlands by open ditches, resulting in rapid rises in water levels in the larger wetlands after summer storms. Most ducks that nest over water are able to elevate their nests with slowly rising water levels, but rapidly rising water levels can flood them out (Stoudt 1982).

Increased nest flooding because of wetland drainage could be one reason for lower brood densities by ducks that nested over water in 1992-93, but this may also have been a problem during 1950-53, because considerable drainage had already been completed.

Wetland Restoration.  Since 1986, the USFWS has restored 28 drained wetland basins in the study area. These restored wetlands may benefit waterfowl populations by increasing the wetland habitat base for breeding pairs, nests, and broods and by decreasing the incidence of nest flooding.

Factors Remaining Unchanged Since 1950-53

Annual Precipitation.  Precipitation data from 1950-93 shows that the fourth wettest summer on record in South Dakota was in 1992 and the wettest was in 1993 (A. R. Bender, South Dakota Weather in 1992 and 1993, Climatological Report). Precipitation data from the Waubay National Wildlife Refuge indicate that the summers of 1950 and 1953 were also wet.

The second coolest summer on record in South Dakota was in 1992, the third coolest was in 1993, and the fourth coolest was in 1951 (A. R. Bender, South Dakota Weather in 1992 and 1993, Climatological Report).

The combination of cool and wet weather during the summers of 1992 and 1993 could have lowered clutch success, brood survival, and overall recruitment of ducklings. Frequent rain and cold temperatures can lower the survival of young waterfowl (Untergasser and Hayward 1972, MacInnes et al. 1974) and reduce clutch success (Stoudt 1971, 1982).

Haying and Grazing Practices.  The amount of land in the study area that was hayed or intensively grazed in 1992-93 was essentially unchanged since 1950-53.

However, new developments in haying equipment since the 1950s enable farmers to harvest forage more efficiently and more frequently. Haying destroys nests, kills hens, and removes residual nesting cover.

Hayed areas are usually dominated by cool-season forbs and grasses that do not regenerate immediately after cutting, thereby reducing the amount of residual cover available to early-nesting mallards and pintails the following spring.

All species of ducks that nest in uplands will begin nesting in mowed areas later than in unmowed areas (Martz 1967). Higher densities of duck nests also are found in residual cover associated with unmowed blocks of highway right-of-way than in mowed blocks in south-central North Dakota (Oetting and Cassel 1971, Voorhees and Cassel 1980).

Grazing can be beneficial or detrimental to ducks (Kirby et al. 1992). Sowls (1955) found that bluegrass pastures intensively grazed by livestock were useless to nesting ducks. Several studies have revealed higher densities of duck nests and higher clutch success in ungrazed fields than in grazed fields (Kirsch 1969, Miller 1971, Kirsch et al. 1978; Kaiser et al. 1979). However, Barker et al. (1990) found that nest densities of ducks and clutch success were nearly equal on idle fields and pastures subjected to twice-over rotation, short duration, and switchback grazing systems. Cattle also trample cattails and create openings in plant-choked wetlands, benefitting breeding waterfowl by increasing the number of loafing sites and the number of potential breeding pairs (Sowls 1955).

Since the amount of land in the Waubay study area being hayed or grazed has not changed since 1950-53, it does not seem that these factors are responsible for lower brood densities in 1992-93.

Breeding Pair Densities.  Breeding pair densities in 1992-1993 were greater than or equal to those during 1950-53 in the Waubay study area. Continent-wide breeding pair estimates did not follow a similar pattern, decreasing from 1992 to 1993 and remaining well below estimates from the 1950s.

Large numbers of breeding ducks in the study area in 1992 and 1993 paralleled good wetland conditions. Numerous studies have revealed sharp declines in pair densities on southern breeding areas during drought years and large increases in Arctic breeding areas during the same year (Smith 1970, Smith 1971, Stoudt 1971, Leitch and Kaminski 1985, Johnson and Grier 1988).

Northern pintails were the only species that failed to increase with the good wetland conditions in the study area in 1992-93. This suggests that the northern pintails are in a serious decline.

Since 1992-93 pair densities and wetland conditions were similar to those during 1950-53, we expected brood densities to be greater than or equal to those in 1950-53. However, 1992-93 brood densities were significantly lower.

When all environmental and cultural changes and their possible impacts on breeding waterfowl are considered, mammalian predation appears to be the factor most responsible for the lower reproductive output from upland and wetland habitats in the Waubay study area during 1992-93.

If mammalian predation is the main factor, further research is needed to find ways to limit their ability to destroy nests. For example, electric predator fences control predation and are socially acceptable (Lokemoen et al. 1982, Greenwood et al. 1990), but they are labor intensive, expensive to construct and maintain, and only applicable on a small scale. Techniques to reduce effects of predators on a larger area are needed, but they must be socially acceptable and economically feasible.

Previous Section -- Results
Return to Contents
Next Section -- Acknowledgments

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 20:03:37 EST
Reston, VA [vaww55]