USGS - science for a changing world

Northern Prairie Wildlife Research Center

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

Duck Nest Success on Conservation Reserve Program
Land in the Prairie Pothole Region

Results and Discussion

Crews found 520 duck nests—211 in 1989, 167 in 1990, and 142 in 1991 (Table 2). Major species were the blue-winged teal (57 percent), mallard (19 percent), gadwall (11 percent), northern pintail (6 percent), and northern shoveler (6 percent). Minor species—green-winged teal (Anas crecca) and American wigeon (A. americana), had less than one percent of total nests. Crews found more nests (62 percent) in the Central Lowlands than in the Great Plains. Central Lowlands plots had 75 percent of blue-winged teal nests and 71 percent of mallard nests, whereas Great Plains plots had 97 percent of northern pintail nests, 93 percent of gadwall nests, and 69 percent of northern shoveler nests. WPAs had about 80 percent of blue-winged teal nests, 75 percent of northern shoveler nests, 65 percent of gadwall nests, and 61 percent of northern pintail nests, and only about 43 percent of mallard nests. Crews found 0 to 43 (average 11.3) nests per WPA during the study, compared with 0 to 14 (average 3.3) nests per CRP plot. Twenty-six nests were unusable for determinations of nest success.

Table 2.  Numbers of duck nests found in prairie pothole region of Minnesota and North Dakota, 1989-1991.
  Central Lowlands (MN) Great Plains (ND)
Species 1989 1990 1991 1989 1990 1991 1989 1990 1991 1989 1990 1991
Green-winged Teal 0 0 0 0 0 1 1 0 0 0 0 0
Mallard 8 9 6 20 10 17 5 4 10 3 3 3
Northern Pintail 0 1 0 0 0 0 7 7 5 5 6 2
Blue-winged Teal 69 85 33 20 16 12 33 5 13 6 2 2
Northern Shoveler 3 2 4 0 0 1 8 2 5 5 2 0
Gadwall 2 0 1 0 0 1 10 6 18 6 7 6
American Wigeon 0 0 1 0 0 0 0 0 0 0 0 1
Total 82 97 45 40 26 32 64 24 51 25 20 14

My goal was to find 100 usable nests in each of the four combinations of cover type and physiographic region, for a total of 1,200 usable nests. I attribute the low number of nests to drought conditions that prevailed during the study, especially in the Great Plains. Indeed, pond numbers in southern North Dakota during 1989-1991 were nearly 50 percent below the long-term (1958-1988) average (U.S. Fish and Wildlife Service, unpubl. data). Only one study field in Stutsman county (Figure 1) was within 1.6 km (1 mi) of ponded water during 1991. Water conditions were also relatively poor in Minnesota, especially in 1989 and 1990 when pond numbers were well below the 1972-1988 average (Minnesota Department of Natural Resources, unpubl. data).

Of the 494 usable nests, observers judged 323 (65.4 percent) destroyed by predators, nearly all by mammals; 159 (32.2 percent) hatched; eight (1.6 percent) abandoned by the hen; and three (0.6 percent) destroyed by machinery. The fate of one nest (0.2 percent) was unknown. Because these numbers are unweighted, they do not reflect true distribution of nest fates. In recent decades, proportions of duck nests destroyed by predators in the prairie pothole region typically ranged from 50-80 percent (4, 6, 7, 13, 14). However, lower loss rates have been reported and often occurred on fields of retired cropland similar to those I studied (4, 8, 18).

Figure 2
Figure 2.  Primary sample unit, 1989-1991.

Nest success of 23.1 percent on CRP fields was significantly (F1, 16 = 7.28, P = 0.02) higher than the 8.2 percent success on WPAs. Interaction between physiographic region and cover was not significant (F1, 16 = 1.73, P = 0.21). This suggests that differences between cover types were consistent for both regions. No main effects or interactions involving year were significant. Nest success on WPAs did not differ between the pre-CRP (10.6 percent) and CRP (8.2 percent) eras or between regions (F3,18 = 0.39, P = 0.75). Nest success on CRP did not differ from that on WPAs during the pre-CRP era or between regions (F3,18 = 3.01, P = 0.057).

Sparse data made tests for differences in nest success among species futile. However, I show success rates for major species for combined years based on samples (fields with usable nests) unweighted for exposure days of clutches in those fields (Table 3). These estimates do not allow statistically valid comparisons among species between cover types. Success rates varied greatly and confidence intervals were wide because of the low numbers of nests found. Nevertheless, the estimates suggest greater nest success on CRP than on WPA cover, especially for the larger species (mallards, northern pintails, and gadwalls).

Table 3.  Mayfield nest success rates (95% confidence interval) for major duck species in the prairie pothole region of Minnesota and North Dakota, 1989-1991.*
Species CRP WPA
Mallard 25.1 (15.1-41.4) 9.7 (3.9-23.5)
Northern Pintail 37.4 (11.8-100.0) 3.9 (0.7-21.3)
Blue-winged Teal 11.6 (5.7-23.4) 9.0 (6.2-13.0)
Northern Shoveler 15.0 (2.1-95.1) 11.7 (3.8-34.9)
Gadwall 59.9 (37.7-94.4) 2.7 (0.7-9.9)
* Estimates were made by pooling all nests and therefore may be biased.

Comparisons of vegetation height-density measurements on the study fields revealed significant two-factor interactions between cover and landform (F1, 17 = 28.23, P < 0.001) and landform and year (F2, 45 = 10.95, P < 0.001). Robel values were higher in CRP cover in the Central Lowlands (X = 4.32, SE = 0.18) than in the Great Plains (X = 2.30, SE = 0.36) and higher than in WPAs in the Central Lowlands (X = 2.37, SE = 0.24). Values in Great Plains WPAs (X = 3.30, SE = 0.35) were higher than in WPAs in the Central Lowlands. No difference in Robel values between CRP and WPA cover was detected in the Great Plains (F1, 45 = 3.69, P = 0.061). Robel values for both CRP and WPA cover in the Central Lowlands, but not in the Great Plains, changed significantly (P < 0.05) from year to year and over the entire period. In the Central Lowlands, values decreased from 1989 to 1990 and then increased in 1991 (F1, 45 = 38.58, P = 0.001).

In 1989 and 1991 Robel values were higher in the Central Lowlands than in the Great Plains (1989 F1, 45 = 4.45, P = 0.041; 1991 F1, 45 = 25.69, P = 0.001); but in 1990, values did not differ between the two landforms (F1, 45 = 2.0, P = 0.165). Correlation analyses showed no evidence that nest success was related to Robel values on the study fields.

The changes observed in vegetation height-density on the study fields likely were related to precipitation, especially amounts received during the growing season, that is the most important factor affecting growth of stands of nesting cover in the prairie pothole region (9). Canopy cover, plant height, and Robel values differ with stand age, but year of maximum growth is variable and unpredictable. Although there is evidence that duck nest success increases with height-density of residual vegetation (9), no management recommendations applicable to various parts of the region have been formulated (11).

We can only speculate on a few of the many factors, all probably related to predation, that resulted in higher nest success on CRP cover. One factor could be distance to water. All WPA fields we studied bordered on at least one semipermanent wetland, whereas many CRP fields did not. CRP fields averaged farther from water than WPAs. Our data suggest that larger ducks had higher nest success on CRP cover than on WPA cover, whereas the smaller ducks, blue-winged teals and northern shovelers, had similar success in the two covers. In addition, we found greater proportions of nests of the small ducks on WPAs. Mallards and gadwalls usually nest farther from water than blue-winged teals (8, 15). The other large ducks we studied probably share this tendency. Wide separation of nests from water may be an adaptation of the larger dabbling ducks to avoid or minimize predation. Predator activity is probably greater near wetlands. Wetland borders are prime hunting areas for wetland-oriented mammals like the mink (Mustela vison) and raccoon (Procyon lotor). These areas are also used by upland-oriented mammals like the red fox (Vulpes vulpes) and striped skunk (Mephitis mephitis) as they travel, as well as hunt, around wetlands.

Another factor that could be responsible for increased nest success on CRP cover is field size. In the Great Plains, annual nest success rates (Mayfield method) for total ducks on CRP plots (30-56 percent) were always higher than on WPAs (4-13 percent). Field sizes in the Great Plains averaged about 64 ha (158 ac) on plots compared with 32 ha (79 ac) on WPAs. Differences in annual success rates between the two covers were much less in the Central Lowlands, where differences in field size were smaller. Although none of the above differences is probably significant, they support Clark and Nudds' (2) call for experiments to provide information on possible relations between patch size and nest success by ducks. This research is urgently needed by managers of upland habitat for breeding waterfowl given today's high costs of habitat acquisition and management.

Stands of seeded nesting cover are believed to be more productive of ducks and other game birds two to eight years after stands are established or rejuvenated (5). No lag effects, where prey establishes before predators, has been documented for prairie nesting waterfowl—newly established habitat is included in existing predator territories. Waterfowl production on older stands has not been evaluated, and vegetation management guidelines for these stands are lacking (9, 11). Of the fields we studied, CRP covers were much younger than WPA covers. Managers of waterfowl habitat need to know whether duck nest success is related to cover age or condition. However, no information is available on the effects of stand age or vigor on predator movements, prey availability, or the establishment of predator populations in fields of seeded nesting cover.

Previous Section -- Methods
Return to Contents
Next Section -- Conclusions

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 19:07:28 EST
Sioux Falls, SD [sdww55]