USGS Home
Contact USGS
Search USGS

Hayland nests were most frequently (82%) in alfalfa. This species is highly attractive to mallards after it has reached about half its mature height and density. Alfalfa, first mowed for hay about 20 June in the study area and again in July, does not produce much residual cover in the following spring. The lack of residual cover followed by rapid growth in late May causes it to become attractive to nesting hens late in the season. The peak period for initiation for all nests was between 21 and 31 May, whereas the peak of hayland nests was between 1 and 10 June. Gates (1965) observed a similar shift of nesting to hayfields as the season progressed. These late nests probably represent second and third attempts for hens that were previously unsuccessful in other habitats. We suspect that in years of good or average rainfall, when alfalfa attains normal height, it attracts a significant proportion of the renesting hens.

Sixteen percent of the nests were in wetlands, and, of these, 34% were in tule bulrush (Scirpus acutus), 17% in cattail (Typha spp.), and 13% in river bulrush (Scirpus fluviatilis). Only 8 (5.6%) of our nests were over water. Although use of wetland was important it did not approach the 66% reported by Krapu et al. (1979b) for the period 1974-77 on a study area adjacent to ours. The high use observed by these authors may be related to an abundance of ponds with dense cover of bulrush and cattail on their study area and a higher rate of nest survival than we observed.

Mallards use a number of sites that cannot be classified as vegetation types. These were lumped into a class called odd areas. Most of the nests in odd areas were in rock piles (36.4%, Fig. 6), followed by shelterbelts (22.7%), and haystacks (13.6%). The remainder were in fencerows, gravel piles, farmsteads, wooded draws, and stumps. Girard (1941) documented a similar use of odd areas. When birds arrived in spring the odd areas frequently represented the only remaining cover on land under intensive agriculture.

Fig. 6. Mallard nest site in intensively farmed habitat of North Dakota.

Only 3 nests (2.1%) were found in cropland. One of these was in standing stubble and 2 were in mulched stubble. No nests were initiated in growing grain although 1 nest, initiated in a small wetland site, survived planting and cultivation and hatched under a canopy of growing sunflowers. Higgins (1977), however, found 20 of 93 cropland nests in growing grain and 8 of these were successful.

Use of habitat for nesting varied by year (x² = 38.9, df = 15, P = 0.001) according to x². Use of grassland in 1977 was much lower than in other years (Table 5) whereas use of odd areas was relatively higher. Lack of nesting in grassland during this drought year probably resulted from the sparse residual cover present at the time of nest initiation. Likewise, dry conditions in 1980 prevented appreciable growth of alfalfa, and alfalfa consequently was not used. The greatest use of wetlands for nesting occurred during the dry years 1977 and 1980 when wetland vegetation provided some of the densest available cover. We were unable to detect any significant difference in the use of the 6 habitat classes between SY and ASY hens (Table 6).

Physiognomy of vegetation within our habitat classes such as grasslands was heterogeneous. We therefore compared habitat use according to life form (Table 7). The high use of woody plants results partly from the use of western snowberry, but nests also occurred in hawthorn (Crataegus spp.), Woods rose (Rosa woodsii), and in shelterbelts. Similar preference for woody vegetation has been observed in Minnesota where many nests were found in clumps of willow (Salix spp.), alder (Alnus spp.), Cassandra leatherleaf (Chamaedaphnae calyculata), and Labradortea ledum (Ledum groenlandicum) growing on sedge mats (Gilmer et al. 1975).