Factors Associated with Duck Nest Success
in the Prairie Pothole Region of Canada
Conclusions
On most of our study areas in the PPR of Canada during 1982-85, native habitats
used by ducks for nesting were extensively reduced in amount and fragmented
because of prior clearing and cultivation.
Annual estimates of nest success were at or above suggested threshold levels
for maintaining stable populations (15%) for mallards and northern pintails
in 8 of 31 area-years and (20%) for gadwalls, blue-winged teals, and northern
shovelers in 4 of 31 area-years.
Approximately 77% of all nests initiated failed directly because of predation.
Although predation was the overwhelming cause of nest failure, predator numbers
did not appear to be the sole determinants of predation. Predation rates were
related to weather variables and other factors that may have strongly affected
nest success (possibly abundance of buffer prey and extent of fragmentation
of nesting habitat).
Large pastures in native grassland were probably the most productive areas
for upland-nesting ducks. Productivity was related to size of pasture, amount
of brush, and mammalian predator community. Large pastures were remote and
had few roads and infrequent visitors—all factors that probably contributed
to habitation by coyotes.
Nest success was negatively correlated with amount of cropland present and
decreased about 4 percentage points for every 10 percentage points increase
in amount of cropland. Under conditions of our study in the prairie physiographic
zone, we predict that the 5 common species cannot maintain local breeding
populations where cropland exceeds about 56% of the habitat.
In areas where cropland is abundant, nesting habitat usually is in fragmented
tracts that tend to be occupied by a mammalian predator community dominated
by red foxes.
Warm weather, precipitation, and abundant wet wetlands in April and May
promoted early nest initiation. Length of nest-initiation period was extended
by precipitation during the nesting season, but high rates of nest success
and above average temperatures tended to shorten it.
Ducks initiating nests toward the end of the breeding season tended to suffer
lower rates of nest predation than earlier nesters, which imparted an advantage
to renesting ducks and late-nesting species.
Relatively high mortality of ducks, especially females, occurred during
the breeding season due to predation, most likely by red foxes, coyotes, northern
harriers, Swainson's hawks, red-tailed hawks, and great horned owls. Mallards
and northern pintails were especially abundant among dead ducks because they
nested early when other prey likely were scant. Females were more vulnerable
to predators during the breeding season than were males because females tended
the nest. Many abandoned nests, at which we found no evidence of egg destruction,
were probably due to death of the female.
Nest success rate, as indexed by the number of successful nests, appeared
to be the most influential factor in determining mallard production. Size
of the breeding population also was strongly influential.
Nest success varied geographically, annually, and in response to other unidentified
factors. Components of variance associated with unexplained variation exceeded
those associated with either geographic or temporal variation for all variables,
except for number and extent of temporary wetlands.
Large areas of the PPR in Canada presently are not producing the sustainable
populations of dabbling ducks for which this Region was known in past decades.
Although we believe our study areas represent habitat conditions across the
PPR of Canada, it is possible conditions on our study areas were better for
duck production than other parts of the Region. If our study areas represent
the best areas for duck production, then much of this important breeding ground
probably is less suitable for nesting ducks than we observed.