Northern Prairie Wildlife Research Center
Effects of Weather on Breeding Ducks in North Dakota
Duck Arrival Dates
The arrival date—the day on which an individual of a species was first observed
on the study area—was subject to some variability. An aberrant value could be
recorded if an unusually early migrant was seen a week or more before others of
that species. Such observations were uncommon among our data; in most years the
arrival of the first single bird was followed by a wave of birds within a few
days.
Mallard
Mallards arrived on the median date of 30 March at Salyer and 20 March at Woodworth
(Table 2). Annual variation in arrival dates was considerable, however, ranging
from 7 March to 10 April at Salyer and from 15 March to 6 April at Woodworth.
The arrival of mallards in spring was strongly influenced by temperatures
during the usual arrival period. Arrival tended to be later when temperatures
were low during 12 March-8 April (Fig. la). Arrival dates were delayed an average
of 1.8 days (P<0.0001) for each Celsius degree difference in the mean temperature.
A more detailed examination showed that mean temperatures during 12-25 March
accounted for most of the variation in arrival dates.
Arrival dates at Salyer and at Woodworth tended to be similar when mean temperatures
were similar (Fig. la). The difference adjusted for mean temperatures was 1.8
days earlier at Woodworth, which was not significant (P=0.30). Thus the 10-day
difference in median arrival dates between the two areas was largely due to
the earlier warming at Woodworth than at Salyer.
Gadwall
The median dates of gadwall arrival were 12-13 April (range, 30 March-2 May) at
Salyer and 7 April (range, 20 March-28 April) at Woodworth (Table 2). The arrival
of gadwalls was also influenced by low temperatures during the usual arrival period
(Fig. lb). Gadwalls arrived an average of 1.46 days later for each Celsius degree
change in mean temperature during 12 March-8 April (P<0.0001). Temperatures during
the latter part of that period were particularly influential.
Unlike the mallard, the gadwall seemed to arrive significantly (P<0.01) earlier
at Woodworth than at Salyer under similar temperature regimes (Fig. lb). The
estimated difference in arrival dates between the two areas, adjusted for mean
temperature, was 6.0 days.
Blue-winged Teal
The median arrival date of blue-winged teal (Table 2) was 15-16 April at Salyer
(range, 1-25 April) and 11-12 April at Woodworth (range, 5-22 April). Blue-winged
teal were influenced in their arrival by low temperatures (Fig. lc), particularly
during 26 March-15 April. Arrival dates were delayed 1.07 days for each Celsius
degree difference in mean temperature during the period (P<0.0001). Under similar
temperatures, blue-winged teal seemed to arrive somewhat earlier at Woodworth
than at Salyer; the adjusted difference was 2.5 days earlier at Woodworth (P =
0.12).
Redhead
Median arrival dates of redheads (Table 2) were 8 April at Salyer (range, 26 March-21
April), and 4-5 April at Woodworth (range, 25 March-21 April). Redheads tended
to be delayed by low temperatures (Fig. 1d), particularly during 12 March-8 April.
They were delayed 2.0 days for each Celsius degree difference in mean temperature
(P<0.0001). Under similar temperature conditions, redheads arrived at Woodworth
somewhat earlier (2.4 days; P=0.14) than at Salyer.
Table 2. Summary statistics of breeding activities of four
duck species at the Salyer (S) and Woodworth (W) study areas. Dates are Julian.
| Activity |
Species and study area |
| Mallard |
Gadwall |
Blue-winged
teal |
Redhead |
| S |
W |
S |
W |
S |
W |
S |
W |
| Arrival (na) |
28 |
13 |
29 |
13 |
28 |
12 |
29 |
10 |
| Median |
89 |
79 |
102.5 |
97 |
105.5 |
101.5 |
98 |
94.5 |
| Mean |
86.7 |
83.6 |
102.2 |
94.8 |
105.4 |
102.5 |
98.3 |
94.3 |
| SDb |
7.7 |
8.8 |
7.9 |
9.9 |
5.8 |
4.9 |
7.6 |
8.5 |
| Initiation (n) |
25 |
- |
25 |
- |
26 |
- |
11 |
- |
| Median |
109 |
- |
137 |
- |
130 |
- |
123 |
- |
| Mean |
108.4 |
- |
132.4 |
- |
128.9 |
- |
126.8 |
- |
| SD |
7.8 |
- |
7.9 |
- |
7.1 |
- |
13.8 |
- |
| Peak hatching (n) |
18 |
- |
19 |
10 |
19 |
12 |
18 |
- |
| Median |
176 |
- |
190 |
180.5 |
185.5 |
180 |
193 |
- |
| Mean |
176.5 |
- |
189.8 |
183.6 |
186.3 |
178.3 |
191.1 |
- |
| SD |
10.6 |
- |
6.3 |
9.6 |
6.9 |
5.6 |
10.1 |
- |
| Span (n) |
18 |
- |
19 |
10 |
19 |
12 |
10 |
- |
| Mean |
46.6 |
- |
32.9 |
30.6 |
34.7 |
38.1 |
37.7 |
- |
| SD |
5.9 |
- |
4.8 |
6.2 |
4.2 |
7.3 |
7.9 |
- |
an=number
of years. bSD=standard deviation. |
General
Below-normal temperatures during or just before the typical arrival period tended
to delay arrivals for each of the four species. The earliest arriving species,
the mallard, was particularly influenced by temperatures during 12-25 March. The
next earliest, the redhead, was apparently affected by temperatures during 12
March-8 April. The third species to arrive, the gadwall, had arrival dates most
strongly related to mean temperatures during 26 March-8 April. The latest of the
four species, the blue-winged teal, had arrival dates correlated with temperatures
during 26 March-15 April.
In addition to effects of mean temperatures, differences in arrival date were
observed between Salyer and Woodworth. Under similar weather regimes, birds
of each species could be expected earlier at Woodworth. The difference was 1.8
days for mallards, 2.4 days for redheads, 2.5 days for blue-winged teal, and
6.0 days for gadwalls, only the last being clearly different from zero.
These findings corroborate the common notion that late, cold springs result
in delayed arrivals of migrating ducks. Sowls (1955) suggested that later-arriving
species were less affected by temperatures than the early-arriving mallard and
pintail, but this contention was not supported by our data. Arrival date is
determined by weather patterns throughout the migrational corridor during the
entire period of migration; temperature at the terminus can only account for
a portion of the variation.
Previous Section -- Methods
Return to Contents
Next Section -- Breeding Pair Populations