Wigeon vigilance on golf courses 107
©Wildfowl & Wetlands Trust
Wildfowl
(2013) 63: 105–114
human activity from golfers, golf carts,
bicyclists, joggers, and pedestrians (see
results). American Wigeon at this site typically
grazed on land, similar to those foraging on
natural coastal pastures or saltmarsh (see Berl
& Black 2011 for description).
Behavioural data were collected from 28
January 2012 to 22 March 2012, and resulted
in 16.5 h of observation. The study
period was kept short to control for
seasonal influences on vigilance behaviour
(Fernández-Juricic & Schroeder 2003).
Vigilance behaviour was recorded using a
digital video camera (Samsung HMX-F80;
65× optical zoom) centred on a focal bird.
Video recordings were transferred to a
computer and later analysed by a single
observer (JLB). Focal birds were recorded for
between 52–97 s, averaging 66 s ± 0.52 s.e
.
(Mayhew 1987; Berl & Black 2011). Vigilance
was expressed as the percentage of time that
the focal bird was vigilant (time vigilant
*100/length of observation period); it also
allowed for direct comparisons with previous
studies (
e.g.
Mayhew 1987; Portugal &
Guillemain 2011). To reduce the likelihood of
pseudo-replication, focal bird recordings
were restricted to 10 per observational
session (Berl & Black 2011), alternating
between sexes when possible. High turnover
rate of wintering waterfowl populations
further diminishes the probability of pseudo-
replication on consecutive observational
events (Portugal & Guillemain 2011). All
observations and video recordings were
conducted from concealed hides at distances
that did not noticeably alter bird behaviour
(75–200 m; Severcan & Yamac 2011).
Several variables were recorded in the
field during each data collection session
using 8× binoculars and 15–40× spotting
scope, including flock size. American
Wigeon typically graze in compact flocks,
and thus only individuals within 30 m of the
focal bird were considered in flock counts
(Portugal & Guillemain 2011). To reduce the
effect of flock size on vigilance, recordings
were limited to flocks of
≥
20 individuals, as
previous studies (Mayhew 1987; Mayhew &
Houston 1989) have shown that flock size
does not tend to have a significant effect on
vigilance rates in wigeon when flocks are
greater than 20 individuals. In all cases, the
bird nearest the centre of the flock was
selected for recording. In instances where it
was impossible to view the centremost bird
(because of flock densities) the position of
the focal bird in relation to the centre of the
flock (in duck lengths) was recorded to
check that an individual’s position within the
flock did not affect vigilance levels (Kruskal-
Wallis analysis of variance:
H
3
= 1.86,
P
= 0.60, n.s.). Furthermore, each focal bird
was grouped into one of four distances
from water categories: 0–20 m, 21–40 m,
41–60 m, and > 61 m.
Focal bird recordings were conducted
during periods of human presence and
absence to determine if American Wigeon
adjust vigilance investment according to
human activity. Each focal recording was
grouped into one of four categories based
on the number of humans present: absent,
low activity (1–2 people), medium activity
(3–4), and high activity (
≥
5). Human activity
was considered irrespective of whether this
caused disturbance to the birds, and
consisted primarily of golfers, golf carts,
bicyclists, joggers and pedestrians which
usually occurred along the golf course paths
