Page 129 - Panuccio2012

90 SHORT COMMUNICATION VOL. 45, NO. 1

Table 2. Comparison of counts of migrating raptors through the Strait of Messina with headwinds (N-NNE) and
tailwinds (WSW-SW-SSW), reported as the hourly mean number of individuals (6SE) and the P value of a test for
significant differences between the two categories. Dates of observation were 27 March–31 May for all Accipitriformes;
25 April–31 May for European Honey-buzzards; and 27 March–22 May for Circus spp.

ACCIPITRIFORMES EUROPEAN HONEY-BUZZARDS CIRCUS SPP.

WIND DIRECTION HOURLY MEAN P HOURLY MEAN P HOURLY MEAN P

Headwind (N-NNE) 24.7 6 3.4 ,0.001 31.4 6 4.9 0.001 4.2 6 0.6 ,0.001
Tailwind (WSW-SW-SSW) 6.2 6 1.5 12.3 6 3.2 0.6 6 0.1

hr, and for 8 hr (1%) wind speed was negligible. Winds et al. (2000) hypothesized that this behavior is adaptive
were considered variable for 0.4% of hr, and data were not because the risk of exhaustion increases with energy con-
available for 0.6% of hr. sumption, and therefore risk is higher when flying against
headwinds. At the island of Ustica, marsh harriers were
The comparison of the mean hourly passage rates of observed to begin crossing the Mediterranean Sea mostly
raptors during the two main wind conditions (headwinds when wind speeds were less than 15 km/h, regardless of
and tailwinds) indicated that wind direction influenced direction (Panuccio et al. 2002). Considering the in-
the number of raptors seen migrating at the site (Table 2). creased length of the crossing at that site (250 km), and
Considering all individuals of all species (n 5 9439), a the associated increase in the risk of changeable weather, it
significant higher number passed with headwinds (N- was not surprising that raptors began the long, powered
NNE) rather than with tailwinds (WSW-SW-SSW; t 5 flight from the island only when weather conditions were
5.023, df 5 430.771, P , 0.001). safer. When tailwinds were blowing, European Honey-buz-
zards crossing the Mediterranean Sea between Cap Bon
This pattern was true for Circus spp. (t 5 6.078, df 5 213.16, and Sicily used a more direct route, when they were ob-
P , 0.001) and for European Honey-buzzards (t 5 3.293, df served at the northern side of the Channel of Sicily passing
5 277.27, P 5 0.001). However, on certain days, large num- over the island of Marettimo. Conversely, in headwinds or
bers of European Honey-buzzards were observed migrating lateral winds, European Honey-buzzards drifted ESE to-
with tailwinds (390, 109, and 126 individuals on 25, 26 and ward the island of Pantelleria where they could use ther-
27 May, respectively). On these days, raptors were observed mal currents to rest from powered flight (Agostini et al.
flying far out over the sea, north of the hawkwatch site. 2007). With tailwinds, a greater number of European Hon-
ey-buzzards were observed at the island of Panarea than at
By contrast, numbers of European Honey-buzzards were the Straits of Messina, although at both sites raptors mi-
not significantly influenced by wind speed (Table 3), ei- grated mostly with headwinds (Agostini et al. 2007).
ther considering all winds regardless of direction (t 5
20.12, df 5 313.18, P 5 0.904), or only headwinds (t 5 The results of this study suggest that raptors may con-
0.294, df 5 207.63, P 5 0.769). centrate at the Strait of Messina when the weather condi-
tions are not suitable for a longer water crossing, particu-
DISCUSSION larly against headwinds. These wind conditions may lead
individuals to fly over land where they can exploit ther-
Much previous research, based on both visual observa- mals, using soaring-gliding flight to minimize the risk
tions and on radar detections, has indicated that raptors and energetic costs of a longer sea crossing. In fact, the
usually prefer to begin water crossings with tailwinds short distance of the crossing at the site (about 3 km)
(Meyer et al. 2000, Agostini et al. 2005, 2007), because allows raptors to reach the continental coast even with
tailwinds should allow a faster and energetically less expen- strong opposing winds. High numbers of harriers were
sive flight over sea. Raptors may also undertake water cross-
ings with lateral winds (Kerlinger 1989, Agostini et al.
1994b), but rarely against strong opposing winds. Meyer

Table 3. Numbers of migrating European Honey-buzzards (hourly mean 6 SE) relative to wind strength, calculated for
all wind directions and for headwinds only (N-NNE), at the Strait of Messina, southern Italy, spring 2004.

WIND DIRECTION Average wind strength (km/h) WEAK WINDS STRONG WINDS P
All directions Honey-buzzards (hourly mean) ,25 km/h .25 km/h 0.94
Average wind strength (km/h) 0.769
Headwinds (N-NNE) Honey-buzzards (hourly mean) 16.2 6 0.5 31.3 6 0.4
23.3 6 4.1 24.2 6 5.1
20.7 6 0.5
32.6 6 6.8 31 6 0.33
29.8 6 6.8

129