|
nif:isString
|
-
a) Study site, birds, and experimental setting: The study site (Rocher des Aigles, Rocamadour, France), overhangs a 120 m-deep canyon, providing natural soaring conditions for raptors. Vultures were trained with falconry techniques to execute free flight shows for the public. During recording sessions, birds were released together from their perch. Just before the flight, the trainer approached the birds and remained close to them for a few minutes before they were released. Once free, birds generally took-off immediately. They could freely choose their flight path, presumably to find an ascending current to climb. They were recalled by the trainer after 10–20 min. However, since the timing of the signal of the trainer was not strictly fixed from one show to another and since birds sometimes responded several minutes after this signal, we do not think that the timing modified the flight behaviour of the birds. Birds landed in an open arena and then had to walk 200–300 m back to their perch. Preliminary experiments indicated that the birds flew similar to wild vultures (C. Tromp, unpublished results), reaching heights >1500 m above ground within 15 minutes. We collected data for one Himalayan griffon (age 15) and one Eurasian griffon (age 30) that performed 11 and 12 flights respectively over four days (mean flight duration 11.0±5.2 min, range 3–19 min). These 2 individuals were chosen for their tame behaviour, tolerance of handling and tolerance for a harness and HR data logger, a prerequisite to obtain normal soaring flight behaviour to reach high altitudes. In contrast birds that were stressed with the harness were more prone to make short flights in the vicinity of the platform where the trainer waited. Although belonging to two closely-related species [31], the two individuals were similar in morphometry (Table 1) and are thus expected to face similar constraints for soaring-gliding flight. Weather conditions were recorded at the site on the day of each experiment and for each flight. Two conditions prevailed during experiments: sunny days with no cloud cover and days with 100% cloud cover.
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0084887.t001 Morphometry and flight characteristics of the two vultures in different weather conditions. Values are mean ± SD with sample size in brackets.
Vultures were equipped with bio-loggers including a GPS, a 3-axial accelerometer, and an electrocardiogram recorder (ECG). The 3D positions of birds were recorded 4 times per second using the high-precision GPS module (Gipsy 1, Technosmart, Italy). We performed immobility tests under open sky to assess the precision of the GPS loggers by calculating standard deviation around mean values of longitude, latitude and altitude (respectively 0.8, and 1.7 and 1.2 m). Three-axial accelerations between −2.3 and +2.3 G or between −9.2 and +9.2 G were acquired at 100 Hz with a LIS302DL accelerometer (STMicroelectronics) and stored in Neurologger 2 [32]. The 52 g ECG logger (UWE-ECG, Little Leonardo, Japan) recorded the ECG at 1000 Hz through 3 cables that were clipped in the skin of the vultures' back with golden-plated safety pins [17]. Two electrodes were clipped on each side of the thorax and one on the middle of the back (ground). The GPS, accelerometer and ECG were taped together and attached to a harness, which consisted of an aluminium plate (100×60 mm) and two Teflon ribbon strips (with silicon tubing for elasticity). The plate was positioned approximately at the centre of gravity on the back of the bird and held by leg-loops [33], [34]. The total mass of the harness and loggers was 215 g which represented <1.5% of total body mass.
ECG and acceleration data were processed using a program developed on Igor Pro for this purpose (Wavemetrics, USA). The vectorial Overall Dynamic Body Acceleration (generally called ODBA or VeDBA) was calculated from the acceleration values, allowing identification of periods of inactivity (resting or soaring-gliding) and activity (walking or flapping) [35]–[37]. At first the dynamic acceleration was calculated for each axis by subtracting the smoothed acceleration from total acceleration. Then the ODBA was calculated as a vectorial sum i.e. the square root of the sum of the squares of the dynamic accelerations. To calibrate ODBA with flight behaviour, vultures were observed and filmed after take-off for as long as possible (birds could not be observed 100% of time during flight since they often went out of sight in the canyon), and again while landing. Preliminary analyses on a larger sample of vultures tracked without an ECG logger (C. Tromp, unpublished results) found a strong correlation between 3D acceleration, ODBA and flight behaviour, as has been observed for other flying birds [38]–[41]. The distinction between flapping flight vs. soaring/gliding was evident from vertical and forward acceleration component because the acceleration fluctuated with each wing beat [38] (see Figure S2). We are confident that 3D accelerometry and the derived value of ODBA detected all flapping bouts. The distinction between soaring and gliding flight was easily derived from the flight paths of the birds: as soon as birds started to fly in circles and have a positive vertical speed (i.e. starting climbing), they were considered as “soaring”, otherwise “gliding” when the flight path was rectilinear and altitude was decreasing [42]. In the raw ECG signal, we identified the PQRS complex (i.e. the complex of electric signals defining a heartbeat, see [17]) by automatically extracting the R peak of each complex. Only peaks positively identified, by visual inspection, as being part of the PQRS complex were kept for analysis, while noise signal from muscle activity was discarded and HR was calculated as inversed R–R interval [17]. HR and ODBA were averaged every minute from 10 min before take-off until 10 min after landing (the notation adopted was related to take off (TO) or landing (LD) and minutes before or after: before flight (TO-10 to TO-1), at take off (TO), during flight (TO+1 to TO+17), upon landing (LD-2 to LD+1) and after landing (LD+2 to LD+10)). On land, we distinguished “perching” and “resting” behaviours based on the atmosphere around the birds. Both perched and resting birds were on their “perching roost”, but tourists made for a noisy atmosphere as the show was going on. In contrast, a resting bird was in a sleeping state during the morning hours when tourists were absent and the atmosphere silent. We defined the “baseline” HR as the value when birds were “resting” and considered it to be a state near Field Basal Metabolic Rate. We considered HR as a proxy for metabolism since HR measured in a resting Eurasian griffon vulture as well as other birds is correlated with oxygen consumption [29] [16]. We collected data for a total of 255 min of flight but variability in HR patterns, inter- and intra-individual variations could not be statistically assessed with only 2 individuals. We thus presented only mean ± SD values.
The study was not specifically approved by an ethical committee as a permit for equipping vultures with loggers was provided as part of the licence of O. Duriez from the Research Centre for Bird Population Studies (CRBPO) of the Natural History Museum (MNHN, Paris). According to the French law of 22 September 2008, the CRBPO has the delegation by the Ministry of Ecology, Energy, Sustainable Development and Land Settlement for allowing the owners of a general bird ringing licence to capture and handle birds from protected species and mark them (with rings or any devices like loggers). The study was conducted under a formal agreement between the animal rearing facility (Rocher des Aigles) and CNRS. Birds were handled by their usual trainer, under the permit of the Rocher des Aigles (national certificate to maintain birds “Certificat de capacité” delivered to the director, Raphaël Arnaud on 4 November 1982). Care was taken to minimize discomfort to the birds and loggers were removed promptly after flights.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}