Data from: Flight initiation distance is repeatable and geographically flexible in greylag geese (Anser anser)
Data files
Nov 28, 2024 version files 61.40 KB
-
Katsis_et_al._(2024)_Supplementary_data.xlsx
54.52 KB
-
README.md
6.88 KB
Abstract
The decision to flee from an approaching threat represents a trade-off between the costs and benefits of flight. In research to date, particular interest has been given to the effects of urbanisation on predator escape behaviour. However, these differences are most often studied at the population rather than individual level, which makes it difficult to disentangle whether site-specific differences in predator avoidance are due to behavioural plasticity or non-random distribution of behavioural phenotypes. In this study, we measured flight initiation distances—a widely used proxy for fearfulness or boldness—in a free-roaming, human-habituated, and colourbanded population of greylag geese (Anser anser). First, we tested whether individuals flexibly adjusted their flight initiation distance based on environmental context, by approaching the same individuals at two sites with different levels of human presence. Second, we examined whether flock members show consistent differences in their flight initiation distance towards unfamiliar humans. Our results suggest that greylag geese can adjust their behaviour based on environmental context and show individual differences in flight initiation distance that are consistent both within and across years. When individuals regularly move between foraging sites, establishing site-specific responses to potential threats may allow them to optimise their escape response and thereby avoid unnecessary energetic and opportunity costs.
https://doi.org/10.5061/dryad.pvmcvdnw5
Description of the data and file structure
Full details for this dataset are given in Katsis et al., (2024, Journal of Avian Biology).
Study sites and species
This study was conducted in a free-roaming population of greylag geese in Grünau im Almtal, Upper Austria. All members of the flock are individually identifiable; almost all geese are marked with a unique combination of aluminum and colored leg bands and any unbanded individuals can be distinguished from their facial markings. Some flock members (n = 11) were hand-raised by a human foster parent until fledging, as part of an ongoing long-term research program. We collected data over two testing periods outside the greylag goose breeding season and approximately 13 months apart: 8–14 September 2022 and 10–29 October 2023. During the first period, the flock comprised 128 individuals (78 male, 50 female; mean ± SE age of tested geese: 7.6 ± 0.6 years, range 0–23); during the second period, it comprised 89 individuals (53 male, 36 female; mean ± SE age of tested geese: 7.3 ± 0.6 years, range 0–20). Data were collected at two sites separated by a linear distance of ~370 m: the former Konrad Lorenz Research Center (47°48'49.7412" N, 13°56'51.72" E) and the northern pond at the Cumberland Wildpark (47°48'37.6704" N, 13°56'53.9196" E). The former Konrad Lorenz Research Center is located alongside the river Alm and surrounded by grassy meadows and mixed forest. The site receives few human visitors aside from goose researchers, whose activities mainly involve passive observation of the flock. Flock members visit the meadows during mornings and evenings to receive a twice-daily supplemental feeding of pellets and grain (approximately from 7:00–8:00 and 16:00–18:00 CEST). Between these periods, the flock typically resides at the northern end of the Cumberland Wildpark, at an intersection of several visitor trails surrounded by ponds and grassy islands.
Approach trials
Prior to each approach trial, we randomly selected a target goose that was stationary and not engaged in behaviours such as preening or foraging, which might distract from the observer’s approach. The observer began their approach from a starting distance of approximately 10 m, walking a direct path towards the front of the goose. The speed of approach was standardised to ~0.5 m/s, with the observer’s pacing assisted by a digital metronome played through earphones. We recorded the goose’s posture during the approach (sitting or standing). Observers maintained continuous eye contact with the target goose during their approach, since this has been shown to influence escape behaviour. We defined FID as the horizontal distance (in m) at which the subject started to move away from the approaching observer. Once the goose started moving away, FID was measured by counting the remaining paces to the goose’s original location and converting this distance to metres (in 2022) or by using a tape measure (in 2023). FID was recorded to the nearest 0.1 m. Shorter FIDs indicated that the target goose allowed the observer to approach closer before starting to flee (i.e. a ‘bolder’ response). If the goose did not move away until the observer’s foot was almost touching its body, it received a minimum FID of 0.1 m. In almost all cases, the fleeing goose moved only a short distance (0–2 m) by foot rather than taking to the air. Across both years, we conducted 530 approach trials with 120 individuals. These comprised n = 274 trials with 104 individuals in 2022 and n = 256 trials with 82 individuals in 2023. Individuals were tested 2.6 ± 0.1 times (mean ± SE; range: 1–7) in 2022 and 3.1 ± 0.2 times (range: 1–9) in 2023, allowing us to calculate the short-term repeatability of FID. In addition, n = 66 geese were tested in both years, allowing us to assess the consistency of FID across years.
Files and variables
File: Katsis_et_al._(2024)_Supplementary_data.xlsx
Description: This file contains flight initiation distances from 530 approach trials, conducted on greylag geese (Anser anser) in 2022 and 2023. Approach trials were conducted at two nearby sites (the Cumberland Wildpark and the meadows near the old Konrad Lorenz Research Center). These data were used to asess the geographic flexibility and repeatability of flight initiation distances. Statistical methods are described in the associated paper by Katsis et al. (2024, Journal of Avian Biology).
Variables
- Observer: The experimenter who approached the geese. AB = Avila Bold, JC = Jonathan Cueva, JL = Jonas Lesigang, JMS = Jana-Marie Schmincke, JP = Johannes Ploderer, MF = Marie Fröhlich
- Year: The year in which the approach trial took place (2022 = September 2022, 2023 = October 2023)
- Date: The date on which the approach trial took place (YYYY-MM-DD)
- Site: The location at which the goose was approached (OldKLF = meadows near the former Konrad Lorenz Forschungsstelle building, Wildpark = Cumberland Wildpark)
- Time: Time at which the approach trial took place
- TrialNo: The trial number per goose, pooled across both years (2022 and 2023)
- TrialNoPerYear: The trial number per goose within each year
- BirdID: The name of the goose that was approached
- Handraised: Whether the goose was hand-raised by humans in its first two months before fledging (Y or N)
- BothSites: The subject was tested at both sites (Y or N)
- YearHatched: The year in which the target goose hatched (1999-2023)
- *Age: *The age of the goose (in complete years, 0-23 years).
- Sex: The sex of the goose (M = male, F = female)
- Posture: The position of the goose (standing or sitting) when approached
- FID: The goose's flight initiation distance (to the nearest 0.1 m), i.e. the distance at which the goose started to move away from the approaching observer.
Code/software
All statistical analyses were performed with the software R v. 4.2.3. Univariate linear mixed models used the ‘lmer’ function in lme4 package v. 1.1-32, while multivariate generalised linear mixed models used the package *MCMCglmm *v. 2.34, which fits generalised linear mixed models in a Bayesian framework using Markov chain Monte Carlo (MCMC) techniques.
File: Katsis et al. (2024) Supplementary R code
This file contains the R script that was used to analyse data and create graphs for the study by Katsis et al. (2024, Journal of Avian Biology)
File: Katsis et al. (2024) Supplementary R code output
This file contains the R script output that was used for the study by Katsis et al. (2024, Journal of Avian Biology)