Birds introduced in new areas show rest disorders

Pierre-Yves Henry1,2,*, Catalina Landeta Salgado3, Fanny Paucar Muñoz1 and Martin C. Wikelski1,4

1Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA.

2UMR 7179, CNRS - Muséum National d’Histoire Naturelle, Brunoy, France.

3Laboratorio de Biología, Pontificia Universidad Católica del Ecuador, Quito, Ecuador.

4Migration and Immuno-ecology, Max Planck Institute for Ornithology, Radolfzell, GermanyDepartment of Biology, University of Konstanz, Konstanz, Germany.

*Author for correspondence ().

Electronic Supplementary Material

2. Material and methods

Procedures for bird geolocation and behavioural observation.

After release, radio-tagged individuals were followed for the entire period of diurnal activity.There were two persons constantly monitoring the focal bird. The first person was in charge of permanently locating it and describing its behaviour every minute. He/she radio-located the tracked sparrow using an AOR AR8200 handheld receiver and two antennas with different precisions for angular determination (respectively, for bands 302 MHz and 906 MHz). The radio-transmitter emitted a 13-ms pulse every 900 ms. For behavioural observations at distance from the bird, (s)he used binoculars (x10) and a scope (x20-45). The second person was in charge of keeping up with the 1-min frequency of data recording, recordings of GPS locations (with a GARMIN GPSMAP76S), counting sparrow calls, and writing down data.

When the bird was not in flight,we took its exact GPS location ( 20 m). When it was flying or not yet precisely located, we recorded the GPS location of our current position, plus an estimate of the distance and the direction to the bird with a compass. Since the precision of GPS location was  20 m, and since we remained most of the time within 5-20 meters from the focal bird, we considered that the position of the bird could be approximated by our position. When we had to remain at more than 20 m from the bird, we waited for the bird to move away, and then one of us went to record the exact GPS position.

House sparrows leave in the close vicinity of humans. We observed several times the birds coming close to us, or voluntarily foraging just a few meters from humans. Hence we have no indication that the behaviour of radio-tracked birds could have beenconfounded by a stress induced by our presence (observer disturbance effect).

Control individuals.

All the 17 birds included in this study were translocated, and dispersed from the release site. Five out of them found established populations of conspecifics and settled with there. Three did so on the day of release (day 1), one the following day (day 2), and one between days 3 and 6 after release. Since these five birds remained within these established populations, and foraged and roosted with the resident sparrows, we assume that they document the normal resting habits of naturally established house sparrows in the study area.

Computationof the relative calling rate.

In the field, we could not distinguish the calls emitted by the radio-tagged individuals from the calls emitted by the other house sparrows. Since the group size was much larger for individuals in naturally established populations (several tens) than for introduced individuals (1-10 individuals), it is meaningless to compare the total number of calls between the two statuses. To obtain an index of vocal activity that could be compared between statuses, we defined a relative calling rate as follows.

First we computed, for each tracked individual,an individual calling rate, defined as the mean number of calls counted per minute per hour of the day (6 a.m. – 7 p.m.) per day (expressed in number of calls.min-1.h-1). With these individual average calling rates, we computed the average hourly calling rate per day per status, and then the average calling rate per day per status (8.5 calls.min-1 for control individuals, 1.4 calls.min-1 for introduced individuals). We divided each individual calling rate by the average calling rate per day per status, obtaining a standardized individual calling rate. Eventually, the mean relative calling rate (± s.d.) per status was obtained by averaging these standardized individual calling rates per hour per status (figure 2).

Definition of the number of days after arrival at a site.

As four introduced birds changed of settling area once during the study period, they were exposed twice to the challenge of finding aroost site in a new,conspecific-free area. Hence, the explanatory variable used to characterize the number of days since arrival at a new site was the number of days that a bird spent per settling area (and not the number of days after release). A settling area was defined as an area equivalent to the daily home range of naturally established house sparrows in the study area (maximum of 1 km of diameter; P.-Y. Henry, unpublished data). If a sparrow moved more than 1 km between two consecutive nights, it was considered to be re-exposed to the challenge of finding a sleeping place in a new, conspecific-free area. The details of changes of settling areas are: bird BMYY moved on day 3 after release, and remained at the new settling area until day 7;bird WMYB moved between days 3 and 6; bird WMYW moved on day 7; and bird WPNM moved on day 2 when it joined a naturally established population and remained there until day 7.

Table S1. Numbers of individuals, of observations, of individuals per total number of observations,and of individuals per day after arrival at a site for each variable characterizing the roosting behaviour of control and introduced house sparrows. (n.a., events that could not be documented given the experimental design).

number of individuals / number of observations / number of individuals with x observations / day after arrival at a site
5 / 4 / 3 / 2 / 1 / 1 / 2 / 3 / 7
(a)hour of arrival at the roost
control / 5 / 12 / n.a. / 1 / 2 / 0 / 2 / 3 / 4 / 2 / 3
introduced / 8 / 20 / n.a. / 2 / 2 / 2 / 2 / 7a / 6 / 5 / 2
(b)hour of departure from the roost
control / 4 / 8 / n.a. / 0 / 1 / 2 / 1 / n.a. / 1 / 3 / 4
introduced / 7 / 11 / n.a. / 0 / 1 / 2 / 4 / n.a. / 3 / 7b / 1
(c)roosting habitat
control / 5 / 17 / 1c / 2 / 0 / 2 / 0 / 2 / 4 / 4 / 7c
introduced / 10 / 25 / 2c / 0 / 3 / 1 / 4 / 8 / 6 / 4 / 7c
(d)relative calling rate between 5 and 6 p.m.
control / 5 / 12 / n.a. / 1 / 2 / 0 / 2 / 2 / 4 / 3 / 3
introduced / 9 / 25 / n.a. / 4 / 1 / 2 / 2 / 8 / 6 / 6 / 5
(e)relative calling rate between 6 and 7 p.m.
control / 4 / 5 / n.a. / 1 / 2 / 0 / 2 / 1 / 2 / 1 / 1
introduced / 9 / 21 / n.a. / 2 / 2 / 2 / 3 / 6 / 6 / 4 / 5

(notes: a two individuals that had changed of settling site contributed with two observations each of roosting hour for day 1; b one of these individuals also contributed with two observations of hour of departure from the roost; cfor three control individuals and two introduced individuals wecould determine the roosting habitat in the morning of day 7, i.e. for the night from day 6 to day 7; these five supplementary data points are here summed with numbers of observations for day 7).

1