Churchill JL, Hannon SJ

S1. Radio transmitter attachment methodology

The radio attachment procedure of Raim (1978) was adapted for use by C Gillies (University of Alberta), and then further modified for use with redstarts in this study. Prior to bird capture, a lightweight-medium-weave piece of nylon material about 50% larger on all sides than the body of the LB-2N transmitter (Holohil Systems Ltd.) was glued with Instant KrazyGlue® to the back of the transmitter. It was allowed to dry for 1 h, and then was trimmed so that its final size was about 10% larger on all sides than the transmitter body, to increase surface area for attachment to the bird. In the field, Ardell® LashGripTMeyelash adhesive (American International Industries) was generously applied to the back of the nylon and allowed to dry for 2-3 min while the bird was measured. This increased the stickiness of the glue and reduced the time a bird needed to be held. A patch of feathers on the spine between and about 1 cm below the scapulae were lifted and held while 5 or 6 feathers were cut at their base to reveal a bare patch of skin slightly smaller than the nylon attached to the transmitter. The transmitter was carefully positioned above the bare patch, and then moved into place, so as not to spread glue onto the feathers. Moderate pressure was applied to the back of the radio to ensure good contact between the skin, glue and nylon, while ensuring that the bird could adequately breathe. After 10 min, pressure was released and the radio was pulled gently to ensure it had dried in place. The held feathers were released:they fell back into place and completely covered the transmitter body. Of radios attached with this method, only 1 transmitter in 2003 and 1 in 2004 fell off before the battery expired, after 15 and 12 days respectively.

REFERENCES

Raim A (1978) A radio transmitter attachment for small passerines. Bird Banding 49:326–332

S2. Correlation matrix for movement investment dependent variables.

off-territory movements/h / intrusions/h / mean trip distance / maximum distance / time off-territory
off-territory
movements/h / 1
intrusions/h / 0.83 / 1
mean trip distance / 0.72 / 0.57 / 1
maximum distance / 0.84 / 0.64 / 0.89 / 1
time off-territory / 0.80 / 0.50 / 0.76 / 0.80 / 1

S3. Correlation matrix for continuous independent variables used in multiple linear regression models.

condition / song / age / density 200 / density 500 / synchrony 200 / synchrony 500 / forest 200 / forest 500
condition / 1
song / 0.01 / 1
age / -0.14 / 0.81 / 1
density 200 / 0.41 / 0.37 / 0.14 / 1
density 500 / 0.34 / 0.31 / 0.12 / 0.81 / 1
synchrony 200 / 0.53 / 0.63 / 0.42 / 0.57 / 0.41 / 1
synchrony 500 / 0.22 / 0.65 / 0.54 / 0.25 / 0.04 / 0.64 / 1
forest 200 / 0.36 / 0.34 / 0.31 / 0.34 / 0.50 / 0.26 / 0.16 / 1
forest 500 / 0.15 / 0.32 / 0.17 / 0.62 / 0.69 / 0.17 / -0.06 / 0.59 / 1

S4. Off territory movement investment for 19 paired American redstart males. One male (WKKS) was considered an outlier (based on model residuals and Cook’s distance) and was excluded from regression models.

Malea / Year / Intrusions/h / Time off-territory (%) / Maximum trip distance (m)
MYRS / 2003 / 0.00 / 3 / 20
WYKS / 2003 / 0.00 / 5 / 30
RSRG / 2003 / 0.41 / 20 / 270
GYMS / 2003 / 0.47 / 23 / 160
RKDS / 2003 / 1.78 / 22 / 380
WKKS / 2003 / 2.89 / 52 / 1170
YOWS / 2004 / 0.64 / 36 / 450
DSDW / 2004 / 0.13 / 19 / 330
WOOS / 2004 / 0.46 / 23 / 270
PYGS / 2004 / 1.30 / 19 / 420
RWOS / 2004 / 0.76 / 11 / 210
POWS / 2004 / 0.38 / 19 / 310
YODS / 2004 / 0.00 / 0 / 20
DYYS / 2004 / 0.51 / 28 / 160
GMMS / 2004 / 0.41 / 15 / 180
LRWS / 2004 / 1.32 / 37 / 400
DRGS / 2004 / 0.52 / 9 / 160
GROS / 2004 / 0.60 / 14 / 110
KLPS / 2004 / 0.76 / 9 / 80

a Male names are coloured leg-band codes.

1