Electronic Appendix

To

Microclimate and microhabitat selection of the Alpine Rock Ptarmigan (Lagopus muta helvetica) during summer

Linda Visinoni a, Claire Agnès Pernollet a,c, Jean-François Desmet b, Fränzi Korner-Nievergelt a, Lukas Jenni a

a Swiss Ornithological Institute, Seerose 1, 6204 Sempach, Switzerland

b Groupe de Recherches et d'Information sur la Faune dans les Ecosystème de Montagne, 74340 Samoëns, France

c present address: Office National de la Chasse et de la Faune Sauvage, CNERA Avifaune Migratrice, La Tour du Valat, Le Sambuc, 13200 Arles, France

Appendix S1: Daily course of ground temperature at four different microtopographic sites

Below a rock, mean ground temperature did not vary much between day and night; it was slightly higher at night and much lower during the day than at the other three sites (Fig. S1a). Also the variance between days was smaller than at the other three sites (Fig. S1b). The highest mean ground temperatures during the day and the lowest during night were reached on rocks with vegetation, probably because of the combined effects of heat absorbed by the rock and protection from cooling winds by the vegetation during the day and efficient heat dissipation during the night. Somewhat lower temperatures during the day were reached in the vegetation and on bare rocks. On bare rocks temperatures were higher in the evening than at the other sites, probably because of the heat stored by the rocks. The variance between days was generally high (SD about 7 °C during the day) and highest in the vegetation (SD about 10 °C in the afternoon). Ground temperature on rocks with vegetation reached up to 45 °C on hot summer days. During five bad-weather days, the diurnal raise in ground temperature was much less pronounced (not shown).

Fig. S1 Daily course of ground temperature at four different microtopographic sites (below rock, in vegetation, on bare rock, on rock with vegetation). (a) Lines are hourly means of 5 different locations over 33 days. (b) Lines are standard deviations and represent the variance between days independent of location (see Methods). The standard deviations are of residuals from a linear model with ground temperature as the dependent variable and location, site, hour and site × hour as independent variables, thus represent the variability between days independent of location. Dotted line = below rock, dashed line= on bare rock, solid line = in the vegetation, dot-dashed line = on rock with vegetation.

Fig. S2 Contour plots of the microclimate variables (a) air temperature, (b) ground temperature, (c) wind speed, (d) solar radiation, (e) relief (depth), (f) vegetation height in one of the two grid-squares. The plane, given by the X- and Y-axes, represents the grid plane of 2 x 2 m with measurement points every 50 cm. The means of seven measurements were used to construct the plot. Light areas represent high values, dark areas low values.

Table S1 Microclimate characterisation of the four microtopographic sites. n = 114 per microtopographic site and microclimate variable. The measurements were taken between 1130 and 1630 hours on days without rain.

Microtopographic site
Below rock / On bare rock / On rock with vegetation / In vegetation
Ground temperature (°C) / Mean ± SD / 16.8 ± 4.2 / 23.8 ± 5.5 / 25.4 ±7.586 / 22.6 ± 5.5
Min − Max / 9.3 − 32.9 / 9.1 − 38.1 / 11.0 − 45.2 / 10.6 − 44.4
Air temperature (°C) / Mean ± SD / 20.1 ± 3.0 / 20.2 ± 2.8 / 20.3 ± 7.6 / 22.3 ± 4.0
Min − Max / 14.6 − 29.1 / 14.7 − 31.7 / 14.6 − 30.7 / 14.0 − 35.2
Wind speed (m/s) / Mean ± SD / 0.19 ± 0.29 / 0.60 ± 0.46 / 0.52 ± 0.49 / 0.14 ± 0.22
Min − Max / 0 − 1.30 / 0 − 2.60 / 0 − 3.20 / 0 − 0.90
Solar radition (W/m2) / Mean ± SD / 48 ± 128 / 769 ± 338 / 667 ± 409 / 449 ± 325
Min − Max / 1 − 772 / 54 − 1413 / 24 − 1410 / 19 − 1250

Table S2 Linear mixed models testing for the effects of vegetation height and relief on wind speed, solar radiation, air and ground temperature in the two grid-squares (n = 350). Depth is the vertical distance from an imaginary horizontal plane at the highest point of the square to the measurement point. In each model, grid identity and date were included as random factors. * = p < 0.05, ** = p < 0.01, *** = p < 0.001

Estimate ± SE / t-value / CI 95 % / Sign.
Air temperature
Depth / 0.05 ± 0.03 / 2.01 / 0.003 − 0.10 / *
Vegetation height / 0.10 ± 0.03 / 4.14 / 0.06 − 0.16 / **
Depth×Vegetation height / -0.002 ± 0.002 / -1.144 / -0.005 − 0.001
Ground temperature
Depth / -0.10 ± 0.05 / -2.12 / -0.20 − -0.009 / *
Vegetation height / -0.46 ± 0.05 / -9.62 / -0.55 − -0.36 / ***
Depth×Vegetation height / 0.006 ± 0.003 / 1.74 / -0.001 − 0.012
Solar radiation
Depth / -0.78 ± 2.38 / -0.33 / -5.14 − 3.79
Vegetation height / -4.86 ± 2.39 / -2.03 / -9.24 − -0.32 / *
Depth×Vegetation height / -0.21 ± 0.16 / -1.31 / -0.53 − 0.07
Wind
Depth / -0.004 ± 0.004 / -1.05 / -0.01 − 0.004
Vegetation height / -0.03 ± 0.004 / -7.08 / -0.04 − -0.02 / ***
Depth×Vegetation height / 0.0001 ± 0.0003 / 0.37 / -0.0005 − 0.0006

Table S3 Results of generalized linear mixed models comparing the topography of points of ptarmigan with their corresponding control point at 5 m or at 30 m (likelihood-ratio tests, 6 separate models). Observation event, triplet and individual were included as random factors. Sample sizes were n = 42 triplets for exposition and n = 59 triplets for microtopography and slope.

c2 / df / p-value
Exposition
Bird versus control 5 m / 20.4 / 4 / < 0.001
Bird versus control 30 m / 20.4 / 4 / < 0.001
Microtopography
Bird versus control 5 m / 32.2 / 2 / < 0.001
Bird versus control 30 m / 17.6 / 2 / < 0.001
Slope
Bird versus control 5 m / 4.9 / 3 / 0.17
Bird versus control 30 m / 4.1 / 3 / 0.25

1