Coping with daily thermal variability: behavioural performance of an ectotherm model in a warming world

Abstract

Global climate change poses one of the greatest threats to species persistence. Most analyses of the potential biological impacts have focused on changes in mean temperature, but changes in thermal variance will also impact organisms and populations. We assessed the effects of acclimation to daily variance of temperature on dispersal and exploratory behavior in the terrestrial isopod Porcellio laevis in an open field. Acclimation treatments were 24 ± 0, 24 ± 4 and 24 ± 8 °C. Because the performance of ectotherms relates nonlinearly to temperature, we predicted that animals acclimated to a higher daily thermal variation should minimize the time exposed in the centre of open field, --i.e. increase the linearity of displacements. Consistent with our prediction, isopods acclimated to a thermally variable environment reduce their exploratory behaviour, hypothetically to minimize their exposure to adverse environmental conditions. This scenario as well as the long latency of animals after releases acclimated to variable environments is consistent with this idea. We suggested that to develop more realistic predictions about the biological impacts of climate change, one must consider the interactions between the mean and variance of environmental temperature on animals' performance.

Figure 1. Dependence of behaviour of space use on the environmental thermal variability.

(A). Dispersion trajectory is made by steps or SL (B).The minimal length is determined by the interval sampling of each behavioural record, i.e. gray line between open circles. Circles represent a sampling event. The absolute angle (α) is defined between the horizontal plane (dx) and the step. The relative angle (ρ) or turning, is the angle between successive steps. Distance (d), is the total distance during each record. The mean distance (R_n) is the net distance between the current location and the first relocation of the trajectory. Modified from .

Figure 2. Autocorrelation mean (± standard error) estimated for three thermal treatments of acclimation for step length (A) and relative angle (B).

Daily thermal variability treatments are: δ0 = 24±0°C, δ4 = 24±4°C and δ8 = 24±8°C. The ACF were calculated for lag  = 0–8. Segmented line indicated the significance limits. Values correspond only to positive correlations. See text for details.

Figure 3. Frequency distribution of turning angle for different thermal treatments.

Daily thermal variability treatments are: δ0 = 24±0°C, δ4 = 24±4°C and δ8 = 24±8°C. The Hartigan's dip statistical test revealed a significant bimodality pattern among all treatments. Namely, δ0: D = 0.0214, P<0.05; δ4: D = 0.0345, P<0.05 and δ8: D = 0.0396, P<0.05.

Figure 4. Frequency distribution of step length for different thermal acclimation conditions.

Daily thermal variability treatments are: δ0 = 24±0°C, δ4 = 24±4°C and δ8 = 24±8°C. The Hartigan's dip statistical test revealed a significant bimodality pattern among all treatments. Namely, δ0: D = 0.013, P>0.05; δ4: D = 0.0123, P>0.05 and δ8: D = 0.0614, P<0.05.

Figure 5. Fractal mean dimension recorded for animals acclimated to different variability thermal regimen.

Daily thermal variability treatments are: δ0 = 24±0°C, δ4 = 24±4°C and δ8 = 24±8°C. Bar showed ±1 standard error.