Skip to main content
. Author manuscript; available in PMC: 2021 Jan 1.
Published in final edited form as: J Anim Ecol. 2019 Jun 7:10.1111/1365-2656.13040. doi: 10.1111/1365-2656.13040

Figure 7.

Figure 7

Theoretical calculations for relationships between mean power and mean DBA for a time period during which a study animal, here a beaver, can spend time spent engaged in either ‘inactivity’ or ‘activity’. Both have defined power costs (from Allers & Culik (1997) cf. Fig. 4) with randomly allocated lengths of times to both conditions. The blue line shows the situation where DBA is the only modulator of power, whereas the red line shows how the slope of the regression decreases dramatically if thermoregulation costs (also given in Allers & Culik 1997) are included. Since it is extremely unlikely that the study animals will have engaged in only one activity, the real spread of mean power and mean DBA values will be compromised accordingly (the inserted box shows the 95% CI for the grand mean of ODBA measurements on 14 wild beavers (data from Graf et al. 2016)), to which system errors and individual variation (cf. Fig. 2a) must be added. All this dramatically decreases the ability of the researcher to ascertain any relationship between DBA and power.