Measuring body temperature in birds - the effects of sensor type and placement on estimated temperature and metabolic rate.

Several methods are routinely used to measure avian body temperature, but different methods vary in invasiveness. This may cause stress-induced increases in temperature and/or metabolic rate and, hence, overestimation of both parameters. Choosing an adequate temperature measurement method is therefore key to accurately characterizing an animal's thermal and metabolic phenotype. Using great tits (Parus major) and four common methods with different levels of invasiveness (intraperitoneal, cloacal, subcutaneous, cutaneous), we evaluated the preciseness of body temperature measurements and effects on resting metabolic rate (RMR) over a 40°C range of ambient temperatures. None of the methods caused overestimation or underestimation of RMR compared with un-instrumented birds, and body or skin temperature estimates did not differ between methods in thermoneutrality. However, skin temperature was lower compared with all other methods below thermoneutrality. These results provide empirical guidance for future research that aims to measure body temperature and metabolic rate in small bird models.