Practical implications of temporal body mass variation for biologging and comparative studies

Abstract

Body mass is an important factor in studying animal behaviour and helps determine the permissible mass of animal-borne devices. However, body mass varies over time, and ignoring temporal variation may risk (i) overestimating the permissible mass to be deployed on animals and (ii) biasing comparative research. Here, we quantify how daily and annual variation in body mass may result in these biases using data from wild populations of birds (Arabian babbler (Argya squamiceps), n = 2281 body-mass measurements, 145 individuals) and mammals (dwarf mongoose (Helogale parvula), n = 10602 body-mass measurements, 170 individuals). First, for both species, we demonstrate that when a 5% body mass threshold is calculated when the animals are usually heavier (e.g. evening), the deployed device may be equivalent to 5.4%−6% of body mass when the animals are lighter (e.g. on the following morning). Second, we simulate our data to quantify the relationship between imbalanced weighing times and type-I and type-II error rates. For example, a true sex difference may be missed, if Arabian babbler females tend to be weighed in the evenings but males in the mornings. We conclude with practical recommendations for how to avoid these issues, thereby improving both data reliability and animal welfare.