Accelerometers can measure total and activity‐specific energy expenditures in free‐ranging marine mammals only if linked to time‐activity budgets

Abstract

Summary Energy expenditure is an important component of foraging ecology, but is extremely difficult to estimate in free‐ranging animals and depends on how animals partition their time between different activities during foraging. Acceleration data have emerged as a new way to determine energy expenditure at a fine scale but this needs to be tested and validated in wild animals. This study investigated whether vectorial dynamic body acceleration (VeDBA) could accurately predict the energy expended by marine predators during a full foraging trip. We also aimed to determine whether the accuracy of predictions of energy expenditure derived from acceleration increased when partitioned by different types of at‐sea activities (i.e. diving, transiting, resting and surface activities). To do so, we equipped 20 lactating northern (Callorhinus ursinus) and 20 lactating Antarctic fur seals (Arctocephalus gazella) with GPS, time‐depth recorders and tri‐axial accelerometers and obtained estimates of field metabolic rates using the doubly labelled water (DLW) method. VeDBA was derived from tri‐axial acceleration, and at‐sea activities (diving, transiting, resting and surface activities) were determined using dive depth, tri‐axial acceleration and travelling speed. We found that VeDBA did not accurately predict the total energy expended by fur seals during their full foraging trips (R2 = 0·36). However, the accuracy of VeDBA as a predictor of total energy expenditure increased significantly when foraging trips were partitioned by activity and when activity‐specific VeDBA was paired with time‐activity budgets (R2 = 0·70). Activity‐specific VeDBA also accurately predicted the energy expenditures of each activity independent of each other (R2 > 0·85). Our study confirms that acceleration is a promising way to estimate energy expenditures of free‐ranging marine mammals at a fine scale never attained before. However, it shows that it needs to be based on the time‐activity budgets that make up foraging trips rather than being derived as a single measure of VeDBA applied to entire foraging trips. Our activity‐based method provides a cost‐effective means to accurately calculate energy expenditures of fur seals using acceleration and time‐activity budgets, that can be transfered to studies on other species.

Processing of acceleration and dive data on‐board satellite relay tags to investigate diving and foraging behaviour in free‐ranging marine predators

“…For example, prey catch attempt (PrCA) behaviours can be identified from peaks in acceleration indicative of rapid head movements (Gallon et al., ; Naito, Bornemann, Takahashi, McIntyre, & Plotz, ; Suzuki, Naito, Folkow, Miyazaki, & Blix, ; Viviant et al., ; Volpov, Hoskins, et al., ), whilst proxies of energetic expenditure (e.g. swimming effort) can be calculated by isolating dynamic movement rates (Jeanniard‐du‐Dot, Guinet, Arnould, Speakman, & Trites, ; Sato, Mitani, Cameron, Siniff, & Naito, ; Volpov, Rosen, Trites, & Arnould, ). The latter of these can additionally be used in tandem with estimations of pitch and vertical speed to give a relative measure of body condition (Aoki et al., ; Richard et al., ).…”

Relationships grow with time: a note of caution about energy expenditure‐proxy correlations, focussing on accelerometry as an example

“…Jeanniard‐du‐Dot et al . (, their fig. ) present strong relationships between total energy expended and summed DBA across foraging trips in both northern and Antarctic fur seals ( C. ursinus and Arctocephalus gazella ).…”

“…; Jeanniard‐du‐Dot et al . ,b). And in recent years, I have refereed a number of submitted manuscripts from a host of research groups where analyses have centred around summed values of energy expenditure and DBA to test for evidence that DBA relates to metabolic rate; sometimes other referees for these manuscripts have suggested summing values as an improvement to the analysis.…”

“…This was achieved by randomly generating the model's input values (metabolic rate, DBA and logger recording duration), assuming a flat underlying distribution, within the ranges for those values reported in one of the aforementioned papers (Jeanniard‐du‐Dot et al . () for northern fur seals. In my view, this paper, for the reason argued in this article, concludes there is a strong relationship between energy expenditure and DBA without presenting sufficient evidence.…”

Jumping in the Night: An Investigation of the Leaping Activity of the Western Tarsier (Cephalopachus bancanus borneanus) Using Accelerometers

“…Data on dynamic acceleration can be used to calculate the vector sum of the dynamic body acceleration (VeDBA) or the overall dynamic body acceleration (ODBA), which indicate body motion when G-force values are higher than 0. VeDBA and ODBA values may also provide information about energy expenditure [e.g., Qasem et al, 2012;Jeanniard-du-Dot et al, 2017]. Accelerometer data can be used to derive daily activity budgets [e.g., when an animal moves; Yoda et al, 1999;Lagarde et al, 2008;Grü newä lder et al, 2012;Williams et al, 2014] and, if accelerometer and viewer-observed behavioural data are collected simultaneously, accelerometer data can also be used to characterise behaviours (e.g., walking, running, and leaping) [Sakamoto et al, 2009;Nathan et al, 2012;Graf et al, 2015] and to estimate the energy expenditure associated with different behaviours [Wilson et al, 2006;Qasem et al, 2012;Jeanniard-du-Dot et al, 2017].…”