Validating accelerometry-derived proxies of energy expenditure using the doubly labelled water method in the smallest penguin species

Determining energy expenditure in a large seabird using accelerometry

The trade off between energy gained and expended is the foundation of understanding how, why and when animals perform any activity. Based on the concept that animal movements have an energetic cost, accelerometry is increasingly being used to estimate energy expenditure. However, validation of accelerometry as an accurate proxy for field metabolic rate in free-ranging species is limited. In the present study, Australasian gannets (Morus serrator) from the Pope's Eye colony (38°16'42″S 144°41'48″E), south-eastern Australia, were equipped with GPS and tri-axial accelerometers and dosed with doubly labelled water (DLW) to measure energy expenditure during normal behaviour for 3-5 days. The correlation between daily energy expenditure from the DLW and vectorial dynamic body acceleration (VeDBA) was high for both a simple correlation and activity-specific approaches (R2=0.75 and 0.80, respectively). Varying degrees of success were observed for estimating at-sea metabolic rate from accelerometry when removing time on land using published energy expenditure constants (R2=0.02) or activity-specific approaches (R2=0.42). The predictive capacity of energy expenditure models for total and at-sea periods was improved by the addition of total distance travelled and proportion of the sampling period spent at sea during the night, respectively (R2=0.61-0.82). These results indicate that accelerometry can be used to estimate daily energy expenditure in free-ranging gannets and its accuracy may depend on the inclusion of movement parameters not detected by accelerometry.

The role of individual variability on the predictive performance of machine learning applied to large bio-logging datasets

Animal-borne tagging (bio-logging) generates large and complex datasets. In particular, accelerometer tags, which provide information on behaviour and energy expenditure of wild animals, produce high-resolution multi-dimensional data, and can be challenging to analyse. We tested the performance of commonly used artificial intelligence tools on datasets of increasing volume and dimensionality. By collecting bio-logging data across several sampling seasons, datasets are inherently characterized by inter-individual variability. Such information should be considered when predicting behaviour. We integrated both unsupervised and supervised machine learning approaches to predict behaviours in two penguin species. The classified behaviours obtained from the unsupervised approach Expectation Maximisation were used to train the supervised approach Random Forest. We assessed agreement between the approaches, the performance of Random Forest on unknown data and the implications for the calculation of energy expenditure. Consideration of behavioural variability resulted in high agreement (> 80%) in behavioural classifications and minimal differences in energy expenditure estimates. However, some outliers with < 70% of agreement, highlighted how behaviours characterized by signal similarity are confused. We advise the broad bio-logging community, approaching these large datasets, to be cautious when upscaling predictions, as this might lead to less accurate estimates of behaviour and energy expenditure.

Wake respirometry allows breath-by-breath assessment of ventilation and CO2 production in unrestrained animals

Quantifying stress and energetic responses in animals are major challenges, as existing methods lack temporal resolution and elevate animal stress. We propose “wake respirometry,” a new method of quantifying fine-scale changes in CO₂ production in unrestrained animals, using a nondispersive infrared CO₂ sensor positioned downwind of the animal, i.e., in its wake. We parameterize the dispersion of CO₂ in wakes using known CO₂ flow rates and wind speeds. Tests with three bird species in a wind tunnel demonstrated that the system can resolve breath-by-breath changes in CO₂ concentration, with clear exhalation signatures increasing in period and integral with body size. Changes in physiological state were detectable following handling, flight, and exposure to a perceived threat. We discuss the potential of wake respirometry to quantify stress and respiratory patterns in wild animals and provide suggestions for estimating behavior-specific metabolic rates via full integration of CO₂ production across the wake.

•

We use open-path nondispersive infrared spectroscopy CO₂ sensor technology

•

We measure ventilation rate and CO₂ production in the wake of unrestrained animals

•

Rapid responses to stressors and recovery from exercise can be measured

•

Metabolic rate could be calculated by full integration of the wake of exhaled CO₂

Quantity over quality? Prey-field characteristics influence the foraging decisions of little penguins (Eudyptula minor)

Quantifying prey characteristics is important for understanding the foraging behaviour of predators, which ultimately influence the structure and function of entire ecosystems. However, information available on prey is often at magnitudes which cannot be used to infer the fine-scale behaviour of predators, especially so in marine environments where direct observation of predator-prey interactions is rarely possible. In the present study, animal-borne video data loggers were used to determine the influence of prey type and patch density on the foraging behaviour of the little penguin (Eudyptula minor), an important predator in southeastern Australia. We found that numerical density positively influenced time spent foraging at a patch. However, when accounting for calorific value in density estimates, individuals spent longer at dense patches of low-quality prey. This may reflect a trade-off between capture effort and calorific gain as lower quality prey were captured at higher rates. During the breeding season, foraging trip distance and duration is constrained by the need to return to the colony each day to feed offspring. The results of the study suggest that, under these spatio-temporal constraints, little penguins maximize foraging performance by concentrating efforts at larger quantities of prey, irrespective of their calorific quality.

Accelerating animal energetics: High dive costs in a small seabird disrupt the dynamic body acceleration - energy expenditure relationship

Accelerometry has been widely used to estimate energy expenditure in a broad array of terrestrial and aquatic species. However, a recent reappraisal of the method showed that relationships between dynamic body acceleration (DBA) and energy expenditure weaken as the proportion of non-mechanical costs increase. Aquatic air breathing species often exemplify this pattern, as buoyancy, thermoregulation and other physiological mechanisms disproportionately affect oxygen consumption during dives. Combining biologging with the doubly-labelled water method, we simultaneously recorded daily energy expenditure (DEE) and triaxial acceleration in one of the world's smallest wing-propelled breath-hold divers, the dovekie (Alle alle). These data were used to estimate the activity-specific costs of flying and diving and to test whether overall dynamic body acceleration (ODBA) is a reliable predictor of DEE in this abundant seabird. Average DEE for chick-rearing dovekies was 604±119 kJ/d across both sampling years. Despite recording lower stroke frequencies for diving than for flying (in line with allometric predictions for auks), dive costs were estimated to surpass flight costs in our sample of birds (flying: 7.24, diving: 9.37 X BMR). As expected, ODBA was not an effective predictor of DEE in this species. However, accelerometer-derived time budgets did accurately estimate DEE in dovekies. This work represents an empirical example of how the apparent energetic costs of buoyancy and thermoregulation limit the effectiveness of ODBA as the sole predictor of overall energy expenditure in small shallow-diving endotherms.