Comparing predicted vs. observed dive behaviours of marine endotherms: Insight into the diving behaviours of Pterodroma gouldi, Megadyptes antipodes, Phocarctos hookeri, and Balaenoptera musculus
Comparing predicted vs. observed dive behaviours of marine endotherms: Insight into the diving behaviours of Pterodroma gouldi, Megadyptes antipodes, Phocarctos hookeri, and Balaenoptera musculus
Abstract
In this study, we utilised equations from Halsey et al. (2006) to estimate the dive behaviours of four marine endotherm species. Predictions included mean dive duration (s), mean maximum duration (s), mean dive depth (m), and maximum dive depth (m). Calculations were based on species mass (kg), applying provided formulas for seabirds and mammals. These predicted values were compared to observed data from existing research. Our results revealed significant discrepancies between predicted and observed dive behaviours across species, with Pterodroma gouldi showing the closest match and Balaenoptera musculus exhibiting the largest deviations. The observed mean dive depth was consistently lower than the maximum, with notable anomalies in P. gouldi and Megadyptes antipodes. These inconsistencies highlight potential limitations in the Halsey formulas, which may not fully account for ecological and behavioural factors such as breeding status and prey availability. Our findings underscore the need for more comprehensive models to predict species-specific diving behaviours better, integrating ecological and behavioural variables to enhance accuracy and inform conservation efforts. Future research should address data errors and incorporate long-term observations to refine predictive models and understand the effects of environmental changes on dive behaviours.
1. Introduction
Breath-hold diving is a remarkable adaptation observed in marine endotherms, enabling these animals to exploit underwater environments for feeding, migration, and predator avoidance. Marine mammals such as whales, sea lions, and seabirds like penguins and petrels exhibit varying degrees of diving proficiency, showcasing a range of physiological and morphological adaptations that support diving for extended periods.
Central to this capability is the dive reflex, a suite of responses that optimise oxygen conservation during submersion. Key components of the dive reflex include bradycardia (slowed heart rate), peripheral vasoconstriction (reduced blood flow to non-essential organs), and increased blood flow to vital organs such as the heart and brain (Trassinelli, 2016). These adjustments reduce overall oxygen consumption and prioritise oxygen supply to critical tissues, extending the duration of dives.
Understanding these morphological adaptations illuminates these animals' capabilities and provides insights into their ecology, feeding habits, breeding characteristics and sometimes anthropological pressures that have shaped their diving abilities. This study reviews the dive behaviours of four marine endotherms: Pterodroma gouldi (grey-faced petrel), Megadyptes antipodes (yellow-eyed penguin), Phocarctos hookeri (NZ sea lion), and Balaenoptera musculus (blue whale) and compares the predicted to observed dive behaviours from various studies.
2. Methods and Materials
Equations from Halsey et al. (2006) were used in Excel to estimate each species' dive behaviours. Four behaviours were predicted, these being mean dive duration (seconds), mean maximum duration (seconds), mean dive depth (meters) and maximum dive depth (meters).
First, we used the species' provided mass (kg) to calculate the respective log mass, log intercept, and log mean dive depth. Using the equation 10log mean dive depth, we gathered the mean dive depth for each species. The same results were obtained using the formula: 10.5M0.389 for seabirds and 3.8M0.389 for mammals, where M = mass. We calculated the maximum dive depth by using the formula 9.4M0.327.
The mean dive duration was calculated using the formula 21.2M0.368, and the mean maximum duration using 35.5M0.326. Each predicted value of the species' dive behaviours was placed into a table and compared to those found from prior research (actual/observed). The maximum observed values were averaged to calculate the mean maximum duration values were not provided. 3. Results
The values calculated using the equations provided by Halsey et al. (2006) were compared with observed values collected from prior research. Table 1 illustrates notable disparities between the predicted and observed values for the recorded species. P. gouldi exhibited the lowest values across all dive behaviours among these species. At the same time, M. antipodes and P. hookeri displayed higher values, with B. musculus exhibiting the highest (smallest - largest mass).
Across all species, the observed mean dive depth consistently fell below the maximum depth. However, irregularities were noted in P. gouldi, with a predicted mean dive depth of 8.38 m and a predicted maximum dive depth of 7.78 m. Similarly, M. antipodes exhibited a predicted mean dive depth of 19.94 m and a predicted maximum dive depth of 16.12 m. However, observed values for both species state otherwise. Notably, P. gouldi and B. musculus displayed lower observed mean dive depth than predicted, while P. hookeri and M. antipodes exhibited higher.P. gouldi demonstrated the most accurate predictions, with a minimal discrepancy of -3.68 seconds in observed mean dive depth. However, variations including mean dive duration (observed = 4 seconds, predicted = 17.12 seconds) and mean dive depth (observed = 4.7 meters, predicted = 8.38 meters) were lower than predicted. In contrast, the mean maximum duration (observed = 55.49 seconds, predicted = 29.39 seconds) and maximum dive depth (observed = 23.6 meters, predicted = 7.78 meters) exceeded predictions.The most significant deviations were in B. musculus, particularly in mean dive duration, with a -1,120.75 second (-18.6 minutes) difference for observed values (observed = 348 seconds, predicted = 1,468.75 seconds). Overall, the observed values for B. musculus were notably lower than predicted. In contrast, P. hookeri and M. antipodes consistently displayed observed values considerably higher than predicted.
Table 1. The predicted vs observed values of dive behaviour between P. gouldi, M. antipodes, P. hookeri, and B. musculus. Predicted values were calculated using formulas published by Halsey et al. (2006), while observed values were collected through the corresponding research and literature.4. Discussion
4.1. Pterodroma gouldi
P. gouldi displayed the most accurate predictions with little discrepancy compared to the observed values from prior research. However, the observed mean dive depth and mean dive duration being lower than predicted may be due to complications within Taylor’s research. Taylor stated that only 51% of the birds deployed with Maximum depth gauges (MDGs) were recollected, and 36% had lost or malfunctioned MDGs. This error could account for the lower observed dive depths and durations (Taylor, 2008). Additionally, breeding males dived deeper than non-breeding males (7.5 ± 6.7 m vs 2.2 ± 0.9 m). The Halsey model may not necessarily account for the breeding status of species; therefore, the large difference between the predicted (7.78 m) and observed (23.6 m) maximum dive depth can be accounted for.
Similarly, the dataset portraying mean dive duration and mean maximum duration was obtained by adults in the incubation phase of the breeding cycle (Dunphy et al., 2015). Interestingly, the observed value of mean dive duration was fairly lower than predicted (observed = 4 seconds, predicted = 17.12 seconds). While the observed mean maximum duration was 55.49 seconds, it is possible that the low value of 4 seconds for mean dive duration is due to the breeding cycle, limiting trips to ensure nest safety. When considering the irregularity between the predicted mean dive depth and maximum dive depth, with the mean being higher than the maximum, it is likely due to a misconception or anomaly in the Halsey formula that does not accurately represent observations. This limitation needs to be studied further to reduce outliers in further research.4.2. Balaenoptera musculus
B. musculus showed the most significant deviations between predicted and observed values, particularly in mean dive duration. The observed value had a discrepancy of -18.6 minutes, possibly due to errors in the dataset during research (Lagerquist, 1997). Although these errors were not included in the dataset, they may have introduced bias, as 19% of transmission errors were received from DEP-1, the whale with the longest dive duration. Additionally, this dataset's shorter mean dive durations may be related to whaling ship avoidance, a prevalent issue in 1997 (Doi, 1974). Newer data indicates that B. musculus have dive durations of up to 1,008 seconds (Davenport et al., 2022), much closer to Halsey’s predicted value.
Similarly, the observed mean and maximum dive depths were much lower than predicted, which can be attributed to transmission errors from DEP-1, the only whale for which dive information was available (Lagerquist, 1997). These errors can skew the data, creating bias and misconceptions. Additionally, 75% of the recorded dives were between 0-16 meters, favouring shallower dives. Newer data shows that B. musculus has a mean dive depth of 189-354 m during deep searching (Davenport et al., 2022). Although these papers do not reference why diving depth may have increased over the last 25 years, it is plausible that rising ocean temperatures could drive krill and other prey into deeper, colder waters. This can also increase the energy conservation of animals, potentially including B. musculus (Kaartvedt, 2010). Further research is required to understand the increase in B. musculus dive depth, promoting ecological awareness and management of these diving behaviours.4.3. Megadyptes antipodes
M. antipodes consistently displayed observed values considerably higher than predicted, with the most notable discrepancies in observed maximum dive depth and mean/mean maximum dive duration. The dataset indicates that all recorded dives were within 10 km of the nesting sites. This proximity may have influenced diving behaviour, causing the birds to dive deeper and longer than predicted due to foraging behaviour during nesting (Chilvers et al., 2014). Additionally, the analysis highlighted the importance of diet, with the bottom-dwelling opal fish (Hemerocoetes monopterygius) being a confounding factor for dive depth and duration.The depth of prey, alongside breeding behaviours, likely contributed to the higher observed values compared to predictions. The Halsey formula may not account for these nesting or dietary behaviours. Like P. gouldi, M. antipodes showed a higher observed mean dive depth of 19.94 m and a lower predicted maximum dive depth of 16.12 m, possibly due to Halsey formula anomalies.4.4. Phocarctos hookeri
All observed values were higher than predicted for P. hookeri, most notably, the mean dive depth (+101.1 m), maximum dive depth (+505.9 m), and mean maximum duration (+332.22 seconds). Each of the recorded sea lions involved healthy females with pups. One interesting observation was that the largest mass sea lion (E-59, 155 kg female) achieved the deepest dive of 550 m and the longest duration of 11.5 minutes or 690 seconds (Costa & Gales, 2000). Additionally, the higher mass sea lions (>100 kg), including E-59, conducted dives close to their calculated aerobic dive limit (cADL). Conversely, lower-mass sea lions (<100 kg) conducted dives exceeding their cADL. These findings support the idea that differences in diving behaviours were due to prey consumption, varying with size and depth. Because the Halsey formula only accounts for mass when constructing predictions, the higher observed values may be due to the lack of prey information.4.5. Conclusions
The study demonstrates that observed dive behaviours in P. gouldi, M. antipodes, P. hookeri, and B. musculus significantly deviate from predictions made using the Halsey et al. (2006) formulas. P. gouldi exhibited the most accurate predictions with minor discrepancies, while B. musculus showed the largest deviations.These findings suggest that the Halsey formula may only partially account for various ecological and behavioural factors influencing dive behaviour, such as prey availability, breeding status, and foraging strategies. This has broader implications for the field, indicating a need for more comprehensive models that integrate these factors. Understanding these discrepancies can enhance our knowledge of species-specific diving behaviours and improve ecological management strategies.While the study provides valuable insights, limitations such as potential data errors, transmission and equipment issues, and comprehensive prey information must be acknowledged. These factors may affect the accuracy and interpretation of the results. Further research should address these limitations to refine predictive models.Future studies should explore the ecological and behavioural variables in predictive models like Halsey’s. Long-term data collection across diverse populations and environments will be crucial to validate these findings and understand the impact of changing ocean conditions on dive behaviours.In conclusion, while the Halsey formula offered a foundational approach to predicting dive behaviours, integrating ecological and behavioural complexities is essential for more accurate and applicable predictions, ultimately aiding in the conservation and management of marine species.References
Chilvers, B., Dobbins, M., & Edmonds, H. (2014). Diving behaviour of yellow-eyed penguins, Port Pegasus/Pikihatiti, Stewart Island/Rakiura, New Zealand. New Zealand Journal of Zoology, 41(3), 161–170. https://doi.org/10.1080/03014223.2014.908931Costa, D. P., & Gales, N. J. (2000). Foraging energetics and diving behavior of lactating New Zealand sea lions, phocarctos Hookeri. Journal of Experimental Biology, 203(23), 3655–3665. https://doi.org/10.1242/jeb.203.23.3655Davenport, A. M., Erbe, C., Jenner, M. M., Jenner, K. C. S., Saunders, B. J., & McCauley, R. D. (2022). Pygmy blue whale diving behaviour reflects song structure. Journal of Marine Science and Engineering, 10(9), 1227. https://doi.org/10.3390/jmse10091227Doi, T. (1974). CHAPTER 16. FURTHER DEVELOPMENT OF WHALE SIGHTING THEORY. In Harvard University Press eBooks (pp. 359–368). https://doi.org/10.4159/harvard.9780674367128.c20Dunphy, B., Taylor, G., Landers, T., Sagar, R., Chilvers, B., Ranjard, L., & Rayner, M. (2015). Comparative seabird diving physiology: first measures of haematological parameters and oxygen stores in three New Zealand Procellariiformes. Marine Ecology. Progress Series, 523, 187–198. https://doi.org/10.3354/meps11195Halsey, L. G., Butler, P. J., & Blackburn, T. M. (2006). A phylogenetic analysis of the allometry of diving. the American Naturalist, 167(2), 276–287. https://doi.org/10.1086/499439Kaartvedt, S. (2010). Diel Vertical Migration Behaviour of the Northern Krill (Meganyctiphanes norvegica Sars). In Advances in marine biology (pp. 255–275). https://doi.org/10.1016/b978-0-12-381308-4.00009-1Lagerquist, B. A. (1997). Dive characteristics of Northeast Pacific blue whales (Balaenoptera musculus) monitored by satellite telemetry. https://ir.library.oregonstate.edu/downloads/gh93h358gTaylor, G. (2008). Maximum dive depths of eight New Zealand Procellariiformes, including Pterodroma species. Papers and Proceedings of the Royal Society of Tasmania, 89–97. https://doi.org/10.26749/rstpp.142.1.89Trassinelli, M. (2016). Energy cost and optimisation in breath-hold diving. Jounal of Theoretical Biology, 396, 42–52. https://doi.org/10.1016/j.jtbi.2016.02.009
