Activity patterns of free-ranging koalas (Phascolarctos cinereus) revealed by accelerometry

A retrospective epidemiological study of sarcoptic mange in koalas (Phascolarctos cinereus) using wildlife carer admission records

Outbreaks of sarcoptic mange are sporadically reported in koala populations across Australia, but disease characteristics (e.g., distribution across the body) remain poorly understood. In an area of Northern Victoria regular cases coming into care suggest mange may have become enzootic, and here we characterise those koala mange admission records. In 18% (n = 10) of mange affected koala reports that had a recorded outcome (n = 55), the animals died before the carers could locate them, and of the remaining 45 koalas that were alive upon carer arrival, 80% (n = 36) had to be euthanised due to severe mange. The number of admissions varied among years (highest observed in 2019), and over 60% of affected koala admissions were male. Male admissions peaked in austral spring and again in late austral summer-autumn (mating and birthing seasons), with female admissions only exhibiting the latter peak (birthing season). Fissures of the epidermis of the front paws occurred in 100% of admitted koalas, with 70% also showing these signs elsewhere on ventral surfaces or limbs. Only male koalas had signs of mange on the chest and face, and only female koalas had signs of mange on their back. Collectively, this study suggests sarcoptic mange can be a severe disease in koalas, and that male koalas may play an important role in seasonal transmission dynamics. We discuss how these findings may help inform intervention strategies.

Comparing the cost-effectiveness of drones, camera trapping and passive acoustic recorders in detecting changes in koala occupancy

Quantifying the cost-effectiveness of alternative sampling methods is crucial for efficient biodiversity monitoring and detection of population trends. In this study, we compared the cost-effectiveness of three novel sampling methods for detecting changes in koala (Phascolarctos cinereus) occupancy: thermal drones, passive acoustic recorders and camera trapping. Specifically, we fitted single-season occupancy-detection models to data recorded from 46 sites in eight bioregions of New South Wales, Australia, between 2018 and 2022. We explored the effect of weather variables on daily detection probability for each method and, using these estimates, calculated the statistical power to detect 30%, 50% and 80% declines in koala occupancy. We calculated power for different combinations of sites (1-200) and repeat surveys (2-40) and developed a cost model that found the cheapest survey design that achieved 80% power to detect change. On average, detectability of koalas was highest with one 24-h period of acoustic surveys (0.32, 95% CI's: 0.26, 0.39) compared to a 25-ha flight of drone surveys (0.28, 95% 0.15, 0.48) or a 24-h period of camera trapping consisting of six cameras (0.019, 95% CI's: 0.014, 0.025). We found a negative quadratic relationship between detection probability and air temperature for all three methods. Our power and cost analysis suggested that 148 sites surveyed with acoustic recorders deployed for 14 days would be the cheapest method to sufficiently detect a 30% decline in occupancy with 80% power. We recommend passive acoustic recorders as the most efficient sampling method for monitoring koala occupancy compared to cameras or drones. Further comparative studies are needed to compare the relative effectiveness of these methods and others when the monitoring objective is to detect change in koala abundance over time.

Divergent evolution of sleep in Drosophila species

Living organisms synchronize their biological activities with the earth's rotation through the circadian clock, a molecular mechanism that regulates biology and behavior daily. This synchronization factually maximizes positive activities (e.g., social interactions, feeding) during safe periods, and minimizes exposure to dangers (e.g., predation, darkness) typically at night. Beyond basic circadian regulation, some behaviors like sleep have an additional layer of homeostatic control, ensuring those essential activities are fulfilled. While sleep is predominantly governed by the circadian clock, a secondary homeostatic regulator, though not well-understood, ensures adherence to necessary sleep amounts and hints at a fundamental biological function of sleep beyond simple energy conservation and safety. Here we explore sleep regulation across seven Drosophila species with diverse ecological niches, revealing that while circadian-driven sleep aspects are consistent, homeostatic regulation varies significantly. The findings suggest that in Drosophilids, sleep evolved primarily for circadian purposes. The more complex, homeostatically regulated functions of sleep appear to have evolved independently in a species-specific manner, and are not universally conserved. This laboratory model may reproduce and recapitulate primordial sleep evolution.

Hot climate, hot koalas: the role of weather, behaviour and disease on thermoregulation

Thermoregulation is critical for endotherms living in hot, dry conditions, and maintaining optimal core body temperature (Tb) in a changing climate is an increasingly challenging task for mammals. Koalas (Phascolarctos cinereus) have evolved physiological and behavioural strategies to maintain homeostasis and regulate their Tb but are thought to be vulnerable to prolonged heat. We investigated how weather, behaviour and disease influence Tb for wild, free-living koalas during summer in north-west New South Wales. We matched Tb with daily behavioural observations in an ageing population where chlamydial disease is prevalent. Each individual koala had similar Tb rhythms (average Tb = 36.4 ± 0.05°C), but male koalas had higher Tb amplitude and more pronounced daily rhythm than females. Disease disrupted the 24-hr circadian pattern of Tb. Koala Tb increased with ambient temperature (Ta). On the hottest day of the study (maximum Ta = 40.8°C), we recorded the highest (Tb = 40.8°C) but also the lowest (Tb = 32.4°C) Tb ever documented for wild koalas, suggesting that they are more heterothermic than previously recognized. This requires individuals to predict days of extreme Ta from overnight and early morning conditions, adjusting Tb regulation accordingly, and it has never been reported before for koalas. The large diel amplitude and low minimum Tb observed suggest that koalas at our study site are energetically and nutritionally compromised, likely due to their age. Behaviour (i.e. tree hugging and drinking water) was not effective in moderating Tb. These results indicate that Ta and koala Tb are strongly interconnected and reinforce the importance of climate projections for predicting the future persistence of koalas throughout their current distribution. Global climate models forecast that dry, hot weather will continue to escalate and drought events will increase in frequency, duration and severity. This is likely to push koalas and other arboreal folivores towards their thermal limit.

The behaviour and activity budgets of two sympatric sloths; Bradypus variegatus and Choloepus hoffmanni

It is usually beneficial for species to restrict activity to a particular phase of the 24-hour cycle as this enables the development of morphological and behavioural adaptations to enhance survival under specific biotic and abiotic conditions. Sloth activity patterns are thought to be strongly related to the environmental conditions due to the metabolic consequences of having a low and highly variable core body temperature. Understanding the drivers of sloth activity and their ability to withstand environmental fluctuations is of growing importance for the development of effective conservation measures, particularly when we consider the vulnerability of tropical ecosystems to climate change and the escalating impacts of anthropogenic activities in South and Central America. Unfortunately, the cryptic nature of sloths makes long term observational research difficult and so there is very little existing literature examining the behavioural ecology of wild sloths. Here, we used micro data loggers to continuously record, for the first time, the behaviour of both Bradypus and Choloepus sloths over periods of days to weeks. We investigate how fluctuations in the environmental conditions affect the activity of sloths inhabiting a lowland rainforest on the Caribbean coast of Costa Rica and examined how this might relate to their low power lifestyle. Both Bradypus and Choloepus sloths were found to be cathemeral in their activity, with high levels of between-individual and within-individual variation in the amounts of time spent active, and in the temporal distribution of activity over the 24-hour cycle. Daily temperature did not affect activity, although Bradypus sloths were found to show increased nocturnal activity on colder nights, and on nights following colder days. Our results demonstrate a distinct lack of synchronicity within the same population, and we suggest that this pattern provides sloths with the flexibility to exploit favourable environmental conditions whilst reducing the threat of predation.

Identification of reindeer fine-scale foraging behaviour using tri-axial accelerometer data

Animal behavioural responses to the environment ultimately affect their survival. Monitoring animal fine-scale behaviour may improve understanding of animal functional response to the environment and provide an important indicator of the welfare of both wild and domesticated species. In this study, we illustrate the application of collar-attached acceleration sensors for investigating reindeer fine-scale behaviour. Using data from 19 reindeer, we tested the supervised machine learning algorithms Random forests, Support vector machines, and hidden Markov models to classify reindeer behaviour into seven classes: grazing, browsing low from shrubs or browsing high from trees, inactivity, walking, trotting, and other behaviours. We implemented leave-one-subject-out cross-validation to assess generalizable results on new individuals. Our main results illustrated that hidden Markov models were able to classify collar-attached accelerometer data into all our pre-defined behaviours of reindeer with reasonable accuracy while Random forests and Support vector machines were biased towards dominant classes. Random forests using 5-s windows had the highest overall accuracy (85%), while hidden Markov models were able to best predict individual behaviours and handle rare behaviours such as trotting and browsing high. We conclude that hidden Markov models provide a useful tool to remotely monitor reindeer and potentially other large herbivore species behaviour. These methods will allow us to quantify fine-scale behavioural processes in relation to environmental events.

Balancing the scales: Preliminary investigation of total energy expenditure and daily metabolizable energy intake in Matschie’s tree kangaroo (Dendrolagus matschiei)

Matschie’s tree kangaroo (Dendrolagus matschiei) is an endangered arboreal marsupial native to Papua New Guinea. Detailed field studies of its behavior and ecology are scarce due largely to its occupation of remote cloud forests and cryptic nature. Although this species has been in human care since the 1950s, much of its biology is still unknown. The current ex situ population is not sustainable due to health and reproductive problems, believed to stem largely from issues with diet and obesity. To better assess potential discrepancies between energy requirements and energy intake, we sought to 1) quantify total energy expenditure (TEE) of two zoo-housed Matschie’s tree kangaroos (body mass = 9.0–9.7 kg) on a diet composed largely of leafy browse; 2) quantify food and macronutrient intake, apparent dry matter macronutrient digestibility, and metabolizable energy (ME) intake over a 14-month period; and 3) test for seasonal changes in ME intake due to seasonal differences in the varieties of leafy browse offered. Using the doubly labeled water method, we determined TEE for the female (288 kcal day ^-1) and male (411 kcal day ^-1). Resulting mean TEE was well below the expected value for marsupials and macropods (i.e., ~60% of the expected value based on body mass). The mean calculated ME intakes for the female and male were 307 kcal day^-1 and 454 kcal day^-1, respectively. There were significant seasonal differences in ME intakes, driven by reduced intake in the autumn. These results demonstrate that Matschie’s tree kangaroos can be maintained at healthy body weights and conditions on fiber-rich and browse-heavy diets. Our findings contribute important insights into tree kangaroo energetics and physiology and can be applied to help reformulate the diet of Matschie’s tree kangaroos at captive facilities to improve population health and sustainability.

Temporal effect of feeding on the body temperature and behaviour of captive koalas (Phascolarctos cinereus)

Free-ranging koalas generally feed at night; however, captive koalas are usually fed during the day in order to encourage activity for display purposes. We studied the temporal effect of feeding on body temperature of captive koalas in Queensland, to determine whether nocturnal feeding may be beneficial for koalas in warmer climates. Six adult koalas were implanted with thermal transmitters and data loggers, waxed together as a single package, to record internal body temperature. Koalas were exposed to two treatments: koalas were fed in the morning (between 0730 and 0830 hours) during the AM treatment or late afternoon (between 1700 and 1800 hours) for the PM treatment. The body temperature of koalas fed in the mornings was on average 0.5°C higher at its peak (P ≤ 0.01) when compared to koalas fed in the evening. Furthermore, the body temperature maxima of morning-fed koalas was reached ~2 h earlier in the afternoon, compared with those fed in the evening. There was no significant difference between behaviours associated with the two feeding regimes: inactivity (P = 0.840), feeding (P = 0.472) and activity (P = 0.634). We postulate that nocturnal feeding by koalas may be an adaptive mechanism that reduces diurnal heat load during times of high environmental temperatures.

Glucocorticoid Production, Activity Levels, And Personality Traits Of Fennec Foxes (Vulpes zerda) Managed For Different Roles In Zoos

Fecal glucocorticoid metabolite (FGM) concentrations, activity, and personality were assessed for 35 fennec foxes (Vulpes zerda) to determine whether animals managed as ambassadors differed from exhibit or off-exhibit animals. A FGM assay, pedometer, and personality assessment tool were validated. Then, fecal samples and movement data were collected during winter and summer periods. Handling was recorded, and the personality of each fox was evaluated. Generalized linear mixed models assessed the relationships between FGM concentrations, activity, personality, handling, sex, season, rearing type, and role. FGM concentrations did not differ in relation to role or handling. Foxes were most active at night; the time of peak activity did not vary with role or handling. Foxes were more active in winter; males were more active than females, and ambassador foxes were more active than off-exhibit animals. Hand-reared foxes were more sociable, and, at one institution, ambassador foxes were more sociable than foxes in other roles. These results suggest that management for ambassador programs is not associated with changes in glucocorticoid production or circadian patterns but may increase activity and be associated with greater sociability.

Quantifying finer-scale behaviours using self-organising maps (SOMs) to link accelerometery signatures with behavioural patterns in free-roaming terrestrial animals

Collecting quantitative information on animal behaviours is difficult, especially from cryptic species or species that alter natural behaviours under observation. Using harness-mounted tri-axial accelerometers free-roaming domestic cats (Felis Catus) we developed a methodology that can precisely classify finer-scale behaviours. We further tested the effect of a prey-protector device designed to reduce prey capture. We aligned accelerometer traces collected at 50 Hz with video files (60 fps) and labelled 12 individual behaviours, then trained a supervised machine-learning algorithm using Kohonen super self-organising maps (SOM). The SOM was able to predict individual behaviours with a ~ 99.6% overall accuracy, which was slightly better than for random forest estimates using the same dataset (98.9%). There was a significant effect of sample size, with precision and sensitivity decreasing rapidly below 2000 1-s observations. We were also able to detect a behaviour specific reduction in the predictability when cats were fitted with the prey-protector device indicating it altered biomechanical gait. Our results can be applied in movement ecology, zoology and conservation, where habitat specific movement performance between predators or prey may be critical to managing species of conservation significance, or in veterinary and agricultural fields, where early detection of movement pathologies can improve animal welfare.

An evaluation of machine learning classifiers for next-generation, continuous-ethogram smart trackers

BACKGROUND

Our understanding of movement patterns and behaviours of wildlife has advanced greatly through the use of improved tracking technologies, including application of accelerometry (ACC) across a wide range of taxa. However, most ACC studies either use intermittent sampling that hinders continuity or continuous data logging relying on tracker retrieval for data downloading which is not applicable for long term study. To allow long-term, fine-scale behavioural research, we evaluated a range of machine learning methods for their suitability for continuous on-board classification of ACC data into behaviour categories prior to data transmission.

METHODS

We tested six supervised machine learning methods, including linear discriminant analysis (LDA), decision tree (DT), support vector machine (SVM), artificial neural network (ANN), random forest (RF) and extreme gradient boosting (XGBoost) to classify behaviour using ACC data from three bird species (white stork Ciconia ciconia, griffon vulture Gyps fulvus and common crane Grus grus) and two mammals (dairy cow Bos taurus and roe deer Capreolus capreolus).

RESULTS

Using a range of quality criteria, SVM, ANN, RF and XGBoost performed well in determining behaviour from ACC data and their good performance appeared little affected when greatly reducing the number of input features for model training. On-board runtime and storage-requirement tests showed that notably ANN, RF and XGBoost would make suitable on-board classifiers.

CONCLUSIONS

Our identification of using feature reduction in combination with ANN, RF and XGBoost as suitable methods for on-board behavioural classification of continuous ACC data has considerable potential to benefit movement ecology and behavioural research, wildlife conservation and livestock husbandry.

Early-life behaviour predicts first-year survival in a long-distance avian migrant

Early-life conditions have critical, long-lasting effects on the fate of individuals, yet early-life activity has rarely been linked to subsequent survival of animals in the wild. Using high-resolution GPS and body-acceleration data of 93 juvenile white storks (Ciconia ciconia), we examined the links between behaviour during both pre-fledging and post-fledging (fledging-to-migration) periods and subsequent first-year survival. Juvenile daily activity (based on overall dynamic body acceleration) showed repeatable between-individual variation, the juveniles' pre- and post-fledging activity levels were correlated and both were positively associated with subsequent survival. Daily activity increased gradually throughout the post-fledging period, and the relationship between post-fledging activity and survival was stronger in individuals who increased their daily activity level faster (an interaction effect). We suggest that high activity profiles signified individuals with increased pre-migratory experience, higher individual quality and perhaps more proactive personality, which could underlie their superior survival rates. The duration of individuals' fledging-to-migration periods had a hump-shaped relationship with survival: higher survival was associated with intermediate rather than short or long durations. Short durations reflect lower pre-migratory experience, whereas very long ones were associated with slower increases in daily activity level which possibly reflects slow behavioural development. In accordance with previous studies, heavier nestlings and those that hatched and migrated earlier had increased survival. Using extensive tracking data, our study exposed new links between early-life attributes and survival, suggesting that early activity profiles in migrating birds can explain variation in first-year survival.

Quantifying koala locomotion strategies: implications for the evolution of arborealism in marsupials

The morphology and locomotor performance of a species can determine their inherent fitness within a habitat type. Koalas have an unusual morphology for marsupials, with several key adaptations suggested to increase stability in arboreal environments. We quantified the kinematics of their movement over ground and along narrow arboreal trackways to determine the extent to which their locomotion resembled that of primates, occupying similar niches, or basal marsupials from which they evolved. On the ground, the locomotion of koalas resembled a combination of marsupial behaviours and primate-like mechanics. For example, their fastest strides were bounding type gaits with a top speed of 2.78 m s^-1 (mean 1.20 m s^-1), resembling marsupials, while the relatively longer stride length was reflective of primate locomotion. Speed was increased using equal modification of stride length and frequency. On narrow substrates, koalas took longer but slower strides (mean 0.42 m s^-1), adopting diagonally coupled gaits including both lateral and diagonal sequence gaits, the latter being a strategy distinctive among arboreal primates. The use of diagonally coupled gaits in the arboreal environment is likely only possible because of the unique gripping hand morphology of both the fore and hind feet of koalas. These results suggest that during ground locomotion, they use marsupial-like strategies but alternate to primate-like strategies when moving amongst branches, maximising stability in these environments. The locomotion strategies of koalas provide key insights into an independent evolutionary branch for an arboreal specialist, highlighting how locomotor strategies can convergently evolve between distant lineages.

Passive acoustic surveys for predicting species' distributions: Optimising detection probability

Surveying terrestrial species across diverse habitats is important for predicting species’ distributions and implementing conservation actions. For vocalising species, passive acoustic monitoring (PAM) is increasing in popularity; however, survey design rarely considers the factors influencing the timing and occurrence of vocalisations and in turn, how they may influence detectability of the species. Here, we use the koala (Phascolarctos cinereus) as a case study to show how PAM can be used to first examine the factors influencing vocalisations, and then use occupancy modelling to make recommendations on survey design for the species. We used automated recording units to monitor koala vocalisations at ten sites between August 2016 and January 2017. The timing of male koala vocalisations was linked to time of sunset with vocalisations increasing two hours prior to sunset and peaking at four hours after sunset. Vocalisations had a seasonal trend, increasing from the early to middle stage of the breeding season. Koala population density and stage of the breeding season had more influence on detection probability than daily sampling schedule. Where population density was low, and during the early stage of the breeding season, 7 survey nights (recording for 6 hours from 20:00h to 02:00h; i.e. the period of peak bellowing activity) were required to be 95% confident of a site-specific absence. Our study provides an approach for designing effective passive acoustic surveys for terrestrial species.

Fine-scale changes in spatial habitat use by a low-density koala population in an isolated periurban forest remnant

Koala (Phascolarctos cinereus) populations in south-east Queensland are in decline. Although various studies have looked at broad-scale tree preference and habitat quality, there has been little attempt to quantify fine-scale activity shifts from one year to the next or examination of activity at the mesoscale. This study quantified koala activity levels in a 909-ha forest patch at Karawatha Forest Park, in south Brisbane. The Spot Assessment Technique was used to quantify activity and tree selection on 33 long-term monitoring plots in 2009 and 2010. In total, 843 trees were searched and koala pellets were found underneath 34 and 47 trees in 2009 and 2010, respectively. A higher proportion of pellets was found underneath Eucalyptus tindaliae and E. fibrosa and there was weak selection for larger trees. A low occurrence of revisits and a minor shift in activity distribution from 2009 to 2010 indicate that the koala population exists at a low density. Rapid declines in koala populations are occurring in primary habitats. Hence, stable low-density koala populations are important for maintaining genetic diversity and connectivity in fragmented urban landscapes.

Modeling activity patterns of wildlife using time-series analysis

The study of wildlife activity patterns is an effective approach to understanding fundamental ecological and evolutionary processes. However, traditional statistical approaches used to conduct quantitative analysis have thus far had limited success in revealing underlying mechanisms driving activity patterns. Here, we combine wavelet analysis, a type of frequency‐based time‐series analysis, with high‐resolution activity data from accelerometers embedded in GPS collars to explore the effects of internal states (e.g., pregnancy) and external factors (e.g., seasonal dynamics of resources and weather) on activity patterns of the endangered giant panda (Ailuropoda melanoleuca). Giant pandas exhibited higher frequency cycles during the winter when resources (e.g., water and forage) were relatively poor, as well as during spring, which includes the giant panda's mating season. During the summer and autumn when resources were abundant, pandas exhibited a regular activity pattern with activity peaks every 24 hr. A pregnant individual showed distinct differences in her activity pattern from other giant pandas for several months following parturition. These results indicate that animals adjust activity cycles to adapt to seasonal variation of the resources and unique physiological periods. Wavelet coherency analysis also verified the synchronization of giant panda activity level with air temperature and solar radiation at the 24‐hr band. Our study also shows that wavelet analysis is an effective tool for analyzing high‐resolution activity pattern data and its relationship to internal and external states, an approach that has the potential to inform wildlife conservation and management across species.

Do koalas really get the blues? Critique of ‘Aversive behaviour by koalas (Phascolarctos cinereus) during the course of a music festival in northern New South Wales, Australia'

It is imperative that the reported results of scientific studies are based on sound data analyses and unbiased interpretation, especially where they may be used to guide government policy and regulation. A recent paper by Phillips (2016) evaluated the behavioural response of radio-collared koalas to an inaugural large music festival held in 2010 in northern New South Wales. The study concluded that six of seven koalas showed an aversive response. However, we regard the paper as misleading because it contains serious errors in the examination of koala home ranges and in the subsequent assessment of ‘aversive behaviour’ during the music festival. We conclude that Phillips’ paper is based on sufficient data to state that three, not six, of the koalas he studied displayed a short-term behavioural response to the music festival. These koalas temporarily moved outside of their estimated ranges during the festival period (10–80 m). Further, Phillips fails to report crucial ecological data regarding the high level of disease and mortality he recorded, which are presented in their entirety in the author’s prior, unpublished report. For the two deaths that are reported, no mention is made that these koalas had pre-existing disease. Rather, the author raises the possibility that the mortalities may have been related to festival-induced stress. The omission of such key data is prejudicial to the interpretation of results. The reported short-term impact of the festival upon some of the study animals is indeed noteworthy; however, it is apparent that the impact has been considerably overstated due to errors of analysis and the omission of critical ecological data. Our critique highlights our concerns by drawing on the author’s unpublished report as well as our own research at the same site over the past five years. Our aim is to ensure that debate over the impact of music festivals on wildlife is based on sound data analyses and unbiased interpretation, to provide guidance to relevant regulators and land managers.