Search

Background: Animal-borne sensors ('bio-loggers') can record a suite of kinematic and environmental data, which are used to elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers, but there exists no common framework for comparing the different machine learning techniques in this domain. This makes it difficult to, for example, identify patterns in what works well for machine learning-based analysis of bio-logger data. It also makes it difficult to evaluate the effectiveness of novel methods developed by the machine learning community. Methods: To address this, we present the Bio-logger Ethogram Benchmark (BEBE), a collection of datasets with behavioral annotations, as well as a modeling task and evaluation metrics. BEBE is to date the largest, most taxonomically diverse, publicly available benchmark of this type, and includes 1654 h of data collected from 149 individuals across nine taxa. Using BEBE, we compare the performance of deep and classical machine learning methods for identifying animal behaviors based on bio-logger data. As an example usage of BEBE, we test an approach based on self-supervised learning. To apply this approach to animal behavior classification, we adapt a deep neural network pre-trained with 700,000 h of data collected from human wrist-worn accelerometers. Results: We find that deep neural networks out-perform the classical machine learning methods we tested across all nine datasets in BEBE. We additionally find that the approach based on self-supervised learning out-performs the alternatives we tested, especially in settings when there is a low amount of training data available. Conclusions: In light of these results, we are able to make concrete suggestions for designing studies that rely on machine learning to infer behavior from bio-logger data. Therefore, we expect that BEBE will be useful for making similar suggestions in the future, as additional hypotheses about machine learning techniques are tested. Datasets, models, and evaluation code are made publicly available at https://github.com/earthspecies/BEBE, to enable community use of BEBE. [ABSTRACT FROM AUTHOR]

Observing animals in the wild often poses extreme challenges, but animal‐borne accelerometers are increasingly revealing unobservable behaviours. Automated machine learning streamlines behaviour identification from the substantial datasets generated during multi‐animal, long‐term studies; however, the accuracy of such models depends on the qualities of the training data. We examined how data processing influenced the predictive accuracy of random forest (RF) models, leveraging the easily observed domestic cat (Felis catus) as a model organism for terrestrial mammalian behaviours. Nine indoor domestic cats were equipped with collar‐mounted tri‐axial accelerometers, and behaviours were recorded alongside video footage. From this calibrated data, eight datasets were derived with (i) additional descriptive variables, (ii) altered frequencies of acceleration data (40 Hz vs. a mean over 1 s) and (iii) standardised durations of different behaviours. These training datasets were used to generate RF models that were validated against calibrated cat behaviours before identifying the behaviours of five free‐ranging tag‐equipped cats. These predictions were compared to those identified manually to validate the accuracy of the RF models for free‐ranging animal behaviours. RF models accurately predicted the behaviours of indoor domestic cats (F‐measure up to 0.96) with discernible improvements observed with post‐data‐collection processing. Additional variables, standardised durations of behaviours and higher recording frequencies improved model accuracy. However, prediction accuracy varied with different behaviours, where high‐frequency models excelled in identifying fast‐paced behaviours (e.g. locomotion), whereas lower‐frequency models (1 Hz) more accurately identified slower, aperiodic behaviours such as grooming and feeding, particularly when examining free‐ranging cat behaviours. While RF modelling offered a robust means of behaviour identification from accelerometer data, field validations were important to validate model accuracy for free‐ranging individuals. Future studies may benefit from employing similar data processing methods that enhance RF behaviour identification accuracy, with extensive advantages for investigations into ecology, welfare and management of wild animals. [ABSTRACT FROM AUTHOR]

Machine learning‐based behaviour classification using acceleration data is a powerful tool in bio‐logging research. Deep learning architectures such as convolutional neural networks (CNN), long short‐term memory (LSTM) and self‐attention mechanism as well as related training techniques have been extensively studied in human activity recognition. However, they have rarely been used in wild animal studies. The main challenges of acceleration‐based wild animal behaviour classification include data shortages, class imbalance problems, various types of noise in data due to differences in individual behaviour and where the loggers were attached and complexity in data due to complex animal‐specific behaviours, which may have limited the application of deep learning techniques in this area.To overcome these challenges, we explored the effectiveness of techniques for efficient model training: data augmentation, manifold mixup and pre‐training of deep learning models with unlabelled data, using datasets from two species of wild seabirds and state‐of‐the‐art deep learning model architectures.Data augmentation improved the overall model performance when one of the various techniques (none, scaling, jittering, permutation, time‐warping and rotation) was randomly applied to each data during mini‐batch training. Manifold mixup also improved model performance, but not as much as random data augmentation. Pre‐training with unlabelled data did not improve model performance. The state‐of‐the‐art deep learning models, including a model consisting of four CNN layers, an LSTM layer and a multi‐head attention layer, as well as its modified version with shortcut connection, showed better performance among other comparative models. Using only raw acceleration data as inputs, these models outperformed classic machine learning approaches that used 119 handcrafted features.Our experiments showed that deep learning techniques are promising for acceleration‐based behaviour classification of wild animals and highlighted some challenges (e.g. effective use of unlabelled data). There is scope for greater exploration of deep learning techniques in wild animal studies (e.g. advanced data augmentation, multimodal sensor data use, transfer learning and self‐supervised learning). We hope that this study will stimulate the development of deep learning techniques for wild animal behaviour classification using time‐series sensor data. [ABSTRACT FROM AUTHOR]

Sophisticated animal‐borne sensor systems are increasingly providing novel insight into how animals behave and move. Despite their widespread use in ecology, the diversity and expanding quality and quantity of data they produce have created a need for robust analytical methods for biological interpretation. Machine learning tools are often used to meet this need. However, their relative effectiveness is not well known and, in the case of unsupervised tools, given that they do not use validation data, their accuracy can be difficult to assess. We evaluated the effectiveness of supervised (n = 6), semi‐supervised (n = 1), and unsupervised (n = 2) approaches to analyzing accelerometry data collected from critically endangered California condors (Gymnogyps californianus). Unsupervised K‐means and EM (expectation–maximization) clustering approaches performed poorly, with adequate classification accuracies of <0.8 but very low values for kappa statistics (range: −0.02 to 0.06). The semi‐supervised nearest mean classifier was moderately effective at classification, with an overall classification accuracy of 0.61 but effective classification only of two of the four behavioral classes. Supervised random forest (RF) and k‐nearest neighbor (kNN) machine learning models were most effective at classification across all behavior types, with overall accuracies >0.81. Kappa statistics were also highest for RF and kNN, in most cases substantially greater than for other modeling approaches. Unsupervised modeling, which is commonly used for the classification of a priori‐defined behaviors in telemetry data, can provide useful information but likely is instead better suited to post hoc definition of generalized behavioral states. This work also shows the potential for substantial variation in classification accuracy among different machine learning approaches and among different metrics of accuracy. As such, when analyzing biotelemetry data, best practices appear to call for the evaluation of several machine learning techniques and several measures of accuracy for each dataset under consideration. [ABSTRACT FROM AUTHOR]

Most studies of collective animal behaviour rely on short-term observations, and comparisons of collective behaviour across different species and contexts are rare. We therefore have a limited understanding of intra- and interspecific variation in collective behaviour over time, which is crucial if we are to understand the ecological and evolutionary processes that shape collective behaviour. Here, we study the collective motion of four species: shoals of stickleback fish (Gasterosteus aculeatus), flocks of homing pigeons (Columba livia), a herd of goats (Capra aegagrus hircus) and a troop of chacma baboons (Papio ursinus). First, we describe how local patterns (inter-neighbour distances and positions), and group patterns (group shape, speed and polarization) during collective motion differ across each system. Based on these, we place data from each species within a 'swarm space', affording comparisons and generating predictions about the collective motion across species and contexts. We encourage researchers to add their own data to update the 'swarm space' for future comparative work. Second, we investigate intraspecific variation in collective motion over time and provide guidance for researchers on when observations made over different time scales can result in confident inferences regarding species collective motion. [ABSTRACT FROM AUTHOR]

Advancements in tracking technologies provide an increasingly important tool in animal monitoring and conservation that can describe animal spatial behavior in native habitats and uncover migratory routes that otherwise may be difficult or impossible to map. In addition, high-resolution accelerometer sensors provide powerful insights into animal activity patterns and can help to identify specific behaviors from accelerometer profiles alone. Previously, such accelerometers were restricted to larger animals due to size and mass constraints. However, recent advances make it possible to use such devices on smaller animals such as the European green toad (Bufotes viridis), the focus of our current study. We deploy custom made tracking devices, that consist of very-high-frequency transmitters and tri-axial accelerometers, to track toads in their native urban environment in Vienna (Austria). A total of nine toads were tracked, ranging from three to nine tracking days per individual during the post-breeding season period. We demonstrate that our devices could reliably monitor toad movement and activity during the observation period. Hence, we confirmed the predominantly nocturnal activity patterns and recorded low overall movement at this urban site. Accelerometer data revealed that toads exhibited brief but intense activity bursts between 10 pm and midnight, resting periods during the night and intermittent activity during the day. Positional tracking alone would have missed the major activity events as they rarely resulted in large positional displacements. This underscores the importance of and value in integrating multiple tracking sensors for studies of movement ecology. Our approach could be adapted for other amphibians or other animals with mass constraints and may become standard monitoring equipment in the near future. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Animal Ecology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Methods in Ecology & Evolution is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Telemetry provides researchers with invaluable data and has contributed to the progress of animal ecology and behavioral studies. However, the impact of biotelemetry devices on animal behavior and welfare has been evaluated in a few species. The telemetry device (GPS or VHF) is attached to a harness made especially for giant anteaters (Myrmecophaga tridactyla) due to its particular anatomy, and this is the best way to access the scheduled spatiotemporal data of the species’ movement. We evaluated the adverse effects of biotelemetry devices on giant anteaters’ behavior and welfare. Behavioral analysis on giant anteaters in captivity (n = 3) was conducted by observing them with and without the GPS-harness. We also include data on GPS-harness dysfunction and animal injuries observed from free-ranging tagged animals (n=74). The GPS-harness influenced the behavior distributions frequency, but no new or atypical act was verified. Free-ranging individuals detached the harness twice, five removed the front part of the harness, and three had skin injuries. This is the first study to document telemetry device effects on giant anteaters. To date, severe adverse effects of harness-attached GPS tracking devices were not reported for this species. Our data supports the continued use of this method for monitoring free-ranging or captive giant anteaters. However, we recommend and stress the importance of continued research that helps improve telemetry and monitoring techniques. [ABSTRACT FROM AUTHOR]

Increasingly, animal behavior studies are enhanced through the use of accelerometry. To allow translation of raw accelerometer data to animal behaviors requires the development of classifiers. Here, we present the "rabc" (r for animal behavior classification) package to assist researchers with the interactive development of such animal behavior classifiers in a supervised classification approach. The package uses datasets consisting of accelerometer data with their corresponding animal behaviors (e.g., for triaxial accelerometer data along the x, y and z axes arranged as "x, y, z, x, y, z,..., behavior"). Using an example dataset collected on white stork (Ciconia ciconia), we illustrate the workflow of this package, including accelerometer data visualization, feature calculation, feature selection, feature visualization, extreme gradient boost model training, validation, and, finally, a demonstration of the behavior classification results. [ABSTRACT FROM AUTHOR]

The use of bio‐logging devices to track animal movement continues to grow as technological advances and device miniaturisation allow researchers to study animal behaviour in unprecedented detail. Balanced against the remarkable data that bio‐loggers can provide is a need to understand the impact of devices on animal behaviour and welfare.Recent meta‐analyses have demonstrated the impacts of device attachment on animal behaviour, but there is a concern about the frequency and clarity with which device effects are reported. One aspect lacking in many studies is assessment of the statistical power of tests of device effects, yet such information would assist the interpretation of results. We address this issue by providing an overview of the statistical power, as well as the Type M (magnitude) and Type S (sign) error rate, of tests of device effects within the avian tracking literature across a range of assumed effect sizes.The median power of statistical tests ranged from 9% to 65% across a range of assumed effect sizes corresponding to benchmark values for small, moderate and large effects (d = 0.2, 0.5, 0.8, respectively). Moreover, when using effect sizes derived from previous a meta‐analysis (d = 0.1), median power was only 6%. When assuming smaller effect sizes, statistical tests were characterised by high Type M and Type S error rates, suggesting that statistically significant results of device effects will tend to exaggerate the size of such effects and may estimate the sign of an effect in the wrong direction.Well‐designed tracking studies will reduce device effects to low levels and consequently issues associated with low power will be commonplace. Nevertheless, assessment of device effects remains important, particularly when embarking on novel tracking studies. We recommend that statistical tests of device effects are reported clearly and are routinely accompanied by assessment of statistical power, including Type M and Type S errors, based upon realistic external estimates of effect size. Reporting the statistical power can help avoid the pitfalls of overstating results from individual studies, shift the emphasis to accurate reporting of effect sizes and guide decisions about the ethical impacts of device attachment. [ABSTRACT FROM AUTHOR]

Numerous bibliographic reviews related to the use of AI for the behavioral detection of farm animals exist, but they only focus on a particular type of animal. We believe that some techniques were used for some animals that could also be used for other types of animals. The application and comparison of these techniques between animal species are rarely done. In this paper, we propose a review of machine learning approaches used for the detection of farm animals' behaviors such as lameness, grazing, rumination, and so on. The originality of this paper is matched classification in the midst of sensors and algorithms used for each animal category. First, we highlight the most implemented approaches for different categories of animals (cows, sheep, goats, pigs, horses, and chickens) to inspire researchers interested to conduct investigation and employ the methods we have evaluated and the results we have obtained in this study. Second, we describe the current trends in terms of technological development and new paradigms that will impact the AI research. Finally, we critically analyze what is done and we draw new pathways of research to advance our understanding of animal's behaviors. [ABSTRACT FROM AUTHOR]

Animal‐borne accelerometers have been used across more than 120 species to infer biologically significant information such as energy expenditure and broad behavioural categories. While the accelerometer's high sensitivity to movement and fast response times present the unprecedented opportunity to resolve fine‐scale behaviour, leveraging this opportunity will require overcoming the challenge of developing general, automated methods to analyse the nonstationary signals generated by nonlinear processes governing erratic, impulsive movement characteristic of fine‐scale behaviour.We address this issue by conceptualising fine‐scale behaviour in terms of characteristic microevents: impulsive movements producing brief (<1 s) shock signals in accelerometer data. We propose a 'seek‐and‐learn' approach: a novel microevent detection step first locates where shock signals occur ('seek') by searching for peaks in envelopes of acceleration data. Robust machine learning ('learn') employing meaningful features then separates microevents. We showcase the application of our method on tri‐axial accelerometer data collected on 10 free‐living meerkats Suricata suricatta for four fine‐scale foraging behaviours – searching for digging sites, one‐armed digging, two‐armed digging and head jerks during prey ingestion. Annotated videos served as groundtruth, and performance was benchmarked against that of a variety of classical machine learning approaches.Microevent identification (μEvId) with eight features in a three‐node hierarchical classification scheme employing logistic regression at each node achieved a mean overall accuracy of >85% during leave‐one‐individual‐out cross‐validation, and exceeded that of the best classical machine learning approach by 8.6%. μEvId was found to be robust not only to inter‐individual variation but also to large changes in model parameters.Our results show that microevents can be modelled as impulse responses of the animal body‐and‐sensor system. The microevent detection step retains only informative regions of the signal, which results in the selection of discriminative features that reflect biomechanical differences between microevents. Moving‐window‐based classical machine learning approaches lack this prefiltering step, and were found to be suboptimal for capturing the nonstationary dynamics of the recorded signals. The general, automated technique of μEvId, together with existing models that can identify broad behavioural categories, provides future studies with a powerful toolkit to exploit the full potential of accelerometers for animal behaviour recognition. [ABSTRACT FROM AUTHOR]

Whether changes in animal behavior allow for short‐term earthquake predictions has been debated for a long time. Before, during and after the 2016/2017 earthquake sequence in Italy, we deployed bio‐logging tags to continuously observe the activity of farm animals (cows, dogs, and sheep) close to the epicenter of the devastating magnitude M6.6 Norcia earthquake (Oct–Nov 2016) and over a subsequent longer observation period (Jan–Apr 2017). Relating 5,304 (in 2016) and 12,948 (in 2017) earthquakes with a wide magnitude range (0.4 ≤ M ≤ 6.6) to continuously measured animal activity, we detected how the animals collectively reacted to earthquakes. We also found consistent anticipatory activity prior to earthquakes during times when the animals were in a building (stable), but not during their time on a pasture. We detected these anticipatory patterns not only in periods with high, but also in periods of low seismic activity. Earthquake anticipation times (1–20 hr) are negatively correlated with the distance between the farm and earthquake hypocenters. Our study suggests that continuous bio‐logging of animal collectives has the potential to provide statistically reliable patterns of pre‐seismic activity that could yield valuable insights for short‐term earthquake forecasting. Based on a priori model parameters, we provide empirical threshold values for pre‐seismic animal activities to be used in real‐time observation stations. [ABSTRACT FROM AUTHOR]

The use of animal‐attached data loggers to quantify animal movement has increased in popularity and application in recent years. High‐resolution tri‐axial acceleration and magnetometry measurements have been fundamental in elucidating fine‐scale animal movements, providing information on posture, traveling speed, energy expenditure, and associated behavioral patterns. Heading is a key variable obtained from the tandem use of magnetometers and accelerometers, although few field investigations have explored fine‐scale changes in heading to elucidate differences in animal activity (beyond the notable exceptions of dead‐reckoning).This paper provides an overview of the value and use of animal heading and a prime derivative, angular velocity about the yaw axis, as an important element for assessing activity extent with potential to allude to behaviors, using "free‐ranging" Loggerhead turtles (Caretta caretta) as a model species.We also demonstrate the value of yaw rotation for assessing activity extent, which varies over the time scales considered and show that various scales of body rotation, particularly rate of change of yaw, can help resolve differences between fine‐scale behavior‐specific movements. For example, oscillating yaw movements about a central point of the body's arc implies bouts of foraging, while unusual circling behavior, indicative of conspecific interactions, could be identified from complete revolutions of the longitudinal axis.We believe this approach should help identification of behaviors and "space‐state" approaches to enhance our interpretation of behavior‐based movements, particularly in scenarios where acceleration metrics have limited value, such as for slow‐moving animals. [ABSTRACT FROM AUTHOR]

Animal behavior is elicited, in part, in response to external conditions, but understanding how animals perceive the environment and make the decisions that bring about these behavioral responses is challenging. Animal heads often move during specific behaviors and, additionally, typically have sensory systems (notably vision, smell, and hearing) sampling in defined arcs (normally to the front of their heads). As such, head‐mounted electronic sensors consisting of accelerometers and magnetometers, which can be used to determine the movement and directionality of animal heads (where head "movement" is defined here as changes in heading [azimuth] and/or pitch [elevation angle]), can potentially provide information both on behaviors in general and also clarify which parts of the environment the animals might be prioritizing ("environmental framing"). We propose a new approach to visualize the data of such head‐mounted tags that combines the instantaneous outputs of head heading and pitch in a single intuitive spherical plot. This sphere has magnetic heading denoted by "longitude" position and head pitch by "latitude" on this "orientation sphere" (O‐sphere). We construct the O‐sphere for the head rotations of a number of vertebrates with contrasting body shape and ecology (oryx, sheep, tortoises, and turtles), illustrating various behaviors, including foraging, walking, and environmental scanning. We also propose correcting head orientations for body orientations to highlight specific heading‐independent head rotation, and propose the derivation of O‐sphere‐metrics, such as angular speed across the sphere. This should help identify the functions of various head behaviors. Visualizations of the O‐sphere provide an intuitive representation of animal behavior manifest via head orientation and rotation. This has ramifications for quantifying and understanding behaviors ranging from navigation through vigilance to feeding and, when used in tandem with body movement, should provide an important link between perception of the environment and response to it in free‐ranging animals. [ABSTRACT FROM AUTHOR]

GPS‐tracking devices have been used in combination with a wide range of additional sensors to study animal behaviour, physiology and interaction with their environment. Tri‐axial accelerometers allow researchers to remotely infer the behaviour of individuals, at all places and times. Collection of accelerometer data is relatively cheap in terms of energy usage, but the amount of raw data collected generally requires much storage space and is particularly demanding in terms of energy needed for data transmission.Here, we propose compressing the raw accelerometer (ACC) data into summary statistics within the tracking device (before transmission) to reduce data size, as a means to overcome limitations in storage and energy capacity.We explored this type of lossy data compression in the accelerometer data of tagged Bewick's swans Cygnus columbianus bewickii collected in spring 2017. Using software settings in which bouts of 2 s of both raw ACC data and summary statistics were collected in parallel but with different bout intervals to keep total data size comparable, we created the opportunity for a direct comparison of time budgets derived by the two data collection methods.We found that the data compression in our case yielded a six times reduction in data size per bout, and concurrent, similar decreases in storage and energy use of the device. We show that with the same accuracy of the behavioural classification, the freed memory and energy of the device can be used to increase the monitoring effort, resulting in a more detailed representation of the individuals' time budget. Rare and/or short behaviours, such as daily roost flights, were picked up significantly more when collecting summary statistics instead of raw ACC data (but note differences in sampling rate). Such level of detail can be of essential importance, for instance to make a reliable estimate of the energy budgets of individuals.In conclusion, we argue that this type of lossy data compression can be a well‐considered choice in study situations where limitations in energy and storage space of the device pose a problem. Ultimately, these developments can allow for long‐term and nearly continuous remote monitoring of the behaviour of free‐ranging animals. [ABSTRACT FROM AUTHOR]

The Mongolian Gobi-Eastern Steppe Ecosystem is one of the largest remaining natural drylands and home to a unique assemblage of migratory ungulates. Connectivity and integrity of this ecosystem are at risk if increasing human activities are not carefully planned and regulated. The Gobi part supports the largest remaining population of the Asiatic wild ass (Equus hemionus; locally called “khulan”). Individual khulan roam over areas of thousands of square kilometers and the scale of their movements is among the largest described for terrestrial mammals, making them particularly difficult to monitor. Although GPS satellite telemetry makes it possible to track animals in near-real time and remote sensing provides environmental data at the landscape scale, remotely collected data also harbors the risk of missing important abiotic or biotic environmental variables or life history events. We tested the potential of animal born camera systems (“camera collars”) to improve our understanding of the drivers and limitations of khulan movements. Deployment of a camera collar on an adult khulan mare resulted in 7,881 images over a one-year period. Over half of the images showed other khulan and 1,630 images showed enough of the collared khulan to classify the behaviour of the animals seen into several main categories. These khulan images provided us with: i) new insights into important life history events and grouping dynamics, ii) allowed us to calculate time budgets for many more animals than the collared khulan alone, and iii) provided us with a training dataset for calibrating data from accelerometer and tilt sensors in the collar. The images also allowed to document khulan behaviour near infrastructure and to obtain a day-time encounter rate between a specific khulan with semi-nomadic herders and their livestock. Lastly, the images allowed us to ground truth the availability of water by: i) confirming waterpoints predicted from other analyses, ii) detecting new waterpoints, and iii) compare precipitation records for rain and snow from landscape scale climate products with those documented by the camera collar. We discuss the added value of deploying camera collars on a subset of animals in remote, highly variable ecosystems for research and conservation. [ABSTRACT FROM AUTHOR]

Observations of animal thermoregulatory behavior are labor‐intensive, and human presence may disturb the normal behavior of the animal. Therefore, we investigated whether a remote biologging technique could be used to detect orientation to solar radiation in savanna antelope. We predicted that when a mammal was orientated perpendicular to solar radiation, the subcutaneous temperature on the side of the body facing the sun would be greater than that on the opposite side, whereas when the mammal was orientated parallel to solar radiation, subcutaneous temperatures on both sides would be similar. A pilot study showed that the difference between left‐ and right‐side temperatures under a pelt reflected orientation to solar radiation if a pelt‐covered cylinder had been orientated for 15 min or longer. In addition, the rate of change in temperature difference could detect orientation that had changed within the previous 5 min. We implanted temperature‐sensitive data loggers subcutaneously into the flanks of eight black (Connochaetes gnu) and eight blue (Connochaetes taurinus) wildebeest. By incorporating both the rate of change and subcutaneous temperature differences and excluding times when wildebeest were lying down, our predictions correctly matched behavioral observations of wildebeest orientation to solar radiation 71% of the time. Our technique tended to fail when wildebeest were lying down, wind speeds were high and the sun was overhead. But those are conditions in which the benefits of manipulating orientation to solar radiation is of diminishing importance to a free‐living mammal. Therefore, subcutaneous temperatures provide physiologically relevant information on the importance of solar radiation to mammals. HIGHLIGHTS: Subcutaneous temperature measured remotely by biologgers could be used to identify orientation to solar radiation of free‐living wildebeest correctly 71% of the time and to quantify heat gain during orientation behaviors. [ABSTRACT FROM AUTHOR]

Bio‐loggers are miniaturized autonomous devices that record quantitative data on the state of free‐ranging animals (e.g. behavior, position and physiology) and their natural environment. This is especially relevant for species where direct visual observation is difficult or impossible. Today, ongoing technical development allows the monitoring of numerous parameters in an increasing range of species over extended periods. However, the external attachment of devices might affect various aspects of animal performance (energetics, thermoregulation, foraging as well as social and reproductive behavior), which ultimately affect fitness. External attachment might also increase entanglement risk and the conspicuousness of animals, leaving them more vulnerable to predation. By contrast, implantation of devices can mitigate many of these undesirable effects and might be preferable, especially for long‐term studies, provided that the many challenges associated with surgical procedures can be mastered. Implantation may then allow us to gather data that would be impossible to obtain otherwise and thereby may provide new and ecologically relevant insights into the life of wild animals. Here, we: (i) discuss the pros and cons of attachment methods; (ii) highlight recent field studies that used implanted bio‐loggers to address eco‐physiological questions in a wide range of species; and (iii) discuss logger implantation in light of ethical considerations. [ABSTRACT FROM AUTHOR]

Two prime issues can detrimentally affect animals that have been equipped with tags: (i) the effect of the capture and restraint process; and (ii) the effect of the tag itself. This work examines some of the issues surrounding quantification of tag effects on wild animals for both restrained and free‐living animals. A new method to quantify stress effects based on monitoring ventilation rates in relation to activity is suggested for restrained animals which may help improve the practice of handling animals. It is also suggested that various metrics, many derived from accelerometers, can be examined in tagged wild animals to examine the change in behaviors over time with a view to having a better understanding of welfare issues, assuring the quality of recorded data and informing best practice. [ABSTRACT FROM AUTHOR]

Decomposing the life track of an animal into behavioral segments is a fundamental challenge for movement ecology. The proliferation of high‐resolution data, often collected many times per second, offers much opportunity for understanding animal movement. However, the sheer size of modern data sets means there is an increasing need for rapid, novel computational techniques to make sense of these data. Most existing methods were designed with smaller data sets in mind and can thus be prohibitively slow. Here, we introduce a method for segmenting high‐resolution movement trajectories into sites of interest and transitions between these sites. This builds on a previous algorithm of Benhamou and Riotte‐Lambert (2012). Adapting it for use with high‐resolution data. The data's resolution removed the need to interpolate between successive locations, allowing us to increase the algorithm's speed by approximately two orders of magnitude with essentially no drop in accuracy. Furthermore, we incorporate a color scheme for testing the level of confidence in the algorithm's inference (high = green, medium = amber, low = red). We demonstrate the speed and accuracy of our algorithm with application to both simulated and real data (Alpine cattle at 1 Hz resolution). On simulated data, our algorithm correctly identified the sites of interest for 99% of "high confidence" paths. For the cattle data, the algorithm identified the two known sites of interest: a watering hole and a milking station. It also identified several other sites which can be related to hypothesized environmental drivers (e.g., food). Our algorithm gives an efficient method for turning a long, high‐resolution movement path into a schematic representation of broadscale decisions, allowing a direct link to existing point‐to‐point analysis techniques such as optimal foraging theory. It is encoded into an R package called SitesInterest, so should serve as a valuable tool for making sense of these increasingly large data streams. We have constructed a new method for identifying sites of interest, tailored specifically for use with high resolution data. This allows for a complicated movement trajectory to be simplified into a "Markov chain‐like" description of behavioural choices, highlighting the animal's broadscale decisions. [ABSTRACT FROM AUTHOR]

The development of multisensor animal‐attached tags, recording data at high frequencies, has enormous potential in allowing us to define animal behaviour.The high volumes of data, are pushing us towards machine‐learning as a powerful option for distilling out behaviours. However, with increasing parallel lines of data, systems become more likely to become processor limited and thereby take appreciable amounts of time to resolve behaviours.We suggest a Boolean approach whereby critical changes in recorded parameters are used as sequential templates with defined flexibility (in both time and degree) to determine individual behavioural elements within a behavioural sequence that, together, makes up a single, defined behaviour.We tested this approach, and compared it to a suite of other behavioural identification methods, on a number of behaviours from tag‐equipped animals; sheep grazing, penguins walking, cheetah stalking prey and condors thermalling.Overall behaviour recognition using our new approach was better than most other methods due to; (1) its ability to deal with behavioural variation and (2) the speed with which the task was completed because extraneous data are avoided in the process.We suggest that this approach is a promising way forward in an increasingly data‐rich environment and that workers sharing algorithms can provide a powerful library for the benefit of all involved in such work. [ABSTRACT FROM AUTHOR]

Simple Summary: This study analyzed the possibility of automatically detecting dairy cow behavior by combining the use of a single triaxial accelerometer applied to the animal's left flank with a machine learning technique. This combination enabled the detection of posture and the main types of behavior that are extremely useful in evaluating the animal's welfare and health such as resting, feeding, and rumination with a high degree of accuracy. The novelty of the study was the success in reaching a high accuracy in detecting five different behaviors and the animal posture by using a single sensor and allowing farmers to save money. To the best of our knowledge, this is the first study that has successfully explored the feasibility of locating a sensor on the animal's left flank, showing the opportunity of automatically measuring some physiological parameters, such as those ones related to respiration and rumen health, in a non-invasive way. The aim of the present study was to develop a model to identify posture and behavior from data collected by a triaxial accelerometer located on the left flank of dairy cows and evaluate its accuracy and precision. Twelve Italian Red-and-White lactating cows were equipped with an accelerometer and observed on average for 136 ± 29 min per cow by two trained operators as a reference. The acceleration data were grouped in time windows of 8 s overlapping by 33.0%, for a total of 35,133 rows. For each row, 32 different features were extracted and used by machine learning algorithms for the classification of posture and behavior. To build up a predictive model, the dataset was split in training and testing datasets, characterized by 75.0 and 25.0% of the observations, respectively. Four algorithms were tested: Random Forest, K Nearest Neighbors, Extreme Boosting Algorithm (XGB), and Support Vector Machine. The XGB model showed the best accuracy (0.99) and Cohen's kappa (0.99) in predicting posture, whereas the Random Forest model had the highest overall accuracy in predicting behaviors (0.76), showing a balanced accuracy from 0.96 for resting to 0.77 for moving. Overall, very accurate detection of the posture and resting behavior were achieved. [ABSTRACT FROM AUTHOR]