Remote monitoring of vigilance behavior in large herbivores using acceleration data

Kröschel, Max; Reineking, Björn; Werwie, Felicitas; Wildi, Felix; Storch, Ilse

doi:10.1186/s40317-017-0125-z

Cited by 36 publications

(45 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ANN with the moving window approach, however, was able to infer caching and walking behaviour much better than the other two. Both RF and SVM generally performed well in inferring behaviour during validation (Table 1) and showed comparable results to other studies (Nathan et al, 2012;Fehlmann et al, 2017;Kröschel et al, 2017). When applied to the wild foxes, however, they both failed to discriminate the different behaviours ( Table 2, Table S5).…”

Section: Discussionsupporting

confidence: 75%

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviour

Rast

Kimmig

Giese

et al. 2019

Preprint

View full text Add to dashboard Cite

ABSTRACTRemotely tracking distinct behaviours of animals using acceleration data and machine learning has been carried out successfully in several species in captive settings. In order to study the ecology of animals in natural habitats, such behaviour classification models need to be transferred to wild individuals. However, at present the development of those models usually requires direct observation of the target animals.The goal of this study was to infer behaviour of wild, free roaming animals from acceleration data by training behaviour classification models on captive individuals, without the necessity to observe their wild conspecifics. We further sought to develop methods to validate the credibility of the resulting behaviour extrapolations.We trained two machine learning algorithms proposed by the literature, Random Forest (RF) and Support Vector Machine (SVM), on data from captive red foxes (Vulpes vulpes) and later applied them to data from wild foxes. We also tested a new advance for behaviour classification, by applying a moving window to an Artificial Neural Network (ANN). Finally, we investigated four strategies to validate our classification output.While all three machine learning algorithms performed well under training conditions, the established methods, RF and SVM, failed in classifying distinct behaviours when transferred from captive to wild foxes. Behaviour classification with the ANN and a moving window, in contrast, inferred distinct behaviours and showed consistent results for most individuals.Our approach is a substantial improvement over the methods previously proposed in the literature as it generated plausible results for wild fox behaviour. We were able to infer the behaviour of wild animals that have never been observed in the wild and to further illustrate the outputs credibility. This framework is not restricted to foxes but can be applied to infer the behaviour of many other species and thus empowers new advances in behavioural ecology.

show abstract

Section: Discussionsupporting

confidence: 75%

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviour

Rast

Kimmig

Giese

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…The ANN with the moving window approach, however, was able to infer caching and walking behaviour much better than the other two. Both RF and SVM generally performed well in inferring behaviour during validation (Table 1) and showed comparable results to other studies [5,38,39]. When applied to the wild foxes, however, they both failed to discriminate the different behaviours ( Table 2).…”

Section: Discussionsupporting

confidence: 75%

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviours

et al. 2020

View full text Add to dashboard Cite

1. Remotely tracking distinct behaviours of animals using acceleration data and machine learning has been carried out successfully in several species in captive settings. In order to study the ecology of animals in natural habitats, such behaviour classification models need to be transferred to wild individuals. However, at present, the development of those models usually requires direct observation of the target animals. 2. The goal of this study was to infer the behaviour of wild, free-roaming animals from acceleration data by training behaviour classification models on captive individuals, without the necessity to observe their wild conspecifics. We further sought to develop methods to validate the credibility of the resulting behaviour extrapolations. 3. We trained two machine learning algorithms proposed by the literature, Random Forest (RF) and Support Vector Machine (SVM), on data from captive red foxes (Vulpes vulpes) and later applied them to data from wild foxes. We also tested a new advance for behaviour classification, by applying a moving window to an Artificial Neural Network (ANN). Finally, we investigated four strategies to validate our classification output. 4. While all three machine learning algorithms performed well under training conditions (Kappa values: RF (0.82), SVM (0.78), ANN (0.85)), the established methods, RF and SVM, failed in classifying distinct behaviours when transferred from captive to wild foxes. Behaviour classification with the ANN and a moving window, in contrast, inferred distinct behaviours and showed consistent results for most individuals. 5. Our approach is a substantial improvement over the methods previously proposed in the literature as it generated plausible results for wild fox behaviour. We were able to infer the behaviour of wild animals that have never been observed in the wild and to further illustrate the credibility of the output. This framework is not restricted to foxes but can be applied to infer the behaviour of many other species and thus empowers new advances in behavioural ecology.

show abstract

“…Though this may be challenging in closed forest environments, one could start by matching behavioral observations of ungulates mainly dwelling in open areas, for instance reindeer in mountainous areas (Mårell et al, 2002) or mountain ibex in alpine grasslands, with the study of plant dispersal. The use of acceleration sensors (Nams, 2014;Kröschel et al, 2017) and its calibration with control animals will help determine activity (active vs. resting) and specific behaviors (e.g., lying, feeding, walking, trotting) of the equipped animals together with its location in open or closed habitats. This could render more realistic the study of the transfer phase of ungulatemediated dispersal that generally combines retention times and associated distances traveled (Westcott et al, 2005;Pellerin et al, 2016).…”

Section: Research Perspectivesmentioning

confidence: 99%

Plants on the Move: Hitch-Hiking With Ungulates Distributes Diaspores Across Landscapes

Baltzinger

Karimi

Shukla³

2019

Front. Ecol. Evol.

View full text Add to dashboard Cite

We here describe the multiple mechanisms by which ungulates distribute diaspores across landscapes. There are three primary and three secondary seed dispersal mechanisms by which ungulate dispersal agents contribute to the spread of plant diaspores, both with and without the intervention of other biotic and abiotic agents. These dispersal mechanisms may be combined in successive interdependent steps. Native, introduced and domestic ungulates co-occur in many ecosystems and frequently interact with numerous plant species, which facilitates long-distance dispersal of both native and exotic plants. However, ungulate taxonomic diversity conceals a much higher diversity in terms of the functional traits involved in ungulate-mediated dispersal (e.g., feeding regime, fur morphology). These traits may strongly affect emigration, transfer and immigration in the animal-mediated plant dispersal, and consequently; they may also impact overall seed dispersal effectiveness, both quantitatively and qualitatively. In this review, we compare internal mechanisms, where seeds must survive digestive treatments (regurgitation, endozoochory), with external mechanisms, where diaspores are carried on the outside of the vectors (epizoochory). We include both primary epizoochory (direct adhesion to fur essentially) and secondary epizoochory (diaspore-laden mud adhering to hooves or the body and, transfer through contact with a conspecific). We addressed the overlap/complementarity of ungulates for the plant species they disperse through a systematic literature review. When two ungulate species co-occur, there is always an overlap in the plant species dispersed by endozoochory or by fur-epizoochory. Further, when we consider the proportion of plant species dispersed both internally and externally by an ungulate, the overlap is higher for grazing than browsing ungulates. We identify two challenges for the field of dispersal ecology: the proportion of all diaspores produced that are carried over long distances by ungulates, and the relative importance of ungulates on the whole as the main dispersal agent for plants. Furthermore, the fact that numerous plants dispersed by fur-epizoochory do not feature any specific adaptations is intriguing. We discuss unsolved methodological challenges and stress research perspectives related to ungulate-mediated dispersal: for example, taking animal behavior and cognition into account and studying how ungulates contribute to the spread of invasive exotic plants and altitudinal plant dispersal.

show abstract

Remote monitoring of vigilance behavior in large herbivores using acceleration data

Cited by 36 publications

References 34 publications

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviour

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviour

Machine learning goes wild: Using data from captive individuals to infer wildlife behaviours

Plants on the Move: Hitch-Hiking With Ungulates Distributes Diaspores Across Landscapes

Contact Info

Product

Resources

About