Directional swimming patterns in jellyfish aggregations

Having a profound influence on marine and coastal environments worldwide, jellyfish hold significant scientific, economic, and public interest.1,2,3,4,5 The predictability of outbreaks and dispersion of jellyfish is limited by a fundamental gap in our understanding of their movement. Although there is evidence that jellyfish may actively affect their position,6,7,8,9,10 the role of active swimming in controlling jellyfish movement, and the characteristics of jellyfish swimming behavior, are not well understood. Consequently, jellyfish are often regarded as passively drifting or randomly moving organisms, both conceptually2,11 and in process studies.12,13,14 Here we show that the movement of jellyfish is modulated by distinctly directional swimming patterns that are oriented away from the coast and against the direction of surface gravity waves. Taking a Lagrangian viewpoint from drone videos that allows the tracking of multiple adjacent jellyfish, and focusing on the scyphozoan jellyfish Rhopilema nomadica as a model organism, we show that the behavior of individual jellyfish translates into a synchronized directional swimming of the aggregation as a whole. Numerical simulations show that this counter-wave swimming behavior results in biased correlated random-walk movement patterns that reduce the risk of stranding, thus providing jellyfish with an adaptive advantage critical to their survival. Our results emphasize the importance of active swimming in regulating jellyfish movement and open the way for a more accurate representation in model studies, thus improving the predictability of jellyfish outbreaks and their dispersion and contributing to our ability to mitigate their possible impact on coastal infrastructure and populations.

Unveiling the roles of temporal periodicity, the spatial environment and behavioural modes in terrestrial animal movement

BACKGROUND

Animal movement arises from complex interactions between animals and their heterogeneous environment. To better understand the movement process, it can be divided into behavioural, temporal and spatial components. Although methods exist to address those various components, it remains challenging to integrate them in a single movement analysis.

METHODS

We present an analytic workflow that integrates the behavioural, temporal and spatial components of the movement process and their interactions, which also allows for the assessment of the relative importance of those components. We construct a daily cyclic covariate to represent temporally cyclic movement patterns, such as diel variation in activity, and combine the three components in a multi-modal Hidden Markov Model framework using existing methods and R functions. We compare the trends and statistical fits of models that include or exclude any of the behavioural, spatial and temporal components, and perform variance partitioning on the model predictions that included all components to assess their relative importance to the movement process, both in isolation and in interaction.

RESULTS

We apply our workflow to a case study on the movements of plains zebra, blue wildebeest and eland antelope in a South African reserve. Behavioural modes impacted movement the most, followed by diel rhythms and then the spatial environment (viz. tree cover and terrain slope). Interactions between the components often explained more of the movement variation than the marginal effect of the spatial environment did on its own. Omitting components from the analysis led either to the inability to detect relationships between input and response variables, resulting in overgeneralisations when drawing conclusions about the movement process, or to detections of questionable relationships that appeared to be spurious.

CONCLUSIONS

Our analytic workflow can be used to integrate the behavioural, temporal and spatial components of the movement process and quantify their relative contributions, thereby preventing incomplete or overly generic ecological interpretations. We demonstrate that understanding the drivers of animal movement, and ultimately the ecological phenomena that emerge from it, critically depends on considering the various components of the movement process, and especially the interactions between them.

A multistate Langevin diffusion for inferring behavior-specific habitat selection and utilization distributions

The identification of important habitat and the behavior(s) associated with it is critical to conservation and place-based management decisions. Behavior also links life-history requirements and habitat use, which are key to understanding why animals use certain habitats. Animal population studies often use tracking data to quantify space use and habitat selection, but they typically either ignore movement behavior (e.g., foraging, migrating, nesting) or adopt a two-stage approach that can induce bias and fail to propagate uncertainty. We develop a habitat-driven Langevin diffusion for animals that exhibit distinct movement behavior states, thereby providing a novel single-stage statistical method for inferring behavior-specific habitat selection and utilization distributions in continuous time. Practitioners can customize, fit, assess, and simulate our integrated model using the provided R package. Simulation experiments demonstrated that the model worked well under a range of sampling scenarios as long as observations were of sufficient temporal resolution. Our simulations also demonstrated the importance of accounting for different behaviors and the misleading inferences that can result when these are ignored. We provide case studies using plains zebra (Equus quagga) and Steller sea lion (Eumetopias jubatus) telemetry data. In the zebra example, our model identified distinct "encamped" and "exploratory" states, where the encamped state was characterized by strong selection for grassland and avoidance of other vegetation types, which may represent selection for foraging resources. In the sea lion example, our model identified distinct movement behavior modes typically associated with this marine central-place forager and, unlike previous analyses, found foraging-type movements to be associated with steeper offshore slopes characteristic of the continental shelf, submarine canyons, and seamounts that are believed to enhance prey concentrations. This is the first single-stage approach for inferring behavior-specific habitat selection and utilization distributions from tracking data that can be readily implemented with user-friendly software. As certain behaviors are often more relevant to specific conservation or management objectives, practitioners can use our model to help inform the identification and prioritization of important habitats. Moreover, by linking individual-level movement behaviors to population-level spatial processes, the multistate Langevin diffusion can advance inferences at the intersection of population, movement, and landscape ecology.

Examination of head versus body heading may help clarify the extent to which animal movement pathways are structured by environmental cues?

Understanding the processes that determine how animals allocate time to space is a major challenge, although it is acknowledged that summed animal movement pathways over time must define space-time use. The critical question is then, what processes structure these pathways? Following the idea that turns within pathways might be based on environmentally determined decisions, we equipped Arabian oryx with head- and body-mounted tags to determine how they orientated their heads - which we posit is indicative of them assessing the environment - in relation to their movement paths, to investigate the role of environment scanning in path tortuosity. After simulating predators to verify that oryx look directly at objects of interest, we recorded that, during routine movement, > 60% of all turns in the animals' paths, before being executed, were preceded by a change in head heading that was not immediately mirrored by the body heading: The path turn angle (as indicated by the body heading) correlated with a prior change in head heading (with head heading being mirrored by subsequent turns in the path) twenty-one times more than when path turns occurred due to the animals adopting a body heading that went in the opposite direction to the change in head heading. Although we could not determine what the objects of interest were, and therefore the proposed reasons for turning, we suggest that this reflects the use of cephalic senses to detect advantageous environmental features (e.g. food) or to detect detrimental features (e.g. predators). The results of our pilot study suggest how turns might emerge in animal pathways and we propose that examination of points of inflection in highly resolved animal paths could represent decisions in landscapes and their examination could enhance our understanding of how animal pathways are structured.

A three-step approach for assessing landscape connectivity via simulated dispersal: African wild dog case study

CONTEXT

Dispersal of individuals contributes to long-term population persistence, yet requires a sufficient degree of landscape connectivity. To date, connectivity has mainly been investigated using least-cost analysis and circuit theory, two methods that make assumptions that are hardly applicable to dispersal. While these assumptions can be relaxed by explicitly simulating dispersal trajectories across the landscape, a unified approach for such simulations is lacking.

OBJECTIVES

Here, we propose and apply a simple three-step approach to simulate dispersal and to assess connectivity using empirical GPS movement data and a set of habitat covariates.

METHODS

In step one of the proposed approach, we use integrated step-selection functions to fit a mechanistic movement model describing habitat and movement preferences of dispersing individuals. In step two, we apply the parameterized model to simulate dispersal across the study area. In step three, we derive three complementary connectivity maps; a heatmap highlighting frequently traversed areas, a betweenness map pinpointing dispersal corridors, and a map of inter-patch connectivity indicating the presence and intensity of functional links between habitat patches. We demonstrate the applicability of the proposed three-step approach in a case study in which we use GPS data collected on dispersing African wild dogs (Lycaon pictus) inhabiting northern Botswana.

RESULTS

Using step-selection functions we successfully parametrized a detailed dispersal model that described dispersing individuals' habitat and movement preferences, as well as potential interactions among the two. The model substantially outperformed a model that omitted such interactions and enabled us to simulate 80,000 dispersal trajectories across the study area.

CONCLUSION

By explicitly simulating dispersal trajectories, our approach not only requires fewer unrealistic assumptions about dispersal, but also permits the calculation of multiple connectivity metrics that together provide a comprehensive view of landscape connectivity. In our case study, the three derived connectivity maps revealed several wild dog dispersal hotspots and corridors across the extent of our study area. Each map highlighted a different aspect of landscape connectivity, thus emphasizing their complementary nature. Overall, our case study demonstrates that a simulation-based approach offers a simple yet powerful alternative to traditional connectivity modeling techniques. It is therefore useful for a variety of applications in ecological, evolutionary, and conservation research.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s10980-023-01602-4.

Defining null expectations for animal site fidelity

Site fidelity-the tendency to return to previously visited locations-is widespread across taxa. Returns may be driven by several mechanisms, including memory, habitat selection, or chance; however, pattern-based definitions group different generating mechanisms under the same label of 'site fidelity', often assuming memory as the main driver. We propose an operational definition of site fidelity as patterns of return that deviate from a null expectation derived from a memory-free movement model. First, using agent-based simulations, we show that without memory, intrinsic movement characteristics and extrinsic landscape characteristics are key determinants of return patterns and that even random movements may generate substantial probabilities of return. Second, we illustrate how to implement our framework empirically to establish ecologically meaningful, system-specific null expectations for site fidelity. Our approach provides a conceptual and operational framework to test hypotheses on site fidelity across systems and scales.

Dynamic winter weather moderates movement and resource selection of wild turkeys at high-latitude range limits

For wide-ranging species in temperate environments, populations at high-latitude range limits are subject to more extreme conditions, colder temperatures, and greater snow accumulation compared with their core range. As climate change progresses, these bounding pressures may become more moderate on average, while extreme weather occurs more frequently. Individuals can mitigate temporarily extreme conditions by changing daily activity budgets and exhibiting plasticity in resource selection, both of which facilitate existence at and expansion of high-latitude range boundaries. However, relatively little work has explored how animals moderate movement and vary resource selection with changing weather, and a general framework for such investigations is lacking. We applied hidden Markov models and step selection functions to GPS data from wintering wild turkeys (Meleagris gallopavo) near their northern range limit to identify how weather influenced transition among discrete movement states, as well as state-specific resource selection. We found that turkeys were more likely to spend time in a stationary state as wind chill temperatures decreased and snow depth increased. Both stationary and roosting turkeys selected conifer forests and avoided land covers associated with foraging, such as agriculture and residential areas, while shifting their strength of selection for these features during poor weather. In contrast, mobile turkeys showed relatively weak resource selection, with less response in selection coefficients during poor weather. Our findings illustrate that behavioral plasticity in response to weather was context dependent, but movement behaviors most associated with poor weather were also those in which resource selection was most plastic. Given our results, the potential for wild turkey range expansion will partly be determined by the availability of habitat that allows them to withstand periodic inclement weather. Combining hidden Markov models with step selection functions is broadly applicable for evaluating plasticity in animal behavior and dynamic resource selection in response to changing weather. We studied turkeys at northern range limits, but this approach is applicable for any system expected to experience significant changes in the coming decade, and may be particularly relevant to populations existing at range peripheries.

Inferring spatially varying animal movement characteristics using a hierarchical continuous-time velocity model

Understanding the spatial dynamics of animal movement is an essential component of maintaining ecological connectivity, conserving key habitats, and mitigating the impacts of anthropogenic disturbance. Altered movement and migratory patterns are often an early warning sign of the effects of environmental disturbance, and a precursor to population declines. Here, we present a hierarchical Bayesian framework based on Gaussian processes for analysing the spatial characteristics of animal movement. At the heart of our approach is a novel covariance kernel that links the spatially varying parameters of a continuous-time velocity model with GPS locations from multiple individuals. We demonstrate the effectiveness of our framework by first applying it to a synthetic data set and then by analysing telemetry data from the Serengeti wildebeest migration. Through application of our approach, we are able to identify the key pathways of the wildebeest migration as well as revealing the impacts of environmental features on movement behaviour.

An integrated path for spatial capture-recapture and animal movement modeling

Ecologists and conservation biologists increasingly rely on spatial capture-recapture (SCR) and movement modeling to study animal populations. Historically, SCR has focused on population-level processes (e.g., vital rates, abundance, density, and distribution), whereas animal movement modeling has focused on the behavior of individuals (e.g., activity budgets, resource selection, migration). Even though animal movement is clearly a driver of population-level patterns and dynamics, technical and conceptual developments to date have not forged a firm link between the two fields. Instead, movement modeling has typically focused on the individual level without providing a coherent scaling from individual- to population-level processes, whereas SCR has typically focused on the population level while greatly simplifying the movement processes that give rise to the observations underlying these models. In our view, the integration of SCR and animal movement modeling has tremendous potential for allowing ecologists to scale up from individuals to populations and advancing the types of inferences that can be made at the intersection of population, movement, and landscape ecology. Properly accounting for complex animal movement processes can also potentially reduce bias in estimators of population-level parameters, thereby improving inferences that are critical for species conservation and management. This introductory article to the Special Feature reviews recent advances in SCR and animal movement modeling, establishes a common notation, highlights potential advantages of linking individual-level (Lagrangian) movements to population-level (Eulerian) processes, and outlines a general conceptual framework for the integration of movement and SCR models. We then identify important avenues for future research, including key challenges and potential pitfalls in the developments and applications that lie ahead.

Modeling abundance, distribution, movement and space use with camera and telemetry data

Studies of animal abundance and distribution are often conducted independently of research on movement, despite the important links between processes. Movement can cause rapid changes in spatial variation in density, and movement influences detection probability and therefore estimates of abundance from inferential methods such as spatial capture-recapture (SCR). Technological developments including camera traps and GPS telemetry have opened new opportunities for studying animal demography and movement, yet statistical models for these two data types have largely developed along parallel tracks. We present a hierarchical model in which both datasets are conditioned on a movement process for a clearly defined population. We fitted the model to data from 60 camera traps and 23,572 GPS telemetry locations collected on 17 male white-tailed deer in the Big Cypress National Preserve, Florida, USA during July 2015. Telemetry data were collected on a 3-4 h acquisition schedule, and we modeled the movement paths of all individuals in the region with a Ornstein-Uhlenbeck process that included individual-specific random effects. Two of the 17 deer with GPS collars were detected on cameras. An additional 20 male deer without collars were detected on cameras and individually identified based on their unique antler characteristics. Abundance was 126 (95% CI: 88-177) in the 228 km² region, only slightly higher than estimated using a standard SCR model: 119 (84-168). The standard SCR model, however, was unable to describe individual heterogeneity in movement rates and space use as revealed by the joint model. Joint modeling allowed the telemetry data to inform the movement model and the SCR encounter model, while leveraging information in the camera data to inform abundance, distribution and movement. Unlike most existing methods for population-level inference on movement, the joint SCR-movement model can yield unbiased inferences even if non-uniform sampling is used to deploy transmitters. Potential extensions of the model include the addition of resource selection parameters, and relaxation of the closure assumption when interest lies in survival and recruitment. These developments would contribute to the emerging holistic framework for the study of animal ecology, one that uses modern technology and spatio-temporal statistics to learn about interactions between behavior and demography.

Multi-mode movement decisions across widely ranging behavioral processes

Movement of organisms plays a fundamental role in the evolution and diversity of life. Animals typically move at an irregular pace over time and space, alternating among movement states. Understanding movement decisions and developing mechanistic models of animal distribution dynamics can thus be contingent to adequate discrimination of behavioral phases. Existing methods to disentangle movement states typically require a follow-up analysis to identify state-dependent drivers of animal movement, which overlooks statistical uncertainty that comes with the state delineation process. Here, we developed population-level, multi-state step selection functions (HMM-SSF) that can identify simultaneously the different behavioral bouts and the specific underlying behavior-habitat relationship. Using simulated data and relocation data from mule deer (Odocoileus hemionus), plains bison (Bison bison bison) and plains zebra (Equus quagga), we illustrated the HMM-SSF robustness, versatility, and predictive ability for animals involved in distinct behavioral processes: foraging, migrating and avoiding a nearby predator. Individuals displayed different habitat selection pattern during the encamped and the travelling phase. Some landscape attributes switched from being selected to avoided, depending on the movement phase. We further showed that HMM-SSF can detect multi-modes of movement triggered by predators, with prey switching to the travelling phase when predators are in close vicinity. HMM-SSFs thus can be used to gain a mechanistic understanding of how animals use their environment in relation to the complex interplay between their needs to move, their knowledge of the environment and navigation capacity, their motion capacity and the external factors related to landscape heterogeneity.

Circular-linear copulae for animal movement data

Animal movement is often modelled in discrete time, formulated in terms of steps taken between successive locations at regular time intervals. Steps are characterized by the distance between successive locations (step lengths) and changes in direction (turn angles). Animals commonly exhibit a mix of directed movements with large step lengths and turn angles near 0 when travelling between habitat patches and more wandering movements with small step lengths and uniform turn angles when foraging. Thus, step lengths and turn angles will typically be cross-correlated.Most models of animal movement assume that step lengths and turn angles are independent, likely due to a lack of available alternatives. Here, we show how the method of copulae can be used to fit multivariate distributions that allow for correlated step lengths and turn angles.We describe several newly developed copulae appropriate for modelling animal movement data and fit these distributions to data collected on fishers (Pekania pennanti). The copulae are able to capture the inherent correlation in the data and provide a better fit than a model that assumes independence. Further, we demonstrate via simulation that this correlation can impact movement patterns (e.g. rates of dispersion overtime).We see many opportunities to extend this framework (e.g. to consider autocorrelation in step attributes) and to integrate it into existing frameworks for modelling animal movement and habitat selection. For example, copulae could be used to more accurately sample available locations when conducting habitat-selection analyses.

Spatial Memory Drives Foraging Strategies of Wolves, but in Highly Individual Ways

The ability of wild animals to navigate and survive in complex and dynamic environments depends on their ability to store relevant information and place it in a spatial context. Despite the centrality of spatial memory, and given our increasing ability to observe animal movements in the wild, it is perhaps surprising how difficult it is to demonstrate spatial memory empirically. We present a cognitive analysis of movements of several wolves (Canis lupus) in Finland during a summer period of intensive hunting and den-centered pup-rearing. We tracked several wolves in the field by visiting nearly all GPS locations outside the den, allowing us to identify the species, location and timing of nearly all prey killed. We then developed a model that assigns a spatially explicit value based on memory of predation success and territorial marking. The framework allows for estimation of multiple cognitive parameters, including temporal and spatial scales of memory. For most wolves, fitted memory-based models outperformed null models by 20 to 50% at predicting locations where wolves chose to forage. However, there was a high amount of individual variability among wolves in strength and even direction of responses to experiences. Some wolves tended to return to locations with recent predation success—following a strategy of foraging site fidelity—while others appeared to prefer a site switching strategy. These differences are possibly explained by variability in pack sizes, numbers of pups, and features of the territories. Our analysis points toward concrete strategies for incorporating spatial memory in the study of animal movements while providing nuanced insights into the behavioral strategies of individual predators.

Movement of Rehabilitated African Elephant Calves Following Soft Release Into a Wildlife Sanctuary

The ability to locate essential resources is a critical step for wildlife translocated into novel environments. Understanding this process of exploration is highly desirable for management that seeks to resettle wildlife, particularly as translocation projects tend to be expensive and have a high potential for failure. African savannah elephants (Loxodonta africana) are very mobile and rely on large areas especially in arid environments, and are translocated for differing management and conservation objectives. Thus, research into how translocated elephants use the landscape when released may both guide elephant managers and be useful for translocations of other species that adjust their movement to social and ecological conditions. In this study, we investigated the movement of eight GPS tracked calves (translocated in three cohorts) following their soft release into a 107 km2 fenced wildlife sanctuary in northern Kenya and compared their movement with that of five tracked wild elephants in the sanctuary. We describe their exploration of the sanctuary, discovery of water points, and activity budgets during the first seven, 14, and 20 months after release. We explored how patterns are affected by time since release, ecological conditions, and social factors. We found that calves visited new areas of the sanctuary and water points during greener periods and earlier post-release. Social context was associated with exploration, with later release and association with wild elephants predictive of visits to new areas. Wild elephants tended to use a greater number of sites per 14-day period than the released calves. Activity budgets determined from hidden Markov models (including the states directed walk, encamped, and meandering) suggested that released calves differed from wild elephants. The first two cohorts of calves spent a significantly greater proportion of time in the directed walk state and a significantly lower proportion of time in the encamped state relative to the wild elephants. Our results represent a step forward in describing the movements of elephant orphan calves released to the wild following a period of profound social disruption when they lost their natal family and were rehabilitated with other orphan calves under human care. We discuss the implications of the elephant behavior we observed for improving release procedures and for defining success benchmarks for translocation projects.

Horizontal-vertical movement relationships: Adélie penguins forage continuously throughout provisioning trips

BACKGROUND

Diving marine predators forage in a three-dimensional environment, adjusting their horizontal and vertical movement behaviour in response to environmental conditions and the spatial distribution of prey. Expectations regarding horizontal-vertical movements are derived from optimal foraging theories, however, inconsistent empirical findings across a range of taxa suggests these behavioural assumptions are not universally applicable.

METHODS

Here, we examined how changes in horizontal movement trajectories corresponded with diving behaviour and marine environmental conditions for a ubiquitous Southern Ocean predator, the Adélie penguin. Integrating extensive telemetry-based movement and environmental datasets for chick-rearing Adélie penguins at Béchervaise Island, we tested the relationships between horizontal move persistence (continuous scale indicating low ['resident'] to high ['directed'] movement autocorrelation), vertical dive effort and environmental variables.

RESULTS

Penguins dived continuously over the course of their foraging trips and lower horizontal move persistence corresponded with less intense foraging activity, likely indicative of resting behaviour. This challenges the traditional interpretation of horizontal-vertical movement relationships based on optimal foraging models, which assumes increased residency within an area translates to increased foraging activity. Movement was also influenced by different environmental conditions during the two stages of chick-rearing: guard and crèche. These differences highlight the strong seasonality of foraging habitat for chick-rearing Adélie penguins at Béchervaise Island.

CONCLUSIONS

Our findings advance our understanding of the foraging behaviour for this marine predator and demonstrates the importance of integrating spatial location and behavioural data before inferring habitat use.

A random walk model that accounts for space occupation and movements of a large herbivore

Animal movement has been identified as a key feature in understanding animal behavior, distribution and habitat use and foraging strategies among others. Large datasets of invididual locations often remain unused or used only in part due to the lack of practical models that can directly infer the desired features from raw GPS locations and the complexity of existing approaches. Some of them being disputed for their lack of biological justifications in their design. We propose a simple model of individual movement with explicit parameters, based on a two-dimensional biased and correlated random walk with three forces related to advection (correlation), attraction (bias) and immobility of the animal. These forces can be directly estimated using individual data. We demonstrate the approach by using GPS data of 5 red deer with a high frequency sampling. The results show that a simple random walk template can account for the spatial complexity of wild animals. The practical design of the model is also verified for detecting spatial feature abnormalities and for providing estimates of density and abundance of wild animals. Integrating even more additional features of animal movement, such as individuals’ interactions or environmental repellents, could help to better understand the spatial behavior of wild animals.

Methodological advances for hypothesis-driven ethnobiology

Ethnobiology as a discipline has evolved increasingly to embrace theory-inspired and hypothesis-driven approaches to study why and how local people choose plants and animals they interact with and use for their livelihood. However, testing complex hypotheses or a network of ethnobiological hypotheses is challenging, particularly for data sets with non-independent observations due to species phylogenetic relatedness or socio-relational links between participants. Further, to account fully for the dynamics of local ecological knowledge, it is important to include the spatially explicit distribution of knowledge, changes in knowledge, and knowledge transmission and use. To promote the use of advanced statistical modelling approaches that address these limitations, we synthesize methodological advances for hypothesis-driven research in ethnobiology while highlighting the need for more figures than tables and more tables than text in ethnobiological literature. We present the ethnobiological motivations for conducting generalized linear mixed-effect modelling, structural equation modelling, phylogenetic generalized least squares, social network analysis, species distribution modelling, and predictive modelling. For each element of the proposed ethnobiologists quantitative toolbox, we present practical applications along with scripts for a widespread implementation. Because these statistical modelling approaches are rarely taught in most ethnobiological programs but are essential for careers in academia or industry, it is critical to promote workshops and short courses focused on these advanced methods. By embracing these quantitative modelling techniques without sacrificing qualitative approaches which provide essential context, ethnobiology will progress further towards an expansive interaction with other disciplines.

A bird's-eye view on turbulence: seabird foraging associations with evolving surface flow features

Understanding physical mechanisms underlying seabird foraging is fundamental to predict responses to coastal change. For instance, turbulence in the water arising from natural or anthropogenic structures can affect foraging opportunities in tidal seas. Yet, identifying ecologically important localized turbulence features (e.g. upwellings approximately 10-100 m) is limited by observational scale, and this knowledge gap is magnified in volatile predators. Here, using a drone-based approach, we present the tracking of surface-foraging terns (143 trajectories belonging to three tern species) and dynamic turbulent surface flow features in synchrony. We thereby provide the earliest evidence that localized turbulence features can present physical foraging cues. Incorporating evolving vorticity and upwelling features within a hidden Markov model, we show that terns were more likely to actively forage as the strength of the underlying vorticity feature increased, while conspicuous upwellings ahead of the flight path presented a strong physical cue to stay in transit behaviour. This clearly encapsulates the importance of prevalent turbulence features as localized foraging cues. Our quantitative approach therefore offers the opportunity to unlock knowledge gaps in seabird sensory and foraging ecology on hitherto unobtainable scales. Finally, it lays the foundation to predict responses to coastal change to inform sustainable ocean development.

Three-dimensional random walk models of individual animal movement and their application to trap counts modelling

BACKGROUND

Random walks (RWs) have proved to be a powerful modelling tool in ecology, particularly in the study of animal movement. An application of RW concerns trapping which is the predominant sampling method to date in insect ecology and agricultural pest management. A lot of research effort has been directed towards modelling ground-dwelling insects by simulating their movement in 2D, and computing pitfall trap counts, but comparatively very little for flying insects with 3D elevated traps.

METHODS

We introduce the mathematics behind 3D RWs and present key metrics such as the mean squared displacement (MSD) and path sinuosity, which are already well known in 2D. We develop the mathematical theory behind the 3D correlated random walk (CRW) which involves short-term directional persistence and the 3D Biased random walk (BRW) which introduces a long-term directional bias in the movement so that there is an overall preferred movement direction. In this study, we focus on the geometrical aspects of the 3D trap and thus consider three types of shape; a spheroidal trap, a cylindrical trap and a rectangular cuboidal trap. By simulating movement in 3D space, we investigated the effect of 3D trap shapes and sizes and of movement diffusion on trapping efficiency.

RESULTS

We found that there is a non-linear dependence of trap counts on the trap surface area or volume, but the effect of volume appeared to be a simple consequence of changes in area. Nevertheless, there is a slight but clear hierarchy of trap shapes in terms of capture efficiency, with the spheroidal trap retaining more counts than a cylinder, followed by the cuboidal type for a given area. We also showed that there is no effect of short-term persistence when diffusion is kept constant, but trap counts significantly decrease with increasing diffusion.

CONCLUSION

Our results provide a better understanding of the interplay between the movement pattern, trap geometry and impacts on trapping efficiency, which leads to improved trap count interpretations, and more broadly, has implications for spatial ecology and population dynamics.

Analysis of local habitat selection and large-scale attraction/avoidance based on animal tracking data: is there a single best method?

BACKGROUND

New wildlife telemetry and tracking technologies have become available in the last decade, leading to a large increase in the volume and resolution of animal tracking data. These technical developments have been accompanied by various statistical tools aimed at analysing the data obtained by these methods.

METHODS

We used simulated habitat and tracking data to compare some of the different statistical methods frequently used to infer local resource selection and large-scale attraction/avoidance from tracking data. Notably, we compared spatial logistic regression models (SLRMs), spatio-temporal point process models (ST-PPMs), step selection models (SSMs), and integrated step selection models (iSSMs) and their interplay with habitat and animal movement properties in terms of statistical hypothesis testing.

RESULTS

We demonstrated that only iSSMs and ST-PPMs showed nominal type I error rates in all studied cases, whereas SSMs may slightly and SLRMs may frequently and strongly exceed these levels. iSSMs appeared to have on average a more robust and higher statistical power than ST-PPMs.

CONCLUSIONS

Based on our results, we recommend the use of iSSMs to infer habitat selection or large-scale attraction/avoidance from animal tracking data. Further advantages over other approaches include short computation times, predictive capacity, and the possibility of deriving mechanistic movement models.

Habitat and resource segregation of two sympatric seals in the North Sea

The study of ecological niche segregation in sympatric species is essential to understand ecosystem functioning and its response to potential changes. In the North Sea, sympatric grey and harbour seals may present competition for food resources sustained by intense fishing activities and recent increase of seal populations. In order to coexist and reduce inter-specific competition, sympatric species must segregate at least one aspect of their ecological niches: temporal, spatial or resource segregation. We aim to study the foraging resources and foraging distributions of grey seals and harbour seals and the potential competition between these species in the North Sea. Therefore, we analysed stable isotopic composition of C, N and S (δ¹³C, δ¹⁵N and δ³⁴S values), and the concentrations of Hg and Se in blood of harbour and grey seals from the North Sea. Blood samples were collected on 45 grey seals and 37 harbour seals sampled along German and Scottish coasts. Stable isotope ratios were performed with an isotope ratio mass spectrometer coupled to an N-C-S elemental analyser for automated analyses. Total mercury concentrations (T-Hg) were determined by atomic absorption spectroscopy and Se concentrations by ICP-MS. The multi-tracer approach shown spatial and resource partitioning within grey and harbour seal living along German and Scottish coasts. Data indicate 1) the offshore foraging distribution of grey seals as reflected by the lower δ¹⁵N values and T-Hg concentrations and higher Se concentrations and 2) the inshore foraging distribution of harbour seals because of higher δ¹⁵N values and T-Hg concentrations and lower Se concentrations. The SIAR mixing model revealed 3) a more selective diet of grey seals compared to harbour seals and 4) the importance of sandeels in grey seal diet reflected by their high δ³⁴S values. Lastly, diet ellipse overlaps between grey seals and harbour seals sampled along the German coasts suggested 5) a potential sharing of food resources, possibly due to the increase number of grey seals number in this area during the foraging season - all year except breeding and moulting periods. The multi-tracer approach of this study provides a more robust discrimination among diet resources and spatial foraging distributions of grey seals and harbour seals in the North Sea.

Identifying resting locations of a small elusive forest carnivore using a two-stage model accounting for GPS measurement error and hidden behavioral states

BACKGROUND

Studies of animal movement using location data are often faced with two challenges. First, time series of animal locations are likely to arise from multiple behavioral states (e.g., directed movement, resting) that cannot be observed directly. Second, location data can be affected by measurement error, including failed location fixes. Simultaneously addressing both problems in a single statistical model is analytically and computationally challenging. To both separate behavioral states and account for measurement error, we used a two-stage modeling approach to identify resting locations of fishers (Pekania pennanti) based on GPS and accelerometer data.

METHODS

We developed a two-stage modelling approach to estimate when and where GPS-collared fishers were resting for 21 separate collar deployments on 9 individuals in southern Oregon. For each deployment, we first fit independent hidden Markov models (HMMs) to the time series of accelerometer-derived activity measurements and apparent step lengths to identify periods of movement and resting. Treating the state assignments as given, we next fit a set of linear Gaussian state space models (SSMs) to estimate the location of each resting event.

RESULTS

Parameter estimates were similar across collar deployments. The HMMs successfully identified periods of resting and movement with posterior state assignment probabilities greater than 0.95 for 97% of all observations. On average, fishers were in the resting state 63% of the time. Rest events averaged 5 h (4.3 SD) and occurred most often at night. The SSMs allowed us to estimate the 95% credible ellipses with a median area of 0.12 ha for 3772 unique rest events. We identified 1176 geographically distinct rest locations; 13% of locations were used on > 1 occasion and 5% were used by > 1 fisher. Females and males traveled an average of 6.7 (3.5 SD) and 7.7 (6.8 SD) km/day, respectively.

CONCLUSIONS

We demonstrated that if auxiliary data are available (e.g., accelerometer data), a two-stage approach can successfully resolve both problems of latent behavioral states and GPS measurement error. Our relatively simple two-stage method is repeatable, computationally efficient, and yields directly interpretable estimates of resting site locations that can be used to guide conservation decisions.

A hierarchical machine learning framework for the analysis of large scale animal movement data

BACKGROUND

In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated.

METHODS

In this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods.

RESULTS

While formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries.

CONCLUSIONS

Multilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.

Spatially Explicit Capture-Recapture Through Camera Trapping: A Review of Benchmark Analyses for Wildlife Density Estimation

Camera traps have become an important research tool for both conservation biologists and wildlife managers. Recent advances in spatially explicit capture-recapture (SECR) methods have increasingly put camera traps at the forefront of population monitoring programs. These methods allow for benchmark analysis of species density without the need for invasive fieldwork techniques. We conducted a review of SECR studies using camera traps to summarize the current focus of these investigations, as well as provide recommendations for future studies and identify areas in need of future investigation. Our analysis shows a strong bias in species preference, with a large proportion of studies focusing on large felids, many of which provide the only baseline estimates of population density for these species. Furthermore, we found that a majority of studies produced density estimates that may not be precise enough for long-term population monitoring. We recommend simulation and power analysis be conducted before initiating any particular study design and provide examples using readily available software. Furthermore, we show that precision can be increased by including a larger study area that will subsequently increase the number of individuals photo-captured. As many current studies lack the resources or manpower to accomplish such an increase in effort, we recommend that researchers incorporate new technologies such as machine-learning, web-based data entry, and online deployment management into their study design. We also cautiously recommend the potential of citizen science to help address these study design concerns. In addition, modifications in SECR model development to include species that have only a subset of individuals available for individual identification (often called mark-resight models), can extend the process of explicit density estimation through camera trapping to species not individually identifiable.

Uncovering ecological state dynamics with hidden Markov models

Ecological systems can often be characterised by changes among a finite set of underlying states pertaining to individuals, populations, communities or entire ecosystems through time. Owing to the inherent difficulty of empirical field studies, ecological state dynamics operating at any level of this hierarchy can often be unobservable or 'hidden'. Ecologists must therefore often contend with incomplete or indirect observations that are somehow related to these underlying processes. By formally disentangling state and observation processes based on simple yet powerful mathematical properties that can be used to describe many ecological phenomena, hidden Markov models (HMMs) can facilitate inferences about complex system state dynamics that might otherwise be intractable. However, HMMs have only recently begun to gain traction within the broader ecological community. We provide a gentle introduction to HMMs, establish some common terminology, review the immense scope of HMMs for applied ecological research and provide a tutorial on implementation and interpretation. By illustrating how practitioners can use a simple conceptual template to customise HMMs for their specific systems of interest, revealing methodological links between existing applications, and highlighting some practical considerations and limitations of these approaches, our goal is to help establish HMMs as a fundamental inferential tool for ecologists.

Machine learning for modeling animal movement

Animal movement drives important ecological processes such as migration and the spread of infectious disease. Current approaches to modeling animal tracking data focus on parametric models used to understand environmental effects on movement behavior and to fill in missing tracking data. Machine Learning and Deep learning algorithms are powerful and flexible predictive modeling tools but have rarely been applied to animal movement data. In this study we present a general framework for predicting animal movement that is a combination of two steps: first predicting movement behavioral states and second predicting the animal's velocity. We specify this framework at the individual level as well as for collective movement. We use Random Forests, Neural and Recurrent Neural Networks to compare performance predicting one step ahead as well as long range simulations. We compare results against a custom constructed Stochastic Differential Equation (SDE) model. We apply this approach to high resolution ant movement data. We found that the individual level Machine Learning and Deep Learning methods outperformed the SDE model for one step ahead prediction. The SDE model did comparatively better at simulating long range movement behaviour. Of the Machine Learning and Deep Learning models the Long Short Term Memory (LSTM) individual level model did best at long range simulations. We also applied the Random Forest and LSTM individual level models to model gull migratory movement to demonstrate the generalizability of this framework. Machine Learning and deep learning models are easier to specify compared to traditional parametric movement models which can have restrictive assumptions. However, machine learning and deep learning models are less interpretable than parametric movement models. The type of model used should be determined by the goal of the study, if the goal is prediction, our study provides evidence that machine learning and deep learning models could be useful tools.

Possibility of Autonomous Estimation of Shiba Goat’s Estrus and Non-Estrus Behavior by Machine Learning Methods

Mammalian behavior is typically monitored by observation. However, direct observation requires a substantial amount of effort and time, if the number of mammals to be observed is sufficiently large or if the observation is conducted for a prolonged period. In this study, machine learning methods as hidden Markov models (HMMs), random forests, support vector machines (SVMs), and neural networks, were applied to detect and estimate whether a goat is in estrus based on the goat’s behavior; thus, the adequacy of the method was verified. Goat’s tracking data was obtained using a video tracking system and used to estimate whether they, which are in “estrus” or “non-estrus”, were in either states: “approaching the male”, or “standing near the male”. Totally, the PC of random forest seems to be the highest. However, The percentage concordance (PC) value besides the goats whose data were used for training data sets is relatively low. It is suggested that random forest tend to over-fit to training data. Besides random forest, the PC of HMMs and SVMs is high. However, considering the calculation time and HMM’s advantage in that it is a time series model, HMM is better method. The PC of neural network is totally low, however, if the more goat’s data were acquired, neural network would be an adequate method for estimation.

Reproductive behavior drives female space use in a sedentary Neotropical frog

Longer-range movements of anuran amphibians such as mass migrations and habitat invasion have received a lot of attention, but fine-scale spatial behavior remains largely understudied. This gap is especially striking for species that show long-term site fidelity and display their whole behavioral repertoire in a small area. Studying fine-scale movement with conventional capture-mark-recapture techniques is difficult in inconspicuous amphibians: individuals are hard to find, repeated captures might affect their behavior and the number of data points is too low to allow a detailed interpretation of individual space use and time budgeting. In this study, we overcame these limitations by equipping females of the Brilliant-Thighed Poison Frog (Allobates femoralis) with a tag allowing frequent monitoring of their location and behavior. Neotropical poison frogs are well known for their complex behavior and diverse reproductive and parental care strategies. Although the ecology and behavior of the polygamous leaf-litter frog Allobates femoralis is well studied, little is known about the fine-scale space use of the non-territorial females who do not engage in acoustic and visual displays. We tracked 17 females for 6 to 17 days using a harmonic direction finder to provide the first precise analysis of female space use in this species. Females moved on average 1 m per hour and the fastest movement, over 20 m per hour, was related to a subsequent mating event. Traveled distances and activity patterns on days of courtship and mating differed considerably from days without reproduction. Frogs moved more on days with lower temperature and more precipitation, but mating seemed to be the main trigger for female movement. We observed 21 courtships of 12 tagged females. For seven females, we observed two consecutive mating events. Estimated home ranges after 14 days varied considerably between individuals and courtship and mating associated space use made up for ∼30% of the home range. Allobates femoralis females spent large parts of their time in one to three small centers of use. Females did not adjust their time or space use to the density of males in their surroundings and did not show wide-ranging exploratory behavior. Our study demonstrates how tracking combined with detailed behavioral observations can reveal the patterns and drivers of fine-scale spatial behavior in sedentary species.

Using approximate Bayesian inference for a "steps and turns" continuous-time random walk observed at regular time intervals

The study of animal movement is challenging because movement is a process modulated by many factors acting at different spatial and temporal scales. In order to describe and analyse animal movement, several models have been proposed which differ primarily in the temporal conceptualization, namely continuous and discrete time formulations. Naturally, animal movement occurs in continuous time but we tend to observe it at fixed time intervals. To account for the temporal mismatch between observations and movement decisions, we used a state-space model where movement decisions (steps and turns) are made in continuous time. That is, at any time there is a non-zero probability of making a change in movement direction. The movement process is then observed at regular time intervals. As the likelihood function of this state-space model turned out to be intractable yet simulating data is straightforward, we conduct inference using different variations of Approximate Bayesian Computation (ABC). We explore the applicability of this approach as a function of the discrepancy between the temporal scale of the observations and that of the movement process in a simulation study. Simulation results suggest that the model parameters can be recovered if the observation time scale is moderately close to the average time between changes in movement direction. Good estimates were obtained when the scale of observation was up to five times that of the scale of changes in direction. We demonstrate the application of this model to a trajectory of a sheep that was reconstructed in high resolution using information from magnetometer and GPS devices. The state-space model used here allowed us to connect the scales of the observations and movement decisions in an intuitive and easy to interpret way. Our findings underscore the idea that the time scale at which animal movement decisions are made needs to be considered when designing data collection protocols. In principle, ABC methods allow to make inferences about movement processes defined in continuous time but in terms of easily interpreted steps and turns.

Intensive and extensive movements of feral camels in central Australia

A better understanding of the movement of feral dromedary camels (Camelus dromedarius) in Australia would be useful for planning removal operations (harvest or culling), because the pattern and scale of camel movement relates to the period they reside in a given area, and thus the search effort, timing and frequency of removal operations. From our results, we suspect that the dune direction influences how camels move across central Australia; particularly effects like the north–south longitudinal dune systems in the Simpson Desert, which appeared to elongate camel movement in the same direction as the dunes. We called this movement anisotropy. Research suggests camel movement in Australia is not migratory but partially cyclic, with two distinctive movement patterns. Our study investigated this further by using satellite tracking data from 54 camels in central Australia, recorded between 2007 and 2016. The mean tracking period for each animal was 363.9 days (s.e.m.=44.1 days). We used a method labelled multi-scale partitioning to test for changes in movement behaviour and partitioned more localised intensive movements within utilisation areas, from larger-scale movement, called ranging. This involved analysing the proximity of movement trajectories to other nearby trajectories of the same animal over time. We also used Dynamic Brownian Bridges Movement Models, which consider the relationship of consecutive locations to determine the areas of utilisation. The mean utilisation area and duration of a camel (n=658 areas) was found to be 342.6km2 (s.e.m.=33.2km2) over 23.5 days (s.e.m.=1.6 days), and the mean ranging distance (n=611 ranging paths) was a 45.1km (s.e.m.=2.0km) path over 3.1 days (s.e.m.=0.1 days).

A continuous-time state-space model for rapid quality control of argos locations from animal-borne tags

BACKGROUND

State-space models are important tools for quality control and analysis of error-prone animal movement data. The near real-time (within 24 h) capability of the Argos satellite system can aid dynamic ocean management of human activities by informing when animals enter wind farms, shipping lanes, and other intensive use zones. This capability also facilitates the use of ocean observations from animal-borne sensors in operational ocean forecasting models. Such near real-time data provision requires rapid, reliable quality control to deal with error-prone Argos locations.

METHODS

We formulate a continuous-time state-space model to filter the three types of Argos location data (Least-Squares, Kalman filter, and Kalman smoother), accounting for irregular timing of observations. Our model is deliberately simple to ensure speed and reliability for automated, near real-time quality control of Argos location data. We validate the model by fitting to Argos locations collected from 61 individuals across 7 marine vertebrates and compare model-estimated locations to contemporaneous GPS locations. We then test assumptions that Argos Kalman filter/smoother error ellipses are unbiased, and that Argos Kalman smoother location accuracy cannot be improved by subsequent state-space modelling.

RESULTS

Estimation accuracy varied among species with Root Mean Squared Errors usually <5 km and these decreased with increasing data sampling rate and precision of Argos locations. Including a model parameter to inflate Argos error ellipse sizes in the north - south direction resulted in more accurate location estimates. Finally, in some cases the model appreciably improved the accuracy of the Argos Kalman smoother locations, which should not be possible if the smoother is using all available information.

CONCLUSIONS

Our model provides quality-controlled locations from Argos Least-Squares or Kalman filter data with accuracy similar to or marginally better than Argos Kalman smoother data that are only available via fee-based reprocessing. Simplicity and ease of use make the model suitable both for automated quality control of near real-time Argos data and for manual use by researchers working with historical Argos data.