Reliable methods for identifying animal deaths in GPS ‐ and satellite‐tracking data: Review, testing, and calibration

Flyways and migratory behaviour of the Vega gull (Larus vegae), a little-known Arctic endemic

Large gulls are generalist predators that play an important role in Arctic food webs. Describing the migratory patterns and phenology of these predators is essential to understanding how Arctic ecosystems function. However, from all six large Arctic gull taxa, including three long-distance migrants, to date seasonal movements have been studied only in three and with small sample sizes. To document the flyways and migratory behaviour of the Vega gull, a widespread but little-studied Siberian migrant, we monitored 28 individuals with GPS loggers over a mean period of 383 days. Birds used similar routes in spring and autumn, preferring coastal to inland or offshore routes, and travelled 4000-5500 km between their breeding (Siberia) and wintering grounds (mainly the Republic of Korea and Japan). Spring migration mainly occurred in May, and was twice as fast and more synchronized among individuals than autumn migration. Migration bouts mainly occurred during the day and twilight, but rates of travel were always higher during the few night flights. Flight altitudes were nearly always higher during migration bouts than during other bouts, and lower during twilight than during night or day. Altitudes above 2000m were recorded during migrations, when birds made non-stop inland flights over mountain ranges and vast stretches of the boreal forest. Individuals showed high inter-annual consistency in their movements in winter and summer, indicating strong site fidelity to their breeding and wintering sites. Within-individual variation was similar in spring and autumn, but between individual variation was higher in autumn than in spring. Compared to previous studies, our results suggest that the timing of spring migration in large Arctic gulls is likely constrained by snowmelt at breeding grounds, while the duration of migration windows could be related to the proportion of inland versus coastal habitats found along their flyways ('fly-and-forage' strategy). Ongoing environmental changes are hence likely in short term to alter the timing of their migration, and in long term possibly affect the duration if e.g. the resource availability along the route changes in the future.

Linking migration and microbiota at a major stopover site in a long-distance avian migrant

Migration is one of the most physical and energetically demanding periods in an individual bird's life. The composition of the bird's gut or cloacal microbiota can temporarily change during migration, likely due to differences in diets, habitats and other environmental conditions experienced en route. However, how physiological condition, migratory patterns, and other drivers interact to affect microbiota composition of migratory birds is still unclear. We sampled the cloacal bacterial microbiota of a long-distance migrant, the steppe buzzard (Buteo buteo vulpinus), at an important spring stopover bottleneck in Eilat, Israel, after crossing the ca. 1800 km Sahara Desert. We examined whether diversity and composition of the cloacal microbiota varied with body condition, sex, movement patterns (i.e., arrival time and migration distance), and survival. Early arrival to Eilat was associated with better body condition, longer post-Eilat spring migration distance, higher microbial α-diversity, and differences in microbiota composition. Specifically, early arrivals had higher abundance of the phylum Synergistota and five genera, including Jonquetella and Peptococcus, whereas the phylum Proteobacteria and genus Escherichia-Shigella (as well as three other genera) were more abundant in later arrivals. While the differences in α-diversity and Escherichia-Shigella seem to be mainly driven by body condition, other compositional differences associated with arrival date could be indicators of longer migratory journeys (e.g., pre-fueling at wintering grounds or stopover habitats along the way) or migratory performance. No significant differences were found between the microbiota of surviving and non-surviving individuals. Overall, our results indicate that variation in steppe buzzard microbiota is linked to variation in migratory patterns (i.e., capture/arrival date) and body condition, highlighting the importance of sampling the microbiota of GPS-tracked individuals on multiple occasions along their migration routes to gain a more detailed understanding of the links between migration, microbiota, and health in birds.

Frontiers in quantifying wildlife behavioural responses to chemical pollution

Animal behaviour is remarkably sensitive to disruption by chemical pollution, with widespread implications for ecological and evolutionary processes in contaminated wildlife populations. However, conventional approaches applied to study the impacts of chemical pollutants on wildlife behaviour seldom address the complexity of natural environments in which contamination occurs. The aim of this review is to guide the rapidly developing field of behavioural ecotoxicology towards increased environmental realism, ecological complexity, and mechanistic understanding. We identify research areas in ecology that to date have been largely overlooked within behavioural ecotoxicology but which promise to yield valuable insights, including within- and among-individual variation, social networks and collective behaviour, and multi-stressor interactions. Further, we feature methodological and technological innovations that enable the collection of data on pollutant-induced behavioural changes at an unprecedented resolution and scale in the laboratory and the field. In an era of rapid environmental change, there is an urgent need to advance our understanding of the real-world impacts of chemical pollution on wildlife behaviour. This review therefore provides a roadmap of the major outstanding questions in behavioural ecotoxicology and highlights the need for increased cross-talk with other disciplines in order to find the answers.

A sex skew in life-history research: the problem of missing males

Life-history strategies are diverse. While understanding this diversity is a fundamental aim of evolutionary biology and biodemography, life-history data for some traits-in particular, age-dependent reproductive investment-are biased towards females. While other authors have highlighted this sex skew, the general scale of this bias has not been quantified and its impact on our understanding of evolutionary ecology has not been discussed. This review summarizes why the sexes can evolve different life-history strategies. The scale of the sex skew is then discussed and its magnitude compared between taxonomic groups, laboratory and field studies, and through time. We discuss the consequences of this sex skew for evolutionary and ecological research. In particular, this sex bias means that we cannot test some core evolutionary theory. Additionally, this skew could obscure or drive trends in data and hinder our ability to develop effective conservation strategies. We finally highlight some ways through which this skew could be addressed to help us better understand broad patterns in life-history strategies.

Compensation for wind drift during raptor migration improves with age through mortality selection

Each year, billions of flying and swimming migrants negotiate the challenging displacement imposed by travelling through a flowing medium. However, little is known about how the ability to cope with drift improves through life and what mechanisms drive its development. We examined 3,140 days of migration by 90 GPS-tagged raptorial black kites (Milvus migrans) aged 1-27 years to show that the ability to compensate for lateral drift develops gradually through many more years than previously appreciated. Drift negotiation was under strong selective pressure, with inferior navigators subject to increased mortality. This progressively selected for adults able to compensate for current cross flows and for previously accumulated drift in a flexible, context-dependent and risk-dependent manner. Displacements accumulated en route carried over to shape the wintering distribution of the population. For many migrants, migratory journeys by younger individuals represent concentrated episodes of trait selection that shape adult populations and mediate their adaptation to climate change.

Machine learned daily life history classification using low frequency tracking data and automated modelling pipelines: application to North American waterfowl

BACKGROUND

Identifying animal behaviors, life history states, and movement patterns is a prerequisite for many animal behavior analyses and effective management of wildlife and habitats. Most approaches classify short-term movement patterns with high frequency location or accelerometry data. However, patterns reflecting life history across longer time scales can have greater relevance to species biology or management needs, especially when available in near real-time. Given limitations in collecting and using such data to accurately classify complex behaviors in the long-term, we used hourly GPS data from 5 waterfowl species to produce daily activity classifications with machine-learned models using "automated modelling pipelines".

METHODS

Automated pipelines are computer-generated code that complete many tasks including feature engineering, multi-framework model development, training, validation, and hyperparameter tuning to produce daily classifications from eight activity patterns reflecting waterfowl life history or movement states. We developed several input features for modeling grouped into three broad categories, hereafter "feature sets": GPS locations, habitat information, and movement history. Each feature set used different data sources or data collected across different time intervals to develop the "features" (independent variables) used in models.

RESULTS

Automated modelling pipelines rapidly developed easily reproducible data preprocessing and analysis steps, identification and optimization of the best performing model and provided outputs for interpreting feature importance. Unequal expression of life history states caused unbalanced classes, so we evaluated feature set importance using a weighted F1-score to balance model recall and precision among individual classes. Although the best model using the least restrictive feature set (only 24 hourly relocations in a day) produced effective classifications (weighted F1 = 0.887), models using all feature sets performed substantially better (weighted F1 = 0.95), particularly for rarer but demographically more impactful life history states (i.e., nesting).

CONCLUSIONS

Automated pipelines generated models producing highly accurate classifications of complex daily activity patterns using relatively low frequency GPS and incorporating more classes than previous GPS studies. Near real-time classification is possible which is ideal for time-sensitive needs such as identifying reproduction. Including habitat and longer sequences of spatial information produced more accurate classifications but incurred slight delays in processing.

Age of First Territory Settlement of Golden Eagles Aquila chrysaetos in a Variable Competitive Landscape

With life-history traits involving high survival, low reproductive output, years of natal dispersal and deferred maturity, the population ecology and behaviour of large raptors which occur at low densities can be difficult to study. The age at which large raptors first settle on a prospective breeding territory receives relatively little attention, but is a key metric in population modelling including, for example, projections of reintroduction projects. It can also be a barometer of the “health” of populations and the availability of breeding opportunities. The advancement of GPS-telemetry has proved invaluable in gaining insights into several aspects of large raptor ecology and behaviour. Age of first territory settlement (AFTS) is one such aspect. AFTS is important in modelling population trajectories and considering individuals’ lifetime reproductive success. We used an algorithm based on GPS-records from dispersing Golden Eagles tagged as nestlings in Scotland to estimate AFTS. While the lifespan of GPS-tags can bias against settlement dates of older birds, they can also potentially reveal settlement ages difficult or impossible to discern from other methods. We found a range of ages for AFTS, including those in their second calendar year; much younger than previously documented by other methods. Ground-truthing – when possible and if inevitably slightly delayed – confirmed territory occupation on field-based survey criteria. We found that eagles settled younger in vacant territories and when older in occupied existing territories. Birds’ sex had no effect on AFTS. Birds which dispersed earlier from their natal territory (indicative of “quality” from some previous research) had no association with AFTS. Our results indicate that within technological temporal limits GPS-data can provide for accurate and precise estimations of AFTS including early settlement not consistently or precisely recorded by other methods. Within our study’s variable competitive landscape we found that AFTS was associated with the availability of territorial opportunities but not with the timing of dispersal. These findings have consequences for studying and understanding large raptor population dynamics.

Animal mortality during fire

Earth's rapidly warming climate is propelling us towards an increasingly fire-prone future. Currently, knowledge of the extent and characteristics of animal mortality rates during fire remains rudimentary, hindering our ability to predict how animal populations may be impacted in the future. To address this knowledge gap, we conducted a global systematic review of the direct effects of fire on animal mortality rates, based on studies that unequivocally determined the fate of animals during fire. From 31 studies spanning 1984-2020, we extracted data on the direct impacts of fire on the mortality of 31 species from 23 families. From these studies, there were 43 instances where direct effects were measured by reporting animal survival from pre- to post-fire. Most studies were conducted in North America (52%) and Oceania (42%), focused largely on mammals (53%) and reptiles (30%), and reported mostly on animal survival in planned (82%) and/or low severity (70%) fires. We found no studies from Asia, Europe or South America. Although there were insufficient data to conduct a formal meta-analysis, we tested the effect of fire type, fire severity, fire regime, animal body mass, ecological attributes and class on survival. Only fire severity affected animal mortality, with a higher proportion of animals being killed by high than low severity fires. Recent catastrophic fires across the globe have drawn attention to the plight of animals exposed to wildfire. Yet, our systematic review suggests that a relatively low proportion of animals (mean predicted mortality [95% CI] = 3% [1%-9%]) are killed during fire. However, our review also underscores how little we currently know about the direct effects of fire on animal mortality, and highlights the critical need to understand the effects of high severity fire on animal populations.

Non-territorial GPS-tagged golden eagles Aquila chrysaetos at two Scottish wind farms: Avoidance influenced by preferred habitat distribution, wind speed and blade motion status

Wind farms can have two broad potential adverse effects on birds via antagonistic processes: displacement from the vicinity of turbines (avoidance), or death through collision with rotating turbine blades. These effects may not be mutually exclusive. Using detailed data from 99 turbines at two wind farms in central Scotland and thousands of GPS-telemetry data from dispersing golden eagles, we tested three hypotheses. Before-and-after-operation analyses supported the hypothesis of avoidance: displacement was reduced at turbine locations in more preferred habitat and with more preferred habitat nearby. After-operation analyses (i.e. from the period when turbines were operational) showed that at higher wind speeds and in highly preferred habitat eagles were less wary of turbines with motionless blades: rejecting our second hypothesis. Our third hypothesis was supported, since at higher wind speeds eagles flew closer to operational turbines; especially-once more-turbines in more preferred habitat. After operation, eagles effectively abandoned inner turbine locations, and flight line records close to rotor blades were rare. While our study indicated that whole-wind farm functional habitat loss through avoidance was the substantial adverse impact, we make recommendations on future wind farm design to minimise collision risk further. These largely entail developers avoiding outer turbine locations which are in and surrounded by swathes of preferred habitat. Our study illustrates the insights which detailed case studies of large raptors at wind farms can bring and emphasises that the balance between avoidance and collision can have several influences.

Demographic modeling to fine-tune conservation targets: importance of pre-adults for the decline of an endangered raptor

Large, long-lived species with slow life histories and protracted pre-breeding stages are particularly susceptible to declines and extinction, often for unknown causes. Here, we show how demographic modeling of a medium-sized raptor, the Red Kite Milvus milvus, can aid to refocus conservation research and attention on the most likely mechanisms driving its decline. Red Kites' survival and reproduction increased through three sequential stages for 1-2, 3-6, and 7-30 yr of age, mainly corresponding to individuals that are dispersing, attempting to gain a territory, and breeding. As typical of long-lived species, elasticities were highest for adult (≥7 yr old) survival, but this was high, with little scope for improvement. Instead, the declines were driven by an extremely low survival of pre-adults in their first years of life, which weakened the whole demographic system by nullifying the offspring contribution of adults and curtailing their replacement by recruits. For example, 27 pairs were necessary to generate a single prime age adult. Simulation of management scenarios suggested that the decline could be halted most parsimoniously by increasing pre-adult survival to the mean levels recorded for other areas, while only the synergistic, simultaneous improvement of breeding success, adult and pre-adult survival could generate a recovery. We propose three actions to attain such goals through selective supplementary feeding of both breeding and non-breeding individuals, and through mortality improvement by GPS remote-sensing devices employed as surveillance monitoring tools. Our results show how improving demographic models by using real, local vital rates rather than "best guess" vital rates can dramatically improve model realism by refocusing attention on the actual stages and mortality causes in need of manipulation, thus building precious time and resources for conservation management. These results also highlight the frequent key role of pre-adult survival for the management of long-lived species, coherent with the idea of demographic systems as integrated chains only as strong as their weakest link.

Differential survival throughout the full annual cycle of a migratory bird presents a life-history trade-off

Long-distance migrations are among the most physically demanding feats animals perform. Understanding the potential costs and benefits of such behaviour is a fundamental question in ecology and evolution. A hypothetical cost of migration should be outweighed by higher productivity and/or higher annual survival, but few studies on migratory species have been able to directly quantify patterns of survival throughout the full annual cycle and across the majority of a species' range. Here, we use telemetry data from 220 migratory Egyptian vultures Neophron percnopterus, tracked for 3,186 bird months and across approximately 70% of the species' global distribution, to test for differences in survival throughout the annual cycle. We estimated monthly survival probability relative to migration and latitude using a multi-event capture-recapture model in a Bayesian framework that accounted for age, origin, subpopulation and the uncertainty of classifying fates from tracking data. We found lower survival during migration compared to stationary periods (β = -0.816; 95% credible interval: -1.290 to -0.318) and higher survival on non-breeding grounds at southern latitudes (<25°N; β = 0.664; 0.076-1.319) compared to on breeding grounds. Survival was also higher for individuals originating from Western Europe (β = 0.664; 0.110-1.330) as compared to further east in Europe and Asia, and improved with age (β = 0.030; 0.020-0.042). Anthropogenic mortalities accounted for half of the mortalities with a known cause and occurred mainly in northern latitudes. Many juveniles drowned in the Mediterranean Sea on their first autumn migration while there were few confirmed mortalities in the Sahara Desert, indicating that migration barriers are likely species-specific. Our study advances the understanding of important fitness trade-offs associated with long-distance migration. We conclude that there is lower survival associated with migration, but that this may be offset by higher non-breeding survival at lower latitudes. We found more human-caused mortality farther north, and suggest that increasing anthropogenic mortality could disrupt the delicate migration trade-off balance. Research to investigate further potential benefits of migration (e.g. differential productivity across latitudes) could clarify how migration evolved and how migrants may persist in a rapidly changing world.

When and where mortality occurs throughout the annual cycle changes with age in a migratory bird: individual vs population implications

The annual cycle of most animals is structured into discrete stages, such as breeding, migration and dispersal. While there is growing appreciation of the importance of different stages of an organism’s annual cycle for its fitness and population dynamics, almost nothing is known about if and how such seasonal effects can change through a species lifespan. Here, we take advantage of the opportunity offered by a long-term satellite/GPS-tracking study and a reliable method of remote death-detection to show that certain stages of both the annual and life cycle of a migratory long-lived raptor, the Black kite Milvus migrans, may represent sensitive bottlenecks for survival. In particular, migratory journeys caused bursts of concentrated-mortality throughout life, but the relative importance of stage-specific survival changed with age. On the other hand, the balance between short-stages of high mortality and long-stages of low mortality made population-growth similarly dependent on all portions of the annual cycle. Our results illustrate how the population dynamics of migratory organisms can be inextricably linked to ecological pressures balanced over multiple stages of the annual cycle and thus multiple areas of the globe, suggesting the frequent need for challenging conservation strategies targeting all portions of a species year-round range.

Patterns of satellite tagged hen harrier disappearances suggest widespread illegal killing on British grouse moors

Identifying patterns of wildlife crime is a major conservation challenge. Here, we test whether deaths or disappearances of a protected species, the hen harrier, are associated with grouse moors, which are areas managed for the production of red grouse for recreational shooting. Using data from 58 satellite tracked hen harriers, we show high rates of unexpected tag failure and low first year survival compared to other harrier populations. The likelihood of harriers dying or disappearing increased as their use of grouse moors increased. Similarly, at the landscape scale, satellite fixes from the last week of life were distributed disproportionately on grouse moors in comparison to the overall use of such areas. This pattern was also apparent in protected areas in northern England. We conclude that hen harriers in Britain suffer elevated levels of mortality on grouse moors, which is most likely the result of illegal killing.

Tackling wildlife crimes requires determining their occurrence and distribution, but they are often difficult to detect. Here, the authors use hen harrier tracking data to show patterns of unexpected tag failure that suggest widespread illegal killing on moors managed for recreational shooting of red grouse.