1 + 1[1] 2Animal Population Spatial Analysis of Black Kites Using GPS Tracking Data
Quarto
Quarto enables you to weave together content and executable code into a finished document. To learn more about Quarto see https://quarto.org.
Running Code
When you click the Render button a document will be generated that includes both content and the output of embedded code. You can embed code like this:
You can add options to executable code like this
[1] 4The echo: false option disables the printing of code (only output is displayed).
1.Context
Knowledge of the spatial ecology and movement behaviour of birds is crucial to the development of ecological theory and to applied conservation. Birds, in particular large gliding raptors like the Black Kite Milvus migrans, are frequently used as indicator of the ecosystem state because of their wide ranging activity patterns and because they are sensitive to disturbance. Their movement data can provide insight into how individuals move across landscapes, choose habitat, respond to anthropogenic threats, and occupy their home ranges – all of which have tangible applications in conservation and land management plans.
Advancements in GPS telemetry in recent years have facilitated the collection of high-resolution movement data, allowing researchers to evaluate fine-scale space use and habitat selection (Rechetelo et al., 2016; Tonra et al., 2019). Commonly used methods including MCP and KDE provide estimates of home range and utilization distributions and hence information about spatial restrictions and resource hotspots that are available within an individual’s home range (Börger et al., 2006). Here we analyze GPS tracking data of several Black Kite individuals
Black Kites show central-place foraging, thermalling during migration or movement and are deterred by anthropogenic structures. For example, Santos et al. (2021) found that during migration close to the Strait of Gibraltar, both adult and juvenile Black Kites veered to avoid wind turbines, generally changing course at ~700–850 m from the turbines. This apparent consistent behavioural response reinforces the potential that distance measures driven by GPS could be deployed as surrogates of collision risk and functional habitat loss in raptors.
4) Indeed, research on other birds demonstrates the value of home range analysis in elucidating the use of space by individuals. Rechetelo et al. (2016) applied both MCP and KDE to estimate the home ranges of the endangered Black-Throated Finch (Poephila cincta cincta) in Australia, and found that while most individuals were relatively stationary in the vicinity of the banding site, some individuals performed long distance movements in response to apparent resource limitations. A similar type of movement variability has been documented in migratory species such as the Eastern Whip-poor-will (Antrostomus vociferous) that exhibited non-stochastic use of bounded areas during non-breeding periods as evidenced by GPS tag data (Tonra et al., 2019).
Together, these studies highlight the necessity of combining fine-scale GPS tracking data with behavioural state models and spatial analytic methods for meaningful ecological inference. As a result, this report applies such approaches to the calculation of home ranges (MCP and KDE), calculation of movement metrics (step length and turning angle), identification of behavioural states, and visualization of the spatial distribution of kite activity.
2. Analytical Approach
2.1 Dataset Description
The dataset used in the present study was based on high-resolution GPS tracking of individuals (n = 135) Black Kites (Milvus migrans) from the Movebank Data Repository (Santos et al., 2021). The birds were followed during post-breeding migration in a high wind turbine density region close to the Strait of Gibraltar, southern Spain. Each record of data contains a timestamp, as well as the latitude, longitude, UTM coordinates, height (meters above ground level), azimuth (degrees from north), and speed (meters per second) of the CopterSonde, bot-specific dust sensor reading, transmitter identifier, host_number and buddy_number of the holding individual, and host sex (male, female or unknown). There were a total of 231,193 location points in the dataset, captured at a regular time interval with the GPS loggers. All subjects used in the analysis had enough valid space-time data (>100 fixes) for good estimates of home range and movement metrics. Before analysis, the data was cleaned by removing duplications, entries with missing coordinates or missing timestamps.
2.2 Data Preprocessing and Tools
All analyses were performed using R (version 4.5.0) with a geospatial and movement ecology suite of packages, including adehabitatHR, amt, move, ggplot2 ,sf etc. The coordinates were projected into the UTM (Universal Transverse Mercator) projection to accurately calculate distances and areas. Records for those with limited or no records were excluded to uphold the soundness of the KDE and MCP analysis.
2.3 Home Range Estimation
Two different methods were used to measure individual spatial utilization:
Minimum Convex Polygon (MCP): We calculated a 95% MCP per individual as a coarse measure of overall utilization of space.
Kernel Density Estimation (KDE): Utilization distributions were generated using KDE with 95% (home range) and 50% (core area) isopleths. The smoothing parameter (bandwidth) was selected by Least Squares Cross Validation (LSCV). We chose KDE as it could show core activity areas and consider spatial intensity of use.Graphical outputs are KDE contour maps and visual comparison between MCP and KDE areas for given subjects.
2.4 Movement statistics and track analysis
The following movement metrics were estimated to assess movement behaviour:
Step Length: Distance flown between GPS locations as a straight path.
Turning Angle: The inclination of direction moving from one step to another.
These measures were used to evaluate movement proclivities and direction persistence. Descriptive statistics were calculated for all subjects, and comparisons for sex were tested by box- and density- plots. Track visualisations were produced to compare movement patterns between individuals, and in detail (e.g. kite1 and kite2).
2.5 Behavioural State Classification
The moving steps were classified into behaviours on the basis of the following rules:
Traveling: Large step size and small turning angle.
Feeding/Resting: Short-stepped paths with wide or irregular turning angles.
These states were estimated from distributions of (averaged over time) movement metrics, and state frequencies calculated for each individual. Maps were produced to illustrate the geographical spread of states, and bar graphs to compare state proportions within the study sample.
3. Outcomes
3.1 Individual Movement and Spatial Patterns
This map shows the distribution of GPS locations and emphasizes the variability in travel distance and direction among individuals. Some birds, such as those in green and orange, moved long distances across the landscape, and others stayed concentrated in core areas. The longitude difference in distribution indicates difference in migratory route or feeding behavior.
The plot represents the variability in the movement behavior. Kite1 undertook long distance travel, kite2 short distance movements. Kite3 demonstrated looping and backtracking in one track, which may have represented area-restricted search behaviour associated with foraging.
The straight flight path of Kite1 seems to show a stable preference for direction, which might have been motivated by exploring a new area or by migratory purposes. The high point density of some sections demonstrates relatively slow movement or residency periods.
This figure is a complete overview of every individual ID and its corresponding colour coding on all tracking and home range plots.
This aggregate map contains all plotted tracks of all animals and is superimposed on a map grid. The centre of high line‐intersection intensity indicates congregation zones or communal areas: roosts or high quality foraging sites. This map is important to offer spatial coverage ratio in home range estimation by KDE and MCP which is explained below.
3.2 Home Range Estimation
Spatial occupancy Minimum Convex Polygon (MCP) and Kernel Density Estimation (KDE) methods were adopted to estimate home ranges of individual BK. MCP estimates offered a simple approximate outer bound of movement, but were sensitive to the outlying cases, hence susceptible to over-estimation. Alternatively, KDE permitted detailed examination of use of space, indicating areas of concentrated activity with probability-based contours.
The KDE analysis has been particularly instructive in delineating areas of high activity (50% contours) and more extensive space use (95% contours). Importantly, some individuals had tightly aggregated home ranges and others had very scattered distributions, indicating behavioural differences.
Full-panel KDE heatmaps of all individuals showed high degrees of variance in the utilisation of space. Some individuals exhibited strong site fidelity, while others were characterized by elongated or multiple activity zones which possibly result from different foraging strategies or specific local conditions.
There was considerable individual variation in home range size estimates (95% KDE), but little ascribed to sex differences. Mean areas (and hence variances) were of a similar order between the two groups (circa 337,000 – 353,000 m²) with large standard deviations indicating substantial overlap in space use. This odd sex group had a slightly higher mean, but only two individuals, making their interpretative power limited. These results are consistent with other studies (e.g., Santos et al., 2021) that reported low sexual dimorphism in spatial ecology of Black Kite.
KDE maps (50%, 90% and 95%) suggested that Black Kites targeted their movements to certain high-use areas. The 50% core areas exhibited a clustering pattern, indicating that site fidelity was relatively high, and the 95% contours exhibited wide but distinct home range boundaries. This spatial pattern places emphasis on areas of focused foraging, and indicates that individual kites have an efficient use of resource-rich sites, consistent with central-place foraging.
3.3 Movement Behaviour: Turning Angle and Step Length
This boxplot compares the turning angle in both sexes. The plot is overlaid with some variation between males, females, and unknown samples and indicates some sex differences in flight maneuverability.
Mean step length differs substantially between sexes and of unknown sex, unknown sex individuals having higher mean step length. The findings also illustrate several outliers that might refer to sporadic long-haul flights.
The histogram shows the entire non-normalized histogram of the turning angles with a fitted density curve. The majority of angles are concentrated at 0 radians in which case straight-line movement is likely; sharp micro-turns are also prevalent.
3.4 Movement State Classification: Travelling vs Feeding/Resting
Diagrams movement states by themselves by location. The driving step points are all red (Feeding and Resting) with some in the diets in cyan (Travelling) which indicates a longer time spent stationary and foraging than in transit.
Bar chart of the number of steps in each state. Feeding and resting is the main activity, at more than 7000 steps, with travelling adding about 640.
Horizontal bar chart of individual activity states. The majority of animals spent over 75% of their time feeding/resting with very few showing a higher proportion of time allocated to traveling.
4.Interpretation of Main Findings
Knowledge on the spatial ecology of long-ranging raptors such as Black Kites is highly important for understanding habitat usage,behavioural strategies and also guiding conservation strategy. Here, GPS tracking data from multiple individuals were analysed to estimate home ranges, look at movement statistics and classify behaviour states. Results show both variability at the individual level and more general behavioural patterns, as expected from avian movement ecology.
4.1 Variation in Home Range and Space Use
Our estimates of home range using estimates of MCP and KDE show considerable variation in utilisation space among individuals. Some birds had large, widely distributed home ranges and others had more limited spatial use. This discrepancy can be reconciled with the measurements of Tanferna et al. (2013), who demonstrated that breeding and non-breeding territorial and floaters Black Kites occupy different spatial requirements, with floaters having much larger home ranges than territorial adults. Similarly, Literák et al. (2022) described mean KDE 95% home ranges of 3,000–4,700 km² for Siberian Black Kites implying relatively large space requirements both at breeding and wintering grounds.
These differences could potentially be attributed to sex, age, reproductive status, or the environment (e.g., resource availability). While we did not assess specific age-related differences directly, these results are coherent with those obtained from other studies showing that floaters and subadults have a wider moving range while searching for territories (Tanferna et al., 2013; Literák et al. 2022).
4.2 Movement measures and interpretation of behaviour
Analysis of movement metrics showed that most of the step lengths were short and the turning angles were tightly clustered around 0 radians , indicating persistent movement interspersed by some tight movement bouts. The turning angle distribution is consistent with what has been reported for other soaring raptors, such straight line segments are common during commuting or gliding flight (Santos and Hansell, 1998).
As a corollary, representatives of the sex unknown had a slightly longer mean step length, causing them to span greater distances. This could be suggestive of these birds being floaters or having a non-breeding status as these can show somewhat more exploratory behaviour (Tanferna et al. Moreover, experience a comparable level of turning across sexes indicates that sex alone is not a pronounced factor in determining flight style or spatial strategy during migratory’s flight, supporting the results found by Santos et al. (2021) who found no such differences in turbine avoidance behaviour in migrating Black Kites.
4.3 Dominant Feeding and Resting Behaviour
The behavioral classification showed that most of the recorded steps were feeding and resting states . That birds exhibit strong spatial fidelity to core habitats in our case is consistent with the general life-history strategy of non-breeding or staging individuals, for whom minimizing energy expenditure is critical (Rechetelo et al. In addition, the spatially clustered nature of these states indicates that habitat features (e.g. fixed food sources, or thermals) may drive movement behaviour.
This kind of stronger non-transit behaviour corresponds to other grain-eaters or scavengers, which usually camp in high quality areas as long as environments stays the same (Rechetelo et al., 2016; Kumar et al., 2020). The low fraction of time in “travelling” states reflects either the post-migratory nature of the data or local foraging strategies.
4.4 Ecological and Conservation Implications
This study further demonstrates the value of high resolution GPS tracking for assessing differences in movement between individuals and fine-scalespace use. The high density of rest and feeding sites around core areas emphasizes the need to conserve certain foraging and roosting sites.”
Further, avoidance behaviour of turbines that has been recorded in similar research (Santos et al., 2021) highlights the need to incorporate movement ecology into wind energy planning. Our home-range analyses also suggest that also non-breeding individuals may need large spatial buffers, in line with previous work in Spain and Central Asia (Tanferna et al., 2013; Literák et al., 2022).
These results highlight the functional relationship between resource abundance and the space-use of Black Kites. The inclination to forage near water, and to use steep and/or rugged terrain for nesting, would appear to be adaptive for behaviour reducing human disturbance and nest predation. Analogous behavioural patterns have been observed in other raptor and granivorous species, where the nearness of feeding sources influences home range selection and breeding success considerably (Rechetelo et al., 2016). In the present study, the home range estimates generated using the KDE and MCP models also indicate that Black Kites adjust their use of space by focusing their movements on high-quality areas (with the guarantee of finding food), which supports the central-place foraging nature of the species.
In addition, when broken down by the state of movement into ‘resting’ and ‘travelling,’ it was found that the subjects spend much more time in resting behaviour than in directed movement. This could reflect habitat-faithful home range use or dependence upon established foraging patches, and is consistent with mechanistic home range theory (Ellison et al., 2020). The ambivalent role of landscape elements as habitat components and stimuli to behaviour highlights the multifaceted nature of spatial ecology. These results not only support previous studies in place-based habitat selection in Black Kites (Sergio et al., 2002), but further our knowledge of how patterns of space-use and behaviour can interface with one another to derive ecological strategies.
5. Conservation Perspectives
Our results emphasize the relevance of fine-grain movement ecology for the conservation management of Black Kites (Milvus migrans). The estimated home ranges and movement patterns showed evidence of the birds’ non-random resource-driven use of landscape, suggesting that the spatial arrangement and composition of habitat are of fundamental importance for maintaining viable populations.
Successful conservation initiatives should ensure that iconic habitats that are frequently used for key activities, such as forage and rest, are conserved. As shown by Sergio et al., 2002) and Black Kites prefer water to nest and feed in as well as to complex relief surfaces with low human influence. These preferences mean that conservation planning should consider the implementation or protection of buffer strips in the surrounding of wetlands, particularly within 1–2 km, because this distance has been record-breaking to be home adequate for both high foraging activity and high breeding performance.
Additionally, overlapping of core areas of activity among individual bears, as found in this study, is consistent with the results of Rechetelo et al. (2016)), which have also showed similar effects in movement patterns as a result of patchy resource distribution. These overlaps demonstrate that landscapes should be managed as functional ecosystems rather than isolated habitat fragments. The application of radio telemetry and KDE analyses shed light on spatiotemporal use, helping conservationists to model how habitat modification will affect species distribution and behaviour.
Behavioural flexibility should also be considered in conservation. Ellison et al. (2020) reported that even non-territorial passerines displayed structured home range use, mediated by spatial memory and landscape properties. For raptors such as Black Kites, the inclusion of behaviourally informed models into larger landscape planning can help adjust the location of reserves and corridors so that birds have access to appropriate habitats without being hemmed in by human-altered boundaries.
In the light of these findings conservation implications for the black kite should include:
Preservation of a mosaic of foraging habitats (e.g., open meadows, water edges) near nesting sites;
Preventing infrastructure (e.g., roads, recreation sites) from being placed in the vicinity of nesting cliffs or core home ranges;
Applying GPS telemetry and remote sensing to assess significant, long-term changes in the use of habitat;
Encouraging agri-environment schemes conducive to extensive grassland cover, with them being positively selected for both foraging and roosting (Sergio et al., 2002).
The knowledge of the spatial ecology of the Black Kite contributes to evidence-based conservation actions. Conservation of both structural habitat diversity and connectivity among the most important resource patches will therefore be crucial for the long-term survival of this species in human-modified landscapes.
- References
Börger, L., Franconi, N., De Michele, G., Gantz, A., Meschi, F., Manica, A., Lovari, S. and Coulson, T., 2006. Effects of sampling regime on the mean and variance of home range size estimates. Journal of Animal Ecology, 75(6), pp.1393–1405. https://doi.org/10.1111/j.1365-2656.2006.01164.x
Ellison, N., Hatchwell, B.J., Biddiscombe, S.J., Napper, C.J. and Potts, J.R., 2020. Mechanistic home range analysis reveals drivers of space use patterns for a non-territorial passerine. Journal of Animal Ecology, 89(11), pp.2642–2653. https://doi.org/10.1111/1365-2656.13292
Kumar, R., Araujo, J., Kumar, A., Khan, J.A., Bensch, S. and Wikelski, M., 2020. GPS-telemetry unveils the regular high-elevation crossing of the Himalayas by a migratory raptor: implications for definition of a Central Asian Flyway. Scientific Reports, 10, p.15988. https://doi.org/10.1038/s41598-020-72970-z
Literák, I., Reif, J., Procházka, P., El-Arabany, N., Al-Shamlih, M., Obuch, J., Šenko, D., Bandžuch, P. and Adamík, P., 2022. Black Kites on a flyway between Western Siberia and the Indian Subcontinent. Scientific Reports, 12, p.5581. https://doi.org/10.1038/s41598-022-09246-1
Rechetelo, J., Grice, A., Reside, A.E., Hardesty, B.D. and Moloney, J., 2016. Movement patterns, home range size and habitat selection of an endangered resource tracking species, the black-throated finch (Poephila cincta cincta). PLOS ONE, 11(11), p.e0167254. https://doi.org/10.1371/journal.pone.0167254
Santos, C.D., Ferraz, R., Muñoz, A.R., Onrubia, A. and Wikelski, M., 2021. Black kites of different age and sex show similar avoidance responses to wind turbines during migration. Royal Society Open Science, 8(1), p.201933. https://doi.org/10.1098/rsos.201933
Santos, C.D., Muñoz, A.R., Onrubia, A., Cortés-Avizanda, A. and Wikelski, M., 2017. Match between soaring modes of black kites and the fine-scale distribution of updrafts. Scientific Reports, 7, p.6421. https://doi.org/10.1038/s41598-017-05319-8
Sergio, F., Pedrini, P. and Marchesi, L., 2002. Adaptive selection of foraging and nesting habitat by black kites (Milvus migrans) and its implications for conservation: a multi-scale approach. Biological Conservation, 104(3), pp.351–364. https://doi.org/10.1016/S0006-3207(02)00332-4
Tanferna, A., Sergio, F., Blanco, G., Tavecchia, G. and Hiraldo, F., 2013. Habitat selection by Black Kite breeders and floaters: Implications for conservation management of raptor floaters. Biological Conservation, 160, pp.1–9. https://doi.org/10.1016/j.biocon.2012.12.028
Tonra, C.M., Reiley, B.M., Cooper, N.W. and Marra, P.P., 2019. Remote estimation of overwintering home ranges in an elusive, migratory nocturnal bird. Ecology and Evolution, 9(22), pp.12586–12597. https://doi.org/10.1002/ece3.5679