Next Article in Journal
Measuring and Modeling Mechanical Ventilation for Long-Term Environmental Monitoring in Large Commercial Laying Hen House
Previous Article in Journal
The Vaginal Microbiome of Mares on the Post-Foaling Day Under Field Conditions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Animals
Volume 14
Issue 22
10.3390/ani14223338
animals-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Spatial Ecology of a Resident Avian Predator During the Non-Breeding Period in Managed Habitats of Southeastern Europe

by
Draženko Z. Rajković
1,*,
Daliborka Stanković
1,
Jelena Šeat
2,
Dejan S. Stevanović
3,
Miona V. Andrejević Stošović
3 and
Stefan Skorić
1
1
Department of Biology and Inland Waters Protection, Institute for Multidisciplinary Research, University of Belgrade, Kneza Višeslava 1, 11030 Belgrade, Serbia
2
MTA-SZTE ‘Momentum’ Applied Ecology Research Group, University of Szeged, Közép Fasor 52, H-6726 Szeged, Hungary
3
Faculty of Electronic Engineering, University of Niš, Aleksandra Medvedeva 14, 18000 Niš, Serbia
*
Author to whom correspondence should be addressed.
Animals 2024, 14(22), 3338; https://doi.org/10.3390/ani14223338
Submission received: 3 November 2024 / Revised: 16 November 2024 / Accepted: 17 November 2024 / Published: 20 November 2024
(This article belongs to the Section Ecology and Conservation)
Browse Figures
Versions Notes

Simple Summary

Using GPS technology, we investigated the spatial ecology of seven male tawny owls in the heterogeneous landscape of Western Serbia during the non-breeding season. Our main goals were to measure home range and core area sizes, understand their habitat choice, and describe roost site characteristics. We caught the owl males and fitted them with lightweight transmitters, which allowed us to track their movements. We found that home ranges varied significantly, with an average size of around 2.2 km2. Males with larger body mass exhibited smaller home ranges. Also, we observed minimal overlap in space use between individuals, indicating quite strong territoriality. In terms of habitat selection, increased area of cultivated land substantially reduced the probability of tawny owl presence, while suburban, built-up areas increased it. Our findings provide a better understanding of the home range and habitat selection of tawny owls on the central Balkan Peninsula. This knowledge could have a considerable implication for land-use practices and effective conservation strategies.

Abstract

Describing home range and resource selection is crucial for understanding ecological needs and creating conservation programs. Still, our knowledge of spatial and behavioural ecology for most species remains limited. Here, we used satellite transmitters to investigate core and home range sizes, habitat selection, and roost characteristics in seven tawny owl males in Western Serbia during the autumn–winter period 2023. Using minimum convex polygon (MCP) and autocorrelated kernel density estimation (AKDE), we found clear variability in core area and home range sizes. Also, adult and heavier males have smaller core area and home ranges than juvenile and lightweight individuals. The Bhattacharyya coefficient showed minor home range overlap in tagged males. The final model for evaluation of habitat selection suggests that the likelihood of owl occurrence was positively correlated with the share of anthropogenic infrastructure and negatively associated with the increase in the proportion of cultivated land within the home range. However, scores of model performance metrics showed moderate predictive accuracy, implying that other unmeasured variables may dictate species presence. Our study illustrates the ecological plasticity and ability of the tawny owl to adapt to a human-modified environment while providing new information about the spatial ecology of this widespread predator in Europe.

1. Introduction

All vertebrates need a specific space where they perform essential activities critical for their survival and reproduction [1]. This space typically comprises both the area and habitat they occupy, including the biotic and abiotic components of their distribution range [2,3]. Understanding how animals utilise spatial and temporal resources remains a fundamental goal in spatial and behavioural ecology [4,5]. Knowledge of animal movements and habitat requirements throughout the life span are crucial metrics to understand ecological dynamics and gene flow [6]. These parameters are also critical for developing effective population management and conservation strategies [7,8], especially as most vertebrate species inhabit landscapes significantly altered and shaped by various human activities [9,10].
Animals typically confine their activities to an explicit area described as a home range [11]. The boundaries of this space include the places where they regularly live and gather vital resources for survival, excluding periods of migration, wandering, or unusual movements [11,12,13]. It is important to note that the concept of a home range is distinct from territory, which is a smaller, actively defended space against conspecific or interspecific competitors [11,14]. A correct home range estimate is essential to define habitat use, resource selection, and movement pattern analyses [15,16]. Additionally, estimating home range enables stakeholders and policymakers to formally preserve the critical habitats frequently utilised by species of conservation interest [17,18]. Factors like prey distribution [19], habitat quality [20], competition [21], and physiological needs [22] usually influence the size, shape, and spatial pattern of a home range.
Habitat selection, on the other hand, refers to how animals interact with the various components, conditions, and resources within their environment [23]. Clearly, a particular habitat type may serve for foraging, nesting, roosting, migration, or other life-history traits, reflecting hunting strategies [24], breeding requirements [25], or avoidance of human disturbance [26]. Roosting sites are an integral part of habitat use because they directly influence birds’ spatial and temporal distribution within their home range. Safe and high-quality roosting sites contribute significantly to a bird’s overall fitness and survival [27], affecting their daily and seasonal movement patterns [28]. For example, a secure roosting site can reduce predation risk, allowing birds to spend more time foraging and enhancing their energy balance and reproductive success [15].
Studies regarding home range and habitat selection have a long tradition in ornithological science, e.g., [29,30,31,32]. Over the past few decades, with the significant advancement of sophisticated technologies, the increasing availability of diverse, lightweight tracking devices, and the proliferation of novel statistical techniques and tools, the number of studies on this topic has been increasing rapidly [33,34,35,36]. However, numerous factors, such as easier access to bird individuals (e.g., known nest locations), higher detectability, and research on the relationship between productivity and prey, contribute to more studies of avian home ranges and habitat use during the breeding compared to the non-breeding period. For instance, by May 2024, around 68,600 articles on breeding and 17,900 on non-breeding birds related to home range and habitat use were listed in the Google Scholar repository (https://scholar.google.com/). Furthermore, the majority of avian ecology research is conducted in economically developed countries, frequently focusing on colonial species, diurnal raptors, or species of high conservation interest, e.g., [37,38,39,40]. These geographical and taxonomic biases are probably driven by better funding, infrastructure, and access to advanced research technologies in developed regions. In contrast, nocturnal birds are relatively understudied, especially those in remote areas and economically less developed parts of the world. As a result, there is a considerable gap in our understanding of the spatial ecology of local populations of crepuscular and nocturnal birds like owls (Strigiformes).
The tawny owl (Strix aluco; Strigiformes: Strigidae) is a common avian predator widely distributed across the Palearctic region with 11 distinct subspecies [41,42]. Two among them, S.a. sylvatica and S.a. aluco, inhabit Europe [42,43]. It is the most common and widespread nocturnal raptor on the European continent [41], with a large, stable population estimated to comprise 632–932 hundred thousand pairs [44]. This primarily nocturnal owl is a non-migratory, sedentary species that predominantly breeds in tree holes, cliffs, and buildings or similar cavities [41,42]. Tawny owls inhabit lowland and hilly areas with a preference for broadleaved and mixed forests; however, local populations successfully adapted to less favourable habitats such as city parks, cemeteries, spacious gardens, orchards and tiny forest patches, as well as alpine coniferous forests and rocky outcrops up to 2800 m [41,42]. Southeastern Europe, particularly the Balkan Peninsula, serves as a regional stronghold for tawny owls, accounting for approximately 12.5% of the continent’s population [44]. In addition, in the biogeographical context, the Balkan Peninsula played a crucial role as a Pleistocene refugium of the species on a European scale [45].
In Europe, the tawny owl is an extensively researched owl species [46]. Most earlier works are devoted to its diet [47], breeding biology [48,49,50,51], distribution and vocal activity [52,53,54,55], and habitat relationship [56,57]. Spatial ecology has also been studied but exclusively in developed regions of Western and Northern Europe [50,58,59,60,61]. This research interest is likely due to its widespread distribution, relatively large population, presence in various habitats, and willingness to occupy artificial nest boxes, e.g., [43,48]. On the other hand, there is still much to learn about the biology and ecology of tawny owls, particularly in remote regions in Eastern and Southeastern Europe. Furthermore, earlier studies were limited to an almost entirely different matrix of habitats and climate conditions that differ significantly from those present in Eastern and Southeastern Europe.
The primary aim of our study is to investigate the spatial ecology of tawny owls (subspecies S.a. aluco), in particular, to examine the home range, habitat use, and roosting site characteristics in heterogeneous landscapes of central Serbia during the non-breeding period using highly accurate VHF-GPS transmitters. We hypothesise that home range and habitat use differ among the tagged individuals. We assume that males in fragmented and human-altered lowland landscapes would move more between suitable habitat patches in search for food and other resources compared to males inhabiting hilly, far-from-settlements woodland. In short, we expect that “lowland” males have more extensive home ranges than those in hilly areas. Also, based on earlier references, e.g., [41,42,43], we expect that tagged adults would predominantly exploit woodland habitats and juvenile, less experienced males would tend to move more, i.e., have more extensive home ranges and core areas than adults.

2. Materials and Methods

2.1. Study Area

Our study took place in part of the Podrinje area (Jadar and Radjevina microregions), covering approximately 950 km2 in Western Serbia (Balkan Peninsula, SE Europe). The investigated plot, c. 40 km2 in size (approx. centroid N44.4964°, E19.3957°), is located 12 km southeast of Loznica town (Figure 1). The terrain altitude ranges from 120 to 500 m above the sea level, creating a diverse vegetation cover in a mosaic landscape. Overall, the northern section of the study area is a lowland (river valley), while hilly terrain characterises the southern section. In the past, this region was completely covered by deciduous forest. However, centuries of human activities, including logging, land cultivation, and grazing, have significantly altered the forest’s uniformity and age. Nowadays, the northern part of the study area is characterised by typical farmland, with cultivated fields commonly bordered and interspaced with hedgerows, forest patches, small villages along local roads, and several minor rivers and streams. In contrast, the southern section of the study area is predominantly covered by natural and semi-natural broadleaf forests with prevalent species like oak (Quercus sp.), linden (Tilia sp.), and hornbeam (Carpinus sp.) in the lower zones and European beech (Fagus sylvatica) in the higher zones, dotted with small watercourses, unpaved roads, and clearings. The human population density in the Podrinje area is about 79 people per km2, with an average settlement density of 9 per 100 km2 [62]. Although cambisols are the dominant type, the soil composition shifts with altitude change. The climate is temperate continental, characterised by mild summers and cold winters. Meteorological data show an average annual air temperature of 11.7 °C, with precipitation of 858 mm and an average of 2041 h of effective sunlight per year [63].

2.2. Capturing, Tagging, and Data Collection

Prior to capturing, we surveyed the study area for tawny owl territories in mid-October 2023. Except for the far north of Europe, this period of a year provides adequate vocal responses from territorial tawny owls, as juvenile owls establish territories and adults defend theirs in preparation for the spring nesting season [52,53,54]. Since the tawny owl’s acoustic activity is notably reduced in windy [55] and rainy weather [64], all field surveys were conducted on calm, dry evenings. The survey methodology followed the call-broadcast techniques demonstrated by [65]. The positions of calling owls were plotted on a Google Maps app (https://play.google.com/store/apps/details?id=com.google.android.apps.maps&pli=1, accessed on 25 August 2024) with accompanying data like type of response and sex.
We used the gathered data on the spatial distribution of eleven discovered territories, defended by a pair or a male, to identify suitable microlocations (e.g., small clearings, prominent ridges, or crossroads) for capturing. A vertical ornithological mist net (Ecotone, Gdynia, Poland) with five shelves and a 70 mm mesh size (3.2 × 18 m) was used to capture the owls. To attract the owls to the mist net, we used the fact that tawny owls fiercely defend their territory against conspecific intruders [49,66]. Thus, we used a decoy, a handmade, black-painted Styrofoam owl model placed at the midpoint of the mist net, accompanied by the broadcast of the male’s territorial hooting from a loudspeaker placed underneath. We set the mist net in the evening and checked it every 20–30 min. All seven owls in this study were captured from late October to early November 2023. Before tagging the males, all individuals were aged, sexed, and ringed and morphological traits were measured. Also, the body mass of each individual was recorded (Table 1) using a spring scale with a 1000 g capacity and an accuracy of ±2 g (PESOLA, Chur, Switzerland). Owls were ringed with distinctively coded aluminium butt-end rings (Natural History Museum, Belgrade, Serbia). Age and sex were determined using the moult stage, morphological measurements [67] and (D.Z. Rajković, unpublished data), and voice expressions in the field [58] during the capturing event. Finally, we equipped captured individuals with lightweight PinPoint VHF-GPS transmitters (models 120 and 240; 5.3 and 8 g; Lotek Wireless Inc., Wareham, UK) with a specimen-unique frequency. The transmitters were X-attached to the birds in a thoracic configuration beneath the plumage by stitching and adhering a 6.3 mm wide Teflon ribbon harness to the transmitter [68,69] (Figure 2).
Single transmitter attachment took 21 to 32 min, respectively. Transmitters were cryptically coloured to prevent potential predation [70]. After tagging and a few minutes of habituation, all trapped individuals were released close to their capture locations.
Although there is no sufficient evidence that proportionally heavier devices possessed more significant consequences on the tagged specimen, e.g., [71], as a rule of thumb, the weight of our transmitter was considerably lower than the typically recommended 3% of the birds’ weight (1–1.6%, respectively) in order to minimise the possible risk of adverse tag effects [72,73]. Furthermore, Sunde [74] reported no negative effects associated with transmitter attachment and tawny owl survival and reproduction parameters.
The VHF-GPS transmitters were pre-programmed to provide timestamps and to record longitude and latitude five times per day: one fix during daylight (midday) and four fixes from dusk till dawn (6:00 p.m., 10:00 p.m., 2:00 a.m., 6:00 a.m.). This schedule was designed to ensure the record of daytime roosting sites and nighttime movements for home range calculations. The data collected and stored by the internal transmitter memory were downloaded remotely using a hand-held 3-element Yagi antenna and a portable PinPoint VHF receiver (Lotek Wireless Inc., Wareham, UK). For further computer analysis, we transformed all collected locations into a rectangular coordinate system-UTM coordinates (WGS 84, EPSG: 32,630).

2.3. Estimation of Home Ranges and Spatial Overlap

In the initial step, the collected data were manually sorted and filtered in Microsoft Excel and visually inspected in Google Earth Pro. We filtered collected movement data in order to exclude all GPS fixes of limited accuracy (less than four satellites) [75,76]. Also, we discarded the fixes from the first night because of the possible influence of the freshly mounted transmitters in the case capturing and handling stress interrupted owls’ normal behaviour and activity patterns. We used a probability utilisation distribution (UD) to estimate the home range and core area. We estimated the UD by employing two different techniques: Minimum Convex Polygon (MCP) [77,78] for 95% isopleth and the Autocorrelated Kernel Density Estimation (AKDE) [79] for 95% (hereafter home range) and 50% isopleth (hereafter core area). However, we only report MCP to enable comparisons with previously published data; this geometric estimator is sensitive to outliers and marginal fixes and tends to underestimate home range areas on autocorrelated data [80,81]. Considering the existing flaws of MCP, we perform AKDE to provide additional information on this matter. This technique represents a relatively novel statistical concept used in ecology, primarily describing the likelihood that an animal can be found at a particular location or area over a given time span [82,83]. We preferred AKDE over other frequently used estimators in ecological studies because it mitigates different biases that can heavily affect home range estimates, see [76,84,85,86]. In brief, AKDE improves upon traditional kernel density estimation (KDE) methods by factoring in the time dependency and autocorrelation of the animal movement data. Also, this method reduces the bias and variance in home range estimates that can be significant in traditional KDE when used with autocorrelated data [84,87].
In order to reproduce the home range, core area, and overlap analysis employing the AKDE method, we used R software version 4.3.2 [88] enhanced by a robust framework of continuous-time stochastic movement models, ctmm web app (https://ctmm.shinyapps.io/ctmmweb/) (accessed on 25 August 2024) [89,90]. After importing the dataset, we employed the integrated ‘Filter Outliers’ tool to mitigate the impact of potential data skewness that could skew the analyses. Further, we assessed the autocorrelation structure of a tracking dataset and temporal stability (range residency) of the tawny owls’ home ranges through variogram analysis [86]. Model selection was performed using a maximum likelihood approach, comparing the suitability of several offered stochastic movement models: Brownian Motion (BM), Integrated Ornstein–Uhlenbeck (IOU), and Ornstein–Uhlenbeck Foraging (OUF). The most supported movement model was selected based on the lowest Akaike’s Information Criterion value [91,92]. As the GPS data collection schedule was slightly altered due to habitat-related signal loss, we used the weighted AKDE, in particular, the ‘Optimal Weighting’ tool [79,86]. This choice allowed us to recalibrate the weightings of the data points, thereby correcting for any inconsistencies introduced by the changes in the fixed collection protocol [79]. Lately, we quantitatively defined and visualised the home ranges using the AKDE approach, mapping the 95% and 50% confidence intervals of the home range contours. The movement distance of tawny owl males was calculated using the distHaversine function in the geosphere package [93].
To quantify the extent of spatial overlap among tagged owl males, we applied a robust, pairwise metric, the Bhattacharyya coefficient (BC) [94,95], also in the ctmm interface. The data for each of the seven owl males were organised into a paired matrix to form dyads (N = 21), which were subsequently utilised to determine home range overlap. A dyad refers to a pair of specific individuals, regardless of whether their home ranges showed any spatial overlap [96]. The BC estimates range from 0 (indicating no overlap) to 1 (complete overlap).

2.4. Defining Habitat Selection and Roosting Sites

After collecting and processing GPS location fixes, we quantified habitat selection by tawny owl males at the third-order selection as proposed by Johnson [97]. To delineate habitat classes within each home range, we employed a combination of field surveys, Google Earth Pro, and QGIS version 3.14 (http://test.qgis.org/html/en/site/index.html) (accessed on 27 May 2024). For simplicity and ecological relevance, we grouped habitat into three main categories: forest, cultivated land, and suburban areas. Forested and suburban areas with buildings provide shelter, roosting sites, foraging, resting, and perching opportunities. At the same time, open habitats such as cultivated fields together with grasslands support large populations of rodents and other small mammals, the tawny owl’s preferred prey in Europe [41,42,43,47]. In addition to environmental, we also included altitude as a topographic variable.
In the next step, we applied a logistic resource selection function (RSF) [98] using a generalised linear model (GLM) with logistic regression. The RSF compares environmental and other conditions at animal-occupied sites with randomly selected available sites, which are the available resources within the study area. This use-availability method uses a weighted distribution to estimate the likelihood that an animal, in our case owl, will select a specific resource based on environmental characteristics [98]. As the RFS model needs used and available points, we generated available points in QGIS randomly spaced within home range borders. We generated the same number of available points as the used points.
Before GLM modelling, we calculated the Spearman’s correlation matrix to screen for potential high collinearity (r > 0.7) among the four predictor variables. Only forest and cultivated land showed a high correlation (Spearman’s r = −0.86). As a result, we excluded the forest variable from the model-building process due to its weaker bivariate association with the dependent variable (r = 0.15 for forest, r = −0.20 for cultivated land). Also, in order to further quantify the degree of multicollinearity, we checked VIF values, which confirmed no significant multicollinearity problem (VIF = 1.12–1.34). Subsequently, we developed all possible combination models (N = 7) based on single, two, or three retained predictors (cultivated land, suburban, and altitude). To choose the best model, we used a widely applied Akaike information criterion (AIC) [91]. We addressed the model selection uncertainty by Akaike weight, which can be explained as the probability of the model being the best in the set of candidate models [91].
Model accuracy was tested using the AUC statistic [99], which evaluates the model’s ability to distinguish used from available locations. AUC values range from 0.5 (random discrimination) to 1.0 (perfect prediction). In order to avoid overfitting and calculate predictive accuracy, we conducted a ten-fold cross-validation with the DAAG package [100], dividing the dataset into ten random subsets, each used once for testing while the other nine were used to fit the model. Roosting site height (m) was measured directly in the field with a laser distance measurer (model PARKSIDE PLEM 50 C3, Parkside, Neckarsulm, Germany; accuracy 1.5 mm) to the nearest 10 cm.

2.5. Additional Statistical Testing

In addition to the statistical procedures described in the previous sections, we employed descriptive statistics (measures of central tendency) and several standard statistical tests to assess correlations and differences and to summarise our findings. Due to their robustness, we used Welch’s t-test, Kruskal–Wallis H test and Fisher’s Exact Test with the Haldane–Anscombe correction. In particular, Fisher’s Exact Test was used to examine the association between habitat (hilly vs. lowland) and roost location type (natural vs. anthropogenic). All mentioned tests are useful when dealing with small sample sizes or when some cells in a contingency table have low or zero counts, as in our study. Mean values are presented with their corresponding standard deviations for all numerical data. Unless otherwise specified, the cut-off point implying statistical significance was p < 0.05. All performed statistical tests were two-tailed and were conducted using R [88].

3. Results

3.1. Home Range and Core Area Estimates

The present study tracked seven GPS-tagged tawny owl males between 35 and 58 days (mean = 47; SD = 3.4). Altogether, 1440 refined, effective locations were obtained (mean = 205.7; SD = 19.6; range = 144–262). The tracking duration (in days) did not show a significant correlation with the size of the core area (log-transformed; Spearman’s correlation r = −0.18, df = 5, p = 0.7) or home range (Spearman’s correlation r = 0, df = 5, p = 1). Also, there was no correlation between the number of GPS fixes per owl male and the estimated home range size (Spearman’s correlation r = −0.11, df = 5, p = 0.82).
The home ranges of the tracked males varied significantly in terms of size and shape (Table 2, Figure 3). The mean home range per male individual was 157.04 ha (SD = 94.69; range = 38.6–285.65) for 95% MCP isopleths. Moreover, the mean size of the home range (95% AKDE) was 217.78 ha (SD = 145.95; range = 56.36–462.23), which is equivalent to approximately a 0.8 km radius around the core area centroid. Range sizes for core area (AKDE) were from 9.95 to 89.74 ha with a mean of 40.43 ha (SD = 31.74; Table 2).
A solid and highly significant relationship exists between core area and home range size (Spearman’s correlation r = 0.96, df = 5, p < 0.001). Specifically, a larger core area is strongly associated with a more extensive home range. Additionally, there is a strong negative relationship between individual body mass with home range (Spearman’s correlation r = −1, df = 5, p < 0.001) and core area size (Spearman’s correlation r = −0.96, df = 5, p = 0.001).
The results of the t-test fail to reject the null hypothesis and indicate an absence of a significant difference between lowland and hilly males in home range size (t = −0.06, df = 4.82, p = 0.95, 95%CI = [−342.10, 326.23]) or core area size (t = −0.06, df = 3.84, p = 0.95, 95%CI = [−85.64, 81.87]). The testing was supported by minor and non-significant effect sizes for both the core area and home range (Hedges’ g = −0.04), suggesting negligible similarity in mean values between the two relief classes.
Further analysis also revealed no significant difference in means between juvenile and adult individuals for home range (t = 2.03, df = 4.46, p = 0.11, 95%CI = [−58.12, 431.32]) and core area (t = 2.54, df = 3.28, p = 0.08, 95%CI = [−8.76, 98.06]). Despite the lack of statistical significance, relatively high values of effect size (Hedges’ g = 1.18 and g = 1.4) and box plot inspection suggest differences in age classes and the superiority of juvenile individuals in home range and core area sizes (Figure 4).
Based on the collected timestamps from transmitters, we found that the average Euclidean movement distance per night for a single male owl was 1088.7 m (SD = 618; median = 1034; range = 45–3082). A strong, positive correlation was detected between the mean night travel distance per individual and the size of the home range (Spearman’s correlation r = 0.82, df = 5, p = 0.02). However, there was no significant difference in the travelled distance between owl males in lowland and hilly areas (t = 1.04, df = 298, p = 0.3, 95% CI = [−0.04, 0.12]). On the other hand, on average, juveniles tended to move more during a single night compared to adults, and this difference was highly significant (t = −5.99, df = 238.3, p < 0.0001, 95% CI = [−0.29, −0.15]). Across the four buffer zones (0–250, 251–500, 501–1000, >1000 m), nearly 60% of movements were within a 250 m radius measured from the core area centre. The results of the Kruskal–Wallis H test showed a statistically significant difference in median distance movement from the core area centre across the tagged males (χ2 = 293.44, df = 6, p < 0.0001).
The analysis of the BC overlap index revealed varying degrees of spatial overlap between the home ranges, with overall minimal intersection observed between neighbouring males. The BC values ranged from 0 to 0.45, with the strongest overlap noticed between males coded 56978 and 56979 (Table 3). In general, juvenile males and those which inhabit hilly areas tend to have more spatial overlap than adult individuals and those from the lowland part of the study area. However, the average overlap BC index was just 0.06 (median = 0; SD = 0.15), representing a minor level of home range sharing among the studied individuals (Table 3).

3.2. Habitat Selection

According to GLM modelling, Model number 4, which included the variables Cultivated land and Suburban, offered the most proper balance between model complexity and goodness-of-fit. Therefore, it was selected as the best-performing model (Table 4 and Table 5).
Nonetheless, the model rankings based on AIC values offer some information about the relative significance of predictors. For instance, Model 7, which ranked second, suggests that topography, combined with other habitat features, may provide valuable information about the potential presence of tawny owl individuals. Notably, cultivated land represents a consistent predictor variable among the generated models with the lowest AIC and ΔiAIC values.
As expected, the likelihood of tawny owls choosing a particular habitat during their nocturnal activities was strongly linked to the amount of agricultural land in the area, with less cultivated land inside the home range being preferred. Despite this, the AUC score (0.601) for the best model indicated only moderate predictive accuracy. Additionally, McFadden’s pseudo-R-squared of 0.043 indicates that the model explains only a tiny portion of the data variance, implying that other unmeasured predictors may significantly influence species presence within the home range boundaries.

3.3. Roosts

During our research, we discovered 15 different roosting locations used by tagged tawny owls. Three individuals consistently used the same roost during the entire study period. The remaining four males moved between roosting sites 26–297 m apart (mean = 174.4; median = 200; SD = 93.8). On average, each male had 2.1 roosting spots (median = 2; SD = 1.2; range = 1–4). Seven roosting sites were situated in holes in tree trunks (46.6%), four inside buildings (26.7%) and four were positioned outside (26.7%) on branches close to the main trunk. With a single exception, all roosts were situated within the core area boundaries. The distances between the nearest neighbouring roosts were closer in males inhabiting higher altitudes (mean = 485; median = 485; SD = 102.8; range = 396–574 m) compared to neighbouring roosts of lowland males (mean = 1399.7; median = 552; SD = 1468.2; range = 552–3095 m).
The mean height of the roosting places from the ground was 10.5 m (median = 10; SD = 3.9; range = 2–17). In the lowlands, the mean height of the roosting place was 7.6 m (median = 8.5; SD = 2.7; range = 2–10), while in a hilly landscape, it was 13.7 m (median = 14; SD = 2; range = 11–17). Statistically, tracked males in the lowlands tend to roost at lower heights compared to those in the hilly landscape (t = 5.06, df = 12.7, p < 0.001, 95%CI = [3.48, 8.69]; Hedges’ g = 2.41). Conversely, there was no significant difference in roosting height between the two age classes (t = 1.43, df = 8.56, p = 0.190, 95%CI = [−1.73, 7.47]). The effect size was moderate (Hedges’ g = 0.73), suggesting a modest difference between the sample means.
The analysis of Fisher’s Exact test yielded a p-value of 0.080, showing that the association between these variables was insignificant. The odds ratio was estimated as infinite, with a 95% confidence interval ranging from 0.810 to infinity, which probably reflects the complete absence of anthropogenic origin roost sites in hilly areas. In sum, although the results approached very close to statistical significance, the data do not provide sufficiently strong evidence of an association between analysed variables.

4. Discussion

Here, we present the results of the first study on the home range and habitat selection of tawny owls in the region of Southeastern Europe. We found considerable variation in home range and core area sizes among tagged males of tawny owl in the mosaic landscape of Western Serbia during the non-breeding period. This variability suggests that the owls, on an individual level, exhibit a high level of adaptability and flexibility in habitat utilisation. Also, we did not notice statistical significance in the size of the core area and home range between individuals inhabiting lowland or hilly regions, which indicates that certain topographical variability within the study area does not influence spatial use patterns. A possible explanation is that both lowland and hilly environments in our study area offer comparable resources (e.g., prey availability). Even though there is no statistically significant difference in home range size between juvenile and adult individuals, box plots of AKDE values show a biologically meaningful discrepancy where juveniles tend to have a more extensive home range than adults (see Figure 4). This inconsistency between the visual representation and statistical results is likely due to the small sample size [101], which reduces the statistical power for recognising a natural variation in home range sizes.
This assumption is also based on the more extensive movement of juvenile males during the average night compared to adults. Such conviction aligns with the hypothesis that juvenile owls may require larger ranges to secure sufficient resources and energetic needs, as they are still establishing their territories and developing foraging skills [102,103]. In addition, such behaviour might be a fundamental mechanism for reducing competition with adults, which, in the end, could influence population dynamics.
Previous works on avian predators demonstrated a significant link between home range size and extrinsic factors like habitat quality and landscape heterogeneity [7,19,20,104,105]. In our study, home range size may be associated with habitat quality, as avian predators in poorer-quality habitats may need to travel further to meet their needs, and vice versa [59,102]. From a predator’s perspective, key components of habitat quality include prey density, abundance, availability, spatial distribution, suitable resting sites, low predator pressure, social structure, territoriality, or a combination of these components [15]. Given that our study area provides a sufficient number of roosting places and that significant predators are completely absent (apart from the Northern Goshawk Accipiter gentilis), the factors that presumably contribute to the variability in home range size are prey availability and territoriality. Indeed, owls’ home ranges barely overlap, particularly among adults, suggesting assertive territoriality behaviour in our study area. Thus, territory guarding and defence could partially explain home range patterns. This is consistent with earlier studies of tawny owls, which have proved a high level of territoriality and minimal home range overlap as part of a strategy to reduce competition [58,59,60]. Interestingly, juveniles and individuals in hilly areas showed slightly more spatial overlap than their adult and lowland counterparts, indicating that territorial boundaries may be more fluid among young, less experienced individuals or in more rugged terrains.
Despite variability and different home range estimation methods, our home range sizes are in accordance with those reported in earlier studies, where ranges vary from 16 to 545 ha [50,59,60,106,107]. Also, a positive correlation between core area and home range size reaffirms the hypothesis that the size of the core area is a pretty reliable predictor of overall home range proportions in avian predators [59,94]. Additionally, it may also reflect the influence of habitat suitability and heterogeneity inside home range boundaries, as these factors can significantly shape the spatial use patterns. Based on the present study and considering that home range size stays relatively consistent throughout the year [60,61], we can conclude that in central Balkans, home ranges are likely to range between 150 and 220 hectares, regardless of habitat quality. This information may be essential for refining regional population size estimates, improving the interpretation of nocturnal survey results and supporting better habitat management and conservation planning.
The negative bivariate relationship between body mass and home range size is particularly noteworthy. This finding could reflect compensatory behaviour, where individuals with lower body mass may need to cover more ground to meet their energy requirements, a pattern already documented in tawny owls [58,108]. Simply, larger home ranges result in greater energy expenditure during foraging and movement. This increased effort could then lead to reduced body mass, especially if prey or other critical resources are limited.
Based upon current knowledge, the tawny owl prefers various types of forest habitats including small patches surrounded by open areas [41,42,43,49,58]. Within forest ecosystems, it primarily favours deciduous and mixed woodlands at lower altitudes [41,42,43,109]. Also, a few studies have shown that altitude and topography are significant determinants of distribution and habitat usage in regions with complex topography, e.g., [109]. However, in the present study, we achieved slightly different results. First, it seems that extensively forested areas did not play such a significant role in meeting the habitat demands of the central Balkan population of tawny owls, at least during the non-breeding period. Second, as for home range size, we found that altitude is not a reliable predictor of owl presence in the studied area. Third, and likely most importantly, we discovered that cultivated land and suburban areas appeared to be the most influential predictors at the third order of selection (Table 5). Altogether, our results suggest that cultivated land surfaces strongly decrease while suburban infrastructure modestly increases the likelihood of tawny owl presence in mosaic landscapes of the central Balkan Peninsula. This may be a consequence of recent agricultural intensification in the region [110], which has led to the removal of isolated trees and a reduction in the surface of hedgerows and treeline rows. Also, these findings support the idea that other habitats and their fragments can be important factors for tawny owl survival, compensating for the absence of extensive forest cover. Several previous studies have also highlighted the ecological plasticity of tawny owls in adapting to human-modified environments and built-up areas [43,51,56]. Consequently, this could reflect regional differences in habitat structure and resource distribution, where small patches, fragmented habitats, and edge effect offset the reduction in primary habitat cover, i.e., woodlands, as concluded in a few other studies [57,108]. However, the interpretation of the results on habitat selection in the central Balkans needs to be further examined with a larger sample size, considering the relatively low explanatory power of our final model. In addition, future studies should incorporate additional predictor variables, such as prey availability and abundance, predation pressure, competition with other nocturnal rodent predators, and human disturbance, into the modelling process.
Daytime roosting sites are an integral part of habitat selection, as location choice, position, and features could significantly impact survival and overall fitness. For example, in ideal conditions, these sites help diminish owls’ exposure to harsh weather conditions and potential predators [111,112,113]. In Western Serbia, tawny owls showed a clear preference for well-sheltered microlocations. Most roosts were found in tree cavities or inside buildings, clustered in core areas. The significantly greater roosting height detected in hilly areas compared to lowland likely reflects the structural characteristics of the various habitat elements in the study area. We assume that taller and thicker trees in these areas provide safer resting places and a better vantage point of the surroundings. This behaviour may also be influenced by reduced human disturbance or, more likely, by the higher availability of suitable natural sites in hilly landscapes. The height of the roosting site in natural cavities in our study area is in agreement with the findings of a detailed study from Poland but is, on average, significantly higher than in the temperate forests of Ukraine [113]. These regional differences suggest that the diversity, age, and density of tree species dictate owl roosting site choice in natural environments, further accenting their broad phenotypic plasticity regarding habitat requirements.

5. Conclusions

Our study represents the first analysis of home range and habitat selection for tawny owls in Southeastern Europe. We found high individual variability in core area and home range size. While forest habitats are traditionally considered essential for tawny owl reproduction and survival, our study demonstrates that other habitat types, such as cultivated lands and suburban areas, can play important roles, at least in certain regions and periods of the year. These findings underline the ecological adaptability of tawny owls as they adjust to various landscapes, especially in areas altered by human activities. Future research should focus on integrating additional predictors, such as prey abundance, habitat structure, or climate and human disturbance, to capture the complexity of habitat selection and assess how these factors interact with forest availability in shaping tawny owl home ranges.

Author Contributions

Conceptualisation, D.Z.R. and D.S.; methodology, D.Z.R., D.S. and D.S.S.; software, D.Z.R. and D.S.S.; validation, D.S.S., M.V.A.S. and S.S.; formal analysis, D.Z.R., D.S., D.S.S. and J.Š.; investigation, D.Z.R. and S.S.; resources, D.Z.R.; data curation, D.Z.R., D.S. and J.Š.; writing—original draft preparation, D.Z.R. and D.S.; writing—review and editing, J.Š., S.S., D.S.S. and M.V.A.S.; visualisation, D.Z.R.; supervision, S.S. and M.V.A.S.; project administration, D.Z.R.; funding acquisition, D.Z.R., S.S. and M.V.A.S. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (grants 451-03-66/2024-03/200053 and 451-03-65/2024-03/200102).

Institutional Review Board Statement

Capturing, ringing, and tagging procedures were conducted following the guidelines and regulations set by the Ministry of Environmental Protection of Serbia (MEPS) under birds ringing license and research permits (ref No 353-01-559/2023-04 and 353-01-117/2023-04) issued by the MEPS. Additionally, supervision by the MEPS inspectors was provided during the fieldwork to ensure ethical and welfare standards.

Informed Consent Statement

Not applicable.

Data Availability Statement

The GPS fixes and other raw data supporting the results and conclusions of this article will be made available by the corresponding author upon request.

Acknowledgments

We thank Vladimir Marković (University of Novi Sad, Faculty of Sciences, Department of Geography, Tourism and Hotel Management) for his help in GIS preparation and analysis. Special thanks to Jovana Koturov for her help with mapping, capturing, and tagging owls in the field. In memory of the late Peter S. Shurulinkov (Bulgarian Academy of Sciences).

Conflicts of Interest

The authors declare no conflicts of interest. The funder had no role in study design, data collection and analysis, or preparation of the manuscript.

References

  1. Morris, D.W. Toward an ecological synthesis: A case for habitat selection. Oecologia 2003, 136, 1–13. [Google Scholar] [CrossRef]
  2. Begon, M.; Townsend, C.R.; Harper, J.L. Ecology: From Individuals to Ecosystems, 4th ed.; Blackwell Publishing: Oxford, UK, 2006; pp. 31–45. [Google Scholar]
  3. Cassini, M.H. Distribution Ecology: From Individual Habitat Use to Species Biogeographical Range, 1st ed.; Springer: New York, NY, USA, 2013; pp. 4–13. [Google Scholar]
  4. Tilman, D.; Lehman, C.L.; Kareiva, P. Population Dynamics in Spatial Habitats; Tilman, D., Kareiva, P., Eds.; Spatial Ecology: The Role of Space in Population Dynamics and Interspecific Interactions; Princeton University Press: Princeton, NJ, USA, 2018; pp. 3–20. [Google Scholar]
  5. Jarić, I.; Lennox, R.J.; Prchalová, M.; Monk, C.T.; Říha, M.; Nathan, R.; Arlinghaus, R. The power and promise of interdisciplinary international research networks to advance movement ecology. Mov. Ecol. 2023, 11, 67. [Google Scholar] [CrossRef] [PubMed]
  6. Silva, I.; Crane, M.; Suwanwaree, P.; Strine, C.; Goode, M. Using dynamic Brownian Bridge Movement Models to identify home range size and movement patterns in king cobras. PLoS ONE 2018, 13, e0203449. [Google Scholar] [CrossRef] [PubMed]
  7. Slaght, J.C.; Surmach, S.G. Blakiston’s Fish-owl Bubo blakistoni and logging: Applying resource selection information to endangered species conservation in Russia. Bird Conserv. Int. 2016, 26, 214–224. [Google Scholar] [CrossRef]
  8. Horne, J.S.; Fieberg, J.; Börger, L.; Rachlow, J.L.; Calabrese, J.M.; Fleming, C.H. Animal home ranges: Concepts, uses, and estimation. In Population Ecology: In Practice, 1st ed.; Murray, D.L., Sandercock, B.K., Eds.; Wiley Blackwell: New York, NY, USA, 2020; pp. 315–332. [Google Scholar]
  9. Barnosky, A.D.; Hadly, E.A.; Bascompte, J.; Berlow, E.L.; Brown, J.H.; Fortelius, M.; Getz, W.M.; Harte, J.; Hastings, A.; Marquet, P.A.; et al. Approaching a state shift in Earth’s biosphere. Nature 2012, 486, 52–58. [Google Scholar] [CrossRef]
  10. Ellis, E.C.; Kaplan, J.O.; Fuller, D.Q.; Vavrus, S.; Goldewijk, K.K.; Verburg, P.H. Used planet: A global history. Proc. Natl. Acad. Sci. USA 2013, 110, 7978–7985. [Google Scholar] [CrossRef] [PubMed]
  11. Burt, W.H. Territoriality and home range concepts as applied to mammals. J. Mammal. 1943, 24, 346–352. [Google Scholar] [CrossRef]
  12. McNab, B.K. Bioenergetics and the Determination of Home Range Size. Am. Nat. 1963, 97, 133–140. [Google Scholar] [CrossRef]
  13. Börger, L.; Dalziel, B.D.; Fryxell, J.M. Are there general mechanisms of animal home range behaviour? A review and prospects for future research. Ecol. Lett. 2008, 11, 637–650. [Google Scholar] [CrossRef]
  14. Maher, C.R.; Lott, D.F. A Review of Ecological Determinants of Territoriality within Vertebrate Species. Am. Midl. Nat. 2000, 143, 1–29. [Google Scholar] [CrossRef]
  15. Newton, I. Population Limitation in Birds, 1st ed.; Academic Press: London, UK, 1998; pp. 45–94. [Google Scholar]
  16. Watson, J. The Golden Eagle, 2nd ed.; Bloomsbury Publishing: London, UK, 2010; pp. 126–133. [Google Scholar]
  17. Johansson, Ö.; Rauset, G.R.; Samelius, G.; McCarthy, T.; Andrén, H.; Tumursukh, L.; Mishra, C. Land sharing is essential for snow leopard conservation. Biol. Conserv. 2016, 203, 1–7. [Google Scholar] [CrossRef]
  18. Jacobson, O.T.; Crofoot, M.C.; Perry, S.; Kosmas, H.; Barrett, B.J.; Finerty, G. The Importance of Representative Sampling for Home Range Estimation in Field Primatology. Int. J. Primatol. 2024, 45, 213–245. [Google Scholar] [CrossRef]
  19. Bühler, R.; Schalcher, K.; Séchaud, R.; Michler, S.; Apolloni, N.; Roulin, A.; Almasi, B. Influence of prey availability on habitat selection during the non-breeding period in a resident bird of prey. Mov. Ecol. 2023, 11, 14. [Google Scholar] [CrossRef]
  20. Williams, P.J.; Gutiérrez, R.J.; Whitmore, S.A. Home range and habitat selection of spotted owls in the central Sierra Nevada. J. Wildl. Manag. 2011, 75, 333–343. [Google Scholar] [CrossRef]
  21. Forsman, E.D. Home Range and Habitat Selection by Northern Spotted Owls on the Eastern Slope of the Cascade Mountains, Washington. JRR 2015, 49, 109–128. [Google Scholar] [CrossRef]
  22. Monsarrat, S.; Benhamou, S.; Sarrazin, F.; Bessa-Gomes, C.; Bouten, W.; Duriez, O. How Predictability of Feeding Patches Affects Home Range and Foraging Habitat Selection in Avian Social Scavengers? PLoS ONE 2013, 8, e53077. [Google Scholar] [CrossRef]
  23. Krausman, P.R. Some Basic Principles of Habitat Use. In Proceedings—Grazing Behaviour of Livestock and Wildlife; Launchbaugh, K.L., Sanders, K.D., Mosley, J.L., Eds.; Idaho Forest, Wildlife and Range Experiment Station Collection; University of Idaho: Moscow, ID, USA, 1999; pp. 85–90. [Google Scholar]
  24. Rutz, C. Home range size, habitat use, activity patterns and hunting behaviour of urban-breeding Northern Goshawks Accipiter gentilis. Ardea 2006, 94, 185–202. [Google Scholar]
  25. Lok, T.; van der Geest, M.; de Goeij, P.; Rakhimberdiev, E.; Piersma, T. Sex-specific nest attendance rhythm and foraging habitat use in a colony-breeding waterbird. Behav. Ecol. 2024, 35, arae020. [Google Scholar] [CrossRef]
  26. Lowe, A.; Rogers, A.C.; Durrant, K.L. Effect of human disturbance on long-term habitat use and breeding success of the European Nightjar, Caprimulgus europaeus. ACE 2014, 9, 6. [Google Scholar] [CrossRef]
  27. Yuan, B.D.; Jiang, A.W.; Li, X.D.; Lu, C.H. Roost habitat selection by birds: A review. Chin. J Ecol. 2012, 31, 2145–2151. [Google Scholar]
  28. Morant, J.; Abad-Gómez, J.M.; Álvarez, T.; Sánchez, Á.; Zuberogoitia, I.; López-López, P. Winter movement patterns of a globally endangered avian scavenger in south-western Europe. Sci. Rep. 2020, 10, 17690. [Google Scholar] [CrossRef] [PubMed]
  29. Grinnell, J. Field tests of theories concerning distributional control. Am. Nat. 1917, 51, 115–128. [Google Scholar] [CrossRef]
  30. Svärdson, G. Competition and habitat selection in birds. Oikos 1949, 1, 157–174. [Google Scholar] [CrossRef]
  31. Hildén, O. Habitat selection in birds: A review. Ann. Zool. Fenn. 1965, 2, 53–75. [Google Scholar]
  32. Kenward, R.E. Wildlife Radio Tagging: Equipment, Field Techniques and Data Analysis, 1st ed.; Academic Press: London, UK, 1987; pp. vii–ix. [Google Scholar]
  33. Kays, R.; Crofoot, M.C.; Jetz, W.; Wikelski, M. Terrestrial animal tracking as an eye on life and planet. Science 2015, 348, aaa2478. [Google Scholar] [CrossRef]
  34. Benson, E.S. Trackable life: Data, sequence, and organism in movement ecology. Stud. Hist. Philos. Sci. Part C Stud. Hist. Philos. Biol. Biomed. Sci. 2016, 57, 137–147. [Google Scholar] [CrossRef]
  35. Hebblewhite, M.; Haydon, D.T. Distinguishing technology from biology: A critical review of the use of GPS telemetry data in ecology. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2010, 365, 2303–2312. [Google Scholar] [CrossRef] [PubMed]
  36. Patterson, T.A.; Thomas, L.; Wilcox, C.; Ovaskainen, O.; Matthiopoulos, L. State–space models of individual animal movement. Trends Ecol. Evol. 2008, 23, 87–94. [Google Scholar] [CrossRef]
  37. García-Ripollés, C.; López-López, P.; Urios, V. Ranging behaviour of non-breeding Eurasian Griffon Vultures Gyps fulvus: A GPS-telemetry study. Acta Ornithol. 2011, 46, 127–134. [Google Scholar] [CrossRef]
  38. Vignali, S.; Lörcher, F.; Hegglin, D.; Arlettaz, R.; Braunisch, V. Modelling the habitat selection of the bearded vulture to predict areas of potential conflict with wind energy development in the Swiss Alps. GECCO 2021, 25, e01405. [Google Scholar] [CrossRef]
  39. Morollón, S.; Urios, V.; López-López, P. Home-Range Size and Space Use of Territorial Bonelli’s Eagles (Aquila fasciata) Tracked by High-Resolution GPS/GSM Telemetry. Diversity 2022, 14, 1082. [Google Scholar] [CrossRef]
  40. Ronconi, R.A.; Lieske, D.J.; McFarlane Tranquilla, L.A.; Abbott, S.; Allard, K.A.; Allen, B.; Black, A.L.; Bolduc, F.; Davoren, G.K.; Diamond, A.W.; et al. Predicting Seabird Foraging Habitat for Conservation Planning in Atlantic Canada: Integrating Telemetry and Survey Data Across Thousands of Colonies. Front. Mar. Sci. 2022, 9, 816794. [Google Scholar] [CrossRef]
  41. König, C.; Weick, F.; Becking, J.-H. Owls—A Guide to the Owls of the World, 1st ed.; Pica Press: Sussex, UK, 1999; pp. 333–335. [Google Scholar]
  42. Butyev, V.T.; Zubkov, N.I.; Ivanchev, V.P.; Koblik, E.A.; Kovshar, A.F.; Kotyukov, Y.V.; Lyuleeva, D.S.; Nazarov, Y.N.; Nechaev, V.A.; Priklonsky, S.G.; et al. Strigiformes, Caprimulgiformes, Apodiformes, Coraciiformes, Upupiformes, Piciformes; KMK Scientific Publications: Moscow, Russia, 2005; p. 487. [Google Scholar]
  43. Mikkola, H. Owls of the Europe; T&AD Poyser: Calton UK, 1983; pp. 136–156. [Google Scholar]
  44. BirdLife International. Strix aluco (Europe assessment). The IUCN Red List of Threatened Species 2021: e.T22725469A166431700. Available online: https://dx.doi.org/10.2305/IUCN.UK.2021-3.RLTS.T22725469A166431700.en (accessed on 18 November 2024).
  45. Brito, P.H. The influence of Pleistocene glacial refugia on tawny owl genetic diversity and phylogeography in western Europe. Mol. Ecol. 2005, 14, 3077–3094. [Google Scholar] [CrossRef] [PubMed]
  46. Derlink, M.; Wernham, C.; Bertoncelj, I.; Kovács, A.; Saurola, P.; Duke, G.; Movalli, P.; Vrezec, A. A review of raptor and owl monitoring activity across Europe: Its implications for capacity building towards pan-European monitoring. Bird Study 2018, 65 (Suppl. S1), S4–S20. [Google Scholar] [CrossRef]
  47. Obuch, J. Spatial and temporal diversity of the diet of the tawny owl (Strix aluco). Slovak Raptor J. 2011, 5, 1–120. [Google Scholar] [CrossRef]
  48. Petty, S.J.; Shaw, G.; Anderson, D.I.K. Value of nest boxes for population studies and conservation of owls in coniferous forests in Britain. J. Raptor Res. 1994, 28, 134–142. [Google Scholar]
  49. Southern, H.N. The natural control of a population of tawny owls (Strix aluco). J. Zool. 1970, 162, 197–285. [Google Scholar] [CrossRef]
  50. Coles, C.F. Breeding, Survival, Movements and Foraging of Tawny Owls Strix aluco in a Managed Spruce Forest: A Spatial Approach. Ph.D. Thesis, Durham University, Durham, UK, October 2000. [Google Scholar]
  51. Solonen, T.; af Ursin, K. Breeding of Tawny Owls Strix aluco in rural and urban habitats in southern Finland. Bird Study 2008, 55, 216–221. [Google Scholar] [CrossRef]
  52. Redpath, S.M. Censusing Tawny Owls Strix aluco by the use of imitation calls. Bird Study 1994, 41, 192–198. [Google Scholar] [CrossRef]
  53. Zuberogoitia, I.A.; Martínez Clement, J.A. Methods for surveying Tawny Owl Strix aluco populations in large areas. Biota 2000, 1, 79–94. [Google Scholar]
  54. Vrezec, A.; Bertoncelj, I. Territory monitoring of Tawny Owls Strix aluco using playback calls is a reliable population monitoring method. Bird Study 2018, 65 (Suppl. S1), S52–S62. [Google Scholar] [CrossRef]
  55. Astaras, C.; Valeta, C.; Vasileiadis, I. Acoustic ecology of tawny owl (Strix aluco) in the Greek Rhodope Mountains using passive acoustic monitoring methods. Folia Oecologica 2022, 49, 110–116. [Google Scholar] [CrossRef]
  56. Pagaldai, N.; Arizaga, J.; Jiménez-Franco, M.V.; Zuberogoitia, I. Colonization of Urban Habitats: Tawny Owl Abundance Is Conditioned by Urbanization Structure. Animals 2021, 11, 2954. [Google Scholar] [CrossRef] [PubMed]
  57. López-Peinado, A.; Lis, Á.; Perona, A.M.; López-López, P. Habitat Preferences of the Tawny Owl (Strix aluco) in a Special Conservancy Area of Eastern Spain. J. Raptor Res. 2000, 54, 402–413. [Google Scholar] [CrossRef]
  58. Redpath, S.M. Habitat fragmentation and the individual: Tawny owls Strix aluco in woodland patches. J. Anim. Ecol. 1995, 64, 652–661. [Google Scholar] [CrossRef]
  59. Burgos, G.; Zuberogoitia, I. A telemetry study to discriminate between home range and territory size in Tawny Owls. Bioacoustics 2018, 29, 109–121. [Google Scholar] [CrossRef]
  60. Sunde, P.; Bølstad, M.S. A telemetry study of the social organization of a tawny owl (Strix aluco) population. J. Zool. 2004, 263, 65–76. [Google Scholar] [CrossRef]
  61. Sunde, P. What do we know about territorial behaviour and its consequences in Tawny Owls? In Ecology and Conservation of European Forest-Dwelling Raptors; Zuberogoitia, I., Je, M., Eds.; Diputación Foral de Bizkaia: Bilbao, Spain, 2011; pp. 253–260. [Google Scholar]
  62. Gajić, M.; Vujadinović, S.; Šabić, D. Contemporary demographic and functional transformations in the settlement net of Jadar. Glas. Srp. Geogr. Drus. 2011, 91, 43–68. [Google Scholar] [CrossRef]
  63. RHMZ. Meteorološki Godišnjak 1: Klimatološki Podaci; Republički Hidrometeorološki Zavod: Beograd, Serbia, 2022; p. 168. [Google Scholar]
  64. Lengagne, T.; Slater, P.J.B. The effects of rain on acoustic communication: Tawny owls have good reason for calling less in wet weather. Proc. R. Soc. B Biol. Sci. 2002, 269, 2121–2125. [Google Scholar] [CrossRef]
  65. Worthington-Hill, J.; Conway, G. Tawny Owl Strix aluco response to call-broadcasting and implications for survey design. Bird Study 2017, 64, 205–210. [Google Scholar] [CrossRef]
  66. Redpath, S.M.; Wyllie, I. Traps for Capturing Territorial Owls. J. Raptor Res. 1994, 28, 115–117. [Google Scholar]
  67. Demongin, L. Identification Guide to Birds in the Hand, 1st ed.; Laurent Demongin: Beauregard-Vendon, France, 2016; pp. 193–194. [Google Scholar]
  68. Kenward, R.E. Raptor Radio-Tracking and Telemetry; ICBP Technical Publication 5; ICBP: Cambridge, UK, 1985; pp. 409–420. [Google Scholar]
  69. Buehler, D.A.; Fraser, J.D.; Fuller, M.R.; McAllister, L.S.; Seegar, J.K.D. Captive and field-tested radio transmitter attachments for Bald Eagles. J. Field Ornithol. 1995, 66, 173–180. [Google Scholar]
  70. Jones, T.M.; Cooper, N.W.; Haradon, H.A.; Brunner, A.R.; Dossman, B.C.; Ward, M.P.; Sillett, T.S.; Kaiser, S.A. Considerations for radio-transmitter specifications on songbirds: Color and antenna length matter too. J. Field Ornithol. 2024, 95, 7. [Google Scholar] [CrossRef]
  71. Barron, D.G.; Brawn, J.D.; Weatherhead, P.J. Meta-analysis of transmitter effects on avian behaviour and ecology. Methods Ecol. Evol. 2010, 1, 180–187. [Google Scholar] [CrossRef]
  72. Gaunt, A.S.; Oring, L.W. Guidelines to the Use of Wild Birds in Research, 2nd ed.; Ornithological Council: Washington, DC, USA, 1999; p. 52. [Google Scholar]
  73. Geen, G.R.; Robinson, R.A.; Baillie, S.R. Effects of tracking devices on individual birds—A review of the evidence. J. Avian Biol. 2019, 50. [Google Scholar] [CrossRef]
  74. Sunde, P. Effects of Backpack Radio Tags on Tawny Owls. J. Wildl. Manag. 2006, 70, 594–599. [Google Scholar] [CrossRef]
  75. Walter, W.D.; Fischer, J.W.; Baruch-Mordo, S.; VerCauteren, K.C. What is the proper method to delineate home range of an animal using today’s advanced GPS telemetry systems: The initial step. Mod. Telem. 2011, 68, 249–268. [Google Scholar]
  76. Fleming, C.H.; Drescher-Lehman, J.; Noonan, M.J.; Akre, T.S.B.; Brown, D.J.; Cochrane, M.M.; Dejid, N.; DeNicola, V.; DePerno, C.S.; Dunlop, J.N.; et al. A comprehensive framework for handling location error in animal tracking data. bioRxiv 2021. [Google Scholar] [CrossRef]
  77. Mohr, C.O. Table of Equivalent Populations of North American Small Mammals. Am. Nat. 1947, 37, 223–249. [Google Scholar] [CrossRef]
  78. Worton, B.J. A review of models of home range for animal movement. Ecol. Model. 1987, 38, 277–298. [Google Scholar] [CrossRef]
  79. Fleming, C.H.; Sheldon, D.; Fagan, W.F.; Leimgruber, P.; Mueller, T.; Nandintsetseg, D.; Noonan, M.J.; Olson, K.A.; Setyawan, E.; Sianipar, A.; et al. Correcting for missing and irregular data in home-range estimation. Ecol. App. 2018, 28, 1003–1010. [Google Scholar] [CrossRef] [PubMed]
  80. Burgman, M.A.; Fox, J.C. Bias in species range estimates from minimum convex polygons: Implications for conservation and options for improved planning. Anim. Conserv. 2003, 6, 19–28. [Google Scholar] [CrossRef]
  81. Walter, W.D.; Onorato, D.P.; Fischer, J.W. Is there a single best estimator? Selection of home range estimators using area-under-the-curve. Mov. Ecol. 2015, 3, 10. [Google Scholar] [CrossRef]
  82. van Winkle, W. Comparison of several probabilistic home-range models. J. Wildl. Manag. 1975, 39, 118–123. [Google Scholar] [CrossRef]
  83. Worton, B.J. Kernel methods for estimating the utilization distribution in home-range studies. Ecology 1989, 70, 164–168. [Google Scholar] [CrossRef]
  84. Fleming, C.H.; Fagan, W.F.; Mueller, T.; Olson, K.A.; Leimgruber, P.; Calabrese, J.M. Rigorous home range estimation with movement data: A new autocorrelated kernel density estimator. Ecology 2015, 96, 1182–1188. [Google Scholar] [CrossRef] [PubMed]
  85. Noonan, M.J.; Tucker, M.A.; Fleming, C.H.; Akre, T.S.; Alberts, S.C.; Ali, A.H.; Altmann, J.; Antunes, P.C.; Belant, J.L.; Beyer, D.; et al. A comprehensive analysis of autocorrelation and bias in home range estimation. Ecol. Monogr. 2019, 89, e01344. [Google Scholar] [CrossRef]
  86. Silva, I.; Fleming, C.H.; Noonan, M.J.; Alston, J.; Folta, C.; Fagan, W.F.; Calabrese, J.M. Autocorrelation-informed home range estimation: A review and practical guide. Methods Ecol. Evol. 2022, 13, 534–544. [Google Scholar] [CrossRef]
  87. Calabrese, J.M.; Fleming, C.H.; Gurarie, E. ctmm: An R package for analyzing animal relocation data as a continuous-time stochastic process. Methods Ecol. Evol. 2016, 7, 1124–1132. [Google Scholar] [CrossRef]
  88. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2023; Available online: https://www.R-project.org/ (accessed on 17 January 2024).
  89. ctmmweb: A Shiny Web App for the ctmm Movement Analysis Package. Available online: https://github.com/ctmm-initiative/ctmmweb (accessed on 25 August 2024).
  90. Calabrese, J.M.; Fleming, C.H.; Noonan, M.J.; Dong, X. ctmmweb: A Graphical User Interface for Autocorrelation-Informed Home Range Estimation. Wildl. Soc. Bull. 2021, 45, 162–169. [Google Scholar] [CrossRef]
  91. Burnham, K.P.; Anderson, D.R. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd ed.; Springer: Berlin/Heidelberg, Germany, 2002; pp. 72–132. [Google Scholar]
  92. Akaike, H. A new look at the statistical model identification. IEEE Trans. Autom. Control 1974, 19, 716–723. [Google Scholar] [CrossRef]
  93. Hijmans, R.J. Geosphere: Spherical Trigonometry. 2022. Available online: https://CRAN.R-project.org/package=geosphere (accessed on 19 July 2024).
  94. Fieberg, J.; Kochanny, C.O. Quantifying Home-Range Overlap: The Importance of the Utilization Distribution. J. Wildl. Manag. 2005, 69, 1346–1359. [Google Scholar] [CrossRef]
  95. Winner, K.; Noonan, M.J.; Fleming, C.H.; Olson, K.; Mueller, T.; Sheldon, D.; Calabrese, J.M. Statistical inference for home range overlap. Methods Ecol. Evol. 2018, 9, 1679–1691. [Google Scholar] [CrossRef]
  96. Zvidzai, M.; Zengeya, F.M.; Masocha, M.; Ndaimani, H.; Murwira, A. Application of GPS occurrence data to understand African white-backed vultures Gyps africanus spatial home range overlaps. Ecol. Evol. 2022, 12, e8778. [Google Scholar] [CrossRef]
  97. Johnson, D.H. The comparison of usage and availability measurements for evaluating resource preference. Ecology 1980, 61, 65–71. [Google Scholar] [CrossRef]
  98. Lele, S.R.; Keim, J.L. Weighted Distributions and Estimation of Resource Selection Probability Functions. Ecology 2006, 87, 3021–3028. [Google Scholar] [CrossRef]
  99. Nemes, S.; Hartel, T. Summary measures for binary classification systems in animal ecology. North-West. J. Zool. 2010, 6, 323–330. [Google Scholar]
  100. Maindonald, J.H.; John Braun, W. DAAG: Data Analysis and Graphics Data and Functions Version 1.25.6. 2024. Available online: https://cran.r-project.org/web/packages/DAAG/index.html (accessed on 22 July 2024).
  101. Jennions, M.D.; Møller, A.P. A survey of the statistical power of research in behavioral ecology and animal behavior. Behav. Ecol. 2003, 14, 438–445. [Google Scholar] [CrossRef]
  102. Kouba, M.; Bartoš, L.; Štastný, K. Differential Movement Patterns of Juvenile Tengmalms Owls (Aegolius funereus) during the Post-Fledging Dependence Period in Two Years with Contrasting Prey Abundance. PLoS ONE 2013, 8, e67034. [Google Scholar] [CrossRef]
  103. Belthoff, J.R.; Sparks, E.J.; Ritchison, G. Home ranges of adult and juvenile Eastern Screech-owls: Size, seasonal variation and extent of overlap. J. Raptor Res. 1993, 27, 8–15. [Google Scholar]
  104. Marzluff, J.M.; Kimsey, B.A.; Schueck, L.S.; McFadzen, M.E.; Vekasy, M.S.; Bednarz, J.C. The Influence of Habitat, Prey Abundance, Sex, and Breeding Success on the Ranging Behavior of Prairie Falcons. Condor 1997, 99, 567–584. [Google Scholar] [CrossRef]
  105. Mirski, P.; Grosberg, J.; Kull, T.; Mellov, P.; Tõnisalu, G.; Väli, V.; Väli, Ü. Movement of avian predators points to biodiversity hotspots in agricultural landscape. R. Soc. Open Sci. 2024, 11, 231543. [Google Scholar] [CrossRef] [PubMed]
  106. La Haye, M.; Mertens, F.; Nieuwenweg, S.; Baeyens, G. Habitat use of Tawny Owls Strix aluco in the Amsterdam Waterwork Dunes, as shown by telemetry. Limosa 1995, 68, 73–76. [Google Scholar]
  107. Choi, W.S.; Sung, H.C.; Park, J.C.; Kim, W.Y. A first Home range and habitat characteristics of a breeding pair tawny owl: Case study using direct tracking in the Korean Peninsula. J. Asia-Pac. Biodivers. 2020, 13, 169–174. [Google Scholar] [CrossRef]
  108. Sunde, P.; Redpath, S.M. Combining information from range use and habitat selection: Sex-specific spatial responses to habitat fragmentation in tawny owls Strix aluco. Ecography 2006, 29, 152–158. [Google Scholar] [CrossRef]
  109. Vrezec, A.; Tome, D. Altitudinal segregation between Ural Owl Strix uralensis and Tawny Owl, S. aluco: Evidence for competitive exclusion in raptorial birds. Bird Study 2004, 51, 264–269. [Google Scholar] [CrossRef]
  110. Đorđević, N.; Popović, Z.; Beuković, D.; Beuković, M.; Đorđević, M. The Importance of Arable Land in Serbia to the Feed Pheasant and Brown Hare and Number of Population. In Proceedings of the XXVI Conference of Agronomists, Veterinarians, Technologists andAgricultural Economists, Belgrade, Serbia, 22–23 February 2012. [Google Scholar]
  111. Pruitt, M.L. Thermal benefits of roost site selection in a small forest owl. Wilson J. Ornithol. 2024, 135, 520–532. [Google Scholar] [CrossRef]
  112. Sunde, P.; Bølstad, M.S.; Desfor, K.B. Diurnal exposure as a risk sensitive behaviour in tawny owls Strix aluco? J. Avian Biol. 2003, 34, 409–418. [Google Scholar] [CrossRef]
  113. Yatsiuk, Y.; Wesołowski, T. Diversity and abundance of large tree holes used by Tawny Owls Strix aluco in lowland temperate forests. Bird Study 2021, 67, 331–343. [Google Scholar] [CrossRef]
Animals 14 03338 g001
Figure 1. Geographical position and map of the study area, Jadar and Radjevina microregions, Podrinje area, Serbia. The GPS positions unique to each individual male tawny owl are represented by coloured dots. The arrow on the inset map indicates the position of the study area.
Figure 1. Geographical position and map of the study area, Jadar and Radjevina microregions, Podrinje area, Serbia. The GPS positions unique to each individual male tawny owl are represented by coloured dots. The arrow on the inset map indicates the position of the study area.
Animals 14 03338 g001
Animals 14 03338 g002
Figure 2. Grey (A) and rufous (B) colour morphs of male tawny owls equipped with X-attached VHF-GPS transmitters, photographed just before release.
Figure 2. Grey (A) and rufous (B) colour morphs of male tawny owls equipped with X-attached VHF-GPS transmitters, photographed just before release.
Animals 14 03338 g002
Animals 14 03338 g003
Figure 3. Non-breeding home range estimates for seven tawny owl males, Strix aluco, tracked in Jadar and Radjevina (Podrinje area, Western Serbia) during the period of autumn–winter of 2023. The outer boundary represents the 95% AKDE, while the inner polygon delineates the 50% AKDE core area.
Figure 3. Non-breeding home range estimates for seven tawny owl males, Strix aluco, tracked in Jadar and Radjevina (Podrinje area, Western Serbia) during the period of autumn–winter of 2023. The outer boundary represents the 95% AKDE, while the inner polygon delineates the 50% AKDE core area.
Animals 14 03338 g003
Animals 14 03338 g004
Figure 4. Comparison of core area (50% AKDE) and home range (95% AKDE) estimates between adult (Ad) and first-year (Juv) tawny owl males, Strix aluco, during the non-breeding season in Jadar and Radjevina microregions, Podrinje area, Western Serbia.
Figure 4. Comparison of core area (50% AKDE) and home range (95% AKDE) estimates between adult (Ad) and first-year (Juv) tawny owl males, Strix aluco, during the non-breeding season in Jadar and Radjevina microregions, Podrinje area, Western Serbia.
Animals 14 03338 g004
Table 1. Summary data of satellite-tracked tawny owls (Strix aluco) in the study area. Ad—adult individual (hatched before study year), Juv—juvenile individual (hatched during study year); HFL—hilly forested landscape; LAL—lowland agricultural landscape with scattered forest patches.
Table 1. Summary data of satellite-tracked tawny owls (Strix aluco) in the study area. Ad—adult individual (hatched before study year), Juv—juvenile individual (hatched during study year); HFL—hilly forested landscape; LAL—lowland agricultural landscape with scattered forest patches.
Individual IDCapture DateAgeSexBody Mass (g)HabitatCollected Fixes (n)Tag ModelTransmission Period (Days)
Male_5697728 October 2023AdMale488HFL26924055
Male_5697828 October 2023JuvMale474HFL26424054
Male_5697929 October 2023JuvMale446HFL18024046
Male_5698029 October 2023JuvMale417HFL28124058
Male_5698130 October 2023AdMale486LAL21312045
Male_5698230 October 2023JuvMale436LAL16212037
Male_569832 November 2023AdMale467LAL15712035
Table 2. Core area and home range estimates for seven tawny owl, Strix aluco, males in the Jadar and Radjevina microregions (Podrinje area, Western Serbia), based on Minimum Convex Polygon (MCP, 95%) and Autocorrelated Kernel Utilisation Distribution (AKDE, 95% and 50%).
Table 2. Core area and home range estimates for seven tawny owl, Strix aluco, males in the Jadar and Radjevina microregions (Podrinje area, Western Serbia), based on Minimum Convex Polygon (MCP, 95%) and Autocorrelated Kernel Utilisation Distribution (AKDE, 95% and 50%).
Individual IDMCP (ha)AKDE Core Area (ha)AKDE Home Range (ha)
Male_5697738.609.9556.36
Male_5697864.3320.80100.84
Male_56979170.5040.43238.08
Male_56980285.6587.33462.23
Male_5698160.6512.5979.22
Male_56982254.5289.74389.85
Male_56983225.0522.20197.87
Table 3. Home range (95% AKDE) overlaps of male tawny owls Strix aluco for the 21 dyads at the 95% confidence level.
Table 3. Home range (95% AKDE) overlaps of male tawny owls Strix aluco for the 21 dyads at the 95% confidence level.
DyadIndividual IIndividual IIHome Range Overlap95% CI
1Male_56977Male_5697800
2Male_56977Male_569790.020.01–0.04
3Male_56977Male_569800.320.27–0.37
4Male_56977Male_569810-
5Male_56977Male_569820-
6Male_56977Male_569830-
7Male_56978Male_569790.450.37–0.53
8Male_56978Male_569800.110.08–0.15
9Male_56978Male_569810-
10Male_56978Male_569820-
11Male_56978Male_569830-
12Male_56979Male_569800.430.34–0.52
13Male_56979Male_569810-
14Male_56979Male_569820-
15Male_56979Male_569830-
16Male_56980Male_569810-
17Male_56980Male_569820-
18Male_56980Male_569830-
19Male_56981Male_569820-
20Male_56981Male_569830.030.02–0.04
21Male_56982Male_569830-
Table 4. The Akaike information criterion (AIC) and performance metrics for tawny owls’ Strix aluco habitat selection models in Jadar and Radjevina microregion, Podrinje area, Western Serbia.
Table 4. The Akaike information criterion (AIC) and performance metrics for tawny owls’ Strix aluco habitat selection models in Jadar and Radjevina microregion, Podrinje area, Western Serbia.
Model NoModel VariablesAICΔiAICAkaike w iAUCCV (%)
ICultivated land3926.1272.3360.1190.5950.602
IISuburban4083.266159.475<0.0010.5150.602
IIIAltitude4079.353155.562<0.0010.5450.602
IVCultivated land + Suburban3923.79103.8410.6010.602
VCultivated land + Altitude3924.9801.1890.2120.6070.602
VISuburban + Altitude4059.943136.152<0.0010.5660.602
VIICultivated land + Suburban + Altitude3924.3910.6000.2840.6070.602
AIC is Akaike’s information criterion; ΔiAIC is (AIC)i − (AIC)min; Akaike w i is the Akaike weight; AUC is the area under the curve; CV is the percent of cases accurately predicted in cross validation. The best model is bolded.
Table 5. Coefficient estimates and statistical significance for top model (Model No IV) of tawny owl Strix aluco habitat selection analysis.
Table 5. Coefficient estimates and statistical significance for top model (Model No IV) of tawny owl Strix aluco habitat selection analysis.
CovariateEstimateSEzp Value
(Intercept)0.2630.0436.117<0.001
Suburban0.3550.1732.0540.040
Cultivated land−1.2310.101−12.118<0.001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rajković, D.Z.; Stanković, D.; Šeat, J.; Stevanović, D.S.; Andrejević Stošović, M.V.; Skorić, S. Spatial Ecology of a Resident Avian Predator During the Non-Breeding Period in Managed Habitats of Southeastern Europe. Animals 2024, 14, 3338. https://doi.org/10.3390/ani14223338

AMA Style

Rajković DZ, Stanković D, Šeat J, Stevanović DS, Andrejević Stošović MV, Skorić S. Spatial Ecology of a Resident Avian Predator During the Non-Breeding Period in Managed Habitats of Southeastern Europe. Animals. 2024; 14(22):3338. https://doi.org/10.3390/ani14223338

Chicago/Turabian Style

Rajković, Draženko Z., Daliborka Stanković, Jelena Šeat, Dejan S. Stevanović, Miona V. Andrejević Stošović, and Stefan Skorić. 2024. "Spatial Ecology of a Resident Avian Predator During the Non-Breeding Period in Managed Habitats of Southeastern Europe" Animals 14, no. 22: 3338. https://doi.org/10.3390/ani14223338

APA Style

Rajković, D. Z., Stanković, D., Šeat, J., Stevanović, D. S., Andrejević Stošović, M. V., & Skorić, S. (2024). Spatial Ecology of a Resident Avian Predator During the Non-Breeding Period in Managed Habitats of Southeastern Europe. Animals, 14(22), 3338. https://doi.org/10.3390/ani14223338

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop